 |
|
|
| |
14th International HIV Drug Resistance Workshop
|
| |
Reported for NATAP by David Margolis, MD, University of North Carolina
TOPICS
--Mother-To-Child Transmission: nevirapine resistance can be reduced
--Response to ritonavir-boosted PIs: Reyataz (genotypic mutation scores & phenotypic parameters of response: fosamprenavir
--Predictive genotypic algorithms: is there a concern interpreting resistance test reports?
-- A small study of ZDV/3TC/TDF
-- The appearance of the K65R mutation: does the L74V predispose for the K65R
-- More concern about the ddI and tenofovir combo
-- Resistance to tipranavir: T-20 improved response; genotypic mutations & phenotypic cutoffs associated with tipranavir non-response
-- When to look for resistance after interruption of therapy
--The complexities of entry inhibitors: 1 patient for whom R5X4 emerged; preliminary cross-resistance reports
-- A few new and novel drugs
--Transmission of resistant HIV: multi-drug resistance, transmission & persistence
--Draining HIV's latent reservoir: can HIV be eradicated?
Although the 14th International HIV Drug Resistance Workshop ("Basic Principles and Clinical Implications") lacked some of the flash and clash of prior meetings, it did not disappoint those looking for new insights into the complexities of this clinically critical area. Workers both in academia and industry strive to better explain how HIV evades antiviral therapy, and to further develop resistance assays that can provide useful clinical guidance to treating clinicians. However, as both the worldwide epidemic and antiviral therapy undergo parallel and interactive evolution, the increasing complexity of the field threatens to leave practicioners on the sidelines scratching their heads.
Each section report below is linked to a second detailed review from NATAP's coverage of the 14th HIV Drug Resistance Workshop.
Mother-To-Child Transmission
The meeting began with studies that addressed the thorny problem of the emergence of non-nucleoside reverse transcriptase inhibitor (NNRTI) resistance following the use of nevirapine to prevent mother-to-child transmission (MTCT) of HIV. Many are vigorously testing the use of other antivirals in combination with or in place of nevirapine (NVP), to prevent HIV transmission and the induction of NNRTI resistance.
Susan Eshleman and colleagues (abstr. 1) found that should a baby become HIV-infected despite MTCT prophylaxis, use of NVP by the mother substantially increased the risk of NVP resistance. A regimen in which the mother was not given NVP, but the newborn was given both NVP and AZT for 7 days resulted in a risk of HIV transmission that was as low as several other regimens tested (single-dose NVP to mother and child, single-dose NVP to child, and single-dose NVP to mother and 7 days of NVP/AZT child) but with the lowest risk of NVP resistance. Eshleman argued that by avoiding exposure of the mother to NVP, NVP could be used by the mother for HAART or future pregnancies.
A South African study (abstr. 2) presented encouraging preliminary results from the beginning of a study comparing the addition of 4 days or 7 days of Combivir (AZT & 3TC) to single-dose NVP given to both mother and child. Preliminarily, they found little NNRTI resistance by standard techniques in mothers and no NNRTI resistance in children who became HIV-infected despite MTCT prophylaxis. Comparison of the 4-day to 7-day regimen is ongoing, as the single-dose NVP arm has been halted.
The group at the NCI Frederick Drug Resistance Program (abstr. 3) used the highly sensitive allele-specific PCR approach to detect tiny subpopulations of drug-resistance HIV in clinical samples. Examining samples from the South African cohort above, the group found that NNRTI resistance in mothers could be detected after NVP alone, but in only 27% after 4 or 7 days of Combivir in addition to NVP. It is known from previous studies that that maternal response to NNRTI-containing HAART is reduced if resistance is detected by standard techniques after MTCT prophylaxis. It is not yet known if the low levels of resistance detected by allele-specific PCR will also lead to poorer treatment responses.
Nevirapine Resistance in Mother-To-Child Transmission (06/13/05)
http://www.natap.org/2005/resistance/drw_1.htm
Response to ritonavir-boosted PIs:
Several presentations examined resistance testing for the protease inhibitor atazanavir (ATZ). Coakley presented the Virologic Phenosense assay cutoff values for ATZ and ritonavir-boosted ATZ (ATZ/r) derived from the datasets of ATZ licensing studies in protease-naïve and protease-experienced subjects. 76% of naïve subjects achieved viral load of <400 copies/ml at 24 weeks if FC <2.2. 77% of protease-experienced subjects achieved a viral load of <400 copies/ml at 24 weeks if FC <5.2. These findings will aid those who use the Phenosense assay, and suggest that modest levels of resistance will thwart unboosted ATZ.
Pellegrin (abstr. 8) studied virological response to ATZ/r based regimens in 90 patients given ATZ/r (300mg/100mg) and TDF/FTC. Resistance to ATZ was scored using a French (ANRS) resistance algorithm counting the protease mutations: L10F/I/V, K20I/M/R, L24I, L33F/I/V, M36I/L/V, M46I/L, G48V, I54L/V, L63P, A71I/V/T, G73A/C/S/T, V82A/F/T/S, I84V, and L90M. A score of <6 was considered sensitive, and 6 mutations resistance. ATZ concentrations were measured, and a genotypic inhibitory quotient (GIQ), meant to combine the information of drug level achieved in the patient and the resistance of the patient's virus, was calculated by dividing the trough ATZ concentration (Cmin) by the number of ANRS ATZ resistance mutations. As might be expected, the extent of viral load decline or the likelihood of clinical success was much greater in those with a sensitive ATZ score or a low GIQ. ATZ drug levels alone were not helpful. Of interest, the protease inhibitor mutations L10I, K20R, M46I/L, 54L/T/V, and L90M, associated with exposure to and resistance to several protease inhibitors were associated with failure.
Reyataz resistance parameters predicting outcomes phenotypic cutoffs and genotypic mutation scores (06/13/05):
http://www.natap.org/2005/resistance/drw_3.htm
Similarly with respect to fosamprenavir, Pelligrin (abstr. 31) found that successful salvage therapy with FosAPV/r (700mg/100mg bid) could be predicted by a FosAPV/r resistance score. Again, failures were associated with the presence of L33F, A71I/T/V/L, and especially L90M, protease-resistance mutations that are not counted towards FosAPV/r resistance in the ANRS score.
Vora (abstr. 7) and colleagues presented a similar study of 62 PI-experienced patients who were switched to ATV/r-containing HAART. 82% of patients had a virological response, and 56% achieved <50 copies/ml at month 3. Vora found 13 PI mutations in the group at baseline associated with reduced response to ATZ/r (10F/I/V, 16E, 33I/F/V, 46I/L, 54L/V/M/T, 60E, 62V, 71I/T/V/L, 82A/T, 84V, 85V, 90M, and 93L). As in the French study above, an ATZ resistance score could be constructed that was predictive of response. Response was seen in 80% or more of patients with 3 or less of the most predictive mutations. However, again as above, mutations previously related to PI exposure but not ATZ resistance was linked to reduced ATZ/r response. Finally, Struble (abstr. 29) presented an analysis performed by the FDA of the number and type of baseline PI mutations that affected response rates to ATZ/r or lopinavir/r (Kaletra). The common PI resistance mutations M46I/V/L, I54V/L/M/A/T, A71V/T/I, V82A/T/F/S, I84V, or L90M at baseline reduced response rates to both ATV/r and LPV/r arms. Response rates were similar between ATV/r and LPV/r-treated subjects with zero to four PI-associated mutations at baseline (75% if <3; <40% if >3 mutations). However, no subjects responded to ATZ/r and only 28% responded to Kaletra if five or more of these PI-associated mutations were present at baseline.
Predictive genotypic algorithms:
Costaglia presented preliminary results (abstr. 9) from the Forum for Collaborative HIV Research, combining databases from ACTG, British Columbia, EuroSIDA, the ARNS, the Swiss cohort, the Stanford and UK databases, and two Italian databases. As genotypic rules-based algorithms for abacavir are thought to be relatively reliable, but those for ddI thought to be less reliable, algorithms for these drugs were compared. Patients had failed a previous regimen and started either abacavir or ddI for the first time. Their genotypes were interpreted to report sensitivity (S), intermediate resistance (I), or resistance (R) to abacavir and ddI using multiple genotypic algorithms (ANRS-V12, Detroit Medical Center-3, Stanford HIV RT and PR Sequence Database-8 and Rega-6.3 for both drugs, and CHL-4.4, Retrogram-1.6, Sao Paulo-2 and VGI-5.0 for abacavir). In brief, when a virus was deemed resistant to ABC or ddI by all algorithms a poor virological response at week 8 was likely. However, for both drugs in nearly all genotypic interpretation systems, the clinical virological response of patients to abacavir or ddI whose viruses were reported to be fully sensitive to abacavir or ddI was not reliably superior to the response of those patients whose viruses were reported to have intermediate sensitivity to those drugs. But, poor response to ABC or ddI was reliably predicted by a genotypic score showing resistance.
These are preliminary results, and largely illustrate the need to pool large datasets and develop innovative statistical approaches to interpret resistance information. As drug exposure over the worldwide population of HIV-infected people expands, differences in viral species worldwide come into play as access to therapies improves, and as new drugs continue to be implemented, this will probably be a necessary but never-ending task. For the time being, it is important for clinicians to know that a test report of resistance is likely to be highly reliable, but clinical decisions on which set of drugs to select for a patient should make use of an array of factors (eg. Adherence, tolerability, etc.) rather than simply by preferring drugs read as fully sensitive to those called partially susceptible. At this point in time, genotypic tests appear unable to distinguish clearly between intermediate and full susceptibility.
A small study of ZDV/3TC/TDF:
Masquelier (abstr. 20) reported the results of an open label pilot study of this combination in 24 patients, 3 with early infection. At baseline 5 had mutations (T215D/C in 4, M41L in 3), the baseline median CD4 cell count was 443/_l, and baseline median VL was 4.38 log10copies/ml. The median decrease in VL was -2.26 log at 1 month (with 2 subjects lost to follow-up), and at 12 months >60% by ITT and >80% on Rx had VL<50copies/ml. In the 4 virologic failures 2 failed with WT virus, one with 210W and M184V added to M41L & T215D at entry, and one with K65R and M184V. Despite the scientific rationale that this drug combination should simultaneously select for antagonistic resistance mutations difficult for the virus to develop, nearly 5% of the subjects (one subject in this small study) failed therapy with a K65R and M184V virus that in theory should be "crippled." The authors conservative conclusion that "this combination could be of interest when other therapy cannot be prescribed" was therefore the correct one.
The appearance of the K65R mutation:
At last year's meeting in Tenerife, one presentation suggested that the L74V mutation, potentially selected by abacavir, could predispose to the selection of the K65R mutation, and therefore resistance to both abacavir and tenofovir. A group from Paris and GSK (Deschamps, abst. 14) used selective PCR to detect K65R, and found that it was very rare. In bulk sequence analysis (as in standard clinical genotyping) 4 patients who failed therapy with ABC and 3TC and a PI or NNRTI, L74V and M184V were found in 3, and only M184V in the fourth. In samples from these 4 patients clonal analysis using selective PCR was used to try to detect minor populations of virus with K65R in these samples. K65R was found in one clone of 720 studied at baseline from one patient, and again in one of 885 clones in this patient at failure with M184V and L74V. In one other patient, one clone with K65R was found in 466 clones at failure with M184V and L74V. No clones with K65R were found in 524 and 270 clones at failure in the other two patients.
Another study (abstr. 97) found the coexistence of K65R and L74V on the same HIV genome was rare but detectable. Of about 3600 individual patient samples in Marseille database, 12 had K65R +L74V/I (0.4%). These patients, however, had high viral loads, calling into question the claim that a virus with these mutations would be very "unfit." In two patients, TAM mutation(s) were found in the same viral clone as the K65R mutation, despite the fact that these mutations are mechanistically antagonistic.
Tenofovir & Nuke Resistance Mutations & Patterns (06/17/05)
http://www.natap.org/2005/resistance/drw_15.htm
Virologic Response to Antiretroviral Therapy in the Setting of the K65R Mutation (06/14/05)
http://www.natap.org/2005/resistance/drw_13.htm
More concern about the ddI and tenofovir combo:
Although these drugs are single-pill, once-a-day drugs several studies raise issues about their use together. One of these was just published (Desmond et al., Early virologic failure in HIV-1 infected subjects on didanosine/tenofovir/efavirenz: 12-week results from a randomized trial. AIDS, July 2005). Gatell et al. reported a high incidence of K65R and NNRTI resistance in a cohort analysis of patients treated with ddI/TDF and an NNRTI or another NRTI (abstr. 18), and suggested avoiding the use of ddI with TDF unless a boosted PI was also used. This was also one of the conclusions also reached by Blanco, Gatell, and others in a study of the addition of ddI to therapy in patients with low-level (<1000 copies/ml) viremia. In this study, ddI intensification was often successful when less than 4 NRTI mutations were present unless both K65R and L74V were present.
Resistance to tipranavir:
An analysis of data from the pivotal phase III RESIST trials that have resulted in licensure of tipranavir provided some insight into factors that predicted non-response to tipranavir (abstr. 27). Less than 14% of study subjects had no more than a 0.5 log drop from baseline in HIV-1 RNA during the first 8 weeks of treatment with ritonavir-boosted tipranavir and an optimized background regimen. Patients with more than two mutations at codons 33, 82, 84, and 90 were excluded from RESIST. A tipranavir resistance score consisted of the number of codons with mutations at 10V, 13V, 20M/R, 33F, 35G, 36I, 43T, 46L, 47V, 54A/M/V, 58E, 69K, 74P, 82L/T, 83D, and 84V. Note that many of these are mutatiosn common to many PIs.
92% had at least a 0.5 log drop from baseline in HIV-1 RNA during the first 8 weeks if less than 3 of these mutations were present. 84% responded if 3 to 5 of these mutations were present, and 72% if six or more were present. Therefore in the setting of optimized therapy, and the absence of more than two of the 33, 82, 84, and 90 mutations, the likelihood of short-term response to tipranavir is good.
Using the Virco Antivirogram phenotype, fold-changes of 0-3 were deemed sensitive, had a high likelihood of response, and correlated with 0-4 mutations contained in the tipranvir score. Fold changes of 3-10 and >10 were deemed intermediate and resistant, respectively, had a lesser likelihood of response, and correlated with 5-7 and 8 or more mutations contained in the tipranavir score.
Tipranavir: predicting viral response; T20 improves response;
Non-response to Tipranavir is associated with pretreatment resistance characterized by tipranavir phenotype or genotypic tipranavir Score (06/13/05)
http://www.natap.org/2005/resistance/drw_4.htm
When to look for resistance after interruption of therapy:
On occasion, antiviral therapy must be interrupted for various reasons, and clinicians sometimes wish to obtain a resistance test after the interruption of successful therapy to "see what is there." Winters and colleagues (abstr. 35) from the ACTG 5102 study report findings their cohort of 46 subjects with HIV RNA <200 copies/ml and >500 CD4 cells who withdrew drugs after a median of 48 months. 43% of subjects developed HIV RNA >500 copies/ml by week 2 post-interruption, 78% by week 4, and 100% at week 8. However, drug resistance mutations were found in only 11 of 46 subjects at viral rebound, and in 10 of these 11 at week 2 after interruption. The authors therefore suggested that in seeking evidence of drug-resistant species in patients previously suppressed on HAART, testing is best performed about 2 weeks after interruption.
The complexities of entry inhibitors:
Foreshadowing the complexities of dealing with resistance to the emerging class of chemokine receptor antagonists, Don Mosier (abstr. 60) found that mutations in the V3 loop of HIV envelope were most often associated with the development of a mixed R5X4 phenotype. Evolution of an R5 receptor-using virus into a virus that can use both the R5 and the X4 receptor is thought to be the likely path by which HIV may evade R5 blockers. These dual-tropic viral intermediates appear to be less efficient at entering cells. As an R5 virus switches to X4 use, there appear to be major barriers to replication. Intermediates are less fit and more susceptible to R5 antagonists and R5 chemokines. Mosier expressed the hope that switching of R5 viruses to X4 may be a less common clinical problem than feared.
The emergence of a dual-tropic (using both R5 and X4 coreceptor) virus was reported in 1 of 39 subjects receiving the GSK CCR5 blocker 873140 for 10days (abstr. 61). This subject was ART Naïve, received the low 200mg QD dose of "140," had an HIV RNA decline of only -0.53 log from baseline (ED Note: I think VL decline was in line with other patients receiving same dose), and drug levels and R5 receptor occupancy (a new assay of how much drug is sticking to cells) comparable to others studied in this dosing group. Careful clonal analysis suggested that this resistant virus emerged by selection of a minority population of dual tropic virus that was present before drug was given. The frequency of this isolate increased for 4% of the total viral population at initiation of therapy, to 27% at day 10, and fell to 8% two weeks after drug was stopped. This virus was sensitive to "140" when tested on cells that possessed only R5 and not X4, suggesting that it was selected in the patient via X4 use. The group from GSK concluded that minor populations of dual tropic virus, undetectable by the Virologic entry assay, could be amplified and selected rapidly during R5 blocker monotherapy.
In a laboratory study (abstr, 59), the Schering R5 inhibitors SCH 351125
and SCH 417690 were used in the lab in an attempt to intentionally select resistant variants by serially culture in PBMCs in the presence of increasing concentrations of SCH 351125 or SCH 417690. Resistant viruses emerged in some but not all cultures after 12-16 weeks of continuous culture. Almost all viruses remained CCR5-tropic in primary cells and were unable to infect PBMCs from donors lacking functional CCR5. Resistance emerged slowly, and resistance patterns were variable. However, cross-resistance was seen among the resistant isolates to the SCH compounds and CPD 167, another R5 blocker.
Further, one isolate grew in cells that lacked R5 but displayed the CXCR4 co-receptor, suggesting switch to X4 use. However, it was reported at the meeting that this virus was resistant to the X4 blocker AMD-3100, suggesting that the virus was using coreceptor other than X4 or R5, or entering via a coreceptor-independent pathway.
On a brighter note, Mike Westby (abstr. 65) showed that Maraviroc-resistant HIV-1 variants, selected by serial passage, were sensitive to other CCR5 antagonists and T-20. Maraviroc is the tongue-twisting name for the Pfizer R5 inhibitor, UK-427,857, now in advanced phase III testing. Maraviroc-resistant biological clones were susceptible to the CCR5 antagonists GW873140, SCH-C, SCH-D, and to the fusion inhibitor T-20.
In summary, these studies suggest that while it might be difficult for the virus to evolve to escape R5 dependence, given enough time (ie, the lack of other effective antiviral on board and functional entry inhibitor monotherapy) the virus may utilize multiple pathways to escape drug pressure: use of R5 despite the presence of inhibitor, use of X4, or use of an alternate chemokine coreceptor.
Resistance to UK-427,857 does not result in cross-resistance to other entry inhibitors (06/13/05)
http://www.natap.org/2005/resistance/drw_7.htm
GSK CCR5 '140', 1 patient had dual tropic R5X4 virus emerge (06/13/05)
http://www.natap.org/2005/resistance/drw_6.htm
Schering CCR5 Drug: Resistant viruses showed cross-resistance to other CCR5 inhibitors (06/13/05)
http://www.natap.org/2005/resistance/drw_5.htm
A few new and novel drugs:
Ray Schinazi introduced another new NRTI from his stable of antiviral compounds (abstr. 58). 1-(beta-D-Dioxolane) thymine or DOT is an orally bioavailable NRTI, reported to have activity against HIV-1 NRTI-resistant variants in the laboratory. A variety of pre-clinical studies suggested that the drug might not have marrow or mitochondrial toxicity, and retained activity against multidrug resistant viruses containing M184V, multiple TAMs, K65R, L74V, the T69S insertion. It was suggested that clinical studies were needed.
BILR 355 BS is a novel NNRTI with potent antiviral activity against wild-type HIV-1 and recombinant viruses encoding >1 mutation conferring phenotypic
resistance to the currently available NNRTIs (abstr. 74). 89% (108/121) of the single and double mutant isolates containing the K103N mutation with or without other NNRTI mutations had Fold Change <10 to BILR.
BILR 355: New NNRTI (06/13/05)
http://www.natap.org/2005/resistance/drw_10.htm
Jochmans and colleagues from Tibotec presented further details (abstr. 83) on a new class of NRTI, first presented at CROI 2005. The nucleotide-competing RT inhibitors (NcRTIs) bind the active site of RT and compete with incoming nucleotides, blocking their engagement of RT. This mechanism of action is distinct from that of the chain terminating NRTIs and the non-competitive NNRTIs. Surprisingly, physiological concentrations of ATP dramatically increase the in vitro activity of NcRTIs, an effect exactly the opposite of the inhibitory effect of ATP on zidovudine-TP activity. Whether these compounds can move forward to become non-toxic antivirals remains to be seen.
Transmission of resistant HIV:
Brenner (abstr. 112) reported transmission events within risk groups following primary HIV-1 infection (PHI) detected in Quebec in the last 7 years. Phylogenetic analysis identified potential transmission events. At least half of PHI infections (n=244/484) segregated into 71 clusters, with a range of 2-11 individuals per cluster. Evolutionary distances of viral isolates within clusters suggested that a major proportion (>50%) of newly transmitted infections may arise during the two year period following PHI. As suggested by others, this establishes public health rationale for earlier initiation of treatment of HIV-infected individuals who would also benefit from lower set points in regard to plasma viraemia.
The frequency of transmission of DMR HIV was debated. Johnson and colleagues from the CDC (abstr. 111) felt that multi-drug resistant HIV-1 was transmitted more frequently than current estimates. Using real-time PCR assays they detected MDR HIV in 9 of 28 (32%) of samples in which MDR was not detected by bulk sequencing. Metzner and colleagues in Germany (abstr. 110) presented quite similar result using similar techniques. However, van de Vijver
(abstr. 113), found that complex multi-class and MDR HIV was transmitted less often than its frequency in the "donor" population. This was not explained by "donor" viral load, and thought to be due to a fitness bottleneck faced during transmission by MDR virus.
In a separate but related study, Kearny and colleagues at the NCI Drug Resistance program (abstr. 143) showed that by using the very arduous and very sensitive single genome sequencing method (SGS) that polymorphisms (mutations) could be detected at sites related to drug resistance in the vast majority of HIV-infected individuals never exposed to therapy. 50% of 27 subjects studied had mutations at a protease drug resistance site by SGS, but 0% by standard genotyping. 77% had a polymorphism at a NRTI drug resistance site, and 73% at an NNRTI drug resistance site. However, viral genomes with these mutations were always in the vast minority, ranging from as little as 0.03% to only as much as 4.3% of the total viral populations (that is 99.97% to 95.7% of the viral genomes were wild type). Of course, this is the reason that combination antiviral therapy is needed. Presumably, these mutations are present in part due to the natural variation of the HIV genome, and perhaps in part due to the transmission of drug-resistant HIV. However, patients with >40% of genomes carrying the same resistance mutation were excluded from analysis, to remove the impact of high-level transmission of drug resistance. Whether these rates of low-level polymorphism will change over time as the epidemic continues to evolve will require follow-up studies.
Reduced susceptibility to protease inhibitors (PI) in the absence of primary PI resistance-associated mutations
http://www.natap.org/2005/resistance/drw_14.htm
However, it has been shown that if transmitted, MDR HIV will often persist as the predominant species in the blood for some time. Smith and colleagues at UCSD (abstr. 115) also found slow reversion of HIV transmitted drug resistance to non-nucleoside reverse transcriptase inhibitors in semen. NNRTI resistance persisted in the blood and semen in the majority of the participants throughout
follow-up (mean 662 days, range 95-1179 days). Only one individual displayed full reversion of NNRTI resistance (K103N) in the seminal compartment, occurring between 512 and 1193 days.
On the other hand, despite the recent report of a patient in NY who had rapid progression to AIDS with the transmission of an MDR HIV, this is not frequently observed. Chaix reported for the PRIMO Cohort Study Group(ANRS CO 06) and the AC11-ANRS ResistanceStudy Group. Their group of 797 patients enrolled in the ANRS Primo Cohort and the ANRS-AC11 Resistance study group. Ten patients (seven treated) out of 797 harboured HIV strains resistant to at least one antiretroviral drug of the three classes (incidence 1.25%). Over 36 months, none of the patient progressed to AIDS, with or without therapy.
Draining HIV's latent reservoir: can HIV be eradicated?
With apologies for self-promotion in this report, we presented preliminary findings from a study to deplete the pool of HIV-infected resting CD4 T cells. The ability of HIV to establish persistent but quiescent infection in long-lived resting CD4 T cells allows the virus to escape the immune response, evade the effects of HAART, archive drug-resistant mutants, and is a primary obstacle to eradication of HIV infection.
We have found that histone deacetylase 1, a member of a family of human enzymes that dampens gene expression, maintains quiescence of HIV that has entered latency within resting CD4 T cells. Valproic acid (VPA), a drug used to treat seizures and depression, is an inhibitor of histone deacetylase.
We tested the ability of VPA to deplete replication-competent HIV in vivo within the resting CD4+ T cell pool in stably aviremic, patients on standard combination antiretroviral therapy (HAART). To prevent the spread of HIV to uninfected cells during the period of VPA therapy, we intensified therapy by the addition of enfuvirtide (T-20) agent prior to VPA treatment. Four volunteers with durable suppression of viraemia (<50 copies/ml for mean of >46 months) added T-20 for 4-6 weeks, and then VPA to their therapy for 3 additional months. The frequency of replication-competent HIV in resting CD4+ T cells (infected units per million cells; IUPM) was enumerated by limiting-dilution culture assays of maximally activated resting CD4+ T cells recovered from these patients before and after exposure to VPA. A significant decline (mean reduction 64%; range 52% to >84%) occurred in three of four patients after treatment with T-20 and VPA. Further studies of the effect of HDAC inhibitors such as VPA and of antivirals with novel mechanisms of action such as T-20 on persistent infection within the resting cell reservoir are warranted.
| |
| |
| |
| |
|
|
|
|
|
