796 
A Randomized Trial of Early vs Delayed Fixed-dose Combination Zidovudine/Lamivudine/Abacavir in Patients Co-infected with HIV and Tuberculosis: Early Findings of the Tuberculosis and HIV Immune Reconstitution Syndrome Trial
Humphrey Shao*1,2, J Crump1,3, H Ramadhani1,2, L Uiso4, O Sendui-Nguyaine1,2, R Kiwera5, E Ndosi1, J Shao1,2, J Bartlett3, and N Thielman3
1Kilimanjaro Christian Med Ctr, Moshi, Tanzania; 2Kilimanjaro Christian Med Coll, Tumaini Univ, Moshi, Tanzania; 3Duke Univ Med Ctr, Durham, NC, US; 4Kibong’oto Natl Tuberculosis Hosp, Sanya Juu, Tanzania; and 5Kikundi cha Wanawake Kilimanjaro cha Kupambana na Ukimwi, Moshi, Tanzania
Background: Because
of its low pill burden and lack of interaction with rifampicin,
fixed-dose combination zidovudine/lamivudine/abacavir
(ZDV/LMV/ABC) has been proposed as an alternative to efavirenz
(EFV)-containing regimens among patients co-infected with HIV and tuberculosis
(TB). We describe the initial tolerability and early clinical outcomes among 70
such patients receiving ZDV/LMV/ABC in the context of a clinical trial in the
Kilimanjaro Region of Tanzania.
Methods: Treatment-naïve
patients with smear-positive pulmonary tuberculosis, documented HIV infection,
and total lymphocyte counts ≤1200/mm3 were randomized to receive
early (2 weeks) vs delayed (8 weeks) initiation of
ZDV/LMV/ABC relative to the start of anti-TB therapy. Subjects received directly
observed ART and anti-TB therapy as inpatients for the initial 8 weeks of ART
and were assessed monthly through home-based or clinic visits thereafter. Laboratory
toxicity monitoring was performed at 4, 8, and 12 weeks, and every 12 weeks
thereafter, with CD4+ lymphocyte measurements every 12 weeks. Subjects
were assessed for evidence of immune reconstitution syndrome (IRS) weekly for
the first 8 weeks of antiretroviral therapy and monthly thereafter.
Results: Between
June 2004 and August 2005, 70 patients (59% male; median age 36 years) entered
the study, and here we report aggregate data from 516 patient-months of study
follow-up. At entry, the median CD4 count was 103/mm3 (IQR 57 to 155,
n = 70). For those reaching 12, 24, 36,
and 48 weeks on study, the median CD4 counts were 129/mm3 (IQR 88 to
179, n =51), 175/mm3 (IQR
127 to 208, n = 38), 201/mm3
(IQR 179 to 252, n = 25), and 276/mm3
(IQR 230 to 347, n = 14), p <0.0001 for temporal trends. The death
of 3 study subjects was not thought to be related to study medications
(disseminated Kaposi’s sarcoma; respiratory failure in a patient with suspected
pneumoconiosis; and cerebral malaria). ZDV/LMV/ABC was discontinued in 6
subjects; 2 with dose-limiting anemia, requiring substitution of stavudine for ZDV, and 4 with suspected ABC
hypersensitivity, requiring substitution of tenofovir
for ABC. To date, no cases of TB-associated IRS have been observed.
Conclusions: ZDV/LMV/ABC
was tolerated in >90% of subjects, and significant increases in CD4
lymphocyte counts were observed over 48 weeks. Clinically apparent
TB-associated immune reconstitution phenomena have yet to be observed in this
cohort, despite 516 patient-months of follow-up.
|
