NUTRISCORE: A Novel Nutrition-Based Nomogram to Predict Severe Neutropenia After Adjuvant Chemotherapy in Gastric Cancer.

IF 3.1 3区医学 Q2 INTEGRATIVE & COMPLEMENTARY MEDICINE

Integrative Cancer Therapies Pub Date : 2026-01-01 Epub Date: 2026-07-03 DOI:10.1177/15347354261447851

Chun Gao, Li Zhu, Yi Xin Tong, Sheng Zhang

{"title":"NUTRISCORE: A Novel Nutrition-Based Nomogram to Predict Severe Neutropenia After Adjuvant Chemotherapy in Gastric Cancer.","authors":"Chun Gao, Li Zhu, Yi Xin Tong, Sheng Zhang","doi":"10.1177/15347354261447851","DOIUrl":null,"url":null,"abstract":"Background: To develop and validate a clinically applicable nomogram (NUTRISCORE) predicting severe neutropenia (Grade 3/4) across cycles 2 to 8 of adjuvant chemotherapy in gastric cancer.Methods: This retrospective cohort study analyzed 391 patients receiving postoperative adjuvant chemotherapy (CAPEOX/XELOX/SOX) following curative laparoscopic D2 gastrectomy. Cohorts were stratified into derivation (n = 318; 1518 cycles) and validation (n = 73; 356 cycles) sets. Generalized Estimating Equations (GEE) were used to account for within-patient correlation of multiple cycles. Multivariable logistic regression identified independent predictors, with subsequent nomogram construction using R. Model performance was quantified through AUC-ROC and decision curve analysis (DCA).Results: Four predictors emerged: age > 60 years (OR = 2.1, P < .01), prealbumin ≤ 180 mg/L (OR = 3.4, P = .002), PNI < 45 (OR = 2.8, P = .008), and prior cycle neutropenia (OR =4.7, P < .001). The nomogram achieved AUCs of 0.825 (95% CI 0.789-0.860) and 0.810 (0.783-0.835) in derivation and validation cohorts, with DCA confirming clinical utility across threshold probabilities (4%-65% in derivation, 3%-63% in validation), supporting its use for guiding G-CSF prophylaxis at clinically relevant risk thresholds (>15%).Conclusions: We establish a nutrition-centric predictive tool for severe neutropenia, demonstrating robust accuracy in stratifying chemotherapy toxicity risk. This model directly informs personalized prophylactic strategies, warranting prospective multicenter validation.","PeriodicalId":13734,"journal":{"name":"Integrative Cancer Therapies","volume":"25 ","pages":"15347354261447851"},"PeriodicalIF":3.1000,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13332284/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Integrative Cancer Therapies","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1177/15347354261447851","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2026/7/3 0:00:00","PubModel":"Epub","JCR":"Q2","JCRName":"INTEGRATIVE & COMPLEMENTARY MEDICINE","Score":null,"Total":0}

引用次数: 0

Abstract

Background: To develop and validate a clinically applicable nomogram (NUTRISCORE) predicting severe neutropenia (Grade 3/4) across cycles 2 to 8 of adjuvant chemotherapy in gastric cancer.

Methods: This retrospective cohort study analyzed 391 patients receiving postoperative adjuvant chemotherapy (CAPEOX/XELOX/SOX) following curative laparoscopic D2 gastrectomy. Cohorts were stratified into derivation (n = 318; 1518 cycles) and validation (n = 73; 356 cycles) sets. Generalized Estimating Equations (GEE) were used to account for within-patient correlation of multiple cycles. Multivariable logistic regression identified independent predictors, with subsequent nomogram construction using R. Model performance was quantified through AUC-ROC and decision curve analysis (DCA).

Results: Four predictors emerged: age > 60 years (OR = 2.1, P < .01), prealbumin ≤ 180 mg/L (OR = 3.4, P = .002), PNI < 45 (OR = 2.8, P = .008), and prior cycle neutropenia (OR =4.7, P < .001). The nomogram achieved AUCs of 0.825 (95% CI 0.789-0.860) and 0.810 (0.783-0.835) in derivation and validation cohorts, with DCA confirming clinical utility across threshold probabilities (4%-65% in derivation, 3%-63% in validation), supporting its use for guiding G-CSF prophylaxis at clinically relevant risk thresholds (>15%).

Conclusions: We establish a nutrition-centric predictive tool for severe neutropenia, demonstrating robust accuracy in stratifying chemotherapy toxicity risk. This model directly informs personalized prophylactic strategies, warranting prospective multicenter validation.

查看原文本刊更多论文

NUTRISCORE：一种新的基于营养的Nomogram预测胃癌辅助化疗后严重中性粒细胞减少症。

背景：开发并验证一种临床适用的nomogram (NUTRISCORE)，用于预测胃癌辅助化疗第2 ~ 8周期的严重中性粒细胞减少症（3/4级）。方法：本回顾性队列研究分析391例根治性腹腔镜D2胃切除术后接受术后辅助化疗（CAPEOX/XELOX/SOX）的患者。将队列分层分为衍生组（n = 318, 1518个周期）和验证组（n = 73, 356个周期）。使用广义估计方程（GEE）来解释多个周期的患者内相关性。多变量逻辑回归识别独立预测因子，随后使用r构建nomogram。通过AUC-ROC和决策曲线分析（decision curve analysis， DCA）对模型性能进行量化。结果：出现了4个预测因子：年龄0 ~ 60岁（OR = 2.1, P = 0.05）。002), p =。008)和前周期中性粒细胞减少（OR =4.7, P = 15%）。结论：我们建立了一个以营养为中心的严重中性粒细胞减少症预测工具，在化疗毒性风险分层方面显示出强大的准确性。该模型直接告知个性化预防策略，保证前瞻性多中心验证。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Integrative Cancer Therapies 医学-全科医学与补充医学

CiteScore

4.80

自引率

3.40%

发文量

审稿时长

>12 weeks

期刊介绍： ICT is the first journal to spearhead and focus on a new and growing movement in cancer treatment. The journal emphasizes scientific understanding of alternative medicine and traditional medicine therapies, and their responsible integration with conventional health care. Integrative care includes therapeutic interventions in diet, lifestyle, exercise, stress care, and nutritional supplements, as well as experimental vaccines, chrono-chemotherapy, and other advanced treatments. Contributors are leading oncologists, researchers, nurses, and health-care professionals.