GIInger predicts homologous recombination deficiency and patient response to PARPi treatment from shallow genomic profiles
Christian Pozzorini, Gregoire Andre, Tommaso Coletta, Adrien Buisson, Jonathan Bieler, Loïc Ferrer, Rieke Kempfer, Pierre Saintigny, Alexandre Harlé, Davide Vacirca, Massimo Barberis, Pauline Gilson, Cristin Roma, Alexandra Saitta, Ewan St. John Smith, Floriane Consales Barras, Lucia Ripol, Martin Fritzsche, Ana Claudia Marques, Amjad Alkodsi, Ray M. Marín, Nicola Normanno, Christoph Grimm, Leonhard Müllauer, Philipp Harter, Sandro Pignata, Antonio González-Martı́n, Ursula Denison, Keiichi Fujiwara, Ignace Vergote, Nicoletta Colombo, Adrian Willig, Éric Pujade-Lauraine, Pierre‐Alexandre Just, Isabelle Ray‐Coquard, Zhenyu Xu
Abstract
Homologous recombination deficiency (HRD) is a predictive biomarker for poly(ADP-ribose) polymerase 1 inhibitor (PARPi) sensitivity. Routine HRD testing relies on identifying BRCA mutations, but additional HRD-positive patients can be identified by measuring genomic instability (GI), a consequence of HRD. However, the cost and complexity of available solutions hamper GI testing. We introduce a deep learning framework, GIInger, that identifies GI from HRD-induced scarring observed in low-pass whole-genome sequencing data. GIInger seamlessly integrates into standard BRCA testing workflows and yields reproducible results concordant with a reference method in a multisite study of 327 ovarian cancer samples. Applied to a BRCA wild-type enriched subgroup of 195 PAOLA-1 clinical trial patients, GIInger identified HRD-positive patients who experienced significantly extended progression-free survival when treated with PARPi. GIInger is, therefore, a cost-effective and easy-to-implement method for accurately stratifying patients with ovarian cancer for first-line PARPi treatment.