The recent vote of the FDA Oncologic Drugs Advisory Committee against recommending the approval of olaparib as a maintenance therapy strategy following the attainment of a second response to platinum-based therapy in either known BRCA mutation-positive or high-grade serous ovarian cancer raises a number of serious questions worthy of both scientific and societal discussion. These include both the level of evidence that should be mandated prior to FDA approval of a new antineoplastic agent and the individual roles of the drug regulatory agency (and its advisory committees) versus that of the individual cancer patient (in consultation with her or his own physician and personal advisors) in the determination of when it is appropriate to consider the administration of such a drug based on the appropropriateness of the end point used and the magnitude of the benefit. The intent of this commentary is not to attempt to explain the thought processes of the committee, other than to say they apparently were not convinced that the existing data demonstrated sufficient evidence of clinical benefit.1 Rather, what follows is a brief discussion of the data, which (in the opinion of this commentator) should unquestionably have resulted in a positive vote at the meeting and subsequent FDA approval of olaparib in this setting.
The theoretical rationale and extensive experimental data supporting the use of PARP (polyadenosine 5' diphosphosphoribose polymerase) inhibitors in BRCA mutation-positive or high-grade serous tumors has been reported elsewhere and will not be discussed in detail here.2-5 In brief, tumors with homologous recombination defects (eg, BRCA mutations) have been shown to be effectively killed by inhibitors of PARP in preclinical studies. Similarly, PARP inhibition in the absence of such defects is inadequate to result in sufficient tumor cell kill.
Of interest, high-grade serous ovarian cancers with wild-type BRCA have been shown to possess a molecular signature that appears similar to that observed in the presence of documented BRCA mutations (so-called, "BRCA-ness").3 This observation is the major justification for examining the clinical utility of PARP inhibition in patients with high-grade serous ovarian cancers in the absence of specific evidence of a BRCA abnormality.
Olaparib has been shown to be an active inhibitor of PARP-1, PARP-2, and PARP-3 in experimental model systems.4 It is well absorbed when taken orally and is removed from the body in both the urine and feces.
In the initial phase 1 study of olaparib conducted among patients with several BRCA-mutant tumor types (ovary, breast, prostate), 9 patients achieved a partial response.6 This study was subsequently expanded to include a total of 50 patients with a germline BRCA mutation, with a total objective response rate (RECIST or CA-125 response criteria) of 40%.7 A difference in the response rate between the platinum-sensitive (69%), platinum-resistant (45%), and platinum-refractory (23%) populations was observed in this trial. These data support the hypothesis that defective homologous recombination pathways are relevant to both sensitivity of platinum agents and PARP inhibitors (such as olaparib).
Subsequent phase 2 trials further confirmed the clinical activity of olaparib in ovarian cancer. One study revealed an overall response rate of 33% when the drug was administered at a dosage of 400 mg twice daily, while a response rate of only 13% was noted when the agent was delivered at 100 mg twice daily.8 A second study that included 65 patients revealed a 41% response rate in the presence of a BRCA mutation versus 24% in the absence of such a mutation.9 Of interest, responses in non-BRCA mutated cancers were limited to tumors that were platinum sensitive (50% response rate in platinum-sensitive cancers vs 4% in platinum-resistant cancers).
In a provocative randomized phase 2 trial, olaparib was directly compared to pegylated liposomal doxorubicin (PLD) as second-line therapy in patients with BRCA-mutated ovarian cancer.10 A reasonable objective response rate for single-agent olaparib was observed (25% and 31% for dosages of 400 mg or 800 mg/day, respectively). However, there was no difference in progression-free survival (PFS) between the olaparib vs the PLD-treated patient population.
Individuals on the PLD control arm experienced a median PFS that was almost double that predicted (7.1 months vs 4 months) in the study design.10 Rather than demonstrating the lack of efficacy for olaparib in this trial, these data provide strong support for the favorable utility of PLD in BRCA-mutant tumors.11
Finally, in an innovative randomized, double-blind, placebo-controlled phase 2 study, investigators examined single-agent olaparib when administered as a maintenance strategy in patients with ovarian cancer who had achieved a response (complete, partial, or stable disease) to a second-line platinum regimen.12 It is important to acknowledge here that patients were not selected for entry into this trial based on their BRCA mutation status. The study revealed a highly statistically significant improvement in PFS for the olaparib regimen (median 8.4 months vs 4.8 months; hazard ratio [HR] = 0.35; P <.001).
The investigators subsequently attempted to retrospectively examine the BRCA mutation status (germline or somatic only) of the patients entered into the trial, ultimately obtaining these relevant molecular data on 95% of study participants.13 When PFS was analyzed based on BRCA status, there was an even greater effect of treatment in the BRCA-mutant patient population (median 11.2 months vs 4.3 months; HR = 0.18; P <.0001). However, it is important to acknowledge that patients with wild-type BRCA also experienced a response to treatment (HR = 0.54; P = .0075).
Although there was no statistically significant improvement in overall survival associated with olaparib treatment on this trial, approximately one-quarter of patients on the study’s placebo arm eventually received a PARP inhibitor in addition to other potentially clinically active antineoplastic agents that may have had a favorable impact on their ultimate survival.14 (The issue of the appropriate clinical trial end points for regulatory approval of antineoplastic agents and the impact of subsequent treatments on an overall survival end point will be discussed by this commentator in more detail in the next issue.)
While it is not possible in this brief commentary to fully discuss the complex and increasingly contentious issue of the role of the FDA, or the agency’s advisory committees, versus cancer patients themselves in the decision to permit an antineoplastic agent to be administered outside the confines of an investigative trial, the determination in the recent olaparib review mandates that the topic be formally acknowledged and highlighted.
In the opinion of this commentator, this debate centers on the fundamental issue of the right of a patient with a very serious medical condition (such as in the current discussion) to make her or his own decision regarding the potential benefits versus risks associated with the utilization of a particular therapeutic strategy. In this specific clinical setting, the benefits associated with a substantial delay in the essentially certain future development of symptomatic disease progression must be balanced by the potential for a variety of relatively common or rarer toxicities (eg, hematologic effects, including myelodysplastic syndrome).
Objectively, who is in the best position to make this determination, to assess the relative utility of such a postponement of the inevitable, versus the discomfort and distress of treatment-related side effects? Is it really the regulator who has studied the data, or the expert physician/researcher/biostatistician panel members who might opine on how patients they have seen in the past or how the population of study patients may theoretically respond?
Or, might it be the individual patient with ovarian cancer who has been fortunate to have achieved a second remission and who has both the personal knowledge of what symptomatic disease progression means to her and has personally experienced the toxicity of antineoplastic treatment? Surely, the answer to this question is obvious.
Affiliation: Maurie Markman, MD, is from Cancer Treatment Centers of America, Philadelphia, PA, and Drexel University College of Medicine, Philadelphia, PA.
Disclosure: Dr Markman has been a consultant or a member of a paid advisory board for Amgen Inc, Celgene Corporation, Caris Life Sciences, Ltd, Novartis Corporation, sanofi-aventis US, LLC, Takeda Pharmaceutical Company, Limited, and Genentech, Inc.