Lung cancer is a primary malignancy originating from the trachea, bronchial mucosa, or glands, and can be classified into small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) based on histological heterogeneity. NSCLC accounts for approximately 80–85% of all lung cancers, with a 5-year survival rate of only 10–16% for patients with advanced NSCLC (1, 2). According to the China’s latest cancer survey in 2022, lung cancer ranks first in incidence and mortality (3). Advances in medical treatment have led to an increasing number of immune checkpoint inhibitors (ICIs) being approved for lung cancer in China, and the application of immunotherapy in the treatment of NSCLC has gradually become more widespread. Preliminary studies (4, 5) have suggested that ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) uptake is associated with the efficacy of immunotherapy, with higher maximum standard uptake value (SUVmax) correlating with better treatment outcomes. However, higher metabolic tumour volume (MTV) and total lesion glycolysis (TLG) are associated with a higher non-response rate and worse prognosis in NSCLC patients treated with ICIs. The objective response rate to ICIs is approximately 20% with 10% of patients experiencing serious immune-related adverse events (irAEs) (6). This study aims to analyse positron emission tomography/ computed tomography (PET/CT) imaging and clinical data of NSCLC patients treated with and without ICIs, comparing differences in progression-free survival (PFS) and overall survival (OS), and identifying characteristics of patients who benefit from ICIs. A survival prediction model based on ¹⁸F-FDG PET/CT parameters is developed to aid in the screening of NSCLC patients for immunotherapy.

NSCLC constitutes over 80% of all lung cancer cases, with most patients experiencing asymptomatic onset in the early stages. Nearly 70% of newly diagnosed NSCLC patients present at an advanced stage, with a 5-year survival rate of only 10 to 16% (2, 11). Currently, immunotherapy is being applied to some patients with advanced NSCLC. Immune checkpoint molecules are protective elements in the human immune system, normally regulating T cell differentiation and proliferation to maintain immune balance. However, overexpression of these immune checkpoint molecules in tumour tissues inhibits T cell activation and proliferation or induces T cell apoptosis, creating an immunosuppressive tumour microenvironment that allows tumour cells to evade immune surveillance and destruction. The use of ICIs, such as those targeting programmed death 1 (PD-1), programmed death-ligand 1(PD-L1) and cytotoxic T lymphocyte-associated antigen-4 (CTLA-4), blocks the interaction between tumour tissues and T cells, thereby restoring normal immune function and providing significant survival benefits to patients (12). In recent years, tumour immunotherapy has expanded treatment options for patients with advanced NSCLC, with ICIs targeting PD-1, PD-L1, and CTLA-4 becoming a clinical hotspot.

At present, the objective remission rate for NSCLC patients undergoing immunotherapy is only 20%, with notable individual variability and a higher incidence of serious irAEs (6). ¹⁸F-FDG PET/CT is widely utilised in the diagnosis, efficacy evaluation, and prognosis assessment of lung cancer due to its non-invasive and comprehensive nature globally (13). This study aims to explore the predictive value of ¹⁸F-FDG PET/CT parameters for ICIs treatment effects in NSCLC patients and to develop a survival prediction model. We retrospectively included 155 NSCLC patients who underwent baseline ¹⁸F-FDG PET/CT at Hebei General Hospital from January 2016 to April 2023. Our findings suggest that ICIs treatment for advanced NSCLC patients provides significant survival benefits (1, 14). Compared to the non-ICIs group, the PFS and OS were significantly improved in the ICIs group, aligning with previous research results (5). Clinical data analysis revealed that elderly males, with a median age of 66 years, were the most common demographic, and adenocarcinoma was the predominant pathological type, consistent with epidemiological data and earlier studies (15). The baseline ¹⁸F-FDG PET/CT metabolic parameters in NSCLC patients are closely linked to general data and lesion morphological characteristics. Therefore, these parameters should be assessed in conjunction with other clinical and scientific data to provide a more personalised and accurate diagnosis and treatment plan.

Univariate analysis of ¹⁸F-FDG PET/CT parameters and clinical characteristics in the ICIs treatment group identified several factors influencing PFS, including smoking history, CT value and maximum diameter, SUVmax, SUVmean, SUVmin, TLR, MTVp, TLGp, MTVwb, and TLGwb. For OS, the influencing factors included age, smoking history, SCC, CT value and maximum diameter, SUVmax, SUVmean, SUVmin, TLR, MTVp, TLGp, and TLGwb. Independent predictors of PFS in the ICIs group were smoking history, primary SUVmax, MTVp, and TLGp. For OS, the independent predictors included age, smoking history, primary SUVmax, MTVp, and TLGp. Smoking history is a well-known risk factor for lung cancer, with studies indicating that the earlier smoking starts, the higher the risk of developing lung cancer, increasing 14 - fold, 8 - fold, and 5 - fold for children, adolescents, and adults, respectively, along with increased mortality (16). In this study, smoking history emerged as an independent predictor of both PFS and OS in NSCLC patients. Xu et al. (17) suggested that smoking status (pack-years), smoking duration, and time to quit smoking could be effective predictors of lung cancer incidence and mortality. However, smoking quantity alone was not an independent risk factor for NSCLC survival in this study, possibly due to the small sample size or inaccuracies in patient smoking data. Ken et al. (5) found that age, primary SUVmax, MTVp, and TLGp were effective predictors of long-term prognosis in NSCLC patients receiving ICIs, aligning with our multifactorial analysis results. Their study also identified carcinomatous lymphangiopathy as an independent predictor, though this finding may vary due to difficulties in diagnosing carcinomatous lymphangiopathy, potentially complicated by other lung conditions like interstitial fibrosis or inflammation. Giulia et al. (15) showed that MTV and TLG are important prognostic factors, positively correlating with disease progression, though SUVmax showed no correlation with PFS and OS, possibly due to differences in study populations and individual disease heterogeneity. In our study, MTVp predicted PFS and OS in ICIs-treated advanced NSCLC patients with AUCs of 0.766 and 0.718, indicating high predictive efficacy. Hye et al. (7) suggested that TLR is an independent prognostic factor for disease recurrence and patient survival in stage IB and IIA NSCLC, though its prognostic value in advanced stages remains unclear and requires further investigation. Karolien et al. (18) found that survival was not related to baseline MTVwb. Similarly, in our study, neither MTVwb nor TLGwb effectively predicted therapeutic response to ICIs in NSCLC patients. The findings indicate that NSCLC patients with no smoking history and primary SUVmax ≤19.2, MTVp ≤20.745cm³, TLGp ≤158.62 g and age ≤ 60 who received ICIs treatment experienced better survival benefits. This information could aid in the clinical screening of patients suited for ICIs treatment, improving the objective response and effective response rates and offering NSCLC patients better prognostic outcomes. Based on these results, PFS and OS prediction models for the ICIs group were established using the Cox proportional hazards regression model, with ROC curve analysis showing AUCs of 0.692 and 0.748, respectively, demonstrating good predictive efficiency.

The non-ICIs group did not undergo ICIs treatment for 3 main reasons: firstly, time reasons prevented some patients from applying ICIs treatment. The 155 patients included in this paper were screened from January 2016. Whilst ICIs therapy has emerged in recent years and gradually entered the clinic for the benefit of patients. Secondly, ICIs therapy requires patients to have some clinical indications. Finally, ICIs therapy is currently expensive and requires a certain economic basis to support patients to receive a sufficient course of treatment to achieve the therapeutic purpose. These leads to some patients not being able to apply ICIs.

This study has certain limitations. Although pathological biopsy is crucial for accurately assessing tumour nature and progression, sampling quality can be affected by various factors, and practical issues arise with multiple and continuous pathological examinations. The study did not fully obtain expression data for PD-1 and other immune checkpoints through immunohistochemistry. As a retrospective study, it may be subject to selection bias. Additionally, the absence of an external validation of monocentric data with small sample size and the lack of subgroup analyses for first-line versus second-line ICIs treatment and specific medication regimens limit the generalizability of the findings. The external validation of larger samples of multicenter data is needed in the future.

(1) Compared with the non-ICIs group, ICIs treatment significantly improves survival and prolongs survival time in NSCLC patients. (2) Smoking history, primary SUVmax, MTVp, and TLGp are independent predictors of PFS in the ICIs group. Age, smoking history, primary SUVmax, MTVp, and TLGp are independent predictors of OS in the ICIs group. All these predictors exhibit good forecasting efficiency. (3) NSCLC patients without a smoking history and with a primary SUVmax ≤19.2, MTVp ≤20.745cm³, TLGp ≤158.62 g and age ≤ 60 experience better survival benefits following ICIs treatment.