Several years ago, the laboratory of Robert Siliciano at Johns Hopkins University reported that the FDA-approved drug disulfiram (Antabuse) could reverse HIV latency in laboratory experiments. The discovery led Steve Deeks and colleagues to conduct a small pilot study in HIV-positive people on ART, which found that a standard dose of 500mg/day of disulfiram showed some hints of latency-reversing activity that appeared to be linked to whether drug levels were detectable in blood. In turn, the pilot study spurred the launch of a larger clinical trial investigating the effects of higher disulfiram doses (up to 2000mg/day) in a cohort of 30 HIV-positive individuals on ART. Results from this trial have now been published in The Lancet HIV.
The researchers report that disulfiram showed evidence of latency-reversing activity at all doses tested. Measurements of unspliced cell-associated HIV RNA—a standard approach for assessing the activity of candidate latency-reversing agents—revealed small but significant increases after dosing compared to baseline. However only the highest dose of 2000mg/day led to a detectable increase in plasma HIV RNA (as measured by an ultrasensitive viral load test with a lower limit of detection of 0.19 copies).
The magnitude of the effect of disulfiram was considerably smaller than has been reported with the HDAC inhibitor romidepsin, but the authors point out that disulfiram may be more suitable for long-term dosing due to its well-established safety record (it is used to induce sensitivity to alcohol as a means to treat chronic alcoholism). In contrast, the toxicity profile of HDAC inhibitors is less favorable to extended dosing, and there are concerns about the safety implications of the long-term changes in gene expression that have been observed in clinical trials. The ability to administer disulfiram for a prolonged period could be an advantage in studies involving combinations of approaches to depleting the HIV reservoir.
The paper also describes an interesting incidental finding from the trial. To establish baseline measures of HIV RNA for each participant, three samples were taken on different days and the results were averaged. The researchers were surprised to discover that the third measurement—taken earlier in the day than the prior two—showed significantly higher levels of unspliced cell-associated HIV RNA. The finding suggests circadian rhythm may influence HIV transcription, and this needs to be considered when attempting to evaluate the effects of latency-reversing agents (these data were also presented at the IAS 2015 conference).
A final point worth highlighting is that the initial discovery of disulfiram’s latency-reversing potential and a significant portion of the subsequent clinical evaluation was made possible by funding from amfAR’s HIV cure research program. Today on World AIDS Day, a further expansion of this program was announced with the launch of the amfAR Institute for HIV Cure Research at the University of California San Francisco (see amfAR’s press release for more information).
Lancet HIV 2015
Published Online November 16, 2015
Julian H Elliott, MBBS, James H McMahon, MBBS, Christina C Chang, MBBS, Sulggi A Lee, MD, Wendy Hartogensis, PhD, Namandje Bumpus, PhD, Rada Savic, PhD, Janine Roney, MPH, Rebecca Hoh, BA, Ajantha Solomon, BSc, Michael Piatak, PhD†, Robert J Gorelick, PhD, Jeff Lifson, PhD, Prof Peter Bacchetti, PhD, Prof Steven G Deeks, MD, Prof Sharon R Lewin, MBBS
†Died on Sept 19, 2014
In vitro, disulfiram activated HIV transcription in a primary T-cell model of HIV latency and in a pilot clinical study increased plasma HIV RNA in individuals with adequate drug exposure. We assessed the effect of disulfiram on HIV transcription in a dose-escalation study.
In this prospective dose-escalation study, to optimise disulfiram exposure we included adults with HIV on suppressive antiretroviral therapy, with plasma HIV RNA of less than 50 copies per mL and a CD4 cell count greater than 350 cells per μL. Participants were allocated sequentially to one of three dosing groups (500 mg, 1000 mg, and 2000 mg) and received disulfiram daily for 3 days. Only the staff who did laboratory assays were masked to group assignment. The primary endpoint was change in cell-associated unspliced HIV RNA in CD4 cells. The primary analysis method was a negative binomial regression, with the number of copies as the outcome variable and the input total RNA or plasma volume as an exposure variable, which is equivalent to modelling copies or input. We used these models to estimate changes from before disulfiram to timepoints during and after disulfiram administration. This study is registered with ClinicalTrials.gov, number NCT01944371.
Of 34 participants screened for eligibility at The Alfred Hospital (Melbourne, VIC, Australia), and San Francisco General Hospital (San Francisco, CA, USA), 30 people were enrolled between Sept 24, 2013, and March 31, 2014. The estimated fold increases in cell-associated unspliced HIV RNA from baseline were 1·7 (95% CI 1·3–2·2; p<0·0001) to the timepoint during disulfiram treatment and 2·1 (1·5–2·9; p<0·0001) to the timepoint after disulfiram in the 500 mg group; 1·9 (1·6–2·4; p<0·0001) and 2·5 (1·9–3·3; p<0·0001) in the 1000 mg group; and 1·6 (1·2–2·1; p=0·0026) and 2·1 (1·5–3·1; p=0·0001) in the 2000 mg group. No deaths occurred, and no serious adverse events were noted. Disulfiram was well tolerated at all doses.
Short-term administration of disulfiram resulted in increases in cell-associated unspliced HIV RNA at all doses, consistent with activating HIV latency. Disulfiram may be suited for future studies of combination and prolonged therapy to activate latent HIV.