Endocrine immune-related adverse events in advanced lung cancer patients receiving immune checkpoint inhibitors: incidence, predictors and outcomes

Introduction

Immune checkpoint inhibitors (ICIs) have revolutionized the therapeutic landscape of lung cancer, targeting pathways mediated by cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), programmed cell death ligand 1 (PD-L1), and programmed cell death 1 (PD-1) (1-3). Several clinical trials have revealed that ICIs improved outcomes by prolonging overall survival (OS) and progression-free survival (PFS) in patients with advanced lung cancer (4,5). Although emerging data confirmed their prognostic benefits, ICIs are linked to a distinct array of immune-related adverse events (irAEs), which can affect any organ or system, such as endocrine gland, skin, lung, and colon (6-8). The majority of irAEs are self-limited or resolve with steroids, whereas they can also be life-threatening in some cases, necessitating permanent treatment discontinuation (7,9).

Endocrine immune-related adverse events (e-irAEs) represent among the most frequent forms of irAEs, including thyroid dysfunction, hypoparathyroidism, hypophysitis, type 1 diabetes mellitus, and adrenal insufficiency (10,11). Early-stage e-irAEs are typically occult, and many patients show no symptoms (12). Unlike other irAEs, the recovery from e-irAEs is not always guaranteed, and may convert to irreversibility with many cases resulting in permanent endocrine insufficiency (13). The median interval from ICI initiation to e-irAE onset is usually several months (11,14-16). Furthermore, previous researches indicated that patients who develop e-irAEs generally exhibit a favorable prognosis, but these conditions can become severe without prompt management (17,18). Hence, early prediction and identification of e-irAEs might help in their prevention and management. However, the potential risk factors of e-irAEs have not been fully elucidated, and there is still controversy regarding their association with prognosis.

In this retrospective cohort study, we sought to assess the incidence of e-irAEs and identify their predictors among ICI-treated patients with advanced lung cancer. Additionally, we evaluated the correlation between e-irAE development and survival outcomes, including PFS and OS. We present this article in accordance with the STROBE reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1064/rc).

Methods

Patients

This single-center retrospective cohort study included adults with lung cancer who received at least two cycles of ICIs from January 1, 2019 to October 1, 2023 at West China Hospital, Sichuan University. ICIs administered comprised PD-1 inhibitors (pembrolizumab, tislelizumab, sintilimab, camrelizumab, nivolumab, serplulimab, or toripalimab), PD-L1 inhibitors (durvalumab, atezolizumab, or sugemalimab) or combined with CTLA-4 inhibitor, ipilimumab. Patients enrolled in clinical trials or with incomplete baseline thyroid function assessment were excluded. A total of 603 patients who met the enrollment criteria were selected for final analysis (60 e-irAE group and 543 no e-irAE group) (Figure 1). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the institutional ethics committee of West China Hospital, Sichuan University (No. 2020-232). Written informed consent from participants was waived owing to its retrospective nature.

Figure 1 Flow diagram of patient enrollment. e-irAE, endocrine immune-related adverse event; ICI, immune checkpoint inhibitor.

Data collection and assessments

Medical records were retrieved from the electronic health record system. We reviewed demographic and clinical variables, such as age, sex, body mass index (BMI), smoking status, Eastern Cooperative Oncology Group performance status (ECOG PS), comorbidities according to the Charlson comorbidity index (CCI), ICI use (treatment line and type), tumor histology and stage, metastasis sites, PD-L1 tumor proportion score (TPS), concomitant medications, and laboratory parameters at baseline. The administration of concomitant medications within 1 month before and after ICI initiation was documented, including systemic corticosteroids, antibiotics, opioids, proton pump inhibitors (PPIs), nonsteroidal anti-inflammatory drugs (NSAIDs), and angiotensin converting enzyme inhibitor (ACEi)/angiotensin II receptor blocker (ARB). Systemic corticosteroid indications were categorized into cancer-unrelated, cancer-related, and irAEs. Cancer-related indications included cancer-related pain, symptomatic brain metastasis, bone metastasis, premedications of chemotherapy, dyspnea, and anorexia. Cancer-unrelated indications were chemotherapy or radiation pneumonitis, allergic contact dermatitis, chronic obstructive pulmonary disease exacerbation, and rheumatoid arthritis. We monitored thyroid function at baseline and every two cycles during immunotherapy, when clinically indicated, adrenocorticotropic hormone, cortisol, sex hormone and thyroid autoantibody were also evaluated. Clinical outcomes, including endocrine immune-related adverse events, disease progression, death, and best treatment response, were also collected. All patients were monitored until either death or the data cutoff (January 1, 2024), whichever came first.

The principal aim was to identify the incidence and predictors of e-irAE, defined as new-onset or worsening immune-related thyroid dysfunction, pituitary dysfunction or adrenal dysfunction occurring subsequent to ICI therapy, as no participant with other endocrine disorder was observed during the study period. The severity of e-irAE was graded according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 (19). Treatments and subsequent clinical outcomes were also assessed. Objective response rate (ORR) was calculated as the combined proportion of patients achieving complete response (CR) or partial response (PR). PFS was determined as the interval from ICI initiation until the earliest occurrence of disease progression or mortality. PFS and best treatment response were assessed by independent physicians using the Response Evaluation Criteria in Solid Tumors (RECIST 1.1) (20). OS was calculated from ICI initiation to death of any cause. Patients were censored at the study cutoff date (January 1, 2024).

Statistical analysis

Qualitative variables were presented as counts (percentages), and quantitative variables were summarized as medians (interquartile ranges). Continuous variables were compared with either t-test or Mann-Whitney U test, whereas categorical variables were analyzed using Chi-square test or Fisher’s exact test. Missing data were handled using multiple imputation. The cumulative incidence of e-irAE was estimated using the Aalen-Johansen method, and group differences were assessed with Gray’s test. Predictors of e-irAEs were assessed using a Fine-Gray subdistribution hazard model, calculating subdistribution hazard ratios (SHRs) and 95% confidence intervals (CIs). Logistic regression models were employed to evaluate predictors of the odds of e-irAE development as a contrast analysis. Multivariate analysis incorporated factors that were univariately associated (P<0.10). Variance inflation factors were calculated to evaluate multicollinearity. Model stability was assessed using 1,000-sample bootstrap resampling, and discrimination was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC). To address the potential immortal-time bias, time-dependent Cox regression analysis was employed, calculating hazard ratios (HRs) and 95% CIs. A sensitivity analysis was also conducted to assess the robustness of the model. Proportional hazards assumptions were verified using Schoenfeld residuals. Survival curves were constructed using the Kaplan-Meier method, and group differences were evaluated with the log-rank test. Two-sided statistical tests were utilized for all comparisons, with a significance level set at P<0.05. All statistical procedures were performed using SPSS (version 26.0) and R software (version 4.5.1).

Results

Patient characteristics

Table 1 presents the baseline characteristics of 603 advanced lung cancer patients meeting the inclusion criteria, stratified by the occurrence of e-irAEs, and the flow diagram of patients enrollment is shown in Figure 1. Of these patients, 88.4% were male, 69.2% were current or former smokers. The majority exhibited an ECOG PS of 0–1 (90.0%), non-squamous histology (57.0%) and stage IV disease (66.0%). The median age was 65.4 years [interquartile range (IQR), 58.0–70.0 years]. Lung metastasis was the most common site (35.3%), followed by bone metastasis (29.9%), brain metastasis (14.6%) and liver metastasis (10.6%). For ICI treatment, most patients received PD-1 inhibitors (81.8%), and ICIs were administered to 77.8% as first-line therapy. Among the cohort, 46.9% of patients received corticosteroids, and most of them were administered for cancer-related indications (41.3%). Demographic profiles of patients with and without e-irAEs were largely comparable. The distribution of sex and lung metastasis differed significantly between the two groups. In the overall cohort, the ORR was 41.8%, a significantly higher ORR was observed in e-irAE group compared to no e-irAE group (61.7% vs. 39.6%; P=0.001).

e-irAE incidence, characteristics, and predictors

A total of 60 (10.0%) patients developed e-irAEs after ICI initiation, with a median follow-up time of 18.0 months (IQR, 11.0–25.0 months). Table 2 displayed the distribution of e-irAEs according to the type of ICI. The incidence of e-irAEs was similar between patients receiving PD-1 inhibitors (9.7%) and PD-L1 inhibitors (10.9%) (P=0.71). The most frequent e-irAEs were hypothyroidism (73.3%) and thyrotoxicosis (23.3%), and most of which were Grade 1 (93.3%). The median duration from ICI start to e-irAE onset was 4.0 months (IQR, 2.0–7.8 months). Figure 2 illustrates the cumulative incidence of e-irAE using the Aalen-Johansen estimator. The estimate 6-month risk was 6.98% (95% CI, 4.94–9.02%), 1-year risk was 9.20% (95% CI, 6.88–11.51%), and 2-year risk was 9.97% (95% CI, 7.55–12.39%). Most e-irAEs tended to occur within 1 year after ICI initiation.

Figure 2 The cumulative incidence of e-irAE in the analysis population. e-irAE, endocrine immune-related adverse event; ICI, immune checkpoint inhibitor.

In the Fine-Gray model, female sex (SHR, 2.27; 95% CI, 1.23–4.21; P=0.009), lung metastasis (SHR, 1.79; 95% CI, 1.07–3.02; P=0.03), elevated thyroid stimulating hormone (TSH) (SHR, 1.04; 95% CI, 1.02–1.06; P<0.001), increased eosinophil count (SHR, 1.66; 95% CI, 1.32–2.10; P<0.001), and ORR (SHR, 2.23; 95% CI, 1.25–3.97; P=0.007) were associated with higher risk of e-irAE development (Table 3). The cumulative incidence curves of e-irAE after ICI treatment are presented in Figure S1. These curves further demonstrated that lung metastasis (Gray’s test P=0.03), elevated TSH (Gray’s test P=0.006), female sex (Gray’s test P=0.01) and increased eosinophil count (Gray’s test P=0.04) were correlated with a higher risk of e-irAE. Multivariate logistic regression analysis yielded similar directions of association, which indicated that lung metastasis [odds ratio (OR), 1.90; 95% CI, 1.08–3.35; P=0.03], female sex (OR, 2.37; 95% CI, 1.18–4.76; P=0.02), eosinophil count (OR, 2.30; 95% CI, 1.06–4.98; P=0.03), TSH (OR, 1.05; 95% CI, 1.00–1.11; P=0.04) and ORR (OR, 2.39; 95% CI, 1.33–4.30; P=0.003) were predictors of e-irAE development. The Hosmer-Lemeshow goodness-of-fit test indicated an adequate model fit (χ²=9.70, P=0.287; Table S1). The bootstrap-corrected AUC was 0.69 (95% CI, 0.62–0.76), which demonstrated acceptable discrimination and no substantial overfitting (Figure S2).

Survival and prognostic predictors

A total of 261 (43.3%) patients died during the follow-up. e-irAE showed a nonsignificant tendency toward increased PFS (median PFS 14.0 vs. 12.0 months; P=0.08; Figure 3A), whereas patients who developed e-irAEs demonstrated significantly prolonged OS compared to those without e-irAEs (median OS 42.0 vs. 27.0 months; P=0.04; Figure 3B). To further explore potential prognostic factors, time-dependent Cox regression analysis was conducted (Figure 4). The occurrence of e-irAE showed a favorable trend for both PFS (HR, 0.77; 95% CI, 0.50–1.19; P=0.24) and OS (HR, 0.84; 95% CI 0.52–1.36; P=0.48), indicating that patients who developed e-irAEs tended to have better clinical outcomes, although the associations did not reach statistical significance. Stage IV disease, neutrophil-to-lymphocyte ratio (NLR) ≥5 and smoking history were associated with poorer OS. In contrast, PD-L1 TPS ≥1% and BMI ≥24 kg/m² were linked to improved OS. Stage IV disease was also associated with worse PFS, whereas PD-L1 TPS ≥1% remained linked to improved PFS. The sensitivity analysis confirmed the robustness of the findings (Figure S3). In terms of corticosteroid indications, no significant differences in e-irAE incidence, PFS, or OS were observed among corticosteroid indication subgroups (Figure S4). In addition, first-line ICI showed a trend toward improved OS, but was not statistically significant (HR, 0.84; 95% CI, 0.63–1.13, P=0.24). The Kaplan-Meier analysis yielded consistent results (Figure S5).

Figure 3 Kaplan-Meier curves showing clinical outcomes, stratified by groups of patients with and without e-irAE. (A) Progression-free survival; (B) overall survival. e-irAE, endocrine immune-related adverse event.

Figure 4 Time-dependent Cox analysis of PFS and OS in all patients. BMI, body mass index; CCI, Charlson comorbidity index; CI, confidence interval; e-irAE_TV, time-varying endocrine immune-related adverse event; HR, hazard ratio; ICI, immune checkpoint inhibitor; NLR, neutrophil-to-lymphocyte ratio; OS, overall survival; PD-L1, programmed cell death ligand 1; PFS, progression free survival; PLR, platelet-to-lymphocyte ratio; TPS, tumor proportion score.

Discussion

The aim of our study was to determine the incidence, predictors and outcomes of e-irAE. A total of 603 patients with advanced lung cancer receiving ICI treatment from January 1, 2019 to October 1, 2023 were included in the final analysis. During the observation period, 10.0% of the patients developed e-irAEs, with hypothyroidism being the most common event. The median time between ICI initiation and e-irAE onset was approximately 4 months, and most e-irAEs occurred within the first year after ICI initiation. Higher subdistribution hazards of e-irAE development were observed in patients with female sex, lung metastasis, elevated TSH, increased eosinophil count, and objective response to ICIs. Moreover, e-irAE occurrence was associated with a higher ORR, and showed a trend toward longer OS and PFS compared with patients without e-irAEs.

According to the literature, the incidence of e-irAE varies from 18.1% to 25.5% (15,21). In our cohort, the overall incidence of e-irAEs was 10.0%, which was generally consistent with these reported rates. This discrepancy may be related to tumor type and ICI regimen (22,23), or may be explained by the possibility of undetected subclinical disease, which could be minimized through regular surveillance. In our study, no significant disparity in e-irAE occurrence was observed between PD-1 and PD-L1 inhibitor users, while others reported the incidence of e-irAEs increased in patients administered a combined regimen of CTLA-4 and PD-1 inhibitors (24,25). This finding relies on a limited dataset from few relevant studies. Additional data from large clinical trials are required to clarify the association. Thyroid toxicities, encompassing hypothyroidism and thyrotoxicosis, accounted for nearly all endocrine toxicities (96.6%). This finding is consistent with previous reports identifying thyroid disorders as the most prevalent endocrine toxicity (14,15), likely reflecting the prevalence of anti-thyroid antibodies, along with a genetic susceptibility to thyroid autoimmunity. In accordance with the literature, most patients with e-irAEs were asymptomatic or could be resolved with hormone replacement therapy (26). We also observed a median onset time of approximately 4 months after ICI initiation, suggesting routine thyroid function monitoring since the start of ICI treatment to facilitate early detection.

At present, no reliable methods exist to predict which individuals possess a high susceptibility to e-irAEs. Taking into account our purpose, we identified female sex, lung metastasis, eosinophil count, TSH, and objective response to ICIs as predictors of e-irAE development in the Fine-Gray subdistribution hazard model, which accounts for death as competing risk. Multivariable logistic regression showed consistent directions of association. Although the model achieved only moderate discriminative performance (AUC =0.69), this level of accuracy is acceptable for a real-world dataset, and was internally validated by 1,000-sample bootstrap resampling. Prior research suggested that low baseline NLR and platelet-to-lymphocyte ratio (PLR) may predispose patients to developing irAEs, but our analysis did not confirm these associations (27,28). Our finding of a higher risk of e-irAE in females suggests potential sex-related differences in immune regulation, consistent with evidence that females often exhibit stronger immune responses and are more susceptible to autoimmune phenomena (29). The observed association between lung metastasis and a higher e-irAE incidence should be interpreted with caution. In our cohort, more than half of the patients with lung metastasis (57.3%) had isolated intrathoracic disease without other organ involvement. This distribution likely reflects selection bias and prognostic heterogeneity. Patients with isolated or limited metastatic disease are generally fitter and may receive longer ICI exposure, which could contribute to the development of e-irAE. Elevated eosinophil counts may amplify systemic immune activation, facilitating the breakdown of self-tolerance in endocrine organs. Similarly, an elevated baseline TSH level was linked to e-irAE occurrence in our analysis, which was in accordance with previous studies (14,30,31). Furthermore, we observed an association between objective response to ICIs and e-irAE development, which could be explained by the fact that patients exhibiting a more robust immune activation in response to ICIs are more likely to achieve antitumor efficacy while simultaneously facing an increased risk of adverse effects. The identification of these predictors provides valuable clinical insights and supports emerging biological hypotheses regarding the mechanisms underlying e-irAEs, while underscoring the need for further investigation.

Within the context of advanced lung cancer, the prognostic role of e-irAEs remains contentious. In this study, e-irAE development showed a directionally favorable but non-significant association with OS and PFS after time-dependent adjustment, this trend aligns with previous reports in ICI therapy (17,18). Nevertheless, other studies have yielded inconsistent results (23), possibly due to methodological and limited sample sizes. In our analysis, the attenuation of statistical significance indicates that the apparent survival advantage associated with e-irAEs may be partly attributable to longer treatment exposure rather than a true causal relationship. Nevertheless, the consistent protective direction across sensitivity analyses supports the robustness of this association, which may reflect an intrinsic link between immune-related endocrine toxicities and enhanced antitumor immune activity. Additional protective factors associated with superior survival included BMI ≥24 kg/m² and PD-L1 TPS positive. In contrast, stage IV disease and NLR ≥5 were risk factors for poorer prognosis, consistent with prior reports (32,33). Although first-line ICI demonstrated a non-significant tendency toward longer OS, this trend aligns with previous trial showing that treatment-naïve patients have better OS than previously treated patients (34). To further ensure robustness, we performed sensitivity analyses that incorporated concomitant medications. In the sensitivity models, opioid use was associated with shorter PFS and OS, consistent with prior evidence that opioids may suppress antitumor immunity and correlate with advanced disease (35,36). Our analyses showed that NSAIDs use was also correlated with poor survival, but its impact remain controversial (37,38). These findings likely represent confounding by indication, as analgesic use often indicates higher symptom burden, systemic inflammation, or poorer ECOG status. Corticosteroid use was associated with a trend toward poorer survival outcomes, though this difference was not statistically significant, consistent with prior reports linking corticosteroid exposure to attenuated ICI efficacy (39,40). To avoid over-adjustment by such severity-related variables, these medications were excluded from the primary model, while the inclusion of these covariates in sensitivity analyses did not materially change the overall conclusions. The results were in line with the main model, supporting the stability of our findings. However, the underlying mechanisms remain poorly understood and should be explored further through large-scale studies.

Several limitations exist in this research. Firstly, this study was conducted in a single Chinese tertiary center, which may limit the generalizability of the findings to other populations. Future multicenter studies in diverse cohorts are warranted for external validation. Secondly, endocrine testing was not conducted in the entire cohort, and only thyroid function was routinely monitored while endocrine toxicities affecting other axes (pituitary, adrenal, and pancreatic) were tested when clinically indicated. Thus, the real-world incidence of e-irAEs may be underestimated. Thirdly, compared with those without e-irAE, participants with e-irAE had longer survival and mean observation time. These findings indicate that the higher incidence of e-irAEs may partially result from the extended survival of certain patients, thereby contributing to the presence of survival time bias. Although a Fine-Gray competing-risk model was applied to mitigate this bias by accounting for death as a competing event, the possibility of residual time-dependent bias cannot be entirely excluded. Next, corticosteroid exposure may influence ICI efficacy and e-irAE incidence. However, detailed information on dosage and duration was not uniformly available. Consequently, residual confounding by corticosteroid exposure cannot be excluded. Moreover, we did not study the dosage of ICIs or chemotherapy regimens. Finally, the retrospective design is susceptible to biases primarily due to the dependence on documentation available for review as well as missing data. In addition, only a small number of cases received CTLA-4 inhibitor. Therefore, further prospective studies with larger cohorts are warranted.

Conclusions

In summary, we demonstrated the precise incidence and predictors of e-irAEs in advanced lung cancer patients, including female sex, lung metastasis, eosinophil count, TSH and ORR. Although e-irAE occurrence was associated with a favorable trend in survival among individuals undergoing ICI treatment, the differences were not statistically significant. Further investigations utilizing larger sample sizes are warranted to confirm and expand these findings.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1064/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1064/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1064/prf

Funding: This research was supported by funding from the Chengdu Science and Technology Project (No. 2023-YF09-00039-SN), the Science and Technology Program of Sichuan (No. 2023NSFSC1939), and the National Natural Science Foundation of China (No. 82173182).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1064/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. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the institutional ethics committee of West China Hospital, Sichuan University (No. 2020-232). Written informed consent from participants was waived owing to its retrospective nature.

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. Riely GJ, Wood DE, Ettinger DS, et al. Non-Small Cell Lung Cancer, Version 4.2024, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2024;22:249-74. [Crossref] [PubMed]
2. Planchard D, Popat S, Kerr K, et al. Metastatic non-small cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2018;29:iv192-237. [Crossref] [PubMed]
3. Wilky BA. Immune checkpoint inhibitors: The linchpins of modern immunotherapy. Immunol Rev 2019;290:6-23. [Crossref] [PubMed]
4. Paz-Ares L, Ciuleanu TE, Cobo M, et al. First-line nivolumab plus ipilimumab combined with two cycles of chemotherapy in patients with non-small-cell lung cancer (CheckMate 9LA): an international, randomised, open-label, phase 3 trial. Lancet Oncol 2021;22:198-211. [Crossref] [PubMed]
5. Mok TSK, Wu YL, Kudaba I, et al. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial. Lancet 2019;393:1819-30. [Crossref] [PubMed]
6. Cortellini A, Chiari R, Ricciuti B, et al. Correlations Between the Immune-related Adverse Events Spectrum and Efficacy of Anti-PD1 Immunotherapy in NSCLC Patients. Clin Lung Cancer 2019;20:237-247.e1. [Crossref] [PubMed]
7. Postow MA, Sidlow R, Hellmann MD. Immune-Related Adverse Events Associated with Immune Checkpoint Blockade. N Engl J Med 2018;378:158-68. [Crossref] [PubMed]
8. Wang Y, Zhou S, Yang F, et al. Treatment-Related Adverse Events of PD-1 and PD-L1 Inhibitors in Clinical Trials: A Systematic Review and Meta-analysis. JAMA Oncol 2019;5:1008-19. [Crossref] [PubMed]
9. Haanen J, Obeid M, Spain L, et al. Management of toxicities from immunotherapy: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol 2022;33:1217-38. [Crossref] [PubMed]
10. Husebye ES, Castinetti F, Criseno S, et al. Endocrine-related adverse conditions in patients receiving immune checkpoint inhibition: an ESE clinical practice guideline. Eur J Endocrinol 2022;187:G1-G21. [Crossref] [PubMed]
11. Wright JJ, Powers AC, Johnson DB. Endocrine toxicities of immune checkpoint inhibitors. Nat Rev Endocrinol 2021;17:389-99. [Crossref] [PubMed]
12. Zhao P, Li J, Yu L, et al. Clinical manifestations and risk factors of immune-related thyroid adverse events in patients treated with PD-1 inhibitors: a case-control study. Front Immunol 2025;16:1581057. [Crossref] [PubMed]
13. Anderson B, Morganstein DL. Endocrine toxicity of cancer immunotherapy: clinical challenges. Endocr Connect 2021;10:R116-24. [Crossref] [PubMed]
14. Barlas T, Sutcuoglu O, Akdogan O, et al. Endocrine immune-related adverse events by immune checkpoint inhibitors and potential predictive markers: a prospective study from a single tertiary center. Endocr Relat Cancer 2024;31:e240101. [Crossref] [PubMed]
15. España S, Pérez Montes de Oca A, Marques-Pamies M, et al. Endocrine adverse events related to immune-oncology agents: retrospective experience of a single institution. Transl Lung Cancer Res 2020;9:103-10. [Crossref] [PubMed]
16. Ferrari SM, Fallahi P, Elia G, et al. Autoimmune Endocrine Dysfunctions Associated with Cancer Immunotherapies. Int J Mol Sci 2019;20:2560. [Crossref] [PubMed]
17. Ishidoya M, Makiguchi T, Tanaka H, et al. Endocrine immune-related adverse event is a prognostic biomarker independent of lead-time bias. Lung Cancer 2024;192:107790. [Crossref] [PubMed]
18. Cheung YM, Wang W, McGregor B, et al. Associations between immune-related thyroid dysfunction and efficacy of immune checkpoint inhibitors: a systematic review and meta-analysis. Cancer Immunol Immunother 2022;71:1795-812. [Crossref] [PubMed]
19. National Institutes of Health, Common Terminology Criteria for Adverse Events (CTCAE) Version 5.0 (accessed 27 November 2017). Available online: https://dctd.cancer.gov/research/ctep-trials/for-sites/adverse-events/ctcae-v5-5x7.pdf
20. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228-47. [Crossref] [PubMed]
21. Iwamoto Y, Kimura T, Iwamoto H, et al. Incidence of endocrine-related immune-related adverse events in Japanese subjects with various types of cancer. Front Endocrinol (Lausanne) 2023;14:1079074. [Crossref] [PubMed]
22. Cukier P, Santini FC, Scaranti M, et al. Endocrine side effects of cancer immunotherapy. Endocr Relat Cancer 2017;24:T331-47. [Crossref] [PubMed]
23. Osorio JC, Ni A, Chaft JE, et al. Antibody-mediated thyroid dysfunction during T-cell checkpoint blockade in patients with non-small-cell lung cancer. Ann Oncol 2017;28:583-9. [Crossref] [PubMed]
24. Nogueira E, Newsom-Davis T, Morganstein DL. Immunotherapy-induced endocrinopathies: assessment, management and monitoring. Ther Adv Endocrinol Metab 2019;10:2042018819896182. [Crossref] [PubMed]
25. Muir CA, Clifton-Bligh RJ, Long GV, et al. Thyroid Immune-related Adverse Events Following Immune Checkpoint Inhibitor Treatment. J Clin Endocrinol Metab 2021;106:e3704-13. [Crossref] [PubMed]
26. Gong W, Zheng E, Liu M, et al. Risk factors and outcomes of thyroid immune-related adverse events following PD-1/PD-L1 inhibitors treatment in a large tertiary Chinese center. BMC Endocr Disord 2025;25:171. [Crossref] [PubMed]
27. Pavan A, Calvetti L, Dal Maso A, et al. Peripheral Blood Markers Identify Risk of Immune-Related Toxicity in Advanced Non-Small Cell Lung Cancer Treated with Immune-Checkpoint Inhibitors. Oncologist 2019;24:1128-36. [Crossref] [PubMed]
28. Peng L, Wang Y, Liu F, et al. Peripheral blood markers predictive of outcome and immune-related adverse events in advanced non-small cell lung cancer treated with PD-1 inhibitors. Cancer Immunol Immunother 2020;69:1813-22. [Crossref] [PubMed]
29. Forsyth KS, Jiwrajka N, Lovell CD, et al. The conneXion between sex and immune responses. Nat Rev Immunol 2024;24:487-502. [Crossref] [PubMed]
30. Luongo C, Morra R, Gambale C, et al. Higher baseline TSH levels predict early hypothyroidism during cancer immunotherapy. J Endocrinol Invest 2021;44:1927-33. [Crossref] [PubMed]
31. Kobayashi T, Iwama S, Yamagami A, et al. Elevated TSH Level, TgAb, and Prior Use of Ramucirumab or TKIs as Risk Factors for Thyroid Dysfunction in PD-L1 Blockade. J Clin Endocrinol Metab 2022;107:e4115-23. [Crossref] [PubMed]
32. Sakata Y, Kawamura K, Ichikado K, et al. The association between tumor burden and severe immune-related adverse events in non-small cell lung cancer patients responding to immune-checkpoint inhibitor treatment. Lung Cancer 2019;130:159-61. [Crossref] [PubMed]
33. Liao S, Sun H, Lu H, et al. Neutrophil-to-lymphocyte ratio-based prognostic score can predict outcomes in patients with advanced non-small cell lung cancer treated with immunotherapy plus chemotherapy. BMC Cancer 2025;25:697. [Crossref] [PubMed]
34. Garon EB, Hellmann MD, Rizvi NA, et al. Five-Year Overall Survival for Patients With Advanced Non-Small-Cell Lung Cancer Treated With Pembrolizumab: Results From the Phase I KEYNOTE-001 Study. J Clin Oncol 2019;37:2518-27. [Crossref] [PubMed]
35. Kavgaci G, Guven DC, Kaygusuz Y, et al. Impact of opioid analgesics on survival in cancer patients receiving immune checkpoint inhibitors. Support Care Cancer 2024;32:467. [Crossref] [PubMed]
36. Guo H, Li Y, Lin J, et al. A novel investigation into the negative impact of opioid use on the efficacy of immune checkpoint inhibitors in advanced non-small cell lung cancer patients. Int Immunopharmacol 2024;129:111611. [Crossref] [PubMed]
37. Kennedy OJ, Glassee N, Kicinski M, et al. Prognostic and predictive value of non-steroidal anti-inflammatory drugs in the EORTC 1325/KEYNOTE-054 phase III trial of pembrolizumab versus placebo in resected high-risk stage III melanoma. Eur J Cancer 2024;201:113585. [Crossref] [PubMed]
38. Prasetya RA, Metselaar-Albers M, Engels F. Concomitant use of analgesics and immune checkpoint inhibitors in non-small cell lung cancer: A pharmacodynamics perspective. Eur J Pharmacol 2021;906:174284. [Crossref] [PubMed]
39. Li N, Zheng X, Gan J, et al. Effects of glucocorticoid use on survival of advanced non-small-cell lung cancer patients treated with immune checkpoint inhibitors. Chin Med J (Engl) 2023;136:2562-72. [Crossref] [PubMed]
40. Rousseau A, Simon-Tillaux N, Michiels S, et al. Concomitant Comedications and Survival With First-Line Pembrolizumab in Advanced Non-Small-Cell Lung Cancer. JAMA Netw Open 2025;8:e2529225. [Crossref] [PubMed]