Adding a targeted agent to chemotherapy failed to improve survival as initial therapy for metastatic urothelial carcinoma, a randomized trial showed.
Median overall survival (OS) was about 14 months with cisplatin-containing chemotherapy when paired with either placebo or the angiogenesis inhibitor bevacizumab (Avastin). Progression-free survival (PFS), a secondary endpoint, favored the bevacizumab arm (8.0 vs 6.7 months).
Grade ≥3 adverse events (AEs) occurred in a similar proportion of patients in each arm, although bevacizumab-related toxicities occurred more often with the addition of the angiogenesis inhibitor, reported Jonathan E. Rosenberg, MD, of Memorial Sloan Kettering Cancer Center in New York City, and colleagues, in the Journal of Clinical Oncology.
"Despite preclinical and clinical data suggesting a benefit for targeting angiogenesis in bladder cancer, bevacizumab did not improve survival in this trial," the authors said. "The results in this study with a PFS but not OS advantage appear similar to those using bevacizumab in combination with chemotherapy in many other solid tumors."
"Although there was a modest improvement in PFS, these results do not change the standard of care, which remains initiation of first-line therapy with cisplatin-based combination chemotherapy for medically eligible patients," they added.
The phase III trial did not confirm promising data from two earlier phase II trials, reinforcing the primacy of data from a randomized trial, said Arjun Balar, MD, of NYU Langone Health in New York City.
"Both of those [earlier] studies were single-arm trials, which means the results were compared against historical control groups," Balar, who was principal investigator in one of the phase II studies, told MedPage Today. "The ultimate test is randomization, and nothing is going to supplant the results from that."
The phase II trials were conducted about 10 years ago. In the interim, immune checkpoint inhibitors and other therapies emerged, and patients whose disease progressed on chemotherapy might have been offered some of the newer options, which could have influenced results in both arms of the phase III trial, he added.
"Despite all of that, if you look at the survival curves, there is a suggestion of a benefit [with bevacizumab], but it's modest," said Balar. "The curves separate and they maintain that separation. That tells us that there's probably an incremental benefit for the addition of bevacizumab to platinum-based chemotherapy, but the benefit probably is not sufficient to say that bevacizumab added to platinum-based chemotherapy is more effective."
Several lines of evidence supported adding bevacizumab to platinum-containing chemotherapy. Vascular endothelial growth factor (VEGF)-A is the primary mediator of cancer growth and progression in urothelial carcinoma. Higher levels of VEGF are associated with increased risk of progression and poor prognosis, and targeting angiogenesis in advanced urothelial cancer showed promise in early-stage clinical trials, Rosenberg and colleagues noted.
In general, single-agent anti-VEGF therapy produced few objective responses in chemorefractory diseases. The VEGF receptor 2 antibody ramucirumab (Cyramza) improved PFS in phase II-III trials of platinum-refractory urothelial cancer but not OS, the authors continued. Neither of the phase II trials of bevacizumab plus chemotherapy for untreated advanced/metastatic urothelial carcinoma met prespecified criteria for PFS, but both demonstrated higher than expected OS.
Rosenberg and colleagues reported primary results from CALGB 90601, a randomized comparison of gemcitabine-cisplatin chemotherapy plus bevacizumab or placebo for locally advanced or metastatic urothelial carcinoma. Eligible patients had received no prior therapy for metastatic disease and no neoadjuvant or adjuvant chemotherapy within the 12 months before enrollment. The primary endpoint was OS, and key secondary endpoints included objective response and toxicity, as well as PFS.
The primary analysis included 506 randomized patients. After a median follow-up of 76.3 months among surviving patients, investigators found no difference in OS (14.5 months with bevacizumab vs 14.3 months with placebo, P=0.14). The small difference in PFS (median follow-up of 46 months) translated into a hazard ratio of 0.77 in favor of bevacizumab, which proved to be statistically significant (95% CI 0.63-0.96, P=0.016). The objective response rate was 40.4% with bevacizumab and 36.4% with placebo, a nonsignificant difference.
Whether a subgroup of patients might benefit from the addition of bevacizumab remains an open question.
"The question is basically asking whether there are biomarkers we could test for a priori," said Balar. "The short answer, right now, is that we don't have any validated biomarkers to predict who is going to benefit from VEGF-targeted therapy. We don't have any pretreatment selection criteria that can tell us that."
The subgroup analysis hinted at a possible bevacizumab benefit for patients with visceral metastases, but the trial design did not address that issue, he added.