In an editorial accompanying publication of the AURELIA study results, Joyce F. Liu, MD, MPH, of Dana-Farber Cancer Institute, and Stephen A. Cannistra, MD, of Beth Israel Deaconess Medical Center, Boston, considered the implications of the benefits observed in the trial and limitations in interpretation of the findings.¹ These authors voiced support for consideration of the bevacizumab/chemotherapy regimen in selected patients with platinum-resistant disease.

Drs. Liu and Cannistra pointed out that AURELIA is the fourth randomized phase III study of bevacizumab (Avastin) in ovarian cancer; the GOG 218 and ICON7 trials assessed bevacizumab/chemotherapy followed by bevacizumab maintenance in newly diagnosed ovarian cancer, and the OCEANS trial evaluated bevacizumab plus carboplatin/gemcitabine followed by bevacizumab maintenance in platinum-sensitive recurrent disease.

Consistent Improvement in Progression-Free Survival

All four studies have shown a significant improvement in progression-free survival with the addition of bevacizumab, with the magnitude of improvement being remarkably consistent, given the different settings in which the combination has been evaluated. The OCEANS trial also showed a significant improvement in objective response rate. However, none of the studies has shown an overall survival benefit. AURELIA raised no new safety issues for bevacizumab/chemotherapy combinations.

The commentators noted that AURELIA is the first of the bevacizumab combination studies to show an improvement in abdominal/gastrointestinal symptoms and other patient-reported outcomes. The GOG 218 and ICON7 studies incorporating bevacizumab into first-line treatment and maintenance showed either no improvement (GOG 218) or a slight worsening (ICON 7) of patient-reported outcomes, and patient-reported outcomes were not examined in OCEANS.

As stated by Drs. Liu and Cannistra, “In the setting of platinum-resistant ovarian cancer, in which therapy is palliative, improvement in [patient-reported outcomes] is of potential importance even in the absence of an [overall survival] benefit.”

Interpreting Findings

Potential problems in interpreting the findings of AURELIA, according to the commentators, include the possibility of bias in the progression-free survival endpoint due to the open-label design of the trial and the absence of a placebo control, particularly since the study allowed for crossover to bevacizumab in progressing chemotherapy patients.

Crossover also likely affected the ability to detect any overall survival difference, with 40% of the chemotherapy-alone group receiving subsequent bevacizumab at the time of overall survival analysis. The commentators noted that the overall survival findings in the trial raise the possibility that using bevacizumab sequentially as monotherapy after progression on chemotherapy alone might yield equivalent overall survival compared with concomitant combined therapy in this disease setting.

The open-label design may have also influenced the patient-reported outcome endpoints, since patients were aware of whether or not they were receiving bevacizumab and may have unintentionally reported better patient-reported outcomes consistent with an expectation that the bevacizumab/chemotherapy regimen was a more effective treatment.

Progression and Patient-Reported Outcomes

Finally, an important difficulty in interpreting patient-reported outcomes derives from the imbalance in the proportions of patients in the two groups who completed assessments at week 8/9. A greater proportion of patients in the bevacizumab group completed assessments, reflecting the lower rate of progression in this group.

In the trial, patients with missing patient-reported outcomes data were scored as patient-reported outcome nonresponders, on the assumption that patients with disease progression did not experience improvements in patient-reported outcomes. As noted by the commentators, this assumption may not have been valid for some patients, who might have experienced improvement in patient-reported outcomes related to such factors as corticosteroid antiemetic treatment during chemotherapy or more effective pain management.

The editorialists stated:

Taking this into account, it is possible that the greater percentage of patients with missing data in the control arm might bias the [patient-reported outcome] results in favor of the bevacizumab group, essentially rendering [patient-reported outcome] assessment a surrogate of disease progression as opposed to a pure metric of symptom improvement. In this regard, it is interesting that by excluding missing data from the [patient-reported outcome] analysis, the effect size in [patient-reported outcome] benefit between the bevacizumab-containing arm and the chemotherapy-alone arm decreased and became nonstatistically significant.

They also noted, however, that differences in patient-reported outcomes still favored bevacizumab/chemotherapy, albeit nonsignificantly, when patients with missing patient-reported outcomes data were excluded from the analysis. This finding suggests that improvement in patient-reported outcomes can be expected with the addition of bevacizumab to chemotherapy in the platinum-resistant setting.

Option in Selected Patients

As to whether the findings in AURELIA support use of bevacizumab/chemotherapy as a new standard of care in the platinum-resistant setting, the commentators observed that the combined improvements in patient-reported outcome, progression-free survival, and objective response rate make it hard to ignore the possibility that such treatment could benefit appropriately selected patients. They noted, however, that patient selection would have to reflect the eligibility criteria of the trial, which excluded patients with more than two prior lines of chemotherapy and those with platinum-refractory disease. Given the risk of bowel perforation with bevacizumab observed in other trials, it also excluded patients with history of bowel obstruction, clinical signs of bowel obstruction, or evidence of bowel involvement on computed tomography.

With regard to this latter consideration, they stated, “Thus, those patients with the greatest need for symptom improvement and response, and who therefore might derive the greatest benefit from a bevacizumab-containing combination, might also be the ones who are at greatest risk for serious toxicity.”

Treatment options for patients with platinum-resistant disease are suboptimal. In acknowledgment of this situation—and notwithstanding the limitations in use and absence of evidence of overall survival improvement with the regimen—the commentators concluded that there is potentially a role for bevacizumab/chemotherapy in this setting. They stated:

…AURELIA raises the possibility that an improvement in [patient-reported outcomes], taken together with an improvement in both [progression-free survival] and [objective response rate], might be sufficient to justify the acceptance of a new regimen in a disease setting in which palliation is an important goal. In our view, it is not unreasonable to consider such evidence supportive of a role for the chemotherapy plus bevacizumab combination in carefully selected patients with symptomatic, platinum-resistant disease, at the same time recognizing that the data to support a [patient-reported outcomes] benefit are suggestive rather than definitive.” ■

Disclosure: Dr. Liu has received research funding from Genentech. Dr. Cannistra reported no potential conflicts of interest.

Reference

1. Liu JF, Cannistra SA: Emerging role for bevacizumab in combination with chemotherapy for patients with platinum-resistant ovarian cancer [editorial]. J Clin Oncol. March 24, 2014 (early release online).