226P - Exploratory analysis of the relationship between HER2 expression (by QRT-PCR) and efficacy with first-line trastuzumab emtansine (T-DM1) vs trastuzu...

Date 30 September 2012
Event ESMO Congress 2012
Session Poster presentation II
Topics Biomarkers
Breast Cancer, Metastatic
Presenter Edith Perez
Authors E.A. Perez1, S. Hurvitz2, L.C. Amler3, V. Ng4, E. Guardino5, L. Gianni6
  • 1Hematology / Oncology, Mayo Clinic, 32246 - Jacksonville/US
  • 2Medical Oncology, UCLA Jonsson Comprehensive Cancer Center and Translational Oncology Research International, Los Angeles/US
  • 3Oncology Biomarker Development, Genentech, Inc., South San Francisco/US
  • 4Biostatistics, Genentech, Inc., South San Francisco/US
  • 5Breast Medical Oncology, Genentech, Inc., South San Francisco/US
  • 6Medical Oncology, San Raffaele Hospital, 20132 - Milano/IT



In TDM4450g (NCT00679341), 137 pts with HER2-positive MBC and no prior treatment for metastatic disease were randomized to T-DM1 or HT. Median progression-free survival (PFS) was 14.2 vs 9.2 mo, respectively (HR = 0.594); objective response rates (ORR) were 64.2% and 58.0%, respectively (Hurvitz 2011). In exploratory analyses of single-arm phase 2 MBC studies of T-DM1, greater expression of HER2 mRNA (assessed by qRT-PCR) was associated with greater ORR and longer PFS (Burris 2011; Krop 2012, in press). We now report a retrospective exploratory analysis of the relationship between qRT-PCR HER2 expression and efficacy outcomes in pts receiving T-DM1 or HT in TDM4450g.


Pts were randomized 1:1 to T-DM1 3.6 mg/kg q3w or H 6 mg/kg q3w (8 kg/mg in cycle 1) + T 75 or 100 mg/m2 q3w and treated until disease progression. HER2 expression was measured by qRT-PCR in archival tissue samples from all randomized pts. Efficacy outcomes included investigator-assessed PFS and ORR. The analysis results presented here are based on the clinical data with a median follow-up of ∼14 mo in each arm.

Table: 226P

Median pre-treatment HER2 levels, concentration ratio units (range)
(n = 61) 8.7 (0.5-105.0) (n = 55) 10.3 (0.4-103.0)
Median PFS, months
HER2 expression HT T-DM1 Hazard ratio relative to HT (95% CI)
Low (n = 32) 9.8 (n = 26) 10.6 0.85 (0.44-1.67)
High (n = 29) 8.8 (n = 29) Not reached 0.39 (0.18-0.85)
ORR, %
HER2 expression HT T-DM1 Odds ratio relative to HT (95% CI)
Low (n = 31) 58.1 (n = 26) 53.8 0.84 (0.30-2.41)
High (n = 29) 65.5 (n = 29) 72.4 1.58 (0.50-4.98)


In total, 122 tumor samples from 137 randomized pts were available for analysis of pre-treatment HER2 by qRT-PCR, resulting in a valid result for 116 pts (6 samples were below the limit of quantification). Efficacy outcomes by low ( Conclusions

These results from an exploratory analysis suggest that the magnitude of efficacy from T-DM1 vs HT may correlate with HER2 expression levels (by qRT-PCR); this effect appears to be most pronounced for PFS. Analyses exploring the relationship between these findings and baseline characteristics are planned. Similar analyses are planned for ongoing phase 3 studies of T-DM1.


S. Hurvitz: SA Hurvitz has received support for study-related travel from Roche and research funding for her institution from Roche.

L.C. Amler: Dr. Amler is employed by Genentech and has stock ownership in F. Hoffmann-La Roche.

V. Ng: Dr. Ng is employed by Genentech and has stock ownership in F. Hoffmann-La Roche.

E. Guardino: Dr. Guardino is employed by Genentech and has stock ownership in F. Hoffmann-La Roche.

L. Gianni: L Gianni has had compensated consultant/advisory relationships with Genentech/Roche, GlaxoSmithKline and Pfizer.

All other authors have declared no conflicts of interest.