Work

Combined Prognostic Value of Pre-Radiotherapy BMI and NRS-2002 for Overall Survival in Patients with Esophageal Cancer

Year: 2026

Type: article

Abstract: Background & Aims: Nutritional status significantly affects the treatment efficacy and survival outcomes of esophageal cancer patients receiving radiotherapy. Although body mass index (BMI) and Nutritional Risk Screening 2002 (NRS-2002) are well-established independent assessment tools, there is a lack of strong evidence validating their combinedprognostic value. This study aims to compare theprognostic value of the combined BMI and NRS-2002 model with that of single-indexmodels (BMI or NRS-2002) for survival in esophageal cancer patients undergoing radiotherapy. Methods: A retrospective analysis was conducted on data from 676 esophageal cancer patients, with a median follow-up of 24 months. Cox regression analysis was used to evaluate the prognostic value of the combined model and single models, andthe Akaike Information Criterion (AIC) and C-index were used to compare model fit and discrimination. Meanwhile, the prognostic consistency of nutritional indicators was evaluated across key clinical subgroupssuch as TNM stage. Results: Multivariate Cox regression analysis identified pre-radiotherapy BMI and NRS-2002 as independent prognostic factors for overall survival (OS). Compared to underweight patients, those with normal BMI (HR=0.599, 95% CI: 0.457-0.785, P<0.001) and overweight status (HR=0.689, 95% CI: 0.493-0.963, P=0.029) had significantly better survival. Conversely, nutritional risk (NRS-2002 ≥3) was associated with poorer survival (HR=1.279, 95% CI: 1.029-1.591, P=0.027). The combined model demonstrated superior model fit (lowest AIC: 5068.882) and discrimination (optimism-corrected C-index: 0.6597) compared to single-indicator models, although the incremental gain in discriminatory power was modest. Conclusion: Combining pre-radiotherapy BMI with the NRS-2002 provides a slight improvement in risk stratification for overall survival among esophageal cancer patients undergoing radiation therapy compared to a model based on a single indicator. Although this combined model offers complementary prognostic information, its incremental value in distinguishing patient risk is limited. In clinical practice, this combined assessment can serve as a practical adjunct to risk stratification, helping to identify high-risk patients who require comprehensive nutritional intervention. Further external validation in prospective cohorts is still needed.

Source: Cancer Management and Research

DOI: 10.2147/cmar.s639380