Biomarker (BM) Analyses of a Phase II Study of Neoadjuvant Pertuzumab and Trastuzumab With and Without Anthracycline (ATC)-Containing Chemotherapy for Treatment of HER2-Positive Early Breast Cancer (BC) (Tryphaena)

ABSTRACT Background TRYPHAENA showed that concurrent administration of trastuzumab (H) plus pertuzumab (P) with ATC in the neoadjuvant setting resulted in similar cardiac tolerability to sequential administration of ATC or to an ATC-free regimen. In addition, neoadjuvant P with H administered in combination with ATC- or carboplatin-based standard chemotherapy (CT) resulted in high pathological complete response (pCR, [ypT0/is]) rates, regardless of the CT regimen, in patients (pts) with HER2-positive BC. The molecular profile of pts who achieved a pCR was compared with those who did not. Methods Baseline core biopsies and sera from over 90% of the 225 pts enrolled in TRYPHAENA were available for BM analyses. The BMs and methods of assessment included: HER1/2/3, PTEN, and IGF1R protein expression by IHC (and FISH for HER2); HER1/2/3 mRNA expression by qRT-PCR; c-myc and TOP2A by FISH and PCR-based PIK3CA analyses detecting 8 mutations. Potential relationships between BMs and pCR were measured primarily using categorical analysis methods (chi-square test of association and Cochran–Mantel–Haenszel statistic). Median values were used as cut-offs except for TOP2A and PTEN where cut-offs were applied following a biologic rationale. Results Besides HER2 protein levels, no other measured BM was associated with pCR when BM samples from pts in each treatment arm were considered separately. TOP2A amplification (≥ 2.0 ratio) was observed in 32% of pts and was not associated with response to ATC regimen. There is no strong biologic rationale for correlation between BMs and pCR outcomes after treatment with different CT regimens; therefore, data from all pts were pooled. In the combined analysis, HER2 protein levels remained significantly correlated with pCR and high levels of HER2 mRNA were also indicative of improved pCR rates. PIK3CA mutations were seen in 24% of pts and were not associated with pCR. Conclusion This analysis supports the use of HER2 overexpression as the strongest predictive and clinically relevant BM for pCR after treatment with H + P + CT neoadjuvant regimen for pts with HER2-positive BC. Disclosure A. Schneeweiss: Prof. Schneeweiss is a member of an advisory board for F. Hoffmann La Roche and also has conducted reearch sponsored by F. Hoffmann La Roche. S. Chia: Dr Chia is a member of an advisory board for F. Hoffmann La Roche and also has conducted reearch sponsored by F. Hoffmann La Roche. R. Deb: -I have received consultancy fees from Roche Pharmaceuticals -I have received a fee for advisory board work for Roche. -Roche have paid for registration, travel and accommodation costs to attend courses. J. Ratnayake: Dr Ratnayake is an employee of F Hoffmann La Roche. A. Kiermaier: Dr Kiermeier is an employee of F Hoffmann-La Roche. V. McNally: Dr McNally is an employee of F Hoffmann-La Roche and owns stock in the company. G. Ross: Dr Ross is an employee of F Hoffmann-La Roche and owns stock in the company. J. Cortes: Dr Cortes has participated in advisory boards and received honoraria from F Hoffmann La-Roche. All other authors have declared no conflicts of interest.