[1547] Enumeration of CD138 Positive Plasma Cells in Bone Marrow Biopsy Specimens by Digital Image Analysis

Pingchuan Zhang, William Koss, Arundhati Rao. Scott and White Memorial Hospital, Temple, TX; Texas A&M University Health Science Center, Temple, TX

Background: Plasma cell percentage evaluation provides the diagnostic cut-off that separates monoclonal gammopathy of undetermined significance (MGUS) from smoldering multiple myeloma in the 2008 WHO diagnostic guidelines. However, the manual count of plasma cells in aspirate smears and estimation from CD138 stained biopsies are subjective and semiquantitative. Our study is intended to demonstrate that digital image analysis (DIA) is more objective and reliable, and can provide more reproducible information than traditional bone marrow aspirate differential count method.
Design: Retrospective file review of all bone marrow cases in 2007 was performed. Patients with immunohistochemistry workup for plasma cell dyscrasia were identified. CD138 Immunostained bone marrow biopsy slides were retrieved from pathology files. The slides are scanned manually at 200X, and multiple images are taken by using the DIA (Applied Imaging Spectrum, Israel). The number of image-taken fields is based on the size of the biopsy specimen. In each field analyzed by DIA, the percentage of positive cells, number of fields and total number of counted cells are recorded. The CD138 positivity is compared with reported plasma cell count of marrow aspirates. Statistical analyses are performed using Pearson correlation, linear regression analysis, and paired student t-test.
Results: Of 536 bone marrow cases, 42 patients have CD138 Immunostained slides for plasma cell dyscrasia work-up. For individual cases, image-taken fields range from 4-28 (average 11) with 4,038-69,195 (average 19,706) cells counted. The plasma cell positivity is 1.2-76.4% (17.1±19.7%) by DIA and 4.0-78.0% (29.0±21.9%) by counting aspirates. Percentage of CD138 positive cells by DIA strongly correlated with the reported numbers (r=0.76, p=0.0001). However reported plasma cell counts of aspirates are slightly higher than DIA by 4.4±2.3%, but not statistically significant (p=0.08). Plasma cell positivity for the same case by using DIA can be as wide as 6.0-94.8%.
Conclusions: DIA provides more reproducible and objective information in enumeration of CD138 positive plasma cells in evaluation of bone marrow biopsy specimens, and may be instrumental in assisting classification of MGUS and smoldering myeloma. Given the wide range of CD138 positivity in certain cases, practical algorithm and better methodology in reporting plasma cell number are needed. In addition, further correlation with clinical outcome will reveal the best method in counting plasma cells.
Category: Hematopathology

Wednesday, March 6, 2013 1:00 PM

Poster Session VI # 228, Wednesday Afternoon


Close Window