Will the use of PARP inhibitors in advanced ovarian cancer with wild-type BRCA lead to improved OS rates?
Click here to read the Cover Story, “PARP inhibitors take big step toward changing practice.”
Significant survival benefits can be hidden in the noise of any clinical trial. Part of this is due to the success we’ve had recently in ovarian cancer. Our patients on average live a long time, and we expand that with every new drug.
OS benefit is harder to prove in BRCA-nonmutated disease, but if it works in BRCA-mutated cancer and almost as well in HRD-positive patients, then it could work for these women.
The problem we have in trying to prove that these will extend OS is that you have to show how long the women live on average after they are treated. If you have a 3-month improvement in PFS, that translates into a 3-month improvement in OS and if patients live 24 months after the trial starts, you would need 3,000 women to have an 80% chance of showing a survival benefit of that magnitude.
Looking at the history of ovarian cancer, we’ve had only one trial — GOG-0213 — that’s shown a survival advantage and met a primary endpoint of OS, just barely, with the addition of bevacizumab to paclitaxel/carboplatin.
In the PRIMA study, when researchers stopped niraparib for BRCA- and HRD-positive women, the Kaplan-Meier curves stayed separated and there was a tail. Even among HRD-proficient women in the PRIMA study, there was a tail on the curve. It was not a large tail, but it was there. Over placebo, 15% of those women on the tail will be treated and continue to be progression-free.
We know that the earlier effective treatments are given, the better off patients are. Now we have drugs that — unlike chemotherapy drugs — have HRs that are unbelievable, under the 0.5 range. They are superior to anything we’ve ever had.
I also believe PARP inhibitors could be used as a monotherapy in some situations, particularly considering the SOLO3 data. I don’t know that it will be the primary therapy, and I doubt PARP inhibitors will be standard over chemotherapy in the upfront setting. In the recurrent setting, however, there are many situations where PARP inhibitors can outperform chemotherapy.
Coleman R, et al. J Clin Oncol. 2019;doi:10.1200/JCO.2018.36.15_suppl.5501.
González-Martín A, et al. N Engl J Med. 2019;doi:10.1056/NEJMoa1910962.
Lowe ES, et al. J Clin Oncol. 2019;doi:10.1200/JCO.2016.34.15_suppl.TPS5598.
R. Wendel Naumann, MD, is professor and associate director of gynecologic oncology at Levine Cancer Institute at Atrium Health. He can be reached at email@example.com. Disclosure: Naumann reports consultant/advisory roles with AstraZeneca, Clovis Oncology, Eisai, Genentech, Merck, OncoMed, Sutro Bioharma and Tesaro and research funding to his institution from BioSutro, Bristol-Myers Squibb and OncoMed.
I am very impressed with the PFS in all three studies presented at ESMO, especially for women with HRD-positive tumors. However, it is too early to make any conclusions about OS.
There have been trials in the ovarian cancer maintenance setting with impressive PFS but for which OS did not pan out, such as the pazopanib (Votrient, Novartis) maintenance trial. It is difficult to change OS in ovarian cancer because many women have a long OS and receive multiple agents during their disease trajectory. Even before 2001, the 5-year survival rates were 33% to 47% for stage III disease and 19% for stage IV disease. Thus, we have long-term survivors, and many of them — if not all of them — receive PARP inhibitors over time for advanced disease. The results of clinical trials presented at ESMO beg the question of how use of PARP inhibitors in the maintenance setting after adjuvant therapy would impact the downstream effect of PARP inhibitors for recurrent disease.
It’s also possible that when the data mature, we will see OS differences among women with HRD-positive vs. -negative tumors. Looking at the Kaplan-Meier curve in the PAOLA trial, it appears that the PFS difference for olaparib vs. olaparib and bevacizumab decreases after 24 months. It would be very interesting to look at the PFS2 data when they become mature.
Although all three trials showed that PARP inhibitors were well-tolerated, I have concerns about late toxicities, especially the risk for myelodysplastic syndrome and acute myelogenous leukemia, which have been observed in 1% to 4% of patients exposed to PARP inhibitors. It is unclear how late these toxicities can occur, but they can occur 2 years after exposure and can be fatal, which is always tragic in a patient who is cured of her cancer.
I am also interested in seeing clinical trials combining immunotherapy and PARP inhibitors for newly diagnosed ovarian cancer. There are ongoing studies looking at neoadjuvant, adjuvant and maintenance use of these combinations that likely have to be redesigned to eliminate the placebo arms for maintenance based on these new data presented at ESMO.
Lastly, previous data have shown that the best predictor of OS is the amount of residual disease after surgery. All our improvements in the adjuvant and first-line maintenance setting, including the use of PARP inhibitors, should not result in decreased surgical effort to achieve no residual disease at the time of surgery completion.
Ray-Coquard IL, et al. Abstract LBA2_PR. Presented at: European Society for Medical Oncology Congress; Sept. 27-Oct. 1, 2019; Barcelona, Spain.
Heintz A, et al. Int J Gynaecol Obstet. 2006;doi:10.1016/S0020-7292(06)60033-7.
Tsibulak I, et al. Crit Rev Oncol Hematol. 2019;doi:10.1016/j.critrevonc.2019.08.002.
Vergote I, et al. Abstract 5518. Presented at: ASCO Annual Meeting; June 1-5, 2018; Chicago.
Lydia Usha, MD, is associate professor of medicine at Rush University and director of Rush Inherited Susceptibility to Cancer Center. She can be reached at firstname.lastname@example.org. Disclosure: Usha reports no relevant financial disclosures.