乌鸦传媒

Another year brings another negative avelumab (Bavencio) trial in ovarian cancer, and more experts are left scratching their heads.

In , median progression-free survival (PFS) for patients who received avelumab with first-line platinum-based chemotherapy was no better than chemotherapy alone and was associated with possible harm, reported Jonathan Ledermann, MD, of University College London.

Median PFS in the phase III trial was 16.8 months for women that received six cycles of carboplatin-paclitaxel plus avelumab maintenance yet was not reached in a control group of chemotherapy followed by observation alone (HR 1.43, 95% CI 1.05-1.95), according to findings presented at the Society of Gynecologic Oncology (SGO) annual meeting webinar series.

In a third group where avelumab was combined with chemotherapy upfront and then continued as maintenance, median PFS reached 18.1 months (HR 1.14, 95% CI 0.83-1.57).

"Futility boundaries were crossed at the interim analysis," said Ledermann. "The trial was terminated early and ongoing treatment of patients was stopped."

No subgroups appeared to benefit from the addition of avelumab, and while overall survival data were immature, there were more deaths in both investigational arms at data cutoff (20 and 21) than in the observation arm (13).

"Other frontline trials with immune checkpoint inhibitors are ongoing and the results are awaited with interest," he added.

Leslie Randall, MD, of Virginia Commonwealth University in Richmond, said the results of JAVELIN Ovarian 100 had been highly anticipated.

"The rationale for this trial was that chemotherapy would indeed heat up this immunologically cold epithelial ovarian cancer tumor," she said. "Unfortunately, we did not see any difference in the experimental arms versus the control arm ... and possibly even somewhat of a decrement, though this trial was not statistically designed to determine that."

In the avelumab maintenance-only arm, patients with CD8-positive tumor infiltrating lymphocytes appeared to be "driving some of that decrement," she noted, a subgroup that researchers anticipated would have improved outcomes with immunotherapy.

At last year's SGO meeting, a phase III trial comparing pegylated liposomal doxorubicin to avelumab alone (or to both in combination), showed that the immunotherapy offered no PFS or overall survival (OS) benefit in unselected patients with platinum-resistant or refractory ovarian cancer, but the combination arm did suggest a possible OS benefit among the PD-L1-positive subgroup.

PD-L1 is an important biomarker of immunotherapy response in numerous tumor types, Randall said, and preliminary data in ovarian cancer led to the design of the current trial, where nearly 50% of patients were PD-L1 positive (≥1%). But in this subgroup, no beneficial effect was observed in either of the avelumab arms.

"Where do we go from here? I think that JAVELIN 100 was really the hope of the first immunotherapy trial in our frontline setting, but maybe we were just ahead of ourselves," said Randall. "Maybe we needed to think through our approach before rushing into this trial."

"Ovarian cancer is clearly different from the other tumor types," she added. "Do we use novel biomarkers, develop non-checkpoint approaches, or look at combinations?"

JAVELIN Ovarian 100 was an open-label, phase III trial that randomized 989 patients with untreated stage III-IV epithelial ovarian cancer. Patient characteristics were well balanced between the three arms, with roughly 40% having received neoadjuvant chemotherapy, and about a fourth of the cohort had residual disease after their primary surgery.

Most patients in each arm of the trial (~85%) were able to complete their planned chemotherapy, and only 7% of patients on avelumab discontinued for toxicity during the maintenance period. Avelumab safety was consistent with previous trials and did not significantly increase toxicities, Ledermann said. Deaths due to adverse events, however, were slightly higher with avelumab, with four and six deaths in the investigational arms compared to three in the control arm.

In the PD-L1-positive subgroup (n=487), median PFS was not reached in any of the arms. Compared to the control arm, the HRs for PFS were 1.23 with maintenance only (95% CI 0.79-1.92) and 0.98 in the upfront plus maintenance avelumab arm (95% CI 0.62-1.54).

For the CD8-positive subgroup (n=332), HRs for PFS were 1.64 in the maintenance-only arm (95% CI 0.95-2.85) and 1.25 in the avelumab-throughout arm (95% CI 0.71-2.22).

No benefit was seen among patients' BRCA1/2 mutations, which made up about 10% of the study population.

Comment