Although there is a significant scientific rationale for targeting the PI3K pathway in breast cancer, research findings presented recently at the 2014 San Antonio Breast Cancer Symposium (SABCS) highlight the inherent complexities in using agents directed at this cell-signaling network.
Although there is a significant scientific rationale for targeting the PI3K pathway in breast cancer, research findings presented recently at the 2014 San Antonio Breast Cancer Symposium (SABCS) highlight the inherent complexities in using agents directed at this cell-signaling network. Given the extensive heterogeneity of breast cancer, the identification of patients most likely to respond to these drugs is becoming increasingly pressing and challenging.
These challenges were illustrated at SABCS in data from the phase II FERGI study, which evaluated the pan- PI3K inhibitor pictilisib (GDC-0941) in combination with fulvestrant in patients with metastatic breast cancer. The addition of pictilisib demonstrat- ed statistically significant progression-free survival (PFS) benefits only in the subset of patients with both estrogen receptor (ER)— and progester- one receptor (PR)–positive disease. Thus far, the most successful strategies in attacking the PI3K pathway have been aimed at the downstream mammalian target of rapamycin (mTOR). Yet, while mTOR inhibitors remain the only class of PI3K path- way-targeting agent to have attained regulatory approval, a number of drugs targeting PI3K and dual inhibitors of PI3K/mTOR are now in the late stages of clinical development in patients with breast cancer (Table 1).
PI3K Conducts Cell-Signaling Symphony
The phosphatidylinositol-3 kinase pathway is the most highly dysregulated signaling network in breast cancer. It has also been implicated in development of resistance to other kinds of breast cancer therapy, most significantly endocrine therapy, human epidermal growth factor receptor 2 (HER2)—directed therapy, and chemotherapy, suggesting that agents directed at PI3K may be promising in counteracting drug resistance that limits current treatments. Mutations in the genes encoding the PI3K pathway are particularly common, making this an extremely attractive point for therapeutic intervention.
PI3Ks encompass a family of kinases that phosphorylate phosphoinositide lipids. Three classes and numerous subclasses have been identified, but the most relevant from an oncology perspective are class IA PI3Ks. Their kinase activity is switched on by receptor tyrosine kinases (RTKs) and, upon activation, they phosphorylate their phosphoinositide target, phosphatidylinositol-4,5-bisphosphate (PIP2), to form phosphatidylinositol-3,4,5-triphosphate (PIP3).
Ks. One of the best characterized is Akt, which in turn regulates the function of a broad range of proteins involved in integral cellular processes. A key downstream effector of Akt is mTOR, which is a master regulator of cell growth and metabolism. So important are these two proteins that the network is often referred to as the PI3K/Akt/mTOR pathway.
Methods of Attacking Pathway
Cell survival, growth, proliferation, metabolism, motility, and adhesion are a small representation of the cellular functions controlled by PI3K and its downstream elements. PI3K and the phosphatase that opposes its action, phosphatase and tensin homolog (PTEN), are conductors of a cell-signaling symphony.
In breast cancer, aberrant activation of at least one component of this pathway can be observed in more than 70% of cases. Mutations in the genes that encode PI3K itself are common, observed in around 18% of patients. PI3K is a heterodimeric protein composed of a catalytic (p110) subunit and a regulatory (p50/p55/p85) subunit. Three isoforms of the catalytic subunit exist (a, β, and δ), which are encoded by three genes: PIK3CA, PIK3CB, and PIK3CD, respectively. The PIK3CA gene is particularly frequently altered in breast cancer (20%-25%). Mutations occur most often in the parts of the gene that encode the helical and kinase domains of the PI3K protein, and three “hotspots” have been identified—E542K, E545K, and H1047R—all of which produce a constitutively active kinase that drives oncogenesis.
The frequency of PIK3CA mutations varies considerably by type and subtype of breast cancer. PIK3CA mutations appear most common in patients with ER-positive or HER2-positive disease (30%-40%) and are much less prevalent in triple-negative breast cancer. Numerous other types of alterations in the PI3K/Akt/mTOR pathway have been implicated in the development of breast cancer, and substantial time and resources have been committed to developing drugs targeting almost every aspect of the pathway. As it stands now, everolimus (Afinitor) remains the only drug targeting this pathway that the FDA has approved in a breast cancer setting. The agency approved everolimus in 2012 as a treatment for postmenopausal women with advanced hormone receptor (HR)—positive, HER2-negative breast cancer in combination with exemestane after failure of treatment with the nonsteroidal aromatase inhibitors (AIs) letrozole or anastrozole.
Various inhibitors of the PI3K protein have now been developed, including pan-isoform inhibitors that target all class I PI3Ks, isoform-specific inhibitors, and, thanks to DNA sequence similarities between mTOR and PI3K, dual inhibitors of both enzymes.
The principal argument for pan-isoform inhibitors of PI3K is that most cancer cells express multiple isoforms of PI3K, which have redundant oncogenic activity. Although these agents continue to be evaluated in clinical trials, they have demonstrated modest activity to date, particularly as monotherapy.
The greatest enthusiasm has been for isoform-specific inhibitors; these agents should offer the advantage of less toxicity, thus allowing them to be tolerated at doses that could result in more complete and reliable inhibition of kinase activity. Dual PI3K/mTOR inhibitors also have shown impressive activity in breast cancer, with tolerable side effect profiles, and offer the advantage of a more complete blockade of the PI3K/Akt/mTOR pathway. Both isoform-specific PI3K inhibitors and dual PI3K/mTOR inhibitors are less advanced in clinical development. The most common adverse events observed with PI3K inhibitors as a class are gastrointestinal toxicity (including diarrhea, nausea, and vomiting), rash, hyperglycemia, and mood changes (anxiety, depression, hallucinations. and affective disorder). The latter have most frequently been reported with buparlisib (BKM120) and are thought to result from the ability of this drug to cross the blood-brain barrier. Rare cardiac events have also been reported with pictilisib.
Combinations Better Than Monotherapy
While single-agent therapy with PI3K inhibitors has proved disappointing, combination therapy appears much more fruitful, and a wide variety of combinations are being clinically evaluated. In fact, evidence suggests that aberrations in the PI3K pathway may be a critical mechanism employed by tumors to enable them to become resistant to various different lines of established therapy in breast cancer, including endocrine therapy, HER2-targeted therapy, and chemotherapy. Thus, blocking the PI3K pathway in combination with other agents could prove extremely useful in counteracting drug resistance that often limits the utility of these standard breast cancer treatments. The most exciting recent developments have come from the combination of PI3K inhibitors and endocrine therapies.
The rationale supporting this combination stems from observations that PIK3CA mutations are significantly related to the expression of HRs—both ER and PR—and it is thought that PI3K signaling may stimulate transcriptional activation of these receptors. Thus, aberrant activation of the PI3K pathway may enable tumors to overcome repression of HR pathways by endocrine therapy. Since a high proportion of breast cancers express HRs, endocrine therapy represents an important class of breast cancer therapy and the development of resistance, a substantial challenge. So far, the only definitive clinical evidence for synergy between PI3K pathway-targeting agents and endocrine therapy comes from trials of mTOR inhibitors. The phase III BOLERO-2 study demonstrated the benefit of adding everolimus to an AI inhibitor in patients with HR-positive disease refractory to AI treatment.
PI3K inhibitors also are under investigation now in combination with endocrine therapies. Buparlisib is the most advanced in clinical development in this respect, and the results of the phase III BELLE trial are eagerly anticipated.
PI3K Findings at SABCS
Swathes of data from early-phase clinical trials reported at SABCS for buparlisib and pictilisib, both pan-isoform inhibitors, and the PI3Ka-specific inhibitors taselisib (GDC-0032) and BYL719 showed the promising efficacy and tolerability of a number of these combinations (Table 2).
Unfortunately, the FERGI study, presented by Ian Krop, MD, PhD, a medical oncologist at the Dana- Farber Cancer Institute in Boston, demonstrated that there was no benefit from the combination of pictilisib and fulvestrant in patients with ER-positive, AI-resistant advanced/metastatic breast cancer.
With 8 therapies approved by the US Food and Drug Administration (FDA) for the treatment of patients with advanced renal cell carcinoma (RCC), practitioners are faced with the challenge of selecting the most appropriate therapies for their patients within this crowded therapeutic landscape.
However, an unplanned posthoc analysis showed that patients with tumors that were positive for both HRs did derive benefit from the combination, with a 56% reduction in the risk of PFS of 7.2 months with pictilisib versus 3.7 months for fulvestrant and placebo (P = .002). Further study is warranted.
Juric et al reported early results from a phase I study of BYL719 and letrozole in combination with LEE011, a cyclin-dependent kinase (CDK) 4/6 inhibitor, a combination considered worth exploring because activation of CDKs has also been implicated in resistance to AIs, and synergy between these agents has been observed in the preclinical setting. Results from the first two arms evaluating LEE011 in combination with letrozole and BYL719 in combination with letrozole demonstrated acceptable safety and preliminary efficacy. The study continues, and once the maximum tolerated dose or recommended phase II dose has been identified, patients will be enrolled in a third arm evaluating the triple combination.
The Search for Biomarkers
Though clearly promising, the optimal strategy for targeting the PI3K pathway in breast cancer remains to be identified. What has become clear is that not all patients benefit, and methods for identifying those patients who are most likely to respond and the molecular mechanisms underlying those responses are urgently needed. This is particularly important since breast cancer is an extremely heterogeneous disease and activation of other pathways can ultimately lead to the development of resistance, which frequently occurs after treatment with PI3K inhibitors, as with other targeted therapies.
In preclinical studies, the biomarkers with the most significant potential to predict treatment efficacy and resistance include PIK3CA mutation and PTEN loss. In clinical studies, however, establishing the significance of PIK3CA mutations has proved challenging.
While some clinical trials have suggested that patients with PIK3CA mutations may derive more benefit from PI3K inhibitors than those without the mutations, the results of the FERGI study somewhat surprisingly demonstrated that PIK3CA mutation status had no effect on the efficacy of pictilisib when combined with fulvestrant.
Meanwhile, several studies have shown that loss of PTEN is a prominent mechanism of resistance to PI3K inhibition. A recent study published in Nature examined the genomic evolution of a patient with metastatic breast cancer with PIK3CA mutations who was treated with BYL719. The patient’s tumor initially responded well to BYL719 but ultimately developed resistance and metastases. The researchers sequenced material from 14 metastatic sites and compared the genetic profile with that of the pretreatment tumor sample. All of the samples that had become refractory to BYL719 had developed alterations that resulted in PTEN loss.
Tumors with PTEN deficiency have been shown to be more dependent on PI3Kβ activity; thus, combined inhibition with inhibitors of PI3Ka and PI3Kβ or use of pan-isoform inhibitors in these tumors may be more effective. Several preclinical studies looking at the antitumor effects of combining PI3Ka and PI3Kβ inhibitors in patients with PTEN loss demonstrated that this combination is synergistic and may reverse resistance.
In HER2-positive breast cancers or those with high levels of RTK signaling, resistance may occur through loss of feedback suppression on RTKs such as HER3 and insulin-like growth factor 1 receptor (IGF1R). Using inhibitors of these drugs in combination with PI3K inhibitors may be effective and, indeed, several phase I studies evaluating IGF1R and HER3 inhibitors in combination with BYL719 are ongoing.