Anaplastic Lymphoma Kinase (ALK) Copy Number Variation (CNV) as a Quality Assurance Measure for Tumor Adequacy in FISH Analysis
Nirali M Patel, Kathleen Kaiser-Rogers, Kathleen Rao, Travis Sapp, Margaret L Gulley. University of North Carolina, Chapel Hill, NC
Background: ALK rearrangement is found in ∼5% of lung adenocarcinomas. Tissue specimens used for ALK testing may contain scant tumor. Therefore, surrogate indicators of specimen adequacy are needed to ensure sufficient numbers of malignant cells are evaluated during testing. We examined whether ALK CNV, which is commonly seen in lung adenocarcinoma and less often in benign cells, could serve to assure that adequate tumor was present in the portion of tissue on which FISH analysis was performed.
Design: Records were obtained for all lung adenocarcinoma specimens undergoing ALK rearrangement testing using Vysis Break Apart FISH Probe at UNC Hospitals Cytogenetics Laboratory from April 2011 to August 2012. Records were reviewed for percent tumor nuclei on H&E stained slides, percent cells with intact ALK signal CNV versus ALK rearrangement, and number of intact ALK signals per cell.
Results: A total of 184 cases were reviewed, with complete information available on 111. Two tumors had ALK rearrangement (1.1%), and one could not be interpreted due to acid decalcification and failed hybridization. 150 cases (81.5%) had aberrant numbers of ALK signals detected in >20% of cells. Nine of these cases had 1 ALK allele, while 141 cases had extra ALK signals (range - 3 to 15 copies per cell). The remaining 33 cases had 2 copies of the ALK gene per cell with neoplastic cells confirmed to be present. Among cases with ALK CNV, correlation between pathologist-estimated tumor burden and proportion of cells with ALK CNV was R2=0.0146. This result suggests no linear relationship between the proportion of malignant cells as assessed on H&E and the proportion with ALK CNV as assessed by FISH.
Conclusions: Among tumors negative for ALK rearrangement, ALK CNV was seen in 81.5%, providing support for the presence of neoplastic cells in analyzed tissue. Possible reasons for lack of association between estimated tumor burden and percent of cells with CNV include acquired CNV in a subset of neoplastic cells, CNV in non-neoplastic cells, and selective enrichment for neoplastic nuclei during FISH. While ALK CNV likely represents a surrogate marker of neoplastic cells and has potential as a real-time quality assurance measure for ALK gene rearrangement testing, additional investigation is needed to identify the reason(s) for the lack of correlation observed between the pathologist-estimated tumor burden and the proportion of cells with ALK CNV.
Category: Quality Assurance
Tuesday, March 5, 2013 9:30 AM
Poster Session III # 284, Tuesday Morning