What are NCI-designated cancer centers using for breast cancer HER2 testing?

Abstract

626 Background: After 15 years of HER2 testing in breast cancer, various testing methods are used in practice. Studies comparing methods are published but no consensus exists on which method is best. Our study provides a benchmark on the use of breast HER2 testing methods in leading U.S. cancer centers. Methods: We conducted an IRB-approved web survey of 58 NCI cancer centers (pathologists and oncologists) providing adult breast cancer care. The survey included 14 questions on breast HER2 testing methods, and reflex and retest practices. We analyzed results using simple frequencies and Fisher's exact test. Results: We achieved a response rate of 98% (57/58 sites). In this cohort, 42% (24/57) of sites conduct HER2 testing for breast cancer using an IHC method with reflex to FISH. Of these, all sites reflex IHC 2+ results and 54% (13/24) automatically reflex IHC results beyond 2+ (see table, results are not mutually exclusive). Concurrent primary FISH and IHC testing is conducted at 32% (18/57) of sites; FISH only testing at 18% (10/57); concurrent primary SISH and IHC testing at 5% (3/57); concurrent primary CISH and IHC testing at 2% (1/57), and CISH only testing at 2% (1/57) of sites. The choice of testing protocol had no correlation with the size of institution, metro vs rural location, NCCN membership, or whether the site acts as a reference lab. However, sites where oncologists always or often request a specific test method were more likely (75%, 21/28) to use FISH as a primary method vs as a reflex method (38%, 9/24), p=.0108. Repeat HER2 testing on surgical tumor samples, after the core biopsy, was reported by 47% of sites (27/57); retesting of relapsed patients by 63% (36/57); and retesting for progressive metastatic disease by 56% (32/57). Conclusions: For HER2 breast biomarker testing, concurrent FISH and IHC testing and expanded reflex testing, beyond IHC 2+ results, have become a common practice at the NCI designated cancer centers. [Table: see text]