Comparative efficacy and safety of PD-1/PD-L1 inhibitors versus docetaxel in the treatment of non-small cell lung cancer: a systematic review and meta-analysis

Introduction

Lung cancer is a predominant cause of cancer-linked death worldwide (1), and non-small cell lung cancer (NSCLC) is one of the most prevalent subtypes (2). According to global cancer statistics, approximately 2.4 million new lung cancer cases are confirmed per year, with a predicted 1.8 million deaths. This disease imposes a heavy economic strain and presents a formidable challenge to public health. The combat against lung cancer has been relentless, leading to the development of various therapeutic approaches like surgery, radiotherapy, chemotherapy, and targeted drug therapy. Identifying and establishing the optimal treatment strategy for lung cancer has become a shared objective of the global medical community.

Among conventional chemotherapeutic agents, docetaxel has played a pivotal role in the treatment of advanced NSCLC. As a taxane-class drug, docetaxel is commonly employed as a therapeutic approach for individuals with unfavorable prognostic outcomes. Because of its favorable tolerability, docetaxel has demonstrated significant efficacy in delaying progression and prolonging survival in NSCLC individuals and metastatic illness (3). In comparison to traditional chemotherapy, cancer immunotherapy offers a broader range of therapeutic options based on distinct mechanisms of action. These strategies primarily include adoptive T-cell transfer, oncolytic viruses, cancer vaccines, as well as monoclonal antibodies. Among these, monoclonal antibodies (4), particularly immune checkpoint inhibitors (ICIs), are extensively adopted in cancer immunotherapy. They function by blocking the interplay between inhibitory receptors and their ligands, thus disrupting suppressive signaling pathways (5). One such inhibitory receptor, programmed cell death-1 (PD-1), is an immune regulatory checkpoint expressed on activated cytotoxic T lymphocytes (CTLs). When PD-1 binds to its ligand (PD-L1), CTLs’ ability to eliminate tumor cells is diminished. Therefore, blocking inhibitory receptors or ligands with ICI may help restore the ability to kill tumor cells (6). The emergence of ICIs, particularly PD-1 and PD-L1 inhibitors, has fundamentally transformed the therapeutic NSCLC method, prompting a reevaluation of conventional multi-line treatment strategies. At this pivotal juncture, rigorous comparative analyses of ICIs versus docetaxel in NSCLC management have become indispensable. This therapeutic evolution has reshaped clinical decision-making frameworks, facilitating biomarker-driven treatment selection based on tumor proportion score (TPS) stratification. Such precision medicine approaches have supplanted empiric regimens, thereby optimizing clinical outcomes within molecularly defined patient subsets and offering personalized therapeutic avenues. Furthermore, this investigation underscores the redefined role of docetaxel as a second-line agent, wherein its clinical utility extends beyond broad-spectrum efficacy to encompass promising applications in the context of immunotherapy rechallenge and rational combinatorial regimens (7). However, while these immunotherapies have demonstrated remarkable efficacy in certain patient populations, their effectiveness and applicability across diverse patient cohorts require further investigation, and the survival benefits they confer remain a subject of ongoing debate.

Research on immunotherapy for NSCLC continues to advance. At present, plenty of clinical trials examining the integration of PD-1/PD-L1 inhibitors with docetaxel for advanced NSCLC have been completed or are ongoing. The present study aims to systematically collect and analyze data from published literature and clinical research via systematic review and meta-analysis, thereby providing evidence-based insights to inform clinical decision-making for NSCLC therapy. We present this article in accordance with the PRISMA reporting checklist (8) (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-976/rc).

Methods

A meta-analysis was undertaken as per an established study protocol (available on request) registered with PROSPERO (ID: CRD42024584679).

Literature search strategy

PubMed, Embase, Cochrane Library, as well as Web of Science were retrieved from inception to July 31, 2024, to identify eligible studies. Our search utilized the following keywords: Nivolumab, ICIs, PD-1, PD-L1, docetaxel, and Carcinoma, NSCLC. The search strategy was formulated as per the PICOS framework, integrating Medical Subject Headings (MeSH) and free-text terms (Table S1). Moreover, manual searches of references from relevant articles and gray literature were executed to identify further eligible studies.

Inclusion criteria

Participants (P): patients of any sex, aged >18 years, with a verified diagnosis of NSCLC via pathological biopsy or clinical, biochemical, imaging, and/or endoscopic examination, with or without complications.

Intervention (I): the experimental group received monotherapy with PD-1/PD-L1 inhibitors, while the control group received monotherapy with docetaxel chemotherapy.

Outcomes (O): data on overall survival (OS) and progression-free survival (PFS) in hazard ratios (HRs) with 95% confidence intervals (CIs); duration of response (DOR) with 95% CI; objective response rate (ORR), expressed as an odds ratio (OR); and treatment-related adverse events (TRAEs), reported as relative risk (RR).

Study design (S): merely randomized controlled trials (RCTs) were encompassed.

Language restriction: only English publications were considered.

Exclusion criteria

The following studies were ostracized:

• Non-English studies.
• Studies lacking valid data (e.g., missing relevant outcome measures).
• Reviews, letters, case reports, conference abstracts, as well as animal studies.
• Those with inappropriate design (e.g., cohort studies, case reports).
• Those with incompatible research objectives (e.g., studies with interventions, comparators, or disease conditions that did not satisfy eligibility criteria, or those involving combination therapy).
• Studies with a sample size of fewer than 100 patients or those published before 2015.

Data extraction

Two authors (S.L. and R.L.) independently screened retrieved literature. Initially, titles and abstracts were reviewed and studies failing to meet the eligibility criteria were ostracized. Full-text articles were then checked for final inclusion. Dissents were addressed via discussion, with a third author (J.Z.) consulted when necessary. Data was gathered independently by two researchers utilizing a pre-designed electronic form and included: first author, publication year, study location, original inclusion criteria, total sample size, OS, PFS, ORR, DOR and TRAEs. Additionally, subgroup data were collected to facilitate subgroup analyses.

Quality assessment

Two authors (S.L. and R.L.) independently examined the quality of encompassed RCTs through the Cochrane Risk of Bias 2.0 for risk of bias (RoB) assessment in intervention studies (9) covering these domains: randomization process, deviations from intended interventions, missing outcome information, outcome measurement, as well as reported result choosing (Figure 1A,1B). The RoB for every outcome was rated as low, high or unclear risk (10).

Figure 1 The Cochrane risk bias assessment tool was used to evaluate bias from five key sources: (I) random process; (II) deviations from the intended interventions; (III) missing outcome data; (IV) measurement of the outcome; (V) selection of the reported result. (A) Risk of bias graph; (B) risk of bias summary.

Statistical analysis

Statistical analyses of interventions were carried out as per the Cochrane Handbook for Systematic Reviews of Interventions 6.5 (11). HRs and 95% CIs were derived for assessing the relations of OS to PFS. ORR and incidence of TRAEs were analyzed via OR or RR as appropriate. 95% CIs of DOR were also recorded. The degree of statistical heterogeneity was detected via the Cochrane I-squared statistic (I²), with P<0.10 or I²>50% denoting substantial heterogeneity. If significant heterogeneity was present, a random-effects model was leveraged, and sensitivity and subgroup or meta-regression analyses were executed. Publication bias was examined via funnel plots and quantified utilizing Egger’s test, with P<0.05 denoting significant publication bias. Statistical analyses were enabled by STATA 18.0.

Results

Search results

Two authors (S.L. and R.L.) independently performed a thorough search of PubMed, Cochrane, Embase, and Web of Science from inception until July 31, 2024, using a predefined strategy. 6,125 articles were identified. The search flowchart is illustrated in Figure 2. After our removal of duplicate records, 4,930 articles remained. Following the exclusion of 2,044 conference abstracts, systematic reviews, book chapters and so on, 2,886 articles were retained. Further screening using the EndNote X9 (version X9.3.3, Build 12062; Clarivate Analytics, Philadelphia, PA, USA) led to the exclusion of cohort studies, case-control studies, and articles that did not align with the study objectives, leaving 101 articles for initial eligibility assessment. Subsequent abstract and full-text screening resulted in the exclusion of 71 articles, along with 7 articles lacking valid research data and 6 articles with insufficient available data. Ultimately, data from 23 articles were synthesized, including 11 primary studies (12-22) and 12 additional studies that provided updates on certain clinical efficacy endpoints based on the original literature (23-34).

Figure 2 Screening process of study selection and inclusion.

Study characteristics

A total of 6,770 patients with advanced NSCLC were enrolled across 11 studies for RoB assessment. Among these studies, one was a Phase II open-label RCT, one was a combined Phase II/III open-label RCT, and the remaining nine were Phase III open-label RCTs. Baseline characteristics of patients, including age, sex, and country of origin, were extracted from each included trial for further analysis. Information on the included studies, patient numbers, interventions, baseline characteristics, and HRs with 95% CIs for OS and PFS are provided in Table 1.

Meta-analysis

OS

A random-effects model analysis showed that relative to docetaxel, PD-1/PD-L1 inhibitors notably increased median OS, with an average survival extension of 2.89 months [weighted mean difference (WMD): 2.89 months, 95% CI: 2.13–3.65, I²: 99.8%, P<0.001]. The PD-1/PD-L1 inhibitor cohort displayed a 28% drop in death risk over the same follow-up period (HR 0.72, 95% CI: 0.65–0.80, I²: 62.0%, P=0.005) with statistical significance (Figure 3A,3B).

Figure 3 Forest plots of continuous variables (A) and HR (B) for OS, continuous variables (C) and HR (D) for PFS, OR for ORR (E), and continuous variables for DOR (F). CI, confidence interval; DL, DerSimonian and Laird random effects models; DOR, duration of response; HR, hazard ratio; OR, odds ratio; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; WMD, weighted mean difference.

PFS

In contrast to docetaxel, PD-1/PD-L1 inhibitors failed to confer an evident amelioration in PFS (WMD: 0.41 months, 95% CI: −0.45 to 1.26, I²: 100%, P<0.001), but they were related to a 19% drop in the likelihood of progression, which was statistically significant (HR 0.81, 95% CI: 0.71–0.93, I²: 80.8%, P<0.001) (Figure 3C,3D).

ORR

Random-effects analysis of ORR showed that receiving PD-1/PD-L1 significantly improved ORR and was more effective than docetaxel (OR 2.67, 95% CI: 1.73–4.10, I²: 83.8%, P<0.001). This statistically significant result proves that PD-1/PD-L1 inhibitors enable superior disease management and potential survival benefits compared to docetaxel (Figure 3E).

DOR

Relative to docetaxel, PD-1/PD-L1 inhibitors markedly prolonged the mean DOR by 12.42 months (WMD: 12.42 months, 95% CI: 8.79–16.06, I²: 98.2%, P<0.001), demonstrating statistically significant efficacy in extending response duration (Figure 3F).

TRAEs

TRAEs across all severity levels (Grade 1–4) were observed in 68.67% of PD-1/PD-L1 inhibitor receivers, relative to a higher rate of 86.62% in those undergoing docetaxel chemotherapy. The pooled RR for TRAEs was 0.80 (RR 0.80, 95% CI: 0.76–0.84, I²: 72.8%, P<0.001), which suggested the link of PD-1/PD-L1 inhibitors to an 80% RR of TRAEs in contrast to docetaxel, with a marked decrease in the overall TRAE incidence.

Furthermore, when the incidence of Grade 3–4 TRAEs was compared between PD-1/PD-L1 inhibitors and docetaxel, the incidence was 14.17% in the former group versus 47.85% in the latter group. The pooled RR was 0.28 (RR 0.28, 95% CI: 0.22–0.36, I²: 85.9%, P<0.001), showing that the risk of Grade 3–4 TRAEs with PD-1/PD-L1 inhibitors was 28% of that noted with docetaxel. These findings highlight the significant drop in the occurrence of severe TRAEs with PD-1/PD-L1 inhibitors, which was statistically significant (Figure 4A,4B).

Figure 4 Forest plots of overall (A) and grade ≥3 (B) TRAEs. CI, confidence interval; DL, DerSimonian and Laird random effects models; TRAEs, treatment-related adverse events.

Subgroup analysis

PD-1/PD-L1 expression

An analysis of OS revealed a significant difference between patients receiving anti-PD-1/PD-L1 with high expression versus low expression. Across three subgroups stratified by TPS (TPS ≥50%, TPS ≥1%, and TPS <1%), people with TPS ≥50% and TPS ≥1% demonstrated greater improvements in median OS and a reduced risk of mortality compared to those with TPS <1%. Specifically, anti-PD-1/PD-L1 treatment in the TPS ≥50% cohort extended OS by an average of 3.88 months (WMD: 7.39 months, 95% CI: 4.01–10.76, I²: 99.8%, P<0.001), with a 41% decrease in death risk compared to docetaxel (HR 0.59, 95% CI: 0.51–0.69, I²: 41.5%, P=0.10). In the TPS ≥1% subgroup, OS was prolonged by an average of 4.11 months (WMD: 4.11 months, 95% CI: 3.27–4.94, I²: 99.7%, P<0.001), with a 29% decrease in mortality risk in contrast to docetaxel (HR 0.71, 95% CI: 0.64–0.77, I²: 31.2%, P=0.18). In the TPS <1% subgroup, OS increased by an average of 1.54 months (WMD: 1.54 months, 95% CI: 0.47–2.62, I²: 99.8%, P<0.001), with a 19% reduction in mortality risk compared to docetaxel (HR 0.81, 95% CI: 0.71–0.92, I²: 0%, P=0.69). Therefore, anti-PD-1/PD-L1 treatment in individuals with high expression notably prolongs OS and reduces mortality risk in comparison to the low-expression cohort, with statistical significance (I²: 47.2%, P=0.01) (Figure 5A,5B).

Figure 5 Subgroup analysis of OS continuous variables (A) and HR (B) and PFS continuous variables (C) and HR (D) for PD-1/PD-L1 inhibitors expression. CI, confidence interval; DL, DerSimonian and Laird random effects models; HR, hazard ratio; OS, overall survival; PD-1, programmed cell death-1; PD-L1, programmed cell death-ligand 1; PFS, progression-free survival; TPS, tumor proportion score; WMD, weighted mean difference.

A subgroup analysis of PFS did not demonstrate an evident variation in PFS between participants with high and low PD-1/PD-L1 expression. However, PFS HR analysis proved that anti-PD-1/PD-L1 in the TPS ≥50% subgroup lowered the likelihood of disease progression by 38% relative to docetaxel (HR 0.62, 95% CI: 0.52–0.74, I²: 11.3%, P=0.32). In the TPS ≥1% cohort, the likelihood of progression dropped by 17% in contrast to docetaxel (HR 0.83, 95% CI: 0.75–0.92, I²: 33.5%, P=0.19). However, an evident drop in progression risk was not noted in the TPS <1% subgroup (HR 0.99, 95% CI: 0.82–1.19, I²: 42.9%, P=0.17). Therefore, anti-PD-1/PD-L1 treatment in people with high PD-1/PD-L1 expression effectively reduces the risk of progression events, with statistical significance (I²: 68.4%, P<0.001) (Figure 5C,5D). In addition, based on each subgroup and pooled I² data, PD-1/PD-L1 expression is likely a source of heterogeneity for PFS HR.

PD-1 versus PD-L1 (target spot)

The 11 eligible studies were split into two subgroups: the PD-1 subgroup, which included trials of nivolumab (12-14,23-28), pembrolizumab (15,16,29,30), tislelizumab (17), and sintilimab (21); and the PD-L1 subgroup, which included trials of atezolizumab (18-20,31-33) and avelumab (22,34).

Subgroup analysis of OS WMD and HRs indicated a marked difference between anti-PD-1 and anti-PD-L1 therapies. Median OS with PD-1 antibodies (WMD: 5.20 months, 95% CI: 3.66–6.75, I²: 99.1%, P<0.001) was higher than with PD-L1 antibodies (WMD: 3.21 months, 95% CI: 0.96–5.47, I²: 99.7%, P<0.001). In contrast to docetaxel, anti-PD-1 therapy lowered the probability of death by 35% (HR 0.65, 95% CI: 0.59–0.72, I²: 18.4%, P=0.29), whereas anti-PD-L1 therapy lowered mortality risk by 16% (HR 0.84, 95% CI: 0.77–0.91, I²: 0%, P=0.55). Therefore, anti-PD-1 treatment enables a greater reduction in death risk than anti-PD-L1 therapy, with statistical significance (I²: 62.0%, P<0.001) (Figure 6A,6B). Additionally, PD-1/PD-L1 targets are likely to be a source of heterogeneity in OS HR.

Figure 6 Subgroup analysis of OS continuous variables (A) and HR (B) and PFS continuous variables (C) and HR (D) for PD-1 and PD-L1 target spot. CI, confidence interval; DL, DerSimonian and Laird random effects models; HR, hazard ratio; OS, overall survival; PD-1, programmed cell death-1; PD-L1, programmed cell death-ligand 1; PFS, progression-free survival; WMD, weighted mean difference.

There was no significant difference between PD-1 antibody and PD-L1 antibody in reducing median PFS. An analysis of PFS HRs demonstrated that anti-PD-1 treatment lowered the progression risk by 29% compared to docetaxel (HR 0.71, 95% CI: 0.61–0.84, I²: 73.9%, P=0.002), whereas anti-PD-L1 therapy did not notably lower progression risk (HR 0.99, 95% CI: 0.89–1.10, I²: 33.8%, P=0.21). These results were statistically significant (I²: 80.8%, P<0.001) (Figure 6C,6D).

ORR analysis revealed a certain degree of difference between the PD-L1 and PD-1 groups. Following treatment, the likelihood of attaining an objective response was 26% higher in the PD-L1 cohort than those receiving docetaxel (OR 1.26, 95% CI: 0.99–1.60, I²: 78.3%, P<0.001), whereas the probability of achieving an objective response in the PD-1 group was 3.84 times that of patients receiving docetaxel (OR 3.84, 95% CI: 2.31–6.38, I²: 0%, P =0.82). Therefore, PD-1 inhibitors are more effective in elevating ORR, with statistically significant results (I²: 83.8%, P<0.001) (Figure 6E).

Sensitivity analysis and RoB

To account for the possible influence of factors including sample size and study quality on model outcomes, a sensitivity analysis was carried out. Pertinent biases were not identified, confirming the robustness of the meta-analysis conclusions (Figure S1). To assess publication bias across studies, STATA 18.0 was employed to perform Egger’s test, Begg’s test (Table 2), and funnel plot analysis (Figure 7). The results of Egger’s test for OS (P=0.007) and ORR (P=0.004), as well as Begg’s test for ORR (P=0.007), suggest possible publication bias. No publication bias was present for other outcome measures (P>0.05).

Figure 7 Funnel plot of OS HR (A), PFS HR (B), ORR (C), overall (D) and grade ≥3 (E) TRAEs. HR, hazard ratio; OR, odds ratio; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; s.e., standard error; TRAEs, treatment-related adverse events.

Meta-regression analysis

To find the sources of heterogeneity, a meta-regression analysis was executed on factors such as age, geographical region, sex, sample size, therapeutic target, and TPS. The findings revealed that TPS (P=0.006) led to heterogeneity in OS, whereas the therapeutic target (P=0.006) was related to heterogeneity in ORR (Table 3).

Discussion

This meta-analysis primarily elucidates the comparative outcomes of PD-1/PD-L1 inhibitors and docetaxel in curing NSCLC based on RCTs published to date. The analyzed endpoints include OS, PFS, ORR, DOR and TRAEs. Overall, PD-1/PD-L1 inhibitors yielded superior outcomes than docetaxel. In most eligible studies, the former was administered as monotherapy, while the effects of the integration with chemotherapy or other targeted therapies were not examined.

Our meta-analysis results demonstrated that relative to second-line docetaxel therapy, PD-1/PD-L1 inhibitor therapy is markedly more effective in ameliorating OS, ORR, and DOR. Although no evident improvement in PFS was found, PD-1/PD-L1 inhibitors correlated with a declined probability of progression. Additionally, analyses of subgroups categorized by PD-1/PD-L1 expression and therapeutic targets revealed certain discrepancies. In NSCLC treatment, PD-1 and PD-L1 expression is pivotal for forecasting the effectiveness of ICIs (35). Within the PD-1/PD-L1 expression subgroup, relative to docetaxel, PD-1/PD-L1 inhibitors conferred an OS benefit in the high TPS expression cohort relative to those with low expression; however, no improvement in PFS was noted. Furthermore, OS and PFS HRs for PD-1/PD-L1 inhibitors were lower in the cohort with high expression than in the low-expression subgroup, suggesting that high PD-1/PD-L1 expression was more notably related to decreased risks of death and progression. In the included RCTs, PD-L1 expression assessment exhibited a degree of complexity, particularly in certain trials like POPLAR and OAK. Additionally, data collection in some recently initiated trials, such as KEYNOTE-033 and IMpower210, remains incomplete. Therefore, a more comprehensive evaluation of PD-1/PD-L1 expression subgroup analyses necessitates additional data. Regarding subgroup analyses based on therapeutic targets, PD-1 inhibitor treatment was related to lower OS and PFS HRs than PD-L1 inhibitor therapy, indicating a greater survival benefit. Moreover, PD-L1 inhibitors did not notably lower disease progression risk. In terms of tumor expression, PD-1/PD-L1 antibodies conferred a significantly greater reduction in HR in tumors with high expression levels (TPS ≥50%) compared to those with low expression levels (TPS <1%). However, no significant differences were found between the groups in terms of delayed WMD. Similarly, in the target-based subgroup analysis, patients receiving PD-1 inhibitors exhibited lower HRs than those treated with PD-L1 inhibitors. It is possible that patient heterogeneity contributed to the amplification of these statistical differences. From a mechanistic perspective, immunotherapy may lead to atypical response patterns in some patients. One such phenomenon is pseudoprogression, wherein the tumor initially appears to increase in size or new lesions emerge following initiation of immunotherapy, only to subsequently stabilize or regress (36). Likewise, a small subset of patients may develop hyperprogression, characterized by accelerated tumor growth during the early phase of treatment (37). These atypical responses may compromise short-term disease control while potentially reducing the risk of long-term progression. Therefore, HR and long-term PFS rates should be prioritized over median PFS when evaluating the therapeutic efficacy of immunotherapy. In addition, PD-1 inhibitors exhibited a greater improvement in ORR, underscoring their superior efficacy in NSCLC treatment.

TRAEs are a critical concern in the treatment of NSCLC, as they not only have the potential to interrupt therapy but also pose serious threats to patients’ safety and health. These adverse events (AEs) possibly arise from T-cell cytotoxicity targeting antigens in normal/non-tumor tissues or from off-target effects, which may be directly caused by immune tolerance disruption (38). This study evaluated the incidence of TRAEs related to PD-1/PD-L1 inhibitors relative to docetaxel. In contrast to docetaxel, PD-1/PD-L1 inhibitors in advanced NSCLC individuals resulted in a reduction of approximately 20% in grade 1–4 TRAEs and a marked decrease of around 72% in grade 3–4 TRAEs, indicating a superior safety profile in mitigating severe adverse reactions. Although certain treatment-related complications, like anemia, fatigue, dermatologic, gastrointestinal, pulmonary, cardiac, and endocrine disorders, have been observed in patients receiving either PD-1/PD-L1 inhibitors or docetaxel, current evidence indicates that docetaxel is associated with a higher risk of serious infections and hematologic AEs. In comparison, patients treated with PD-1/PD-L1 inhibitors exhibit a relatively lower incidence of these complications (39,40). However, due to the discrepancy between the data of clinical trials and the real situation (41,42), the relationship between infection and immunotherapy in patients with cancer is difficult to clarify (43). Therefore, clinicians should closely monitor the onset and progression of these symptoms throughout the course of treatment.

The aforementioned differences may stem from the distinct mechanisms of action of the two drug classes. Regarding therapeutic efficacy, resistance to docetaxel is often associated with mechanisms like the overexpression of P-glycoprotein (P-gp)-mediated drug efflux pumps in tumor cells (44), which may lead to chemotherapy resistance. In contrast, resistance to PD-1/PD-L1 antibodies primarily arises from the disruption of immune cell interactions within tumor microenvironment. This resistance is primarily linked to the upregulation of immune checkpoints like lymphocyte-activation gene 3 (LAG-3) and T-cell immunoglobulin and mucin domain-containing protein 3 (TIM-3). Such dysregulation may be mitigated by combination therapies targeting different checkpoints to synergistically activate tumor immune responses while preventing compensatory pathway activation, thereby reducing the development of resistance (45,46). For instance, the combination of PD-1/PD-L1 with CTLA-4 antibodies demonstrated improved efficacy in treating NSCLC. However, this combined therapy is also linked to an elevated incidence of TRAEs (47).

Regarding drug safety, docetaxel exerts its effect by binding to β-tubulin, thereby inhibiting the proper assembly of microtubules into mitotic spindle and arresting cell cycle at the G2/M phase. While this effectively eradicates malignantly proliferating tumor cells, it also affects rapidly proliferating normal cells, such as hematopoietic progenitors in the bone marrow and epithelial cells of the gastrointestinal tract, leading to significant hematologic toxicities and chemotherapy-related adverse effects such as mucositis (3). In contrast, PD-1/PD-L1 antibodies may disrupt immune tolerance by excessively activating peripheral T cells and impairing the regulation of adaptive immune responses. This can lead to rapid diversification and clonal expansion of cytotoxic T lymphocytes, thereby inducing rapid diversification and clonal expansion of cytotoxic T cells, as well as exacerbating non-tumor-associated inflammation and autoimmunity (48,49). This pathogenic immune activation may result in irAEs (50), including pneumonitis, colitis, and endocrine disorders. Clinical evidence suggests that the toxicities associated with ICIs can be managed without compromising their antitumor immune response (51). Their toxicity profile can be modulated through immunoregulatory interventions such as corticosteroids (52), so they have a notable advantage in balancing efficacy and safety.

There are limitations in this study. First, the methodological quality of encompassed trials is crucial in meta-analysis, and studies with larger sample sizes are typically assigned greater weight. Small-scale meta-analyses may be limited to summarizing available information and generating hypotheses for future research (53). Our meta-analysis included some studies with limited sample sizes. Although no abnormalities were identified in sensitivity analyses, caution is warranted when interpreting the findings. Second, due to the unavailability of certain outcome data, eligible research on PD-1/PD-L1 antibody expression and DOR was limited, restricting the scope of our meta-analysis. Moreover, despite no evident publication bias in most encompassed studies, there was potential bias in certain outcomes like OS and ORR. The pooled HR for OS may be overestimated due to the potential omission of unpublished negative trials. Moreover, ORR assessments in open-label studies are susceptible to evaluator bias, with a tendency to overestimate response rates in the experimental group. These potential sources of bias may compromise the objectivity and accuracy of the results, thereby diminishing the reliability of the conclusions (54). Therefore, while the current findings support a favorable trend for PD-1/PD-L1 inhibitors over docetaxel, the presence of bias may attenuate the strength of the evidence, particularly with respect to the magnitude of OS benefit. Lastly, significant heterogeneity was observed in some outcome measures. Although subgroup analyses and meta-regression analyses identified certain sources of heterogeneity, undiscovered sources may still exist. Future studies should aim to increase sample sizes, standardize outcome measures, and enhance data transparency. Multicenter collaborations should be encouraged to improve sample diversity and representativeness, allowing for a more comprehensive exploration of sources of heterogeneity. These improvements will strengthen the scientific rigor and credibility of research, ultimately presenting more evidence for utilizing PD-1/PD-L1 antibodies clinically and advancing tumor immunotherapy.

Conclusions

In general, our study findings demonstrate that, in contrast to docetaxel, PD-1/PD-L1 antibodies for treating NSCLC are linked to amelioration in OS, ORR, and DOR. PD-1/PD-L1 antibodies lower the risk of progression events, but no significant amelioration in PFS was observed. Subgroup analysis proved that PD-1/PD-L1 antibodies with high TPS expression, as well as PD-1 monotherapy, were associated with lower combined HRs for OS and PFS, with PD-1 monotherapy demonstrating superior ORR improvement. Compared with docetaxel, PD-1/PD-L1 antibodies better reduce the occurrence of both overall and severe TRAEs. Nevertheless, continued vigilance is warranted regarding the incidence and progression of AEs.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-976/rc

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-976/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-976/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Hendriks LEL, Remon J, Faivre-Finn C, et al. Non-small-cell lung cancer. Nat Rev Dis Primers 2024;10:71. [Crossref] [PubMed]
3. Farha NG, Kasi A. Docetaxel. In: StatPearls. Treasure Island (FL): StatPearls Publishing; June 8, 2024.
4. Zhang M, Liu C, Tu J, et al. Advances in cancer immunotherapy: historical perspectives, current developments, and future directions. Mol Cancer 2025;24:136. [Crossref] [PubMed]
5. Dine J, Gordon R, Shames Y, et al. Immune Checkpoint Inhibitors: An Innovation in Immunotherapy for the Treatment and Management of Patients with Cancer. Asia Pac J Oncol Nurs 2017;4:127-35. [Crossref] [PubMed]
6. Jia H, Jiang L, Shen X, et al. Post-translational modifications of cancer immune checkpoints: mechanisms and therapeutic strategies. Mol Cancer 2025;24:193. [Crossref] [PubMed]
7. Wang Y, Hao Q, Nie J, et al. Nivolumab combined docetaxel versus nivolumab in patients with previously treated nonsmall cell lung cancer: a phase 2 study. Anticancer Drugs 2024;35:412-7. [Crossref] [PubMed]
8. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372: [Crossref] [PubMed]
9. Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898. [Crossref] [PubMed]
10. Liu K, Sun DQ, Liao X, et al. Interpretation of the Randomized Controlled Trials Risk of Risk of Bias, 2.0 Revision. Chinese Journal of Evidence-Based Cardiovascular Medicine 2019;11:284-91.
11. Higgins JPT, Thomas J, Chandler J, et al. editors. Cochrane Handbook for Systematic Reviews of Interventions version 6.5 (updated August 2024). Cochrane, 2024. Available online: www.training.cochrane.org/handbook
12. Wu YL, Lu S, Cheng Y, et al. Nivolumab Versus Docetaxel in a Predominantly Chinese Patient Population With Previously Treated Advanced NSCLC: CheckMate 078 Randomized Phase III Clinical Trial. J Thorac Oncol 2019;14:867-75. [Crossref] [PubMed]
13. Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus Docetaxel in Advanced Squamous-Cell Non-Small-Cell Lung Cancer. N Engl J Med 2015;373:123-35. [Crossref] [PubMed]
14. Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus Docetaxel in Advanced Nonsquamous Non-Small-Cell Lung Cancer. N Engl J Med 2015;373:1627-39. [Crossref] [PubMed]
15. Ren S, Feng J, Ma S, et al. KEYNOTE-033: Randomized phase 3 study of pembrolizumab vs docetaxel in previously treated, PD-L1-positive, advanced NSCLC. Int J Cancer 2023;153:623-34. [Crossref] [PubMed]
16. Herbst RS, Baas P, Kim DW, et al. Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial. Lancet 2016;387:1540-50. [Crossref] [PubMed]
17. Zhou C, Huang D, Fan Y, et al. Tislelizumab Versus Docetaxel in Patients With Previously Treated Advanced NSCLC (RATIONALE-303): A Phase 3, Open-Label, Randomized Controlled Trial. J Thorac Oncol 2023;18:93-105. [Crossref] [PubMed]
18. Wu YL, Lu S, Chen G, et al. IMpower210: A phase III study of second-line atezolizumab vs. docetaxel in East Asian patients with non-small cell lung cancer. Chin J Cancer Res 2024;36:103-13. [Crossref] [PubMed]
19. Rittmeyer A, Barlesi F, Waterkamp D, et al. Atezolizumab versus docetaxel in patients with previously treated non-small-cell lung cancer (OAK): a phase 3, open-label, multicentre randomised controlled trial. Lancet 2017;389:255-65. [Crossref] [PubMed]
20. Fehrenbacher L, Spira A, Ballinger M, et al. Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer (POPLAR): a multicentre, open-label, phase 2 randomised controlled trial. Lancet 2016;387:1837-46. [Crossref] [PubMed]
21. Shi Y, Wu L, Yu X, et al. Sintilimab versus docetaxel as second-line treatment in advanced or metastatic squamous non-small-cell lung cancer: an open-label, randomized controlled phase 3 trial (ORIENT-3). Cancer Commun (Lond) 2022;42:1314-30. [Crossref] [PubMed]
22. Barlesi F, Vansteenkiste J, Spigel D, et al. Avelumab versus docetaxel in patients with platinum-treated advanced non-small-cell lung cancer (JAVELIN Lung 200): an open-label, randomised, phase 3 study. Lancet Oncol 2018;19:1468-79. [Crossref] [PubMed]
23. Vokes EE, Ready N, Felip E, et al. Nivolumab versus docetaxel in previously treated advanced non-small-cell lung cancer (CheckMate 017 and CheckMate 057): 3-year update and outcomes in patients with liver metastases. Ann Oncol 2018;29:959-65. [Crossref] [PubMed]
24. Lu S, Wang J, Cheng Y, et al. Nivolumab versus docetaxel in a predominantly Chinese patient population with previously treated advanced non-small cell lung cancer: 2-year follow-up from a randomized, open-label, phase 3 study (CheckMate 078). Lung Cancer 2021;152:7-14. [Crossref] [PubMed]
25. Horn L, Spigel DR, Vokes EE, et al. Nivolumab Versus Docetaxel in Previously Treated Patients With Advanced Non-Small-Cell Lung Cancer: Two-Year Outcomes From Two Randomized, Open-Label, Phase III Trials (CheckMate 017 and CheckMate 057). J Clin Oncol 2017;35:3924-33. [Crossref] [PubMed]
26. Chang J, Wu YL, Lu S, et al. Three-year follow-up and patient-reported outcomes from CheckMate 078: Nivolumab versus docetaxel in a predominantly Chinese patient population with previously treated advanced non-small cell lung cancer. Lung Cancer 2022;165:71-81. [Crossref] [PubMed]
27. Borghaei H, Gettinger S, Vokes EE, et al. Five-Year Outcomes From the Randomized, Phase III Trials CheckMate 017 and 057: Nivolumab Versus Docetaxel in Previously Treated Non-Small-Cell Lung Cancer. J Clin Oncol 2021;39:723-33. [Crossref] [PubMed]
28. Antonia SJ, Borghaei H, Ramalingam SS, et al. Four-year survival with nivolumab in patients with previously treated advanced non-small-cell lung cancer: a pooled analysis. Lancet Oncol 2019;20:1395-408. [Crossref] [PubMed]
29. Herbst RS, Garon EB, Kim DW, et al. Five Year Survival Update From KEYNOTE-010: Pembrolizumab Versus Docetaxel for Previously Treated, Programmed Death-Ligand 1-Positive Advanced NSCLC. J Thorac Oncol 2021;16:1718-32. [Crossref] [PubMed]
30. Herbst RS, Baas P, Perez-Gracia JL, et al. Use of archival versus newly collected tumor samples for assessing PD-L1 expression and overall survival: an updated analysis of KEYNOTE-010 trial. Ann Oncol 2019;30:281-9. [Crossref] [PubMed]
31. Hida T, Kaji R, Satouchi M, et al. Atezolizumab in Japanese Patients With Previously Treated Advanced Non-Small-Cell Lung Cancer: A Subgroup Analysis of the Phase 3 OAK Study. Clin Lung Cancer 2018;19:e405-15. [Crossref] [PubMed]
32. Fehrenbacher L, von Pawel J, Park K, et al. Updated Efficacy Analysis Including Secondary Population Results for OAK: A Randomized Phase III Study of Atezolizumab versus Docetaxel in Patients with Previously Treated Advanced Non-Small Cell Lung Cancer. J Thorac Oncol 2018;13:1156-70. [Crossref] [PubMed]
33. Bordoni R, Ciardiello F, von Pawel J, et al. Patient-Reported Outcomes in OAK: A Phase III Study of Atezolizumab Versus Docetaxel in Advanced Non-Small-cell Lung Cancer. Clin Lung Cancer 2018;19:441-449.e4. [Crossref] [PubMed]
34. Park K, Özgüroğlu M, Vansteenkiste J, et al. Avelumab Versus Docetaxel in Patients With Platinum-Treated Advanced NSCLC: 2-Year Follow-Up From the JAVELIN Lung 200 Phase 3 Trial. J Thorac Oncol 2021;16:1369-78. [Crossref] [PubMed]
35. Tang Q, Chen Y, Li X, et al. The role of PD-1/PD-L1 and application of immune-checkpoint inhibitors in human cancers. Front Immunol 2022;13:964442. [Crossref] [PubMed]
36. Tong Y, Long Y, Zhang F, et al. Advances in Pseudoprogression of Immune Checkpoint Inhibitors in Non-small Cell Lung Cancer. Zhongguo Fei Ai Za Zhi. 2024;27:306-20. [Crossref] [PubMed]
37. Li G, Choi JE, Kryczek I, et al. Intersection of immune and oncometabolic pathways drives cancer hyperprogression during immunotherapy. Cancer Cell 2023;41:304-322.e7. [Crossref] [PubMed]
38. Dimitriou F, Cheng PF, Saltari A, et al. A targetable type III immune response with increase of IL-17A expressing CD4(+) T cells is associated with immunotherapy-induced toxicity in melanoma. Nat Cancer 2024;5:1390-408. [Crossref] [PubMed]
39. Wang S, Cai M, Xiong Y, et al. Grade ≥ 3 hematologic adverse events of immunotherapy in advanced NSCLC patients: a systematic review and meta-analysis. Eur J Clin Pharmacol 2025;81:479-93. [Crossref] [PubMed]
40. Du Q, Jiang G, Li S, et al. Docetaxel increases the risk of severe infections in the treatment of non-small cell lung cancer: a meta-analysis. Oncoscience 2018;5:220-38. [Crossref] [PubMed]
41. Redelman-Sidi G, Michielin O, Cervera C, et al. ESCMID Study Group for Infections in Compromised Hosts (ESGICH) Consensus Document on the safety of targeted and biological therapies: an infectious diseases perspective (Immune checkpoint inhibitors, cell adhesion inhibitors, sphingosine-1-phosphate receptor modulators and proteasome inhibitors). Clin Microbiol Infect 2018;24:S95-S107. [Crossref] [PubMed]
42. Morelli T, Fujita K, Redelman-Sidi G, et al. Infections due to dysregulated immunity: an emerging complication of cancer immunotherapy. Thorax 2022;77:304-11. [Crossref] [PubMed]
43. Luo YH, Shen CI, Chiang CL, et al. Dynamic immune signatures of patients with advanced non-small-cell lung cancer for infection prediction after immunotherapy. Front Immunol 2024;15:1269253. [Crossref] [PubMed]
44. Amjad MT, Chidharla A, Kasi A. Cancer Chemotherapy. 2023 Feb 27. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025.
45. Andrews LP, Butler SC, Cui J, et al. LAG-3 and PD-1 synergize on CD8(+) T cells to drive T cell exhaustion and hinder autocrine IFN-γ-dependent anti-tumor immunity. Cell 2024;187:4355-4372.e22. [Crossref] [PubMed]
46. Koyama S, Akbay EA, Li YY, et al. Adaptive resistance to therapeutic PD-1 blockade is associated with upregulation of alternative immune checkpoints. Nat Commun 2016;7:10501. [Crossref] [PubMed]
47. Alifu M, Tao M, Chen X, et al. Checkpoint inhibitors as dual immunotherapy in advanced non-small cell lung cancer: a meta-analysis. Front Oncol 2023;13:1146905. [Crossref] [PubMed]
48. Yin Q, Wu L, Han L, et al. Immune-related adverse events of immune checkpoint inhibitors: a review. Front Immunol 2023;14:1167975. [Crossref] [PubMed]
49. Sharpe AH, Pauken KE. The diverse functions of the PD1 inhibitory pathway. Nat Rev Immunol 2018;18:153-67. [Crossref] [PubMed]
50. Pauken KE, Dougan M, Rose NR, et al. Adverse Events Following Cancer Immunotherapy: Obstacles and Opportunities. Trends Immunol 2019;40:511-23. [Crossref] [PubMed]
51. Siwicki M, Gort-Freitas NA, Messemaker M, et al. Resident Kupffer cells and neutrophils drive liver toxicity in cancer immunotherapy. Sci Immunol 2021;6:eabi7083. [Crossref] [PubMed]
52. Chhabra N, Kennedy J. A Review of Cancer Immunotherapy Toxicity: Immune Checkpoint Inhibitors. J Med Toxicol 2021;17:411-24. [Crossref] [PubMed]
53. Vetter TR. Systematic Review and Meta-analysis: Sometimes Bigger Is Indeed Better. Anesth Analg 2019;128:575-83. [Crossref] [PubMed]
54. Lin L, Chu H. Quantifying publication bias in meta-analysis. Biometrics 2018;74:785-94. [Crossref] [PubMed]