This article is more than 5 years old. Information may no longer be current.
PIK3CA mutations may predict response to combination treatment for HER-2–positive breast cancer
Click Here to Manage Email Alerts
Women with HER-2–positive breast cancer who harbor PIK3CA mutations appeared less likely to benefit from neoadjuvant treatment with trastuzumab, lapatinib and chemotherapy than women without the mutations, according to the results of a prospective analysis of phase 2 study results.
“We are seeing more and more anti-HER-2 treatments becoming available to treat patients with HER-2–positive breast cancer,” PierFranco Conte, MD, chief of medical oncology at Istituto Oncologico Veneto in Padava, Italy, said in a press release. “By identifying new markers that help us to predict which patients are more sensitive to which anti-HER-2 treatments, we are moving closer to treatment personalization.”
Data from the randomized, phase 2 CHER-LOB study showed the combination of lapatinib (Tykerb, GlaxoSmithKline), trastuzumab (Herceptin, Genentech) and chemotherapy increased the pathologic complete remission (pCR) rate in this patient population compared with chemotherapy plus lapatinib or trastuzumab alone.
Conte and colleagues conducted a prospective biomarker analysis to identify possible predictive factors for sensitivity to different treatments and to evaluate treatment effect based on biomarkers.
The CHER-LOB study included 121 women with HER-2–positive breast cancer randomly assigned to neoadjuvant chemotherapy plus the trastuzumab–lapatinib combination or chemotherapy plus either trastuzumab or lapatinib. The researchers collected tissue samples pre- and post-treatment and evaluated them for HER-2, p95–HER-2, phosphorylated AKT (pAKT), phosphatase and tensin homolog, Ki67 and PIK3CA mutations.
The investigators used fresh-frozen tissue samples to conduct genomic analyses.
Twenty percent of patients had a mutation in PIK3CA exon 20 or 9.
Patients with PIK3CA wild-type and PIK3CA–mutated tumors experienced similar overall pCR rates (33.3% vs. 22.7%); however, when researchers analyzed outcomes among patients assigned to the combination therapy, they observed higher pCR rates among patients with PIK3CA wild-type tumors (48.4% vs. 12.5%).
The researchers observed a significant reduction in Ki67, pAKT and apoptosis measured on residual disease from baseline. Results showed significantly higher mean Ki67 suppression after combined HER-2 blockade therapy (–19.9 vs. –8.5; P = .0036), as well as higher mean pAKT inhibition in patients with hormone receptor negativity (–40.3 vs. –8; P = .0005).
The researchers could not confirm p95–HER-2 or markers of phosphoinositide 3-kinase pathway deregulation as markers of different sensitivity to trastuzumab or lapatinib.
“The presence of PIK3CA mutations may identify patients who are less likely to benefit from dual anti-HER-2 inhibition,” Conte and colleagues wrote. “Taking into account the availability of agents specifically targeting the PIK3 pathway, PIK3CA status seems to be the only marker in this analysis to be studied further in HER-2–positive disease.” – by Cameron Kelsall
Disclosure: Conte reports no relevant financial disclosures. Other researchers report employment with and ownership interests in GlaxoSmithKline and Pfizer, as well as honoraria from Eisai and Merck.
Perspective
Back to Top
Debu Tripathy, MD
The CHER-LOB study correlative data presented in this report add to our understanding of predictive biomarkers for HER-2–targeted therapy, but not quite enough to change clinical practice. Neoadjuvant trials present a unique opportunity to learn about biology and resistance mechanisms that are not afforded by cell line or animal models. In this case, FEC chemotherapy with either trastuzumab (Herceptin, Genentech), lapatinib (Tykerb, Novartis) or the combination was compared using pathologic complete response (pCR) as the primary endpoint. This outcome was analyzed with respect to aberrations known to mediate HER-2 signaling and response based on preclinical and clinical data. For example, in the CLEOPATRA study, activating mutations in the PIK3CA catalytic domain of PI3 kinase were associated with resistance to trastuzumab alone and with pertuzumab (Perjeta, Genentech) in the metastatic setting. PIK3CA mutations also were associated with pCR in the NeoALTTO trial with trastuzumab, lapatinib or the combination. On the other hand, PTEN loss — which also activates the PI3K pathway — did not influence outcome with adjuvant trastuzumab. Inconsistent results also have been seen with the truncated HER-2 peptide p95, which signals autonomously in laboratory models, leading to resistance to trastuzumab but not kinase inhibition. So far, this suggests that no single biomarker is a strong predictor of benefit to trastuzumab or other HER-2–targeted therapies, perhaps with the exception of PIK3CA. But even this marker appears to be dependent on other factors, such as hormone receptor status.In this report, PIK3CA mutation only predicted lower response in the combination trastuzumab–lapatinib arm, and p95 was not predictive at all. A better approach might be to develop a predictive signature that incorporates the activity of several components of relevant networks. A 50-gene expression signature on a subset of the study samples did predict response to lapatinib. Several such signatures developed in other HER-2–treatment tissue sets based on either transcriptomic, phosphoproteomic or other indices have been developed and will need to be validated in additional studies. It is likely that such classifiers also will need to be developed specific to treatment and receptor subtypes. These should then be tested prospectively to determine if treatment guidance using the assay actually improves outcome or maintains overall benefit treating a more select population.
References:
Baselga J, et al. J Clin Oncol. 2014;doi:10.1200/JCO.2013.54.5384.Majewski IJ, et al. J Clin Oncol. 2015;doi:10.1200/JCO.2014.55.2158.
Perez EA, et al. J Clin Oncol. 2013;doi:10.1200/JCO.2012.42.2642.
Scaltriti M, et al. J Natl Cancer Inst. 2007;99:628-638.
Click Here to Manage Email Alerts