To the Editor:
Hepatitis C virus (HCV) has been shown to be the main etiologic agent of type II mixed cryoglobulinemia (MC),1,2 a disease associated with an underlying B-cell clonal proliferation, predisposing to B-cell malignancy, in 2% to 11% of the cases.3 Italian studies have recently reported a 9% to 35% prevalence of chronic HCV infection in patients with B-cell non-Hodgkin’s lymphomas (B-NHL), compared with a 2% to 13% prevalence in patients with other blood malignancies, and a 1% to 5.4% prevalence in healthy controls.4-9 In addition, a recent American controlled study10 showed a 22% prevalence of HCV infection markers in patients with B-NHL versus 5% in age-matched controls. However, studies from other countries, including the United Kingdom and Germany,11-14 did not confirm such an association. The aim of the present study was to determine the prevalence of chronic HCV infection in a large cohort of unselected patients with B-NHL in a French center.
Between January 1994 and July 1997, 201 patients with B-NHL (109 men, 92 women; mean age, 56.0 years; range, 18 to 90) were seen at the hematology clinic of our hospital. During the same period, 94 patients with Hodgkin’s disease were seen (54 men, 40 women; mean age, 37.5 years; range, 18 to 75) and served as a control group for this study. Lymphoma was diagnosed on the basis of morphologic evaluation of lymph node tissue or extranodal tissues, including bone marrow specimens. Immunophenotypic analysis of B- and T-lymphocytic markers was performed and B-NHL was histologically classified using the Revised European American Lymphoma (R.E.A.L.) classification.15Anti-HCV antibodies were prospectively sought in all patients using second-generation enzyme-linked immunosorbent assay (ELISA), confirmed by second-generation recombinant immunoblot assay (RIBA) (Ortho Clinical Diagnostics, Raritan, NJ).
The overall prevalence of HCV markers in the B-NHL patients was 4/201 (2.0%; 95% confidence interval: 0.1% to 3.9%). Table1 shows the prevalence of HCV markers according to B-NHL histotypes. The 4 HCV+ patients were HCV-RNA+ by polymerase chain reaction (PCR). HCV+ patients had a lymphoplasmatocytoid lymphoma in 1 case, a mucosa-associated lymphoid tissue (MALT) lymphoma in 1 case, and a diffuse large B-cell lymphoma in 2 cases. The prevalence of HCV infection in the patients with Hodgkin’s disease (control group) was 1/94 (1.1%; 95% confidence interval: 0.0% to 3.2%). It was not significantly different from the prevalence in patients with B-NHL.
Overall, the present results do not support the existence of a significant relationship between HCV infection and B-NHL in France. The marked discrepancies among studies from different geographic areas might be explained by the following: (1) the studies4-10 in favor of a relationship between HCV infection and B-NHL originated from areas with high endemicity for HCV. However, because of the higher HCV prevalence in older subjects in these areas, age-matched control groups should have been used systematically for comparison.5,8 (2) in contrast, the studies suggesting a lack of association11-14 originated from countries with a lower prevalence of HCV infection. Thus, HCV appears to be associated with B-NHL only in areas where HCV is highly prevalent, suggesting that HCV is not essentially involved in the pathogenesis of B-NHL. However, some specific B-NHL histotypes could be preferentially associated with HCV infection. It is noteworthy that, in three studies,6-8 HCV was principally found in patients with lymphoplasmacytoid lymphoma/immunocytoma associated with type II MC, with 30% to 45% of these patients being HCV-RNA+.6-8 The lymphoplasmatocytoid lymphoma/immunocytoma originates from a CD5-peripheral B lymphocyte able to differentiate into a plasma cell. Sites involved include bone marrow, lymph nodes, spleen and, less frequently, peripheral blood or extranodal sites. This condition is commonly associated with type II MC.16 In the United States study,10 monocytoid nodal B-cell lymphoma accounted for 23% of all lymphomas and 67% of all low-grade lymphomas among the HCV+ patients. Furthermore, another Italian report17 pointed out an additional subset of overt B-NHL in HCV-infected individuals, diffuse large B-cell lymphoma (23 of 83, 27.7% of cases were HCV+).
The overrepresentation of lymphoplasmatocytoid lymphoma in several Italian studies (14% to 33%)6,7 might result from a selection bias of the patients, due to the follow-up of a significant number of patients with MC in these centers. This could account for the high prevalence of HCV infection in B-NHL patients in these series. The lower prevalence of lymphoplasmatocytoid lymphoma in our series of unselected patients with B-NHL (4.5%) is in agreement with the 1.2% prevalence, as estimated by R.E.A.L.,15 and could explain the low prevalence of HCV markers observed. Furthermore, the low prevalence of HCV infection (2 of 118, 1.7%) in the large cohort of our patients with diffuse large B-cell lymphoma does not confirm the recently suggested association between HCV and diffuse large B-cell lymphoma.17 Prospective case-control studies of the prevalence of HCV in B-NHL patients from different countries, including all R.E.A.L. histological subtypes in sufficient numbers, are now mandated to confirm such an association.
