首页|University of Texas MD Anderson Cancer Center Reports Findings in Personalized M edicine (Personalized Composite Dosimetric Score- Based Machine Learning Model of Severe Radiation-Induced Lymphopenia among Esophageal Cancer Patients)
University of Texas MD Anderson Cancer Center Reports Findings in Personalized M edicine (Personalized Composite Dosimetric Score- Based Machine Learning Model of Severe Radiation-Induced Lymphopenia among Esophageal Cancer Patients)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Drugs and Therapies - Personalized Medicine is the subject of a report. According to news originating from Houston, Texas, by NewsRx correspondents, research stated, “Radiation-induc ed lymphopenia (RIL) is common among patients undergoing radiotherapy (RT), and severe RIL has been linked with adverse outcomes. The severity and risk of RIL c an be predicted from baseline clinical characteristics and dosimetric parameters .” Our news journalists obtained a quote from the research from the University of T exas MD Anderson Cancer Center, “However, dose-volume (DV) indices are highly co rrelated with one another and are only weakly associated with RIL. Here we intro duce the novel concept of ‘composite dosimetric score’ (CDS) as the index that c ondenses the dose distribution in immune tissues of interest to study the dosime tric dependence of RIL. We derived an improved multivariate classification schem e for risk of grade 4 (G4) RIL, based on this novel RT dosimetric feature, for p atients receiving chemoRT for esophageal cancer. DV indices were extracted for 7 34 patients who received chemoRT for biopsy-proven esophageal cancer. Non-negati ve matrix factorization was used to project the DV indices of lung, heart, and s pleen into a single CDS; XGBoost was employed to explore significant interaction s among predictors; and logistic regression was applied to combine the resultant CDS along with baseline clinical factors and interaction terms to facilitate in dividualized prediction of immunotoxicity. Five-fold cross-validation was applie d to evaluate the model performance. The CDS for selected immune organs at risk (OARs, i.e., heart, lung, and spleen) (1.791, 95 CI [1.350,2. 377]) was a statistically significant risk determinant for G4 RIL. Pearson correlation coefficients for CDS vs. G4RIL risk for individual immu ne OARs were greater than any single DV indices. Personalized prediction of G4RI L based on CDS and 4 clinical risk factors yielded an area under the curve value of 0.78. Interaction between age and CDS revealed that G4RIL risk increased mor e sharply with increasing CDS for patients 65. Risk of immunotoxicity for patien ts undergoing chemoRT for esophageal cancer can be predicted by CDS. The CDS con cept can be extended to immunotoxicity in other cancer types and in dose-respons e models currently based on DV indices.”
HoustonTexasUnited StatesNorth and Central AmericaCancerCyborgsDrugs and TherapiesEmerging TechnologiesE sophageal CancerGastroenterologyHealth and MedicineHematologic Diseases an d ConditionsHemic and Lymphatic Diseases and ConditionsImmune System Disease s and ConditionsImmunologic Deficiency SyndromesLeukocyte DisordersLeukope niaLymphopeniaMachine LearningOncologyPersonalized MedicinePersonalize d TherapyRisk and Prevention
NETL
NSTL
