Analysis of thromboembolic events in head and neck cancer patients who underwent concurrent chemoradiotherapy with cisplatin: A real-world study

Background The risk of thromboembolic events (TEE) in cancer patients, especially those receiving cisplatin-based chemotherapy, is higher than that of general population. Methods The study investigated the incidence of TEE in 257 head and neck (H&N) cancer patients, during or within 6 months of completion of concurrent chemoradiotherapy (CCRT) with cisplatin, and analyzed factors affecting TEE occurrence. Results TEE occurred in 5 patients, an incidence rate of 1.9%. Khorana score was the only factor associated with TEE occurrence (p = 0.010). The incidence of TEE in H&N cancer patients who underwent CCRT with cisplatin was relatively low when compared to other types of cancer. However, patients with a high Khorana score require more careful surveillance for possible TEE occurrence.


Introduction
While malignant tumor cells induce hypercoagulation through several mechanisms, the administration of chemotherapeutic agents may be associated with thrombotic events (1,2). As a result, the risk of thromboembolic events (TEE) in cancer patients is about four times higher than in the general population, and that risk increases up to 6.5 times for patients receiving chemotherapy (3,4). In particular, TEE incidence in various cancer types of patients treated with chemotherapy including cisplatin, one of the most frequently used agents, was reported to be higher compared to those with other regimens (5)(6)(7)(8)(9).
Head and neck (H&N) cancer refers to malignancy, usually squamous cell carcinoma, in areas such as the nasopharynx, nasal cavity, oropharynx, oral cavity, hypopharynx, and larynx, and incidence is relatively high in many countries (12,13). In locally advanced H&N cancer, concurrent chemoradiotherapy (CCRT) with cisplatin is the established standard therapy for de nitive treatment and adjuvant therapy (14)(15)(16)(17). Nonetheless, there are small number of reports regarding the incidence of TEE in H&N cancer with con icting results, without any real-world study which investigated TEE in patients receiving CCRT with cisplatin (18)(19)(20).
For that reason, we investigated the incidence of TEE in H&N cancer patients who received CCRT with cisplatin, and analyzed the factors affecting TEE occurrence.

Study Population
To begin, all patients who started CCRT with cisplatin for H&N cancer between January 2005 and December 2019 were retrospectively identi ed. The eligibility criterion was diagnosis with H&N squamous cell carcinoma, including other types of carcinoma in cases of nasopharyngeal cancer. Patients who underwent CCRT with cisplatin as de nitive therapy following diagnosis or at the time of local recurrence as well as adjuvant therapy after surgical resection, were included in the analysis.
Exclusion criteria included patients with non-squamous histology except for nasopharyngeal cancer; those with distant metastases other than cervical lymph nodes; those who experienced TEE within 3 months prior to initiation of CCRT; those who underwent chemotherapy such as induction therapy before CCRT; those who received other anticancer drugs in addition to cisplatin during CCRT; and those who had started CCRT at other hospitals before continuing of treatment at our institution. Patients with history of chemotherapy for H&N cancer or other types of cancer were included, with the exception of those who had previously received platinum agents (cisplatin, carboplatin, oxaliplatin).
The study protocol was approved by the Ajou University Hospital Institutional Review Board (IRB No. AJIRB-MED-MDB-20-456).

Clinical Review
Patient information was collected retrospectively from medical records and radiological reports, including age, gender, smoking history, previous TEE history, cancer site, cancer stage, Eastern Cooperative Oncology Group performance status at the start of CCRT, purpose of CCRT, interval of cisplatin administration, radiation dose, Khorana score, and occurrence of TEE (21,22). Tumor stage classi cation was based on the American Joint Committee on Cancer 8th edition (23).
For the purpose of this study, TEE was de ned as deep vein thrombosis (DVT), pulmonary embolism (PE), other types of venous thrombosis, myocardial infarction, cerebral artery thrombosis, and other types of arterial thrombosis, that occurred during or within 6 months of completion of CCRT. Con rmation of TEE occurrence was based on patient radiological reports, including computerized tomography, magnetic resonance imaging, doppler ultrasonography, and coronary angiography.
During CCRT, cisplatin was usually administered, with a dosage of 100 mg/m 2 every 3 weeks for 3 cycles or 30 mg/m 2 weekly for 7 weeks. The radiation treatment dose was 200cGy per day, with a target of 7000cGy at 35 fractions. Treatment dose and schedule were modi ed at the discretion of the treating physicians.

Khorana Score
The Khorana score is a model that predicts the risk of venous thromboembolism (VTE) associated with chemotherapy (22). The score is composed of ve clinical and laboratory variables: site of cancer (very high-risk site: 2 points; high-risk site: 1 point); pretreatment leukocyte count (more than 11 x 10 9 /L); pretreatment hemoglobin (less than 10g/dL) and/or use of erythropoiesis stimulating agents; pretreatment platelet count (more than 350 x 10 9 /L); and body mass index (BMI) of 35 kg/m 2 or more (all 1 point each). H&N cancer is assigned a score of 0 at the cancer site.

Statistical analysis
Continuous variables are presented as mean ± standard deviation and categorical variables as frequencies and percentages. Comparisons of continuous and categorical variables were performed by Mann-Whitney test and Fisher`s exact test, respectively. All statistical analyses were two-sided and performed using SPSS version 23.0 for Windows.

Results
Of the 265 H&N cancer patients who underwent CCRT with cisplatin during the study period, 257 patients were included in the analysis while 8 patients were excluded for the following reasons: previous history of chemotherapy including platinum (4 patients); history of TEE within 3 months before the start of treatment (2 patients); salivary gland adenocarcinoma (1 patient); and induction chemotherapy before CCRT (1 patient).

Discussion
To our knowledge, the present study is the rst to investigate the incidence of TEE in H&N cancer patients who underwent CCRT with cisplatin in real-world practice. The observed incidence of TEE was 1.9%, and Khorana score was the only factor associated with the risk of TEE.
In cancer patients, various risk factors increase the incidence of TEE. Among patient-related factors, TEE is known to be associated with age, obesity, thrombocytosis, leukocytosis, hemoglobin < 10g/dL before chemotherapy, and a history of previous TEE (22,24). Treatment-related factors such as chemotherapy are also associated with risk of TEE (2).
Con icting results have been reported regarding the incidence of TEE in H&N cancer patients. According to one meta-analysis, the incidence of VTE in H&N cancer patients was 0.16-3.13% (18). Incidence of VTE in H&N cancer patients after surgery is generally low (0 ~ 8%), although one study reported higher rate of 26.3% (18,20). While one study reported that H&N cancer is the least common type of cancer diagnosed following VTE, another large study identi ed H&N cancer as the second highest risk for VTE among 18 types of carcinoma (18,19). In a large population-based study, the incidence of TEE in noncancer population was 1.4%, which is similar to the rate of occurrence among H&N cancer patients reported in several studies (18,25).
In the present study, the observed incidence rate of TEE was 1.9%, which is not that dissimilar to the rate in the general population and aligns with previous studies regarding the incidence of TEE in H&N cancer (18). However, most previous studies of TEE in H&N cancer analyzed heterogenous populations consisting mainly of a surgical treatment alone group, with small number of reports about patients who received chemotherapy including various regimens (18,20). The present study investigated patients who underwent CCRT with cisplatin, which is the essential chemotherapeutic agent for management of H&N cancer. Although the mechanism of cisplatin-induced hypercoagulability is not entirely clear, several studies have proposed possible pathophysiologic processes; these include impaired autoregulation of the vascular system, an altered balance between thrombosis and brinolysis, direct endovascular damage, increased procoagulant activity of red blood cells, and modulation of tissue factor on human monocytes (26)(27)(28).
There have been several reports of increased TEE incidence in patients treated with cisplatin-based regimens. Among these, Russell et al. reported an 18.1% incidence of TEE related to cisplatin-based chemotherapy in 932 patients with various types of malignancy (8). The reported incidence of TEE in non-small cell lung cancer patients who underwent a cisplatin-based regimen was 8.0 ~ 17.6%, with 10.2% of the 1-year incidence rate of TEE in small cell lung cancer patients treated with cisplatin-based chemotherapy (5)(6)(7)29). In the present study, the incidence of TEE in H&N cancer patients during or after CCRT with cisplatin seems to be low when compared to patients with other types of cancer who received cisplatin-based regimens. A larger-scale study is required to determine the possible explanation for these ndings.
In the present study, one interesting result is the signi cant association between the TEE occurrence and Khorana score despite of the low incidence of TEE itself. As a means of predicting the risk of chemotherapy-associated thrombosis in cancer patients, the Khorana scoring system has some limitations. For example, since BMI > 35kg/m 2 is rare among Asian patients, accurate evaluation is di cult (5,30). In the present study cohort, only one patient (0.4%) had a BMI of > 35 kg/m 2 , which gure was not signi cantly different from that of the previous Asian study (5,30). Nonetheless, Khorana score was the only factor associated with the TEE incidence in present study. Therefore, in cases of H&N cancer patients with high Khorana score receiving cisplatin-CCRT, treating physician should be aware of possible TEE occurrence during or for a period of time after CCRT. In addition, these patients should be informed about the somewhat higher risk of TEE before starting treatment.
The current study has several limitations. As a retrospective study based on a review of medical records, selection bias may occur, resulting in possible underestimation of TEE incidence. The small number of patients with TEE means that statistically signi cant results are less reliable, especially in terms of the relationship between TEE incidence and clinical characteristics. Finally, the present study cohort comprised patients from a single institution over a fairly long period. Nevertheless, this study seems to be clinically meaningful because it investigated all patients during the de ned period who underwent CCRT with cisplatin, which is the mainstay of treatment for locally advanced H&N cancer, re ecting real-world clinical practice in Korea.
In conclusion, while the incidence of TEE in H&N cancer patients who underwent CCRT with cisplatin was relatively low (1.9%) when compared to other types of cancer, patients with a high Khorana score require more careful surveillance for possible occurrence of TEE. Declarations Ethics approval and consent to participate The protocol was reviewed and approved by the Institutional Review Board of Ajou University Hospital (IRB approval no. AJIRB-MED-MDB-20-456). Informed consent was waived by the same ethics committee that approved the study (Institutional Review Board of Ajou University Hospital). In present study, all methods were performed in accordance with the relevant guidelines and regulations.

Consent for publication
Not applicable Availability data and materials The datasets generated and/or analyzed during the current study are not publicly available due to the con dentiality of the data of patient but are available from the corresponding authors on reasonable request.

Competing interests
The authors declare no competing interests. Funding