Roche’s pictilisib, a pan inhibitor of the PI3 kinase enzyme has failed to improve outcomes when combined with fulvestrant in patients with estrogen receptor positive (ER+) metastatic breast cancer in a mid-stage trial.

“The progression-free survival results for those patients with PIK3CA mutations were essentially identical to that in the wild-type population,” concluded the lead investigator of the FERGI study, Ian Krop of the Dana-Farber Cancer Institute, speaking at the recent San Antonio Breast Cancer conference.

The rationale for doing the study is this: 40-45% of ER+ breast cancer tumours harbour the PIK3CA mutation; an alteration in the PI3K pathway, via the signalling partner mTOR, is thought to be responsible for the emergence of resistance mechanisms in patients treated with anti-estrogen therapies.

The idea was to combine a PI3K inhibitor with hormonal therapy to overcome this resistance and improve outcomes. “We observed previously that alterations in the PI3K pathway predicted sensitivity to pictilisib in vitro and in vivo,” said Dr Krop, “and that a combination of pictilisib and AstraZeneca’s Faslodex (fulvestrant) is synergistic in hormone receptor-positive xenograph models.”

To test this hypothesis in humans, the FERGI Phase II study enrolled post-menopausal, ER+, HER2- women with advanced or metastatic breast cancer in one of two treatment arms: pictilisib/fulvestrant vs fulvestrant/placebo (N=169).

Patients were stratified by PIK3CA alteration status, resistance to hormone therapy, and/or measureable disease and all had prior treatment with hormonal therapy and chemotherapy. After a median follow-up of 17.5 months, Dr Krop reported the final results for FERGI at the conference.

The median PFS was 5.1 months for fulvestrant/placebo, and 6.6 months for the active combination. Dr Krop said: “I should point out that the FERGI design was hypothesis-generating and would only able to detect large differences between the two arms.”

Toxicity issues

Not only was the combination not effective, it was much more toxic than the comparator. “There were more discontinuations in pictilisib arm,” said Dr Krop.  “In fact, 50% of those on the active study arm discontinued, or had their dose reduced because of drug toxicity.” The most common adverse events were gastro-intestinal related, as well as rash – a similar toxicity profile to other pan-PI3K inhibitors.

Of note, a sub-analysis of patients according to mutational status showed that individuals with the PIK3CA mutation were no more likely to benefit from treatment with pictilisib than those with no mutation. “There was a low response in both arms: 7.9% for the combination vs 6.3% for the placebo arm.”

That said, a further sub-analysis showed that patients who were both ER+, and progesterone receptor-positive (PR+) did experience significant benefit, in fact, a doubling of PFS with the combination as compared to placebo: 3.7 months vs. 7.4 months for pictilisib/fulvestrant.

Should the drug go forward?

“Whether or not this particular drug is the right drug remains to be seen,” commented Eric Winer, also of Dana-Farber. “There are now ongoing Phase III trials of agents that are thought to be more potent inhibitors of the PI3 kinase pathway in combination with hormonal therapy.”

Some of the newer agents, referred to as alpha selective PIK3 inhibitors, appear to be more active in patients who have 3CA mutations, “so, I don’t think we’ll see an end to testing as part of either inclusion criteria or as part of evaluating results of a trial with these agents.”

As to the sub-group that did demonstrate efficacy in FERGI, “patients that are ER/PR-positive…that’s worth looking into further, and there are studies ongoing to do just that,” Dr Winer said. However, even in this group, patients who had PIK3CA mutations – the mutation thought to activate the oncogenic pathway – these patients fared no better with pictilisib. “That in and of itself was a surprise.”

“In all likelihood, where the field is going is to these selective agents,” Dr Winer said. Selective, as in targeting the alpha portion of PI3K overall complex. “I think there is a reasonable likelihood that those agents will actually make it into the commercial setting. In terms of the pan inhibitors – I think it is probably somewhat less likely.”

And pictilisib? “My guess is that it will probably not go forward.”