Mark Mascolini (

An FDA advisory panel left little doubt that Truvada, the once-daily pill combining tenofovir and emtricitabine, will get its license extended to cover pre-exposure prophylaxis (PrEP) of HIV infection in adults with a high risk of picking up HIV sexually. But the thorny deliberations leading to that endorsement in a marathon 12.5-hour session posed several tough questions about how Truvada PrEP will work in practice
-- Will people take Truvada consistently enough to protect themselves from HIV?

-- If they take Truvada haphazardly and get infected, will resistance mutations evolve quickly?

-- Will providers religiously test people for HIV before prescribing PrEP, and how often should they retest?

-- Will Truvada PrEP work in women as well as men?

-- Will taking PrEP make people abandon condoms and other HIV-evading tactics?

-- How often must PrEP takers be monitored for kidney and bone problems?

-- Will the side effects seen with Truvada prove too high a price for healthy HIV-negative people?

The FDA's Antiviral Drugs Advisory Committee wagered that the answer to the last question will be no, at least not for sexually active people with risky liaisons. Panelists voted 19 to 3 to recommend Truvada PrEP for HIV-negative men who have sex with men (MSM) and 19 to 2 (with 1 abstention) for HIV-negative people with a positive partner (HIV-discordant couples). The vote favoring PrEP for "other individuals at risk for acquiring HIV through sexual activity" was 12 to 8 with 2 abstentions.

The tepid endorsement of PrEP for "other individuals" reflected committee discomfort with a relative lack of evidence that daily Truvada wards off HIV in people not studied in the two big placebo-controlled trials Gilead Sciences relied on to buttress its PrEP application: iPrEx enrolled 2499 MSM in South and North America, South Africa, and Thailand [1], and Partners PrEP signed up 4747 HIV-discordant couples in Kenya and Uganda [2]. A smaller trial, TDF2, assessed daily Truvada PrEP in heterosexual Batswana women and men not necessarily in partnerships [3].

HIV risk reduction with Truvada in primary analyses of those three placebo-controlled trials stood at 44% in iPrEx, 75% in Partners PrEP, and 62% in TDF2. In Partners PrEP tenofovir alone cut HIV acquisition risk 67%. But Truvada did not protect high-risk African women from HIV in the FEM-PrEP trial, largely because of poor adherence to the daily regimen in these 2120 women [4]. The VOICE trial in 5029 African women closed its tenofovir-only arm when early results showed this strategy wasn't working [5]. VOICE also shut down its 1% tenofovir vaginal gel arm, but a blinded comparison of Truvada and placebo continues.

PrEP has not been tested in US women or heterosexual men, so no one can say certainly how well Truvada PrEP will work for them or whether they will even consider taking a sometimes-toxic pill every day to shield themselves from HIV. But there seems little doubt that we'll find out. Although an advisory committee recommendation does not oblige the FDA to follow suit, the FDA probably will: Scrutinizing PrEP trial data, FDA analysts determined that those findings support the "safety and efficacy of [Truvada] for the prevention of HIV-1 infection in high-risk individuals."

The FDA presentation leading to that conclusion did dwell deliberately on toxicity data, and the agency underlined the risk of kidney trouble in its premeeting briefing conclusion: "The decision to prescribe [Truvada] for the prevention of sexual acquisition of HIV infection should carefully weigh the individual risks for acquiring HIV, their understanding of the importance of adherence to medication, and their potential for development of renal toxicity" [6].

Can PrEPers withstand the side effects?
Not hurting people more than you help them is a hallowed tenet of medical practice, cited more than once at the FDA PrEP hearing. Clinicians rarely prescribe medicines for otherwise healthy people. Probably the most notable exception relevant to HIV care is prescription of hormonal contraceptives to women living in areas with high HIV prevalence because the family-planning benefit seems to outweigh a possibly heightened risk of HIV infection, the World Health Organization says [7].

Emtricitabine is among the least toxic drugs on pharmacy shelves, but tenofovir carries two much-studied risks--kidney trouble and bone density loss. Partly because of these threats, Gilead scientists are racing to develop a potentially gentler tenofovir prodrug, labeled GS-7340, to replace tenofovir in QUAD, the 4-in-1 tablet also containing emtricitabine, elvitegravir, and cobicistat [8-10], and possible in a 2-in-1 tablet with emtricitabine. But tenofovir-containing Truvada is the PrEP pill people will have to take for now.

In the successful PrEP trials completed so far--iPrEx [1], Partners PrEP [2], and TDF2 [3]--people took Truvada for an average of about 2 years, so you can't judge the long-term side effect potential. But you can reckon the 2-year risk in PrEP takers, and you can gauge longer-term risks in HIV-positive people taking tenofovir alone, in Truvada, or in Atripla. (from Jules: see my report on 2nd generation PrEP where there are numerous additional PrEP therapies in research now, of course the efficacy & safety may or may not be better than Truvada or GS-7340. Although intermittent Truvada is an alternate being researched, discussed in my 2nd Generation PrEP Report, and may have less safety concerns regarding renal/bone side effects, but this remains to be seen & intermittent Truvada means timing it before & after sex so there are inherent risks of poor timing.)
Hardly any men in iPrEx had gastrointestinal trouble after starting Truvada, and rates of diarrhea, nausea, vomiting, and all GI disorders were similar in the Truvada and placebo groups. Men randomized to Truvada did endure abdominal pain more often than placebo takers (4% versus 2%, P = 0.01), and they lost weight slightly more often than the placebo group (3% versus 2%, P = 0.04). But all side effect rates in iPrEx must be judged in light of the poor overall adherence in that trial (see "Will people without HIV take Truvada regularly?" below). In the TDF2 study of heterosexual men and women in Botswana [3], nausea, vomiting, and dizziness affected significantly more people randomized to Truvada than placebo [6].

In its review at the PrEP hearing, the FDA noted that 7 iPrEx men interrupted Truvada because of climbing creatinine, compared with 3 men assigned to placebo. Six of these 7 Truvada takers restarted Truvada "without further incident." Four Partners PrEP participants stopped tenofovir or Truvada because creatinine clearance fell to or below 50 mL/min, while 1 stopped placebo for that reason. Creatinine clearance climbed back above 50 mL/min after tenofovir stopped. In both trials, the FDA said, creatinine changes were not linked to clinical problems or other lab abnormalities.

Hypophosphatemia was the leading moderate to severe adverse event related to study drugs in Partners PrEP, affecting 11% assigned to tenofovir alone, 14% assigned to Truvada, and 13% assigned to placebo [6]. Only 6% in any study arm had serum phosphorus levels below 2 mg/dL at any point during follow-up. Neutropenia affected slightly more people taking Truvada (18%) than tenofovir alone (15%) or placebo (13%).

Seven people permanently stopped their study drug in Partners PrEP, 6 because of grade 2 renal toxicity: 3 people taking tenofovir, 2 taking Truvada, and 1 taking placebo [6]. Five of these 6 toxicities occurred in women, and all 5 had normal estimated creatinine clearance. Treatment-emergent creatinine elevations were recorded in 5% randomized to tenofovir, 7% randomized to Truvada, and 5% randomized to placebo. Only 0.2% of Partners PrEP participants had creatinine elevations judged to be drug-related, 3 people taking tenofovir alone and 5 taking placebo.

In an iPrEx bone mineral density (BMD) substudy, average BMD rose in the placebo group during the trial, while the Truvada group had small but significantly greater losses in hip and spine BMD compared with placebo takers [6]. DEXA scans done after Truvada stopped showed BMD values inching back toward baseline levels.

In a DEXA-based BMD analysis of the CDC 4323 PrEP study in 210 MSM [11,12], 10% of study participants had low BMD (a Z score at or below -2.0) when they entered the trial [12]. Median age in these men was 41. About 70% were white and 5% black. Men who used amphetamines had almost a 6 times higher risk of low BMD, and men who used inhalants had more than a 4 times higher risk. Men taking multivitamins, calcium, or vitamin D had about a 75% lower risk of low BMD.

Over time, men randomized to tenofovir had an average 1.1% drop in BMD at the femoral neck compared with the placebo group, an average 0.8% drop at the total hip, and an average 0.7% drop at the lumbar spine. After 24 months, 13% of men randomized to tenofovir and 6% randomized to placebo had more than 5% BMD loss at the femoral neck, but that difference stopped short of statistical significance (P = 0.13).