One of the aims of this study was to analyze the efficacies of different monoclonal anti-PD-1 antibodies in northern China. In a previous study, a comparison of the amino acid sequences of nivolumab and pembrolizumab showed that they are basically the same except for the variable region [8]. In addition, Fesses et al. compared the molecular, preclinical, and early clinical characteristics of nivolumab and pembrolizumab. They found significant molecular similarities between these drugs, indicating that the differences observed in the clinical data were not likely to be drug-dependent and likely to be drug-independent. The observed differences may be due to variations between the participant populations in the clinical trials. In our study, toripalimab was found to have a higher ORR and longer PFS than the other 2 PD-1 agents, especially at the 3-month evaluation.
The PD-L1 of tumor cells is currently the most widely studied protein and the only Federal Drug Agency–approved biomarker for clinical use. According to the KEYNOTE-042 study results published at the 2018 American Society of Clinical Oncology conference, for advanced NSCLC patients with PD-L1 ≥ 1% and negative for EGFR or ALK mutations, the overall survival(OS) of those receiving first-line pembrolizumab therapy was higher than that of those receiving chemotherapy. This finding showed that pembrolizumab is more beneficial for NSCLC patients with PD-L1 ≥ 1% [9]. The TMB is the sum of non-synonymous mutations in the coding region of the somatic genome. Non-synonymous somatic mutations can alter the amino acid sequence of the protein encoded by the affected gene, thereby forming a neoantigen and helping to enhance the immunogenicity of tumor cells [10]. There are various methods to calculate the TMB, and they generally require whole-exome sequencing of ≥ 200 cancer-related genes [11]. The TMB was first reported as a biomarker in the first edition of the 2019 National Comprehensive Cancer Network guide for screening NSCLC patients for whom nivolumab combined with ipilimumab or nivolumab mono-therapy may be effective. In the present study, higher TMB and PD-L1 expression levels were also demonstrated to benefit PD-1 blockade therapy. We found that the TMB influenced the ORR significantly after 6 months of therapy and that the PD-L1 expression influenced the ORR significantly after 3 months of therapy.
The correlation between mutations of cancer driver genes and the efficacy of PD-1 monoclonal antibody therapy has also attracted attention. An EGFR mutation in adenocarcinoma lung cancer is a good predictive factor for the efficacy of EGFR tyrosine kinase inhibitors [12–14]. In the KEYNOTE-001 study, the median OS of EGFR mutation–positive patients treated with pembrolizumab was significantly shorter than that in wild-type EGFR patients [15]. Gainor et al. also found that compared with wild-type EFGR NSCLC patients, EGFR mutation–positive NSCLC patients had a much lower ORR [16]. However, our study found that a non-classical EGFR mutation, such as G719C or E709V, did not significantly influence the efficacy of PD-1 therapy for NSCLC patients in northern China. According to the results of the functional enrichment analysis of co-mutated genes in patients with an EGFR mutation, the commonly enriched terms were “proteoglycans in cancer,” “response to osmotic stress,” “response to oxidative stress,” and “myeloid leukocyte activation” (Fig. 8B). And more the co-mutations, better the prognosis (Fig. 8A). These pathways associated activation of cell signaling pathways of proliferation, angiogenesis, and cell motility (Fig. 8C). The most enriched signal pathway of these mutations is "proteoglycans in cancer". Proteoglycans (PGs) are heterogeneous glycoproteins, expressed in cells of the tumor microenvironment and on tumor cells, which could regulation immunosensitization [17]. “Response to osmotic stress”, “response to oxidative stress” and “myeloid leukocyte activation” pathways were also significantly enrichment herein. Both osmotic stress and oxidative stress are associated with inflammation. And inflammation was associated with immune activation [18]. Myeloid leukocyte activation may activate B cells [19] and improve the tumor microenvironment to make it sensitive to PD-1 agents [20]. Take above all, these gene characteristic maybe influence the patients response to PD-1 therapy. As we all known the expression of CD47 is involved in imparting resistance to programmed cell death (PD)-1/PD-ligand 1 inhibitors [21]. According to our study, the EGFR-mutated patients who were achieved PR after 6 m-therapy showed a lower CD47 expression. But the patients with disease progression exhibited a higher CD47 expression. Maybe CD47 expression is the potential biomarker to patients with EGFR mutation treated with PD-1 agents.
Another important driver gene of NSCLC is ALK. The ATLANTIC study showed that for patients with ALK rearrangement, if the PD-L1–positive cell expression rate is ≥ 25%, immunotherapy will still work [22]. Moreover, it has been reported that NSCLC patients with ALK gene fusion might benefit from PD-1 treatment [22–24]. However, according to our clinical observations, those with ALK gene fusion showed a poor response to PD-1 therapy. Previous studies have indicated that a high TMB and PD-L1 expression level were both independently correlated with survival benefits from anti-PD-1 therapy. In a recent study, researchers also observed that the median OS was longer in the TP53-mutated group than in the wild-type TP53 group (18.1 months vs. 8.1 months). Furthermore, the median PFS was significantly longer and the ORR was higher in the TP53-mutated patients [24]. In accordance with these findings, we also found that a TP53 mutation was beneficial for PD-1 therapy.
This study had several limitations. The sample size of the group of participants treated with PD-1 antibodies in our study was small, especially when classified according to the therapy. Additionally, the OS was not determined for most of the participants, and the efficacy of PD-1 therapy was evaluated using only the ORR and the PFS. It can take years before an effective improvement or death occurs to enable further assessment of efficacy. When available, the final set of data, including the effectiveness and late toxicity endpoints, will provide valuable insight into the differences in survival and quality of life of the NSCLC patients.