The 41st Interscience Conference on Antimicrobial Agents and Chemo-therapy (ICAAC), held in Chicago from December 16 to 19, had been postponed due to the events of September 11, 2001. As a result of the awkward timing, attendance was somewhat sparse. Antiretroviral news was somewhat sparse, as well, which probably has more to do with the growing number of conferences at which data pertaining to HIV infection and its treatment are presented.

Resistance

Not surprisingly, the most widely publicized presentation at ICAAC was Doug Richman's analysis of drug resistance from the HIV Cost and Service Utilization Study (HCSUS) [Abstract LB-17]. The media were quick to trumpet the bad news that drug resistance is common, but most of the time they got the story wrong. HCSUS is a longitudinal cohort of patients receiving care at a variety of U.S. sites that was established in 1996. Results from 1906 plasma samples were weighted and modeled to represent 208,724 patients with HIV infection under care in the U.S. Viral load was undetectable (<500 c/mL) in 698 of the 1906 samples, and was >500 c/mL in 1208 samples (63.4%). Of those with detectable viremia, phenotypic resistance testing could only be performed in 1080 samples, representing 117,976 U.S. patients. Drug-resistant virus was found in 78% of those samples, a figure that led to the erroneous headline that "three-quarters of U.S. patients have drug resistance." If one made the unreliable assumption that those with undetectable virus had no resistance, the proportion with resistance would be approximately 50%. Among those with viremia, nucleoside analog resistance was observed in 70% of the samples, the majority having 3TC resistance. PI and NNRTI resistance was observed in 42% and 31%, respectively. Over 50% of samples had resistance to two classes, and 14% had resistance to all three classes. Resistance was noted in 87% currently on therapy and 41% of those not currently under treatment, as well as in 20% of those who claimed to have never received anti-retrovirals. Predictors of resistance in a multivariate model included lower CD4 nadir, higher baseline viral load, male sex, being treated in a smaller practice, and geographic region, with the lowest levels of resistance being observed in patients from the Midwest.

While these findings are disturbing, the gloom and doom message has been exaggerated. First, it is important to remember that the presence of drug resistance does not mean that patients are untreatable. 3TC resistance, for example, was the most common type of nucleoside resistance observed in this study and is extremely common in patients who have been treated with that drug. While it decreases the antiviral activity of 3TC, it may have beneficial effects with respect to the activity of thymidine analogs and tenofovir. Second, the susceptibility cut-offs used in this study to define resistance to PIs and NNRTIs was a 2.6-fold change, which is lower than the cut-offs used clinically for some of these agents. This may have resulted in an over-estimate of resistance to some drugs. Finally, this cohort enrolled in 1996, a time when many patients had been treated with sub-optimal therapy and when our understanding of how to use HAART was primitive at best. The temporal effect may limit our ability to generalize these findings to patients beginning therapy today. It should also be noted that this study once again emphasizes the importance of care by experts. Patients seen in larger HIV practices had less resistance, a finding contrary to what might be expected, since patients are often referred to HIV experts only after they have failed initial regimens.

Return to top

When to Start Therapy

Matthias Egger reported combined data from 13 European and U.S. cohorts looking at 12,574 patients initiating antiretroviral therapy [Abstract LB-18]. Mean duration of follow-up was 2 years, representing an impressive 24,310 person-years. The goal of the study was to estimate the probability of progression to a new AIDS event or death for naïve patients starting therapy and to determine factors associated with AIDS-free survival. There was a total of 1045 AIDS events and 344 deaths among patients in these cohorts during the follow-up period. As has been seen in other cohorts, the most important predictor of AIDS-free survival was CD4 count at initiation of therapy. Those who started therapy with CD4 counts >200 cells/mm³ had an estimated 3-year AIDS-free survival of 95%, compared with 75% in those with baseline CD4 counts <50 cells/mm³. There was no significant difference in AIDS-free survival for patients who started therapy in CD4 strata above 200 cells/mm³, but those with CD4 counts <200 cells/mm³ had a lower AIDS-free survival than those with higher CD4 counts at baseline. Using patients with baseline CD4 count <50 cells/mm³ as a comparator (RH= 1.0), the relative hazard for AIDS events or death was 0.75 for those with CD4 counts of 50-99 cells/mm³, 0.53 for those with 100-199 cells/mm³, 0.25 for those with 200-349 cells/mm³, and 0.18 for those initiating therapy with >350 cells/mm³. In contrast, baseline viral load was not an important predictor of AIDS-free survival unless it was over 100,000 c/mL, in which case estimated 3-year AIDS-free survival was 85%, compared to 90-95% with a viral load <100,000 c/mL. Other factors significantly associated with AIDS events or death included age >50 years, injection drug use, and CDC stage C disease.

These data from a very large cohort may have greater relevance to today's practice than data from cohorts in the pre-HAART era. They support current treatment guidelines which recommend deferral of therapy and emphasize the importance of the CD4 count as the most important indicator of need for therapy. However, it should be noted that the amount of follow-up time in this study is limited, as it is in a number of the other cohort studies that have led to a change in treatment practices and recommendations. Ongoing follow-up from these large and important cohorts will help to determine the safety of deferral of therapy, since we may never answer the question of when to start therapy based on randomized, controlled clinical trials.

Return to top

Switching Therapy

Christine Katlama presented data from the TRIZAL study, in which patients on HAART with viral loads <400 c/mL for at least 6 months and viral loads <50 c/mL at screening were randomized to continue their current therapy (n=103) or to switch to AZT/3TC/ABC (Trizivir, n=106) [Abstract I-671]. At 48 weeks 22% of the patients in both arms had experienced treatment failure (virologic failure or premature discontinuation) by intent-to-treat analysis. However, at 24 weeks virologic failure had occurred in 5 patients in the AZT/3TC/ABC compared to 1 patient in the continued HAART arm. Three of the 5 patients failing triple nucleoside therapy at 24 weeks achieved undetectable viral loads at 48 weeks, one with the addition of efavirenz. Minimal data were presented regarding resistance patterns in patients experiencing virologic failure. Another question that needs to be answered by such switch studies is whether patients with baseline viral loads >100,000 c/mL, who appear to have poorer responses to initial therapy with AZT/3TC/ABC, can safely switch to a triple nucleoside regimen after achieving virologic suppression on other HAART regimens. In the SWATCH study, regimens were proactively switched at pre-determined time intervals. Patients were randomized to receive d4T/ddI/EFV (n=52), AZT/3TC/ NFV (n=54), or to alternate between these two regimens at 3 month intervals (n=55) [Abstract I-672]. By intent-to-treat analysis at 48 weeks, 69% of those in the switch arm had a viral load <400 c/mL compared to only 57% of those on stable therapy (100% vs 85% by as-treated analysis). Adherence levels were high and did not appear to differ among the three arms. The reason for the improved response in the alternating therapy group is unclear.

Return to top

Stopping Therapy

This author presented data from an observational study of 62 patients without a history of CDC-defined AIDS who discontinued therapy with the intention of restarting based on laboratory parameters [Parish M, et al. Abstract I-673]. At the time of the analysis, 74% remained off therapy after a mean interruption of 64 weeks. The remaining 26% resumed therapy after a mean interruption of 33 weeks. The estimated rate of CD4 decline was 409 cells/yr among those who resumed therapy vs 124 cells/yr among those who remain off therapy. The median estimated time to CD4 <200 cells/mm³ was 1.7 yrs vs 4.6 yrs, respectively. While 88% of resumers and 70% of non-resumers met earlier DHHS criteria for initiation of antiretroviral therapy (p=0.154), only 50% of resumers and 26% of non-resumers met criteria according to 2001 DHHS guidelines (p=0.077). Factors significantly associated with resumption of therapy and rate of CD4 decline included pre-HAART viral load and CD4 nadir as well as viral load rebound following treatment interruption. Although one patient in the cohort experienced symptomatic viral rebound, no opportunistic infections or other HIV-related complications occurred during treatment interruption. Most of the patients who resumed therapy achieved viral resuppression, and there was a suggestion of improvement in some of the metabolic complications of HAART during treatment interruption. These findings suggest that for a subset of patients, especially those with relatively high pre-treatment CD4 counts and/or low pre-treatment viral loads, prolonged treatment interruption may be possible, with resumption of therapy based on the same criteria that would be used in a naïve patient. This strategy is often referred to as "pulse therapy," and is being tested in large-scale randomized clinical trials. However, it is also likely that these results can be explained in part by a temporal effect: Many of the patients in this study may have done well off therapy because they had marginal indications for therapy to begin with. As we move toward later initiation of therapy, it is possible that fewer patients will be good candidates for prolonged treatment interruption.

Return to top

Tenofovir

The 41st ICAAC was an important conference for tenofovir disoproxil fumarate (TDF, Viread^TM, Gilead Sciences), though the data would have been less anticlimactic had they been presented as planned in September before the drug was approved by the FDA. Tenofovir is a nucleotide reverse transcriptase inhibitor that is administered as a single 300 mg tablet once daily with food. Kathleen Squires presented data from the Gilead 907 study, a phase III randomized, double-blind, placebo-controlled trial in which 552 patients experiencing virologic failure (HIV RNA 400-10,000 c/mL) on a stable antiretroviral regimen were randomized to add TDF (n=368) or placebo (n=184) to their existing regimen [Abstract I-666]. At 24 weeks, patients who added TDF had a drop in viral load of 0.61 log10 c/mL compared to a drop of 0.03 in the placebo group (p<0.0001). Viral load was <400 c/mL in 42% of the TDF recipients and 13% of the placebo recipients (p<0.0001), and 22% vs 1% had viral loads <50 c/mL (p<0.0001).

Michael Miller presented data from a resistance sub-study involving 253 evaluable patients from the 907 trial [Abstract I-1928]. Patients with an isolated M184V mutation at baseline had a somewhat more pronounced drop in the viral load than the group overall, but when M184V was combined with thymidine analog mutations (TAMs), this difference disappeared. Both the K65R mutation and the T69SSS insertion mutation are known to cause loss of susceptibility to TDF, but in the 907 sub-study, patients with 3 or more TAMs that included either M41L or L210W also had a poor response to addition of TDF. There is probably no single clinically relevant susceptibility cut-off for TDF. Patients with a <1-fold change in susceptibility have the best response, and those with >4-fold decrease in susceptibility can be considered resistant. However, those with a 1-4-fold loss of susceptibility may still respond to TDF, not unlike abacavir, where a 4.5-6.5-fold change in susceptibility indicates intermediate resistance and partial activity. Fortunately, the development of RT mutations appears to be uncommon in patients treated with TDF for 96 weeks [Schooley R, et al. Abstract I-1929].

The demonstration of cross-resistance between TDF and nucleoside analogs suggests that this is not a salvage drug. It is currently being used as a nucleoside alternative in patients with prior nucleoside experience or for intensification in patients with low-level virologic failure, as in study 907. However, its simple once-daily dosing, excellent tolerability [Becker S, et al. Abstract I-1930; Schooley R, et al. Abstract I-1929], and apparent lack of mitochon-drial toxicity make it a promising agent for use in initial therapy. The results of the ongoing Gilead study 903 will help us to determine the role of this drug for the treatment of naïve patients.

Although TDF appears to have few drug interactions, it also appears to increase the AUC of buffered ddI by 40% [Flaherty, et al. Abstract I-1729]. This is concerning, as co-administration of these two drugs could result in increased ddI toxicity, including neuropathy and pancreatitis. Studies have not been carried out yet using the enteric coated formulation of ddI (Videx EC). It may also be possible to offset this interaction by administering both drugs with food. If this approach is supported by ongoing studies, it would make it much easier to use TDF and ddI together in once-daily regimens.

Return to top

Investigational Antiretroviral Agents

• TMC 125 is a "second generation" NNRTI being developed by Tibotec-Virco. It demonstrates in vitro activity against highly NNRTI-resistant virus. Results of a phase 2 trial were presented by van't Klooster, in which 12 treatment naïve patients received 7 days of monotherapy with TMC 125 at a dose of 900 mg bid, compared with 6 patients who received placebo [Gruzdev B, et al. Abstract I-668]. All doses were directly observed. Patients receiving active drug experienced an impressive 2 log drop over the one-week treatment period (range 1.13-3.30 log₁₀ c/mL). Eight achieved viral loads <400 c/mL in 7 days, and two achieved viral loads <50 c/mL. Some formulation issues will have to be worked out before this promising agent is brought to large scale clinical trials, however. Patients on active drug swallowed 18 pills twice daily!
• Tipranavir (TPV) is a non-peptidic protease inhibitor being developed by Boehringer Ingelheim. It has promising activity against a wide variety of PI-resistant strains. Charles Farthing reported 16-week data from an open-label, randomized trial comparing two doses of TPV/RTV (500/100 mg bid and 1250/100 mg bid) with SQV/RTV (400/400 mg bid) in patients who had failed a single PI (HIV RNA >1000 c/mL) and who had at least 2 NRTIs available [Slater L, et al., Abstract LB-15]. By a modified intent-to-treat analysis, viral load reduction was greater in the TPV/RTV arms than in the SQV/RTV arm (-1.41 vs -0.87 log10 c/mL). Although the data were not presented orally, the abstract indicated that the proportion of patients who achieved viral loads <400 c/mL was somewhat higher in the high-dose TPV/RTV arm (55%) than in the low dose or SQV/RTV arms (39% and 40% respectively). GI toxicity was more common in the high dose TPV/RTV arm. Of note, 42% of the patients in this protocol had no demonstrable PI resistance despite experiencing virologic failure on a PI-based regimen. Clearly, the real test of this agent will be in patients with more extensive PI resistance than was seen in this trial. Dosing issues also need to be worked out: TPV/RTV doses currently being studied are 500/100, 500/200, and 750/200 mg bid.
• Atazanavir (Zrivada^TM, Bristol-Myers Squibb) will most likely be the next protease inhibitor approved by the FDA. It is administered once daily, and appears to have little if any adverse effect on lipid profiles. Sanne presented results from BMS 008, in which naïve patients (mean baseline viral load ~4.7 log10 c/mL) were randomized to receive atazanavir (TAZ, 400 or 600 mg qd) or nelfinavir (NFV, 1250 mg bid) plus d4T/3TC. By intent-to-treat analysis (last observation carried forward), the proportion of patients in the 3 arms with viral load <400 c/mL at 48 weeks was 65%, 62%, and 59% respectively. The proportion with viral load <50 c/mL was 31%, 36%, and 38%. By as-treated analysis, the proportion with viral load <400 c/mL was 74%, 75%, and 60%, respectively (p<0.05 for TAZ vs NFV), and the proportion with viral load <50 c/mL was 40%, 41%, and 39% (NS).Increased cholesterol was seen in 5-7% of patients randomized to receive TAZ compared to 20-25% of those taking NFV, and diarrhea was less common among TAZ recipients. Indirect hyperbilirubinemia, including clinically apparent jaundice, was more common in patients in the 600 mg arm, and further development will proceed with the 400 mg qd dose.
      Like ritonavir, TAZ increases levels of other protease inhibitors through its inhibitory effect on the CYP 3A4 enzyme system. The possibility of using TAZ to boost levels of other PIs is attractive, since it is less likely to cause hyperlipidemia than ritonavir. David Haas presented results of BMS 009, in which patients experiencing virologic failure on HAART (HIV RNA 2,000-100,000 c/mL) received TAZ/SQV at a dose of 400/1200 mg qd (n=34) or 600/1200 mg qd (n=28) or RTV/SQV at a dose of 400/400 mg bid (n=23) [Abstract LB-16]. Patients in each arm experienced a 1-1.5 log drop in viral load at 24 weeks, but those on the TAZ-based regimens had more favorable lipid profiles.
• Emtricitabine (FTC, Coviracil^TM, Triangle Pharmaceuticals) is a nucleoside analog with a resistance pattern that is essentially identical to that of 3TC. Its long-term safety profile also appears to be comparable to that of 3TC [Benson C, et al. Abstract I-1931], and patients who switched from 3TC to FTC had equivalent virologic results compared to those who remained on 3TC [Van der Horst C, et al. Abstract I-1932]. Although it doesn't appear to have any obvious advantages over 3TC (both have pharmacokinetics that support once-daily dosing), it might have a niche if it were to be co-formulated with other once-daily agents.

Return to top

Boosted PI Regimens

• SQV/RTV: Julio Montaner presented data from the FOCUS trial, an open-label trial comparing once daily SQV/RTV (1600/100 mg) vs EFV, both in combination with 2 nucleoside analogs. The trial involved 161 naïve patients. EFV was clearly better tolerated: Eight patients discontinued therapy in the SQV/RTV arm vs one in the EFV arm. Nausea was seen in 22% vs 1%, and vomiting in 6% vs 0%. As a result, EFV was superior by intent-to-treat analyses, with 81% achieving viral loads <50 c/mL compared to 60% in the SQV/RTV arm (p=0.008). By as-treated analysis, the results were 90% and 81% (p=0.224). It was postulated that the hard-gel formulation of SQV (Invirase), which achieves similar drug concentrations when combined with RTV, might be better tolerated than the soft-gel formulation (Fortovase). Interestingly, there was no difference in potency in either arm between patients with baseline viral loads <100,000 c/mL or those with >100,000 c/mL, a finding that has previously been demonstrated with EFV and LPV/RTV (Kaletra), but not with SQV/RTV.
• IDV/RTV: Young presented data on 89 naïve patients treated with IDV/RTV (800/100 mg bid) plus d4T/3TC [Abstract I-1923]. By an intent-to-treat, non-completer=failure analysis, viral load was <50 c/mL in 42% of patients at 24 weeks, due in part to a high non-completion rate. Approximately half withdrew consent, and half had side effects, including a high incidence of nephrolithiasis. Katner presented data on 63 treatment experienced patients who received IDV/RTV at a dose of 800/200 mg bid in combination with 2 NRTIs [Abstract I-1922]. By as-treated analysis, 78% had viral loads <400 c/mL and 54% had VL <50 c/mL at 24 weeks.
• APV/RTV: A comparison of two APV/RTV regimens (600/100 and 900/ 100 mg bid) suggested that the lower dose is preferred because of similar efficacy and APV concentrations with decreased toxicity and discontinuation [Schooley R, et al. Abstract I-1924].
• LPV/RTV: Danner presented data assessing response rates to LPV/RTV among 57 treatment-experienced patients treated with standard dose LPV/RTV plus EFV and NRTIs [Abstract I-1925]. He found that patients with less than a 10-fold reduction in phenotypic susceptibility to LPV had a 93% response rate (<400 c/mL) at 72 weeks, and those with up to 5 PI resistance mutations had a 91% response. After 14 days, patients were randomized to increase the dose of LPV/RTV to 533/133 mg bid or to remain on the standard dose. It is now recommended that the higher dose be used in combination with EFV or NVP, especially in experienced patients where reduced susceptibility to LPV is expected.

Return to top

Safety of Antiretroviral Therapy During Pregnancy

• Reassuring data from the CDC's Antiretroviral Pregnancy Registry were presented by Garcia [Abstract I-1325]. In an analysis of 1630 evaluable cases, the prevalence of birth defects after first trimester exposures was 12 per 638 live births (1.9%). Results were similar for AZT or 3TC, given either alone or in combination with other agents. This is similar to what is seen in the general population, suggesting that there is no significant increase in birth defects with first trimester exposure to antiretroviral agents overall or for AZT or 3TC in particular.

20020101
JH20020101

©1997,1998,1999,2000,2001, 2002. The Johns Hopkins University AIDS Service, Division of Infectious Diseases. Permission to use and reproduce portions of this newsletter is hereby granted provided that author and publication are fully credited and both copyright and permission notice appear with reprinted material. Inquiries may be directed to Sharon McAvinue, Managing Editor. Website: Johns Hopkins AIDS Service.

AEGiS is made possible through unrestricted grants from Boehringer Ingelheim, iMetrikus, Inc., the National Library of Medicine, and donations from users like you. Always watch for outdated information. This article first appeared in 2002. This material is designed to support, not replace, the relationship that exists between you and your doctor.

AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1980, 2002. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content.