PARP Inhibitor Increases Survival for Subset of Patients With Ovarian Cancer

Lynparza (olaparib) was found to significantly increase overall survival (OS) when given as a maintenance therapy for patients with platinum-sensitive, relapsed serous ovarian cancer, according to the third survival analysis of 265 patients from the phase 2 Study 19 trial. These data were presented at the 2016 Annual Meeting of the American Society of Clinical Oncology (ASCO), a gathering of 30,000 oncology professionals in Chicago.

Results of the study's extension showed that the PARP inhibitor demonstrated the greatest OS advantage in women who had a BRCA mutation, according to lead author Jonathan Ledermann, professor of Medical Oncology at the University College London Cancer Institute and director of the Cancer Research UK & UCL Cancer Trials Centre. However, this survival advantage did not meet the criterion for statistical significance.

Additional key findings were that both time to first subsequent therapy or death (TFST) and time to second subsequent therapy or death (TSST) were substantially prolonged in patients taking Lynparza; again, the greatest benefit was seen in those with a BRCA mutation. Additionally, the unprecedented long-term exposure meant that 13 percent of all trial patients (15 percent of BRCA-mutated patients) received maintenance Lynparza for at least five years. In three years of follow-up since the 2012 analysis, there were no new safety findings.

Study 19 assessed patients with platinum-sensitive, recurrent high-grade serous ovarian cancer. They had received two or more prior regimens of platinum-based chemotherapy and experienced complete response (CR) or partial response (PR) to their most recent regimen. In a double-blind, one-to-one randomization, half the patients received Lynparza 400 mg capsules twice daily as maintenance therapy (136 patients) and half received placebo capsules twice daily (129 patients).

BRCA testing occurred for all patients in the form of case reports that contained the results of previous local germline BRCA testing or retrospective germline BRCA testing or tumor BRCA testing. The division between patients was nearly even between those BRCA mutations (136 patients) and those with wild type BRCA findings (118 patients), meaning that they either did not have a detected BRCA mutation or they had a BRCA mutation of unknown significance.

The primary endpoint was progression free survival (PFS) as measured by RECIST 1.0. Secondary endpoints included OS, safety and tolerability. Exploratory endpoints included TFST and TSST.

In Study 19, the median PFS for patients taking maintenance Lynparza was 8.4 months, compared to 4.8 months for the control group. The difference in the BRCA mutation subgroup was even more pronounced: 11.2 months with Lynparza and 4.3 months with placebo.

The third data analysis of Study 19, with data cut-off at September 30, 2015, was performed with data at 77 percent maturity, OS in the overall study population (265 patients) was a median 29.8 months in the Lynparza group and 27.8 months in the placebo group. In the BRCA mutation group (136 patients), where the data were at 70 percent maturity, median OS was 34.9 months for the Lynparza group and 30.2 months for placebo.

“Statistical tests did not provide sufficient evidence to dismiss the proportional hazards assumption for OS, so a restricted means analysis was performed to compare mean survival,” Ledermann said. “The difference in mean survival adds further support to the OS advantage of Lynparza.” In the overall study population, the restricted mean OS was 40.1 for the Lynparza group and 34.9 for the control group, for a difference of 5.2 months. In the BRCA subgroup, the restricted mean OS was 44.3 months on Lynparza (74 patients) versus 36.9 months with placebo (62 patients), a difference of 7.4 months.

There were 118 patients in the BRCA wild type subgroup. Of these, 47 (82 percent) experienced TFST events. The median TFST was 12.9 months. Of the 61 women who received placebo, 60 (98 percent) experienced TFST events; for them, the median TFST was 6.9 months. These data were at 91 percent maturity.

The updated exploratory analysis also examined TSST events. Of the 57 patients in the BRCA wild type subgroup treated with maintenance Lynparza, 47 (82 percent) experienced TSST. The median TSST was 17.0 months. In the placebo group, 59 (97 percent) experienced a TSST; the median TSST was 14.7 months. These data were at 90 percent maturity.

The somatic BRCA mutation consisted of 20 patients. Their OS data were inconsistent with the overall study population, but conclusions were limited by the small group size.

With a median follow-up interval of 5.9 years, 15 patients (11 percent) were still receiving Lynparza (eight with BRCA mutations). One BRCA-mutated patient was still receiving placebo. Twelve percent of BRCA wild-type patients on Lynparza achieved median follow-up of between five and six years.

The adverse events (AEs) and dose modifications remained consistent from 2012 and included three cases of myelodysplastic syndrome/acute myeloid leukemia. Two of these patients were in the Lynparza group (on treatment for 57 and 10 months) and one was on placebo for 44 months. In women receiving Lynparza for at least two years, frequencies of common AEs (nausea, fatigue, vomiting, anemia) were consistent with the overall population. AEs were generally reported during the first two years of treatment.

In the overall study population, 59 patients (43 percent) experienced an AE of grade 3 or above, while only 28 (22 percent) of those on placebo did. Those percentages remained fairly consistent among patients who remained on trial for two years or more: 15 (47 percent) in the Lynparza arm and one (20 percent) in the control arm. Likewise, the percentages of dose reductions due to AEs were consistent over time: 34 (25 percent)/five (4 percent) active versus control in the overall study population and eight (25 percent) and 0, respectively, among those on treatment for two years or longer. There were few treatment discontinuations due to AEs: eight active and two control patients overall and three Lynparza patients from the long-term treatment group. No patients from the long-term placebo group discontinued treatment due to AEs.