Pathway Alterations Predict Response to HER2-Targeted Therapies in Breast Cancer

October 19, 2016
Dave Levitan
Dave Levitan

Response to the HER2-targeted therapies lapatinib and trastuzumab are correlated with pathway-level genetic alterations, but not specific gene mutations.

An analysis from the NeoALTTO trial showed that response to the HER2-targeted therapies lapatinib and trastuzumab are correlated with pathway-level genetic alterations, but not specific gene mutations. The combined mutation status of the PIK3CA-related pathway and the RhoA pathway could help guide treatment decisions.

Earlier reports from the NeoALTTO trial showed that women with early-stage HER2-positive breast cancers had better rates of pathologic complete response (pCR) when trastuzumab and lapatinib are combined (though there was no difference with regard to event-free or overall survival). Several genetic predictors of response have been previously suggested. “However, due to their modest positive and negative predictive values, none of these markers is clinically useful in ruling out patients from receiving HER2 targeted therapies or in selecting one therapy over another,” wrote study authors led by Lajos Pusztai, MD, DPhil, of Yale Cancer Center in New Haven, Connecticut.

The new analysis involved whole-exome sequencing of 203 patients included in the NeoALTTO trial; the study had one group receiving trastuzumab alone, one receiving lapatinib alone, and one receiving a combination of the two drugs. The results were published in Annals of Oncology.

A total of 12 individual genes were found to have higher than background mutation rates. Of those, only one-PIK3CA-was associated with response in all three treatment arms of the trial, with an odds ratio (OR) for pCR of 0.42 (P = .0185).

On the pathway level, mutations in 33 out of 714 different pathways showed significant associations with response. Specifically in the trastuzumab arm of the trial, 23 pathways showed an association with residual disease (RD); all of these pathways included PIK3CA, though the gene itself was not mutated in most patients.

In the lapatinib-only arm, three pathways were significantly associated with higher pCR rates, and none were associated with RD. Mutations in the “regulation of RhoA activity” pathway, involving 48 individual genes, had the highest correlation with pCR, with an OR of 14.8 (P = .0083). In the full study cohort, it was still associated with pCR, with an OR of 3.77 (P = .0009), though it was not in the trastuzumab-alone arm.

Patients who had mutations in what authors called the “trastuzumab resistance network” (the 23 pathways associated with RD), but intact RhoA pathways, had a 2% pCR rate with trastuzumab alone; adding lapatinib, though, increased the pCR rate to 45%. The 46 patients who had no mutations in either of the two gene sets had a 6% pCR rate with lapatinib alone, but had the highest rate of pCR (52%) with trastuzumab alone.

“Mutations in the RhoA pathway are associated with high sensitivity to lapatinib therapy and mutations in a PIK3CA gene network are associated with relative resistance to trastuzumab,” the authors concluded. “The mutation status of these two pathways, in combination, could define a patient population with very low response rate to trastuzumab alone that can be augmented by adding lapatinib or substituting trastuzumab with lapatinib.”