The Lymphoma Classification Project

Shortly after its publication, an international study was convened to determine whether the REAL classification could be readily applied by a group of six expert hematopathologists who, with one exception, were not associated with the original proposal.22 The aims of the project were to judge whether the classification could be used in practice, to test its interobserver repro-ducibility, to assess the need for immunophenotyping in making a diagnosis (one of the criticisms following its publication having been that the REAL classification was not cost effective in this respect!), to determine whether the constituent diseases were clinically distinctive either at presentation or in terms of clinical outcome, and to determine the relevant frequency of these diseases in the study populations. The participating pathologists, assisted by clinicians and statisticians, studied 1400 cases of lymphoma comprising 80-210 cases in each of eight centers in North America, Europe, Asia, and Africa.

The participants found that the REAL classification was highly practical, allowing the ready classification of 95% of cases. Interobserver reproducibility was greater than 85% for most entities, which was a substantial improvement over previous studies using other classifications where reproducibility was frequently in the region of only 60% or less. Immunophenotyping was not necessary for the classification of certain diseases including follicular lymphoma and small lym-phocytic lymphoma/chronic lymphocytic leukemia but essential for the classification of T-cell lymphomas and particularly helpful for some B-cell disorders including mantle cell lymphoma and diffuse large B-cell lymphoma (DLBCL).

In undertaking clinicopathological correlation, the Lymphoma Classification Project showed that the different diseases recognized by the REAL classification did indeed differ in terms of clinical presentation and survival, supporting the contention that they were distinct biological entities. However, an important finding of the study was that classification is not the only predictor of clinical outcome of any individual case. In this respect, the power of the IPI was confirmed. For example, patients with follicular lymphoma and an IPI score of 1-3 have a median survival of 7-10 years, while the median survival for the minority with an IPI of 4 or 5 is significantly reduced to 1.5 years.23

0 0

Post a comment