Canadian ALK (CALK): A pan-Canadian multicenter study to optimize and standardize ALK immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) for ALK gene rearrangements.
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
8096 Background: ALK gene rearrangement (ALK+) has been found in 3-5% of advanced non-small cell lung cancer patients. FISH is considered the “gold standard” for identification of ALK+ tumors, but its cost-effectiveness and adoption as a screening assay has been debated. Recent reports suggested that ALK IHC may serve as an alternative screening or possibly a diagnostic method. In this context, CALK was initiated to assess the feasibility of implementing ALK IHC and/or FISH assays across Canadian hospitals. Methods: FISH-confirmed 22 ALK+ and 6 ALK- tumors were used as study samples. Unstained sections and scanned images of HE-stained slides from each tumor block were distributed to participating centres. IHC protocols with best signal to noise ratio using the 5A4 (Novocastra) or ALK-1 (Dako) antibodies were developed for various auto-stainers and implemented to suit the existing conditions of the participating centres. A common FISH protocol using the ALK break-apart probe (Abbott Molecular, Chicago, IL) was developed based on published reports. H-score was used to assess IHC. FISH signals were scored in 100 tumor cells/case by 2-3 pre-trained persons. A second round IHC study using newly distributed slides was completed by 8 centres. Results: Independent IHC scores from 12 centres and FISH scores from 11 centres were collected and analysed. The intraclass correlation coefficients (ICC) between centres for IHC and FISH were 0.84 and 0.68, respectively. Following the analysis of initial IHC results, a second round study resulted in improved ICC of 0.94. One of 23 tumors revealed IHC-/FISH+ discrepancy, with the FISH revealing unusual signal configurations that suggested an atypical rearrangement. However, the sensitivity and specificity of FISH results across centres using the 15 aberrant signals cut-off ranged from 86.7-100% and 100%, respectively. Conclusions: Standardization across multiple centres for ALK testing by IHC and FISH can be achieved. IHC detected all FISH+ ALK tumors, except for one discrepant case with atypical FISH finding of unknown clinical implication. The study was supported by a Pfizer Canada grant.
