Immune checkpoint inhibitors (ICIs), which interact with the anti-programmed death-1 (PD-1)/programmed death-ligand 1 (PD-L1) axis to reactivate the immune response, have emerged as a novel therapeutic option in patients with non-small cell lung cancer (NSCLC) (1). Even though ICIs therapy as a first-line and second-line treatment has been found to produce superior tumor remissions for NSCLC, a significant number of patients do not benefit from such approaches alone (2, 3). To address this issue, ICI-based combination therapies were developed after several clinical studies were undertaken to investigate the efficiency of the combination regimen of ICIs.

The phenomenon of tumor angiogenesis, which refers to the recruitment of endothelial cells to tumors, is considered to play a significant role in the development, growth, and metastasis of tumors (4, 5). In addition to accelerating disease development, aberrant tumor vasculature also creates an immune-suppressive environment via impeding antigen-specific T-cell maturation, inhibiting T-cell infiltration, and attracting myeloid-derived suppressor cells to the tumor sites (6). Accordingly, angiogenesis inhibition could remodel the tumor immune microenvironment and improve the efficacy of immunotherapy (7). Several scientific studies have demonstrated the synergy between immunotherapy and anti-angiogenic treatment (8–13). In particular, the IMPOWER-150 study has shown the medical benefits of atezolizumab combined with bevacizumab and chemotherapy in the first-line treatment of advanced NSCLC (14). It is essential not to underestimate the possibility of adverse effects associated with chemotherapy despite clinical outcomes improving. Therefore, the exploration of chemotherapy-free regimens in NSCLC is necessary. And some of the recent clinical trials for treating advanced NSCLC involve ICIs combined with anti-angiogenic therapy as an alternative to immunochemotherapeutic combinations (8, 10). This combined regimen yielded promising clinical outcomes with a favorable safety profile in several clinical trials. However, long-term follow-up studies with large samples validating the durability of this “chemotherapy-free” combination are quite limited. Furthermore, a more optimal combination needs to be explored regarding drug selection, tumor type, and risk-benefit assessment.

Recombinant human(rh)-endostatin, a multiple-target angiogenesis inhibitor, blocks proliferation and abnormal angiogenesis of tumor endothelial cells (15). Endostatin outperforms other anti-angiogenic medications in terms of safety, compliance, and a broad spectrum of indications covering all pathological types (16, 17). The China National Medical Products Administration authorized endostatin with chemotherapy as first-line treatment for advanced NSCLC based on a phase III study (18). As first-line therapy for NSCLC patients, endostatin plus camrelizumab and chemotherapy showed favorable efficacy and safety in a recent retrospective study (19). Moreover, an in vivo study found that endostatin, in combination with ICIs, inhibited tumor development in lung cancer mouse models with brain metastasis (20). A recent phase II trial showed the effectiveness and acceptable tolerance of nivolumab in combination with endostatin for previously treated NSCLC patients (10). However, studies comparing anti-angiogenic and chemotherapy have not been conducted to determine which has a more favorable effect in conjunction with immunotherapy. In this retrospective study, we evaluated the efficacy and safety of ICIs combined with endostatin or chemotherapy as a second-line treatment for advanced NSCLC to determine which combination was more effective.

The introduction of immunotherapy has altered the therapeutic paradigm for advanced NSCLC. Despite the fact that the FDA has officially approved ICIs alone as a second-line treatment for NSCLC, the ORR for unidentified persons was only 10-20% (21), highlighting a severe concern with ICIs monotherapy as a second-line treatment. Thus, concepts for ICIs combination treatment were starting to take shape. Several studies have shown that chemotherapy and anti-angiogenesis can increase the efficacy of immunotherapy, but it is still unclear which combination regimen is the most suitable or optimal strategy.

In our study, immunotherapy combined with endostatin outperformed immunotherapy combined with chemotherapy in patients with advanced, previously treated NSCLC. The median PFS in the IE group was 7.1 months (95% CI, 4.64-9.56) compared to 5.13 months (95% CI, 4.29-5.97) in the IC group. As a second-line therapy, the ORR and median PFS of IE treatment were superior to those of ICIs monotherapy from Checkmate017, Checkmate057, keynote001, and OAK Table 5 (22–25). As a consequence of the proportion of patients with squamous histology, aging, and poor ECOG scores enrolled in the study, ICIs plus endostatin exhibits a wide range of clinical indications. It is an efficient second-line treatment for advanced NSCLC.

Despite the absence of sizeable randomized control studies, the combination of anti-angiogenesis and immunotherapy was considered a potential strategy for NSCLC and showed favorable outcomes in several studies Table 6 . The satisfactory effectiveness of PFS and ORR in the IE group, as seen in our investigation, was roughly consistent with the treatment of Nivolumab with endostatin in a phase II study and pembrolizumab plus lenvatinib in the LEAP study (10, 12). The ORR of Nivolumab plus endostatin was marginally higher than the ORR of the IE group. And the reason for this discrepancy might be the different pathological composition between the two studies, with 64.7% of enrolled patients being adenocarcinoma, whereas squamous cell carcinoma (80%) was the predominant pathological type in the IE group. Previous research indicated that patients with squamous NSCLC (sq-NSCLC) had a significantly poorer prognosis than patients with non-squamous NSCLC due to the characteristics of sq-NSCLC (26). In several studies that used ICIs plus chemotherapy as first-line therapy, the survival time of sq-NSCLC was shorter compared with non-sq NSCLC (27–30).

Moreover, it was also clear that the different results were caused by different baseline pathological makeups in a retrospective study of ICIs with low-dose anlotinib (13). However, in the JVDF study, ramucirumab plus pembrolizumab showed a relatively higher median PFS than the IE group (11). The reason for this might be that the JVDF trial excluded patients with ECOG 2 or above, and Caucasians were the primary population in this study. With 32% Asian participants, the JVDJ study evaluated the efficacy of ramucirumab plus durvalumab for pretreated NSCLC. With an ORR of 11%, a median PFS of 2.7 months, and a median OS of 11 months, the results were unsatisfactory (31). This implied that the outcome of therapy could be influenced by racial differences. However, whether Asians had a lower efficacy of ICIs plus anti-angiogenesis may require further research.

Contrary to our findings, previous studies of pembrolizumab or nivolumab with docetaxel in pretreated advanced NSCLC showed promising outcomes and significant improvements than in the IC group (32, 33). And compared to the above two studies, the IE group had comparable outcomes but was superior in safety. Apart from the pathological difference, a younger study population with better performance may lead to better findings than our study. Above all, our study showed that ICIs plus endostatin was associated with clinical benefits independent of tumor histological type, age, and ECOG score.

A subgroup analysis based on baseline characteristics favored the IE treatment, except for patients with brain and pleural metastases. There were significant improvements in PFS for advanced NSCLC in the subgroups of sq-NSCLC, ECOG-2, and malignant pleural effusion (MPE) treated with ICIs plus endostatin (P = 0.02, P = 0.008, P = 0.037). It is indicated that Sq-NSCLC is more likely to invade large blood vessels and form tumor cavitation, which means that sq-NSCLCs are more likely to develop pulmonary hemorrhage (26). Because of such characteristics, anti-angiogenesis drugs such as bevacizumab were limited in squamous NSCLC. The fact remains that ICIs plus endostatin have outperformed ICIs plus chemotherapy in the sq-NSCLC subgroup not only by a wider margin but also by a more significant margin from a safety standpoint, regardless of the type of disease. ICIs plus bevacizumab did not result in an advantage for NSCLC patients compared with ICIs plus chemotherapy. In one retrospective study, ICIs plus bevacizumab failed to demonstrate an advantage for NSCLC patients over ICIs plus chemotherapy (34). This might be because endostatin has a wide range of inhibitory activities against angiogenesis, whereas bevacizumab appears to target a specialized area of angiogenesis (15, 16, 35). Interestingly, our subgroup analysis showed that patients with MPE assigned to the IE group had a better PFS than those assigned to the IC group. However, the effectiveness of ICIs plus endostatin in treating MPE requires further research.

The safety profile of the IE group was in accordance with previous research and was more favorable than that of the IC group. Additionally, there were no new adverse events attributable to either treatment group. The frequency of nonhematologic adverse events, including fatigue, rash, hepatic toxicity, and immune-mediated pneumonitis, was lower in the IE group than in the IC group, except for hypothyroidism and proteinuria. It is worth noting that patients in the IE group experience a higher frequency of adverse arrhythmia events. Therefore, the pros and cons of IE treatment for patients with cardiovascular disease should be carefully considered. In total, chemotherapy-free treatment reduced the incidence of these TAREs significantly, increasing patient compliance, reducing treatment-related deaths, and allowing patients to receive follow-up treatment after progression.

The research we conducted has several limitations. Firstly, our study was a retrospective observational study with limited sample size, so it was impossible to eliminate selection bias. Secondly, the immunotherapy with chemotherapy used in our study was not the standard second-line treatment for advanced NSCLC. As a result, we should include a monotherapy group to compare each combination pattern. Additionally, a portion of the patients in our study had previously received anti-angiogenesis therapy, so the relationship between prior anti-angiogenesis therapy and subsequent anti-angiogenesis is unclear. Further clinical trials will need to examine whether prior anti-angiogenesis therapy affects the efficacy of the subsequent immunotherapy plus angiogenesis therapy. Lastly, the patients with gene mutations were not included in the study, making it difficult to investigate the efficacy of immunotherapy combination therapy in these patients. It is also important to gather data on PD-1 expression, but the majority of the patients in our study have not undergone this test. Thus, the relationship between the ICIs combination method and the above two sides still needs to be explored.

In conclusion, the combination of PD-1 and rh-endostatin has shown encouraging survival benefits with a favorable safety profile as second-line therapy for advanced NSCLC patients. Although rh-endostatin was found to be more effective as an immunotherapy synergist than chemotherapy, more prospective randomized controlled studies are needed to corroborate these findings.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by the research ethics committee of the First Affiliated Hospital of Nanchang University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

LC and ZL contributed to the conception and design of the study. HH, PZ, and XZ collected the data and organized the database. PZ, SF, SL, and SP performed the statistical analysis. HH and PZ wrote the first draft of the manuscript. HH, PZ, and YL revised the manuscript critically for important content. HH, LC, and ZL completed the final review of the manuscript. All authors contributed to the manuscript revision, reading, and approving the submitted version.