Machine learning prediction of anemia events before and after talazoparib dose modification using TALAPRO-2 phase 3 trial data

Study type

• Study topic:
  Human medicinal product
• Study type:
  Non-interventional study
• Scope of the study:
  Drug utilisation

  Method development or testing
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Medicinal product name:
  TALZENNA
• Study drug International non-proprietary name (INN) or common name:
  TALAZOPARIB
• Anatomical Therapeutic Chemical (ATC) code:
  (L01XK04) talazoparib

  talazoparib
• Medical condition to be studied:
  Anaemia

Population studied

• Short description of the study population:
  TALAPRO-2 trial data of metastatic castrate-resistant prostate cancer (mCRPC) patients allocated to the talazoparib + enzalutamide

Study design details

• Study design:
  NI, observational, retrospective cohort study of mCRPC patients who received talazoparib in the treatment arm
• Main study objective:
  Primary objectives: ; 1. Describe patient characteristics, dosing patterns and hemoglobin trajectories with respect to anemia events in talazoparib-treated metastatic castrate-resistant prostate cancer (mCRPC) patients from the TALAPRO-2 trial data. ; 2. Develop and evaluate machine learning prediction models of anemia risk after treatment initiation and preceding dose modification. ; 3. Develop and evaluate machine learning prediction models of anemia recovery after dose modification. ; Exploratory objective: ; 1. Describe patient characteristics and hemoglobin trajectories with respect to anemia events in control-arm mCRPC patients from the TALAPRO-2 trial data
• Data analysis plan:
  For Primary Objective 1, descriptive statistics will be used to summarize baseline patient characteristics (demographic, clinical, treatment history, lab values) overall and stratified by subsequent anemia status; to summarize early dose patterns with respect to initial post-index anemia status and dose modification; to summarize baseline patient characteristics and subsequent anemia status by early dose pattern strata; to summarize the proportion of patients who improved, remained the same, or worsened in anemia severity following dose modification; and to summarize the rate of hemoglobin change prior to and after dose modification. Time to dose modification, time to anemia, and time to anemia improvement will be assessed using Kaplan-Meier survival analyses. For Objectives 2 and 3, a variety of machine learning models including LASSO Cox regression, ; extreme gradient boosting, survival support vector machine, and deep cox proportional hazards model will be applied to identify the best performing model for predicting anemia risk and recovery based on performance metrics such as concordance index (test c-index) and dynamic area under the curves (AUCs). Exploratory objective 1 will be assessed similarly as Objective 1, with descriptive statistics to be used to summarize baseline patient characteristics (demographic, clinical, treatment history, lab values) overall and stratified by subsequent anemia status; and to summarize the rate of hemoglobin change after study initiation. Time to anemia will be assessed using Kaplan-Meier survival analyses. No dose modification-related concepts will be assessed for ; exploratory objective 1 since dosing is not applicable for the control arm.