Immunotherapy is becoming increasingly important in the treatment of metastatic HER2-negative and HER2-positive gastric cancer [11]. To our knowledge, this NIVOFGFR2 study is the first to include patients with metastatic gastric cancer co-expressing FGFR2 and PD-L1 who received first-line nivolumab in combination with chemotherapy. Notably, about 50% of participants exhibited strong expression of both FGFR2 (3+) and PD-L1 (CPS ≥ 20). FGFR2 expression and amplification are associated with worse outcomes in patients with gastric cancer. This has been observed in various studies where higher levels of FGFR2 correlate with more aggressive disease and poorer prognosis [12–14]. Understanding the impact of FGFR2 on treatment outcomes is crucial, as it may influence treatment decisions and the development of targeted therapies.
Regarding prognostic factors in our study, patient characteristics were similar to those observed in real-world populations as well as in other studies [15, 16]. The majority of patients had an ECOG PS of 1, a primary tumor site in the stomach, metastases in two or more organs, and no history of surgical treatment, mirroring the patient demographics of the CheckMate 649 and ATTRACTION-4 studies [10, 17]. This similarity in patient characteristics underscores the relevance of our preliminary findings to broader clinical practice. The alignment with real-world and randomized study populations ensures that our results are applicable and can be interpreted. This is important for validating the generalizability of our outcomes and for understanding how our findings fit into the broader landscape of metastatic gastric cancer treatment.
Despite the promising rationale for combining nivolumab with chemotherapy in this particular subgroup of gastric cancer patients with FGFR2 and PD-L1 expression, the results of our study were not as anticipated, and the primary endpoint of 1-year PFS rate was not met. Only a third of patients remained progression-free at 12 months, with a median PFS of 6.0 months. These outcomes are inferior to those reported for the nivolumab plus chemotherapy group in the randomized CheckMate 649 trial, where better PFS was observed [10]. The median PFS data in our study are more comparable to the chemotherapy alone group from CheckMate 649, which also had a median PFS of 6.0 months.
However, OS results in NIVOFGFR2 study were very similar to the figures demonstrated in the nivolumab group with CPS ≥ 5 in the CheckMate 649. In our cohort, the median OS was 15.1 months with a 95% CI of 13.2 to 16.8 months. This is comparable to the phase 3 study, where the median OS was 14.4 months with a 95% CI of 13.1 to 16.2 months.
On the one hand, these findings suggest that despite the lack of improvement in PFS, nivolumab combined with chemotherapy may still confer a survival benefit for patients with PD-L1 CPS ≥ 5. The impact of nivolumab on OS, even after disease progression or discontinuation, has been repeatedly described in other studies across various tumor types [18–21]. Thus, the alignment in OS outcomes highlights the potential role of nivolumab in extending survival in gastric cancer as well, even if the immediate control of disease progression, as measured by PFS, is not markedly improved.
On the other hand, the role of subsequent treatment cannot be excluded. In our study, all patients who experienced disease progression on the first-line therapy received ramucirumab in combination with paclitaxel as a second-line, and 40% received third- and subsequent lines of therapies. These additional treatments may have mitigated the shortcomings of the first-line and positively influenced the OS outcome [22, 23]. The potential influence of subsequent therapies underscores the complexity of assessing the true impact of the first-line treatment regimen. It is possible that the observed OS benefit may be partially attributable to the efficacy of second-line and later treatments. This highlights the need for a holistic approach to treatment planning that considers the entire continuum of care for patients with metastatic gastric cancer.
As for the response rate in the NIVOFGFR2 study, it was almost three times lower than in the CheckMate 649 and ATTRACTION-4 studies [2, 17], while the disease control rate was nearly the same across all studies. Despite the lower response rate, the development of a response to nivolumab in combination with chemotherapy was statistically significantly correlated with an increase in median PFS, which was twice as high in responders compared to non-responders. This finding highlights the importance of achieving a response to treatment in improving PFS outcomes. The comparable disease control rates to other studies indicate that even though fewer patients achieved a complete or partial response in NIVOFGFR2, many still maintained stable disease, contributing to overall disease management and clinical benefits.
Finally, the safety profile of the treatment was satisfactory. Despite a high rate of all TRAEs, grade ≥ 3 adverse events were reported in only a third of patients and were primarily associated with chemotherapy rather than nivolumab. Importantly, no patients discontinued treatment due to TRAEs, and no treatment-related deaths occurred.
In conclusion, while the primary endpoint of improved 1-year PFS rate was not met, our study provides valuable insights into the potential benefits of combining nivolumab with chemotherapy in patients with metastatic gastric cancer co-expressing FGFR2 and PD-L1. The OS results were comparable to other significant studies, and the safety profile was manageable. Future research should focus on optimizing patient selection, assessing the potential of combined immunotherapy and targeted anti-FGFR2 therapy, and further investigating the role of subsequent treatments to maximize the therapeutic benefits.