PHILADELPHIA—Two trials reported at the American Association for Cancer Research Annual Meeting featured increasingly promising results for olaparib, a drug that inhibits poly ADP-ribose polymerase. Both trials involved research from the PI3 Kinase Dream Team of Stand Up To Cancer (SU2C).
Men with metastatic, castration-resistant prostate cancer with genomic aberrations in DNA repair function had significantly better responses with single-agent olaparib compared with the responses for patients whose tumors were without those mutations, according to the investigator-initiated Phase II TOPARP study of 49 patients (Abstract CT322).
Still, some men without the mutation also responded to olaparib, calling into question whether that mechanism will be useful as a biomarker for response, the researchers said.
The first author, Joaquin Mateo, MD, MSc, a clinical research fellow in the Prostate Targeted Therapy Group and Drug Development Unit at the Institute for Cancer Research (ICR) and the Royal Marsden NHS Foundation Trust in London, reported that olaparib had a response rate of about 33 percent.
In many men the response lasted more than six months, said Mateo, a doctoral candidate in the laboratory of Johann S. de Bono, MD, PhD, Professor of Experimental Cancer Medicine at ICR.
Mateo said next-generation sequencing detected mutations in genes associated with DNA repair in tumor samples from 16 of the 49 patients evaluated. Of these, 14 had a response to olaparib.
“Metastatic castrate-resistant prostate can harbor genomic defects in DNA repair genes, both somatic and germline, and these results may arguably represent the first molecular treatment stratification of metastatic castration-resistant prostate cancer.”
He said the investigators are now further testing this stratification in the second stage of TOPARP, TOPARP-B, which is enrolling only patients who screen positive for the DNA repair mutations linked to response in TOPARP-A. That study uses an adaptive design, with biomarker selection based on response rate, with a test set for all patients to prove activity and a validation set to identify molecular signatures for PARP-inhibitor sensitivity.
Patients enrolled in the trial had prior treatment with one or two lines of chemotherapy, good performance status, and good organ function. No patient had prior exposure to platinum drugs or PARP inhibitors.
A biopsy was taken before starting olaparib, and another after two to four weeks of treatment. Half of the biopsies were taken from bone marrow and the rest from liver or from new metastases. Forty three of 49 biopsies (87%) were positive and suitable for next-generation sequencing, Mateo reported.
Of 50 patients enrolled, 49 were evaluable for the primary endpoint, and 16 of the 49 patients (33%) responded to olaparib. A majority of the responses lasted at least six months, with four patients responding for approximately one year—“That is quite a big achievement in a late-stage population,” Mateo noted.
Six patients also achieved a confirmed radiological response. Mateo explained that patients with germline BRCA mutations are known to respond to olaparib, but that these patients were not selected based on any family history nor were they known to be mutant for BRCA. In fact, most of the mutations were actually somatic.
Anemia, fatigue, and minor fever were the most common toxicities. “Compared with the studies in ovarian cancer, the drug was better tolerated because we were not seeing all the gastrointestinal toxicities associated with ovarian cancer,” he said.
Thirteen of 50 patients (26%) required a dose reduction to 300 mg BID, and three required a second dose reduction to 200 mg BID. One patient discontinued treatment due to myelosuppression.
TOPARP is being conducted in the U.K. but the investigators worked with U.S. researchers on the SU2C-Prostate Cancer Foundation Dream Team to identify the biomarkers of response to olaparib, Mateo said.
The study drug olaparib was provided by AstraZeneca.
Discussant: PFS Impressive, Questions about Preselection
The Discussant for the study, William K. Kelly, DO, Professor of Medical Oncology, Director of the Division of Solid Tumor Oncology, and Leader of the Biology of Prostate Cancer Program at Thomas Jefferson University's Kimmel Cancer Center, said the most impressive outcome in this trial is the median progression-free survival of 38 weeks: “For patients treated with third-line treatment, that's the most meaningful endpoint.”
He said that very few prostate cancer trials have published data in a third-line setting, but one important one that did tested abiraterone in 30 patients with progressive metastatic castration-resistant prostate cancer, all of whom had had disease progression on docetaxel and on enzalutamide (Noonan et al: Ann Oncol 2013;7:1802-1817).
“All those patients had a great response to enzalutamide, but abiraterone had minimal effect—only 10 percent of the patients responded,” Kelly said.
Median progression-free survival in that study was 15.4 weeks, compared with 38 weeks in the olaparib study—i.e., more than double. He said this clearly indicates that olaparib has clinical activity in patients with advanced castrate-resistant prostate cancer.
“What is important about olaparib is that we are treating a very advanced population with a third-line therapy,” he said. “All 50 patients had prior docetaxel, all had prior abiraterone, half had prior cabetaxel, a quarter had prior enzalutamide, and a quarter had palliative radiotherapy.”
Kelly said the TOPARP study endpoints were well described and based on radiologic response, PSA decline, and circulating tumor cell (CTC) conversion. Conversion from more than five or more CTCs per 7.5 ml of blood to less than five has been used in many trials to show prognostic significance, he explained.
“But then what they did was combine these three endpoints together to get the overall response, so one has to be very cautious making comparisons between this and other trials. If we look at endpoints individually, the PSA decline in response to therapy was 18 percent, CTC conversion was 29 percent, and overall response in the composite was 32 percent. So you just have to be cautious when you actually compare it.”
Kelly said he also differed from the TOPARP authors because he believes this was actually a very highly selected patient population: “Anybody who has third-line treatment comes into a major treatment center, has two phases of eligibility criteria that they go through, MRIs, and biopsies—that's a highly pre-selected population of patients.”
Problem with Tissue Biopsies
He also noted that although tissue biopsies were a key element in this study, they are cumbersome and not practical in large-scale trials.
Regarding the search for a reliable biomarker for this drug, Kelly noted that DNA repair defects were seen in 14 of 16 evaluable patients. “But the two patients who did not have these deletions also had responses, so this suggests there might be some other underlying method of action for these PARP inhibitors. We need to have prospective validation for this model, and further investigations to confirm activity in prostate cancer.”
Meanwhile, DNA repair defects are “not quite ready for prime time” as a clinical biomarker, and further pre-clinical and clinical validation are needed.
Kelly concluded by commending the TOPARP investigators for doing an excellent job on a very difficult study. “And perhaps we should start thinking about the PARPs a little differently, instead of confining ourselves to late-stage disease.”
Ovarian, TNBC Associated with BRCA Mutations
In another study reported at the meeting, a combination of olaparib and a PI3K inhibitor produced responses in both breast and ovarian cancer (Abstract 324).
The lead author of the Phase I trial, Ursula A. Matulonis, MD, Medical Director of Gynecologic Oncology at Dana-Farber Cancer Institute and Associate Professor of Medicine at Harvard Medical School, explained that olaparib was combined with the PI3K inhibitor BKM120 for patients with either high-grade serous ovarian cancer or triple negative breast cancer (TNBC) and that preclinical data had shown synergy between oral BKM120 and olaparib.
“Several years ago, my colleagues on the Stand Up To Cancer Targeting the PI3K Pathway in Women's Cancers Dream Team found that olaparib and BKM120 were more effective in mouse models of BRCA-mutant breast cancer and BRCA-wildtype triple-negative breast cancer than either drug alone,” she said. “Using SU2C funding, we then initiated this clinical trial to test whether the preclinical data would hold true in patients.”
Similarities exist between high-grade serous ovarian cancer and TNBC, and both are associated with BRCA mutations and are sensitive to platinum agents, she said.
The primary objective of the trial was to determine the maximal tolerated dose of the combination given orally on a continuous basis—which was:
- 50 mg per day for BKM120; and
- 300 mg given twice daily for olaparib.
There was little overlap in toxicities between the two drugs, Matulonis reported. The most common toxicities were nausea and fatigue, occurring in 80 and 66 percent of the patients, respectively, with most grade 1.
“PARP inhibitors will cause myelosuppression, whereas the PI3K inhibitors cause more rash and hyperglycemia. And BKM120 does cross the blood-brain barrier, so we were watching it for central nervous system toxicities as well.”
Approximately 25 percent of patients had grade 1 hyperglycemia, an-on target effect of PI3 kinase inhibitors—“That was certainly reassuring to us, that we were hitting the target,” she said.
The mean age of the ovarian cancer patients was 60, most of whom had high-grade serous ovarian cancer and germline BRCA mutation. The 24 breast cancer patients were slightly younger—mean age of 47—and two thirds of the patients had triple-negative breast cancer.
Among the 46 patients with ovarian cancer, 12 (26%) had a partial response, and 22 (48%) had stable disease.
And among the 24 breast cancer patients, five (21%) had a partial response, and 12 (50%) had stable disease. Four of the five partial responses were in patients with TNBC, and one who had ER/PR-positive disease.
Matulonis concluded by saying that the combinations of biologic agents will require establishment of target patient populations using biomarkers in order to predict sensitivity as well as determine mechanisms of resistance.
In another trial now starting, the alpha-specific PI3 kinase inhibitor BYL719 will be combined with olaparib, and this may be better tolerated because it does not cross the blood-brain barrier, she explained.