Krishnansu S. Tewari, MD, discussed the results from the phase III clinical trial investigating the addition of bevacizumab to standard chemotherapy treatment following surgery in patients with advanced ovarian cancer, and the next steps for research regarding the role of bevacizumab in ovarian cancer treatment.
Krishnansu S. Tewari, MD
Bevacizumab (Avastin) has demonstrated benefit in the treatment of patients with ovarian cancer in the past as vascular epithelial growth factor (VEGF) is an important promoter of ovarian cancer progression.
GOG-0218, a randomized phase III trial, investigated the use of bevacizumab in combination with chemotherapy in the treatment of patients with ovarian cancer following surgery.
Over 1800 women with stage III to IV newly diagnosed ovarian, fallopian tube, and primary peritoneal carcinoma who had undergone an incomplete resection were enrolled in the trial. Following primary debulking surgery, 3 arms were randomized to chemotherapy treatment with maintenance.
The control arm received chemotherapy plus placebo, followed by placebo maintenance. Arm 2 received chemotherapy plus bevacizumab, followed by placebo maintenance. Arm 3 received chemotherapy in combination with bevacizumab, followed by bevacizumab maintenance. The protocol-specified primary endpoint was overall survival (OS) but was changed during the course of the study to progression-free survival (PFS).
According to a previous report of the trial findings, the median PFS in the arm receiving bevacizumab plus chemotherapy followed by maintenance bevacizumab was 18.2 months versus 12.8 months in arm 2 and 12.0 months in the chemotherapy arm.1,2These data led toan FDA approval in June 2018 for the use of bevacizumab in combination with carboplatin and paclitaxel, followed by bevacizumab maintenancein patients with advanced ovarian cancer following resection.
The final OS data were recently published in theJournal of Clinical Oncologyin June 2019.3At a median follow-up of 102.9 months, neither investigational arm demonstrated a survival advantage compared to the control arm. However, patients with stage IV disease had a median OS of 42.8 months versus 32.6 months in the control arm and 34.5 months in arm 2. This OS rate observed in patients with stage IV disease is comparable to what’s been observed in patients with more favorable stage III disease in other studies.
Additionally,BRCA1/2mutations were not predictive of bevacizumab activity in these patients, but an analysis of blood specimens demonstrated that theBRCAmutations or homologous recombination deficiency (HRD) had prognostic value in patients with advanced disease. The hazard ratio for death inBRCA-mutated cancers was 0.62 versus 0.65 in nonBRCA-mutated or HRD cancers. The investigators noted that testing for bothBRCA1/2mutations and HRD is essential in ovarian cancer management.
In an interview withTargeted Oncology, Krishnansu S. Tewari, MD, professor and director of the Division of Gynecologic Oncology at the University of California, Irvine, discussed the results from the phase III clinical trial investigating the addition of bevacizumab to standard chemotherapy treatment following surgery in patients with advanced ovarian cancer, and the next steps for research regarding the role of bevacizumab in ovarian cancer treatment.
TARGETED ONCOLOGY: What was the rationale for investigating bevacizumab in this patient population?
Tewari:In the first decade of the twenty-first century, a VEGF emerged as an important therapeutic target in a number of solid tumors, including colorectal cancer and nonsquamous nonsmall cell lung cancer, where bevacizumab, which is a VEGF monoclonal antibody, gained a label. We felt that patients with ovarian cancer with bulky, residual disease after primary surgery would [benefit as well] as the tumor would be secreting relatively high levels of VEGF; we thought we would take advantage of that and target these cancers by integrating bevacizumab with chemotherapy as part of primary treatment after surgery, then continue [bevacizumab] after chemotherapy was completed as a maintenance therapy.
TARGETED ONCOLOGY: How was this trial designed?
Tewari:The trial was designed to study women with advanced ovarian cancer who had undergone primary surgery. They were expected to have residual disease, and the trial was a 3-arm placebo-controlled randomized clinical trial. It was a multicenter, multinational study, and there were 3 arms. One arm was to have just standard chemotherapy plus placebo, followed by maintenance placebo. The second arm was chemotherapy plus bevacizumab 15 mg/kg on a 21-day schedule, followed by maintenance placebo. The third arm was chemotherapy plus bevacizumab, followed by maintenance bevacizumab. There were 3 arms to the study with bevacizumab appearing in 2 of the investigational arms.
TARGETED ONCOLOGY: What were the results from this trial?
Tewari:The primary endpoint was originally OS. That was changed to PFS during the conduct of the study, mainly to assist with patients that progressed.
The initial results demonstrated that the incorporation of bevacizumab, specifically in the third arm of the study where it was given as treatment with chemotherapy and as maintenance, significantly improved PFS by 4 to 6 months, depending on whether blinded independent radiology review was being performed or not. The bottom-line is that it met its primary endpoint, and it led to a label for bevacizumab in the European Union back in 2011.
TARGETED ONCOLOGY: What did this trial demonstrate in terms of the importance ofBRCAand HRD testing?
Tewari:During the study of the protocol-specified final OS analysis, which was just published in theJournal of Clinical Oncologyin July, we also wanted to see if patients with germline or somaticBRCAmutations and/or HRD tracked with survival. Although theBRCAmutations or HRD did not predict bevacizumab activity, what was important was that in the study, we had over 1100 specimens that were obtained prospectively. This demonstrated that not only was BRCA1andBRCA2prognostic in the advanced ovarian population under study, but also HRD mutations in homologous recombination repair (HRR) genes were also significant for prognosis.
The reason this is important is, based on earlier work, we always suspected women harboringBRCAmutations had an improved prognosis. They tended to respond to platinum-based chemotherapy and tend to be platinum-sensitive at recurrence, but we never showed that. Now for the first time in a large trial with over 1100 specimens, we were able to demonstrate the prognostic value of bothBRCAand HRD. I think it’s important because in the final analysis, although neitherBRCAor HRD were predictive of bevacizumab activity, they are prognostic, and these data would support doing HRD testing at diagnosis in order to inform patients on prognosis and counsel them.
TARGETED ONCOLOGY: What are the key points to take away from this final OS analysis?
Tewari:In the final analysis in the intent-to-treat population, bevacizumab did not improve OS compared to the control. However, it should be pointed out, though, that post-progression survival in this disease is relatively long and because investigators were not able to control post-progression therapy, the OS endpoint is often times diluted and essentially meaningless. We feel that PFS is a validated clinical primary endpoint in advanced ovarian cancer, but the key takeaway is bevacizumab did not improve survival.
A second important takeaway is that in the stage IV patients, patients treated on the third arm with bevacizumab with chemotherapy followed by bevacizumab maintenance, it was an exploratory analysis and wasn’t prespecified, but it did appear to have a survival advantage compared to the other 2 arms. In fact, the median survival of 42 months approximately mirrored what we are seeing in the stage III patients. It’s important, even though it was exploratory and not prespecified in the protocol; it does align itself with the ICON7 sub-analysis involving stage IV patients, so there are 2 studies both demonstrating similar findings for stage IV patients. I think this would argue, for patients with stage IV disease, to consider receiving bevacizumab with chemotherapy and as maintenance therapy if they turn out not to have aBRCAmutation.
The third take away is the [importance of]BRCAand HRD analysis.
TARGETED ONCOLOGY: What are the next steps for bevacizumab in gynecologic cancers?
Tewari:As I mentioned, based on the PFS endpoint, bevacizumab was given a label in the European Union and in other non-United States countries in 2011. Last June, the FDA approved bevacizumab in the United States for this specification based on this trial.
The next steps are that we need to learn more about HRD testing; in the PRIMA study, which is a PARP inhibitor study looking at niraparib [Zejula] as frontline maintenance therapy, the principal population being studied is the HRD-positive population. That study met its primary endpoint based on a press release that was issued last month. We are anticipating seeing the results in detail at the upcoming ESMO meeting in September this year. We will have to see what this demonstrates.
Other next steps are that we still need to determine and clarify a role for antiangiogenesis therapy in this disease. For patients that don’t have a mutation, they should be counseled about receiving bevacizumab as a maintenance therapy. Also, combinations are being studied. The PAOLA-1 study is also being presented at ESMO, we believe, and that’s combining bevacizumab with another PARP inhibitor, olaparib [Lynparza], which received a label for either germline- or somatic-mutated patients.
We don’t have a validated predictive biomarker for bevacizumab in this disease, so we still need to continue investigations for that and keep moving the field forward. There are studies combining bevacizumab with checkpoint immunotherapy in this disease. The NOVA study being done in the platinum-sensitive recurrent population is an important study. Whether we can replicate those results in the frontline setting, it still has to be seen.
I think the next steps are about evaluating combinations, further clarifying HRD in this disease, and still searching for a biomarker. Even though this study did not show a significant improvement in survival, the OS endpoint in this space is often obscure due to post-progression time for people to remain alive and inability to access post-progression therapy. PFS is an important clinical endpoint, especially when it’s accompanied by other measures demonstrating benefit and [improvement in] quality of life.