Section written by Graeme Moyle MD, MBBS, with selected contributions by Harvey S. Bartnof, MD, Jules Levin, David Margolis, MD, Nancy Shulman, MD, and David Alain Wohl, MD.

New drugs presented at the 8th Annual Retrovirus Conference hold much promise in both the setting of ‘salvage’ (detectable viral load with current anti-HIV regimen) and treatment simplification. However, questions remain as to whether these agents will deliver in terms of pharmacology (active drug concentrations in various bodily compartments), their efficacy against strains of virus resistant to current agents and whether side effects will be fewer.

Presentations at the Conference suggest a shift in interest towards immune-based therapies as additional components of HIV therapy. This shift was led by interest in both immune modulators like interleukin-2 (IL-2, Proleukin) and potential vaccine approaches. While in the 1980s the key marker used in managing HIV was the absolute CD4 cell count, in the ‘90s the treatment mantra was the "patronizing squeal" from virologists of ‘It’s the virus, stupid.’ In this decade, we may be seeing a more balanced approach; making the CD4 count go up and viral load go down is not enough. Potentially restoring the immune system, and controlling HIV without the need for sustained drug therapy, could be the achievement for this decade.

Several new NRTI, NNRTI and PI drugs were reported, mostly pre-clinical or early clinical. The main issue for drug developers in both NNRTI and PI classes is the extent of cross-resistance that exist between the available drugs of these classes, such that when resistance to one agent has developed it can render an entire class of inhibitors ineffective. Two approaches may be used to manage this problem: using boosted drug levels (particularly trough or lowest levels) to overcome low level resistance (mild decreased sensitivity) such as with the PI drug combination lopinavir/ritonavir (Kaletra) or low-dose ("baby-dose") ritonavir with saquinavir (Fortovase), indinavir (Crixivan) or amprenavir (Agenerase). The other choice would be to develop new agents that do not show cross-resistance to the current ones. Examples include the PI drug tipranavir and TMC-126 or possibly the NNRTI drugs capravirine, TMC-120 and DPC-083. Ideally, new agents in these classes would possess both high potency against mutant variants of HIV and show higher levels of free drug in blood serum (liquid portion) throughout the dosing period, but safety and tolerability are also needed.

DPC 681 & DPC 684:
Effective for Protease Inhibitor-Resistant Virus
Dr. Susan Erickson-Viitanen of DuPont discussed ‘second-generation’ PI drugs in pre-clinical development, including in vitro (laboratory) results of DPC 681 and DPC 684 (abstract 11). The agents have a chemical design ("substituted sulfonamides") that bears some resemblance to amprenavir (Agenerase, PI drug). These new compounds are potent inhibitors of HIV protease enzyme with 90% "inhibitory concentrations" (IC90) for wild-type HIV of 4-8 nM (nanomolar, similar to lopinavir and up to 10-times more potent than indinavir). As with currently available PI drugs, DPC 681 and DPC 684 showed no loss in potency against mutant virus with the "D30N" nelfinavir (Viracept) mutation and maximal 5.6-fold loss in potency towards variants with up to 5 mutations. (Greater than 4-fold shifts are often considered relevant to clinical drug resistance.) Viruses derived from patients who have failed PI-containing regimens and having 6-7 mutations (including positions "10, 63, 71, 82, 84 and 90") showed only a 3- to 6-fold loss in sensitivity (resistance) to DPC 681 or DPC 684. In contrast, these mutant variants showed a 13- to 190-fold loss in sensitivity to currently marketed PI drugs. Potent exposures of these agents should be achievable in humans. Following the successful completion of safety and pharmacology studies in animals, Phase I clinical studies of these promising agents in healthy volunteers, are now underway.

TMC126:
High Potency Against HIV Highly-Resistant to PI Drugs
Dr. John Erickson of Tibotec in Rockville, Maryland presented data on two potentially exciting PI compounds (abstract 12). Interestingly, their methodology was to pursue a "structure- and fitness-based strategy" using computer-based drug design, in combination with drug resistance profiling, to discover PI drugs that possess not only high potency and activity against viral mutants resistant to current PI drugs, but also a so-called ‘resistance repellent’ design. The intent of this approach is to design a drug that retains the ability to limit or ‘suppress’ the evolution of drug-resistant variants along genetic pathways commonly observed with the approved PI drugs. Also, such a drug would still interact with HIV protease enzyme at highly conserved points that would lead to dramatic impact on viral fitness if mutated. The lead compound in their stable, called TMC126, shows previously unreported high levels of potency. When directly tested against the protease enzyme, this and related drugs possess the ability to inhibit wild type and multi-drug resistant clinical isolates of HIV with IC50 values that represent an extremely low concentration. The drug also was shown to establish barriers to the emergence of common active site mutations by limiting enzyme "fitness," thereby resulting in resistant strains emerging more slowly and along new genetic pathways. Pharmacological properties of this agent were not described, but TMC126 and several related compounds in this series are being evaluated for suitability as candidates for clinical development with at least one agent having decent bioavailability in dogs.

BMS-232632:
Once-a-Day PI Drug
The PI drug in most advanced development is Bristol Myers Squibb’s BMS-232632 (abstract 15). This agent is a potent PI drug with significant activity in vitro (laboratory testing) and a pharmacokinetic (drug distribution, metabolism) profile that supports once daily dosing (without "ritonavir boosting’) and with a low tablet load. Drug absorption is improved by food. Interestingly, so far no increases in cholesterol & triglycerides have been seen. Kathleen Squires, MD of the University of Southern California presented the results of a dose-finding study in treatment-naïve (no previous treatment) patients with baseline viral loads of at least 2000 copies per milliliter. This randomized study compared the safety and anti-HIV activity of 3 doses of BMS-232632 once daily with nelfinavir (NFV, Viracept, PI drug) 750 mg 3-times daily both as monotherapy for 2 weeks, followed by the addition of didanosine (ddI, Videx, NRTI drug) and stavudine (d4T, Zerit, NRTI drug). Of note, since both NFV and BMS-232623 are dosed with food and ddI is dosed on an empty stomach, the regimens studied are relatively complicated. (A newer formulation of ddI under development may avoid this problem; see below.) The study included 92 patients initially (through 48 weeks), with another 166 patients added after satisfactory preliminary results (results through 24 weeks).

The results at week 24 indicated that 65-68% of BMS-232632-treated patients and 63% of NFV-treated patients achieved an undetectable viral load (limit 400 copies per milliliter). CD4 cell count increases were similar in both treatment groups. The most common adverse events that were possibly drug-related included diarrhea (17% for BMS-232632 vs. 51% NFV) and nausea (13% BMS-232623, 7% NFV). An increased bilirubin (bile pigment) level usually without obvious jaundice (yellow coloration of skin) or other symptoms and not associated with other abnormal liver tests was seen with BMS-232623. This occurred in 28% of patients taking 400 mg daily, and more often among those taking 500 mg daily, the highest tested dose. (This occurrence is similar to that which occurs with indinavir [Crixivan], PI drug.) No significant increases in cholesterol, LDL- (low density lipoprotein or "bad") cholesterol, or triglycerides (fats) were observed in the BMS arm, whereas in the NFV arm, a sustained increase of 20-50 mg per deciliter was observed.

This agent in now progressing to further studies in a range of clinical settings at the 400 mg once-daily dose and in comparison to current standard of care regimens. In another presentation, the once daily combination of BMS-232632 200-400 mg with ritonavir (RTV, Norvir) 100-200 mg "boosting" significantly increased blood concentrations of BMS, was safe and well-tolerated, according to Edward O’Mara, MD of Bristol Myers Squibb (abstract 740). "Icterus" (yellowing of the eye whites) occurred in 22%, was associated with increases in blood bilirubin (bile pigment) levels without liver enzyme changes and resolved after the drug was discontinued.

GW433908:
Amprenavir PI Pro-Drug
Another experimental PI drug presented at the Conference was GW433908 (GlaxoSmithKline and Vertex), a pro-drug of amprenavir (APV, Agenerase) (abstract 333). Dr. R. Wood of Somerset Hospital in Capetown, South Africa reported a 42-day "cross-over" study of 85 treatment-naïve (no previous treatment) patients randomized to take 3-4 tablets twice daily of GW433908 (465 mg tablets) or APV 1,200 mg twice daily (8 pills bid). The dosing of the APV prodrug is 2 pills (new formulation) twice daily in phase 3 studies and for FDA approval. The current study of "908" once daily in combination with 200 mg RTV uses 2 908 pills (equivalent to 600 mg of APV per pill). 908 plus RTV once and twice daily is also being studied versus Kaletra in patients with protease inhibitor resistance (and NNRTI experienced), as APV has a unique resistance profile. The 908 + RTV twice daily dosing regimen being studied is the equivalent of 600mg APV + 100mg RTV. In the study described in abstract 333, all patients also took twice daily lamivudine (3TC, Epivir) and abacavir (ABC, Ziagen), both NRTI drugs. The results revealed a near equivalent 2 log (100-fold) copies per milliliter decrease in the HIV viral load (VL) in all 3 arms, with an approximate 100 cells per microliter increase in the CD4 count, also in all 3 arms. The percentages of VL undetectability were not presented. Blood APV levels in all 3 arms were quite similar with an equal total exposure ("area-under-the-curve") and maximal concentrations that were 18-28% lower in the prodrug arms. It's hoped that lower maximal concentration and reduced pill count will translate into reduced GI side affects associated with the current formulation of APV. Minimal or trough concentrations were not presented. Adverse events were similar in the 3 treatment arms, with the GW433908 arms having less nausea, diarrhea, abdominal (stomach area) pain and "gas" symptoms, but with somewhat more headache and sleep problems. There were no premature discontinuations due to adverse events associated with study drugs in all 3 arms.

TMC120:
for NNRTI Drug-Resistant HIV
Dr. R. Pauwels of Tibotec discussed a new NNRTI drug under development (abstracts 13, 304). The agent is TMC120 (also called R147681) and is distantly related to some previously discontinued NNRTI drugs, but unrelated to approved agents. It’s in vitro activity reveals a potent IC50 (1-10 nanomolar) against wild-type HIV, similar to efavirenz (Sustiva, NNRTI drug). However, activity against HIV that is resistant to the approved NNRTI drugs (efavirenz, nevirapine and delavirdine, including those with single mutations "K103N, Y181C or G190A/S") is not diminished. However, modest, greater than 5-fold shifts in sensitivity (resistance) were seen with infrequently-observed mutations "100I and 188L." This drug has now completed a first clinical evaluation and is proceeding with further activity and safety studies.

The phase II study was randomized and double-blind (drug or placebo unknown by patient and treating physician). It compared two doses of TMC120 (50 or 100 mg, twice daily) as monotherapy to placebo for 7 days in treatment-naïve (no previous treatment) HIV positive patients with a baseline viral load between 3.7-5.1 log (5,000 -125,000) copies per milliliter. Triple therapy with FDA-approved ART was offered thereafter. The study was conducted in Russia and Poland and included 43 patients (21% women). The median baseline CD4 cell count was 571 cells per microliter, and the viral load was 4.6 log (40,258) copies per milliliter. The results showed 8-day decreases in viral load by treatment group (placebo, TMC 50 mg, and 100 mg) in log copies per milliliter were –0.17, -1.44 and –1.55, respectively. No resistance has been detected in samples at the end of therapy. Pharmacokinetic (drug concentration) results suggested drug "exposure" at trough (lowest level) up to 50-times greater than the "protein-corrected" in vitro IC50. TMC120 was well tolerated with 3 patients reporting possible drug-related adverse events: mild somnolence (sleepiness) or insomnia (difficulty sleeping). One subject on the 50 mg arm withdrew from study after 4 days because of an HIV-related problem unrelated to study drug. Since the antiviral effects of the two doses of TMC120 were not significantly different, additional studies will be required to determine which dose will be used in larger studies.

Capravirine:
NNRTI Drug
The phase II results of capravirine (CPV, AG1549, NNRTI drug, Agouron) combination therapy were presented by Dr. P. Wolfe of Pacific Oaks Medical Group in Beverly Hills, California (abstract 323). CPV retains activity against the common "K103N" NNRTI drug mutation that leads to resistance among the 3 marketed NNRTI drugs. The current study was double-blind (drug or placebo not known by physician or patient) and placebo-controlled. A total of 50 patients with virologic failure associated with an NNRTI drug regimen were randomized to receive placebo (inactive drug) or CPV at a dose of either 1,400 or 2,100 mg twice daily, in conjunction with nelfinavir (NFV, Viracept, PI drug) 1,250 mg twice daily and 2 NRTI drugs. All patients were naïve to (never took) PI drug therapy. The median baseline HIV viral load (VL) was 4.2 log (13,803) copies per milliliter, with a CD4 count of 362 cells per microliter. The interim results at 12 weeks showed that more than half of patients in the CPV arms achieved HIV VL undetectability (limit 400 copies per milliliter). The most common moderate or worse adverse events were nausea, vomiting and diarrhea that were associated with a 14% discontinuation rate in the CPV arms. Further development of CPV is on hold due to "vasculitis" (blood vessel inflammation) in a dog study.

DPC-083:
NNRTI Drug
Although no updates were reported on this new NNRTI drug at the Retrovirus Conference, DPC-083 from DuPont is in phase II development. In preliminary studies reported at last year’s 13th International AIDS Conference in Durban, South Africa, it has shown activity against selected isolates with NNRTI drug resistance. It has a half-life (amount of time until half is remaining) of 140 hours, but is expected to be dosed once daily. It appears to be more potent than EFV.

Emtricitabine (FTC):
New NRTI Drug
There remains a need for new NRTI drugs both because of concerns regarding the long-term toxicity of some drugs in this class and the extent of cross-resistance that is increasingly appreciated between different FDA-approved members of this class.

Data on emtricitabine (FTC, Coviracil) were presented by Charles Van Der Horst, MD of the University of North Carolina in Chapel Hill (abstract 18). FTC is an NRTI drug of similar chemical structure to 3TC (lamivudine, Epivir, NRTI drug), but with in vitro potency against HIV that is 4-10 times greater. This drug is under development by Triangle and Abbott Pharmaceuticals. It can be dosed once daily, but, as with 3TC, it selects for the "M184V" resistance mutation. Also like 3TC, it is active against hepatitis B virus (HBV). An initial 2-week monotherapy study with FTC showed an approximately 2 log (100-fold) copies per milliliter median decrease in viral load with a dose of 200 mg once daily. Data from two randomized, controlled "equivalence" trials of 150 mg 3TC twice daily compared with 200 mg FTC once daily, in triple therapy combination regimens, were reported. In study FTC-303, 440 patients with viral loads no greater than 400 copies per milliliter on a 3TC-containing triple therapy regimen for at least 12 weeks were randomized (2:1) in an open-label manner to switch to FTC or continue on 3TC. The results showed that 8% of the patients in each arm have experienced viral rebound greater than 400 copies per milliliter to date, with no differences through 48 weeks between drugs and few drug-related adverse events. John Bartlett, MD of Duke University in Durham, North Carolina presented another study with FTC (abstract 19). FTC-302 was randomized and double blind (drug name not known by patient or treating physician) with 468 South African patients (59% women and 87% Black) who were treatment-naïve (no previous treatment). The study compared FTC to 3TC plus stavudine (d4T, Zerit, NRTI drug) and either nevirapine (NVP, Viramune, NNRTI drug if the screening viral load was no greater than 100,000 copies per milliliter) or efavirenz (EFV, Sustiva, NNRTI drug, if the screening viral load was greater than that). The results showed that viral rebound occurred in 12% of FTC and 6% of 3TC patients, a significant difference. However, adherence may be the reason, since 60% of rebounders on FTC (vs. 23% on 3TC) had "wild-type" (no mutations) virus and only 15% of those taking FTC had the "184V" mutation compared to 54% of 3TC rebounders, a significant difference. (Drug levels to confirm decreased adherence were not reported.) Adverse events were predominantly mild-to-moderate in both groups. Of note, all 58 patients in this study who experienced grade 3 or grade 4 (severe or life-threatening) elevations in liver enzymes were taking NVP. For more detailed information about liver toxicity in this study, see page 20.

These studies support equivalent antiviral activity and safety of once-daily FTC compared to twice daily 3TC in HIV-infected patients initiating therapy or switching from 3TC to FTC with a baseline HIV RNA no greater 400 copies per milliliter. However, given that other data suggest that 3TC may also be given once daily (due to a 15-hour long "half-life" of its active "tri-phosphate" form in cells) and that adverse events clearly related to 3TC are rare, the place of FTC in anti-HIV treatment remains to be determined. (A presentation by Dr. M. Sension at this Conference reports on patients successfully switching to 3TC once-daily after being undetectable on 3TC twice-daily regimen, see page 14, abstract 317.) Another clinical study using FTC in an anti-HIV regimen that is entirely once daily is reported on page 14.

Other NRTI Drugs for Resistance:
DAPD, L-Fd4C, BCH-10618
Other experimental NRTI drugs or compounds presented at the Retrovirus Conference include DAPD (abstracts 306, 752) (from Triangle and Abbott Pharmaceuticals), L-Fd4C (abstract 303, Achillion Pharmaceuticals), BCH-10618 (abstract 472, BioChem Pharma) and the "ethinyl nucleoside analogs" (abstract 305). DAPD data has been previously presented showing in small studies that it has antiviral activity against some NRTI resistant viruses. In one small brief study DAPD yielded a 1 log reduction in viral load in NRTI experienced patients, and a 1.5 log reduction in patients naïve to HIV treatment. Dr. Lisa Dunckle of Achillion presented data indicating the L-Fd4C (ACH-126,443) led to minimal decreases in "mitochondrial DNA" when compared to no drug (control) in vitro (laboratory tests, abstract 303). And, combining L-Fd4C with either ddI, d4T, or ddC led to improvements in mitochondrial DNA, when compared to any one of them alone. The drug appears to have a favorable resistance profile against NRTI resistant virus. Once daily dosing might be possible, and human studies will begin early this year. BCH-10618 development was previously stopped due to toxicities but the development program appears to be re-started.

Tenofovir DF(PMPA):
Once Daily NtRTI Drug
The final 48-week results of Study 902 were presented by Michael Miller, PhD of Gilead (abstract 441). Tenofovir DF is an NtRTI (nucleotide reverse transcriptase inhibitor) drug with activity against HIV and HBV (hepatitis B virus) that is dosed once daily. In this dose-ranging study, 189 heavily pretreated patients were randomized to add to their stable HAART regimen either placebo (inactive drug) or tenofovir, dosed at 75 mg, 150 mg or 300 mg once daily for 24 weeks. After that time, all placebo patients rolled over into 300 mg once daily. The mean duration of prior anti-HIV therapy was 4.6 years, with a median baseline CD4 count of 375 cells per microliter and a viral load (VL) of 3.7 log (5,011) copies per milliliter. At baseline, 94% had some NRTI drug genotypic resistance and 57% had some PI drug genotypic resistance. For all 54 patients in the 300 mg arms, the 48-week results showed a mean HIV VL decrease of –0.6 log copies per milliliter. CD4 count changes and percentage with an undetectable VL were not presented; however, a VL reduction of –0.6 log would be expected to be associated with a slower progression of HIV disease, some CD4 count increase and possibly, decreased "fitness" in the dominant HIV strain. For those with baseline resistance to lamivudine (3TC, Epivir, "M184V"), the VL decrease was –0.8 log copies per milliliter. 3TC resistance increases PMPA activity in preliminary studies. But the 3TC M184V mutation must be maintained while on PMPA, so a patient needs to take a drug that maintains the mutation such as 3TC. Developing genotype resistance to tenofovir ("K65R") occurred in only 2% and was not associated with VL rebound. It will be useful to see this drug in other studies of HIV patients. No kidney toxicity has been reported, unlike high doses (60-120 mg) of adefovir, another NRTI drug being developed by Gilead to treat chronic hepatitis B (see report on page 22). Immediately prior to the Conference, Gilead announced an Expanded Access Program for tenofovir DF in the US. For more information, physicians may call 1-800-GILEAD-5.

Entry Inhibitors:
Fusion Inhibitors T-20, T-1249, T-649
Several presentations addressed the next class of anti-HIV drugs, "entry" inhibitors, that include attachment and fusion inhibitors. Existing classes of drugs work inside the infected cell. Nucleoside reverse transcriptase inhibitors such as AZT and abacavir and non-nucleoside reverse transcriptase inhibitors like nevirapine and efavirenz all act by shutting down the reverse transcriptase enzyme that HIV uses to replicate itself once inside the cell. Protease inhibitors shut down the viral protease enzyme HIV uses to package itself for export. By contrast, entry inhibitors are drugs that interfere with the processes involved in the virus' initial assault on the cell's outer membrane--the process that permits attachment and entry into the cell. The fusion inhibitor T-20, developed by Trimeris and Roche, is the furthest along in development and genuinely an exciting agent. It is currently dosed twice daily with subcutaneous (under the skin) injections (like diabetics administer insulin).

Jay Lalezari, MD of Quest Research in San Francisco presented interim, 16-week results of a phase II, randomized, controlled trial of 71 treatment-experienced patients who took T-20 (abstract LB5). All had previously taken PI and NRTI drugs, but were naïve to (never took) NNRTI drugs. Patients were randomized to placebo (inactive drug) or T-20 at a dose of 50, 75 or 100 mg twice daily by subcutaneous (under the skin) injection (50 mg per milliliter), also with abacavir (ABC, Ziagen, NRTI drug), amprenavir (APV, Agenerase, PI drug, standard dosing), ritonavir (RTV, Norvir 200 mg "booster dose" twice daily), and efavirenz (EFV, Sustiva, NNRTI drug). The overall median baseline viral load (VL) was 4.3 log (18,620) copies per milliliter with a CD4 count of 232 cells per microliter; levels were similar across the study arms.

The interim results after 16 weeks revealed the median change in VL at week 16 in the 100 mg bid T-20 treatment arm was -2.80 logs and -2.30 log in the 50 mg bid arm. This was compared to a -2.15 log reduction in the arm not receiving T-20 (as treated analysis, excluding drop-outs). The 0.65 log difference in reduction in viral load between the group receiving T-20 and those not receiving T-20 is the key finding of the study, and the difference in viral load reduction was greater if patients with viral load <20,000 copies/ml were not included in the analysis. CD4 count increases in the highest 2 T-20 dosing arms were 64-74 cells per microliter. There was a trend towards increased suppression of HIV and higher CD4 counts in the highest two doses. This study was a pilot exploratory study and as such was not designed to compare antiviral efficacy. 65% of patients had pain or redness at the T-20 injection sites, otherwise side affects between the T-20 and control arms were similar. Completing 16 weeks included 75% of patients. Using a strict "ITT" analysis (all enrollees included who were randomized and received at least one dose of T-20), 48% of the T-20 arms and 37% of the placebo control arm achieved <50 copies/ml, and 58% in the placebo arm vs 71% in the T-20 arm had <400 copies/ml. There were 6 serious adverse events determined to be possibly or probably related to study drugs by the investigators. Yet no single event occurred more than once suggesting no consistent T-20 toxicity, except for one injection site abscess ("pus pocket"). Phase III studies are ongoing. Application to the FDA is expected in 2002 if all goes well. Dr. Lalezari indicated that a new formulation of T-20 had 100 mg/ml that would allow for 1 injection twice per day at the highest dose used in this study.

In a small study in children, with viral load >10,000 on a failing regimen, T-20 was dosed at 30 and 60 mg/m2 (abstract 561). Viral load log reductions of 0.8 and 0.9 were seen, and again the most frequent treatment related adverse event was local injection site reactions occurring in 8/12 children. They were graded as "Mild" in all but 1 child. Further testing in children is planned.

Resistance to T-20 has been reported in in vitro studies, although current commercial resistance tests do not report T-20 resistance. However, such HIV isolates are fully sensitive to another experimental fusion inhibitor, T-1249, again from Trimeris/Roche. This agent already is in clinical development, and appears to allow for once daily dosing. Like T-20, T-1249 is a peptide (protein) fusion inhibitor and as such will not be suitable for oral administration. In laboratory and animal studies, T-1249 has demonstrated to be more potent than T-20.

T-1249 phase I/II study results were presented by Joe Eron, MD of the University of North Carolina at Chapel Hill (abstract 14). The study involved 72 HIV-positive, treatment experienced adults on no concomitant therapy and HIV RNA greater than 5,000 copies per milliliter. Patients received T-1249 as monotherapy by sub-cutaneous (under the skin) injection for 14 days at doses ranging from 6.25 mg daily to 50 mg daily, given as either once or twice daily. The drug demonstrated dose-dependent activity (i.e., the more drug given, the greater the fall in viral load). On day 14, the median change from baseline ranged from -0.1 log at the lowest dose up to -1.4 log copies per milliliter at 50 mg per day. Since T-1249 appears to be eliminated more slowly than T-20, this will allow for once-daily dosing. Only two serious adverse events assessed as possibly related to study drug, occurred during the study: an allergic or ‘hypersensitivity’ reaction (involving oral ulcers, a rash and fever) in one patient and a marked fall in white cell count ("neutropenia") in another. Otherwise, important clinical or laboratory abnormalities did not occur. Injection site reactions (soreness, localized swelling and redness) were mild and reported in 40% of patients. As a maximal tolerated or effective dose was not established in this study, evaluation of higher doses appears warranted.

In a separate presentation about sensitivity to T-20 and T-1249, Dr. M. Greenberg of Duke University in Durham, North Carolina reported that either drug’s effectiveness is independent of the fusion "co-receptors CCR5 or CXCR4" (abstract 473). This suggests that either drug would be effective in early HIV infection (when CCR5 co-receptors dominate) as well as late infection (when CXCR4 co-receptors dominate, allowing HIV entry in T cells). Another fusion inhibitor, T-649, was presented by Dr. C.A. Derdyn of the University of Alabama at Birmingham (abstract 475).

Other Entry Inhibitors (SCH-C)
One of the most exciting new classes of drugs in development was presented by Dr. Gregory Reyes of Schering-Plough at a State-of-the-Art lecture (abstract L11). Schering has 2 new "entry" inhibitor drugs that block the binding of HIV to the "CCR5 co-receptor," a step required before the virus can fuse and enter into cells. People who have a genetic mutation in both of their CCR5 co-receptors are resistant to HIV infection ("homozygotes"), and HIV positives who carry 1 defective gene ("heterozygotes") have a slower rate of disease progression. "Homozygotes" are otherwise totally healthy, thus it is believed that blocking the CCR5 co-receptor should be safe (although there are some concerns about safety). The 2 compounds (SCH-C and SCH-D) can be taken by mouth, are well absorbed into the bloodstream, are potent inhibitors of HIV in vitro (laboratory) and in animal models, and have no significant toxicities in those animals. The drugs also have a long "half-life" (time until an original amount is metabolized by half), suggesting likely dosing at once daily. They are "synergistic" (enhanced anti-HIV effects by combining) with NRTI drugs and should also be synergistic with other drugs that are entry inhibitors, but that block entry at different steps, including T-20, T-1249, AOP-RANTES, NNY-RANTES and the CXCR4 co-receptor inhibitor AMD3100 (abstracts 309 - 311). In vitro, resistance to SCH-C was demonstrated, but fortunately, it was not due to a switch to the "CXCR4" co-receptor that is associated with more severe HIV disease. The only side affect seen at high doses of SCH-C was "prolongation of the QT interval" (longer electrical wave on "ECG, electrocardiogram," heart wave test) that might cause an abnormal heart rhythm ("Torsades") that can be fatal.

Dr. K. Nakashima of Progenics Pharmaceuticals reported their development of a "gp120" inhibitor PRO 140 and an attachment inhibitor PRO 542 (abstract 310). Dr. M.J. Root of the Massachusetts Institute of Technology described the "gp41" entry inhibitor 5-Helix that is extremely potent in laboratory tests (abstract LB1). However, the drug may be difficult to develop, and Dr. Root suggested it might have utility in developing an anti-HIV vaccine.

Integrase Inhibitor
In his Plenary presentation, "New Antiretroviral Drugs," Roy Gulick, MD of Cornell University in New York City included experimental integrase inhibitors under development by Schering (abstract S25). Dr. D. S. Strayer of Jefferson Medical College in Philadelphia, Pennsylvania documented enhanced control of HIV when using a "gene vector" ("SV Aw") that contains antibodies to integrase in "SCID-hu" mice (mice with "severe combined immune deficiency" that have human thymus gland transplants, abstract 312).

Interleukin-2:
Promising But Preliminary Clinical IL-2 Data
Interleukin-2 (IL-2, aldesleukin, Proleukin) is an FDA-approved drug to treat "metastatic" (spread throughout) kidney cancer and melanoma (skin cancer), but not for an HIV indication. This immune modulating drug was the topic in several presentations. IL-2 already is recommended for persons with no or modest CD4 cell responses after 24 weeks of HAART in the current French National HIV Treatment Guidelines. A variety of smaller studies performed in different patient populations and clinical settings, have demonstrated that adjunctive IL-2 therapy can result in substantial rises in CD4 cell count. There were updates that document this in particular for patients who use IL-2 with HAART during or shortly after primary (acute) HIV infection (abstracts 406 , 407). Also, IL-2’s toxicities and side effects are well known.

IL-2 is now being evaluated in a large, 4,000 patient, 5-year global study assessing whether this drug has clinical benefits in persons on HAART with CD4 counts greater than 300 cells per microliter (ESPRIT study, telephone 1-800-772-5464 extension 58008). A similar study for those with a CD4 count less than 300 cells per microliter is underway called SILCAAT. The endpoints of these studies will include changes in CD4 counts and HIV viral loads, clinical progression to AIDS, toxicities and costs.

Ronald Mitsuyasu, MD of the University of California at Los Angeles reported the final results of ACTG 328 (abstract 17). This prospective, randomized study evaluated IL-2 added to HAART versus HAART alone in 204 patients naïve to (never took) PI drug therapy. The study enrolled patients with CD4s between 50-350. The baseline median CD4 count was 199 prior to receiving 12 weeks of HAART, and 249 after 12 weeks of HAART (equivalent in both IL-2 dosing arms). For those who achieved a viral load less than 3.7 log or 5,000 copies per milliliter after 12 weeks of indinavir (Crixivan, PI drug) plus 2 NRTI drugs, randomization then occurred to continue HAART alone or to add IL-2 in cycles. At baseline median viral load was 4.5 log (31,000 copies/ml), and was 1.7 log after 12 weeks of HAART. The dose of IL-2 was either continuous IV (CIV, intravenous, 9 MIU [million international units] daily for 5 days every 8 weeks, 54 patients) or subcutaneous (SC, 7.5 MIU twice daily for 5 days every 8 weeks, 55 patients). Follow-up was for 84 weeks.

The results showed that approximately 2/3 of patients completed 6 cycles of IL-2. After 60 and 84 weeks of therapy, IL-2 resulted in significantly greater increases in CD4 counts compared to HAART alone. At the end of the study, CD4 cell counts were higher in the IV arm (800 cells) than the SC arm (614 cells), which were both significantly higher than the no IL-2 (HAART alone) arm (396 cells per microliter). HIV viral loads were not significantly different comparing the three arms. Only 1 AIDS-related infection was seen in both of the IL-2 arms, compared to 5 in the HAART only arm, suggesting that IL-2 may have clinical benefits. However, direct proof of this still awaits the completion of other ongoing clinical-endpoint studies. (The current study was not "powered statistically" to answer the question of potential clinical benefits.) As allowed by protocol, 76% of patients on CIV switched to SC IL-2 after 3 or 6 cycles due to adverse events related to IL-2 ("flu"-like symptoms), which also commonly led to dose reductions in both IL-2 arms. Serious grade 3-4 (serious or life-threatening) toxicities were infrequent and likely related to IL-2: nausea, vomiting, fever, and fatigue.

Dr. P. Saint-Mezard of Pitie-Salpetriere Hospital in Paris, France reported that SC IL-2 added to HAART led to increased production of naïve (have not been exposed to "foreign" antigens or proteins) T cells from the thymus gland (behind the breastbone) in advanced HIV patients (abstract 350). This is promising, but still to be determined is if these naïve cells can be programmed to be clinically effective. This can likely only be determined through the planned SILCAAT & ESPRIT clinical endpoint studies. The 14 patients with CD4 counts less than 200 cells per microliter received adjuvant SC IL2 (4.5 MIU twice daily in 4 cycles every 6 weeks). The proportion of T cells recently produced by the thymus was measured by counting "TRECs" (T cell receptor rearrangement excision circles) that occurs during development of these cells there. (TREC is a piece of DNA remnant left in a new T cell that has developed in the thymus. A cell that is TREC positive is a new T cell. As a cell ages and divides into multiple generations of "daughter" cells, TREC decreases, since its signal wanes by dilution. The test uses PCR to measure TREC).

IL-2 therapy led to a significant increase in the percentage and the number of naïve and memory CD4 cells, paralleled by an increase of the number of TRECs in the CD4 cell subset. This study provided evidence that IL-2 may do more than just increase the number of T cells; it may contribute to real immune reconstitution.

Nadir CD4 Predicts IL-2 Response Suggesting Immune Defects Persist Even After HAART; Nice CD4 Increases Even When Nadir CD4 <250
Dr. Neal Markowitz of Henry Ford Hospital in Detroit, Michigan reported the interim results of a community study wherein SC IL-2 was added in patients who had responded to ART (antiretroviral therapy) (abstract 343). CPCRA 059 was a multicenter, open-label, randomized trial of SC IL-2 added at 7.5 MIU (126 patients) or 4.5 MIU (130 patients) twice daily for 5 consecutive days every 8 weeks, for at least 3 cycles. A total of 192 patients completed at least 3 cycles and 245 patients had a month-12 CD4 cell count. At baseline, the HIV VL was undetectable (limit 50 copies per milliliter) in 60%. The median pre-enrollment nadir (lowest) and baseline CD4 counts were 293 cells and 538 cells per milliliter, respectively, while the median time on ART at entry was nearly 4 years. Even in the worst cohort, those with CD4 nadirs < 250/µl and baseline CD4 counts < 480/µl, CD4 response at 12 months was +363 cells. If IL-2 can be shown to confer a clinical benefit, this study would suggest that it may work in community settings, even in patients with long histories of HIV infection and therapy. The interim results showed that the nadir (lowest) CD4 cell count was the strongest significant predictor of responding to IL-2 (higher count associated with better response). This suggests that immune defects may persist even after HAART leads to a CD4 count increase. However, longer follow-up and treatment of these poor responders may demonstrate that immune reconstitution with IL-2 might require longer than 12 months in some subjects. The results also suggested that BMI (body mass index, body weight divided by height squared) should be considered in selecting IL-2 dosing.

IL-2 Dosing: 7.5 MIU or 4.5-5 MIU SC?
Dr. R. Arduino of The University of Texas at Houston presented a meta-analysis (combining studies) of the CD4 cell response to 3 doses of SC IL-2 in 3 Vanguard studies performed in Houston, Argentina and Thailand (abstract 346). SC IL-2 was given twice daily in 5-day cycles every 8 weeks for 3 cycles to patients already taking ART and who had CD4 counts greater than 350 cells per microliter. The results were a significant IL-2 dose-dependent increase in CD4 cell counts. Specifically, the analysis supported the ability of the 7.5 MIU dose to achieve the most rapid CD4 count increases greater than 100 cells per microliter. However, this dose was also associated with higher rates of dropout and toxicity than lower dosing. Nonetheless, the authors concluded that their results "supported the initial use of 7.5 MIU dose" of SC IL-2. The French ANRS 079 study group used 5 MIU twice daily for 5 days every 4 weeks for 3 cycles and felt this leads to less dose reductions.

4-year follow-up of added SC IL-2 in cycles from 3 studies was presented by Doreen Chait, RN, MPH of the US NIAID (abstract 347). Among 97 patients originally enrolled in IL-2 studies (added to ART) between 1993-97, 77 (100% men) entered the extension phases and 63 remain on active follow-up. Two studies required a baseline CD4 count of at least 200 cells per microliter, while the 3rd required at least 500 cells. The mean duration of study participation for the 77 patients is 4.1 (range: 1 –7.1) years. The mean baseline CD4 cell count was 543 cells (28%). The results showed that the current mean CD4 count after 4.1 years is 1,030 cells per microliter (40%). During their most recent cycle of SC IL-2, patients tolerated a mean total daily dose of 11.6 MIU (range: 3.0-15.0). To maintain that CD4 count level, the mean number of IL-2 cycles has been 10 (range 3-28). However, the mean interval since the last required IL-2 cycle was 26 months (range 2-60). One patient has developed lymphoma, an AIDS-defining illness. Yet, no AIDS-defining opportunistic infections have occurred. For 56 of the patients with viral load results, the median baseline level was 3.4 log (2,511) copies and has decreased to 1.69 log (49) copies per milliliter.

Two Presentations Addressed the Potential for Decreasing Side-Effects from IL-2
Etanercept (Enbrel, FDA-approved drug for rheumatoid arthritis, "recombinant tumor necrosis factor [TNF] receptor," a blocker of "TNF-alpha) when given prior to a cycle of interleukin-2 (IL-2, aldesleukin, Proleukin) against a background of HAART blunted "immune activation" of "pro-inflammatory cytokines" known to be induced by IL-2 (abstract 66). Dr. B. Sha of Rush Medical College in Chicago, Illinois presented the prospective study of 11 patients. Overall, etanercept did not appear to be a particularly effective strategy for increasing the tolerability of IL-2.

In a separate presentation, Dr. J.A. Tavel of the US NIH reported that prednisone (immune blocker) reduced moderate or severe IL-2 toxicities, but that it significantly blunted CD4 cell increases (poster 348). The dose of SC IL-2 was 7.5 MIU twice a day for 5 days every 8 weeks in combination with placebo or prednisone, 0.5 mg per kilogram divided into two daily doses for 7 days (during and two days after IL-2). (Note weight in pounds times 0.454 = weight in kilograms.) All patients previously had been taking stable HAART. IL-2 was decreased by 1.5 MIU per dose as needed for adverse events. The results revealed that the 10 patients completed a total of 37 cycles of IL-2 with prednisone or placebo, with a mean follow-up of 7 months. The mean total cycle dose of IL-2 was insignificantly different between the 2 arms (approximately 12 MIU daily). In a strict ITT analysis (all patients included), significantly fewer toxicities were seen in the group receiving prednisone compared to placebo. Specifically, 31 grade 3-4 (severe or life threatening) toxicities occurred among the patients receiving IL-2 plus placebo, compared to only 3 grade 3-4 toxicities in the patients receiving IL-2 plus prednisone. However, prednisone co-administration significantly blunted CD4 cell count increase (mean increase of 0.7 CD4 cells per day), when compared to the no-prednisone arm (increase of 2.4 cells per microliter per day). HIV-RNA decreased or remained undetectable in all patients, with no significant difference between the groups. The results also showed that while fewer grade 3-4 toxicities occurred among IL-2 recipients also taking prednisone, this trend does not permit higher rIL-2 dosing. Note that chronic steroids usage, including prednisone is associated with decreased immune response to opportunistic infections.

Other Experimental Drugs
Other experimental drugs to treat HIV were presented at the Conference. They included the following:

Thalidomide (Thalomid, FDA-approve to treat leprosy and used experimentally to treat HIV wasting and oral ulcers) was tested in 36 HIV-positive patients with only moderate baseline immune compromise (median CD4 count 338 cells per microliter (abstract 352). David Wohl, MD of the University of North Carolina at Chapel Hill authored the dose-escalation study that used a daily dosing of 50-150 mg for 8 weeks. The results were that thalidomide did not significantly affect the median HIV viral load or lead to any dose-related changes in the median CD4 count. Known side effects included sleepiness and neuropathy. In a separate presentation by Dr. P.A.J. Haslett of The Rockefeller University in New York, thalidomide therapy for 21 days increased T lymphocytes directed against HIV and CMV (cytomegalovirus, CMV, cause of HIV-blindness) in 2 patients (abstract 173).

Mycophenolate mofetil (MM, CellCept, FDA-approved to help prevent immune rejection of transplanted organs, same class as ribavirin used to treat chronic hepatitis C) was shown to increase the anti-HIV affect of ddI (didanosine, Videx, NRTI drug), abacavir (Ziagen, NRTI drug), and tenofovir DF (experimental NtRTI drug) in vitro (in the laboratory), including "wild-type " (no mutations) & NRTI-resistant HIV (Dr. M. Hossain, abstract 307). In a small, pilot study, low dose (250-500 mg twice daily) MM, when added to HAART that included ABC, led to additional VL reductions in 3 HIV-infected individuals, even with multi-drug-resistant HIV (D. Margolis, MD, abstract 351). In a separate presentation of similar patients, MM reduced abnormal lymphocyte "activation" (Dr. J. J. Coull, abstract 56). However, there are concerns about MM toxicity.

Hydroxyurea (Hydrea, Droxia, FDA-approved to treat certain cancers and sickle cell anemia) was discussed as experimental anti-HIV co-therapy in several presentations, without any newly reported, definitive benefits (abstracts 356 , 363 , 408 , 445 , 617 , 620 , 747). Due to documented risks of severe and life-threatening toxicity to the liver (abstract 617) and pancreas (abstract 620), FDA has modified the "black box" warnings when combining hydroxyurea with either of 2 NRTI drugs, didanosine (ddI, Videx) and stavudine (d4T, Zerit).

Cyclosporine (several brand names, FDA-approved for organ transplants) was added to HAART for primary HIV infection and led to significantly greater increases in CD4 counts than HAART alone, and it prevented "activation" and movement of those cells into lymphoid organs (Dr. G.P. Rizzardi, abstract 759).

Other experimental treatments presented at the Retrovirus Conference included DCK (abstract 302), interleukin-7 (abstracts 375 , 376 ), interleukin-12 (abstracts 65 , 421 ), CP96,345 ("substance P antagonist, abstract 41), PKF050-638 (HIV "rev" inhibitor, abstract 283), "SV Aw" ("gene delivery" with "anti-integrase" antibody, abstract 312), cyclophosphamide (Cytoxan cancer drug, (abstract 16), no benefits when added to HAART), and actinomycin D, abstract 64, although it increases HIV growth in vitro [laboratory]).