PARP inhibitors for the treatment of HR-deficient breast and ovarian cancer

Read Also

                Based on the preclinical observations by Bryant and Farmer demonstrating that BRCA1 or BRCA2 deficient cell lines were sensitive to PARP inhibition (31, 32), there has been a great deal of interest in testing PARP inhibitors in these populations of patients.  Following the impressive phase I results showing single agent activity of olaparib in BRCA1/2 deficient tumors, a phase II study was initiated evaluating olaparib in patients with advanced ovarian cancer with known or suspected BRCA1 or BRCA2 mutations.  Overall, 40% of these participants obtained at least a PR and 46% in total obtained clinical benefit (defined as either response or SD ≥ 4 months).  In addition, olaparib appeared to offer more clinical benefit in patients defined as platinum sensitive compared to those who were platinum resistant or refractory (69% vs. 45% vs. 23% respectively) (64).  This has lead to the hypothesis that platinum sensitivity in ovarian cancer may be mechanistically related in some way to loss of HR within the tumor cell and that platinum sensitivity may be a predictive factor for clinical benefit from PARP inhibitors.

   

                Olaparib was then definitively evaluated as a single agent in BRCA1 and BRCA2 deficient breast and ovarian cancer, with participants requiring a confirmed, rather than suspected, germline mutation in order to participate.  In the breast cancer study, patients with confirmed germline BRCA1 or BRCA2 mutations were enrolled to either the RP2D of olaparib (400 mg twice daily) or a lower dose with evidence of PARP inhibitory activity (100 mg twice daily).  This trial suggested that single agent olaparib in active in this breast cancer population, with an overall response rate (ORR) of 41% in those 400 mg twice a day cohort.  Clinical activity, although less robust, was also observed in the lower dose cohort (ORR=11%) (65).  Using an identical design to the breast cancer study, evaluation of olaparib in BRCA1 or BRCA2 deficient ovarian cancer showed that 400 mg twice a day cohort appeared to have superior clinical activity compared to 100 mg twice a day, with an ORR of 33% compared to 13% respectively (66).  In both studies, although the higher dose cohort had a higher rate of grade 3 or greater toxicities compared to the lower dose cohort, overall toxicities were manageable; therefore, the data strongly suggests that, given the apparent therapeutic advantage of 400 mg twice a day, this dose should be used in future trials.

                Given that TNBC and high grade ovarian cancer share a number of pathologic features to breast and ovarian tumors with confirmed germline BRCA1 and BRCA2 mutations, it has been hypothesized that these sporadic cancers may have acquired deficiencies in HR repair and may also be sensitive to single agent PARP inhibitors (67). To evaluate this hypothesis, Gelmon and colleagues enrolled patients with TNBC or high grade serous ovarian cancer to receive olaparib 400 mg twice a day.  Following enrollment, patients were then stratified based on the presence or absence of BRCA1/2 mutations.  In the study, a total of 91 patients were enrolled (65 ovarian, 26 breast).  In patients with confirmed BRCA1/2 mutations, an ORR of 41% was observed.  Surprisingly, an ORR of 24% was observed in patients without BRCA1/2 mutations.  This finding is the first result in clinical trials demonstrating single agent PARP activity in non-BRCA mutation carriers and offers the first evidence that a “BRCAness” phenotype may be relevant clinically (68).  Gelmon and colleagues did not identify any responses in the breast cancer patients enrolled, regardless of mutation status.  Given that there was a total of 26 patients TNBC patients enrolled with only 10 possessing confirmed BRCA1/2 mutations, this unexpected observation is more likely due a combination of low enrollment numbers and heavy pretreatment (3 median prior lines of treatment) than PARP inhibitor inactivity in BRCA1/2 mutated breast cancer.

                As normal tissues are spared the toxicity of PARP inhibition due to functional HR activity, PARP inhibitors may represent a better tolerated and efficacious treatment that standard chemotherapy options in patients with germline BRCA1/2 mutations.  To assess this question, 97 patients with ovarian cancer with confirmed BRCA1/2 mutations were randomized in a 1:1:1 fashion to receive either olaparib 400 mg twice a day, olaparib 200 mg twice a day or liposomal doxorubicin at its standard dose of 50 mg/m² every 28 days.  The overall response rate was 31%, 25% and 18% in each of the three arms respectively.  Unfortunately, olaparib failed to demonstrate a progression free survival (PFS) benefit compared to liposomal doxorubicin (HR=0.88; 95% CI: 0.51-1.56; p=0.66) (69).  In their editorial discussing the study, Konstantinopoulos and Cannistra commented that the failure to see a PFS benefit may have been due to selecting doxorubicin as the standard agent.  As a topoisomerase II inhibitor, doxorubicin induces DSBs in DNA; theoretically, these types of agents are potentially more efficacious in the BRCA1/2 population than what might be predicted in a non-selected ovarian cancer population (70).  A retrospective review of doxorubicin in the BRCA1/2 population offers further support for this hypothesis (71).  This study illustrates some of the challenges that investigators face in designing future trials evaluating PARP inhibitors.  First, because of the biology and chemotherapy responsiveness of BRCA1/2-deficient tumors, particularly early on in treatment, it will be challenging to demonstrate a PFS benefit for PARP inhibitors compared to standard chemotherapy agents.  One could address this issue by choosing an equivalency rather than a superiority endpoint; however, one drawback to this approach is that equivalency trials require a much larger number of candidates compared to superiority trials to reach the necessary statistical power, resulting in a higher cost and length of time to obtain a result.  Second, because of the large number of identified standard agents for both breast and ovarian cancer, identifying an OS benefit will be difficult, if even possible, and will take a long follow up period to detect.

                Due to their toxicity profiles, PARP inhibitors may represent a potential maintenance therapy for BRCA1/2-deficient ovarian cancer.  In addition, given that high grade serous ovarian cancers may also have a “BRCAness” phenotype due to sporadic loss of HR repair (67, 68), it is possible that these agents may be effective in an unselected ovarian cancer population.  To asses this hypothesis, 265 patients with high grade serous ovarian cancer were randomized to receive either olaparib 400 mg twice a day or placebo.  To enroll in the study, patients were required to have platinum sensitive disease as previous work had correlated platinum sensitivity with increased PARP inhibitor efficacy (64).  Although patients were stratified based on ethnicity as some populations are more frequent carriers of germline BRCA1/2 mutations, the study did not specifically stratify based on BRCA mutation status.  This study demonstrated that although olaparib improved progression-free survival (8.4 months vs. 4.8 months; HR=0.35; 95% CI: 0.25-0.49; p=<0.001), at the time of publication, no overall survival benefit was identified (HR=0.94; 95% CI: 0.63-1.39; p=0.75).  A subgroup analysis performed by the study team suggests that although patients with germline BRCA1/2 mutations received more benefit from olaparib maintenance compared to patients without mutations, benefit was observed regardless of mutation status (72).  Given that patient with ovarian cancer patients have a number of treatment options available, it is not surprising that no overall survival benefit has been observed at this point in time.  Maintenance olaparib in combination with carboplatin is currently being evaluated; reported preliminary results indicated that the combination of olaparib and carboplatin resulted in an improved PFS compared to carboplatin alone (12.2 months vs. 9.6 months; HR=0.51; 95% CI: 0.34-0.77; p=0.0012) (73).