In the early 1990’s we described a series of patients from Kumasi, Ghana, West Africa who appeared to have a distinctive lymphoproliferative disorder (LPD) characterised by massive splenomegaly and lymphocytosis with characteristic morphology. At the time this was termed tropical splenic lymphoma (TSL). These patients, in contrast to most patients with LPD’s, tended to be female and relatively young and were frequently diagnosed as having hyper-reactive malarial splenomegaly. An aetiological role for chronic malaria infection has been postulated but remains as yet unproven. Only limited phenotypic and genotypic characterisation of such patients has been possible previously. Given the clinical features and potential aetiological role for malaria infection we wanted to determine whether TSL represented a distinct entity or may be more appropriately considered as splenic marginal zone lymphoma (SMZL). We have therefore evaluated the clinical, immunophentypic and genotypic features of 19 pts (median age 63 years, range 40–78) with a clinical diagnosis of TSL. Peripheral blood, bone marrow aspirate and trephine biopsies were obtained in all patients and the laboratory assessment were performed in the HMDS laboratory, Leeds, UK as part of an on-going collaborative project investigating lymphomas in Ghana. 14 of the 19 patients (73%) were female and all had significant splenomegaly (median length below the costophrenic margin of 17cm, range 6–30 cm). 5/19 patients (26%) had lymphadenopathy and 17/19 patients (89%) had a lymphocytosis - median 30.1×09/l (range 5.4×09/l - unrecordable). Bone marrow infiltration was evident in the trephine biopsies of all patients and was extensive in the majority. Immunophenotyping was performed primarily by immunohistochemistry on the trephine sections although flow cytometry was possible in a proportion of patients. The majority of cases were characterised by a CD5− CD10− CD20+ CD23− CD79+ BCL2+ BCL6− MUM1/IRF4- cyclin D1- immunophenotype. The rate of cell proliferation was low in all cases. FISH studies were performed and these demonstrated del7q31 in 4/18 cases and del6q21 in 1/19 cases. There was no evidence of the t(11;14), t(9;14) or MALT1 rearrangements. IgH sequence analysis was also performed in 16 cases and this demonstrated that 9/16 cases (56%) were germline (>98% homology) and 7/16 (44%) were mutated (<98% homology). Within the mutated group the overall mutation load appeared to be relatively low - median 3.3% (range 3–6%) with most utilising the VH3 family genes. This is the first detailed clinicopathological assessment of patients with a clinical diagnosis of TSL. These patients clearly have some pathological features seen in patients with SMZL such as a CD5− CD10− CD23− immunophenotype, 7q31 deletions and cases with both germline and mutated Ig genes. Definitive phenotypic and genotypic features are unfortunately lacking in SMZL and it remains uncertain whether patients with a clinical diagnosis of TSL should be considered as having SMZL.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author