This study explored the combination of sintilimab plus apatinib and chemotherapy. With an ORR reaching 53.6% and a long median PFS (8.5 months), this prospective trial met the primary endpoint. The median OS time was > 1 year, reaching 12.5 months. These results indicate a significant improvement over previous second-line treatments, suggesting that the combination of a PD-1 inhibitor, VEGFR inhibitor, and chemotherapy has a certain synergistic effect.
For the first-line treatment of advanced GC, the anti-PD-1 combination therapy has been listed in the latest treatment guidelines. The CheckMate 649 study showed that nivolumab combined with chemotherapy improved the OS compared with chemotherapy alone33. Sintilimab plus chemotherapy provided a new standard first-line treatment option for Chinese patients with gastric or GEJ cancer, demonstrating superior OS and PFS with an acceptable safety profile according to the ORIENT-16 study. However, there are relatively few studies on second-line therapy for advanced GC.
Previous studies have also shown a synergistic effect between chemotherapy and apatinib34. In a study by Xu et al.13, apatinib improved the efficacy of paclitaxel and 5-fluorouracil both in vitro and in vivo. In addition, apatinib has immunomodulatory effects mediated by decreases in regulatory T cells and myeloid-derived suppressor cells as well as enhanced dendritic cell maturation and effector T-cell infiltration. Liang et al.31 found that the combination of anti-PD-1 antibody camrelizumab and apatinib for advanced solid tumors was more effective than immunotherapy alone (ORR, 34.2% vs 18.8%; median PFS, 6.0 vs 4.5 months, P = 0.002). Xu et al.32 explored the safety and tolerability of apatinib at different doses (125–500 mg) in combination with camrelizumab in patients with advanced GC and advanced liver cancer. The phase Ia-identified recommended phase II dose was 250 mg, and 250 mg of apatinib combined with 200 mg of camrelizumab was used for a expansion study. The results showed that safety was controllable. On this basis, we conducted this exploratory study of anti-PD-1 antibody plus apatinib and chemotherapy in patients with advanced GC.
Our study of the combination of sintilimab plus apatinib and chemotherapy (irinotecan or paclitaxel) in an unselected population showed an ORR of 53.6%, median PFS of 8.5 months, and median OS of 12.5 months. These results are better than those of ramucirumab combined with paclitaxel in the RAINBOW study (ORR of 28%, median PFS of 4.4 months, and median OS of 9.6 months)11, paclitaxel or pembrolizumab in the KEYNOTE-061 study (ORR of 16% vs 14%, median PFS of 1.5 vs 4.1 months, and median OS of 9.1 vs 8.3 months)35, and apatinib combined with camrelizumab (ORR of 17.4%, median PFS of 2.9 months, and median OS of 11.4 months)32. In a phase I/II study of nivolumab combined with paclitaxel plus ramucirumab as second-line treatment in 43 patients with advanced GC, the ORR was 37.2%, the median PFS was 5.1 months, and the median OS was 13.1 months36. The results are similar to our study, suggesting that the combination of the three drugs is a treatment option worthy of further verification and has promising anti-tumor activity.
Although recent studies have indicated that PD-L1 levels are related to the efficacy of PD-1 inhibitors, the CheckMate 649 study showed that nivolumab combined with chemotherapy improved the OS, especially in patients with a Combined Positive Score (CPS) of ≥ 533. In the unselected population from the ATTRACTION-4 study, nivolumab combined with chemotherapy failed to improve OS compared with placebo combined with chemotherapy, and advantages only existed in terms of the ORR and PFS37. In addition, pembrolizumab plus chemotherapy in the KEYNOTE-062 study did not result in a significant improvement in efficacy, regardless of the CPS38. Therefore, CPS is not the only predictor of the efficacy of PD-1 antibody, especially in the mode of combination therapy. Other relevant influencing factors need to be further explored.
The univariate Cox regression analysis of this study showed that the Lauren classification and liver metastasis were potential prognostic factors affecting both PFS and OS, whereas the multivariate analysis showed that liver metastasis was an independent prognostic factor for PFS and OS. Patients with liver metastasis had significantly better PFS and OS and significantly better ORR than those with non-liver metastasis. This result suggests that patients with liver metastasis are more likely to benefit from the treatment regimen in this study. However, Tumeh et al.39 found that patients with liver metastasis of melanoma or non-small cell lung cancer who were treated with pembrolizumab showed reduced response and PFS and that liver metastasis was associated with reduced marginal CD8 + T-cell infiltration, providing a potential mechanism for the worse prognosis. Akinori et al.40 also found that liver metastasis, ECOG performance status of 1 or 2, and a large sum of target lesion diameters at baseline were significantly associated with hyper-progressive disease in 62 patients with advanced GC treated with nivolumab. The results of the two above-mentioned studies are seem to be contradictory to those in our studies. The reasons for this discrepancy may be as follows. First, there may have been differences among the tumor types. Furthermore, the regimen in this study involved immunotherapy combined with anti-angiogenesis and chemotherapy, but the regimens in the above two studies involved immunotherapy monotherapy. The combination of three drugs may change the treatment outcome of liver metastasis. However, because of the small number of patients, the conclusions of this study need to be verified by expanding the sample size.
The ORR obtained by patients with intestinal type tumors according to the Lauren classification was significantly better than that of patients with diffuse type, mixed type, or unknown tumors, and there was also a significant difference in the median PFS (11.3 vs 4.1 months, P = 0.030). Although the median OS was prolonged, the difference did not reach statistical significance (not reached vs 6.2 months, P = 0.266). However, a retrospective study by Nie et al.24 showed that the efficacy of sintilimab in patients with intestinal subtype GC (ORR, 30.0%; DCR, 80.0%) was superior to that in patients with non-intestinal subtype GC (ORR, 6.3%; DCR, 56.3%); additionally, patients with the intestinal subtype obtained a longer PFS (4.0 vs 1.9 months) and OS (9.0 vs 4.1 months) than those with the non-intestinal subtype. These findings are basically consistent with the conclusions of our study, suggesting that patients with the intestinal type of GC according to the Lauren classification are more likely to benefit from immunotherapy. However, because of the small number of patients in the above studies and the present study, this conclusion requires further verification.
No unexpected adverse events occurred in this study. Most of the adverse events were grade 1 and 2. The most common grade 3–4 treatment-related adverse event was neutropenia, which was mainly caused by irinotecan. The most common adverse event caused by sintilimab was increased liver transaminases, followed by hypothyroidism; most of these cases were grade 1 and 2. The most common adverse events related to apatinib were proteinuria and hypertension, most of which were also grade 1 and 2. The overall tolerance of this treatment regimen was good, likely because of the reduced dose of chemotherapy drugs and apatinib used in this study compared with the standard dose when used as monotherapy.
This study has some limitations. Firstly, this was a single-arm, single-center study without a randomized control group; thus, it lacked sufficient persuasiveness. Secondly, patients enrolled in this study were HER-2-negative and had a proficient mismatch repair status; however, the lack of more molecular feature detections may bring some limits for further mechanism analysis, such as PD-L1 levels and so on. In addition, the number of patients was small, which may lead to some bias in the study results. A prospective expansion study of this regimen is underway.