Accurate measurement of HER2 amplification and/or overexpression is vital due to the prognostic and potential therapeutic implications of being HER-2 positive.
Two main methods exist for the determination of HER2 gene amplification and protein expression in breast cancer specimens, fluorescence in situ hybridisation or FISH (direct or indirect) and IHC. The former measures gene amplification in breast cancer specimens whilst the latter measures protein expression.
In the FISH technique, fluorescence labelled cDNA probes for HER2 (chromosome 17 q11.2–q12.0) and chromosome 17 centromeres [(chromosome enumeration probe 17 (CEP17)] are used. The HER2 gene appears as a red/orange signal and the CEP17 appears as a green. A ratio of HER2:CEP17 copy number > 2 denotes amplification when taken over an average of at least 60 invasive cancer cells.
In the IHC technique, membrane staining of malignant cells is assessed using the appropriate antibody in fixed tumour blocks. It is a semi-quantitative technique with the intensity of staining reflecting the amount of protein present. From the therapeutic standpoint, the recommended scoring system to evaluate IHC staining is shown in Table below.
The IHC scoring system of HER2
This system has been used in pivotal trials [24
] evaluating the efficacy of a humanised anti-HER2 monoclonal antibody therapy in woman with advanced breast cancer and has been approved by the appropriate authorities in the USA and the EU.
Ridolfi et al
(23) tested 750 consecutive invasive carcinomas for HER2 overexpression using both IHC and FISH techniques. He found that whereas the concordance rate between FISH and IHC (positive = 3+, negative = 1+, 0) was 98.7%. FISH was positive in 36% of specimens scored 2+ by IHC (Figure ) which would have been scored negative by this technique. A pictorial comparison between the two testing methods is shown in figure [25
HER2-positive testing (a) by IHC (courtesy of Dc M. J. Kornstein, Medical College of Virginia) and (b) by FISH.
Hoang et al
] also reported a high concordance rate for HER2 positivity between IHC score 3+ and FISH (89%). However, there was a low interobserver reproducibility in separating 2+ from 3+cases. This data suggests that IHC is a useful initial test and cases scoring 2+ should be considered for FISH testing.
Mass from Genetech [27
] recently presented the concordance rates between FISH and the clinical trial assay (CTA), a immunohistochemical technique in 623 samples randomly selected from the two pivotal Herceptin trials. FISH positivity was observed in 4.2%, 6.7%, 23.9% and 89.3% of CTA 0, 1+, 2+ and 3+, respectively.
Conflicting results regarding the best antibody to use in determining the IHC status of HER2 were presented by Falo et al
]and Bartlett et al
], respectively. The former found that monoclonal antibody CB11 was more reliable than the Dako polyclonal antibody compared against FISH, whereas Bartlett reported a higher accuracy for the polyclonal antibody (87.4%) compared with the monoclonal (83.8%). Gancberg et al
] studied the sensitivity of three frequently used antibodies and found that the monoclonal antibody TAB250 had the lowest misclassification rate against FISH in 160 breast cancer specimens.
In view of such conflicting results and the significant inter-observer variation IHC should be performed in standardised reference laboratories with a large caseload. In the United Kingdom, all laboratories performing IHC as a predictive test must participate in an external quality assurance scheme and guidelines recommend that they should be performing at least 250 assays per year (100 for Her2 FISH testing. Alternatively, FISH testing could be performed as the gold standard. The latter option may need to be introduced gradually until the technique and required expertise are established in breast cancer centres.
FISH has been described as not only being both more accurate in determining HER2 status but also is a better predictor of prognosis and response to Herceptin. In addition, DNA is more stable than protein, interpretation is easier and there is less inter-operator error. FISH however is both more labour intensive and expensive.
For these reasons in the United Kingdom, a two-phase testing regimen exists for assessing HER2 status. Initial IHC testing of 0 or 1+ is reported as negative with 3+ being reported as positive. A score of 2+ is reported as intermediate with these cases being referred for FISH to establish a definitive diagnosis.