Non-small cell lung cancer (NSCLC) is still a dominant cause of cancer-related deaths globally. Immune checkpoint inhibitors (ICIs), especially ICIs targeting PD-1/PD-L1 pathways, have revolutionized the treatment landscape for NSCLC, demonstrating significant clinical benefits and improving patient outcomes. Tislelizumab, a humanized anti-PD-1 monoclonal antibody, has shown promising efficacy and safety in various settings of lung cancer. Previous meta-analyses concerning tislelizumab in lung cancer have primarily focused on classical clinical endpoints such as overall survival (OS) and progression free survival (PFS). For instance, a systematic review and meta-analysis of randomized controlled trials (RCTs) evaluated tislelizumab monotherapy or in combination with chemotherapy for lung cancer, demonstrating significant improvements in both OS (HR: 0.72, 95% CI: 0.63-0.81) and PFS (HR: 0.61, 95% CI: 0.54-0.68) compared to control therapies (1). Another systematic review and network meta-analysis compared tislelizumab with other anti-PD-(L)1 agents in localized advanced or metastatic NSCLC, concluding that tislelizumab, with or without chemotherapy, was comparable or more favorable in terms of OS, PFS, and treatment-related adverse events (2). Similarly, a systematic review of clinical trials on tislelizumab’s safety and effectiveness in NSCLC also highlighted its association with improved PFS and objective response rate (ORR), particularly when combined with chemotherapy (3).

While OS and PFS are crucial long-term indicators of treatment efficacy, comprehensive response assessments, including complete response (CR) and pathological complete response (pCR), offer valuable insights into the immediate and deep tumor responses, especially in the context of neoadjuvant and resectable settings. CR, defined as the absence of detectable tumor, is a significant endpoint in clinical trials and can serve as a surrogate endpoint for accelerated or traditional approval, depending on the disease and context. pCR, defined as the absence of residual invasive and in situ cancer in resected specimens after neoadjuvant therapy, is a strong prognostic factor for improved long-term outcomes, including event-free survival (EFS) and OS in various cancers, including NSCLC. Studies have consistently demonstrated that patients achieving pCR after neoadjuvant treatment exhibit significantly better long-term survival rates in NSCLC. Specifically, pCR following neoadjuvant chemo- or chemoradiotherapy has been associated with marked improvements in recurrence and OS. Similarly, rewarding long-term results, including a 5-year OS of 56.18% and disease-free survival of 48.84%, have been observed in locally advanced NSCLC patients who achieve pCR after induction therapy followed by surgery (4). Furthermore, the crucial role of pCR has been highlighted as a surrogate endpoint for predicting and improving clinical outcomes, allowing for shorter duration clinical trials by providing an early indicator of treatment effectiveness (5). The association between neoadjuvant ICI-chemotherapy and meaningful improvement in 2-year EFS and pCR further underscores the importance of this endpoint as a strong surrogate for long-term survival in early-stage NSCLC (6). Thus, increasing the rate of pCR has become a critical endpoint in neoadjuvant trials, with the expectation of translating into improved survival outcomes.

Despite the growing body of evidence supporting the efficacy of tislelizumab in NSCLC, a systematic evaluation of its impact on CR and pCR, integrating data from both RCTs and real-world observational studies, remains an important gap in the literature. Therefore, this meta-analysis aims to specifically address this gap by comprehensively analyzing the available data on CR and pCR rates associated with tislelizumab-based therapies for NSCLC. This study will incorporate evidence from both randomized controlled trials and retrospective/prospective cohort studies to provide a robust assessment of these critical response endpoints.

232 records in total were initially identified through comprehensive database searching across PubMed, EMBASE, and Web of Science. An additional 43 records were found through other sources, bringing the preliminary total to 275 records. After the removal of 19 duplicate entries, 256 unique records remained for initial screening. All 256 records underwent title and abstract review, which led to the exclusion of 142 records. Consequently, 114 full-text articles were retrieved and subsequently assessed for eligibility. During this full-text review stage, a further 107 articles were excluded for various reasons: 7 were identified as meta-analyses or reviews, 86 lacked sufficient data for inclusion, 12 possessed an unsuitable study design, and 2 could not be formatted into a table. Following this rigorous screening process, a final count of 7 studies met the predefined eligibility criteria, which were ultimately included in this meta-analysis ( Figure 1 ).

This meta-analysis provides a comprehensive assessment of the efficacy and safety of tislelizumab for NSCLC, with a focus on pCR and CR rates. In line with the principles of evidence-based medicine, our findings have been systematically evaluated using the GRADE approach to determine the certainty of the evidence for each key outcome. This formal assessment revealed significant variations in the certainty of our findings. The evidence for pathological complete response (pCR) was rated as very low certainty, primarily due to a high risk of bias across the included studies, substantial heterogeneity, and imprecision stemming from wide confidence intervals that included the null effect. This very low certainty underscores the need for extreme caution when interpreting the pooled pCR data. In contrast, the evidence supporting a significant advantage for tislelizumab plus chemotherapy in achieving CR compared to chemotherapy alone was rated as low certainty. This outcome benefited from minimal heterogeneity and a precise effect estimate, but the overall rating was downgraded due to serious limitations in the study designs, including the prevalence of non-randomized and high-bias studies. Lastly, the evidence for the comparison of tislelizumab plus chemotherapy against other immunotherapy plus chemotherapy regimens for CR was also of very low certainty, a result of both serious study limitations and a wide, imprecise confidence interval that failed to demonstrate a statistically significant difference. These nuanced certainty ratings are critical for contextualizing our results and guiding the subsequent discussion of potential contributing factors and study limitations ( Table 2 ).

The meta-analysis on pCR rates for Tislelizumab plus chemotherapy in NSCLC revealed substantial heterogeneity, indicating considerable variability in treatment effects across the enrolled studies. Additionally, the lack of standardization in pathological assessment across studies, including variability in the number of sections examined, sampling thoroughness, and pathologist experience, may have contributed to the observed heterogeneity in pCR rates. Differences in pathological review protocols could introduce unmeasured variability in pCR determination, potentially influencing reported outcomes. Future studies should adhere to standardized pathological assessment guidelines, such as those proposed by the International Association for the Study of Lung Cancer (IASLC), to minimize such heterogeneity and improve comparability across trials. This high degree of heterogeneity necessitates a thorough discussion of potential contributing factors. One significant source of heterogeneity likely stems from differences in patient populations and disease characteristics, particularly the histological subtypes of NSCLC. While our meta-analysis provided a subgroup analysis for NSCLC (encompassing various histological types) versus SCC only, the broader NSCLC group itself still exhibited high heterogeneity (I² = 93.4%). This suggests that even within the NSCLC category, variations in the precise proportions of LUAD versus squamous cell carcinoma (SCC/LUSC) among studies, or the inclusion of other less common NSCLC subtypes, could impact pCR rates differently. For instance, the Huang et al., 2024 study included both LUAD and LUSC/SCC patients, while Yue et al., 2024 included SCC and non-squamous NSCLC (LUAD). The intrinsic biological differences between these subtypes, including their unique molecular pathways and responsiveness to specific therapeutic agents, could lead to divergent pCR outcomes. Furthermore, the meta-analysis did not provide a detailed breakdown of disease stage within NSCLC. While all enrolled studies focused on resectable stage II-IIIA/IIIb NSCLC, subtle differences in the distribution of specific substages (e.g., II vs. IIIA/B) across studies could influence resectability and pathological response, thereby contributing to heterogeneity. Another critical factor contributing to heterogeneity is the diversity in control group interventions. Our subgroup analysis by comparator drug type highlighted a statistically significant difference in effects between studies using Pembrolizumab+Chemotherapy as a comparator versus those using Chemotherapy alone. Studies like Hu et al., 2025, compared Tislelizumab+Chemotherapy against Pembrolizumab+Chemotherapy, introducing another active immune checkpoint inhibitor in the control arm. In contrast, Huang et al., 2024, compared Tislelizumab+Chemotherapy to chemotherapy alone. The inclusion of another immunotherapy agent in the control arm (Pembrolizumab) inherently provides a more robust comparison against a highly effective regimen, potentially narrowing the observed treatment effect of Tislelizumab, or even showing a non-significant difference as seen in the Pembrolizumab+Chemotherapy subgroup (OR 0.6968, 95% CI: 0.3803 to 1.2767). Conversely, comparing Tislelizumab+Chemotherapy to chemotherapy alone (as in Huang et al., 2024, and Yue et al., 2024 (placebo+chemotherapy)) is likely to demonstrate a more pronounced advantage for the immunotherapy-containing arm, which is reflected in the much higher OR observed in the Chemotherapy subgroup (OR 7.3123, 95% CI: 2.9204 to 18.3092). This fundamental difference in control arm design is a major driver of the observed overall heterogeneity. Beyond the type of comparator drug, the specific chemotherapy regimens themselves, which were not finely stratified in this meta-analysis, could also contribute to variability. Although all studies generally employed platinum-based doublet chemotherapy, the choice of platinum agent (cisplatin vs. carboplatin) and the companion cytotoxic drug (e.g., paclitaxel, pemetrexed, nab-paclitaxel) can influence response rates. For instance, Yue et al., 2024, specified cisplatin or carboplatin plus paclitaxel for squamous NSCLC and cisplatin or carboplatin plus pemetrexed for non-squamous NSCLC. Similarly, Huang et al., 2024, used nab-paclitaxel plus carboplatin for LUSC and pemetrexed plus carboplatin for LUAD. Even subtle differences in chemotherapy dosages, administration schedules, or the number of neoadjuvant cycles (e.g., 2–4 cycles in Huang et al., 2024; 3–4 cycles in Yue et al., 2024) across studies could affect the overall pCR. Finally, unmeasured or unaddressed factors, often categorized as other variables, might also contribute to the high heterogeneity. These could include variations in patient characteristics such as PD-L1 expression levels (though some studies like Yue et al., 2024, stratified by this), genetic mutations (e.g., EGFR/ALK status, though excluded in some studies), baseline tumor burden, patient comorbidities, and differences in surgical techniques or post-operative management protocols across the different clinical centers and regions. Furthermore, variability in surgical techniques (e.g., open vs. minimally invasive approaches), completeness of resection (R0 vs. R1/R2), and the interval between neoadjuvant therapy and surgery may contribute to unmeasured heterogeneity in pCR rates across studies. For instance, longer intervals between neoadjuvant therapy and surgery have been associated with higher pCR rates in some NSCLC studies, while variations in surgical quality and perioperative management may independently influence pathological outcomes. Although these factors were not systematically reported in the enrolled studies, they represent important sources of heterogeneity that should be addressed in future prospective trials through standardized surgical and pathological reporting protocols. Methodological differences, such as retrospective versus prospective study designs (e.g., Hu et al., 2025 and Huang et al., 2024 were retrospective, while Yue et al., 2024 was a randomized, double-blind, placebo-controlled phase 3 trial), could also introduce variability in reported outcomes. The inherent differences in study design and execution further complicate the interpretation of pooled results and contribute to the high I² value observed.

This subgroup meta-analysis specifically examined the CR rates of Tislelizumab combined with chemotherapy compared to other immunotherapy agents combined with chemotherapy in NSCLC patients. The analysis included 6 datasets from two studies: Chen et al., 2024, and Yan et al., 2024. These studies involved comparisons against Nivolumab, Pembrolizumab, Sintilimab, Toripalimab, and Camrelizumab, all combined with chemotherapy. The I² value was 0.0%, and the test for heterogeneity yielded a Q statistic of 1.52 with 5 degrees of freedom, resulting in a non-significant p-value of 0.9113. This indicates a high degree of consistency in the observed effects across the studies. However, despite the consistency, the meta-analysis did not demonstrate a statistically significant difference in CR rates when Tislelizumab plus chemotherapy was compared to other immunotherapy plus chemotherapy regimens. Under the CEM, the OR for CR was 1.4444, which was not statistically significant. Similarly, the REM also did not show a statistically significant effect, with an OR of 1.4570, a Z-value of 0.65, and a p-value of 0.5159. The heterogeneity parameters and tau = 0 further supported the low heterogeneity, meaning that the lack of significance was not attributable to substantial variability between studies. These findings suggest that, in terms of achieving CR, Tislelizumab combined with chemotherapy appears to have comparable efficacy to other PD-1 inhibitors (Nivolumab, Pembrolizumab, Sintilimab, Toripalimab, and Camrelizumab) when each is combined with chemotherapy as first-line treatment for advanced NSCLC. The studies included in this analysis, such as Chen et al., 2024, and Yan et al., 2024, represent real-world clinical practice in China, where these agents are commonly used. While some preclinical studies and network meta-analyses might suggest subtle differences in binding affinities or pharmacokinetics among PD-1 inhibitors, this specific meta-analysis on CR in combination with chemotherapy does not indicate a distinct advantage for Tislelizumab over the other immunotherapies in this outcome. The results align with broader real-world evidence often showing comparable efficacy profiles across various PD-1 inhibitors in similar clinical settings for NSCLC, particularly when combined with chemotherapy.

This subgroup analysis specifically investigated CR rates in studies where Tislelizumab combined with chemotherapy was compared against chemotherapy alone. Both fixed and random effects models were employed, and the results consistently indicated minimal heterogeneity among these studies, with an I² value of 0.0% and a non-significant heterogeneity test. This low heterogeneity suggests that the effects observed were consistent across the different studies within this subgroup. Under the CEM, Tislelizumab combined with chemotherapy demonstrated a statistically significant advantage in achieving CR, with an OR of 4.2887. This finding was supported by a Z-value of 3.18 and a p-value of 0.0015. The REM similarly yielded a statistically significant result, showing an OR of 3.8690, with a Z-value of 2.88 and a p-value of 0.0039. The heterogeneity parameters for the REM, with a tau² of 0 and a tau of 0, further reinforced the conclusion of minimal heterogeneity. The enrolled studies in this subgroup were Huang et al., 2024, Lu et al., 2024, and Wang et al., 2024 (contributing two separate comparisons: Tislelizumab plus paclitaxel plus carboplatin versus paclitaxel plus carboplatin, and Tislelizumab plus nab-paclitaxel plus carboplatin versus paclitaxel plus carboplatin). These results strongly suggest that the addition of Tislelizumab to chemotherapy significantly improves CR rates when compared to chemotherapy alone in NSCLC patients.

This meta-analysis, incorporating both RCTs and real-world retrospective cohort studies from China, provides a comprehensive overview of tislelizumab’s efficacy in NSCLC. Our findings regarding pCR demonstrate significant heterogeneity across the enrolled studies, with a high I² value of 92.5%. Despite this heterogeneity, when examining the pCR of tislelizumab combined with chemotherapy, our REM yielded an OR of 2.1103. This is further supported by individual study data, such as the single-arm study where 76% (35 of 46 patients) of the ITT population achieved MPR, and 52% (24 of 46 patients) achieved pCR in the ITT population, significantly exceeding the predefined threshold of 30% (10). The rate of complete surgical resection (R0) was also notably high in this single-arm study, achieved in 96% (44 of 46) of patients (11). These individual study outcomes lend qualitative support to the potential for favorable pathological responses with tislelizumab-based regimens.

Our subgroup analysis for pCR by disease type revealed a statistically significant difference between NSCLC and SCC only subgroups under the CEM (p-value = 0.0010). However, this difference was not statistically significant under the REM (p-value = 0.0632). This suggests that while there might be a trend, further high-quality studies, especially in SCLC populations, are needed to confirm differential benefits as also highlighted by a recent systematic review.

Regarding CR, our meta-analysis demonstrated consistent and statistically significant benefits of tislelizumab combined with chemotherapy, with minimal heterogeneity across the enrolled studies. The overall OR for CR was 2.6277 under the REM. This positive outcome aligns with findings from another systematic review and meta-analysis which indicated that tislelizumab significantly improved OS and PFS in lung cancer patients. This review, encompassing four Phase III RCTs with 1,837 patients, reported consistent benefits across NSCLC and SCLC populations, and similar efficacy for tislelizumab as monotherapy or in combination with chemotherapy.

While the efficacy data are encouraging, it is crucial to consider the safety profile and potential adverse events associated with tislelizumab. Our enrolled studies examine AEs and immune-related adverse events (irAEs) as secondary endpoints. The interim analysis from the phase 3 RATIONALE-315 study supports a manageable safety profile for tislelizumab when used in combination with neoadjuvant chemotherapy, with exposure-adjusted adverse event rates in the tislelizumab arms being lower than those in the placebo arm [PMID-39461775]. This is consistent with prior reports, suggesting that tislelizumab may have a favorable safety profile compared to classical chemotherapy in certain aspects. Furthermore, a dedicated analysis of immune-related adverse events (irAEs) across various PD-1 inhibitors, including tislelizumab, revealed no statistically significant difference in the incidence of irAEs at any grade among the agents, indicating a comparable and predictable irAE profile for tislelizumab. Notably, this analysis also identified a statistically significant association between the occurrence of irAEs and improved progression-free survival and objective response rates, suggesting that the presence of manageable irAEs may serve as a potential biomarker for treatment efficacy. Case reports further highlight the importance of vigilance when using ICIs. For instance, a case of pulmonary artery pseudoaneurysm leading to hemoptysis was reported in a 65-year-old male with squamous cell carcinoma after receiving chemotherapy combined with tislelizumab. Another case documented a 71-year-old male with extensive-stage small cell lung cancer developing type 1 diabetic ketoacidosis (new-onset) induced by tislelizumab after three cycles of treatment, despite no prior history of diabetes (12). These case reports underscore the potential for rare but serious irAEs, emphasizing the necessity of routine glucose monitoring and early recognition of high blood sugar and C-peptide deficiency during ICI treatment to prevent life-threatening endocrine irAEs (13). The overall bias analysis revealed a high or unclear risk of bias in a majority of our enrolled studies, especially in areas of randomization, allocation concealment, and blinding. This underscores the need for continued high-quality RCTs to validate these findings, particularly in specific subgroups, and to better characterize patient-specific factors such as age, gender, and comorbidities for optimized treatment strategies.

This meta-analysis provides compelling evidence that tislelizumab, particularly when combined with chemotherapy, significantly improves CR rates in NSCLC patients compared to chemotherapy alone, with a high degree of consistency across studies. However, it is important to acknowledge that radiological CR serves as a valuable short-term efficacy marker but is not a validated surrogate endpoint for long-term survival in NSCLC. In contrast, pCR has been established as a validated surrogate for long-term survival, particularly in the neoadjuvant setting. While the impact on pCR demonstrates higher heterogeneity, individual study data highlight tislelizumab’s potential for favorable pathological outcomes. In comparisons against other immunotherapy-plus-chemotherapy regimens, tislelizumab shows comparable efficacy in achieving CR. Furthermore, existing pharmacoeconomic evaluations predominantly from China suggest that tislelizumab represents a cost-effective treatment option in various cancer settings, including NSCLC. These findings underscore the clinical utility and economic value of tislelizumab in the management of NSCLC, though the generalizability of these findings is limited by the predominance of data from the Chinese healthcare landscape and the inherent biases and heterogeneity of the included studies. To address these limitations, future high-quality randomized controlled trials with diverse patient populations and standardized reporting are warranted. These studies are crucial to better confirm these benefits globally, explore nuanced differences in response across specific NSCLC subtypes, and further refine optimal treatment strategies.