Immunotherapy has emerged as a major therapeutic modality in oncology. However, the majority of patients with women’s cancers do not derive benefit from immune checkpoint blockade monotherapy, creating the need to optimize combination treatment strategies [1].

Data suggest increased DNA damage by radiation or DNA repair inhibitors promotes local antigen release resulting in systemic anti-tumor immune responses [2]. Such neoantigen release and high tumor mutational burden (TMB) are shown to be associated with clinical response to immune checkpoint inhibition in solid tumors [3].

PARP inhibitors (PARPi) are one of the active new drug families in the drug armamentarium for women’s cancers and may modulate immune responses. PARPi, such as olaparib and talazoparib were shown to induce PD-L1 expression in breast cancer in vitro and in vivo models [4, 5]. Also, increased DNA damages from PARPi exposure may yield greater TMB, potentially increasing neoantigens, and affect the immune milieu, complementing the clinical benefit of immune checkpoint blockade in subsets of recurrent women’s cancer.

Angiogenesis pathways interact with both DNA repair mechanisms and immune activity. Tumor hypoxia induces downregulation of genes involved in DNA repair, e.g., RAD51 and BRCA1, leading to further DNA damages, genomic instability, and cell death [6]. VEGF suppresses lymphocyte trafficking across endothelia into tumor deposits and sites of inflammation to promote vessel growth [7]. Combining inhibition of DNA repair and angiogenesis pathways therefore may modulate the immune response by increasing DNA damage and TMB and by attenuating immunosuppressive microenvironments.

The combination of olaparib and cediranib, a VEGFR1–3 inhibitor, has been demonstrated to be clinically superior to olaparib monotherapy in recurrent platinum-sensitive ovarian cancer [8]. We extended this concept with our hypothesis that reduced VEGF signaling by cediranib and increased DNA damages by olaparib would complement anti-tumor activity of the immune checkpoint inhibitor. Durvalumab (MEDI4736) is a selective, high-affinity human IgG1 monoclonal antibody that blocks PD-L1 binding to PD-1 and CD80, thereby enhancing the function of tumor-directed T cells [9]. We previously reported the safety data and recommended phase 2 doses (RP2D) of the doublets of durvalumab in combination with olaparib or intermittent cediranib in recurrent women’s cancers [10]. We now report safety, RP2D and pharmacokinetics (PK)/pharmacodynamics (PD) findings of the 3-pathway modulation using durvalumab in combination with olaparib and cediranib.

This is the first reported 3-drug combination therapy of immune checkpoint blockade with a PARPi and a VEGFR inhibitor. This study established the RP2D of 3-drug combination and demonstrates preliminary evidence of activity in heavily pretreated recurrent gynecologic cancer patients, particularly in ovarian cancer. We previously reported durvalumab with an intermittent cediranib schedule had greater tolerability without attenuating clinical benefit observed in the daily schedule [10]. We now observe a full dose olaparib can be added to that combination in a safe manner using prospective toxicity mitigation approaches. Unlike our prior observations where durvalumab affected Cmax and clearance of daily cediranib [10], PK parameters of cediranib and olaparib were not affected by the presence of durvalumab in this 3-drug combination.

The current study presents durable long-term activity and safety outcome of the 3-pathway modulation with a median of 8 months follow-up (range 1 to 26+ months) and a RR worthy of further investigation. Single agent olaparib has limited activity (4%) in heavily pretreated BRCA wild-type patients with platinum-resistant ovarian cancer [15]. Cediranib monotherapy also demonstrated a RR of 17% and a median PFS of 5.2 months in patients with platinum-resistant ovarian cancer, treated in the 2nd or 3rd line setting [16]. However, our data should be interpreted with a caution due to its small sample size and phase 1 single-arm dose escalation design, thereby preventing direct comparison of the combination with either drug alone.

There are limited published data on anti-tumor activity and safety of PD-1/PD-L1 blockade combinations with a PARPi or VEGF/VEGFR inhibition in women’s cancer. Preliminary results from two PARPi and PD-1/PD-L1 blockade combination trials were presented showing early clinical activity in subsets of recurrent ovarian cancer; 25% RR of niraparib and pembrolizumab in platinum-resistant ovarian cancer [17] and 72% RR of durvalumab and olaparib in germline BRCA mutant platinum-sensitive ovarian cancer patients [18]. Our group reported, at the 2018 European Society for Medical Oncology (ESMO) meeting, a 14% RR and 37% clinical benefit rate (as defined in Methods) of the combination durvalumab and olaparib, both in full doses, in heavily pretreated ovarian cancer patients, predominantly composed of BRCA wild-type and platinum-resistant disease [19]. A RR of 25% with niraparib and pembrolizumab was shown in women with platinum-resistant ovarian cancer [17], and in germline BRCA mutant platinum-sensitive ovarian cancer, a 72% RR with durvalumab and olaparib [18]. The single arm phase II study of bevacizumab and nivolumab in recurrent ovarian cancer presented at ESMO 2018 reported 16.7% RR and 55.3% CBR (defined by CR + PR + SD ≥ 6 months), with a median PFS of 5.3 months in platinum-resistant patients (18 patients) and 40% RR and 75% CBR, with a median PFS of 9.4 months in those with platinum-sensitive disease (20 patients), respectively [20]. These findings suggest a modest benefit of addition of either PARPi or bevacizumab to PD-1/PD-L1 blockade in germline BRCA wild-type ovarian cancer patients, warranting further investigation of our 3-drug combination strategy. Separately, phase 3 randomized trials of 3-pathway modulation (immune checkpoint inhibitor in combination with PARPi and VEGF inhibition) are now being investigated in the front-line setting for ovarian cancer treatment (e.g., avelumab (PD-L1 inhibitor) + talazoparib (PARPi) + bevacizumab, ClinicalTrials.gov identifier: NCT03642132).

Defining predictive biomarkers of immune checkpoint inhibitor-based combination therapy is a major challenge. Our study suggests that presence of PD-L1 positive carcinoma cells in archival tissues may be associated with clinical benefit. In contrast, peripheral immune subsets analysis showed a transient increase of PD-L1 expression on total CD14+ monocytes after 2 weeks treatment of olaparib and cediranib although it did not last after 2 months treatment. An association of tumoral PD-L1 expression and response to immune checkpoint inhibitor therapy has been reported in ovarian carcinoma [21] and triple negative breast cancer (TNBC) [22]. The correlation of clinical benefit and expression of PD-L1 using archival tissues needs further validation, particularly in recurrent women’s cancers given preanalytical variabilities (e.g., different IHC techniques, score systems, different antibodies [23, 24]) and possible dynamic changes of PD-L1 expression affected by prior therapies [25]. The present study is now being expanded to a phase 2 study for patients with recurrent ovarian cancer. We will prospectively examine the association of PD-L1 induction and TMB by PARPi with clinical response to immune checkpoint blockade using fresh core biopsies collected before and on-treatment.

In summary, our study demonstrates that the 3-drug combination is tolerable and active in heavily pretreated recurrent gynecologic cancer patients without germline BRCA mutation. This response may be associated with carcinoma cell PD-L1 labeling, but this requires further validation. The preliminary activity findings warrant further investigation and a single-arm phase II expansion study is now open to accrual for recurrent ovarian cancer patients (ClinicalTrials.gov identifier: NCT02484404).