CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

Methods: We enrolled all consenting adults treated for lymphoma between 10/2010 and 9/2014 at two referral hospitals in Gaborone, Botswana. Standard treatment included 6 cycles of CHOP (recently with rituximab) for NHL and 6 cycles of ABVD for HL. ART-naïve HIV-infected patients or those who started ART within 6 months of lymphoma diagnosis were categorized as off ART. Exact methods were used for categorical measures and survival was assessed using Cox modeling adjusted for lymphoma type (NHL or HD). Results: Seventy-two patients were enrolled (56 NHL, 16 HL) and followed for a median 12.9 months. Fifty-two (72%) patients were HIV-infected and 20 (28%) were HIV- uninfected. The median age of HIV-infected and HIV-uninfected patients was 37 and 52 years with NHL (P=0.059) and 40 and 32 years with HL (P=0.38), respectively. Among HIV-infected, 42% of lymphomas were diagnosed in patients on ART (median duration 2.6 years), with median CD4 count of 255 cells/ m L (IQR 147-390 cells/ m L) for NHL and 401 cells/ m L (IQR 311-525 cells/ m L) for HL. In those patients off ART, 41%were eligible to have been initiated on ART by CD4 count alone ( ≤ 250 cells/ m L). Fifty percent of NHL and 36% of HL patients presented at advanced stages (Ann Arbor stage III or IV). Chemotherapy records were analyzed in 43 (60%) cases and all patients were started on standard treatment regimens. Treatment modifications were common and due to toxicity (22 instances) and chemotherapy stock out (5 instances). Sixteen NHL patients (28%) and 1 HL patient (6%) died during follow-up. The one-year survival was 74% (95% CI 58 – 84%) for NHL and 93% (95% CI 59-99%) for HL. While power was low, no significant differences were detected in stage at presentation (p=0.76), treatment modification (p=0.60), or survival (HR 0.76, 95% CI 0.26-2.2) by HIV status.

Poster Abstracts

Conclusions: Despite ART coverage exceeding 90% by government estimates, the majority of lymphomas were diagnosed in patients off ART with relatively high CD4 cell counts. Persistently high incidence may be related to low CD4 thresholds for ART eligibility. HIV-infection was not associated with reduced survival, although sample size was limited. 719 CHOP Is Feasible for HIV-Associated Lymphoma in the ART Era in Malawi Satish Gopal 2 ;Yuri Fedoriw 1 ; Agnes Moses 2 ; Nathan Montgomery 1 ;Wongani Kaimila 2 ; Coxcilly Kampani 2 ; Robert Krysiak 2 ; Kristy Richards 1 ;Thomas Shea 1 ; George Liomba 2 1 University of North Carolina, Chapel Hill, NC, US; 2 University of North Carolina Project–Malawi, Lilongwe, Malawi Background: Although it is the standard treatment throughout the region, there is no prospective study describing use of CHOP for lymphoma patients in sub-Saharan Africa in the antiretroviral therapy (ART) era. Methods: We describe a prospective longitudinal cohort of adult lymphoma patients receiving CHOP at a national teaching hospital in Malawi between June 2013 and August 2014. Chemotherapy and supportive care are given using standardized monitoring and dose adjustment, and HIV+ patients receive concurrent ART. Results: Thirty-five patients (34 non-Hodgkin, 1 Hodgkin; 19 HIV+, 16 HIV-) were treated with CHOP, with median age 48 years (range 22-77), 26 being (74%) male, and 20 (57%) with stage III/IV disease. Clinical characteristics were overall similar between HIV+ and HIV- patients. Seventeen HIV+ patients (90%) were on ART for a median 18.9 months (range 0.2-98.8) at lymphoma diagnosis. Median CD4 count among HIV+ patients was 138 cells/ μ L (range 32-1,013), and 10 (53%) had suppressed HIV RNA. HIV+ patients experienced greater leukopenia and neutropenia during chemotherapy, with similar hemoglobin and platelet count levels (Figure). However, median absolute neutrophil count (ANC) remained ≥ 1x10 3 / μ L during treatment even for HIV+ individuals. Among HIV+ patients completing chemotherapy, white blood cell (WBC) and ANC levels returned to values similar to HIV- patients by six months. Grade 3/4 neutropenia occurred in more HIV+ than HIV- patients (83% vs 31%, p=0.015). Of 32 grade 3/4 neutropenia events, 29 were grade 3 and three were grade 4. For HIV+ patients, median CD4 count increased to 180 cells/ μ L (range 44-369) at six months. Cumulative dose and dose intensity were similar for HIV+ and HIV- patients, as assessed by CHOP cycles per patient (median 5 HIV+ vs 4 HIV-, p=0.43), days between cycles (median 21 vs 21, p=0.15), cyclophosphamide dose (mg/m2) per cycle (median 723 vs 710, p=0.48), and doxorubicin dose (mg/m2) per cycle (median 49 vs 49, p=0.38). Eighteen patients (60%) received <6 CHOP cycles (53% HIV+ vs 67% HIV-, p=0.71), with reasons being death (n=11, 4 HIV+, 7 HIV-), toxicity (n=4, 3 HIV+, 1 HIV-), social barriers (n=2, 1 HIV+, 1 HIV-), and disease progression (n=1, HIV-).

442

CROI 2015