Primary infection with drug-resistant HIV
Resistance Workshop Review: Part 4
Cohort studies from North America, Western Europe, and Australia suggest that primary infection with antiretroviral-resistant HIV has stabilized at around a 10%, but statistics vary from country to country and sometimes within larger countries. Going home with your sex partner's resistant virus can spoil chances of responding to a first antiretroviral combination, even after the resistant mutants have faded to a small slice of a your viral population-a slice so thin that standard genotyping can't spot it. One workshop study tied methamphetamine use to infection with resistant virus, while another found that people claiming no treatment experience may have surreptitiously sampled one or more antiretrovirals-a ready recipe for resistance.
Hidden mutations can wreck first-line response
Standard resistance testing may not be good enough to screen antiretroviral-naive people for mutations that can ruin their response to a first regimen, according to results of a Centers for Disease Control (CDC) study . A highly sensitive real-time polymerase chain reaction (PCR) assay smoked out more than twice as many potentially harmful mutations in 138 previously untreated people enrolled in a 2001-2002 GlaxoSmithKline trial. Everyone who started treatment with the K103N or Y181C nonnucleoside (NNRTI) mutation or the M184V mutation evoked by 3TC or FTC had a virologic failure.
Jeffrey Johnson and CDC colleagues focused on 138 people who had a viral load above 50 copies/mL in the first 48 weeks of treatment with abacavir, 3TC, and efavirenz and 138 people who kept HIV under wraps for 48 weeks. Technicians blinded to 48-week response status scrutinized pretreatment samples first with a standard genotyping test, then with the PCR assay aimed at spotting peewee viral populations of K103N, Y181C, M184V, and K70R, a thymidine analog mutation that would not affect response to this regimen.
Conventional gene sequencing saw K103N in 2 people (1.4% of the 138 people with virologic failure), Y181C in 1 (0.7%), and M184V in 1 (0.7%) (Table 1). Real-time PCR rousted out pretreatment K103N in another 2 people, Y181C in another 1, and M184V in another 2. That meant standard genotyping picked up only 4 of 12 critical mutations (33%) that may sabotage a first-line regimen.
The standard assay tracked down K70R in 1 person, while real-time PCR unmasked K70R in 9 others. Although K70R has no immediate impact on people starting a combination lacking the thymidine analog AZT or d4T, it could well imperil response to a later regimen containing those drugs.
All 10 people who got infected with resistant virus had a poor response to abacavir, 3TC, and efavirenz. Two of them never reined in viral replication, 4 had a failure within 2 months, 2 more within 4 months, and the last 2 within 6 months. Treatment flunked within 4 months in the 1 person with dual-class resistance (M184V plus K103N plus Y181C).
Statistical analysis showed that infection with resistant virus correlated with virologic failure independently of pretreatment viral load and CD4 count.
Johnson and coworkers believe their findings imply that more sensitive pretreatment resistance testing can improve responses to first-line therapy in places with frequent transmission of resistant virus. And, as studies reviewed below show, those places include many areas of Western Europe, North America, and Australia.
Another crystal meth risk: resistance from day 1
Everyone suspects that crystal methamphetamine poses grave threats to people at risk for HIV infection, and research has begun confirming that suspicion. At the 2005 Resistance Workshop, a group from the University of California, San Francisco linked crystal use to a 4 times higher risk of unprotected sex involving men or women with HIV . Around the time of this year's workshop, a team from the University of California, San Diego (UCSD) tied the drug to higher scores on a sexual compulsivity index, then tied higher scores to meth use before or during sex, cruising sex clubs and street corners for sex partners, lower self-esteem, higher scores on a disinhibition scale, and more HIV-negative or unknown-serostatus partners .
At this year's Resistance Workshop, Lydia Drumright and UCSD coworkers verified the logical consequence of risky sex with meth-infection with resistant HIV -a unprepossessing plight heretofore seen only in an isolated case report . From May 2002 through March 2006, Drumright asked 214 recently infected men who have sex with men to complete computer-assisted surveys about substance use with their last 3 sex partners. She genotyped virus from these men with a standard assay and defined resistance as a score of 60 or more according to the Stanford University HIV resistance algorithm.
Thirty-two men (15%) had mutations indicating primary infection with resistant HIV, and 57% of the study group reported using illicit substances with one or more of their last 3 partners-32% used nitrites, 31% meth, 26% marijuana, 12% gamma hydroxy butyrate.(GHB), 6% cocaine, and 2% ketamine. Men infected with resistant virus did not differ from those infected with drug-susceptible virus in sexual history or demographics such as age, ethnicity, education, or employment.
A statistical analysis that weighs various resistance risk factors at the same time determined that methamphetamine use before sex independently raised the risk of infection with resistant virus 4.29 times (95% confidence interval 1.3 to 13.1, P = 0.01). No other party drug significantly boosted the risk of primary resistance, but lower income did. Compared with men who earned $51,000 or more yearly, those who made $10,000 to $30,000 had a 4.8 times higher risk of primary resistance (P = 0.03), and those who made less than that had a 5.5 times higher risk (P = 0.01).
Drumright cautioned that her findings are preliminary, but they seem unlikely to surprise anyone. And as the CDC study  suggests, Drumright's use of standard genotyping may have missed smaller populations of resistant virus that could affect response to therapy.
When Martin Markowitz and colleagues reported a case of infection with multidrug-resistant HIV in a crystal meth fancier , many dismissed the finding as a blip on the resistance radar screen. In light of Drumright's findings, Robert Grant (University of California, San Francisco) asked at the workshop whether the Markowitz case was no blip, but a "sentinel of an epidemic of resistance in meth users."
Stable rates of resistance transmission
Nine studies of antiretroviral-naive people in developed countries-most of them diagnosed in 2000 or later-found the prevalence of transmitted resistant virus hovering around 10% [6-14]. In country- (or continent-) wide studies, rates of resistance to any antiretroviral ranged from 8% in Switzerland (1996-2005)  to 9% in Germany (2001-2005)  and across Europe (2002-2003) , to 10% in a US study (2000-2004) , and to 13% in a North American-Australian survey (2000-2006) .
The pan-European SPREAD study of 1083 newly diagnosed people in 17 countries saw more resistance to nucleosides (NRTIs) (5%) than to protease inhibitors (PIs) (3%) or NNRTIs (3%) , while German  and Swiss  studies mirrored that rank order. The North American-Australian study turned those trends on end, finding higher rates of resistance to NNRTIs than to the other two classes . But it turned out that people from California, who made up 51% of the cohort, explained that difference from the European cohorts.
The North American-Australian survey, presented by Susan Little (University of California, San Diego), involved viral samples from 1542 untreated people, 1250 of whom had a genotypic test done an average 104 days after their estimated date of infection and 1033 of whom had their virus phenotyped an average 88 days after infection . The study group was largely male (94%) and non-Hispanic white (71%) and picked up HIV sexually (88%) (see note 15 for the study sites).
From 1996 to 2006 overall genotypic resistance climbed from 9% to 13% in this group (P = 0.04), reflecting an overall rise in genotypic resistance to NNRTIs from 5% to 11% (P = 0.002). Phenotypic resistance to NNRTIs jumped from 4% to 9% over the course of observation (P = 0.003), while phenotypic resistance to NRTIs fell from 9% to 3% (P < 0.001).
When Little and colleagues split the cohort into three groups-California (51%), New York (17%), and others (32%)-they found that the overall jump in genotypic resistance and the rise in genotypic resistance to NNRTIs proved statistically significant only in California, whereas the falloff in phenotypic resistance to NRTIs reached statistical significance only in New York and the other sites. Virus resistant to more than one antiretroviral class held steady during the study period at rates of 2% to 4%.
Genotyping virus from 1795 treatment-naive people who signed up for GlaxoSmithKline trials in the US from 2000 through 2004, Glaxo's Lisa Ross confirmed a significant vault in any primary resistance mutation (5% to 13%, P = 0.009) and in NNRTI-induced mutations (2% to 7%, P = 0.0387). Resistance rates in this analysis did not vary greatly from one region to another-11% in the South, 9% in the Northeast, 8% in the Midwest, and 7% in the West.
Like the North American-Australian cohort , most of these Glaxo trial enrollees got infected sexually (66% homosexual and 34% heterosexual) and were mostly men (84%). But the Glaxo cohort was more diverse ethnically-48% white, 37% black, and 14% Hispanic. Although Ross found a significantly lower overall resistance rate in black study participants, in 2004 resistance rates varied little between whites (13%) and blacks (11%). In that year 17% of treatment-naive Hispanics had resistant HIV. Also as in the North American-Australian study, rates of two- or three-class resistance remained low throughout Ross's study.
Because of the surging rate of primary infection with NNRTI-resistant virus, Ross and colleagues suggested that "if baseline genotyping cannot be done, then use of an NNRTI-sparing first-line regimen should be considered." This is not just agitprop from a company that markets no NNRTI. Others including Daniel Kuritzkes (Brigham and Women's Hospital, Boston) mulled the same idea at the workshop.
Infection with enfuvirtide-resistant virus
When researchers studying transmission of resistant HIV talk about multiclass resistance, they mean resistance to two or three of the first three antiretroviral families-NRTIs, NNRTIs, and PIs. But now that term must extend to the fourth class, entry inhibitors, Bordeaux University's Bernard Masquelier reported in a study analyzing the Aquitaine cohort .
Masquelier genotyped 55 antiretroviral-naive people, 50 of them born in France and 5 in Africa. Eleven people (20%) turned out to be infected with drug-resistant virus. Five had mutations conferring resistance to NRTIs, 5 had NNRTI mutations, 2 had PI mutations, and 2 had mutations in the HR1 region of HIV-1 gp41 that make virus resistant to enfuvirtide.
Virus from 1 person carried the N42D mutation in gp41, along with NRTI and PI mutations. Another had the G36D mutation in gp41 and no other mutations. Masquelier estimated that both men picked up HIV in the summer of 2004, and both did so during sex with other men. He tracked the sex partner of the man with triple-class resistance and determined that the partner-who also never took antiretrovirals-carried a closely related virus. In all probability the source partner had himself been infected by a treated person with multiclass-resistant virus.
Research shows that a failing enfuvirtide regimen selects the gp41 mutations Masquelier identified in these men. These substitutions do not appear as natural viral variants in enfuvirtide-naive people.
Don't sample your friend's efavirenz
Standard genotyping of 103 HIV-infected people in San Diego County who reported no antiretroviral experience found that 26 of them (25%) harbored virus with at least one resistance mutation, that 6 (6%) had two-class resistance, and that 1 (1%) had three-class resistance . But these disturbingly high rates are not the key finding of this man-bites-self story detailed by Davey Smith (University of California, San Diego).
When Smith saw that two viral samples bearing identical mutation patterns came from the same clinic, he called the physician in charge and asked whether that group mistakenly submitted a sample from the same person twice. Further checking showed that the two samples did indeed come from two individuals, who happened to be roommates. The men confessed borrowing a friend's antiretrovirals when they were feeling "poorly," just to swat down the virus for a day or so.
Surprised to learn that his study group included people who purportedly got infected with resistant virus but actually bred their resistance "the old-fashioned way," Smith began questioning people in his own clinic. He learned that borrowing antiretrovirals for short-term self-treatment was not unique to the first 2 men who divulged this stratagem.
As an example, Smith mentioned the case of one previously untreated HIV-infected man who took his partner's efavirenz when some wheezing struck him as the first sign of Pneumocystis pneumonia-even though he had a robust CD4 count.
Do people who borrow antiretrovirals suffer from limited education?
"They're smarter than I am," Smith averred, except, apparently, when it comes to HIV resistance.
Mark Mascolini writes about HIV infection (email@example.com).
References and Notes
1. Johnson JA, Li J-F, Wei X, et al. Baseline detection of low-frequency drug resistance-associated mutations is strongly associated with virological failure in previously antiretroviral-na•ve HIV-1-infected persons. Antivir Ther 2006;11:S79.
2. Chin-Hong PV, Deeks SG, Liegler T, et al. Methamphetamine use is associated with ongoing high-risk sexual behaviors among HIV-infected individuals with drug-resistant virus. XIV International HIV Drug Resistance Workshop. June 7-11, 2005. QuŽbec. Abstract 117.
3. Semple SJ, Zians J, Grant I, Patterson TL. Sexual Compulsivity in a Sample of HIV-Positive Methamphetamine-using Gay and Bisexual Men. AIDS Behav June 6, 2006; Epub ahead of print.
4. Drumright LN, Gorbach PM, Frost SDW, et al. Transmitted HIV drug resistance is associated with methamphetamine use among recently HIV infected MSM in Southern California, USA. Antivir Ther 2006;11:S112.
5. Markowitz M, Mohri H, Mehandru S, et al. Infection with multidrug resistant, dual-tropic HIV-1 and rapid progression to AIDS: a case report. Lancet 2005;365:1031-1038.
6. Wensing AMJ, Vercauteren J, van de Vijver DA, et al. Transmission of drug-resistance in Europe is characterized by single mutations and revertants. Antivir Ther 2006;11:S111.
7. Garcia-Diaz A, Booth C, Nebbia G, et al. Transmitted drug resistance clusters with the
infecting HIV-1 subtype: a single centre analysis of all new HIV-1 diagnoses in London. Antivir Ther 2006;11:S126.
8. Oette M, Kaiser R, Daumer M, et al. Trends of primary drug resistance in chronically
HIV-infected patients in Germany, 2001-2005. Antivir Ther 2006;11:S125.
9. Yerly S, von Wyl V, Boni J, et al. Transmission of HIV-1 drug resistance in
Switzerland: a 10-year molecular epidemiology survey. Antivir Ther 2006;11:S118.
10. Little SJ, May S, Hecht F, et al. Increase in transmitted NNRTI drug resistance
among recently HIV infected patients from North America and Australia. Antivir Ther 2006;11:S110.
11. Ross LL, Florance A, Wine B, et al. Prevalence of HIV-1 drug resistance-associated
mutations in a large cohort of antiretroviral therapy (ART) naive HIV-infected individuals in the United States from 2000-2004. Antivir Ther 2006;11:S120.
12. Eshleman SH, Husnik M, Hudelson S, et al. Analysis of antiretroviral drug resistance and HIV-1 subtype among men who have sex with men recently infected with HIV-1 in the United States: the EXPLORE Study. Antivir Ther 2006;11:S122.
13. Bennett DE, Smith AJ, McCormick L, et al. Categorization of transmitted HIV drug resistance using the WHO/CDC HIV drug resistance threshold survey method. Antivir Ther 2006;11:S116.
14. Truong HM, Klausner JD, Hecht FM, Grant RM. Reduced levels of primary resistance to nRTIs in San Francisco is discernable using two independent sentinel populations. Antivir Ther 2006;11:S115.
15. Sites that accounted for more than 5% of the total cohort were San Diego (22.5%), San Francisco (17%), New York (17%), Los Angeles (11%), Australia (7%), and Montreal (6%). Sites that contributed fewer than 5% were Baltimore, Birmingham (Alabama), Boston, Brazil, Brooklyn, Dallas, Denver, Miami, Seattle, and Vancouver.
16. Peuchant O, Capdepont S, Ragnaud JM, et al. Primary resistance to enfuvirtide in recently infected, antiretroviral-naive patients, ANRS CO3 Aquitaine cohort. Antivir Ther 2006;11:S108.
17. Smith D, Pesano R, Cachay E, et al. Prevalence of HIV drug resistance among antiretroviral naive individuals of unknown infection duration. Antivir Ther 2006;11:S121.