In this study, Pucotenlimab achieved an ORR of 20.17% (24/119, 95% CI, 13.370%-28.506%) in ≥ second line setting, which was slightly higher than the other anti-PD-1 products approved for advanced melanoma in a similar setting, among them were Pembrolizumab with an ORR of 16.7% (95% CI, 10.0%-25.3%) and Toripalimab with an ORR of 17.3% (95% CI, 11.2–25.0%) (13, 14). Unlike melanoma patients in the white population, which the cutaneous subtype dominates, the non-white patient population is mainly composed of acral or mucosal subtypes. Unfortunately, mucosal melanoma was usually refractory to treatment with a poor prognosis (15). Compared with Toripalimab for mucosal melanoma with an ORR of 0% (13, 14), some patients with mucosal melanoma could benefit from Pucotelimab, as the ORR for this subgroup in this study was 8.7%.
The patients in the current study achieved a median OS of 16.59 (95% CI, 13.963–26.973) months, which was significantly longer than those of the conventional therapies. According to the follow-up study after Pucotenlimab failure, merely 2 out of 55 patients achieved SD, and none achieved CR or PR from the follow-up treatments. Therefore, the longer mOS was contributed by Pucotenlimab to a great extent instead of the follow-up treatments.
In our study, the cutaneous subtype achieved the highest ORR among different known primary subtypes, followed by the acral, and then the mucosal subtype. The same patterns were observed in mPFS and mOS. Thus, ORR and PFS might be appropriate surrogate endpoints for OS, the gold-standard endpoint for efficacy in a similar setting.
ADA post Pucotenlimab treatment was detected in 4 out of 119 (3.36%). No ADA-related AE or aggravation was observed in these ADA-positive patients. Two of them achieved SD as the best response to Pucotenlimab (data was not showed). shownthe influence of ADAs in the current study was minimal.
According to James C. Lee, et al., anti-P-1/PD-L1 therapy demonstrated low effectiveness for various cancers with liver metastasis including melanoma. The underlying mechanisms were CD8 + T cell depletion in liver and distal immunosuppression induced by regulatory T-cell activation (16). Pucotenlimab also showed similar pattern. The ORR for the melanoma patients with liver metastasis was 6.67% (95% CI, 0.169%-31.948%), while for those without was 22.12% (95% CI, 14.566%-31.313%).
It was acknowledged that BRAF mutation did not affect the benefit melanoma patients got from anti-PD-1 treatment (17). Our study also indicated that quite a proportion of patients (5/13, 38.46%) with BRAF mutation achieved CR/PR after Pucotenlimab treatment, which was comparable to that previously reported.
In our study, the rates of TRAEs of any Grade and of Grade ≥ 3 were 77.3% and 15.1%, respectively, which were similar to those reported in Pembrolizumab of 84.5% and 8.7%, and Toripalimab of 90.6% and 19.6% in the same setting. Besides, no new types of TRAEs appeared in our study compared with Pembrolizumab and Toripalimab during long-term safety follow-up (13, 14). Therefore, the safety profile of HX008 was consistent with other anti-PD-1 antibodies.
In the current study, serum Eotaxin level was positively correlated with PR status in patients treated with Pucotenlimab, which supported the efficacy-predicting role of Eotaxin and its effector cells eosinophil for anti-PD-1 treatment in various types of cancer including melanoma (18–22).
A variety of cancers (lung, breast, liver cancer) highly expressed CCL2, and usually it was associated with poor prognosis. Besides, Megan M. Tu, et al showed that the efficacy of anti-PD-1 therapy was boosted by inhibiting CCL2 receptor, CCR2 (23). Generally, CCL2 promotes tumor-growth through process, like inducing angiogenesis, recruiting MDSCs and metastasis-promoting monocytes, while it also has anti-tumor capability by driving tumor cell apoptosis (23, 24). Interestingly, Tianqian Zhang, et al revealed that CCL2 secreted by melanoma cells in 3-dimensional organoids as a Chemoattractant induced migration of cytotoxic T lymphocytes by CCR4 towards melanoma cells (25). The prognostic role of CCL2 in melanoma has not been revealed yet. In the current study, it seemed that higher serum CCL2 level indicated a good outcome of Pucotenlimab treatment, which contradicted with the results in other cancers mentioned above. Thus, the evidence provided by Tianqian Zhang, et al. may support our opinion that CCL2 functioned as a good prognostic marker for PD-1 blockade in melanoma.
Previous studies showed that TGF-α was upregulated in many types of cancers including melanoma. Its higher serum level was associated with poorer prognosis. The underlying mechanism was its pro-differentiation capability for melanoma cancer cells (26). Our study further supported this conclusion, as higher serum levels of TGF-α were detected when the patients were experiencing progressed disease.
Serum VEGF level has been suggested as a poor prognostic marker for ipilimumab for the treatment of advanced melanoma, but not for PD-1 inhibitors or in combination with ipilimumab (27). Our study seemed to support VEGF serum levels as a poor prognostic factor for anti-PD-1 in the treatment of advanced melanoma.
Due to the limited sample size in the current study, the conclusions drawn above may be inaccurate, which needs further investigation in future phase III study that has been approved by the Chinese National Medical Products Administration (NMPA) with a large sample size involved.