Aims To critically evaluate HER2 testing data for 3500 consecutive cases over 28 months for a single laboratory and review these findings in the light of current UK reporting guidelines.

Methods We have reviewed all data relating to the HER2 testing service including reagents and analytical machine, HER2 immunohistochemistry (IHC) and fluorescent in situ hybridisation (FISH) scoring profiles. We have examined the place of double counting and rapid screening of FISH and when it is and is not safe to do so. This analysis has been facilitated greatly by the inclusion of custom scripts embedded in a spreadsheet used for recording the data.

Results There were no differences in scoring profiles in relation to testing machine or reagent batch for both HER2 IHC and FISH. There was excellent concordance in scoring by the biomedical scientists and pathologists providing the service. There is a significant difference between the proximity of scores in double-scored cases with HER2 copy number >6.0 compared with those with copy number <6.00. It is safe to single score with rapid screening of HER2 FISH following a first count with a HER2/CEP17 ratio <1.5 and a HER2 copy number <3.5.

Conclusions Single counting of HER2 FISH supported by a second rapid screen is safe and practicable within strict parameters. Any assessment of proximity of scores in double-scored cases should take into account the HER2 copy number, with greater spreads experienced at higher copy numbers. Centralising HER2 testing in a regional centre allows detailed audit of the entire analytical process.