Pembrolizumab Fails to Show Survival Benefit in PD-L1 ≥1 Gastric/GEJ Cancers

Kohei Shitara, MD

Second-line treatment with pembrolizumab (Keytruda) did not significantly improve overall survival (OS) or progression-free survival (PFS) in patients with advanced or metastatic gastric or gastroesophageal junction (GEJ) cancer with a PD-L1 combined positive score (CPS) ≥1 when compared with paclitaxel. These findings, which were from the phase III KEYNOTE-061 study, were presented at the 2018 World Congress on GI Cancer.

“Pembrolizumab did not reach the prespecific level of statistical significance for improving OS over paclitaxel as second-line therapy for advanced or metastatic gastric or GEJ cancer with a PD-L1 CPS ≥1,” said lead study author Kohei Shitara, MD, from National Cancer Center Hospital East, Kashiwa, Japan.

In the trial, 592 patients with metastic or locally advanced and unresectable gastric or GEJ cancer were randomized 1:1 to receive pembrolizumab at 200 mg every 3 weeks or paclitaxel at 80 mg/m² on days 1, 8, and 15 of 4-week cycles.

OS and PFS in patients with PD-L1 CPS ≥1 was the primary endpoint of this study. At a median follow-up of 7.9 months and a data cutoff date of October 26, 2017, Shitara reported that in patients with a PD-L1 CPS ≥1, the median OS was 9.1 months (95% CI, 6.2-10.7) with pembrolizumab and 8.3 months (95% CI, 7.6-9.0) with paclitaxel. This was associated with a 18% reduction in the risk for death with pembrolizumab (HR, 0.82; 95% CI, 0.66-1.03).

Additionally, the median OS for patients with PD-L1 CPS <1 with pembrolizumab was 4.8 months (95% CI, 3.9-6.1), compared with 8.2 months (95% CI, 6.8-10.6) with paclitaxel. Patients with a PD-L1 CPS ≥10 treated with pembrolizumab experienced a median OS of 10.4 months (95% CI, 5.9-17.3) compared with 8.0 months (95% CI, 5.1-9.9) with paclitaxel.

Blinded central independent review (BICR) by RECIST v1.1 criteria showed a median OS for patients with CPS ≥1 was 1.5 months (95% CI, 1.4-2.0), and 4.1 months (95% CI, 3.1-4.2) with paclitaxel. For patients with a CPS ≥10 treated with pembrolizumab, median OS was 2.7 months (95% CI, 1.4-3.1), and 3.4 months (95% CI, 2.7-4.1) with paclitaxel.

Median OS for patients with an ECOG performance status (PS) of 0 was 12.3 months (95% CI, 9.7-15.8) for pembrolizumab and 9.3 months (95% CI, 8.3-10.5) for paclitaxel. For patients with ECOG PS of 1, the median OS was 5.4 months (95% CI, 3.7-7.7) with pembrolizumab and 7.5 months (95% CI, 5.3-8.4) with paclitaxel.

Shitara noted that the pembrolizumab treatment effect for OS was greater in patients with ECOG PS 0, PD-L1 CPS ≥10 tumors, and microsatellite instability-high (MSI-H) tumors. However, these were not the primary endpoints of the study.

Median PFS for patients with CPS ≥1 treated with pembrolizumab was 1.5 months (95% CI, 1.4-2.0) compared with 4.1 months with paclitaxel (95% CI, 3.1-4.2). For patients with CPS ≥10 treated with pembrolizumab, median PFS was 2.7 months (95% CI, 1.4-3.1) compared with 3.4 months with paclitaxel (95% CI, 2.7-4.1).

Secondary endpoints of the trial were median duration of response (DOR) and overall response rate (ORR) in patients with PD-L1 CPS ≥1, which was evaluated by RECIST v1.1 BICR. The ORR for pembrolizumab was 15.8% compared with 13.6% for paclitaxel. Patients treated with pembrolizumab had a DOR of 18.0 months compared with 5.2 months for those treated with paclitaxel.

The OS, PFS, and ORR for patients with MSI-H tumors were also reported. The median OS for patients with MSI-H tumors treated with pembrolizumab was not reached (95% CI, 5.6 months-NR) versus 8.1 months (95% CI, 2.0-16.7) with paclitaxel. Median PFS for pembrolizumab was 17.8 months (95% CI, 2.7-NR) versus 3.5 months (95% CI, 2.0-9.8) for paclitaxel.

Of the 15 patients with MSI-H tumors treated with pembrolizumab, 7 responded, with an ORR of 46.7% (95% CI, 21.3-73.4). Comparatively, 2 of 12 patients with MSI-H tumors treated with paclitaxel responded with an ORR of 16.7% (95% CI, 2.1-48.4).

Although pembrolizumab did not significantly improve OS or PFS, the safety profile was deemed safer than that of paclitaxel. Of those who received pembrolizumab (n = 294), 155 experienced treatment-related adverse events (AEs). These included fatigue (11%), decreased appetite (8.2%), nausea (5.8%), diarrhea (5.4%), anemia (3.4%), alopecia (0.3%), and peripheral neuropathy (0.3%).

The rate of treatment-related AEs with paclitaxel were significantly higher, with 232 of 276 patients experiencing higher rates of alopecia (40.2%), fatigue (23.2%), nausea (18.1%), decreased appetite (15.5%), peripheral neuropathy (14.5%), anemia (14.1%), and diarrhea (13.8%).

Treatment-related AEs led to 3 deaths in the pembrolizumab arm and 1 in the paclitaxel arm. Nine patients (3.1%) discontinued treatment with pembrolizumab due to AEs, while 15 (5.4%) discontinued treatment with paclitaxel.

Additionally, immune-mediated AEs and infusion reactions occurred in 54 patients (18.4%) in the pembrolizumab arm, which led to 2 deaths, and 21 (7.6%) in the paclitaxel arm, which did not lead to any deaths. Immune-mediated AEs and infusion reactions with pembrolizumab included hypothyroidism (7.8%), hyperthyroidism (4.1%), pneumonitis (2.7%), hepatitis (1.4%), hypophysitis (1.4%), colitis (1.0%), and severe skin reactions (0.3%).

Prior findings did show clinical activity and manageable safety in patients with advanced gastric or GEJ cancer, as seen in the KEYNOTE-059 trial. Here, the ORR was 15.5% in those who received more than 2 prior lines of therapy with a PD-L1 CPS ≥1.2 The FDA approved pembrolizumab for this patient population in September 2017.

Although PFS and OS were not improved with pembrolizumab in patients with PD-L1 CPS ≥1, responses were more durable than the responses observed with paclitaxel, explained Shitara.

“Data support further exploration to identify patients likely to benefit from pembrolizumab monotherapy and ongoing development of pembrolizumab-based combination therapy,” concluded Shitara.

References

1. Shitara K, Ozguroglu M, Bang YJ, et al. KEYNOTE-061: phase 3 study of pembrolizumab vs paclitaxel for previously treated advanced gastric or gastroesophageal junction (G/GEJ) cancer. In: Proceedings from the 2018 World Congress on Gastrointestinal Cancer; June 20-23, 2018; Barcelona, Spain. Abstract LBA-005.
2. FDA Approves Merck’s KEYTRUDA^® (pembrolizumab) for Previously Treated Patients with Recurrent Locally Advanced or Metastatic Gastric or Gastroesophageal Junction Cancer Whose Tumors Express PD-L1 (CPS Greater Than or Equal to 1). Merck. http://bit.ly/2wberGs. Accessed September 22, 2017.