The annual medical meeting called the Interscience Conference on Antimicrobial Agents and Chemotherapy, or ICAAC, is one of the research community's most important venues for reporting on the progress of new anti-infective drugs, including those for HIV. This year's event, originally scheduled for late September in Chicago, was postponed for three months due to travel uncertainty following the terror attack on New York. But the book containing abstracts of the meeting's presentations and posters was already in the mail. Rather than let this information sit unread until the New Year, we present a two-part look at some of what ICAAC might have offered. In this issue we review a number of studies concerning a new antiretroviral drug, tenofovir disoproxil fumarate (TDF). Next month we'll look at some research on salvage therapy and drug interactions.
Keep in mind that abstracts are submitted many months in advance of the conference and often report preliminary or partial data compared to the final poster presentation.
A notable innovation in Gilead's development plan for tenofovir was the decision to first evaluate the drug's benefit in a treatment-experienced population. This is significant to the HIV community because data from so-called "salvage" studies will be particularly relevant to those most likely to be among the first wave of tenofovir users. The decision to study the drug in a trial of treatment intensification among a population with advanced disease also probably helped accelerate tenofovir towards approval. Finally, Gilead's decision to develop trials for a population with advanced disease and extensive treatment experience gave many people an opportunity to gain early access to one of the few new drugs likely to help them.
However, this strategy came with risks. No HIV drug has been fully approved without data showing efficacy in previously untreated individuals. While there is no scientific rationale why a drug that lowers viral load in treatment experienced people with drug-resistant virus would not also lower viral load in those who are treatment naïve, there have been concerns that the FDA will not grant full approval to tenofovir for use in both naïve and experienced adults. Despite this concern, it's not clear that the abstract status of partial approval (which will not prevent doctors from prescribing the drug for any patient) would outweigh the convenience of an easy-to-take, once-a-day pill with much-needed activity against drug-resistant strains of virus. On October 26, the FDA approved tenofovir without restrictions for use in adults with HIV.
One common frustration of people who take and prescribe HIV drugs is the lack of data about possible differences in absorption and clearance of drugs due to an individual's weight, age or gender. A new drug's key pharmacokinetic (PK) investigations are most often performed using HIV-negative individuals -- usually males. Studies to evaluate demographic factors on dosing and blood levels of HIV drugs are rare, especially after a drug has been approved for sale.
(Abstract references are listed at the end of this article.)
Gilead analyzed results from PK studies among 17 people with HIV and 36 without, to evaluate PK characteristics of the drug by gender, serostatus, weight and age. Although not a prospective study, the analysis found no significant associations between these demographic factors and the pharmacokinetics of tenofovir. For a drug about to enter widespread usage in combination with any of fifteen other approved HIV drugs, even this slim amount of data is helpful. Hopefully, this sort of research will become a part of all drug development plans. Physicians and community members should impress upon drug company representatives at every opportunity the crucial need for this kind of data to be made available whenever a new drug comes to market. Gilead also has studies planned or in progress to investigate the PK performance of tenofovir for patients with renal or liver disease.
Gilead investigators also conducted a study of tenofovir to address concerns about potential PK interactions with ddI due to a shared route of elimination through the kidneys. Fourteen HIV-negative individuals were randomized into two groups, each group to receive either tenofovir or ddI individually for one week. Then, after another week off drugs as a washout period, all participants took the combination of tenofovir and ddI for one final week. The pharmacokinetics of tenofovir were not affected by ddI, although mean blood levels of ddI over time (AUC) increased by about 40 percent. A retrospective examination of safety data from two 24-week studies in which ddI was given with or without tenofovir did not find a significant difference between the groups in the incidence of pancreatitis or neuropathy, the two most important toxicities of ddI.
Gilead investigators also reported on a 291-person open-label safety study in highly treatment-experienced patients with CD4 cell counts below fifty. Tenofovir was part of a regimen that, in 91 percent of patients, included lopinavir/ritonavir (Kaletra). With a mean duration on tenofovir of 25 weeks, 15 percent experienced a serious adverse event (SAE), 8 percent dropped out due to adverse events, and 3 percent died of AIDS-related causes. Severe (Grade 3) adverse events included pneumonia and diarrhea (4 percent each). About a quarter of these patients had Grade 3 elevation of triglycerides.
An earlier, similar drug developed by Gilead called adefovir was abandoned as an AIDS drug after kidney damage developed in some people at doses necessary for activity against HIV (at lower doses, adefovir remains promising as an anti-hepatitis B drug). There has been a great deal of concern that tenofovir might also cause or exacerbate existing renal damage. This study included both people who had previously taken adefovir and those who hadn't; no significant changes in serum creatinine or phosphorus were observed in either group.
One of the highlights of ICAAC this year was to have been a presentation of data from an extension of Gilead's 48-week trial of tenofovir in 189 patients who were, again, treatment experienced. Individuals on stable antiretroviral therapy were initially randomized to add tenofovir at daily doses of 75mg, 150mg, or 300mg or placebo. Those on the 300mg dose experienced a mean -0.62 log reduction of viral load from baseline. The ICAAC paper was to present data on 135 patients who continued therapy with 300mg open label tenofovir beyond 48 weeks. Their mean reduction in viral load from baseline remained at -0.6 log by week 72. The safety profile at two years also remained consistent with earlier reports.
A report from a 253-patient virology substudy of Gilead's large Phase III trial in treatment-experienced individuals with unsuppressed viral load explains why tenofovir is likely to be embraced by people with drug-resistant virus. At baseline, nearly all had a nucleoside RT resistant mutation; 69 percent with viral mutations associated with loss of susceptibility to AZT and the other thymidine analogs; 68 percent with the 3TC-resistance mutation; and 45 percent with both. Despite this broad cross-resistance to nucleoside RT inhibitors, at 24 weeks, those receiving tenofovir had a mean viral load -0.59 log below that of the placebo group. Of 171 who had a genotypic test performed at the end of 24 weeks, only 68/171 had a sufficiently high viral load to allow sequencing. Of these, only 5/68 patients had the characteristic K65R RT mutation associated with resistance to tenofovir. In addition, significantly fewer people on tenofovir subsequently developed mutations that confer resistance to protease inhibitors, most likely because the rate of viral replication was held to a lower level by improved suppression.
Loss of susceptibility to DNA chain terminating RT inhibitors occurs generally via two mechanisms. One collection of RT mutations known as thymidine analog mutations, or TAMS (D67N, K70R, T215Y, and others), tends to allow more frequent dislodging of chain terminating drug molecules, which unblocks the DNA chain and lets reverse transcription continue. AZT is particularly prone to this kind of resistance. A different RT mutation (M184V) reduces the activity of 3TC by impeding the drug molecule's fit into the enzyme's active site. Two posters reported on the effects of these mechanisms on the activity of tenofovir.
A three-dimensional "snapshot" of tenofovir caught in the act of terminating a DNA chain in the clutches of RT was obtained by X-ray crystallography. Although these kinds of pictures are somewhat fuzzy, the investigators report observing a void in an area of the tenofovir/RT complex where the M184V RT mutation typically blocks 3TC from binding. The authors propose that the lack of a bulky structure in this region of the tenofovir molecule allows its activity against 3TC-resistant HIV.
As an extension of this research, Gilead offered a poster ranking the ease with which mutated RT can remove nucleoside chain terminators. As expected, AZT led the list, followed by D4T, ddC and to a lesser extent, abacavir. Least affected were 3TC, ddI and tenofovir, in that order. The report concludes by restating the observation that the activity of tenofovir appears to be little affected by either of these two mechanisms for NRTI resistance.
Finally, another Gilead sponsored study examined phenotypic resistance (Virco method) to tenofovir in HIV isolates derived from 1000 treatment-naïve and 5000 treatment-experienced individuals. Since nearly 98 percent of treatment-naïve isolates were susceptible to tenofovir at levels less than 3-fold greater than that for a wild-type control virus, the dividing line, or cutoff between tenofovir resistance and susceptibility was set at 3-fold. With this level established, the 5000 clinical isolates from treatment-experienced patients were tested. Only 5 percent of the samples had greater than 5-fold losses in susceptibility to tenofovir and 88 percent were within the "normal" 3-fold range. Eighty-five percent of samples resistant to 3TC were susceptible to tenofovir; for AZT-resistant isolates, the proportion was 71 percent. Results with isolates resistant to other NRTI drugs were similar.
Interestingly, the M184 mutation associated with 3TC resistance, in the presence of TAMS, may actually increase the susceptibility of RT to tenofovir. It's not known if 3TC therapy must be continued for this hypersusceptability effect to be useful as part of a treatment strategy.
These abstracts represent only a sample of the research that will contribute to our knowledge base about tenofovir. Additional studies and fresher data will emerge during the coming months and results from the large confirmatory trials should be known in about a year. Gilead is to be commended for taking a risky path in collecting data for tenofovir's approval; we know more about this drug in the populations in which it is most likely to be used than any previously approved HIV treatment. But even after approval for tenofovir is granted -- for whatever the indication -- studies must be continued that address open questions about possible toxicity, interactions, and the drug's effectiveness over extended periods of time.
Back to the GMHC Treatment Issues October 2001 contents page.