Lymphoma discrimination by computerized triple matrix analysis of list mode data from three-color flow cytometric immunophenotypes of bone marrow aspirates

Background: The goal of this study was to evaluate a self-learning algorith m for the computer classification of information extracted from flow cytome tric immunophenotype list mode files from high-grade non-Hodgkin's lymphoma (NHL), Hodgkin's disease CHD), and multiple myeloma (MM). Materials and Methods: Bone marrow aspirates (BMA) were obtained from untre ated NHL (n = 51), HD (n = 9), or MM (n = 13) patients. Bone marrow aspirat es were not infiltrated in NHL and HD patients as confirmed by thorough his tologic and cytologic investigation; however, MM patients showed an infiltr ation rate >50% by malignant myeloma cells. Peripheral blood leukocyte (PBL ) samples were taken from age-matched healthy volunteers (n = 44) as easily available control material. A second control group of 15 healthy volunteer s, from whom BMA and PBL samples were available, allowed us to differentiat e whether the observed classification results on malignant samples were due to the malignant process or simply to the inherent differences between BMA and PBL. Bone marrow aspirates and PBL were analyzed by the same immunophe notyping antibody panel (CD45/14/20, CD4/8/3, kappa/CD19/5, lambda/CD19/5). The acquired list mode data files were analyzed and classified by the self -learning triple matrix classification algorithms CLASSIF1 following a prio ri separation of the data into a learning set and unknown test set. After c ompletion of the learning phase, known patient: samples were reclassified a nd unknown samples prospectively classified by the algorithm. Results: Highly discriminatory information was extracted for the various ly mphoma entities. The most discriminating information was encountered in ant ibody binding, antibody binding ratios, and relative antibody surface densi ty parameters of leukocytes rather than in percentage frequencies of discre te leukocyte subpopulations. Samples from healthy controls were classified as normal in 97.2% of the cases, whereas those of NHL, HD, and MM patients were on average correctly classified in 80.8% of the cases. Conclusions: Although no detectable lymphoma cells were present in BMA of N HL and HD patients, the CLASSIF1 classification of the immunophenotypes of morphologically normal cells provided a surprisingly good disease discrimin ation equal or better than that obtained by examining pathological lymph no des according to the respective literature. The results are suggestive for a lymphoma-related and disease-specific antigen expression shift on normal hematopoietic bone marrow cells that can be used to discriminate the underl ying disease (specificity of unspecific changes), i.e., in this case NHL fr om HD. Multiple myeloma patients were discriminated by changes on malignant as well as on normal bone marrow cells. (C) 2000 Wiley-Liss, Inc.