Robert T Novo, Samantha M Thomas, Michel G Khouri, Fawaz Alenezi, James E Herndon, Meghan Michalski, Kereshmeh Collins, Tormod Nilsen, Elisabeth Edvardsen, Lee W Jones, Jessica M Scott
{"title":"机器学习驱动的表型分组与转移性乳腺癌的心肺功能反应","authors":"Robert T Novo, Samantha M Thomas, Michel G Khouri, Fawaz Alenezi, James E Herndon, Meghan Michalski, Kereshmeh Collins, Tormod Nilsen, Elisabeth Edvardsen, Lee W Jones, Jessica M Scott","doi":"10.1200/CCI.24.00031","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>The magnitude of cardiorespiratory fitness (CRF) impairment during anticancer treatment and CRF response to aerobic exercise training (AT) are highly variable. The aim of this ancillary analysis was to leverage machine learning approaches to identify patients at high risk of impaired CRF and poor CRF response to AT.</p><p><strong>Methods: </strong>We evaluated heterogeneity in CRF among 64 women with metastatic breast cancer randomly assigned to 12 weeks of highly structured AT (n = 33) or control (n = 31). Unsupervised hierarchical cluster analyses were used to identify representative variables from multidimensional prerandomization (baseline) data, and to categorize patients into mutually exclusive subgroups (ie, phenogroups). Logistic and linear regression evaluated the association between phenogroups and impaired CRF (ie, ≤16 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup>) and CRF response.</p><p><strong>Results: </strong>Baseline CRF ranged from 10.2 to 38.8 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup>; CRF response ranged from -15.7 to 4.1 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup>. Of the n = 120 candidate baseline variables, n = 32 representative variables were identified. Patients were categorized into two phenogroups. Compared with phenogroup 1 (n = 27), phenogroup 2 (n = 37) contained a higher number of patients with none or >three lines of previous anticancer therapy for metastatic disease and had lower resting left ventricular systolic and diastolic function, cardiac output reserve, hematocrit, lymphocyte count, patient-reported outcomes, and CRF (<i>P</i> < .05) at baseline. Among patients allocated to AT (phenogroup 1, n = 12; 44%; phenogroup 2, n = 21; 57%), CRF response (-1.94 ± 3.80 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup> <i>v</i> 0.70 ± 2.22 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup>) was blunted in phenogroup 2 compared with phenogroup 1.</p><p><strong>Conclusion: </strong>Phenotypic clustering identified two subgroups with unique baseline characteristics and CRF outcomes. The identification of CRF phenogroups could help improve cardiovascular risk stratification and guide investigation of targeted exercise interventions among patients with cancer.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":null,"pages":null},"PeriodicalIF":3.3000,"publicationDate":"2024-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11407741/pdf/","citationCount":"0","resultStr":"{\"title\":\"Machine Learning-Driven Phenogrouping and Cardiorespiratory Fitness Response in Metastatic Breast Cancer.\",\"authors\":\"Robert T Novo, Samantha M Thomas, Michel G Khouri, Fawaz Alenezi, James E Herndon, Meghan Michalski, Kereshmeh Collins, Tormod Nilsen, Elisabeth Edvardsen, Lee W Jones, Jessica M Scott\",\"doi\":\"10.1200/CCI.24.00031\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Purpose: </strong>The magnitude of cardiorespiratory fitness (CRF) impairment during anticancer treatment and CRF response to aerobic exercise training (AT) are highly variable. The aim of this ancillary analysis was to leverage machine learning approaches to identify patients at high risk of impaired CRF and poor CRF response to AT.</p><p><strong>Methods: </strong>We evaluated heterogeneity in CRF among 64 women with metastatic breast cancer randomly assigned to 12 weeks of highly structured AT (n = 33) or control (n = 31). Unsupervised hierarchical cluster analyses were used to identify representative variables from multidimensional prerandomization (baseline) data, and to categorize patients into mutually exclusive subgroups (ie, phenogroups). Logistic and linear regression evaluated the association between phenogroups and impaired CRF (ie, ≤16 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup>) and CRF response.</p><p><strong>Results: </strong>Baseline CRF ranged from 10.2 to 38.8 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup>; CRF response ranged from -15.7 to 4.1 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup>. Of the n = 120 candidate baseline variables, n = 32 representative variables were identified. Patients were categorized into two phenogroups. Compared with phenogroup 1 (n = 27), phenogroup 2 (n = 37) contained a higher number of patients with none or >three lines of previous anticancer therapy for metastatic disease and had lower resting left ventricular systolic and diastolic function, cardiac output reserve, hematocrit, lymphocyte count, patient-reported outcomes, and CRF (<i>P</i> < .05) at baseline. Among patients allocated to AT (phenogroup 1, n = 12; 44%; phenogroup 2, n = 21; 57%), CRF response (-1.94 ± 3.80 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup> <i>v</i> 0.70 ± 2.22 mL O<sub>2</sub>·kg<sup>-1</sup>·min<sup>-1</sup>) was blunted in phenogroup 2 compared with phenogroup 1.</p><p><strong>Conclusion: </strong>Phenotypic clustering identified two subgroups with unique baseline characteristics and CRF outcomes. The identification of CRF phenogroups could help improve cardiovascular risk stratification and guide investigation of targeted exercise interventions among patients with cancer.</p>\",\"PeriodicalId\":51626,\"journal\":{\"name\":\"JCO Clinical Cancer Informatics\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.3000,\"publicationDate\":\"2024-09-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11407741/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"JCO Clinical Cancer Informatics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1200/CCI.24.00031\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ONCOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"JCO Clinical Cancer Informatics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1200/CCI.24.00031","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ONCOLOGY","Score":null,"Total":0}
引用次数: 0
摘要
目的:抗癌治疗期间心肺功能(CRF)受损的程度以及CRF对有氧运动训练(AT)的反应存在很大差异。本辅助分析的目的是利用机器学习方法来识别CRF受损和CRF对有氧运动训练反应不佳的高风险患者:我们评估了 64 名转移性乳腺癌女性患者 CRF 的异质性,她们被随机分配到为期 12 周的高度结构化 AT(33 人)或对照组(31 人)。我们使用无监督分层聚类分析从随机化前(基线)的多维数据中识别出代表性变量,并将患者分为相互排斥的亚组(即表型组)。逻辑回归和线性回归评估了表型组与受损的CRF(即≤16 mL O2-kg-1-min-1)和CRF反应之间的关联:基线 CRF 为 10.2 至 38.8 mL O2-kg-1-min-1;CRF 反应为 -15.7 至 4.1 mL O2-kg-1-min-1。在 n = 120 个候选基线变量中,确定了 n = 32 个代表性变量。患者被分为两个表型组。与表型组 1(n = 27)相比,表型组 2(n = 37)中既往未接受过转移性疾病抗癌治疗或抗癌治疗次数大于 3 次的患者人数较多,且基线时静息左心室收缩和舒张功能、心输出量储备、血细胞比容、淋巴细胞计数、患者报告结果和 CRF 均较低(P < .05)。在分配到 AT 的患者中(表型组 1,n = 12;44%;表型组 2,n = 21;57%),与表型组 1 相比,表型组 2 的 CRF 反应(-1.94 ± 3.80 mL O2-kg-1-min-1 v 0.70 ± 2.22 mL O2-kg-1-min-1 )减弱:表型聚类确定了两个具有独特基线特征和 CRF 结果的亚组。确定 CRF 表型组有助于改善心血管风险分层,并指导对癌症患者进行有针对性的运动干预研究。
Machine Learning-Driven Phenogrouping and Cardiorespiratory Fitness Response in Metastatic Breast Cancer.
Purpose: The magnitude of cardiorespiratory fitness (CRF) impairment during anticancer treatment and CRF response to aerobic exercise training (AT) are highly variable. The aim of this ancillary analysis was to leverage machine learning approaches to identify patients at high risk of impaired CRF and poor CRF response to AT.
Methods: We evaluated heterogeneity in CRF among 64 women with metastatic breast cancer randomly assigned to 12 weeks of highly structured AT (n = 33) or control (n = 31). Unsupervised hierarchical cluster analyses were used to identify representative variables from multidimensional prerandomization (baseline) data, and to categorize patients into mutually exclusive subgroups (ie, phenogroups). Logistic and linear regression evaluated the association between phenogroups and impaired CRF (ie, ≤16 mL O2·kg-1·min-1) and CRF response.
Results: Baseline CRF ranged from 10.2 to 38.8 mL O2·kg-1·min-1; CRF response ranged from -15.7 to 4.1 mL O2·kg-1·min-1. Of the n = 120 candidate baseline variables, n = 32 representative variables were identified. Patients were categorized into two phenogroups. Compared with phenogroup 1 (n = 27), phenogroup 2 (n = 37) contained a higher number of patients with none or >three lines of previous anticancer therapy for metastatic disease and had lower resting left ventricular systolic and diastolic function, cardiac output reserve, hematocrit, lymphocyte count, patient-reported outcomes, and CRF (P < .05) at baseline. Among patients allocated to AT (phenogroup 1, n = 12; 44%; phenogroup 2, n = 21; 57%), CRF response (-1.94 ± 3.80 mL O2·kg-1·min-1v 0.70 ± 2.22 mL O2·kg-1·min-1) was blunted in phenogroup 2 compared with phenogroup 1.
Conclusion: Phenotypic clustering identified two subgroups with unique baseline characteristics and CRF outcomes. The identification of CRF phenogroups could help improve cardiovascular risk stratification and guide investigation of targeted exercise interventions among patients with cancer.