Note: This article contains discussion of off-label use of some drugs; not all have been approved for that use by the FDA.

The Conference on Retroviruses and Opportunistic Infections (CROI), despite its awkward and outmoded name, has become the most important venue for the dissemination of results from HIV-related research in the U.S. and, arguably, the world.  The 10th CROI held in Boston on February 10-14 was more clinically oriented than most of the preceding conferences and saw results from several important clinical trials presented, as well as preliminary data from studies of new antiretroviral agents and updates on the epidemiology of the epidemic.

Clinical Trial Results
2NN Study
A presentation of the large Boehringer Ingleheim-sponsored 2NN Study1 by International AIDS Society President Joep Lange was of key interest to many attending the conference. These long-awaited results detailed the efficacy and safety of initial antiretroviral therapy with stavudine (d4T) and lamivudine (3TC), plus one of the following: 

• nevirapine (NVP) 400 mg QD, 
• NVP 200 mg BID,
• efavirenz (EFV) 600 QD, or 
• a combination of NVP 400 mg QD + EFV 800 mg QD

A total of 1,216 subjects enrolled in 17 countries. All subjects were treatment naïve; median HIV viral load and CD4+ cell count were 4.7 log10 copies/mL (~50,000 copies/mL) and 109/uL, respectively. Treatment failure was defined as less than a 1 log10 decline in HIV RNA level by week 12, two consecutive viral load measurements above 50 copies/mL after week 24, a new Centers for Disease Control and Prevention (CDC) category C diagnosis or death, or change of study-assigned therapy. 

About 84% of subjects completed the study. After 48 weeks, success (absence of above-defined failure) was fairly comparable across arms with 56% of the NVP QD arm, 56% of the NVP BID arm, 62% of the EFV arm and 47% of the NVP+EFV arm achieving this endpoint. A statistically significant difference was only seen between the EFV and the NVP+EFV arms. A change in treatment occurred most commonly in the dual Non-Nucleoside Reverse Transcriptase Inhibitor (NNRTI) assigned subjects (34.5%), followed by NVP QD (29%) and less commonly NVP BID (22%) and EFV (20%). In an intent to treat analysis, the proportion of subjects with a viral load below 50 copies/mL at week 48 was 70% in both the NVP QD and the EFV arms, 65% in the NVP BID arm and 63% in the dual NNRTI arm. Response to treatment based on baseline viral load demonstrated a decline in treatment success rates in all the study arms when the HIV RNA level was >100,000 copies/mL at entry with a suggestion that the EFV arm tended to be superior compared to the NVP arms. CD4+ cell counts increased in all the groups not dissimilarly with approximately a 140-150 cell/uL rise experienced from baseline to week 48. 

While virologic and immunologic efficacy was not wildly different among arms, adverse events did distinguish the study regimens. Overall, there was more toxicity reported in the NVP containing arms than the EFV arm. The proportion of subjects who had treatment limiting adverse events in the four arms ranged from 30% in the dual NNRTI arm to 15.5% in the EFV QD arm; 24% of the NVP QD and 21% of the NVP BID discontinued therapy during the study. Hepatotoxicity comprised the lion's share of grade 3 (severe) and grade 4 (life threatening) laboratory toxicity with 13.2% of the NVP QD, 7.8% of the NVP BID, 4.5% of the EFV QD and 8.6% of the dual NNRTI subjects experiencing elevations of ALT and/or AST (p<0.001 NVP QD vs. EFV QD). About 10% of the cohort was HCV seropositive and 5% had active HBV infection and these patients were distributed equally across the study arms. CNS adverse events were not surprisingly significantly higher among those taking EFV (5.5% had grade 3 or 4 CNS toxicity in this arm). 

During the trial, 25 people died. Eleven of these deaths were associated with HIV disease and another 11 from non-HIV or study treatment related causes (including a number of murders occurring in South Africa). In the remaining three cases, death was attributed to study medication: one died from lactic acidosis associated with d4T use, one woman developed fatal toxic hepatitis while on NVP without evidence of viral hepatitis and another man also receiving NVP developed Stevens-Johnson Syndrome and subsequent fulminate sepsis. 

This long-awaited study demonstrated parity in potency between NVP and EFV and the feasibility of once daily dosing with NVP. The higher rates of serious toxicity, specifically hepatotoxicity, in the NVP arms may dampen some clinicians' enthusiasm for once a day NVP, although others may opt for this convenient alternative to EFV coupled with close monitoring of liver functions.

903 Study
Another study with implications for clinical care, the Gilead 903 Study2, pitted tenofovir (TDF) against d4T as part of a three-drug regimen with 3TC and EFV in 600 antiretroviral naïve patients. In a presentation of 96-week results from this trial, the extremely high rates of viral suppression reported after 48 weeks have persisted. In both study arms the proportion of subjects with a viral load less than 50 copies/mL was between 74% and 78% - a response that is all the more impressive considering the median baseline viral load was 81,000 copies/mL and the median entry CD4+ cell count 276/mm3. 

However, as in the 2NN Study, it is the toxicities that set the regimens apart. Fasting triglyceride levels rose significantly higher in the d4T+3TC+EFV arm compared to those assigned to TDF+3TC+EFV (100 mg/dL vs. 5 mg/dL, p=0.001). Likewise, total cholesterol increased 50 mg/dL in the d4T group and 30 mg/dL in the TDF group. Reflecting these changes in lipids and indicating their clinical significance, rates of new statin and fibrate use during the study was much higher in the d4T subjects than the TDF patients (10% vs. 2%). Other metabolic complications were also compared. Investigator-defined lipodystrophy was reported in four times as many d4T subjects. Differences between limb fat at 96 weeks by DEXA scanning were also seen. Those assigned TDF had about eight pounds more peripheral fat than d4T subjects. DEXAs were also obtained longitudinally to assess bone density. These results were released after the conference and demonstrated slight declines in bone mineralization in both study arms. However, although the differences between arms were statistically significant, thus far, they are of limited clinical significance. 

This important study indicates that tenofovir can hold its own against d4T in suppressing viral replication and does so with less toxicity. That the study compared two regimens that can both now be administered once daily was prescient - and certainly increases the attractiveness of the TDF regimen within corrections. 

Pegylated Interferon for Early-Stage HIV
Researchers from Germany presented results of a study3 that treated early-stage HIV-positive patients with weekly injections of pegylated interferon Alfa-2b (PegIntron). Ten patients with early-stage HIV and no symptoms who had not yet received treatment were studied; five patients received weekly PegIntron injections while the other five patients received no treatment. 

At baseline the average CD4 count in the group that received therapy was 462; the control group CD4 count averaged 535. After six months, the average CD4 count in the treatment group had increased to 611 while the average CD4 in the control group dropped to 450. Viral load in the group receiving therapy dropped from 22,158 copies/mL to 3,039 copies/mL. The viral load in the control group went from 7,136 copies/mL to 40,092 copies/mL.

The presenters reported no serious side effects in the treatment group, perhaps due to the fact that the PegIntron dose was half the standard dose used for hepatitis C therapy.

This study of a small number of patients offers some intriguing data on the favorable immunologic and virologic effects of interferon therapy. Further studies are needed to assess the longer-term effects and toxicities when using interferon to treat patients with early-stage HIV infection.

Drug-Drug Interactions
Tenofovir and ddI
In addition to results of large clinical trials, several reports focused on important drug interactions of the agents currently used to treat HIV infection. As many providers know, co-administration of tenofovir and didanosine (ddI) leads to elevated plasma levels of ddI and, therefore, potential heightened risk of ddI-related toxicity. In contrast, ddI does not alter tenofovir levels. In response to the effect of tenofovir on ddI levels, clinicians have been using a reduced dose of 250 mg of ddI-EC (enteric coated) along with tenofovir. 

The pharmacokinetics of this dose of ddI-EC when coupled with tenofovir in a fed and fasted state was studied4 in HIV-uninfected volunteers. Dosing of 250 mg ddI-EC on an empty stomach followed two hours later by 300 mg tenofovir and a light meal resulted in a ddI exposure identical to that seen when ddI-EC is administered as 400 mg by itself. Concomitant administration of 250 mg ddI-EC and tenofovir without food led to a slightly lower ddI exposure relative to 400 mg of ddI-EC (88.6% geometric mean exposure) but when the two were taken with food, ddI levels increased by 114%. Generally, these data indicate that a dose reduction of enteric-coated ddI to 250 mg is appropriate when this agent is used with tenofovir however, it should be noted that all subjects weighed over 60 kg. The differences in exposure in the various scenarios tested were not major and the convenience of being able to take both medications with a light meal has led many to recommend this dosing schedule to their patients. Appropriate dosing in those who weigh less than 60 kg is unclear and certainly requires further study.

Lopinavir/ritonavir and Phenytoin
Kashuba and colleagues examined a potential drug-drug interaction between lopinavir/ritonavir and the anticonvulsant phenytoin (Dilantin)5. Lopinavir/ritonavir is an inhibitor of CYP3A4 and, like phenytoin, a CYP450 inducer. Phenytoin is a substrate for CYP2C9 and CYP2C19. To investigate the interactions between these agents, 24 HIV-uninfected volunteers were randomized to two groups, the first of which received lopinavir/ritonavir 400 mg BID for 10 days followed by concomitant administration of the PI with 300 mg QD of phenytoin for 12 more days. The other arm started the study taking phenytoin alone and then added lopinavir/ritonavir at day 11 through day 22.  Intensive pharmacokinetics were performed and revealed a two-way interaction between these drugs in which levels of both are decreased. Clinically, this may mean that phenytoin levels may drop if lopinavir/ritonavir is added; therefore levels of the anticonvulsant should be checked soon after the PI is added. Levels of lopinavir may also drop despite the low dose ritonavir present in the formulation. Whether this reduced exposure can be compensated for by the addition of more lopinavir/ritonavir is unclear as this could lead to even greater reduction in phenytoin.  Follow-up studies of the complex interactions between these drugs and possible regimen adjustments to compensate are planned.

The interactions between phenytoin and lopinavir/ritonavir are complex enough that they should be not be co-administered in the same patient.

Metabolic Complications
The centerpiece of the session on metabolic complications associated with HIV was the D:A:D Study6, a large meta-cohort study of cardiovascular disease involving 23,468 subjects from the U.S., Europe and Australia.  At baseline, participants had a median age of 39 years, 24% were women, 55% had viral loads below the limits of assay detection, 67% were PI experienced, 34% NNRTI experienced, 20-30% had elevated total cholesterol or triglycerides and 60% were current smokers. During the study, 126 subjects had probable or definite myocardial infarctions (MIs), 28% of which were fatal. Men experienced all but 10% of the MIs. In an analysis of exposure to highly active antiretroviral therapy (combination treatment which includes a PI and/or an NNRTI plus NRTIs) and outcome of MI, there was a linear trend of increased risk of MI with greater exposure to HAART. The relative risk of MI increased by 1.26 (95% CI, 1.12-1.41) per year of HAART exposure (i.e. a 26% increased annual risk of MI). Other factors that placed participants at risk for MI included increased age, male gender, previous cardiovascular disease, smoking and elevated total cholesterol. Enigmatically, the presence of clinician-defined lipodystrophy was associated with a decrease in MI risk. 

This important study is one of the first well-designed studies to demonstrate an increased risk of cardiovascular disease among persons treated with HIV therapies. Despite this risk, it remains important to note that the overall life-sustaining benefits of combination HIV treatment vastly outweigh the risks of therapy defined in this study. Measures to reduce other cardiovascular disease risk factors such as smoking and treatment of dyslipidemia in antiretroviral treated patients can tilt the balance even further in favor of HIV therapy.

Epidemiology
New data from the CDC presented in Boston7 indicate that the number of AIDS cases nationwide has increased for the first time in several years. In addition, the number of new HIV infections in 25 reporting states increased by 8% from 1999 to 2001. The greatest increases in new HIV diagnoses were seen in both men who have sex with men and heterosexuals. These are disturbing data and suggest that HIV prevention measures are falling far short of the mark. 

Further exacerbating the spread of the epidemic is the failure to identify a significant proportion of persons with living with HIV infection. It is estimated that as many as a quarter of all HIV-infected persons in the U.S. do not know that they are HIV positive. Enhanced HIV testing in clinics and correctional facilities can allow for the diagnosis, counseling and treatment of persons who are infected. In addition, interventions aimed at transmission risk reduction can be most effective when applied to those who are aware of their HIV serostatus. Appropriate diagnosis and treatment of infected individuals can reduce viral burden in the blood and genital secretions, likely reducing transmissibility. 

Conclusion
At a time where every several weeks it seems there is another "major" HIV-related conference, when there are dedicated meetings for everything from antiretroviral resistance to metabolic complications to treatment adherence, the 10th CROI proved to be impressively comprehensive. There were few data specific to the confluence of HIV and corrections, but information regarding strategies to optimize treatment success and reduce drug toxicity, and trends in the epidemiology of the epidemic will serve physicians who treat HIV-infected patients in jails and prisons well. While no major breakthroughs were reported and the news was not always good, the data presented clearly inched our understanding of HIV and its management forward - and, of course, there is always next year. 

*Disclosures: Speaker's Bureau: GlaxoSmithKline, Gilead, Merck, Roche, and Abbott. Grant support from Roche.
FOOTNOTES:
1. Results of the 2NN Study: A Randomized Comparative Trial of First-line Antiretroviral Therapy with Regimens Containing Either Nevirapine Alone, Efavirenz Alone or Both Drugs Combined, Together with Stavudine and Lamivudine. F van Leth, E Hassink, P Phanuphak, S Miller, B Gazzard, P Cahn, R Wood, K Squires, C Katlama, B Santos, P Robinson, R van Leeuwen, F Wit, J Lange, for the 2NN Study Group. Oral Abstract 176. 10th CROI, February 10-14, 2003.
2. Efficacy and Safety of Tenofovir DF (TDF) Versus Stavudine (d4t) When Used in Combination with Lamivudine and Efavirenz in Antiretroviral Naïve Patients: 96-week Preliminary Interim Results. S Staszewski, J E Gallant, A L Pozniak, J M A H Suleiman, E DeJesus, B Lu, J Sayre, A Cheng. Poster 564b. 10th CROI, February 10-14, 2003.
3. Pegylated Interferon Alfa-2b: A New Therapeutic Option in the Treatment of Early-stage HIV Infection. I Schugt, A Kreuter, R Schlottmann, A Bader, P Altmeyer, N H Brockmeyer. Oral Abstract 59. 10th CROI, February 10-14, 2003.
4. Didanosine and Tenofovir DF Drug-Drug Interaction: Assessment of Didanosine Dose Reduction. B P Kearney, E Isaacson, J Sayre, H Namini, A Cheng. Poster 533. 10th CROI, February 10-14, 2003.
5. A Two-way Drug Interaction Between Lopinavir/Ritonavir and Phenytoin. M L Lim, S S Min, J J Eron, R Bertz, M Robinson, A Gaedigk, A D M Kashuba. Poster 535. 10th CROI, February 10-14, 2003.
6. Exposure to HAART Is Associated with an Increased Risk of Myocardial Infarction: The D:A:D Study. N Friis-Moller, R Weber, A D'Arminio Monforte, W El-Sadr, P Reiss, F Dabis, L Morfeldt, S De Wit, C Pradier, G Calvo, M Law, O Kirk, C Sabin, J D Lundgren. Oral Abstract 130. 10th CROI, February 10-14, 2003.
7. Preventing New HIV Infections in the U.S.: What Can We Hope to Achieve? R Valdiserri. Plenary Lecture 4. 10th CROI, February 10-14, 2003.

RETURN TO THE TOP OF PAGE

ABOUT HEPP REPORT HEPP News is published ten times a year by the: HIV and Hepatitis Education Prison Project Brown University Box G-B4 Providence, Rhode Island 02912 401-277-3651 fax 401-277-3656 heppreport@brown.edu