Cytochemical and/or Immunohistochemical Staining for Myeloperoxidase Are Essential in the Work-Up of Mixed Phenotype Acute Leukemia.
Jacqueline Emmons, Jason S Willis, Nitin J Karandikar, Franklin S Fuda. University of Texas Southwestern Medical School, Dallas
Background: The 2008 WHO Classification for hematolymphoid tumors designates myeloperoxidase (MPO) expression as the lone marker for establishing myeloid differentiation in mixed phenotype acute leukemia (MPAL) in the absence of monocytic differentiation. While 3% positivity by cytochemistry is the widely accepted threshold for determining MPO expression in leukemic blasts, defined cutoffs for flow cytometry (FCM) do not exist. In our study, we evaluated MPO expression by FCM, cytochemistry (CC) and/or immunohistochemistry (IHC) in leukemic blasts in an effort to identify a reliable flow cytometric cutoff.
Design: We queried our database for cases of newly diagnosed AML and MPAL as defined by either the 2001 or 2008 WHO criteria. Morphologic diagnoses were collected along with CC and IHC studies when available. FCM %MPO expression was determined using isotypic controls.
Results: FCM MPO expression in the 67 AMLs ranged from 0.00% to 99.99%. Of these, 42 cases had MPO CC studies available. Of the 10 cases that showed FCM MPO at 0-20%, 8 (80%) were negative by CC. Of the 8 cases with FCM MPO 20-60%, only 4 (50%) were MPO+ by CC. Cases with FCM MPO >60% were consistently MPO+ by CC. While only 15 of our 39 MPAL cases had concurrent CC or IHC studies, we observed a similar trend where the majority of leukemias with FCM MPO between 20-60% were MPO-neg by either CC or IHC.
Conclusions: Our results indicate that an FCM cutoff of 20% (as used conventionally for most markers) is not a reliable predictor of MPO expression, which may need cutoffs as high as 60%. Both CC and IHC showed similar discordance, suggesting that this discrepancy was not due to non-functional protein. Considering the importance of MPO expression in MPAL diagnosis, it may be essential for laboratories to establish their own FCM cutoffs measured against another technique or to include CC and/or IHC studies in the evaluation of these cases. Finally, the issue of sole reliance on MPO for diagnosing MPAL may need to be revisited through prospective studies.
Monday, February 28, 2011 1:00 PM
Poster Session II # 189, Monday Afternoon