Read the Full Video Transcript

Christopher Wallis: Hello, and thank you for joining us for this UroToday Journal Club. Today, we are discussing a recent publication in The Journal of Clinical Oncology, entitled, Randomized Phase 3 Trial of Gemcitabine and Cisplatin with Bevacizumab or Placebo in Patients with Advanced Urothelial Carcinoma. These are the results of the CALGB 90601 trial. I'm Chris Wallis a Fellow in Urological Oncology at Vanderbilt. With me today is Zach Klaassen, an Assistant Professor in the Division of Urology at the Medical College of Georgia. This is the citation for this recent publication in JCO.

If you look here, we see that in the last few years, we have had a dramatic proliferation of treatment options for patients with advanced bladder cancer, and in fact, since 2016, we've had seven new drug approvals in addition to new indications for existing drugs. However, if you look at the NCCN guidelines for patients who are cisplatin eligible, cisplatin with gemcitabine or as a part of the dose-dense MVAC regime remains the preferred treatment option, followed by avelumab maintenance therapy on the basis of the JAVELIN data.

However, we know that angiogenesis across all, or at least many, disease states remains fundamental for cancer growth and progression, and there is evidence that in urothelial cancer, VEGF-A activity is critical to cancer progression and growth. And so, we've seen correlational studies demonstrating that VEGF expression and HIF-1 alpha expression are associated with poor prognosis. As a result, the idea of targeting angiogenesis has been of interest in the field of carcinoma. So, there are three studies highlighted here, among many, looking at the use of various anti-angiogenic agents, tyrosine kinase inhibitors in patients with metastatic urothelial disease.

Bevacizumab, in particular, is a humanized monoclonal antibody, which targets VEGF-A, and this slide shows that demonstrated efficacy in combination with chemotherapy across a variety of disease sites, and so, I've highlighted here, data on cervical cancer, data for patients with advanced lung cancer, and ovarian cancer. And in early single-arm phase 2 trials, there was a suggestion of benefit for the use of bevacizumab in combination with chemotherapy compared to historical controls for patients with urothelial cancer.

This study enrolled patients who are 18 years or older with histologically or cytologically confirmed urothelial cancer, who had radiologically documented metastatic or unresectable locally advanced disease. They had to have a good performance status and adequate organ function to be included. There was a variety of exclusion criteria here, including prior chemo, although this was allowed if it was used in the adjuvant or neoadjuvant setting and at least 1 year had passed, prior use of angiogenic agents, CHF, bleeding or arterial embolic events, brain metastases, peripheral arterial disease, GI perforation, non-healing wounds or fractures, peripheral neuropathy, and carcinomatosis, and these were all considered exclusion criteria for this study.

Once included, this was a randomized, double-blind phase 3 trial with 1:1 randomization going to either gemcitabine-cisplatin and placebo or gemcitabine-cisplatin and bevacizumab, and randomization was stratified according to the presence of visceral disease and the prior use of chemotherapy. Up to six cycles of chemotherapy were allowed and bevacizumab was administered during chemotherapy, as well as on an ongoing maintenance basis until progression.

The primary endpoint was overall survival and secondary endpoints included progression-free survival, objective response rate, and adverse events. The assessment was performance radiologic imaging with cross-sectional studies at baseline in every three cycles, and this continued until disease progression, relapse, or the initiation of non-protocol-based therapy, and all imaging was assessed according to standard RECIST criteria.

In terms of statistical analyses, the authors sought to enroll 500 patients in order to have 454 death events. And on the basis of this sample size, they would have 87% power to detect for a hazard ratio of 0.74, with a two-sided alpha of 0.05. This is premised on the assumption of median overall survival of just under 14 months in the chemotherapy alone arm, and 18.6 months in those who received bevacizumab. There were a number of interim analyses, both for futility as well as superiority, but none of those reached stopping boundaries, and so this continued until this present final analysis.

Intention-to-treat analyses were undertaken for time-to-event outcomes, with safety and objective response rates assessed among patients who received any therapy. The proportional hazards models were used to calculate hazard ratios for time-to-event outcomes, and a number of subgroup analyses were performed according to the stratification factors we described before, as well as performance status, patient sex, and primary tumor site.

At this point in time, I am now going to hand it over to Zach to take us through the remainder of the results.

Zachary Klaassen: Thanks, Chris. This is the CONSORT diagram for this trial. There were 506 patients that were allocated and randomized to the two arms. As you can see on the left, this is the gemcitabine-cisplatin and bevacizumab arm with 252 patients, and there were 254 patients allocated to the gemcitabine-cisplatin and placebo arm. In the efficacy analysis, you can see there where there were 252 patients in the bevacizumab arm and 254 patients in the placebo arm with the same breakdown for PFS analysis. And for the objective response, there were 213 patients in the bevacizumab arm and 209 patients in the placebo arm.

Looking at this baseline characteristics table, we can see here that there was a fairly balanced breakdown between the two arms, with about three-quarters of the patients being less than 70 years of age, the majority of patients being white, at 91% and 93%, the majority of patients being male, at 79% and 78%, the majority of patients receiving no preoperative chemotherapy, at 87%, and about two thirds in each arm having visceral metastases at 69%. The baseline performance status was slightly more than 50% for ECOG 0, and the rest of the patients essentially being ECOG 1. About three-quarters in each arm were bladder cancer primary tumor site, with about one-quarter of these patients being upper tract. The majority of these patients did have excellent kidney function with a creatinine clearance greater than or equal to 60 in more than 75% of these patients, and the majority of patients did have measurable disease at baseline, at 89% and 92%.

This is the Kaplan-Meier curve for overall survival. You can see here that in the bevacizumab arm, the median overall survival was 14.5 months, very comparable to the placebo arm at 14.3 months, which is reflected here in the Kaplan-Meier curves, essentially almost overlapping. The hazard ratio, not statistically significantly favoring bevacizumab at a hazard ratio of 0.87 and a 95% confidence interval of 0.72 to 1.05.

This is the forest plot for the subgroup analyses of overall survival. You can see here that the majority of these hazard ratios and 95% confidence intervals do cross 1, however, several notes to make here. We can see that in the prior chemotherapy group, there was a hazard ratio of 0.55 and a 95% confidence interval of 0.33 to 0.92, and there was a hazard ratio in males of 0.82 and a 95% confidence interval of 0.66 to 1.02.

Looking at the Kaplan-Meier curve for progression-free survival, the median progression-free survival for the bevacizumab arm was 8.0 months and for the placebo arm was 6.7 months. This was statistically significant, favoring bevacizumab with a hazard ratio of 0.77 and a 95% confidence interval of 0.63 to 0.95.

Again, this is a subgroup analysis for progression-free survival. We can see here that several points of note in the visceral metastases patients, a hazard ratio, favoring the bevacizumab arm of 0.73 and a 95% confidence interval of 0.57 to 0.93.

In terms of objective responses, we can see that for both arms, there was an 8% complete response rate. The partial response was 33% for the bevacizumab arm and 28% for the placebo arm. Stable disease in both arms was 42%, for an overall confirmed response rate of 40%, in the bevacizumab arm and 36% in the placebo arm.

Looking at hematological adverse events, we'll focus on the grade 3, 4 adverse events. They are essentially the same for anemia, 18% in the bevacizumab arm and 19% in the placebo arm. More thrombocytopenia in the bevacizumab arm at 27% and 18% in the placebo arm. Looking at the non-hematological adverse events, again, focusing on the grade 3 to 5 adverse events, fatigue, 11% in the bevacizumab arm compared to 7% in the placebo arm, hemorrhage, 10% in the bevacizumab arm with 3% in the placebo arm, hypertension was quite common in the bevacizumab arm at 21% compared to only 5% in the placebo arm. And finally, infection, 16% in the bevacizumab arm and 15% in the placebo arm.

Several discussion points from the CALGB study. The addition of bevacizumab to standard first-line chemotherapy for locally advanced and metastatic urothelial carcinoma did not improve overall survival. importantly, the standard of care for first-line therapy remained unchanged during the five-year accrual period for this trial, and even today, cisplatin-based chemotherapy without the addition of biological agents remains the standard of care. When the CALGB 90601 study was designed, phase 2 data from two single-arm trials suggested the addition of bevacizumab to gemcitabine plus platinum chemotherapy yielded longer than expected median survival compared to historic controls. And this is important because it highlights the importance of phase 3 randomized trials testing the addition of novel agents, and not just relying on phase 2 single-arm trials. Finally, an improved understanding of the association of molecular biomarkers and response to chemotherapy may allow for optimal patient selection for treatment.

In conclusion, the addition of bevacizumab did not improve overall survival over gemcitabine-cisplatin chemotherapy alone. And although there was a modest improvement in progression-free survival, as demonstrated in these slides, these results do not change the standard of care, which remains as the initiation of first-line therapy with cisplatin-based combination chemotherapy for medically eligible patients.

Thank you very much and we hope you enjoyed this UroToday discussion on the CALGB study.