GI Oncology Daily Digest

OrigAMI-1 is a phase Ib/II trial (NCT05379595) of amivantamab — an EGFR-MET bispecific antibody already approved in NSCLC — given as monotherapy to 94 patients with chemorefractory (2-3 prior lines), centrally confirmed RAS/BRAF/EGFR-ectodomain wild-type mCRC. Objective response rates were 29% (5/17) in left-sided, anti-EGFR–naive disease (cohort A), 19% (10/54) in left-sided patients with prior anti-EGFR therapy (cohort B), and 22% (10/23) in right-sided disease (cohort C), with median duration of response of 9.0, 6.1, and 9.8 months respectively. Median PFS was 5.7, 4.6, and 3.7 months across cohorts. The safety profile mirrored NSCLC experience — the most common grade ≥3 treatment-related events were rash (7%), dermatitis acneiform (4%), and hypoalbuminemia (4%), with only one treatment-related discontinuation. Durable single-agent activity regardless of prior anti-EGFR exposure or primary tumor location is the headline; amivantamab plus chemotherapy is now being tested in two phase III mCRC studies.

The single-arm phase II PLATIC trial treated 57 patients with unresectable HCC using the PD-1 inhibitor sintilimab plus lenvatinib combined with TACE and hepatic arterial infusion chemotherapy (TACE-HAIC). After a median of three cycles, 44 patients (77.2%; 95% CI 0.64-0.87) were converted to surgical resection — the primary endpoint. Objective response was 80.7% by mRECIST and 43.8% by RECIST 1.1; among resected patients median PFS was 17.3 months and median OS was not reached. Grade ≥3 treatment-related adverse events occurred in 64.9% with no treatment-related deaths. Against an inverse-probability-weighted historical TACE-HAIC cohort, PLATIC showed longer PFS (HR 0.55, p=0.020) and OS (HR 0.50, p=0.029). The results are encouraging, though the single-arm design with historical comparison tempers interpretation pending randomized confirmation.

OPTICAL-2 is an open-label, randomized phase II trial (NCT05571644) that assigned 123 patients with proficient mismatch repair/microsatellite-stable (pMMR/MSS) locally advanced colorectal cancer 1:1:1 to neoadjuvant mFOLFOXIRI plus cadonilimab (a PD-1/CTLA-4 bispecific), mFOLFOXIRI alone, or mFOLFOX6, followed by surgery. Adding cadonilimab significantly improved pathologic complete response versus mFOLFOX6 (26.8% vs 9.8%; OR 3.4, 95% CI 1.04-13.24, p=0.046), whereas chemotherapy intensification with mFOLFOXIRI alone did not (14.6%, p=0.50). Major pathologic response (68.3% vs 43.9%, p=0.026) and downstaging to ypStage 0-I (66% vs 41%, p=0.027) also favored the cadonilimab arm. Grade ≥3 events — chiefly liver-enzyme elevation and neutropenia — were manageable. The trial offers an early but important signal that dual immune checkpoint blockade, not added chemotherapy, can generate pathologic responses in MSS colorectal cancer, a setting historically resistant to immunotherapy.