This article is more than 5 years old. Information may no longer be current.
Maintenance pazopanib extended PFS, not OS in ovarian cancer
Click Here to Manage Email Alerts
Maintenance therapy with pazopanib extended PFS by about 45% among women with ovarian cancer whose disease did not progress during first-line chemotherapy, according to results of a phase 3 study.
However, the maintenance regimen did not appear to confer an OS benefit.
Pazopanib (Votrient, GlaxoSmithKline) is an orally administered multi-targeted tyrosine kinase inhibitor of the vascular endothelial growth factor receptor and platelet-derived growth factor receptor. It is approved to treat renal cancer and soft tissue sarcoma, and a previous phase 2 study showed pazopanib has activity in ovarian cancer.
The current study included 940 patients with ovarian cancer whose disease did not progress during first-line chemotherapy.
Andreas du Bois, MD, professor of gynecologic oncology at Kliniken Essen Mitte in Essen, Germany, and colleagues randomly assigned 472 patients to a maintenance regimen of 800 mg pazopanib once daily for up to 2 years. The other 468 patients received placebo.
Median time to diagnosis and median interval from the last cycle of chemotherapy to study entry were comparable between study arms.
PFS served as the primary outcome measure. Median follow-up was 24.3 months.
Patients assigned pazopanib demonstrated significantly longer median PFS (17.9 months vs. 12.3 months; HR=0.77; 95% CI, 0.64-0.91). However, results of two planned interim analyses showed no OS benefit (HR=1.08; 95% CI, 0.87-1.33).
Results of exploratory post-hoc analyses of protocol-prespecified subgroups showed pazopanib’s benefit appeared to be derived primarily by non-East Asian patients, who comprised 78% of the cohort. In those analyses, the HR for PFS in this subgroup was 0.69 (95% CI, 0.57-0.84), compared with an HR of 1.16 (95% CI, 0.78-1.73) for patients recruited in East Asia.
However, results of a second interim survival analysis showed a nonsignficant difference in benefit in non-East Asian patients (HR=0.98; 95% CI, 0.77-1.24), as well as a significant harmful effect in the East Asian population (HR=1.71; 95% CI, 1.01-2.89).
Patients assigned pazopanib were considerably more likely to discontinue treatment due to grade 3 or grade 4 adverse events (33.3% vs. 5.6%). Grade 3 or grade 4 adverse events reported in the pazopanib arm included hypertension (30.8%), neutropenia (9.9%), liver-related toxicity (9.4%), diarrhea (8.2%), fatigue (2.7%), thrombocytopenia (2.5%) and palmar-plantar erythrodysesthesia (1.9%).
More patients assigned placebo received post-treatment anticancer therapy (61% vs. 50%), but time to second-line therapy was significantly longer in the pazopanib arm.
“The observed prolongation of PFS is worthwhile and resulted in a significant delay of the time to second-line cytotoxic chemotherapy,” du Bois and colleagues wrote. “On the other hand, we could not demonstrate any survival benefit, and toxicity led to a significant proportion of patients not tolerating the planned treatment schedule … Pazopanib cannot be recommended for broad clinical use in ovarian cancer. Further analysis may identify another clinical setting or specific subgroups of patients who may derive a significant clinical benefit of this active antiagiogenesis drug.”
The increased toxicity observed with some antiangiogenic agents, as well as the potential that those agents could harm certain subsets of patients, highlight the risks associated with “indiscriminate use of surrogate endpoints such as PFS for drug approval,” Kate E. Oliver, MD, of Walter Reed National Military Medical Center in Bethesda, Md, and William P. McGuire, MD, of Inova Fairfax Hospital in Annandale, Va., wrote in an accompanying editorial.
“Perhaps the putative gold standard of OS, especially when combined with quality-of-life metrics, should be embraced once again,” Oliver and McGuire wrote. “Alternatively, molecular separation of tumors may allow for identification of patients who are most likely to benefit from these targeted therapies, or the corollary, for selective withholding of targeted therapies from those who are most likely to be harmed. Until we better understand and define the important aspects of tumor biology that may govern effects of targeted therapies, it is premature to claim victory for the addition of these new agents to standard therapy for advanced ovary cancer solely on the basis of significant improvement in PFS.”
Disclosure: The researchers report employment/leadership positions with, consultant/advisory roles with and stock ownership in Amgen, AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline, GlaxoSmithKline Japan, Merck, Morphotek, MSD, Roche and Zeria Pharmaceuticals.
Perspective
Back to Top
James Tate Thigpen, MD
Treatment for ovarian carcinoma has dramatically improved over the past 30 years with the addition of platinum compounds and taxanes. Overall response rates in patients with bulky advanced disease have risen from 25% to 30% with single-agent melphalan in the 1970s to 65% to 75% with paclitaxel/carboplatin today, and clinical complete response rates in the same group of patients with bulky disease have been reported to be as high as 50% (Omura G. Cancer. 1986;57:1725-1730). In the same period, median PFS and OS have improved by a similar magnitude.With the combination of aggressive surgical bulk reduction followed by paclitaxel/carboplatin, approximately 75% of patients with stage III or stage IV ovarian carcinoma will end initial therapy in a clinical complete response. This makes ovarian carcinoma one of the most responsive solid tumors to systemic therapy.
The problem is that, essentially, 75% of those with a clinical complete response eventually relapse and ultimately die of their disease. A number of clinical trials have sought to identify effective methods to prevent or delay relapse. Three agents to date have demonstrated in phase 3 trials the ability to improve outcomes when employed as maintenance therapy: paclitaxel, demonstrated by a Gynecologic Oncology Group/Southwest Oncology Group collaborative study in patients in clinical complete response after initial treatment (Markman M. J Clin Oncol. 2003;21:2460-2465); bevacizumab (Avastin, Genentech), demonstrated in four trials that included more than 4,000 patients with at least stable disease after initial therapy (Aghajanian C. J Clin Oncol. 2012;30:2039-2045; Burger RA. N Engl J Med. 2011;365:2473-2483; Perren TJ. N Engl J Med. 2011;365:2484-2496; Pujade-Lauraine E. J Clin Oncol. 2014;32:1302-1308); and now pazopanib, as demonstrated by du Bois and colleagues in their study of 940 patients with at least stable disease after initial treatment (du Bois A. J Clin Oncol. 2014;32:3374-3382).
The major criticism of these observations is that all six studies show improved PFS without concomitant improvement in OS. This fails to take into account the evolution of treatment for ovarian carcinoma during the past 2 decades. Prior to 1989, there was little effective therapy for those whose disease relapsed. Observations in 1989 (Blackledge G. Br J Cancer. 1989;59:650-653) and 1991 (Markman M. J Clin Oncol. 1991;9:389-393) established that retreatment of recurrent disease with platinum-based therapy if the treatment-free interval exceeded 6 months improved response, PFS and OS. Subsequently, at least 20 additional agents have been shown to produce responses in recurrent disease. Most patients receive multiple additional lines of therapy after the initial recurrence. All of this post-progression therapy renders the OS endpoint impossible to interpret and makes PFS the optimal endpoint for trials in ovarian carcinoma. The demonstration in the trial by du Bois and colleagues of a clear improvement in PFS should be sufficient to make pazopanib one legitimate option for maintenance therapy in those patients with at least stable disease. The physician and the patient then will have to weigh toxicity vs. potential benefit.
Click Here to Manage Email Alerts