Scholarlink

Purpose <p>To investigate risk factors for cancer therapy-related cardiac dysfunction (CTRCD) in patients receiving anthracycline-based chemotherapy and develop a predictive model for assessing the risk of CTRCD.</p> Materials and methods <p>Patients with breast cancer or lymphoma who underwent chemotherapy were retrospectively enrolled. Independent risk factors were identified for CTRCD, and a clinical prediction model (nomogram) was constructed and evaluated.</p> Results <p>324 patients were randomly divided into the training and validation cohorts, and patients in the training cohort were further divided into the CTRCD positive (<i>n</i> = 48) and negative cohort (<i>n</i> = 222). Body mass index (BMI) ≥ 25, brain natriuretic peptide (BNP) acuity 52 pg/ml, pathologic stage, low density lipoprotein (LDL) or greater tendency for 3.37 / L, combined HER-2 class drugs, and cumulative dose of anthracycline of 550&#xa0;mg /m<sup>2</sup> were independent risk factors for CTRCD. The C-index of the CTRCD prediction map was 0.872 (95% confidence interval or CI: 0.821–0.923) in the training cohort, similar to that of 0.886 (95% CI: 0.799–0.974) in the validation cohort. The cut-off value of the receiver operating characteristics (ROC) curve analysis in the training cohort was 0.164. Good agreement was shown in the predicted data of the training group (<i>P</i> = 0.837) and the verification group (<i>P</i> = 0.700). The observed probability predicting CTRCD was close to the actual probability. The decision curve analysis (DCA) showed that the constructed nomogram had a high clinical application prospect.</p> Conclusion <p>BMI ≥ 25, history of diabetes, BNP ≥ 52 pg/ml, LDL ≥ 3.37 mmol/L, pathological stage, and cumulative anthracycline dose ≥ 550&#xa0;mg/m² were independent risk factors for CTRCD. The nomogram model demonstrates excellent predictive accuracy, with a C-index of 0.872 in the training cohort and 0.886 in the validation cohort, offering a reliable tool for identifying high-risk for CTRCD.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Risk factors and prediction models for cardiotoxicity induced by anthracyclines in malignant chemotherapy

  • 期刊论文

  • 发表日期：

    2025年7月7日

查看全文
  • Pengxiang Zhang,
  • Yan Zhang,
  • Zheng Xue,
  • Dongchao Liu,
  • Dongliang Li,
  • Bing Duan,
  • Meina Zhao,
  • Liang Yin,
  • Hongjie Gao,
  • Bulang Gao,
  • Jie Mi

摘要

Purpose

To investigate risk factors for cancer therapy-related cardiac dysfunction (CTRCD) in patients receiving anthracycline-based chemotherapy and develop a predictive model for assessing the risk of CTRCD.

Materials and methods

Patients with breast cancer or lymphoma who underwent chemotherapy were retrospectively enrolled. Independent risk factors were identified for CTRCD, and a clinical prediction model (nomogram) was constructed and evaluated.

Results

324 patients were randomly divided into the training and validation cohorts, and patients in the training cohort were further divided into the CTRCD positive (n = 48) and negative cohort (n = 222). Body mass index (BMI) ≥ 25, brain natriuretic peptide (BNP) acuity 52 pg/ml, pathologic stage, low density lipoprotein (LDL) or greater tendency for 3.37 / L, combined HER-2 class drugs, and cumulative dose of anthracycline of 550 mg /m2 were independent risk factors for CTRCD. The C-index of the CTRCD prediction map was 0.872 (95% confidence interval or CI: 0.821–0.923) in the training cohort, similar to that of 0.886 (95% CI: 0.799–0.974) in the validation cohort. The cut-off value of the receiver operating characteristics (ROC) curve analysis in the training cohort was 0.164. Good agreement was shown in the predicted data of the training group (P = 0.837) and the verification group (P = 0.700). The observed probability predicting CTRCD was close to the actual probability. The decision curve analysis (DCA) showed that the constructed nomogram had a high clinical application prospect.

Conclusion

BMI ≥ 25, history of diabetes, BNP ≥ 52 pg/ml, LDL ≥ 3.37 mmol/L, pathological stage, and cumulative anthracycline dose ≥ 550 mg/m² were independent risk factors for CTRCD. The nomogram model demonstrates excellent predictive accuracy, with a C-index of 0.872 in the training cohort and 0.886 in the validation cohort, offering a reliable tool for identifying high-risk for CTRCD.