Camrelizumab Plus Famitinib in Recurrent or Metastatic Head and Neck Squamous Cell Carcinoma: A Phase 2 Study Authors: Bei Xu, Guopei Zhu, Jin Lu, Ximei Zhang, Xiaozhong Chen, Qinjiang Liu, Song Qu, Jia Fan, Tianshu Liu Published: 2026 ABSTRACT ABSTRACTBackgroundTo evaluate the outcomes of patients with refractory recurrent or metastatic head and neck squamous cell carcinoma (R/M HNSCC) treated with camrelizumab combined with famitinib following platinum‐based chemotherapy.MethodsIn this multicenter, phase 2, single‐arm trial, patients with R/M HNSCC who had documented progression after platinum‐based chemotherapy were administered camrelizumab 200 mg every 3 weeks and famitinib at 20 mg once daily (cohort 1). The primary endpoint was the objective response rate (ORR). The key secondary endpoints included progression‐free survival (PFS), duration of response (DOR), time to objective response (TTR), disease control rate (DCR), overall survival (OS), and safety.FindingsBetween August 07, 2020 and June 22, 2022, a total of 20 patients with R/M HNSCC were enrolled. The median age of the patients was 61 years (range, 37–69). The ORR was 35.0% (95% CI, 15.4–59.2) and the DCR was 55.0% (95% CI, 31.5–76.9). The median TTR was 2.23 months (95% CI, 2.0–8.3), the median DOR was 8.0 months (95% CI, 2.2 to not reach), the median PFS was 4.1 months (95% CI, 1.7–10.3), and the median OS was 8.4 months (95% CI, 3.0–17.0), with a median follow‐up duration of 7.2 months (range 0.1 months to 19.6 months). Treatment‐related adverse events (TRAEs) of grade ≥ 3 included decreased white blood cell count (15.0%), hypertension (15.0%), pneumonia (15.0%), and tumor bleeding (15.0%). The most common TRAEs of any grade were decreased white blood cell count (50.0%), anemia (50.0%), hypertension (50.0%), and decreased platelet count (45.0%).InterpretationIn this single‐arm phase II trial, camrelizumab plus famitinib demonstrated encouraging efficacy and a manageable safety profile in patients with refractory R/M HNSCC. These preliminary findings warrant further investigation.Trial Registration ClinicalTrials.gov identifier: NCT04346381Keywords: Camrelizumab, Famitinib, head and neck squamous cell carcinoma, PD‐1 inhibitor, VEGFR Introduction Most patients with head and neck squamous cell carcinoma (HNSCC) present with locally advanced disease at diagnosis and are treated with surgery, definitive (chemo)radiation, or platinum‐based chemotherapy [1]. Despite ongoing advancements in surgical techniques and chemotherapeutic strategies, the prognosis for most patients with advanced disease remains poor [2]. Currently, no standard treatment regimen has been established for patients with platinum‐refractory R/M HNSCC who progress after first‐line therapy, highlighting a significant unmet medical need for improving clinical outcomes in this population. The immune system plays a pivotal role in the development and progression of R/M HNSCC [3]. For patients with high tumor burden or negative expression of immune markers, immunotherapy combined with chemotherapy remains a critical therapeutic option [4]. Currently, first‐line treatment for R/M HNSCC consists of pembrolizumab‐based immunotherapy (monotherapy or with chemotherapy) according to the KEYNOTE‐048 trial [5]. In patients with platinum‐refractory R/M HNSCC who have not previously treated with first‐line immunotherapy, immune checkpoint inhibitor monotherapy demonstrates the most favorable overall survival benefit and the lowest incidence of serious adverse events in the second‐line setting [6]. However, resistance to immunotherapy has emerged as an increasingly prominent clinical challenge [7]. In recent years, research exploring the combination of immunotherapy with targeted agents to enhance anti‐tumor activity has gained significant attention [8]. Among small‐molecule targeted drugs, EGFR TKIs (gefitinib and erlotinib), mTOR inhibitors (e.g., everolimus, temsirolimus), pan‐human epidermal growth factor receptor (pan‐HER) inhibitors (dacomitinib), and EGFR/HER2 inhibitors (lapatinib and afatinib) have failed to demonstrate clear efficacy benefits in clinical studies, whether used as monotherapy or in combination with chemotherapy [9]. Camrelizumab is a high‐affinity, humanized IgG4‐κ anti‐PD‐1 monoclonal antibody. Famitinib is a small molecule, multi‐targeted tyrosine kinase inhibitor with significant inhibitory activity against vascular endothelial growth factor receptor 2/3 (VEGFR 2/3), stem‐cell factor receptor (c‐KIT), platelet‐derived growth factor receptor β (PDGFRβ), proto‐oncogene tyrosine‐protein kinase receptor and FMS‐like tyrosine kinase‐1/3 receptor. Given that multi‐targeted tyrosine kinase inhibitors can remodel the tumor immune microenvironment and promote immune cell infiltration, thereby enhancing the anti‐tumor efficacy of immune checkpoint inhibitors [10], the combination of camrelizumab and famitinib has demonstrated favorable efficacy and a manageable safety profile in solid tumors such as recurrent/metastatic cervical cancer [11], providing a rationale for its investigation in head and neck squamous cell carcinoma. We launched a phase 2 basket trial to assess the efficacy and safety of camrelizumab in combination with famitinib in patients with various advanced solid tumors. Here, we report the results of camrelizumab plus famitinib in patients with previously treated R/M HNSCC (Cohort1). Methods Study Design and Patients This is a multicenter, open‐label phase 2 basket study conducted in China to evaluate the efficacy and safety of camrelizumab plus famitinib for the treatment of advanced solid tumors (ClinicalTrials. gov registration: NCT04346381). Here, we present the data from cohort 1 which included patients with R/M HNSCC. Eligible patients aged 18–75 years with an Eastern Cooperative Oncology Group performance status (ECOG PS) of 0 or 1 were enrolled. All patients had histologically or cytologically confirmed incurable R/M HNSCC and had received no more than one line of systemic therapy, specifically for R/M disease. Patients who had progressed after prior platinum‐based treatment, whether administered as part of multimodal therapy for locally advanced disease (e.g., concurrent chemoradiotherapy) or as first‐line palliative treatment for R/M disease, were eligible. Other key eligibility criteria included measurable disease according to the Response Evaluation Criteria in Solid Tumors (version 1.1), an estimated life expectancy of at least12 weeks, and adequate organ and hematological function. PD‐L1 expression testing was not mandated for study entry. Main exclusion criteria included active autoimmune disease, application of immunosuppressants or systemic hormones within 2 weeks prior to study treatment, poorly controlled hypertension or active infection, untreated central nervous system metastases, uncontrolled hypertension, bleeding risk, abnormal coagulation, radiological evidence of tumor invasion into major blood vessels, prior treatment with small‐molecule vascular endothelial growth factor receptor (VEGFR) inhibitors, active infection or non‐infectious pneumonitis, and a history of another malignancy within the past 5 years. Prior treatment with any immune checkpoint inhibitor, including anti‐PD‐1, anti‐PD‐L1, or anti‐CTLA‐4 antibodies, was not permitted. The protocol and all amendments were approved by the ethics committee of each participating site (Table S1). The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. All patients provided written informed consent. The overall study design has been published, and the results for camrelizumab plus famitinib in the cohorts of non‐small cell lung cancer (cohorts 2 and 7) and nasopharyngeal carcinoma (cohort 9) have been reported separately [12, 13, 14]. All enrolled patients received camrelizumab 200 mg via intravenous infusion (30–60 min) on day 1 of each 21‐day cycle, and oral famitinib at an initial dose of 20 mg once daily. Treatment was continued until disease progression, unacceptable adverse events, patient withdrawal from the study, or loss to follow‐up, whichever occurred first. The maximum total duration of camrelizumab exposure was 2 years. Imaging evaluations were performed at baseline and subsequently every 9 weeks. Disease progression was assessed using the Response Evaluation Criteria in Solid Tumors (RECIST version 1.1). Toxicity assessments were conducted every 3 weeks prior to each treatment cycle, following the National Cancer Institute's Common Terminology Criteria for Adverse Events (NCI‐CTCAE version 5.0; National Institutes of Health). Procedures The study treatment was continued until intolerable toxicity, confirmed disease progression per RECIST version 1.1, investigator decision, withdrawal of consent, cumulative camrelizumab exposure of 24 months, poor compliance, or loss to follow‐up, whichever occurred first. Treatment beyond confirmed disease progression was permitted only for patients with clinically stable conditions, at the investigator's discretion. To manage adverse events, treatment interruption for camrelizumab (which should be permanently discontinued if immune‐mediated adverse events [irAEs] are not resolved within 15 weeks of the last dose) or famitinib was allowed; however, dose reduction of camrelizumab was not permitted. Outcomes The primary endpoint was the objective response rate (ORR), defined as the percentage of patients achieving a best overall response of confirmed complete response (CR) or partial response (PR) according to RECIST version 1.1. The secondary endpoints were progression‐free survival (PFS), defined as the time from the first dose of study treatment to the first documented disease progression or death, whichever occurred first. The duration of response (DOR) was defined as the time from the first documentation of a response to the first documentation of disease progression or death from any cause. Time to objective response (TTR), defined as the time from the initiation of study treatment to the first documentation of a response, was calculated for patients with a best‐confirmed CR or PR. Disease Control Rate (DCR) was defined as the proportion of patients achieving a best response of CR, PR, or stable disease (SD). Overall survival (OS) was defined as the time from the first day of treatment to death due to any cause. Additionally, 12‐month OS rates were assessed. Safety and tolerability were evaluated as secondary endpoints in accordance with the NCI‐CTCAE V5.0 criteria. Exploratory endpoints included tumor responses assessed per RECIST V.1.1 by the investigator, stratified by PD‐L1 tumor proportion score (TPS) and combined positive score (CPS) expression status. Statistical Analysis A Simon (minimax) two‐stage design was employed to determine the sample size. In the cohort 1, the desirable target ORR was set at 35%, while the uninteresting level of ORR under the null hypothesis was set at 15%. Based on these assumptions, with a one‐sided α‐error of 2.5% and a power of 80%, the study required 21 evaluable patients to be treated in stage 1. Stage 2 would proceed if at least 4 responders were obtained during stage 1, involving the treatment of an additional 13 evaluable patients. Overall, the combination regimen would be considered active if 10 or more responders were observed among the 34 patients. In cohort 1, the actual sample size for stage 1 was 20, and 7 of the 20 patients enrolled in stage 1 achieved objective responses. No patients were enrolled in stage 2. As no enrollment occurred in the second stage, no further statistical inference was conducted. The full analysis set (FAS) and safety analysis included all patients who received at least one dose of study treatment. ORR and DCR, along with their 95% confidence intervals (CIs), were calculated using the Clopper–Pearson method. For time‐to‐event endpoints, medians were estimated using the Kaplan–Meier method, with corresponding 95% CIs calculated using the Brookmeyer‐Crowley method. The Kaplan–Meier method was also used to estimate the probabilities of the 12‐month OS rate, with corresponding 95% CIs calculated using the log–log transformation (based on the normal approximation method) and back transformed to a CI on the untransformed scale. TTR was summarized using median, minimum, and maximum values. All statistical analyses were performed using SAS (V.9.4). Results Demographics and Baseline Characteristics Between August 2020 and June 2022, 20 patients with R/M HNSCC received camrelizumab plus famitinib. The median age of the patients was 61 years (range, 37–69). Baseline characteristics are summarized in Table 1. TABLE 1Demographics and baseline characteristics.CharacteristicsPatients (N = 20)Age, years, median (range)61.0 (37, 69)Sex, n (%)Male17 (85.0)Female3 (15.0)ECOG performance status, n (%)04 (20.0)116 (80.0)Primary tumor site, n (%)Oral cavity10 (50.0)Laryngeal carcinoma3 (15.0)Hypopharyngeal carcinoma4 (20.0)Oropharyngeal carcinoma3 (15.0)Metastases, n (%)No4 (20.0)Yes16 (80.0)Metastatic site, n (%)Neck2 (10.0)Lung10 (50.0)Lymph node5 (25.0)Bone3 (15.0)Mediastinum1 (5.0)Soft tissue1 (5.0)Initial clinical TNM stage, n (%)I2 (10.0)II1 (5.0)III4 (20.0)IIIB2 (10.0)IV3 (15.0)IVA4 (20.0)UK4 (20.0)PD‐L1 status (TPS), n (%)< 1%17 (85.0)1%–49%3 (15.0)PD‐L1 status (CPS), n (%)< 115 (75.0)≥ 15 (25.0)≥ 101 (5.0)Abbreviations: CPS, combined positivity score; ECOG, Eastern Cooperative Oncology Group; TPS, tumor proportion score; UK, unknown. Demographics and baseline characteristics. Abbreviations: CPS, combined positivity score; ECOG, Eastern Cooperative Oncology Group; TPS, tumor proportion score; UK, unknown. Four (20.0%) patients had an ECOG PS of 0, while 16 (80.0%) patients had an ECOG PS of 1. Among the 20 patients, 16 (80%) had distant metastatic lesions. The most common metastatic sites were the lungs (10 patients, 50.0%), lymph nodes (5 patients, 25.0%), and bones (3 patients, 15.0%). All 20 enrolled patients had tumor samples available for PD‐L1 TPS and CPS status assessments at baseline. All 20 patients had received prior antitumor therapy. A total of 13 (65%) patients underwent surgical treatment for the primary lesion, 2 (10%) patients underwent palliative resection, 1 (5%) patient underwent radiofrequency ablation, and 1 (5%) patient underwent radiation particle implantation. A total of 16 (80%) patients had previously received radiation therapy, with 7 (35%) undergoing radical radiotherapy, 3 (15%) receiving palliative radiotherapy, and 6 (30%) patients had received adjuvant radiotherapy. The majority of patients (80.0%) had received at least 1 prior line of systemic chemotherapy: 13 (65.0%) patients underwent concurrent chemoradiotherapy, 5 (25%) received induction chemotherapy, 4 (20%) received adjuvant therapy, 2 (10%) received neoadjuvant therapy, and 2 (10%) received palliative chemotherapy. None of the patients had received prior immune checkpoint inhibitors. (Table 2). TABLE 2Prior antitumour therapy.Characteristics, n (%)Patients (N = 20)Prior antitumour therapy in R/M settings20 (100.0)1st line2 (10.0)Prior surgery for primary tumor13 (65.0)Palliative surgery2 (10.0)Local treatment history2 (10.0)Radiofrequency ablation (RFA)1 (5.0)Radioactive particle implantation1 (5.0)Prior radiotherapy16 (80.0)Radical radiotherapy7 (35.0)Palliative radiotherapy3 (15.0)Adjuvant radiotherapy6 (30.0)Previous systemic therapy16 (80.0)Neoadjuvant2 (10.0)Adjuvant4 (20.0)Palliative2 (10.0)Induction chemotherapy5 (25.0)Concurrent chemoradiotherapy13 (65.0)Type of systemic therapyChemotherapy16 (80.0)Targeted therapy5 (25.0) Prior antitumour therapy. At the time of the data cutoff, the median follow‐up duration was 7.2 months (range, 0.1–19.6 months). All patients discontinued treatment due to radiologic disease progression (n = 9), adverse events (n = 2), death (n = 4), physician decision (n = 2), or patient withdrawal (n = 3) (Figure 1). Trial profile. Antitumor Activity in All Patients Among the 20 patients in the FAS, no patient achieved a CR, 7 (35.0%) patients achieved a PR, and 4 (20.0%) patients had SD. The confirmed ORR was 35.0% (95% CI, 15.4–59.2), and the DCR was 55.0% (95% CI, 31.5–76.9) in patients treated with camrelizumab and famitinib (Table 3, Figure 3A). TABLE 3Tumor response to treatment, as assessed by the investigators.Patients (N = 20)Best overall response, n (%)Complete response0Partial response7 (35.0)Stable disease4 (20.0)Progressive disease3 (15.0)Not available6 (30.0)ORR, % (95% CI)35.0 (15.4, 59.2)DCR, % (95% CI)55.0 (31.5, 76.9)Time to response, months, median (range)2.23 (2.0, 8.3)Duration of response, months, median (95% CI)8.0 (2.2, NR)Progression‐free survival, months, median (95% CI)4.1 (1.7, 10.3)Overall survival, months, median (95% CI)8.4 (3.0, 17.0)12‐month rate, % (95% CI)42.2 (20.4, 62.6)Abbreviations: PD, progressive disease; PR, partial response; SD, stable disease. Tumor response to treatment, as assessed by the investigators. Abbreviations: PD, progressive disease; PR, partial response; SD, stable disease. Kaplan–Meier estimates of survival (N = 20) (A) Duration of response. (B) Progression‐free survival. (C) Overall survival. A reduction in target lesion size from baseline was observed in 18 (90.0%) patients, and the decreased tumor burden was sustained across multiple assessments (Figure 2). Tumor measurements and response (N = 20) (A) Change in target lesion size from baseline based on response. (B) Waterfall plot of maximum change in target lesion size from baseline in each patient. (C) Progression events and duration of response according to RECIST (version 1.1). The median time to response (TTR) was 2.23 months (95% CI, 2.0–8.3), and the median DOR was 8.0 months (95% CI, 2.2 to not reached). The median PFS was 4.1 months (95% CI, 1.7–10.3), while the median OS was 8.4 months (95% CI, 3.0–17.0). The estimated 12‐month OS rate was 42.2% (95% CI, 20.4–62.6) (Table 3, Figure 3B,C). Antitumor Activity in Subgroups by PD‐L1 Expression Subgroup analyses based on baseline characteristics, including sex, age, and ECOG PS, did not identify specific factors associated with clinical activity of the combination treatment. Higher tumor PD‐L1 expression levels were associated with improved outcomes across all key endpoints, including ORR, DCR, and PFS. Tumor PD‐L1 expression data were available for all patients, with 17 patients having PD‐L1 TPS < 1% and three having PD‐L1 TPS ≥ 1%, 15 patients had PD‐L1 CPS < 1, and 5 had PD‐L1 CPS ≥ 1. The ORR was 66.7% (95% CI, 9.4–99.2) in PD‐L1 TPS ≥ 1% compared to 29.4% (95% CI, 10.3–56.0) in patients with PD‐L1 TPS < 1%. The DCR was 100.0% (95% CI, 29.2–100.0) in PD‐L1 TPS ≥ 1% versus 47.1% (95% CI, 23.0–72.2) in those with PD‐L1 TPS < 1%. The PFS was not reached (95% CI, 10.3 to not reached) in patients with PD‐L1 TPS ≥ 1% compared to 3.8 months (95% CI, 1.4–6.2) in those with PD‐L1 TPS < 1%. In patients with PD‐L1 CPS ≥ 1, the ORR was 60.0% (95% CI, 14.7–94.7) compared to 26.7% (95% CI, 7.8–55.1) in those with PD‐L1 CPS < 1. The DCR was 80.0% (95% CI, 28.4–99.5) in patients with PD‐L1 CPS ≥ 1 versus 46.7% (95% CI, 21.3–73.4) in those with PD‐L1 CPS < 1. The PFS was 10.3 months (95% CI, 4.1 to not reached) in patients with PD‐L1 CPS ≥ 1 compared to 3.6 months (95% CI, 1.4–10.2) in those with PD‐L1 CPS < 1 (Figure 4). Subgroup analysis of Tumor response and survival (N = 20) (A) Objective response rate. (B) Duration of response. (C) Progression‐free survival. Safety All 20 patients were included in the safety analysis. The median number of camrelizumab treatment cycles was five (range, 1–24), with a median relative dose intensity (defined as the ratio of delivered dose intensity to the standard dose intensity) of 99.9% (range, 88.9%–107.4%). The median duration of famitinib exposure was 2.7 months (range, 0.1–13.1), with a median relative dose intensity of 94.3% (range, 47.9%–100.0%). All patients experienced at least one treatment‐related adverse event (TRAE) (Table 4). A total of 1 patient (5%) discontinued camrelizumab due to TRAEs, 2 (10%) discontinued famitinib, 6 (30%) interrupted camrelizumab, 13 (65%) interrupted famitinib, and 7 (35%) required a dose reduction of famitinib (Table 4). TABLE 4Treatment‐related adverse events.TRAEs, n (%)Patients (N = 20)Any TRAE20 (100.0)irAE9 (45.0)TRAEs leading to camrelizumab discontinuation1 (5.0)TRAEs leading to famitinib discontinuation2 (10.0)TRAEs leading to camrelizumab interruption6 (30.0)TRAEs leading to famitinib interruption13 (65.0)TRAEs leading to famitinib dose reduction7 (35.0)Treatment‐related SAEs3 (15.0)Abbreviations: SAEs, serious adverse events; TRAEs, treatment‐related adverse events. Treatment‐related adverse events. Abbreviations: SAEs, serious adverse events; TRAEs, treatment‐related adverse events. The most common TRAEs of any grade were decreased white blood cell count (9 [45.0%]), decreased platelet count (9 [45.0%]), decreased neutrophil count (8 [40.0%]), and hypertension (8 [40.0%]). The most frequently observed TRAEs were grade 1 or 2. The most common TRAEs of grade 3 or higher were decreased white blood cell count (3 [15.0%]), palmar–plantar erythrodysesthesia syndrome (3 [15.0%]), and hypertension (2 [10.0%]). Three grade 5 fatal events (15.0% of the patients) were reported, including one case of bacterial pneumonia, one case of pharyngeal bleeding, and one death from an unknown cause. (Table 5). TABLE 5Treatment‐related adverse events.TRAEs, n (%)Patients (N = 20)Grade 1–2Grade 3Grade 4Grade 5Any gradeAny TRAE5 (25.0)10 (50.0)1 (5.0)3 (15.0)19 (95.0)Hematological toxicityDecreased platelet count8 (40.0)1 (5.0)009 (45.0)Decreased white blood cell count6 (30.0)3 (15.0)009 (45.0)Decreased neutrophil count6 (30.0)2 (10.0)008 (40.0)Anemia6 (30.0)0006 (30.0)Lymphopenia1 (5.0)1 (5.0)002 (10.0)Non‐hematological toxicityHypertension6 (30.0)2 (10.0)008 (40.0)Hypothyroidism7 (35.0)0007 (35.0)Hypoalbuminemia5 (25.0)0005 (25.0)Proteinuria3 (15.0)1 (5.0)004 (20.0)Palmar–plantar erythrodysesthesia syndrome03 (15.0)003 (15.0)Skin lesions2 (10.0)1 (5.0)003 (15.0)Diarrhea2 (10.0)1 (5.0)003 (15.0)AST elevation2 (10.0)1 (5.0)003 (15.0)CK‐MB elevation3 (15.0)0003 (15.0)Hypercholesterolemia3 (15.0)0003 (15.0)Hyperuricemia3 (15.0)0003 (15.0)Hyperlipidemia3 (15.0)0003 (15.0)Hyperglycemia3 (15.0)0003 (15.0)Reactive cutaneous capillary endothelial3 (15.0)0003 (15.0)Occult blood positive3 (15.0)0003 (15.0)Bleeding in the pharynx1 (5.0)001 (5.0)2 (10.0)Hypertriglyceridemia1 (5.0)01 (5.0)02 (10.0)Wound infection02 (10.0)002 (10.0)ALT elevation1 (5.0)1 (5.0)002 (10.0)TSH elevation2 (10.0)0002 (10.0)Hemobilirubin elevation2 (10.0)0002 (10.0)r‐GT01 (5.0)001 (5.0)Bacterial pneumonia0001 (5.0)1 (5.0)Unknown death0001 (5.0)1 (5.0)Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma‐glutamyl transferase; LDH, lactate dehydrogenase. Treatment‐related adverse events. Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma‐glutamyl transferase; LDH, lactate dehydrogenase. Discussion Platinum‐based combination chemotherapy incorporating cetuximab or PD‐1/PD‐L1 inhibitors constitutes the current standard of care for R/M HNSCC [15, 16]. Nevertheless, the overall survival of patients with R/M HNSCC remains poor, particularly among those with platinum‐resistant disease [17, 18]. Evidence indicates that, compared with treatment‐naïve tumors, recurrent HNSCC tissues display significantly reduced numbers of CD8+ T cells, NK cells, and B cells, along with marked downregulation of B cell–related signaling pathways and genes implicated in tertiary lymphoid structure formation. These findings suggest that prior treatment may disrupt the pre‐existing anti‐tumor immune microenvironment, thereby potentially diminishing the efficacy of subsequent immunotherapy [19]. Multi‐kinase inhibitors (MKIs) exert simultaneous inhibition of multiple signaling pathways involved in tumor proliferation, angiogenesis, and immune evasion [20]. Beyond their direct anti‐tumor effects, MKIs can induce vascular normalization, alleviate hypoxia, reduce the release of immunosuppressive mediators, and remodel the tumor immune microenvironment through blockade of pathways such as VEGFR, thereby enhancing immune effector functions [21]. Consequently, the combination of MKIs with ICIs has emerged as a pivotal strategy to augment anti‐tumor immunity and extend survival [22, 23, 24, 25]. In this phase II study, we assessed the efficacy and safety of camrelizumab in combination with famitinib in patients with refractory R/M HNSCC. The regimen demonstrated notable anti‐tumor activity, with an ORR of 35.0%, median PFS of 4.1 months, and median OS of 8.4 months. Although three grade 5 treatment‐related adverse events were reported, the overall safety profile was considered manageable, with hematologic toxicities being the most frequent. These results indicate that this chemotherapy‐free combination may represent a potential therapeutic option for this heavily pretreated, difficult‐to‐treat population. Our results are comparable to those reported for other anti‐angiogenic plus PD‐1 inhibitor combinations in similar patient populations. In a phase Ib/II trial, lenvatinib (20 mg/day) combined with pembrolizumab in a pre‐specified phase II expansion cohort of eight HNSCC patients achieved an ORR of 36% (95% CI, 17.2–59.3) at week 24 [26]. Chen et al. further reported an ORR of 28.6% (95% CI, 5.0–52.2), a median PFS of 4.6 months, and a median OS of 6.2 months with the same combination in anti–PD‐1‐refractory R/M HNSCC patients who had received at least two prior lines of systemic therapy and radiotherapy [27]. The efficacy observed in our study (ORR 35.0%, median PFS 4.1 months) is broadly consistent with these lenvatinib–pembrolizumab data. In contrast, a phase II study evaluating cabozantinib (40 mg QD) plus pembrolizumab in 33 R/M HNSCC patients reported substantially higher ORR (52%), median PFS (14.6 months), and median OS (22.3 months) [28]. However, that cohort exhibited more favorable baseline characteristics, with 50% of patients having a PD‐L1 CPS of 20 or higher, which may have contributed to the improved outcomes. Our exploratory analyses indicated that higher PD‐L1 TPS and CPS were associated with improved survival, supporting PD‐L1 as a potential predictive biomarker. Nevertheless, the current use of PD‐L1 CPS ≥ 1 as a selection criterion for first‐line immunotherapy remains imperfect, with notable false‐positive and false‐negative rates [5, 29]. High tumor mutational burden (TMB ≥ 10 mut/Mb) and HPV‐positive status have also been associated with favorable responses to immunotherapy in HNSCC [30, 31]. A multidimensional biomarker approach integrating clinical features, immune microenvironment factors, and genomic alterations (e.g., TP53/TSG mutations) may facilitate identification of primarily resistant populations and guide novel combination strategies [32]. Further investigation into reliable biomarkers for R/M HNSCC remains imperative. The combination of ramucirumab (a selective VEGFR2 monoclonal antibody) and pembrolizumab as first‐line therapy for R/M HNSCC was recently evaluated in a single‐center phase 1/2 trial [33]. Among 33 evaluable patients, the ORR was 55% (95% CI, 38–70), with 33% achieving complete response. Median PFS was 5.5 months (95% CI, 3.2–13.3) and median OS was 14.6 months (95% CI, 7.5–29.6). Notably, no treatment‐related deaths occurred, and the safety profile was consistent with the known toxicities of each agent, with hypertension and proteinuria being common but manageable. This regimen demonstrated favorable activity, particularly in patients with HPV‐positive disease and even in those with PD‐L1 CPS 0 disease. The complete response rate (33%) exceeded that observed with our camrelizumab–famitinib regimen, possibly reflecting differences in patient populations and mechanisms of action (selective VEGFR2 inhibition versus MKI). Nonetheless, the encouraging results of ramucirumab–pembrolizumab underscore the potential of anti‐angiogenic/immunotherapy combination strategies in R/M HNSCC and support further investigation of such regimens. The recent negative results of the phase III LEAP‐010 trial, which demonstrated that adding lenvatinib to pembrolizumab improved PFS and ORR but failed to enhance OS in first‐line PD‐L1 CPS ≥ 1 R/M HNSCC, raise important questions regarding the anti‐VEGFR/PD‐1 combination strategy [34]. The discrepancy between PFS/ORR and OS observed in LEAP‐010 suggests that early tumor control may not translate into a durable survival benefit, potentially due to the emergence of resistant clones, cumulative toxicity, or insufficient sustained immune activation. For our camrelizumab–famitinib regimen, these findings underscore the need for cautious interpretation of phase II results and highlight OS as the definitive endpoint in future randomized trials. Although the LEAP‐010 outcome tempers enthusiasm for unselected application of such combinations, the preclinical rationale for targeting VEGFR to potentiate immunotherapy remains compelling, and selective VEGFR2 inhibition with ramucirumab may confer a more favorable therapeutic index compared with MKIs [33]. Ongoing trials, such as LEAP‐009, will further clarify the role of these combination strategies. Regarding safety, the most common TRAEs in our study were hematologic toxicities and hypertension. This toxicity profile differs from those reported for other combinations: lenvatinib–pembrolizumab was predominantly associated with fatigue, diarrhea, and hypothyroidism [26]; cabozantinib–pembrolizumab was characterized by anemia, hypertension, and elevated alanine aminotransferase [28]; whereas ramucirumab–pembrolizumab commonly resulted in proteinuria, hypertension, and hypothyroidism, with no treatment‐related deaths [33]. These distinct patterns likely reflect pharmacologic differences among the various anti‐angiogenic agents. Lenvatinib is a multi‐kinase inhibitor targeting VEGFR1‐3, FGFR1, PDGFRα, RET, and c‐kit [35]; cabozantinib also inhibits MET [36]; famitinib is a structural analog of sunitinib targeting c‐kit, VEGFR2/3, PDGFR, FLT3, and RET [37]; whereas ramucirumab selectively inhibits VEGFR2. Moreover, all patients in our study had previously received platinum‐based chemotherapy, which may have contributed to the observed hematologic toxicities. Racial differences and the limited sample size may also have influenced the adverse event profile. Tolerability was a notable challenge in our study, with 65% of patients requiring famitinib interruption and 35% requiring dose reduction, primarily due to hematologic toxicities. This underscores the importance of proactive adverse event management and close monitoring to preserve treatment adherence. In the ramucirumab–pembrolizumab trial, 68% of patients completed treatment without dose holds or discontinuations, and only 32% required dose modifications, suggesting that selective VEGFR2 inhibition may be better tolerated than multi‐kinase inhibition in this context [33]. Despite the higher rate of dose adjustments in our study, the overall risk–benefit profile of camrelizumab–famitinib appears acceptable for a heavily pretreated R/M HNSCC population, with efficacy comparable to that of lenvatinib–pembrolizumab in similar cohorts. In summary, camrelizumab combined with famitinib exhibits promising antitumor activity and a manageable safety profile in refractory R/M HNSCC, warranting further investigation. The variability in efficacy and toxicity across different anti‐angiogenic/immunotherapy regimens underscores the necessity of biomarker‐driven patient selection and rigorously designed randomized trials with OS as the primary endpoint. The LEAP‐010 results serve as a cautionary note, but the positive outcomes from our study and the ramucirumab–pembrolizumab trial support continued exploration of this therapeutic approach. This study has several limitations. First, as an early‐phase exploratory clinical trial, it adopted a single‐arm design with a limited sample size and a relatively short follow‐up period. Second, given the substantial heterogeneity and complex genetic mutation landscape of R/M HNSCC, patient‐specific parameters were not incorporated into the survival analysis. Third, human papillomavirus (HPV) status—an important prognostic and predictive factor in oropharyngeal cancer—was neither systematically collected nor analyzed. Fourth, due to insufficient baseline tumor tissue samples from certain patients, definitive conclusions regarding biomarkers could not be drawn. Fifth, the small sample size limits the ability to perform a comprehensive safety assessment; although the attribution of three fatal events remains uncertain, their occurrence highlights the need for caution and further validation in subsequent investigations. Future phase III randomized controlled trials should incorporate detailed biomarker analyses of patient tumor and blood samples to refine patient selection strategies. Despite these limitations, this phase II study demonstrates that camrelizumab combined with famitinib exhibits promising antitumor activity (ORR 35.0%) and an acceptable safety profile in patients with refractory R/M HNSCC. Preliminary data suggest that PD‐L1 TPS or CPS merits further investigation as a potential predictive biomarker. These findings indicate that a chemotherapy‐free regimen combining a PD‐1 inhibitor with a multi‐kinase inhibitor may offer a novel therapeutic option for this patient population. However, the generalizability of these results is constrained by the study design and sample size. Large‐scale randomized controlled trials are warranted to confirm these findings and to determine the survival benefit of such combination regimens. Author Contributions Bei Xu: investigation (equal), writing – original draft (equal). Guopei Zhu: investigation (equal). Jin Lu: investigation (equal). Ximei Zhang: investigation (equal). Xiaozhong Chen: investigation (equal). Qinjiang Liu: investigation (equal). Song Qu: investigation (equal). Jia Fan: funding acquisition (equal), methodology (equal), supervision (equal). Tianshu Liu: project administration (equal), supervision (equal), writing – review and editing (equal). Funding This work was supported by Jiangsu Hengrui Medicine. Ethics Statement This trial was registered with ClinicalTrials.gov and approved by the institutional ethics committee review board at each site, and conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki. Consent All patients provided written informed consent. Conflicts of Interest Jiangsu Hengrui Pharmaceuticals provided camrelizumab and famitinib for the study. All other authors declare no conflicts of interest. Supporting information Table S1: List of participating sites, ethics committees and approval IDs. Acknowledgements We are grateful to all patients and their families who participated in the study. We appreciate the support from Jiangsu Hengrui Pharmaceuticals. Contributor Information Jia Fan, Email: fan.jia@zs-hospital.sh.cn. Tianshu Liu, Email: liu.tianshu@zs-hospital.sh.cn. Data Availability Statement Data collected for the study, including individual participant data, de‐identified participant data, and a data dictionary, can be requested by qualified researchers from the corresponding author. A signed data access agreement with the sponsor is required before accessing shared data. References Associated Data Supplementary Materials Table S1: List of participating sites, ethics committees and approval IDs. Data Availability Statement Data collected for the study, including individual participant data, de‐identified participant data, and a data dictionary, can be requested by qualified researchers from the corresponding author. A signed data access agreement with the sponsor is required before accessing shared data.