AbstractAbstract 4156 Introduction:Human Immunodeficiency Virus (HIV) and hepatitis C virus (HCV) infections are both associated with an increased risk of B-cell non-Hodgkin lymphoma (B-NHL). They are known to be preferentially diffuse large B-cell lymphoma (DLBCL), and Burkitt lymphoma subtypes in HIV infection, while marginal zone lymphoma (MZL) and DLBCL represent the main subtypes in HCV infected patients. The absence of increased risk of B-NHL in HIV-HCV co-infected patients compared to HIV infected patients was reported in epidemiological studies, but the clinico-pathological characteristics of B-NHL in the setting of such co-infection particularly in the era of HAART remains unclear. Patients and methods:We present the data of adult patients with B-NHL and HIV-HCV co-infection from the French ANRS CO16 Lymphovir cohort. This multicentric prospective cohort includes HIV infected patients with lymphoma. Each patient is followed every 6 months during 5 years. At each follow-up, a blood sample is withdrawn allowing ancillary studies. Data collection concerns clinical presentation, treatment and evolution of B-NHL and HIV and HCV infections. Pathological and cytological materials are centralized in order to allow their review and a concerted analysis by a group of expert hematopathologists (MR, SP, DC), haematologists and immunologists. Results:Among the 49 patients with B-NHL included in the cohort, 6 were HIV-HCV coinfected, [5 men and 1 woman, median age 47 years (range 36–67)]. They were included between october 2007 and august 2009 in 6 French centres. Transmission risk was intravenous drug abuse (n=4) and heterosexual transmission (n=2). Median duration from HIV diagnosis to B-NHL was 11 years (range 1–20), and the median nadir CD4 T-cell count was 194/mm3 (range 151–746) (nadir not available for 2 patients). Only one patient had developed AIDS before the diagnosis of NHL. HCV genotypic distribution was: genotype 1 (n=3), 2 (n=1), 4 (n=1), and not available (n=1). Median HCV viral load was 6.0 log (range 5.4–7.1). One patient had compensated cirrhosis and none of the patients were tested for cryoglobulinemia. Immunological assays showed the presence of monoclonal IgM Kappa in 1 patient and no positive rheumatoid factor.At the diagnosis of B-NHL, 4 patients received HAART and had an undetectable HIV load. In contrast, none of the patients were treated for HCV infection. The median interval between B-NHL diagnosis and last follow-up is 17 months (range 0.1–32 months). The histological subtype distribution is 4 MZL -including 2 extranodal MZL of MALT, 1 splenic lymphoma with villous lymphocytes (SLVL) and 1 lymphoplasmacytic lymphoma- and 2 EBV-negative DLBCL. All patients have extranodal involvement, including digestive tract (n=3), liver (n=2), bone marrow (n=2) and spleen (n=1). Median CD4 T-cell count at B-NHL diagnosis was 582/mm3 (range 235–1322) in MZL patients and 274/mm3 (200 and 347) in DLBCL patients.Following diagnosis, all patients were treated with HAART, associated with peg-interferon plus ribavirin in 1 patient. The 2 patients with DLBCL received R-CHOP, associated with methotrexate in 1 case. One patient with a partial response exhibited a neuromeningeal relapse and was treated with COPADM and CYVE regimen, allowing a complete and sustained response. The other patient died of serious infection after 2 courses of chemotherapy. Among patients with MZL, one patient received R-CVP and one received chlorambucil, allowing a complete response in both patients. The patient with SLVL received HAART then peg-interferon plus ribavirin, allowing normalization of lymphocytes count (5.89 at baseline vs. 1.69 × 109/L at 12 weeks of anti-HCV therapy) with a correlation between HCV virological and hematological response. Patient with lymphoplasmacytic lymphoma died of cardiac ischemia before the initiation of chemotherapy. Conclusions:This study describes the resurgence of MZL in HIV-HCV co-infected patients with restored immunity under HAART. Such B-NHL, very rarely described in HIV patients, occurs mainly in patients with control of HIV without history of AIDS but with active HCV infection. Histological subtypes and localizations resemble more to that observed in HCV infected patients than in HIV infected patients. These findings suggest the role of chronic antigenic stimulation by HCV and/or HIV and question about the interest of HCV antiviral therapy on NHL in co-infected patients. Disclosures:No relevant conflicts of interest to declare.