Skip to main content
Download PDF

Thyroid dysfunction as a predictor of PD- 1/PD-L1 inhibitor efficacy in advanced lung cancer

BMC Cancer volume 25, Article number: 791 (2025) Cite this article

Abstract

Purpose

To investigate the correlation between thyroid dysfunction (TD) and the efficacy of programmed cell death protein 1 (PD-1) and programmed death ligand 1 (PD-L1) inhibitors in the treatment of advanced lung cancer, and the possible influencing factors for TD occurrence, providing insights that could guide individualized therapeutic approaches.

Methods

The data of 120 advanced lung cancer patients from January 2019 to August 2024 were retrospectively collected. Then, the patients were divided into TD and non-TD subgroups according to whether TD occurred or not, to analyse the possible factors influencing the occurrence of TD and the correlation between TD and PD-1/PD-L1 inhibitor efficacy.

Results

For all cases, the baseline TSH level was significantly higher in the TD subgroup than in the non-TD subgroup (median: 2.33 mIU/L vs. 1.58 mIU/L, p = 0.001). The progression-free survival (PFS) was significantly longer in the TD subgroup than in the non-TD subgroup (mPFS: 7.90 months vs. 4.87 months, p = 0.003), and the patients in the TD subgroup had a lower HR for progression (0.499, 95% CI (0.317–0.766)). For the PD-1/PD-L1 inhibitor group, the baseline TSH level was also significantly higher in the TD subgroup than in the non-TD subgroup (median: 2.16 mIU/L vs. 1.52 mIU/L, p = 0.009). The PFS was also significantly longer in the TD subgroup than in the non-TD subgroup (mPFS: 8.83 months vs. 6.50 months, p = 0.041).

Conclusions

The baseline TSH level was the predictive factor for the occurrence of TD. The occurrence of TD was positively associated with a favorable prognosis for patients with advanced lung cancer.

Peer Review reports

Introduction

Lung cancer remains the malignancy with the highest mortality rate worldwide, with about 1.8 million people dying from lung cancer every year [1]. Due to the lack of obvious symptoms and diagnostic means in the early stage, more than 60 percent of patients are diagnosed at an advanced stage, losing the chance for surgery, and the five-year survival rate is less than 5 percent [2]. Conventional chemotherapy has limited effect in improving survival, and the 5-year survival rate is only 20–30 percent [3]. The emergence of immune checkpoint inhibitors (ICIs) has changed this situation. Wang et al. [4] found that either ICIs alone or ICIs in combination with chemotherapy significantly prolonged the progression-free survival (PFS) of non-small-cell lung cancer (NSCLC) patients compared with chemotherapy alone. The combination of PD-L1 inhibitors with platinum-based chemotherapy in the treatment of extensive-stage SCLC could improve survival, and combination chemoimmunotherapy is now approved as the standard of treatment [5]. However, about 60–80% of ICI-treated patients suffered immune-related adverse events (irAEs), including lung, dermatologic, gastrointestinal, renal, ophthalmic, neurologic, endocrine, musculoskeletal, hematologic, and cardiovascular toxicity [6,7,8]. For the patients with severe irAEs, ICI therapy should be discontinued immediately or even permanently due to the severity and high possibility of recurrence [9]. PD- 1/PD-L1 inhibitors are the most widely used ICIs in clinical application, which can significantly improve patients’ prognosis while causing few toxic side effects [10]. At present, a variety of PD- 1 and PD-L1 inhibitors have been incorporated into the first-line treatment of advanced lung cancer by the Chinese Society of Clinical Oncology (CSCO) Guidelines.

In addition to the efficacy, PD- 1/PD-L1 inhibitors-induced unique irAEs are of increasing concern to clinicians. Thyroid dysfunction is one of the most common irAEs, accounting for approximately 6–8% of all irAEs [11]. According to relevant studies, TD occurs in about one-fifth of advanced lung cancer patients treated with PD- 1/PD-L1 inhibitors. However, real-world studies have found TD rates as high as 40–50% [11, 12]. Interestingly, some studies have found that the occurrence of TD may activate the immune system, improving the efficacy of PD- 1/PD-L1 inhibitors [13,14,15], but this correlation has not yet been confirmed in lung cancer [12, 14]. This suggests that the correlation between TD and good prognosis may be related to tumor type. Previous studies have found positive correlations in renal-cell carcinoma, head and neck cancer, and uroepithelial cancer, but no correlation in lung cancer, melanoma, and gastrointestinal carcinoma [16]. It should be noted that the subjects of the previous studies are not first-line patients with advanced lung cancer, so their study process and results may be affected by a variety of confounding factors. Furthermore, previous studies did not exclude patients with pre-existing thyroid diseases from the study population, and thyroid function was not monitored closely, which might influence their study results to a certain degree. While this study will overcome these limitations by excluding patients with thyroid disease at the time of inclusion population, while closely monitoring thyroid function and testing thyroid function at each cycle of treatment. In addition, there is a lack of reliable biomarkers to predict the development of TD. Therefore, the primary objective of this study was to explore whether there was a correlation between TD occurrence and the PD- 1/PD-L1 inhibitor efficacy for advanced lung cancer. The secondary objective is to discover possible predictive factors for the occurrence of TD. This study will help explore the molecular mechanisms leading to TD in PD- 1/PD-L1 inhibitor treatment for lung cancer. By analysing the correlation between TD and therapeutic efficacy and the possible factors influencing the occurrence of TD, we hope to establish a predictive model of therapeutic efficacy based on the occurrence of TD. This will guide the clinical diagnosis and treatment, and provide supports for clinicians to formulate personalized diagnosis and treatment plan.

Research methods

Research targets and inclusion criteria

In this retrospective study, the data of 120 advanced lung cancer patients with primary treatment in the First Affiliated Hospital of Shihezi University and the People’s Hospital of Shihezi City from January 2019 to August 2024 were collected. The PD- 1/PD-L1 inhibitor group consisted of 60 patients who were treated with PD- 1/PD-L1 inhibitors in combination with a platinum-based chemotherapy regimen (some patients with adenocarcinoma were treated with bevacizumab in combination with platinum), and the non-PD- 1/PD-L1 inhibitor group consisted of 60 patients who were treated without PD- 1/PD-L1 inhibitors. The PD- 1/PD-L1 inhibitor group included 21 squamous cell carcinomas, 17 adenocarcinomas, and 22 small cell carcinomas, and the non-PD- 1/PD-L1 inhibitor group included 6 squamous cell carcinomas, 30 adenocarcinomas, 22 small cell carcinomas, and 2 specific types of carcinomas. The PD- 1/PD-L1 inhibitor treatment group included 42 patients treated with PD- 1 inhibitors and 18 patients treated with PD-L1 inhibitors. All were treated in accordance with the standardized treatment of CSCO guidelines for lung cancer. This study was approved by the Ethics Committee of the First Affiliated Hospital of Shihezi University under the ethical number: KJX 2022–081 - 01. All patients (or their family members) were duly informed and provided their written consent, authorizing access to their clinical data for the purposes of this study. This study was compliant with the Declaration of Helsinki.

The inclusion criteria were as follows: (1) Patients older than 18 years of age; (2) Patients with advanced lung cancer (tumor node metastasis (TNM) stage 3 (inoperable) or 4 confirmed by cytological and imaging tests of pathological tissues; (3) Patients who were first treated and had not experienced any antitumor therapy before; (4) Patients who had been treated with at least 2 cycles of PD- 1/PD-L1 inhibitors combined with chemotherapy or received therapies without PD- 1/PD-L1 inhibitor (only chemotherapy or chemotherapy combined with bevacizumab); (5) Patients with a PS lower than 2 (including 2); (6) Patients with a survival period greater than 3 months; (7) Patients with complete clinical and hematological results and follow-up records.

The exclusion criteria were as follows: (1) Patients treated with TKI-targeting therapy; (2) Patients with abnormal thyroid function, thyroid-related disease, or autoimmune disease, or received thyroid surgery or radiation to the thyroid area prior to treatment; (3) Patients with TD induced by toxic goiter; (4) Patients with multiple malignancies or previously received any antitumor therapy; (5) Patients with combined severe cardiac insufficiency, impaired hepatic and renal function, or hematopoietic disorders; (6) Patients with clinical evidence of chronic inflammatory disease, comorbid severe acute infection, or inflammation prior to treatment or at the time of efficacy assessment; (7) Patients without complete follow-up records or complete clinical information.

Clinical indicators

The following clinical details were collected from the 120 patients 1–2 days before treatment: (1) Thyroid function test results (TSH, free triiodothyronine (fT3), free thyroxine (fT4)); (2) Complete blood count (neutrophil count, platelet count, and lymphocyte count were used to calculate NLR and PLR); (3) Liver function (albumin level was used to calculate PNI (albumin level (g/L) + 5 × lymphocyte count). Follow-up visits were conducted every 2 cycles to evaluate efficacy and collect data, until disease progression, discontinuation of the drug for intolerable adverse effects, loss of follow-up, or deadline reached (October 2024).

The patients were divided into TD and non-TD subgroups according to whether they had a TD or not, and the difference in therapeutic efficacy between the two subgroups was analyzed. Meanwhile, the differences in general information, basal TSH level, NLR, PLR, and PNI before and after the treatment between TD and non-TD subgroups were analyzed, to explore the influencing factors of TD occurrence.

Evaluation tools

The efficacy for all patients was strictly evaluated using the Response Evaluation Criteria in Solid Tumors Version 1.1, and the efficacy results were recorded, including complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD) (Table S1a, b).

TD was assessed according to the following criteria: one or more consecutive abnormal TSH values, irrespective of the levels of fT4 and fT3 (those with reduced fT3 only were also recorded as TD). Hypothyroidism: increased TSH level and decreased fT4 or fT3 level. Subclinical hypothyroidism: increased TSH level only. Hyperthyroidism: decreased TSH level and increased fT4 or fT3 level. Subclinical hyperthyroidism: decreased TSH level only. Low-T3 syndrome: decreased fT3 level only, normal TSH and fT4 level. The ranges for normal thyroid function: TSH: 0.27–4.2 mIU/L; fT3: 3.1–6.8 pmol/L; fT4: 12.0–22.0 pmol/L. Levothyroxine replacement therapy was received when TSH level was greater than 10 mIU/L. Besides, TD was graded by severity according to the Common Terminology Criteria for Adverse Events Version 5.0 (Table S2).

Outcome indicators

Primary outcome indicator: PFS (time from initiation of treatment to disease progression or death). Secondary outcome indicators: ORR ((CR + PR)/total number of cases), DCR ((CR + PR + SD)/total number of cases), and duration of response (DOR) (time from the first assessment of CR and PR to disease progression or death).

Analysis and workflow

In this study, we gathered clinical data from 120 patients who underwent early treatment for advanced lung cancer, adhering to specific inclusion and exclusion criteria. All patients had normal baseline thyroid function at the start. They were categorized based on their treatment regimens into two groups: one receiving PD- 1/PD-L1 inhibitors (60 patients) and the other not receiving these inhibitors (60 patients). We conducted follow-ups and assessments at the end of every two treatment cycles to compare the incidence of TD between the groups. Patients were further classified into those who developed TD and those who did not during the course of their treatment. We analyzed differences in various parameters such as pre-treatment demographics, baseline TSH levels, NLR, PLR, and PNI before and after treatment, aiming to identify potential factors associated with TD development. Additionally, we examined the relationship between TD occurrence and treatment efficacy by comparing PFS, ORR, and DCR between the subgroups. We also explored the impact of different types of TD on PFS. All the work steps are illustrated in Fig. 1.

Fig. 1
figure 1

Workflow

Full size image

Statistical analysis

SPSS software version 22.0 was used for statistical analysis. Data with normal distribution were expressed as mean ± standard deviation (SD) (x ± s). T-test was used for comparison between two groups, and analysis of variance (ANOVA) was used for comparison between multiple groups. Continuous variables that were not normally distributed were expressed as median and interquartile range (M, IQR), and non-parametric tests were used for comparison between groups. Count data were expressed as the number of cases and the rate (%), the x2 test was used for comparison between groups, and the rank-sum test was used for comparison of rank data. The Kaplan–Meier method was used for survival analysis, and the Log-rank method was used for comparison of survival between groups. The multifactorial Cox regression model was used to evaluate the HR and 95% CI for the factors that were statistically significant in the one-way Cox analysis. Odds ratios (OR) were calculated to assess the risk associated with baseline TSH level and the occurrence of TD. Receiver operating characteristic (ROC) curves were drawn to assess the threshold value of baseline TSH level associated with the risk of TD occurrence. Differences were considered statistically significant at p < 0.05 (Cox regression analysis was performed to avoid missing meaningful influences, and p-value levels were increased appropriately. All single factors with p ≤ 0.1 were included in the multifactorial analysis).

Results

Comparison of general information between PD- 1/PD-L1 inhibitor and non-PD- 1/PD-L1 inhibitor groups

There were no significant differences in age, gender, BMI, smoking history, clinical stage, tumor type, PS score, distant metastasis (e.g., bone metastasis, brain metastasis, and liver metastasis), and therapy (chemotherapy or chemotherapy + bevacizumab) between PD- 1/PD-L1 inhibitor and non-PD- 1/PD-L1 inhibitor groups (Table S3).

Incidence of TD in the PD- 1/PD-L1 inhibitor and non-PD- 1/PD-L1 inhibitor groups

Throughout the course of treatment, 29 patients (48.3%) in the PD- 1/PD-L1 inhibitor group developed TD, whereas only 10 patients (16.7%) in the non-PD- 1/PD-L1 inhibitor group developed TD (p < 0.001). The difference in the median time to TD between the two groups was not significant (p = 0.079). Subclinical hypothyroidism was the most common type of TD occurring in both groups, but the incidence of low-T3 syndrome was higher in the non-PD- 1/PD-L1 inhibitor group. The majority of TDs occurring in both groups throughout the course of treatment were grade 1, and only five patients with a grade 2 score (TSH > 10 mIU/L) received levothyroxine replacement therapy (Table 1).

Table 1 Incidence of TD in the PD- 1/PD-L1 inhibitor and non-PD- 1/PD-L1 inhibitor groups
Full size table

Comparison of clinical data between TD and non-TD subgroups

General information

Of the 120 patients, 39 (32.5 percent) patients developed TD. Low-T3 syndrome had the earliest onset, with a median onset time of 45 days, followed by hyperthyroidism (55 days), hypothyroidism (124 days), subclinical hypothyroidism (127 days), and subclinical hyperthyroidism (147 days). Of the five patients who developed hyperthyroidism, two transformed into hypothyroidism (one grade 2), one returned to normal, and one transformed into subclinical hypothyroidism after a short period of hyperthyroidism (Table S4a). There were no significant differences in gender, age, BMI, smoking history, clinical stage, tumor type, PS score, distant metastasis, and underlying diseases between TD and non-TD subgroups (Table 2).

Table 2 Comparison of general information between TD and non-TD subgroups
Full size table

In the PD- 1/PD-L1 inhibitor group, the study also found that subclinical hyperthyroidism (182 days) occurred latest, followed by subclinical hypothyroidism (172 days), and thyroid function became normal after a period of time in 9 patients (Table S4b). The general information of the TD and non-TD subgroups was also compared and the results were similar to the results in Table 2. There was no significant difference in the TD occurrence between PD- 1 inhibitor and PD-L1 inhibitor groups (p = 0.464) (Table S4c).

Comparison of baseline TSH level, NLR, PLR and PNI between TD and non-TD subgroups

Among the 120 patients, the baseline TSH level was significantly higher in the TD subgroup than in the non-TD subgroup (median: 2.33 mIU/L vs. 1.58 mIU/L) (p = 0.001) (Table 3). The ROC curve suggested a significantly increased risk of TD (OR: 1.729, 95% CI (1.252–2.386), p = 0.001) when TSH was greater than 1.775 mIU/L (area under curve (AUC): 0.701, Sensitivity: 0.794, Specificity: 0.592) (Figure S1a). There were no differences in NLR, PLR, and PNI between the two subgroups (p > 0.05) (Table 3).

Table 3 Comparison of baseline TSH level, NLR, PLR and PNI between TD and non-TD subgroups
Full size table

For the 60 patients in the PD- 1/PD-L1 inhibitors group (Table S5), the baseline TSH level in the TD subgroup was significantly higher than that in the non-TD subgroup (median: 2.16 mIU/L vs. 1.52 mIU/L, p = 0.009). The ROC curve suggested a significantly increased risk of TD (OR: 1.761, 95% CI (1.086–2.857), p = 0.022) when TSH was greater than 1.765 mIU/L (AUC: 0.712, Sensitivity: 0.808, Specificity: 0.692) (Figure S1b). The differences in NLR, PLR, and PNI between the two subgroups were not significant (p > 0.05) (Table S5).

Within-group analysis of NLR, PLR, and PNI before and after treatment

For the 120 patients, there were no differences in the PLR and PNI before and after treatment in the TD and non-TD subgroups (p > 0.05). However, the non-TD subgroup had a significant decrease in NLR after treatment compared to before (p = 0.042) (Table 4). In the PD-L1/PD-L1 inhibitor group, it was found that there were no significant differences in the NLR, PLR, and PNI between the TD subgroup and the non-TD subgroup before and after treatment (Table S6).

Table 4 Within-group analysis of NLR, PLR, and PNI before and after treatment
Full size table

Comparison of PFS between TD and non-TD subgroups

Comparison of PFS

Of the 120 patients, progression was observed in 98 patients. Besides, 19 patients did not progress by the cut-off date, 2 were lost to follow-up, and 1 was discontinued due to a serious adverse reaction. Patients in the TD subgroup had a longer PFS (mPFS: 7.90 months vs. 4.87 months, p = 0.002), and a lower HR of progression (0.499, 95% CI (0.317–0.786)), compared with those in the non-TD subgroup (Fig. 2). The 6-month progression rate was significantly lower in the TD subgroup than in the non-TD subgroup (23.7% vs. 47.3%, p = 0.015). However, there was no difference in the 1-year progression rate between the two subgroups (65.8% vs. 78.3%, p = 0.160).

Fig. 2
figure 2

Comparison of PFS between TD and non-TD subgroups

Full size image

Of the PD- 1/PD-L1 inhibitors group, progression was observed in 49 patients, 10 did not progress by the cut-off date, and 1 was lost to follow-up. Similarly, compared with the non-TD subgroup, patients in the TD subgroup had a longer PFS (mPFS: 8.83 months vs. 6.50 months, p = 0.041) and a lower risk of progression (HR: 0.541, 95% CI (0.298–0.985)) (Figure S2a). The 6-month progression rate in the TD subgroup was significantly lower than that in the non-TD subgroup (10.3% vs. 40.7%, p = 0.009). However, there was no significant difference in the 1-year progression rate between the two subgroups (62.1% vs. 73.1%, p = 0.385).

TD occurred in 10 patients in the non-PD- 1/PD-L1 inhibitor group. There was no difference in median PFS between TD and non-TD subgroups (mPFS: 4.50 months vs. 3.767 months, p = 0.361, HR: 0.682, 95% CI (0.299–1.559)) (Figure S2b).

Analysis of confounding factors

Cox univariate analysis was performed on 120 patients to explore the effect of other confounding factors on PFS. The results showed that the TD occurrence, BMI, smoking history, PS score, bone metastasis, liver metastases, clinical stage, and PD- 1/PD-L1 inhibitor therapy may affect PFS (p ≤ 0.1) (Table 5). The significant factors mentioned above were included in a multivariable Cox regression model. The results showed that the occurrence of TD was a protective factor for PFS, and the risk of progression in the TD subgroup was 0.47 times that in the non-TD subgroup (HR: 0.470, 95% CI (0.296–0.747), p = 0.001). The PS score, bone metastasis, and liver metastasis were risk factors for PFS (p < 0.05) (Table 6).

Table 5 One-way COX regression analysis of PFS
Full size table
Table 6 Multivariable COX regression for PFS
Full size table

Using the same methodology, the results of the Cox univariate analysis of the 60 patients who in the PD- 1/PD-L1 inhibitor group showed that the occurrence of TD and PS scores impacted the PFS (p ≤ 0.1) (Table S7). After including the significant factors mentioned above in the multivariable Cox regression model, it was found that the occurrence of TD remained a protective factor for the occurrence of progression after controlling for confounding factors (HR: 0.536, 95% CI (0.294–0.980), p = 0.043), and PS score was an independent risk factor for the occurrence of progression (HR: 1.962, p = 0.029) (Table S8).

Comparison of recent efficacy between TD and non-TD subgroups

Among the 120 patients, the recent efficacy of patients who developed TD was significantly better than that of patients who did not develop TD (p = 0.006), and the ORR in the TD subgroup was significantly higher than that in the non-TD subgroup (64.1% vs. 33.3, p = 0.001). There was no difference in DCR (87.2% vs. 81.5, p = 0.433) and median duration of remission (DOR) (7.2 months vs. 5.23 months, p = 0.100) between TD and non-TD subgroups (Table 7).

Table 7 Comparison of recent efficacy between TD and non-TD subgroups
Full size table

The analysis of recent efficacy for the PD- 1/PD-L1 inhibitors group found that the ORR in the TD subgroup was significantly higher than that in the non-TD subgroup (69% vs. 38.7, p = 0.019). Similarly, there was no difference in DCR and DOR between TD and non-TD subgroups (p > 0.05). There was one case of CR on assessment in each of the TD and non-TD subgroups, but a significantly greater proportion of the TD subgroup achieved PR than the non-TD subgroup (65.5% vs. 35.5%). Overall, the best assessment was better in the TD subgroup than in the non-TD subgroup (p = 0.025) (Table S9).

Subgroup analysis

Comparison of PFS for different types of TD

The analysis of differences in PFS between different types of TD in all 39 patients who developed TD showed that there was no difference in the risk of progression among patients with overt hypothyroidism, overt hyperthyroidism, subclinical hypothyroidism, and subclinical hyperthyroidism (p > 0.05). HR was significantly higher in patients with low-T3 syndrome than in patients with other types of TD (HR: 14.268, p = 0.001) (Table 8). Besides, the risk of progression was higher in the low-T3 syndrome patients than in the non-TD subgroup (HR: 3.307, 95% CI (1.006–10.871)) (Figure S3a).

Table 8 Subgroup analysis of whether PFS differed by type of TD
Full size table

Subgroup analysis of the 29 patients who developed TD in the PD- 1/PD-L1 inhibitor group showed that there was no difference in the risk of progression among patients with overt hypothyroidism, overt hyperthyroidism, subclinical hypothyroidism, and subclinical hyperthyroidism (p > 0.05). Similarly, patients who developed low-T3 syndrome had the highest risk of progression compared with patients who developed other types of TD (HR: 24.495, p = 0.024) (Table S10). There was no significant difference in the risk of progression between patients with low T3 syndrome and those in the non-TD subgroup (HR: 1.748, 95% CI (0.233–13.696)) (Figure S3b).

Factors influencing PFS within the TD subgroup

Of all 39 patients who developed TD, 10 (25.6%) returned to normal after treatment, with a median time of regression of 1.7 months. The results of Cox univariate analysis showed that the risk of progression decreased by 0.093 for every 1-unit increase in the duration of TD (HR: 0.907, 95% CI (0.844–0.975), p = 0.008). It was also found that the PFS within the TD subgroup was not relevant to the baseline TSH level, time of TD occurrence, severity of TD, TD regression, and time to regression (Table 9).

Table 9 One-way COX regression analysis of PFS within the TD subgroup
Full size table

Of the 29 patients who developed TD in the PD- 1/PD-L1 inhibitor group, 9 patients (31.0%) recovered after a period of time, with a median time of regression of 1.63 months. After including factors with p ≤ 0.1 in the multivariable Cox regression analysis, it was found that the risk of progression decreased by 0.523 for each 1-unit increase in time of TD occurrence (HR: 0.477, 95% CI (0.354–0.645), p < 0.001), and by 0.494 for each 1-unit increase in the duration of TD (HR: 0.506, 95% CI (0.384–0.667), p < 0.001). Factors such as baseline TSH, severity of TD, TD regression, and time to regression within the TD subgroup did not have a significant effect on PFS (Table S11).

Discussion

Thyroid dysfunction is one of the most common (40%) endocrine toxicities associated with the use of PD- 1/PD-L1 inhibitors in the treatment of various tumors [17]. This study found that the incidence of TD was significantly higher in the PD- 1/PD-L1 inhibitor group than in the non-PD- 1/PD-L1 inhibitor group (48.3% vs. 16.7%), and most were grade 1. Only 5 patients (grade 2) received levothyroxine replacement therapy because their TSH was greater than 10 mIU/L. The mechanism of PD- 1/PD-L1 inhibitor therapy-inducing TD has not been fully elucidated. The possible mechanisms are as follows: Firstly, TD occurs as a destructive thyroiditis mediated by helper T cells, Treg cells, and a variety of cytokines [18,19,20]. It is now mostly considered to be a bystander effect of activated T cells, with the possibility of autoimmune toxicity greater in patients responding to ICIs [21]. Kotwal et al. [22] reported that PD1+ T-lymphocytes in thyroid tissue were significantly higher in patients with TD than in patients without TD, which might be one of the reasons for the frequent attacks on the thyroid gland. Secondly, autoimmunity mediated by thyroid autoantibodies is also involved [23], but the mechanism of its occurrence is not exactly the same as that of autoimmune thyroiditis [24]. Yamauchi et al. [25] reported that not all patients who developed TD expressed thyroid autoantibodies, so cross-antigens might exist in thyroid and lung tissues, which could explain the correlation between the development of TD and the good prognosis of patients with advanced lung cancer. However, the specific mechanism still needs to be further explored. Finally, a decrease in the number of immunosuppressive cells (CD14+ HLA-DR+lo/neg monocytes) could also induce TD [26]. In terms of embryonic development, although the thyroid gland and lung originate from different endoderm, thyroid transcription factor 1 (TTF- 1) is expressed in both lung and thyroid tissues, which can control the embryonic development of the thyroid gland and lung. Therefore, both have homology [27]. The development of TD has been found to be associated with low-level TTF- 1 expression. Koyama et al. found that TTF- 1-negative lung cancer patients with TD had a longer PFS than TTF- 1-negative patients without TD and TTF- 1-positive patients with/without TD (10.3 months vs. 2.4 months vs. 4.2 months vs. 1.4 months) [28]. Traditionally, chemotherapy has been considered to be immunosuppressive [29], but recent studies have found that platinum-based chemotherapy can reduce the level of PD-L2 in tumor cells, as well as the levels of PD-L1 and PD-L2 on dendritic cells, thus activating T-cells to kill tumor cells [30, 31]. This is the evidence for recommending PD- 1/PD-L1 inhibitors in combination with chemotherapy, and may be the reason why chemotherapy leads to TD. Some patients with lung adenocarcinoma are treated with PD- 1/PD-L1 inhibitors in combination with bevacizumab, but no studies have shown that this therapy can cause TD [32,33,34]. Therefore, TD is mainly caused by PD- 1/PD-L1 inhibitors or, less likely, by chemotherapy, whereas bevacizumab has no significant effect on the thyroid gland.

Predictors of TD occurrence may help alert clinicians to closely monitor the thyroid function of patients and formulate individualized diagnostic and therapeutic plans. It may also help explore the molecular mechanisms of TD occurrence. However, there is a lack of reliable biomarkers to predict the occurrence of TD currently. This study found that TSH was a key indicator of thyroid function. Specifically, the baseline TSH level was higher in the TD subgroup than in the non-TD subgroup, with or without taking into account the effect of non-PD- 1/PD-L1 inhibitors (p < 0.05). Pollack et al. [35] reported that the risk of TD was significantly increased when the baseline TSH level was greater than 2.19 mIU/L. The present study found that the risk of TD was significantly increased when the baseline TSH level was greater than 1.775 mIU/L (or 1.765 mIU/L). This suggests that clinicians should closely monitor thyroid function, especially the TSH level. Although baseline TSH levels may still within the normal range, patients with higher baseline TSH levels are more likely to develop TD after a period of treatment with PD- 1/PD-L1 inhibitors. Especially, when the baseline TSH level is greater than 1.765 mIU/L, clinicians should pay attention to monitoring changes in thyroid function. Pollack et al. [36] found that the mean BMI of patients who developed thyroid dysfunction was significantly higher than that of the group with normal thyroid function. However, in the present study, it was found that there was no significant difference in the mean BMI levels of the TD group and the non-TD group. Further analysis found that the proportion of obese patients in the study of Pollack et al. was as high as 20%, whereas the proportion of obese patients in the present study was only 11.7%. Therefore, racial differences may explain the difference in the results of the two studies (Pollack et al.’s study was conducted in Israel, with a predominance of Caucasian, whereas the present study was conducted in China, with a predominantly yellow population). Besides, it is necessary to increase the number of enrolled obese patients. Previous studies have found an association between gender [37], type of PD- 1/PD-L1 inhibitors [38], hypertension [39], smoking history [40] and the development of TD. However, significant correlations were not detected in the present study. NLR, PLR, and PNI are often used to reflect the body’s active inflammation and inflammatory depletion [41]. Besides, PNI is often used to assess nutritional status and immune status [42]. Previous studies have found that patients who developed irAEs have significantly decreased NLR and PLR [43] and significantly increased PNI [44]. TD is the most common irAE of the endocrine system. However, in the present study, a significant decrease in NLR was found in the non-TD subgroup in the analysis of all cases, but no significant NLR change was found in the analysis of PD- 1/PD-L1 inhibitor group. Considering that the analysis on all cases was strongly influenced by the effect of the non-PD- 1/PD-L1 inhibitor group, therefore, the analysis results of the PD- 1/PD-L1 inhibitor group were more credible. Then, it was speculated that there was no significant difference in NLR between the TD subgroup and the non-TD subgroup before and after PD- 1/PD-L1 inhibitor treatment. This study also did not find significant differences in the PLR and PNI between the TD and non-TD subgroups before and after treatment. This suggests that the predictive role of NLR, PLR, and PNI for irAEs may be system or organ-specific. NLR, PLR, and PNI may be correlated with the irAEs in the whole [43] or organs such as stomach and lung [45], but are not correlated with the occurrence of TD.

Chmielewska et al. [46] observed that lung cancer patients undergoing a second-line treatment with nivolumab and experiencing endocrine-related toxicity (thyroid dysfunction predominantly) exhibited significantly prolonged PFS compared to those without such toxicity (9 months vs. 2 months). However, in the study of Wu et al. [14], the occurrence of thyroid dysfunction was not significantly correlated with the PD- 1/PD-L1 inhibitor efficacy and prognosis of lung cancer patients. Therefore, it is still controversial whether there is a correlation between the occurrence of TD after PD- 1/PD-L1 inhibitor treatment and the outcome of advanced lung cancer patients. While in the present study, the analysis of the relationship between TD occurrence and PFS found that PFS was significantly longer in the TD subgroup than in the non-TD subgroup, and the occurrence of TD was a protective factor for the occurrence of progression. This suggests that patients who develop TD have a better prognosis, and the development of TD can be used as a clinical biomarker to predict a good prognosis. This is similar to the results of Cheung et al. [47]. ORR is often used as an important indicator to assess the recent efficacy. The results of this study showed that the ORR of the TD subgroup was significantly higher than that of the non-TD subgroup. This suggests that when using PD- 1/PD-L1 inhibitors for the treatment of advanced lung cancer, patients who developed TD can better activate the immune system, resulting in a better anti-tumor effect, compared with those who did not develop TD [48]. In the PD- 1/PD-L1 inhibitor group, the PFS and ORR in the TD subgroup were higher than those in the non-TD subgroup. However, in the non-PD- 1/PD-L1 inhibitor group, there was no significant difference in PFS between TD and non-TD subgroups. This suggests that the TD is positively correlated with a good outcome only when using PD- 1/PD-L1 inhibitors for treatment. As for why the occurrence of TD is positively correlated with good efficacy, some studies believe that it is the bystander effect of the activated T cells; The more pronounced response to ICIs, the more activated T cells. It play a destructive role in other immune organs such as the thyroid gland while providing a good anti-tumor effect [21]. Besides, to explain why thyroid dysfunction occurs when treating lung cancer, Yamauchi et al. [25] held that there might be not only lung-associated antigens but also cross-antigens in the thyroid tissue when treating lung cancer with PD- 1/PD-L1 inhibitors.

In clinical research, the survival time of the observed subjects is often affected by several factors, such as treatment, age, gender, disease severity, etc. Besides, the distribution of the survival data often does not conform to the normal distribution. Therefore, it is necessary to carry out multivariable Cox regression analysis to find out whether there is still a certain degree of correlation between the independent variables and the outcome indicators after controlling for the confounding factors. The results of multivariable Cox regression analysis are more convincing than those of univariate analysis [49]. Ran et al. [50] reported that the higher the PS score, the shorter the PFS for immunotherapy. Landi et al. [51] reported that advanced lung cancer patients who developed bone metastases received limited benefit from immunotherapy, and their overall survival was not as high as that of those without bone metastases (7.4 months vs. 15.3 months). The multivariable Cox regression analysis of all cases and only the PD- 1/PD-L1 inhibitor group found that the occurrence of TD was an independent protective factor for progression, whereas PS score was an independent risk factor for the occurrence of progression. This is consistent with the findings of Landi et al. [51]. Although bone metastases and liver metastases were independent risk factors for progression when analyzing all cases, this correlation was not found when analyzing the PD- 1/PD-L1 inhibitor group. Since the main purpose of this study was to investigate the correlation between TD and efficacy of PD- 1/PD-L1 inhibitor treatment, and the analysis results of all cases were susceptible to be influenced by the non-PD- 1/PD-L1 inhibitor group, the analysis results of the PD- 1/PD-L1 inhibitor group were more indicative of the factors affecting PFS after PD- 1/PD-L1 inhibitor treatment. Therefore, this study only considered the PS score and the occurrence of TD as the independent influences on PFS after PD- 1/PD-L1 inhibitor treatment. The association of TD with a good prognosis has also been confirmed in other studies. In a study by Baek et al. [52], 191 patients treated with ICIs, including lung cancer, melanoma, urothelial carcinoma and various gastrointestinal tumors, were found that patients with TD had a better prognosis than patients with no TD (OS: 25 months vs. 18 months, respectively, p = 0.005), which has been confirmed in other studies [53, 54].

Low-T3 syndrome is characterized by reduced fT3 level and normal TSH and fT4 levels [55]. Thuillier et al. [56] excluded low-T3 syndrome from the syndromes of TD. However, according to recent studies, low-T3 syndrome is an abnormal thyroxine metabolism that involves a variety of complex physio-pathological processes and is associated with acute critical illnesses and chronic diseases [57]. It may be an early stage of thyrotoxicosis due to the lack of fT3 in tissues induced by strong immune response [17]. Therefore, this study categorized low-T3 syndrome as a type of TD. Baek et al. [52] reported a significant improvement in prognosis in the hypothyroidism group compared with other types of TD (p = 0.002). However, Yu et al. [58] found that patients in the hyperthyroidism group had the most significant improvement in prognosis, followed by subclinical hyperthyroidism. Thuillier et al. [56] reported that there was no significant difference in survival indices between different types of TD. The present study found no significant difference in PFS among patients who developed hypothyroidism, hyperthyroidism, subclinical hypothyroidism, and subclinical hyperthyroidism. However, patients with low-T3 syndrome had a higher risk of progression than patients with other types of TD and without TD (HR > 1). Therefore, this study concluded that the correlation between TD and good prognosis is not affected by the type of TD, but it should be noted that low-T3 syndrome may be a predictor of poor prognosis. Low T3 syndrome was the earliest to appear in this study. This reminds clinicians to monitor patients’ thyroid function and detect low T3 syndrome at the early stage.

The TDs identified in the study were all at grade 1–2, which were not clinically significant. Besides, there were no discontinuations or deaths due to severe TDs, and 25.6% (PD- 1/PD-L1 inhibitor group: 31.0%) of patients with TDs returned to normal after a period of time. This is consistent with the results of Thuillier et al. [56], that is, irAEs-associated hyperthyroidism is transitory, which may progress to hypothyroidism or return to normal after a period of time. This study found that baseline TSH level, TD severity, regression or not of TD, and time to regression were not associated with PFS. When analyzing all cases, only the duration of TD was found to be an influencing factor for PFS, but when analyzing the PD- 1/PD-L1 inhibitor group, both the duration of TD and the time of occurrence of TD were found to be able to influence the PFS of the patients. Considering that the latter was not affected by the non-PD- 1/PD-L1 inhibitor group and the results were accurate, the results of the latter were more convincing. The later the occurrence of TD, and the longer the duration of TD, the lower the risk of progression (HR < 1, p < 0.05). This has not been reported in previous studies. Therefore, the time of occurrence and duration of TD may be influential factors affecting the prognosis within the TD subgroup.

This study has following limitations. Firstly, the patients included in the study were not randomly sampled, and the small sample size might lead to selection bias. Since this study is retrospective, there may be recall bias. To mitigate this, we employed rigorous data collection methods, including cross-referencing multiple sources of information to ensure consistency in data gathering. Secondly, not all patients had a complete thyroid function test at each follow up, so it is not possible to accurately judge the detailed thyroid function changes. Thyroid autoantibodies are closely related to the occurrence of TD [23]. However, in this study, autoantibody detection is not routinely used in the clinic, so the relevant data could not be collected for further analysis. PD-L1 expression levels may be an indicator of PFS [59]. However, in this study, PD-L1 expression level was not detected for most patients due to the fact that this detection was at patients’ own expense. Since this study was only conducted in China, these findings may vary globally due to population differences (e. g., race, genetics, environment, etc.). Finally, due to the time constraints of this study, the follow-up was not long enough. However, follow-up will be continued until the patient has OS, to further assess the correlation between TD and prognosis. Moreover, subsequent prospective studies will continue to further validate the conclusion that TD is associated with a good prognosis.

Conclusions

The use of PD- 1/PD-L1 inhibitors in the treatment of advanced lung cancer often led to the occurrence of TD. However, most TDs were mild, and only a small number needed symptomatic treatment. The higher the baseline TSH level, the more likely to develop TD, so the baseline TSH level may be predictive factors for the development of TD. Clinicians can detect thyroid function, especially the TSH level, before treatment with ICIs to predict the risk of developing TD. The development of TD is associated with a good prognosis and may be not affected by the type of PD- 1/PD-L1 inhibitors and the common types of TD. However, it should be noted that the occurrence of low-T3 syndrome may be associated with a poor prognosis, and it is the earliest to occur. The later the onset and the longer the duration of TD, the longer the PFS of the patients.

Data availability

The data and materials for this study are available from the corresponding author on reasonable request.

Abbreviations

AUC:

Area under the curve

ANOVA:

Analysis of variance

BMI:

Body mass index

CI:

Confidence interval

CR:

Complete response

CSCO:

Chinese Society of Clinical Oncology

CTCAE 5.0:

Common Terminology Criteria for Adverse Events Version 5.0

DCR:

Disease control rate

DOR:

Duration of response

FT3:

Free triiodothyronine

FT4:

Free thyroxine

HR:

Hazard ratio

ICI:

Immune checkpoint inhibitor

IrAEs:

Immune-related adverse events

IQR:

Interquartile range

M:

Median

mOS:

Median overall survival

NSCLC:

Non-small-cell lung cancer

NLR:

Neutrophil/lymphocyte ratio

ORR:

Objective response rate

OR:

Odds ratios

PD-L1:

Programmed death ligand 1

PLR:

Platelet/lymphocyte ratio

PNI:

Prognostic nutrition index

PFS:

Progression-free survival

PS Score:

Physical status score

PD-L2:

Programmed death ligand 2

PD:

Progressive disease

PR:

Partial response

RECIST 1.1:

Response Evaluation Criteria in Solid Tumors Version 1.1

ROC:

Receiver operating characteristic

SD:

Stable disease

SPSS 22.0:

Statistical Product and Service Solutions Version 22.0

SD:

Standard deviation

SCLC:

Small-cell lung cancer

TD:

Thyroid dysfunction

TTF- 1:

Thyroid transcription factor 1

T3:

Triiodothyronine

TNM:

Tumor-node-metastasis

χ:

Mean

References

  1. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209–49.

    Article  PubMed  Google Scholar 

  2. Lu S, Yu Y, Yang Y. Retrospect and Prospect for Lung Cancer in China: Clinical Advances of Immune Checkpoint Inhibitors. Oncologist. 2019;24(Suppl 1):S21-s30.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Min HY, Lee HY. Mechanisms of resistance to chemotherapy in non-small cell lung cancer. Arch Pharm Res. 2021;44(2):146–64.

    Article  CAS  PubMed  Google Scholar 

  4. Wang C, Qiao W, Jiang Y, et al. The landscape of immune checkpoint inhibitor plus chemotherapy versus immunotherapy for advanced non-small-cell lung cancer: A systematic review and meta-analysis. J Cell Physiol. 2020;235(5):4913–27.

    Article  CAS  PubMed  Google Scholar 

  5. Scott SC, Hann CL. Immunotherapy for small cell lung cancer: established applications and novel approaches. Clin Adv Hematol Oncol. 2021;19(10):654–63.

    PubMed  Google Scholar 

  6. Okiyama N, Tanaka R. Immune-related adverse events in various organs caused by immune checkpoint inhibitors. Allergol Int. 2022;71(2):169–78.

    Article  CAS  PubMed  Google Scholar 

  7. Chuzi S, Tavora F, Cruz M, et al. Clinical features, diagnostic challenges, and management strategies in checkpoint inhibitor-related pneumonitis. Cancer Manag Res. 2017;9:207–13.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Zhou X, Yao Z, Bai H, et al. Treatment-related adverse events of PD-1 and PD-L1 inhibitor-based combination therapies in clinical trials: a systematic review and meta-analysis. Lancet Oncol. 2021;22(9):1265–74.

    Article  CAS  PubMed  Google Scholar 

  9. Jing Y, Zhang Y, Wang J, et al. Association Between Sex and Immune-Related Adverse Events During Immune Checkpoint Inhibitor Therapy. J Natl Cancer Inst. 2021;113(10):1396–404.

    Article  PubMed  Google Scholar 

  10. Guo J, Mu D, Han Y. Research Progress on the Mechanism and Clinical Data of Cereblon in Reversing the Resistance of Lung Cancer to PD-1 Antibody by T cells. Zhongguo Fei Ai Za Zhi. 2021;24(1):49–55.

    PubMed  Google Scholar 

  11. Lisberg A, Tucker DA, Goldman JW, et al. Treatment-Related Adverse Events Predict Improved Clinical Outcome in NSCLC Patients on KEYNOTE-001 at a Single Center. Cancer Immunol Res. 2018;6(3):88–294.

    Article  Google Scholar 

  12. Kotwal A, Kottschade L, Ryder M. PD-L1 Inhibitor-Induced Thyroiditis Is Associated with Better Overall Survival in Cancer Patients. Thyroid. 2020;30(2):177–84.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Rubino R, Marini A, Roviello G, et al. Endocrine-related adverse events in a large series of cancer patients treated with anti-PD1 therapy. Endocrine. 2021;74(1):172–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Wu Y, Wang Z, Bai H, et al. Thyroid dysfunction during PD-1 inhibitor treatment in patients with cancer: Incidence and association with progression-free survival. Oncol Lett. 2022;24(3):309.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. D’Aiello A, Lin J, Gucalp R, et al. Thyroid Dysfunction in Lung Cancer Patients Treated with Immune Checkpoint Inhibitors (ICIs): Outcomes in a Multiethnic Urban Cohort. Cancers (Basel). 2021;13(6):1464.

    Article  PubMed  Google Scholar 

  16. 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.

    Article  CAS  PubMed  Google Scholar 

  17. Percik R, Liel Y, Urban D, et al. Thyroid dysfunction and survival in cancer patients treated with immune checkpoint inhibitors: analyses from a large single tertiary cancer center database. Acta Oncol. 2021;60(11):1466–71.

    Article  CAS  PubMed  Google Scholar 

  18. Yamauchi I, Sakane Y, Fukuda Y, et al. Clinical Features of Nivolumab-Induced Thyroiditis: A Case Series Study. Thyroid. 2017;27(7):894–901.

    Article  CAS  PubMed  Google Scholar 

  19. Kurimoto C, Inaba H, Ariyasu H, et al. Predictive and sensitive biomarkers for thyroid dysfunctions during treatment with immune-checkpoint inhibitors. Cancer Sci. 2020;111(5):1468–77.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Luty J, Ruckemann-Dziurdzińska K, Witkowski JM, et al. Immunological aspects of autoimmune thyroid disease - Complex interplay between cells and cytokines. Cytokine. 2019;116:128–33.

    Article  CAS  PubMed  Google Scholar 

  21. Das S, Johnson DB. Immune-related adverse events and anti-tumor efficacy of immune checkpoint inhibitors. J Immunother Cancer. 2019;7(1):306.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Kotwal A, Gustafson MP, Bornschlegl S, et al. Immune Checkpoint Inhibitor-Induced Thyroiditis Is Associated with Increased Intrathyroidal T Lymphocyte Subpopulations. Thyroid. 2020;30(10):1440–50.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  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(3):583–9.

    Article  CAS  PubMed  Google Scholar 

  24. Mazarico I, Capel I, Giménez-Palop O, et al. Low frequency of positive antithyroid antibodies is observed in patients with thyroid dysfunction related to immune check point inhibitors. J Endocrinol Invest. 2019;42(12):1443–50.

    Article  CAS  PubMed  Google Scholar 

  25. Yamauchi I, Yasoda A, Matsumoto S, et al. Incidence, features, and prognosis of immune-related adverse events involving the thyroid gland induced by nivolumab. PLoS ONE. 2019;14(5): e0216954.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Delivanis DA, Gustafson MP, Bornschlegl S, et al. Pembrolizumab-Induced Thyroiditis: Comprehensive Clinical Review and Insights Into Underlying Involved Mechanisms. J Clin Endocrinol Metab. 2017;102(8):2770–80.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Ikeda K, Clark JC, Shaw-White JR, et al. Gene structure and expression of human thyroid transcription factor-1 in respiratory epithelial cells. J Biol Chem. 1995;270(14):8108–14.

    Article  CAS  PubMed  Google Scholar 

  28. Koyama J, Horiike A, Yoshizawa T, et al. Correlation between thyroid transcription factor-1 expression, immune-related thyroid dysfunction, and efficacy of anti-programmed cell death protein-1 treatment in non-small cell lung cancer. J Thorac Dis. 2019;11(5):1919–28.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Schirrmacher V. From chemotherapy to biological therapy: A review of novel concepts to reduce the side effects of systemic cancer treatment (Review). Int J Oncol. 2019;54(2):407–19.

    Article  CAS  PubMed  Google Scholar 

  30. Hato SV, Khong A, de Vries IJ, et al. Molecular pathways: the immunogenic effects of platinum-based chemotherapeutics. Clin Cancer Res. 2014;20(11):2831–7.

    Article  CAS  PubMed  Google Scholar 

  31. Vansteenkiste J, Wauters E, Reymen B, et al. Current status of immune checkpoint inhibition in early-stage NSCLC. Ann Oncol. 2019;30(8):1244–53.

    Article  CAS  PubMed  Google Scholar 

  32. Zhou Y, He M, Li R, et al. The Safety and Effectiveness of Bevacizumab in the Treatment of Nonsquamous Non-Small-Cell Lung Cancer: A Meta-Analysis of Randomized Controlled Trials. Biomed Res Int. 2021;2021:5537899.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Liu M, Luo N, Fang Z, et al. The efficacy and toxicity of maintenance therapy with bevacizumab plus pemetrexed versus bevacizumab/pemetrexed alone for stage IIIB/IV nonsquamous non-small cell lung cancer: A meta-analysis of randomized controlled trials. J Clin Pharm Ther. 2022;47(2):157–67.

    Article  CAS  PubMed  Google Scholar 

  34. Wu Y, Du B, Lv C, et al. First-generation EGFR-TKI plus bevacizumab and chemotherapy for advanced EGFR-mutated non-squamous non-small-cell lung cancer: a retrospective study. Ann Med. 2023;55(2):2243967.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Pollack RM, Kagan M, Lotem M, et al. Baseline Tsh Level Is Associated with Risk of Anti-Pd-1-Induced Thyroid Dysfunction. Endocr Pract. 2019;25(8):824–9.

    Article  PubMed  Google Scholar 

  36. 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.

    Article  PubMed  Google Scholar 

  37. Triggianese P, Novelli L, Galdiero MR, et al. Immune checkpoint inhibitors-induced autoimmunity: The impact of gender. Autoimmun Rev. 2020;19(8): 102590.

    Article  CAS  PubMed  Google Scholar 

  38. Barroso-Sousa R, Barry WT, Garrido-Castro AC, et al. Incidence of Endocrine Dysfunction Following the Use of Different Immune Checkpoint Inhibitor Regimens: A Systematic Review and Meta-analysis. JAMA Oncol. 2018;4(2):173–82.

    Article  PubMed  Google Scholar 

  39. Coelho MA, de Carné TS, Rana S, et al. Oncogenic RAS Signaling Promotes Tumor Immunoresistance by Stabilizing PD-L1 mRNA. Immunity. 2017;47(6):1083-1099.e1086.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Calles A, Liao X, Sholl LM, et al. Expression of PD-1 and Its Ligands, PD-L1 and PD-L2, in Smokers and Never Smokers with KRAS-Mutant Lung Cancer. J Thorac Oncol. 2015;10(12):1726–35.

    Article  CAS  PubMed  Google Scholar 

  41. Yamamoto T, Kawada K, Obama K. Inflammation-Related Biomarkers for the Prediction of Prognosis in Colorectal Cancer Patients. Int J Mol Sci. 2021;22(15):8002.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Jiang P, Li X, Wang S, et al. Prognostic Significance of PNI in Patients With Pancreatic Head Cancer Undergoing Laparoscopic Pancreaticoduodenectomy. Front Surg. 2022;9: 897033.

    Article  PubMed  PubMed Central  Google Scholar 

  43. von Itzstein MS, Khan S, Gerber DE. Investigational Biomarkers for Checkpoint Inhibitor Immune-Related Adverse Event Prediction and Diagnosis. Clin Chem. 2020;66(6):779–93.

    Article  Google Scholar 

  44. 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(9):1813–22.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Fujisawa Y, Yoshino K, Otsuka A, et al. Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab. J Dermatol Sci. 2017;88(2):225–31.

    Article  CAS  PubMed  Google Scholar 

  46. Chmielewska I, Dudzińska M, Szczyrek M, et al. Do endocrine adverse events predict longer progression-free survival among patients with non-small-cell lung cancer receiving nivolumab? PLoS ONE. 2021;16(9): e0257484.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. 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(8):1795–812.

    Article  PubMed  PubMed Central  Google Scholar 

  48. Kim HI, Kim M, Lee SH, et al. Development of thyroid dysfunction is associated with clinical response to PD-1 blockade treatment in patients with advanced non-small cell lung cancer. Oncoimmunology. 2017;7(1): e1375642.

    Article  PubMed  PubMed Central  Google Scholar 

  49. van der Net JB, Janssens AC, Eijkemans MJ, et al. Cox proportional hazards models have more statistical power than logistic regression models in cross-sectional genetic association studies. Eur J Hum Genet. 2008;16(9):1111–6.

    Article  PubMed  Google Scholar 

  50. Zeng R, Liu F, Fang C, et al. PIV and PILE Score at Baseline Predict Clinical Outcome of Anti-PD-1/PD-L1 Inhibitor Combined With Chemotherapy in Extensive-Stage Small Cell Lung Cancer Patients. Front Immunol. 2021;12: 724443.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Landi L, D’Incà F, Gelibter A, et al. Bone metastases and immunotherapy in patients with advanced non-small-cell lung cancer. J Immunother Cancer. 2019;7(1):316.

    Article  PubMed  PubMed Central  Google Scholar 

  52. Baek HS, Jeong C, Shin K, et al. Association between the type of thyroid dysfunction induced by immune checkpoint inhibitors and prognosis in cancer patients. BMC Endocr Disord. 2022;22(1):89.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Lima Ferreira J, Costa C, Marques B, et al. Improved survival in patients with thyroid function test abnormalities secondary to immune-checkpoint inhibitors. Cancer Immunol Immunother. 2021;70(2):299–309.

    Article  CAS  PubMed  Google Scholar 

  54. Yoon JH, Hong AR, Kim HK, et al. Characteristics of Immune-Related Thyroid Adverse Events in Patients Treated with PD-1/PD-L1 Inhibitors. Endocrinol Metab (Seoul). 2021;36(2):413–23.

    Article  CAS  PubMed  Google Scholar 

  55. Lee S, Farwell AP. Euthyroid Sick Syndrome. Compr Physiol. 2016;6(2):1071–80.

    Article  PubMed  Google Scholar 

  56. Thuillier P, Joly C, Alavi Z, et al. Thyroid dysfunction induced by immune checkpoint inhibitors is associated with a better progression-free survival and overall survival in non-small cell lung cancer: an original cohort study. Cancer Immunol Immunother. 2021;70(7):2023–33.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Fliers E, Boelen A. An update on non-thyroidal illness syndrome. J Endocrinol Invest. 2021;44(8):1597–607.

    Article  CAS  PubMed  Google Scholar 

  58. Zhou Y, Xia R, Xiao H, et al. Thyroid function abnormality induced by PD-1 inhibitors have a positive impact on survival in patients with non-small cell lung cancer. Int Immunopharmacol. 2021;91: 107296.

    Article  CAS  PubMed  Google Scholar 

  59. Chen R, Tao Y, Xu X, et al. The efficacy and safety of nivolumab, pembrolizumab, and atezolizumab in treatment of advanced non-small cell lung cancer. Discov Med. 2018;26(143):155–66.

    PubMed  Google Scholar 

Download references

Acknowledgements

We thank the First Affiliated Hospital of Shihezi University in Xinjiang for providing data support, and especially thank the Talent Boosting Plan of Shanghai Fourth People’s Hospital Affiliated to Tongji University (NO. SY-XKZT- 2019 - 3007) for financial support. We also thank all the authors for their selfless contribution to this article.

Funding

This study was supported by the Talent Boosting Plan of Shanghai Fourth People’s Hospital Affiliated to Tongji University (NO. SY-XKZT- 2019–3007).

Author information

Author notes

  1. Xiaoping Ma and Jia Ma equal first author.

Authors and Affiliations

  1. Department of Radiotherapy, Shanghai Fourth People’s Hospital, School of Medicine, Tongji University, Shanghai, 200434, People’s Republic of China

    Wei Gao & Ping Dai

  2. Department of Oncology, The First Affiliated Hospital of School of Medicine Shihezi University, Shihezi, Xinjiang, 832000, People’s Republic of China

    Yanling Wang, Xiaoping Ma, Jing Li & Zhiyi Lin

  3. School of Medicine, Shihezi University, Shihezi, Xinjiang, 832000, People’s Republic of China

    Ping Gong

  4. Department of General Surgery, Shanghai Jian Gong Hospital, Shanghai, 200083, People’s Republic of China

    Jia Ma

Authors
  1. Yanling Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Xiaoping Ma
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Jia Ma
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Jing Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Zhiyi Lin
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Wei Gao
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Ping Gong
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Ping Dai
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Yanling Wang: Conceptualization, Formal analysis, Investigation, Methodology, Software, Visualization, Data curation, Validation, Writing-original draft. Xiaoping Ma: Methodology, Data curation, Writing-original draft. Jia Ma: Software, Methodology. Jing Li: Investigation. Zhiyi Lin: Investigation. Wei Gao: Conceptualization, Resources. Ping Gong: Conceptualization, Supervision, Resources, Project administration, Validation. Ping Dai: Supervision, Project administration, Funding Acquisition, Methodology, Validation, Writing-review & editing. All authors reviewed the manuscript.

Corresponding authors

Correspondence to Wei Gao, Ping Gong or Ping Dai.

Ethics declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the First Affiliated Hospital of Shihezi University under the ethical number: KJX 2022–081 - 01. All participants have given informed consent to participate in this study, and all the participants were aware of the study’s purpose, risks, and benefits.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Ping Dai is the lead corresponding author author.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, Y., Ma, X., Ma, J. et al. Thyroid dysfunction as a predictor of PD- 1/PD-L1 inhibitor efficacy in advanced lung cancer. BMC Cancer 25, 791 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12885-025-14097-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12885-025-14097-w

Keywords

BMC Cancer

ISSN: 1471-2407

Contact us