Abstract 1400
Background
Predictive modeling of CIN/FN prophylaxis with Zarzio®, a biosimilar filgrastim, in the pan-European MONITOR-GCSF study identified several risk factors at the patient level (“ever” during chemotherapy) and cycle level (during cycle) of 5 outcomes: CIN grade 4, FN, CIN/FN-related hospitalizations or chemotherapy disturbances, and “worst case” composite (Aapro et al, Ann Oncol 2016). We used these modeling results as proof-of-concept data for the development of a calculator of patient and cycle level risk for these outcomes.
Methods
Using the 2016 modeling generated odds ratios (OR), we specified an analytics algorithm to calculate probabilities for best to worst scenarios by 1) taking the OR natural logarithm, 2) applying log odds to the predictors, 3) summing intercept and log odds to obtain total log odds, 4) then converted to total OR, 5) to compute outcome probability, and 6) to estimate best to worst scenarios. Best scenario assumed absence of all negative and presence of all positive predictors; and vice versa for worst scenario.
Results
Patient level: risk of “ever” experiencing any of 5 outcomes ranged from
Conclusions
This analysis based on the proof-of-concept data have yielded an intriguing model. The statistical analytics and the results thus generated show the feasibility of estimating the risk of CIN/FN episodes, CIN/FN-related hospitalizations and chemotherapy disturbances at the start (“ever”) of a line of chemotherapy, and at start of each cycle. These risk calculations are based on real-world evidence, enhancing external validity, and conform to recent ESMO guidelines for G-CSF usage (Klastersky et al, Ann Oncol 2016), assuring internal validity. A risk calculator for use in daily clinical practice is being developed and will be submitted to validation studies.
Clinical trial identification
N/A
Legal entity responsible for the study
Sandoz Biopharmaceuticals
Funding
Sandoz Biopharmaceuticals
Disclosure
C. Bokemeyer: Hexal AG (personal fees). M. Aapro: Amgen (honoraria, speakers bureau, expert testimony), Helsinn Healthcare (advisory role, speakers bureau, research funding), Hospira (advisory role, speakers bureau, research funding), Teva (advisory role, speakers bureau), Merck KGaA (advisory role), Merck (advisory role), Sandoz (advisory role, speakers bureau, research funding), Pierre Fabre Medicament (advisory role, speakers bureau, research funding), Vifor Pharma (advisory role, speakers bureau), Tesaro (advisory role, speakers bureau), Novartis (speakers bureau, research funding), Roche (speakers bureau), Johnson & Johnson (speakers bureau). H. Ludwig: Celgene (speaker’s honoraria), Takeda (research funding, speaker’s honoraria), Amgen (honoraria for advisory boards and speakers board). P. Gascón: Sandoz (speakers bureau). M. Boccadoro: Celgene (personal fees), Novartis (personal fees), Amgen (personal fees), Sanofi (personal fees). K. Denhaerynck: Matrix45 (employment). Matrix45 has been under contract for scientific consulting services to GCSF manufacturers including Hospira, Hexal/Sandoz and Therapeutic Proteins International. By company policy, employees cannot hold equity in sponsor organizations and cannot receive direct personal benefits, financial or other, from sponsor organizations. A. Krendyukov: Hexal AG (employment). K. Macdonald, I. Abraham: Matrix45 (employment). Matrix45 is under contract with study Sponsor, Hexal/Sandoz, for scientific support services including study design, statistical analysis and manuscript preparation for the MONITOR-GCSF study. Matrix45 has been under contract for scientific consulting services to other GCSF manufacturers including Hospira. By company policy, employees cannot hold equity in sponsor organizations and cannot receive direct personal benefits, financial or other, from sponsor organizations. Matrix45 provides similar services to other biopharmaceutical companies without exclusivity constraints.