Pembrolizumab may potentiate subsequent therapy in patients with advanced gastric or gastroesophageal junction adenocarcinoma

An exploratory analysis of the KEYNOTE-061 trial explored the differences in overall survival (OS) and response to immediate subsequent therapy following disease progression in patients who received pembrolizumab or paclitaxel in the second-line setting with a focus on ramucirumab + paclitaxel versus other regimens. Both second-line pembrolizumab followed by any subsequent therapy and followed by ramucirumab + paclitaxel were associated with longer OS than second-line paclitaxel followed by any subsequent therapy, suggesting that pembrolizumab may potentiate subsequent therapy and a long-term OS benefit may be possible after disease progression on pembrolizumab monotherapy.

Background

KEYNOTE-061 was a randomised, open-label phase III study of pembrolizumab versus paclitaxel in the second-line treatment of patients with in advanced gastric or gastroesophageal junction (GEJ) cancer. In the primary analysis of this trial, focussing on patients with CPS ≥1, pembrolizumab did not significantly improve the OS or progression-free survival (PFS). In an updated analysis with approximately two years of additional follow-up, a numerically improved OS was seen with pembrolizumab compared to paclitaxel (no difference in PFS). At the 2020 ESMO World Congress on Gastrointestinal cancers, an analysis of OS in response to immediate subsequent line of therapy following recurrence or relapse on KEYNOTE-061 study treatment was presented. The rationale for this analysis can be found in recent reports in multiple cancer types that demonstrated better outcomes on subsequent therapies following anti-PD-1/PD-L1 therapy. These findings suggest the possibility that PD-1/PD-L1 blockade potentiates subsequent therapy. Ramucirumab (anti-VEGFR-2) in combination with paclitaxel is generally considered as the standard therapy for patients with advanced gastric/GEJ cancer whose disease progressed on prior therapy. Ramucirumab + paclitaxel, administered after progression on pembrolizumab, has been associated with favourable outcomes in single-institution observational studies, including post-study observations for the global phase II KEYNOTE-059 trial. Key questions even though remain whether or not pembrolizumab potentiates the effects of immediate subsequent therapy and if pembrolizumab potentiates the effects of immediate subsequent ramucirumab + paclitaxel more than other regimens?

In KEYNOTE-061, 395 patients were randomised to pembrolizumab (N= 196) or paclitaxel (N= 199). After disease progression, patients in the pembrolizumab arm received either no third-line therapy (N= 103), third-line ramucirumab + paclitaxel (N= 26) or third-line other treatment (N= 67) while patients in the paclitaxel arm received no third-line treatment (N= 82) or third-line therapy (N= 117).

Results

In total, 19% of patients whose disease progressed on second line pembrolizumab remained alive for at least 18 months (i.e. long-term OS). All these long-term survivors received some sort of subsequent third-line therapy. After disease progression on paclitaxel, 12% of patients were still alive at 18 months. In addition, more patients in the pembrolizumab arm who received no third-line treatment were still alive at 18 months (9%) compared to patients treated with paclitaxel without third-line treatment (3%). Patients in the pembrolizumab group without subsequent therapy did not have a longer median OS as compared to patients in the paclitaxel group who did not receive subsequent therapy (3.2 vs. 4.2 months, respectively). However, at 18-months 17.5% of patients in the pembrolizumab group were still alive, as compared to 7.3% of the patients treated with paclitaxel. For patients who did receive third-line treatment, baseline characteristics were well balanced between the pembrolizumab and paclitaxel groups. Taxanes were the most commonly given third-line therapy in the pembrolizumab arm while patients in the paclitaxel arm most frequently received irinotecan in third line.

Results from the first analysis indicated that second line pembrolizumab followed by any subsequent therapy was associated with a longer OS than second-line paclitaxel followed by any subsequent therapy (13.5 vs. 10.1 months, respectively). Furthermore, second-line pembrolizumab followed by ramucirumab + paclitaxel was associated with a longer OS than second-line paclitaxel followed by any subsequent therapy (13.1 vs. 10.1 months). Finally, second-line pembrolizumab followed by ramucirumab + paclitaxel was not associated with longer OS from randomisation than second-line pembrolizumab followed by any other subsequent therapy (13.1 vs. 14.7 months, respectively). The median OS from the start of subsequent therapy for patients treated with pembrolizumab followed by ramucirumab + paclitaxel was 9.0 months as compared to 8.0 months and 6.0 months for patients who received pembrolizumab followed by any other therapy and patients on paclitaxel followed by any subsequent therapy, respectively. Limitations of the study include the small subgroup sizes, the fact that this was an exploratory analysis with no adjustments for multiplicity and unknown factors for whether or which subsequent therapy was given. In addition, the comparator arm in the KEYNOTE-061 trial was not ramucirumab + paclitaxel and other efficacy outcomes such as objective response rates, duration of response and PFS were not available for patients who received third-line therapy.

Conclusions

These data suggest that pembrolizumab may potentiate subsequent therapy and indicate that a long-term overall survival (>18 months) is possible after disease progression on pembrolizumab monotherapy. Based on these limited data, there may be potential for a greater antitumour effect when pembrolizumab is sequentially combined with anti-VEGF/VEGFR and a taxane. However, this hypothesis warrants further examinations as these results should be interpreted with caution and are only hypothesis-generating.

Reference

Yoon HH, Fuchs CS, Özgüroglu M, et al. KEYNOTE-061: Response to Subsequent Therapy Following Second-Line Pembrolizumab or Paclitaxel in Patients With Advanced Gastric or Gastroesophageal Junction Adenocarcinoma. Presented at ESMO World GI 2020; Abstract oral-12.