Wolff, Daynna J.

Abstract Context. —Fluorescence in situ hybridization (FISH) is a common method used to determine HER-2 status in breast cancer. Limited information is available concerning reproducibility of FISH in determining HER-2 gene amplification. Objective. —To present proficiency testing results of FISH for HER-2 conducted by the Cytogenetics Resource Committee of the College of American Pathologists/American College of Medical Genetics. Design. —During the past 5 years, unstained sections from 9 invasive breast carcinomas were used for HER-2 FISH proficiency testing, allowing for comparison of FISH results among a large number of laboratories. Additional data were collected using an educational (ungraded) challenge and supplemental questions in the surveys. Results. —The number of laboratories participating in HER-2 FISH proficiency testing has increased steadily during the past 5 years (from 35 in 2000 to 139 in 2004). Reproducibility of test results among laboratories was excellent for breast tumors with low copy number (no HER-2 amplification) and for breast tumors with high copy number ( HER-2 amplification). However, there was considerable variation in interpretation of results for a tumor with low-level HER-2 amplification that was tested on 2 separate occasions. Responses to supplemental questions indicated that there was a need for consensus on the use of a separate equivocal/borderline interpretative category and the need for standardization of cutoff values used to define interpretative categories. Conclusions. —The College of American Pathologists proficiency survey programs provide useful information concerning the reproducibility of clinical testing for HER-2 by FISH and reflect clinical interpretation of HER-2 FISH analyses from laboratories across the country.

AbstractInfant leukemias are a rare group of neoplasms that are clinically and biologically distinct from their pediatric and adult counterparts. Unlike leukemia in older children where survival rates are generally favorable, infants with leukemia have a 5‐year event‐free survival rate of <50%. The majority of infant leukemias are characterized by KMT2A (MLL) rearrangements (~70 to 80% in acute lymphoblastic leukemia), which appear to be drivers of early leukemogenesis. In this report, we describe three cases: a 9‐month‐old female infant with B‐acute lymphoblastic leukemia (B‐ALL), an 8‐month‐old female presenting with B/myeloid mixed phenotype acute leukemia (MPAL), and a 16‐month‐old male with B‐ALL. The first case had a normal karyotype and B‐ALL FISH results consistent with an atypical KMT2A rearrangement. The second case had trisomy 10 as the sole chromosomal abnormality and a normal KMT2A FISH result. Case 3 had trisomy 8 and a t(11;15)(q23;q21), an atypical KMT2A rearrangement by FISH studies, and a focal deletion of 15q with a breakpoint within the USP8 gene by chromosomal microarray. Mate pair sequencing was performed on all three cases and identified a KMT2A‐USP2 rearrangement (cases 1 and 2) or a KMT2A‐USP8 rearrangement (case 3). These recently characterized KMT2A fusions have been described exclusively in infant and pediatric leukemia cases where the incidence varies vary according to leukemia subtype, are considered high‐risk, with a high incidence of central nervous system involvement, poor response to initial prednisone treatment, and poor event free survival. Additionally, approximately half of cases are unable to be resolved using standard cytogenetic approaches and are likely under recognized. Therefore, targeted molecular approaches are suggested in genetically unresolved infant leukemia cases to characterize these prognostically relevant clones.