Phenotypic overlaps between pleomorphic malignant T‐cell lymphomas and mixed‐cellularity Hodgkin's disease

Abstract

Histologically diagnosed, or in part questionable, malignant pleomorphic peripheral T-cell lymphomas (pPTCLs, n = 16) and mixed-cellularity Hodgkin's disease (MCHD, n = 12) were objectively compared by the use of combined immunohistochemistry on paraffin sections, test-point analysis of tissue components, and semi-automated nuclear morphometry on semi-thin resin sections. Classical, qualitative histomorphological distinction of these sub-types of lymphomas proved to be valid and is probably still the best method. Quantitative discriminant features, in order of decreasing significance, were: (i) expression by large atypical cells (LACs) of CD45R0, CD43 and CD45 in pPTCLs, and of CD30 and CD15 in MCHD; (ii) means and standard deviations (SDs) of LAC nuclear-profile areas (greater in MCHD than in pPTCLs); (iii) expression of CD3 by LACs in pPTCLs; (iv) prominence of small lymphoid cells in MCHD; (v) higher percentage of medium-sized lymphoid cells in pPTCLs; and (vi) higher SDs of nuclear-profile circularity factor of small lymphoid cells in MCHD. The medians of the largest nucleolar profile areas in LACs per field did not differ in pPTCLs and MCHD, but dispersion of individual values towards higher levels was significantly greater in the latter. Stepwise discriminant analysis of test point and nucleometric variables that best distinguished pPTCLs from MCHD revealed considerable overlaps, and questionable cases tended to be intermediate between the two. In conclusion, our results confirm and expand the notion of intra-group heterogeneity, with indistinct borders and the existence of intermediate phenotypes between these two taxonomic categories of malignant lymphomas.

Growth vs. DNA strand breaks in Hodgkin's disease: Impaired proliferative ability of Hodgkin and Reed‐Sternberg cells

“…We conclude from these findings that the identification of presumed mitotic or damaged H-RS cells in sections not stained for CD30 can be accepted with greater confidence than anticipated. The percentage of distinct non-neoplastic cell types was close to the values reported previously (Tosi et al, 1992; details not shown).…”

“…We carried out differential counts in parallel and expressed the proportion of distinct cell types (Tosi et al, 1992) as percentages of all reactive cells. In sections immunostained for CD30 only, all CD30+ cells in 60 randomly chosen HPF (approximately 25,000 cells of all types per case) were registered and the percentages and MI (in %) per case of CD30+ large cells and CD30-small lymphoid cells were established.…”