– PARIS

In high-grade serous ovarian cancer (HGSOC) there is an urgent medical need for effective treatment options for the patients with a limited response to both standard chemotherapy and emerging PARP inhibitors. Apart from the mutations in DNA repair pathways, targetable point mutations in tumour suppressors are uncommon in HGSOC tumours. Instead, patients present unique and highly variable combinations of copy number aberrations, structural variations and additional subclonal changes, significantly complicating classification and targeting of intratumoral chemoresistance-associated heterogeneity across patients. Thus, innovative personalised approaches are required to help these patients.

To overcome treatment resistance in HGSOC, we propose to implement a novel concept: to unbiasedly detect the distinct tumour subpopulations that confer resistance to current therapies, and then to predict and test which drugs they respond to. We propose to implement this concept by combining [i] a novel, unbiased, and robust algorithm to reliably identify the resistant tumour cell subpopulations from single cell transcriptomic data of HGSOC tumours and matched organoids, [ii] drug response prediction and screening in organoid cultures, using single cell readouts, and [iii] validation of the resulting personalised drug predictions in independent samples, in both organoid cultures and in vivo patient-derived xenografts, co-treated with chemotherapy.

Our novel approach enables the design of effective personalised treatments of HGSOC to complement chemotherapy by targeting the specific chemoresistance mechanisms active in each patient’s tumour cell subpopulations. The resulting tools and biomarkers can be used for screening and prioritisation; first, in early-phase patient-centric studies, and then in larger trials. The overall approach can be extended to other diseases where subpopulations provide treatment resistance, towards a paradigm shift in personalised oncology.