The PI3K pathway receives plenty of hopeful attention as an alternative for ER+ patients in whom hormonal therapy has come to a dead end, with aromatase inhibitor resistance. But oncologists are rightfully tentative about this, because PI3K is central to so many processes beyond the mammary epithelium.
Now, while several drugs that target PI3K are in the earliest stages of clinical testing, cancer researchers at Washington University in St. Louis have published an elegant set of experiments that propose how inhibitors of PI3K could (or should) be used, and for whom. The studies presented in Breast Cancer Research involve tumor cell lines and biopsy samples, not women with breast cancer. But they offer a compelling strategy for chasing down and eradicating ER+ cells, perhaps "obviating the requirement for long-term endocrine treatment" in some patients.
Not slower growth, but death
Hormone treatment slows the growth and proliferation of breast cancer cells, but it does not kill them. The hope is that PI3K inhibitors can have that effect.
Responding to extracellular signals including the HER family and the insulin growth factor receptor, PI3K (phosphoinositide 3-kinase) catalyzes the first step in a pathway that is integral to cellular metabolism, differentiation, and survival. At least a fourth of women with breast cancer have mutations in this pathway, most of them in PIK3CA, the gene that encodes the catalytically active subunit of PI3K. The result, often, is hyper-activation of the enzyme.
Two years ago, Matthew Ellis and the St. Louis team reported that inhibiting PI3K in combination with estrogen deprivation is a way to kill ER+ breast cancer cells, at least those growing in cell culture. Now they have expanded these studies to include breast-cancer cell lines with differing mutations in the PI3K pathway (including PIK3CA), and challenging them with 3 drugs that target the effects of PI3K. Two of these drugs are still in testing (BKM120 and BGT226, both Novartis). The third is Everolimus, which acts against mTOR, downstream of PI3K.
The new studies show that BKM120 and BGT226 both induced high levels of apoptosis in ER+ breast cancer cell lines, notably those with mutations in PIK3CA but also some with mutations in PTEN, which reverses the catalytic actions of PI3K. Everolimus showed only a modest effect. Adding estrogen to the cell lines reduced this tumor-killing activity, but then pumping up the dose of the drugs progressively restored it.
Emulating aromatase inhibitors
To emulate the clinical situation, the team also grew some samples of the tumor cell lines in a medium that had been stripped of hormones, either for 9 months or for 1-3 weeks, before exposing them to the PI3K inhibitors. It required higher concentrations of BKM120 and BGT226 to reach the same level of apoptosis in these estrogen-deprived cells, but adding fulvestrant (which down-regulates the estrogen receptor) strongly potentiated cell death. Re-exposing the cells to estradiol had the same effect.
"A combination of fulvestrant and a PI3K pathway inhibitor may be an effective approach for aromatase inhibitor resistant advanced breast cancer," write the Washington University researchers, especially for relapsed ER+ patients with a mutation in PI3KCA.
The modest effects of Everolimus on these estrogen-deprived cells could help to explain the negative results of a recent randomized Phase 2 trial that added the aromatase inhibitor letrozole to Everolimus, they point out. Perhaps a PI3K inhibitor might be more effective in this situation.
In 2009, researchers at Memorial Sloan-Kettering Cancer Center complicated the picture by demonstrating an association between the PI3KCA mutation and improved survival. Does this mean that few of the women threatened by aromatase inhibitor resistance actually have the mutation that is most vulnerable to PI3K inhibitors?
Quite the contrary; about half of biopsy samples from 51 patients with recurrent or metastatic breast cancer show that mutation, according to the Washington University studies. Patients with this mutation tend to relapse later than others, although analysis of this small sample suggested that their overall survival is no better.
For the one-fifth of ER+ patients with the PI3KCA mutation, suggests oncologist Jose Baselga of Massachusetts General Hospital, "it could well be that a less aggressive therapy, including PI3K inhibitors and hormonal therapies, should be considered." The Washington University team has reached the same conclusion: Estrogen deprivation plus PI3K inhibitors for estrogen-dependent ER+ tumors with the mutation, and for ER+ patients with aromatase inhibitor resistance, fulvestrant plus a PI3K inhibitor.
Predictably, that's not the end of the story. Vanderbilt University researchers reported a few weeks ago that breast tumor cells responded to the Exelexis PI3K inhibitor XL147 by pulling other receptors than HER2 into play, allowing downstream events in the P13K pathway to recover somewhat. Thus, at least in HER2+ patients, still another drug may be necessary.