RNASEH2B loss and PARP inhibition in advanced prostate cancer
Juliet Carmichael, … , Adam Sharp, Johann de Bono
Juliet Carmichael, … , Adam Sharp, Johann de Bono
Published June 4, 2024
Citation Information: J Clin Invest. 2024;134(21):e178278. https://doi.org/10.1172/JCI178278.
View: Text | PDF
Clinical Research and Public Health Oncology

RNASEH2B loss and PARP inhibition in advanced prostate cancer

Abstract

BACKGROUND Clinical trials have suggested antitumor activity from PARP inhibition beyond homologous recombination deficiency (HRD). RNASEH2B loss is unrelated to HRD and preclinically sensitizes to PARP inhibition. The current study reports on RNASEH2B protein loss in advanced prostate cancer and its association with RB1 protein loss, clinical outcome, and clonal dynamics during treatment with PARP inhibition in a prospective clinical trial.METHODS Whole tumor biopsies from multiple cohorts of patients with advanced prostate cancer were interrogated using whole-exome sequencing (WES), RNA-Seq (bulk and single nucleus), and IHC for RNASEH2B and RB1. Biopsies from patients treated with olaparib in the TOPARP-A and TOPARP-B clinical trials were used to evaluate RNASEH2B clonal selection during olaparib treatment.RESULTS Shallow codeletion of RNASEH2B and adjacent RB1 — colocated at chromosome 13q14 — was common, deep codeletion infrequent, and gene loss associated with lower mRNA expression. In castration-resistant prostate cancer (CRPC) biopsies, RNASEH2B and RB1 mRNA expression correlated, but single nucleus RNA-Seq indicated discordant loss of expression. IHC studies showed that loss of the 2 proteins often occurred independently, arguably due to stochastic second allele loss. Pre- and posttreatment metastatic CRPC (mCRPC) biopsy studies from BRCA1/2 WT tumors, treated on the TOPARP phase II trial, indicated that olaparib eradicated RNASEH2B-loss tumor subclones.CONCLUSION PARP inhibition may benefit men suffering from mCRPC by eradicating tumor subclones with RNASEH2B loss.TRIAL REGISTRATION Clinicaltrials.gov NCT01682772.FUNDING AstraZeneca; Cancer Research UK; Medical Research Council; Cancer Research UK; Prostate Cancer UK; Movember Foundation; Prostate Cancer Foundation.

Authors

Juliet Carmichael, Ines Figueiredo, Bora Gurel, Nick Beije, Wei Yuan, Jan Rekowski, George Seed, Suzanne Carreira, Claudia Bertan, Maria de Los Dolores Fenor de La Maza, Khobe Chandran, Antje Neeb, Jon Welti, Lewis Gallagher, Denisa Bogdan, Mateus Crespo, Ruth Riisnaes, Ana Ferreira, Susana Miranda, Jinqiu Lu, Michael M. Shen, Emma Hall, Nuria Porta, Daniel Westaby, Christina Guo, Rafael Grochot, Christopher J. Lord, Joaquin Mateo, Adam Sharp, Johann de Bono

×

Figure 4

Evaluation of RB1 and RNASEH2B protein expression at CRPC by IHC.

Options: View larger image (or click on image) Download as PowerPoint
Evaluation of RB1 and RNASEH2B protein expression at CRPC by IHC. (A) Gr...
(A) Graphical representation of RB1 and RNASEH2B protein expression in 93 CRPC biopsies (HALO-generated H Score, OD and % negative cells) and intrasample heterogeneity, quantified by Shannon’s diversity index, across biopsy sites. Samples are matched, displayed in order of increasing RNASEH2B nuclear H score for both plots. (B) Scatter plot showing association between RNASEH2B and RB1 IHC quantification by HALO-generated % negative cells and OD. Scatterplots on the right distribute samples according to biopsy site, in bone marrow alone and nonbone marrow (soft tissue, liver, lymph node, prostate) samples. r and P values were calculated using Spearman correlation. (C) Representative micrographs of RB1 and RNASEH2B detection by IHC in matched, same-patient CRPC biopsies. Examples of concordant RNASEH2B and RB1 expression (case 1), heterogeneous loss of both RB1 and RNASEH2B (case 2), RB1 loss alone (case 3), RB1 loss with heterogeneous RNASEH2B (case 4) and RNASEH2B loss alone (case 5) at various biopsy sites are shown. IHC depicted here; magnification × 10; scale bar: 100 μm. While in a whole biopsy RB1 and RNASEH2B protein loss correlate, with both proteins being commonly heterogenously lost, surprisingly the data indicate that different cells in a biopsy often lose one protein or the other with only a minority of cells having coloss of both proteins. IHC, immunohistochemistry; IQR, interquartile range; CRPC, castration resistant prostate cancer; OD, optical density.