PARP inhibitors in Ovarian Cancer

Abstract

Poly-ADP-ribose polymerase inhibitors (PARPis) were first approved for the treatment of epithelial ovarian cancer (EOC), where as a maintenance therapy they transformed clinical management of this disease in both patients with and without homologous recombination deficiency (HRD). In this review, we provide a historical overview of PARPi use in EOC and discuss recent updates on overall survival (OS) data, highlighting their impact on regulatory approvals. We explore their potential as combination regimens with antiangiogenic and cell-cycle checkpoint inhibitors, as well as other small molecule inhibitors, to overcome resistance mechanisms and enhance therapeutic efficacy, providing a future perspective on the use of PARPis in EOC treatment.

Introduction

Epithelial ovarian cancer (EOC) is the most lethal gynecological malignancy in Western countries, with a five-year survival rate of around 50% [1]. Due to diffuse symptoms at presentation and lack of reliable biomarkers, patients are often diagnosed at an advanced stage. Approximately 75% of newly diagnosed patients have distant or regional spread, frequently involving other peritoneal organs including the omentum, intestines, and peritoneal wall [1]. Despite this, the combination of upfront treatment with carboplatin and paclitaxel and debulking surgery can effectively minimize residual disease in the majority of patients leading to long periods of remission. Recurrence, however, is near universal for late-stage disease.

Ovarian cancer treatment has relied on platinum analogues since the late 1970s, when clinical trials revealed that cisplatin produced nearly double the overall response rates (ORR) and complete responses when compared to non-platinum agents. Platinum agents (initially cisplatin, then carboplatin) combined with taxanes (typically paclitaxel) have since formed the cornerstone of chemotherapy for EOC and have served as the control arms for most clinical trials conducted on this disease [2]. An important contributor to the remarkable sensitivity of EOC tumors to platinum is thought to be the underlying deficiency in homologous recombination-mediated DNA repair. High-grade serous ovarian cancer (HGSOC) is the most common type of EOC, accounting for about 70% of cases. In HGSOC, genetic and epigenetic alterations of genes involved in the homologous recombination pathway are frequently observed, particularly in BRCA1 and BRCA2 genes, along with copy number variation and ubiquitous TP53 mutations. Homologous recombination deficiency (HRD) is a tumor characteristic defined by the inability to repair double-strand breaks (DSBs) in DNA via homologous recombination. Targeting this HRD is crucial in the treatment of EOC, as exemplified by the success of platinum agents and inhibitors of poly-ADP ribose polymerase (PARPis), a class of anticancer drugs that effectively kill cells with HRD through synthetic lethality.

In 2014, PARPi received their first regulatory approval as a fourth line treatment for recurrent EOC harboring BRCA mutations—an indication which was recently withdrawn — and subsequently for other histologies characterized by HRD or BRCA mutations [3]. This review aims to provide an overview of the current indications (Text box 1) and major clinical trials using PARPi in newly diagnosed EOC, both as an adjunct to chemotherapy and as a maintenance therapy (Table 1). We will also briefly discuss the long-term survival data that failed to show an overall survival (OS) benefit, resulting in the recent changes of PARPi approvals in the treatment of recurrent EOC. Importantly, we explore the latest opportunities for PARPi therapy and ongoing clinical trials.

TEXT BOX.

PARP INHIBITORS IN OVARIAN CANCER USFDA APPROVALS

Olaparib [LYNPARZA^®]

• Maintenance treatment of adult patients with deleterious or suspected deleterious germline or somatic BRCA-mutated advanced epithelial ovarian, fallopian tube or primary peritoneal cancer who are in complete or partial response to first-line platinum-based chemotherapy. Select patients for therapy based on an FDA-approved companion diagnostic for Lynparza.

• In combination with bevacizumab for the maintenance treatment of adult patients with advanced epithelial ovarian, fallopian tube or primary peritoneal cancer who are in complete or partial response to first-line platinum-based chemotherapy and whose cancer is associated with homologous recombination deficiency (HRD)-positive status defined by either:

  • a deleterious or suspected deleterious BRCA mutation, and/or
  • genomic instability. Select patients for therapy based on an FDA-approved companion diagnostic for Lynparza.

• For the maintenance treatment of adult patients with deleterious or suspected deleterious germline or somatic BRCA-mutated recurrent epithelial ovarian, fallopian tube or primary peritoneal cancer, who are in complete or partial response to platinum-based chemotherapy. Select patients for therapy based on an FDA-approved companion diagnostic for Lynparza.

Rucaparib [RUBRACA^®]

• For the maintenance treatment of adult patients with a deleterious BRCA mutation (germline and/or somatic)- associated recurrent epithelial ovarian, fallopian tube, or primary peritoneal cancer who are in a complete or partial response to platinum-based chemotherapy.

Niraparib [ZEJULA^®]

• ZEJULA is a poly (ADP-ribose) polymerase (PARP) inhibitor indicated:

• For the maintenance treatment of adult patients with advanced epithelial ovarian, fallopian tube, or primary peritoneal cancer who are in a complete or partial response to first-line platinum-based chemotherapy.

• For the maintenance treatment of adult patients with deleterious or suspected deleterious germline BRCA-mutated recurrent epithelial ovarian, fallopian tube, or primary peritoneal cancer who are in a complete or partial response to platinum-based chemotherapy. Select patients for therapy based on an FDA-approved companion diagnostic for ZEJULA.

Table 1:

Landscape of clinical trials investigating PARP inhibitors in epithelial ovarian cancer.

Study	Treatment Arm	Control Arm	No.	Phase	Indication	% BRCA+	Platinum status
PAOLA-1 [4]	Olaparib + bevacizumab	Placebo + bevacizumab	806	III	First-line	30%	PS
SOLO1 [6]	Olaparib	Placebo	391	III	First-line	100%	PS
Oza et al. [20]	Olaparib + chemotherapy	Chemotherapy	162	III	Relapsed	38%	PS + PPS
NRG-GY004 [22]	Olaparib + cediranib	Chemotherapy	565	III	Relapsed	24%	PS
SOLO 3 [19]	Olaparib	Chemotherapy	266	III	Relapsed	96%	PS + PPS
ICEBERG 3 [18]	Olaparib	Placebo	97	III	Relapsed	100%	PS + PPS
STUDY 19 [17]	Olaparib	Placebo	265	III	Relapsed	22%	PS + PPS
SOLO2 [16]	Olaparib	Placebo	295	III	Relapsed	97%	PS
CLIO [42]	Olaparib	Chemotherapy	100	III	Relapsed	NR	PS + PR
STUDY 42 [43]	Olaparib	None	193	II	Relapsed	100%	PS + PPS + PR
PRIMA [5]	Niraparib	Placebo	733	III	First-line	30%	PS
AVANOVA2 [44]	Niraparib + bevacizumab	Niraparib	97	II	Relapsed	34%	PS + PPS
NOVA [14]	Niraparib	Placebo	533	III	Relapsed	38%	PS + PPS
QUADRA [28]	Niraparib	None	463	II	Relapsed	19%	PS
NORA [15]	Niraparib	Placebo	265	III	Relapsed	100%	PS
VELIA [7]	Veliparib + chemotherapy	Placebo + chemotherapy	1140	III	First-line	30%	PS
Kummar et al. [21]	Veliparib + cyclophosphamide	Placebo + cyclophosphamide	75	III	First-line	40%	PS
ATHENA-MONO [8]	Rucaparib	Placebo	538	III	First-line	22%	PS
ARIEL4 [23]	Rucaparib	Chemotherapy	349	III	Relapsed	100%	PS + PR
ARIEL3 [13]	Rucaparib	Placebo	564	III	Relapsed	35%	PS + PPS
ARIEL2 part 1 [45]	Rucaparib	None	256	II	Relapsed	24%	PS
STUDY 10 [46]	Rucaparib	None	98	I/II	Relapsed	100%	PS

Abbreviations: BRCA+, BRCA mutation; NR, not reported; PS, platinum-sensitive, PPS, partially platinum-sensitive; PR, platinum-resistant

PARPi in EOC

Newly diagnosed EOC

Five randomized trials investigated the effectiveness of PARPis in patients with newly diagnosed advanced EOC who achieved a partial or complete response to first-line platinum-based treatment (PAOLA-1 [4], PRIMA [5], SOLO1 [6], VELIA [7], and ATHENA-MONO [8]) (Table 2). These were industry-sponsored, multi-center trials that used progression-free survival (PFS) as the primary outcome. SOLO1 required a BRCA mutation as an inclusion criterion and did not assess HRD. The other studies reported a BRCA mutation rate of around 30% and HRD rates of 63% in VELIA (Myriad myChoice CDx, genomic instability score [GIS≥33]) and about 50% in PAOLA-1 and PRIMA (myChoice HRD Plus, [GIS≥42]).

Table 2:

PARP inhibitor clinical trials

Clinical Trial	Results	Additional data and Comments
Newly diagnosed advanced EOC: PARPi as maintenance therapy after achieving a partial or complete response to 1st line platinum-based treatment. [PFS as primary endpoint]
PAOLA-1 [4]	• Maintenance olaparib + bevacizumab, PFS = 22.1 months. Maintenance placebo + bevacizumab, PFS = 16.6 months. • Lacked olaparib monotherapy arm for comparison so not clear if benefit from the combination or from olaparib alone.	• Prespecified 5-year OS analysis showed benefit from olaparib + bevacizumab for the HRD population. • HR 0.62, 95%CI: 0.45–0.85 • OS at 5 years, 65.5 versus 48.4%
SOLO1 [6, 34]	• Patients with newly diagnosed advanced ovarian cancer and a BRCA mutation who received olaparib as first-line maintenance therapy. • Olaparib maintenance therapy x 2 years substantially extended PFS of patients with tumors harboring BRCA mutations. • Approvals: USFDA approval 12/2018; EMA 5/2019 • Lower median age at diagnosis may have favorably affected the outcomes	7-year follow-up: • Survival benefit for patients with tumors harboring BRCA mutations. • Achieved clinically meaningful, albeit not statistically significant according to prespecified criteria, improvement in OS with maintenance olaparib. • HR for OS of 0.55, 95%CI: 0.40–0.76
PRIMA [5]	• Extended results in SOLO1 • Improvement in PFS with maintenance olaparib in patients with HRD not just BRCA mutation carriers.	• OS data pending
ATHENA-MONO [8]	• 1st-line maintenance treatment in a broad patient population, including those without BRCA mutations or other evidence of HRD, or high-risk clinical characteristics such as residual disease. • Maintenance rucaparib extended PFS for all patients with EOC, although effect less pronounced in patients without HRD. • 28.7 months with rucaparib versus 11.3 months with placebo in the HRD-positive population; and 12.1 months versus 9.1 months in the HRD-negative population.	• OS data pending
Newly diagnosed EOC: PARPi in combination with chemotherapy
VELIA [7]	• Design to assess efficacy of veliparib added to 1st line induction chemotherapy with carboplatin + Paclitaxel and continued as maintenance monotherapy in patients with previously untreated stage III or IV HGSOC. • 1:1:1 randomization:  — Chemotherapy + placebo followed by placebo maintenance (control)  — Chemotherapy + veliparib followed by placebo maintenance (veliparib combination only)  — Chemotherapy + veliparib followed by Veliparib maintenance (veliparib throughout) • Primary end point investigator-assessed PFS in the veliparib-throughout group versus control group, analyzed sequentially in the BRCA-mutation cohort, the cohort with HRD (which included the BRCA-mutation cohort), and the ITT population. • Median PFS (veliparib-throughout group versus control group):  — BRCA-mutation cohort: mPFS 34.7 and 22.0 months (HR 0.44, 95%CI: 0.28–0.68, P<0.001)  — HRD cohort: mPFS 31.9 versus 20.5 months (HR 0.57, 95CI: 0.43–0.76, P<0.001)  — ITT population: mPFS 23.5 and 17.3 months (HR 0.68, 95%CI: 0.56–0.83, P<0.001) • Veliparib + chemotherapy led to higher incidence of anemia and thrombocytopenia as well as of nausea and fatigue.	• As noted by the authors, “the independent value of adding veliparib during induction therapy without veliparib maintenance was less clear”. While front-line continuation-maintenance treatment with veliparib showed an improvement in PFS, no significant differences were seen when veliparib was administered in combination with chemotherapy (without maintenance) versus the placebo arm, demonstrating the benefit of veliparib in the concurrent/maintenance arm is driven mainly by veliparib maintenance. • Given the value of veliparib during induction without maintenance not clear, it is not common practice to use PARPi concurrently with chemotherapy. • Veliparib led to a higher incidence of anemia and thrombocytopenia when combined with chemotherapy as well as of nausea and fatigue overall.
Recurrent, platinum-sensitive EOC: PARPi in maintenance phase during or following chemotherapy.
[Were in CR or PR to their most recent platinum-based regimen, ≥2 prior treatments, and randomized to receive PARPi maintenance or placebo]
SOLO2 [16]	• Olaparib as maintenance therapy in patients with platinum-sensitive, relapsed ovarian cancer. • Only enrolled patients with known gBRCAmt • Significant improvement in PFS with olaparib. • mPFS with gBRCAmt (HR 0.18, 95%CI: 0.10–0.31, P<0.0001):  — Olaparib: 11.2 months (95%CI: 8.3-not calculable)  — Placebo: 4.3 months (3.0–5.4) • mPFS with gBRCAwt (HR 0.54, 95%CI: 0.34–0.85, P=0.0075):  — Olaparib 7.4 months (5.5–10.3)  — Placebo: 5.5 months (3.7–5.6) • Second interim analysis of OS (58% maturity):  — All patients: HR 0.88, 95%CI: 0.64–1.21, P=0.44  — gBRCAmt: HR 0.73, 95%CI: 0·45–1.17, P=0.19  — gBRCAwt: HR 0.99, 95%CI: 0.63–1.55, P=0.96 • Most common grade ≥3 AE in the olaparib group were fatigue (in ten [7%] patients in the olaparib group versus four [3%] in the placebo group) and anemia (seven [5%] versus one [<1%]).	Authors concluded: • Results support the hypothesis that patients with platinum-sensitive recurrent EOC with a BRCA mutation have the greatest likelihood of benefiting from olaparib treatment. • Apart from anemia, toxicities with olaparib were low grade and manageable. Note: PFS advantage without OS gains
Study 19 [17]	• Olaparib as maintenance therapy in patients with platinum-sensitive, relapsed ovarian cancer. • Significant improvement in PFS with olaparib. • mPFS – (HR 0.30, 95% CI: 0·22–0·41, P<0·0001):  — Olaparib: 19.1 months (95%CI: 16.3–25.7)  — Placebo: 5.5 months (95%CI: 5.2–5.8) • mOS HR 0.74 (95%CI: 0.54–1.00), P=0.054, unadjusted for the 38% of patients in the placebo group who received subsequent PARPi therapy:  — Olaparib: 51.7 months (95%CI: 41.5–59.1)  — Placebo: 38.8 months (95%CI: 31.4–48.6) • Most common grade ≥3 AE:  — Olaparib (n=195): Anemia, 38 (19%); fatigue or asthenia, 8 (4%); neutropenia 10 (5%)  — Placebo (n=99): Anemia 2 (2%); fatigue or asthenia 2 (2%); neutropenia 4 (4%) • SAEs:  — Olaparib: 35 (18%)  — Placebo: 8 (8%) • Benefit independent of mutation or HRD status but benefit of olaparib greater in the BRCA mutant population (HR 0.18, 95%CI: 0.10–0.31, P < 0.0001) than the non-BRCA mutant population (HR 0.54, 95%CI: 0.34–0.85, P = 0.007).	• Authors concluded “Olaparib provided a median overall survival benefit of 12.9 months compared with placebo in patients with platinum-sensitive, relapsed ovarian cancer and a BRCA1/2 mutation. Although statistical significance was not reached, these findings are arguably clinically meaningful and support the use of maintenance olaparib in these patients.” • Broad approval based on platinum sensitivity rather than specific biomarkers.
ARIEL3 [13, 33]	• Enrolled patients with platinum-sensitive, recurrent ovarian carcinoma responsive to the last line of platinum therapy. • PFS2 significantly longer with rucaparib than placebo, with HRs and 95%CIs <1, indicating clinical PFS benefit among all subgroups:  — ITT population: HR 0.703, 95%CI: 0.579–0.854  — HRD subgroup: HR 0.718, 95%CI: 0.558–0.923  — BRCA-mutant cohort: HR 0.672, 95%CI: 0.480–0.941	• Significant improvement in PFS benefit independent of mutation or HRD status initially led to broad approval for rucaparib based on platinum sensitivity rather than specific biomarkers. • Median OS:  — BRCA-mutant cohort: 45.9 months (95% CI: 37.7–43.2–59.6) in patients who received rucaparib versus 47.8 months (95% CI: 43.2–55.8) among patients who received placebo (HR 0.832, 95% CI: 0.581–1.192).  — Tumors with HRD: 40.5 months (95%CI: 36.6–48.4) with rucaparib compared with 47.8 months (95%CI: 42.7–53.0) with placebo (HR 1.005, 95%CI: 0.766–1.320).  — ITT population: 36.0 versus 43.2 months in favor of placebo (HR 0.995, 95%CI: 0.809–1.223). • Based on the above data, the FDA requested Clovis Oncology to voluntarily revise the second-line maintenance treatment indication, restricting rucaparib maintenance treatment to only patients with tumors harboring BRCA mutations. • In June 2022, in the US and Europe, Clovis voluntarily restricted the indication of rucaparib for the maintenance treatment of adult patients with recurrent epithelial ovarian, fallopian tube, or primary peritoneal cancer who are in complete or partial response to platinum-based chemotherapy received in the second or later line setting to patients with a tumor BRCA (tBRCA) mutation only. • In the entire non-tBRCA subgroup (n=368), the median OS was 32.2 months for patients treated with rucaparib compared to 38.3 months for patients treated with placebo (HR 1.08, 95%CI: 0.84–1.40).
NOVA [14]	• 553 enrolled patients, 203 in gBRCAmt cohort and 350 in non- gBRCAmt cohort • Significant improvement in PFS with niraparib independent of mutation or HRD status initially led to a broad approval based on platinum sensitivity rather than specific biomarkers. • mPFS gBRCAmt (HR 0.27, 95%CI: 0.17–0.41, P<0.001):  — Niraparib: 21.0 months  — Placebo: 5.5 months • mPFS non-gBRCAmt with HRD (HR 0.38, 95% CI: 0.24–0.59, P<0.001):  — Niraparib: 12.9 months  — Placebo: 3.8 months • mPFS overall non-gBRCAmt (HR 0.45, 95% CI: 0.34–0.61, P<0.001):  — Niraparib: 9.3 months  — Placebo: 3.9 months • Most common grade 3 or 4 AEs with niraparib: Thrombocytopenia (in 33.8%), anemia (in 25.3%), and neutropenia (in 19.6%), which were managed with dose modifications.	Before OS data became available, authors concluded: Among patients with platinum-sensitive, recurrent ovarian cancer, the median duration of PFS was significantly longer among those receiving niraparib than among those receiving placebo, regardless of the presence or absence of gBRCAmt or HRD status, with moderate bone marrow toxicity”. • Median OS in long-term follow-up data:  — gBRCAmt cohort: 40.9 months versus 38.1 months for niraparib and placebo, respectively (HR 0.85, 95% CI:0.61–1.20).  — Non-gBRCAmt cohort: 31.0 months versus 34.8 months for niraparib and placebo, respectively (HR 1.06, 95%CI: 0.81–1.37).  — non- gBRCAmt with HRD group: 35.6 months compared to 41.4 months in the niraparib and placebo groups respectively, HR of 1.29 (95%CI: 0.85–1.95).  — HRD not determined: 29.8 months vs 20.2 months for patients who received niraparib and placebo, respectively, HR of 0.62 (95%CI: 0.29–1.35).  — Homologous recombination proficient cohort: 27.9 months in both the niraparib and placebo arms, HR of 0.93 (95%CI:0.61–1.41). • Due to the above results, niraparib was withdrawn as maintenance therapy in ≥2nd line of therapy in patients whose tumors did not harbor BRCA mutations.
NORA [15, 32]	• Significant improvement in PFS with niraparib. • 265 patients were randomized to receive niraparib. 249 patients received an individualized starting dose (ISD) (300 mg, n = 14; 200 mg, n = 235) as per protocol. • mPFS, ITT population (HR 0.32, 95%CI: 0.23–0.45, P<0.0001):  — Niraparib: 18.3 months (95%CI: 10.9-not evaluable)  — Placebo: 5.4 months (95%CI: 3.7–5.7 months) • PFS benefit observed in patients receiving an ISD, regardless of BRCA mutation status. • Grade ≥3 TEAE:  — Niraparib: 50.8%  — Placebo: 19.3% • Most common AEs; neutropenia (20.3% versus 8.0%) and anemia (14.7% versus 2.3%). • Benefit independent of mutation or HRD status led initially to a broad approval based on platinum sensitivity rather than specific biomarkers.	• At OS maturity of 44% mOS not significantly longer with niraparib (46.3 months) versus placebo (43.4 months) (HR 0.826, 95%CI: 0.56–1.21). • Reported a favorable but still insignificant trend for OS with niraparib in the IPCW-adjusted analysis, given 43% rate (38/88) of subsequent use of a PARPi after progression on placebo with mOS of 46.3 and 34.3 months with niraparib and placebo respectively, HR 0.692 (95%CI: 0.447–1.072).
Recurrent, platinum-sensitive EOC: PARPi against non-platinum containing chemotherapy regimens
ICEBERG3 [18]	Multicenter open label randomized phase 2 trial comparing olaparib 200 mg, olaparib 400 mg and pegylated liposomal doxorubicin (PLD). • Enrolled patients with ovarian cancer that recurred within 12 months of prior platinum therapy and with confirmed gBRCAmt. • 1ry Efficacy Endpoint: Median PFS (Difference not statistically significant. mPFS for combined olaparib doses versus PLD: HR 0.88, 95%CI: 0.51–1.56, P=0.66):  — Olaparib 200 mg: 6.5 months (95%CI: 5.5–10.1 months)  — Olaparib 400 mg: 8.8 months (95%CI: 5.4–9.2 months)  — PLD: 7.1 months (95%CI: 3.7–10.7 months) • 2ry Efficacy Endpoint: RECIST-assessed ORRs (Differences not statistically significant):  — Olaparib 200 mg: 25%  — Olaparib 400 mg: 31%  — PLD: 18%	• BRCA-mutant tumors; 50% likely not platinum-sensitive. PFS not significantly different but unexpectedly high PLD activity. • Olaparib 400 mg dose deemed suitable.
SOLO3 [19, 30]	Patients with platinum-sensitive relapsed ovarian cancer harboring gBRCAmt who had received ≥2 prior lines of platinum-based chemotherapy. • Olaparib versus non-platinum chemotherapy physician’s choice (TPC; paclitaxel [P], topotecan [T], gemcitabine [G], or pegylated liposomal doxorubicin [PLD]) Primary and Secondary Endpoints: • Olaparib monotherapy provided statistically significant improvements in ORR (primary endpoint) PFS (secondary endpoint) Following disease progression: • 119/178 [67%] patients assigned to olaparib received subsequent anticancer therapy versus 54/88 [61%] TPC patients received ≥1 subsequent anticancer regimen. • 3/178 (2%) assigned to olaparib versus 23/88 (26%) TPC patients received a PARPi in their first subsequent line of therapy. • Median PFS2 23.6 (olaparib) versus 19.6 months (TPC): HR 0.80 (95%CI: 0.56–1.15), P=0.229 • Median OS: 34.9 (olaparib) versus. 32.9 months (TPC): HR 1.07 (95%CI: 0.76–1.49), P=0.714	• Potential olaparib detrimental effect compared to standard treatment in patients with gBRCAmt-associated EOC and ≥3 prior lines of chemotherapy (HR 1.33, 95%CI: 0.84–2.18). • Neither PFS2 nor OS significantly different leading to voluntary withdrawals for this indication.
Recurrent, platinum-sensitive EOC: Concurrent/Combination PARPi + chemotherapy
Oza et al, 2015 [20]	• Randomized, open-label, phase 2 study. • Enrolled platinum-sensitive, recurrent, HGSOC who had received up to three previous courses of platinum-based chemotherapy and were progression free for at least 6 months before randomization. • Regimens:  — Olaparib 200 mg bid days 1–10 of each 21-day cycle + 175 mg/m2 paclitaxel on day 1 + carboplatin AUC 4 mg/mL/min on day 1 then olaparib 400 mg bid until progression.  — 175 mg/m2 paclitaxel on day 1 + carboplatin AUC 6 mg/mL/min on day 1 then no further treatment. • Median PFS (HR 0.51, 95%CI: 0.34–0.77, P=0.0012):  — Olaparib + chemotherapy: mPFS, 12.2 months (95%CI: 9.7–15.0).  — Chemotherapy alone: mPFS 9.6 months (95%CI: 9.1–9.7) • Subgroup analysis: Patients with BRCA mutations (HR 0.21, 95%CI: 0.08–0.55, P=0.0015) • AEs: Reported more frequently with olaparib + chemotherapy than with chemotherapy alone, most were of mild-to-moderate intensity and attributable to paclitaxel and carboplatin.	• PFS significantly longer in the olaparib plus chemotherapy group. However, the trial was not designed to measure the contribution of each treatment, and the late separation of the PFS curves suggest maintenance key contributor to the improvement. • Results similar to those seen with front-line combination/maintenance treatment with veliparib that showed improved PFS however, no significant differences were seen when veliparib was administered in combination with chemotherapy (without maintenance) versus the placebo arm, by inference demonstrating the benefit of veliparib in the concurrent/maintenance arm is driven mainly though post-platinum maintenance. • The difference in OS of 2.6 months at expense of toxicity attributable to the chemotherapy. • Benefit was more pronounced in tumors with BRCA mutations.
Kummar et al, 2015 [21]	• Veliparib + oral cyclophosphamide • Randomized phase II trial in patients with pretreated ovarian cancer harboring BRCA mutations, primary peritoneal cancer, fallopian tube cancer, or HGSOC. • Comparing response rate of veliparib + cyclophosphamide compared with cyclophosphamide alone. • One CR observed in each arm, with 3 PRs in combination arm and 6 PRs in the cyclophosphamide alone arm. PFS and ORR not improved.	• 1st trial evaluating single-agent, low-dose cyclophosphamide in HGSOC, peritoneal, fallopian tube, and BRCA-mutant ovarian cancers shown to be well-tolerated with clinical activity was observed. • However, addition of veliparib at 60 mg daily did not improve ORR or mPFS. • Genetic sequence and expression analyses performed for 211 genes involved in DNA repair; none of the detected genetic alterations were significantly associated with treatment benefit.
Olaparib alone or in combination with cediranib against platinum-based chemotherapy regimens.
NRG-GY004 [22]	Randomized phase 3 study evaluating olaparib or cediranib + olaparib to platinum-based chemotherapy in platinum-sensitive recurrent ovarian cancer • BROCA-HR, a targeted next generation sequencing assay identifying all classes of mutations in 88 DNA repair or related genes was performed on germline and tumor DNA and was evaluable in 491/565 randomized patients. • Core HRR genes wild-type (HRRwt) in 323 pts, mutant (HRRmt) in 147, and not assessable in 21. >90% of HRRmt were BRCA mutation. • Across all patients, mPFS for HRRmt was 13.7 months versus 8.3 months HRRwt (HR 0.41, p <0.0001) • Median PFS in patients with HRRmt:  — Chemotherapy: 12.3 months  — Olaparib: 13.1 months  — Olaparib + cediranib: 20.4 months  (Olaparib versus chemotherapy, HR 0.78, 95%CI: 0.48–1.27)  (Olaparib + cediranib versus chemotherapy: HR 0.55, 95%CI: 0.32–0.95) • Median PFS in patients with HRRwt:  — Chemotherapy: 9.0 months  — Olaparib: 6.4 months  — Olaparib + cediranib: 8.5 months  (Olaparib versus chemotherapy, HR 1.56, 95%CI: 1.15–2.12)  (Olaparib + cediranib versus chemotherapy: HR 0.93, 95%CI: 0.68–1.27) •HRR status was predictive of olaparib response versus chemotherapy (P=0.0176) but not of cediranib + olaparib versus chemotherapy (P=0.1009).	• HRR status driven by BRCA mutation, correlated with overall prognosis, and predictive of olaparib response versus chemotherapy.
Tumors harboring BRCA mutations and varying degrees of platinum sensitivity: Rucaparib vs. standard-of-care chemotherapy
ARIEL4 [23]	• Open-label, randomized, controlled, phase 3 study in patients with ovarian carcinoma harboring germline or somatic BRCA1 or BRCA2 mutations who had received ≥2 prior lines of chemotherapy. • Aimed to compare rucaparib versus platinum-based and non-platinum-based chemotherapy. • Eligible patients stratified by PFS interval after most recent platinum-containing therapy, to 600 mg rucaparib bid or chemotherapy. • Platinum resistant or partially platinum-sensitive disease given 60–80 mg/m2 paclitaxel on days 1, 8, and 15); those with fully platinum-sensitive disease received platinum-based chemotherapy (single-agent cisplatin or carboplatin, or platinum-doublet chemotherapy). Patients were treated in 21-day or 28-day cycles. • mPFS in the efficacy population:  — Rucaparib group: 7.4 months (95%CI: 7.3–9.1)  — Chemotherapy group: 5.7 months (95%CI: 5.5–7.3) (HR 0.64, 95%CI: 0·49–0.84, P=0·0010) • mPFS in the ITT population:  — Rucaparib group: 7.4 months (95%CI: 6.7–7·9)  — Chemotherapy group: 5.7 months (95%CI: 5.5–6.7) (HR 0.67, 95%CI: 0.52–0.86, P=0.0017) • Most common ≥grade 3 TEAE was anemia or decreased hemoglobin (in 52 (22%) of 232 patients in the rucaparib group versus six (5%) of 113 in the chemotherapy group. • Only trial to prospectively assess the role BRCA reversion mutations on treatment response to PARPi or platinum therapy.	• Authors argued results “support rucaparib as an alternative treatment option to chemotherapy for patients with relapsed, BRCA1-mutated or BRCA2-mutated ovarian carcinoma”. • However, despite PFS advantage with rucaparib, median OS were 19.4 months with rucaparib compared to 25.4 months with standard treatment (HR 1.313, 95%CI: 0.999–1.725). [Note: While 69% crossover rate potentially confounded interpretation of OS data, there was no difference in PFS2 (HR 0.860, 95%CI: 0.674–1.098). • As a result of this, Clovis Oncology voluntarily withdrew rucaparib for the treatment of adult patients with a deleterious germline and/or somatic BRCA mutation associated epithelial ovarian, fallopian tube, or primary peritoneal cancer who have been treated with ≥2 chemotherapies. This decision was made in consultation with the USFDA after the detrimental effect in terms of OS in the ARIEL4 Study.
PARPi in tumors with varying degrees of HRD or chemotherapy sensitivity.
ARIEL2 part 1 [45]	• Assessed if in addition to BRCA1/2 mutations, genomic LOH quantified with a next-generation sequencing assay, could predict response to rucaparib. • On the basis of tumor mutational analysis, classified patients with recurrent, platinum-sensitive, HGSOC into one of three predefined HRD subgroups (Cutoff of 14% or more genomic LOH was prespecified for LOH high):  — Deleterious germline or somatic BRCA mutation  — BRCA wild-type and LOH high (LOH high group)  — BRCA wild-type and LOH low (LOH low group) • Rucaparib 600 mg bid for continuous 28-day cycles until disease progression or any other reason for discontinuation. • Primary endpoint: PFS • 192/256 patients could be classified into one of the three predefined subgroups. • mPFS was significantly longer in BRCA mutant (HR 0.27, 95%CI: 0.16–0.44, P<0.0001) and LOH high (HR 0.62, 95%CI: 0.42–0.90, P=0.011) subgroups compared with the LOH low subgroup:  — BRCA mutation: 12.8 months (95% CI: 9.0–14.7)  — LOH high group: 5.7 months (95% CI: 5.3–7.6)  — LOH low group: 5.2 months (95% CI: 3.6–5.5) • Most common grade ≥3 TEAE were anemia or decreased hemoglobin (45 [22%] patients), and elevations in ALT or AST (25 [12%]).	• Authors concluded: Our results suggest that assessment of tumor LOH can be used to identify patients with BRCA wild-type platinum-sensitive ovarian cancers who might benefit from rucaparib. • As noted above guided by the results in ARIEL3, Clovis voluntarily restricted the indication of rucaparib for the maintenance treatment of adult patients with recurrent epithelial ovarian, fallopian tube, or primary peritoneal cancer who are in complete or partial response to platinum-based chemotherapy received in the second or later line setting to patients with a tumor BRCA mutation only. Likewise, guided by the results in ARIEL4 Clovis Oncology voluntarily withdrew rucaparib for the treatment of adult patients with a deleterious germline and/or somatic BRCA mutation associated epithelial ovarian, fallopian tube, or primary peritoneal cancer who have been treated with ≥2 chemotherapies.
ARIEL4	See above
PARPi in combination with anti-angiogenic agents
NRG-GY004 [22]	• Examined the safety and efficacy of adding cediranib (a VEGFR1–3 TKI) to olaparib (C+O), versus olaparib monotherapy (O) versus SOC platinum-based chemotherapy (carboplatin + PLD, gemcitabine, or paclitaxel) in patients with recurrent platinum-sensitive ovarian cancer, fallopian tube, or primary peritoneal cancer patients, with or without BRCA mutation. • The trial did not meet the primary endpoint in the ITT population of a statistically significant improvement in PFS with C added to olaparib versus platinum-based chemotherapy. • Median follow-up 66.5 months • Median OS:  — Olaparib (O): 31.0 months  — Cediranib + Olaparib (C+O): 33.5 months  — Platinum-based chemotherapy (SOC): 32.7 months • HRs for OS:  — O versus SOC: 1.27 (95% CI: 0.99–1.62, P = 0.06)  — C+O versus SOC: 1.12 (95% CI: 0.87–1.43, P = 0.38) • Median OS patients with gBRCAmt:  — Olaparib (O): 41.3 months  — Cediranib + Olaparib (C+O): 44.8 months  — Platinum-based chemotherapy (SOC): 43.2 months • HRs for OS patients with gBRCAmt:  — O versus SOC: 1.39 (95%CI: 0.80–2.42)  — C+O versus SOC: 1.24 (95%CI: 0.94–1.63) • HRs for OS patients with gBRCAmt:  — O versus SOC: 1.26 (95%CI: 0.71–2.21)  — C+O versus SOC: 1.07 (95%CI: 0.82–1.40)	• Author’s conclusions: In NRG-GY004, neither olaparib nor cediranib + olaparib improved OS compared to SOC as treatment for relapsed platinum sensitive ovarian cancer. HRs for OS for both olaparib and for cediranib + olaparib exceeded 1 with wide 95% CIs that included 1. These findings must be interpreted with caution given the proportion of patients terminating follow-up early and the number of patients on the SOC arm who received off-protocol PARPi maintenance.
AVANOVA2 [44]	• Niraparib 300 mg alone or with intravenous bevacizumab 15 mg/kg once every 3 weeks until disease progression in patients previously treated with platinum-containing therapy for primary disease and one or fewer non-platinum-containing regimens for recurrent disease. • Median PFS:  — Niraparib + bevacizumab: 12.5 months (95%CI: 8.5–16.7)  — Niraparib alone: 5.5 months (95%CI: 3.8–6.3) • HRs:  — All patients, adjusted HR 0.34 (95%CI: 0.21–0.55, P<0·0001)  — Patients with HRD-positive tumors: HR 0.41, 95%CI: 0.23–0.76  — Patients with HRD-negative disease: HR 0.40, 95%CI: 0.20–0.79  — Time to First Subsequent Therapy: HR 0.4, 95%CI: 0.25–0.64  — PFS2: HR 0.55, 95%CI: 0.35–0.88  — OS (49 events only) HR 0.77, 95%CI: 0.42–1.41 • Grade ≥3 AE in 31/48 (65%) patients who received niraparib + bevacizumab and 22/49 (45%) who received single-agent niraparib. • Most common grade ≥3 AE in both groups were anemia (7/48 [15%] versus 9/49 [18%]) and thrombocytopenia (5/48 [10%] versus 6/49 [12%]), and hypertension in the combination group (10/49 [21%] versus 0).	• The authors note that unfortunately the study was not powered to detect differences in OS or any other efficacy endpoints. While other endpoints are encouraging verification will be needed.
Recurrent platinum-resistant EOC: PARPi monotherapy [Limited number of trials due to shared mechanisms of resistance between platinum drugs and PARPis, such as BRCA reversion mutation]
ARIEL2 part 2	• Part 2 enrolled patients who had received three to four prior chemotherapies, including patients with platinum-sensitive or platinum-resistant/refractory disease. • Confirmed ORR, the study’s primary endpoint: with ORR among heavily pretreated patients in Part 2:  — BRCA mutation (n=138): 31.0% (95%CI: 21.3–42.0)  — BRCAwt/LOH-high (n=156): 6.8% (95% CI: 2.3–15.3)  — BRCAwt/LOH-low (n=168): 5.6% (95% CI: 2.1–11.8) • ORR and PFS among molecular subgroups:  — BRCA mutation: 45.7% (95%CI: 37.2–54.3) / 7.8 months  — BRCAwt/LOH-high: 16.7% (95%CI: 11.2–23.5) / 4.3 months  — BRCAwt/LOH-low: 7.7% (95%CI: 4.2–12.9) / 4.0 months • Mutations in RAD51C (n=4), and RAD51D (n=3) identified as possible vulnerable alteration, although 6/7 were platinum sensitive, suggesting that as long as cross-resistance is not present, rucaparib can be highly effective in HGSOC with RAD51C/D mutations, even in late lines of treatment.	Authors argue: • Analysis highlights significant overlap between molecular mechanisms resulting in platinum and PARPi sensitivity and the extent of cross-resistance that exists between these two drug classes. • RAD51C/D mutations and high level BRCA1 promoter methylation are strong predictors of sensitivity to a PARPi. • Administration of rucaparib, should be considered in earlier lines of therapy before emergence of platinum resistance.
ARIEL4	See above.
CLIO [42]	• Evaluated single-agent olaparib versus SOC chemotherapy in platinum-resistant ovarian cancer (recurrence within 6 months after last platinum) • Olaparib monotherapy, 300 mg bid or physician’s choice chemotherapy (PLD, topotecan, paclitaxel, or gemcitabine) • Primary endpoint: ORR • ORR similar for the platinum-resistant cohort, with favorable disease response to olaparib only in patients with ≤4 prior lines of treatment. • mDOR:  — Olaparib: 5.4 months  — Chemotherapy: 4.5 months • mPFS:  — Olaparib: 2.9 months  — Chemotherapy: 3.4 months	• Compared with chemotherapy, olaparib monotherapy appeared comparably effective in platinum-resistant ovarian cancer including patients with gBRCAwt • However, >90% of patients PARPi naive which could account for superior ORR of olaparib compared with chemotherapy.
ICEBERG 3 [18]	See above
QUADRA [28]	• Investigated activity of niraparib monotherapy as ≥4th line of therapy in patients whose disease recurred <6 months after last platinum-based chemotherapy. • 151/463 (33%) of tumors were resistant and 161/463 (35%) of tumors were refractory to the last administered platinum therapy. • Primary efficacy population:  — ORR: 13/47 (28%) (95%CI: 15.6–42.6; one-sided P=0.00053) - possible benefit in heavily pre-treated tumors using niraparib. • Not surprisingly, effect more pronounced with HRD, platinum-sensitive disease.	• Despite select efficacy population, authors argued they observed clinically relevant activity of niraparib among women with heavily pretreated ovarian cancer, especially in patients with HRD-positive platinum-sensitive disease, which includes not only patients with a BRCA mutation but also a population with BRCA wild-type disease. Our data support expansion of the treatment indication for PARPi to include patients with HRD-positive ovarian cancer beyond those with BRCA mutations.
STUDY 42 [43]	• Advanced ovarian cancer who had received ≥3 prior lines of chemotherapy. • Olaparib 400 mg bid monotherapy until disease progression. • gBRCAmt-associated ovarian cancer:  — ORR: 46/137 (34%) (95%CI: 26–42)  — mDoR 7.9 (95%CI: 5.6–9.6) months • Efficacy according to platinum sensitivity:  — ORR in platinum-resistant tumors: 30%  — mDoR for platinum-sensitive tumors: 8.2 months (95%CI: 5.6–13.5)  — mDOR for platinum-resistant tumors: 8.0 months (95%CI: 4.8–14.8)	• Olaparib monotherapy demonstrated notable antitumor activity in patients with gBRCAmt advanced ovarian cancer who had received ≥3 prior lines of chemotherapy.
ARIEL2 part 1 [45]	See above
STUDY 10 [46]	• Investigated activity of rucaparib in patients with HGSOC and a deleterious germline or somatic BRCA1 or BRCA2 mutation who received ≥2 prior chemotherapies and whose disease was sensitive, resistant, or refractory to platinum-based chemotherapy. • Primary endpoint: investigator-assessed ORR • Secondary endpoints: DOR and PFS • Efficacy population (n=106)  — ORR: 53.8% (95%CI: 43.8–63.5) [CR: 8.5%; PR: 45.3%]  — mDOR: 9.2 months (95%CI: 6.6–11.6) • Safety population (n=377):  — Most frequent TEAEs: Nausea, asthenia/fatigue, vomiting, and anemia  — Most common grade ≥3 TEAE: Anemia  — TEAEs led to treatment interruption (58.6%), dose reduction (45.9%), and treatment discontinuation (9.8%)	• Rucaparib has antitumor activity in advanced BRCA-mutated HGSOC and a manageable safety profile in a very select efficacy population

Abbreviations: AE, adverse event; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CI, confidence interval; CR, complete response; DOR, duration of response; EMA, European Medicines Agency; EOC, epithelial ovarian cancer; gBRCA, germline BRCA; gBRCAwt, wild type germline BRCA status; gBRCAmt, mutant germline BRCA status; HGSOC, high grade serous ovarian carcinoma; HR, hazard ratio; HRD, homologous recombination deficient; ITT, intention to treat; IPCW, inverse-probability-of-censoring weighting; LOH, loss of heterozygosity; mOS, median overall survival; mPFS, median progression-free survival, mDOR, median duration of response; ORR, overall response rate; OS, overall survival; PARP, poly(ADP-ribose) polymerase: PARPi, PARP inhibitor; PFS, progression-free survival; PFS2, time to second objective disease progression; PLD, pegylated liposomal doxorubicin; PR, partial response; SAE, serious adverse event; SD, stable disease; SOC, standard of care; TEAE, treatment emergent adverse event; TKI, tyrosine kinase inhibitor; USFDA, United States Food and Drug Administration; VEGFR, vascular endothelial growth factor receptor.

PARPi as maintenance therapy for newly diagnosed EOC

Four of these studies evaluated the PARPi as maintenance therapy after platinum-based chemotherapy with the goal of increasing the durability of response. The SOLO1 trial was the first to report results, specifically for EOC with BRCA mutations. It showed that maintenance therapy with olaparib for two years substantially extended PFS leading to its approval by the US Food and Drug Administration (FDA) in December 2018 and the European Medicines Agency in May 2019. The PFS benefit continued at 3-year follow-up in the experimental arm despite stopping olaparib after 2 years. This trial set a new standard of care for the narrow set of patients with BRCA1/2 mutations. The median age in SOLO1 was markedly lower (53 years) than other contemporary studies (63 years) and the median age of diagnosis may have favorably affected the outcomes [1]. In 2019, the PRIMA trial extended these results beyond BRCA mutation carriers, showing a significant improvement in PFS in patients with HRD. Most recently, ATHENA-MONO showed that maintenance rucaparib extended PFS for all patients with EOC, although this effect was less pronounced in patients without HRD [8]. In 2018, the results of GOG218 [9] and ICON7 [10] had led to the approval of the anti-angiogenic agent bevacizumab for first-line chemotherapy and continuation as maintenance therapy. Based on this, the PAOLA-1 trial evaluated the efficacy of adding olaparib to bevacizumab maintenance after platinum–taxane chemotherapy [4]. Bevacizumab was given at its full single-agent dose of 15 mg/kg every three weeks for a total duration of up to 15 months. The addition of maintenance olaparib provided a significant improvement in PFS to 22.1 months with olaparib plus bevacizumab from 16.6 months with placebo plus bevacizumab. This effect was greater in patients with HRD-positive tumors (37.2 versus 17.7 months) although it is not clear if this benefit was derived from the combination or from olaparib alone since there was no olaparib monotherapy arm for comparison.

PARPi in combination with chemotherapy for newly diagnosed EOC

VELIA investigated the concurrent administration of a PARPi and chemotherapy for first-line treatment of advanced EOC [7]. In this study, veliparib was given together with carboplatin and paclitaxel chemotherapy during induction therapy, followed by veliparib maintenance therapy or placebo. The rationale for this approach was that the combination of a PARPi and platinum-based chemotherapy would impair the cell’s ability to repair DNA damage caused by the chemotherapy, resulting in synergy and synthetic lethality [7]. The addition of veliparib to induction and maintenance therapy significantly improved PFS compared to carboplatin plus paclitaxel induction therapy alone. However, the independent value of adding veliparib during induction therapy without maintenance therapy was less clear, and it is not common practice to use PARPi concurrently with chemotherapy [3].

Recurrent, platinum-sensitive EOC

Despite undergoing optimal debulking surgery, first-line chemotherapy, and in some cases maintenance therapy, most patients diagnosed with EOC will experience a recurrence of disease clonally related to the original tumor [11]. If this happens within 6 months after completion of platinum-based chemotherapy, the patient’s tumor is considered platinum-resistant and in general further treatment with platinum drugs is deferred. If the duration of response exceeds 6 months, the patient’s tumor is considered platinum-sensitive and is a candidate for platinum-based combination therapy as second line [12]. However, the exact definition of platinum-sensitivity has varied over time and differs between studies with some considering recurrence within 6–12 months as partially platinum-sensitive [12].

For recurrent, platinum-sensitive disease, five studies tested PARPis in the maintenance phase during or following chemotherapy (ARIEL3 [13], NOVA [14], NORA [15], SOLO2 [16], and Study 19 [17]), two tested PARPi against non-platinum containing chemotherapy regimens (ICEBERG 3 [18] and SOLO3 [19]), and two tested concurrent PARPi and chemotherapy (Oza et al, 2015 [20] and Kummar et al, 2015 [21]). NRG-GY004 tested olaparib alone or in combination with cediranib against platinum-based chemotherapy regimens [22]. ARIEL4 was a randomized phase 3 trial testing rucaparib versus standard of care chemotherapy in patients with tumors harboring BRCA mutations and varying degrees of platinum sensitivity [23]. Patients were stratified according to their platinum-free interval and based on this were randomized to physician’s choice standard of care which included platinum if the patient’s tumor was deemed fully platinum-sensitive.

PARPi as maintenance therapy for recurrent platinum-sensitive EOC

Five studies evaluated PARPis as maintenance therapy in recurrent disease, two studied olaparib (Study 19 and SOLO 2), one rucaparib (ARIEL3), and two niraparib (NOVA and NORA). In each of these studies, participants had high-grade platinum-sensitive relapsed EOC, were in complete or partial response to their most recent platinum-based regimen, had ≥2 prior treatments, and were randomized to receive a PARPi as maintenance or a placebo. All showed a significant improvement in PFS compared with placebo, establishing a new standard of care for women with relapsed platinum-sensitive EOC. SOLO2 only enrolled patients with known germline BRCA mutations but Study 19, ARIEL3, NORA, and NOVA demonstrated benefit independent of BRCA mutation or HRD status leading to broad approval of these agents based on platinum sensitivity rather than specific biomarkers.

PARPi as chemotherapy for recurrent platinum-sensitive EOC

PARPis have also been investigated as monotherapy in comparison to other chemotherapeutic agents in recurrent EOC. For instance, ICEBERG 3 was a phase II trial with three arms, two of which involved different doses of olaparib and the third was the control arm of single agent pegylated liposomal doxorubicin (PLD) in BRCA mutant recurrent EOC [18]. Notably, in about 50% of the participants in this study, their disease had recurred within six months of platinum-based chemotherapy, not meeting the strict definition of platinum-sensitive. Although PFS was not significantly different between the groups in this trial, this may have been due to the unexpectedly high activity of PLD in BRCA mutant tumors and the small sample size.

SOLO3 also compared olaparib to single-agent, non-platinum-based chemotherapy in platinum-sensitive and BRCA mutation carriers who had received ≥2 prior lines of platinum-based chemotherapy. Options for chemotherapy in the control arm were limited to PLD, paclitaxel, gemcitabine, or topotecan. In the primary analysis, olaparib monotherapy improved ORR and PFS compared with single-agent non-platinum chemotherapy but in the final analysis, only time to second objective disease progression (PFS2) favored olaparib monotherapy and the OS was similar in both groups (median OS 34.9 (olaparib) versus 32.9 months (chemotherapy) HR 1.07, P=0.714) [24].

Oza el al. [20] tested the concurrent administration of olaparib with carboplatin and paclitaxel followed by olaparib alone as maintenance therapy. PFS was significantly longer in the olaparib plus chemotherapy group but the difference in median survival of 2.6 months came at the expense of additional toxicity. The benefit was more pronounced in the subgroup of patients with tumors harboring BRCA mutations. Similarly, Kummar et al. [21] examining veliparib in combination with non-platinum oral chemotherapy (cyclophosphamide) failed to show an increase in either PFS or ORR with the combination compared to cyclophosphamide alone.

PARPi in combination with anti-angiogenic agents

Apart from using PARPis alone as a maintenance therapy or as a component of a treatment regimen, PARPis have been tested in combination with anti-angiogenic agents in platinum-sensitive disease. NRG-GY004 was an NCI-sponsored, phase III trial of olaparib and cediranib, a VEGFR1–3 tyrosine kinase inhibitor, compared with olaparib alone or a control arm involving of physician’s choice carboplatin plus PLD, gemcitabine, or paclitaxel [22]. The combination olaparib/cediranib did not improve PFS compared with standard of care platinum-based chemotherapy in the intention-to-treat (ITT) population though it did show some clinical activity in patients with tumors harboring BRCA mutations [22]. In a post-hoc analysis, patients with HRD as assessed by BROCA-HR, had a median PFS of 12.3, 13.1, and 20.4 months for chemotherapy, olaparib, and olaparib/cediranib respectively [22].

Bevacizumab was also tested with niraparib in AVANOVA2, a randomized phase II superiority trial, which examined the combination in patients who had previously received platinum-containing therapy for their primary disease and one or fewer non-platinum-containing regimens for recurrent disease. The combination treatment resulted in a significant improvement in PFS, regardless of HRD status or the interval since the last chemotherapy treatment.

PARPi in tumors with varying degrees of HRD or chemotherapy sensitivity

Finally, ARIEL2 and ARIEL4 trials are unique in that they enrolled patients with varying degrees of HRD. ARIEL2 assessed other measures of HRD such as tumor genomic loss of heterozygosity in addition to BRCA mutation to predict response to rucaparib. ARIEL4 only enrolled patients with known BRCA mutations but with varying degrees of platinum sensitivity (platinum-sensitive [21%], partially platinum-sensitive [28%], and platinum-resistant [51%]). It was the only trial to prospectively assess the role BRCA reversion mutations on treatment response and platinum sensitivity. Patients in the chemotherapy arms whose tumors were deemed partially platinum-sensitive, or platinum-resistant received weekly paclitaxel, while patients whose tumors were fully platinum-sensitive were treated with a platinum single agent or combination therapy (such as carboplatin/paclitaxel, carboplatin/gemcitabine, or cisplatin/gemcitabine). Rucaparib treatment led to a higher PFS of 7.4 months compared to 5.7 months for chemotherapy, suggesting that rucaparib could be a viable alternative to chemotherapy for patients with relapsed ovarian carcinoma and tumors harboring BRCA1 or BRCA2 mutations.

Despite the positive signals in PFS from PARPis in the treatment of recurrent EOC, mature OS data from ARIEL4 and SOLO3 showed a potential detrimental effect of third-line rucaparib monotherapy and fourth-line olaparib monotherapy in EOC leading to voluntary withdrawals for these indications.

Recurrent platinum-resistant EOC

Platinum-resistant EOC is a hard-to-treat disease with limited treatment options. Due to the shared mechanisms of resistance between platinum drugs and PARPis, such as BRCA reversion mutation or drug efflux [25, 23, 26], only a limited number of clinical trials have tested a PARPi as monotherapy in those with platinum-resistant disease. However, five PARPi trials, ARIEL2 part 2, ARIEL4, CLIO, ICEBERG 3, and QUADRA allowed patients whose disease recurred less than six months after their last platinum-based chemotherapy to enroll. In the CLIO trial [27], olaparib was compared to a physician’s choice of chemotherapy, including PLD, topotecan, paclitaxel, or gemcitabine for patients with platinum-resistant disease, and carboplatin plus PLD or gemcitabine for patients with platinum-sensitive disease. Although the ORR was nearly double for the chemotherapy arm in the platinum-sensitive cohort versus olaparib alone (65% versus 35%), this result was not statistically significant. The ORR were similar for the platinum-resistant cohort, with a favorable disease response to olaparib found only in patients with disease that had been pre-treated with more than four prior lines of treatment. The disease in more than 90% of patients in the study was PARPi naive which could account for the superior ORR of olaparib compared with chemotherapy but limits the external validity of the findings in a post-PARPi age. The single-arm QUADRA [28] trial also demonstrated meaningful clinical benefit for patients whose disease had been heavily pre-treated (median 4 lines) using the PARPi niraparib. This trial included patients with both platinum-sensitive and platinum--resistant diseases. Not surprisingly, the effect was more pronounced in the population with HRD, platinum-sensitive disease.

Long-term survival data and recent changes to FDA approvals

Despite years of success in the treatment of EOC, PARPis have faced unexpected concerns recently, pertaining to long-term OS data from the clinical trials that led to their initial approvals. As a result, several sponsors voluntarily withdrew their PARPi approvals from regulatory agencies limiting options for EOC. The withdrawn approvals mainly involve the recurrent setting, both as maintenance therapy after completion of platinum-based treatment and as monotherapy replacing chemotherapy. There still is considerable uncertainty surrounding the future of PARPis in the treatment of EOC, particularly regarding their efficacy and their role within the standard of care.

The final OS analysis of ARIEL4 demonstrated a shorter median OS in the rucaparib arm (19.4 months) compared to standard treatment (25.4 months, HR 1.313; 95%CI 0.999–1.725) in patients with germline or somatic BRCA mutation and who had received at least two prior lines of chemotherapy [23]. However, the high crossover rate (69%) confounds interpretation of OS data, as nearly 90% of patients received rucaparib from randomization or after crossover. Despite this, no differences in PFS2 were observed (HR 0.860, 95%CI: 0.674–1.098) [29].

The SOLO3 study also showed similar concerning signals in their post-hoc subgroup analysis, where olaparib had a potential detrimental effect compared to standard treatment in patients with germline BRCA mutation-associated EOC who had received at least three prior lines of chemotherapy (HR 1.33, 95%CI: 0.84–2.18) [30]. In the final analysis, PFS2 slightly favored olaparib compared to placebo, although not significantly (23.6 months versus 19.6 months, HR 0.80; P=0.229) [24]. As a result, the approvals for rucaparib, olaparib, and niraparib for patients with heavily pre-treated BRCA-mutated EOC were withdrawn.

Concern extended to the recurrent maintenance setting as well. NOVA’s long-term follow-up data (median follow-up time: 5.6 years [68% mature]) revealed that patients whose tumors did not have BRCA mutation and received niraparib had a median OS (mOS) of 31.1 months compared to 36.5 months in the placebo group (HR 1.10, 95%CI: 0.831–1.459) and in the BRCA-mutated cohort, 43.6 months for the niraparib arm and 41.6 months for the placebo arm (HR 0.93, 95%CI: 0.633–1.355). Due to these results, niraparib was withdrawn as maintenance therapy in second or later lines of therapy in patients whose tumors did not harbor BRCA mutations. Post-progression treatment information was not available for 28% of patients, and about 25% of patients in the placebo arm received a PARPi in subsequent lines of therapy, confounding the survival data. To address this, the authors adjusted the OS analysis (inverse probability of censoring weighted [IPCW]) to control for subsequent PARPi use. After adjustment, no significant difference in mOS was observed between the niraparib and placebo arms in patients with BRCA wild-type (BRCAwt) tumors, instead, a trend for improved OS was seen with niraparib (43.8 versus 34.1 months; HR, IPCW: 0.66, 95%CI: 0.44–0.99) compared to the placebo arm [31].

In the ad-hoc, interim survival analysis of the NORA trial, however, no detrimental effect was observed [15]. NORA evaluated niraparib in platinum-sensitive recurrent EOC. At an OS maturity of 44% at data cut-off, the mOS was not significantly longer with niraparib (46.3 months) versus placebo (43.4 months) in the ITT population (HR 0.821, 95%CI: 0.558–1.207) showing that regardless of the subgroup, there were no detrimental effects with PARPi maintenance, but also no clear benefit. They reported a favorable trend for OS with niraparib in the IPCW-adjusted analysis, given the 43% rate of subsequent use of a PARPi after progression in the placebo and 20% in niraparib arms (mOS of 46.3 and 34.3 months in the niraparib and placebo arms respectively, HR 0.692 (0.447–1.072) [32].

Similarly, in the final OS analysis from the ARIEL 3 study, no OS benefit was observed with rucaparib. In the nested primary analysis of the ITT population, a mOS of 36 months was observed for patients treated with rucaparib as maintenance therapy versus mOS of 43.2 months in the placebo arm (HR 0.995, 95%CI: 0.809–1.223). The lack of an OS benefit in the HRD cohort with a HR of 1.005 (95%CI: 0.766–1.320) was not anticipated, with only an insignificant trend of benefit in the subgroup with BRCA mutations (HR 0.832, 95%CI: 0.581–1.192). Furthermore, in BRCAwt non-nested cohorts, HR>1 were found in both the HRD and non-HRD subgroups; in addition, the CIs for all HRs crossed 1 in all the non-nested cohorts, indicating no difference between the treatment arms (BRCAwt/HRD: HR 1.280, 95%CI: 0.841–1.948; BRCAwt/non-HRD: HR 1.153, 95%CI: 0.784–1.695; BRCAwt/HR-unknown: HR 0.673, 95%CI: 0.305–1.483). Based on those data, the FDA requested the sponsor to voluntarily revise the second-line maintenance treatment indication, restricting rucaparib maintenance treatment to only patients with tumors harboring BRCA mutations. Favorable post-progression outcomes were observed in the nested analysis among all the cohorts, but nearly half of the patients assigned to the placebo arm received a PARPi in the subsequent lines. PFS2 was significantly longer for patients treated with rucaparib, compared with placebo, with HRs and 95%CIs <1, indicating a clinical PFS benefit among all the subgroups (ITT population: HR 0.703, 95%CI, 0.579–0.854; HRD subgroup: HR 0.718, 95%CI: 0.558–0.923; BRCA-mutant cohort: HR 0.672, 95%CI: 0.480–0.941) [33].

While updated survival data in the recurrent setting has been underwhelming, the mature data from the first-line maintenance setting offers some reassurance. The 7-year follow-up of the SOLO1 study revealed a clinically meaningful survival benefit for patients with tumors harboring BRCA mutations who received olaparib as first-line maintenance therapy, with a median treatment duration of 24.6 months in the olaparib group compared to 13.9 months in the placebo group (HR for OS of 0.55; 95%CI: 0.40–0.76) [34]. Additionally, the prespecified 5-year OS analysis of the PAOLA-1 study showed a clinically meaningful benefit from the olaparib and bevacizumab combination therapy for the HRD population (HR 0.62, 95%CI: 0.45–0.85; OS at 5 years, 65.5% versus 48.4%), consistent with previously reported PFS data [35]. When OS data from PRIMA and ATHENA-MONO [8] are available we can assess whether the benefit of PARPis in the first-line setting extends to all patients.

Conclusions

The mature OS data suggests that some patients will not derive a benefit from PARPis. There is a need for biomarker-driven treatments in molecularly selected subgroups in the post-PARPi setting [24]. The post-hoc analysis of SOLO2 showed a reduced efficacy of subsequent platinum therapy in patients who had received olaparib maintenance versus placebo (7 months versus 14.6 months, HR 2.33, 95%CI: 1.27–4.28), suggesting a cross-resistance between PARPi and platinum agents [36]. Patients may transiently respond to a PARPi, manifested as an increased PFS, but if it comes at the expense of the tumor’s platinum sensitivity, then their OS may suffer and be reduced [37].

OS is a challenging outcome to measure, particularly for patients with EOC with extended post-progression survival rates. The constantly evolving advancements in the management of recurrent disease, and unexpectedly high mOS rates in the placebo groups, may have led to biased OS data in the PARPi treatment arms. In these studies, OS was a secondary and underpowered endpoint and high crossover rates in the control group and/or the use of a PARPi in subsequent treatment lines for patients randomized to the placebo complicates long term data interpretation.

Accurate endpoints are needed in the recurrent setting where PARPi may be given indefinitely. PFS2, or the time to subsequent therapy, have been proposed as potentially valuable surrogate outcomes to measure clinical benefit, when looking to decide if OS data that is not statistically significant can be argued to be clinically meaningful.

Moving forward, prospective, pre-planned biomarker analysis should be used to select patients who are most likely to benefit and to establish treatment sequence and strategies to overcome resistance (reviews on resistance mechanisms to PARPi including EOC specific considerations [26, 38] and general resistance mechanisms [39, 40] are outside the scope of this article but the topic is covered elsewhere in this special edition). Ongoing Phase I/II trials testing a PARPi in combination with cell cycle checkpoints blockade (ATR inhibitors (i.e., ceralasertib, M4344, berzosertib, BAY1895344, camonsertib), CHK1 inhibitor (i.e., prexasertib (a.k.a. ACR-3680), and WEE1 inhibitor i.e., adavosertib) are being explored as one therapeutic approach for cancers resistant to PARPis [41]. Given the clinically meaningful benefit of a PARPi as a first-line maintenance therapy, extensive use in this setting is expected.

Considering the emerging evidence of cross-resistance, alternative options for platinum-based chemotherapy in patients whose disease progresses on or after a PARPi should be investigated. In this context, translational research will be a crucial tool to guide the subsequent treatment strategy based on the specific mechanism of resistance detected in patient samples. Finally, OS data from trials evaluating PARPis in the non-HRD population are needed to elucidate the magnitude of clinical benefit in those patients with the worst prognosis among EOC patients and to decide whether new treatment strategies should be explored to improve survival rates even in this population.

Abbreviations—

BRCAwt

BRCA wild-type

EOC

epithelial ovarian cancer

FDA

Food and Drug Administration

HGSOC

high-grade serous ovarian cancer

HRD

homologous recombination deficiency

mOS

median overall survival

ORR

overall response rate

OS

overall survival

PARPi

poly-ADP- ribose polymerase inhibitor

PFS

progression free survival

PFS2

time to second objective disease progression

PLD

pegylated liposomal doxorubicin