Synthetic Lethality of PARP Inhibitors in Prostate Cancer

Today we are speaking with Dr. Emmanuel Antonarakis, MD, a medical oncologist and associate professor of oncology at Johns Hopkins University who specializes in the treatment of genitourinary cancers, including prostate, bladder, and testicular cancers. We discuss his novel clinical trial of the PARP inhibitor olaparib in men with high-risk and biochemically recurrent prostate cancer who have previously undergone a radical prostatectomy.

-Interviewed by Anna Azvolinsky

Cancer Network: First, can you define what a PARP inhibitor is and why this class of drugs is being tested in patients with prostate cancer in general? Are there existing data [showing] PARP inhibitors [are effective against] certain types of prostate cancer?

Dr. Antonarakis: PARP inhibitors work on the theory that cancer cells accumulate DNA damage as they are dividing. There are two ways that a cancer cell can fix DNA damage. The first is called base excision repair. This is a way that the cancer cell can fix single-stranded DNA damage and the main protein, an enzyme, is called the PARP, for poly (ADP-ribose) polymerase, family of proteins. The second way that a cancer cell can fix DNA damage is called homologous repair, and that primarily fixes double-stranded breaks.

The main proteins involved in this pathway are the BRCA1 [breast cancer type 1 susceptibility protein] and BRCA2 proteins as well as the ATM [ataxia-telangiectasia mutated] protein. If a cell doesn’t have the ability to fix a double-stranded DNA break (a form of DNA damage) because it has a mutation in a gene that codes for BRCA1, BRCA2, or ATM, then its homologous repair function is compromised, and the cell can rescue or recover from the unrepaired DNA damage by using the PARP enzyme as a second option to fix the DNA breaks, so it can divide and not die. The PARP inhibitors work based on a concept called synthetic lethality. Synthetic lethality means you are now using a drug that inhibits the cancer cell’s other mechanisms by which it can repair its DNA damage when one form of DNA repair is already compromised.

By blocking the PARP activity in a cancer cell with a BRCA1, BRCA2, or ATM mutation, you are blocking the ability of the cancer cell to fix its DNA damage, which causes catastrophic DNA damage to accumulate in that cell, and therefore that cell has no choice but to die.

The concept is called synthetic lethality, and the goal is to block every which way that the cancer cell can fix its DNA damage-both because the cell has a BRCA1, BRCA2, or ATM mutation and secondly [because] adding a PARP inhibitor blocks another mechanism by which the cancer cell could repair its DNA damage. In prostate cancer, there has been a lot of excitement [about] using PARP inhibitors in castration-resistant patients.

The first study that led to this excitement was a trial called the TOPARP-A study, which was done in the United Kingdom by Joaquin Mateo, MD, and Johann S. de Bono, PhD. It was published in the New England Journal of Medicine more than 2 years ago.

In that study, [Mateo and de Bono] used an oral PARP inhibitor called olaparib. It was given to 49 patients with metastatic castration-resistant prostate cancer, and the investigators tried to determine what genetic mutations in the tumor cells would predict a favorable response to olaparib versus a lack of a response to the drug. The first interesting observation the authors made was that 16 out of the 49 patients (33%) had a mutation in one of homologous repair genes; the most common ones [were] in BRCA1, BRCA2, or ATM. In patients who had one of these so-called homologous repair gene mutations in their tumors, 88% had a response to olaparib, which was a quite high response rate. In the remaining [patients] who did not have a homologous repair mutation, such as one in BRCA1, BRCA2, or ATM, the response rate was only 6%.

That was the first trial that [found] PARP inhibitors, and olaparib specifically, [to be associated with] an 88% chance of producing a clinical response if a mutation in a homologous repair gene (such as in one of the BRCA genes) was present-but that the success rate [was] much lower, only about 6%, if one of those mutations was not present. This led to a lot interest in developing PARP inhibitors for prostate cancer, and has led to multiple other trials being conducted right now.

Cancer Network: Can you tell us about the design of this study you are leading at Johns Hopkins?

Dr. Antonarakis: We designed a trial, which is currently enrolling patients, focusing on a very early prostate cancer disease state called the biochemical recurrence, which is a prostate-specific antigen (PSA) level that is rising after a patient has had local therapy in the form of radical prostatectomy or radiotherapy, [but] there is no evidence of metastases on conventional scans while the PSA level is rising.

Our trial [is evaluating] 50 men [who have] a biochemical recurrence without metastatic disease [and] a PSA level that is rising and has to be above 1.0 ng/dL. [Also], the PSA doubling time has to be less than 6 months. Patients who have a biochemical recurrence with a PSA doubling time of less than 6 months typically develop metastases after a median of 2 years, and typically [die] from prostate cancer after a median of 6 years. This is an aggressive risk group even though these patients do not yet have metastatic disease.

We decided to do a molecularly unselected trial. What that means is that we chose patients without a specific DNA biomarker, such as a mutation. We decided to open this trial up to 50 patients regardless of the genetic status of their tumors. What we are doing is giving the men olaparib, a PARP inhibitor that was used in the TOPARP-A trial, and we are giving the drug as a monotherapy without androgen deprivation. Our goal here is to see if olaparib will work in these patients without a concurrent use of hormonal therapy.

The primary endpoint of the trial is to see what percentage of patients achieve a PSA50 response, which means a 50% reduction in PSA. We designed the trial seeking [a response rate of] at least 30%, so we want to see at least 30% of the 50 patients achieving a PSA response of 50% of more. The trial is designed as two stages. In the first stage, we enroll 20 patients, and if we see 3 or more PSA responses in these patients, we will expand the trial to stage II. If we see two or fewer responses, we will terminate the trial. In the second stage, if we get there, we will enroll the full 50 patients and want to see at least 15 patients of the 50 achieving a PSA response.

Cancer Network: What [are] the clinical and biological questions that you are testing, and [are] there preclinical data to support this trial?

Dr. Antonarakis: The clinical questions we are seeking to answer [are]: Can olaparib be used as a noncastrating therapy for biochemical recurrence? And if we do see responses, which patients benefit the most? The second clinical question is: Will olaparib be safe and tolerable in the biochemically recurrent population?

The biological question we are trying to answer is: What genomic mutation status predicts a favorable response to olaparib in the biochemically recurrent patient population? We are also trying to learn what transcriptome signature-in other words, what RNA transcription signature-predicts a favorable response to olaparib.

The interesting thing about this trial is that we are mandating that all patients have available tissue from their radical prostatectomy, in order for us to molecularly characterize [not only] the DNA mutations but also the RNA transcription profiles of all 50 patients.

Although we are selecting patients without requiring any particular genetic or transcriptomic inclusion criteria, we do want to have the ability to retrospectively analyze the DNA and RNA from the prostatectomy specimens of all 50 patients. The eligibility criterion that is very important to us is that patients must have archival tumor tissue available for these analyses. Interestingly, there is not much preclinical data [on the use of] olaparib without hormone therapy. Most of the olaparib preclinical and clinical data have focused on the castration-resistant prostate cancer population, and in those patients olaparib is always given in conjunction with androgen deprivation therapy (and this hormone therapy is continued and the olaparib is added on).

For our trial, we decided that the best way to test the hypothesis of whether PARP inhibition can be used alone, without hormonal therapy, would be to go directly into a human trial-which, on the one hand may seem risky because the human data are limited, but we have also seen, anecdotally, in a number of patients, that responses can occur with PARP inhibitors without the use of hormonal therapy. This encouraged us to design this trial to more conclusively address this question. This will be the first data on PARP inhibitors without hormonal therapy in prostate cancer. 

Cancer Network: That sounds like this makes this trial unique in the context of PARP inhibitor trials for prostate cancer. Is there anything else that makes this trial a unique prostate cancer trial?

Dr. Antonarakis: There are currently at least 15 clinical trials using various PARP inhibitors in advanced prostate cancer. The vast majority are focusing on men [who] have metastatic prostate cancer and specifically, those [who] have castration-resistant prostate cancer (refractory to hormones).

This trial is unique in the sense that we are looking for nonmetastatic patients, those [who] have a biochemical recurrence. We are attempting to treat patients with hormone-sensitive prostate cancer, but without hormone therapy. Also, the vast majority of PARP inhibitor trials in men with prostate cancer are currently requiring molecular selection based on the presence of a BRCA1, BRCA2, or ATM mutation, or a number of other gene mutations that are involved in the homologous repair pathway.

This trial is also unique because this is a molecularly unselected trial in which we allow patients to enroll and then retrospectively determine whether there is a particular molecular signature that is linked to a response or not.

We are also planning a second study called the TRIUMPH trial, and [it] will be even more unique and ambitious in that we are going one step further and testing a PARP inhibitor, in this case, rucaparib, in patients with metastatic hormone-sensitive prostate cancer. In TRIUMPH, patients will get rucaparib as a monotherapy without androgen deprivation. The only difference is that for this TRIUMPH study, patients need to have a germline (in other words, inherited), mutation in one of the homologous repair genes such as BRCA1, BRCA2  or ATM. The question for this trial is whether it is possible, in a patient with metastatic prostate cancer, [to] treat a subset of these patients without requiring androgen deprivation therapy at all.

If those study results show promise, that would be a bold new step in trying to define and discover nonhormonal therapies to treat, and hopefully cure, metastatic hormone sensitive prostate cancer. So, we are excited about both studies.