Predictive factors for the efficacy of immune checkpoint inhibitors in advanced non-small cell lung cancer: a retrospective study

Introduction

Lung cancer is a highly prevalent and fatal type of tumor encountered throughout the world (1), with non-small cell lung cancer (NSCLC) being the predominant subtype. In recent years, immune checkpoint inhibitors (ICI), particularly programmed cell death protein and programmed death-ligand 1 (PD-1/PD-L1) inhibitors, have significantly improved the prognosis of patients with driver-negative advanced NSCLC. Multiple phase III randomized controlled trials (RCTs) targeting advanced NSCLC have demonstrated that PD-1/PD-L1 inhibitors prolong progression-free survival (PFS) and overall survival (OS) as compared to chemotherapy (2-6). Moreover, combining PD-1/PD-L1 inhibitors with chemotherapy with or without antiangiogenic agents can further enhance treatment efficacy (7-22).

Despite the significant efficacy achieved by ICI in treating NSCLC, several unresolved issues remain, including the identification of predictive factors for treatment efficacy. For instance, the efficacy of PD-1/PD-L1 inhibitors varies among NSCLC patients with different driver gene alterations. Patients with epidermal growth factor receptor (EGFR) mutations or anaplastic lymphoma kinase (ALK) rearrangements tend to exhibit a poor response (23-25), while those with Kirsten rat sarcoma (KRAS) mutations show comparable efficacy to patients with wild-type KRAS (24,26,27). Additionally, the sensitivity to PD-1/PD-L1 inhibitors differs between individuals with rare gene alterations (24,25). Furthermore, various other biological markers, such as PD-L1 expression and tumor mutational burden (TMB), may influence the efficacy of PD-1/PD-L1 inhibitors (28,29). Clinical factors, including a high number of metastatic sites (30), specific metastatic locations such as bone metastasis (31), occurrence of immune-related adverse events (irAEs) (32-34), and the use of antibiotics during initial treatment (35), may also potentially impact the efficacy of immunotherapy. Given the growing application of PD-1/PD-L1 inhibitors in the management of advanced NSCLC, it is essential to summarize the real-world data that can account for the greater complexity of patient characteristics as compared to those examined in clinical trial settings.

This study thus aimed to retrospectively analyze the efficacy, survival outcomes, and incidence irAEs in patients with advanced NSCLC treated with PD-1/PD-L1 monotherapy or in combination with other antitumor drugs at our institution. Additionally, we aimed to identify the factors associated with treatment efficacy. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-868/rc).

Methods

Inclusion and exclusion criteria

The inclusion criteria for patients were as follows: pathologically diagnosed with NSCLC and ineligible for surgery or curative radiotherapy at stage III or IV, administration of at least one cycle of PD-1/PD-L1 inhibitor treatment, and completion of efficacy evaluation. Meanwhile, patients who underwent curative surgery after immunotherapy were excluded.

Data collection

Retrospective data collection was conducted from March 2019 to June 2023 at the Department of Medical Oncology, Peking University First Hospital. All data were collected from medical records and via telephone follow-up and included the following: (I) patient demographics, including gender, age, smoking history, Eastern Cooperative Oncology Group performance status (ECOG PS), and date of death; (II) tumor characteristics, including histological type, stage, lesions confined to the thoracic cavity, site of tumor metastasis, status of driver genes, tumor cell proportion score (TPS), and tumor mutation burden (TMB); and (III) treatment-related information, including start and end dates of immunotherapy, progression time, use of antibiotics in the month before and after immunotherapy, line of immunotherapy, combination antitumor medications including chemotherapy or antiangiogenic targeted therapy, history of radiation therapy, and occurrence of irAEs. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and was approved by the Ethics Committee of Peking University First Hospital (ethical approval No. 2024-035-001). The requirement for individual consent was waived due to the retrospective nature of this study.

Efficacy assessment and outcomes definition

The efficacy assessment was performed based on the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1. The evaluation included overall response rate (ORR), disease control rate (DCR), and PFS. Overall response was defined as a complete response (CR), partial response (PR), stable disease (SD), or progressive disease (PD). ORR was calculated as follows: (CR + PR)/(all evaluated cases) × 100%. Meanwhile, DCR was calculated as follows: (CR + PR + SD)/(all evaluated cases) × 100%. PFS was considered to be the time from the initiation of ICI treatment to disease progression or death. For patients temporarily or permanently discontinued immunotherapy due to irAEs, PFS was still defined as the time from the initiation of immunotherapy to disease progression or death, rather than to treatment discontinuation. Patients were classified as immunotherapy sensitive (PFS ≥24.0 weeks) or non-sensitive (PFS <24.0 weeks) at a cutoff of 24.0 weeks (36-38). OS was defined as the time from the start of ICI treatment to death.

IrAEs assessment

The assessment of irAEs was conducted according to the American Society of Clinical Oncology (ASCO) guideline for irAEs (39).

Statistical analyses

All statistical analyses were performed with SPSS software version 23.0 (IBM Corp., Armonk, NY, USA). Descriptive statistics and tables were used to summarize the baseline characteristics, response rates, PFS, and irAEs evaluations. Median PFS was estimated using Kaplan-Meier curves. The associations between potential risk factors and PFS were assessed using univariate Cox regression analysis. Factors with a P value <0.10 in the univariate analysis were selected for multivariate Cox regression analysis (stepwise regression). A P value <0.05 was considered statistically significant. The hazard ratios (HRs) from Cox regression and the corresponding 95% confidence intervals (CIs) are reported. Logistic regression analysis was used to examine the distribution differences of potential risk factors between the immunotherapy-sensitive and non-sensitive groups, as well as the associations of these risk factors with ORR and DCR. Similar to the PFS analysis, logistic regression analysis was initially performed with univariate analysis. Factors with a P value <0.10 were selected for further multivariate analysis. A P value <0.05 was considered statistically significant. The odds ratios (OR) from logistic regression and their corresponding 95% CIs are reported. To minimize immortal time bias, landmark analyses were conducted to assess the association between efficacy and irAEs in patients who received ICI treatment for at least 6, 12, and 18 weeks. The figures were drawn by GraphPad Prism.

Results

Baseline characteristics

A total of 123 patients were included in the study, and their baseline demographic and clinical characteristics are shown in Table 1.

The median age was 65 years, and there was a higher proportion of males (74.8%). The majority (63.4%) had a smoking history, and 33.3% of patients had an ECOG PS score of 2–3. Adenocarcinoma accounted for the highest proportion (62.6%) of cases, followed by squamous cell carcinoma (33.3%). Stage IV was the predominant stage (81.3%), approximately 52.8% of patients had extrathoracic metastases at the start of immunotherapy, and 41.5% of patients had metastases involving three or more organs, with bone metastasis accounting for 35.8%. Genetic testing for driver gene alterations was performed in 105 patients: wild type was the most common (44.8%), followed by EGFR activating mutation (21.0%) and KRAS mutation (17.1%). Other rare mutations included mesenchymal-epithelial transition (MET), human epidermal growth factor 2 (HER2), v-raf murine viral oncogene homolog B1 (BRAF), rearranged during transfection (RET), ALK, and ROS proto-oncogene1 (ROS1) alterations. Among the 18 patients without genetic testing, 15 had squamous cell carcinoma. TPS was evaluated in 96 patients, with scores of <1%, 1–49%, and ≥50% accounting for 34.4%, 37.5%, and 28.1% of patients, respectively. TMB was assessed in 29 patients: the median number of mutations per megabase was 5.98, and 31.0% of patients had ≥10 mutations per megabase.

Treatment

Among the patients, 61.8%, 21.1%, and 17.1% received first-line, second-line, or third-line-and-beyond treatment, respectively. Patients received four immunotherapy regimens: chemotherapy combined with immunotherapy (80 cases), chemotherapy combined with immunotherapy and anti-angiogenesis therapy (17 cases), immunotherapy combined with anti-angiogenesis therapy (11 cases), or single-agent immunotherapy (15 cases). Overall, 78.9% of patients received immunotherapy combined with chemotherapy (± antiangiogenic agents), while 21.1% received immunotherapy alone or with antiangiogenic agents (without chemotherapy). Radiotherapy was administered to 52.0% of patients at different stages of treatment. Whether antibiotics were used within 1 month before and after immunotherapy was recorded for 109 patients, among whom 19.3% were administered antibiotics (Table 1).

Efficacy and survival

The median follow-up time was 50.0 weeks (12.5 months). The percentages of patients achieving CR or PR, SD, DCR, and PD were 42.3%, 24.4%, 66.7%, and 33.3% respectively (Table 1). Regarding, PFS assessment, 72.4% of patients experienced disease progression, and the median PFS was 24.0 weeks (Table 1 and Figure 1). The PFS rates at 1 and 2 years were 30.5% and 17.1%, respectively. The 118 patients with a PFS ≥24 weeks or those who had progressed or died of tumor progression before 24 weeks were categorized into immunotherapy-sensitive and non-sensitive groups based on the 24-week PFS threshold (36-38), with proportions of 49.2% and 50.8%, respectively (Table 1).

Figure 1 Kaplan-Meier analysis of PFS in all patients. The mPFS was 24 weeks. mPFS, median progression-free survival; PFS, progression-free survival.

OS assessment was not conducted as only 37.4% of the patients died.

Analysis of factors associated with efficacy

Firstly, according to univariate Cox regression analysis for PFS, there were 11 factors associated with PFS (P<0.10). Multivariate Cox regression analysis indicated that ECOG PS 2–3 (HR =4.496; P<0.001), non-KRAS driver alterations (HR =2.040; P=0.01), TPS ≥50% (HR =0.276; P<0.001), and the occurrence of irAEs (HR =0.435; P=0.002) were significantly associated with PFS (Table 2 and Figure 2).

Figure 2 Kaplan-Meier analysis of PFS in the different subgroups. Correlation between different factors and PFS: (A) ECOG PS scores, (B) the occurrence of irAEs, (C) driver gene alterations, and (D) TPS levels. ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio; irAEs, immune-related adverse events; KRAS, Kirsten rat sarcoma; mPFS, median progression-free survival; PFS, progression-free survival; TPS, tumor cell proportion score.

Univariate logistic regression analysis for immunotherapy-sensitive (PFS ≥24 weeks) versus non-sensitive (PFS <24 weeks) patients indicated that all the factors mentioned above except radiotherapy had P values <0.10. In the multivariate logistic regression analysis, four of these factors remained significant (Table 3).

In patients who received first-line immunotherapy, Cox regression analysis for PFS identified 7 factors with P<0.10. Multivariate Cox regression analysis showed that ECOG PS 2–3 (HR =2.793; P=0.01), TPS ≥50% (HR =0.278; P=0.004), irAEs occurrence (HR =0.255; P<0.001), and disease stage (HR =2.503; P=0.050) were significantly associated with PFS (Table S1).

Secondly, according to the univariate logistic regression analysis for ORR, there were 9 factors associated with ORR (P<0.10). Multivariate logistic regression analysis identified 3 factors that were significantly associated with ORR: ECOG PS 2–3 (OR =0.286; P=0.02), third-line-or-beyond therapy (OR =0.069; P=0.01), and irAEs (OR =3.541; P=0.009) (Table 4).

Thirdly, according to the univariate logistic regression analysis for DCR, there were 12 factors associated with DCR (P<0.10). Multivariate logistic regression analysis revealed that ECOG PS 2–3 (OR =0.107; P=0.001), lesions limited to the thorax (OR =4.179; P=0.03), TPS ≥50% (OR =7.82; P=0.02), and irAEs (OR =7.122, P=0.005) were significantly associated with DCR (Table 5).

Fourthly, in the landmark analysis for the association of irAEs with efficacy, the efficacy of ICI therapy for patients with durations of ≥6, ≥12, and ≥18 weeks, respectively, was determined (Figure 3 and Table 6).

Figure 3 Landmark analysis of the correlation between occurrence of irAEs and PFS. Patients with ICI treatment durations of (A) ≥6 weeks, (B) ≥12 weeks, and (C) ≥18 weeks. HR, hazard ratio; ICI, immune checkpoint inhibitor; irAEs, immune-related adverse events; mPFS, median progression-free survival; PFS, progression-free survival.

For patients treated for ≥6 weeks (n=107), univariate Cox analysis indicated that irAEs were associated with a longer PFS (HR =0.419; P<0.001), and logistic regression analysis indicated that irAEs were associated with a higher ORR (OR =3.324; P=0.003) and DCR (OR =8.675; P=0.001).

For patients treated for ≥12 weeks (n=83), univariate Cox analysis indicated that irAEs were associated with a longer PFS (HR =0.560; P=0.03); however, logistic regression analysis indicated no statistical difference in ORR (53.85% vs. 68.18%; P=0.18) or DCR (94.87% vs. 97.73%; P=0.50) between the non-irAEs group and irAEs group, but numerically better results were observed in the irAEs group.

for patients treated for ≥18 weeks (n=66), there were no statistically significant differences in PFS (HR =0.610; P=0.10), ORR (58.62% vs. 70.27%; P=0.33), or DCR (93.10% vs. 97.30%; P=0.43) between the non-irAE group and irAE group, but numerically better results were observed in the irAE group.

irAE analysis

irAEs occurred in 49 (39.8%) of cases. Skin irAEs had the highest incidence rate (22.0%), mainly presenting as rash and/or itching. The next most common irAEs were thyroid dysfunction (13.0%) and pneumonia (12.2%). Rare irAEs included hypophysitis and cardiac toxicities (3.3% each), arthritis and renal impairment (2.4% each), colitis and primary adrenal insufficiency (1.6% each), and myositis (0.8%). Immunotherapy was temporarily paused or switched to other anticancer drugs in 20 (16.3%) cases due to irAEs, including 10 cases of pneumonia, 3 cases each of severe rash and cardiac toxicity, and 1 case each of renal toxicity, recurrent ankylosing spondylitis, dermatomyositis recurrence, and enteritis. Steroid treatment for irAEs was administered in 11 (8.9%) cases, including 5 cases of pneumonia and 2 cases of severe rash, with 1 case each of nephritis, enteritis, myocarditis, and dermatomyositis recurrence. No deaths due to irAEs were reported.

Discussion

This study summarized the real-world data on the efficacy and irAEs of monotherapy or combination therapy with PD-1/PD-L1 inhibitors in the treatment of patients with advanced NSCLC and analyzed the factors associated with ICI efficacy.

In the overall population, the median PFS was 24.0 weeks, with an ORR of 42.3% and a DCR of 66.7%. Multiple phase III RCTs have reported PFS ranging from 2.3 to 5.8 months, an ORR of 14.0–33.0%, and a DCR of 49.0–66.0% for first-line or second-line ICI monotherapy (2-6). First-line ICI in combination with chemotherapy have been found to yield a PFS of 5.5 to 11 months, an ORR of 43.0–75.0%, and a DCR of 81.0–90.0% (7-20). Moreover, combination of first-line ICI with chemotherapy and bevacizumab were shown to provide a PFS of 8.3–12.1 months, an ORR of 61.5–63.5%, and a DCR of 85.3–87.3% (21,22). These RCTs had consistent baseline characteristics, such as ECOG PS scores of 0–1, and most studies excluded patients with EGFR mutation and ALK rearrangement who are less responsive to PD-1/PD-L1 inhibitors. Additionally, there was consistency in the administration of the ICI in these RCTs. In contrast, real-world studies involve a variability in characteristics. In our study, 33.3% of patients had an ECOG PS score of 2–3, 21.0% had EGFR gene mutations, and 17.1% received third-line-beyond treatment, with ICI administration being highly diverse. Nevertheless, the PFS, ORR, and DCR obtained fell within the range of data reported in the RCTs.

This real-world study enrolled 41 patients with ECOG PS 2–3, among whom 27 received chemotherapy-containing regimens. Chemotherapy was used because these patients had factors associated with poor prognosis when receiving immunotherapy alone: including driver gene mutations, low or negative PD-L1 expression, high tumor burden; additionally, in some patients with bone metastases, PS deterioration was caused by mobility impairment. Younger age, which enhances patients’ tolerance to chemotherapy, is also a reason for our choice to use chemotherapy. All these patients received reduced chemotherapy doses, and active supportive care was provided to ensure their safety. We found that an ECOG PS score of 2–3 was the most significant negative prognostic factor for ICI efficacy. This is consistent with in the relevant literature. RCTs of immunotherapy generally exclude patients with PS scores ≥2, and a few prospective phase II studies have indicated inferior efficacy in patients with PS scores of 2 compared to those with scores of 0–1 (40,41). In 2021, Facchinetti et al. conducted a meta-analysis of patients with advanced NSCLC receiving ICI monotherapy and a TPS ≥50%. They reported that an ECOG PS score of 2 predicted poorer efficacy and shorter survival time (42). Subsequent large-sample, multicenter retrospective studies have also reached similar conclusions (43). However, a recently published RCT comparing single-agent immunotherapy and single-agent chemotherapy in patients with advanced NSCLC with an ECOG PS score of 2–3 found that immunotherapy conferred survival advantage over chemotherapy (44). Research into patients with ECOG PS scores ≥2 receiving combination chemotherapy and immunotherapy are relatively scarce. A large-sample retrospective study found worse survival outcomes after a first-line combination of chemotherapy and immunotherapy (45). One prospective phase II study compared the efficacy of ICI monotherapy to that of immunotherapy combined with low-dose chemotherapy and reported higher response rates in the combination treatment group (7/10, 70.0%) as compared to the monotherapy group (2/10, 20.0%) (46). Additionally, the efficacy of immunotherapy in patients with PS scores of 2 may vary depending on different clinical conditions. For example, patients with PS scores of 2 not attributable to malignancy may still benefit from immunotherapy (47), while those with significant respiratory symptoms may experience rapid tumor progression after ICI monotherapy (41).

We identified the occurrence of irAEs as a positive predictive factor for superior efficacy in our analysis. Numerous retrospective studies and meta-analyses on a variety of tumor types have also reported this association (32-34). However, due to the correlation between the occurrence of irAEs and time, conventional statistical analysis may exaggerate the association of irAEs with efficacy. This phenomenon is known as immortal time bias. To reduce this bias, researchers commonly employ time-dependent Cox regression or landmark analysis, with the latter being more common.

Dall’Olio et al. conducted a meta-analysis involving various solid tumors and compared the results between landmark and non-landmark approaches. The study found that the association between irAEs and outcomes remained significant across both approaches, but the effect size was smaller in landmark approach (34). Socinski et al. performed a retrospective analysis of three phase III studies examining the relationship between irAEs and OS. They used both time-dependent Cox regression and landmark analysis, and both methods showed that irAEs could predict a longer OS (33). In our study, landmark analysis was also conducted for patients with an ICI treatment duration of ≥6-, 12-, and 18-week, respectively. The results showed that irAEs were predictive or tended to be associated with improved efficacy at each time point, but the correlation was weaker than that in the non-landmark analysis. This suggests that immortal time bias indeed affects the predictive value of irAEs to some extent. The mechanisms underlying the predictive value of irAEs for ICI efficacy remain unclear. One possible mechanism is the presence of shared antigens between tumor cells and normal cells (48).

Another predictive factor for superior ICI efficacy in our multivariate analysis was high PD-L1 expression: patients with TPS ≥50% had a longer PFS and a higher DCR. These findings are consistent with those of several RCTs that have been referenced in authoritative guidelines (2-4,7-18,20,49-51). RCTs on ICI monotherapy have demonstrated PD-L1 expression to be an important factor in predicting efficacy and survival in patients with advanced NSCLC (2-4). When PD-1/PD-L1 inhibitors are combined with chemotherapy, the correlation weakens. Nonetheless, patients with PD-L1 <1% still experience superior efficacy compared to chemotherapy, and efficacy and survival time increase with higher levels of PD-L1 expression (7-18,20). Although our retrospective study from real-world data included both monotherapy and various combination regimens, it results were consistent with those from RCTs, indicating that high PD-L1 expression is a relatively stable predictor of PD-1/PD-L1 inhibitor efficacy in both clinical research and practice.

In the multivariate analysis, we further found that oncogenic driver alterations, except for KRAS mutation, were associated with a shorter PFS. The relationship between driver alterations and ICI efficacy is complex. Studies have suggested that EGFR mutations and ALK rearrangements are predictive of poor efficacy for monotherapy with PD-1/PD-L1 inhibitors (23,24). The subgroup analysis of patients with EGFR mutations in the IMPOWER150 study, a phase III clinical trial of chemotherapy combined with bevacizumab and PD-L1 inhibitor for advanced NSCLC, showed that this more aggressive combination therapy yielded a higher ORR and PFS as compared to chemotherapy combined with bevacizumab, with somewhat improved OS (52). The ORIENT-31 study, a phase III RCT specifically conducted in patients with EGFR mutations, also found that combination therapy with PD-1 inhibitors provided superior ORR and PFS as compared to chemotherapy, but not an improvement in OS (53). However, in the IMPOWER151 study, reported at the 2023 World Congress of Lung Cancer (WCLC), the analysis of the EGFR mutation subgroup did not indicate superior efficacy of chemotherapy combined with PD-1 inhibitors and bevacizumab as compared to chemotherapy combined with bevacizumab (54). The CheckMate722 and KEYNOTE789 studies, two phase III RCTs targeting patients with EGFR mutation-positive advanced NSCLC, similarly reported no significant or only marginal improvement of PFS and OS with chemotherapy combined with PD-1 inhibitors as compared to chemotherapy (55). Therefore, there is sufficient evidence to suggest that, overall, EGFR mutations are associated with insensitivity to ICI, which may be attributed to an inactive tumor immune microenvironment (25). On the other hand, several studies have shown that patients with KRAS mutations, as compared to wild-type patients, do not experience inferior efficacy with immunotherapy and in fact may receive better efficacy (24,26,27), which may be related to their more active tumor immune microenvironment (56). Other rare oncogenic driver alterations (e.g., HER2, MET exon 14 skipping, BRAF mutations, and RET, ALK, and ROS1 rearrangements) have varying levels of PD-L1 expression. BRAF and MET mutations exhibit higher PD-L1 expression compared to others, while RET and ROS1 rearrangements remain unclear due to their low incidence (25). The TMB is generally low for other rare diver alterations, with the BRAF V600E mutation being the exception (25). The retrospective, multicenter IMMUNOTARGET study found that patients with KRAS and BRAF mutations had a similar ORR to those with wild-type genes, while patients with EGFR, HER2, MET mutations, as well as ALK, RET, and ROS1 rearrangements, had a lower ORR (24). Combination therapy with chemotherapy and immunotherapy ± bevacizumab may improve the efficacy for these patients (57). Further expansion of the sample size may provide a clearer understanding of the precise efficacy in different subpopulations with rare driver alterations.

In our study, we found that patients who received third-line-or-beyond immunotherapy had a lower ORR compared with those receiving first- and second-line therapies in the multivariate analysis. The majority (17 out of 21 cases) of patients having driver gene alterations other than KRAS mutations and a higher proportion of patients with a PS 2–3 (10 out of 21 cases) may be potential contributing factors.

Multiple phase III clinical studies on PD-1/PD-L1 inhibitor monotherapy in patients with advanced NSCLC reported incidence rates of overall and severe irAEs ranging from 26.1% to 74.9% and from 10.2% to 15.7%, respectively (2-6); among other studies on chemotherapy combined with PD-1/PD-L1 inhibitor therapy, the corresponding incidence rates were from 18.9% to 74.0% and from 2.9% to 15.6%, respectively (7-9,11,13-16,18-20). This high degree of variability in the data might be due to difficulties in distinguishing between irAEs and tumor-related symptoms, as well as adverse reactions caused by other anticancer drugs. The incidence of irAEs appears to be higher under the combination of chemotherapy with bevacizumab and PD-1/PD-L1 inhibitors, with the overall incidence rates ranging from 77.4% to 82.8% and grade ≥3 irAE incidence ranging from 14.1% to 24.8% (21,22). In contrast to clinical trials, our study included a diverse range of treatment patterns for PD-1/PD-L1 inhibitors, including monotherapy or combination with chemotherapy or antiangiogenic agents. The overall incidence rate of irAEs, at 39.8%, reflects real-world data and falls within the range of results reported in RCTs. The most common irAEs observed were rash and/or pruritus, thyroid dysfunction, and pneumonia, which is consistent with the results reported in most clinical trials. We found that checkpoint inhibitor pneumonia (CIP) was the most common type of irAEs that led to treatment discontinuation and required glucocorticoid therapy. This is consistent with previous research findings: one meta-analysis reported that patients with NSCLC are more prone to developing CIP than are those with other tumor types, with the CIP also tending to be of higher grade (58). CIP is among the irAEs associated with high morbidity and mortality (48). In our patient cohort, there were no deaths attributed to irAEs. Close follow-up, timely diagnosis, prompt discontinuation of immunotherapy, and the use of glucocorticoids are crucial for ensuring safety. A limitation of this real-world study in the context of irAEs is the failure to perform toxicity grading, while the grading of irAEs remains a critical criterion for assessing the severity of treatment-related toxicity in immunotherapy. Additionally, further analysis of risk factors associated with the occurrence of high-grade irAEs also holds important clinical value.

This single-center, real-world study demonstrated that the efficacy and safety of PD-1/PD-L1 inhibitors as monotherapy or in combination with other drugs are consistent with the data reported in RCTs. Factors such as an ECOG PS score of 2–3, non-KRAS driver gene alterations, TPS ≥50%, and the occurrence of irAEs were significantly associated with PFS. ECOG PS scores of 2–3, occurrence of irAEs, and third-line or later immunotherapy were significantly associated with ORR. Landmark analysis also indicated that irAEs could predict better immunotherapy outcomes.

This study involved several limitations. First, as we employed a retrospective design, the clinical data collected were not sufficiently precise; for example, data on irAE grade were absent. Second, the inclusion of patients with heterogeneous clinical conditions and tumor characteristics, along with a limited sample size, made accurate analyses of specific patient populations with distinct features difficult. Additionally, due to the low number of deaths, an analysis related to OS could not be conducted.

Conclusions

This single-center real-world study found that the efficacy and safety of PD-1/PD-L1 inhibitors as monotherapy or in combination with other drugs are consistent with the data reported in RCTs. Factors such as ECOG PS scores of 2–3, non-KRAS driver gene alterations, TPS ≥50%, and the occurrence of irAEs were significantly associated PFS. An ECOG PS score of 2–3, occurrence of irAEs, and third-line-or-beyond immunotherapy were significantly associated with the ORR. Landmark analysis further demonstrated that the occurrence of irAEs was correlated with better efficacy of immunotherapy.

Acknowledgments

We appreciate all the patients and their families who participated in this study.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-868/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-868/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-868/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-2025-868/coif). T.U. received payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Amgen, AstraZeneca, Boehringer Ingelheim, Chugai, Daiichi Sankyo, Eki Lilly, Kyowa Kirin, MSD, Norartis Pharma, Ono Pharmaceutical, Taiho, Takeda. The other 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. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and was approved by the Ethics Committee of Peking University First Hospital (ethical approval No. 2024-035-001). The requirement for individual consent was waived due to the retrospective nature of this study.

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. Leiter A, Veluswamy RR, Wisnivesky JP. The global burden of lung cancer: current status and future trends. Nat Rev Clin Oncol 2023;20:624-39. [Crossref] [PubMed]
2. de Castro G Jr, Kudaba I, Wu YL, et al. Five-Year Outcomes With Pembrolizumab Versus Chemotherapy as First-Line Therapy in Patients With Non-Small-Cell Lung Cancer and Programmed Death Ligand-1 Tumor Proportion Score ≥ 1% in the KEYNOTE-042 Study. J Clin Oncol 2023;41:1986-91. [Crossref] [PubMed]
3. Jassem J, de Marinis F, Giaccone G, et al. Updated Overall Survival Analysis From IMpower110: Atezolizumab Versus Platinum-Based Chemotherapy in Treatment-Naive Programmed Death-Ligand 1-Selected NSCLC. J Thorac Oncol 2021;16:1872-82. [Crossref] [PubMed]
4. 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]
5. Mazieres J, Rittmeyer A, Gadgeel S, et al. Atezolizumab Versus Docetaxel in Pretreated Patients With NSCLC: Final Results From the Randomized Phase 2 POPLAR and Phase 3 OAK Clinical Trials. J Thorac Oncol 2021;16:140-50. [Crossref] [PubMed]
6. 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]
7. Garassino MC, Gadgeel S, Speranza G, et al. Pembrolizumab Plus Pemetrexed and Platinum in Nonsquamous Non-Small-Cell Lung Cancer: 5-Year Outcomes From the Phase 3 KEYNOTE-189 Study. J Clin Oncol 2023;41:1992-8. [Crossref] [PubMed]
8. West H, McCleod M, Hussein M, et al. Atezolizumab in combination with carboplatin plus nab-paclitaxel chemotherapy compared with chemotherapy alone as first-line treatment for metastatic non-squamous non-small-cell lung cancer (IMpower130): a multicentre, randomised, open-label, phase 3 trial. Lancet Oncol 2019;20:924-37. [Crossref] [PubMed]
9. Nishio M, Barlesi F, West H, et al. Atezolizumab Plus Chemotherapy for First-Line Treatment of Nonsquamous NSCLC: Results From the Randomized Phase 3 IMpower132 Trial. J Thorac Oncol 2021;16:653-64. [Crossref] [PubMed]
10. Zhang L, Wang Z, Fang J, et al. Final overall survival data of sintilimab plus pemetrexed and platinum as First-Line treatment for locally advanced or metastatic nonsquamous NSCLC in the Phase 3 ORIENT-11 study. Lung Cancer 2022;171:56-60. [Crossref] [PubMed]
11. Lu S, Wang J, Yu Y, et al. Tislelizumab plus chemotherapy as first-line treatment of locally advanced or metastatic nonsquamous non-small-cell lung cancer (final analysis of RATIONALE-304: a randomized phase III trial). ESMO Open 2024;9:103728. [Crossref] [PubMed]
12. Zhou C, Chen G, Huang Y, et al. Camrelizumab plus carboplatin and pemetrexed as first-line therapy for advanced non-squamous non-small-cell lung cancer: 5-year outcomes of the CameL randomized phase 3 study. J Immunother Cancer 2024;12:e009240. [Crossref] [PubMed]
13. Novello S, Kowalski DM, Luft A, et al. Pembrolizumab Plus Chemotherapy in Squamous Non-Small-Cell Lung Cancer: 5-Year Update of the Phase III KEYNOTE-407 Study. J Clin Oncol 2023;41:1999-2006. [Crossref] [PubMed]
14. Jotte R, Cappuzzo F, Vynnychenko I, et al. Atezolizumab in Combination With Carboplatin and Nab-Paclitaxel in Advanced Squamous NSCLC (IMpower131): Results From a Randomized Phase III Trial. J Thorac Oncol 2020;15:1351-60. [Crossref] [PubMed]
15. Zhou C, Wu L, Fan Y, et al. Sintilimab Plus Platinum and Gemcitabine as First-Line Treatment for Advanced or Metastatic Squamous NSCLC: Results From a Randomized, Double-Blind, Phase 3 Trial (ORIENT-12). J Thorac Oncol 2021;16:1501-11. [Crossref] [PubMed]
16. Wang J, Lu S, Yu X, et al. Tislelizumab plus chemotherapy versus chemotherapy alone as first-line treatment for advanced squamous non-small-cell lung cancer: final analysis of the randomized, phase III RATIONALE-307 trial. ESMO Open 2024;9:103727. [Crossref] [PubMed]
17. Zhou C, Ren S, Chen J, et al. First-line (1L) camrelizumab plus chemotherapy (chemo) for advanced squamous non-small cell lung cancer (sqNSCLC): 4-yr update from the phase III CameL-sq trial. European Lung Cancer Congress; 2024:62P. Available online: https://oncologypro.esmo.org/meeting-resources/european-lung-cancer-congress-2024/first-line-1l-camrelizumab-plus-chemotherapy-chemo-for-advanced-squamous-non-small-cell-lung-cancer-sqnsclc-4-yr-update-from-the-phase-iii-c
18. Makharadze T, Gogishvili M, Melkadze T, et al. Cemiplimab Plus Chemotherapy Versus Chemotherapy Alone in Advanced NSCLC: 2-Year Follow-Up From the Phase 3 EMPOWER-Lung 3 Part 2 Trial. J Thorac Oncol 2023;18:755-68. [Crossref] [PubMed]
19. Zhong J, Fei K, Wu L, et al. Toripalimab plus chemotherapy for first line treatment of advanced non-small cell lung cancer (CHOICE-01): final OS and biomarker exploration of a randomized, double-blind, phase 3 trial. Signal Transduct Target Ther 2024;9:369. [Crossref] [PubMed]
20. Zhou C, Wang Z, Sun M, et al. Interim survival analysis of the randomized phase III GEMSTONE-302 trial: sugemalimab or placebo plus chemotherapy as first-line treatment for metastatic NSCLC. Nat Cancer 2023;4:860-71. [Crossref] [PubMed]
21. Socinski MA, Nishio M, Jotte RM, et al. IMpower150 Final Overall Survival Analyses for Atezolizumab Plus Bevacizumab and Chemotherapy in First-Line Metastatic Nonsquamous NSCLC. J Thorac Oncol 2021;16:1909-24. [Crossref] [PubMed]
22. Kim HR, Sugawara S, Lee JS, et al. First-line nivolumab, paclitaxel, carboplatin, and bevacizumab for advanced non-squamous non-small cell lung cancer: Updated survival analysis of the ONO-4538-52/TASUKI-52 randomized controlled trial. Cancer Med 2023;12:17061-7. [Crossref] [PubMed]
23. Zhou N, Leung CH, William WN Jr, et al. Impact of select actionable genomic alterations on efficacy of neoadjuvant immunotherapy in resectable non-small cell lung cancer. J Immunother Cancer 2024;12:e009677. [Crossref] [PubMed]
24. Mazieres J, Drilon A, Lusque A, et al. Immune checkpoint inhibitors for patients with advanced lung cancer and oncogenic driver alterations: results from the IMMUNOTARGET registry. Ann Oncol 2019;30:1321-8. [Crossref] [PubMed]
25. Dantoing E, Piton N, Salaün M, et al. Anti-PD1/PD-L1 Immunotherapy for Non-Small Cell Lung Cancer with Actionable Oncogenic Driver Mutations. Int J Mol Sci 2021;22:6288. [Crossref] [PubMed]
26. Noordhof AL, Damhuis RAM, Hendriks LEL, et al. Prognostic impact of KRAS mutation status for patients with stage IV adenocarcinoma of the lung treated with first-line pembrolizumab monotherapy. Lung Cancer 2021;155:163-9. [Crossref] [PubMed]
27. Landre T, Justeau G, Assié JB, et al. Anti-PD-(L)1 for KRAS-mutant advanced non-small-cell lung cancers: a meta-analysis of randomized-controlled trials. Cancer Immunol Immunother 2022;71:719-26. [Crossref] [PubMed]
28. Ricciuti B, Wang X, Alessi JV, et al. Association of High Tumor Mutation Burden in Non-Small Cell Lung Cancers With Increased Immune Infiltration and Improved Clinical Outcomes of PD-L1 Blockade Across PD-L1 Expression Levels. JAMA Oncol 2022;8:1160-8. [Crossref] [PubMed]
29. Giustini N, Bazhenova L. Recognizing Prognostic and Predictive Biomarkers in the Treatment of Non-Small Cell Lung Cancer (NSCLC) with Immune Checkpoint Inhibitors (ICIs). Lung Cancer (Auckl) 2021;12:21-34. [Crossref] [PubMed]
30. Banna GL, Cantale O, Muthuramalingam S, et al. Efficacy outcomes and prognostic factors from real-world patients with advanced non-small-cell lung cancer treated with first-line chemoimmunotherapy: The Spinnaker retrospective study. Int Immunopharmacol 2022;110:108985. [Crossref] [PubMed]
31. Ham A, Lee Y, Kim HS, et al. Real-World Outcomes of Nivolumab, Pembrolizumab, and Atezolizumab Treatment Efficacy in Korean Veterans with Stage IV Non-Small-Cell Lung Cancer. Cancers (Basel) 2023;15:4198. [Crossref] [PubMed]
32. Hussaini S, Chehade R, Boldt RG, et al. Association between immune-related side effects and efficacy and benefit of immune checkpoint inhibitors - A systematic review and meta-analysis. Cancer Treat Rev 2021;92:102134. [Crossref] [PubMed]
33. Socinski MA, Jotte RM, Cappuzzo F, et al. Association of Immune-Related Adverse Events With Efficacy of Atezolizumab in Patients With Non-Small Cell Lung Cancer: Pooled Analyses of the Phase 3 IMpower130, IMpower132, and IMpower150 Randomized Clinical Trials. JAMA Oncol 2023;9:527-35. [Crossref] [PubMed]
34. Dall'Olio FG, Rizzo A, Mollica V, et al. Immortal time bias in the association between toxicity and response for immune checkpoint inhibitors: a meta-analysis. Immunotherapy 2021;13:257-70. [Crossref] [PubMed]
35. Abdelhamid A, Tuminello S, Ivic-Pavlicic T, et al. Antibiotic treatment and survival in non-small cell lung cancer patients receiving immunotherapy: a systematic review and meta-analysis. Transl Lung Cancer Res 2023;12:2427-39. [Crossref] [PubMed]
36. Kluger H, Barrett JC, Gainor JF, et al. Society for Immunotherapy of Cancer (SITC) consensus definitions for resistance to combinations of immune checkpoint inhibitors. J Immunother Cancer 2023;11:e005921. [Crossref] [PubMed]
37. Rizvi N, Ademuyiwa FO, Cao ZA, et al. Society for Immunotherapy of Cancer (SITC) consensus definitions for resistance to combinations of immune checkpoint inhibitors with chemotherapy. J Immunother Cancer 2023;11:e005920. [Crossref] [PubMed]
38. Atkins MB, Ascierto PA, Feltquate D, et al. Society for Immunotherapy of Cancer (SITC) consensus definitions for resistance to combinations of immune checkpoint inhibitors with targeted therapies. J Immunother Cancer 2023;11:e005923. [Crossref] [PubMed]
39. Schneider BJ, Naidoo J, Santomasso BD, et al. Management of Immune-Related Adverse Events in Patients Treated With Immune Checkpoint Inhibitor Therapy: ASCO Guideline Update. J Clin Oncol 2021;39:4073-126. [Crossref] [PubMed]
40. Felip E, Ardizzoni A, Ciuleanu T, et al. CheckMate 171: A phase 2 trial of nivolumab in patients with previously treated advanced squamous non-small cell lung cancer, including ECOG PS 2 and elderly populations. Eur J Cancer 2020;127:160-72. [Crossref] [PubMed]
41. Mark M, Froesch P, Eboulet EI, et al. SAKK 19/17: safety analysis of first-line durvalumab in patients with PD-L1 positive, advanced nonsmall cell lung cancer and a performance status of 2. Cancer Immunol Immunother 2021;70:1255-62. [Crossref] [PubMed]
42. Facchinetti F, Di Maio M, Perrone F, et al. First-line immunotherapy in non-small cell lung cancer patients with poor performance status: a systematic review and meta-analysis. Transl Lung Cancer Res 2021;10:2917-36. [Crossref] [PubMed]
43. Chouaid C, Thomas M, Debieuvre D, et al. Effectiveness of Nivolumab in Second-Line and Later in Patients with Advanced Non-Small Cell Lung Cancer in Real-Life Practice in France and Germany: Analysis of the ESME-AMLC and CRISP Cohorts. Cancers (Basel) 2022;14:6148. [Crossref] [PubMed]
44. Lee SM, Schulz C, Prabhash K, et al. First-line atezolizumab monotherapy versus single-agent chemotherapy in patients with non-small-cell lung cancer ineligible for treatment with a platinum-containing regimen (IPSOS): a phase 3, global, multicentre, open-label, randomised controlled study. Lancet 2023;402:451-63. [Crossref] [PubMed]
45. Waterhouse D, Lam J, Betts KA, et al. Real-world outcomes of immunotherapy-based regimens in first-line advanced non-small cell lung cancer. Lung Cancer 2021;156:41-9. [Crossref] [PubMed]
46. Bonomi M, Ahmed T, Addo S, et al. Circulating immune biomarkers as predictors of the response to pembrolizumab and weekly low dose carboplatin and paclitaxel in NSCLC and poor PS: An interim analysis. Oncol Lett 2019;17:1349-56. [PubMed]
47. Facchinetti F, Mazzaschi G, Barbieri F, et al. First-line pembrolizumab in advanced non-small cell lung cancer patients with poor performance status. Eur J Cancer 2020;130:155-67. [Crossref] [PubMed]
48. Ghanbar MI, Suresh K. Pulmonary toxicity of immune checkpoint immunotherapy. J Clin Invest 2024;134:e170503. [Crossref] [PubMed]
49. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: Non-Small Cell Lung Cancer. Version 3. January 14, 2025. Available online: https://www.nccn.org/guidelines/guidelines-detail?category=1&id=1450
50. Hendriks LE, Kerr KM, Menis J, et al. Non-oncogene-addicted metastatic non-small-cell lung cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol 2023;34:358-76. [Crossref] [PubMed]
51. Hendriks L, Cortiula F, Mariamidze E, et al. ESMO Non-Oncogene-Addicted Metastatic Non-Small-Cell Lung Cancer Living Guideline. Version 1.2. January 2025. Available online: https://www.esmo.org/living-guidelines/esmo-non-oncogene-addicted-metastatic-non-small-cell-lung-cancer-living-guideline
52. Nogami N, Barlesi F, Socinski MA, et al. IMpower150 Final Exploratory Analyses for Atezolizumab Plus Bevacizumab and Chemotherapy in Key NSCLC Patient Subgroups With EGFR Mutations or Metastases in the Liver or Brain. J Thorac Oncol 2022;17:309-23. [Crossref] [PubMed]
53. Lu S, Wu L, Jian H, et al. Sintilimab plus chemotherapy for patients with EGFR-mutated non-squamous non-small-cell lung cancer with disease progression after EGFR tyrosine-kinase inhibitor therapy (ORIENT-31): second interim analysis from a double-blind, randomised, placebo-controlled, phase 3 trial. Lancet Respir Med 2023;11:624-36. [Crossref] [PubMed]
54. Zhou C, Dong X, Chen G, et al. OA09.06 IMpower151: Phase III Study of Atezolizumab + Bevacizumab + Chemotherapy in 1L Metastatic Nonsquamous NSCLC. J Thorac Oncol 2023;18:S64-S65. [Crossref]
55. Qin BD, Jiao XD, Yuan LY, et al. Immunotherapy-based regimens for patients with EGFR-mutated non-small cell lung cancer who progressed on EGFR-TKI therapy. J Immunother Cancer 2024;12:e008818. [Crossref] [PubMed]
56. Xu M, Zhao X, Wen T, et al. Unveiling the role of KRAS in tumor immune microenvironment. Biomed Pharmacother 2024;171:116058. [Crossref] [PubMed]
57. Bylicki O, Tomasini P, Radj G, et al. Atezolizumab with or without bevacizumab and platinum-pemetrexed in patients with stage IIIB/IV non-squamous non-small cell lung cancer with EGFR mutation, ALK rearrangement or ROS1 fusion progressing after targeted therapies: A multicentre phase II open-label non-randomised study GFPC 06-2018. Eur J Cancer 2023;183:38-48. [Crossref] [PubMed]
58. Ma K, Lu Y, Jiang S, et al. The Relative Risk and Incidence of Immune Checkpoint Inhibitors Related Pneumonitis in Patients With Advanced Cancer: A Meta-Analysis. Front Pharmacol 2018;9:1430. [Crossref] [PubMed]

(English Language Editor: J. Gray)