Hemastatic biomarkers as prognostic factors in locoregionally advanced nasopharyngeal carcinoma treated with nimotuzumab plus chemoradiotherapy

Background: The prognostic value of hemostatic biomarkers in many tumors are well recognized. However, it remains unclear in locoregionally advanced nasopharyngeal carcinoma (NPC) patients treated with nimotuzumab plus chemoradiotherapy . In this study, we investigated pretreatment hemostatic biomarkers as prognostic factors for short-term efficacy after treatment. Methods: In total, 101 patients with stage Ⅲ or stage Ⅳa NPC were enrolled from March 2016 to December 2018. All patients received chemoradiotherapy with concurrent nimotuzumab (200 mg/week) administration. Platelet count (PLT), prothrombin time (PT), international normalized ratio (INR), thrombin time (TT), activated partial thromboplastin time (APTT), and fibrinogen (FIB) were collected before treatment. Results: There was a statistically significant association between APTT and gender ( P =0.006), while PT and INR were found to be correlated with age ( P = 0.020 and P = 0.011, respectively). Prolonged PT or elevated INR was associated with a higher CR rate after treatment ( P = 0.008 and P = 0.003, respectively). Binary logistic regression analysis demonstrated that age, T category, N category, PT and INR were independent prognostic factors for primary tumor response, whereas N category and the overall stage could independently predict cervical lymph node response. Conclusion: Pretreatment PT and INR may serve as useful prognostic biomarkers for predicting the primary tumor response to combined nimotuzumab and chemoradiotherapy in patients with locoregionally advanced NPC.


3
Radiotherapy is the mainstay for early-stage disease, while concurrent chemoradiotherapy with or without neoadjuvant chemotherapy or adjuvant chemotherapy is recommended for locoregionally advanced disease. Currently, the prognosis of patients with NPC depends primarily on TNM staging system. However, patients with the same stage and similar treatment strategy may undergo various clinical courses and have different clinical outcomes, due to the heterogeneity of tumor. Hence, TNM staging system is insufficient for the prognostication for patients with NPC.
There is a strong correlation between cancer and aberrant hemostasis [2]. Cancer cells activate the coagulation system, meanwhile, the hemostatic factors play an important role in tumor progression [3]. A substantial amount of data suggests that a hypercoagulable state is associated with a large tumor burden, clinical progression, low rates of response to chemotherapy, and a poor prognosis. In this case, hemostatic biomarkers may serve as a potential tool in predicting cancer outcomes [4]. Mounting evidence shows that hemostatic biomarkers are associated with prognosis in NPC [5][6][7], and help to improve the prediction accuracy of the TNM staging system [8].
Huang et al. [9] reported a multi-center phase Ⅱ clinical trial of nimotuzumab combined with radiotherapy for locoregionally advanced NPC. Its results demonstrated that the complete response of the combined therapy group was significantly higher than that of the radiotherapy alone group, indicating that nimotuzumab synergized with radiotherapy presumably by increasing the radiosensitivity of tumor cells. Thus, nimotuzumab is recommended as an adjunct to radiotherapy for locoregionally advanced NPC.
Subsequently, several retrospective researches confirmed that combining nimotuzumab with chemoradiotherapy increased treatment efficacy and improved prognosis with tolerable toxicities [10][11][12]. However, it is important to note that not all patients benefit from the concurrent treatment of nimotuzumab. So, the challenge is to develop a reliable 4 and preferably non-invasive methodology to screen patients for nimotuzumab. The aim of this retrospective study is to evaluate the prognostic potential of pretreatment hemostatic biomarkers for short-term efficacy in locoregionally advanced NPC patients treated with nimotuzumab, and find the best combination of prognostic markers to identify suitable patients for this additional treatmen  Table 1. Written informed consent was obtained from all patients. This study was approved by the Ethics Board of the Meizhou People's Hospital.

Treatment responses assessment
At the end of radiotherapy, fiberoptic nasopharyngoscopy, chest X-ray/CT, ultrasound/CT of upper abdomen, MRI of the nasopharynx and neck were performed. Treatment responses were assessed according to the Response Evaluation Criteria in Solid Tumors (version 1.1) [13].

Statistical analysis
Continuous variables were categorized using median values as cut-off point. Assessments of association of pretreatment hemastatic biomarkers with clinicopathological features, and short-term efficacy after treatment, were accomplished using Chi-square test. Binary logistic regression analysis was carried out to determined the significant prognostic factors for short-term efficacy. All statistical analysis was performed using SPSS 25.0 software, and P values of <0.05 were considered statistically significant.

The correlations between pretreatment hemastatic biomarkers and clinicopathological features
There was a statistically significant association between APTT and gender. Male patients exhibited prolonged APTT when compared with female subjects (P = 0.006; Table 2).
Patients aged ≤47 had significant prolonged PT and elevated INR than those aged >47 (P = 0.020 and P = 0.011, respectively). Advanced N category was associated with elevated platelet counts and INR, although the differences between two groups nearly reached statistical significance (P = 0.066 and P = 0.064, respectively). However, neither pretreatment TT nor FIB was correlated with any clinicopathological features of patients.

The correlations between pretreatment hemastatic biomarkers and short-term
At the end of radiotherapy, the response rates of the primary disease were as follows: 69 patients had a complete response (CR) (68.3%), 30 patients had a partial response (PR) (29.7%), and 2 patients had a stable disease (SD) (2.0%), giving an objective response rate of 98.0%. For 97 NPC patients with lymphoadenopathy, CR, PR, and SD rates of the cervical lymph nodes were 80.4% (78/97), 18.6% (18/97), 1.0% (1/97), respectively. As only 3 patients had a stable disease after radiotherapy, they were combined with those who had a partial response, when the association between pretreatment hemastatic biomarkers and tumor response was evaluated. At the end of study, 81.3% of patients with PT >11.2 sec and 56.6% of patients with PT ≤11.2 sec experienced CR of primary tumor after treatment (P = 0.008; Table 3). The CR rate of patients with INR 1.05 was significantly higher than that of patients with INR ≤1.05 (83.0% vs. 55.6%, P = 0.003). No significant correlations were observed between the short-term efficacy and pretreatment PLT, TT, APTT, or FIB.

Prognostic factors
Binary logistic regression analysis demonstrated that age, T category, N category, PT and INR were independent prognostic factors for primary tumor response. Younger age, early T category, advanced N category, prolonged PT and elevated INR were associated with a higher response rate of primary tumor. N category and overall stage were independent prognostic factors for cervical lymph nodes response. Patients with advanced N category or overall stage would have relatively lower response rates of cervical lymph nodes (Table   4).

Discussion
Malignancy affects the hemastatic system, and in turn, the hemastatic system affects 8 malignancy [3]. Subclinical hypercoagulation condition is common in cancer patients, and almost all patients with advanced tumors present with low-grade systemic coagulation activation, as indicated by abnormalities in laboratory coagulation tests [4,14].
Meanwhile, systemic activation of hemostasis has been exhaustively implicated in the pathogenesis, progression, and metastasis of cancer [15]. In the current analysis, advanced N category was associated with aberrant hemostasis (i.e. elevated platelet counts and INR), and the differences were nearly reached statistical significance.
PT is the time taken for plasma to clot when exposed to tissue factors. It is often used to evaluate the extrinsic pathway of coagulation, and reflects the activity of coagulation factors Ⅰ, Ⅱ, Ⅴ, Ⅶ, and Ⅹ. Given the results are coming from different instruments and reagents, it is difficult to compare the PT measurements among different laboratories.
Nonetheless, INR makes the results comparable. Mounting evidence demonstrated that prolonged PT was associated with poor prognosis in lung cancer [16,17], pancreatic cancer [18], colorectal cancer [19], and hepatocellular carcinoma [20]. The exact mechanisms underlying this association remain unclear. One of the possible explanations is that the activation of the blood clotting cascade is common in the tumor microenvironment [21], and systemic activation of hemostasis and thrombosis promotes the progression and metastasis of cancer [15]. During this process, a large number of coagulation factors may be consumed, resulting in the prolongation of PT. In the current study, our result showed that pretreatment PT and INR were independent prognostic factors for primary tumor response. There was a significantly positive correlation between PT/INR and the response extent of primary tumor. Interestingly, those with prolonged PT or elevated INR had a higher CR rate after treatment. We speculate on the reasons for the discrepancy between others' findings and ours. Firstly, NPC has distinct biological characteristics from other solid tumors, so that the prognostic factors may be different; Secondly, chemoradiotherapy is the standard treatment modality for locoregionally advanced NPC currently, whereas patients in this retrospective study received combined nimotuzumab with chemoradiotherapy, so that the treatment outcomes may be changed with the addition of nimotuzumab. Several previous studies concluded that nimotuzumab combined with chemoradiotherapy benefited patients, and was more effective than chemoradiotherapy alone [10][11][12]. According to our findings, prolonged PT and elevated INR before treatment can predict short-term efficacy in locoregionally advanced NPC patients for combing nimotuzumab treatment with chemoradiotherapy.
It is well known that platelets play a vital role in hemostasis and thrombosis. But mounting evidence suggests that platelets also mediate tumor cell growth, dissemination, and angiogenesis. Platelets facilitate tumor metastases through direct interaction with tumor cells or through secreting bioactive proteins. Platelets also form aggregates with tumor cells, thereby protecting them from host immune surveillance. Meanwhile, plateletsecreted growth factors stimulate tumor cells proliferation to micrometastatic foci [22][23].
FIB is a key glycoprotein synthesized by hepatocytes, which can turn into fibrin by reacting with thrombin. Fibrin provides a scaffold for tumor cell anchorage and invasion, and protects tumor cells from immune system recognition and destruction [4]. Several studies have confirmed that an increase in platelet number and FIB level were correlated with poor prognosis in many solid tumor types [24][25][26][27][28][29]. A similar conclusion was also drawn in NPC [5][6][30][31][32][33]. Thrombocytosis and hyperfibrinogenemia were found to be unfavorable prognostic factors. In addition, Gao et al. [30] reported that platelet count showed significant correlation with T stage and clinical stage, while He et al. [6] showed that a high pretreatment plasma FIB level was associated with advanced TNM stage.
Nevertheless, in the current study, neither pretreatment PLT nor FIB level was significantly associated with TNM stage, or tumor response after radiotherapy. One of the possible explanations is that the assay process is not harmonized across different laboratories, and cut-off values for variables are determined by different methods, leading to different results. In most of the published studies, cut-off values were established using receiver operating characteristic (ROC) curve analysis or upper limit of normal. However, continuous variables were categorized using median values as a cut-off point in the present analysis. Due to the lack of assay standardization, the cut-off points have varied across reported studies, making it difficult to compare the results across studies [34].
Hence, seeking a harmonized assay with standard procedures is warranted.
Several limitations of the present study should be noted. Firstly, the study was a retrospective cohort study conducted at a single center with a relatively small sample size. Secondly, plasma D-dimer, an important hemostatic biomarker, was not included in the present study. Previous study has shown that elevated pretreatment plasma D-dimer level was a highly significant predictor for a worse outcome in patients with NPC [7].
Thirdly, only short-term tumor response after radiotherapy was evaluated in our study, but long-term survival outcomes are yet to be followed up. Fourthly, as some patients declined MRI three months after radiotherapy, the results were only evaluated at the end of treatment with no follow-up evaluation. In this case, the prognostic values of hemostatic biomarkers might have been underestimated.

Conclusions
In summary, our results suggest that pretreatment PT and INR were independent