Apex AI Institute / OncoLattice team
Clinical Development Critiques
Critiques of clinical development news, compiled from public sources.
23 min read
EA2251: Summit taps ECOG-ACRIN to rerun a first-line biliary tract cancer OS trial in the US, supplying the US data the FDA wants for HARMONi-GI1
[Fact] EA2251 (NCT07765433) is a US Phase 2/3 of ivonescimab plus GemCis vs pembrolizumab or durvalumab plus GemCis, with OS as the primary endpoint and an estimated 336 patients. [Inference] It is a direct answer to the FDA's concerns about single-country data, but its open-label design and cooperative-group pace will decide how fast it reads out.
- BTC
- Biliary
- Bispecific
- PD-1/VEGF
- Cooperative group
- FDA
34 min read
HARMONi-GI1: ivonescimab plus GemCis beats durvalumab plus GemCis on OS in first-line biliary tract cancer (HR 0.75), but it is a single-region Chinese interim analysis with imbalanced subsequent therapy
[Fact] HARMONi-GI1 (n=682, China): mOS 14.9 vs 12.1 months (HR 0.75), PFS HR 0.61, ORR 53.8% vs 32.5%. [Inference] It is the first trial in BTC to win OS head-to-head against IO plus chemotherapy; the NMPA path is clear, but the FDA path depends on the US EA2251 trial.
- BTC
- Biliary
- Bispecific
- PD-1/VEGF
- GI
- ESMO 2026
21 min read
Krascendo 1: divarasib beats sotorasib/adagrasib head-to-head with PFS HR 0.58 and OS HR 0.56, challenging first-generation KRAS G12C inhibitors
[Fact] Krascendo 1 (n=338): divarasib vs sotorasib/adagrasib, mPFS 9.8 vs 5.7 months (HR 0.58), mOS 17.7 vs 12.0 months (HR 0.56). [Inference] If approved it would challenge the later-line KRAS G12C standard, but 91% of the control arm received sotorasib; the next battle is first line, in Krascendo 2.
- NSCLC
- Lung
- KRAS G12C
- Targeted
- Head-to-head
- ESMO 2026
22 min read
DASL-HiCaP: darolutamide added to ADT plus radiotherapy gives MFS HR 0.42 in very-high-risk prostate cancer, but no OS difference yet
[Fact] n=1106, MFS-CI HR 0.42 (95% CI 0.27–0.66), 4-year MFS 95% vs 88%; OS HR 0.88 (0.49–1.59). [Inference] The result reinforces the trend toward ARPI intensification for high-risk localized disease, but with an absolute difference of 7% and only 90 events, whether the FDA accepts MFS remains the key question.
- Prostate
- Darolutamide
- Radiotherapy
- GU
- ESMO 2026
19 min read
DeLLphi-305: first-line maintenance tarlatamab plus durvalumab lifts median OS in ES-SCLC from 12.7 to 21.1 months, and bispecific TCEs formally enter first line
[Fact] n=563, mOS 21.1 vs 12.7 months (HR 0.42, 95% CI 0.32–0.54), mPFS 7.4 vs 1.9 months (HR 0.34); CRS 49.6% (1 grade 3 case). [Inference] Significance at the interim analysis will change the SOC; the challenges are putting CRS monitoring into practice and the selection bias of enrolling only patients without progression.
- SCLC
- Lung
- Bispecific
- TCE
- ESMO 2026
22 min read
iza-bren PANKU-Breast01: the EGFR×HER3 bispecific ADC meets its PFS primary endpoint in HR+/HER2- breast cancer, but the trial was run only in China
[Fact] n=383 (190 vs 193), BICR mPFS 8.2 vs 4.0 months (HR 0.41, 95% CI 0.33–0.53), cORR 40.5% vs 20.2%, adjudicated ILD 1.1%; OS immature, grade ≥3 TRAE rate not disclosed. [Inference] Meeting the PFS primary endpoint supports validating the dual-target design in global trials; a single Chinese population and a chemotherapy control are not enough on their own to support FDA approval.
- Breast
- ADC
- Bispecific
- EGFR
- HER3
- China
26 min read
LITESPARK-012: adding belzutifan or quavonlimab to Keytruda plus Lenvima both failed, and triplet stacking in first-line ccRCC has reached its end
[Fact] n=1688; belzutifan triplet PFS 20.8 vs 23.7 months (HR 1.08, 95% CI 0.93–1.26), OS HR 1.04 (0.87–1.25); the quavonlimab arm also failed, with lower ORR. [Inference] The main cause of failure may be toxicity-driven discontinuation and lower dose intensity; first-line HIF-2α needs a different partner or a different population.
- RCC
- Kidney
- HIF-2α
- GU
- ESMO 2026
27 min read
VOLGA: perioperative durvalumab plus 3 doses of EV wins both EFS and OS in cisplatin-ineligible MIBC, but pCR of 45% vs 57% in EV-303 leaves the partner question open
[Fact] In VOLGA the doublet arm had EFS HR 0.51 and OS HR 0.58, with pCR 45% vs 10% (local review); the triplet arm's OS was not significant. [Inference] AZ has a second perioperative regimen for cisplatin-ineligible MIBC, but its only differentiation from Keytruda plus Padcev is fewer EV cycles.
- MIBC
- Bladder
- ADC
- Perioperative
- GU
- ESMO 2026
22 min read
FDA grants accelerated approval to dostarlimab in dMMR/MSI-H locally advanced rectal cancer: can cCR12-based organ preservation move to the bladder?
[Fact] FDA granted accelerated approval to dostarlimab in dMMR/MSI-H locally advanced rectal cancer on single-arm cCR12 data; some patients may avoid chemotherapy, radiotherapy and surgery. [Inference] This sets a regulatory template for an immunotherapy + watch-and-wait organ-preservation path.
- Rectal cancer
- Organ preservation
- FDA
- Immunotherapy
22 min read
GSK B7-H3 ADC ris-rez first global data: the 61% ORR at 8 mg/kg is 11 of 18 patients, and GU should watch the mCRPC Phase 3
[Fact] At 8 mg/kg the confirmed ORR was 61% (11/18) with 42% grade ≥3 TRAEs; at 6 mg/kg, 32% and 15%. [Inference] The sample is small and intervals wide; taking 8 mg/kg into Phase 3 trades higher toxicity for a higher response rate.
- ADC
- B7-H3
- SCLC
- mCRPC
24 min read
Summit/Akeso start SMT112-216, a global first-line ccRCC Phase 2 of ivonescimab: an exploratory design that screens combinations by ORR
SMT112-216 is a small, open-label combination-selection trial with investigator-assessed ORR as its primary endpoint, not a registrational path. It can answer which ivonescimab IO combination deserves Phase 3, not whether it can beat nivo + ipi.
- ccRCC
- Bispecifics
- Trial design
- GU
18 min read
Roche returns to Hansoh: $400 million upfront for an ADC platform and early pipeline
In substance, Roche is using US$510 million in committed payments to buy an ADC production line in which Roche picks the targets and Hansoh takes them through global Phase Ia. The US$15 billion-plus in milestones is a ceiling, not an expected value.
- ADC
- Deals
- China biotech
22 min read
FDA extrapolates CIS data to approve Anktiva + BCG in BCG-unresponsive papillary-only NMIBC: the regulatory precedent matters more than the drug
[Fact] FDA extrapolated from CIS-cohort evidence to approve Anktiva + BCG in BCG-unresponsive papillary-only NMIBC. [Inference] The news is the regulatory precedent: single-arm papillary-only DFS trials lose value and well-powered CIS cohorts gain it.
- NMIBC
- FDA
- Regulatory
- GU
23 min read
Gilead pays up to $1.9 billion for three Vir dual-masked T-cell engagers: a platform deal whose GU angle is VIR-5500
This is an early platform-option deal with a small upfront, large milestones and no clinical data. Its value rests on VIR-5500, a PSMA×CD3 molecule, so for GU the thing to watch is the design of VIR-5500's 2027 Phase 3, not Gilead's three undisclosed targets.
- T-cell engagers
- Deals
- PSMA
- GU
21 min read
BMS RYZ101 (Ac-225 DOTATATE) ACTION-1 hits at interim: the first randomized report card for an alpha radioligand, and how GU should read it
ACTION-1 shows that an alpha radioligand can still extend PFS after beta agents fail, but the effect size is entirely undisclosed. For GU, the real lesson is that a post-Lu-177 population can support a registrational trial, and the bottleneck will shift from efficacy to Ac-225 supply and nuclear medicine site capacity.
- Radioligands
- Ac-225
- NET
39 min read
Two mRNA cancer vaccine Phase 3 trials, one win and one loss: the difference is trial design, not the platform
INTerpath-001 succeeded and AHEAD-MERIT missed on PFS. The outcome was decided mainly by disease stage, antigen strategy and endpoint choice, not by the mRNA platform itself.
- mRNA vaccines
- Phase 3
- Trial design