US-NJ, 2018 Internship Program - Biostatistic Intern 234774BR-MZ
2018 Internship Program - Biostatistic Intern
Global Drug Development
East Hanover, NJ
USA Novartis Pharmaceuticals Corporation, East Hanover, NJ
Research & Development
You are applying to be part of the Novartis Talent Pool. We are not currently recruiting for this role but we are building a pipeline for future opportunities. If you would like to be considered for a similar position in future, then please submit your CV.
Large confirmatory phase III studies with time-to-event endpoints are regularly designed with exponential survival model under proportional hazard assumptions and log-rank test for simplicity. However, it is well-recognized that in adjuvant studies and immuno-oncology studies, delayed treatment effect and/or non-proportional hazard situation can occur due to time varying treatment effect associated with the underlying disease and the mechanism of action of the drug. This has been widely observed in the therapeutic areas, such as adjuvant lung cancer, adjuvant breast cancer and immuno-oncology (IO).
Design of such studies while properly accounting for delayed treatment effect is critical. The specification of the delay time and the test statistics has to be pre-specified in such designs. The use of weighted log rank tests have been proposed with pre-specified weights in this setting. But, the impact of the delay time misspecification at the design stage when the log rank test or the weighted log rank test with pre-specified weights are used, is not well understood. Moreover, with the delayed treatment effect interim analysis (IA) timing and futility/efficacy boundaries derived based on proportional hazard model may be inaccurate, and may even result in substantial increase in false negative and false positive rate.
The intern is expected to perform extensive simulations and explore internal existing data, e.g. past adjuvant/IO trial data, to identify appropriate models for analyzing time-to-event data under delayed treatment effect and/or non-proportional assumptions, including cure rate model with delayed effect, weighted log rank tests and landmark analysis. Subsequently, impact of delay time misspecification, optimal timing and boundaries for IA can be investigated and recommendations will be made for designing further adjuvant and IO studies.
The Novartis Group of Companies are Equal Opportunity Employers and take pride in maintaining a diverse environment. We do not discriminate in recruitment, hiring, training, promotion or any other employment practices for reasons of race, color, religion, gender, national origin, age, sexual orientation, marital or veteran status, disability, or any other legally protected status.
-Strong knowledge in theoretical statistics and hands-on experience in computational statistics. The intern is expected to work independently and have excellent presentation skills, including using graphics.
-Proficiency in R is required.
-Good background in survival analysis, and basic knowledge in drug development, clinical trial and group sequential design.
* If a direct employer requests that you go to their web site and complete your application there in order to be
considered, please do so. Applications for all positions are subject to each employer's specific requirements.