AMC075: The Combination of Vorinostat with Chemotherapy and Rituximab Is Tolerable and Feasible in HIV-Related B-Cell Non-Hodgkin’s Lymphoma with High-Risk Features

Background: Individuals infected with human immunodeficiency virus (HIV) are at an increased risk of developing highly aggressive non-Hodgkin’s lymphoma (NHL). Superior outcomes have been recently reported for rituximab (R) plus infusional etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin (R-EPOCH) over R-CHOP chemotherapy in patients with HIV-associated NHL, including those with high-risk features such as high International Prognostic Index (IPI) score, non-germinal center (GC) type diffuse large B-cell lymphoma (DLBCL), and high Ki-67 expression. The proportion of NHLs harboring γ-herpesviruses, such Epstein Barr virus (EBV) and human herpesvirus 8 (HHV-8), is higher in patients with HIV-AIDS, warranting the investigation of chemotherapy in combination with new drugs that target the biology of HIV-NHL.

At the AIDS Malignancy Consortium (AMC), we developed AMC075, a CTEP-sponsored phase I/randomized phase II trial using the histone deacetylase (HDAC) inhibitor (HDI) vorinostat (V) in combination with R-DA-EPOCH for the treatment of high-risk HIV-NHL. Through acetylation HDIs alter chromatin and protein function, facilitating the expression of silenced cellular genes and latent viruses in neoplasms. We hypothesized that V would enhance anti-tumor chemotherapy effects in HIV-NHL, and transiently activate latent HIV and γ-herpesviruses, leading to the elimination of virally infected reservoirs under antiviral treatment.

Methods:Eligibility: HIV+ subjects with untreated NHL (non-Burkitt) with CD4 count ≥50 cell/mm³. Subjects with high-risk disease (age adjusted-IPI 2-3, or Ki-67 ≥ 80%, or DLBCL-ABC subtype, or other aggressive NHL) were eligible. Phase I study: 16 subjects were enrolled between Dec-2010 and Nov-2012. Patient characteristics were as follows: Median age 48.5 years (range 27-65), 82 % males, 63 % non-Hispanic white, 19 % black, and 13% Hispanic or Latino. Baseline absolute CD4 counts were <100 and >100 cells/mm³ in 13% and 88% respectively. The median absolute CD4 count was 263 cells/mm³(range 91-754). Eight subjects had undetectable viral loads at baseline. Three were withdrawn from study after enrolment due to eligibility or compliance issues. Twelve were treated with V-R-dose-adjusted (DA)-EPOCH. The starting dose of V was 300 mg on Days 1-5 on each of six 21-day cycles and escalated to a maximum of 400 mg on a 3 + 3 design. One subject was treated with V-R-CHOP under low-risk early stage arm, which was closed due lack of accruals.

Results:Phase I study: One subject completed treatment with V-R-CHOP, under low-risk arm. Twelve high-risk subjects completed treatment with V-R-DA-EPOCH. The recommended phase II dose (RPTD) was determined at 300 mg based on dose-limiting toxicities (mainly grade 4 thrombocytopenia) observed at 400 mg level. Toxicities. Hematologic and non-hematologic toxicities after cycle 1 are summarized in table below. Pharmacokinetics. In a preliminary analysis, V did not alter doxorubicin (0.018±0.023 µg/mL), etoposide (1.60±0.49 µg/mL) or vincristine (2.86±3.10 ng/mL) steady-state concentrations across dose levels. Efficacy. Twelve of 13 subjects are still alive after completion of treatment and are on active follow-up. One subject who had central nervous system (CNS) involvement at the time of enrollment relapsed after achieving a CR and died of CNS disease progression.

Conclusions: We are conducting a national trial for HIV-NHL with high-risk features combining standard chemotherapy with vorinostat designed to target the biology of these tumors. Preliminary results from the phase I portion of the study show the combination treatment is tolerable with acceptable toxicities and highly efficacious. A randomized phase II study (90 subjects) of R-DA-EPOCH with and without V is currently enrolling subjects. The primary objectives are to determine the overall toxicity and efficacy of R-DA-EPOCH with and without V in subjects with high-risk HIV-NHL. Correlative studies were designed to assess the effect of R-DA-EPOCH +/- V on latent HIV-infected reservoirs, EBV and HHV-8. Complete phase I study results will be presented at the meeting.