Emmanuel S. Antonarakis, MD, discusses the prevalence of mismatch repair and homologous recombination deficiency genes and how they may play a role in determining a patient’s treatment plan in addition to promising agents that have been introduced to the field in clinical trials.
Emmanuel S. Antonarakis, MD
Pivotal updates surrounding genetic testing have been incorporated into the National Comprehensive Cancer Network (NCCN) guidelines in prostate cancer, with a focus on germline testing in patients with metastatic or high-risk localized nonmetastatic disease and somatic testing in those with lymph node or distant metastases, explained Emmanuel S. Antonarakis, MD.
In a presentation during the 2019 NCCN Annual Conference, Antonarakis introduced the latest additions to the NCCN guidelines in terms of managing patients with prostate cancer, speaking to the role of genetic testing and how this can help determine treatment decisions.
The genes detected in patients with prostate cancer fall into 2 classes: mismatch repair (MMR) and homologous recombination deficiency (HRD). In some cases, the understanding of these mutations can help in determining the treatment course for a patient, he said, adding that the results can help define whether or not a patient’s family may also be at risk for developing cancer.
“These tests can help to form subsequent therapy if patients develop advanced or metastatic prostate cancer,” said Antonarakis. “They will not necessarily inform the first-line treatment for these patients, but they may inform the second- or third-line treatment.”
In an interview with OncLive, Antonarakis, associate professor of oncology, John Hopkins University, discussed the prevalence of these different genes and how they may play a role in determining a patient’s treatment plan. He also highlights promising agents that have been introduced to the field in clinical trials, including PARP inhibitors olaparib (Lynparza) and rucaparib (Rubraca).Antonarakis: The purpose of my talk this year was to update the audience with the new information about genetics in prostate cancer and how this information helps us to make treatment decisions. There are 2 main types of genetic testing that need to be considered. The first is called germline testing—looking for inherited gene mutations that can predispose patients to getting cancer in the first place, some of which can be used to make therapeutic decisions. The second type of testing is somatic testing, where we are looking for mutations in the tumor specimen itself.
Again, some of those can also be used to make decisions about treatments. The 2 newest things on the guidelines related to that are that germline inherited testing is being recommended for every patient with metastatic prostate cancer, so that’s a broad spectrum, and for every patient with high-risk, localized nonmetastatic prostate cancer. The only people right now that the guidelines are not recommending for germline inherited testing is men who have very low-risk prostate cancer. Secondly, who should we do the somatic testing in? The guidelines say that anyone who has lymph node metastases should also undergo tumor somatic testing.
This leads to the next question of, “How prevalent these genes are in prostate cancer, and secondly, what can you do with this information?” The first thing to address is the prevalence; the most informative type of gene mutations involves 2 classes of genes. The first class is MMR genes and the second class is HRD genes.
MMR genes are the ones that are responsible for Lynch syndrome if inherited, but they can also be acquired. These genes are MSH2, MSH6, MLH1, and PMS2. These are the genes that cause these hypermutated tumors; tumors have hundreds, if not thousands, of mutations. The therapeutic reason that this is important is that in prostate cancer and in other cancer types, if you have an MMR deficiency or mutation, you can receive pembrolizumab (Keytruda), a PD-1 inhibitor, which is now FDA approved for all cancer types that have the MMR gene mutation.
Unfortunately, in prostate cancer, this only represents about 3% to 5% of all advanced prostate cancers. Pembrolizumab can be used and is on the guidelines for prostate cancer; however, it only applies to the 3% to 5% of patients with this MMR mutation.
The second class of mutations, the HRD genes, are the ones that include BRCA2, BRCA1, ATM, and a number of other genes. Collectively, these gene mutations make up between 20% to 25% of all advanced prostate cancers, so these are far more common than the MMR mutations. Even though testing for these HRD genes is recommended on the NCCN prostate guidelines, we do not currently have any FDA-approved therapies specifically targeting those patients. There is a lot of interest in using PARP inhibitors like olaparib or rucaparib, and others for those patients.
However, those drugs remain in clinical trials right now, and the purpose of finding out if a patient has an HRD mutation, such as BRCA2, is so you can refer that patient to 1 of the PARP inhibitor trials. The NCCN guidelines do not endorse the off-label use of a PARP inhibitor at this time for a patient with an HRD-positive prostate cancer.The best data that supports the use of pembrolizumab in patients with MRR comes from a series of studies that started at John Hopkins University. Those studies were published in 2015 in the New England Journal of Medicine and 2016 in Science, and those studies showed that for the first time, irrespective of what type of cancer you have, whether it’s colorectal cancer, prostate cancer, or breast cancer, if you have an MMR mutation, those patients have an extremely high chance of responding to pembrolizumab. The response rates are roughly in the 50% range, so just because you have an MMR mutation, it doesn’t necessarily mean or guarantee that you will respond to pembrolizumab, but it greatly increases the chance.
In prostate cancer more specifically, we have done large studies using pembrolizumab in unselected patients with castration-resistant prostate cancer. The largest of them was a study called the KEYNOTE-199 trial, using pembrolizumab in 258 patients with metastatic castration-resistant prostate cancer. In that study, the objective response rate for single-agent pembrolizumab was in the order of 4% to 5%.
However, a more recent single-center experience published by researchers at Memorial Sloan Kettering Cancer Center showed that out of 11 patients with prostate cancer with MMR mutations, approximately half of them had an objective tumor response or a prostate-specific antigen response. This does suggest that if a PD-1 inhibitor is used in the MMR deficient prostate cancer population, it could increase that patient’s chance of responding from about 5% overall to maybe 50% in the context of an MMR mutation.One of the most exciting things in the management of prostate cancer is the investigational use of PARP inhibitors. PARP inhibitors have been approved now for a few other cancer types by the FDA, but not for prostate cancer. The two agents that have received the most attention, which are being studied in the phase III setting, are olaparib and rucaparib. What we are beginning to learn from these phase II studies that have been partially presented or published is that it doesn’t seem to be enough to have a mutation in 1 of the HRD genes. In fact, the actual gene that is mutated seems to matter.
There are some mutated genes, such as BRCA2, in which the chance of responding to olaparib or rucaparib, is quite high—in the 50% to 70% range—but there are other genes that were also previously considered to be HRD genes, the notable example being the ATM gene where we are seeing very few responses to olaparib and rucaparib.
Another gene that is relatively common is called CHEK2. We used to think that patients with CHEK2 mutations might respond to PARP inhibitors in the context of prostate cancer, and again response rates in those mutations are also quite low. The emerging data are pointing to the fact that a patient with BRCA1/BRCA2-mutated prostate cancer will probably have a very good chance of response to a PARP inhibitor—maybe 50% to 70% chance—but perhaps some of the other genes like ATM, CHEK2, and other more rare ones may not respond as well.We are still struggling to answer the question of the germline mutation versus a somatic mutation. For example, if you sequence a patient’s tumor and you find a BRCA2 mutation, there are 2 ways that a mutation can get there. One, the patient can get the mutation from the parent, and it is passed from the germline DNA to every cell in that patient’s body—including the tumor cell. However, the second way is that the mutation can spontaneously arise after the patient’s birth, only in the tumor cell. That is called the somatic-only mutation. One of the things that we don’t yet know is whether, for example, a germline BRCA2 mutation will have a higher response to a PARP inhibitor in prostate cancer compared with a somatic-only mutation. Not only does the gene itself matter, but also whether the germline tumor DNA only may be an additional factor that might predict sensitivity or perhaps resistance to PARP inhibitors.The take-home message from my talk is that both inherited and tumor genetic testing has entered the primetime clinical care for patients with prostate cancer. These tests are now commercially available, FDA approved in many cases, and reimbursed by insurance companies, so patients and physicians should not be scared to order these tests. Most of the time, especially in the Medicare population, these tests are fully reimbursed.
The final take-home message is that if one of these genes is found to be mutated in the tumor, you need to take a saliva or blood sample to determine whether this mutation was inherited or not; this can have profound implications for the patient’s first-degree relatives—siblings, children, or parents. If you begin by doing tumor sequencing and find a BRCA2 mutation, it’s paramount to determine if that BRCA2 mutation was inherited or not because if it was inherited, that patient’s family should be counseled about their risk of developing other cancers.