Lucy Hutchinson

Title

Optimization of biopsy scheduling in oncology clinical studies through an integrated modeling, simulation and digital pathology approach

Abstract

In the context of cancer immunotherapy clinical trials, baseline and on-treatment tumour biopsies may provide important insight into whether a treatment is working as expected, and furthermore whether efficacy is anticipated. For tumour-retained antibodies that perturb the behaviour of immune cells, such as T cell bispecific antibodies (TCBs), information derived from biopsy images may be particularly insightful. On-treatment biopsies in clinical trials are usually scheduled at a time point that is considered convenient for the study design and when therapeutic effects, such as T cell infiltration, are expected to be distinguishable in tumour tissue. Our integrated workflow is designed to select the time point at which on-treatment biopsies could be most informative. Leveraging baseline and on-treatment digitized biopsy images from patients undergoing treatment with immune stimulating TCBs, we train an agent-based model to simulate tumour cell/T cell interactions in the tumor microenvironment. The model produces an enriched dataset of “virtual” biopsy images corresponding to predictions at intermediate time points. The virtual biopsies are evaluated based on their ability to predict treatment response. 

Short bio 

As a mathematical modeller in Clinical Pharmacology at Roche, Lucy provides modelling support to inform the dose and schedule of oncology molecules in early-phase clinical trials. Since completing her D.Phil. on angiogenesis modelling in 2017, Lucy has moved into the area of cancer immunotherapy, and is now co-supervising a D.Phil. student with Philip Maini and Helen Byrne. Lucy and her teammates, Roisin Stephens (WCMB) and Oliver Grimm (Roche), have recently secured funding to further develop their model via an internal competition at Roche.