Bevacizumab is an intravenously (i.v.) administered recombinant humanized monoclonal IgG1 antibody that targets VEGF-A, with clinical benefits in patients with metastatic colorectal cancer, non-small cell lung cancer, and breast cancer [13]. This drug binds and neutralizes all biologically active forms of VEGF-A (e.g., VEGF-A165), thus suppressing tumor growth and inhibiting metastatic disease progression by inhibiting neovascularization and inducing existing microvessel regression [14, 15]. Bevacizumab also normalizes tumor vessels that are structurally and functionally abnormal. These morphological changes lead to functional changes (e.g., decreased interstitial fluid pressure, increased tumor oxygenation, improved drug penetration in tumors) that may enhance the effects of chemotherapy [16].

The phase II trials Gynecologic Oncology Group (GOG)-0170D and AVF2949 evaluated bevacizumab as a monotherapy for recurrent ovarian cancer and yielded favorable results, with response rates of 16–21% [17, 18] and hypertension and proteinuria as the most common grade 3/4 adverse events. Although no gastrointestinal (GI) perforation was observed in patients of the GOG-0170D study who had received one or two previous regimens, the AVF2949 trial observed GI perforation in five patients (11.4%) previously subjected to heavy treatment (three or more prior regimens).

Two landmark phase III trials of bevacizumab for ovarian cancer, GOG-0218 and International Collaborative Ovarian Neoplasm (ICON) 7, were conducted in a first-line/adjuvant chemotherapy setting (Table 3.1) [19, 20]. In the GOG-0218 trial, patients who received combination chemotherapy (paclitaxel/carboplatin) plus bevacizumab (15 mg/kg) for six cycles and maintenance bevacizumab for 16 cycles had a significantly longer progression-free survival (PFS) than those who received first-line chemotherapy alone (median PFS: 10.3 vs. 14.1 months) [19]. However, no statistically significant difference was observed in overall survival (OS). Similarly, patients in the ICON 7 trial were randomized to chemotherapy alone (carboplatin/paclitaxel) or plus bevacizumab (7.5 mg/kg) for six cycles, with 12 cycles of maintenance bevacizumab [20]. The latest report observed a significantly prolonged restricted mean survival time among poor-prognosis patients in the bevacizumab group vs. the chemotherapy group (39.3 vs. 34.5 months), although no OS benefit of bevacizumab was recorded [21].

A randomized trial, GOG-0262, evaluated the optimal combination of bevacizumab with dose-dense therapy (weekly paclitaxel plus carboplatin every 3 weeks) and conventional dose therapy (paclitaxel/carboplatin every 3 weeks; Table 3.1) [22]. Both groups of patients who opted to receive bevacizumab had a similar PFS, although among patients who did not receive bevacizumab, the medium PFS was 3.9 months longer with dose-dense therapy vs. conventional dose therapy (14.2 vs. 10.3 months).

Two phase III trials, OCEANS (Ovarian Cancer Study Comparing Efficacy and Safety of Chemotherapy and Anti- Angiogenic Therapy in Platinum-Sensitive Recurrent Disease) and AURELIA (Avastin Use in Platinum-Resistant Epithelial Ovarian Cancer), were conducted to evaluate recurrent disease (Table 3.1). Both trials evaluated the effect of bevacizumab in combination with chemotherapy and observed improvements in PFS [23, 24]. In the OCEANS study, platinum-sensitive recurrent ovarian cancer patients received six cycles of carboplatin/gemcitabine in combination with bevacizumab, followed by maintenance bevacizumab, and patients receiving bevacizumab had a significantly longer PFS vs. the control arm (median PFS: 12.4 vs. 8.4 months) [23]. However, the final OS analysis revealed no significant difference between the treatment arms [25]. In the AURELIA study, platinum-resistant recurrent ovarian cancer patients received single-agent chemotherapy [pegylated liposomal doxorubicin (PLD), weekly paclitaxel, or topotecan] alone or with bevacizumab. PFS was significantly improved in the chemotherapy plus bevacizumab arm vs. the chemotherapy arm (median PFS: 6.7 vs. 3.4 months). However, the trend toward improved OS (median: 16.6 vs. 13.3 months) was not statistically significant. GI perforation was observed only in the bevacizumab arm (2.2%), although the risk was lower than expected.

Several other ongoing phase III trials are investigating the optimal use of bevacizumab. The GOG-0252 study is evaluating the efficacy of bevacizumab in combination with intraperitoneal (i.p.) chemotherapy (i.v. paclitaxel and i.p. cisplatin or carboplatin) vs. i.v. chemotherapy (paclitaxel/carboplatin). The AGO-OVAR 17 (BOOST) trial is investigating the optimal bevacizumab treatment duration (15 vs. 30 months) with first-line chemotherapy (paclitaxel/carboplatin). GOG-0213, a study on platinum-sensitive recurrent disease, is comparing chemotherapy (paclitaxel/carboplatin) alone vs. with bevacizumab; surgical candidates in this cohort will undergo secondary randomization to surgery or no surgery.

Aflibercept (VEGF Trap) is a fusion protein of the Fc region of immunoglobulin G1 with domain two of VEGFR1 and domain three of VEGFR2 (VEGFR_δ1R2). This decoy receptor binds with high affinity to VEGF-A, thus preventing VEGFR1 and VEGFR2 binding and subsequent stimulation [26]. Aflibercept also exhibits a strong binding affinity for VEGF-B and PlGF.

Two phase II studies of platinum-resistant disease and symptomatic malignant ascites have been conducted (Table 3.2) [27, 28], and both demonstrated effective control of malignant ascites with aflibercept, evidenced by a reduction in the interval between repeat paracenteses (e.g., 55.1 vs. 23.3 days) [28]. However, one study observed a higher frequency of fatal GI events in the aflibercept arm (3/29 patients) vs. the placebo arm (1/26 patients) [28]. In the other study, platinum-resistant ovarian cancer patients were randomized to receive aflibercept at different doses (2 or 4 mg/kg) (Table 3.2) [29]. Although aflibercept was generally well tolerated at both doses, the response rate was low.

A phase I–II study of aflibercept in combination with docetaxel was conducted in patients with recurrent ovarian cancer [30]. The objective response rate was 54%, and grade 1–2 hypertension (11%) and grade 2 hypotension (2%) were adverse events specifically associated with aflibercept. Therefore, the combination of aflibercept and docetaxel seems safe and active for patients with recurrent ovarian cancer.

Cediranib is a highly potent, small-molecule, oral tyrosine kinase inhibitor of all three VEGF receptors (VEGFR1–3) and c-Kit, which competes for the ATP-binding site within the receptor kinase domain [31, 32]. A phase II trial of cediranib in patients with recurrent ovarian cancer reported a partial response (PR) rate of 17% and median PFS of 5.4 months [33]. These promising results led to a phase III study (ICON6) of patients with platinum-sensitive recurrent disease (Table 3.1) in which the median PFS was significantly prolonged in the platinum-based chemotherapy plus concurrent and maintenance cediranib arm (arm 3) vs. the chemotherapy and placebo (arm 1) (11.0 vs. 8.7 months) [34]. Although the OS analysis is ongoing, early median OS durations for arms 1 and 3 were 21.0 months and 26.3 months, respectively (P = 0.11).

Pazopanib is a potent, selective oral multi-targeted receptor tyrosine kinase inhibitor of VEGFR1–3, platelet-derived growth factor receptor (PDGFR)-α and PDGFR-β, and fibroblast growth factor receptor (FGFR) 1–3 [35]. A phase II study (VEG104450) of pazopanib in patients with recurrent ovarian cancer reported a PR rate of 18% [36]. In a phase III trial (AGO-OVAR16), patients with International Federation Gynecology Obstetrics (FIGO) stage II–IV ovarian cancer received maintenance pazopanib or placebo for up to 24 months (Table 3.1) [37]. PFS was significantly prolonged for patients in the maintenance pazopanib arm vs. those in the placebo arm (median PFS: 17.9 vs. 12.3 months), although OS did not differ significantly between the arms at the interim analysis. In a randomized phase II trial (MITO 11), patients with platinum-resistant ovarian cancer received weekly paclitaxel with or without pazopanib; PFS was significantly longer in the paclitaxel/pazopanib group vs. the paclitaxel-only group (median PFS: 6.35 vs. 3.49 months) (Table 3.2) [38]. Two randomized phase II trials to evaluate chemotherapy (paclitaxel or gemcitabine) and combined effects with pazopanib are ongoing in patients with platinum-resistant ovarian cancer.

Nintedanib (BIBF 1120) is a potent, oral tyrosine kinase inhibitor of VEGFR1–3, PDGFR-α and -β, and FGFR1–3. In a placebo-controlled randomized phase II trial of post-chemotherapy maintenance therapy in patients with relapsed ovarian cancer, nintedanib was well tolerated and associated with a potential improvement in PFS (Table 3.2) [39]. A phase III trial (AGO-OVAR12) investigated the combination of standard chemotherapy (paclitaxel/carboplatin) with nintedanib or placebo in patients with newly diagnosed FIGO stage IIB–IV ovarian cancer (Table 3.1) [40] and observed a significantly longer median PFS in the nintedanib group vs. the placebo group (17.2 vs. 16.6 months). The efficacy of nintedanib was particularly notable in patients with a low postsurgical disease burden (FIGO stage IIB–III, ≤1 cm residual postoperative tumor). Although the OS results are pending, further studies are needed to assess the clinical value of nintedanib, particularly in cohorts with lower tumor burdens.

Sunitinib is a potent, oral multi-tyrosine kinase inhibitor that targets VEGFR1–3, PDGFR-α and -β, Flt-3, and c-Kit [41]. Three phase II trials were conducted to evaluate the efficacy and safety of this inhibitor in patients with recurrent ovarian cancer (Table 3.2) [42–44]. However, efficacy seemed to be limited, with response rates of 3–17%, and the common adverse events included hypertension, gastrointestinal events, fatigue, and hand–foot syndrome.

Sorafenib is an oral bis-aryl urea that inhibits c-Raf and b-Raf kinases and VEGFR-2 and -3, PDGFR-β, Flt-3, and c-Kit [45]. In a phase II trial (GOG-0170F) of sorafenib for patients with recurrent ovarian cancer, PR rate of 3% and median PFS and OS of 2.1 months and 16.3 months, respectively, were achieved [46]. However, a randomized phase II study of sorafenib maintenance therapy observed no significant difference in PFS between the sorafenib and placebo arms (Table 3.2) [47].

In tumor angiogenesis, angiopoietin-1 and angiopoietin-2 interact with the tyrosine kinase with immunoglobulin-like and EGF-like domains (Tie) 2 receptor, which is expressed on endothelial cells, to mediate blood vessel maturation and stabilization in a VEGF axis-independent pathway [48]. Trebananib (AMG 386), a neutralizing peptibody (i.e., peptide-Fc fusion protein), blocks the binding of both angiopoietin-1 and angiopoietin-2 to the Tie2 receptor, thereby inhibiting angiogenesis [49].

In a randomized phase II trial, trebananib combined with weekly paclitaxel prolonged PFS in patients with recurrent ovarian cancer (Table 3.2) [50]. A phase III trial, Trebananib in Ovarian Cancer-1 (TRINOVA-1), investigated trebananib in addition to single-agent weekly paclitaxel for patients with recurrent ovarian cancer (Table 3.1) [51]. Median PFS was significantly longer in the paclitaxel/trebananib group vs. the paclitaxel/placebo group (7.2 vs. 5.4 months), although the median OS did not statistically differ. Two subsequent phase III trials are ongoing: TRINOVA-2, which evaluates trebananib plus PLD for recurrent, partially platinum-sensitive ovarian cancer, and TRINOVA-3, which investigates trebananib plus first-line chemotherapy (carboplatin/paclitaxel) for FIGO stage III–IV ovarian cancer.

PARPs have multiple functions, including DNA repair, cell dysfunction and necrosis, and inflammation (Fig. 3.2) [52]. PARP-1, the most abundant nuclear isoform, plays a vital role in DNA single-strand break (SSB) repair through the base excision repair pathway, whereas residual PARP activity (approximately 10%) is attributed to PARP-2. PARP inhibition causes an accumulation of DNA SSBs and consequent DNA double-strand breaks (DSBs) at replication forks. In normal cells, such DSBs are generally repaired by the BRCA1- and BRCA2-dependent homologous recombination (HR) DNA repair pathway. However, these lesions are not repaired in BRCA1- or BRCA2-deficient tumor cells, leading to genomic instability and cell death despite the existence of an alternate non-homologous end-joining pathway for DSB repair.

Female carriers of germline mutations in BRCA1 on chromosome 17q21 or BRCA2 on chromosome 13q31 have a higher risk of breast and ovarian cancer development. The lifetime risks of ovarian cancer are 54% for BRCA1 and 23% for BRCA2 mutation carriers [53]. Although germline mutations in those genes are seen in 5–10% of all ovarian cancer patients, a loss of HR function (BRCAness), either via genetic or epigenetic events in BRCA1 or BRCA2 or alterations in other genes (e.g., EMSY, PTEN, RAD51C, ATM, ATR, Fanconi anemia genes), are observed in approximately half of high-grade serous ovarian carcinomas [54]. In ovarian cancer, a BRCAness profile may correlate with responses to platinum-based chemotherapy and PARP inhibitors.

The epidermal growth factor receptor (EGFR; HER in humans) family comprises four distinct transmembrane tyrosine kinase receptors: HER-1 (EGFR/erbB1), HER-2/neu (erbB2), HER-3 (erbB3), and HER-4 (erbB4) [64]. These receptors are activated via C-terminal autophosphorylation by ligand binding (although HER2 has no known ligand) and multiple receptor homo- or hetero-dimerization combinations, thus triggering downstream signaling pathways such as the MAPK and PI3K/Akt pathways and thus inducing cancer-cell proliferation, blocking apoptosis, activating invasion and metastasis, and stimulating tumor-induced neovascularization. Accordingly, HERs are attractive targets for anticancer therapies (Fig. 3.3) [64].

HER family members are expressed in many human malignancies, including ovarian cancers, in which a wide range of HER family expression has been reported [EFGR, 4–100% (average, 48%); HER-2, 0–100% (average, 40%); HER-3, 3–90% (average, 48%); and HER-4, 45–92% (average, 71%)] [65]. Overexpression of HER, particularly EGFR and HER-2, may correlate with poor prognosis and decreased therapeutic response, although clinical data are contradictory. Several EGFR and HER-2 inhibitors have been tested in patients with ovarian cancer.