In 2009, a College of American Pathologists expert panel published supplemental HER2 testing recommendations suggesting that cases with between 5% and 50% individual cells amplified by fluorescence in situ hybridization be reported as "heterogeneous for HER2 gene amplification." We examined the implications of applying these recommendations to clinical practice in 1,329 consecutive breast cancer cases. By ratio criteria, 23.2% of cases met the proposed criteria for heterogeneity, of which 81.6% were not amplified and 15.5% were equivocal by standard criteria. In contrast, the proposed criteria based on HER2 signals per cell classified only 6.5% of cases as heterogeneous, of which only 8% (7/87) were not amplified and 79% (69/87) were equivocal by standard criteria. These results show that the 2 proposed criteria sets are not equivalent and that the ratio-based definition results in large numbers of nonamplified cases being classified as heterogeneous. Further definition of optimal criteria with clinical relevance is needed before HER2 heterogeneity reporting is adopted in routine practice.