More Clinical Trials Needed to Overcome PARP Inhibitor Resistance in Ovarian Cancer

Gottfried E. Konecny, MD

While PARP inhibitors have played a large part in improving progression-free survival (PFS) in patients with ovarian cancer, long-term use of these agents often leads to resistance that are often quite challenging to overcome, according to Gottfried E. Konecny, MD.

“We need to do studies that do biospecimen collection to analyze the reasons for the lack of efficacy of a PARP inhibitor but also to make intelligent decisions on how to treat them subsequently,” explained Konecny, an associate professor of medicine and lead clinician for gynecologic oncology in the Department of Medicine at the University of California, Los Angeles.

Several trials have come along examining the use of PARP inhibitors in the frontline treatment of recurrent ovarian cancer, but data are more limited when it comes to retreatment with PARP or what to do in those who develop acquired resistance to these agents. While there currently is no optimal strategy to overcome PARP resistance, methods such as repairing double-strand breaks by recovering homologous recombination proficiency, utilizing antiangiogenic agents, such as cediranib or bevacizumab (Avastin), and combining PARP inhibitors with immunotherapy are all possible options, added Konecny.

Future studies with biomarker research are additionally needed to overcome this unmet need.

“The bottom line is that there are more possible drug combinations,” said Konecny. “It is an effort of our laboratory to study these drug interactions to quantify the level of interaction and to follow leads that show pronounced synergy when combined with a PARP inhibitor and bring these to clinic as fast as possible, with the hope that [we] can overcome PARP resistance or restore the sensitivity to the drug.”​

In an interview with OncLive® during the 2020 Institutional Perspectives in Cancer webinar on ovarian cancer, Konecny further discussed potential strategies to overcome PARP resistance, future combinations with PARP inhibitors, and other unmet needs in ovarian cancer.

OncLive®: What are the biggest challenges with PARP inhibitor resistance?

Konecny: The addition of PARP inhibitors to the treatment of patients with early-stage ovarian cancer and recurrent disease has made a very large impact on improving PFS. Recent data from the 2020 ESMO Virtual Congress showed an improvement in overall survival (OS), particularly in patients with BRCA1/2 mutations. Despite these successes, unfortunately, many patients face [progression on] these drugs over time. That leads us to focus on mechanisms of drug resistance and understanding how we can overcome these mechanisms of drug resistance.

About half of all ovarian cancers have an intrinsic primary drug resistance to PARP inhibitors, as they have a competent high-fidelity double-strand break repair mechanism that's intact. Single-strand break repair disruption through a PARP inhibitor is not that relevant in these cases because the backup DNA repair mechanism steps in and the drug does not work through synthetic lethality in these patients.

Of those patients who are initially deficient of double-strand break repair, we now understand that many patients can develop resistance by a recovery of double-strand break repair. That means recovery of homologous recombination, and this is due to demethylation of 1 of the silenced BRCA1 genes. It has also been described by an increase in copy numbers that occur during disease progression of either mutated or unmutated BRCA allele, which leads to an increased expression of the protein. Other mechanisms are secondary mutations that occur, particularly in those who have mutations in BRCA1/2.

[There] are very unique findings on how tumors can adapt to PARP inhibitor treatment by selecting cells that actually have the ability to splice out mutations or areas of the gene that contained the mutations and express an incomplete but partially functional BRCA1 or BRCA2 protein. This is basically alternative splicing of a hypomorphic BRCA1; that's particularly pertaining to exon 11, so if mutations are in exon 11, it can be that some of these cell clones splice out the mutation and they have restoration of homologous recombination.

Then, there is a large area where negative regulators of homologous recombination can be lost, and there is loss of negative regulators are basically involving the proteins 53BP1 or REV7. Beyond that, PARP inhibitors have shown to be effective by leading to the replication for degradation; PARP inhibitors also function not just through enzymatic inhibition of PARylationand recruiting DNA repair factors, but they blocks the replication fork and leads to its degradation.

There are a number of proteins that execute this degradation, and when you lose these executers that are supposed to complete the command of destroying the replication, then PTIP, MRE11, or EZH2 can also be lost. Therefore, a PARP inhibitor loses its ability to degrade the replication fork and gets less effective. Lastly, there are mechanisms, such as drug efflux pumps that just simply decrease the intracellular drug concentrations by pumping the drug out efficiently through the well-described P-glycoprotein pump, which is also responsible for chemotherapy resistance in many instances.

Currently, we don't have a good understanding of how clinically relevant these described mechanisms are. Many of them have been described in cell lines or preclinical models, and to date, it has only been demethylation of silence BRCA promoters over time, as well as increased copy numbers of the BRCA gene mutated or wild-type and secondary mutations that have been described in clinical samples. The remaining reasons for PARP resistance are preclinical. We need to do a better job in looking at samples and finding whether these mechanisms truly play an important clinical role.

What is the clinical relevance of reversing mutations?

Reversing mutations has been [found in] about 20% of patients who develop resistance to a PARP inhibitor. How do patients do after the occurrence of a reversion mutation? Some studies suggest that they do not respond to PARP inhibitors. A retrospective study from ARIEL2 showed cases that had a reversion mutation in the screening biopsy and responded poorly to the PARP inhibitor.

However, there are contrasting studies showing that if you have resistance to a PARP inhibitor, you still have a sensitivity towards platinum-based chemotherapy. A British study looked at the efficacy of chemotherapy after PARP inhibitor–failure. It shows an objective response rate of 45% and 49% if they were retreated with a platinum-[containing agent], suggesting that even with resistance mechanisms to a PARP inhibitor, they may still respond to a platinum-[containing agent].

That leads to the question: Are reversion mutations that clinically relevant? They should confer resistance to [a PARP inhibitor], but also platinum-based chemotherapy. Possibly, [if] you detect a reversion mutation, it may be pretty infrequent in the tumor. Some studies have shown that you can find a number of reversion mutations with different allele frequencies. That means they are only present in a small subsets of cells, indicating that this is a clonal selection process. You develop a reversion mutation and then it takes time until this population outgrows the otherwise sensitive cell lines. We don't understand what to make out of reversion mutations. We need more clinical studies to understand that.

How can these resistance mechanisms be overcome?

There are a number of theoretical approaches to overcome these resistance mechanisms and most of them pertain to mechanisms that are not geared toward overcoming genomic resistance, such as secondary mutations or a copy number changes. They’re geared towards affecting reversible recovery of homologous recombination. There are preclinical studies that have shown that when you add a MEK inhibitor, PI3K inhibitor, or epigenetic therapy to a PARP inhibitor that you can actually increase the DNA repair deficiency by downregulating the expression of proteins involved in double-strand break repair. You're actually switching a cell that is proficient in homologous recombination back into one that is homologous repair deficient. (HRD) This can be done by antiangiogenic agents. By simply inducing hypoxia in the tumor microenvironment, some studies have shown that that decreases the expression of proteins involved in double-strand break repair, and increases sensitivity towards PARP inhibitors again.

These are all hypotheses of ongoing clinical studies. Looking at combinations of various PARP inhibitors with either olaparib (Lynparza), niraparib (Zejula), rucaparib (Rubraca), or talazoparib (Talzenna), you can group them into 3 main groups. First, all of those drugs presumably increase DNA repair deficiency; then, there's another strategy that uses a group of drugs that just simply increases the amount of DNA damage. These are mostly inhibitors that interfere with remaining repair pathways. One important remaining pathway functions later in the cell cycle; there are numerous inhibitors in this pathway that may increase the activity of a single-agent PARP inhibitor. The Wee1, ATR, and ATM inhibitors are currently all in clinical trials.

Another strategy is to combine PARP with chemotherapy to overwhelm the DNA repair system. Most of these studies have to use low-dose chemotherapy because of the potentiation of cytotoxicity or myelotoxicity [GM5] with the combination. The last area of combination studies to overcome PARP inhibitor resistance or diminished activity is combining it with checkpoint inhibitors to leverage pathway interactions. Combining a checkpoint inhibitor with a PARP inhibitor doesn't restore or increase DNA damage or decrease the ability of DNA damage repair. It just means that those who are characterized by double-strand break repair deficiency may have some characteristics that suggest that they respond better to immunotherapy.

Could you expand on the rationale to combine PARP inhibitors with immunotherapy?

It makes sense to combine both approaches, and the data that have been published regarding that are threefold. When you have increased double-strand break repair, you have increased double-strand breaks. The cell recognizes this as a double-stranded viral DNA fragment, which induces an immune response mechanism, which triggers a signaling pathway called STING that then leads to an upregulation of interferon signaling and activation of immune cells. That is a well-described mechanism, suggesting that cells with increased HRD may respond better to immunotherapy. In addition, studies show that tumors with double-strand break repair deficiency, or those that are BRCA1/2 mutated, have higher neoantigen loads. A significantly higher neoantigen load has been associated with an improved response to immune therapy. There's great promise in the combination of PARP inhibitors with immune checkpoint inhibitors but, again, these studies are in early-phase testing.

Some examples are the MEDIOLA study, which showed outstanding activity with an over 70% response rate when combining olaparib with durvalumab (Imfinzi). Likewise, a similar study combining niraparib with pembrolizumab (Keytruda) showed promising activity; however, this was in predominantly BRCA wild-type tumors. These studies are leading the way and are the rationale for a number of frontline studies that are comparing PARP plus checkpoint inhibitors, such as ATHENA study or DUO-O.

If a patient has a certain type of resistance, is it too early in practice to direct them towards a chemotherapy/PARP combination versus an immunotherapy combination?

The only clinically useful way of detecting resistance markers right now are the reversion mutations. [This is] because we send specimens for sequencing and they suddenly reveal additional mutations in some alleles, or differences from the initial sequencing report. This can be done on tumor tissues and can now be done on circulating tumor DNA.

This is clinically available and it's still unclear what we should do as clinicians. Should we withhold the PARP inhibitor? Possibly or very likely. Should we withhold a platinum-based chemotherapy? We don't know that yet because there may not be a complete overlap in the resistance. These are questions that we have to address.

If [a patient] received a PARP inhibitor and if they have been on a PARP inhibitor for more than 6 months, it's worth trying retreatment with PARP as maintenance therapy. But, again, these are not unfounded suggestions; [they’re] not backed by clinical studies. These clinical studies are currently ongoing and a number of studies going forward, specifically in randomizing patients who were on a PARP maintenance therapy, have progressed, and have platinum-sensitive disease [are looking at] another line of platinum-based chemotherapy, and then rerandomizing them to either a PARP inhibitor again or a combination. For example, [this could be] a PARP inhibitor with another other cell cycle checkpoint inhibitor or placebo.