There
are concerns that the use of PrEP may lead to the spread of resistance. The
data supporting this comes, at present, from animal studies.
In a
study reported at the 13th CROI Conference in 2006, the potential of
tenofovir to produce high-level anti-HIV drug resistance in already infected
monkeys was measured.1
This found that four out of eleven monkeys dosed with high levels of tenofovir
developed the K65R mutation within just one week, and another four within six
weeks. At this point, the three non-resistant animals were taken off tenofovir
for one week and then had it reinstated, in imitation of a treatment
interruption; they too had all developed K65R by the ninth week. The initial
proportion of resistant virus was below 10% of the whole viral population in
five animals, and would not have been detected by conventional resistance
tests, though it had increased to detectable levels within four weeks.
In
a previously cited CDC study,7 some monkeys were infected, despite
receiving high doses of FTC monotherapy. These animals continued to receive FTC
after infection. Two out of six infected monkeys developed the M184V FTC
resistance mutation, even though one of them was also taking tenofovir, and did
not go on to achieve an undetectable viral load.
Veteran
virologist and anti-HIV drug resistance expert John Mellors discussed
resistance at the 2010 Microbicides Conference in Pittsburgh.2 He
admitted that, so far, “there are highly divergent opinions on whether PrEP
will spread resistance,” and added that there was only one person in a PrEP
trial with documented seroconversion while taking PrEP whose drug resistance
was measured, and this person turned out to have non-resistant ‘wild type’
virus.
However,
if one of the four PrEP due to produce results by the end of 2010 reports a
significant efficacy there would be pressure to roll out national programmes,
he said, and adopting programmes without careful monitoring could spread
resistance to tenofovir and FTC. In treatment programmes, although viral
suppression rates are at least as high as they are in high-income countries,
80% of people who are not virally suppressed acquire resistance.
Mellors
said that individuals put on PrEP who were HIV-positive would probably develop
K65R unless PrEP was given as triple-combination therapy, which at present was
probably impractical due to cost (and is untested).
He
was also concerned that the seeding of resistance into the viral population
might not be detected if resistant virus existed at very low levels, only to
reappear if treatment or PrEP regimens put HIV under selective drug pressure.
He
gave, as an example, the OCTANE trial of triple-combination therapy for the
prevention of mother-to-baby transmission of HIV. This used tenofovir/FTC in
all participants and then randomised them to receive either lopinavir/ritonavir
or nevirapine as their third drug. There were a lot more transmissions in the
nevirapine arm.
At
first these did not appear to be related to pre-existing nevirapine resistance
in the mother, but when mothers were retested using a hypersensitive resistance
assay, it was found that proportions of nevirapine resistance as low as 0.8% of
the viral population led to treatment failure and transmission. Could PrEP seed
similarly low levels of, say, M184V into the viral population?
To
compute the possibility of avoiding this scenario, some key questions had to be
answered: how transmissible was virus with M184V, K65R, or both? Would
high-concentration gels still work against tenofovir-resistant virus? And how
soon would resistant viruses decline and to what level?
Mellors
said that in order to minimise resistance:
- Pre-exposure
prophylaxis should only be prescribed in the context of voluntary counselling
and testing, in order to minimise the number of people who unwittingly take
PrEP while HIV-positive.
- People
taking PrEP should be closely monitored for seroconversion.
- People
should be educated on the dangers of sharing their drugs.
- Alternatives
to tenofovir and FTC badly needed to be developed.
- Whether
topical microbicides still worked with resistant virus needed to be evaluated
(one monkey study had shown that animals with M184V still responded to a
tenofovir/FTC microbicide).
The
best answer would be to use a combination of drugs that did not share
resistance patterns with drugs used for treatment.
Anna
Forbes, former director of the Global Campaign for Microbicides, commented that
in a situation where antiretrovirals (ARVs) for treatment were not universally
available, it was difficult to see how PrEP would not create a black market in
ARVs.
Mellors
agreed, saying an ideal country scenario would be a simultaneous rollout of the
most effective measures. If care was taken, then although there might be a rise
in resistance in the initial stages, the eventual development of more
sophisticated and powerful prevention regimens would have a chance to overcome
resistance in the end. When highly active antiretroviral therapy (or HAART) had
first started, huge levels of resistance had been feared, leading to HAART
failing within years; the development of better drugs has stopped this
happening.
However,
he said that the danger of resistance was not widely understood by public
health policy bodies, even in the USA.
Not
all experts agree that drug resistance is a likely consequence of PrEP. Robert
Grant, Principal Investigator of the iPrEx Trials, said at the Adherence
Conference in Miami
in 2010 that he did not think PrEP would lead to much resistance.3
He
acknowledged that if a person started or restarted PrEP in the first weeks of
infection, the selection of resistant mutations would be probable. But it was
unclear that poor adherence to daily PrEP (or deliberate intermittent dosing)
would lead to drug resistance. When treating a person with HIV, the virus
population is huge and so there are many opportunities for mutations to generate.
“The prevention setting is different: we’re not treating a multi-million
population of virus, we’re treating about 15ml of genital secretion,” he said.