Head and neck cancer is the sixth most common cancer worldwide, and more than 600,000 new cases are diagnosed annually (1). Head and neck squamous cell carcinomas (HNSCCs) are the most common type of head and neck tumor worldwide and represent approximately 95% of all head and neck cancers. Tobacco exposure, alcohol consumption, and human papillomavirus (HPV) infection are known risk factors for HNSCC (2, 3). The therapeutic options for patients with HNSCC include surgery, chemotherapy, radiation, and immunotherapy (4–6). However, little progress has been made in the treatment of HNSCC, with a 5-year overall survival (OS) rate of ~50% (1), due to a propensity for locoregional recurrence and distant metastases (3). Therefore, it is necessary to identify effective predictive and prognostic biomarkers that may help to improve the prognosis of HNSCC.

A number of novel prognostic biomarkers for HNSCC have been explored (7–9). However, none of these has been recommended for routine clinical testing. A number of blood markers are routinely measured before and during routine clinical treatments, including C-reactive protein (CRP). Cancer-related systemic inflammation has been regarded as a pivotal determinant of disease progression and survival in most cancers (10–13); CRP is among these inflammatory markers. Several recent studies investigated the association between serum CRP level and HNSCC patient survival, but they yielded conflicting results (14–29). It is unclear whether serum CRP levels can be used to predict the survival of Han Chinese patients with HNSCC. Therefore, this study examined the role of serum CRP in the prognosis of patients with HNSCC.

In this study, we found that serum CRP level was associated with UICC stage, tumor localization, T stage, N stage, and HPV infection. In addition, N1–3 stage and a high serum CRP level (>11.3 mg/L) were shown to be associated with poorer OS and PFS in patients with HNSCC.

Salas et al. first explored the association between CRP level and survival of patients with unresectable HNSCC and found that CRP influenced the OS of HNSCC on univariate analysis, but not on multivariate analysis (25). However, they reported that, among all biological nutritional factors, CRP level correlated with the response to radiochemotherapy (25). Peter et al. suggested that a higher CRP level was a negative prognostic factor for OS in HNSCC patients in Germany (23). A single-center analysis of 246 elderly (≥65 years) HNSCC patients undergoing (chemo)radiotherapy could show that CRP levels were a prognostic parameter in the univariate Cox analysis, but not in the multivariate analysis (24). However, a subsequent bi-center study, in which a survival score for elderly HNSCC patients was developed and validated, demonstrated that CRP levels were an independent prognosticator for OS in elderly HNSCC patients (31). Magnes et al. reported that a higher CRP level was associated with OS in Australian patients with recurrent or metastatic HNSCC treated with cetuximab and chemotherapy (22). A Canadian study revealed that serum CRP level was an independent predictor of relapse-free survival and OS in advanced-stage HNSCC despite p16-positive status (27). Two Asian studies from Korea and Japan yielded inconsistent findings (19, 28); the Korean study showed that CRP level was not associated with the survival of HNSCC patients treated with chemoradiotherapy (19, 28), while the Japanese study indicated that a low CRP level was significantly associated with better OS in patients with recurrent and/or metastatic HNSCC receiving salvage chemotherapy following treatment with nivolumab (28).

With regard to oral cancer, Khandavilli et al. found that preoperative CRP level was associated with OS in patients with oral squamous cell carcinoma (OSCC) treated with primary surgery (18), but this was not observed in other studies (14, 15). Kruse et al. showed that there were no correlations between preoperative CRP level and either recurrence or metastasis of oral cancer (21). In addition, a high pretreatment CRP level was a consistent prognostic indicator of poor OS and cancer-specific survival (CSS) in patients with orohypopharyngeal squamous cell carcinoma (17, 20). In an Australian study of tongue squamous cell carcinoma (TSCC), Graupp et al. reported that the CRP level was related to the OS, but not DFS, in TSCC (16). Two Chinese studies indicated that serum CRP level predicted a poor prognosis in patients with nasopharyngeal carcinoma (26, 29). Some other groups shed light on the effects of the combination of CRP and other markers on the survival of HNSCC patients. Two Taiwanese studies observed that a squamous cell carcinoma antigen level of 2.0 ng/ml and a CRP level of 5.0 mg/L were associated with the prognosis of OSCC (32) and pharyngolaryngeal carcinoma (33). Ko et al. reported that the preoperative CRP–lymphocyte ratio was related to the prognosis of patients with OSCC (34). The CRP/albumin ratio could predict poor survival in patients with hypopharyngeal and laryngeal cancer (35).

Due to the discrepancies in the results of the studies outlined ( Supplementary Table 1 ) above, Chen et al. conducted a meta-analysis and reported that an elevated pretreatment CRP level is an indicator of poor prognosis in HNSCC (36). To date, no Chinese studies have investigated the effects of serum CRP level on the prognosis of patients with HNSCC, including cancers of the hypopharynx, nasopharynx, larynx, and oropharynx. Therefore, we designed this cohort study and found that a high serum CRP level was associated with poorer OS and PFS in Han Chinese HNSCC patients. There are several potential explanations for the partial inconsistencies of our findings with those of previous studies (22–25, 27, 28). One, clinical heterogeneity, including cancer location, tumor stage, and treatment method, may have affected the final results. Two, the primary outcomes, including OS, DFS, and CSS, were different between studies. Three, the cutoff value for the serum CRP level differed between studies, which might result in conflicting results. Four, the sample sizes varied between studies; studies with a limited sample size may yield false-negative results. Lastly, there were differences in the ethnicity of the study populations between studies.

Notably, the PFS of patients with HNSCC was only investigated in one previous study in South Korea by Kim et al. (19). They found that CRP level was not an independent predictor of PFS and tumor recurrence on multivariate analysis (19). In this study, we observed that a high CRP level (>11.3 mg/L) was an unfavorable prognostic factor for PFS in patients with HNSCC, which was inconsistent with the findings of Kim et al. (19). To our knowledge, this is the first study to demonstrate an association of high CRP level (>11.3 mg/L) with poorer PFS in Han Chinese patients with HNSCC.

The mechanisms underlying the association between CRP level and survival of HNSCC patients are unclear, but the following factors may play roles. Inflammation promotes the development of cancer and all stages of tumorigenesis (37). Inflammation exerts a major effect on the composition of the tumor microenvironment, particularly on the plasticity of both stromal and tumor cells (38). Inflammation is an intricate process involving dozens of molecules, including interleukin 6 (IL-6) and CRP (39). CRP, an acute-phase inflammatory protein, is synthesized in the liver in response to inflammation. Many studies have demonstrated that CRP was associated with cancer survival (36, 40–45). In addition, the production of CRP in the liver is stimulated by IL-6 (46). IL-6 accelerates angiogenesis, which promotes the progression and metastasis of tumors (47). Cancer cells could, in turn, produce several chemokines and cytokines, thus increasing the serum CRP level (48). In addition, CRP may mediate carcinogenesis and cancer progression via activation of the FcgRs/MAPK/ERK, FcgRs/NF-κB/NLRP3, and FcgRs/IL-6/AKT/STAT3 pathways (12). She et al. found that the network of CRP-interacting proteins may activate the PI3K/Akt signaling pathway, thereby contributing to the pathogenesis of hepatocellular carcinoma (49).

Several limitations were shown in this study. First, the study population was relatively small. Therefore, additional studies with larger sample sizes in China are required. Second, the clinical heterogeneity of HNSCC was inevitable (e.g., diverse treatment strategies were applied in the patients). Third, other pivotal factors affecting the prognosis of HNSCC were not explored in this study. Fourth, as this was a retrospective study of the effects of serum CRP level on survival in HNSCC patients at a single center, not all laboratory indexes were available for all HNSCC patients. Prospective studies at multiple centers should therefore be designed in the future. Lastly, a further study is needed to determine the mechanisms underlying the results of this study, which will help to identify new targets for innovative biological therapy in HNSCC patients.

In total, this study shows that a higher CRP level is associated with poorer PFS and OS in patients with HNSCC. Therefore, CRP level may be useful as a prognostic indicator for HNSCC patients. Serum CRP level can be measured simply and repeatedly and therefore could be regarded as a routine clinical marker in patients with HNSCC.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by Huaian No. 1 People’s Hospital. The patients/participants provided their written informed consent to participate in this study.

Conception and design: YZ and DG. Administrative support: DG. Provision of study materials or patients: YZ and DG. Collection and assembly of data: YZ. Data analysis and interpretation: YZ and DG. Manuscript writing: YZ and DG. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.