Examining the Underpinnings of PARP Inhibition in Recurrent Ovarian Cancer

In Partnership With:

Partner | Cancer Centers | <b>UCLA Health Jonsson Comprehensive Cancer Center</b>

Gottfried E. Konecny, MD, shares the known aspects of PARP inhibition in recurrent ovarian cancer.

Gottfried E. Konecny, MD

Whether germline or somatic, patients with recurrent ovarian cancer who harbor BRCA1/2 mutations have shown a unique sensitivity to PARP inhibition. As the field moves forward, optimization of the 3 FDA-approved PARP inhibitors in this setting will come from learning more about the genetic makeup of extreme responders, the features of BRCAness, and the underpinnings of adaptive resistance mechanisms, said Gottfried E. Konecny, MD.

“Not every BRCA mutation carrier is alike; there is more to learn, but in the next few years we will understand more reasons for the failure of PARP inhibitors or ultimate resistance by collecting clinical data and performing biopsies in patients who recur,” said Konecny. “That research will also provide us with information on how we can continue to best treat a patient with a BRCA mutation or underlying homologous recombination deficiency.”

In an interview during the 2019 OncLive® State of the Science Summit™ on Ovarian Cancer, Konecny, an associate professor of medicine at the University of California, Los Angeles, shared the known aspects of PARP inhibition in recurrent ovarian cancer.

OncLive: What are the indications for the 3 FDA-approved PARP inhibitors in recurrent ovarian cancer?

Konecny: Currently, we have 3 PARP inhibitors approved for the treatment of patients with recurrent ovarian cancer. Olaparib (Lynparza) is approved for use in patients who have had 3 prior lines of chemotherapy as an active treatment, as well as maintenance therapy in those who have had either a partial response (PR) or a complete response (CR) to platinum-based chemotherapy. Likewise, rucaparib (Rubraca) is approved for the treatment of patients who have had at least 2 prior lines of chemotherapy or as maintenance therapy in patients who have had a CR or PR to chemotherapy. Finally, niraparib (Zejula) is approved as maintenance therapy in patients who have had a CR or PR to platinum-based chemotherapy.

We have a new class of drugs available for broad use in recurrent ovarian cancer, either as maintenance therapy in those who have platinum-sensitive disease or as treatment in those who have a germline or somatic mutation or increased deficiency in homologous recombination.

How do you choose among the available PARP inhibitors? Are there certain aspects that make patients amenable to 1 but not another?

With the currently available data, all 3 PARP inhibitors have comparable efficacy. They are distinguishable from each other in terms of their toxicity profiles and in their interactions with other drugs as well as the ease of taking them. Niraparib is dosed once daily, whereas rucaparib and olaparib are dosed twice daily. All 3 agents cause mild to moderate nausea, which is amenable to dose reductions. All 3 PARP inhibitors induce anemia in approximately 20% of patients. Initial hypertension or thrombocytopenia are more often seen with niraparib; however, these adverse events (AEs) are well managed with appropriate dose reductions or treatment intermissions. The major difference is seen in the toxicities of these drugs and differences in dosing schedules.

At a preclinical level, there are differences between PARP inhibitors. Now, we understand that PARP inhibitors are either enzymatic inhibitors of PARP itself or trap the replication fork onto the DNA through a mechanism called PARP trapping. At a preclinical level, there are differences in PARP trapping. It’s said from preclinical studies, for example, that niraparib is a stronger trapper than rucaparib or olaparib would be. Talazoparib (Talzenna), which is not approved for ovarian cancer, but is in breast cancer, is described as the most potent trapper. Whether this ability to trap as opposed to enzymatically inhibit parylation is clinically relevant has not been studied, and we have no clear head-to-head clinical comparisons between the PARP inhibitors. In my opinion, we really don't need that. We shouldn't waste resources in doing head-to-head comparisons in this regard.

Dose modifications with niraparib did not sacrifice outcomes, according to a study presented at the 2019 SGO Annual Meeting. Do the other agents in this class require dose modifications?

The rucaparib data from the ARIEL3 trial show that approximately 50% of patients could not continue treatment on the recommended dose of 600 mg twice daily. With olaparib, the number is somewhat lower. Approximately 25% of patients required dose reductions. There was a higher frequency of dose delays in the olaparib data, but it applies to all 3 PARP inhibitors that dose reductions allow you to minimize the known moderate or severe AEs, such as fatigue, anemia, nausea, or gastrointestinal toxicities.

Patients with BRCA1/2 mutations benefit from PARP inhibitors, but what about those with BRCAness?

BRCAness is a description of tumors that have the same inability to repair DNA double strand breaks, which is a feature of tumors that have either germline or somatic BRCA mutations. BRCA is an essential enzyme in repairing double-strand break repairs. However, BRCA is part of a protein complex, which includes approximately 30 other proteins, and these others can also have mutations such as BRIP1 or PALB2 so that components of the double-strand break repair complex can be affected by other mutations that are, by themselves, less frequent.

However, if you look at the totality of these DNA repair mutations, approximately 50% of ovarian cancers have deficiency in an important player of double-strand break repair. The question is, “Do you detect that by sequencing all potentially relevant genes or do you just screen for the net result of double-strand break repair deficiency by looking at genomic scarring?”

Genomic scarring can be assessed by studying loss of heterozygosity [via] next-generation sequencing assays. By assessing the percentage of DNA that is affected by genomic scarring or DNA repair deficiency, you can draw conclusions on whether the tumor is sensitive to PARP or not.

Have clinical trials examined patients with this BRCA-like signature?

Yes. The best data exist with rucaparib, where the value of BRCAness was prospectively evaluated in the ARIEL2 study. These patients were stratified at baseline. Likewise, in the ARIEL3 study, patients were stratified prior to treatment by either being biomarker-negative or BRCA-like, using Foundation Medicine's assay. An homologous recombination deficiency (HRD) score above 14% [meant] being truly BRCA-mutant, either somatic or germline. The study clearly showed that patients who are BRCA-like have higher responses to single-agent PARP inhibitors compared with those who are truly biomarker-negative. However, the responses were lower compared with those who had somatic or germline mutations. It may not be that strong of a predictor, but it does allow enrichment.

The NOVA study with maintenance niraparib, as well as the ARIEL3 trial, clearly showed that the degree of improvement in progression-free survival (PFS) [with the PARP inhibitors] was most pronounced in those who had germline or somatic mutations. The PFS benefit was highly significant in those who had BRCAness or increased HRD scores, and it was least [pronounced] in those who were biomarker-negative. Biomarker-negative patients did not harbor a germline or somatic mutation, or any other features that led to double-strand break repair deficiency.

You spoke about a patient with metastatic recurrent disease who has been on rucaparib for 6 years during the State of the Science Summit™. Is that response the exception to the rule?

Yes, it is the exception. If we look at published sets, approximately 15% of patients derive a long-term benefit from PARP inhibitors in the recurrent setting. These are patients who failed prior chemotherapy, have metastatic disease burden, and go into remission with a PARP inhibitor. This has been shown for rucaparib as well for olaparib. In fact, patients who have BRCA wild-type disease and other deficiencies in HRD may respond very well to a PARP inhibitor.

We're currently studying the reasons for these extreme responders. It may be that the underlying mutation that predicts response does not allow adaptive responses that are otherwise seen that lead to resistance. Certain BRCA mutations do not allow the occurrence of reversion mutations. There are events in which the other allele has been completely lost, and, therefore, there cannot be epigenetic upregulation of other alleles.

There are some [mutations] that will predict a very pronounced response to PARP inhibition and others where the success of the response is more temporary, because the resistance of adaptive mechanisms may restore double-strand break repair again and decrease the sensitivity or the activity of a PARP inhibitor.

What else is known about resistance to PARP inhibitors?

We're learning more about resistance as we are collecting samples from patients who are now failing treatment to PARP inhibitors that are either being used as a maintenance or as [primary] treatment for recurrent disease. A well-studied mechanism is reversion mutations. Here, secondary mutations occur that restore a reading frame, and a normally truncated BRCA protein is transcribed again and restored. There are very interesting mechanisms of restoring an incomplete BRCA protein. [For example], by losing an alternative splicing site which essentially splices out exon 11 where a truncating mutation might have been, an incomplete protein is restored.

There are studies [that have shown] that the contralateral allele that's not mutated can be epigenetically silenced as a mechanism of inactivating BRCA. If that is re-expressed again, you can have increased expression. There are genetic rearrangements that can restore a missing allele. There are other proteins that are involved in double-strand break repair like BP53, which, when lost, can restore double-strand break repair activity again. There's upregulation of drug efflux pumps. There are multiple mechanisms that can explain why PARP inhibition loses its efficacy, generally summarized in that a tumor basically restores double-strand break repair ability as a backup mechanism. That leads to novel treatment strategies in that combining a PARP inhibitor with a drug that decreases HRD again may actually restore the sensitivity to the PARP inhibitor.

Is there anything else you would like to emphasize?

We know from these retrospective studies that there are subtypes that greatly benefit from [PARP] and then there are so called biomarker-negative platinum-sensitive patients who don't respond to the therapy. The question is, “Should every patient who responds to a platinum agent and experiences a PR or CR be treated with maintenance therapy with a PARP inhibitor?” We have to individualize treatment approaches and weigh the toxicities and the cost of the treatment with the anticipated benefit. There are cases where a 2- to 3-month improvement in PFS may not justify the AEs that patients experience or the cost in some cases.

The question of selecting the patients who derive the greatest benefit will become more important when we move PARP inhibitors into the frontline setting where we don't have the surrogate of platinum sensitivity. Having molecular tests that can guide a clinician to assess the likelihood of a response or the degree of benefit to expect [with an agent] is important, as those are strategies to [use] these drugs in the most meaningful way.