Pathological and immunophenotypic features of adult non-Hodgkin's lymphomas by age group

To elucidate age-related differences in non-Hodgkin's lymphoma (NHL), the authors evaluated 950 consecutive, human immunodeficiency virus-negative patients (age range, 15 to 96 years) observed between July 1988 and June 1995 in the same Italian cancer institute. Patients were grouped into six age groups and cross-tabulated by Working Formulation (WF) categories and other newly recognized entities according to the Revised European American Lymphoma (REAL) classification, cell immunophenotype, and nodal or extranodal location. There was a tendency of the low-grade category to increase with increasing age (16.8% in the age group 15 to 34 years to 32.4% in the age group 65 to 74 years), although a subsequent decline was seen at age 75 years or older (23.2%). Also the intermediate-grade category was more frequent in the elderly (46.6% and 49.4% at 65 to 74 years and at 75 years or older, respectively). High-grade category showed compared with low and intermediate grade ones, a significant downward trend with age (X2 for trend = 25.31; P < .001), interrupted in only the oldest age group. The relative excess of low-grade NHL in patients older than 55 years. of age was accounted for by the high proportion of small lymphocytic lymphomas, which, however, somewhat declined at age 75 years or older. Conversely, the relative excess of high-grade NHL below age 35 years chiefly derived from the high percentage (28.4%) of CD30-positive anaplastic large cell lymphomas. B- and T-cell lymphomas accounted for 85.9% and 9.0% of all cases, respectively. B- and T- and non-B, non-T-cell and histiocytic NHL accounted for the remaining 5.1%. A highly significant trend of increase in the proportion of B-cell lymphomas with age increase was noted (X2 for trend = 21.90; P < .001); chiefly attributable to the excess of T-cell (15.1%) and undetermined phenotype (18.6%) in patients younger than 35 years of age. Extranodal location was not significantly related to age groups. Thus, the present study showed some interesting differences in NHL morphology and cell phenotype according to age, avoiding, at the same time, the arbitrariness of patients' dichotomization into elderly and nonelderly.