PARP Inhibitor Resistance: Closing a Door, Opening a Window

Recently, those of us who treat prostate cancer have become comfortable with testing for Breast Cancer Gene 2 (BRCA2) and acting on the results. It’s now time we took a slightly deeper dive into this gene and how knowing more can help us counsel our patients better and design better clinical trials.

With two poly ADP ribose polymerase (PARP) inhibitors on the market, we expect to see many patients on these therapies in the coming years, most likely with great benefit for a substantial proportion of them, prolonging life and delaying or preventing further misery from this disease. I I have commented before on our need to study the patients with primary resistance to PARP inhibitors and suggested that these may be individuals with complex tumors and other key driving mutations that render intervention with a PARP inhibitor to be too little, or too late.

The other side of the coin is the patient who is treated with a PARP inhibitor, benefits, and then develops acquired resistance. A few years ago observations trickled in regarding the presence of “reversion mutations” in PARP inhibitor-treated prostate and ovarian cancer. This observation was as intellectually fascinating as it was clinically devastating: BRCA2 mutant prostate cancers, when treated with PARP inhibitors, can ‘fix’ their own mutation when another mutation occurs that results in an intact functioning BRCA2 gene and resistance to PARP inhibitors. Hence the term ‘reversion’. In genetic and functional terms, these reversion mutations signal restored homologous recombination.

It’s a classic case of treatment mediated selection pressure, and a key paradigm of how we should think about the battle against the disease – tumors with drug sensitivity based on mutation may develop drug resistance based on mutation. We see this elsewhere in oncology, most notably in the treatment of Epidermal Growth Factor Receptor (EGFR) mutated lung cancer.

Thus far our knowledge of reversion mutations has come through relatively small-scale analyses, either small case series or even case reports. Now we have some additional detail on the reversion mutation story that might actually help us in the clinic and avoid the sense of exasperation that we feel when our precision-guided, personalized and targeted therapy fails. Stephen Pettitt and colleagues, based at the Institute for Cancer Research in London, did some heavy lifting by aggregating 300 cases of Breast Cancer Gene 1 (BRCA1) and BRCA2 mutated cancers that were treated with PARP inhibitors and developed resistance, and provide us with more detail on how this discovery can be used therapeutically.1

Let’s focus on BRCA2. It is a big gene with approximately 8,000 codons. Exactly where the mutation occurs in that gene will govern the risk for cancer in the first place, and the likelihood of the development of reversion mutations. Interestingly, where a mutation occurs on BRCA2 also governs the type of cancer it may be associated with.2  For example, there is an ovarian cancer cluster region (OCCR), but mutations in the OCCR are far less common in patients with prostate or breast cancer. Similarly, two potential prostate cancer cluster regions (PCCR) from codon 7339 to the 3’ end of the gene or from c6373 to 6492 have been identified. 3,4 Not all of the ovarian cancers occur in the OCCR, but they cluster there, and the same occurs for prostate cancer and the PCCR.

Likewise, it turns out there is a reversion mutation ‘cluster region’ as well. The clinical implication is that we might be able to identify or even predict by a patients original mutation, whether he is at risk to develop a reversion mutation. Pettitt’s paper suggests that the cluster region for reversion mutations is at c750-775, at a very early point in the gene. Reversions typically occur, for obvious reasons, very close to the original mutation that is driving the disease.

If not all BRCA mutations carry the same risk for the development of reversion mutations, these data may be actionable. If we could identify which patients with BRCA alterations are likely to develop PARP inhibitor resistance through reversion, we could target them with alternate clinical approaches whether that be intermittent PARP inhibitor therapy, ATR inhibitors, platinum or other targeted approaches.

So, as we move forward, we may need to consider not only IF a patient has a BRCA mutation but WHERE they have a BRCA mutation. This will require greater curation of our genomic reports from the various companies and institutions who provide them and will require potentially further educational efforts from clinicians like myself. I look forward to learning and thinking about this problem.

But the paper from Pettitt and colleagues suggests to us that the closing of the door might lead to the opening of a window; while PARP inhibitor-based therapies can propagate mutations that later cause their own futility, a better understanding of the products of PARP-resistant reversion mutations could lead to new clinical approaches, in particular immunotherapies. 

In certain cases, the development of the reversion mutations, while it drives resistance to PARP inhibitors, is an epiphenomenon of a more large-scale genomic process that involves far more than this one important gene. As it turns out, concurrent with the development of the reversion mutations in the development of multiple other ‘smaller’ mutations that may drive the expression of spontaneous neo-antigens. Basically, when the BRCA gene ‘re-mutates’ to fix its dysfunction, it doesn’t create a normal-looking protein, but rather one with ‘out of frame sequences’ which can create small peptide appendages. Given their novelty, these out of frame sequences are unlikely to have been previously seen by the immune system and may act as neoantigens. Whether or not that observation is enough to determine if an immune response can be generated is unknown, but it is intriguing to think that there may be a glimmer of hope for a potential therapeutic angle for these patients with PARP inhibitor refractory disease.

Written by: Charles Ryan, MD, B.J. Kennedy Chair in Clinical Medical Oncology, Director and Professor of Medicine, Division of Hematology, Oncology and Transplantation, University of Minnesota, Minneapolis, Minnesota

  1. Pettitt, S.J., et al., Clinical BRCA1/2 reversion analysis identifies hotspot mutations and predicted neoantigens associated with therapy resistance. Cancer Discov. 2020 Jul 22;CD-19-1485. doi: 10.1158/2159-8290.CD-19-1485
  2. Patel, V.L., et al., Association of Genomic Domains in BRCA1 and BRCA2 with Prostate Cancer Risk and Aggressiveness. Cancer Res, 2020. 80(3): p. 624-638. doi: 10.1158/0008-5472.CAN-19-1840.
  3. Nielsen, H.R., et al., BRCA1/BRCA2 founder mutations and cancer risks: impact in the western Danish population. Fam Cancer, 2016. 15(4): p. 507-12. doi: 10.1007/s10689-016-9875-7.
  4. Nyberg, T., et al., Prostate Cancer Risk by BRCA2 Genomic Regions. Eur Urol. 2020 Jan;77(1):24-35. doi: 10.1016/j.eururo.2019.08.025
Published Date:  September 2020

Related Content:
Clinical Understanding of PARP Inhibitors in Prostate Cancer - Elena Castro, Wassim Abida, and Karen Knudsen
DNA Repair and PARP Inhibitor Therapy in Prostate Cancer - Veda Giri, Patrick Pilié, and Arpit Rao
Integrating Germline and Somatic Mutations Tests into Clinical Practice Workflows - Wassim Abida
PARP Inhibitors, Prostate Cancer and a Promise Fulfilled
Molecular Diagnostics and Genetic Testing in Genitourinary Cancers - Colin Pritchard and Heather Cheng