Lung cancer is the predominant form of cancer globally and the primary cause of mortality connected to cancer, resulting in about 1.7 million fatalities annually (1). About 80%-85% of lung cancers are pathologically classified as non-small cell lung cancer (2). Currently, surgery is the main approach used to treat early-stage non-small cell lung cancer. Nevertheless, a significant proportion of individuals experience the formation of either local or distant metastases (3). The therapeutic advancement of immune checkpoint blockade has significantly transformed the approach to treating and predicting outcomes for individuals diagnosed with NSCLC (4). Tumor cells possess many methods to resist immune system attacks, including the expression of immunosuppressive molecules on their cell surface, secretion of immunosuppressive substances, and recruitment of other immune cell populations with suppressive properties (5). Specific inhibitors against checkpoint receptors can block this immunosuppression, thereby increasing the specific immune response of T lymphocytes and eliciting an antitumor response (6, 7). Pembrolizumab is a humanized monoclonal antibody targeting programmed death 1 (PD-1), which has been shown to have antitumor activity in advanced non-small cell lung cancer (NSCLC). Single-agent pembrolizumab as first-line therapy is approved for tumors with high expression of PD-L1 (≥50%) while immunotherapy and chemotherapy are approved for any PD-L1 (8).

Pembrolizumab has demonstrated encouraging outcomes in recent clinical studies, particularly in cases where PDL1 staining is equal to or greater than 50% of tumor cells (9). Clinical trials and meta-analyses have shown that this treatment regimen can, in some cases, significantly improve patients’ overall survival and progression-free survival, while having a low toxicity profile (10). Additionally, the utilization of pembrolizumab in conjunction with chemotherapy has garnered considerable interest. The benefit of this combo treatment is its ability to achieve a wider range of effectiveness in patients with limited PD-L1 expression. Combining multiple therapies may offer longer-lasting disease management and improved survival advantages compared to using a single medication. Nevertheless, it is crucial to consider the potential toxicities and medication resistance associated with it (11). The KEYNOTE-042 study was an open label phase II-III randomized trial comparing pembrolizumab monotherapy with chemotherapy in the treatment of advanced NSCLC. The findings indicated that pembrolizumab outperformed chemotherapy in terms of overall survival and progression-free survival in the overall population. Moreover, there were more substantial enhancements in OS and PFS specifically for the subset of tumors with PDL1≥50% (12). The KEYNOTE-189 study was a randomized phase III trial that evaluated the efficacy of pembrolizumab in combination with chemotherapy compared to chemotherapy alone for the treatment of advanced NSCLC. The findings demonstrated that the combination of pembrolizumab and chemotherapy outperformed chemotherapy alone in terms of OS and PFS in both the overall population and the subgroup of tumors with a PDL1≥50% expression (13).

However, there is currently a lack of clinical trials of chemotherapy combined with pembrolizumab versus pembrolizumab monotherapy to determine whether chemotherapy combined with pembrolizumab has a higher benefit than pembrolizumab monotherapy in metastatic non-small cell carcinoma. Network meta-analysis enables the comparison of treatment arms in randomized controlled trials (RCTs) by utilizing relative and absolute measures of treatment efficacy and common treatment arms (14–16). Prior network meta-analyses (17–19) had shown the effectiveness and safety of combining pembrolizumab with chemotherapy compared to using pembrolizumab alone. However, since the original articles only presented short-term outcomes, they failed to provide information on the long-term outcomes of RCTs. Over the course of the last three years, multiple randomized controlled trials (RCTs) have updated their long-term outcomes (13, 20–23). Hence, it is both possible and essential to perform a meta-analysis that compares the long-term outcomes of pembrolizumab plus chemotherapy versus pembrolizumab monotherapy as the initial treatment for metastatic non-small-cell lung cancer.

The aim of our study was to indirectly compare the long-term outcomes of pembrolizumab in combination with chemotherapy versus pembrolizumab monotherapy as first-line therapy. We additionally assessed the disparities in survival rates among patients with tumors exhibiting PD-L1 expression ranging from 1% to 49%.

Activated T cells, B cells, natural killer cells, monocytes, and dendritic cells express programmed cell death receptor-1 (PD-1), a type I transmembrane glycoprotein belonging to the Ig superfamily (26). As an important immune checkpoint proteins, PD-1 interacts with two ligands, PD-L1 (B7-H1) and PD-L2(B7-H2), respectively. Immune and epithelial cells inductively express PD-L1, while antigen-presenting cells express PD-L2. In a physiological sense, PD-1 prevents immune system dysregulation by interacting with antigen-presenting cell surface PD-L1 and PD-L2. By overexpressing PD-L1, tumor cells encourage PD-1 binding to surface-expressed PD-L1 molecules, which in turn impairs immune surveillance of T cells, making it more difficult for tumor cells to be recognized and killed, and encouraging tumor immune escape (27). By disrupting interactions of PD-1/PD-L1, tumor immune tolerance can be broken, tumor specific T cells can regain their killing ability, and tumor clearance can be achieved by PD-1/PD-L1 monoclonal antibodies (28). Pembrolizumab, a PD-L1 inhibitor, has received approval for the treatment of NSCLC due to its notable clinical efficacy (29). The KEYNOTE-189 study was a randomized phase III trial that assessed the effectiveness of pembrolizumab in conjunction with chemotherapy in comparison to chemotherapy alone for treating advanced NSCLC. The results showed that the combination of pembrolizumab plus chemotherapy was more effective than chemotherapy alone in terms of OS and PFS in both the entire study population and the subset of tumors with a PDL1≥50% expression (13).

The results of our study offer robust evidence-based recommendations about the long-term prognosis for choosing between pembrolizumab monotherapy or pembrolizumab in combination with chemotherapy in clinical practice. Notably, the IPDformKM software was used to reconstruct Kaplan-Meier curves for OS and PFS, providing a clear and intuitive representation of oncological outcomes. Our results indicate that in patients with advanced NSCLC who have high levels of PD-L1 expression, the combination of pembrolizumab and chemotherapy is more efficacious than pembrolizumab alone as the first-line therapy in terms of PFS (median survival time: 10.41 months versus 7.41 months) and ORR (RR:1.74). However, there was no statistically significant distinction between the two groups in terms of OS. The notable enhancement in terms of ORR and PFS might be attributed to the stimulation of neoantigen release induced by chemotherapy, as well as the synergistic impact of immunotherapy and chemotherapy. However, the chemotherapy regimens are generally maintained for only three to four months(every 3 weeks, 4 to 6 cycles), while the pembrolizumab regimens are generally maintained for two years (every 3 weeks, 35 cycles). Over time, the residual effects of chemotherapy will progressively diminish, leaving only the lingering effects of pembrolizumab. In addition, pembrolizumab in combination with chemotherapy offers a more favorable long-term survival benefit in relation to OS (median survival time: 20.88 months versus 13.60 months) when compared to pembrolizumab monotherapy in patients with PD-L1 expression levels ranging from 1% to 49%. Since the data was not available, we could not provide the progression-free survival data among the PD-L1 TPS 1-49%. Regarding safety, there was no statistically significant difference between the two groups in terms of any irAEs or Grade≥ 3 irAEs. However, the combination of chemotherapy and pembrolizumab was associated with a higher probability of experiencing any trAEs and Grade≥ 3 trAEs compared to using pembrolizumab alone, suggesting that the enhanced effectiveness of pembrolizumab plus chemotherapy came with the drawback of increased adverse reactions.

Chemotherapeutics possess the capacity to enhance the immune system’s capability to identify and react to malignancies, or they can eliminate cells that inhibit the immune system. Furthermore, they have the potential to alter certain elements of the tumor microenvironment (30). It is crucial for us to distinguish these effects as we progress since chemotherapeutics have the ability to postpone the development of drug resistance, which could potentially change the chances of survival. Our meta-analysis consistently confirms that combining chemotherapy with first-line immune checkpoint medicines, such as pembrolizumab, enhances the efficacy of treatment for patients with advanced NSCLC. This phenomenon can be partially elucidated by the synergistic impact of immunotherapies and the induction of neoantigen release prompted by chemotherapy (31). There is variability among immune checkpoint inhibitors. Nivolumab and Pembrolizumab are immunosuppressive medications that inhibit the PD-1 protein, while Durvalumab and Atezolizumab were specifically engineered to target the PD-1, PD-L1 ligand (32). This approach has the potential to be used more extensively in order to reduce the impact on particular subgroups of NSCLC. However, further research is required. Genetic alterations of PD-L1 have been observed, which often result in the over-expression of PD-L1 (33). Xianhuo Wang et al. discovered that the genetic mutations of PD-L1 were specifically positioned in the exons of PD-L1 (34). These changes could impact the function of immunoglobulins and the transmembrane action of PD-L1, thus altering the immune response against tumors. Multiple studies have indicated that the expression of PD-L1, a protein associated with lung and other solid tumors, can be modified following treatment with platinum-based chemotherapy or concomitant chemoradiation (35). Fujimoto et al. conducted a study where they found that the expression of PD-L1 dropped dramatically after concurrent chemoradiation therapy in patients with locally advanced non-small lung cancer. This decrease in PD-L1 expression was linked to a positive prognosis (36). Toshiaki Takahashi et al. observed a considerable decrease in PD-L1 expression after treatment with pembrolizumab (37). This connection could potentially be one of the contributing factors to resistance against ICI and warrants additional exploration in extensive investigations.

An major strength of this study is that it is the first network meta-analysis to compare the long-term outcomes of pembrolizumab plus chemotherapy versus pembrolizumab monotherapy as first-line therapy for metastatic non-small-cell lung cancer. Previous network meta-analyses (17–19) exclusively reported outcomes that were limited to the short-term. Our findings support the existing scientific evidence about the long-term prognosis for pembrolizumab plus chemotherapy as first-line therapy for metastatic non-small-cell lung cancer. Besides, the Kaplan-Meier curves for OS and PFS were recreated to allow for a clear and comprehensible representation of the oncological outcomes.

This network meta-analysis possesses inherent limitations. The IPDformKM software was used to obtain reconstructed individual patient data from published KM curves of different quality. The quality of the analysis may be impacted. it is important to exercise caution when interpreting the results due to the potential for errors in reconstructing individual data. However, earlier research has demonstrated that HR obtained from rebuilt data has exhibited superior accuracy compared to published HR (38). Inconsistent baseline characteristics of patients in different clinical trials, such as doses and schedules of chemotherapeutic regimens, PD-L1 expression, gender, ECOG PS, smoking status, histology, metastases, and neoadjuvant therapy, may lead to heterogenicity in term of efficacy assessment and long-term survival assessment. Chemotherapy alone showed similar treatment effects in the five RCTs in terms of median PFS and OS ( Supplementary Material 2 ), implying the that the types of chemotherapeutic agent would not lead to no obvious impact. By stratifying patients based on PD-L1 TPS, the heterogeneity caused by PD-L1 expression was minimized, leading to improved reliability in pooling the results. Unfortunately, this meta-analysis did not have access to data on individual patients, which means that it was not possible to conduct subgroup analysis based on other inconsistent baseline characteristics. To address these constraints, it is imperative to conduct controlled randomized trials to directly assess the effectiveness of pembrolizumab combination chemotherapy in comparison to pembrolizumab monotherapy.

In conclusion, the present network meta-analysis reported comparative long-term outcomes of pembrolizumab plus chemotherapy versus pembrolizumab monotherapy as first-line therapy for metastatic non-small-cell lung cancer. Pembrolizumab plus chemotherapy resulted in enhanced PFS and ORR among patients with advanced NSCLC who had a PD-L1 expression level of 50% or above. Nevertheless, there was no discernible advantage in terms of OS when pembrolizumab was combined with chemotherapy in comparison to using pembrolizumab alone. Furthermore, pembrolizumab plus chemotherapy provided a more advantageous long-term survival advantage in terms of OS compared to using pembrolizumab alone in patients with PD-L1 expression levels ranging from 1% to 49%. In addition, the heightened efficacy of pembrolizumab in combination with chemotherapy was accompanied by a rise in undesirable side effects.

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

SH: Conceptualization, Data curation, Investigation, Software, Writing – original draft. ZH: Conceptualization, Data curation, Investigation, Software, Writing – original draft. XH: Data curation, Formal analysis, Writing – original draft. RL: Data curation, Formal analysis, Writing – original draft. WL: Funding acquisition, Methodology, Writing – review & editing. TQ: Funding acquisition, Methodology, Writing – review & editing.