New in Oncology - June 25, 2026
ASCO aftershocks keep rolling: melanoma vaccines, MSI-H colon de-escalation, HER2 gastric survival, bladder ADCs, prostate intensification, and head-and-neck IO boosts—plus 32 new trials.
🏆 THIS WEEK’S BREAKTHROUGHS:
25 standouts among 102 notable findings
Melanoma
🏆 Intismeran autogene + pembrolizumab in resected melanoma: The personalized mRNA vaccine story keeps getting harder to dismiss, with 5-year RFS/DMFS benefit suggesting neoantigen vaccination can add real durability to adjuvant PD-1.
⚡️ Moderna / Merck🏆 Ipilimumab + nivolumab–relatlimab neoadjuvant triplet: Very high pCR/MPR rates show how much tumor can melt with triple checkpointing—but one fatal myocarditis case in a tiny cohort is the safety warning in neon lights.
⚡️ Bristol Myers Squibb🏆 Nivolumab + relatlimab in 1L melanoma: Five-year RELATIVITY-047 data make PD-1/LAG-3 feel like a durable middle lane between PD-1 alone and higher-toxicity CTLA-4 combinations.
⚡️ Bristol Myers Squibb
Colon / Colorectal Cancer
🏆 Ivonescimab + mFOLFOX6 in 1L unresectable mCRC: PD-1×VEGF plus chemo posts a striking ~71% ORR and 100% DCR, raising the question of whether this class can finally push immune biology into MSS-leaning CRC.
⚡️ Akeso / Summit🏆 Nivolumab + ipilimumab in MSI-H/dMMR mCRC: Long-term CheckMate 8HW data strengthen the case that dual checkpoint blockade may buy more patients durable treatment-free time than PD-1 alone.
⚡️ Bristol Myers Squibb🏆 Tislelizumab neoadjuvant in dMMR/MSI-H colon cancer: A ~79% pCR and rapid ctDNA MRD clearance push colon cancer closer to the organ-preservation logic already transforming rectal cancer.
⚡️ BeOne Medicines🏆 Onvansertib + FOLFIRI/bevacizumab in RAS-mutant mCRC: A 72% ORR and strong early PFS signal suggest PLK1 inhibition may become a real chemo-sensitizing strategy in a large RAS-mutant population.
⚡️ Cardiff Oncology
Gastric / GEJ Cancer
🏆 Zanidatamab + tislelizumab + chemotherapy in 1L HER2+ mGEA: A 26.4-month median OS and benefit across PD-L1 subgroups make zanidatamab look like a serious challenger to trastuzumab-based frontline therapy in HER2+ gastric/GEJ cancer.
⚡️ Jazz Pharmaceuticals / BeOne Medicines🏆 Durvalumab + perioperative FLOT in resectable gastric/GEJ cancer: MATTERHORN gives curative-intent gastric cancer one of its clearest perioperative immunotherapy wins yet, with EFS benefit on top of FLOT where prior IO efforts have mostly disappointed.
⚡️ AstraZeneca
Bladder Cancer
🏆 Avelumab + sacituzumab govitecan maintenance in aUC: Adding SG to avelumab extends PFS across visceral, liver, lung, bone, and nodal subgroups—but the very high grade ≥3 toxicity rate makes patient selection the key question.
⚡️ Gilead🏆 LY4052031 after enfortumab vedotin in mUC: A next-generation Nectin-4 topo-I ADC showing post-EV responses suggests the target may still be usable if the payload changes.
⚡️ Eli Lilly🏆 EV + pembrolizumab in cisplatin-ineligible MIBC: A ~57% pCR rate and strong EFS/OS subgroup signals make perioperative EV+pembro look like a true potential standard for cisplatin-ineligible patients.
⚡️ Astellas / Pfizer; Merck🏆 Erdafitinib chemoablation in FGFR3-mutant NMIBC: Short-course oral FGFR blockade producing high CR rates before TURBT suggests a possible “break-the-resection-cycle” strategy for recurrent FGFR3-mutant NMIBC.
⚡️ Johnson & Johnson
Glioma / Glioblastoma
🏆 Vorasidenib in grade 2 IDH-mutant glioma: Longer follow-up keeps reinforcing the central clinical point: IDH inhibition can buy years before RT/chemo while preserving function in carefully selected patients.
⚡️ Servier
Ovarian / Gynecologic Cancer
🏆 Azenosertib + weekly paclitaxel in platinum-resistant ovarian cancer: WEE1 inhibition may re-sensitize taxane-exposed PROC, with response rates above weekly paclitaxel expectations—though cytopenias and randomized PFS/OS are the real test.
⚡️ Zentalis Pharmaceuticals
🏆 Relacorilant + nab-paclitaxel in platinum-resistant ovarian cancer: A positive Phase 3 OS result in PROC is rare enough to matter, especially because it builds on a familiar weekly taxane backbone.
⚡️ Corcept Therapeutics🏆 TUB-040 in platinum-resistant ovarian cancer: A biomarker-unselected ~61% confirmed ORR immediately makes this NaPi2b ADC one of the most eye-catching PROC signals of the week, pending dose optimization and durability.
⚡️ Tubulis
Prostate Cancer
🏆 CAN-2409 + valacyclovir with curative EBRT: A rare positive Phase 3 immuno-gene therapy result in localized prostate cancer, with DFS benefit and no major safety penalty.
⚡️ Candel Therapeutics🏆 Apalutamide + perioperative ADT before/after prostatectomy: PROTEUS gives surgery-centered high-risk localized prostate cancer a real MFS win, though the absolute benefit and added toxicity make selection important.
⚡️ Johnson & Johnson🏆 Talazoparib + enzalutamide in HRR-altered mHSPC: TALAPRO-3 delivers a major rPFS gain in HRR-altered metastatic hormone-sensitive disease, with anemia and dose management as the practical tradeoffs.
⚡️ Pfizer / Astellas🏆 ABBV-969 in heavily pretreated mCRPC: Dual PSMA/STEAP1 ADC targeting shows high PSA50 and RECIST response activity, suggesting a possible post-taxane/post-radioligand ADC lane if marrow toxicity is manageable.
⚡️ AbbVie
Head and Neck Cancer
🏆 NBTXR3 + SBRT reirradiation followed by anti–PD-1: A radioenhancer strategy produces response rates well above historical PD-1 monotherapy benchmarks, especially in recurrent-only disease.
⚡️ Nanobiotix🏆 Pembrolizumab + definitive RT in unfavorable-risk p16+ HNSCC: KEYCHAIN raises the possibility of a cisplatin-sparing curative-intent path, though the Phase 2 design and wide confidence intervals mean Phase 3 confirmation is essential.
⚡️ Merck🏆 Tosposertib + pembrolizumab in 1L R/M HNSCC: A small but striking 75% first-line ORR makes dual TGFβ/VEGFR modulation one of the more interesting IO-boosting ideas in HNSCC.
⚡️ TiumBio🏆 Ficerafusp alfa + pembrolizumab in HPV-negative R/M HNSCC: EGFR targeting plus TGF-β trapping continues to look like a serious way to deepen PD-1 responses in HPV-negative disease.
⚡️ Bicara Therapeutics
🏆 THIS WEEK’S NEW TRIALS:
6 spotlights among 32 new trials
AML: Acute Myeloid Leukemia
🏆 Ziftomenib + olutasidenib in NPM1+/IDH1+ R/R AML: A rational double-differentiation strategy pairing menin inhibition with mutant-IDH1 inhibition in a genomically defined AML subset. (NCT07411586)
⚡️ Kura Oncology
Ovarian, Fallopian, or Primary Peritoneal Cancer
🏆 Azenosertib vs chemotherapy in Cyclin E1+ platinum-resistant ovarian cancer: WEE1 inhibition gets the biomarker-selected Phase 3 test this replication-stress population has been waiting for. (NCT07612215)
⚡️ Zentalis Pharmaceuticals
Pancreatic Ductal Adenocarcinoma (PDAC)
🏆 Atebimetinib + gem/nab-paclitaxel in 1L metastatic PDAC: A rare frontline Phase 3 targeted-therapy swing in PDAC, testing whether MEK inhibition can improve a broad chemo backbone. (NCT07562152)
⚡️ Immuneering
Non-Small Cell Lung Cancer (NSCLC)
🏆 NBTXR3 + radiation in unresectable stage I–III NSCLC: A radioenhancer strategy for patients who cannot receive chemoradiation—small concept, but a large unmet need if local control improves. (NCT07224152)
⚡️ Nanobiotix
Kidney (Renal Cell) Carcinoma
🏆 NBTXR3 + radiation in locally advanced or metastatic renal cell carcinoma: This trial explores whether a radioenhancer can improve tumor control when combined with radiotherapy in RCC, a setting where radiation has historically had limited impact. (NCT07227402)
⚡️ Nanobiotix
Breast Cancer
🏆 RxFINE-Low ribociclib guided by Oncotype DX: A pragmatic Phase 3 attempt to use genomic risk to decide who truly needs CDK4/6 escalation in early HR+/HER2− disease. (NCT07604571)
⚡️ Novartis
32 NEW TRIALS
Prostate Cancer (6) • 🏆 AML: Acute Myeloid Leukemia (1) • Hematological Disease (2) • Large B-cell Lymphoma (LBCL) (1) • Neuroendocrine Tumors (1) • Neuroblastoma (1) •
Leiomyosarcoma (smooth muscle sarcoma) (1) • 🏆 Ovarian, Fallopian, or Primary Peritoneal Cancer (2) • Uterine (Endometrial) Cancer (2) • 🏆 Pancreatic Ductal Adenocarcinoma (PDAC) (2) • Small-Cell Lung Cancer (SCLC) (1) • 🏆 Non-Small Cell Lung Cancer (NSCLC) (3) • 🏆 Kidney (Renal Cell) Carcinoma (1) • 🏆 Breast Cancer (1) • Colorectal Cancer (1) • Head and Neck Cancer (1) • Ewing Sarcoma / Osteosarcoma (1) • Multiple Solid Tumors (4)
Get a Personalized Clinical Trial Shortlist
Patients and clinicians can now get a free, confidential, oncologist-reviewed clinical trial shortlist from Sagely Health. Using Sage, our AI assistant, we gather the key details in about 5–10 minutes and send a clear, personalized report with relevant trial options—usually within 24 hours.
102 New Notable Cancer Findings
Melanoma
🏆 SPOTLIGHT
For decades, personalized cancer vaccines have been more promise than reality—Intismeran autogene, the personalized mRNA cancer vaccine, may be the study that finally changes that narrative.
Intismeran autogene + pembrolizumab sustains 5-year RFS/DMFS benefit in resected melanoma
Intismeran autogene, also known as V940 or mRNA-4157, is Moderna/Merck’s individualized neoantigen mRNA therapy encoding up to 34 patient-specific tumor neoantigens, designed to broaden tumor-specific T-cell responses when combined with PD-1 blockade. In the randomized Phase IIb KEYNOTE-942 trial (NCT03897881), patients with completely resected high-risk stage IIIB–IV cutaneous melanoma received intismeran autogene 1 mg IM Q3W for up to 9 doses plus pembrolizumab 200 mg IV Q3W for up to 18 cycles (n=107) vs pembrolizumab alone (n=50); at ~5 years of planned follow-up, the combination sustained improved RFS (HR 0.510, 95% CI 0.294–0.887) and DMFS (HR 0.411, 95% CI 0.200–0.843), with an OS trend (HR 0.471, 95% CI 0.165–1.345). Safety remained manageable with no new signals reported, and the translational data showed increased TCR clonality/novel clonotypes, which matters because adjuvant pembrolizumab alone is already active in resected stage III melanoma, improving RFS vs placebo in KEYNOTE-054 with HR 0.57.
Take Home: This is one of the clearest long-follow-up signals that a personalized neoantigen vaccine can add something real to adjuvant PD-1, but Phase III confirmation and manufacturing practicality will decide whether this becomes a broadly usable curative-intent strategy.
🔗 https://pubmed.ncbi.nlm.nih.gov/42223134
🏆 SPOTLIGHT
Spotlight for this triple ICI neoadjuvant treatment. When you throw the kitchen sink at non-metastatic melanoma, it works, but the price of 26% G3 AEs and one death seems high.
Ipilimumab (CTLA-4) + nivolumab–relatlimab (PD-1/LAG-3) drives 69% MPR (63% pCR) in resectable stage III melanoma—at the cost of 26% G3 and 1 G5 myocarditis
Neoadjuvant INR combines low-dose ipilimumab (CTLA-4 blockade) with fixed-dose nivolumab–relatlimab (PD-1/LAG-3 blockade) to deepen pathologic responses while keeping treatment short. In this Phase II trial in high-risk resectable AJCC8 stage IIIB–D melanoma (NCT06295159), patients received ipilimumab 1mg/kg once plus nivolumab–relatlimab 480/160mg q4w for 2 doses, then surgery; 20 patients enrolled (Aug 2024–Jan 2026), with 16 resected and evaluable for pathology (median time to surgery 73 days). Among resected patients, pCR was 10/16 (63%) and MPR (pCR+near-pCR) was 11/16 (69%); by contrast, the most-used neoadjuvant benchmark regimen (ipi 1mg/kg + nivo 3mg/kg x2 in OpACIN-neo) delivered a ~77% pathologic response rate with ~20% G3–4 irAEs, and Phase III NADINA established neoadjuvant ipi/nivo as superior to adjuvant PD-1 with 1-year EFS 83.7% vs 57.2% (though cross-trial comparisons are imperfect). Safety: among 19 treated, 26% had grade 3 AEs (colitis, encephalitis, headache, hyperglycemia, rash) and there was 1 grade 5 myocarditis; median follow-up was 9.8 months, and no post-surgery recurrences were reported to date (immature EFS/OS).
Take Home: If INR’s very high pCR/MPR rates hold up with longer follow-up, it could be a real “short-course, no-adjuvant-needed” neoadjuvant strategy—but the single grade 5 myocarditis in just 19 treated patients is a loud reminder that triplet checkpointing may narrow the therapeutic window versus today’s neoadjuvant ipi/nivo standard.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.9501
🏆 SPOTLIGHT
Five-year melanoma survival approaching 50% is becoming the new normal. RELATIVITY-047 confirms that PD-1/LAG-3 blockade delivers a durable overall survival advantage and cements nivolumab-relatlimab as a true frontline standard rather than merely a lower-toxicity compromise.
Nivolumab + Relatlimab (PD-1/LAG-3) Adds 10% Absolute 5-Year OS in 1L Melanoma
Nivolumab is a PD-1 inhibitor and relatlimab is a LAG-3 inhibitor evaluated as a fixed-dose combination versus nivolumab alone in the randomized, double-blind Phase 3 RELATIVITY-047 first-line trial in unresectable/metastatic melanoma (n=714; NCT03470922). At 57.3 months minimum follow-up, nivolumab+relatlimab improved median PFS to 10.2 vs 4.6 months (HR 0.77) and median OS to 54.8 vs 33.2 months (HR 0.78), with 60-month OS 48% vs 38% and ORR 44% vs 34%. Grade 3–4 TRAEs were higher than nivolumab alone (23% vs 12%; TRAE discontinuation 17% vs 10%) but below historical nivolumab+ipilimumab toxicity, where grade 3–4 TRAEs are roughly 55%.
Take Home: Five-year data make PD-1/LAG-3 feel less like a “gentler PD-1 plus” and more like a real middle option for patients who need more than PD-1 alone but may not tolerate CTLA-4.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.9532
TIL adoptive cell therapy (autologous) delivers 22% ORR in metastatic uveal melanoma—responses track with UMIS-selected, tumor-reactive products
Autologous tumor-infiltrating lymphocyte (TIL) adoptive cell therapy—generated from resected metastases, expanded ex vivo, then given after cyclophosphamide/fludarabine lymphodepletion with high-dose IL-2—was tested in a single-center Phase 2 study in heavily pretreated metastatic uveal melanoma (NCT03467516). In 32 evaluable patients (n=33 treated; median 2 prior lines; 64% prior checkpoint blockade; 33% prior tebentafusp; 76% elevated LDH; 24% CNS disease), ORR was 22% (7/32) with median duration of response 10.8 months (max ongoing 58 months); by comparison, published benchmarks in this disease are typically low for immunotherapy—tebentafusp has a modest ORR (often ~7–10% despite OS benefit in HLA-A*02:01+ patients) and ipilimumab+nivolumab is commonly ~10–20% ORR across series/trials. Product potency was the differentiator: tumor-reactive TIL products had 37% ORR (7/19) vs 0% (0/12) for nonreactive (P=0.026), with higher peripheral persistence of infused clones in responders (R/NR ratio 10.69, P=0.020); in a pooled dataset (n=54; 52 evaluable), reactive products drove 43% ORR (12/28) vs 0% (0/20) (P=5.06E-4). The in situ transcriptomic biomarker UMIS predicted tumor-reactive culture growth (rho=0.55, P=2.72E-5), final product reactivity (P=0.040), and clinical response (P=0.004), suggesting a practical way to “pick the right lesion” for harvest when multiple metastases exist.
Take Home: If UMIS can reliably steer surgeons to metastases that yield tumor-reactive TIL, this approach could turn TIL from a “sometimes works” option into a more consistently effective rescue strategy in metastatic uveal melanoma—especially for post-tebentafusp or non–HLA-A02:01 patients where SOC response rates remain frustratingly low.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2512
Adaptive encorafenib/binimetinib (BRAF/MEK) + nivolumab cuts Grade 3+ AEs while preserving long-run disease control in BRAF V600E melanoma
Encorafenib (BRAF inhibitor) + binimetinib (MEK inhibitor) were given either continuously or via patient-specific adaptive scheduling (ctDNA- and imaging-calibrated mathematical model), with nivolumab (anti–PD-1) dosed continuously in both arms, in a randomized pilot feasibility study in previously untreated advanced BRAF V600E melanoma with detectable baseline plasma ctDNA (n=10; NCT03543969). Drug exposure diverged sharply—median time on ENCO+BINI was 61 weeks (89% of on-study time) with continuous dosing vs 11 weeks (24%) with adaptive dosing—yet at 45.5 months’ median follow-up, PFS was 26.9 months in Arm A vs not reached in Arm B (log-rank p=0.5) and OS was 32 months vs not reached (2-year OS 60% vs 75%; p=0.18), with fewer treatment-related Grade 3 AEs in the adaptive arm (6 vs 16). Efficacy looks directionally within the range of frontline PD-1–based standards (e.g., pembrolizumab ORR ~33% and median PFS ~5–8 months in KEYNOTE-006; nivolumab+ipilimumab median PFS ~15.5 months in the BRAF-mutant subgroup of CheckMate 067), but this tiny n and imbalanced ORR (100% vs 60%, all PRs) make cross-trial comparisons—and the “non-inferiority” claim—statistically fragile.
Take Home: If a larger trial confirms that ctDNA-guided “minimum effective” BRAF/MEK exposure can maintain long-term control while materially reducing Grade 3 toxicity, this could be a practical way to keep the immunotherapy backbone doing the heavy lifting—while sparing patients the cumulative BRAF/MEK burden and cost.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.9538
Recent anti-CTLA-4 (ipilimumab) linked to IL-6–high immunosuppressive profile and ~6× shorter OS on PD-L1 + CD40/4-1BB adenoviral gene therapy in PD-1–refractory melanoma
LOKON003 combines an anti–PD-L1 antibody with a replication-competent adenoviral immunogene therapy designed to stimulate CD40 and 4-1BB signaling (NCT04123470) in anti–PD-1–refractory metastatic melanoma (Phase I/II; n=24 enrolled; baseline clustering reported for n=23). At baseline, an “hyperactive immunosuppressive” plasma protein signature (n=8) tracked with markedly worse survival: median OS 4.0 months vs 28.1 months without the signature (n=15); independently, recent anti-CTLA-4 exposure within 12 months (n=10) also separated outcomes (mOS 5.3 vs 30.7 months for later/none, n=14) and was associated with an IL-6–centered upregulated network. Safety and response rates were not reported in this abstract, and the OS deltas are hypothesis-generating in a very small, biomarker-defined subset—still notable because salvage outcomes after PD-1 failure are commonly poor, with retrospective ipilimumab-after–anti–PD-1 series reporting mOS ~5 months.
Take Home: If validated, this argues that “what you just gave” (recent CTLA-4 blockade) may materially shape the next immunotherapy’s immune set-point—making IL-6 blockade and/or a deliberate washout strategy worth testing prospectively in PD-1–refractory melanoma rather than reflexively sequencing more checkpoint-on-checkpoint.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2634
Brenetafusp (PRAME×CD3 ImmTAC) shows 17% ORR at 160µg in late-line melanoma, with mostly low-grade CRS and no irAE recurrence
Brenetafusp (IMC-F106C) is a PRAME-targeted ImmTAC bispecific (PRAME×CD3) designed to redirect polyclonal T cells into PRAME-expressing tumors; in this Phase 1 study (NCT04262466), HLA-A*02:01+ advanced melanoma patients who had exhausted standard options (median 2.5 prior lines; 100% prior anti–PD-1; 30% PD-1 refractory; 68% stage IVc/IVd) received weekly IV brenetafusp monotherapy across doses including 40µg (n=16) and 160µg (n=24), plus a small combination cohort of brenetafusp 160µg weekly + pembrolizumab 400mg Q6W (n=10). In the full monotherapy melanoma set (n=66), DCR was 52% with ORR 12%, 6-month OS 83%, and ctDNA molecular response 38% (ctDNA-evaluable n=50); by dose, 160µg trended higher than 40µg for DCR (67% vs 56%), ORR (17% vs 6%), and ctDNA response (33% vs 20%) despite baseline imbalances favoring the 40µg cohort, while PD-1–refractory patients still showed ORR 10% and ctDNA response 53% (ctDNA-evaluable n=15). Safety was characterized by Grade 1/2 CRS (56% at 40µg; 42% at 160µg) and rash (44% vs 42%), with Grade 3/4 AEs more frequent at 160µg (54% vs 25%) mainly due to transient lymphocyte decreases; notably, among 10 patients with prior ICI Grade 3/4 irAEs, none recurred, and the combo cohort showed DCR 70% and ORR 20% with no new signals—an AE profile that appears qualitatively different from CTLA-4–based salvage strategies where ≥Grade 3 toxicity is often a central limitation. For efficacy context post–anti–PD-1, published salvage outcomes with ipilimumab-based approaches are typically in the ~10–25% ORR range but with substantial ≥Grade 3 irAEs (often ~30%+; higher in many series), so a 17% ORR signal at 160µg lands “in-family” on response while potentially differentiating on irAE phenotype (needs confirmation in larger, controlled data).
Take Home: If the Phase 3 PRISM-MEL strategy (brenetafusp+nivolumab vs standard nivolumab regimens) reproduces these response/ctDNA signals without importing CTLA-4–like irAEs, this could become a genuinely patient-friendly way to re-introduce T-cell killing in PD-1–experienced melanoma—but the current dataset is small, non-randomized, and dose cohorts weren’t prognostically balanced, so the “160µg is better” story still needs clean confirmatory evidence.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.9527
Botensilimab (Fc‑enhanced anti‑CTLA‑4) + balstilimab (anti‑PD‑1) delivers 22% ORR in dual ICI‑refractory cutaneous melanoma
Botensilimab (BOT) is an Fc‑engineered anti‑CTLA‑4 antibody designed to enhance T‑cell priming and Treg depletion, and balstilimab (BAL) is an anti‑PD‑1; both were evaluated in the open-label global Phase 2 C‑800‑23 trial (NCT05529316) in unresectable stage III/IV cutaneous melanoma refractory/resistant to prior anti–PD-(L)1 ± CTLA‑4. As of December 13, 2025, BOT+BAL showed ORR 22% (95% CI 10–39) and CBR 33% with median DOR not reached, 12‑mo PFS 29%, and median OS 16.6 mo in a heavily primary‑resistant population (n=36; 51% M1c/M1d; 39% LDH>ULN); BOT monotherapy produced ORR 6% and 12‑mo PFS 7% (n=138; median OS 12.9 mo). Grade ≥3 TRAEs were 36% with BOT+BAL (diarrhea 42%) and 30% with BOT alone (diarrhea 33%), with one possibly treatment-related death on BOT monotherapy (immune-mediated enterocolitis); in context, the approved post–PD‑1 option lifileucel has reported ~31.4% ORR in advanced melanoma after anti–PD‑1/PD‑L1 (and BRAF±MEK if BRAF V600+), albeit with very different logistics/toxicity expectations versus antibody therapy.
Take Home: If durability holds up, a ~22% ORR with BOT+BAL in PD-(L)1±CTLA‑4–refractory, primary‑resistant melanoma is a real “salvage ICI” signal (even if not yet at lifileucel-like response rates), but the safety/colitis risk and the small n—especially in the ipi‑exposed subset driving the numerically higher 29% ORR—mean this still needs confirmatory, better-controlled readouts before anyone declares a new post‑checkpoint standard.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.9543
Imneskibart (AU-007) IL-2–CD25 “decoy” antibody + low-dose aldesleukin drives tumor shrinkage in CPI-refractory melanoma (early PFS signal)
Imneskibart (AU-007) is an AI-designed anti–IL-2 monoclonal antibody that blocks IL-2 binding to CD25 (trimeric IL-2R) while preserving signaling through CD122/CD132 (dimeric IL-2R), aiming to expand Teff/NK without expanding Tregs; it was tested with low-dose subcutaneous aldesleukin (± nivolumab) in an ongoing Phase 1/2 study in locally advanced/metastatic cutaneous melanoma with confirmed progression after prior anti–PD-1 + anti–CTLA-4 or anti–PD-1 + anti–LAG-3 doublet CPI (n=15 treated; n=14 efficacy-evaluable; NCT05267626). Across imneskibart + aldesleukin, 9/14 had tumor reductions (-1% to -100%), median PFS was 7.6 months (median follow-up 9.3 months), and 6 patients remained on treatment; in the imneskibart + aldesleukin + nivolumab cohort to date, 2/6 efficacy-evaluable achieved PR (33%) with -35% and -31% shrinkage and 67% had disease control at 4 months—numerically above historical single-agent high-dose IV IL-2 ORR ~16% (with substantial toxicity) and notably encouraging for a post–doublet-CPI-refractory population. Pharmacodynamically, treatment decreased peripheral Tregs and increased CD8/Treg ratios (up to 2–3× over baseline), with >2× increases associated with longer time on treatment/PFS/OS; safety looked favorable with Grade 3/4 drug-related AEs in 14% overall and 0% in the +nivolumab subgroup, and no discontinuations for drug-related AEs (subcutaneous low-dose IL-2 schema, not classic inpatient high-dose IL-2).
Take Home: If the 7.6-month mPFS and early 33% PR rate with nivolumab hold up in larger CPI-refractory cohorts, “IL-2 reprogramming” via CD25 blockade could become a genuinely practical way to re-sensitize cold, post–doublet-IO melanoma—especially since the tolerability here looks far more clinic-friendly than legacy high-dose IV IL-2.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.9526
IO102-IO103 (PD-L1/IDO peptide vaccine) + nivolumab–relatlimab (PD-1/LAG-3) posts 60% ORR in 1L unresectable melanoma
IO102-IO103 is a subcutaneous peptide vaccine targeting PD-L1 and IDO-expressing tumor and immunosuppressive cells, combined here with nivolumab–relatlimab (dual PD-1/LAG-3 checkpoint blockade) in a multicenter Phase 2 trial in previously untreated, unresectable non-uveal melanoma (NCT05912244). In 43 evaluable patients, best overall response rate (RECIST v1.1) was 60% (95% CI 44-75%; 16% CR, 44% PR); at Dec 1, 2025 data lock with 10.1 months median follow-up, median PFS was 8.2 months (95% CI 6.8-NR) and median DOR was NR (95% CI 14-NR). Grade 3-4 TRAEs occurred in 21% (no treatment-related deaths); for context, nivo–rela alone in RELATIVITY-047 reported ORR ~43% and median PFS ~10.1-10.2 months with grade 3-4 TRAEs in the high-teens range—so the response rate looks higher here, but the PFS signal is not clearly better (and follow-up/sample size are limited), including in PD-L1–negative tumors where ORR was 52% (n=25) and median PFS remained 8.2 months.
Take Home: If IO102-IO103 can reproducibly lift response depth (more CRs) on top of nivo–rela without materially worsening immune toxicity, it could become a pragmatic “upgrade” for patients who want dual-checkpoint efficacy but need better odds of tumor shrinkage—however, with only 43 evaluable patients and short follow-up, the key question is whether the higher ORR translates into longer PFS/OS versus nivo–rela’s established ~10-month median PFS in Phase 3.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.9519
OBX-115 (ACZ-inducible mbIL15–armored TIL) drives 67% ORR in post-ICI advanced melanoma without high-dose IL2
OBX-115 is an autologous tumor-infiltrating lymphocyte (TIL) cell therapy engineered to express membrane-bound IL15 (mbIL15) under small-molecule control with acetazolamide (ACZ), designed to avoid the traditional high-dose IL2 support used with “classic” TIL. In the single-arm Agni-01 Phase 2 cohort (NCT06060613) at the RP2D, 15 patients with advanced non-uveal melanoma progressing on/after immune checkpoint inhibitors (93% prior ICI doublet; ≤2 prior lines; 75% of BRAF-mutant patients previously received BRAF/MEK) achieved a confirmed ORR of 67% (1 CR, 9 PR) and DCR of 93%, with median DOR not reached at a median 18.6-week follow-up (range 10.1–73.4). Safety appeared differentiated for a TIL approach (low-dose Cy/Flu lymphodepletion; no DLTs/ICU/TRM; no ICANS; grade 3 CRS in 1 patient and grade 3 hypoxia in 1, both steroid-responsive); by contrast, the FDA-approved post–anti–PD-1 TIL lifileucel (Amtagvi) reported an ORR of 31.5% (n=73) in its registrational dataset, but uses non-regulatable IL2 support and more intensive inpatient-style administration in many centers.
Take Home: If this high response rate holds up with mature duration-of-response, OBX-115 could become a major next-generation TIL contender in post-checkpoint melanoma—especially because it avoids high-dose IL-2 and appears more operationally manageable than classic TIL approaches. The key question is whether the early 67% ORR persists with longer follow-up and broader treatment-center experience.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.9507
Atezolizumab + Tiragolumab (PD-L1/TIGIT) Shows Index Node pCR Can Miss Melanoma
Atezolizumab is an anti–PD-L1 antibody and tiragolumab is an anti-TIGIT antibody evaluated in a preplanned NeoACTIVATE Arm C substudy (NCT03554083) asking whether clipped index lymph node (ILN) pathology can substitute for full nodal-basin pathology after neoadjuvant doublet immunotherapy in resectable clinical stage III macroscopic nodal melanoma. Among 30 patients who underwent therapeutic lymph node dissection (TLND), ILN and basin pathology were discordant in 2 patients (6.7%); both had ILN pCR but residual non-ILN disease, yielding a 13.3% false-negative rate for ILN pCR as a basin-clearance surrogate. No new safety signal was reported, but the relevant benchmark is surgical adequacy: PRADO-style response-directed surgery after neoadjuvant ipilimumab/nivolumab explored TLND omission only after index-node MPR, while TLND remains the conventional fallback for clinically evident nodal disease.
Take Home: This is a surgical de-escalation caution flag—patients can have a “clean” clipped node and still harbor residual basin disease, so ILN-only surgery needs outcomes validation before replacing TLND.
🔗 https://pubmed.ncbi.nlm.nih.gov/42218345
Colorectal Cancer
🏆 SPOTLIGHT
Can aPD-1/VEGF finally crack MSS CRC? Ivonescimab plus FOLFOX generated response rates approaching 71% and disease control in every treated patient, but because of the heavy contribution from China, even ASCO presenters seemed dubious of the results until repeated in the US.
Ivonescimab (PD-1×VEGF bispecific) + mFOLFOX6 posts 70.8% ORR and 100% DCR in 1L unresectable mCRC (Phase 2)
Ivonescimab is an investigational tetrameric bispecific antibody co-targeting PD-1 and VEGF, here combined with oxaliplatin/5-FU/leucovorin (mFOLFOX6) as first-line therapy in a multiregional Phase 2 trial (NCT05382442) enrolling untreated unresectable mCRC in China and the US (n=49; 20 mg/kg n=25, 10 mg/kg n=24; data cutoff December 31, 2025). With 7-month median follow-up, ORR was 70.8% in both arms (all partial responses) and DCR was 100%; median PFS was not reached, with 6-month PFS rates of 95.2% (20 mg/kg) and 84.8% (10 mg/kg)—numerically above typical first-line mFOLFOX6 ± bevacizumab ORR (50–60% range in older datasets) but still early and single-study, with no randomized SOC comparator here. Grade ≥3 ivo-related TEAEs occurred in 40.0% vs 33.3%, serious ivo-related TEAEs in 24.0% vs 16.7%, and ivo discontinuation due to TEAEs was 4.0% vs 0% (no TEAE-related deaths), suggesting meaningful added toxicity that will need to be weighed against benefit versus established chemo±VEGF strategies.
Take Home: If these response and early PFS signals hold up in a randomized setting, PD-1×VEGF “all-in-one” biology could be a practical way to push immunotherapy benefit into largely immunotherapy-resistant MSS mCRC—but with only 49 patients and ~7 months follow-up, the key question is whether this translates into durable PFS/OS beyond what we already get with FOLFOX-bevacizumab.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3576
🏆 SPOTLIGHT
Long-term CheckMate 8HW data suggest dual Ipi/Nivo checkpoint blockade is creating a growing population of MSI-H CRC patients who remain alive and therapy-free years after initiation.
Nivolumab + ipilimumab (PD-1 + CTLA-4) improves long-term treatment-free survival vs nivolumab in MSI-H/dMMR mCRC
Nivolumab + ipilimumab (NIVO + IPI), a dual immune checkpoint regimen targeting PD-1 and CTLA-4, was compared with nivolumab (PD-1) monotherapy in the Phase 3 CheckMate 8HW trial in metastatic colorectal cancer with locally tested MSI-H/dMMR (centrally confirmed in 84% vs 81% of randomized patients; median follow-up 55.1 months; NCT04008030). Disease control was numerically higher with NIVO + IPI vs NIVO in 1L (85% vs 81%) and across all lines (84% vs 76%); subgroup medians favored NIVO + IPI for PFS in patients with SD in 1L (33.1 vs 9.2 months; HR 0.52) and in all lines regardless of metastatic burden (1 site: mPFS NR vs 60.8 months; HR 0.51; ≥2 sites: 59.2 vs 23.0 months; HR 0.70), with OS benefits emerging in key subgroups (e.g., SD: mOS NR vs 26.8 months; HR 0.43) and disease-specific OS HR 0.47. Post-treatment “functional cure” signals were more frequent with NIVO + IPI: 82% (185/226) were alive and treatment-free vs 67% (125/186) on NIVO, and 90% vs 79% were alive and systemic-treatment–free—positioning the combo as a higher-intensity alternative to PD-1 monotherapy benchmarks like 1L pembrolizumab (KEYNOTE-177 mPFS 16.5 vs 8.2 months vs chemo; OS not statistically significant at final analysis).
Take Home: If these treatment-free and systemic-treatment–free deltas hold up across full datasets, NIVO + IPI looks like it may buy more patients durable off-therapy time than PD-1 alone in MSI-H/dMMR mCRC—but the subgroup framing (CR/PR/SD-conditioned outcomes and many “NR” medians) means clinicians will still want the complete, unconditioned OS/PFS curves and late immune-toxicity/discontinuation details before universally favoring the combo upfront.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3532
🏆 SPOTLIGHT
It’s great to see that neoadjuvant PD-1 melts MSI-H colon cancer with ~80% pCR and also prevents surgeries (just like rectal cancer).
Tislelizumab (PD-1) neoadjuvant drives 78.6% pCR + ctDNA MRD clears fast in dMMR/MSI-H colon cancer
Tislelizumab is an anti–PD-1 monoclonal antibody engineered to minimize Fcγ receptor binding; here it was given as neoadjuvant monotherapy (200mg IV Q3W ×4, then up to 8+ cycles per MDT) with tumor-informed ctDNA MRD tracking in a prospective pilot in locally advanced dMMR/MSI-H colon adenocarcinoma (n=40 enrolled; 39 treated; NCT06262581; median follow-up 13.6 months). Among resected patients (n=28), pCR was 78.6% (22/28) and MPR was 92.9% (26/28); 9 patients chose non-operative management after cCR, yielding an overall CR rate of 79.5%—numerically above historical surgery-first pCR rates (typically low single digits) and even higher than published neoadjuvant dual-checkpoint benchmarks in dMMR colon cancer like NICHE-2 (pCR 68%, MPR 95%). Grade 3 irAEs occurred in 10.3% (4/39; pneumonitis n=2, renal insufficiency n=1, stroke n=1) with no treatment-related deaths; MRD was positive at baseline in 96.6% (28/29 tested) and cleared in 48.3% after 1 cycle and 75.9% after 2 cycles, with 1 patient MRD-positive post-treatment after pneumonitis-driven discontinuation.
Take Home: If ctDNA MRD clearance continues to correlate with durable control, this “PD-1 alone + MRD-guided NOM” approach could push dMMR/MSI-H colon cancer toward organ-preservation pathways—though with only ~14 months’ follow-up, NOM durability and the real-world risk of immune pneumonitis still need longer, harder endpoints.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3651
🏆 SPOTLIGHT
Spotlight: PLK1 inhibition with standard chemo/bev result in a 72% ORR for RAS mutant MSS mCRC, in a clear response breakthrough with manageable toxicity. Over time we’ll get to see how long the durability lasts.
Onvansertib (PLK1 inhibitor) + FOLFIRI/Bev lifts ORR to 72% and improves PFS signal in 1L RAS-mut mCRC
Onvansertib (Onv) is an oral, selective polo-like kinase 1 (PLK1) inhibitor combined here with standard doublet chemotherapy plus bevacizumab (FOLFIRI+Bev or FOLFOX+Bev) in the randomized Phase 2 CRDF-004 trial (NCT06106308) in previously untreated KRAS/NRAS-mutant metastatic colorectal cancer (ITT n=110; data cutoff Jan 22, 2026; median follow-up ~10 months). The strongest activity was in the Onv 30mg + FOLFIRI+Bev arm: BICR ORR 72.2% (13/18) vs 42.1% with FOLFIRI+Bev (8/19) and 43.2% across combined SoC controls (16/37), with confirmed ORR narrowly missing conventional significance vs SoC (p=0.051) but a 6‑month ORR advantage vs FOLFIRI+Bev (55.6% vs 21.1%, p=0.045); median PFS was not reached vs ~11 months for controls, with HR 0.38 vs FOLFIRI+Bev and HR 0.37 vs combined SoC (p=0.048)—directionally better than the typical ~9–10 month PFS historically reported for bevacizumab + oxaliplatin/fluoropyrimidine regimens in 1L mCRC. Grade ≥3 AEs were described as manageable with no unexpected toxicities; neutropenia was the most common grade ≥3 event.
Take Home: If the PFS HR ~0.37 holds up in a larger, mature dataset, Onvansertib could become a real differentiator for the ~50%+ of mCRC patients who are RAS-mutant (and therefore not candidates for anti‑EGFR benefit), but the small n in the key Onv30+FOLFIRI+Bev cohort and borderline p-values mean Phase 3 confirmation is doing the heavy lifting here.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3510
ABFOLFOX (PD-L1 + VEGF + FOLFOX) drives immune-activated CRLM states, with 35% durable control
Atezolizumab (anti–PD-L1) + bevacizumab (anti-VEGF) + leucovorin/5-FU/oxaliplatin (FOLFOX) was evaluated as neoadjuvant therapy for unresectable colorectal liver metastases (CRLM) in a prospectively registered cohort with serial sampling (cohort B, n=20; NCT03698461) alongside a cross-sectional tissue cohort (cohort A, n=60). Durable disease control (PFS≥24 months) occurred in 35% (7/20) on ABFOLFOX; multi-omic profiling suggested stepwise tumor-ecosystem immune remodeling with added agents (notably bevacizumab), with the largest shifts in the ABFOLFOX-treated serial-sampling cohort, and benefit associated with monocyte-lineage programs plus SP140 regulon activation in responders. Stool metagenomics linked higher Prevotella abundance to “good response” and a more immune-active tumor microenvironment, and mediation analysis suggested the gut microbiome partially bridges treatment response and intratumoral immune changes—while clinically, the 24-month disease-control landmark looks meaningfully above typical first-line FOLFOX–bevacizumab median PFS (~9–12 months) reported in mCRC, though the endpoints aren’t directly comparable across designs.
Take Home: If this gut–liver axis signal holds up, ABFOLFOX could become less of a “treat everyone” triplet-plus-IO idea in MSS-leaning CRC and more of a biomarker-enriched strategy (tumor monocyte states/SP140, plus Prevotella-high microbiomes) to identify the minority who can truly achieve long disease control.
🔗 https://pubmed.ncbi.nlm.nih.gov/42218533
Ozekibart (INBRX-109) DR5 agonist + FOLFIRI delivers 23% ORR in 4L+ metastatic CRC
Ozekibart (INBRX-109) is a next-generation tetravalent death receptor 5 (DR5) agonist designed to induce tumor cell apoptosis; here it was combined with FOLFIRI in an open-label Phase 1 expansion cohort (C4d; NCT03715933) for locally advanced/metastatic, unresectable colorectal adenocarcinoma after 2–3 prior systemic lines (44 treated; 26 response-evaluable; >70% treated in 4th line; 81.8% prior irinotecan). At the October 15, 2025 cutoff, ORR was 23% (all PRs) with 92% disease control; 63.6% remained on treatment, with 1 AE-related discontinuation and 11 progressions. Safety looked broadly “FOLFIRI-like” (anemia 36.4% any-grade/9.1% G3+, diarrhea 34.1%/9.1%, nausea 31.8%/0%, fatigue 27.3%/2.3%), with grade ≥3 hepatotoxicity limited to 1 case of ALT increase—worth noting given the DR5 class history; efficacy appears well above typical late-line SOC ORRs of ~1% with regorafenib and ~1.6% with trifluridine/tipiracil (TAS-102) in refractory mCRC.
Take Home: If this ~23% ORR holds up in a larger, controlled dataset, a DR5 agonist “re-sensitizing” chemo in 4L+ CRC would be a real shift from today’s low-response salvage standards—but the key question is durability (PFS/DoR) and whether the apparent benefit is driven by patient selection in a small, response-evaluable subset.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3533
Ad5.F35-GUCY2C-PADRE (GUCY2C-directed adenoviral vaccine) drives dose-dependent T-cell immunity and signals improved 2-year RFS in resected CRC
Ad5.F35-GUCY2C-PADRE is an off-the-shelf chimeric Ad5/35 vector vaccine targeting the intestinal tumor antigen GUCY2C and incorporating PADRE to enhance CD4+ help, designed to bypass pre-existing anti-Ad5 neutralizing antibodies. In this open-label Phase 2A dose-finding study (NCT04111172) in resected high-risk GI adenocarcinomas with no evidence of disease post-adjuvant therapy (46 evaluable; CRC n=29; includes oligometastatic stage IV CRC n=8), three IM vaccinations q4w were given at 1×10^11, 1×10^12, or 5×10^12 viral particles; GUCY2C-specific ELISpot responses at 4 weeks post–dose 1 rose with dose (33%→94%→100% responders; p<0.0001) with higher-dose responses largely independent of baseline NAb status. Safety was benign (no DLTs, no TRSAEs; mostly grade 1–2 flu-like AEs), and among stage II–III CRC (excluding stage IV), recurrence rates at ~2 years numerically improved with dose (50% [4/8] at 1×10^11 vs 17% [1/6] at 1×10^12 vs 0% [0/7] at 5×10^12; p=0.051) with “median RFS not reached” at the two higher doses; immune responders also had significantly better RFS than non-responders (p<0.05). Versus current post-resection standard (surveillance after completion of adjuvant fluoropyrimidine/oxaliplatin where appropriate), this looks directionally better than the typical ~75% 3-year DFS seen in large Stage III adjuvant datasets—though this trial is small, noncomparative, and enriched for immunologic endpoints.
Take Home: If this immune-response/RFS association holds up in a larger randomized setting, a well-tolerated, NAb-agnostic GUCY2C vaccine could become a true “maintenance-like” recurrence-prevention strategy after curative-intent CRC therapy—where today’s SOC is essentially “finish chemo, then watch,” and the bar is meaningful DFS gains without chronic toxicity.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3617
Zongertinib (selective irreversible HER2 TKI) drives 45% ORR with mostly grade 1 diarrhea in HER2+ CRC
Zongertinib is an irreversible, HER2-selective tyrosine kinase inhibitor designed to spare wild-type EGFR (aiming to reduce classic EGFR-TKI rash/diarrhea) and was evaluated as monotherapy in a pooled subgroup of HER2-positive colorectal cancer across a Phase Ia dose-escalation/expansion study (NCT04886804) and the ongoing Phase II Beamion PANTUMOR-1 trial (NCT06581432). In 20 heavily pretreated HER2-positive CRC patients (n=20; median 2.5 prior systemic lines; 40% had prior HER2-directed therapy including trastuzumab/pertuzumab/T-DXd), confirmed investigator-assessed ORR was 45% (9/20; all PRs) with stable disease in 45% (DCR 90%) and median treatment duration 5.1 months (range 0–18); 35% remained on treatment at cutoff. TRAEs occurred in 70% with only 5% grade ≥3 (one grade 3 lipase increase); most common TRAEs were diarrhea 45% (all grade 1) and rash 15% (all grade 1)—notably milder than HER2-targeted standards in this setting like tucatinib+trastuzumab (FDA-approved for RAS WT HER2+ mCRC) with ORR ~39% and median PFS ~8 months, or T-DXd in HER2+ mCRC with ORR ~38% and grade ≥3 drug-related AEs ~41% (cross-trial comparisons). HER2 amplification/overexpression is a relatively small slice of mCRC (≈3% overall; ~5% in RAS WT), so this is a niche but clinically important biomarker-defined population.
Take Home: If durability holds up (PFS/DoR weren’t reported here), a 45% ORR with very low grade ≥3 toxicity suggests zongertinib could become a “cleaner,” chemo-free HER2 option in HER2+ mCRC—especially attractive for patients who have already seen antibody-based HER2 therapy and want an oral regimen, but confirmation in larger cohorts (and clarity on RAS status and survival endpoints) will be the real differentiator.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3538
Sotorasib (KRAS G12C inhibitor) + panitumumab (EGFR mAb) links early ctDNA clearance to longer survival in chemorefractory mCRC
Sotorasib is an oral covalent KRAS G12C inhibitor, and panitumumab is an anti-EGFR monoclonal antibody; in the Phase 3 CodeBreaK 300 trial (NCT05198934) in chemorefractory KRAS G12C–mutant mCRC, this analysis tested whether early ctDNA “clearance” predicts response and prognosis. Among patients with baseline and Cycle 2 Day 1 ctDNA (n=142), ≥80% KRAS G12C VAF clearance occurred in 77.3% (960 mg; n=44) and 65.9% (240 mg; n=44) vs 15% with investigator’s choice trifluridine/tipiracil or regorafenib (n=40)—therapies that typically deliver very low ORR and short mPFS (~2 months) in this chemorefractory setting. In the 960 mg arm, patients achieving ≥80% clearance had ORR 35.3% (12/34) by VAF (and 32.3% by cTF) and substantially longer outcomes (mPFS 5.8 months vs 1.9 months without clearance; mOS not evaluable vs 4.9 months), while increasing the stringency beyond ≥80% did not add prognostic value.
Take Home: If a simple Cycle-2 ctDNA readout can reliably flag “on-track” patients (and, just as importantly, early non-benefiters), KRAS G12C mCRC—a ~2%–4% subset of CRC—may finally get a practical response-adaptation tool to complement imaging, but we still need prospective, ctDNA-guided decision studies to prove acting on non-clearance improves patient outcomes.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3511
ST316 β-Catenin/BCL9 Antagonist Reaches 46.7% ORR With 2L FOLFIRI/Bev in MSS mCRC
ST316 is a first-in-class IV peptide antagonist of the β-catenin–BCL9/9L interaction, designed to suppress oncogenic Wnt/β-catenin transcription while sparing physiologic Wnt signaling. In the ongoing Phase 1/2 ST316-101 study (NCT05848739), Phase 2 expansion in evaluable second-line metastatic CRC patients (n=15; 100% MSS; 73% RAS-mutant) treated with ST316 plus FOLFIRI/bevacizumab produced confirmed ORR 46.7% and DCR 93.4%. Common AEs included nausea, fatigue, ALT/AST elevations, and neutropenia, with grade 3 transaminitis and neutropenia; the ORR is well above historical second-line FOLFIRI/bevacizumab expectations around ~11% in EAGLE.
Take Home: A Wnt-pathway drug finally showing tumor shrinkage in MSS mCRC would be a big deal, but the current dataset is only 15 evaluable patients and needs randomized confirmation fast.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e15622
Telisotuzumab adizutecan (Temab-A; ABBV-400) c-Met–targeted topo-1 ADC posts 37% ORR in platinum-resistant ovarian cancer
Telisotuzumab adizutecan (Temab-A; ABBV-400) is a c-Met protein–targeting antibody–drug conjugate delivering a topoisomerase 1 inhibitor payload; in this Phase 1 open-label basket cohort (NCT06084481), patients with platinum-resistant ovarian/primary peritoneal/fallopian tube cancer received 2.4mg/kg Q3W after ≥1 to ≤5 prior systemic lines (n=41; median 3 prior lines). At the September 11, 2025 cutoff (median follow-up 5.1 months), confirmed ORR was 37% (15/41; 95% CI 22–53) with a clinical benefit rate of 85% (95% CI 71–94); DOR/PFS/OS were immature, but the ORR is above the typical ~10–20% range seen with single-agent non-platinum chemotherapy options like PLD or topotecan in PROC. Safety was primarily hematologic/GI: any-grade anemia 59%, nausea 59%, fatigue 39%; Grade ≥3 anemia 34%, neutrophils decreased 17%, WBC decreased 15%; adjudicated ILD/pneumonitis was 5% (Grade 1 only), with discontinuation 5%, interruption 39%, and dose reduction 37%—a meaningful dose-modification burden that could matter in a frail, heavily pretreated population.
Take Home: If durability holds up, a 37% ORR without biomarker selection is competitive for PROC (and starts to encroach on biomarker-selected ADC territory like FRα-high mirvetuximab), but the short follow-up plus frequent interruptions/reductions make DOR and tolerability the real make-or-break for whether ABBV-400 can become a practical “workhorse” option beyond trials.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5514
Leronlimab (CCR5 mAb) + TAS-102/Bev shows early ctDNA drops and 100% stable disease in refractory mCRC (CLOVER)
Leronlimab (LRM) is a subcutaneous anti-CCR5 humanized monoclonal antibody being added to the current late-line standard regimen trifluridine/tipiracil (TAS-102) plus bevacizumab in the open-label Phase 2 CLOVER study (NCT06699836) for CCR5+ refractory mCRC. As of abstract drafting, 23/37 (62.2%) screened patients were enrolled; no LRM-related dose-limiting toxicities were seen at 350mg, and 700mg dosing had started, with early efficacy in 6 centrally imaged patients showing stable disease in 6/6 (median tumor-size change -10.2%, range +12.5% to -28.4%) and ctDNA declines by Week 2 (n=11; median -85%, range -9% to -99%). With TAS-102/Bev alone (Phase 3 SUNLIGHT), the benchmark is modest tumor shrinkage/very low ORR but improved survival (mOS 10.8mo vs 7.5mo for TAS-102; mPFS 5.6mo vs 2.4mo), so CLOVER’s early “all-SD” signal is directionally consistent with expected disease control—however, it’s too early (n=6 imaged) to claim additive clinical benefit beyond the TAS-102/Bev backbone, and the PD-L1 increases on CTCs/CAMLs (12/17, 70.6% by Week 5) remain hypothesis-generating.
Take Home: If the ctDNA declines translate into longer PFS/OS (not just early disease control that TAS-102/Bev already delivers), CCR5 blockade could become a pragmatic “bolt-on” to an FDA-approved late-line backbone—otherwise this may end up being a biomarker-rich but clinically incremental study until randomized data show separation.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e15525
ctDNA (MRD)-guided ACT (capecitabine±oxaliplatin) cuts 3-year recurrence in stage II MSS colon cancer
CIRCULATE (AIO KRK-0217/ABCSG; NCT04089631) used a tumor-informed NGS ctDNA assay post-op in stage II pMMR/MSS colon cancer: ctDNA+ patients were randomized to adjuvant chemotherapy (6 mo capecitabine or 3–6 mo CAPOX) vs observation, while ctDNA− patients were largely observed (blinded to ctDNA status). Among ctDNA+ patients, per-protocol chemo (n=26 randomized; some untreated excluded in PP) reduced 3-year recurrence from 62% to 19% (HR 0.21; p=0.009) and improved 3-year DFS from 38% to 77% (HR 0.28; p=0.022), but ITT did not reach significance (3-year DFS 61% vs 38%; HR 0.52; p=0.12); there was 1 treatment-related death. Versus “unselected” stage II where fluoropyrimidine ACT yields only small absolute benefit overall (and many low-risk patients do well with surgery alone), this trial suggests ctDNA identifies a small, very-high-risk slice that may actually derive meaningful DFS/TTR benefit from ACT—if the signal holds despite the tiny ctDNA+ sample and PP dependence.
Take Home: If validated in larger ctDNA+ cohorts (and with modern, high-sensitivity assays), this is the kind of MRD-enrichment that could finally make adjuvant chemo in “low-risk” stage II colon cancer feel like treating residual disease rather than treating in the dark.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.17_suppl.LBA3500
EC (encorafenib+BRAF inhibitor) + cetuximab (EGFR mAb) + FOLFIRI cuts PFS risk ~56% in 1L BRAF V600E mCRC
Encorafenib (BRAF inhibitor) plus cetuximab (EGFR antibody) was combined with FOLFIRI chemotherapy as first-line treatment in BREAKWATER Cohort 3 (Phase 3, open-label; NCT04607421) for untreated BRAF V600E–mutant metastatic colorectal cancer with ECOG 0–1 (n=147; EC+FOLFIRI n=73 vs control FOLFIRI±bevacizumab n=74; data cutoff Jan 6, 2026). EC+FOLFIRI improved median PFS to 15.2 months vs 8.3 months (HR 0.44, one-sided p=0.0002) and showed an OS signal (median OS not estimable vs 20.3 months; HR 0.56; 18-mo OS 72.0% vs 54.5%). Chemo discontinuation due to AEs was similar (14% vs 10%), with serious TEAEs 49% vs 44% and max grade 3/4 TEAEs 70% vs 81%—notably, efficacy appears well beyond typical FOLFIRI±bevacizumab benchmarks in this biomarker-defined high-risk subtype (BRAF V600E in ~7–10% of mCRC), while tolerability looks broadly comparable to control.
Take Home: If EC+FOLFIRI uptake is practical, this gives clinicians a second “targeted-triplet + chemo” 1L backbone alongside EC+mFOLFOX6—doubling median PFS versus FOLFIRI±bev is hard to ignore, but the OS is still immature (mOS NE) and we’ll want longer follow-up plus clarity on how to choose FOLFIRI vs FOLFOX up front for individual patients.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.17_suppl.LBA3503
Lycopene (tomato-carotenoid) intake linked to ~40% lower recurrence/death risk in stage III colon cancer
In a prospective cohort nested within the Phase III CALGB (Alliance)/SWOG 80702 adjuvant trial (NCT01150045), post-diagnosis lycopene (a non–provitamin A carotenoid largely derived from tomato-based foods; diet ± supplements) was assessed by FFQs at ~6 weeks and again at 14–16 months after randomization, modeled as time-varying cumulative average intake in resected stage III colon cancer (n=1666). Over median follow-up 6.0 years, highest vs lowest quintile total lycopene intake was associated with improved DFS (HR 0.60, 95% CI 0.44–0.83; Ptrend=0.006), RFS (HR 0.58, 95% CI 0.41–0.82; Ptrend=0.007), and OS (HR 0.60, 95% CI 0.40–0.89; Ptrend=0.02); dietary lycopene showed similar directionality. There were no treatment-emergent toxicity signals here (dietary exposure study), but confounding is a major concern—especially because modern adjuvant stage III colon cancer outcomes on oxaliplatin-based regimens are already strong (e.g., ~72–76% 3-year DFS with mFOLFOX6 in IDEA France; ~75–76% 3-year DFS reported for FOLFOX-era stage III cohorts), making it hard to know whether lycopene is causal vs a marker of healthier behaviors.
Take Home: If this association holds up with tighter control for “healthy user” bias (or, better, an intervention trial), increasing tomato-based foods could become a low-cost, patient-friendly survivorship lever alongside standard adjuvant therapy—but right now it’s hypothesis-generating rather than practice-changing.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3630
Bevacizumab + FOLFOXIRI (VEGF blockade + triplet chemo) misses PFS bar vs bevacizumab doublets in right-sided mCRC
Bevacizumab (anti-VEGF antibody) plus FOLFOXIRI (5-FU/leucovorin, oxaliplatin, irinotecan) was tested against bevacizumab plus investigator’s choice FOLFOX/FOLFIRI as first-line therapy in the randomized, open-label Phase II KCSG CO19-18 trial in unresectable metastatic right-sided colon cancer (cecum to splenic flexure; n=61; NCT03641976). The primary endpoint was negative: 6-month PFS rate 100% vs 87.1% (HR 1.05, 95% CI 0.53–1.71; p=0.87), with similar median PFS (14.7 vs 16.6 months) despite a numerically higher ORR with triplet (82.8% vs 64.6%; p=0.245). Grade ≥3 neutropenia was more frequent with FOLFOXIRI, while neuropathy/GI toxicity/hypertension were comparable or slightly higher with the doublet arm; efficacy here looks directionally in-line with historical bevacizumab+doublet first-line benchmarks (mPFS often ~10–11 months), and does not clearly recapitulate the broader all-comers TRIBE signal where FOLFOXIRI+bevacizumab improved OS vs FOLFIRI+bevacizumab (29.8 vs 25.8 months).
Take Home: If you’re choosing between bevacizumab-based triplet vs doublet specifically for right-sided primaries, this small Phase II doesn’t show a PFS advantage—and with higher neutropenia, the case for routine upfront intensification probably hinges on clearly defined “needs-maximum-shrinkage” subsets (and confirming whether the exploratory “no palliative resection” signal holds up).
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e15562
AtezoTRIBE: ctDNA (ctF) non-responders show larger PFS gain with PD-L1 atezolizumab add-on in 1L mCRC
In the Phase 2 AtezoTRIBE trial (NCT03721653), adding the PD-L1 inhibitor atezolizumab to first-line FOLFOXIRI/bevacizumab in untreated metastatic colorectal cancer was evaluated here with a translational liquid-biopsy analysis using the Tempus xF 105-gene cfDNA assay to measure plasma TMB (pTMB; high ≥26 mut/Mb) and circulating tumor fraction (ctF) dynamics from baseline (T0) to end of induction (T1). Baseline pTMB was evaluable in 123/155 (79%), with 16% pTMB-high (20/123); pTMB-high was prognostic for longer PFS (HR 0.69; 90%CI 0.43–1.10; p=0.095) but not predictive for atezolizumab benefit (treatment interaction p=0.871)—consistent with the broader reality that pMMR/MSS mCRC rarely benefits meaningfully from ICIs outside niche biomarker-defined pockets. Among patients with paired ctF (n=89), “molecular responders” (top 30% relative ctF reduction/clearance) were common (71%; 63/89), yet the larger relative benefit from adding atezolizumab appeared in ctF non-responders (HR 0.33; 90%CI 0.16–0.69) vs molecular responders (HR 0.76; 90%CI 0.03–3.95), with an interaction signal using the authors’ pre-specified lenient threshold (p_int=0.132; sensitivity analysis p_int=0.108). For context, first-line FOLFOXIRI/bevacizumab already delivers median PFS ~12.3 months in TRIBE vs ~9.7 months with FOLFIRI/bevacizumab (HR 0.77), so any incremental ICI effect needs to be clearly distinguishable from chemotherapy-driven ctDNA clearance.
Take Home: If validated, ctF “lack of early molecular response” could become a practical enrichment tool—paradoxically flagging the patients least helped by induction chemo as the ones most likely to gain from adding atezolizumab—though the interaction is modest by conventional stats and the MR subgroup’s CI is so wide that this could still be a sample-size/selection artifact rather than true immune-sensitization.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2575
Nivolumab+ipilimumab (PD-1+CTLA-4) PFS benefit holds with single MSI PCR or MMR IHC CDx in MSI-H/dMMR mCRC
Nivolumab (PD-1 inhibitor) plus ipilimumab (CTLA-4 inhibitor) was evaluated in a post hoc CheckMate-8HW (Phase 3; NCT04008030) analysis asking whether central companion diagnostics (Idylla MSI PCR and/or Dako MMR IHC) change the observed first-line PFS benefit versus chemotherapy in locally called MSI-H/dMMR unresectable/metastatic CRC. In the CTA+ population (local MSI-H/dMMR; n=301 of 303 randomized to nivo+ipi [n=202] vs chemo [n=101]), PFS was NR vs 6.2 months (HR 0.32, 95% CI 0.22–0.45); among central single-assay positives, the HRs were similar or slightly lower (MSI+ HR 0.20; dMMR HR 0.22) and matched central dual-positive pts (HR 0.20; n=208 dual-positive), with PFS NR in the nivo+ipi arms and ~6 months with chemo across these subsets. The “discordant” centrally MSS or pMMR groups were small (e.g., MSS n=45 total; pMMR n=39 total) and showed no benefit (HRs ~1.4–1.5), but tissue-missingness and low discordant counts limit inference; importantly, this does not alter the current clinical bottom line that immunotherapy (not chemo) is the SOC in true MSI-H/dMMR mCRC, consistent with the centrally confirmed CheckMate-8HW primary analysis (nivo+ipi PFS NR vs chemo 5.8 months; HR 0.21).
Take Home: If your lab has access to either a validated MSI PCR CDx or a validated MMR IHC CDx (but not both), these data suggest you’re unlikely to “miss out” on the nivo+ipi PFS effect by not requiring dual-positive confirmation—though the small discordant (MSS/pMMR) subsets and post hoc design mean we still need to be careful about over-reading the HR differences.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3531
Panitumumab (EGFR mAb) maintenance benefit holds regardless of baseline dNLR in RAS-WT mCRC—dNLR mainly prognostic for OS
Baseline derived neutrophil-to-lymphocyte ratio (dNLR; cut-off 2.2) was assessed post hoc in the PANAMA/AIO KRK0212 randomized Phase 2 trial (NCT01991873) of first-line RAS WT mCRC receiving FU/FA ± panitumumab (EGFR-targeting monoclonal antibody) maintenance after induction FOLFOX+panitumumab. Among 241 patients with baseline counts (dNLR ≤2.2: n=140; >2.2: n=101), dNLR was prognostic for OS from maintenance start (30.8 vs 22.7 months; HR 1.56, 95% CI 1.15–2.12; p=0.004) but not for maintenance PFS (10.1 vs 9.7 months; p=0.25). dNLR did not predict benefit from adding panitumumab to FU/FA during maintenance (interaction p=0.72 for PFS; p=0.294 for OS), with PFS in the expected PANAMA range for FU/FA+panitumumab vs FU/FA alone (~8.8 vs ~5.6 months in dNLR ≤2.2; ~9.7 vs ~5.8 months in dNLR >2.2), but reinduction PFS showed a signal: in dNLR ≤2.2, reinduction PFS was shorter after FU/FA+panitumumab maintenance than after FU/FA maintenance (2.6 vs 7.4 months; interaction p=0.036), raising the possibility of EGFR-pressure biology or treatment-sequencing effects rather than a pure “predictive biomarker” story; key anti-EGFR maintenance toxicities (rash, hypomagnesemia) are not detailed in this abstract and would be needed to weigh sequencing tradeoffs versus the established PANAMA maintenance PFS gain over FU/FA alone.
Take Home: If dNLR is validated prospectively, it looks more useful for counseling prognosis than choosing FU/FA±panitumumab maintenance—but that reinduction interaction is a yellow flag that sequencing (continuous EGFR pressure vs “EGFR holiday”) might matter for patients with lower inflammatory burden, and it deserves a biology-driven follow-up rather than overcalling it as predictive today.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e15552
Panitumumab (EGFR mAb) drives ctDNA “High→Low” conversions that track ~51–61mo OS in 1L mCRC
ctDNA dynamics were analyzed in PARADIGM (Phase 3; mFOLFOX6 + panitumumab vs mFOLFOX6 + bevacizumab) using paired baseline and post-treatment plasma from 556 patients with mCRC, stratified into High/Mid/Low ctDNA by Gaussian mixture modeling (NCT02394795; NCT02394834). Baseline ctDNA was strongly prognostic for OS (Low: 50.7mo; Mid: 34.1mo; High: 27.8mo), and post-treatment ctDNA further refined risk: patients converting to Low ctDNA had the best survival even from a High baseline (High→Low OS 50.9mo), while persistently high ctDNA did worst (High→High OS 23.3mo). In the High-baseline subset, more patients reached Low ctDNA with panitumumab vs bevacizumab (19.1% vs 10.8%, p=0.053) alongside more discontinuations after curative resection (e.g., High→Low: 62.1% vs 43.8%), with directionally better OS/PTS for panitumumab (High→Low OS 61.0 vs 43.1mo; HR 0.49), consistent with PARADIGM’s established OS advantage for panitumumab over bevacizumab in left-sided RAS WT disease (~37.9 vs 34.3mo) despite similar imaging-based PFS.
Take Home: If prospectively validated with standardized thresholds and earlier on-treatment timepoints (not just “at discontinuation”), ctDNA trajectory could become a more patient-relevant barometer than RECIST PFS—flagging who is actually on track for long survival and potential conversion to resectability, where PARADIGM already suggests EGFR-first can outperform bevacizumab-first in RAS WT/left-sided mCRC.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3573
Zanzalintinib (multi-kinase TAM/MET/VEGFR inhibitor) + atezolizumab (PD-L1) shows OS/PFS benefit—and atezo exposure appears to matter—in refractory mCRC
Zanzalintinib (zanza) is an oral multi-kinase inhibitor (TYRO3/AXL/MER, MET, VEGFRs) combined with atezolizumab (atezo), an anti–PD-L1 antibody, in the Phase 3 STELLAR-303 trial in previously treated metastatic CRC (largely MSS/pMMR). In this immunogenicity/exposure analysis among 451 pts on zanza+atezo (394 TEADA-evaluable), both TEADA-positive (n=143) and TEADA-negative (n=251) subgroups retained benefit vs regorafenib (n=450): PFS HR 0.81 (95% CI 0.67–0.99) and 0.59 (0.51–0.71), respectively; OS HR 0.79 (0.63–0.98) and 0.72 (0.60–0.86), respectively, with the strongest efficacy in TEADA-negative pts. Exposure–response modeling (n=376) linked higher time-varying atezo trough concentrations with better outcomes (95th vs 5th percentile HR 0.82 [0.80–0.83] for PFS; 0.73 [0.70–0.75] for OS; p<0.001), while safety appeared comparable by TEADA status; in the parent STELLAR-303 readout, the combo improved OS vs regorafenib (median OS 10.9 vs 9.4 months; stratified HR 0.80, p=0.0045), an incremental gain over an established late-line SOC with modest survival expectations (regorafenib historically mOS ~6.4 months vs placebo in CORRECT).
Take Home: If the atezo exposure–efficacy signal holds up (and isn’t just “healthier patients clear drug more slowly”), it strengthens the case that the PD-L1 backbone is doing real work here—making this combo more than just another regorafenib-like TKI swap in a tough, largely MSS mCRC setting.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3574
Gastric / GEJ
🏆 SPOTLIGHT
Can first-line HER2-positive gastric cancer finally break the 2-year median survival barrier? Zanidatamab plus tislelizumab and chemotherapy delivered a remarkable 26.4-month median OS and outperformed trastuzumab-based therapy across PD-L1 subgroups.
Zanidatamab (dual HER2 bispecific; ECD2/4) + chemo in 1L HER2+ mGEA shows early ERBB2 ctDNA clearance tied to 12.5-mo mPFS
Zanidatamab is a dual HER2-targeted bispecific antibody (binding HER2 extracellular domains 2 and 4) evaluated with platinum/fluoropyrimidine chemotherapy (mFOLFOX6, CAPOX, or 5-FU/cisplatin) in a Phase 2, 1L HER2+ advanced/metastatic gastroesophageal adenocarcinoma trial (NCT03929666; n=46; 89% centrally confirmed HER2+). Confirmed ORR was 76.2% (95% CI 60.5–87.9) with median PFS 12.5 months (95% CI 8.2–21.8)—numerically above historical trastuzumab + chemo benchmarks from ToGA (ORR ~47%, mPFS 6.7 months) in a similar 1L HER2+ population, noting cross-trial differences and that modern practice often adds PD-1 for eligible patients. Baseline plasma ERBB2 amplification by ctDNA NGS showed 94% concordance with tissue FISH (34/36); among patients with detectable baseline ERBB2 amplification, median PFS was 18.6 months (95% CI 12.0–NE), and all responders/SD with available samples had undetectable on-treatment ERBB2 amplification, while PD/SD on-treatment samples showed persistence/recurrence of ERBB2 alterations plus PI3K/AKT and RAS/MAPK co-alterations; 94% (15/16) of matched samples had >80% ctDNA drop on-treatment (Guardant Infinity panel).
Take Home: If these early ctDNA dynamics (rapid ERBB2 clearance; emergence of PI3K/RAS co-alterations in non-responders) replicate prospectively, they could become a practical way to (1) confirm “true” HER2-driven disease up front and (2) flag patients who need escalation/switch strategies earlier than imaging—though the clinical utility hinges on standardized timepoints, adequate matched sampling, and outcome-linked thresholds.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4050
🏆 SPOTLIGHT
The biggest immunotherapy advance in resectable gastric cancer is MATTERHORN showing durvalumab can improve event-free survival on top of perioperative FLOT, bringing meaningful checkpoint blockade benefit into a curative-intent population where prior perioperative IO efforts have largely disappointed.
Durvalumab (PD-L1 blockade) + perioperative FLOT boosts EFS across completion strata in resectable G/GEJ cancer (MATTERHORN)
Durvalumab (anti–PD-L1 monoclonal antibody) was added to perioperative FLOT (5-FU/leucovorin/oxaliplatin/docetaxel) in MATTERHORN, a global, double-blind Phase 3 trial in untreated, resectable gastric/gastroesophageal junction adenocarcinoma (NCT04592913). In this treatment-period analysis, EFS favored durvalumab + FLOT vs placebo + FLOT regardless of which perioperative components were started/completed, with the strongest outcomes in patients completing all adjuvant immunotherapy (completed all adjuvant D/Pbo: 24-mo EFS 91.8% vs 87.2%; HR 0.63 [95% CI 0.41–0.97]; received any adjuvant but did not complete D/Pbo: 24-mo EFS 46.7% vs 29.8%; HR 0.65 [0.46–0.91]; neoadjuvant + surgery but no adjuvant: 24-mo EFS 49.2% vs 33.3%; HR 0.65 [0.38–1.13]; neoadjuvant started but no surgery: 24-mo EFS 10.0% vs 6.2%; HR 0.80 [0.55–1.18]) (subgroup/landmark readouts; interpret cautiously). Safety burden was largely chemotherapy-driven overall (Grade 3–4 AEs 71.6% vs 71.2%; SAEs 48.2% vs 44.1%), while the adjuvant monotherapy period was markedly lower in both arms (Grade 3–4 AEs 21.2% vs 20.5%; SAEs 14.5% vs 14.5%), with discontinuations of durvalumab/placebo in adjuvant monotherapy at 5.8% vs 2.7%—a modest added drop-off that looks within the expected range for PD-(L)1 maintenance in curative-intent settings.
Take Home: If the headline MATTERHORN EFS/OS win holds up broadly, these period-by-period curves suggest the biggest “patient-level unlock” may simply be getting more people through surgery and into adjuvant durvalumab—because the incremental toxicity during IO-only maintenance looks modest, but the EFS separation is most compelling in those who can actually stay on treatment.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4070
Tislelizumab (PD-1) + chemo adds ~2.6 months Q‑TWiST in peritoneal mets gastric/GEJ cancer
Tislelizumab (TIS), an anti–PD-1 monoclonal antibody engineered to reduce FcγR binding, was evaluated with fluoropyrimidine/platinum chemotherapy versus placebo (PBO) + chemotherapy in the global Phase 3, double-blind RATIONALE-305 trial (NCT03777657) in previously untreated advanced gastric/gastroesophageal junction adenocarcinoma; this post hoc analysis used long-term follow-up (data cutoff February 28, 2024; max follow-up 57 months) and applied Q‑TWiST to quantify quality-adjusted survival. In patients with peritoneal metastasis (PM), mean Q‑TWiST was 13.4 months with TIS+CT vs 10.8 months with PBO+CT (Δ +2.6 months; 95% CI 0.1–5.0; p=0.04), with relative gains of 16.8% in PD‑L1 TAP ≥1% and 33.0% in PD‑L1 TAP ≥5% (Δ +4.7 months; 95% CI 1.1–8.6; p=0.02); a sensitivity analysis using trial-derived EQ‑5D utilities favored TIS+CT similarly (PM ITT Δ +3.1 months; p<0.01). While this abstract doesn’t report PM-specific median OS/PFS or grade ≥3 AE rates, the Q‑TWiST lift is directionally consistent with the parent RATIONALE-305 OS benefit in the overall population (mOS 15.0 vs 12.9 months; HR 0.80), and it helps contextualize value vs the current 1L US SOC of PD‑1 + platinum/fluoropyrimidine where pembrolizumab+chemo improved mOS to 12.9 vs 11.5 months (HR 0.78) in KEYNOTE‑859.
Take Home: If this Q‑TWiST signal holds up in prospective PM-focused analyses, it suggests PD‑1+chemo isn’t just “adding months” in a notoriously symptomatic subgroup—it may be adding better months, but we still need the PM subgroup’s hard endpoints (median OS/PFS, toxicity/discontinuation) to judge whether tislelizumab truly differentiates from other PD‑1+chemo options.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4034
SHR-A1904 (garetatug rezetecan) CLDN18.2 ADC posts 25.5% ORR in heavily pretreated CLDN18.2+ solid tumors
SHR-A1904 is an investigational claudin-18.2 (CLDN18.2)–targeting antibody–drug conjugate evaluated in a global Phase 1/2a dose-escalation/optimization study in CLDN18.2-expressing relapsed/refractory solid tumors (NCT05277168). As of December 1, 2025, 51 patients were enrolled (GC/GEJC n=41; pancreatic cancer n=9; lung adenocarcinoma n=1); most had ≥2 prior lines (72.5%), and 42 received ≥3.6mg/kg Q3W (24 Asian; 18 non-Asian). Across all patients, ORR was 25.5% with median DoR 5.7 months, median PFS 2.8 months, and median OS 9.8 months—numerically in the ballpark of typical later-line chemo outcomes in unselected GC/GEJ (often ORR ~10–20% and PFS ~2–4 months with options like taxanes/irinotecan ± ramucirumab), while the non-Asian subgroup showed higher ORR 44.4% but with small numbers and wide CIs. Safety was dominated by GI toxicity: TRAEs 94.1% with nausea 66.7% and vomiting 52.9%; Grade ≥3 TRAEs 41.2% and serious TRAEs 17.6%, with 1 DLT (Grade 3 vomiting) at 6.0mg/kg and no treatment-related deaths—important given CLDN18.2 programs (e.g., zolbetuximab) are also notorious for nausea/vomiting.
Take Home: If this ORR/DoR holds up in a larger, GC/GEJ-focused expansion (and especially if it consistently beats the ~10–20% ORR typical of later-line chemo), SHR-A1904 could become a meaningful CLDN18.2 option beyond first-line zolbetuximab—but the very high nausea/vomiting burden and the eye-catching non-Asian subgroup efficacy both need confirmation in bigger, more homogeneous cohorts.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e15018
Avatrombopag (TPO-R agonist) sharply improves platelet recovery and “on-time” chemo delivery in persistent CIT (ACT‑GI)
ACT‑GI (NCT05772546) randomized patients with gastrointestinal malignancies and persistent chemotherapy‑induced thrombocytopenia (Plt <85×10^9/L on day 1 of a planned cycle) to avatrombopag 40mg PO daily vs placebo in a lead‑in (chemo held up to 2 weeks) then one on‑cycle full‑dose chemo cycle if Plt recovered to ≥100×10^9/L. At a prespecified interim analysis, the DSMB stopped enrollment for efficacy: 70% (16/23) on avatrombopag met the composite primary endpoint (Plt ≥100×10^9/L within lead‑in and maintained ≥100×10^9/L at cycle end) vs 17% (4/24) on placebo (p<0.001); end‑of‑cycle median platelets were 157 vs 72×10^9/L, and no patients required platelet transfusion. Safety showed more AEs/SAEs with avatrombopag (74%/13% vs 46%/0%), with no study‑drug–related SAEs and no discontinuations/deaths attributed to study drug; bleeding events occurred in 2 avatrombopag patients (including 1 intracranial hemorrhage). Versus current “SOC” for CIT—dose delays/reductions and transfusions mainly reserved for severe thrombocytopenia/bleeding risk (often prophylaxis around ~10×10^9/L in stable patients)—these data suggest a pharmacologic path to preserve dose intensity in a population where there is no FDA‑approved CIT therapy.
Take Home: If this platelet “rescue + maintenance” effect holds in larger, chemo‑regimen–stratified studies (and thrombotic/bleeding risks are clarified), avatrombopag could become a practical way to keep metastatic GI patients on schedule rather than repeatedly dose‑reducing or delaying therapy for platelet counts stuck in the 50–100×10^9/L range.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3580
Zanidatamab (dual-HER2 bispecific) + PD-1 tislelizumab delivers OS gains regardless of PD-L1 in 1L HER2+ mGEA
Zanidatamab is a HER2-targeted bispecific antibody that binds two HER2 epitopes; in HERIZON-GEA-01 (Phase 3; NCT05152147), it was combined with platinum/fluoropyrimidine chemotherapy (CAPOX or FP) plus the anti–PD-1 tislelizumab versus trastuzumab + chemotherapy in previously untreated HER2+ locally advanced/metastatic gastroesophageal adenocarcinoma (zanidatamab+tisl+CT n=302; tras+CT n=308; 26-month median follow-up). In the ITT population, zanidatamab+tisl+CT improved mPFS to 12.4 vs 8.1 months (HR 0.63, p<0.001) and mOS to 26.4 vs 19.2 months (HR 0.72, p=0.004)—an OS that exceeds the current 1L benchmark of pembrolizumab+trastuzumab+chemo in KEYNOTE-811 (mOS 20.0 vs 16.8 months; HR 0.80) and materially outperforms historical trastuzumab+chemo ToGA-era expectations (mOS 13.8 months). PD-L1 subgroup analyses (VENTANA SP263; TAP and CPS) suggested benefit was maintained in PD-L1–negative (e.g., TAP<1%: mPFS 18.5 vs 7.9 months; mOS 29.7 vs 15.8 months) and PD-L1–positive disease (TAP≥1%: mPFS 11.3 vs 8.3 months; mOS 26.4 vs 21.2 months), though post-progression therapy imbalances (more subsequent checkpoint inhibitor and HER2-targeted therapy in the tras+CT arm) complicate clean OS interpretation across PD-L1 strata.
Take Home: If the NEJM-level survival signal holds up with longer follow-up, zanidatamab+tislelizumab+chemo looks like it could become the new “PD-L1-agnostic” 1L HER2+ mGEA default—especially notable because today’s SOC immunotherapy add-on is often most compelling in PD-L1–enriched populations, and cross-trial comparisons suggest this regimen may be clearing the KEYNOTE-811 bar rather than merely matching it.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4010
ONO-4578 (EP4 antagonist) + nivolumab + oxaliplatin/fluoropyrimidine extends PFS in 1L HER2-negative G/GEJ
ONO-4578 is an oral prostaglandin E2 (PGE2)–EP4 receptor antagonist intended to reverse myeloid-driven immunosuppression and potentially deepen anti–PD-1 benefit when added to nivolumab + chemotherapy. In the randomized, double-blind Phase 2 ONO-4578-08 trial (NCT06256328) across Japan/Korea/Taiwan, chemo-naïve HER2-negative unresectable adv/rec gastric/GEJ cancer patients received ONO-4578 40mg QD vs placebo (2:1) with nivolumab 360mg Q3W + SOX/CAPOX (n=226; 150 vs 76): median PFS 9.0 vs 6.9 months (HR 0.67; 90% CI 0.48–0.92; p=0.040), ORR 62.0% vs 48.7%, and early OS signal (not reached vs 12.7 months; HR 0.60; 95% CI 0.37–0.96) at ~8.5-month median follow-up. Safety looked directionally tougher on the triplet—diarrhea (55.7%), anemia (55.0%), neuropathy (50.3%) and serious TEAEs (53.7% vs 42.7%)—while efficacy benchmarks appear at least competitive with the current anti–PD-1+chemo backbone (e.g., pembrolizumab+chemo in KEYNOTE-859: median PFS 6.9 vs 5.6 months; ORR 51% vs 42%).
Take Home: If the OS trend holds with longer follow-up, EP4 blockade could become a practical way to “tune” the myeloid microenvironment on top of PD-1+chemo in 1L HER2-negative gastric/GEJ—but the higher serious-TEAE rate means any Phase 3 needs to prove this is more than a modest PFS bump.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4007
Pembrolizumab + Trastuzumab (PD-1/HER2) Sustains OS Gain in HER2+ G/GEJ Cancer
Pembrolizumab is a PD-1 inhibitor added to trastuzumab, an anti-HER2 antibody, plus platinum/fluoropyrimidine chemotherapy in the randomized Phase 3 KEYNOTE-811 first-line trial for untreated advanced/metastatic HER2-positive gastric/GEJ adenocarcinoma (n=698; NCT03615326). At 70.0 months median follow-up, OS improved to 20.0 vs 16.8 months in ITT (HR 0.81) and 20.1 vs 15.7 months in PD-L1 CPS≥1 tumors (HR 0.79), while PFS was 10.0 vs 8.1 months in ITT (HR 0.73) and 10.9 vs 7.3 months in CPS≥1 (HR 0.72); ORR was 72.6% vs 60.1% in ITT. Grade ≥3 TRAEs were higher with pembrolizumab (58.9% vs 51.2%) with no new safety signals, and the OS exceeds historical ToGA trastuzumab+chemo mOS of 13.8 months, though the incremental pembrolizumab gain remains modest.
Take Home: This looks like a durable but incremental win—important for patients with CPS≥1 HER2+ disease, but not a dramatic leap beyond the trastuzumab-chemo backbone.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4040
Bladder Cancer (Urothelial, Urethral, Ureteral)
🏆 SPOTLIGHT
Adding sacituzumab govitecan to avelumab added ~7 mo progression-free survival across every metastatic-site subgroup examined, including liver and bone metastases where durable disease control remains notoriously difficult.
Ave + SG (TROP2 ADC) extends PFS across visceral/liver/bone subgroups vs avelumab alone in 1L maintenance aUC
Avelumab is a PD-L1 inhibitor, and sacituzumab govitecan (SG) is a TROP2-directed ADC delivering SN-38; this randomized Phase 2 JAVELIN Bladder Medley trial (NCT05327530) tested Ave + SG vs avelumab monotherapy as 1L maintenance in unresectable/metastatic urothelial carcinoma without progression after 4–6 cycles of platinum chemo (n=111; 74 vs 37; randomized 2:1; stratified by visceral metastases at start of chemo). At the Apr 28, 2025 cutoff (median PFS follow-up 15.7 vs 25.1 months), PFS favored Ave + SG in every metastatic-site subgroup reported: visceral 9.03 vs 2.20 months (HR 0.49), nonvisceral 14.69 vs 7.33 (HR 0.61), bone 9.26 vs 2.33 (HR 0.40), lung 8.77 vs 1.81 (HR 0.43), liver 8.77 vs 2.69 (HR 0.48), lymph node–only 14.00 vs 7.59 (HR 0.65); OS was immature, and the control arm’s visceral/nonvisceral estimates were extended using propensity score–weighted JAVELIN Bladder 100 data per protocol. Toxicity was the clear tradeoff: grade ≥3 TRAEs were 57.9%–82.4% with Ave + SG across metastatic-site subgroups vs 0%–9.1% with avelumab alone; for context, SOC 1L maintenance avelumab (vs best supportive care) improved OS to 23.8 vs 15.0 months (HR 0.76) in JAVELIN Bladder 100, but did not carry the same high grade ≥3 AE burden seen with adding SG here.
Take Home: If the ~7-month absolute PFS gain in visceral disease (9.0 vs 2.2 months) holds up in a confirmatory trial, Ave + SG could become a serious “go-to” maintenance intensification for the worst-prognosis metastatic patterns—but the very high grade ≥3 toxicity rate means durability (OS), dose optimization, and patient selection will have to do the heavy lifting.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4568
🏆 SPOTLIGHT
The next-generation topo-I ADC LY4052031 produced a 40% response rate in enfortumab-pretreated urothelial cancer, suggesting that changing the payload—not the target—may unlock a new treatment sequence.
LY4052031 (Nectin-4 topo1 ADC) delivers 48% ORR in metastatic urothelial cancer, including post-enfortumab activity
LY4052031 is a next-generation anti–Nectin-4 antibody-drug conjugate (ADC) with a cleavable linker and a novel topoisomerase I inhibitor payload (Camp98; DAR 8) evaluated in the NEXUS-01 Phase 1 dose-escalation study (NCT06465069) in previously treated or standard-therapy-ineligible locally advanced/metastatic urothelial carcinoma (mUC) and other solid tumors. As of Nov 25, 2025, 70 patients were treated (47 mUC), and in efficacy-evaluable mUC with CYP2D6 activity score (AS) ≥0.5 receiving 2.4–4.8mg/kg Q3W (n=21), ORR was 48% (1 CR, 9 PR) with DCR 81% and median DoR 7 months (95% CI, 2.7–NE); in prior enfortumab vedotin (EV)-treated mUC (n=15), ORR was 40% (DCR 80%), which sits above typical post-platinum/ICI benchmarks like sacituzumab govitecan’s ~27% ORR in TROPHY-U-01 (not an EV-post population, but a useful “late-line” yardstick). Safety was chemo-like and topo1-flavored (nausea 43%, alopecia 39%, fatigue 37%, mucositis 27%), with anemia the most common grade ≥3 TEAE (13%); a notable development was CYP2D6 low-AS association with increased DLTs at higher doses, prompting a protocol amendment requiring baseline CYP2D6 genotyping and dedicated dose-finding for low AS (<0.5).
Take Home: If the 40% ORR in EV-pretreated mUC holds up with longer follow-up (and across CYP2D6 strata), LY4052031 could become a credible “same-target, different-payload” way to recycle Nectin-4 after EV—though the need for genotype-guided dosing may complicate real-world rollout and combination plans.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4508
🏆 SPOTLIGHT
Can more than half of cisplatin-ineligible bladder cancer patients achieve pathologic complete response before surgery? Perioperative enfortumab vedotin plus pembrolizumab delivered pCR rates approaching 60% and cut the risk of recurrence or death by more than half in key MIBC subgroups, positioning the regimen as a major new standard.
Enfortumab vedotin (Nectin-4 ADC) + pembrolizumab (PD-1) delivers ~57% pCR and halves death risk in cisplatin-ineligible MIBC
Enfortumab vedotin (a Nectin-4–directed MMAE antibody–drug conjugate) plus pembrolizumab (anti–PD-1) was evaluated as perioperative (neoadj-adj) therapy in the Phase 3 KEYNOTE-905/EV-303 randomized trial (NCT03924895) in cisplatin-ineligible or cisplatin-declining MIBC (T2–T4aN0M0 or T1–T4aN1M0; n=344; data cutoff June 6, 2025; median follow-up 25.6 months). Across clinically relevant subgroups by baseline stage, EV+pembro improved EFS and deepened pathologic response vs radical cystectomy + pelvic lymph node dissection (RC+PLND) alone: T3–T4aN0 had EFS HR 0.43 with mEFS NR vs 17.2mo, OS HR 0.54 with mOS NR vs 41.7mo, and pCR 57.1% vs 6.8%; smaller cohorts also favored EV+pembro (T2N0 EFS HR 0.26; pCR 60.0% vs 18.8%; T1–4aN1 mEFS 36.7 vs 7.2mo; pCR 42.9% vs 0%), though some OS reads are immature (<10 events). As a practical benchmark, neoadjuvant cisplatin-based chemo (for eligible patients) typically yields pT0 rates in the ~20–40% range, so the ~57% pCR here is above historical chemo expectations—while also addressing the cisplatin-ineligible population where RC alone has low pCR (8.6% in this trial).
Take Home: If the OS benefit holds up with longer follow-up, perioperative EV+pembro looks like a true practice-changer for cisplatin-ineligible MIBC—turning a “surgery-only” population with single-digit pCR into one where >1 in 2 patients are ypT0N0, but we still need the full safety/discontinuation story and durability by nodal stage before calling it universally applicable.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4613
🏆 SPOTLIGHT
What if a month of oral targeted therapy could eliminate recurrent bladder tumors before TURBT? Erdafitinib produced a 75% response rate and 55% complete response rate in FGFR3-mutant NMIBC (mFGFR3 occurs in ~50% of NMIBC) without any reported grade ≥3 toxicity, suggesting a potential new paradigm.
Erdafitinib (pan-FGFR inhibitor) delivers 55% CR chemoablation in FGFR3-mutant recurrent NMIBC
Oral erdafitinib, a pan-FGFR1–4 tyrosine kinase inhibitor approved in metastatic urothelial carcinoma, was tested as short “chemoablation” in a Phase II window-of-opportunity study before TURBT in recurrent FGFR3-altered NMIBC (NCT04917809). Across 20 patients with FGFR3-mutant NMIBC (70% prior low-grade Ta; all had ≥1 prior intravesical therapy and 55% had multiple prior intravesical courses), 28 days of erdafitinib 6mg daily produced a RECIT-Bladder ORR of 75% (15/20) including 55% CR (11/20), with no grade ≥3 toxicities reported—an attractive tolerability profile versus intravesical approaches (e.g., BCG/mitomycin/gemcitabine) that reduce recurrence risk but are not typically “chemoablative” in a single pre-TURBT window and can drive local/systemic AEs and treatment discontinuations in real-world practice. FGFR3 alterations are enriched in NMIBC (often cited up to ~80% in some NMIBC series, though estimates vary by cohort/assay), making this a biomarker-defined strategy with potentially meaningful reach if durability holds.
Take Home: If the CRs prove durable (and not just a short-term debulking effect before TURBT), a 4-week, well-tolerated oral FGFR blocker could become a pragmatic “break-the-resection-cycle” option for FGFR3-mutant recurrent NMIBC—especially for patients worn out by repeated intravesical courses.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4623
Zelenectide pevedotin (BT8009) Nectin-4–targeted BDC + pembrolizumab hits 50% cORR and 13.0-mo mPFS in 1L cisplatin-ineligible la/mUC
Zelenectide pevedotin (zele; BT8009) is a Nectin-4–targeted Bicycle Drug Conjugate delivering MMAE (small 4.2 kDa construct with short plasma half-life), combined here with PD-1 blockade (pembrolizumab) in Duravelo-1 expansion cohort B7 (Phase 1/2; NCT04561362) for previously untreated, cisplatin-ineligible locally advanced/metastatic urothelial carcinoma (n=22; median age 77; ECOG PS 2 in 45%). At a July 1, 2025 cutoff (median follow-up 11.8 months), confirmed ORR was 50% (11/22) with 5 CRs and DCR 82%, and mPFS was 13.0 months (95% CI 3.8–NE); responses are numerically above historical pembrolizumab monotherapy in this setting (KEYNOTE-052 ORR 29% with ~9%–10% CR), though cross-trial comparisons are shaky and DOR is not mature here. Grade ≥3 TRAEs included ALT increase (18%) and neutropenia/asthenia (14% each); treatment-related Grade 3 AECIs included peripheral neuropathy (14%, leading to withdrawal in 2 patients) and skin reactions (9%) with no Grade 4/5 treatment-related AECIs—contrasting with the established Nectin-4 ADC + pembrolizumab standard (enfortumab vedotin+pembrolizumab) that delivers high responses and OS/PFS benefit in Phase 3 but is clinically limited by rash and neuropathy risk.
Take Home: If durability holds up, a “fast in/fast out” Nectin-4 MMAE conjugate that can pair with pembrolizumab while keeping skin toxicity manageable could be meaningful for frail 1L cisplatin-ineligible mUC patients—but with only 22 patients and immature DOR, Duravelo-2 will need to show it can compete with (or carve out a safer niche than) EV+pembro rather than just beat historical pembro-alone.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4564
Avelumab (PD-L1) 1L maintenance extends OS regardless of NECTIN4/HER2 RNA in la/mUC
Avelumab is an anti–PD-L1 monoclonal antibody used as first-line (1L) maintenance after platinum-based chemotherapy (PBC) in unresectable locally advanced/metastatic urothelial carcinoma (la/mUC). In the Phase 3 JAVELIN Bladder 100 trial (NCT02603432), progression-free patients after 4–6 cycles of 1L PBC with RNAseq biomarker data (n=560) were stratified by tumor NECTIN4 and HER2 (ERBB2) RNA expression (median split): OS favored avelumab + best supportive care (BSC) vs BSC alone across all combined-expression groups—high/high mOS 23.0 vs 14.3 months (HR 0.73), low/low 29.8 vs 18.8 (HR 0.78), high/low 18.4 vs 17.4 (HR 0.84), with the abstract also stating benefit by each marker alone (safety by subgroup not reported here). This is directionally consistent with the overall JAVELIN Bladder 100 SOC result (avelumab maintenance improved OS vs BSC after response/SD on 1L platinum), suggesting NECTIN4/HER2 RNA—while relevant for ADC targeting biology—doesn’t look like a practical selection biomarker for avelumab maintenance based on these exploratory splits.
Take Home: If anything, the “pronounced” OS in the low-NECTIN4 strata reads more like prognostic noise/biology (and small subgroup N) than a clean predictive signal—so until we see interaction p-values and multivariable modeling, don’t over-rotate on NECTIN4/HER2 RNA to decide who should (or shouldn’t) get avelumab maintenance.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4575
Durvalumab (PD-L1) + tremelimumab (CTLA-4) neoadjuvant combo hits 44% pCR—but higher CTLA-4 dose triggers serious irAEs in cisplatin-ineligible MIBC
Durvalumab (anti–PD-L1) plus tremelimumab (anti–CTLA-4) was evaluated in a pilot neoadjuvant study for cisplatin-ineligible, high-risk muscle-invasive urothelial carcinoma (NCT02812420), with cystectomy planned 4–6 weeks after therapy. In the higher-dose tremelimumab cohort (D 1500mg + T 300mg; n=13), 9 patients underwent cystectomy; among those, pCR was 44% (4/9) with 1-year RFS 67% (SE 16%) and 1-year OS 62% (SE 14%), but 3 grade ≥3 irAEs (colitis, hepatitis, and myositis/myocarditis) led to early cohort termination and 2 patients became unable to proceed to surgery due to toxicity (both later died). With longer follow-up in the lower-dose cohort (D 1500mg + T 75mg; n=28), 5-year OS was 82% (SE 7%) and 5-year RFS was 74% (SE 9%); as a benchmark, single-agent neoadjuvant atezolizumab in cisplatin-ineligible/refusing MIBC (ABACUS) delivered pCR 31% and 1-year RFS 79%, suggesting the pCR signal here is in-range-to-higher while the higher-dose CTLA-4 toxicity looks less tolerable for a curative-intent setting. Baseline tertiary lymphoid structures (TLS) associated with downstaging/response (TLS-high responders 69% vs 29% TLS-low; OR 0.18, 95% CI 0.04–0.75) and trended toward improved RFS/OS (RFS HR 0.24, p=0.047; OS HR 0.32, p=0.068).
Take Home: If dosing can be de-risked (or CTLA-4 swapped/attenuated), this reinforces “some patients can be cured with neoadjuvant ICI alone” in cisplatin-ineligible MIBC—and TLS density could become a practical way to pick who’s worth exposing to dual-checkpoint toxicity versus a gentler PD-(L)1-only approach.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e16584
Zelenectide pevedotin (BT8009) Nectin‑4 BDC + pembrolizumab delivers ~55–58% ORR in 1L la/mUC (Duravelo‑2 IA)
Zelenectide pevedotin (zele; BT8009) is a Nectin‑4–targeted Bicycle Drug Conjugate (BDC) evaluated with pembrolizumab in the Phase 2/3 Duravelo‑2 trial (NCT06225596; Cohort 1, previously untreated la/mUC; randomized 1:1:1 vs gemcitabine + cisplatin/carboplatin ± avelumab). At 27 weeks’ follow-up, confirmed BICR ORR was 55% (16/29; CR 24%) with zele 5mg/m² D1/8/15 + pembro and 58% (15/26; CR 31%) with zele 6mg/m² D1/8 + pembro (n=30 per zele arm; median zele duration 6.21 months); by comparison, current 1L SOC enfortumab vedotin + pembrolizumab has reported BICR ORR ~67.5% (CR ~30.4%), while historical platinum/gemcitabine control arms have ORR ~44%. Zele-related AEs occurred in 90% (Gr≥3 40%) at 6mg/m² (dose reductions 20%, discontinuations 3%); common Gr≥3 zele-related AEs at 6mg/m² (≥5%) were neutropenia (10%), neutrophil count decreased (7%), and anemia (7%); AECIs at 6mg/m² included peripheral neuropathy 37% (Gr≥3 3%), skin reactions 17% (no Gr≥3), and eye disorders 10% (no Gr≥3), with no zele-related severe skin reactions reported at 6mg/m².
Take Home: If durability holds up, zele’s “EV-like” Nectin‑4 biology with a potentially cleaner skin-toxicity profile could matter for real patients—especially those who struggle on EV+pembro—though the bar is now EV‑302’s ~67.5% ORR with similar CR rates, so differentiation will likely need either better tolerability in practice and/or standout DOR/PFS.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4516
LY4101174 (Nectin-4–targeted exatecan ADC) shows 18% ORR post-enfortumab vedotin in mUC—but with heavy hematologic toxicity
LY4101174 is a next-generation anti–Nectin-4 antibody-drug conjugate delivering the topoisomerase I inhibitor exatecan (DAR 8; maleimide–β-glucuronide poly-sarcosine linker) being evaluated in the EXCEED Phase 1 dose-escalation study (NCT06238479) in heavily pretreated locally advanced/metastatic urothelial carcinoma and selected solid tumors. As of December 3, 2025, 143 patients were treated; among 66 efficacy-evaluable mUC patients receiving 2.4–4.0mg/kg Q3W (92% EV-pretreated; median 4 prior lines), ORR was 18% (12/66; 12 PRs) and DCR was 70% (46/66), with short median follow-up (4.9 months). Toxicity was dominated by cytopenias (grade≥3 anemia 47%, neutropenia 29%, thrombocytopenia 21%, febrile neutropenia 10%), including DLTs at 3.2mg/kg Q2W prompting a move to Q3W dosing; G-CSF prophylaxis was required at 4.0mg/kg Q3W, while TRAEs drove dose reductions in 24% and discontinuations in 2%—a tolerability profile that looks at least as myelosuppressive as other topo-I ADCs and contrasts with 1L EV+pembrolizumab efficacy benchmarks (EV-302 ORR ~68%, mPFS 12.5 months), though that’s a very different (frontline) population.
Take Home: If LY4101174 can find a tolerable dose/schedule (or smarter supportive-care strategy), an 18% ORR in predominantly EV-exposed mUC suggests Nectin-4 is still “drug-able” after EV—but right now the hematologic toxicity burden may limit how far this exatecan ADC can differentiate in a post-EV landscape that badly needs durable responses, not just incremental disease control.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.4517
Nogapendekin alfa inbakicept-pmln (IL‑15 superagonist) + BCG tops pembrolizumab (PD‑1) on 12‑mo CR in BCG‑unresponsive NMIBC CIS (MAIC)
Nogapendekin alfa inbakicept-pmln (NAI; an IL‑15 superagonist cytokine complex that expands/activates NK and CD8+ T cells) plus intravesical BCG from the Phase 3 QUILT‑3.032 trial (NCT03022825) was compared vs pembrolizumab (PD‑1 inhibitor) from KEYNOTE‑057 (NCT02625961) using an unanchored MAIC/STC in BCG‑unresponsive NMIBC with CIS ± papillary disease. After weighting on 5 shared baseline variables, 12‑month CR favored NAI+BCG vs pembrolizumab (47.3% vs 18.8%; OR 3.88, 95% CI 1.77–8.50) and modeled DOR was longer by +10.65 months (26.85 vs 16.20 months); by contrast, the single‑arm KEYNOTE‑057 publication reports an initial CR of ~41% with median DOR ~16.2 months, with ~46% of responders maintaining CR ≥12 months—i.e., ~19% of the full cohort in durable CR at 12 months, broadly in the same ballpark as the MAIC’s 12‑mo CR estimate for pembrolizumab. At a 43‑month cutoff, PFS trended toward NAI+BCG (median not reached) vs pembrolizumab (39.9 months; HR 0.45, 95% CI 0.17–1.20), with no OS difference; cross‑trial TRAE comparisons were limited, but non‑bladder grade 3–4 TRAEs numerically favored NAI+BCG (OR 0.33, 95% CI 0.00–4.09).
Take Home: If this durability signal holds up outside an unanchored MAIC, NAI+BCG’s higher 12‑month CR could meaningfully reduce cystectomy pressure vs pembrolizumab in CIS‑heavy BCG‑unresponsive NMIBC—but with only 5 matching covariates and wide CIs (PFS/SAE), this is still hypothesis‑generating rather than practice‑changing.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e16619
Durvalumab (PD-L1) + ddMVAC Hits ~49% pCR in Neoadjuvant MIBC; Tremelimumab (CTLA-4) Adds Toxicity
NEMIO (Phase I/II, randomized, noncomparative; NCT03549715) tested ddMVAC plus durvalumab with or without tremelimumab ahead of radical cystectomy in cisplatin-eligible cT2–T4a N0–1 muscle-invasive bladder cancer (119 treated; 113 cystectomy). Bayesian posterior mean pCR (ypT0N0) was 48.70% with ddMVAC+durvalumab (95% CI 35.93–61.56) and 46.27% with ddMVAC+durvalumab+tremelimumab (95% CI 33.92–58.85); in exploratory PD-L1-high tumors, pCR was 68.25% vs 33.49% in PD-L1-low/negative (PD-L1 data n=103). Grade ≥3 TRAEs were 30.48% in the doublet vs 49.63% in the triplet (overall 40.95%); immune-related AEs occurred in 26.9% (grade ≥3 4.2%), and 2-year EFS/OS were 75%/85% (doublet) and 77%/88% (triplet)—directionally above historical neoadjuvant cisplatin chemo benchmarks where pCR is often in the ~25–40% range depending on regimen and series, but the triplet misses the trial’s ≤30% grade ≥3 TRAE bar.
Take Home: If perioperative durvalumab is now the direction of travel in MIBC (post-NIAGARA), these data argue for “chemo + PD-(L)1” rather than adding CTLA-4—pCR looks similar while grade ≥3 toxicity clearly worsens, and the PD-L1-high enrichment signal is hypothesis-generating for smarter patient selection rather than more drugs.
🔗 https://pubmed.ncbi.nlm.nih.gov/42190158
Glioma / Glioblastoma
🏆 SPOTLIGHT
A win for IDH targeting - Vorasidenib pushed mPFS to nearly 50 months in grade 2 IDH-mutant glioma, allowing most patients to defer radiation and chemotherapy for years.
Vorasidenib (dual mutant IDH1/2 inhibitor) delivers ~50-month PFS and delays next intervention in grade 2 IDH-mutant glioma
Vorasidenib (VOR) is an oral, brain-penetrant dual inhibitor of mutant IDH1/2, evaluated in the global, randomized, double-blind Phase 3 INDIGO trial (NCT04164901) in residual/recurrent grade 2 IDH1/2-mutant oligodendroglioma or astrocytoma after surgery only (no prior RT/chemo; no immediate need for chemoradiotherapy), with 331 patients randomized (VOR n=168; placebo n=163). With updated follow-up to January 17, 2025 among patients randomized to VOR (median follow-up 41.6 months), investigator-assessed median PFS was 49.9 months (95% CI 39.6–NE) and median time to next intervention (TTNI) was not reached (95% CI 52.0–NE); at 42 months, 59.2% were progression-free and 75.1% had not yet required chemoradiotherapy or surgery, with ORR 31.5% (95% CI 24.6–39.2) and OS not reached. Versus the original INDIGO blinded comparison (BIRC) at earlier cutoff—median PFS 27.7 vs 11.1 months (HR 0.39) and delayed TTNI with placebo crossover allowed—this update mainly reinforces durability on continuous VOR; safety was consistent with prior reports with no new safety signals (but updated Grade ≥3 AE rates/discontinuations were not provided in the abstract for benchmarking).
Take Home: If these “watch-and-wait after surgery” curves hold up with longer follow-up (and despite placebo-to-VOR crossover), vorasidenib looks like a practical way to buy years before RT/PCV/TMZ—an outcome patients care about—though we still need granular updated toxicity/discontinuation data and mature OS/quality-of-life to understand the true tradeoff versus early definitive chemoradiation strategies.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2010
Vorasidenib (dual mutant IDH1/2 inhibitor) cuts seizure rates ~80% and delays QoL decline in INDIGO grade 2 glioma
Vorasidenib (VOR) is an oral, brain-penetrant dual inhibitor of mutant IDH1/2 evaluated vs placebo in the Phase 3 INDIGO trial (NCT04164901) in treatment-naïve (post-surgery only) grade 2 mIDH1/2 diffuse glioma patients aged ≥12 with controlled baseline seizures. In this exploratory seizure/QoL analysis using pre- and post-unblinding data (placebo patients after crossover excluded), 110/331 (33.2%) reported ≥1 seizure during study; among those with seizure burden, on-treatment seizure rates were 13.2 vs 64.9 per person-year for VOR vs placebo (rate ratio 0.20; p=0.0008), and VOR delayed time to first worsening of FACT-Br (HR 0.38, p=0.0435) and the brain cancer subscale BrCS (HR 0.32, p=0.0263). In context, the “SOC” for this exact INDIGO population is typically active surveillance until progression (no systemic antiseizure benefit expected), while definitive RT+chemo is commonly deferred to avoid neurocognitive toxicity—so a pharmacologic seizure-burden reduction on top of INDIGO’s disease-control benefit (median PFS 27.7 vs 11.1 months vs placebo in the primary report) is clinically additive rather than a head-to-head vs RT/temozolomide comparison.
Take Home: If these exploratory patient-reported seizure and FACT-Br findings hold up in longer follow-up/real-world use, vorasidenib starts to look like more than a “delay RT/chemo” drug—reducing seizure burden (a daily lived outcome) could become a key reason to treat earlier rather than watch-and-wait.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2043
NeoVax (personalized neoantigen long-peptide vaccine + poly-ICLC) plus PD-1 blockade (pembrolizumab) extends OS signals in newly diagnosed GBM
NeoVax is a personalized synthetic long-peptide neoantigen vaccine (up to 20 patient-specific peptides) formulated with the TLR3 agonist adjuvant poly-ICLC, designed to prime mutation-specific T cells; here it was combined with pembrolizumab (anti–PD-1) after gross total resection and radiotherapy in a Phase 1, four-cohort study in newly diagnosed GBM (n=39 enrolled; n=37 initiated vaccination; NCT02287428). Median OS was 36.9 months in MGMT-methylated patients and 19.0 months in MGMT-unmethylated patients, compared with 25.3 and 16.7 months in propensity score–matched historical controls, respectively; neoantigen-specific T-cell responses were seen ex vivo in 65% and in vitro in 97%, and ex vivo immune responders had better OS than non-responders (HR 0.26, 95% CI 0.09–0.77; p=0.015). Safety looked clean (no serious AEs reported), and paired tumor profiling (snRNA-seq/scTCR-seq) showed vaccine-reactive clonotypes and increased intratumoral vaccine-specific T-effector populations post-vaccination—suggesting true CNS trafficking rather than peripheral-only immunogenicity.
Take Home: If this survival separation holds up in a randomized setting, a personalized neoantigen vaccine + PD-1 backbone could finally move the needle in a disease where MGMT status heavily dictates outcomes (and only ~40–45% of GBMs are MGMT-methylated), but the reliance on historical controls and the “responder vs non-responder” analysis mean we should watch closely for selection/immortal-time biases and confirmatory efficacy in larger trials.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2006
DOC1021 (dubodencel) DC vaccine expands memory T cells with low tox in GBM chemoradiation combo
DOC1021 is an autologous dendritic-cell immunotherapy made by tandem loading patient DCs with autologous tumor mRNA plus tumor lysate to drive broad antigen presentation and a cDC1-like, CD8+ effector program; in a pooled Phase I (Study 8148; NCT04552886) + expanded-access cohort, it was given after resection and standard chemoradiation in newly diagnosed/recurrent IDH-wt GBM (N=25; 88% MGMT unmethylated; 24% subtotal resection; 16% recurrent). Safety was largely flu-like symptoms/injection-site reactions with no DLTs; one grade 3 cerebral edema resolved and dosing continued, and pseudo-progression within ~6 months post-vaccination was “common.” Efficacy was not reported as mOS/mPFS/ORR, but at data cut 7/21 newly diagnosed patients were alive ~9–30 months post-surgery and 2/4 recurrent patients were alive ~16–20 months post–second resection; immunologically, post-vaccine PBMC showed CD4+ TCM expansion in 17/20 (mean +28.6%, p=0.0006), CD8+ TCM expansion in 15/20 (mean +26.8%, p=0.001), and higher CD127 on CD8+ T cells in 18/20 (mean MFI +72.2%, p=0.03). In context, contemporary SOC for newly diagnosed GBM remains maximal resection + RT/TMZ with historical mOS ~14.6 months in the Stupp regimen (and real-world MGMT-unmethylated outcomes often ~12 months), while recurrent GBM systemic options typically yield mOS around ~10 months in randomized datasets—so the survival “alive at timepoint” signals here are hard to benchmark without Kaplan–Meier endpoints and matched controls.
Take Home: If the ongoing randomized Phase 2 (NCT06805305) can turn these immune-pharmacodynamic signals (and the pseudo-progression pattern) into a clear OS/PFS win over RT/TMZ—especially in the predominantly MGMT-unmethylated population with limited TMZ benefit—it would be a meaningful step for a setting where “better than ~12 months” is still frustratingly uncommon.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2076
CART‑EGFR‑IL13Rα2 (EGFR/IL13Rα2 bivalent CAR‑T) shows 12.0‑mo mOS with no late safety signals in recurrent GBM
Eighteen patients with recurrent EGFR-amplified GBM received a single intracerebroventricular dose of CART‑EGFR‑IL13Rα2, an autologous bivalent CAR‑T designed to recognize EGFR (806 epitope) and IL13Rα2 (Phase 1, 3+3 dose escalation; data cut-off January 16, 2026; NCT05168423). With median follow-up 18.5 months, median OS was 12.0 months (95% CI 7.4–23.2) and 3/18 patients lived >18 months—numerically above typical post-recurrence medians reported with lomustine±bevacizumab (8.6–9.1 months) and regorafenib (7.4–7.6 months) in largely first-relapse settings, though cross-trial comparisons are confounded by eligibility/line-of-therapy differences. Safety remained consistent with earlier reporting: aside from one prolonged Grade 1 neurotoxicity, there were no prolonged/late-onset/unexpected toxicities, no on-target off-tumor signal, and no secondary cancers; neurologic function (NANO) transiently worsened at day 1 after first infusion (mean +1.9; p=0.09) and returned to baseline by day 4, with no meaningful change at 1 month (+0.41; p=0.998) or 2 months (+0.54; p=0.992), and retreatment at progression (n=7) did not reproduce the day‑1 NANO spike.
Take Home: If durability holds in larger cohorts, an ICV-delivered, dual-antigen CAR‑T that preserves (and quickly restores) neurologic function could be one of the more credible shots at moving survival beyond the ~8–10‑month “glass ceiling” seen with common salvage regimens in recurrent GBM—but the small, single-arm Phase 1 dataset (and wide OS CI) still leaves plenty of room for selection effects and regression-to-the-mean.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2013
Liposomal curcumin (LC) + RT/TMZ hits RP2D 350mg/m² with early survival signal in newly diagnosed IDH-wt GBM
Liposomal curcumin (LC) is an IV nano/liposomal curcumin formulation intended to improve curcumin exposure and modulate inflammatory/oxidative and oncogenic signaling (including kynurenine-pathway biology) alongside standard radiotherapy + temozolomide (RT/TMZ) for newly diagnosed high-grade glioma. In this open-label Phase I/II, TITE-BOIN dose-escalation study (NCT05768919), 25 patients with newly diagnosed WHO grade 4 glioblastoma/molecular glioblastoma, IDH-wildtype (median age 56; median KPS 90; MGMT: 7 methylated, 17 unmethylated, 1 indeterminate) received weekly LC (planned 34 doses) with standard RT (60Gy in 30 fractions) and concomitant TMZ, then adjuvant TMZ; two DLTs at 400mg/m² (grade 4 pancytopenia; grade 1 hemolysis) led to selection of 350mg/m² as RP2D, with LC-related grade ≥3 AEs in 16% and no LC-related deaths. With median follow-up 12.9 months (data cutoff 1/23/2026), median PFS/OS were not reached and 6- and 12-month overall survival rates were 95% and 87% among uncensored patients (n=18 and n=13), which looks directionally higher than historical Stupp-regimen benchmarks (median OS ~14.6 months; median PFS ~6.9 months) and is particularly notable given the high proportion of MGMT-unmethylated tumors, where outcomes are typically worse (often ~11–12 months median OS in large real-world datasets).
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2069
Lucicebtide/ST101 (C/EBPβ Antagonist) Shows Early Survival Signal in GBM
Lucicebtide (ST101) is a first-in-class peptide antagonist of C/EBPβ designed to suppress the mesenchymal GBM program and reprogram immunosuppressive tumor-associated myeloid cells. In a Phase 2 window-of-opportunity study (NCT04478279), newly diagnosed GBM patients received perioperative lucicebtide then continued lucicebtide with standard chemoradiation (n=9), while recurrent GBM patients received perioperative lucicebtide then monotherapy to progression (n=9). Safety was described as well tolerated without detailed grade ≥3 rates; pharmacodynamic data showed BBB penetration and target engagement, while efficacy signals included ndGBM 18- and 24-month OS of 67% and 44% versus historical Stupp-style 18- and 24-month OS around 39–45% and 26–31%, and rGBM mPFS 4.0 months/mOS 12.3 months versus common salvage benchmarks often near mPFS 2–4 months and mOS 8–9 months.
Take Home: The biology is compelling, especially if C/EBPβ targeting can shift the GBM microenvironment, but n=9 cohorts and historical comparisons mean this still needs a randomized test.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2079
GammaTile Cs-131 Brachytherapy + Stupp Cuts Rapid Early Progression to 6% in GBM
GammaTile is a cesium-131 collagen-tile brachytherapy platform implanted at resection to deliver immediate local radiation, then followed by abbreviated EBRT with concurrent/adjuvant temozolomide in the prospective GESTALT feasibility/safety study for newly diagnosed GBM across 15 US centers (mITT n=67; NCT05342883). Feasibility was met, with 74% starting EBRT+TMZ within 21–35 days post-op (median 29 days), and centrally reviewed rapid early progression was 6%, far below literature suggesting rapid early progression can occur in a large fraction of patients during the postoperative gap. Grade ≥3 TRAEs occurred in 39% (22% surgery-only, 18% radiation-only, 13% surgery+radiation), comparable to the ~38% cited for SOC in NRG BN007; median OS was 16.5 months in MGMT-unmethylated disease and 28.0 months in methylated disease, with median PFS 9.3 months and not reached, respectively.
Take Home: Immediate cavity radiation is an intuitive way to attack the “waiting-to-start-RT” problem, but the key question is whether fewer early progressions translate into better randomized survival.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2011
Ovarian, Fallopian, or Primary Peritoneal Cancer
🏆 SPOTLIGHT
Azenosertib follows rela in a resensitization strategy with weekly taxanes. Can it mature fast enough to grab a piece of the weekly taxane pie with 2 new approvals?
Azenosertib (WEE1 inhibitor) + weekly paclitaxel delivers 39% ORR in platinum-resistant ovarian cancer
Azenosertib (ZN-c3) is an oral, selective WEE1 kinase inhibitor designed to exploit replication stress and impair DNA-damage checkpoint control; in the MUIR Phase 1b open-label dose-escalation trial (NCT04516447), it was combined with weekly paclitaxel 80mg/m² (days 1/8/15 q28d) in platinum-resistant high-grade serous ovarian/fallopian tube/primary peritoneal cancer after 1–4 prior systemic lines (n=46; all previously exposed to paclitaxel). As of December 1, 2025, confirmed ORR was 39.1% (95% CI 25.1–54.6) with median DOR 5.6 months (95% CI 5.6–9.2), which sits above the ~25–30% ORR typically reported with weekly paclitaxel monotherapy in PROC and ~29–30% in a Phase III weekly paclitaxel control arm, while DOR appears broadly in the range seen with paclitaxel-based approaches. Safety was mainly myelosuppression and fatigue: any-grade fatigue 61%, anemia 59%, nausea 52%, neutropenia 50%; grade ≥3 neutropenia 30% and anemia 20%, with serious TRAEs in 20%—important given PROC patients often need “dose-density” therapy to stay on treatment.
Take Home: If this ORR holds up in a randomized setting, adding WEE1 inhibition could be a practical way to “re-sensitize” taxane-exposed PROC patients beyond what weekly paclitaxel alone typically delivers, but the bar is a controlled PFS/OS benefit (and manageable cytopenias) rather than single-arm response rate inflation.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5529
🏆 SPOTLIGHT
Relacorilant just delivered one of the few positive overall survival trials ever reported in platinum-resistant ovarian cancer and is now approved!
Relacorilant (glucocorticoid receptor antagonist) + nab-paclitaxel extends OS by 4.1 months in platinum-resistant ovarian cancer
Relacorilant is a first-in-class, selective glucocorticoid receptor (GR) antagonist intended to counter GR-mediated chemotherapy resistance, and was combined with weekly nab-paclitaxel in the Phase 3 ROSELLA trial (GOG-3073; NCT05257408) in platinum-resistant ovarian cancer (n=381) versus nab-paclitaxel alone (dose-reduced in the combo arm: 80 mg/m² vs 100 mg/m²). At final OS analysis (288 deaths; 76% maturity; median follow-up 24.8 months), the combination improved OS to 16.0 vs 11.9 months (HR 0.65, 95% CI 0.51–0.83; p=0.0004); as a benchmark, historical nab-paclitaxel monotherapy in platinum-resistant disease has shown ORR ~23% with median PFS ~4.5 months and OS ~17.4 months in a GOG Phase 2 dataset (cross-trial comparisons apply). Taxane pretreatment was essentially universal (99.5%), and the OS benefit was directionally consistent across “taxane-free interval” (≤6 months: HR 0.60; >6 months: HR 0.66) and whether a taxane was part of the most recent regimen (taxane last: HR 0.67; no taxane last: HR 0.63); safety was stated as similar to nab-paclitaxel alone, but Grade 3+ AE details were not provided in this abstract.
Take Home: If the tolerability truly holds up with full AE reporting, a statistically significant 4.1-month OS gain on top of weekly taxane—still a workhorse in platinum-resistant ovarian cancer—would be a meaningful, practice-relevant advance, but the subgroup splits here are small and not powered, so the “prior taxane” story should stay descriptive rather than definitive.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5503
🏆 SPOTLIGHT
TUB-040 delivers a confirmed response rate and tolerability in heavily pretreated platinum-resistant ovarian cancer that immediately says it’s a winner.
TUB-040 (NaPi2b Exatecan ADC) Delivers 61% cORR in Platinum-Resistant Ovarian Cancer
TUB-040 is a NaPi2b-targeted antibody–drug conjugate with a DAR8 exatecan topoisomerase I payload via a cleavable P5 linker, evaluated as every-3-week monotherapy in biomarker-unselected platinum-resistant ovarian cancer in NAPISTAR 1-01 (Phase 1/2a; NCT06303505). As of December 1, 2025, efficacy-evaluable patients treated at 1.67–3.3 mg/kg (n=46; median 4 prior lines; 76% prior PARPi; 84% prior bevacizumab; 13% prior mirvetuximab) achieved uORR 63%, cORR 60.9% including 2 cCRs, DCR 96%, and CA-125 response in 81%, with 90% of responses ongoing at cutoff. Safety was mainly GI and cytopenias, including nausea 78%, fatigue 54%, grade ≥3 neutropenia 26%, grade ≥3 anemia 13%, no fatal TEAEs, no TEAE-related discontinuations, and grade 1 pneumonitis in 6.5%; the response rate is well above typical single-agent chemo ORRs around 10–15% and above mirvetuximab’s 32.4% ORR in FRα-high PROC.
Take Home: A biomarker-unselected 61% confirmed ORR in PROC is hard to ignore, but dose optimization, follow-up, and pneumonitis monitoring will decide whether this becomes a real ADC contender.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5513
SBS5 (clock-like) mutational signature flags longer PFS after cisplatin HIPEC in EOC
In a single-center WES biomarker analysis of patients with primary or recurrent epithelial ovarian cancer (EOC) treated with cisplatin HIPEC during optimal cytoreductive surgery (cisplatin 75mg/m², 60 min, 42°C; NCT01970722), SBS5 (unknown etiology; “clock-like”) was enriched in “good responders” defined by PFS ≥18 months (primary) or ≥12 months (recurrent) vs “poor responders” (n=36 tumors; 21 primary/15 recurrent; p=0.017). Clinically, good responders had longer OS vs poor responders (NR vs NR; p=0.02), but this is a retrospective, threshold-defined endpoint rather than randomized HIPEC vs no-HIPEC predictive evidence; in first-line interval-debulking HIPEC trials, median PFS benchmarks have typically been in the mid–teens months range (e.g., ~14 months in the van Driel setting), so the “good responder” cutoffs are directionally consistent with clinically meaningful separation rather than a marginal tweak. Recurrence pattern signal: peritoneal-only recurrence (64%) showed higher SBS15 (MMR-associated) burden than extraperitoneal recurrence (36%) (p=0.048), while median PFS was similar by recurrence site (15.3 vs 14.3 months), suggesting SBS15 may be more about metastatic pattern than overall disease control in this dataset.
Take Home: If validated, SBS5/SBS15 could become practical “who benefits/where will it fail” genomic context for HIPEC—but right now the study is small, non-randomized, and uses post-hoc PFS cutoffs, so it’s best viewed as a hypothesis-generating biomarker signal rather than a patient-selection tool.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e17577
Delayed interval debulking (6-cycle NAC) shows no DFS gain vs standard 3-cycle timing in CHRONO ovarian cancer
CHRONO (NCT03579394) tested whether delaying complete cytoreductive surgery after neoadjuvant platinum-taxane chemotherapy improves outcomes in FIGO stage IIIB–IVA high-grade epithelial ovarian cancer that was deemed amenable to complete surgery after 3 NAC cycles: standard interval debulking after 3 cycles plus adjuvant chemo (control; n=103) vs 3 additional NAC cycles then surgery plus fewer post-op cycles (experimental; n=106), with maintenance per local SOC. With 40.4 months’ median follow-up, median DFS was 20.2 vs 23.4 months (HR 0.88, p=0.48) and CC0 (complete macroscopic resection) rates were high in both arms (83.2% vs 90%); major 30-day post-op complications numerically increased with delay (5% vs 11%; p=0.11) with no 30-day deaths and no meaningful QoL differences. In context, the long-standing SOC for patients not suitable for complete primary debulking is 3-cycle NAC → interval debulking → chemo (supported by EORTC 55971/CHORUS showing non-inferior survival vs primary debulking), and CHRONO suggests that extending NAC to 6 cycles in “chemosensitive, resectable-at-3-cycles” patients does not clearly improve DFS over that benchmark.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5505
Mirvetuximab soravtansine (FRα ADC) + carboplatin fails to improve PFS vs standard platinum doublets in FRα-high recurrent ovarian cancer
Mirvetuximab soravtansine (MIRV) is an FRα-targeted antibody–drug conjugate delivering the maytansinoid payload DM4; in MIROVA/AGO-OVAR 2.34 (Phase II; NCT04274426), platinum-eligible recurrent ovarian cancer with centrally confirmed FRα-high (≥75% PS2+) and platinum-free interval >3 months were randomized to carboplatin AUC5 + MIRV q3w ×6 then MIRV maintenance vs carboplatin + investigator’s choice (paclitaxel/gemcitabine/PLD) ×6 then PARP inhibitor maintenance if applicable. In 145 randomized patients (77.2% prior bevacizumab; 66.9% prior PARPi; 15.2% BRCAmut; 39.15% received PARPi maintenance in the control arm), median PFS was 9.53 months with carbo+MIRV vs 9.79 months with standard therapy (HR 1.00; 95% CI 0.68–1.46; p=0.996), squarely within historical PFS ranges for platinum-based combinations in platinum-sensitive relapse (e.g., carbo-PLD vs carbo-paclitaxel in CALYPSO showed PFS ~11 vs ~9 months, depending on analysis).
Take Home: With a literal HR of 1.00 and slightly shorter median PFS, carbo+MIRV doesn’t look like a better “platinum backbone” than existing carbo doublets in this setting—so the next question is whether any meaningful subgroup (e.g., ultra–FRα-high, prior PARPi-exposed, or chemo-sparing tolerability) can justify further development here.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5506
Sacituzumab govitecan (TROP2 ADC delivering SN-38) posts 35% ORR and 8.0‑mo mPFS in platinum‑resistant ovarian cancer
Sacituzumab govitecan (SG) is a TROP2-directed antibody–drug conjugate that delivers SN-38 (irinotecan’s active metabolite) to TROP2-expressing tumor cells; in this single-institution, open-label Phase II study (NCT06028932), PROC patients received SG 10mg/kg IV on days 1 and 8 q21d. In 20 heavily pretreated patients (median 3 prior lines; range 1–8), investigator-assessed ORR was 35% (7/20) with 0% CR and 40% stable disease; median PFS was 8.0 months (95% CI 3.8–14.8) and median OS was not reached at 9.9 months’ median follow-up—numerically above common single-agent chemo benchmarks in PROC where ORR is often ~10–20% (e.g., PLD ~19.7% and topotecan ~17.0% in recurrent EOC trials) and above the ~3–4‑mo PFS seen with chemo-alone control in AURELIA (mPFS 3.4 mo; adding bevacizumab improved mPFS to 6.7 mo). (pubmed.ncbi.nlm.nih.gov) Safety was consistent with the known SG profile (no new signals): grade 3–4 neutropenia was common (15/20), with other grade 3–4 events including hypokalemia (n=3) and anemia (n=2); low-grade GI toxicity (diarrhea, nausea/vomiting) and alopecia were frequent.
Take Home: If this 35% ORR/8‑month mPFS holds up in larger, multi-center cohorts (and ideally against a chemo±bevacizumab control), SG could carve out a meaningful option for PROC—though the very high grade 3–4 neutropenia rate looks like the practical limiter, and we’ll need clearer discontinuation/G-CSF and hospitalization details to judge real-world tolerability.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5575
ETX-19477 (oral PARG inhibitor) posts 29% PR rate in heavily pretreated BRCA-mut HGSOC
ETX-19477 is an oral, potent, selective poly(ADP-ribose) glycohydrolase (PARG) inhibitor designed to exploit replication-fork DNA repair vulnerabilities (mechanistically distinct from PARP inhibitors). In the first-in-human ERADIC8 Phase 1/2 dose-escalation/expansion study (NCT06395519), 45 patients with advanced solid tumors were treated across 11 dose levels; in BRCA-mutant high-grade serous ovarian cancer (HGSOC) treated at BID dose levels, 2/7 evaluable patients (29%) achieved a partial response and 3/7 (43%) had stable disease despite a median 4 prior lines (range 2–11), including platinum-resistant and prior PARP inhibitor–exposed responders. Safety was largely GI and hematologic (nausea 53%, vomiting 38%, fatigue 24%; Grade 3 neutropenia 16%, one Grade 4 neutropenia 2%); BID dosing achieved >85% PARG inhibition across the dosing interval, while typical single-agent chemo options in platinum-resistant ovarian cancer often land in the low–mid teens for ORR, and biomarker-selected mirvetuximab (FRα-high) is ~32% ORR (so this early BRCA-enriched signal looks “in range,” albeit in a tiny n and without DOR/PFS yet).
Take Home: If ETX-19477 can sustain responses (DOR/PFS) with manageable neutropenia, PARG inhibition could become a real “post-PARP” DNA-repair strategy for BRCA-mut platinum-resistant HGSOC—but with only 7 evaluable ovarian patients so far, the expansion cohort needs to prove this isn’t just small-n noise.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.3109
BI-1808 (TNFR2 IgG1 mAb; Treg-depleting) + pembrolizumab delivers 24% ORR in heavily pretreated ovarian cancer
BI-1808 is an anti-TNFR2 IgG1 monoclonal antibody designed to block TNF-α/TNFR2 signaling and, via FcγR engagement, deplete tumor Tregs and reprogram myeloid cells—mechanistically complementary to PD-1 blockade with pembrolizumab. In the Phase 2a ovarian cancer sub-cohort of 19-BI-1808-01 (NCT04752826), 24 patients were treated with BI-1808 1000 mg Q3W + pembrolizumab 200 mg Q3W; among 17 response-evaluable patients, confirmed ORR was 24% (4 PR) with DCR 65%, including multiple prolonged SD cases ongoing >10 months (activity noted in both high-grade serous and clear cell). Safety was described as generally tolerable with manageable immune-related AEs and no discontinuations; numerically, the ORR exceeds pembrolizumab monotherapy in recurrent ovarian cancer (ORR 8% in KEYNOTE-100), but cross-trial comparisons are limited and mature PFS/OS are not yet reported.
Take Home: If durability holds up (and especially if clear cell signals persist), TNFR2-driven Treg depletion could be one of the more plausible “chemo-free” immunotherapy strategies in ovarian cancer—but with only 17 evaluable patients and no PFS/OS yet, this still reads as an early signal that needs the planned expansion and biomarker-defined enrichment to avoid another IO false start in an immune-cold disease.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2605
YL205 (NaPi2b-targeted topo1 ADC) posts 46.5% ORR in heavily pretreated ovarian cancer
YL205 is a NaPi2b-directed antibody–drug conjugate delivering a topoisomerase I inhibitor payload (TMALIN platform) evaluated as monotherapy in the first-in-human, multicenter, open-label Phase I/II YL205-CN-101-01 trial (NCT06459973) at 1.0–3.0mg/kg IV Q3W in advanced ovarian cancer (data cutoff: Dec 12, 2025). In 48 dosed patients (median 5.5-month follow-up; 37.5% with ≥4 prior lines; prior bevacizumab 89.6%, PARPi 43.8%, prior mirvetuximab 2.1%), no MTD was reached; Grade 3 TRAEs occurred in 27.1% and related SAEs in 14.6%, with common TRAEs neutropenia (60.4%; Grade ≥3 12.5%), anemia (60.4%; Grade ≥3 4.2%), nausea (47.9%), vomiting (33.3%), and thrombocytopenia (25.0%; Grade ≥3 6.3%). Among 43 efficacy-evaluable patients, confirmed ORR was 46.5% (95% CI 31.2–62.3) and DCR 95.3% (95% CI 84.2–99.4), with mPFS immature and 62.5% (15/24) of responders still on treatment—well above typical single-agent chemo benchmarks in platinum-resistant/refractory disease (historically 12–14% ORR with agents like weekly paclitaxel/topotecan/PLD in refractory settings) and numerically higher than FRα-high mirvetuximab monotherapy in PROC (32% ORR, albeit in a biomarker-selected, 1–3 prior-line population). NaPi2b H-score was high (201–300) in most baseline samples (35/47 with tissue), but no response trend by expression level was observed.
Take Home: If durability holds up, a ~46% confirmed ORR in a broadly enrolled, later-line ovarian cancer population would put YL205 in “best-in-class topo1 ADC” territory versus the usual low-teen chemo response rates—though the lack of a NaPi2b expression–response relationship raises questions about the right biomarker strategy and how to enrich for benefit going into expansion/registrational planning.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5551
Prostate Cancer
🏆 SPOTLIGHT
A rare positive Phase 3 immuno-viral-therapy result in localized prostate cancer for CAN-2409.
CAN-2409 (HSV‑tk adenoviral gene therapy) + valacyclovir improves DFS when added to curative EBRT in localized prostate cancer
Aglatimagene besadenovec (CAN-2409) is an intraprostatic, replication-defective adenoviral vector expressing HSV‑thymidine kinase; with valacyclovir it drives tumor cell kill and immunogenic cell death, and was tested here as an “in situ vaccination” add-on to radiotherapy. In this Phase 3, multicentre, randomized, double-blind trial in intermediate/high-risk localized prostate cancer (n=745; 2:1 randomization; optional ADT), CAN-2409+valacyclovir added to EBRT improved disease-free survival vs placebo+valacyclovir (median DFS not reached vs 86.1 months; HR 0.70, 95% CI 0.52–0.94; p=0.016) at median follow-up 50.3 months—an effect size that’s directionally meaningful vs contemporary EBRT±ADT benchmarks where ~10–20% of intermediate-risk patients recur biochemically within 5 years. Grade ≥3 TEAEs were similar (8% vs 7%); the most common grade ≥3 AE was acute kidney injury (2% in each arm), with treatment-related serious AEs 2% in both groups and no treatment-related deaths (NCT01436968).
Take Home: If these DFS gains hold up with longer follow-up and translate into fewer salvage therapies (and their QoL hit), CAN-2409 could become a rare example of a broadly applicable, non–biomarker-selected “immune-RT amplifier” in localized prostate cancer—but DFS definitions/ADT heterogeneity mean we’ll want to see metastasis-free survival, salvage-therapy rates, and subgroup consistency before calling it practice-changing.
🔗 https://pubmed.ncbi.nlm.nih.gov/42225101
🏆 SPOTLIGHT
Has Apalutamide plus ADT finally added to surgery-centered treatment with a win for metastasis-free survival benefit compared to radiation? 6 cycles still causes significant side effects, so we’ll look at the uptake by oncologists.
Apalutamide (androgen receptor inhibitor) + perioperative ADT improves 5-year MFS in high-risk localized/locally advanced prostate cancer undergoing prostatectomy
Apalutamide is an oral androgen receptor pathway inhibitor added to standard androgen-deprivation therapy (ADT) to intensify perioperative systemic control around radical prostatectomy. In the Phase 3 double-blind PROTEUS trial (NCT03767244), men with newly diagnosed high-risk localized or locally advanced prostate cancer were randomized to ADT+apalutamide 240mg daily vs ADT+placebo for 6×28-day cycles pre-op and post-op (n=2109; median follow-up 61.7 months). Dual primaries both hit: pCR/minimal residual disease was 8.9% vs 1.0% (OR 10.17; p<0.001) and 5-year metastasis-free survival was 78.2% vs 73.5% (HR for distant metastasis or death 0.80; p=0.02); event-free survival, time to first subsequent treatment, and time to distant metastasis also favored apalutamide (all p<0.001). Grade 3/4 AEs were higher with apalutamide (39.6% vs 31.0%), driven mainly by rash—so the efficacy delta comes with a real tolerability tax versus perioperative ADT alone.
Take Home: If you believe MFS as a perioperative “north star,” a ~4.7% absolute 5-year gain (HR 0.80) is meaningful in a population where recurrence risk is high—but the modest size of the effect and higher Grade 3/4 toxicity make patient selection (and durability/OS follow-up) the key next questions.
🔗 https://pubmed.ncbi.nlm.nih.gov/42223077
🏆 SPOTLIGHT
Talazoparib delivers a 52% reduction in progression risk in metastatic hormone-sensitive HRD prostate cancer that is the kind of result that can redefine a treatment paradigm.
Talazoparib (PARP inhibitor) + enzalutamide (AR inhibitor) boosts 3-year PFS to 77% in HRR-altered mHSPC
Talazoparib is an oral PARP1/2 inhibitor combined with enzalutamide, a potent androgen-receptor signaling inhibitor, on top of ADT in TALAPRO-3 (Phase 3, double-blind; NCT04821622) for metastatic androgen pathway modulation–sensitive prostate cancer with homologous recombination repair (HRR) gene alterations (n=599). At 3 years, investigator-assessed imaging-based PFS was 77% vs 56% with placebo + enzalutamide (HR 0.48, 95% CI 0.36–0.65; P<0.001), while interim 3-year OS was 78% vs 72% (HR 0.77, 95% CI 0.56–1.04; not yet significant); as a benchmark, enzalutamide + ADT already delivers strong rPFS benefit vs ADT alone in unselected mHSPC (ARCHES rPFS HR 0.39; median rPFS NR vs 19.0 mo), so TALAPRO-3 is effectively asking whether PARP adds incremental disease control on top of an established ARPI backbone in biomarker-selected patients. Toxicity clearly increased: serious AEs 42% vs 32%, grade ≥3 anemia 51% (notable “PARP-class” marrow suppression), and 2 treatment-related deaths—so the efficacy gain comes with substantial hematologic burden versus ARPI + ADT alone.
Take Home: If the OS curve eventually separates, this could cement “ARPI + PARP” as an upfront standard specifically for HRR-altered mHSPC—but with 51% grade ≥3 anemia, patient selection, transfusion/growth-factor logistics, and dose management may matter as much as the hazard ratio.
🔗 https://pubmed.ncbi.nlm.nih.gov/42223064
🏆 SPOTLIGHT
ABBV-969 delivered response rates that suggest dual-antigen ADC targeting may open an entirely new therapeutic lane in late-stage mCRPC.
ABBV-969 (PSMA/STEAP1 dual-targeting Top1i ADC) drives 45% ORR and 67% PSA50 in heavily pretreated mCRPC
ABBV-969 is a first-in-class dual-targeting antibody–drug conjugate (ADC) against PSMA and STEAP1 delivering a topoisomerase-1 inhibitor payload, evaluated as Q3W monotherapy in the Phase 1 FIH dose-escalation study (NCT06318273) in heavily pretreated metastatic castration-resistant prostate cancer (mCRPC) (n=49; median 5 prior lines; 100% prior ARPI, 84% docetaxel, 39% cabazitaxel, 47% prior Lu-PSMA-617). At doses ≥3mg/kg, confirmed PSA50 and PSA90 rates were 67% and 28%, and among RECIST-evaluable patients (n=29) investigator-assessed confirmed ORR was 45% (radiographic PFS pending); for context, post-ARPI/post-taxane PSMA-targeted radioligand Lu-PSMA-617 improved median OS to 15.3 vs 11.3 months (HR 0.62) and is a key SOC option, but ADC cross-trial comparisons are limited without rPFS/OS. Safety was dominated by cytopenias—Grade ≥3 TEAEs in 63% (notably Grade 3 anemia in 24/49), DLTs in 6% (anemia at ≥8mg/kg), TEAE-driven discontinuation in 6%, and 1 treatment-related death from pneumonitis (2%).
Take Home: If the rPFS holds up, a 45% RECIST ORR plus 67% PSA50 in a cohort where nearly half previously received Lu-PSMA-617 hints that dual-antigen PSMA/STEAP1 ADCs could become a meaningful post-radioligand/post-taxane option—but anemia (and rare pneumonitis) may be the dose-limiting tax to pay, so the RP2D choice will matter.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5014
COMPPARE Proton Therapy vs IMRT (photon) Shows No 2-Year Bowel QOL Advantage in Localized Prostate Cancer
COMPPARE (NCT03561220) is a large prospective, non-randomized comparative effectiveness study testing proton therapy (PT; particle RT with Bragg-peak dose deposition) versus photon IMRT in treatment-naïve localized prostate cancer (excluding very high-risk), allowing conventional and moderately hypofractionated schedules and discretionary ADT/rectal spacers across 51 sites (analyzed PT n=1,404; IMRT n=939; median follow-up 4.0 years). The 2-year primary endpoint showed no difference in EPIC bowel urgency (p=0.324) or bowel frequency (p=0.514) trajectories with inverse probability of treatment weighting; clinician-reported 2-year CTCAE v5 grade ≥2 GI toxicity was similarly low (5.2% PT vs 5.6% IMRT; HR 0.91, 95% CI 0.65–1.28; p=0.6). Exploratory 3-year freedom from biochemical progression was essentially identical (98.0% PT vs 97.9% IMRT; HR 0.95, 95% CI 0.52–1.71; p=0.90), landing squarely in the high tumor-control range expected with contemporary definitive external-beam RT for favorable/intermediate-risk populations.
Take Home: If the big selling point for PT in routine localized prostate cancer is “meaningfully better bowel outcomes,” COMPPARE’s early readout doesn’t support that—so the differentiators may need to be longer-term endpoints (late toxicity, secondary malignancy) or narrower subgroups where rectal dose reductions translate into patient-noticeable benefit.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.17_suppl.LBA5012
Apalutamide (AR antagonist) + ADT periop boosts pCR/MRD and improves 5-year MFS after prostatectomy in high-risk localized/locally advanced disease
Apalutamide (APA), an androgen receptor inhibitor, was added to androgen deprivation therapy (ADT) perioperatively (6 months neoadjuvant, then 6 months adjuvant) around radical prostatectomy (RP) in the double-blind Phase 3 PROTEUS trial (NCT03767244) in newly diagnosed high-risk localized or locally advanced prostate cancer (n=2109; GS≥8 in 95.8%; cN0/cN1 allowed on conventional imaging; median follow-up 61.7 months). APA+ADT improved both co-primary endpoints vs placebo+ADT: pCR/MRD (≤ypT2 and ≤5 mm) 8.9% vs 1.0% (OR 10.17; p<0.0001) and BICR MFS HR 0.80 (p=0.0169) with 5-year MFS 78.2% vs 73.5% (median NR); key secondaries also favored APA+ADT (EFS HR 0.71; median 57.1 vs 38.4 months; TTST1 HR 0.65; TTDM HR 0.68). Safety moved as expected for intensified AR-pathway blockade: grade 3/4 TEAEs 39.6% vs 31.0% and discontinuations due to TEAEs 7.4% vs 2.7%; importantly, this is a meaningful advance over surgery±(neo)adjuvant conventional ADT strategies that historically improved pathologic features without clear survival benefit.
Take Home: If this MFS benefit holds up with longer follow-up and integrates cleanly with post-RP salvage/adjuvant radiation practices, perioperative APA+ADT could become the first broadly practice-shifting systemic intensification strategy “built around surgery” in high-risk localized prostate cancer—but the absolute 5-year MFS gain (~4.7%) and higher discontinuation rate mean patient selection and sequencing with RT will matter.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.17_suppl.LBA1
Talazoparib (PARP inhibitor) + enzalutamide (AR inhibitor) halves rPFS risk in HRR-altered mCSPC
Talazoparib is an oral PARP inhibitor that exploits homologous recombination repair (HRR) deficiency, and enzalutamide is an androgen-receptor signaling inhibitor; in the double-blind Phase 3 TALAPRO-3 trial (NCT04821622), men with HRR-altered mCSPC on ADT were randomized 1:1 to talazoparib + enzalutamide vs placebo + enzalutamide (n=599; cutoff Feb 18, 2026; median follow-up ~37.6–37.7 mo). Investigator-assessed rPFS improved with talazoparib + enzalutamide (HR 0.481; 95% CI 0.357–0.647; p<0.0001; median not reached vs 45.8 mo), with a larger effect in BRCA-mutated disease (HR 0.368; n=207; 34.6%) than non-BRCA HRR alterations (HR 0.567; n=392; 65.4%); interim OS numerically favored the combo but was not statistically significant (74 vs 91 deaths; HR 0.767; 95% CI 0.564–1.044). Compared with enzalutamide + ADT as a mCSPC benchmark (ARCHES rPFS HR ~0.39 vs ADT alone), TALAPRO-3 is a different question—whether adding PARP inhibition on top of enzalutamide improves outcomes in an HRR-selected population—at the cost of substantial hematologic toxicity: anemia 71.2% (all-grade), neutropenia 22.1% (all-grade), and talazoparib discontinuation due to TEAEs in 18.7%; “no new safety signals” were reported.
Take Home: If the OS signal matures, TALAPRO-3 could make “PARP + ARPI upfront” a new precision standard for HRR-altered mCSPC (especially BRCA), but the real-world tradeoff will be whether clinicians and patients accept high anemia rates and nearly 1-in-5 PARP discontinuations for an rPFS win when enzalutamide-based intensification already performs strongly in unselected mCSPC.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.17_suppl.LBA5007
MMAI digital pathology AI biomarker predicts abiraterone benefit (CYP17 inhibition) in very high-risk M0 prostate cancer
MMAI is a locked multimodal AI model built from digital pathology plus PSA, T-stage, and age, tested post hoc across two sequential randomized STAMPEDE Phase 3 abiraterone comparisons in clinically very high-risk, non-metastatic prostate cancer (NCT00268476). Among 1,137 patients (LT-ADT n=583; LT-ADT+abiraterone n=554), the MMAI “very high-risk” subgroup (n=268) had a large metastasis-free survival (MFS) gain with abiraterone (HR 0.47; 95% CI 0.31–0.70), with 5-year MFS 62%→81%; in the MMAI “standard high-risk” subgroup (n=869), benefit was limited (HR 0.83; 95% CI 0.63–1.09), with 5-year MFS 82% vs 84% (interaction p=0.02). For context, in the overall STAMPEDE non-metastatic high-risk population, adding abiraterone to SOC improved 6-year MFS from ~69% to ~82% (HR ~0.53), but with higher rates of grade ≥3 hypertension and transaminitis than SOC—so a biomarker that concentrates benefit into a subset could materially improve the risk/benefit and cost equation.
Take Home: If this MMAI interaction holds up prospectively, it’s a practical way to “de-intensify” abiraterone in the ~75% with already-high 5-year MFS on LT-ADT±RT (82–84% here) while prioritizing intensification for the subgroup with the biggest absolute MFS delta (62%→81%), where avoiding metastases is the whole game.
🔗 https://pubmed.ncbi.nlm.nih.gov/42250604
Enzalutamide + ribociclib (CDK4/6) misses PSA50 bar but extends rPFS/OS in RB+ mCRPC
Enzalutamide is an androgen receptor (AR) inhibitor and ribociclib is an oral CDK4/6 inhibitor intended to “re-awaken” RB tumor suppressor function by preventing RB hyperphosphorylation; RiboX (Phase Ib/II; NCT02555189) enrolled taxane‑naïve mCRPC with retained RB expression and established RP2D as ribociclib 600mg QD (D1–21) + enzalutamide 160mg QD (D1–28). In Phase II, PSA50 at 12 weeks was 75% with enzalutamide alone (n=12) vs 82% with the combination at RP2D (n=34), which did not beat the pre-specified PSA50 null (≤78%) despite numerically higher activity; by comparison, in chemo‑naïve mCRPC enzalutamide’s benchmark rPFS in PREVAIL was ~20.0 months (vs placebo). Combination outcomes were numerically longer for median rPFS (27.0 vs 10.9 months; evaluable n=30 vs 12), PSA PFS (14.8 vs 10.1 months; n=34 vs 12), and OS (58.1 vs 31.2 months; n=34 vs 12), but interpretation is constrained by small, modified design and wide CIs; key combination TEAEs included fatigue (59%), nausea (26.9%), neutropenia (26.7%; grade 3 12%), diarrhea (20.9%), plus infrequent arrhythmia (2.9%) and seizure (2.9%).
Take Home: If the rPFS/OS separation holds up in a larger, properly randomized dataset, CDK4/6 inhibition could be a “disease-stabilization” add-on for a biologically selected RB-intact mCRPC subset—but with PSA50 essentially in the enzalutamide-expected range and non-trivial CDK4/6-class toxicity (notably neutropenia), the bar for a confirmatory trial and sharper biomarkers beyond RB IHC is high.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5057
Enzalutamide-era mCRPC: ctDNA AR enhancer/AR copy gains sharpen rPFS risk stratification (Alliance A031201)
A031201 randomized chemotherapy-naïve mCRPC patients to the androgen receptor (AR) pathway inhibitor enzalutamide ± abiraterone acetate/prednisone (n=776 with baseline ctDNA profiled by the AR-ctDETECT assay; NCT01949337), and this analysis built a circulating tumor DNA (ctDNA) clinical-genetic (CG) prognostic model for radiographic PFS (rPFS). AR enhancer gain and AR gain were among the strongest adverse markers (median rPFS 13.7 vs 27.3 months and 13.5 vs 27.0 months with vs without the alteration, respectively); the random survival forest model selected gains (AR, AR enhancer, MYC, CCND1, FOXA1) and losses (PTEN, TP53, LRP1B) plus clinical labs (hemoglobin, PSA, alkaline phosphatase), improving discrimination vs a clinical-only model (mean itAUC 0.73 vs 0.66; NRI 0.30 at 22 months). The model stratified patients into low/intermediate/poor risk with median rPFS 39.2/25.2/12.8 months (HR low vs poor 0.25; intermediate vs poor 0.46); as a benchmark, first-line ARSI-era trials report rPFS on the order of ~20 months for enzalutamide vs placebo in PREVAIL and ~16.5 months for abiraterone in COU‑AA‑302 (investigator assessment), underscoring how much baseline biology can swing outcomes even within “ARSI-treated” mCRPC populations.
Take Home: If externally validated, this kind of ctDNA-first risk model could become the practical way to (1) enrich intensification trials for the ~13‑month poor-risk biology (AR enhancer/MYC/PTEN/TP53 patterns) and (2) spare clearly low-risk patients (median rPFS ~39 months here) from “more drugs up front” strategies that A031201 itself did not show to materially improve outcomes over enzalutamide alone.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5018
ST-DoxPCa (AI “virtual spatial transcriptomics”) identifies an OS-benefiting subgroup for adjuvant docetaxel in high-risk localized prostate cancer
ST-DoxPCa is an AI/computational pathology biomarker built from routine H&E biopsies using a Vision Transformer to infer spot-level expression for a prostate gene panel (distilled to a locked 26-gene signature; includes genes such as PTEN, NKX3-1, ACPP, FASN, TMPRSS2) and aggregate to patient-level risk features. In NRG/RTOG 0521 (Phase 3; high-risk localized prostate cancer treated with RT+long-term ADT ± adjuvant docetaxel; n=350, RT+ADT n=169 vs RT+ADT+DTX n=181), ST-DoxPCa-positive patients had a significant OS benefit from docetaxel (HR=0.38, 95% CI 0.18–0.83; p=0.012), while ST-DoxPCa-negative patients had no OS benefit (HR=1.05, 95% CI 0.67–1.63; p=0.84)—notably aligning with the parent trial’s “modest/limited” unselected benefit signal that attenuated with longer follow-up (i.e., adding docetaxel to RT+ADT has not been a universally durable OS win in the full population).
Take Home: If this locked, biopsy-based signature holds up in additional cohorts, it’s exactly the kind of “treat-more vs treat-less” classifier that could keep docetaxel for the subset that actually lives longer with intensification—while sparing the biomarker-negative majority the toxicity and inconvenience of chemo with no OS upside.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5111
Rucaparib (PARP inhibitor) extends rPFS in BRCA-mutant mCRPC regardless of when prior ARPI was used
Rucaparib is an oral PARP inhibitor evaluated in the Phase 3 TRITON3 trial (NCT02975934) versus physician’s choice of docetaxel or an ARPI (abiraterone/enzalutamide) in BRCA1/2-mutated mCRPC after progression on 1 prior second-generation ARPI (any setting). In this post-hoc subgrouping by where the prior ARPI was given, rucaparib improved median rPFS vs physician’s choice both after prior ARPI in mCSPC (n=68: 13.6 vs 8.2 months; HR 0.60 [95% CI, 0.32–1.15]) and after prior ARPI in mCRPC (n=234: 11.2 vs 5.8 months; HR 0.43 [95% CI, 0.30–0.61]); median OS was similar in both settings (mCSPC: 23.2 vs 21.7 months; HR 0.81; mCRPC: 23.2 vs 21.0 months; HR 0.91) with comparable treatment-related AE rates (89.6% vs 86.1%). In practical SOC terms, this is consistent with the core TRITON3 message that rucaparib’s rPFS advantage in BRCA1/2-mutated, post-ARPI, chemo-naïve mCRPC generally beats ARPI “switch” and is competitive with docetaxel control performance in this trial, while OS remains hard to move in crossover-permitted designs.
Take Home: If you’re sequencing PARP inhibition in BRCA-mutant mCRPC, these data argue the rPFS lift with rucaparib isn’t “timing-fragile” to whether the first ARPI was used in mCSPC vs mCRPC—but the wide CI in the smaller mCSPC subgroup and the flat OS signal mean we’re still mostly learning about delay-of-progression rather than survival.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5054
Xaluritamig (STEAP1×CD3 T-cell engager) + abiraterone posts 58% PSA50 and 43% ORR in chemo‑naïve mCRPC
Xaluritamig (xalu) is a STEAP1×CD3 bispecific T-cell engager combined with abiraterone acetate + predniso(lo)ne in an open-label Phase 1 dose exploration/expansion study (NCT04221542) in taxane-naïve mCRPC with ≤2 prior ARPIs (no prior abiraterone permitted); 39 patients were treated, with an expansion cohort at 1.5 mg Q2W (n=20; 19 PSA-evaluable, 7 RECIST-evaluable). In the expansion cohort, confirmed PSA50/PSA90 were 58%/47% and RECIST ORR was 43%, with median PSA response duration 6.7 months (95% CI 3.8–NE) and median rPFS 10.5 months (95% CI 8.1–NE); by comparison, abiraterone + prednisone in pre-chemo mCRPC historically delivered median rPFS ~16.5 months vs ~8.3 months for prednisone alone (COU‑AA‑302), so this early rPFS signal does not yet exceed what you’d expect from potent ARPI therapy in an earlier-line population. Safety was immune-/on-target heavy: myalgia 92% (G≥3 41%), CRS 64% (G≥3 7.7%), anemia 46% (G≥3 12.8%), with 2 G4 events and no G5 treatment-related AEs; 4 DLTs occurred across dose levels, and the program moved forward with 1.5 mg Q2W (aligned to the monotherapy RP3D) and is now in randomized Phase 3 (XALience; NCT07213674).
Take Home: If Phase 3 confirms that adding a STEAP1×CD3 engager meaningfully improves rPFS/OS beyond what abiraterone-class ARPIs already deliver (while keeping CRS/myalgia manageable), xaluritamig could become a real “chemo-sparing” immunotherapy option in earlier mCRPC—but the current single-arm rPFS (10.5 months) looks more like “activity present” than “SOC-beating” in this setting.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5070
Cabazitaxel/carbo + nivolumab/ipilimumab (PD-1/CTLA-4) hits 78% 6-mo rPFS in metastatic NEPC/AVPC
CHAMP (NCT04709276) is a multicenter, single-arm Phase 2 trial testing cabazitaxel (taxane microtubule inhibitor) + carboplatin (platinum) with dual checkpoint blockade nivolumab (PD-1) and ipilimumab (CTLA-4), followed by maintenance immunotherapy, in metastatic neuroendocrine prostate cancer or aggressive-variant prostate cancer (NEPC/AVPC) on ongoing ADT. In 40 enrolled patients (n=38 efficacy-evaluable; 75% visceral mets; 58% liver; 85% prior ARPI; 70% prior chemo), the primary endpoint was met with 6-mo iRECIST/PCWG3 rPFS 78% (one-sided 90% CI 69–100%) and median rPFS 12 months; ORR was 34% (CR 2.6%, PR 32%), median OS 12 months with 12-mo OS 46%. Grade 3–4 toxicity was substantial (50% G3; 18% G4), with common G3–4 AEs including anemia (n=15), neutropenia (n=5), sepsis (n=5), thrombocytopenia (n=5), colitis (n=5), and febrile neutropenia (n=4), but no treatment-related deaths—numerically longer disease control than historical platinum doublets in AVPC/NEPC where rPFS/PFS is often ~4–6 months and OS commonly ~7–11 months depending on setting and case-mix.
Take Home: If a randomized study confirms a ~12-month rPFS signal in this heavily pretreated, visceral-metastasis–rich NEPC/AVPC population (where platinum regimens typically buy only a few months), chemo + dual ICI could become a real “durability play,” but the sepsis/colitis/myelosuppression burden looks non-trivial and will need careful patient selection and supportive-care optimization.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5016
Enzalutamide + PARP inhibitor rucaparib adds PSA drops—but no clear rPFS/OS win in all-comer 1L mCRPC (CASPAR)
Alliance A031902 (CASPAR; NCT04455750) tested enzalutamide (androgen receptor inhibitor) plus rucaparib (PARP inhibitor) vs enzalutamide + placebo as first-line therapy for metastatic castration-resistant prostate cancer regardless of HRR mutation status, but the Phase III trial stopped early (10/2023) after enrolling just 61 patients (Enz/Ruca n=30; Enz/Plac n=31; ~50% HRR status unknown). PSA declines favored Enz/Ruca (PSA≥50%: 79% vs 45%; PSA≥90%: 60% vs 43%), yet time-to-event outcomes were statistically inconclusive: median rPFS 17.1 vs 11.7 months (HR 0.72, p=0.44), median PFS 17.1 vs 14.2 months (HR 1.13, p=0.81), and OS not reached vs 27.0 months (HR 0.65, p=0.29)—directionally within (not clearly above) what you’d expect from effective AR-pathway therapy alone in chemo-naïve mCRPC (e.g., enzalutamide in PREVAIL had median rPFS ~20 months vs placebo). Grade 3+ anemia was substantially higher with Enz/Ruca (33% vs 3%), with more AE-driven discontinuations (11.5% vs 0%), consistent with the known hematologic tax of PARP/ARPI combinations.
Take Home: If PARP–AR co-targeting is going to earn a broad “all-comers” 1L mCRPC role, it likely needs a convincingly significant rPFS/OS delta (and/or biomarker-selected enrichment) to justify the added grade 3+ anemia burden seen here—this dataset is simply too underpowered to settle that question.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5069
Fuzuloparib (PARP inhibitor) + abiraterone extends rPFS in 1L mCRPC, with biggest lift in DRD+
FUZUPRO (NCT04691804) is an international, randomized, double-blind Phase 3 trial testing fuzuloparib (oral PARP inhibitor) 150mg BID plus abiraterone acetate/prednisone (AA-P) vs placebo + AA-P as first-line therapy for mCRPC (n=496). At the prespecified interim analysis (data cut March 23, 2025; median follow-up 33.3 months), BICR-assessed median rPFS improved to 24.8 vs 19.9 months (HR 0.71, 95% CI 0.55–0.91; 1-sided p=0.0034); in DRD+ disease (n=116) rPFS was 27.7 vs 13.9 months (HR 0.51), while DRD-negative/unknown (n=380) showed a smaller separation (22.8 vs 21.2 months; HR 0.81). OS was immature/neutral at this look (median 41.9 vs 36.8 months; HR 0.96), and the main added toxicities were hematologic—grade ≥3 anemia 20.1% plus grade ≥3 leukopenia 5.6% and thrombocytopenia/neutropenia 5.2% each (TRAEs 81.9% vs 76.0%); the efficacy magnitude looks directionally similar to the established PARP+abiraterone class signal (e.g., PROpel olaparib+abiraterone HR ~0.66 in an all-comer population), while the clearest “why treat” subgroup again appears to be HRR/DRD-enriched disease rather than biomarker-negative patients.
Take Home: If the final analysis holds up, fuzuloparib+AA-P looks like another credible “PARP + abiraterone” entrant for 1L mCRPC—clinically meaningful in DRD+ patients (rPFS roughly doubling vs AA-P), but with a biomarker-negative signal that may be too modest to justify routine added hematologic toxicity without sharper patient-selection and clearer OS separation.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5008
A-DREAM (Alliance A032101): Intermittent ADT+ARPI interruption delivers 41% 18‑mo treatment-free eugonadal recovery in mHSPC
A-DREAM is a single-arm Phase 2 de-escalation strategy testing interruption of testosterone suppression (ADT/TS) plus an androgen receptor pathway inhibitor (ARPI) after an “exceptional response” (PSA<0.2ng/mL on therapy with castrate testosterone) in metastatic hormone-sensitive prostate cancer; eligible patients had already received TS for 540–750 days and ARPI for ≥360 days before stopping both (n=78). At 18 months after treatment interruption, 58% (45/78) remained treatment-free and 67% (52/78) recovered testosterone; the primary endpoint—treatment-free at 18 months with testosterone recovery—was 41% (32/78; 80% CI 33.5–48.9; one-sided p=0.0249), with 35% restarting protocol TS+ARPI by a median follow-up of 21.2 months. There’s no direct SOC comparator because current mHSPC SOC is continuous ADT intensification with an ARPI (±docetaxel), but historically intermittent vs continuous ADT in metastatic HSPC (pre-ARPI era) raised concerns about outcomes in some subgroups (e.g., SWOG S9346 suggested worse OS with intermittent therapy in “minimal disease,” HR 1.23), so this modern ARPI-era “stop after deep response” approach will likely need longer-term PFS/OS plus QoL and subgroup (volume/risk) readouts before it can be viewed as broadly practice-changing.
Take Home: If longer follow-up shows that this 41% eugonadal, treatment-free rate doesn’t trade away disease control—especially in low-volume patients—A-DREAM could become a realistic path to less cumulative ADT/ARPI toxicity for a carefully selected “super-responder” subset, but the field will rightly demand durable time-to-CRPC/OS and PRO gains before embracing interruption as a new default.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5004
Darolutamide (AR antagonist) + ADT drives deep PSA suppression across older/comorbid mHSPC subgroups
Darolutamide (DARO), an androgen receptor inhibitor, was evaluated with androgen deprivation therapy (ADT) vs placebo + ADT in the Phase 3 ARANOTE trial (NCT04736199) in metastatic hormone-sensitive prostate cancer (mHSPC), with these post hoc analyses stratified by age, baseline comorbidity burden, and concomitant medications (DARO n=446; placebo n=223). PSA <0.2ng/mL at any time was consistently more frequent with DARO vs placebo across age (<65: 62% vs 21%; 65–74: 62% vs 21%; ≥75: 64% vs 12%), comorbidities (<5: 66% vs 21%; ≥5: 66% vs 14%), and concomitant meds (<5: 67% vs 22%; ≥5: 62% vs 16%); among those who achieved PSA <0.2ng/mL, PSA progression was 8% with DARO vs 33% with placebo, and median time to PSA progression was 421 vs 256 days. In context, ADT alone is a suboptimal modern approach in mHSPC (current standards typically intensify ADT with an AR pathway inhibitor and/or docetaxel), so these PSA gains vs ADT monotherapy align with the broader shift away from ADT-only backbones; ARANOTE’s primary endpoint also showed improved rPFS with DARO + ADT (HR 0.54).
Take Home: If deep PSA nadirs truly translate into delayed mCRPC and better long-term outcomes, DARO’s consistency in ≥75-year and higher-comorbidity/conmed subgroups could be clinically meaningful—though this is still a post hoc PSA analysis, and we’ll want mature OS and patient-relevant endpoints (symptoms/QoL) before calling it “comorbidity-proof.”
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5103
BAT (Supraphysiologic Testosterone) + Sipuleucel‑T Drives 100% PA2024 ELISPOT Responses in mCRPC
Bipolar androgen therapy (BAT; testosterone cypionate 400mg IM q4w with ongoing ADT) was given for two cycles before sipuleucel‑T (autologous PAP–GM‑CSF–primed cellular immunotherapy) in a Phase II, Simon two‑stage study in progressive mCRPC selected for “sip‑T-appropriate” features (PSA<20ng/mL, no visceral/hepatic mets, no prior mCRPC chemo, minimal symptoms) (n=11 evaluable; NCT06100705). The primary endpoint—PA2024 IFN‑γ ELISPOT positivity—was 11/11 (100%) vs a ~50% historical ELISPOT positivity rate cited for sipuleucel‑T alone; mean ELISPOT was 232 spots (95% CI 147–317), and T‑cell proliferation to PA2024 was also 100% with mean stimulation index 23 (95% CI 17–30). Early clinical signals were limited (RECIST‑evaluable n=2 with 1 confirmed PR; PSA50 2/11), but tolerability looked clean in this small cohort (no Grade ≥3 AEs/SAEs), whereas sipuleucel‑T in IMPACT improved median OS by ~4.1 months (25.8–25.9 vs 21.4–21.7 months; HR ~0.78) with ~6.7% Grade ≥3 events within 1 day of infusion—so this combo’s differentiator would need to be downstream clinical benefit, not just immune assay uplift.
Take Home: If this “immune-priming” effect translates into harder endpoints (time to progression/OS) in a larger, well-controlled dataset, BAT→sipuleucel‑T could be a pragmatic way to make sip‑T work for more patients—right now, 100% ELISPOT in n=11 is intriguing but still a surrogate win looking for a clinical correlate.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e17081
YL201 (B7-H3–targeted ADC) posts 33% ORR and 11.8-mo rPFS in heavily pretreated mCRPC (2.4 mg/kg)
YL201 is an anti-B7H3 antibody–drug conjugate (ADC) evaluated as monotherapy in a Phase 2 study in metastatic castration-resistant prostate cancer after ≥1 androgen receptor axis-targeted therapy (ARAT) and ≤2 chemotherapy lines (NCT06241846); 82 patients were treated Q3W at 2.0 mg/kg (n=34) or 2.4 mg/kg (n=48), with high-risk features including visceral metastases in 43.9% (liver 22.0%) and a median 4 prior therapy lines. Across doses, confirmed PSA50 response was 38.5% and confirmed ORR was 29.5% with median DoR 9.9 months and median rPFS 9.1 months; activity numerically favored 2.4 mg/kg (PSA50 48.9%, ORR 33.3%, rPFS 11.8 months), including in visceral disease at 2.4 mg/kg (PSA50 54.5%, ORR 28.6%, rPFS 11.8 months). Safety was primarily myelosuppression (grade ≥3 TRAEs 43.9%: neutropenia 29.3%, anemia 22.0%, leukopenia 18.3%, thrombocytopenia 9.8%); discontinuation for TRAEs was 1.2% and no ILD/pneumonitis was reported—an important differentiator vs some ADC classes, while efficacy benchmarks in this post-ARAT/post-taxane space include cabazitaxel rPFS ~8.0 months (CARD) and ^177Lu-PSMA-617 + SOC rPFS 8.7 months with ORR 29.8% in PSMA-selected patients (VISION), making YL201’s 2.4 mg/kg rPFS/ORR look at least “in-family” despite cross-trial differences.
Take Home: If these single-arm data hold up in a randomized setting, a B7-H3 ADC could become a practical option for patients who’ve already “cycled through” ARAT ± taxanes—especially notable here is the visceral-metastasis subgroup signal and the low discontinuation rate, but OS is immature and the dose–efficacy/safety trade-off at 2.4 mg/kg will matter.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5015
CONV01-α (Ac-225 PSMA–mAb alpha-radioconjugate) drives 45% PSA50 in Lu-PSMA–pretreated mCRPC
Ac-225 rosopatamab tetraxetan (CONV01-α) is an actinium-225 alpha-emitting radionuclide conjugated to a PSMA-targeting monoclonal antibody, evaluated in the Phase 2 CONVERGE-01 Part 3 dose-escalation/initial expansion study in PSMA PET–positive (VISION criteria) mCRPC previously treated with ≥1 ARPI, 0-1 taxane regimens, and 1-6 cycles of Lu-PSMA (NCT06549465). In 22 treated patients (DL45 n=12; DL55 n=10; 21/22 taxane-exposed, mostly in the CRPC setting), no DLTs were seen; at DL45, thrombocytopenia was mostly grade 1-2 (5/12) and xerostomia was limited to grade 1-2 (6/12), while Grade ≥3 TEAEs across cohorts included lymphopenia 36%, thrombocytopenia 23% (notably 50% at DL55), neutropenia 18%, and anemia 14%. Among PSA-evaluable patients at DL45, 45.4% achieved PSA50 (5/11), leading DL45 (45 kBq/kg; two doses on D1 and D15) to be selected as RP2D—numerically in the ballpark of cabazitaxel PSA50 rates in post-docetaxel mCRPC (~37% in TheraP), but here in a more treatment-exhausted post–Lu-PSMA population where benchmarks are generally lower and cross-trial comparisons are shaky.
Take Home: If durability and symptomatic/radiographic endpoints track with the PSA50 signal, an Ac-225 PSMA antibody approach could become a credible “post–Lu-PSMA” option—but the real differentiation will hinge on confirming manageable marrow toxicity (especially thrombocytopenia/lymphopenia) and demonstrating rPFS/OS benefit in the planned pivotal study.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5011
ST-DoxPCa H&E Biomarker Identifies mHSPC Patients Most Likely to Benefit From Docetaxel
ST-DoxPCa is an AI computational pathology biomarker that infers spatial gene-expression patterns from routine H&E to predict docetaxel benefit in metastatic hormone-sensitive prostate cancer, using a Vision Transformer trained on paired H&E/spatial transcriptomics and distilled to a 26-gene signature including AR, PTEN, FASN, FOXA1, and ACPP. In CHAARTED (ECOG-ACRIN E3805; NCT00309985), the OS model was trained and validated in the ADT-alone arm (n=154; 77/77 split), locked, and applied to the held-out ADT+docetaxel arm (n=129). ST-DoxPCa-positive patients derived OS benefit from adding docetaxel to ADT (HR 0.53, 95% CI 0.31–0.90; p=0.018), while ST-DoxPCa-negative patients did not (HR 1.32, 95% CI 0.74–2.34; p=0.34), refining CHAARTED’s broader overall docetaxel benefit where low-volume disease historically showed little/no benefit.
Take Home: If externally validated, this could help spare some men upfront docetaxel toxicity while identifying those likely to get real survival benefit beyond crude disease-volume categories.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.5111
Head and Neck Cancer
🏆 SPOTLIGHT
Could a nanoparticle radioenhancer turn local radiation into systemic immunotherapy? NBTXR3 combined with SBRT and PD-1 blockade generated response rates far above historical checkpoint inhibitor monotherapy, hinting at a powerful local-to-systemic immune effect, but as other abstracts show, if aPD-1 came first, it might be even better.
NBTXR3 (hafnium oxide radioenhancer) + SBRT reirradiation followed by anti–PD-1 drives 48% ORR in recurrent/metastatic HNSCC
NBTXR3 is an intratumoral hafnium oxide nanoparticle “radioenhancer” designed to locally amplify radiation effects and potentially augment immune priming; in Study 1100 (Phase I dose escalation/expansion; NCT03589339), patients with previously irradiated recurrent (R) or recurrent+metastatic (R+M) HNSCC received NBTXR3 injection (22% or 33% of gross tumor volume), SBRT 35Gy/5 fractions, then nivolumab or pembrolizumab (30 treated; 27 efficacy-evaluable; 16 ICI-naïve and 14 ICI-resistant; 55% HPV-negative). Grade ≥3 events included 13.3% injection-related, 26.7% RT-related, and 20% NBTXR3-related toxicities; notably, carotid injury was not observed, while injection-site pain (10%) and soft tissue necrosis (6.7%) were among the more frequent injection/NBTXR3-related AEs. Across all disease sites (injected + non-injected), ORR was 48.1% (13/27) and DCR 77.8% (21/27); outcomes separated sharply by pattern of disease—R only: ORR 64.7% (11/17), DCR 88.2% vs R+M: ORR 20% (2/10), DCR 60%—and these response rates look well above anti–PD-1 monotherapy benchmarks in platinum-refractory R/M HNSCC (ORR ~13%–16% in CheckMate 141 nivolumab and KEYNOTE-040 pembrolizumab), though cross-trial comparison is confounded by the addition of SBRT and different eligibility.
Take Home: If the R-only cohort signal holds up in a larger, controlled dataset, NBTXR3-enabled SBRT reirradiation could become a meaningful “local-to-systemic” strategy for in-field recurrences—while the much lower ORR in R+M disease suggests this approach may not be enough to overcome widespread, more immunotherapy-resistant biology without additional systemic intensification.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.6034
🏆 SPOTLIGHT
Can immunotherapy finally replace cisplatin in curative-intent HPV-positive head and neck cancer? Pembro’s KEYCHAIN delivered striking improvements in 2-year PFS and OS versus standard chemoradiation.
Pembrolizumab (PD-1) + Definitive RT Improves 2-Year PFS vs Cisplatin CRT in Unfavorable-Risk p16+ HNSCC (KEYCHAIN)
KEYCHAIN was an open-label, randomized Phase II trial across 6 US academic centers testing definitive RT (70 Gy/35 fx) plus concurrent/adjuvant pembrolizumab (anti–PD-1; 200 mg starting 2 weeks pre-RT, then q3w up to 20 cycles) vs standard concurrent cisplatin-RT (100 mg/m² q3w) in newly diagnosed unresected “unfavorable-risk” p16+ HNSCC (mostly oropharynx; modified ITT n=102: pembro n=50, cis n=52; median follow-up 26.5 months). Two-year PFS was 84% vs 70% (HR 0.57; 95% CI 0.25–1.31; 1-sided p=0.09, meeting the protocol’s 1-sided α=0.15), and 2-year OS was 98% vs 85% (HR 0.33; 95% CI 0.09–1.28; 1-sided p=0.048); grade ≥3 treatment-related AEs were 36% vs 46%—not a dramatic de-tox signal, but directionally favorable. Versus definitive cisplatin-RT benchmarks in HPV+/p16+ OPC (commonly ~97% 2-year OS in low-risk populations), KEYCHAIN’s cisplatin arm looks meaningfully worse—consistent with the “unfavorable-risk” enrichment—making the absolute PFS/OS deltas for pembro-RT clinically interesting if they hold up.
Take Home: If a Phase III confirms this PFS/OS separation in a genuinely higher-risk p16+ population (and especially if late swallowing/renal/ototoxicity endpoints improve), pembrolizumab-RT could become a cisplatin-sparing definitive option—but the wide CIs, liberal 1-sided α, and immature long-term toxicity/HPV-risk stratification mean this still needs a “real” registration-grade win.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.6016
🏆 SPOTLIGHT
Tosposertib has a small study, but one of the largest efficacy signals in the entire HNSCC dataset.
Tosposertib (TGFβRI/VEGFR2 inhibitor) + pembrolizumab drives 75% ORR in 1L R/M HNSCC
Tosposertib (TU2218) is an oral dual TGFβRI/ALK5 and VEGFR2 inhibitor designed to counter TGF-β–mediated immune exclusion and angiogenesis, here combined with pembrolizumab (anti–PD-1) in an open-label, multicenter, non-randomized Phase IIa trial in PD-(L)1–naïve R/M HNSCC with PD-L1 CPS ≥1 (NCT05784688). As of December 31, 2025, 29 patients enrolled and 26 were efficacy-evaluable: ORR was 57.7% overall, including 2 CRs (7.7%); ORR was 75.0% in 1L (9/12; 1 CR) and 42.9% after ≥1 prior systemic therapy (6/14; 1 CR), with higher ORR in CPS ≥20 vs CPS 1–10 (66.7% vs 52.9%). Grade ≥3 TEAEs included rash (20.7%), mucosal inflammation (13.8%), and AST/ALT elevations (each 3.4%); 3 patients discontinued due to TEAEs and there were no treatment-related deaths—efficacy benchmarks in this setting for pembrolizumab-based 1L therapy are typically far lower for pembro monotherapy (ORR ~15–25% depending on CPS), while pembro + platinum/5-FU historically lands around ~35–40% ORR (KEYNOTE-048), so a 75% ORR signal (even at n=12) is well above what we’d expect with SOC.
Take Home: If this early 1L ORR signal holds with longer follow-up (DoR/PFS/OS not yet reported), dual TGFβ/VEGFR modulation may be one of the more credible “IO-boosting” approaches in HNSCC—but the small, non-randomized dataset and short median follow-up (6.0 months) make durability and true differentiation vs pembrolizumab + chemo the key questions.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.2635
🏆 SPOTLIGHT
HPV-negative HNSCC remains one of immunotherapy’s toughest battlegrounds—yet ficerafusp alfa may be breaking through by combining dual EGFR targeting and TGF-β trapping with PD-1 blockade.
Ficerafusp alfa (EGFR×TGF-β trap) + pembrolizumab drives deeper shrinkage and ~10-mo mPFS in 1L HPV− R/M HNSCC
Ficerafusp alfa (BCA101) is a bifunctional EGFR-directed antibody engineered to trap TGF-β, combined here with pembrolizumab (anti–PD-1) in two Phase 1/1b first-line dose-expansion cohorts for PD-L1 CPS≥1, HPV-negative recurrent/metastatic HNSCC (NCT04429542). Among efficacy-evaluable patients (750mg n=30; 1500mg n=28), confirmed ORR was similar (57% vs 54%) but the 1500mg cohort showed more “deep” responses (≥80% shrinkage in 80% vs 47% of responders), longer median DOR (21.7mo vs NR), and longer mPFS (9.9mo vs 6.9mo), with mOS 21.3mo (NR at 750mg); paired biopsies/blood also showed dose-related pharmacodynamic shifts consistent with stronger TGF-β pathway inhibition and immune activation (e.g., pSMAD2 −33.2% vs −16.4%). Versus pembrolizumab monotherapy in KEYNOTE-048 for CPS≥1 disease (ORR ~19% and mOS 12.3mo) and pembrolizumab+chemo (mOS benefit vs EXTREME in CPS≥1), this single-arm Phase 1/1b dataset looks efficacy-competitive on response depth/durability—though cross-trial comparisons and selection effects are major caveats.
Take Home: If the durability signal (median DOR 21.7 months and mPFS ~10 months) holds up in randomized testing, “TGF-β trapping” could be a real way to make PD-1 work better in HPV− HNSCC—but the near-identical ORR across doses and small cohorts make FORTIFI-HN01 (NCT06788990) the place where this either becomes practice-changing or fades as early-phase noise.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.6040
Pembrolizumab (PD-1) + definitive CRT improves 5-year EFS in PD-L1 CPS≥10 unresected LA HNSCC
Pembrolizumab is a PD-1 checkpoint inhibitor added concurrently to definitive cisplatin-based chemoradiotherapy (CRT) and continued as maintenance (17 total q3w cycles) in the randomized, double-blind Phase 3 KEYNOTE-412 trial (NCT03040999); this post hoc subgroup focuses on PD-L1 CPS≥10 unresected, high-risk locally advanced HNSCC. In the CPS≥10 population (n=382; pembrolizumab n=194 vs placebo n=188) at the August 21, 2024 cutoff with 74.1 months median follow-up, median EFS was NR vs 61.4 months (HR 0.71, 95% CI 0.53-0.97) with 60-month EFS 62.4% vs 50.3%; OS was NR in both arms with 60-month OS 70.7% vs 60.9% (HR 0.75, 95% CI 0.53-1.04). Against current SOC (definitive cisplatin-RT), these 5-year survival/EFS rates sit in the expected long-term outcome range for CRT in mixed-risk populations, but the absolute ~12% higher 5-year EFS and ~10% higher 5-year OS here suggest enrichment by PD-L1 CPS≥10 may identify a subset where adding PD-1 meaningfully shifts the curve despite the overall KEYNOTE-412 being statistically negative at its prespecified threshold.
Take Home: If CPS≥10 truly enriches for benefit, pembrolizumab+CRT could be a “biomarker-rescued” strategy in unresected LA HNSCC—but because this is post hoc and OS HR crosses 1.0, the field still needs a prospectively powered PD-L1–selected trial (or a confirmatory signal elsewhere) before treating this as practice-changing.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.6063
ASP-1929 (EGFR–IR700 photoimmunotherapy) lifts OS to 15.7mo vs 9.6mo in heavily pretreated recurrent HNSCC, but underpowered
ASP-1929 photoimmunotherapy (PIT) is an anti‑EGFR monoclonal antibody–IR700 dye conjugate activated by 690nm light to cause localized tumor cell kill (and potentially immune priming); it was tested in a global Phase 3, randomized (2:1), open-label trial in locoregionally recurrent HNSCC after ≥2 prior lines (planned n=275; enrolled n=135; data cutoff April 30, 2025; NCT03769506). Efficacy numerically favored PIT with median OS 15.7 vs 9.6 months (HR 0.83; 95% CI 0.50–1.36) and ORR 25.8% vs 15.2% (DCR 68.5% vs 43.5%), while PFS was similar (HR 0.91; 95% CI 0.24–3.45); however, enrollment stopped early for recruitment/SOC shifts, so estimates are imprecise and cross-trial comparisons to current ICI-era sequences are shaky. Grade ≥3 TEAEs were higher with PIT (58.3% vs 36.4%), though fewer TEAEs drove dose modification/delay/interruption (15.5% vs 30.3%); as a benchmark, “physician’s choice” chemo arms in the pre-ICI era typically delivered ~6.9–8.4 month median OS in platinum-refractory R/M HNSCC trials, highlighting that the control OS here (9.6 months) is not unusually low for a late-line population but may not reflect today’s post–PD-1 treatment mix.
Take Home: If ASP-1929’s OS signal holds up in ASP-1929-381, PIT could become a niche but meaningful “local control + survival” option for patients with function-threatening locoregional recurrences who’ve already exhausted multiple systemic lines—though the early stop and wide CIs mean we should treat this as hypothesis-generating rather than practice-changing.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.6080
VV1 (VSV‑IFNβ‑NIS) intratumoral oncolytic virus + PD‑1 blockade (cemiplimab) hits 30% ORR in 1L R/M HNSCC (Simon’s stage 1)
VV1 is an intratumoral oncolytic vesicular stomatitis virus engineered to express IFNβ and NIS (VSV‑IFNβ‑NIS), combined here with the anti–PD‑1 antibody cemiplimab in a Phase 2, Simon two-stage, first-line chemotherapy-free regimen for PD‑L1 CPS≥1 recurrent/metastatic HNSCC (NCT04291105). In stage 1, 10 treated patients had 3 confirmed PRs (ORR 30%) with PR durations of 6.2 and 9.9 months before progression and 1 ongoing PR at 7.7 months; 8/10 had tumor shrinkage (mean −16%; range −57% to +77%). Safety included 4/10 patients with any treatment-related grade ≥3 AE (notably lymphopenia in 3/10; elevated ALK in 1/10) and 1 case of grade 2 immune nephritis responsive to steroids—efficacy looks directionally above the ~15–20% ORR typically seen with first-line PD‑1 monotherapy in PD‑L1–positive disease in KEYNOTE‑048, while grade ≥3 AEs may be higher than PD‑1 monotherapy expectations, though cross-trial comparisons are fragile.
Take Home: If the ~30% ORR and multi-month PR durability hold up in stage 2, IT VV1 + cemiplimab could carve out a chemo-sparing option for injectable 1L R/M HNSCC—potentially improving on PD‑1 monotherapy’s modest response rate, but the regimen will need clearer durability/benefit vs pembrolizumab-based standards and a better read on whether grade ≥3 cytopenias are an oncolytic-virus “cost of doing business.”
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e18026
Metronomic carboplatin/paclitaxel + PD-1 blocker cemiplimab delivers 42.5% ORR (15% CR) in 1L R/M HNSCC, with cytokine shifts tied to outcome
Cemiplimab is an anti–PD-1 monoclonal antibody combined here with metronomic weekly carboplatin (AUC1) and paclitaxel (25 mg/m²) to potentially retain chemo–immune priming with less toxicity than standard-dose regimens. In a single-arm Phase II trial (NCT04862650) in first-line recurrent/metastatic HNSCC, 40 evaluable patients (35% HPV-positive) achieved ORR 42.5% with 15% CR and median OS 14.8 months (95% CI, 10.3–23.9) at 10-month median follow-up; as a benchmark, KEYNOTE-048 reported median OS 13.0 months with pembrolizumab + platinum/5-FU in the total population (and ~10.7 months with cetuximab + platinum/5-FU), with best results in PD-L1 CPS≥20 (14.7 months). Cytokine profiling (baseline, week 3, week 6) linked higher IFN-β to response (P=.008–.04), baseline IL-4/IL-5/IL-10 to improved ORR, and early rises in IFN-γ and IL-4Rα (baseline→week 3) to worse OS (HR 3.23, P=.006; HR 3.06, P=.008); notably, safety/tolerability outcomes were not reported in this abstract, which is a central question given the “metronomic = better tolerated” premise.
Take Home: If the tolerability really is improved versus standard pembrolizumab + platinum/5-FU (and ideally without giving up OS), this “metronomic chemo + PD-1” template could be a practical way to keep sick, symptomatic R/M HNSCC patients on effective 1L therapy—but the cytokine signals are hypothesis-generating and the single-arm efficacy looks broadly “in-family” with existing pembrolizumab-based standards rather than clearly practice-changing.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.6044
Cemiplimab (PD-1) + cisplatin/docetaxel induction delivers 83% ORR in stage IV LA SCCHN (Phase 1)
Cemiplimab is an anti–PD-1 checkpoint inhibitor combined here with induction cisplatin + docetaxel (TP) ahead of definitive standard-of-care surgery and/or concurrent chemoradiation, followed by adjuvant cemiplimab, in a Phase 1, non-randomized 3+3 feasibility study (NCT05376553) in previously untreated stage IV locally advanced SCCHN (n=24; 15 oropharynx [HPV+ 9], 3 oral cavity, 3 hypopharynx, 2 larynx, 1 nasopharynx). ORR was 83.33% (CR 58.33% [14/24], PR 16.67% [4/24]) with 6/24 progressions; 16/24 (67%) were alive in remission at a median 21-month follow-up, and relapse rate was reported as 18.18%; in the oral cavity subset, 3/5 achieved pathologic CR after resection. No dose-limiting toxicities (DLTs) attributable to cemiplimab were seen in cycle 1; interpretation is constrained by small n, mixed definitive local therapies, and the fact that multiple prior Phase 3 “add PD-1/PD-L1 to CRT” strategies have not clearly improved event-level endpoints versus CRT alone in unselected LA HNSCC populations (e.g., KEYNOTE-412).
Take Home: If this induction + perioperative/peri-CRT PD-1 strategy can reproduce its high CR rate in a randomized setting, it might carve out a role in organ-preservation/intensification niches—but given the historically negative “IO + CRT” Phase 3 experience, the bar should be clear survival/EFS benefit (not just response) and a clean “doesn’t compromise curative local therapy” safety story.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.e18085
Cetuximab (EGFR IgG1) After PD-1 Failure Delivers 33% PR Rate in R/M HNSCC
Cetuximab is an EGFR-targeting IgG1 monoclonal antibody with potential immune-effector activity, evaluated as monotherapy in a non-randomized, single-institution cohort of recurrent/metastatic HNSCC progressing on or intolerant to PD-1 therapy ± chemotherapy (n=27; NCT04375384). Best overall response was 33.3% PR (9/27) and 37.0% SD (10/27), with median PFS 4.4 months and median OS 11.6 months; historical cetuximab monotherapy in platinum-refractory R/M HNSCC typically produced ORR around 10–13%, median PFS/TTP around 2–3 months, and OS around 5–6 months. Safety was notable for 2 first-infusion anaphylaxis events and universal acneiform rash, including 68% grade 3 rash requiring oral antibiotics, which appears higher than older cetuximab monotherapy experience.
Take Home: Cetuximab may have a useful post–PD-1 sequencing niche, but the rash/anaphylaxis burden and single-center design mean this should be confirmed before changing expectations for routine salvage therapy.
🔗 https://ascopubs.org/doi/10.1200/JCO.2026.44.16_suppl.6055
32 New Trials This Week
🆕 — 9 First-in-human
🔄 — 20 Expansion
✅ — 3 Registrational
Prostate Cancer Trials
🔄 Phase 2 ADT + ARPI androgen-receptor signaling inhibition with risk-adapted de-escalation in mHSPC
Georgetown University and the MedStar Health network are running this prospective, interventional, non-randomized Phase 2 pragmatic study (up to 108 patients) to test whether a risk-adapted intensification strategy followed by planned treatment de-escalation can preserve disease control while reducing long-term hormonal side effects in metastatic hormone-sensitive prostate cancer. Core systemic therapy uses androgen deprivation therapy (ADT; suppresses gonadal androgen production) plus an androgen receptor pathway inhibitor (ARPI; blocks androgen receptor signaling), with high-risk patients also receiving docetaxel (a microtubule/tubulin inhibitor chemotherapy) unless contraindicated. Local therapy (prostate radiation and, in some patients, brachytherapy and/or radiation to metastatic sites) is incorporated, then patients who sustain deep PSA response (≤0.2 ng/mL at 36 months) can elect surveillance off ARPI with protocol-defined restart triggers. The key scientific/commercial angle is pragmatic “real-world” de-intensification (intermittent-style therapy) in an era where many patients receive lifelong ADT + ARPI ± docetaxel; today’s care for mHSPC is continuous ADT combined with an ARPI (often plus docetaxel in fit, higher-risk patients) with radiation sometimes used for select low-volume disease, and this study asks whether some responders can safely stop therapy to improve quality-of-life and sexual function.
Washington, DC (MedStar network) | NCT07645326
🔄 Phase 2 MR-guided SBRT uses MR-Linac image guidance (no drug; radiation dose de-escalation).
This Memorial Sloan Kettering Cancer Center–sponsored Phase 2 randomized trial (DESTINATION 2) tests whether intra-prostatic dose de-escalation can reduce urinary/bowel side effects while preserving tumor control in localized prostate adenocarcinoma. Both arms use MR-guided stereotactic body radiotherapy (SBRT) delivered on an MR-Linac in 2 fractions, leveraging daily MRI for alignment/adaptation and tighter margins than conventional CT-guided RT. The control arm treats the prostate more uniformly, while the investigational arm lowers dose to MRI-benign prostate tissue and maintains a high dose to MRI-visible, biopsy-proven tumor focus(es) (PIRADS 3–5 lesions that are boostable and <2.5 cm). Eligible patients include Grade Group 1–3 disease with PSA <20 ng/mL and largely T1–T2 (allowing up to radiologic T3a on MRI), without nodal or distant metastases; short-course ADT may be used when clinically appropriate (generally ≤6 months). Today’s care for Grade Group 1 often includes active surveillance, while many Grade Group 2–3 patients receive surgery or external-beam RT (often moderate hypofractionation or 5-fraction SBRT) ± short-course ADT; this study asks if MR-guided focal boosting plus de-intensifying uninvolved prostate can improve tolerability in an ultra-short 2-treatment regimen.
Multiple New York sites (Middletown, Montvale, Commack, Harrison, Uniondale, Basking Ridge) | NCT07635238
🔄 Phase 2 sacituzumab tirumotecan TROP2-directed ADC (topoisomerase I payload) in post-platinum NEPC.
This Phase 2, open-label, single-arm study tests sacituzumab tirumotecan (a TROP2-targeting monoclonal antibody–drug conjugate carrying a belotecan-derived topoisomerase I inhibitor payload) in metastatic neuroendocrine prostate cancer that has progressed after platinum-based chemotherapy. The purpose is to define anti-tumor activity and safety in an aggressive subtype where responses to available regimens are often short-lived. Eligible patients have metastatic NEPC with progression on prior platinum, aligning the trial with a high-unmet-need, chemo-refractory population. Virginia Commonwealth University sponsors the study at its Richmond site, while the scientific angle is extending the clinical “TROP2-ADC” playbook (validated commercially in other tumors) into NEPC, where targetable drivers are limited and drug delivery to rapidly proliferative disease may matter. Today’s care after platinum typically moves to additional cytotoxic regimens (often borrowing from small-cell paradigms) and supportive care, with few established targeted options for NEPC.
Richmond, VA (single-site, US) | NCT07639086
🔄 Phase 2 61Cu-NU101 PSMA-targeted PET radiotracer (PSMA ligand) vs Pylarify in prostate cancer
This Phase 2, two-center, single-arm diagnostic comparison study tests whether the investigational PET radiotracer 61Cu-noDAGa-PSMA I&T (61Cu-NU101) can match or outperform the current standard 18F-piflufolastat (Pylarify) for detecting metastatic prostate cancer. Participants must have biopsy-proven prostate cancer with at least one lesion already seen on a prior Pylarify PET/CT, creating a “head-to-head within the same patient” imaging comparison. The intervention is a single IV bolus of 61Cu-NU101 (148 MBq / 4 mCi ±10%) followed by PET/CT at ~1 hour and ~4 hours post-injection (optional dynamic imaging around injection). Mechanistically, 61Cu-NU101 is a PSMA-binding small-molecule radiotracer, aiming to leverage copper-61 imaging kinetics to potentially enable higher-contrast or delayed-timepoint lesion detection than 18F agents. Clinically, men with oligometastatic/PSMA-avid disease are typically worked up and managed with PSMA PET–guided staging plus metastasis-directed RT and systemic therapy (and, when appropriate, prostate-directed local therapy), and this trial probes whether a new PSMA PET tracer could refine where disease truly is before treatment decisions. Hoag Memorial Hospital Presbyterian sponsors the study; the NU101 program is associated with Nuclidium’s copper-based PSMA portfolio in the broader field.
Newport Beach, CA and St Louis, MO | NCT07635524
🔄 Phase 2 Microboost SBRT focal SIB (DNA damage via ionizing radiation) with whole-gland de-escalation
Georgetown University’s Phase 2 MIDAS trial tests whether MRI-defined “microboost” SBRT can safely intensify dose to up to 4 intraprostatic lesions (45 Gy in 5 fractions) while de-escalating dose to the remainder of the prostate to reduce toxicity. The intervention is external-beam stereotactic body radiation therapy, whose mechanism is ionizing radiation–induced DNA double-strand breaks leading to tumor cell death, delivered as a simultaneous integrated boost (SIB) to dominant MRI lesions with adaptive-risk whole-gland dosing. It is non-randomized, open-label, parallel-assignment, enrolling an estimated 58 men with NCCN low-risk (Arm 1: whole gland 30 Gy/5) or intermediate-risk disease (Arm 2: whole gland 35 Gy/5), with lesion selection driven by mpMRI (PI-RADS 4+ or selected PI-RADS 3 with GG2-3) and restrictions on higher-risk features (no GG4+, PSA ≤20, no T3a+). A key commercial/scientific angle is focal intensification plus whole-gland de-escalation in a curative setting—trying to preserve the disease-control benefits of lesion boosting while improving urinary/rectal tolerability over conventional full-dose whole-gland RT. Today’s care for low/intermediate-risk localized prostate cancer commonly includes active surveillance (selected low-risk), prostatectomy, or definitive radiation (moderate hypofractionation or 5-fraction SBRT) ± short-course ADT for higher-risk intermediate disease, and this study tests a more personalized intraprostatic dose-painting approach.
Washington, DC (single-site, Georgetown University) | NCT07644598
🆕 Phase 1/2 GDC-1261 trial in prostate cancer; no disclosed MOA/target.
No disclosed MOA. We do not include details of trials without this basic information.
NCT07567846
AML: Acute Myeloid Leukemia Trial
🏆 TRIAL SPOTLIGHT
Moving engineered TCR-T into MRD-positive AML is a smart setting: lower disease burden, clearer immune target, and a potentially meaningful chance to deepen remission before relapse declares itself. HLA-gating limits reach, but the concept is strong.
🆕 Phase 1 FH-WT1-E50 WT1-targeted TCR‑T plus azacitidine for MRD+ AML
This Phase 1 study at Fred Hutchinson Cancer Center evaluates FH‑WT1‑E50, an autologous CD4+/CD8+ T-cell product genetically transduced to express an HLA‑A02:01–restricted, WT1-specific T-cell receptor (TCR‑T; MOA: MHC-presented WT1 peptide recognition leading to T-cell cytotoxic killing of WT1-expressing leukemia cells), given with azacitidine (DNA hypomethylating agent that inhibits DNA methyltransferases and promotes epigenetic reprogramming). Patients first undergo leukapheresis, then receive azacitidine, followed by azacitidine plus FH‑WT1‑E50 infusion; if enough cells are available, a second azacitidine + TCR‑T infusion can occur as early as day 28 and up to 1 year. The target population is minimal residual disease (MRD)–positive AML with WT1 expression and HLA‑A02:01 positivity (HLA‑A*02:01 is common in many populations, roughly ~40–50%, making the approach broadly applicable but still HLA-gated). The key angle is moving engineered TCR‑T into the MRD setting, where tumor burden is lower and immune therapies may have a better chance to deepen remission; today’s care for MRD+ AML commonly involves additional consolidation strategies such as hypomethylating agents (including azacitidine), venetoclax-based therapy in some contexts, and/or consideration of allogeneic stem cell transplant depending on risk, fitness, and prior therapy.
Seattle, WA (single-center, Fred Hutch) | NCT07645469
Hematological Disease Trials
🔄 Phase 2 automated VMAT-based TMLI conditioning refines marrow / lymphoid-targeted irradiation for allo-HCT.
Stanford University is running this non-randomized, open-label Phase 2 study to confirm feasibility and safety of an automated total marrow and lymphoid irradiation (TMLI) platform when embedded into a previously validated reduced-toxicity conditioning approach for allogeneic hematopoietic cell transplant (allo-HCT). Patients receive fludarabine (a purine analog that inhibits DNA synthesis) plus cyclophosphamide (an alkylator that crosslinks DNA) together with VMAT-based TMLI (external-beam radiation conformally targeting marrow and lymphoid tissues), followed by allogeneic peripheral blood stem cell transplantation. GVHD prophylaxis uses post-transplant cyclophosphamide (PTCy; immunomodulatory T-cell depletion), then mycophenolate mofetil (IMPDH inhibitor blocking lymphocyte proliferation) and tacrolimus (calcineurin inhibitor suppressing T-cell activation). Cohort A focuses on first allo-HCT for high-risk myeloid malignancies (AML with adverse biology or persistent blasts; MDS/MPN/MPN with high blasts and/or TP53 mutation, noting TP53 alterations are a clinically important minority but strongly prognostic), while Cohort B enrolls patients relapsing after a prior allo-HCT with a lower-dose TMLI regimen. The key angle is operational: automation aims to make technically demanding TMLI planning/QA faster and more scalable without losing organ-sparing advantages; today’s care for these high-risk AML/MDS/MPN settings is induction/salvage systemic therapy followed by allo-HCT using chemotherapy-based conditioning (with or without broader-field irradiation) and PTCy-based GVHD prophylaxis in many centers.
Palo Alto, CA (Stanford) | NCT07634536
🆕 Phase 1 JNJ-95804306 (no disclosed MOA/target) in relapsed/refractory AML/MDS and CLL/SLL
No disclosed MOA. We do not include details of trials without this basic information.
NCT07572006
Large B-cell Lymphoma (LBCL) Trial
🔄 Phase 1 axatilimab CSF1R-blocking monoclonal antibody conditions macrophages alongside commercial CAR-T.
This open-label Phase 1 study (“MAC-SHIFT”) at Northside Hospital evaluates whether adding axatilimab—a CSF1R-directed monoclonal antibody that inhibits macrophage/monocyte CSF-1R signaling and can reduce proinflammatory, tissue macrophage activity—can improve outcomes around commercially available CAR-T therapy in high-risk large B-cell lymphoma (LBCL). Patients receive axatilimab plus a commercial CAR-T product (the CAR-T component is an autologous chimeric antigen receptor T-cell therapy designed to recognize a B-cell antigen such as CD19, depending on the commercial product selected). The core angle is tumor-microenvironment “macrophage conditioning” to potentially enhance CAR-T expansion, persistence, and depth/durability of response in biologically or clinically high-risk LBCL, where relapse after CAR-T remains a key unmet need. Today’s care for high-risk relapsed/refractory LBCL typically includes salvage chemoimmunotherapy and/or antibody-based options as bridges, followed by CD19 CAR-T in eligible patients, with bispecific antibodies increasingly used in later lines; this trial tests whether targeted CSF1R blockade can make CAR-T work better rather than replacing it.
Atlanta, GA (single-site, Northside Hospital) | NCT07638982
Neuroendocrine Tumors Trial
🔄 Phase 2 AUR103 calcium oral CD47 antagonist tests innate-immunity reactivation in NETs.
This proof-of-concept, open-label Phase 2 study (BHARAT-3) evaluates single-agent AUR103 calcium—an orally bioavailable small-molecule CD47 antagonist that aims to block the “don’t-eat-me” signal and promote macrophage-mediated phagocytosis—in advanced, well or moderately differentiated (G1–G2) neuroendocrine tumors. Aurigene Discovery Technologies Limited is sponsoring a twice-daily dosing strategy with two dose levels (200 mg or 300 mg BID) continued until disease progression or unacceptable toxicity, with a small estimated enrollment (~30). Eligible patients are relapsed/refractory and must have progressed after at least one prior systemic line; the protocol also expects prior use of at least one FDA-approved NET therapy appropriate to tumor context (examples listed include everolimus, sunitinib, cabozantinib, or lutetium Lu-177 dotatate). The commercial/scientific angle is bringing CD47 biology—most validated with antibodies—into an oral small-molecule format, which could improve convenience and combination flexibility if activity is seen. Today, progressive metastatic well-differentiated NET care typically sequences somatostatin analogs, targeted agents (e.g., everolimus; sunitinib for pancreatic NET), peptide receptor radionuclide therapy (Lu-177 dotatate) for somatostatin-receptor–positive disease, liver-directed approaches when appropriate, and chemotherapy for selected higher-burden/aggressive cases—this trial explores an immune-innate strategy after those options.
Jacksonville, FL; Rochester, MN; Phoenix, AZ; Newport Beach, CA; Philadelphia, PA | NCT07640737
Neuroblastoma Trial
🆕 Phase 1 CIML-NK allogeneic NK-cell immunotherapy (IL-12/15/18–primed “memory-like” NK cells)
This first-in-human, single-site Phase 1 study at Cincinnati Children’s tests whether cytokine-induced memory-like natural killer cells (CIML-NK)—an allogeneic NK-cell product generated from a patient’s haploidentical donor apheresis—can be manufactured and infused safely for relapsed/refractory high-risk neuroblastoma during dinutuximab-based anti-GD2 therapy. The CIML-NK approach uses brief priming with IL‑12/IL‑15/IL‑18 to create “memory-like” NK cells with enhanced persistence and cytotoxicity, aiming to intensify antibody-dependent cellular cytotoxicity when paired with dinutuximab (anti‑GD2 mAb; originally developed by NCI and commercialized as Unituxin by United Therapeutics). The study is primarily a feasibility/safety platform (dose and schedule refinement, monitoring for infusion toxicity, cytokine-release–type syndromes, and graft-versus-host–type complications that are uncommon with NK cells but still monitored). It targets patients whose disease has progressed after prior intensive multimodal therapy; today’s care for R/R neuroblastoma is iterative salvage chemotherapy/chemoimmunotherapy (often irinotecan/temozolomide-based regimens) plus anti‑GD2–based combinations when tolerated, with select patients receiving MIBG radiotherapy, ALK inhibitors for ALK-altered tumors, or clinical trials. The scientific angle is a donor-derived, “trained” innate-cell add-on designed to make anti‑GD2 therapy more effective without introducing engineered receptors (e.g., CARs), potentially simplifying manufacturing relative to gene-modified cell therapies.
Cincinnati, OH (single US site) | NCT07635056
Leiomyosarcoma (smooth muscle sarcoma) Trial
🔄 Phase 1b REC-617 selective CDK7 inhibitor (transcription/cell-cycle kinase target) in RB1– LMS
MD Anderson Cancer Center is running this Phase 1b, single-group, open-label study to test whether REC-617—an oral small-molecule inhibitor of CDK7 (a cyclin-dependent kinase that supports transcriptional regulation and cell-cycle progression)—can control metastatic or unresectable leiomyosarcoma that is RB1-negative by IHC (0% staining cutoff). Patients must have had at least one prior systemic therapy line (so a typical 2nd-line+ setting) and have measurable disease by RECIST 1.1; the trial also requires lesions amenable to biopsy (excluding bone) for pharmacodynamic and response workups. The intervention is continuous REC-617 10 mg once daily, with endpoints including objective response rate plus clinical benefit rate, PFS, OS, and CTCAE toxicity. The drug is owned/developed by Recursion Pharmaceuticals, while the trial is sponsored by MD Anderson, reflecting an academic-center expansion into a genomically/biologically defined LMS subset (RB1 loss is common in LMS, but not universal). Today’s care after progression is sequential systemic therapy such as doxorubicin-based regimens up front, then options like gemcitabine/docetaxel, trabectedin, pazopanib, eribulin, or dacarbazine-based therapy, with responses often limited—so this tests whether CDK7 dependence in RB1– disease can be exploited with a targeted oral agent.
Houston, TX (MD Anderson only) | NCT07633756
Ovarian, Fallopian, or Primary Peritoneal Cancer Trials
🏆 TRIAL SPOTLIGHT
Solid-tumor CAR-T still needs better persistence, better trafficking, and better local control. An IL-18–armored MUC16 CAR-T with IV plus intraperitoneal delivery is exactly the kind of ovarian cancer experiment worth following closely.
🆕 Phase 1 GB-5267 IL-18–armored anti-MUC16 CAR T cell therapy in platinum-resistant ovarian cancer.
This first-in-human, open-label Phase 1 dose-escalation study tests GB-5267, an autologous CAR T-cell product engineered to recognize MUC16 (a surface antigen commonly expressed in epithelial ovarian cancers) and “armored” with IL-18 to enhance local immune activation and persistence in solid tumors. The trial’s primary purpose is to define safety/tolerability and a recommended dose, with secondary evaluation of early antitumor activity and cellular kinetics/biologic effects. Participants have platinum-resistant ovarian, fallopian tube, or primary peritoneal cancer (O/F/P), a setting where current care is sequential systemic therapy such as non-platinum chemotherapy (often with bevacizumab when appropriate), PARP inhibitors for selected biomarker-defined patients, and later-line options including antibody-drug conjugates in some jurisdictions—yet responses are frequently short-lived. The protocol includes an IV CAR T approach and a planned cohort exploring combined IV + intraperitoneal (IP) delivery to better address peritoneal-dominant disease, a practical anatomic challenge for cellular therapies in ovarian cancer. The study is sponsored by Roswell Park Comprehensive Cancer Center, with GB-5267 developed in partnership with Generate Biomedicines.
Buffalo, NY (single-site, US) | NCT07489287
🔄 Phase 2 pembrolizumab PD-1 antibody plus TLR3 agonist rintatolimod and autologous αDC1 vaccine
This Phase 2, nonrandomized, sequential-design neoadjuvant study at UPMC Hillman Cancer Center tests whether “locoregional chemoimmunotherapy” can improve immune activation and clinical outcomes in chemo-naive stage III–IV ovarian cancer. Patients receive intraperitoneal (IP) cisplatin (DNA crosslinker) with intravenous paclitaxel (microtubule stabilizer), followed by an IP chemokine-modulatory regimen that includes rintatolimod (TLR3 agonist dsRNA) plus interferon alfa-2b (type I interferon immune cytokine), along with oral celecoxib (COX-2 inhibitor) to shape inflammatory signaling. Systemic pembrolizumab (Keytruda; Merck/MSD) is added to block PD-1 and reduce T-cell exhaustion, and one arm adds intradermal autologous αDC1 dendritic-cell vaccines (a personalized cellular vaccine platform) before interval debulking surgery and subsequent therapy. The key angle is the heavy emphasis on peritoneal, immune-directed “conditioning” plus checkpoint blockade—trying to turn a typically immunologically “cold” peritoneal disease compartment more inflamed and T-cell–permissive. Today’s care for advanced-stage ovarian cancer in the neoadjuvant setting is platinum-taxane chemotherapy followed by interval cytoreductive surgery, then completion chemotherapy and (in many patients) maintenance therapy such as PARP inhibition when BRCA/HRD-positive and/or bevacizumab in selected cases; this trial explores whether adding IP innate stimulation + interferon signaling + PD-1 blockade + a dendritic-cell vaccine can deepen response before surgery.
Pittsburgh, PA (single-site, UPMC Hillman) | NCT07634094
Uterine (Endometrial) Cancer Trials
🔄 Phase 2 elacestrant oral SERD (estrogen receptor degrader) + everolimus mTOR inhibitor in ER+ endometrial cancer
This Phase 2, single-arm interventional study (ELAVERA) at UCLA/Jonsson Comprehensive Cancer Center evaluates elacestrant (an oral selective estrogen receptor degrader that binds ER and promotes receptor degradation) combined with everolimus (an mTORC1 inhibitor that suppresses PI3K/AKT/mTOR signaling and can help reverse endocrine resistance) in recurrent, persistent, advanced, or metastatic ER+ endometrial cancer. Participants take both drugs by mouth once daily continuously on days 1–28 of each 28-day cycle, with imaging (CT ± PET or MRI) and blood collection to track response and safety. The primary endpoint is the 24-week progression-free survival rate, with secondary endpoints including ORR, clinical benefit rate, PFS, and safety. Eligibility includes ER positivity (≥1% by IHC) across histologies; patients must have had prior platinum chemotherapy, and those with dMMR tumors generally must also have received prior immunotherapy. The commercial/scientific angle is endocrine-therapy intensification in a setting where many patients after platinum ± immunotherapy move to options like lenvatinib + pembrolizumab (for pMMR), additional immunotherapy approaches (for dMMR/MSI-H), or cytotoxic chemotherapy, with limited endocrine-based standards despite frequent ER expression in endometrioid tumors.
Los Angeles, CA (single US site so far) | NCT07634601
🔄 Phase 1/2 MR‑guided adaptive IMRT (MR‑linac–based pelvic radiation) after surgery in high‑risk endometrial cancer
Rutgers Cancer Institute of New Jersey (Rutgers University) is running this Phase 1/2 feasibility study to test whether daily MRI‑guided, online adaptive pelvic IMRT can be delivered safely and consistently as adjuvant therapy for higher‑risk endometrial cancer features (including high‑grade disease, pelvic node positivity, and/or cervical stromal invasion). The intervention is MR‑guided adaptive external‑beam radiotherapy on an MR‑linac: onboard MRI each day is used to adapt (replan) the IMRT dose to anatomic changes, aiming to improve target coverage while reducing dose to bowel, bladder, and other organs at risk. The regimen is moderately hypofractionated (42 Gy in 15 fractions), with optional vaginal brachytherapy afterward per institutional practice. The key scientific angle is operational and dosimetric: demonstrating that “daily adaptive” MR‑guided pelvic RT is practical in the post‑op setting where bowel/bladder motion is a major toxicity driver. Today’s care for these post‑hysterectomy higher‑risk patients is adjuvant pelvic EBRT (often CT‑planned IMRT) with selective use of vaginal brachytherapy and systemic chemotherapy depending on stage/risk, and this study probes whether MR‑guided adaptation can maintain disease control while potentially lowering GI/GU toxicity.
New Brunswick, NJ (single‑site, US) | NCT07643012
Pancreatic Ductal Adenocarcinoma (PDAC) Trials
🏆 TRIAL SPOTLIGHT
This is the PDAC KRAS G12D trial to watch: oral targeted therapy, a registration-directed Phase 2 design, and a rational EGFR-combo arm. If VS-7375 works here, it could open a badly needed targeted lane in pancreatic cancer.
🔄 Phase 2 VS-7375 oral KRAS G12D (ON/OFF) inhibitor ± EGFR antibody cetuximab in metastatic PDAC.
Verastem, Inc. is running TARGET-D 201 (NCT07644559), an open-label, multicenter Phase 2 study testing VS-7375—an oral, KRAS G12D–selective “ON/OFF” small-molecule inhibitor—given as monotherapy and in combination with cetuximab, an EGFR-directed monoclonal antibody (marketed by Eli Lilly; originally from ImClone/Bristol Myers Squibb). The purpose is to define safety and preliminary efficacy in metastatic pancreatic ductal adenocarcinoma (PDAC) with a confirmed KRAS G12D mutation, including treatment-naïve/near–treatment-naïve patients in a 1L cohort and a separate combination cohort for previously treated (2L) metastatic disease. Scientifically, the commercial angle is “registration-directed” Phase 2 development in a high-unmet, KRAS G12D–driven tumor where EGFR feedback signaling can limit KRAS-pathway suppression—hence the rationale for adding cetuximab. Today’s care for metastatic PDAC is systemic combination chemotherapy up front (modified FOLFIRINOX or gemcitabine + nab-paclitaxel), with subsequent lines often including nanoliposomal irinotecan/5-FU–based regimens and targeted therapy only for small biomarker-defined subsets (e.g., BRCA/PALB2).
West Valley City, UT (single US site listed so far) | NCT07644559
🔄 Phase 1b samuraciclib trial tests CDK7 inhibitor transcription/cell-cycle blockade in basal PDAC
This single-institution, single-arm, open-label Phase 1b window-of-opportunity study evaluates whether samuraciclib (CT7001)—an oral small-molecule CDK7 inhibitor that disrupts transcriptional CDK activity and cell-cycle control—changes tumor biology and shows early activity in localized “basal” pancreatic ductal adenocarcinoma. Patients with resectable, borderline resectable, or locally advanced disease without metastases receive samuraciclib by mouth once daily for 14 days, positioned before planned chemotherapy and/or surgery, with on-study research EUS/FNB plus CT imaging and blood collection. The biomarker-defined population is basal PDAC (GATA6-negative and HMGA2-positive by IHC, with positivity defined as >10% tumor epithelial nuclei), a subgroup associated with worse outcomes and reduced benefit from gemcitabine-based adjuvant therapy in published cohorts (HMGA2+/GATA6− was the poor-prognosis group). Commercially, the key angle is using a short, pre-treatment “window” to rapidly learn whether CDK7 inhibition meaningfully perturbs basal PDAC programs, potentially enabling biomarker-driven development. Today’s care for localized PDAC is surgery when feasible plus perioperative multi-agent chemotherapy (commonly modified FOLFIRINOX; alternatives include gemcitabine-based regimens), with chemoradiation selectively used for borderline/locally advanced cases.
Seattle, WA (Fred Hutch/University of Washington Cancer Consortium) | NCT07645651
Small-Cell Lung Cancer (SCLC) Trial
🆕 Phase 1/2 DJI136 (LB2102) DLL3-targeted autologous CAR‑T for extensive-stage SCLC.
This Phase 1/2, open-label, non-randomized, multi-center study sponsored by Novartis Pharmaceuticals is testing DJI136 (also described as LB2102), an autologous CAR‑T cell therapy engineered to recognize DLL3 on tumor cells; DLL3 is highly prevalent in SCLC (commonly reported ~80%+), making it a rational surface target for cellular therapy. The Phase 1 portion dose-escalates DJI136 to define the recommended dose(s) and characterize safety/tolerability and early efficacy in relapsed/refractory ES‑SCLC after at least 1 prior chemotherapy regimen (the listing notes prior platinum-doublet chemo plus a PD‑L1 inhibitor per local practice unless not eligible). The Phase 2 portion then treats DLL3-therapy–naïve ES‑SCLC at the recommended dose(s) to better estimate activity and safety. The key angle is bringing a DLL3-directed CAR‑T approach into a disease where today’s front-line care is platinum–etoposide plus a PD‑L1 inhibitor (atezolizumab or durvalumab), and post-progression options are limited (often lurbinectedin, topotecan, or clinical trials), so durable responses from a cell therapy would be commercially and clinically meaningful.
Singapore; and Houston, TX | NCT07564401
Non-Small Cell Lung Cancer (NSCLC) Trials
🏆 TRIAL SPOTLIGHT
KRAS G12D is quickly becoming one of oncology’s hottest targets, and this is a true Phase 3 test in lung cancer. A degrader beating docetaxel would make the “KRAS G12D moment” feel very real outside PDAC.
✅ Phase 3 setidegrasib (ASP3082) KRAS G12D–targeted protein degrader vs docetaxel
Astellas Pharma Global Development is running this randomized, open-label Phase 3 study comparing setidegrasib (ASP3082) with docetaxel in adults with locally advanced unresectable or metastatic NSCLC harboring KRAS G12D (~5% of NSCLC). Setidegrasib is a KRAS G12D–selective protein degrader designed to eliminate the mutant KRAS protein rather than simply inhibit signaling, positioning it as a differentiated approach among KRAS-directed agents. Participants must have progressed on or after platinum-based chemotherapy and checkpoint inhibitor therapy, and have had no more than two prior systemic lines; key exclusions include actionable alternative drivers and uncontrolled/symptomatic CNS disease. The primary efficacy focus is progression-free survival (and overall survival), with crossover from docetaxel to setidegrasib allowed after progression—commercially important because docetaxel (with or without ramucirumab in practice) remains a common later-line option when targeted therapies aren’t available, but durability is limited.
Glendale, CA and Arlington, TX (so far) | NCT07566052
🔄 Phase 2 olomorasib KRAS G12C covalent inhibitor + pembrolizumab PD‑1 blockade in 1L NSCLC.
This UNC Lineberger Comprehensive Cancer Center study (NCT07639242) tests first-line olomorasib plus pembrolizumab in locally advanced/metastatic NSCLC with a KRAS G12C mutation and PD-L1 TPS 1–49%. Olomorasib (Eli Lilly) is an oral, selective next-generation KRAS G12C covalent inhibitor designed to lock KRAS G12C in an inactive state; pembrolizumab (Merck/MSD) is an anti–PD‑1 monoclonal antibody that restores T-cell anti-tumor activity. The goal is to estimate progression-free survival, with additional endpoints including response rate, duration of response, overall survival, safety, and exploratory biomarker work; the protocol also emphasizes tougher-to-study groups such as patients with brain metastases and less favorable ECOG performance status. Scientifically, the key angle is whether a better-tolerated, “next-gen” KRAS G12C inhibitor can be combined up front with PD‑1 therapy—an approach historically limited by overlapping toxicities seen with some KRAS inhibitor + immunotherapy combinations. Today, many PD‑L1 1–49% KRAS G12C NSCLC patients start with pembrolizumab plus platinum-doublet chemotherapy, with KRAS G12C inhibitors typically used after progression.
Chapel Hill, NC (single-site) | NCT07639242
🔄 Phase 2 PF-08634404 trial tests PD-1×VEGF bispecific antibody in stage II–III NSCLC
This open-label Phase 2 Pfizer study (NCT07489066) evaluates PF-08634404—an investigational bispecific antibody that targets PD-1 (to relieve T-cell inhibition) and VEGF/VEGF-A (to reduce pro-angiogenic, immunosuppressive signaling)—given either with chemotherapy or as monotherapy across three peri-definitive settings. Part A gives neoadjuvant PF-08634404 plus chemotherapy before surgery in resectable stage II–IIIB NSCLC without actionable genomic alterations, aiming to improve pathologic response at resection. Part B gives adjuvant PF-08634404 monotherapy after prior neoadjuvant chemo-immunotherapy and surgery for patients who did not achieve a pathologic complete response, addressing a high-relapse-risk group despite modern perioperative PD-1/PD-L1 regimens. Part C gives PF-08634404 consolidation after definitive concurrent chemoradiotherapy in unresectable stage III NSCLC without progression, testing an alternative to today’s typical approach of platinum-based cCRT followed by durvalumab consolidation when eligible. The commercial/scientific angle is “PD-(1) plus VEGF” dual-pathway modulation in earlier-stage disease, a strategy supported by biology and multiple programs in advanced NSCLC, now being moved into neoadjuvant/adjuvant/consolidation windows.
New Lenox and Hinsdale, IL; and Manatí, Puerto Rico | NCT07489066
Kidney (Renal Cell) Carcinoma Trial
🏆 TRIAL SPOTLIGHT
Post-adjuvant IO recurrence in RCC is becoming a real clinical category. Belzutifan plus zanzalintinib versus cabozantinib is a clean, practice-relevant Phase 3 test of whether HIF-2α + VEGFR/MET targeting can beat a common standard.
✅ Phase 3 belzutifan HIF-2α inhibitor + zanzalintinib MET/VEGFR TKI vs cabozantinib
Merck Sharp & Dohme (MSD) is running LITESPARK-033 (MK-6482-033), a Phase 3, randomized, open-label, multicenter study in recurrent advanced clear cell RCC after prior adjuvant anti–PD-1/PD-L1 therapy. Participants are randomized 1:1 to belzutifan (MK-6482), an oral HIF-2α inhibitor, plus zanzalintinib, an oral multikinase TKI targeting MET (c-Met) and VEGF receptors, versus Exelixis’s cabozantinib, an oral multikinase inhibitor (including MET, AXL/RET, and VEGFR2 pathways). Co-primary endpoints are progression-free survival by blinded independent central review (RECIST 1.1) and overall survival; key secondaries include objective response rate and duration of response. The scientific/commercial angle is “post-adjuvant IO recurrence” as a distinct and growing segment (especially after pembrolizumab adjuvant use), testing whether dual hypoxia signaling blockade (HIF-2α) plus angiogenesis/MET inhibition can outperform a widely used VEGFR/MET TKI. Today’s care after recurrence following adjuvant anti–PD-(L)1 often shifts to VEGFR TKIs (cabozantinib, axitinib, lenvatinib combinations) and selected IO/TKI regimens depending on prior IO exposure, timing, and risk group, so this trial directly challenges a common TKI standard in that setting.
Global, multi-center (US + Canada, Europe, Latin America, and Asia-Pacific) | NCT07227402
Breast Cancer Trial
🏆 TRIAL SPOTLIGHT
A head-to-head Phase 3 against today’s CDK4/6 standards is a major commercial and clinical swing. The key question is whether CDK4 selectivity can preserve first-line efficacy while improving tolerability or resistance dynamics.
✅ Phase 3 BGB-43395 CDK4-selective inhibitor plus letrozole vs CDK4/6 inhibitors in 1L HR+/HER2- MBC
This open-label, randomized, multicenter Phase 3 study (BeOne Medicines) tests whether BGB-43395—an oral, CDK4-selective small-molecule inhibitor—combined with letrozole (aromatase inhibitor) improves outcomes versus standard CDK4/6 inhibitors (abemaciclib [Eli Lilly], palbociclib [Pfizer], or ribociclib [Novartis]) combined with letrozole in previously untreated advanced/metastatic HR+/HER2- breast cancer. Participants are treated in the first-line metastatic setting, where current care for most patients is endocrine therapy (often an aromatase inhibitor like letrozole) paired with a CDK4/6 inhibitor, with later-line options including fulvestrant- or SERD-based regimens and targeted therapies when biomarkers apply. The scientific/commercial angle is the head-to-head concept: replacing “CDK4/6” with a more selective “CDK4-only” strategy that could preserve efficacy while changing tolerability and/or resistance dynamics in a very large market segment. The comparator arm reflects real-world practice by allowing investigator choice among the three approved CDK4/6 inhibitors.
US sites (MO, NJ, OH, AK, NE) plus Australia, Japan, and South Korea | NCT07492641
Colorectal Cancer Trial
🔄 Phase 2 denikitug (GS-1811) trial tests CCR8-targeted Treg-depleting antibody in MSS CRC.
This Phase 2, open-label, multicenter, randomized study from Gilead Sciences evaluates denikitug (GS-1811), a CCR8-directed monoclonal antibody designed to target CCR8-expressing tumor-infiltrating regulatory T cells (a strategy intended to reduce local immune suppression). Participants have unresectable locally advanced/recurrent or metastatic microsatellite-stable (MSS) / pMMR colorectal cancer in the third line or later setting, where checkpoint inhibitors alone usually have low activity and most patients cycle through later-line systemic options such as trifluridine-tipiracil (Lonsurf; Taiho) and regorafenib, often with bevacizumab (Avastin; Genentech/Roche) added to improve outcomes. Arms include denikitug monotherapy; denikitug + nivolumab (Opdivo; Bristol Myers Squibb), a PD-1 antibody intended to restore effector T-cell function; and denikitug + trifluridine-tipiracil (oral DNA synthesis inhibitor combination) + bevacizumab (VEGF antibody) after a safety run-in, versus the control of trifluridine-tipiracil + bevacizumab alone. The primary endpoint is investigator-assessed objective response rate per RECIST 1.1, making the key commercial/scientific question whether CCR8-directed immune remodeling can “convert” immunologically cold MSS CRC into a setting where PD-1 therapy and/or chemo-anti-VEGF backbones yield more meaningful tumor shrinkage.
North Haven, CT; Boston, MA; and sites in Australia (Camperdown, Herston) | NCT07527858
Head and Neck Cancer Trial
🔄 Pilot (early feasibility) ConformalFLASH ultra-high–dose-rate proton therapy (FLASH radiotherapy) for HNSCC reirradiation
This pilot interventional study (C-FLASH-01) evaluates whether ultra-high dose-rate conformal proton therapy (ConformalFLASH®) can be planned and delivered safely and reproducibly for patients who need head and neck re-irradiation and are not candidates for surgical resection. The investigational intervention is ConformalFLASH irradiation, which delivers proton radiation at ultra-high dose rates intended to harness the “FLASH effect” (normal-tissue sparing at very high instantaneous dose rates) while keeping target coverage conformal, using workflows described as comparable to stereotactic body radiation therapy (SBRT). The sponsor is IBA Proton Therapy, Inc., positioning this as a clinically oriented test of a novel delivery paradigm rather than a new systemic agent. The key scientific/commercial angle is translation of proton FLASH from physics/planning promise into a real-world, protocolized re-irradiation workflow in a population with tight organ-at-risk constraints. Today’s care for unresectable recurrent/second primary head and neck cancer after prior definitive radiation is individualized salvage—often systemic therapy (e.g., immunotherapy-based regimens when appropriate) with or without highly conformal re-irradiation such as SBRT/IMRT/proton therapy—because cumulative toxicity limits curative dosing.
Philadelphia, PA (single-site) | NCT07644585
Ewing Sarcoma / Osteosarcoma Trial
🔄 Phase 2 eflornithine (DFMO) ornithine decarboxylase inhibitor added/maintenance in Ewing and osteosarcoma
This Phase 2, non-randomized, open-label basket study tests eflornithine (DFMO)—a small-molecule ornithine decarboxylase (ODC) inhibitor that suppresses polyamine synthesis—as concurrent therapy with multi-agent chemotherapy and/or as long maintenance after completion of local control/consolidation in Ewing sarcoma and osteosarcoma. Milton S. Hershey Medical Center sponsors the master protocol with five parallel cohorts spanning relapsed/refractory, metastatic-at-diagnosis, and post-local-control settings, including DFMO alone for resected lung-metastasis relapse osteosarcoma and DFMO added during post-surgery MAP-based consolidation (high-dose methotrexate + doxorubicin + cisplatin) in osteosarcoma cohorts. Ewing cohorts incorporate DFMO alongside commonly used regimens such as cyclophosphamide + topotecan and multi-agent VDC/IE-style backbones (vincristine/doxorubicin/cyclophosphamide alternating with ifosfamide/etoposide). The commercial/scientific angle is repositioning a metabolism-targeting oral agent as a long-duration maintenance strategy in AYA bone sarcomas where recurrence risk remains high despite intensive chemo plus surgery ± radiation. Today’s care for Ewing and osteosarcoma is multi-agent chemotherapy with definitive surgery when feasible and radiation in selected cases, with limited proven maintenance options after remission—this trial asks whether sustained ODC inhibition can extend event-free and overall survival versus historical outcomes.
Tampa, FL and Hershey, PA | NCT07321912
Multiple Solid Tumors Trials
🆕 Phase 1 BSI-082 anti-SIRPα macrophage checkpoint antibody ± T-DXd HER2 ADC
This open-label Phase 1a/1b, two-part dose-escalation and dose-expansion study evaluates BSI-082 (a fully human anti–SIRPα monoclonal antibody that blocks the CD47–SIRPα “don’t eat me” signal to enhance macrophage-mediated phagocytosis and downstream T/NK anti-tumor immunity) as monotherapy and in combination with trastuzumab deruxtecan (T-DXd; Daiichi Sankyo/AstraZeneca’s HER2-directed antibody-drug conjugate delivering a deruxtecan topoisomerase I inhibitor payload). The Phase 1a portion escalates IV BSI-082 given every 21 days to define safety, pharmacokinetics, and a recommended dose; the Phase 1b combination evaluates BSI-082 with fixed-dose T-DXd and then expands in patients who have an FDA-approved indication for T-DXd. Eligible patients have locally advanced or metastatic solid tumors that are relapsed/refractory, progressed after available systemic therapy, or are ineligible for standard options. The key angle is pairing HER2-targeted ADC “eat me” biology with SIRPα blockade to counter adaptive anti-phagocytic signaling and potentially deepen responses beyond what ADC chemotherapy delivery alone achieves. Today’s care in these late-line solid tumors is histology- and biomarker-driven systemic therapy (chemo, targeted therapy when actionable alterations exist, and ADCs like T-DXd for appropriate HER2-expressing indications), with clinical trials commonly pursued after progression.
San Antonio, TX (single-site, University of Texas Health Science Center at San Antonio) | NCT07086768
🔄 Phase 1/2 PM54 + pembrolizumab tests tubulin (microtubule) inhibition paired with PD-1 blockade.
This multicenter, open-label Phase 1/2 trial from PharmaMar evaluates PM54 (a small-molecule microtubule/tubulin inhibitor) combined with pembrolizumab (Keytruda; Merck/MSD’s PD-1–blocking monoclonal antibody) in adults with previously treated, advanced/metastatic, progressive, unresectable solid tumors. Part 1 is a safety run-in with PM54 dose escalation alongside pembrolizumab to establish a recommended combination dose; Part 2 expands at that dose to estimate antitumor activity using investigator-assessed endpoints such as objective response rate and clinical benefit rate. The scientific/commercial angle is a cytotoxic “backbone” designed to trigger tumor cell death and potentially increase immunogenicity, layered onto a widely used checkpoint inhibitor platform. Today, patients with refractory metastatic solid tumors typically cycle through tumor-specific options (targeted therapy when actionable alterations exist, immunotherapy in responsive histologies/biomarker-defined subsets, and multiple sequential chemotherapies/ADCs), so this study explores whether a new tubulin agent can deepen or restore responses when prior systemic therapy has failed.
Fort Worth & Houston, TX; Fairfax, VA; New York, NY; Irving, TX | NCT07644039
🆕 Phase 1/2 VMD-102 selective PKCε (protein kinase C epsilon) inhibitor in HCC and solid tumors.
This first-in-human, open-label, multicenter Phase 1/2 study sponsored by VM Discovery evaluates VMD-102, a selective PKC epsilon (PKCε) kinase inhibitor intended to suppress oncogenic PKC signaling that can promote tumor cell survival, growth, and invasion. Phase 1 uses dose escalation (25 participants; 21-day cycles) to characterize safety/tolerability and pharmacokinetics and to establish the maximum tolerated dose and recommended Phase 2 dose. Phase 2 expands using a Bayesian Optimal Phase II (BOP2) design (86 total across two cohorts) to estimate preliminary anti-tumor activity and further safety in advanced/unresectable hepatocellular carcinoma (Child-Pugh up to 7 depending on phase) and a mixed cohort including metastatic uveal melanoma, renal cell carcinoma, non-small cell lung cancer, and colorectal cancer; biomarker-guided enrollment may be added later. Clinically, these populations typically receive systemic therapy matched to disease context—e.g., HCC often uses immunotherapy + anti-VEGF or TKIs; RCC often uses IO/TKI combinations; metastatic uveal melanoma has limited effective systemic options—so the commercial/scientific angle is testing the first selective PKCε inhibitor in humans across multiple hard-to-treat settings.
Scottsdale, AZ (single US site listed so far) | NCT07636785
🆕 Phase 1 MSK‑TCR5 CD8α/β‑armored KRAS/HRAS/NRAS G12D TCR‑T therapy
This Memorial Sloan Kettering Cancer Center (MSKCC) investigator-initiated Phase 1, open-label, dose-escalation trial tests MSK‑TCR5—an autologous T‑cell receptor–engineered T‑cell product designed to recognize the RAS G12D neoantigen presented by HLA‑A11:01. The “armoring” concept is to enhance engineered T‑cell function/persistence beyond a standard TCR‑T approach by adding CD8 alpha/beta co-receptor support (i.e., strengthening recognition of the target peptide–HLA complex), with the goal of improving activity in solid tumors. The study enrolls adults with unresectable, advanced/metastatic solid tumors (including CRC and NSCLC cohorts described in the record) whose cancers harbor KRAS, HRAS, or NRAS G12D and who carry HLA‑A11:01, generally after at least second-line therapy (some parts require prior immune checkpoint inhibitor exposure and/or biomarker-defined subsets such as dMMR/MSI‑H). The intervention is MSK‑TCR5 given at escalating cell doses, aiming to establish safety and a recommended dose for expansion. Today’s care for these patients is tumor-specific systemic therapy (often immunotherapy and/or targeted therapy when eligible, otherwise sequential chemotherapy), and this trial explores a precision, HLA-restricted cellular therapy option for the RAS G12D population where durable options are limited beyond standard lines.
Multiple New York sites (Montvale, Commack, Harrison, Uniondale, Basking Ridge, Middletown) | NCT07638371
This summary is produced by Sagely Health. We salute the scientists, doctors, professionals, and patients who are advancing cancer treatment. If you have any questions, comments, suggestions, or corrections, we’re listening at info@sagelyhealth.com.