Chemoradiotherapy Increases the Incidence Rate of Cachexia in Head and Neck Cancer: A Cohort Study

Article information

J Korean Cancer Assoc. 2025;.crt.2025.482

Publication date (electronic) : 2025 October 30

doi : https://doi.org/10.4143/crt.2025.482

Nan-Chin Lin¹, Kuo-Ting Sun^2,3, Yu-Hsin Tseng⁴, Tzong-Ming Shieh^,3, Yin-Hwa Shih^,5

¹Department of Oral and Maxillofacial Surgery, Show-Chwan Memorial Hospital, Changhua, Taiwan

²Department of Pediatric Dentistry, China Medical University Hospital, Taichung, Taiwan

³School of Dentistry, China Medical University, Taichung, Taiwan

⁴Department of Pediatrics, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan

⁵Department of Healthcare Administration, Asia University, Taichung, Taiwan

Correspondence: Yin-Hwa Shih, Department of Healthcare Administration, Asia University, 500, Lioufeng Road, Wufeng, Taichung 41354, Taiwan Tel: 886-4-23323456 #20040 E-mail: evashih@gm.asia.edu.tw

Co-correspondence: Tzong-Ming Shieh, School of Dentistry, China Medical University, 500, Lioufeng Rd., Wufeng, Taichung 404332, Taiwan Tel: 886-423323456 #20040 E-mail: tmshieh@mail.cmu.edu.tw

*Nan-Chin Lin and Kuo-Ting Sun contributed equally to this work.

Received 2025 May 6; Accepted 2025 October 15.

Abstract

Introduction

According to the Taiwanese national cancer registry database spanning 1980 to 2019, oral cancer is the fifth most prevalent cancer among men. The incidence of oral cancer in Taiwan is strongly linked to risk behaviours such as smoking, alcohol consumption, and betel nut chewing [1-3]. Data from 2021 indicate that the incidence rate of lip and oral cavity cancer in men was 29.77 per 100,000 individuals, with a mortality rate of 11.25 per 100,000 individuals [4]. These statistics highlight the importance of prevention and effective treatment for head and neck cancer.

The management of head and neck cancer requires a collaborative approach by a multidisciplinary team. Surgery is the primary treatment modality for non-invasive and early-stage cancers, with attempts made to minimise morbidity related to the surgical procedure. For patients at high risk of recurrence, adjuvant therapies such as radiotherapy (RT) or chemoradiotherapy are commonly implemented [5,6]. When chemotherapy is administered in conjunction with RT as a strategy for organ preservation, it may exacerbate toxicities, such as oral mucositis [7-9]. It may cause the patient to have a poor quality of life. In addition, Cetuximab (Erbitux), a monoclonal antibody against the epidermal growth factor receptor, is used as adjuvant treatment in patients with locally advanced disease [10]. It is administered intravenously and commonly combined with chemotherapy or radiation therapy.

Cancer treatment can lead to a range of complications, such as xerostomia (dry mouth), oral ulcers, pain, and gastrointestinal (GI) disturbances. These complications can contribute to malnutrition, fatigue, and anaemia [11,12], which may culminate in cachexia syndrome, a risk factor for poor prognosis, and low survival rate [13,14]. Nearly 40% of patients with head and neck cancer develop cachexia immediately after treatment (within 6 months). Persistent cachexia, or cachexia that develops after 6 and 12 months of treatment, is significantly correlated with higher rates of cancer recurrence, non-cancer-related health events, and overall mortality (p < 0.05) [15].

Several studies have explored the prevalence, causation, and risk factors associated with cachexia in patients with head and neck cancer [16-18]. However, research examining the link between the clinical characteristics of patients and risk factors specific to Taiwanese behaviours (such as betel nut chewing) within Taiwan is lacking. Upon examination of the cancer registry database at the Show-Chwan Memorial Hospital from 2021 to 2022, we observed a high incidence of cachexia in patients following initial treatment. The objective of this study was to identify risk factors contributing to cachexia and provide clinical practitioners with suggestions or indicators for early intervention to prevent treatment-induced cachexia.

Materials and Methods

1. Data source and collection

Medical records were collected from the cancer registry database at Show-Chwan Memorial Hospital between 2021 and 2022. We screened eligible patients who underwent initial treatment for head and neck cancer and assessed the presence of cachexia at the 6-month post-treatment mark. During this period, a total of 362 individuals were diagnosed with head and neck cancer at the hospital. Of these, 272 patients remained at the hospital for treatment. Among them, the pre- and post-treatment body weight records were available for 115 patients. Finally, 50 patients experienced cachexia, whereas 65 patients did not exhibit cachexia symptoms after initial treatment (Fig. 1). We collected clinical data pertaining to sex, cancer type, pathological stage, pathological differentiation, recurrence, types of treatment received, complications after treatment, and cancer-related risk behaviours (alcohol drinking, betel nut chewing, and cigarette smoking).

Fig. 1.

Flow chart of patient selection in the cohort study.

Results

A total of 115 patients (108 men and 7 women) were enrolled in this study. After treatment, 50 patients developed cachexia, while 65 did not. In this cohort, 40.0% of patients were in the late stage of the disease, and 67.0% were diagnosed with oral cancer. Recurrence was detected in 21% of patients. Morphologically, the tumour cells mainly showed moderate differentiation. The distribution of patients according to the type of treatment was as follows: 60% underwent surgery, 27.8% received chemotherapy, 28.7% underwent RT, 52.2% were treated with concurrent chemoradiotherapy (CCRT), and 20% received cetuximab (Erbitux) targeted therapy. The chief complications reported by the patients were as follows: pain and ulcers in the oral cavity, 80.9%; GI symptoms, 24.3%; dysphagia, 32.3%; cough or sputum production, 30.4%; and difficulties with ingestion, 50.4%. Additionally, more than 50% of the patients engaged in cancer-associated risk behaviours, including smoking, alcohol consumption, and betel nut chewing (Table 1).

Table 1.

Patient’s characteristics (n=115)

We categorised the variables into sex, pathological data, treatment, complications, and risk behaviours. We used the chi-square test to assess the relationship and differences between the incidence of cancer cachexia and the clinical data. We found that a significantly higher proportion of patients with cachexia received CCRT (p < 0.001) rather than only RT (p < 0.001). Except for the treatment modality, no significant associations were evident between cachexia and the other clinical data (Table 2).

Table 2.

Association between cachexia and clinical characteristics

We further examined the effect of CCRT on cachexia using binary logistic regression. Initially, all the clinical parameters were analyzed using the forced entry method, which identified chemoradiotherapy, RT, CCRT, and problem with ingestion as significant predictors of cachexia based on the results of the “Variables Not in the Equation” table. Subsequently, these four parameters were entered into the model using forced entry again. As shown in Table 3, the overall model fit chi-square value was 17.726 (p=0.001), indicating that the predictors as a set reliably distinguished between outcome categories. The Hosmer-Lemeshow goodness-of-fit test value was 7.61 and nonsignificant (p > 0.05), suggesting adequate model fit. Among the individual predictors, the RT group was significantly associated with 67% lower odds of the event compared to the non-RT group. In contrast, the CCRT group was significantly associated with 173% higher odds of the event compared to the non-CCRT group. No significant differences in odds were observed for the chemoradiotherapy group or the problem with ingestion group compared to their respective reference groups. Overall, the results suggest that RT and CCRT were meaningful predictors (Table 3).

Table 3.

Overall model fit test and individual parameter significance test results

In the third round of binary logistic regression, we conducted a binary logistic regression with a “forward: conditional” model to identify the most parsimonious set of predictors while controlling for redundancy. In step 1, CCRT was entered first. Patients receiving CCRT had 289% higher odds of cachexia compared to those not receiving CCRT. The significant chi-square value indicated CCRT alone significantly improved prediction. In step 2, RT entered as the second significant variable. Patients receiving RT had 68% lower odds of cachexia compared to those not receiving RT. While CCRT remains significant in the model (p < 0.05), it was associated with 154% higher odds of cachexia when RT is also in the model. The significant chi-square value indicated that adding RT significantly improved predictive accuracy. The stepwise logistic regression identified CCRT as the strongest initial predictor of cachexia, increasing odds by 3.889-fold (Table 4). These two treatment variables significantly and independently predicted the occurrence of cachexia.

Table 4.

Stepwise logistic regression results

Almost 50% of head and neck cancer patients seek medical assistance at an advanced stage of the disease. To prevent tumour recurrence and preserve oral functionality, CCRT is frequently utilized. However, our retrospective cohort study has revealed that CCRT is a primary risk factor for the onset of treatment-induced cachexia, a serious and debilitating condition. Therefore, it is imperative for healthcare providers to be vigilant regarding this risk and to design comprehensive care plans tailored to address the nutritional and supportive needs of this patient group, ensuring better management of potential cachexia and improving overall prognosis.

Discussion

This retrospective cohort study included 115 individuals, 56.5% of whom did not exhibit cachexia characteristics. Since head and neck cancer is more prevalent in men, 93% of the participants were male. Post-treatment follow-up revealed that 21.7% of patients experienced recurrence. Nearly 60% of patients with cancer underwent surgical treatment, 52% received adjuvant treatment with CCRT, and 20% received targeted therapy (cetuximab). The most common complications after cancer treatment were pain and eating problems caused by the treatment. Over 50% of patients had unhealthy behaviors such as smoking, drinking alcohol, and chewing betel nuts. Our analysis of the risk factors for cachexia revealed a robust correlation between patients who received CCRT and the occurrence of cachexia (Table 2). The incidence probability of cachexia was 3.889 times higher in patients who underwent CCRT compared to patients who did not receive CCRT (Table 4). Betel nut chewing, a common habit in Taiwanese patients with head and neck cancer, was not related to the incidence of cachexia.

Before 2021, the medical records of Show-Chwan Memorial Hospital did not record body weight data before or after 6 months of treatment. The lack of body weight data hindered us from including more participants in this research and was a bias in estimating the prevalence of cachexia in the head and neck cancer cohort. Since this study only covers data spanning 2 years, the patient’s survival or recurrence data after 2 years were not accessible, and we could not evaluate the impact of treatment-induced cachexia on the long-term prognosis. In addition, missing data on pathological staging (n=53), degree of tissue differentiation (n=79), type of disease (n=10), and risk behaviours (n=7) could potentially bias the statistical analysis of the risk factors. Despite these limitations, this study serves as a preliminary investigation into treatment-induced cachexia among patients with head and neck cancer in Taiwan. In future research, we intend to access a more extensive database that will provide adequate evidence and shed light on treatment-related cachexia to improve clinical practice and address this challenge effectively.

One study reported that the incidence of cachexia within 6 months of treatment is higher than that after 6 and 12 months of treatment. However, cachexia occurring only after 6 to 12 months post-treatment was strongly associated with tumour recurrence and non-cancer health events and was accompanied by a higher mortality rate [15]. Our study only evaluated patients who developed cachexia within 6 months of treatment. In our analysis (Table 2), the occurrence of cachexia was not significantly correlated with tumour recurrence, consistent with the previous study’s findings.

Diagnosing cachexia requires a multifactorial approach, and while professional organizations have proposed overlapping diagnostic criteria, no universally accepted guideline currently exists. Based on our literature review, cancer-related cachexia is primarily diagnosed by body weight loss [19-21]. In contrast, non-cancer-specific cachexia is diagnosed by body weight loss accompanied by at least three of five clinical symptoms (fatigue, anorexia, decreased muscle strength, reduced fat-free mass, or systemic signs of inflammation) [22]. In other words, muscle loss is not necessarily a required criterion; body weight loss is. Therefore, in our study, cachexia was diagnosed based on body weight loss occurring within 6 months after treatment, which is appropriate for the study design.

Our data indicate that oral cancers account for 67% of head and neck cancer cases. In the Asian region (including China, India, and Taiwan), betel nut chewing is a recognized risk factor for oral cancer. Previous studies have suggested that chemicals in betel nuts can alter the body’s metabolism [23,24]. In our study, we also analysed the association between cachexia and behaviours related to head and neck cancer, such as smoking, alcohol consumption, and betel nut chewing. However, our results showed that these deleterious habits were not correlated with the development of cachexia after treatment.

In Taiwan, public health policy promotes early screening for oral cancer, as early detection and treatment are associated with improved prognosis [25]. However, real-world clinical data indicate that most patients are diagnosed at advanced pathological stages (Table 1). When there is a risk of recurrence, RT and CCRT are the preferred treatment approaches. When organ preservation is required, CCRT is often administered [26]. Patients who receive CCRT and RT may reflect the severity of the disease, and the side effects after treatment may be the confounding factors for the statistical outcomes. We provided sufficient evidence to exclude the relation between CCRT and severity of disease. First, in Table 2, there was no significant difference in pathological stage and complications after treatment between the cachexia and non-cachexia groups. Second, in Tables 3 and 4, we conducted the stepwise logistic regression analysis to exclude the confounding factors, and RT was negatively correlated to the odds of cachexia. That is to say, the CCRT is an independent risk factor of cancer treatment–induced cachexia and has no association with disease severity.

The 5-year survival rate for oral cancer is approximately 50-60%, whereas untreated patients have a median survival of 12 months [27,28]. It is imperative to plan effective posttreatment prognostic strategies for these individuals. Head and neck cancer management involves a multidisciplinary team approach, including surgeons, oncologists, radiologists, pathologists, speech therapists, nurses, social workers, and dietitians [29]. This study found that CCRT affected the development of cachexia in patients. We suggested that head and neck cancer treatment teams should be vigilant about this in clinical practice. A collaborative care plan should be formulated by radiology experts, oncology specialists, nurses, and dietitians, specifically designed to support patients undergoing CCRT. The objective of this plan should be to reduce post-treatment weight loss and decrease the risk of cachexia onset. Such a strategy is not only likely to yield better patient outcomes but also to improve the quality of life for individuals following treatment significantly [11].

Article information Continued