At the same time as the immunophenotype was being defined, two clinical staging systems for CLL were being developed. In New York, Kanti Rai defined five groups and gave them Roman numerals (Rai et al, 1975), while in Paris Jacques-Louis Binet designated three groups alphabetically (Binet et al, 1977). In reality, both systems were saying the same thing, namely that the more disease you had the worse the prognosis and if the bone marrow started to fail then the outlook was dire. It was no surprise to see the International Workshop on CLL (IWCLL) recommend that the two systems be amalgamated (IWCLL, 1981, 1989), although in reality the Americans continue to use Rai and the Europeans use Binet. A number of other prognostic indicators have since been recognized, including lymphocyte doubling time (Montserrat et al, 1986), bone marrow histology (Rozman et al, 1984) and chromosomes (Juliusson et al, 1990).
Most of the large series of patients with CLL published before 1990 were contaminated with diseases that we would now recognize as not being CLL. Lymphosarcoma cell leukaemia was recognized early as something rather different (Isaacs, 1937), but criteria for its diagnosis varied between different institutions (Zacharski & Linman, 1960; Schwartz et al, 1965). Aisenberg &Wilkes (1976) recognized a type of spillover lymphosarcoma with sparse surface immunoglobulin, similar to CLL (well-differentiated lymphocytic lymphoma), that, by the revised European American classification of lymphoid neoplasms (REAL), we would probably recognize as the same disease as CLL. Those tumours with denser surface immunoglobulin were clearly different, but lymphoma classification was so uncertain that we can only speculate on how different. Many will have been follicular lymphomas (Spiro et al, 1975) or what we may perhaps refer to in the future as t(14;18) disease (Cleary et al, 1986).
Prolymphocytic leukaemia (PLL) was recognized as a separate entity by Galton et al (1974). Because prolymphocytes may accumulate in CLL, there has been confusion, but it should now be clear that CLL does not transform into PLL.
Hairy cell leukaemia was recognized in 1958, albeit under a different name (Bouroncle et al, 1958). It is difficult to believe it could be confused with CLL, although perhaps more excuse could be made for confusing the hairy cell variant (Cawley et al, 1980).
Splenic lymphoma with villous lymphocytes (SLVL) was first recognized in 1979 (Neiman et al, 1979), although again under a different alias. Since the REAL classification gained popularity, this is being recognized as a form of splenic marginal zone lymphoma. Other cases of this disease may not have very obvious villi and may constitute what has become known as CD5-negative CLL (Salomon-Nguyeu et al, 1995).
Because it is CD5 positive, mantle cell lymphoma is the last of the lymphosarcoma cell leukaemias to be separated from CLL. It has emerged as a distinct entity via many different name changes, although its existence in a leukaemic phase was only lately recognized (De Oliveira et al, 1989). It has dense surface immunoglobulin and lacks CD23, but most distinctive is the t(11;14) translocation (Raffeld & Jaffe, 1991). It is a matter of faith in our laboratory that no case of CLL has this translocation.
T-cell CLL was first described by Brouet et al (1975). Several subtypes have now been recognized and the term is no longer used. Although T-PLL had been recognized in a number of prior publications, including some by the Royal Marsden team, the defining description was by Matutes et al (1986). It is usually a very malignant disease with a characteristic karyotype. Despite its distinctive cellular morphology, it is still confused with CLL (Hoyer et al, 1995).
McKenna et al (1977) first described large granular lymphocytic leukaemia. This condition usually has CD8-positive lymphocytes. It exists in CD3-positive and CD3-negative forms (Loughran, 1993). Non-clonal proliferations are also seen. There remain the CD4-positive leukaemias that are either the disease associated with HTLV-1 or part of the mycosis fungoides/Sezary syndrome complex (Matutes et al, 1988). The term T-cell CLL should no longer be used and therefore it is no longer necessary to call CLL `B-cell CLL.