Why I Voted “No” to Truvada PrEP FREE

There is an urgent need for new prevention approaches to HIV infection (1). The principal interventions used to date have included condoms, risk-reduction counseling, and voluntary HIV testing. However, these interventions have been inadequate and there has been a cry for “new tools in the HIV prevention toolbox.” One such new tool is HIV preexposure prophylaxis (PrEP), the treatment for persons who are HIV-negative and at high risk for infection with antiretroviral drugs, which has now been studied in several randomized clinical trials, using Truvada (Gilead Sciences, Foster City, California), a licensed combination pill consisting of emtricitabine (FTC) and tenofovir disoproxil fumarate (TDF). On the basis of these and other studies, the manufacturer of Truvada, Gilead Sciences, submitted a new drug application to the U.S. Food and Drug Administration (FDA) for approval of a PrEP indication. On 10 May 2012, the FDA Antiviral Drugs Advisory Committee, of which I was a member, met to advise the agency on this application.

The committee considered data from 3 clinical trials in its individual and population PrEP risk–benefit assessment: iPrEx (Preexposure Prophylaxis Initiative), Partners PrEP, and CDC (Centers for Disease Control and Prevention) 4323 (2). The iPrEx study examined daily oral FTC–TDF in 2499 high-risk adult men who have sex with men (MSM) conducted in Brazil, Ecuador, Peru, South Africa, Thailand, and the United States (3). Partners PrEP was conducted in Kenya and Uganda, comparing daily oral TDF or FTC–TDF in 4758 heterosexual couples who are serodiscordant (4). The small CDC 4323 was conducted in the United States, in which 373 adult MSM were randomly assigned to receive immediate versus delayed treatment with daily oral TDF. The committee considered the results from the iPrEx and Partners PrEP studies for the efficacy indication of reducing HIV transmission, whereas data from all 3 trials were considered for safety evaluation. The FDA briefing document (5) provides an excellent and comprehensive summary of the results of oral TDF, oral FTC–TDF, and vaginal TDF clinical trials published or presented to date. Several critical issues resulted in my voting against all 3 approval indications considered by the committee.

In a modified intention-to-treat analysis done by the FDA, the iPrEx trial documented a relative risk reduction of 42% (95% CI, 22% to 63%) (3,5). Although it was a statistically significant finding, it did not exclude the possibility that the true risk reduction was 30% or less, a threshold previously discussed at scientific meetings (5). In contrast to iPrEx, the Partners PrEP study unequivocally ruled out an efficacy lower than 30%. Compared with placebo, TDF reduced the risk for HIV acquisition by 67% (CI, 44% to 81%), whereas FTC–TDF reduced the risk by 75% (CI, 55% to 87%), with no statistically significant difference seen in treatment effect between TDF and FTC–TDF (4). In both studies, by plasma and intracellular drug concentration analysis, overall adherence seemed to be low and age younger than 25 years seemed to correlate with worse adherence and reduced PrEP efficacy (5).

Of importance, data were presented that documented a statistically significant increase in condom use, a decreased number of sexual partners or acts of unprotected receptive anal intercourse, and reduced rates of syphilis for both studies. In this context, I found it difficult to get a sense of the additional benefit contributed by Truvada PrEP in reducing HIV transmission and would have liked to have had the effects of PrEP confirmed in a multiple logistic regression analysis of the data.

In contrast, the FEM-PrEP (Preexposure Prophylaxis Trial for HIV Prevention among African Women) and VOICE (Vaginal and Oral Interventions to Control the Epidemic) trials in African women, evaluating daily dosing with oral FTC–TDF (FEM-PrEP) or daily oral TDF or 1% TDF vaginal gel (VOICE), were halted because of lack of a protective effect of FTC–TDF or TDF PrEP (5 - 6). The comparison between the oral FTC–TDF and placebo groups of the VOICE trial is ongoing. The reasons for the vastly different outcomes in women in the oral PrEP trials are the focus of intense investigation.

In summary, the objective evidence for the benefit of PrEP is not consistent across all PrEP studies, and the benefit is even more questionable for women among studies conducted to date.

A second key issue is inadequate adherence based on a subset analysis of measured plasma TDF concentrations done in many studies. For iPrEx, adherence was less than 10% because 44 of 48 participants (92%) had levels that were below the limit of quantitation (5). Adherence for Partners PrEP was only slightly better at 15% because 11 of 13 participants (85%) had TDF levels that were sometimes or never measurable (5). Despite substantial counseling efforts, drug adherence was also documented to be low in the FEM-PrEP study (6).

A representative of Gilead Sciences publicly stated that greater than 90% adherence would be necessary to ensure success of Truvada PrEP. However, based on TDF drug levels, adherence was far less than 90% in both efficacy trials reviewed for the pivotal indication.

Although it was assumed that lack of measured drug equated to nonadherence, I also believe that lower or undetectable plasma drug concentrations could reflect increased TDF clearance based on body weight–serum creatinine ratios, as documented in population pharmacokinetic studies conducted in persons with HIV (7 - 8). In the safety data presented, low body weight was associated with an increased risk for TDF nephrotoxicity, which seems to confirm the potential influence of body weight on TDF clearance. Of importance, neither iPrEx nor Partners PrEP examined the pharmacokinetics of FTC–TDF in genital tract secretions, an important parameter that may have further influenced the observed HIV transmission outcomes.

Across studies, no acute safety issues were identified and rates of serious adverse events were similar between treatment and placebo groups. However, there was still extensive discussion by the committee related to concerns of long-term renal and bone toxicity associated with long-term drug exposure and how best to monitor for them. Although not part of the formal safety analysis, statistically significant declines in bone mineral density were seen in participants treated with FTC–TDF in the TDF2 study (9) and an increased risk for irreversible renal events documented in patients who have HIV with long-term TDF exposure (10) raises additional safety concerns about its long-term use for PrEP.

Concerns remain that inadvertent PrEP use in persons already infected with HIV or those who become infected will lead to resistance that could compromise first-line treatment regimens for HIV that universally recommend Truvada. Preexisting HIV infection was documented in every PrEP study; fortunately, treatment-related resistance developed in a small amount of HIV seroconverters. Although the numbers were small, the documented transmission of HIV in persons despite measurable TDF levels warrants further scientific investigation.

An effective tool used incorrectly or inconsistently is reduced to an ineffective tool. The data presented by Gilead Sciences objectively document that most persons enrolled in the 2 pivotal efficacy trials on which the PrEP approval indication is based did not use the Truvada “tool” consistently or correctly, thereby substantially limiting its potential efficacy and overall utility in preventing HIV infection.

This article was published on www.annals.org on 22 July 2012.