irTD predictive model and a real cohort study on the survival benefits in advanced tumor immunotherapy

doi:10.3760/cma.j.cn371439-20250719-00074

Abstract

Abstract:

Objective To investigate the influencing factors of immune-related thyroid dysfunction （irTD） and its impact on survival prognosis in patients with advanced malignant tumors， and to analyze the regulatory role of metabolic factors. Methods A total of 70 patients with advanced tumors who received immunotherapy at the Third Affiliated Hospital of Kunming Medical University from January 1， 2018 to January 31， 2024 were selected as the study subjects. Based on the occurrence of thyroid dysfunction after immune checkpoint inhibitors （ICIs） treatment， patients were divided into the irTD group （n=53） and the non-irTD group （n=17）. The relevant indicators of the patients in the irTD group before and after immunotherapy were compared. Receiver operator characteristic （ROC） curves were plotted to evaluate the predictive efficacy of each indicator for irTD in advanced tumor patients after ICIs treatment. Multivariate logistic regression was used to identify the risk factors for irTD in tumor patients after ICIs treatment. Kaplan-Meier survival curves were plotted for survival analysis， and the log-rank test was performed. A stratified Cox proportional hazards regression model was used to evaluate the influencing factors of prognosis in patients with advanced tumors. Results In tumor patients with irTD， serum levels before and after immunotherapy were as follows： thyroid-stimulating hormone （TSH） was 3.24 （2.16， 4.01） and 12.20 （0.03， 73.55） μIU/ml； thyroglobulin antibody （TGAb） was 19.87 （12.82， 216.00） and 29.47 （15.92， 436.80） IU/ml； thyroid peroxidase antibody （TPOAb） was 4.04 （2.00， 130.43） and 42.28 （2.00， 400.00） IU/ml； and thyroglobulin （TG） was 5.67 （1.48， 17.81） and 22.80 （2.55， 52.15） ng/ml， with statistically significant differences （Z=4.36， P<0.001； Z=4.81， P<0.001； Z=4.64， P<0.001； Z=4.30， P<0.001）. ROC curve analysis showed that baseline TSH， TGAb， TPOAb， and suppressor T （Ts） cell levels had areas under the curve （AUCs） for predicting irTD after immunotherapy of 0.663， 0.704， 0.737， and 0.691， respectively， but pairwise comparisons revealed no significant differences in predictive efficacy （all P>0.05）. Multivariate analysis identified TSH≥1.87 μIU/ml （OR=4.12， 95%CI： 1.10-15.46， P=0.036） and baseline antibody positivity （OR=10.68， 95%CI： 1.21-94.43， P=0.033） as independent risk factors for irTD after immunotherapy in tumor patients. Survival analysis showed that the median overall survival （mOS） was 29 months in the irTD group and 22 months in the non-irTD group， with 1-year overall survival （OS） rates of 88.5% and 76.0%， respectively， with a statistically significant difference （χ²=4.85， P=0.028）. The subgroup analysis showed that in patients aged ≥65 years， receiving first-line therapy， with comorbid hyperlipidemia， without hyperuricemia， without diabetes， without combined surgery， with combined targeted therapy， with C-reactive protein-albumin-lymphocyte score ≤3， body mass index （BMI）<18.5 kg/m² or 18.5 kg/m²≤BMI<24 kg/m²， the mOS in the irTD group was significantly longer than that in the non-irTD group （all P<0.05）. The mOS for patients with lung adenocarcinoma， lung squamous cell carcinoma， small cell lung cancer， and other histological subtypes in the irTD group was 43， 29， 18， and 15 months， respectively， with a statistically significant difference （χ²=8.67， P=0.034）. However， pairwise comparisons showed that there were no statistically significant differences after Bonferroni correction （all adjusted P>0.008 3）. Multivariate analysis revealed that among all patients receiving ICI treatment， sex （HR=0.39， 95%CI： 0.18-0.84， P=0.016）， treatment line （HR=0.33， 95%CI： 0.13-0.85， P=0.021）， occurrence of irTD （HR=0.43， 95%CI： 0.19-0.95， P=0.037）， and surgery （HR=4.38， 95%CI： 1.26-10.55， P=0.017） were all independent influencing factors for OS. Among patients who developed irTD， further subgroup analysis stratified by TSH level showed that all Eastern Cooperative Oncology Group performance status score （HR=0.19， 95%CI： 0.05-0.83， P=0.027）， BMI grade （HR=0.54， 95%CI： 0.29-0.98， P=0.043）， and TSH>100 μIU/ml （HR=0.04， 95%CI： 0.00-0.60， P=0.021） were independent influencing factors for OS in this population. Conclusions Baseline TSH≥1.87 μIU/ml and elevated thyroid autoantibodies are independent predictors of irTD in patients with advanced tumors. The occurrence of irTD is associated with prolonged mOS， and this survival benefit persists in patients with comorbid hyperlipidemia， as well as in those without hyperuricemia or diabetes， suggesting that metabolic status may influence the prognostic value of irTD. Among patients with irTD， TSH>100 μIU/ml is an independent protective factor for OS.

Key words: Immune checkpoint inhibitors, Thyrotoxicosis, Hypothyroidism, Biomarkers, tumor, Survival analysis, Metabolic diseases, Real-world study

Li Xueqin, Gong Quan, Zhang Lijuan, Chen Xi, Li Shijuan, Zhou Chunyan, Li Jun, Zhuang Li. irTD predictive model and a real cohort study on the survival benefits in advanced tumor immunotherapy[J]. Journal of International Oncology, 2026, 53(8): 452-463.

Figures/Tables 14

指标	irTD组（n=53）	非irTD组（n=17）	t/χ²/Z值	P值
年龄（岁，$ \bar{x} \pm s$）	56.13±10.18	58.00±6.84	0.71	0.483
性别[例（%）]
男	41（77.4）	10（58.8）	2.24	0.135
女	12（22.6）	7（41.2）	2.24	0.135
病种分类[例（%）]
肺鳞状细胞癌	17（32.1）	5（29.4）	0.78	0.854
肺腺癌	17（32.1）	4（23.5）
小细胞肺癌	9（17.0）	4（23.5）
其他癌种	10（18.9）	4（23.5）
TNM分期[例（%）]
Ⅲ	16（30.2）	3（17.6）	1.03	0.312
Ⅳ	37（69.8）	14（82.4）	1.03	0.312
ECOG PS评分[分，例（%）]
0~1	45（84.9）	14（82.4）	<0.01	>0.999
≥2	8（15.1）	3（17.6）	<0.01	>0.999
治疗线数[例（%）]
一线	40（75.5）	14（82.4）	0.07	0.798
二线及以上	13（24.5）	3（17.6）	0.07	0.798
BMI等级[kg/m²，例（%）]
BMI<18.5	8（15.1）	2（11.8）	1.91	0.613
18.5≤BMI<24	26（49.1）	6（35.3）
24≤BMI<28	17（32.1）	8（47.1）
BMI≥28	2（3.8）	1（5.9）
高血压[例（%）]
是	4（7.5）	3（17.6）	0.55	0.457
否	49（92.5）	14（82.4）	0.55	0.457
高脂血症[例（%）]
是	38（71.7）	13（76.5）	0.01	0.943
否	15（28.3）	4（23.5）	0.01	0.943
高尿酸血症[例（%）]
是	21（39.6）	5（29.4）	0.58	0.448
否	32（60.4）	12（70.6）	0.58	0.448
冠心病[例（%）]
是	3（5.7）	0（0）	-	>0.999
否	50（94.3）	17（100）	-	>0.999
糖尿病[例（%）]
是	7（13.2）	3（17.6）	<0.01	0.955
否	46（86.8）	14（82.4）	<0.01	0.955
手术[例（%）]
是	8（15.1）	3（17.6）	<0.01	>0.999
否	45（84.9）	14（82.4）	<0.01	>0.999
化疗[例（%）]
是	50（94.3）	16（94.1）	<0.01	>0.999
否	3（5.7）	1（5.9）	<0.01	>0.999
放疗[例（%）]
是	13（24.5）	4（23.5）	<0.01	>0.999
否	40（75.5）	13（76.5）	<0.01	>0.999
靶向治疗[例（%）]
是	25（47.2）	7（41.2）	0.19	0.666
否	28（52.8）	10（58.8）	0.19	0.666
疗效评价[例（%）]
疾病控制	50（94.3）	12（70.6）	5.02	0.025
客观缓解	39（73.6）	11（64.7）	0.50	0.481
免疫药物种类[例（%）]
替雷利珠单抗	22（41.5）	7（41.2）	4.95	0.911
信迪利单抗	7（13.2）	4（23.5）
斯鲁利单抗	8（15.1）	4（23.5）
阿替利珠单抗	1（1.9）	0（0）
特瑞普利单抗	4（7.5）	1（5.9）
卡瑞利珠单抗	2（3.8）	1（5.9）
帕博利珠单抗	5（9.4）	0（0）
纳武利尤单抗	2（3.8）	0（0）
舒格利单抗	1（1.9）	0（0）
度伐利尤单抗	1（1.9）	0（0）
甲状腺功能基线正常[例（%）]
是	51（96.2）	15（88.2）	0.40	0.526
否	2（3.8）	2（11.8）	0.40	0.526
基线抗体阳性[例（%）]
是	23（43.4）	1（5.9）	6.46	0.011
否	30（56.6）	16（94.1）	6.46	0.011
基线抗体阳性类型[例（%）]
无	30（56.6）	16（94.1）	7.77	0.035
TGAb升高	8（15.1）	1（5.9）
TPOAb升高	3（5.7）	0（0）
TGAb+TPOAb升高	12（22.6）	0（0）
TSH[μIU/ml，M（Q₁，Q₃）]	3.24（2.16，4.01）	1.83（1.17，3.36）	2.01	0.045
FT₃（pmol/L，$ \bar{x} \pm s$）	4.98±0.71	5.05±0.67	0.35	0.727
FT₄[pmol/L，M（Q₁，Q₃）]	15.49（14.30，16.71）	14.90（13.33,17.34）	0.58	0.560
TGAb[IU/ml，M（Q₁，Q₃）]	19.87（12.82，216.00）	12.33(10.00，18.25)	2.52	0.012
TPOAb[IU/ml，M（Q₁，Q₃）]	4.04（2.00，130.43）	2.00（2.00，2.00）	3.22	0.001
TG[ng/ml，M（Q₁，Q₃）]	5.67（1.48，17.81）	8.53（7.00，15.51）	1.76	0.078
白蛋白（g/L，$ \bar{x} \pm s$）	44.02±4.36	40.86±4.97	2.51	0.015
IL-6[pg/ml，M（Q₁，Q₃）]	7.21（5.50，12.70）	10.55（5.50，16.70）	0.98	0.327
PCT[ng/ml，M（Q₁，Q₃）]	0.04（0.01，0.06）	0.04（0.01，0.15）	0.83	0.409
CRP[mg/L，M（Q₁，Q₃）]	4.49（2.19，10.00）	2.77（1.55，12.53）	0.07	0.945
NLR[M（Q₁，Q₃）]	2.91（1.83，4.51）	2.90（2.43，4.27）	0.19	0.853
LMR[M（Q₁，Q₃）]	2.76（1.92，3.87）	3.30（2.27，4.28）	0.77	0.439
CAR[M（Q₁，Q₃）]	0.10（0.05，0.22）	0.06（0.04，0.32）	0.14	0.885
CALLY>3[例（%）]
是	15（28.3）	6（35.3）	0.30	0.584
否	38（71.7）	11（64.7）	0.30	0.584
T细胞[cells/μl，M（Q₁，Q₃）]	645.00（512.00，870.00）	582.00（474.00，655.00）	1.13	0.258
Th细胞[cells/μl，M（Q₁，Q₃）]	404.00（305.00，510.00）	350.00（312.00，467.00）	0.62	0.538
Ts细胞[cells/μl，M（Q₁，Q₃）]	324.00（241.00，410.00）	226.00（159.00，286.00）	2.36	0.018

References 30

[1]	Kennedy LB, Salama AKS. A review of cancer immunotherapy toxicity[J]. CA Cancer J Clin, 2020, 70(2): 86-104. DOI: 10.3322/caac.21596.
[2]	中华医学会内分泌学分会免疫内分泌学组. 免疫检查点抑制剂引起的内分泌系统免疫相关不良反应专家共识(2020)[J]. 中华内分泌代谢杂志, 2021, 37(1): 1-16. DOI: 10.3760/cma.j.cn311282-20200627-00475.
[3]	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[J]. Cancer Immunol Immunother, 2022, 71(8): 1795-1812. DOI: 10.1007/s00262-021-03128-7.
[4]	Li ZJ, Xia Y, Xia MZ, et al. Immune-related thyroid dysfunction is associated with improved long-term prognosis in patients with non-small cell lung cancer treated with immunotherapy: a systematic review and meta-analysis[J]. J Thorac Dis, 2023, 15(2): 690-700. DOI: 10.21037/jtd-23-254. pmid: 36910117
[5]	Han XY, Chen YC, Xie H, et al. Organ-specific immune-related adverse events and prognosis in cancer patients receiving immune checkpoint inhibitors[J]. BMC Cancer, 2025, 25(1): 139. DOI: 10.1186/s12885-025-13566-6. pmid: 39856626
[6]	Iwama S, Kobayashi T, Yasuda Y, et al. Immune checkpoint inhibitor-related thyroid dysfunction[J]. Best Pract Res Clin Endocrinol Metab, 2022, 36(3): 101660. DOI: 10.1016/j.beem.2022.101660.
[7]	焦盼盼, 薛丽娟, 詹娟. 免疫检查点抑制剂相关不良反应的危险因素与预测因素[J]. 国际肿瘤学杂志, 2023, 50(12): 739-744. DOI: 10.3760/cma.j.cn371439-20230810-00139.
[8]	Lechner MG, Cheng MI, Patel AY, et al. Inhibition of IL-17A protects against thyroid immune-related adverse events while preserving checkpoint inhibitor antitumor efficacy[J]. J Immunol, 2022, 209(4): 696-709. DOI: 10.4049/jimmunol.2200244. pmid: 35817515
[9]	Zhao PF, Li J, Yu LH, et al. Clinical manifestations and risk factors of immune-related thyroid adverse events in patients treated with PD-1 inhibitors: a case-control study[J]. Front Immunol, 2025, 16: 1581057. DOI: 10.3389/fimmu.2025.1581057.
[10]	高雅, 叶婷, 武凌鸽, 等. 基线TPOAb和(或)TgAb阳性与恶性肿瘤患者免疫检查点抑制剂治疗后发生甲状腺免疫相关不良反应的关系[J]. 中华医学杂志, 2024, 104(12): 963-969. DOI: 10.3760/cma.j.cn112137-20231011-00706.
[11]	Zhou X, Iwama S, Kobayashi T, et al. Risk of thyroid dysfunction in PD-1 blockade is stratified by the pattern of TgAb and TPOAb positivity at baseline[J]. J Clin Endocrinol Metab, 2023, 108(10): e1056-e1062. DOI: 10.1210/clinem/dgad231.
[12]	Wang Q, Wu TT, Zhao R, et al. Development and validation of a prediction model for thyroid dysfunction in patients during immunotherapy[J]. Endocr Pract, 2024, 30(10): 943-950. DOI: 10.1016/j.eprac.2024.07.006.
[13]	von Itzstein MS, Gonugunta AS, Wang YQ, et al. Divergent prognostic effects of pre-existing and treatment-emergent thyroid dysfunction in patients treated with immune checkpoint inhibitors[J]. Cancer Immunol Immunother, 2022, 71(9): 2169-2181. DOI: 10.1007/s00262-022-03151-2.
[14]	Muir CA, Wood CCG, Clifton-Bligh RJ, et al. Association of antithyroid antibodies in checkpoint inhibitor-associated thyroid immune-related adverse events[J]. J Clin Endocrinol Metab, 2022, 107(5): e1843-e1849. DOI: 10.1210/clinem/dgac059.
[15]	González MM, Sanz-Pastor AG, Gómez RA, et al. Endocrine adverse events related to immune checkpoint inhibitor treatment: relationship between antibodies and severity of thyroid dysfunction[J]. Endocr Metab Immune, 2024, 24(14): 1628-1639. DOI: 10.2174/0118715303280679240206100137.
[16]	Wang YL, Yang XX, Ma J, et al. Thyroid dysfunction (TD) induced by PD-1/PD-L1 inhibitors in advanced lung cancer[J]. Heliyon, 2024, 10(5): e27077. DOI: 10.1016/j.heliyon.2024.e27077.
[17]	Shi YQ, Fang J, Zhou CZ, et al. Immune checkpoint inhibitor-related adverse events in lung cancer: real-world incidence and management practices of 1905 patients in China[J]. Thorac Cancer, 2022, 13(3): 412-422. DOI: 10.1111/1759-7714.14274.
[18]	Yoo WS, Ku EJ, Lee EK, et al. Incidence of endocrine-related dysfunction in patients treated with new immune checkpoint inhibitors: a meta-analysis and comprehensive review[J]. Endocrinol Metab (Seoul), 2023, 38(6): 750-759. DOI: 10.3803/EnM.2023.1785.
[19]	郭玮, 王冬慧, 王振华, 等. 晚期肿瘤患者PD-1单抗治疗后免疫相关甲状腺功能异常与生存情况的相关性[J]. 国际肿瘤学杂志, 2024, 51(8): 481-486. DOI: 10.3760/cma.j.cn371439-20240215-00080.
[20]	Choi MG, Choi CM, Lee DH, et al. Impact of gender on response to immune checkpoint inhibitors in patients with non-small cell lung cancer undergoing second- or later-line treatment[J]. Transl Lung Cancer Res, 2022, 11(9): 1866-1876. DOI: 10.21037/tlcr-22-146.
[21]	Guo G, Jing ZH, Dou WR, et al. Immune-related thyroid dysfunction as a positive prognostic factor for patients with lung cancer in China: a real-world retrospective study[J]. Front Immunol, 2024, 15: 1495460. DOI: 10.3389/fimmu.2024.1495460.
[22]	Wang YL, Ma XP, Ma J, et al. Thyroid dysfunction as a predictor of PD-1/PD-L1 inhibitor efficacy in advanced lung cancer[J]. BMC Cancer, 2025, 25(1): 791. DOI: 10.1186/s12885-025-14097-w.
[23]	Kopanos S, Filippatos C, Rousakis P, et al. Prognostic significance of endocrine-related adverse events in patients with melanoma, non-small cell lung cancer and urothelial cancer after treatment with immune checkpoint inhibitors: a systematic review and meta-analysis[J]. Cancers (Basel), 2025, 17(22): 3675. DOI: 10.3390/cancers17223675.
[24]	Pollack R, Ashash A, Cahn A, et al. Immune checkpoint inhibitor-induced thyroid dysfunction is associated with higher body mass index[J]. Clin Endocrinol Metab, 2020, 105(10): dgaa458. DOI: 10.1210/clinem/dgaa458.
[25]	Guo H, Lin XY, Feng S, et al. Prognostic value of obesity in patients with cancer treated with immune checkpoint inhibitors: an updated meta-analysis and systematic review[J]. Mol Clin Oncol, 2024, 20(1): 5. DOI: 10.3892/mco.2023.2703.
[26]	Qin WH, Yang ZS, Li M, et al. High serum levels of cholesterol increase antitumor functions of nature killer cells and reduce growth of liver tumors in mice[J]. Gastroenterology, 2020, 158(6): 1713-1727. DOI: 10.1053/j.gastro.2020.01.028.
[27]	Acar C, Yüksel HÇ, Şahin G, et al. Prognostic utility of the CALLY index in metastatic melanoma: building a nomogram for patients on anti-PD-1 therapy[J]. Clin Transl Oncol, 2025, 27(9): 3770-3780. DOI: 10.1007/s12094-025-03888-z. pmid: 40091005
[28]	Nguyen NN, Nguyen THH, Hoang KD, et al. The prognostic significance of diabetes in non-small cell lung cancer patients treated with immune checkpoint inhibitors: a meta-analysis[J]. Diabetes Res Clin Pract, 2024, 218: 111930. DOI: 10.1016/j.diabres.2024.111930.
[29]	Garstka MA, Kedzierski L, Maj T. Diabetes can impact cellular immunity in solid tumors[J]. Trends Immunol, 2025, 46(4): 295-309. DOI: 10.1016/j.it.2025.02.014.
[30]	Feng Q, Tang LJ, Luo DH, et al. Metabolic syndrome-related hyperuricemia is associated with a poorer prognosis in patients with colorectal cancer: a multicenter retrospective study[J]. Cancer Manag Res, 2021, 13: 8809-8819. DOI: 10.2147/CMAR.S338783. pmid: 34866939

指标	AUC	95%CI	敏感性	特异性	约登指数	截断值	P值
TSH	0.663	0.512~0.813	0.811	0.529	0.341	1.87 μIU/ml	0.045
TGAb	0.704	0.579~0.829	0.541	0.824	0.317	18.53 pmol/L	0.012
TPOAb	0.737	0.621~0.853	0.453	1.000	0.453	15.83 pmol/L	0.003
Ts细胞	0.691	0.553~0.829	0.547	0.824	0.371	304 cells/μl	0.018

变量	β值	SE值	Wald χ²值	OR值	95%CI	P值
TSH（≥1.87 μIU/ml）	1.58	0.60	6.92	4.84	1.95~15.66	0.009
基线抗体阳性	2.51	1.07	5.52	12.27	1.51~99.38	0.019
Ts细胞（≥304 cells/μl）	1.73	0.69	6.22	5.64	1.45~21.95	0.013

变量	β值	SE值	Wald χ²值	OR值	95%CI	P值
TSH（≥1.87 μIU/ml）	1.42	0.67	4.41	4.12	1.10~15.46	0.036
基线抗体阳性	2.37	1.11	4.54	10.68	1.21~94.43	0.033

变量	β值	SE值	Wald χ²值	HR值	95%CI	P值
性别	0.62	0.34	3.38	0.54	0.28~1.04	0.066
治疗线数	0.42	0.38	1.22	0.66	0.31~1.39	0.269
发生irTD	0.82	0.39	4.38	0.44	0.21~0.95	0.037
手术	1.06	0.53	3.98	2.89	1.01~8.15	0.046

变量	β值	SE值	Wald χ²值	HR值	95%CI	P值
病种分类	0.27	0.13	4.44	1.31	1.02~1.69	0.035
ECOG PS评分	0.63	0.57	1.21	0.53	0.17~1.64	0.271
BMI等级	0.33	0.24	1.91	0.72	0.45~1.15	0.167