Bavituximab/Pembrolizumab Combo Is Active in Advanced Gastric/GEJ Cancer

Ian Chau, MD, FRCP

The combination of bavituximab and pembrolizumab (Keytruda) demonstrated synergistic antitumor activity and was well tolerated in patients with pretreated advanced gastric or gastroesophageal junction cancer, according to findings from a phase 2 study (NCT0409641) that were presented during the virtual 2020 ESMO Congress.

Among 36 evaluable patients, the objective response rate (ORR) was 19.4% (n = 7), which consisted of a 5.5% complete response (CR) rate (n = 2) and a 13.9% partial response (PR) rate (n = 5). The stable disease rate was 19.4% (n = 7), which translated to a disease control rate of 38.9% (n = 14).

“Antitumor activity was observed in this high unmet need population of both second- and third-line patients with lower historical checkpoint inhibitor response rates, such as microsatellite stable [MSS; patients] and [those with a] combined positive score [CPS] of less than 1,” lead study author, Ian Chau, MD, FRCP, of the Royal Marsden NHS Foundation Trust, and co-authors, wrote in the poster that was presented during the meeting. “These data support the proposed hypothesis that inhibition of immunosuppressive cells with bavituximab can potentiate checkpoint inhibitor activity.”

Inhibition of PS receptor mediated activation in both exhausted myeloid and lymphoid immune cells with bavituximab could inhibit immunosuppressive myeloid cells, restimulate T cells, or render exhausted T cells that are vulnerable to checkpoint inhibitors.​

The international, multicenter, open-label, single-arm study enrolled patients with checkpoint inhibitor–naïve advanced gastric or GEJ cancer. Patients with known microsatellite instability–high disease were excluded from enrollment.

The study consisted of a safety run-in phase, with a 21-day dose-limiting toxicity observation period, followed by an expansion phase.

A total of 3 patients were enrolled in the safety run-in phase, where the recommended expansion dose was determined to be 3 mg/kg of bavituximab weekly plus 200 mg of pembrolizumab every 3 weeks. The expansion phase enrolled an additional 37 patients.

The population was predominantly male (n = 30; 75%), and the median age was 62.5 years. The primary site in the majority of patients was gastric (n = 25; 62.5%) versus GEJ (n = 15; 37.5%), and 70% of patients (n = 28) had received 1 line of prior therapy. Non-Asian and Asian patients were well represented, at 55% (n = 22) and 45% (n = 18), respectively.

Among responders, the duration of treatment ranged from approximately 63 to 315 days or 3 to 15 cycles, respectively.

Regarding safety, no dose-limiting toxicities were observed in the safety run-in phase. As of the data cutoff date of July 6, 2020, the most common treatment-emergent adverse effects (TEAEs) included decreased appetite (n = 16; 40%), nausea (n = 15; 37.5%), fatigue (n = 12; 30%), constipation (n = 10; 25%), vomiting (n = 10; 25%), anemia (n = 8; 20%), and weight decrease (n = 8; 20%). The majority of TEAEs (90%) were of grades 1 to 3. However, five grade 5 TEAEs (10%) were observed and included disease progression (n = 3), aspiration pneumonia (n = 1), and respiratory failure (n = 1).

Fourteen patients had at least 1 serious AE (SAE). All SAEs were unrelated to the combination, with the exception of 1 case of pneumonitis (since recovered/resolved), which was found to be related to bavituximab and pembrolizumab. Investigators noted that no new safety signals were reported with the combination.

Thirty-two patients discontinued treatment due to disease progression (n = 24), AEs (n = 5), and other (n = 3). Eight patients remained on treatment as of the data cutoff date of August 14, 2020.

Subset antitumor analyses were also performed per MSS status, PD-L1 expression, tumor microenvironment (TME) status, and neutrophil to lymphocyte (NLR) ratio.

Among 23 evaluable patients with MSS disease, the ORR was 17.4% (n = 4). The CR and PR rate was 8.7% each (n = 2 each), and the SD rate was 17.4% (n = 4). Thirteen patients had unknown status, and the ORR in this cohort was 23.1% (n = 3), with a PR and SD rate of 23.1% each (n = 3 each).

Higher PD-L1 expression did not appear to correlate with improved responses. Among patients with a CPS less than 1 (n = 10), 1 to 9 (n = 9), 10 or greater (n = 10), and unknown (n = 7), the ORRs were 20% (n = 2), 0%, 20% (n = 2; PR only), and 42.9% (n = 3; PR only), respectively. Of the 2 patients with a CPS less than 1, both responses were CRs.

Notably, antitumor activity was enriched in patients with reduced NLR and in patients whose TME RNA expression profile displayed hallmarks of an active or suppressed immune response.

All responders had an NLR of less than 4 (n = 21). The ORR in patients with an NLR less than 4 was 33% (n = 7), which compared favorably to 0% in 15 evaluable patients with an NLR of 4 or greater.

Among 14 TME-1 biomarker panel–positive patients, 21.4% responded to treatment (n = 3) versus 11.1% (n = 1) of 9 biomarker panel–negative patients.

“Overall, the bavituximab/pembrolizumab combination in advanced gastric cancer patients appears promising, especially in TME-1 biomarker positive patients with NLR values less than 4 where the combined ORR is 42.9% [n = 3],” concluded the study authors.

Reference

Chau I, Park H, Lee J, et al. Initial safety and efficacy findings with bavituximab plus pembrolizumab in patients with advanced gastric or gastroesophageal cancer. Presented at 2020 ESMO Congress; September 19-September 21, 2020; Virtual. Abstract 1446P.