Ezabenlimab (BI 754091), an anti-PD-1 antibody, in combination with BI 836880, a VEGF/Ang2-blocking nanobody, in patients (pts) with advanced colorectal cancer (CRC).

98 Background: Anti-PD-1 antibodies may have synergistic effects with other immunomodulatory or targeted agents. This open-label, Phase II platform trial is investigating ezabenlimab, an anti-PD-1 antibody, combined with other agents. Module C of the platform is assessing ezabenlimab plus BI 836880, a humanized bispecific nanobody targeting VEGF/Ang2. Pts are being enrolled into 5 advanced solid tumor cohorts: gastric/gastroesophageal adenocarcinoma; solid tumors (except non-squamous NSCLC or melanoma) with secondary resistance to anti-PD-(L)1 treatment (progression after at least SD for ≥4 months); solid tumors with primary resistance to anti-PD-(L)1 treatment; microsatellite stable (MSS) CRC; mismatch repair-proficient/MSS endometrial carcinoma. Here, we report data from the CRC cohort which has completed recruitment. Methods: Pts with locally advanced, unresectable or metastatic, MSS CRC were enrolled. Patients had received ≥1 line of prior systemic therapy for metastatic disease but were anti-PD-(L)-1 therapy-naïve. Prior anti-angiogenic therapy was permitted. Pts received BI 836880 720 mg plus ezabenlimab 240 mg iv q3w for 1 year or until disease progression, consent withdrawal or undue toxicity. Primary endpoint: investigator-assessed OR (CR or PR per RECIST v1.1). Secondary endpoints: duration of response, disease control, and PFS; safety is also being assessed. Results: 30 pts have been treated: 57% male; median age 61.5 years. All pts had received prior chemotherapy; most pts (23 [77%]) had received prior bevacizumab. At data cut-off (Sep 2021), median duration (range) of treatment was 115.5 (28–295) days; 6 pts remain on treatment. 1 (3%) pt (who had not received prior bevacizumab) achieved a confirmed PR; 16 (53%) pts had SD. Median duration (range) of SD was 128.5 (42–242) days. 29/17/2 (97/57/7%) pts had an AE (any/G3/G4). The most frequent AEs (any/G3) were nausea (40/10%), fatigue (30/3%), peripheral edema (30/0%), vomiting (27/7%), and hypertension (27/17%). There were two G4 AEs (hypertension; platelet count decreased) and no G5 AEs. 24/10/2 (80/33/7%) pts had a drug-related AE (any/G3/G4); most commonly (any/G3) nausea (33/7%), fatigue (27/3%) and hypertension (27/17%). 3 (10%) pts had an infusion-related reaction (G1, n = 1; G2, n = 2). 2 (7%) pts had an AE leading to discontinuation (G3 bile duct stone and G2 peripheral edema). Immune-related AEs were reported in 6 (20%) pts and serious AEs occurred in 13 (43%) pts. Conclusions: BI 836880 plus ezabenlimab had a manageable safety profile in pts with advanced MSS CRC; however, anti-tumor activity was limited in these pts, the majority of whom had received prior bevacizumab. Clinical trial information: NCT03697304.