The present study was a single-center, retrospective study. Patients with driver gene-negative advanced lung adenocarcinoma who received ICIs combined with antiangiogenic agents and chemotherapy as first-line treatment at The Sixth Affiliated Hospital (School of Medicine, South China University of Technology, Foshan, China) between January 1, 2022 and December 31, 2024, were enrolled. The present study was a retrospective observational study and its protocol was approved by the Medical Ethics Committee of The Sixth Affiliated Hospital, School of Medicine, South China University of Technology (approval no. 2023080).

Inclusion criteria were i) Age: ≥18 years; ii) disease stage and treatment history: Patients with stage IIIb or IV lung adenocarcinoma as defined by the American Joint Committee on Cancer (AJCC)/Union for International Cancer Control (UICC) 8th edition TNM staging system (https://ajccstaging.org/en) with no prior systemic treatment for advanced disease; iii) driver gene-negative; iv) Physical status: Eastern Cooperative Oncology Group (ECOG) performance status (PS) ≤2 (ecog-acrin.org/scale); v) complete imaging data and laboratory test indicators and vi) treatment contraindication exclusion: No pneumonia induced by systemic treatment.

Exclusion criteria were i) severe organ dysfunction or cachexia: Patients who were definitively diagnosed with severe hepatic or renal insufficiency before treatment or those with terminal cancer cachexia (characterized by severe weight loss, muscle wasting, and inability to tolerate antitumor treatment), ii) insufficient treatment duration: Patients who received <3 cycles of combination therapy; iii) prolonged treatment interruption: Patients whose treatment was interrupted for >1 month during the course of combination therapy course; iv) missing key data: Patients with missing imaging data or laboratory test data; v) chronic use of immunosuppressive drugs: Patients with long-term continuous use of glucocorticoids or other immunosuppressants before or during treatment.

Patient baseline characteristics are given in Table I.

All enrolled patients were definitively diagnosed with NSCLC via integrated multimodal assessment, including imaging (CT/MRI), pathological biopsy, and clinical symptom evaluation. Lung cancer staging was performed according to the AJCC/UICC 8th TNM staging system, with all patients categorized as stage IIIb or IV. PD-L1 expression was detected using the Dako 22C3 assay, and a tumor proportion score ≥1% was defined as positive.

All patients received first-line combination therapy consisting of ICIs, antiangiogenic agents, and platinum-based chemotherapy, which was administered in a synchronous manner unless contraindicated. The specific regimens were as follows:

ICIs: Tislelizumab, 200 mg every 3 weeks; toripalimab, 240 mg every 3 weeks; sintilimab, 200 mg every 3 weeks; camrelizumab, 200 mg every 3 weeks.

Antiangiogenic agents: bevacizumab, 15 mg/kg every 3 weeks; recombinant human endostatin, 15 mg/m² on Days 1–3 of each 21-day cycle; anlotinib, 12 mg orally once daily for 2 weeks, followed by a 1-week break (21-day cycle).

Chemotherapy: Pemetrexed 500 mg/m²; paclitaxel 175 mg/m²; cisplatin 75 mg/m²; carboplatin AUC 5–6.

In the present retrospective study, the monitoring of adverse events (AEs), such as blood pressure, proteinuria and hepatic function during treatment was performed using routine clinical tests based on archived medical records (blood pressure was measured with a standard electronic sphygmomanometer, proteinuria was quantified by dry chemistry-based urinalysis and overall hepatic function was evaluated by routine serum liver biochemical tests). All AEs were graded according to the Common Terminology Criteria for Adverse Events, version 5.0 (8).

Statistical analysis of the baseline data was performed using IBM SPSS Statistics 27 (version 27.0.1). Patients' baseline characteristics were summarized using frequency (n) and the constituent ratio (%) for descriptive statistics. Continuous variables that conformed to a normal distribution were analyzed using the independent samples t test, and nonnormally distributed variables were analyzed using the Mann-Whitney test. In the influencing factor analysis, univariate analysis was performed using the log-rank test to explore the associations of variables (including sex, tumor location, comorbid underlying diseases, smoking history and treatment drugs) with PFS and OS. Progression-free survival and overall survival curves were plotted using the Kaplan-Meier method, ensuring consistency with the survival endpoint definitions in the present study. A two-tailed P<0.05 was considered to indicate a statistically significant difference.

All patients were identified from the hospital medical records system, diagnosed with advanced lung adenocarcinoma and received first-line treatment at The Sixth Affiliated Hospital (School of Medicine, South China University of Technology) between January 1, 2022 and December 31, 2024. Follow-up was initiated from the date of the first combined regimen administration and ended at the time of censoring, the present study cutoff date (March 31, 2025), or the occurrence of the primary endpoint (patient mortality). The median follow-up duration was 26.6 months (95% CI: 21.3–31.8 months). During the follow-up period, seven patients were censored and six patients succumbed. The median age of the patients was 71 years (range: 45–76 years), including two patients (16.7%) aged <60 years and ten patients (83.3%) aged ≥60 years. There were eleven male patients (91.7%) and one female patient (8.3%). Among them, ten patients (83.3%) had no smoking history, and eight patients (66.7%) had comorbid underlying diseases such as diabetes mellitus, hypertension, and coronary heart disease. The patients were staged according to the tumor TNM staging system, with nine patients (75.0%) at stage IV and three patients (25.0%) at stage IIIb. (Fig. 1; Table I).

All patients received ≥3 cycles of ICI-based triple combination therapy comprising ICIs, antiangiogenic agents, and chemotherapy. After treatment, therapeutic efficacy was assessed by integrating imaging examinations and other relevant tests: the primary pulmonary lesions were defined as target lesions, and the sizes of the baseline lesions and posttreatment lesions were compared. The evaluation was conducted in accordance with the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 (9). The results were as follows: a PR of 58.3%, an SD of 33.3%, a PD of 8.3%, an ORR of 58.3%, and a DCR of 83.3% (Tables II and III).

To investigate the effect of baseline data (such as age, sex, tumor location and comorbid underlying diseases) on the ORR of short-term efficacy, univariate analysis was performed. Patients whose efficacy was evaluated as CR or PR were classified into the high-sensitivity group, while those whose efficacy was evaluated as SD or PD were classified into the low-sensitivity group. The analysis revealed that there was no statistically significant difference in the effect on short-term efficacy among factors such as age, sex, tumor location, smoking status, chemotherapeutic agents, ICIs and targeted agents (P>0.05). By contrast, comorbid underlying diseases had a statistically significant effect on short-term treatment efficacy (P<0.05; Table IV).

All patients developed various adverse reactions. The specific manifestations and incidence were as follows: Seven patients (58.3%) experienced toxicity and six patients (50.0%) experienced nausea and vomiting. All adverse reactions were alleviated after symptomatic treatment, and no severe adverse events (SAEs) occurred (Table V).

The results of the analysis revealed that there was no statistically significant difference in survival status between the high-sensitivity group (CR, PR) and the low-sensitivity group (SD, PD) (P>0.05). In accordance with AJCC/UICC 8th TNM staging system (10), staging was performed on the basis of the maximum tumor diameter: two patients (16.7%) were in the T1 stage, four patients (33.3%) were in the T2 stage, four patients (33.3%) were in the T3 stage, and two patients (16.7%) were in the T4 stage. Of the patients, eight (66.67%) were in the N2 stage and four (33.33%) were in the N3 stage. The data revealed no statistically significant difference in the effect of tumor size or lymphatic metastasis on survival status (Table VI).

PFS and OS curves of patients were plotted using the Kaplan-Meier method. The median PFS of patients reached 25.4 months (95% CI: 12.7–38.0 months), as shown in Fig. 2. The median OS of patients reached 26.6 months (95% CI: 21.3–31.8 months), as shown in Fig. 3.

As shown in the figures, the median PFS and median OS were not reached in either the high-sensitivity group or the low-sensitivity group. The log-rank test was used to compare intergroup survival differences, and the results revealed that, compared with the low-sensitivity group, the high-sensitivity group had a longer PFS; however, there was no statistically significant difference in PFS between the two groups (P=0.332). Additionally, there was no statistically significant difference in OS between the two groups (P=0.691; Figs. 4 and 5). This was primarily due to the limited sample size (n=12) and the relatively small number of events (five deaths) in the present study, which resulted in insufficient endpoint events occurring in some subgroups during the observation period, making accurate estimation of the median survival times challenging.

In recent years, clinical trials investigating the combination of ICIs, antiangiogenic agents, and chemotherapy for advanced driver gene-negative lung cancer have yielded promising results. The IMpower150 study (19) demonstrated significant improvements in survival outcomes among driver gene-negative patients, the ICI-containing arm achieved a median progression-free survival (mPFS) of 8.3 months and a median overall survival (mOS) of 19.2 months, both of which were markedly longer than those in the control arm. Lee et al (20) reported that the addition of bevacizumab to ICIs further improved mPFS and mOS, supporting the value of salvage therapy with this combination. In a multicenter, phase II prospective cohort study (21), revealed that the combination of Rh-endostatin with PD-1 inhibitors and chemotherapy could yield favorable clinical outcomes and tolerable toxicity profile, the ORR was 57.7 vs. 34.2%, DCR was 88.5 vs. 84.2%, and the mPFS was 16.7 months vs. 16.4 months.

In studies exploring small-molecule multitarget tyrosine kinase inhibitors combined with ICIs, anlotinib plus ICIs conferred survival benefits in terms of PFS and OS (22). Specifically, first-line treatment with this combination in advanced driver gene-negative NSCLC achieved a high mPFS of 15 months; however, the incidence of grade ≥3 adverse events was 54.5% (23), highlighting a trade-off between efficacy and toxicity. By contrast, the TISAL-FE-01 study (24) reported exceptional efficacy with anlotinib plus ICIs, with an ORR of 75% and a DCR of 100%.

Notably, some studies investigating other antiangiogenic agents combined with ICIs, such as lenvatinib plus ICIs and cabozantinib plus ICIs (25,26), failed to meet their primary endpoints. This discrepancy may be attributed to excessive modification of the tumor vasculature by antiangiogenic agents, which induces intratumoral hypoxia and subsequently exacerbates immunosuppression in the TME.

In this real-world, single-center, retrospective study, the ICI-based triple combination therapy showed promising preliminary antitumor activity in patients with advanced driver gene-negative lung adenocarcinoma, with an ORR of 58.3%, a DCR of 83.3%, a mPFS of 25.4 months, and a median OS of 26.6 months. Numerically, the median PFS and OS in the cohort were longer than those reported in previously published trials of similar combination regimens, and the ORR and DCR suggested clinically relevant antitumor effects. For historical reference, the NCT02039674 study (27) (evaluating pembrolizumab plus bevacizumab, paclitaxel, and carboplatin) reported an ORR of 56%, mPFS of 7.1 months, and median OS of 16.7 months; the present study revealed a numerically prolonged median PFS and OS compared with these historical data. Notably, 75% of patients in the present cohort were PD-L1-negative, which may suggest that the ICI-based triple combination could help overcome immunotherapy resistance in patients with negative PD-L1, although this hypothesis requires validation in larger, well-controlled studies.

The clinical benefits observed in the present study may be attributed to two factors: Strategic drug selection and unique patient population characteristics. First, all the patients received pemetrexed. High tissue specificity for lung adenocarcinoma is well documented for pemetrexed. Some studies have shown that pemetrexed enhances the efficacy of immunotherapy by reducing the infiltration of myeloid-derived suppressor cells (MDSCs) in the TME (28,29). MDSCs are a major component of the immunosuppressive TME, and their depletion by pemetrexed creates a more favorable environment for ICI-mediated antitumor immune responses, thereby synergizing with ICIs and antiangiogenic agents. Second, the majority of patients were treated with tislelizumab, an anti-PD-1 antibody with proprietary Fc region engineering. This modification reduces the phagocytosis of T cells by macrophages (30), which prevents the loss of effector T cells and may further amplify the synergistic antitumor effect, contributing to its enhanced efficacy in this cohort. Univariate analysis revealed that comorbid underlying diseases (such as diabetes mellitus and hypertension) were critical factors affecting short-term treatment efficacy (P<0.05). The potential mechanism underlying this association involves chronic inflammatory states induced by comorbidities: Persistent elevation of proinflammatory cytokines (such as IL-6 and TNF-α) can alter TME homeostasis (31). These changes ultimately reduce the responsiveness to the ICI-based triple combination, highlighting the need for optimized comorbidity management in clinical practice. By contrast, age, sex, and PD-L1 expression status did not markedly affect treatment efficacy (all P>0.05). Notably, the effect of smoking history on efficacy also did not reach statistical significance (P>0.05), which may be attributed to insufficient statistical power due to the small sample size. Larger-scale studies are therefore needed to further validate the association between smoking history and the efficacy of this ICI-based triple combination.

In terms of safety, the majority of adverse reactions were grade 1 or 2 in the present study and patients exhibited good tolerance to the ICI-based triple combination. The most frequently reported AEs were hematological toxicity (58.3%), followed by nausea and vomiting (50.0%), fatigue and anorexia (33.3%), myalgia (33.3%) and chest tightness with dyspnea (33.3%). Of the patients. three developed grade 3 adverse events, including hematological toxicity and myalgia, all of which improved after symptomatic treatment. Myalgia was mainly transient. Prophylactic interventions for myelosuppression were actively administered during subsequent antitumor therapy, and the patients successfully completed the remaining treatment cycles. As a limitation, retrospective data failed to clarify the monitoring frequency of adverse reactions, so the present study was unable to calculate their median onset time.

Hematological toxicity and gastrointestinal reactions were the primary treatment-related adverse events, which is consistent with the common toxicity profile of platinum-based chemotherapy (32). Notably, hypertension and proteinuria, two typical adverse events associated with antiangiogenic agents, were not observed in the present cohort, further supporting the favorable safety of the regimen. Compared with the IMpower150 study (19), which reported a grade 3–4 AEs incidence of 67% in the ICI-containing arm, the severity of AEs in the present study was substantially lower. This discrepancy may be attributed to differences in drug doses, specific drug types, or patient baseline characteristics. The low incidence of severe hypertension and proteinuria in the present study may also be associated with the long-term use of antihypertensive drugs in some patients with preexisting hypertension, which masked the onset of such adverse reactions. Regardless, the milder toxicity profile observed here reinforces the good tolerability of the ICI-based triple combination (22), particularly providing a safe basis for elderly patients with advanced lung adenocarcinoma. Importantly, all AEs were alleviated after intervention and no treatment discontinuation or dose reduction was attributed to SAE.

The present study has several limitations that should be acknowledged. First, owing to its single-center retrospective nature and relatively small sample volume, the present investigation may be subject to inherent selection bias. Furthermore, non-typical baseline demographic and clinical characteristics could impair the generalizability of the research outcomes across diverse patient populations. Additionally, the small sample size limited the statistical power of the subgroup analyses, leading to insufficient statistical stability and making it difficult to draw reliable conclusions about these subgroups. Although the drug types included in the patient treatment regimens were the same, there were differences in specific drugs for some patients. Additionally, some patients changed their treatment regimens during the treatment process. Since no subgroup analysis was performed on the drugs, there may be potential effects of specific drug combinations on efficacy. This could affect the accuracy of the experimental results. In the future, it is planned to increase the number of included patients and perform subgroup analysis of patient medications to exclude the effects produced by specific drug combinations. The limited sample size and number of events resulted in reduced statistical power, which not only made it difficult to detect statistically significant differences in survival between groups but also restricted the application of multivariable Cox proportional hazards regression models (as the model failed to converge when multiple covariates were included). Therefore, The present study primarily employed the Kaplan-Meier method for survival analysis. Second, the present study lacked a control group, which precluded direct comparison of the ICI-based triple combination with other commonly used combination regimens. Without such a comparison, it is not possible to definitively determine whether the survival benefits observed in the present study were unique to the ICI-based triple combination or comparable to those of alternative therapies. Third, one patient was administered anlotinib. Although anlotinib exerts targeted inhibitory effects on VEGFR, it also suppresses multiple receptor tyrosine kinases including PDGFR, FGFR and c-Kit. Due to the limited sample size, sensitivity or subgroup analyses to eliminate the influence of anlotinib-related drug heterogeneity on the outcomes could not be performed, which may partially affect the results. Finally, radiotherapy-related factors were not included in the analysis. A subset of patients may have received local radiotherapy during treatment, which could have confounded survival outcomes. Subsequent studies should explicitly account for radiotherapy use and adjust for this variable to exclude potential confounding effects on efficacy assessments.

In summary, as a first-line therapy for patients with advanced driver-gene-negative lung adenocarcinoma, ICI-based triple combination therapy results in relatively high response rates and a manageable safety profile to a certain extent. To further validate the efficacy of the regimens and identify factors influencing their therapeutic outcomes, multicenter, large-sample randomized controlled trials are needed in the future to support the integration of this ICI-based triple combination into personalized treatment strategies for advanced lung adenocarcinoma, ultimately improving clinical outcomes for patients with this disease. In addition, it is planned to perform basic research for direct verification of the underlying mechanism and detection of critical biomarkers in subsequent investigations.