Abstract
Purpose
Given the critical roles of body mass index (BMI) and age in cancer prognosis, this study investigated their impact on non-metastatic nasopharyngeal carcinoma (NPC).
Methods
Patients aged > 18 years with newly diagnosed non-metastatic NPC treated at the First Affiliated Hospital of Guangxi Medical University between June 2016 and March 2020 were retrospectively enrolled. BMI at diagnosis was classified into underweight, normal, overweight, and obese categories according to the WHO Asian BMI criteria, and patients were stratified into 20-year age intervals.
Results
A total of 674 patients were included in the study. There were 310 (46.0%), 244 (36.2%), 45 (6.7%), and 75 (11.1%) patients in the normal, overweight, obese, and underweight BMI groups, respectively. Age distribution was 205 (30.4%) young, 393 (58.3%) middle-aged, and 76 (11.3%) older. The median follow-up duration was 72 months (range, 3–98 months). The 5-year overall survival (OS) and event-free survival (EFS) rates were 79.2% and 77.0%, respectively. Overweight patients had significantly better OS than those with normal BMI (85.2% vs. 77.4%, p = 0.021) or underweight BMI (85.2% vs. 69.3%, p < 0.001). Underweight patients had a worse EFS than overweight (66.7% vs. 81.6%, p = 0.002) or normal patients (66.7% vs. 76.8%, p = 0.029). Survival outcomes declined with age (p < 0.05). Cox multivariate analysis identified underweight BMI and advanced age as independent adverse prognostic factors (p < 0.05).
Conclusion
Underweight BMI and advanced age at diagnosis are associated with poor prognosis in non-metastatic NPC. These findings confirm previous observations in a modern IMRT-treated cohort and emphasize that these associations are hypothesis-generating, without implying direct causal effects.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12885-026-16276-9.
Keywords: Body mass index, Age, Survival, Non-metastatic nasopharyngeal carcinoma, Prognostic
Introduction
Nasopharyngeal carcinoma (NPC) is a common head and neck malignancy with endemic prevalence in East and Southeast Asia [1]. NPC shows notable geographic disparities in incidence and mortality, with Asia comprising 83.3% of global cases and 83.6% of related deaths.With advancements in intensity-modulated radiotherapy and chemotherapy, the 5-year overall survival (OS) rate for early-stage NPC now exceeds 90% [2]. However, the prognosis of advanced and metastatic stages remains poor. Induction chemotherapy followed by cisplatin-based concurrent chemoradiotherapy remains the standard first-line treatment for advanced NPC [3]. However, the prognosis of locoregionally advanced disease, particularly stage T4 or N3, remains poor [4]. Stage T4 NPC carries the highest risk of local failure, with a 5-year local failure-free survival rate of 76%, compared to over 90% for T1–3 disease. Likewise, distant failure-free survival is significantly lower in N3 disease (approximately 66%) than in N1–2 (79.8–88.6%) [5]. Therefore, pre-treatment prognostic factors are vital for predicting survival prediction.
Body mass index (BMI) is a key indicator of body weight. An abnormal BMI, whether present as obesity or cachexia, is frequently observed in various malignancies. Previous studies have shown that the BMI is associated with the prognosis of various cancer types [6–10]. Obesity accelerates tumor progression and metastasis in cancers [8, 9], and a low BMI is a critical factor associated with a heightened risk of adverse outcomes and mortality in cancer patients [6, 7]. However, the relationship between pretreatment BMI and prognosis in patients with NPC is inconsistent. A cohort study suggested that being overweight or obese at diagnosis is linked to lower all-cause and cause-specific mortality in patients with NPC, whereas lean body composition is associated with higher mortality compared to individuals with normal weight [11]. Another study examining the relationship between BMI and prognosis in patients with locally advanced NPC yielded conflicting results [12], indicating that BMI may not be an independent prognostic determinant of NPC.
The incidence of NPC varies with age and typically peaks in middle age in endemic populations [13]. Globally, the number of NPC cases among middle-aged and elderly individuals has increased by 58.2%, with mortality increasing by 33.8% and disability-adjusted life-years increasing by 42.1% [14]. Compared with younger patients with NPC, middle-aged or elderly patients typically have a worse prognosis and reduced survival rates [15], which may be associated with diminished immunity, increased complications, and higher nutritional needs in the elderly [16]. For instance, a later disease stage at presentation, more aggressive tumor characteristics, suboptimal treatment, and risk factors such as smoking and alcohol consumption in the elderly may contribute to survival outcomes. As the global population continues to age, the disease burden in this age group poses a significant challenge. However, the impact of age on BMI and its subsequent effect on NPC prognosis remain unclear.
This retrospective study aimed to evaluate the prognosis of patients with non-metastatic NPC at diagnosis using BMI and age as indicators. We hypothesize that BMI and age independently and jointly affect survival in patients with non-metastatic NPC. BMI is a readily available, albeit crude, anthropometric marker that reflects overall nutritional status, and is distinct from detailed body composition measures such as sarcopenia. This study focused on the relationship between age and BMI, with the goal of providing a rational classification of patients with non-metastatic NPC and offering further evidence to support long-term survival management strategies.
Materials and methods
Patients
Patients with complete clinical pathological data and newly diagnosed NPC treated at the First Affiliated Hospital of Guangxi Medical University between June 2016 and March 2020 were enrolled in the study. This retrospective observational study was conducted in accordance with the STROBE guideline. The exclusion criteria were as follows: (1) age < 18 years (n = 5), (2) radiologically or histopathologically confirmed distant metastases (n = 11), (3) prior anticancer treatment or concomitant malignancies (n = 16), (4) severe cardiovascular or cerebrovascular diseases affecting treatment (n = 9), (5) lack of magnetic resonance imaging assessment preventing definitive tumor staging (n = 15), (6) interruption of chemoradiotherapy (n = 7), and (7) loss to follow-up (n = 29). After excluding 92 patients based on these criteria, a total of 674 patients were included in the final cohort.
This study complies with the 1964 Declaration of Helsinki and its revised versions. The waiver for informed consent and the study design were approved by the Ethics Committee of the First Affiliated Hospital of Guangxi Medical University(approval number: 2025-E0234).
Data collection and follow-up
Clinical data included patient information, such as stage, treatment, pathological type, sex, age, BMI, history of smoking and drinking, comorbidities (including hypertension, diabetes, hepatitis, healed tuberculosis, and rhinitis), and serum albumin level before treatment. All patients underwent definitive intensity-modulated radiation therapy, whereas those with locally advanced NPC received concurrent radiochemotherapy along with induction or adjuvant chemotherapy.
The follow-up period ended in March 2025. Patients were followed up every three months for the first two years post-treatment, semi-annually from years three to five, and annually thereafter. Follow-ups will be conducted via telephone or outpatient visits.
Variable definitions
BMI was calculated as body weight (kg) divided by height squared (m²). The BMI classification used in this study was based on the World Health Organization (WHO) guidelines for Asian populations [17]. These cut-offs were recommended because, at the same BMI, Asians tend to have a higher percentage of body fat and obesity-related health risks compared with Western populations. Accordingly, patients were categorized into four groups: normal BMI (18.5–22.9 kg/m2), overweight BMI (23.0–27.4 kg/m2), obese BMI (≥ 27.5 kg/m2), and underweight BMI (< 18.5 kg/m2). Based on the five age group classifications in the study by Yu and Zhang [18], the participants were grouped into 20-year intervals. The age range of this study was 18–78 years, with only 7 participants aged < 20 years and no patients aged > 80 years. Participants were divided into three age groups: 18–39 years (young), 40–59 years (middle-aged), and ≥ 60 years (older). According to the WHO pathological classification, NPC was divided into keratinizing squamous cell carcinoma (WHO type I), non-keratinizing squamous cell carcinoma—subclassified into differentiated (WHO type II) and undifferentiated (WHO type III) forms—and basaloid squamous cell carcinoma [15, 19]. The Ninth Version of the AJCC and UICC Nasopharyngeal Cancer TNM Staging Classification [20] was used to re-determine the tumor stage for all patients.
Primary endpoints were OS and event-free survival (EFS). OS was defined as the duration from diagnosis to death from any cause or the last follow-up in surviving patients. EFS was defined as the time from diagnosis to the first event of disease progression, relapse, second malignancy, death from any cause, or last follow-up in patients without such events.
Statistical analysis
Statistical analyses were performed using the IBM SPSS Statistics (version 24.0; IBM Corp., Armonk, NY, USA) and GraphPad Prism 8.3.0 (GraphPad Software, San Diego, CA, USA). Chi-square or Fisher’s exact tests were used to compare variables between the groups. Kaplan-Meier survival curves were generated for OS and EFS, and the log-rank test was used to evaluate the differences in adverse event risk between the groups. Multivariate Cox proportional hazards regression was used to identify the prognostic factors for survival. Variables with statistical significance in univariate analysis (p < 0.05) were included in the multivariate model to control for potential confounding. The proportional hazards assumption was verified using Schoenfeld residuals; all covariates satisfied the assumption. To assess potential selection bias, baseline characteristics of included and excluded patients were compared using Chi-square test for categorical variables. No adjustment for multiple comparisons was performed, given the limited number of variables included and the hypothesis-driven nature of the analyses. Statistical significance was defined as a two-sided p value of < 0.05.
Results
Characteristics
A total of 674 patients were enrolled, including 650 patients with non-keratinizing subtype carcinomas. Among them, 479 were male, with a male-to-female ratio of 2.5:1. The median age at diagnosis was 46 years (range: 18–76 years), and 205 (30.4%) patients were classified into young, middle-aged, and older age groups were 205 (30.4%), 393 (58.3%), and 76 (11.3%), respectively. The median serum albumin level was 38.9 g/L, and 24.8% of the patients had hypoalbuminemia (< 35 g/L). A total of 142 (21.2%) patients had comorbidities, with hepatitis B and hypertension being the most common. A total of 4 (0.6%), 273 (40.5%), and 397 (58.9%) patients were stratified into clinical stages I, II, and III, respectively. Based on BMI classification, 310 patients (46.0%) were of normal weight, 244 (36.2%) were overweight, 45 (6.7%) were obese, and 75 (11.1%) were underweight. The clinical characteristics are summarized in Table 1. Comparison of baseline characteristics between included and excluded patients is summarized in Supplementary Table S1. Most variables, including sex, T stage, N stage, clinical stage, and BMI, did not differ significantly ( p > 0.05). Age ( p = 0.014) and albumin level ( p = 0.04) showed minor differences, indicating minimal selection bias.
Table 1.
Clinical characteristics across BMI groups in non-metastatic NPC patients
| Item | Body mass index (BMI, kg/m2) | p value | |||
|---|---|---|---|---|---|
| Normal (18.5–22.9) |
Overweight (23.0-27.4) |
Obese (≥ 27.5) |
Underweight (< 18.5) |
||
| Sex | 0.032 | ||||
| Male | 209 (43.6%) | 189 (39.5%) | 33 (6.9%) | 48 (10.0%) | |
| Female | 101 (51.8%) | 55 (28.2%) | 12 (6.2%) | 27 (13.8%) | |
| Age/year | 0.003 | ||||
| 18–39 | 93 (45.4%) | 60 (29.3%) | 15 (7.3%) | 37 (18.0%) | |
| 40–59 | 182 (46.3%) | 154 (39.2%) | 28 (7.1%) | 29 (7.4%) | |
| ≥ 60 | 35 (46.1%) | 30 (39.5%) | 2 (2.6%) | 9 (11.8%) | |
| Non-keratinising subtype carcinoma | 0.412 | ||||
| No | 14 (58.3%) | 7 (29.2%) | 0 (0%) | 3 (12.5%) | |
| Yes | 296 (45.5%) | 237 (36.5%) | 45 (6.9%) | 72 (11.1%) | |
| Drinking | 0.141 | ||||
| No | 222 (45.2%) | 175 (35.6%) | 31 (6.3%) | 63 (12.8%) | |
| Yes | 88 (48.1%) | 69 (37.7%) | 14 (7.7%) | 12 (6.6%) | |
| Smoking | 0.058 | ||||
| No | 182 (43.9%) | 146 (35.2%) | 32 (7.7%) | 55 (13.3%) | |
| Yes | 128 (49.4%) | 98 (37.8%) | 13 (5.0%) | 20 (7.7%) | |
| Comorbidity | 0.009 | ||||
| No | 258 (48.5%) | 188 (35.3%) | 28 (5.3%) | 58 (10.9%) | |
| Yes | 52 (36.6%) | 56 (39.4%) | 17 (12.0%) | 17 (12.0%) | |
| Induction chemotherapy | 0.572 | ||||
| No | 117 (44.7%) | 101 (38.5%) | 19 (7.3%) | 25 (9.5%) | |
| Yes | 193 (46.8%) | 143 (34.7%) | 26 (6.3%) | 50 (12.1%) | |
| T stage | 0.248 | ||||
| T1 | 2 (40.0%) | 2 (40.0%) | 0 (0%) | 1 (20.0%) | |
| T2 | 33 (48.5%) | 29 (42.6%) | 4 (5.9%) | 2 (2.9%) | |
| T3 | 133 (45.7%) | 108 (37.1%) | 23 (7.9%) | 27 (9.3%) | |
| T4 | 142 (45.8%) | 105 (33.9%) | 18 (5.8%) | 45 (14.5%) | |
| N stage | 0.717 | ||||
| N0 | 15 (50.0%) | 9 (30.0%) | 4 (13.3%) | 2 (6.7%) | |
| N1 | 90 (46.6%) | 68 (35.2%) | 10 (5.2%) | 25 (13.0%) | |
| N2 | 130 (44.5%) | 106 (36.3%) | 21 (7.2%) | 35 (12.0%) | |
| N3 | 75 (47.2%) | 61 (38.4%) | 10 (6.3%) | 13 (8.2%) | |
| Clinical stage | 0.251 | ||||
| I | 3 (75.0%) | 1 (25.0%) | 0 (0%) | 0 (0%) | |
| II | 121 (44.3%) | 106 (38.8%) | 23 (8.4%) | 23 (8.4%) | |
| III | 186 (46.9%) | 137 (34.5%) | 22 (5.5%) | 52 (13.1%) | |
| Serum albumin | 0.334 | ||||
| < 35 g/l | 79 (47.3%) | 63 (37.7%) | 6 (3.6%) | 19 (11.4%) | |
| ≥ 35 g/l | 231 (45.6%) | 181 (35.7%) | 39 (7.7%) | 56 (11.0%) | |
Clinical features of patients with BMI
Significant variations in BMI were observed across different age groups, sex, and comorbidity statuses (no/yes) (p < 0.05; Table 1). Subgroup analysis showed a significant difference in BMI between the young and middle-aged groups, with a higher proportion of overweight BMI and a lower proportion of underweight BMI in the middle-aged group (p = 0.001). Distribution of BMI at diagnosis according to sex and age is shown in Fig. 1. BMI was not significantly correlated with serum albumin levels (Spearman r = 0.024, p = 0.531).
Fig. 1.
Distribution of BMI at diagnosis by age (a) and sex (b)
Analysis for survival
The 5-year OS and EFS rates of the entire group were 79.2% and 77.0%, respectively, and the median follow-up duration was 72 months (range, 3–98 months). 23 (3.4%) patients experienced locoregional relapse, 62 (9.2%) had distant metastasis (predominantly lung metastases), and four (0.6%) experienced a second tumor (leukemia, lung cancer, hypopharyngeal cancer, and liver cancer). The 5‐year OS rates for patients with stages I, II, and III were 100%, 88.6%, and 72.5%, respectively (p < 0.001), and the 5‐ year EFS rates were 100%, 87.2%, and 69.8%, respectively (p < 0.001; Fig. 2a-b).
Fig. 2.
Kaplan−Meier curves of 5-year overall survival (OS) and event-free survival (EFS) stratified by TNM clinical stage (a-b), age (c-d), and body mass index (e-f)
The 5-year OS rates were 88.3%, 78.6%, and 57.9% in the young, middle-aged, and older age groups, respectively (p < 0.001; Fig. 2c). The corresponding 5-year EFS rates were 84.4%, 76.8%, and 57.9%, respectively (p < 0.001; Fig. 2d). As patient age increased, both OS and EFS worsened, and subgroup analyses across all age groups revealed statistically significant differences (p < 0.05).
The 5-year OS rates for patients with normal, overweight, obese, and underweight BMI at diagnosis were 77.4%, 85.2, 75.6%, and 69.3%, respectively (p = 0.005; Fig. 2e). The 5‐year OS rate for patients with a normal or underweight BMI at diagnosis was worse than those with overweight BMI (p = 0.021, p < 0.001, respectively). The 5‐year EFS rates for patients with normal, overweight, obese, and underweight BMI at diagnosis were 76.8%, 81.6%, 71.1%, and 66.7%, respectively (p = 0.015; Fig. 2f). The 5‐year EFS rate in patients with underweight BMI at diagnosis was worse than that in those with normal or overweight BMI (p = 0.029, p = 0.002, respectively).
Univariate and multivariate analysis
Univariate analysis revealed significant differences in the effects of age, BMI, comorbidities, T stage, N stage, and clinical stage on the survival time of NPC patients (p < 0.05, Table 2). Additionally, the 5-year EFS rates were lower in patients of older age, male, underweight BMI, comorbidities, and advanced T, N, and clinical stages. Age, sex, BMI, comorbidities, and clinical stage were included in the Cox model. T and N stages were analyzed in the univariate analysis but excluded from the multivariate Cox model due to collinearity with clinical stage, which is a standard approach to avoid redundancy and ensure model stability. The results indicated that age, BMI, comorbidity, and clinical stage were significant prognostic factors in non-metastatic NPC patients (p < 0.05; Fig. 3). The proportional hazards assumption was formally tested for all covariates and was satisfied (p > 0.05).
Table 2.
Univariate analyses for the whole cohort
| Variables | No. of patients | OS | EFS | ||
|---|---|---|---|---|---|
| No. of events | p value | No. of events | p value | ||
| Sex | 0.09 | 0.035 | |||
| Male | 479 | 108 | 121 | ||
| Female | 195 | 32 | 34 | ||
| Age/year | < 0.001 | < 0.001 | |||
| 18–39 | 205 | 24 | 32 | ||
| 40–59 | 393 | 84 | 91 | ||
| ≥ 60 | 76 | 32 | 32 | ||
| Non-keratinising subtype carcinoma | 0.98 | 0.771 | |||
| No | 24 | 5 | 5 | ||
| Yes | 650 | 135 | 150 | ||
| Drinking | 0.482 | 0.5 | |||
| No | 491 | 98 | 109 | ||
| Yes | 183 | 42 | 46 | ||
| Smoking | 0.642 | 0.65 | |||
| No | 415 | 88 | 97 | ||
| Yes | 259 | 52 | 58 | ||
| Comorbidity | < 0.001 | 0.001 | |||
| No | 532 | 94 | 108 | ||
| Yes | 142 | 46 | 47 | ||
| Induction chemotherapy | 0.431 | 0.279 | |||
| No | 262 | 50 | 54 | ||
| Yes | 412 | 90 | 101 | ||
| T stage | < 0.001 | < 0.001 | |||
| T1 | 5 | 0 | 0 | ||
| T2 | 68 | 9 | 9 | ||
| T3 | 291 | 42 | 49 | ||
| T4 | 310 | 89 | 97 | ||
| N stage | < 0.001 | < 0.001 | |||
| N0 | 30 | 3 | 3 | ||
| N1 | 193 | 21 | 28 | ||
| N2 | 292 | 62 | 63 | ||
| N3 | 159 | 54 | 61 | ||
| Clinical stage | < 0.001 | < 0.001 | |||
| I | 4 | 0 | 0 | ||
| II | 273 | 31 | 35 | ||
| III | 397 | 109 | 120 | ||
| Serum albumin | 0.654 | 0.432 | |||
| < 35 g/l | 167 | 33 | 35 | ||
| ≥ 35 g/l | 507 | 107 | 120 | ||
| BMI | 0.005 | 0.015 | |||
| Normal | 310 | 70 | 72 | ||
| Overweight | 244 | 36 | 45 | ||
| Obese | 45 | 11 | 13 | ||
| Underweight | 75 | 23 | 25 | ||
Abbreviations: OS Overall survival, EFS Event-free survival, BMI Body mass index
Fig. 3.
Multivariate analysis of prognostic factors for OS (a) and EFS (b) in patients with non-metastatic NPC
Aging and underweight BMI were associated with an increased risk of adverse events. Specifically, in the stratified analysis, an underweight BMI at diagnosis (HR = 1.815, 95% CI: 1.120–2.942, p = 0.016) was associated with a higher risk of mortality, whereas an overweight BMI at diagnosis (HR = 0.602, 95% CI: 0.402–0.902, p = 0.014) was associated with a lower risk (Table 3). A formal age × BMI interaction analysis was conducted for both OS and EFS. The interaction terms were not statistically significant for either endpoint, indicating that the effect of BMI on survival did not differ significantly across age groups (Supplementary Table 2).
Table 3.
Multivariate analyses in non-metastatic nasopharyngeal carcinoma
| Variables | B | SE | HR (95% CI) | p value |
|---|---|---|---|---|
| OS | ||||
| Age/year | < 0.001 | |||
| 18–39 | Ref. | 1 | ||
| 40–59 | 0.778 | 0.238 | 2.176 (1.365–3.469) | 0.001 |
| ≥ 60 | 1.686 | 0.277 | 5.397 (3.138–9.281) | < 0.001 |
| BMI | < 0.001 | |||
| Normal | Ref. | 1 | ||
| Overweight | -0.507 | 0.206 | 0.602 (0.402–0.902) | 0.014 |
| Obese | 0.34 | 0.331 | 1.405 (0.734–2.690) | 0.304 |
| Underweight | 0.596 | 0.246 | 1.815 (1.120–2.942) | 0.016 |
| Comorbidity | 0.714 | 0.182 | 2.041 (1.428–2.918) | < 0.001 |
| Clinical stage | 0.971 | 0.204 | 2.641 (1.769–3.944) | < 0.001 |
| EFS | ||||
| Age/year | < 0.001 | |||
| 18–39 | Ref. | 1 | ||
| 40–59 | 0.521 | 0.212 | 1.684 (1.112–2.552) | 0.014 |
| ≥ 60 | 1.305 | 0.256 | 3.686 (2.231–6.092) | < 0.001 |
| BMI | 0.002 | |||
| Normal | Ref. | 1 | ||
| Overweight | -0.317 | 0.192 | 0.728 (0.500–1.060) | 0.098 |
| Obese | 0.461 | 0.309 | 1.586 (0.866–2.905) | 0.135 |
| Underweight | 0.579 | 0.239 | 1.784 (1.118–2.847) | 0.015 |
| Sex | -0.343 | 0.196 | 0.710 (0.484–1.041) | 0.08 |
| Comorbidity | 0.607 | 0.177 | 1.834 (1.296–2.596) | 0.001 |
| Clinical stage | 0.954 | 0.194 | 2.596 (1.776–3.793) | < 0.001 |
Abbreviations: OS Overall survival, EFS Event-free survival, BMI Body mass index
To further evaluate the robustness of the findings, a sensitivity analysis was performed after excluding 6 patients with a follow-up duration of less than 12 months (n = 668; Supplementary Table S3). Multivariate Cox regression analyses for both OS and EFS in this cohort showed effect directions consistent with the primary analysis, with no meaningful changes in statistical significance across key prognostic variables. Overall, the sensitivity analysis confirmed the robustness of the findings.
Discussion
This retrospective study of 674 non-metastatic NPC patients treated with IMRT confirms the prognostic relevance of BMI and age, using BMI cut-offs specific to Asian populations. Underweight BMI and advanced age at diagnosis were associated with poorer survival outcomes.
Nutritional status in cancer patients is a major public health concern and has been strongly linked to overall survival [21]. BMI, a key nutrition-related factor, has been identified as a prognostic marker for NPC [22]. An abnormal BMI is becoming increasingly common in patients with NPC. A previous study reported that over 50% of patients with non-disseminated NPC were overweight, obese, or malnourished [23]. The proportion of patients with abnormal BMI in our study was 54%. The relationship between BMI and NPC has been widely recognized [22]. In this cohort, the prevalence of underweight and obesity was lower than that of normal BMI across age groups. With increasing age, the proportion of overweight individuals increases, whereas the prevalence of obesity declines. Underweight was most common in the 18–39 year age group and least prevalent in the 40–59 year age group. Male patients were significantly more likely to be overweight or obese than female patients, whereas the prevalence of being underweight was lower in males compared than in females. These BMI-related clinical characteristics may partly reflect tumor-induced metabolic competition.
In this study, patients with an underweight BMI at diagnosis exhibited significantly lower 5-year OS (69.3%) and EFS (66.7%) than those in other BMI groups. The incidence of tumor progression during treatment was markedly higher in underweight patients (33.3%) than in those with normal BMI (23.2%). Multivariate analysis identified underweight BMI as an independent prognostic factor associated with an increased mortality risk in non-metastatic NPC. These findings suggest that underweight patients are more susceptible to treatment resistance, potentially resulting in inferior clinical outcomes. Previous studies have found that malnourished patients with cancer have an inferior quality of life, reduced tolerance to treatment, and poorer prognosis [24, 25]. Patients with head and neck cancer experience malnutrition due to multifactorial causes, including tumor-induced hypermetabolism, gastrointestinal obstruction, and treatment-related toxicities [11, 24]. Cells within the tumor microenvironment secrete inflammatory mediators such as interleukin-1, interleukin-6, tumor necrosis factor-α, and interferon-γ, which accelerate protein degradation via the ubiquitin-proteasome pathway, leading to muscle atrophy [26–28]. These inflammatory mediators not only contribute to endocrine, metabolic, and central nervous system dysfunction in cancer patients but also promote tumor cell invasion and immune evasion [26]. Consistently, underweight patients in this study presented with more advanced disease stages potentially linked to these mechanisms. However, these mechanistic explanations remain speculative and should be interpreted with caution. It should be noted that BMI is a crude anthropometric measure and does not capture body composition, sarcopenia, or longitudinal weight change. In this study, serum albumin levels showed no significant correlation with BMI (r = 0.024, p = 0.531), and body composition data were not available; therefore, BMI’s utility as a surrogate for nutritional status is limited and should be interpreted cautiously. Accordingly, serum albumin was not included in the multivariate Cox regression model, as predefined variable selection prioritized stronger prognostic factors such as BMI and age. The mechanistic explanations discussed above are therefore speculative and should be interpreted with caution. Although obese patients also had worse OS and EFS than those with a normal BMI, the difference was not statistically significant (p > 0.05). This may be due to the limited sample size and retrospective nature of the study as well as the possibility that obese patients tolerate treatment better than underweight patients. It should be noted that the small number of obese patients in our cohort limits statistical power, and null findings should be interpreted with caution. This observation is likely related to the relatively low prevalence of hypoalbuminemia (24.8%) and the stronger prognostic effects of BMI and age in this population. These findings suggest that BMI may provide complementary information to serum albumin in reflecting nutritional status in non-metastatic NPC patients treated with IMRT. The sensitivity analysis yielded consistent findings, supporting the robustness of the primary results.
Interestingly, overweight patients demonstrated favorable prognosis, with 5-year OS and EFS rates of 85.2% and 81.6%, respectively. Stratified analysis showed a significantly longer OS in overweight patients than in those with a normal BMI (p = 0.014, HR = 0.602, 95% CI = 0.402–0.902). Previous studies have indicated that 70%-90% of patients with NPC experience weight loss during treatment [29, 30]. In overweight individuals, this may bring BMI closer to normal levels, whereas sufficient energy reserves reduce the risk of cancer-associated cachexia. Consequently, overweight patients may avoid the adverse effects of excess adiposity on tumor progression and malnutrition-related treatment intolerance, achieving prognoses comparable to or better than those of patients with a normal BMI. Our findings highlight the importance of comprehensive nutritional assessment and support before and during treatment, rather than relying solely on BMI, to help mitigate adverse outcomes in underweight and malnourished patients.
Age is a major risk factor for tumors and previous studies have shown that advanced age is associated with poor clinical outcomes in patients with colorectal cancer [31]. Our study identified age as a significant prognostic factor (p < 0.001), with middle-aged and older patients exhibiting lower 5-year OS rates (78.6% and 57.9%) compared to younger patients, consistent with previous findings [14, 15]. Elderly patients exhibit a decline in immune function, characterized by reduced antigen uptake by dendritic cells, impaired T cell expansion and differentiation, and decreased effector molecule function [32, 33]. Poor nutritional status and a weaker host immune response to tumors are closely linked to an unfavorable prognosis, partially explaining the poor clinical outcomes in elderly patients. In addition, older patients are more susceptible to comorbidities, such as hypertension, diabetes, and cardiovascular diseases [34], which may also contribute to the poor prognosis of older patients with NPC. Therefore, a scientifically rigorous and comprehensive assessment of the physical condition, coupled with more aggressive treatment strategies for elderly patients with cancer, requires further investigation. We further evaluated whether the effect of BMI on survival differed across age groups. Formal interaction analysis showed no significant interaction between age and BMI for either OS or EFS ((p > 0.05), suggesting that the prognostic effect of BMI is independent of age.
Patients with multiple comorbidities exhibit diminished organ reserves, reduced treatment tolerance, and poorer prognosis [35]. In this study, 21.2% of the patients had comorbidities, with hepatitis B and hypertension being the most common. Patients with comorbidities who completed chemoradiotherapy had significantly worse OS and EFS than those without comorbidities, similar to the results of previous research [35]. Our study confirms that, for patients with comorbidities who opt for radical treatment, providing intensive support during both radiotherapy and the early post-radiotherapy period is of paramount importance. Additionally, the TNM stage was strongly associated with patient outcomes.
The limitations of this study are as follows. First, as a retrospective analysis, potential selection and information biases cannot be entirely ruled out, although uniform imaging protocols and strict inclusion criteria were applied to minimize their impact. Second, data on several relevant factors, including weight loss, body composition, longitudinal BMI changes, and sarcopenia, were unavailable, which may have limited further exploration of nutritional effects. Third, the BMI and age classifications used in this study may have affected subgroup comparisons, though these criteria were adopted to ensure consistency with established standards. Fourth, the relatively small sample size, particularly in the underweight and obese subgroups, may have reduced the statistical power. Nevertheless, the observed differences in OS and EFS remained significant, supporting the reliability of the findings. Fifth, although the median follow-up duration was relatively long, the minimum follow-up time was 3 months because a small number of patients experienced early events shortly after diagnosis or treatment. This variation in follow-up duration may introduce potential bias and should therefore be interpreted with caution. Finally, although we formally evaluated the potential interaction between age and BMI using multivariate Cox models, the study may still be limited by sample size in certain subgroups, and future prospective studies are warranted to validate these results. Additionally, established prognostic factors, including EBV DNA levels, treatment compliance, and dynamic weight loss, were not included in our analyses and may represent potential confounders influencing the observed associations.
In conclusion, this study shows that non-metastatic NPC patients with underweight BMI or advanced age at diagnosis are associated with poorer survival outcomes. These findings highlight observed associations rather than causal effects, and may serve as a basis for generating hypotheses for future research.
Supplementary Information
Acknowledgements
We would like to acknowledge and thank the patients who participated in the study and Editage (www.editage.cn) for the scientific editing of this manuscript.
Authors’ contributions
Yuanling Yang: Conceptualization, Methodology, Writing – original draft; Yingyi Luo: Conceptualization, Supervision，Writing – review & editing; Tianyu Wu: Data curation, Investigation; Guangren Huang: Formal analysis, Data curation; Yifan Xu: Statistical analysis, Formal analysis; Xinting Peng: Statistical analysis; Weiyi Liu: Validation, Writing – review & editing; Zisan Zeng: Validation, Writing – review & editing.
Funding
The research was supported by National Natural Science Foundation of China (82460337), Joint Project on Guangxi Key Research and Development Plan (No. GK AB24010172), and Regional High-Incidence Diseases Research of Guangxi Natural Science Foundation under Grant (No.2024GXNSFAA010315).
Data availability
Data supporting the results of this study are available from the corresponding author upon reasonable request. Access may be restricted to safeguard patient privacy and in accordance with applicable ethical guidelines, as this is a retrospective study.
Declarations
Ethics approval and consent to participate
The waiver for informed consent and the study design were approved by the Ethics Committee of the First Affiliated Hospital of Guangxi Medical University(approval number: 2025-E0234).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Yuanling Yang and Yingyi Luo contributed equally to this work.
Contributor Information
Weiyi Liu, Email: liuweiyiupday@163.com.
Zisan Zeng, Email: zengzisan@aliyun.com.
References
- 1.Chang ET, Ye W, Zeng YX, Adami HO. The Evolving Epidemiology of Nasopharyngeal Carcinoma. Cancer Epidemiol Biomarkers Prev. 2021;30(6):1035–47. 10.1158/1055-9965.Epi-20-1702. [DOI] [PubMed] [Google Scholar]
- 2.Su ZY, Siak PY, Lwin YY, Cheah SC. Epidemiology of nasopharyngeal carcinoma: current insights and future outlook. Cancer Metastasis Rev. 2024;43(3):919–39. 10.1007/s10555-024-10176-9. [DOI] [PubMed] [Google Scholar]
- 3.Chen YP, Ismaila N, Chua MLK, Colevas AD, Haddad R, Huang SH, et al. Chemotherapy in Combination With Radiotherapy for Definitive-Intent Treatment of Stage II-IVA Nasopharyngeal Carcinoma: CSCO and ASCO Guideline. J Clin oncology: official J Am Soc Clin Oncol. 2021;39(7):840–59. 10.1200/jco.20.03237. [DOI] [PubMed] [Google Scholar]
- 4.Ng WT, Corry J, Langendijk JA, Lee AWM, Mäkitie A, Mendenhall WM, et al. Current management of stage IV nasopharyngeal carcinoma without distant metastasis. Cancer Treat Rev. 2020;85:101995. 10.1016/j.ctrv.2020.101995. [DOI] [PubMed] [Google Scholar]
- 5.Au KH, Ngan RKC, Ng AWY, Poon DMC, Ng WT, Yuen KT, et al. Treatment outcomes of nasopharyngeal carcinoma in modern era after intensity modulated radiotherapy (IMRT) in Hong Kong: A report of 3328 patients (HKNPCSG 1301 study). Oral Oncol. 2018;77:16–21. 10.1016/j.oraloncology.2017.12.004. [DOI] [PubMed] [Google Scholar]
- 6.Zheng HL, Lin J, Shen LL, Yang HB, Xu BB, Xue Z, et al. The GLIM criteria as an effective tool for survival prediction in gastric cancer patients. Eur J Surg Oncol. 2023;49(5):964–73. 10.1016/j.ejso.2023.01.009. [DOI] [PubMed] [Google Scholar]
- 7.Zhou CJ, Lin Y, Liu JY, Wang ZL, Chen XY, Zheng CG. Malnutrition and visceral obesity predicted adverse short-term and long-term outcomes in patients undergoing proctectomy for rectal cancer. BMC Cancer. 2023;23(1):576. 10.1186/s12885-023-11083-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zou Y, Pitchumoni CS. Obesity, obesities and gastrointestinal cancers. Dis Mon. 2023;69(12):101592. 10.1016/j.disamonth.2023.101592. [DOI] [PubMed] [Google Scholar]
- 9.Zimta AA, Tigu AB, Muntean M, Cenariu D, Slaby O, Berindan-Neagoe I. Molecular Links between Central Obesity and Breast Cancer. Int J Mol Sci. 2019;20(21). 10.3390/ijms20215364. [DOI] [PMC free article] [PubMed]
- 10.Gupta A, Majumder K, Arora N, Mayo HG, Singh PP, Beg MS, et al. Premorbid body mass index and mortality in patients with lung cancer: A systematic review and meta-analysis. Lung Cancer. 2016;102:49–59. 10.1016/j.lungcan.2016.10.017. [DOI] [PubMed] [Google Scholar]
- 11.Du Y, Feng R, Chang ET, Yin L, Huang T, Li Y, et al. Body mass index and body shape before treatment and nasopharyngeal carcinoma prognosis: A population-based patient cohort study in southern China. Int J Cancer. 2023;153(2):290–301. 10.1002/ijc.34524. [DOI] [PubMed] [Google Scholar]
- 12.Hutajulu SH, Howdon D, Taroeno-Hariadi KW, Hardianti MS, Purwanto I, Indrasari SR, et al. Survival outcome and prognostic factors of patients with nasopharyngeal cancer in Yogyakarta, Indonesia: A hospital-based retrospective study. PLoS ONE. 2021;16(2):e0246638. 10.1371/journal.pone.0246638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bossi P, Chan AT, Licitra L, Trama A, Orlandi E, Hui EP, et al. Nasopharyngeal carcinoma: ESMO-EURACAN Clinical Practice Guidelines for diagnosis, treatment and follow-up(†). Annals oncology: official J Eur Soc Med Oncol. 2021;32(4):452–65. 10.1016/j.annonc.2020.12.007. [DOI] [PubMed] [Google Scholar]
- 14.Liu Q, Wang H, Chen Z, Xiong J, Huang Y, Zhang S, et al. Global, regional, and national epidemiology of nasopharyngeal carcinoma in middle-aged and elderly patients from 1990 to 2021. Ageing Res Rev. 2025;104:102613. 10.1016/j.arr.2024.102613. [DOI] [PubMed] [Google Scholar]
- 15.Chua MLK, Wee JTS, Hui EP, Chan ATC. Nasopharyngeal carcinoma. Lancet (London England). 2016;387(10022):1012–24. 10.1016/s0140-6736(15)00055-0. [DOI] [PubMed] [Google Scholar]
- 16.Wu SG, Liao XL, He ZY, Tang LY, Chen XT, Wang Y, et al. Demographic and clinicopathological characteristics of nasopharyngeal carcinoma and survival outcomes according to age at diagnosis: A population-based analysis. Oral Oncol. 2017;73:83–7. 10.1016/j.oraloncology.2017.08.006. [DOI] [PubMed] [Google Scholar]
- 17.Consultation WE. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet (London England). 2004;363(9403):157–63. 10.1016/s0140-6736(03)15268-3. [DOI] [PubMed] [Google Scholar]
- 18.Yu W, Zhang ZZ. [Age-Based Grouping Criteria in Medicine]. Zhongguo Yi Xue Ke Xue Yuan Xue Bao. 2023;45(2):285–9. 10.3881/j.issn.1000-503X.15133. [DOI] [PubMed] [Google Scholar]
- 19.Chen YP, Chan ATC, Le QT, Blanchard P, Sun Y, Ma J. Nasopharyngeal carcinoma. Lancet (London England). 2019;394(10192):64–80. 10.1016/s0140-6736(19)30956-0. [DOI] [PubMed] [Google Scholar]
- 20.Pan JJ, Mai HQ, Ng WT, Hu CS, Li JG, Chen XZ, et al. Ninth Version of the AJCC and UICC Nasopharyngeal Cancer TNM Staging Classification. JAMA Oncol. 2024;10(12):1627–35. 10.1001/jamaoncol.2024.4354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Dewys WD, Begg C, Lavin PT, Band PR, Bennett JM, Bertino JR, et al. Prognostic effect of weight loss prior to chemotherapy in cancer patients. East Coop Oncol Group Am J Med. 1980;69(4):491–7. 10.1016/s0149-2918(05)80001-3. [DOI] [PubMed] [Google Scholar]
- 22.Jin X, Hu R, Guo H, Ding C, Pi G, Tian M. Pretreatment Body Mass Index (BMI) as an Independent Prognostic Factor in Nasopharyngeal Carcinoma Survival: A Systematic Review and Meta-Analysis. Nutr Cancer. 2022;74(10):3457–67. 10.1080/01635581.2022.2084557. [DOI] [PubMed] [Google Scholar]
- 23.Du XJ, Tang LL, Mao YP, Guo R, Sun Y, Lin AH, et al. Value of the prognostic nutritional index and weight loss in predicting metastasis and long-term mortality in nasopharyngeal carcinoma. J Transl Med. 2015;13:364. 10.1186/s12967-015-0729-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Laursen A. Nutritional Screening: Development and Implementation of a Protocol in Patients With Head and Neck Cancer. Clin J Oncol Nurs. 2020;24(4):415–20. 10.1188/20.Cjon.415-420. [DOI] [PubMed] [Google Scholar]
- 25.Järvinen T, Ilonen I, Kauppi J, Salo J, Räsänen J. Loss of skeletal muscle mass during neoadjuvant treatments correlates with worse prognosis in esophageal cancer: a retrospective cohort study. World J Surg Oncol. 2018;16(1):27. 10.1186/s12957-018-1327-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Baracos VE, Martin L, Korc M, Guttridge DC, Fearon KCH. Cancer-associated cachexia. Nat Rev Dis Primers. 2018;4:17105. 10.1038/nrdp.2017.105. [DOI] [PubMed] [Google Scholar]
- 27.Loumaye A, de Barsy M, Nachit M, Lause P, Frateur L, van Maanen A, et al. Role of Activin A and myostatin in human cancer cachexia. J Clin Endocrinol Metab. 2015;100(5):2030–8. 10.1210/jc.2014-4318. [DOI] [PubMed] [Google Scholar]
- 28.Fearon KC, Glass DJ, Guttridge DC. Cancer cachexia: mediators, signaling, and metabolic pathways. Cell Metab. 2012;16(2):153–66. 10.1016/j.cmet.2012.06.011. [DOI] [PubMed] [Google Scholar]
- 29.Miao J, Wang L, Ong EHW, Hu C, Lin S, Chen X, et al. Effects of induction chemotherapy on nutrition status in locally advanced nasopharyngeal carcinoma: a multicentre prospective study. J Cachexia Sarcopenia Muscle. 2023;14(2):815–25. 10.1002/jcsm.13196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Li G, Jiang XY, Qiu B, Shen LJ, Chen C, Xia YF. Vicious circle of acute radiation toxicities and weight loss predicts poor prognosis for nasopharyngeal carcinoma patients receiving intensity modulated radiotherapy. J Cancer. 2017;8(5):832–8. 10.7150/jca.17458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Mima K, Kurashige J, Miyanari N, Morito A, Yumoto S, Matsumoto T, et al. Advanced Age Is a Risk Factor for Recurrence After Resection in Stage II Colorectal Cancer. Vivo. 2020;34(1):339–46. 10.21873/invivo.11779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Li G, Smithey MJ, Rudd BD, Nikolich-Žugich J. Age-associated alterations in CD8α + dendritic cells impair CD8 T-cell expansion in response to an intracellular bacterium. Aging Cell. 2012;11(6):968–77. 10.1111/j.1474-9726.2012.00867.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Castle SC. Clinical relevance of age-related immune dysfunction. Clin Infect Dis. 2000;31(2):578–85. 10.1086/313947. [DOI] [PubMed] [Google Scholar]
- 34.Wang C, Yuan Y, Zheng M, Pan A, Wang M, Zhao M, et al. Association of Age of Onset of Hypertension With Cardiovascular Diseases and Mortality. J Am Coll Cardiol. 2020;75(23):2921–30. 10.1016/j.jacc.2020.04.038. [DOI] [PubMed] [Google Scholar]
- 35.Sze HC, Ng WT, Chan OS, Shum TC, Chan LL, Lee AW. Radical radiotherapy for nasopharyngeal carcinoma in elderly patients: the importance of co-morbidity assessment. Oral Oncol. 2012;48(2):162–7. 10.1016/j.oraloncology.2011.08.019. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data supporting the results of this study are available from the corresponding author upon reasonable request. Access may be restricted to safeguard patient privacy and in accordance with applicable ethical guidelines, as this is a retrospective study.