Camrelizumab and apatinib in Chinese patients with locally advanced or metastatic radioiodine-refractory differentiated thyroid cancer: an exploratory open-label, single-arm trial Authors: Xin Zhang, Di Sun, Yingqiang Zhang, Zhuanzhuan Mu, Cong Shi, Yuqing Sun, Xinqi Cheng, Chao Meng, Xing Wei, Jun Liang, Yansong Lin Published: 2025 ABSTRACT BackgroundRadioactive iodine-refractory differentiated thyroid cancer (RAIR-DTC) remains a therapeutic challenge, especially after progression on prior systemic therapies. While immune checkpoint inhibitors combined with antiangiogenic agents have shown synergy in other tumors, data in RAIR-DTC are limited. This study evaluated camrelizumab plus apatinib in first-line and salvage settings.MethodsThis open-label, single-arm, phase II trial (NCT04560127) enrolled patients with locally advanced or metastatic RAIR-DTC. Patients received intravenous camrelizumab (200 mg every 2 weeks) and oral apatinib (250 mg daily) in 4-week cycle until progression or intolerable toxicity. The primary endpoint was progression-free survival (PFS). Secondary endpoints included objective response rate (ORR), overall survival (OS), disease control rate (DCR), and safety.ResultsBetween September 2020 and August 2021, 20 patients were enrolled and treated. As of the data cutoff, the median follow-up was 28.4 months (IQR, 22.3–31.2). Median PFS was 11.0 months (95% CI, 6.0–14.2), with 11.6 months (95% CI, 4.8–NE) in treatment-naïve patients and 11.0 months (95% CI, 3.7–19.4) in pretreated patients. Median OS was not reached. The confirmed ORR was 30.0% (95% CI, 11.9–54.3), with 6 patients achieving partial response, and DCR was 95.0% (95% CI, 75.1–99.9). The most common grade ≥ 3 treatment-related adverse events were increased gamma-glutamyltransferase (40.0%) and increased alanine aminotransferase (30.0%).ConclusionsCamrelizumab plus apatinib demonstrated promising efficacy and manageable toxicity in RAIR-DTC across both first-line and salvage settings, supporting further evaluation as a potential second-line option in the absence of established standard therapies in this setting.Trial registrationThe study was registered at clinicaltrials.gov with the identifier NCT04560127.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12916-025-04550-9. Background Thyroid carcinoma (TC) ranks as the seventh most prevalent malignancy globally, with China alone accounting for more than half of newly diagnosed cases each year [1]. Among its subtypes, differentiated thyroid cancer (DTC) predominates, comprising over 90% of cases. The majority of DTC patients respond well to conventional treatments, including thyroidectomy, radioactive iodine (RAI) therapy, and thyroid-stimulating hormone (TSH)-suppressive therapy [2]. However, a subset of patients, particularly those with locally advanced disease or distant metastases, may develop resistance to RAI therapy, termed RAI-refractory DTC (RAIR-DTC) [3]. The definition of RAIR-DTC is based primarily on clinical response to RAI therapy, rather than on precise pathological criteria, and there is a lack of worldwide consensus among clinical guidelines. Furthermore, its identification is influenced by clinicians’ interpretations of RAI avidity and treatment efficacy [4–6] and is consistently associated with a markedly poor prognosis, with a 10-year survival rate of approximately 10% [7]. These factors collectively make the management of RAIR-DTC a significant challenge in thyroid cancer, highlighting an urgent need for more effective therapeutic strategies. The emergence of tyrosine kinase inhibitors (TKIs) has significantly expanded treatment options for RAIR-DTC. Lenvatinib and sorafenib, both multitargeted TKIs, are currently the standard first-line systemic therapies, as demonstrated by their efficacy in the DECISION [8] and SELECT [9] trials. More recently, highly selective agents targeting RET or NTRK fusions have shown remarkable efficacy in patients with corresponding fusions [10, 11]. However, these alterations are present in only a small fraction of DTC patients, leaving a substantial unmet need for the majority who lack these actionable targets. Most RAIR-DTC patients developed resistance to these multikinase inhibitors over the following 1 to 2 years [12]. The COSMIC-311 trial established cabozantinib as an effective second-line option following prior targeted therapy, leading to its approval as the standard treatment in this setting [13]. Despite these advancements, disease progression remains a major challenge in many advanced RAIR-DTC cases. To date, no established standard of care exists in China for patients who experience progression on TKI-targeted therapy, and clinical studies providing guidance on salvage treatments remain limited [14]. This underscores the urgent need to explore novel and more effective treatment approaches for these patients, irrespective of their prior treatment history. While immune checkpoint inhibitors (ICIs) have revolutionized cancer treatment, their efficacy as monotherapy in advanced DTC is limited due to the tumor’s low mutational burden and low programmed death-ligand 1 (PD-L1) expression [15]. The modest response rates observed with PD-1 inhibitors, such as pembrolizumab, in advanced DTC patients further confirm these limitations [16, 17]. Antiangiogenic agents have been shown to enhance ICI efficacy, based on the “cancer-immunity cycle” [18], where abnormal tumor vasculature leads to hypoxia and acidosis, fostering an immunosuppressive microenvironment. Anti-angiogenic therapy can induce vascular normalization, thereby reversing this immunosuppression, promoting T-cell infiltration, and improving drug delivery [19]. This mechanistic synergy forms the biological rationale for combining ICIs with VEGF-targeted therapy, which has been clinically validated in multiple solid tumors, including unresectable hepatocellular carcinoma and advanced renal cell carcinoma [20, 21]. However, clinical evidence for this therapeutic strategy in DTC is only beginning to emerge. Given the shared immunologic mechanisms between DTC and other solid tumors, applying this combination strategy to RAIR-DTC may offer a promising means to overcome immune resistance and improve long-term outcomes. Encouraging preliminary data from the combination of lenvatinib and pembrolizumab in poorly differentiated thyroid cancer (PDTC) [22] and RAIR-DTC [23] suggest that integrating VEGFR TKI with ICI may improve treatment outcomes, both as a first-line and salvage therapy. Apatinib, a highly selective VEGF receptor 2 (VEGFR-2) inhibitor, has demonstrated significant clinical benefits in RAIR-DTC, with phase II and III trials reporting a median progression-free survival (PFS) of 22.2 months and an overall survival (OS) of 51.6 months [24, 25]. Camrelizumab, a humanized IgG4 anti-PD-1 monoclonal antibody, has shown distinctive activity in solid tumors and was approved in China for the treatment of classical Hodgkin lymphoma, nasopharyngeal carcinoma, and esophageal squamous cell carcinoma [26–28]. Notably, the combination of apatinib and camrelizumab has shown promising efficacy in advanced hepatocellular carcinoma, establishing a precedent for this synergistic therapeutic approach [29]. A preliminary case report has suggested the feasibility of this combination in progressive RAIR-DTC [30], yet its clinical application in DTC remains largely unexplored. Therefore, this study aims to evaluate the efficacy and safety of camrelizumab combined with apatinib in RAIR-DTC patients, both in the first-line and salvage settings. Methods Study design and patients This was an open-label, single-arm, exploratory phase II study performed at Peking Union Medical College Hospital. This study followed the Consolidated Standards of Reporting Trials (CONSORT) 2025 checklist for reporting (Additional file 1). Patient were recruited between September 2020 and August 2021 in accordance with the Declaration of Helsinki and in compliance with good clinical practice (ClinicalTrials.gov identifier NCT04560127). The trial was approved by the Ethics Committee of Peking Union Medical College Hospital, and all participants provided written informed consents. Key eligibility criteria included patients aged 18 years or older with pathologically or cytologically confirmed locally advanced or metastatic RAIR-DTC, encompassing papillary, follicular, and poorly differentiated subtypes; had at least one measurable lesion according to RECIST v1.1; an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2; adequate organ function and a life expectancy of 12 weeks or more; and serum thyroid-stimulating hormone (TSH) < 0.5 mIU/L. Eligible patients were also required to have radiographic evidence of disease progression within 12 months prior to enrollment according to RECIST v1.1. Prior systemic anti-cancer therapies, including but not limited to TKIs, chemotherapy, or radiotherapy, were permitted if discontinued for at least 28 days before study initiation; Patients with prior apatinib monotherapy who progressed radiographically (per RECIST v1.1) can continue apatinib and enroll. RAIR-DTC was defined by at least one of the following: complete loss of iodine uptake in the target lesion; disease progression within 12 months after receiving a single radioactive iodine treatment of ≥ 3.7 gigabecquerels (GBq) (≥ 100 mCi); disease progression 12 months after at least one dose of radioactive iodine, despite repeated treatments with intervals < 12 months and doses ≥ 3.7 GBq; or a cumulative radioactive iodine dose ≥ 22.2 GBq (≥ 600 mCi). Exclusion criteria encompassed uncontrolled cardiac or hypertensive conditions, any factors affecting oral drug absorption, increased bleeding risks, thrombotic or embolic events within 6 months, active autoimmune diseases, active infections, and pregnancy or breastfeeding. Study treatment Camrelizumab 200 mg was administered intravenously once every 2 weeks and apatinib 250 mg was administered orally once daily (equal to half-dose of apatinib monotherapy for RAIR-DTC in REALITY trial). Treatment was continued until unacceptable toxicities or disease progression. Four weeks were considered a treatment cycle. Dose interruptions, discontinuation for all drugs and dose reductions for apatinib were allowed for the management of adverse events (AEs). Drug administration of camrelizumab or apatinib alone was at the discretion of the investigator per the study protocol. Endpoints The primary endpoint was PFS, defined as the time from the date of treatment initiation to the date of disease progression assessed by the investigator according to RECIST v1.1 or death due to any cause, whichever occurs earlier. The secondary endpoints included confirmed objective response rate (ORR), defined as the proportion of patients with confirmed complete response (CR) or partial response (PR) according to RECIST v1.1; disease control rate (DCR), defined as the proportion of patients who achieved CR, PR, or stable disease (SD) based on RECIST v1.1; OS, defined as the time from the date of treatment initiation to the date of death due to any cause; duration of response (DoR), defined as the time from the first documented response (CR or PR) to the date of the first disease progression or death due to any cause, whichever came first; time to response (TTR), the time from the date of treatment initiation to first documented tumor response; time to progression (TTP), defined as the time from the date of treatment initiation to the date of first disease progression. Tumor responses were evaluated by the investigators without an independent radiologic review. For safety evaluation, all AEs were assessed and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 4.03. Treatment was continued for grade 1 and tolerable grade 2 AEs. For intolerable grade 2 AEs or ≥ grade 3 AEs, treatment interruption was permitted. The dose modification of apatinib followed a predefined sequential adjustment schedule: the first reduction was to a “5 days on, 2 days off” (5–2) schedule, followed by a reduction to “1 day on, 1 day off” (1–1) schedule. Dose re-escalation to a previous level was not permitted for apatinib. Dose modification of camrelizumab was not allowed. Statistical analysis This is an exploratory study and twenty patients are planned to be enrolled and treated in our trial. Efficacy and safety analysis included patients who received at least one dose of study drug. Continuous variables were summarized using the median (range), and categorical variables were presented as counts and percentages (n [%]). ORR and DCR were presented along with their corresponding 95% Clopper-Pearson confidence interval (CI). PFS, DoR, TTR, TTP, and OS were analyzed using Kaplan–Meier method, and the 95% CIs of median were calculated based on Brookmeyer-Crowley method. All statistical analyses were performed using SAS version 9.4. Results Baseline characteristics of patients Between September 2020 and August 2021, 20 patients were enrolled and received treatment. All patients were included in the full analysis population (FAS) and safety analysis population (SS). As of the data cutoff date of September 08, 2024, the median follow-up time was 28.4 months (IQR, 22.3–31.2), and two patients remained on the treatment (Fig. 1). The most common reason for treatment discontinuation was disease progression (n = 15). Demographics of the treated patients were summarized in Table 1. Median age was 57 years (range, 35–77). The most common type of thyroid cancer was papillary (70.0%), followed by follicular (25.0%) and mixed follicular/PDTC (5%). Regarding previous therapy, 10 patients (50.0%) had received systemic anticancer therapy, including 7 who had received a single regimen and 3 who had been treated with at least two systemic treatment regimens.Fig. 1Patient flowchartTable 1Baseline characteristicsCharacteristicsPatients (n = 20)Age (years), median (range)57.0 (35–77)Gender, n (%) Male11 (55.0) Female9 (45.0)ECOG performance status, n (%) 07 (35.0) 112 (60.0) 21 (5.0)Time since RAI (month), median (range)36.9 (1.1–150.4)Histologic subtype, n (%) Papillary14 (70.0) Follicular5 (25.0) Mixed follicular/poorly differentiated1 (5.0)Site of metastases, n (%) Lung19 (95.0) Brain2 (10.0) Pancreas1 (5.0) Lymph nodes8 (40.0) Pleura2 (10.0) Bone2 (10.0) Othera3 (15.0)No. of prior treatment line, n (%) 010 (50.0) ≥ 110 (50.0)Prior systemic anticancer therapy, n (%) Sorafenib4 (20.0) Lenvatinib2 (10.0) Apatinib3 (15.0) Donafenib3 (15.0) Anlotinib1 (5.0) Cabozantinib1 (5.0) RX2082 (10.0)ECOG Eastern Cooperative Oncology Group, RAI radioactive iodineaOther includes subcutaneous tissue (n = 1), muscle (n = 1), and liver (n = 1) Patient flowchart Table 1Baseline characteristicsCharacteristicsPatients (n = 20)Age (years), median (range)57.0 (35–77)Gender, n (%) Male11 (55.0) Female9 (45.0)ECOG performance status, n (%) 07 (35.0) 112 (60.0) 21 (5.0)Time since RAI (month), median (range)36.9 (1.1–150.4)Histologic subtype, n (%) Papillary14 (70.0) Follicular5 (25.0) Mixed follicular/poorly differentiated1 (5.0)Site of metastases, n (%) Lung19 (95.0) Brain2 (10.0) Pancreas1 (5.0) Lymph nodes8 (40.0) Pleura2 (10.0) Bone2 (10.0) Othera3 (15.0)No. of prior treatment line, n (%) 010 (50.0) ≥ 110 (50.0)Prior systemic anticancer therapy, n (%) Sorafenib4 (20.0) Lenvatinib2 (10.0) Apatinib3 (15.0) Donafenib3 (15.0) Anlotinib1 (5.0) Cabozantinib1 (5.0) RX2082 (10.0)ECOG Eastern Cooperative Oncology Group, RAI radioactive iodineaOther includes subcutaneous tissue (n = 1), muscle (n = 1), and liver (n = 1) Baseline characteristics ECOG Eastern Cooperative Oncology Group, RAI radioactive iodine aOther includes subcutaneous tissue (n = 1), muscle (n = 1), and liver (n = 1) Efficacy At data cutoff, 15 PFS events had occurred. The median PFS by investigator assessment was 11.0 months (95% CI, 6.0–14.2), with 11.6 months (95% CI, 4.8–NE) for treatment-naïve patients and 11.0 months (95% CI, 3.7–19.4) for previously systemic anticancer therapy patients (Fig. 2A–B). The median TTP was 11.0 (95% CI, 6.0–14.2). Median OS was not reached (95% CI, NE–NE) in either group; specifically, treatment-naïve patients had a median OS of NR (95% CI, NE–NE), and pretreated patients had NR (95% CI, 21.1–NE). The OS rates at 12, 24, and 36 months were 100.00% (95% CI, 100.00, 100.00); 93.75% (95% CI, 63.23, 99.10); and 85.94% (95% CI, 54.04, 96.34), respectively.Fig. 2Tumor response according to RECIST V1.1. A–B Kaplan–Meier curves for progression-free survival in the full analysis set (A) and stratified by treatment line (treatment-naïve vs. prior systemic anticancer therapy; post hoc) (B), nominal p value. C Waterfall plot showing confirmed objective responses based on the maximum percentage change in target lesions from baseline. D Duration of treatment with different treatment lines in each patient Tumor response according to RECIST V1.1. A–B Kaplan–Meier curves for progression-free survival in the full analysis set (A) and stratified by treatment line (treatment-naïve vs. prior systemic anticancer therapy; post hoc) (B), nominal p value. C Waterfall plot showing confirmed objective responses based on the maximum percentage change in target lesions from baseline. D Duration of treatment with different treatment lines in each patient The distribution of tumor responses was illustrated in Fig. 2C, presenting the maximum percentage change in target lesions from baseline. Six patients (30.0%) achieved confirmed PR, and no patients achieved CR, resulting in an ORR of 30.0% (95% CI, 11.9–54.3), with 40.0% (95% CI, 12.2–73.8) in treatment-naïve and 20.0% (95% CI, 2.5–55.6) in pre-treated patients. Meanwhile, SD was observed in 13 patients (65.0%). And the DCR was 95.0% (95% CI, 75.1–99.9) (Table 2). Furthermore, the median time to confirmed objective response (TTR) was 3.7 (95% CI, 1.0–NE) months and duration of confirmed response (DoR) was 5.7 (95% CI, 3.8–NE) months. Duration of treatment with different treatment lines in each patient was demonstrated in Fig. 2D. Table 2Confirmed objective responseResponsePatients (n = 20)Best response, n (%) Complete response0 Partial response6 (30.0) Stable disease13 (65.0) Progressive disease0 Not evaluable a1 (5.0) ORR, % (95% CI)30.0 (11.9–54.3) DCR, % (95% CI)95.0 (75.1–99.9)ORR objective response rate, DCR disease control rate, OS overall survival, PFS progression-free survivalaBest overall response could not be evaluated with available assessments owing to early withdrawal after a single efficacy assessment Table 2Confirmed objective responseResponsePatients (n = 20)Best response, n (%) Complete response0 Partial response6 (30.0) Stable disease13 (65.0) Progressive disease0 Not evaluable a1 (5.0) ORR, % (95% CI)30.0 (11.9–54.3) DCR, % (95% CI)95.0 (75.1–99.9)ORR objective response rate, DCR disease control rate, OS overall survival, PFS progression-free survivalaBest overall response could not be evaluated with available assessments owing to early withdrawal after a single efficacy assessment Confirmed objective response ORR objective response rate, DCR disease control rate, OS overall survival, PFS progression-free survival aBest overall response could not be evaluated with available assessments owing to early withdrawal after a single efficacy assessment Median percentage change in thyroglobulin (Tg) from baseline demonstrated a sustained downward trend over 14 weeks in both treatment-naïve and pre-treated patients (Additional file 2: Fig. S1). Safety and treatment All patients experienced treatment-related adverse events (TRAEs) of any grade, with the most common ones being proteinuria (n = 14, 70.0%); hypertension (n = 13, 65.0%). Grade 3 or higher-level TRAEs occurred in 13 patients (65.0%), with gamma-glutamyltransferase increased (n = 8, 40.0%), alanine aminotransferase increased (n = 6, 30.0%) being the most common events, which was likely due to apatinib (Table 3). Immune-related AEs (irAEs) and AEs related to apatinib were listed separately in the Additional file 3: Table S1 and Table S2. Among irAEs, reactive cutaneous capillary endothelial proliferation (RCCEP) was the most common, occurring in 10 patients (50%) at any grade. Serious AEs (SAEs) occurred in 3 patients (15.0%), of which one case was camrelizumab related. Table 3Summary of treatment-related adverse events (TRAEs)TRAE occurring in ≥ 10% of patientsPatients (n = 20)Any grade, n (%)Grade ≥ 3, n (%)Proteinuria14 (70.0)3 (15.0)Hypertension13 (65.0)1 (5.0)ACTH increased10 (50.0)0GGT increased10 (50.0)8 (40.0)RCCP10 (50.0)0ALT increased9 (45.0)6 (30.0)Total blood cortisol increased9 (45.0)0AST increased8 (40.0)5 (25.0)Hypocalcemia7 (35.0)2 (10.0)Hyperuricemia7 (35.0)0Alkaline phosphatase increased7 (35.0)2 (10.0)Hand-foot syndrome7 (35.0)0Neutrophil count decreased7 (35.0)0TSH increased6 (30.0)0WBC count decreased6 (30.0)0Diarrhea6 (30.0)0Weight decreased5 (25.0)0Blood bilirubin increased4 (20.0)2 (10.0)Anorexia4 (20.0)0Fatigue3 (15.0)0Platelet count decreased3 (15.0)0Toothache3 (15.0)0Periodontal disease3 (15.0)0Fever2 (10.0)0Edema2 (10.0)0Arthralgia2 (10.0)0Myalgia2 (10.0)0Urinary hyaline cast2 (10.0)0Electrocardiogram QT prolonged2 (10.0)0Hematuria2 (10.0)0Total blood cortisol decreased2 (10.0)0Sore throat2 (10.0)0ACTH adrenocorticotropic hormone, GGT gamma-glutamyltransferase, ALT alanine aminotransferase, AST aspartate aminotransferase, TSH thyroid-stimulating hormone, WBC white blood cell, RCCP reactive cutaneous capillary proliferation Table 3Summary of treatment-related adverse events (TRAEs)TRAE occurring in ≥ 10% of patientsPatients (n = 20)Any grade, n (%)Grade ≥ 3, n (%)Proteinuria14 (70.0)3 (15.0)Hypertension13 (65.0)1 (5.0)ACTH increased10 (50.0)0GGT increased10 (50.0)8 (40.0)RCCP10 (50.0)0ALT increased9 (45.0)6 (30.0)Total blood cortisol increased9 (45.0)0AST increased8 (40.0)5 (25.0)Hypocalcemia7 (35.0)2 (10.0)Hyperuricemia7 (35.0)0Alkaline phosphatase increased7 (35.0)2 (10.0)Hand-foot syndrome7 (35.0)0Neutrophil count decreased7 (35.0)0TSH increased6 (30.0)0WBC count decreased6 (30.0)0Diarrhea6 (30.0)0Weight decreased5 (25.0)0Blood bilirubin increased4 (20.0)2 (10.0)Anorexia4 (20.0)0Fatigue3 (15.0)0Platelet count decreased3 (15.0)0Toothache3 (15.0)0Periodontal disease3 (15.0)0Fever2 (10.0)0Edema2 (10.0)0Arthralgia2 (10.0)0Myalgia2 (10.0)0Urinary hyaline cast2 (10.0)0Electrocardiogram QT prolonged2 (10.0)0Hematuria2 (10.0)0Total blood cortisol decreased2 (10.0)0Sore throat2 (10.0)0ACTH adrenocorticotropic hormone, GGT gamma-glutamyltransferase, ALT alanine aminotransferase, AST aspartate aminotransferase, TSH thyroid-stimulating hormone, WBC white blood cell, RCCP reactive cutaneous capillary proliferation Summary of treatment-related adverse events (TRAEs) ACTH adrenocorticotropic hormone, GGT gamma-glutamyltransferase, ALT alanine aminotransferase, AST aspartate aminotransferase, TSH thyroid-stimulating hormone, WBC white blood cell, RCCP reactive cutaneous capillary proliferation Dose modification was the main strategy for AE management. Nine patients (45.0%) had a dose reduction of apatinib, including proteinuria (n = 3) and hepatic dysfunction (n = 6). All these patients reduced to a 5–2 schedule: 3 with proteinuria, 6 with hepatic dysfunction. Further reduction to a 1–1 schedule occurred in 3 patients: 2 with proteinuria, 1 with hepatic dysfunction. Four patients (20.0%) discontinued using apatinib, 3 patients (15%) discontinued camrelizumab, and 3 patients (15.0%) discontinued both study drugs. Detailed data on TRAE-related dose reductions and discontinuations were provided in Additional file 3: Table S3. Discussion This study explored the potential benefits of combining apatinib with camrelizumab in the treatment of RAIR-DTC, including both treatment-naïve patients and those who had failed first-line therapy. Our preliminary results demonstrate that this combination therapy offers reliable oncologic efficacy to some extent, with a manageable AE profile. Notably, the PFS benefit associated with this combination was sustained both as an initial and salvage treatment. To date, there are no publicly approved treatments for RAIR-DTC patients with first-line TKI resistance in China, highlighting the need for further clinical investigations to refine treatment strategies for progressive RAIR-DTC in the era of targeted cancer therapies. To the best of our knowledge, this is the first clinical study to investigate apatinib combined with camrelizumab as a salvage therapy for RAIR-DTC, revealing its potential role in treating these challenging cases. PD-L1 expression is absent in normal thyroid tissue but is upregulated in DTC, providing a rationale for ICIs targeting the PD-1/PD-L1 axis [31]. While pembrolizumab monotherapy has shown modest antitumor activity, with reported ORRs of 9% in KEYNOTE-028 [16] and 6.8% in KEYNOTE-158 [17], combination strategies have been explored to enhance ICI efficacy. Given that VEGFR-targeted therapies remained the backbone of systemic treatment for RAIR-DTC, their combination with ICIs represents a logical therapeutic strategy to potentially improve patient outcomes. Lenvatinib plus pembrolizumab was the first exploration of a VGFR-TKI combined with an ICI for RAIR-DTC patients, and encouraging results were observed in both TKI-naïve (ORR 65.5%, PFS 26.8 months) and TKI-treated patients (ORR 16%, PFS 10.0 months) [23]. In our study of apatinib plus camrelizumab, promising outcomes were also observed in TKI-treated patients (ORR 20.0%, PFS 11.0 months), despite the patients enrolled in our study tended to be more advanced, with higher proportion of ECOG-PS 1, brain metastases and more prior lines of TKI therapy (baseline characteristics of both trials were provided in Additional file 3: Table S4). Notably, given that apatinib was administered at a reduced dose of 250 mg (half of the standard 500 mg monotherapy dose), suggesting that the efficacy may be attributed more to camrelizumab than to apatinib. In contrast, in the lenvatinib plus pembrolizumab regimen, lenvatinib was administered at a nearly full dose of 20 mg, and the efficacy was similar to that of lenvatinib monotherapy in the SELECT study (24 mg once daily, ORR 64.8%, PFS 18.3 months), implying that the efficacy might be primarily driven by lenvatinib rather than the ICI. Furthermore, the efficacy of apatinib plus camrelizumab appeared comparable to that reported in a recent second-line combination trial of CaboNivoIpi (cabozantinib plus nivolumab/ipilimumab, ORR 10%, PFS 9.0 months) [32] Thus, considering the relatively higher ORR and promising PFS observed in our study, apatinib plus camrelizumab may provide salvage benefit for patients with prior TKI exposure. Although the rationale for combing VEGFR-TKIs with ICIs centers on remodeling the tumor microenvironment and boosting immune activity, clinical efficacy and safety vary considerably across regimens. In our study, the overall efficacy of apatinib plus camrelizumab (ORR 30.0%, PFS 11.0 months) was broadly comparable to that observed with VEGFR-TKI monotherapies including sorafenib, lenvatinib, apatinib, and donafenib, with PFS ranging from 10.8 to 22.2 months and ORR from 12.2% to 54.3% [8, 9, 24, 33]. In salvage settings, it appeared not superior to monotherapy in SELECT study (PFS 15.1 months) [9], but similar to both DIRECTION [33] and COSMIC-311 studies (PFS 11.0 months). Future studies should focus on optimizing patient selection, refining VEGFR-ICI combination strategies, and identifying predictive biomarkers to improve the personalized treatment landscape for RAIR-DTC. Importantly, cross-trial comparisons should be interpreted with caution, particularly within the exploratory context of this open-label phase II design. In addition to the use of a half-dose of apatinib, other factors may have confounded the results, including differences in study populations, for example, only 25.3% of patients in SELECT study had received prior systemic therapy, a proportion substantially lower than that in our study, as well as the inherent variability associated with the relatively small sample size. In this study, the TRAE spectrum was generally consistent with the known camrelizumab and apatinib safety profile in other solid tumors [34–36], and no new signals were detected. The most common AEs, including hypertension and proteinuria, were consistent with those among patients receiving other VEGF and VEGFR inhibitors [37, 38]. In our current analysis, 13 patients (65.0%) experienced grade ≥ 3 TRAEs, which were comparable to the results of REALITY and SELECT trials (73.9% and 75.9%) [9, 24]. Specifically, 15% of patients in our study experienced SAEs, and no patient died from TRAEs, compared with 50% of patients in SELECT. In a trial involving Chinese patients receiving lenvatinib, the rate of grade 3 and 4 AEs was 85.4%, with 10 fatal TRAEs [39]. This finding suggests the manageable safety profile of camrelizumab plus apatinib for RAIR-DTC treatment. The limitations of this study should be acknowledged. First, the study design was limited to a single-arm with a small sample size and was conducted at a single center, which may affect the generalizability of the results. Second, the cohort included patients with different histological subtypes of DTC, which could contribute to variability in treatment response and prognosis. Third, molecular and immunologic biomarkers were not systematically assessed. The absence of genomic profiling (e.g., RET and NTRK fusions, BRAF or RAS mutations) and PD-L1 expression data limited our ability to explore potential predictors of treatment response. Incorporating these molecular and immunologic characteristics as stratification factors in future studies may help identify patient subgroups most likely to benefit from camrelizumab plus apatinib therapy. Finally, the follow-up period was insufficient to provide definitive OS data, and extended follow-up is ongoing to determine long-term outcomes. Conclusions In conclusion, camrelizumab plus apatinib shows preliminary antitumor activity and acceptable safety profile in patients with RAIR-DTC, including those with TKI-treated disease. Further investigation with larger sample size and longer follow up is warranted to assess whether the addition of camrelizumab to apatinib offers further therapeutic benefit beyond VEGFR TKI monotherapy, either in the first-line or salvage settings. Supplementary Information Additional file 1. CONSORT 2025 checklist. Additional file 1. CONSORT 2025 checklist. Additional file 2. Thyroglobulin (Tg) levels in both bothtreatment-naïve and pre-treated patients. The x-axis represents time from treatment initiation (weeks), and the y-axis represents the median value of percentage change from baseline (%). Additional file 2. Thyroglobulin (Tg) levels in both bothtreatment-naïve and pre-treated patients. The x-axis represents time from treatment initiation (weeks), and the y-axis represents the median value of percentage change from baseline (%). Additional file 3. Summary of specific adverse events (AEs) and related management. Table S1. Summary of the incidence of any-grade and ≥ Grade 3 immune-related AEs associated with camrelizumab. Table S2. Incidence of any-grade and ≥ Grade 3 AEs attributed to apatinib. Table S3. Description of AE-related dose modifications. Table S4. Demographic and clinical characteristics of the “camrelizumab + apatinib” cohort and the “pembrolizumab + lenvatinib” cohort. Additional file 3. Summary of specific adverse events (AEs) and related management. Table S1. Summary of the incidence of any-grade and ≥ Grade 3 immune-related AEs associated with camrelizumab. Table S2. Incidence of any-grade and ≥ Grade 3 AEs attributed to apatinib. Table S3. Description of AE-related dose modifications. Table S4. Demographic and clinical characteristics of the “camrelizumab + apatinib” cohort and the “pembrolizumab + lenvatinib” cohort.