Publication Year: 2000 Visited: 100

297

Baseline Serum HER2 (sHER2) Levels in the Pivotal Herceptin Breast Cancer Trials: A Comparison of 2 ELISA Methods. Wai Lee Wong, Alex Bajamonde, Betty Nelson, Walter Carney, Robert D Mass, Genentech, Inc, South San Francisco, CA; Oncogene Science Diagnostics, Inc, Cambridge, MA.

The extracellular domain (ECD) of the tyrosine kinase receptor HER2 is variably cleaved or "shed" from the cell membrane. Factors influencing the rate of shedding are not fully understood, but the 105kD sub-unit can be detected in serum using an ELISA. sHER2 was measured at baseline in a subset of subjects enrolled in two, pivotal Herceptin® metastatic breast cancer trials utilizing an ELISA coated with monoclonal antibody 4D5, the murine precursor of Herceptin®. All subjects in these trials overexpressed HER2 in tumor samples at the 2+ or 3+ level determined by immunohistochemistry using the Clinical Trials Assay. We are reporting these results together with sHER2 in these samples as measured with a commercially available ELISA from Oncogene Science Diagnostics, Inc. (OSDI). For both assays, samples were run with control standards of recombinant HER2 ECD. The primary detection antibody in the OSDI assay is NB-3. The 4D5 assay does not detect sHER2 in normal serum and the lower limit of detection is 3.4 ng/ml. The OSDI assay does detect sHER2 in normal serum and the cut-point for elevation was set at > 16 ng/ml. For the 4D5 ELISA, 64% of 438 samples demonstrated elevated sHER2; 33% of the 2+ group and 74% of the 3+ group. The median is 10.6 ng/ml, 25-75 percentile is <3.4-60.7 ng/ml, range is ><3.4-1880.0 ng/ml. For the OSDI assay, 74% of 378 samples demonstrated elevated sHER2; 44% of the 2+ group and 84% of the 3+ group. The median is 41.6 ng/ml, 25-75 percentile is 15.8-185.3 ng/ml, range is 3.4-9766.4 ng/ml. In univariate analyses, only IHC score (2+ vs. 3+) was associated with sHER2 levels. There was a non-significant trend toward higher sHER2 levels with bulk of tumor volume as measured by the number of metastatic sites. These associations were consistent across both assays. The OSDI assay showed higher sensitivity than the original 4D5 assay, although a false negative rate persists. A false positive rate cannot be determined from this analysis. Data correlating sHER2 and Herceptin® benefit will be presented. Further evaluation of sHER2, measured by the OSDI ELISA, as a method to select patients for Herceptin® therapy is indicated. >