There is a growing need for new anti-HIV medications for people who have developed resistance to available drugs in every class of therapy and are no longer able to construct a suppressive regimen. These so-called salvage patients are part of an expanding population who, whether because of improper prescribing, inadequate pharmacokinetics, or poor adherence, find themselves chronically awaiting the next ARV to enter the marketplace. It is now understood that simply adding the latest drug down the pipeline to a failing regimen -- a practice called serial monotherapy -- will quickly result in resistance to the newcomer as well, and that unless at least two new active ARV agents are added to a salvage regimen, the chain of virologic failure due to resistance can not be broken.
The next investigational HIV agent likely to hit this treadmill is a protease inhibitor (PI) called tipranavir. The drug has been specifically developed by its sponsor, Boehringer-Ingelheim, to address the urgent need for new agents in salvage therapy for people with limited treatment options due to multi-drug resistance. Tipranavir is active against many strains of HIV that have developed resistance to currently available PIs. At the 44th annual ICAAC meeting in Washington, DC, in October 2004, Charles Hicks, of Duke University, reported on a planned, 24 week, interim analysis of a phase III clinical trial of tipranavir, dubbed RESIST-1. The trial enrolled 620 patients in the U.S., Canada and Australia.
In the randomized RESIST-1 study, tipranavir was added to a physician-selected, standard-of-care regimen that was prescribed for all participants at entry. At 24 weeks, regimens that added tipranavir outperformed the comparator regimens, producing a treatment response (defined as a > 1.0 log drop in HIV RNA) in 41.5% of patients, compared to an equivalent response in 22.3% of those who did not add tipranavir to their "best available" regimens (P<0.0001). Viral load was suppressed below 400 copies/mL in 34.7% of those receiving tipranavir versus 16.5% of patients in the comparator arm (P<0.001) and fell below 50 copies/mL in 25.1% and 10.0% of patients in these respective groups (P<0.001). The median viral load drop was -0.88 log copies/mL in the tipranavir arm compared to -0.26 log copies/mL in the control group (P<0.001). Changes in CD4 cell counts at 24 weeks were modest, with a median rise of 36 cell/mm3 experienced by those receiving tipranavir versus an added 6 cells/mm3 by those in the comparison arm (P<0.001).
Grade 3 or 4 adverse events were more common on the tipranavir arm, mainly due to increased rates of nausea, and there were more discontinuations due to adverse events than in the comparator arm. Grade 3 or 4 elevations of ALT were 3 times more frequent on the tipranavir arm than the comparator regimen (6.9% vs. 1.3%) although ALT and AST elevations were reported as asymptomatic and not resulting in discontinuations. The incidence of elevated cholesterol (4.2% vs. 0) and triglycerides (21.0% vs. 12.5%) were also significantly more common in those receiving tipranavir.
At the 7th International Congress on Drug Therapy in HIV Infection, held in Glasgow in November 2004, preliminary 24-week results were presented from a sister study to RESIST-1. RESIST-2 was conducted in 863 subjects in Europe and Latin America. Not surprisingly, its results served to reinforce the findings reported in RESIST-1.
In RESIST-2, adding tipranavir produced a >1.0 log drop in HIV RNA in 41.0% of subjects compared to only 14.9% of those in the comparator arm (P<0.001). Viral load was suppressed below 400 copies/mL in 33.6% of those receiving tipranavir versus 13.1% of patients in the comparator arm (P<0.0001) and fell below 50 copies/mL in 22.5% and 8.6% of patients in these respective groups (P<0.0001). The median viral load drop was -0.72 log copies/mL in the tipranavir arm compared to -0.22 log copies/mL in the control group (P<0.001). Changes in CD4 cell count in RESIST-2 were also similar to the modest changes seen in RESIST-1. The impact of using enfuvirtide in combination with tipranavir in this study was less impressive than in RESIST-1, with only 38.5% of subjects who used the two drugs together achieving viral load suppression below 400 copies.
As in RESIST-1, patients receiving tipranavir in RESIST-2 had significantly greater incidence of grade 3 or 4 laboratory abnormalities than those receiving the comparator regimens; these included ALT (5.2% vs. 2.2%), cholesterol (2.5% vs. 0.5%) and triglycerides (20.1% vs. 10.2%).
The sponsor had previously reported serious drug-drug interactions between tipranavir and other protease inhibitors, which will likely make using this drug as part of a dual boosted PI strategy quite tricky. BI says it plans to deliver a comprehensive analysis of the best way to use -- and not use -- tipranavir in salvage therapy at the next annual Retrovirus Conference in February 2005. An expanded access program is slated to begin soon, although limitations on using other investigational drugs while receiving tipranavir may mean in effect that, if adding enfuvirtide is not an option, then many individuals may have no second active drug to add to their salvage regimen, thus perpetuating serial monotherapy and the attendant risks of new resistance. If all goes well, FDA approval could come as soon as mid-Spring 2005. Due to the likely complexity of using this drug in very advanced patients with few options, the FDA is urged to convene an advisory committee meeting to discuss tipranavir and air these issues before approval.
Gilead Sciences announced in October 2004 that it was abandoning the development of its protease inhibitor (PI) candidate, GS 9005 due to lack of bioavailability. While the drug itself was not revolutionary -- it offered a resistance profile similar to that of tipranavir -- it was meant to serve as the payload for Gilead's novel prodrug technology that specifically targets drug molecules to lymphocytes -- the cells that HIV prefers to infect. The idea is that when the prodrug complex (comprised of the targeting molecule joined to the drug molecule) enters a lymphocyte, it is cleaved by a host enzyme specific to those cells, which releases and activates the drug precisely where it can do the most good. Meanwhile, other types of cells are spared from any collateral toxicity. At least that is the idea.
Gilead proved the concept with a prodrug version of tenofovir, but looked to protease inhibitors as the therapeutic class that could most benefit from the reduction in toxicity and boost in effective concentration expected with more precise targeting. Unfortunately, the PI molecule they selected turned out to be too large to efficiently make it past the gut and into the bloodstream. In a proof-of-concept study of GS 9005 in people with HIV, no significant viral load reduction was seen. Although a similar study of GS 7340, a Viread version of the prodrug, was successful, the company thinks there is little to be gained by tweaking tenofovir. The search is on now for a PI candidate better suited for this novel approach. The disappointing outcome of this agent demonstrates why many companies are reluctant to talk about drug candidates before they have proved themselves in people. But as community treatment advocates have become more sophisticated about the drug development process, we have learned to season our optimism with the salt of caution.
It increasingly appears that there is something not quite right about combining ddI and tenofovir. However, exactly what that is, isn't clear. It's been known for some time that tenofovir can raise blood levels of ddI, which (unless managed with ddI dose reduction) can cause characteristic ddI toxicities. Recently, some small studies have identified early virologic failure in patients taking the drugs together. One theory for this is that tenofovir may cause inadequate metabolism of ddI, possibly by inhibiting PNP, an obscure enzyme in the gut that metabolizes ddI, thereby eliminating one active drug and allowing viral failure. Still other groups have recently reported that HIV-infected patients combining ddI with tenofovir may experience a paradoxical decline in CD4+ T-cell count despite achieving complete viral suppression. This interpretation, reported by Vincent Soriano, from Hospital Carlos III, in Madrid, and presented at the 44th ICAAC in Washington DC, found that T-cell loss began to be evident by about 6 months on the combination and increased over time. The paradoxical T-cell loss continued for some patients despite initiating or switching to a lower dose of ddI.
The investigators conducted a chart review of 570 patients who had initiated a PI-sparing regimen between September 2002 and June 2003 and who had maintained viral load below 400 copies/mL for at least a year. Changes in CD4+ cell counts were analyzed by treatment experience status and by whether the third drug in the regimen was an NNRTI or another nucleoside. The impact of ddI and tenofovir when used separately was also evaluated.
Dose reduction of ddI from 400 mg per day to 250 mg per day has been recommended when used in combination with tenofovir due to a drug-drug interaction that raises ddI levels and has been observed to produce characteristic ddI toxicity. Yet the drug combination remains attractive because both agents are regarded as potent antiretrovirals with a high genetic barrier to resistance and offer convenient once daily dosing.
By 12 months of ddI/tenofovir therapy, CD4+ cell declines greater than 50 cells/mm3 had occurred in 78 of 175 (45%) treatment experienced patients who were also taking an NNRTI. About a third of this group had CD4+ cell counts reduced by over 100, and 14% had losses greater than 200 cells/mm3. Patients on their first NNRTI-containing regimen fared better; 20% lost at least 50 cells and none lost as many as 200. Yet the CD4 loss was dramatically worse for patients employing a third nucleoside in their regimens: 91% of treatment experienced patients on triple NRTIs that included ddI and tenofovir had CD4 losses in excess of 100 and 64% lost more than 200 cells/mm3. No treatment naive patients received a triple nucleoside/nucleotide combination.
With regard to ddI dose, 38% of treatment experienced patients on a NNRTI-containing regimen who received a lower dose of ddI, experienced CD4 cell loss greater than 50 cells/mm3. However, this proportion is not dramatically different from the equivalent degree of CD4 cell loss observed in 30% of treatment experienced patients on an NNRTI who did not combine ddI with tenofovir. Nevertheless, of the 8 patients who experienced a change in absolute CD4 count from over 350 cells/mm3 to below 200, 2 had received a lower dose of ddI, which suggests caution in using this combination, particularly in patients with extensive treatment experience and low baseline CD4 counts.
However, several prominent researchers are said to have doubts about the existence of a paradoxical loss of T-cells despite viral suppression. Indeed, two studies presented at the Glasgow meeting found fault with the ddI/tenofovir combo for not adequately suppressing viral load. Based on these two studies, Bristol-Myers Squibb issued a Dear Doctor letter warning not to use ddI EC tablets in combination with tenofovir and either nevirapine or efavirenz. This situation is rife with contradictions. Until someone can definitively explain what is going on with these drugs, the prudent thing may be to consult a specialist HIV physician before starting or continuing this combo.