46
IASLC ATLAS OF ALK TESTING IN LUNG CANCER
2010). In one study, 1+ and 3+ IHC scores defined ALK negative and ALK positive, respectively, with a 2+ score considered to be an equivocal result (Yi 2011). This classification is similar to that for HER2 testing in breast cancer. The use of tyramide amplification was recommended to increase the sensitivity of detection, but nonspecific staining of tumor and nontumor cells was reported at a high concentration of ALK1 antibody (Mino-Kenudson 2010). D5F3 IHC versus FISH IHC with the D5F3 clone has been compared with FISH in three studies (Table 2) (Mino-Kenudson 2010, Martinez 2013, Minca 2013). The sensitivity was 83% to 100%, and the specificity was 99% and 100%. Among six specimens that were ALK positive by FISH in one study, one specimen was ALK negative by IHC, but none of the specimens that were ALK negative by FISH were ALK positive by IHC (Martinez 2013). In addition to these studies, an international panel of pathologists evaluated D5F3 IHC in a series of lung adenocarcinoma with known ALK genotype. Overall, the sensitivity was 90% and the specificity was 95% for IHC compared with FISH, and the interobserver concordance in scoring was high (Hirsch 2013).
Table 2. Staining Features of IHC with D5F3 Antibody
Study MinoKenudson et al., 2010 Martinez et al., 2013 Minca et al., 2013a Hirsch et al., 2013b* No. of D5F3 Specimens Dilution 153 1:100 Antigen Retrieval EDTA (pH 8.0) in pressure cooker Standard on BenchMark XT Heat mediated with BenchMark XT Standard on BenchMark XT Detection and Amplification System EnVision+ Scoring IHC Positive Threshold IHC Sensitivity (vs. FISH) 100% IHC Specificity (vs. FISH) 99%
0, 1+, 2+, or >10% 3+ and % of positive tumor cells tumor cells 0 vs. + ?10% positive tumor cells Positive
79
1:50
ultraView
83%
100%
231
1:100
OptiView
0 vs. +
94%**
100%
98
Prediluted
OptiView and OptiView Amplification
0 vs. +
Any strong cytoplasmic staining
90%
95%
a
Two false-negative results on IHC testing o