Moving Beyond Standard of Care: Novel Treatments in Gynecologic Cancer

,
Contemporary Oncology®, Spring 2010, Volume 2, Issue 1

In the United States, gynecologic cancers affect nearly 80,000 women annually and are the cause of death in more than 25,000 women.

In the United States, gynecologic cancers affect nearly 80,000 women annually and are the cause of death in more than 25,000 women.1 Cancers arising in the uterus are the most common, with 42,000 cases in 2009, followed by 21,500 cases of ovarian cancer, and 11,270 cases of cervical cancer.1 Although ovarian cancer is not the most common gynecologic cancer, it remains the most lethal. More women die each year from ovarian cancer than from uterine and cervical cancers combined.

Chemotherapy in the adjuvant setting is the standard of care for most gynecologic malignancies. For ovarian cancer, chemotherapy consists of a platinum and taxane. It is indicated in women with grade 3 stage IB or IC disease and in all women with a diagnosis of stage II disease or higher.2 In advanced uterine cancer, randomized trials have shown that chemotherapy is superior to radiation therapy in the adjuvant setting. Based on completed clinical trials, standard of care consists of a 3-drug combination of doxorubicin, cisplatin, and paclitaxel.3 This regimen, however, is associated with excessive toxicities, so most centers use carboplatin and paclitaxel in an alternative regimen. Recently, a randomized trial comparing these 2 treatments completed enrollment and results are eagerly anticipated.4

In cervical cancer, chemoradiation with cisplatin is a curative regimen in women with locally advanced disease and can be used as primary therapy in lieu of radical hysterectomy. A recent trial from South America compared chemoradiation with a combined approach of chemoradiation followed by systemic combination chemotherapy using cisplatin and gemcitabine (Gemzar).5 At the American Society of Clinical Oncology (ASCO) meeting in 2009, the study investigators reported that adjuvant systemic chemotherapy improved disease-free and overall survival in women with cervical cancer. This raises the possibility that adjuvant chemotherapy may someday play a larger role in the treatment of cervical cancer.

The use of adjuvant or first-line chemotherapy has led to improvement in survival outcomes for women with gynecologic cancers. Still, for those women who experience relapse, the disease is often incurable and most will likely succumb to it. While there are treatment options for recurrent gynecologic cancers, trials have generally focused on ovarian cancer, leaving it up to patients and their providers to extrapolate from the data how the treatments might apply to cervical or uterine cancers.A number of agents are approved to treat first or subsequent ovarian cancer relapses. In endometrial cancer, however, the US Food and Drug Administration has not approved any agent for second-line therapy. Thus, there is a desperate need to broaden options and improve the outcomes for women with recurrent disease. Several promising treatments are in development for gynecologic cancers.

Bevacizumab (Avastin)

Bevacizumab is a recombinant humanized monoclonal antibody directed against vascular endothelial growth factor (VEGF)-alpha.6,7 It is approved for use in colorectal cancer, non—small cell lung cancer, and breast cancer. Recently completed trials indicate it may also be highly active in gynecologic cancers (Table 1).

Burger and colleagues reported the results of Gynecologic Oncology Group (GOG) 170D, a phase II trial of bevacizumab in women with recurrent ovarian cancer.8 Eligible patients had measurable or assessable disease and had received up to 2 previous chemo-therapy regimens. Platinum-sensitive patients with a treatment-free interval of ≥12 months were required to have been re-treated with a platinum-based regimen. The study accrued 64 patients, of which 62 were evaluable. The overall response rate (ORR) was 21% (90% confidence interval [CI], 12.9%-31.3%), and 32 (52%) patients had stable disease as their best response. At 6-month follow-up, 40% had progression-free survival (PFS; 90% CI, 30%-53.6%). In this study, grade 3-4 gastrointestinal (GI) toxicity was rare, but 9.7% of patients developed grade 3-4 hypertension. Overall, toxicities were primarily grade 1-2. The most common adverse events were pain (50%), constitutional symptoms (45.6%), hepatic toxicity (33%), renal complications (30%), proteinuria (30%), anemia (30.6%), leukopenia (29%), hemorrhage (23%), diarrhea (21%), constipation (20%), coagulation problems (18%), nausea (15%), and hypertension (13%). While there were no reports of GI perforation in this study, a similar study conducted by Genentech was halted after 11% of participants experienced GI perforation.9 Prior treatment appeared to correlate with an increased risk of GI perforation; 24% of patients treated with 3 previous regimens experienced perforation, while no perforations occurred in patients who received ≤2 prior regimens.

Two large randomized trials have recently closed; one evaluated bevacizumab as part of an upfront treatment strategy with carboplatin and paclitaxel (GOG 218), and the other (OCEANS) evaluated bevacizumab in women with relapsed platinum-sensitive recurrent ovarian cancer using a chemotherapy backbone of carboplatin and gemcitabine.10,11 Bevacizumab is also being used in GOG 252, a phase III, randomized, 3-arm trial evaluating 2 different intraperitoneal regimens compared with standard intravenous therapy. Bevacizumab will be administered in all arms concomitant to chemotherapy and for 52 weeks after chemotherapy (Figure 1).12 The GOG 213 study will randomize the “platinum-sensitive” population of women who experience relapse ≥6 months from the end of primary treatment to carboplatin and paclitaxel, with or without bevacizumab (Figure 2).13

The data on the effects of bevacizumab on gynecologic malignancies extends beyond ovarian cancer. In GOG 227C, Monk and colleagues studied single-agent bevacizumab in women with previously treated cervical cancer.14 Eligible patients had recurrent squamous or adenosquamous carcinomas of the cervix with measurable disease and were permitted to have had prior treatment with cisplatin-based chemoradiation and ≤2 prior cytotoxic regimens. Investigators enrolled 50 patients, of which 46 were treated. The ORR was 11%, with 24% of patients having PFS of ≥6 months. Toxicities in this study were similar to those reported by Burger and colleagues, with few grade 3 or 4 events. There was one death due to a grade 5 infection, possibly treatment related. These results led to GOG 240, a randomized trial that is currently enrolling patients with recurrent cervical cancer. Participants will be randomized to topotecan (Hycamtin)/paclitaxel versus cisplatin/paclitaxel, with a second randomization to bevacizumab or placebo (Figure 3).15

Finally, at the 2009 ASCO annual meeting, Aghajanian and colleagues presented data from GOG 229E, a phase II trial of bevacizumab in 56 women who had received ≤2 prior lines of cytotoxic treatment for recurrent or metastatic endometrial cancer.16 There were 53 evaluable patients, and 53% had received radiation therapy. Bevacizumab produced a 13% ORR and a 6-month PFS rate of 40%. Further evaluation in this population is ongoing in GOG 86P (Figure 4),

a phase II randomized trial evaluating a carboplatin/tubulin-stabilizer backbone with either bevacizumab or temsirolimus (Torisel).17

Aflibercept (VEGF-Trap)

Unlike bevacizumab, aflibercept is a fusion protein between the Fc region of human immunoglobulin (Ig) G1 and the extracellular domains of human VEGF-receptors 1 (VEGFR1) and 2 (VEGFR2). In 2007, Tew reported preliminary results of a multicenter trial, after 162 women out of a planned 200 were accrued and treated with afilbercept.18 All the women had platinum-resistant ovarian cancer and had been treated with ≤4 prior lines of chemotherapy. Radiologic criteria indicated that 13 patients (8%) achieved partial response (PR); 21 patients (13%) had at least a 50% decline in levels of cancer antigen (CA)-125. Adverse effects were similar to those associated with bevacizumab.

Another pilot study evaluated the ability of aflibercept to reduce the rate of repeated paracentesis in women with recurrent ovarian cancer and recurrent ascites.19 Of 10 patients enrolled at the time of presentation, 8 met the primary criteria for response, defined as a doubling in time to first paracentesis on study compared with the baseline average before the study. There appeared to be considerable toxicity, with a 40% rate of bowel obstruction, 30% rate of severe nausea and vomiting, and 20% rate of anorexia, edema, or global clinical deterioration. While data on the activity of afilbercept in other gynecologic neoplasms has yet to be presented, GOG 229F—a phase II trial investigating afilbercept in women with recurrent or metastatic endometrial cancer—has completed enrollment.20

Fosbretabulin (Zybrestat)

Moving beyond the antiangiogenic approach of VEGFR inhibition to block the formation of tumor vasculature, a new mechanism has emerged. This new class of drugs, termed vascular disrupting agents (VDAs), targets pericytes and endothelial cells of established tumor vasculature.21 Fosbretabulin, which is a combretastatin A4 phosphate (CA4P), is one such VDA in clinical development. It causes microtubule destabilization by binding the beta subunit of endothelial tubulin, resulting in disruption of the endothelial cytoskeleton. Additionally, fosbretabulin selectively disrupts cell-cell interaction of the vascular endothelial cadherin/beta-catenin complex, resulting in increased vascular permeability and reduced blood flow to the tumor.22

Fosbretabulin has been under investigation in ovarian cancer, with results presented at the 2009 ASCO Annual Meeting. A phase II trial enrolled women with platinum-resistant ovarian cancer, which was strictly defined as a treatment-free interval of <6 months from the end of previous platinum treatment to the start of trial therapy.23 The regimen called for carboplatin area under the curve equal to 5 and paclitaxel 175 mg/m2 delivered on day 1, with fosbretabulin 63 mg/m2 delivered 24 hours later. Of the 44 patients treated, 20% were taxane-naïve. The ORR by radiographic assessment or CA-125 criteria was 25%. Grade 3-4 pain was seen in >10% of patients. Nearly one-quarter (23%) of patients developed hypertension ~1 hour after administration of fosbretabulin, but all cases were grade 1-2 in severity and resolved within 4 hours. No significant or progressive increase in blood pressure was observed with successive courses of therapy, nor was there significant hematologic toxicity beyond what would be expected with chemotherapy alone. The GOG plans to further test this agent.

Numerous other anti-vascular therapies are also under development, with more recent agents broadening beyond sole blockade of VEGF and exerting inhibition on other proteins, such as those against receptors for platelet-derived growth factor or fibroblast growth factor.Some target other angiogenic receptors, such as the platelet-derived growth factor receptor and fibroblast growth factor receptor. Others operate on other pathways, including those mediated by c-Kit and RAS/RAF (Table 2).

Belinostat (PXD-101)

Belinostat is a hydroxymate type I histone deactylase (HDAC) inhibitor that leads to the acetylation of H3 and H4 histones and tubulin.24 In vitro and in vivo studies showed that belinostat was highly active in ovarian cancer cell lines, including multidrug-resistant cell lines, and showed synergy with platinums and taxanes.24 The mechanism of synergy with carboplatin is believed to result from HDAC inhibition causing relaxation of DNA, which increases the ability of platinum salts to access the DNA.24,25Tubulin is a common target of both belinostat and taxanes; hence, belinostat may work by enhancing tublin’s activity.24

At the 2008 ASCO annual meeting, Finkler presented data from a phase II study of carboplatin, paclitaxel, and belino-

stat in a cohort of 31 women with relapsed ovarian cancer who had received ≤3 prior regimens.26 The ORR was 31%, and 1 patient had complete remission. When results were stratified by platinum-resistant status, women considered platinum-resistant following primary therapy had a PR rate of 17%. Toxicity was generally mild; the most commonly seen adverse effects were nausea (80%), vomiting (63%), fatigue (63%), and diarrhea (27%). The only grade 3-4 toxicity reported in >10% of participants was fatigue (17%). The GOG 126-T study is currently evaluating belinostat in combination with carboplatin in platinum-resistant patients.27

Epothilones

The epothilones are secondary metabolite macrolides produced by the myxobacterium Sorangium cellulosum.28 Their mechanism of action is similar to taxanes, in so far as they inhibit disassembly of microtubules and induce apoptosis.29 Two agents in this class are under evaluation in gynecologic cancers: patupilone (epothilone B, EPO906) and ixabepilone (Ixempra).

Patupilone

This macrocytic polyketide was originally found to demonstrate potent cytotoxic activity in mammalian cells.30,31 It differs from paclitaxel in two significant ways: it is not formulated with polyoxyethylated castor oil (Cremophor), so infusion or hypersensitivity reactions may be less likely; and it is not a substrate for P-glycoprotein nor do tubulin mutations affect its efficacy (both are associated with resistance to paclitaxel).29,31

The results of an open-label phase I trial in women with advanced ovarian, fallopian tube, or primary peritoneal cancers was published in 2009.32 Patupilone was administered every 3 weeks, and response was measured by radiologic and CA-125 criteria. The ORR in this previously treated population was 20%, with median time to progression (TTP) of 16 months. Doses up to 11.0 mg/m2 were considered tolerable, with the most common toxicities being diarrhea, peripheral neuropathy, and fatigue. Grade 3 diarrhea was the most common serious adverse event reported. A phase III trial of patupilone versus pegylated liposomal doxorubicin (Doxil) in patients with recurrent ovarian cancer has completed accrual and results are eagerly anticipated.33

Ixabepilone

This drug is also a semisynthetic analog of epothilone B. In preclinical studies, it was shown to be active in paclitaxel-sensitive and paclitaxel-resistant cell lines. It appears unaffected by known mechanisms of drug resistance, such as P-glycoprotein overexpression. This may be related to the drug’s ability to suppress the dynamic instability of alpha/beta III microtubules in cells with increased levels of beta III tubulin.34 Ixabepilone is currently approved for use in metastatic breast cancer as a single agent following treatment with anthracyclines, taxanes, and capecitabine; and in combination with capecitabine after anthracyclines and taxanes.35

Trials have been conducted with ixabepilone in ovarian and endometrial cancers. In GOG 126-M, 49 women with platinum-resistant ovarian cancer were given ixabepilone on a 3-week on, 1-week off schedule.36 The ORR was 14%, with a median TTP of 4.4 months. The incidence of grade 3-4 neutropenia was 20%, 3% of patients experienced grade 3 neuropathy.

In a phase II study, 50 women with locally advanced, recurrent, or metastatic endometrial cancer (94% had previously received paclitaxel) were given ixabepilone every 3 weeks.37 The ORR was 12%, with 6 of the 50 women responding. At 6 months, 20% were progression free. The primary toxicity was myelosuppression, with a 52% incidence of severe neutropenia. Only 10% of patients reported grade 3-4 neuropathy. The findings have led to registration of a phase III randomized trial of ixabepilone versus either paclitaxel or doxorubicin as a second-line agent in women with recurrent, advanced, or metastatic endometrial cancer.38

Conclusion

The ongoing clinical research in gynecologic malignancies could result in a better biologic understanding of these complex disease processes, allowing us to move beyond the current standard of care, which relies heavily on cytotoxic therapy. The findings from the bevacizumab studies in gynecologic cancers have confirmed that angiogenic targeting is an important and clinically relevant strategy. The current phase III studies should define the role for this agent in standard management. In the future, it is likely that the incorporation of biologics into a chemotherapy backbone will be of interest to clinical researchers; however, the role of cytotoxic therapy in the management of most of these solid tumors will remain important in the short term.

Scanning the clinical-trial landscape, it is gratifying to see how much work is going into finding effective treatments for ovarian cancer, the most fatal of all gynecologic neoplasms. Still, we must not forget that women are also living with and dying from cancers of the cervix and uterus, and, to a lesser extent, the vulva and vagina. We strongly urge the scientific community to broaden efforts into conducting much-needed trials in these other tumor types, to improve the standard of care for all gynecologic cancers.