Emtricitabine (FTC) or tenofovir (TFV) resistance was not observed among pre-exposure prophylaxis (PrEP) users with incident HIV infections in PrEP trials that demonstrated efficacy [1–4]. Resistance may occur more frequently when adherence to daily PrEP is ineffective, as occurred in the FEM-PrEP trial . Such resistance could undermine the activity of first-line therapies, which frequently include these agents. Concerns that suboptimal PrEP use could engender drug resistance among PrEP users and their partners have inspired caution in recommending PrEP.
Standardized clinical resistance testing methods may underestimate the true incidence of drug resistance. HIV-1 populations within people are diverse, and clinical assays characterize only variants that comprise the majority of the virus population. Nonetheless, minor variants are known to occur frequently in some clinical settings, such as people who have recently stopped therapy , have transmitted resistance , or mothers exposed to nevirapine monotherapy for prevention of vertical transmission . Such minor variants have been associated with impaired virological responses even while they are not detectable with standard assays . More sensitive detection systems have been developed based on allele-specific PCR (ASPCR) and deep sequencing.
Possible virological consequences of ineffective PrEP use in FEM-PrEP were evaluated including plasma HIV-1 RNA level, time from RNA appearance to antibody detection, and drug resistance occurrence using standardized and sensitive assays for minor viral variants. Drug concentrations near the time of infection were measured directly to help distinguish primary resistance from resistance that may have been selected for during PrEP use.
FEM-PrEP was a randomized, blinded, placebo-controlled trial of daily oral FTC/tenofovir disoproxil fumarate (TDF) among women in four sites in South Africa, Kenya, and Tanzania as previously described . All seroconverters detected during the randomized treatment phase were included in this analysis. All research participants provided written informed consent in accordance with protocols that were approved by all applicable regulatory agencies .
Viral nucleic acid testing was done with the Cobas AmpliPrep/Cobas TaqMan HIV-1 v2.0 having a lower limit of quantitation of 20 RNA copies/ml (Roche Molecular Systems, Branchburg, New Jersey, USA). Quantitative HIV RNA PCR was performed on samples from the seroconversion visit and weeks 4, 8, 12, 16, 24, 36, and 52 after seroconversion. Plasma specimens collected at visits prior to the seroconversion visit were tested retrospectively until the last HIV RNA PCR-negative plasma specimen was detected. The upper layer packed cells (ULPCs) of plasma specimens testing negative using the HIV RNA PCR were analyzed for the detection of HIV-1 DNA (see Supplement, http://links.lww.com/QAD/A621). The infection window period was defined by the last visit with no detectable HIV-1 infection by nucleic acid testing and the first visit with any laboratory evidence of infection, whether it be antibody or nucleic acids or both. CD4+ absolute cell count was performed locally using Trucount tubes on BD Facscalibur.
Clinical resistance testing was done in a manner blinded to randomization group using the TRUGENE genotyping kit (Siemens Healthcare Diagnostics, Inc. Tarrytown, New York, USA) and the PhenoSense Assay (Monogram Bioscience, South San Francisco, San Francisco, USA; see supplement for interpretation). The PhenoSense assay also provides estimates of viral replication capacity, expressed as the percentage of replication relative to a drug-susceptible control in the absence of drug . If resistance to FTC or TDF was detected, subsequent specimens were tested through the end of the study. HIV-1 subtypes were assigned from the population sequences (see Supplement, http://links.lww.com/QAD/A621). Drug-resistant mutants present in a minority of the virus population were evaluated using an ASPCR designed to control for sequence heterogeneity in the primer target regions and by deep sequencing using the 454 platform  (see Supplement, http://links.lww.com/QAD/A621).
Tenofovir and FTC were measured in plasma samples using protein precipitation and LC-MS/MS detection as previously described . The lower limit of quantification was 0.25 ng/ml. Intra and interassay accuracy ranged from 96 to 112%, and intra and interassay precision ranged from 5 to 13%. The intracellular active metabolites tenofovir diphosphate (TFV-DP) and emtricitabine triphosphate (FTC-TP) were measured in ULPC samples using an LC-MS/MS method as previously described . Concentrations from these two matrices were utilized to develop an adherence algorithm to PrEP based on previously published data of TFV and TFV-DP concentrations in plasma and red blood cells (see Supplement, http://links.lww.com/QAD/A621 and [11,12]).
Participants and seroconversion
As reported previously , we randomized 2120 women, having a mean age of 24 years. Thirty-three infections occurred in the TDF/FTC arm (incidence density 4.7/100 person-years) and 35 in the placebo arm (incidence density 5.0/100 person-years), leading to an estimated hazard ratio of 0.94 [95% confidence interval (CI) 0.59–1.52, P = 0.81). The percentage of seroconverters in whom nucleic acids were first detected prior to seroconversion visit was 61% in the active arm and 54% in the placebo arm, and the median time from HIV-1 RNA detection to seroconversion was 4 weeks in both groups.
Viral load and CD4+ cell counts
At the time of first detection of infection, the mean viral load was 4.9 (N = 33, SD 1.8) and 5.1 (N = 35, SD 1.7) log10 RNA copies/ml in the active and placebo groups, respectively (P = 0.77), and did not differ in the subsequent 52 weeks. Among women in the active arm at the earliest time of HIV detection, either by nucleic acid testing or antibody testing or both, there was a trend toward less plasma HIV-1 RNA level among those with concentrations of drug commensurate with use of four or five tablets per week versus those with detectable drug at lower concentrations and those with no detectable drug in the active arm and those in the placebo arm (mean viral loads 3.5 versus 4.8 versus 5.2 log10 HIV-1 RNA copies/ml, respectively; P = 0.19). FTC/TDF was discontinued at the time of seroconversion in all participants and plasma RNA levels were similar at subsequent visits. At the seroconversion visit, CD4+ T-lymphocyte counts were 561 (N = 31, SD 243) versus 614 (N = 31, SD 288) in the active and placebo arms, respectively (P = 0.59).
Clinical drug resistance
Tenofovir resistance was not detected in genotypic or phenotypic assays. FTC-associated resistance mutations (RT M184 V/I) were detected in viruses from five study participants at the time of the seroconversion visit (Table 1): one in the placebo arm and four in the FTC/TDF arm, as reported previously . Phenotypic resistance to FTC and lamivudine (3TC) was confirmed in viruses from four of these participants, and the assay was indeterminate in one.
Phenotypic tenofovir hypersusceptibility and viral subtypes
Two of the four FTC-resistant viruses were hypersusceptible to TFV (fold change in IC50 ≤ 0.4), and the other two tended to have greater susceptibility to TFV (fold change in IC50 > 0.4 and < 1.0; Table 2); three of three with valid results had reduced replication capacity (Table 2). The HIV-1 subtype of the PR-RT sequence was C in 41 (60%), A1 in 20 (29%), D in 3 (4%), and AC or AD mosaics in 4 (6%).
Minor variant drug resistance
Allele-specific PCR and ultradeep sequencing yielded highly concordant results (Fig. 1). The resistant mutants were predominant in the virus population at the time when HIV-1 infection was first detected, either by nucleic acids or antibodies, and waned to levels below the biological cut-off (BCO) of the virus population after FTC/TDF PrEP was discontinued. Among seroconverters without evidence of FTC resistance in clinical resistance assays, analysis of resistance using minor variant assays detected the RT M184I mutation (associated with FTC resistance) at the seroconversion visit for one participant randomized to the FTC/TDF arm at a frequency just above the BCO (0.66% by PCR, <BCO by 454): FTC, TFV, and their intracellular metabolites were undetectable at this and two previous visits. Three participants in the placebo arm had virus showing minor variant FTC resistance by a single assay, one with M184I (1.9% by 454, <BCO by PCR) and two showing M184V (1.4% by 454, <BCO by PCR; <BCO by 454, and 0.7% by PCR).
Extracellular and intracellular drug concentrations
Drug concentration analysis was performed on plasma and ULPC at all active arm visits prior to seroconversion. Of the 33 seroconverters in the active arm at the end of the infection window, 21 (64%) had drug concentrations that were negligible or not detected (including three whose infection window period started with enrollment), four (12%) had drug concentrations commensurate with use of 1–2 tablets per week, eight (24%) had concentrations commensurate with use of 4–5 tablets per week, and none had drug concentrations commensurate with use of more than 5 tablets per week (Fig. 2). At the beginning of the infection window period, 20 (61%) had drug concentrations that were negligible or not detected (including 6 that were not tested because their enrollment visit was the beginning of the window period), six (18%) had drug concentrations suggesting use of 1–2 tablets per week, four (12%) had drug concentrations suggesting use of 4–5 tablets per week, and none had drug concentrations commensurate with use of more than 5 tablets per week. Three of the 33 (9%) seroconverters had cellular drug concentrations suggesting average use of 4 to 5 tablets per week at the beginning and end of the window period, of which two had no detectable TFV in plasma at the end of the infection period, indicating no recent use of PrEP.
Correlates of drug-resistant cases in the active arm
Clinical FTC resistance occurred in two of eight (25%) women with indicators of average use of four to five tablets per week at the end of the window period, in none of 4 with detectable but lower concentrations of drug, and in 2 of 21 (9%) of women with no drug detected. Of the six participants in the active arm whose infection window period started with enrollment (the visit of PrEP initiation), three (50%) had clinical FTC resistance. Two infections with FTC-resistant viruses had concomitant resistance to drugs other than TFV and FTC, suggesting primary or transmitted resistance: one infection in the placebo arm had mutations indicating resistance to zidovudine (ZDV) (RT M41L, T215DY), FTC (RT M184V), and non-nucleoside RT inhibitors (NNRTIs; RT K103N; Table 1). In two additional infections, viruses had mutations associated with primary NNRTI resistance and no evidence of FTC or TFV resistance.
The time course of clinical resistance detection is given in Fig. 1, and detailed case descriptions are provided in supplemental materials [http://links.lww.com/QAD/A621 (see supplement, http://links.lww.com/QAD/A621)].
There were five cases (7%) of FTC-resistant infections among 68 seroconverters in the FEM-PrEP study detected using clinical assays, representing an overall prevalence that is similar to the prevalence of primary or transmitted resistance in Africa (1 to 12%) . Analysis using highly sensitive assays identified one more case in the active arm, and three additional cases in the placebo arm. As with trials that showed evidence of oral FTC/TDF PrEP efficacy [1–4], this trial reveals no excess in the rate of occurrence of FTC resistance even when more sensitive assays are used .
Drug resistance can be transmitted between people or selected within a person by antiretroviral medications used for treatment or prophylaxis. One of the clinical FTC-resistant infections had several characteristics associated with recent drug selection: RT M184I was present and typically indicates recent selection by FTC or 3TC, the drug levels were among the highest observed in seroconverters, and the resistant mutant waned quickly in blood plasma (within 4 weeks) after PrEP was stopped. A second FTC-resistant infection also had moderate concentrations of drug and resistance isolated only to FTC, although the month-long delay in the waning of her RT M184V mutant is more typical of primary or transmitted resistance. The two other infections with clinical FTC resistance occurred with no recent drug exposure at the time of seroconversion, although one woman had missed her visit prior to seroconversion, so drug exposure during the infection window could not be ruled out. One infection had evidence of transmitted drug resistance to NNRTIs, indicating that these mutants were likely to have been transmitted rather than selected. There were no cases of TFV resistance.
Other PrEP trials indicated that the risk of drug resistance was limited to those in the viral RNA-positive/antibody-negative window period before starting PrEP [1–3]. In the FEM-PrEP study, both FTC-resistant infections that occurred in the presence of detectable PrEP drug concentrations had detectable infection by week 4 of PrEP use. In these two participants, the very rapid appearance of systemic infection in the face of detectable concentrations of PrEP drugs suggests that infection had been incubating at the time that PrEP was started. In contrast to other studies, retrospective use of sensitive standard and investigational assays for viral RNA and DNA in blood specimens did not identify incubating infection in these cases. The sensitivity of such assays may have been impaired by processing, cryopreservation, and shipment required for remote testing. Alternatively, the infections may have been in the eclipse phase between viral exposure and the first expression of virus in the blood, thought to last for 4 to 11 days [14–17].
The clinical implications of the observed FTC resistance are not yet known. The RT M184V/I mutations primarily affect 3TC and FTC while increasing susceptibility to ZDV and TFV, and have no direct impact on the activity of protease inhibitors, NNRTIs, integrase inhibitors, entry inhibitors, or fusion inhibitors. The FTC-resistant viral mutants had diminished replication capacity, which has been associated with better long-term immunological responses to therapy  and decreased rates of vertical transmission . The FTC-resistant mutations are present in untreated infections, as discerned by their rapid out-growth after starting 3TC  or by deep sequencing that revealed their prevalence between 0.07 and 0.09% . Selective enrichment of these pre-existing mutants by PrEP, even if for a few months, may increase their eventual fitness and their prevalence in the virus population after PrEP is stopped. If so, the prior circulation of M184V/I mutants in these women, whether they were acquired by transmission or enriched by PrEP, could require use of additional active agents to assure the best chance for a treatment response. Use of such second-line regimens could increase costs of treatment and side effects for the patient.
The low frequency of drug resistance in the setting of PrEP failure contrasts with the experience from nevirapine (NVP) monotherapy to prevent mother to child transmission, in which resistance is detected in 25% of women using standard clinical assays that characterize the majority viral variant , and in nearly 90% of infants when sensitive assays for minor resistant variants are used . The relatively low occurrence of drug resistance in the setting of oral FTC/TDF PrEP reflects its use in HIV-uninfected people, its high efficacy when used consistently, and the high fitness barrier to TFV and FTC resistance. Seroconversions among those offered PrEP are associated with no or low level of exposure to the drugs [1,2,24]; such low drug concentrations appear to be insufficient to select for FTC or TFV resistance in most people. Unlike NNRTI resistance which preserves viral replication capacity and persists after therapy is stopped [25,26], selected resistance to FTC impairs viral replication capacity and wanes after drug exposure ends, although likely it remains archived in tissue reservoirs [25,27].
The concentrations of drug in this trial were not sufficient to select for TFV resistance; TFV resistance was also not observed in nonhuman primate models [28,29]. This suggests that the concentration of drug required to select for TFV resistance is higher than the level required for prevention of infection. In contrast, the trend toward more frequent appearance of clinical FTC resistance among women with moderate drug concentrations suggests that risk of clinical FTC resistance occurs primarily with higher levels of oral FTC/TDF use, especially if infection is incubating when PrEP is started (as previously observed [1–3]). This finding is also consistent with nonhuman primate studies showing M184V mutations appearing when in some animals dosed daily with FTC, but not in animals receiving two doses per week .
Drug resistance was rare in the context of PrEP trials, even when adherence was low and sporadic, as in this study. In practice settings where follow-up and HIV testing may be less well controlled, drug resistance may occur more frequently. Ultimately, the risks of FTC resistance need to be weighed against the benefits of HIV prevention and engagement in care, and monitored during clinical practice.
Funding statement: This work was supported by grants from the United States Agency for International Development; the Bill and Melinda Gates Foundation; and the National Institutes of Health [AI062333, AI027763, and the Centers for AIDS Research at UCSF/Gladstone and UNC]. Gilead Sciences, Inc. donated FTC/TDF and placebo. The views expressed in this publication do not necessarily reflect those of FHI360 or the agencies funding the study. The corresponding author confirms that he had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Authors’ contributions: L.V.D., J.D., T.L., and R.G. conceived and designed the study; W.A., K.A., L.V.D., J.D., and P.D. performed the clinical study and managed specimens; A.K., M.A., G.B., K.F., I.B., and T.C. designed or conducted laboratory testing; D.T. and T.L. analyzed the data; R.G. wrote the paper; all authors critically reviewed, revised, and approved the final manuscript.
Conflicts of interest
R.M.G. receives consulting fees from Siemens, the manufacturer of the clinical genotypic test used in this report. Other authors did not report conflicts of interest.
1. Grant RM, Lama JR, Anderson PL, McMahan V, Liu AY, Vargas L, et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men
. N Engl J Med
2. Baeten JM, Donnell D, Ndase P, Mugo NR, Campbell JD, Wangisi J, et al. Antiretroviral prophylaxis for HIV prevention in heterosexual men and women
. N Engl J Med
3. Thigpen MC, Kebaabetswe PM, Paxton LA, Smith DK, Rose CE, Segolodi TM, et al. Antiretroviral preexposure prophylaxis for heterosexual HIV transmission in Botswana
. N Engl J Med
4. Choopanya K, Martin M, Suntharasamai P, Sangkum U, Mock PA, Leethochawalit M, et al. Antiretroviral prophylaxis for HIV infection in injecting drug users in Bangkok, Thailand (the Bangkok Tenofovir Study): a randomised, double-blind, placebo-controlled phase 3 trial
5. Van Damme L, Corneli A, Ahmed K, Agot K, Lombaard J, Kapiga S, et al. Preexposure prophylaxis for HIV infection among African women
. N Engl J Med
6. Deeks SG, Wrin T, Liegler T, Hoh R, Hayden M, Barbour JD, et al. Virologic and immunologic consequences of discontinuing combination antiretroviral-drug therapy in HIV-infected patients with detectable viremia
. N Engl J Med
7. Metzner K, Rauch P, Walter H, Boesecke C, Zollner B, Jessen H, et al. Detection of minor populations of drug-resistant HIV-1 in acute seroconverters (Abstract 110)
. Antiviral Ther
8. Arrive E, Newell ML, Ekouevi DK, Chaix ML, Thiebaut R, Masquelier B, et al. Prevalence of resistance to nevirapine in mothers and children after single-dose exposure to prevent vertical transmission of HIV-1: a meta-analysis
. IntJ Epidemiolo
9. Johnson J, Li J, Brant A, Bennett D, Cong M, Spira T, et al. Multidrug resistant HIV-1 are transmitted more frequently than current estimates (abstract 111)
. Antiviral Ther
10. Liegler T, Abdel-Mohsen M, Bentley LG, Atchison R, Schmidt T, Javier J, et al. HIV-1 drug resistance in the iPrEx preexposure prophylaxis trial
. J Infect Dis
11. Patterson KB, Prince HA, Kraft E, Jenkins AJ, Shaheen NJ, Rooney JF, et al. Penetration of tenofovir and emtricitabine in mucosal tissues: implications for prevention of HIV-1 transmission
. Sci Transl Med
12. Adams JL, Sykes C, Menezes P, Prince HM, Patterson KB, Fransen K, et al. Tenofovir diphosphate and emtricitabine triphosphate concentrations in blood cells compared with isolated peripheral blood mononuclear cells: a new measure of antiretroviral adherence?
. J Acquir Immune Defic Syndr
13. Hamers RL, Wallis CL, Kityo C, Siwale M, Mandaliya K, Conradie F, et al. HIV-1 drug resistance in antiretroviral-naive individuals in sub-Saharan Africa after rollout of antiretroviral therapy: a multicentre observational study
. Lancet Infect Dis
14. Niu MT, Jermano JA, Reichelderfer P, Schnittman SM. Summary of the National Institutes of Health workshop on primary human immunodeficiency virus type 1 infection
. AIDS Res Hum Retroviruses
15. Busch MP, Satten GA. Time course of viremia and antibody seroconversion following human immunodeficiency virus exposure
. Am J Med
1997; 102:117–124.discussion 125–116.
16. Kahn JO, Walker BD. Acute human immunodeficiency virus type 1 infection
. N Engl J Med
17. Cohen MS, Gay CL, Busch MP, Hecht FM. The detection of acute HIV infection
. J Infect Dis
2010; 202 (Suppl 2):S270–S277.
18. Barbour JD, Hecht FM, Wrin T, Segal MR, Ramstead CA, Liegler TJ, et al. Higher CD4+ T cell counts associated with low viral pol replication capacity among treatment-naive adults in early HIV-1 infection
. J Infect Dis
19. Eshleman SH, Lie Y, Hoover DR, Chen S, Hudelson SE, Fiscus SA, et al. Association between the replication capacity and mother-to-child transmission of HIV-1, in antiretroviral drug-naive Malawian women
. J Infect Dis
20. Frost S, Nijhuis M, Schuurman R, Boucher C, Leigh-Brown A. Evolution of lamivudine resistance in human immunodeficiency virus type 1-infected individuals: the relative roles of drift and selection
. J Virol
21. Hedskog C, Mild M, Jernberg J, Sherwood E, Bratt G, Leitner T, et al. Dynamics of HIV-1 quasispecies during antiviral treatment dissected using ultra-deep pyrosequencing
. PLoS One
22. Eshleman SH, Guay LA, Mwatha A, Cunningham SP, Brown ER, Musoke P, et al. Comparison of nevirapine (NVP) resistance in Ugandan women 7 days vs. 6-8 weeks after single-dose nvp prophylaxis: HIVNET 012
. AIDS Res Hum Retroviruses
23. Loubser S, Balfe P, Sherman G, Hammer S, Kuhn L, Morris L. Decay of K103N mutants in cellular DNA and plasma RNA after single-dose nevirapine to reduce mother-to-child HIV transmission
24. Anderson PL, Glidden DV, Liu A, Buchbinder S, Lama JR, Guanira JV, et al. Emtricitabine-tenofovir concentrations and preexposure prophylaxis efficacy in men who have sex with men
. Sci Transl Med
25. Barbour JD, Hecht FM, Wrin T, Liegler TJ, Ramstead CA, Busch MP, et al. Persistence of primary drug resistance among recently HIV-1 infected adults
26. Little SJ, Frost SD, Wong JK, Smith DM, Pond SL, Ignacio CC, et al. Persistence of transmitted drug resistance among subjects with primary human immunodeficiency virus infection
. J Virol
27. Jain V, Sucupira MC, Bacchetti P, Hartogensis W, Diaz RS, Kallas EG, et al. Differential persistence of transmitted HIV-1 drug resistance mutation classes
. J Infect Dis
28. García-Lerma JG, Cong ME, Mitchell J, Youngpairoj AS, Zheng Q, Masciotra S, et al. Intermittent prophylaxis with oral truvada protects macaques from rectal SHIV infection
. Sci Transl Med
29. Garcia-Lerma JG, Otten RA, Qari SH, Jackson E, Cong ME, Masciotra S, et al. Prevention of rectal SHIV transmission in macaques by daily or intermittent prophylaxis with emtricitabine and tenofovir
. PLoS Med