SAN ANTONIO (GenomeWeb) – Although a study of Novartis' buparlisib found that patients with PIK3CA mutations were 50 percent less likely to experience disease progression, three patients receiving the PI3K inhibitor in a Phase III clinical trial attempted suicide, according to research presented at the San Antonio Breast Cancer Symposium.
Although the study suggests there might be a subpopulation of patients that might benefit from a more targeted PI3K inhibitor, the toxicity profile of buparlisib seems to have dampened leading oncologists' willingness to prescribe that particular therapy if it came to market.
"Buparlisib is probably not the best compound to be used in this particular setting," Angelo Di Leo from the Ospedale Misericordia e Dolce in Italy said at the meeting yesterday while reporting results from the BELLE-3 study. "Despite the improvement in progression-free survival, the higher rate of toxicity reported in this trial for patients receiving buparlisib," such as increases in liver enzymes and greater psychiatric symptoms, anxiety, and depression, "may represent a clinically relevant challenge for the future development of this compound," he said.
Di Leo led the randomized Phase III study of 432 hormone receptor-positive, HER2-negative advanced breast cancer patients who had received treatment with an aromatase inhibitor and progressed after treatment with endocrine therapy and mTOR inhibitor everolimus. They were randomized to receive buparlisib plus AstraZeneca's fulvestrant or just fulvestrant.
Median progression-free survival was 3.9 months in the buparlisib arm and 1.8 months in the chemo-only arm. But when researchers considered PIK3CA mutation status, as determined by PCR testing of primary tumor tissue, patients with mutations experienced median progression-free survival of 4.7 months with buparlisib compared to 1.4 months on just fulvestrant, but there wasn't a different in progression-free survival between the two arms in wild-type patients.
They also tested 349 patients' circulating tumor DNA at study entry using a blood test based on BEAMing (beads, emulsion, amplification, and magnetics) technology and found that 147 patients with PIK3CA mutations had 4.2 months median progression-free survival on the buparlisib arm compared to 1.6 months on just fulvestrant. In wild-type patients, median progression-free survival was 3.9 months on buparlisib versus 2.7 months on the comparator arm.
Ruth O'Regan from the University of Wisconsin and a researcher on BELLE-3 noted that one-third of patients in the trial were considered PIK3CA mutation positive by tumor or ctDNA testing. "In the tumors that had wild-type PIK3CA, if you looked at the primary tumor, there was no difference [between] the arms," O'Regan said. "But if you looked at the wild-type tumors based on ctDNA analysis, there still appears to be benefit over buparlisib over the control group."
PIK3CA status was assessed in primary tumor tissue and matched ctDNA samples in 257 patients, and 44 patients had discordant results. Di Leo hypothesized that 7 percent of patients with detected mutations in their tumor but wild-type ctDNA may be due to the fact that most patients had pretreatment with everolimus, which could have targeted the pathway and cleared the PIK3CA mutated ctDNA.
Regardless of benefit seen in the PIK3CA mutation-positive patients, the pan-PI3K inhibitor caused more psychiatric issues, such as depression and anxiety, and three cases of suicidal attempts were reported in the buparlisib arm. More patients in the buparlisib arm experienced dose interruptions and reductions, and decided to stop treatment due to toxicities.
Carlos Arteaga from the Vanderbilt-Ingram Cancer Center said if buparlisib became available he probably wouldn't prescribe it for his patients. "But I should not go further and say that PI3K is not a good target," he said. Although a pan-PI3K inhibitor, like buparlisib, may cause too many toxicities, a drug with a more specific target, such as a PI3K-alpha specific agent, may be worth exploring. "I think we desperately need a PI3K-mutant specific drug," he said.
Novartis is studying alpelisib plus fulvestrant in the SOLAR-1 study in an effort to explore whether p110 alpha-selective PI3K inhibition is efficacious in breast cancer patients without these toxicities. Researchers will investigate whether this regimen improves progression-free survival compared to just fulvestrant in patients with PIK3CA mutant status. They will also assess progression-free survival in ctDNA, which is an important exploration, O'Regan said, since it is more difficult to obtain tumor biopsies in metastatic disease.
"There is still room for improvement," Di Leo said, noting that the results can change based on the subgroup evaluated. "We shouldn't focus on the average result, but we should focus on the subgroups where we see some signals of a specific interaction between the treatments and the biomarkers or the clinical characteristics."