We thank the authors for their supportive comments and questions regarding our study, “Olaparib Monotherapy in Patients With Advanced Cancer and Germline BRCA1/2
McNamara et al2
question the study design, end points, and dose modifications. This was a signal-seeking phase II study of olaparib with the inherent limitations of all phase II trials. The impact of olaparib on overall survival as well as on patient-reported outcomes such as symptom benefit is best addressed in randomized phase III trials. The primary aim of this trial was to look for an efficacy signal in heavily pretreated patients with germline BRCA
mutations and a wide range of cancers that included ovarian, breast, pancreatic, and prostate, as well as others. These patients typically have limited treatment options and the results of phase II trials can drive progress. Indeed, the results of this study were included in the consideration by the US Food and Drug Administration for the approval of olaparib for women with BRCA-
associated recurrent ovarian cancer after more than three lines of therapy (an unmet clinical need).
McNamara et al2
also question the dose modifications for toxicity and whether patients received adequate dose intensity. Future studies are needed to determine minimally effective dose. This is particularly relevant in patients who have been heavily pretreated who may be at risk of increased toxicities and require dose reductions.
We agree that knowledge of poststudy treatment after olaparib would be helpful. Approximately 50% of individuals overall (40% with pancreatic cancer) received subsequent therapy but detailed information is not available. McNamara et al2
ask whether an undiscovered PALB2
mutation could have contributed to hematologic adverse events. We think that this is unlikely given the rarity of coexistent homologous recombination (HR) gene mutations in BRCA
mutation carriers and it is more likely that prior chemotherapy contributed to the toxicity seen.3
Copur et al4
asked about the impact on response rates of women with platinum-sensitive ovarian cancer who could not receive further platinum. Unfortunately, these data were not collected prospectively but are being retrospectively ascertained. A preliminary analysis suggests that approximately 70% had platinum-resistant/-refractory ovarian cancer. Once all data are available, we will report the response rates in ovarian patient by platinum status as well as address the percentage of patients with high-grade serous cancer. It is however likely that the majority had high-grade serous ovarian cancer in keeping with the literature.
Although not reported in the original article, an equal number of patients with breast cancer had estrogen receptor (ER) –positive and ER-negative disease (each group with 31, 50%). Four patients in each group had a confirmed RECIST partial response. This is consistent with data from Tutt et al,5
who did not observe difference in response rates to olaparib based on ER status.
Copur et al4
suggests that platinum sensitivity alone may be more important than BRCA
status to predict response to poly(ADP–ribose) polymerase inhibitors, but this is not supported by the literature. There are certainly data that that platinum sensitivity is an important predictor of response to olaparib in patients with BRCA1/2
germline mutation-associated ovarian cancer, but responses have been previously reported in patients with BRCA-
related platinum-resistant ovarian cancer.6,7
Gelmon et al8
reported that patients with BRCA-related recurrent ovarian cancer had a higher response to olaparib compared with platinum-sensitive wild-type ovarian cancer. In addition in the maintenance trial of olaparib versus placebo which included women with platinum-sensitive relapsed ovarian cancer, those with BRCA
mutations had a significantly longer progression-free survival than patients with BRCA wild type.9
Both groups raise the pertinent question of whether the findings in this study could be applied to patients with germline mutations in other HR genes or in tumors with somatic mutations. This is an important and active area of research. Swisher et al10
recently presented the results of a study of rucaparib in recurrent ovarian cancer which found a response rate of over 60% in patients with BRCA
mutations but importantly a response rate of over 30% in patients with high genomic loss of heterozygosity in their tumors. Pennington et al11
have reported that germline and somatic mutations in HR genes predict platinum response in ovarian cancer (although the number of mutations in non-BRCA1/2
HR genes was small). Clinical trials are in progress to address these important questions. These studies will also help inform the clinical utility, from a therapeutic perspective, of the use of multigene panel tests to identify patients most likely to benefit from treatment with a poly(ADP–ribose) polymerase inhibitor.