Highlights from

ASCO 2019

American Society of Clinical Oncology annual meeting

Chicago, USA 31 May - 4 June 2019

Olaratumab trial in soft tissue sarcoma fails to meet overall survival endpoint

Medical writer: Tim Donald, ELS

The ANNOUNCE trial, assessing the overall survival (OS) benefit of olaratumab in patients with advanced soft tissue sarcoma (STS), failed to meet its dual primary endpoint of OS in all STS histologies or in a subset of patients with leiomyosarcoma (LMS), according to a Late-Breaking Abstract (Abstract LBA3) presented at the June 2 Plenary Session. William D. Tap, MD, of Memorial Sloan Kettering Cancer Centre, presented the study data.

Dr. Tap explained that, after the data readout from this study, the trial sponsor and global regulatory agencies recommended that no new patients be started on olaratumab. Withdrawal of the drug from the market is in progress. Patients who are currently receiving treatment with olaratumab with apparent benefit may continue to be treated with the drug.

“ANNOUNCE was a well-controlled and conducted phase III trial, which failed to meet its overall survival primary endpoint in all STS histologies or the LMS population,” he said, adding that the trial did not confirm the benefit seen in an earlier trial.

In that previous phase Ib/II clinical trial in patients with advanced STS, a regimen of doxorubicin plus olaratumab improved progression-free survival (PFS) and OS in comparison with doxorubicin alone. After that trial, the U.S. Food and Drug Administration granted accelerated approval to olaratumab, conditional on the performance of a follow-up study; ANNOUNCE was that study.

Possible confounding factors in the phase Ib/II study, Dr. Tap said, included that the trial was unblinded; that there was a large discrepancy between results in OS and PFS, the study’s primary endpoint; and that the drug had an unknown mechanism of action in sarcoma subtypes.

ANNOUNCE enrolled adult patients with unresectable locally advanced or metastatic STS, with ECOG status 0 or 1, and with any number of previous treatments but no anthracycline. Patients were randomly assigned 1:1 to olaratumab 20 mg/kg for cycle 1, then olaratumab 15 mg/kg for cycles 2 through 8, or placebo, on days 1 to 8 of each 21-day cycle. Patients in both arms received doxorubicin 75 mg/m2 on day 1 for up to eight cycles. After eight cycles, patients with disease control continued on olaratumab or placebo as monotherapy until progression or toxicity. Treatment with dexrazoxane was allowed to mitigate doxorubicin-related cardiotoxicity.

Randomization was stratified by number of previous treatments, histology, and ECOG status. The primary endpoint was OS by intent-to-treat in the total STS and the LMS populations, and the study was designed to be positive if either population, or both, had a statistically significant improvement in OS. The secondary endpoints were PFS, safety, pharmacokinetics, objective response rate, and patient-reported outcomes.

Of 509 patients randomly assigned, 258 were in the investigational regimen and 251 in the control arm. Baseline characteristics were well balanced.

In the total STS population, median OS was 20.4 months in the investigational arm and 19.7 months in the control arm (HR 1.05, 95% CI [0.84, 1.30]; P = 0.6945). In the LMS population, median OS was 21.6 months in the investigational vs. 21.9 months in the control arm (HR 0.95, 95% CI [0.69, 1.31]; P = 0.7618).

Why was the study negative? Dr. Tap listed several possible reasons, including that olaratumab is not effective in combination with doxorubicin in STS, or that olaratumab has some activity in STS, but it was masked by heterogeneity of the study populations within and between studies or by differences in study designs.

Another consideration is that patients in the control arm in ANNOUNCE performed better than expected. “Median OS in the doxorubicin and placebo arm was 19.7 months,” Dr. Tap said. “This is the highest survival rate described to date in any phase III sarcoma study.”

In light of the large number of subtypes, Dr. Verweij questioned the predictive value of phase II randomized studies investigating the addition of drugs to doxorubicin for STS. Several randomized phase II studies in STS to date have overestimated PFS and OS benefits in comparison to subsequent phase III studies of the same drug combinations, he said.

When STS subtypes are lumped together in phase II studies, even when those studies are randomized, they can provide misleading results, Dr. Verweij said. Phase II studies should be considered a screening tool, and they always need confirmation in a phase III trial, even when activity data in phase II are compelling.

He suggested that phase II studies in specific STS subtypes could yield better predictions of outcomes for subsequent phase III studies. Global collaboration now allows the performance of rapidly accruing clinical studies in STS, and investigators should favour subtype-specific, biomarker-driven studies whenever possible.

The content and interpretation of these conference highlights are the views and comments of the speakers/authors.