The development of drug resistance among patients on antiretroviral therapy (ART) remains an important challenge in the management of HIV disease. Patterns of drug resistance mutations in response to antiretroviral drugs and their impact on ART management have been well characterized in patients with HIV-1 subtype B in the developed world,1 but less frequently in patients from resource-limited settings who bear a disproportionate burden of the AIDS epidemic and are infected by a diverse pool of non-B HIV-1 subtypes. The President's Emergency Plan For AIDS Relief (PEPFAR) was established in 2004 in Nigeria and genotypic analysis of patients with ART failure has been conducted since September 2005, initially supported by the National Institutes for Health. We describe preliminary results from these genotype analyses and provide a more detailed description of mutational and clinical characteristics in a smaller cohort of patients in which the K65R mutation was observed. The K65R mutation is selected by tenofovir (TDF), didanosine, abacavir, and possibly stavudine (d4T) usage and causes a variable loss of susceptibility to TDF, didanosine, and abacavir depending on the presence of thymidine analogue mutations (TAM).2-4 It has been described frequently in both populations with HIV-1 subtype B5-8 and non-B subtype infection.2,9-11 More recently, the K65R mutation has also been reported in non-subtype B patients receiving non-TDF-containing first-line antiretroviral regimens.9,12 This finding has significant implications for second-line and third-line ART regimen choices in resource-limited settings, where TDF is often reserved for use due to its relatively high cost and resultant sustainability considerations. Responses to these TDF-containing second-line or third-line regimens among patients in whom K65R developed while on first line therapy may thus be significantly impacted.
The AIDS Prevention Initiative in Nigeria Plus (APIN Plus) Harvard PEPFAR program has been providing highly active antiretroviral therapy to eligible HIV-infected patients in Nigeria since September 2004. ART is provided free of charge, and funding is provided for additional program activities including baseline assessment, clinical, immunological, and virological monitoring, and prophylaxis and treatment of opportunistic infections. APIN Plus supports and collaborates with a total of 29 sites, including 11 tertiary care hospitals in 10 different Nigerian states. Patients enrolled in the program are treated according to Nigerian national ART guidelines and international standards. Standard first-line ART regimens include d4T or zidovudine (AZT), lamivudine (3TC), plus efavirenz (EFV) or nevirapine (NVP). More recently, Truvada (TDF + emtricitabine) has been recommended as a first-line alternative in Nigeria, particularly, for patients coinfected with hepatitis B. Since program enrollment began, 46,975 patients have been initiated on or provided with ART. The Government of Nigeria's ART program began in 2002 with the provision of generic ART at many of the APIN Plus clinical centers; 6920 patients who were already on ART provided through the Nigerian Government program or privately were enrolled and continued on therapy.
All HIV-infected adult patients on ART who were enrolled at one of the APIN Plus PEPFAR-funded ART clinics and had genotype testing performed were considered for this analysis. Genotype testing has been performed as a part of patient clinical follow-up in the program since September 2005, supported by the United States National Institutes of Allergy and Infectious Diseases/National Institutes for Health. Patients who were eligible for genotype testing included those with evidence of virological failure (detectable viral load >1000 copies/mL after 6 months on first-line ART), to which they were adherent (defined as adhering to scheduled drug pickups 3 months before failure). Criteria for inclusion in this analysis were eligibility for resistance testing and the presence of one or more reverse transcriptase resistance mutations. If more than 1 genotype was performed, the first result fitting these inclusion criteria was used in the analysis.
Patients were recruited for participation and enrolled in the APIN Plus ART treatment program after written informed consent, which was subject to ethical review by the Institutional Review Boards of the University of Ibadan/University College Hospital, National Institute for Medical Research, Lagos, Jos University Teaching Hospital, University of Maiduguri Teaching Hospital, and the Harvard School of Public Health.
Study Design and Data Collection
We describe HIV-1 subtype and drug resistance mutation characteristics in a cohort of eligible patients and further characterize nucleoside reverse transcriptase inhibitor (NRTI) mutations in the subset with the K65R mutation. We also compared genotypic mutation patterns among patients with and without the K65R mutation and among patients with the K65R mutation on standard first-line ART, which did not contain TDF to those on TDF-containing first-line ART. Finally, we describe immunologic and virologic outcomes among patients with the K65R mutation, who were switched to second-line ART. Genotypic analysis was performed retrospectively on samples collected from the patient closest to the time virological failure was detected. For genotypic analysis, plasma samples from EDTA-separated blood were aliquotted and shipped in liquid nitrogen to the Harvard School of Public Health where testing was performed using Abbott's ViroSeq HIV-1 Genotyping System 2.0 assay. RNA was isolated from plasma, reverse transcribed and amplified, and sequenced on an ABI 3100 capillary system. The sequence chromatograms were then edited using Abbott's ViroSeq software. Clinical data on each patient with the K65R mutation was recorded from standardized electronic forms which capture demographic, clinical, laboratory, and therapeutic information at baseline and at each visit. Patients enrolled into the APIN Plus program who are on ART have scheduled visits every 3 months or as medically indicated until they are clinically stable and the viral load is undetectable. Thereafter, visits occur every 6 months. CD4+ cell count and viral loads, in addition to laboratory tests for toxicity monitoring, are routinely collected at each visit. Data collected for the purposes of this analysis included date of antiretroviral failure, date of first genotypic analysis, ART regimen and duration before each genotypic analysis, and ART regimen and date of ART regimen switch if switch to second-line ART occurred. CD4+ cell count (cells/mm3) and viral load (copies/mL) were also collected at the time of antiretroviral failure, initial genotypic testing, and at 6 months after ART switch.
Summary statistics were performed in Stata v.9 (StataCorp, College Station, TX). Comparative analyses were performed using Fisher exact test.
Genotype testing and subtype determination was performed on 338 patients with virologic failure. HIV-1 subtypes included: CRF02_AG [152 of 338 (44.9%)], G [128 of 338 (37.9%)], CRF06 [15 of 338 (4.4%)], A [12 of 338 (3.6%)], and other subtypes or recombinants [31 of 338 (9.2%)]. Three hundred seven of 338 patients (90.8%) had received ARV regimens containing d4T usually in combination with 3TC and NVP or EFV before genotypic testing. Eighteen patients (5.3%) had previous exposure to abacavir or didanosine, and 41 patients (12.1%) had received a prior regimen containing TDF. The most common NRTI mutations observed were M184V (301, 89.1%), K70R (91, 26.9%), D67N (75, 22.2%), T215Y (61, 18.0%), T215F (51, 15.1%), M41L (46, 13.6%), K219Q (45, 13.3%), S68G (43, 12.7%), and K65R (37, 10.9%). The most common nonnucleoside reverse transcriptase inhibitor mutations included: Y181C (168, 49.7%), K103N (123, 36.4%), G190A (89, 26.3%), A98G (66, 19.5%), K101E (59, 17.5%), V108I (52, 15.4%), and V90I (41, 12.1%).
The K65R mutation was detected in 37 of 338 patients (10.9%) (Table 1). The most prevalent HIV-1 subtype among patients with K65R was CRF02_AG (21/37; 56.8%). Twenty-four of 37 patients (64.9%) with K65R were on TDF-containing first-line ART. The remaining 13 patients (35.1%) with K65R had only received a combination of d4T or AZT, 3TC, and NVP and had no documented prior exposure to TDF. Eight of these patients were previously enrolled in the Nigerian government ART program and had been receiving the Government of Nigeria first-line regimen of d4T, 3TC, and NVP before PEPFAR enrollment. Among patients with the K65R mutation, the median duration of ART therapy before the first genotypic test was 13.4 months (range 6.4-45.2 months), the median time between detection of failure and first genotypic test was 5.72 months (range 0-26.2), and the median viral load and CD4+ cell count at the time of genotypic testing was 70,469.5 copies/mL (range 2318-1,037,171) and 84 cells per cubic millimeter (range 4-469), respectively.
A variety of NRTI mutations were observed in association with the K65R mutation (Table 2), with some notable differences between patients with and without TDF in their first-line ART regimen. Among the 13 patients on non-TDF ART, the following NRTI mutations were observed in association with the K65R mutation: Q151M, F77L, F116Y, V75I, M184V, K219R, T69del, and S68G. Among 21 patients primarily on TDF-containing ART, the following mutations were observed: M184V, M184I, S68G, A62V and Y115F, and K219E. Mutations that occurred significantly more commonly in patients on non-TDF ART than those on TDF-containing ART included Q151M complex mutations (P < 0.05) and the combination of K219R and T69del mutations, which always appeared together [6 of 13 (46.2%) vs. 0 of 21 (0%); P < 0.01]. The S68G mutation was also seen more frequently in patients on non-TDF ART, but this finding was not significant [9 of 13 (69.2%) vs. 9 of 21 (42.9%); P > 0.05]. In contrast, the M184V [15 of 21 (71.4%) vs. 3 of 13 (23.1%); P = 0.01] and Y115F mutations [7 of 21 (33.3%) vs. 0 of 13 (0%); P = 0.03] were significantly more common among patients on TDF-containing ART compared with patients on non-TDF ART. TAMs were very infrequent among both groups of patients with the K65R mutation, and there were no significant differences between them (P > 0.05).
A comparison of drug resistance mutations between patients with (n = 37) and without (n = 301) the K65R mutation was also performed. The following mutations were observed in significantly more patients with K65R: S68G, A62V, Y115F, Q151M complex, T69del, and K219R [all p ≤ 0.01]. In contrast, the M184V mutation [280 of 301 (93.0%) vs. 21 of 37 (56.8%), P < 0.01)] and TAMs, M41L [45 of 301 (15.0%) vs. 1 of 37 (2.7%), P = 0.04] and K219Q [44 of 301 (14.6%) vs. 1 of 37 (2.7%), P = 0.04], were significantly more prevalent among patients without K65R.
Twenty-five patients were switched to second-line ART after the diagnosis of antiretroviral failure to first-line ART (Table 1). The median time to switch from ART failure detection was 10.5 months (range 1.8-31.3). Nineteen of the switch regimens included TDF. The median viral load at 6 months of follow-up after ART switch among 15 of these switch patients decreased to 200 (range 200-7918) copies per milliliter. The median CD4 at 6 months among these patients increased to 221 (range 90-651) cells per cubic millimeter. There was 1 reported death in our K65R cohort with no further information. Missed drug pickups (≥1 missed monthly drug pickup during time on ART before initial genotypic test) were observed in all the patients. In addition, interruptions to therapy as a result of stockouts could not be ruled out among 6 of 8 patients who initiated first-line ART before enrollment in our program; there have been no stockouts, however, during the APIN Plus PEPFAR program.
We report results of genotype testing in patients with non-B HIV subtypes who failed first-line ART. In general, drug resistance mutations most frequently observed were similar to those described in patients infected with HIV subtype B13; however, a notable finding was the significant proportion of patients in this cohort with the K65R mutation, a number of whom had no previous exposure to TDF or other antiretroviral drugs commonly known to select for K65R. The development of K65R among patients on non-TDF ART is uncommon, especially among patients on first-line combinations commonly used in resource-limited settings (d4T or AZT plus lamividine and NVP or EFV). To our knowledge, only 2 other studies have reported K65R developing in patients on these antiretroviral combinations in resource-limited settings. Patients in these studies were predominantly infected with HIV-1 subtypes, CRF01_AE (Thailand) and C (Malawi). In the study by Sungkanuparph et al,12 the K65R mutation was observed in 8 of 122 failing patients (7%) on a first-line regimen of fixed-dose d4T, 3TC, and NVP, which was provided by the Thailand national government. Factors associated with the presence of K65R included viral load at virologic failure and duration of ART. In a second study by Ferradini et al,9 of a cohort of Malawian patients, 10 of 52 patients (5%) failing first-line d4T or AZT plus lamividine and NVP or EFV developed the K65R mutation.
The route of K65R acquisition among our patients on non-TDF ART is unknown. Transmission of the K65R mutation remains rare, especially in resource-limited settings.14,15 Selection of K65R by d4T, which was part of initial ART in all but one of the patients on non-TDF-containing regimens, has been described previously in patients with HIV-1 subtype B. In a study by Garcia-Lerma et al,16 which investigated the in vitro pathways of acquisition of resistance to d4T in a panel of viruses, the K65R mutation was selected for by d4T in 7 of 9 viruses under investigation. Clinical studies to support this finding include a study by Margot et al,17 which compared the development of resistance among treatment-naive patients receiving TDF or d4T with 3TC and EFV. The K65R mutation was noted to develop in 2 of 301 patients (0.7%) in the d4T-containing arm. In another study by Valer et al,7 2 of 53 patients (3.8%) on d4T and 3TC developed K65R. The K65R has also been observed in 2 studies conducted in Botswana and South Africa of patients infected with HIV subtype C on d4T-containing regimens. It should be noted that patients in these studies were also receiving didanosine.18,19
Novel NRTI mutation patterns were also detected in our patient cohort with K65R. They included the Q151M complex, the T69 deletion, and K219R and S68G mutations. All of these mutations were significantly more prevalent than among patients without K65R and significantly more prevalent among patients on non-TDF-containing regimens. The Q151M complex, a set of mutations known to confer multi-NRTI resistance, occurred exclusively among patients in this group. Prior studies with HIV-1 subtype B have demonstrated a strong association between K65R and Q151M.7,20-22 The combination of K65R and Q151M mutations has also been observed more frequently among patients on non-TDF-containing ART, suggesting that this particular resistance pattern may be linked to multidrug resistance (MDR) more specifically than to TDF.23,24 To support these observations, in 23 of 24 patients in our cohort on TDF-containing ART, Q151M was not present.
The T69 deletion was also observed in a significant proportion of our non-TDF-exposed patients (6 of 13; 46.2%). The significance of this unusual combination is unknown. In subtype B, it has been reported to occur in close association with S163I.25 In our patient population, with a G or CRF02_AG background, the T69 deletion appeared in association with K219R. Deletions in the β3-β4 region (codons 62-78) of HIV-1 reverse transcriptase, which confer reduced susceptibility to nucleoside analogs, have been reported previously; however, insertions in this region are more common.26 Deletions have been shown to occur most commonly in association with the Q151M complex.25,26 In a recent study, the T69 deletion increased the replicative capacity of HIV-1 variants with a MDR background.27
Twenty-one of 37 patients (56.8%) in our cohort with K65R had the M184V mutation. In most published studies, the M184V has occurred in association with K65R, although other studies have observed that the presence of M184V may actually be protective for the development of K65R.13,21 The number of patients in our cohort where this mutation occurred simultaneously with the K65R was notable, even though it was less frequent than in patients without the K65R.
Other common mutations that occurred in our patient cohort included the S68G mutation. A strong correlation between K65R and S68G was observed in a previous study by Trotta et al21 and may be one of the steps in the mutational pathway toward Q151M mutation. This may explain why it was seen more frequently among patients on non-TDF ART; those in whom the Q151M complex mutation also predominated. The mutation, S68G, was also found in high frequency in conjunction with K65R in a study by Boucher et al.23
An important observation in our analysis was the relative absence of TAMS in most of our patients with the K65R mutation. The prevalence of 2 or more TAMS was significantly higher among patients lacking the K65R mutation compared with those with the K65R mutation. The absence of TAMs with the K65R mutation has been previously reported. In a study by Parikh et al,28 which examined the frequency of K65R in a large genotype database, there was a strong negative association between TAMS and K65R with in vitro confirmation of bidirectional phenotypic antagonism. Recently, the rare association between TAMS and K65R was confirmed in vivo; furthermore, when K65R was present with TAMS, it was only found on the same genome with T215F/Y and ≥2 other TAMS in the presence of Q151M.29
The effect of K65R on responses to second-line regimens is yet to be fully ascertained. An interesting finding among our patients was that all the patients who switched to second-line TDF-containing regimens responded well to therapy at 6 months of follow-up. The successful short-term response to second-line therapy observed among these patients is likely due to the inclusion of AZT in the majority of the second-line regimens to which K65R viruses are exquisitely susceptible and the effect of lopinavir/ritonavir.30 It will be important, however, to determine whether this response is sustained over time.
Finally, a diverse number of HIV-1 subtypes were observed in our patients; however, due to small patient numbers, it is not possible to determine whether there was preferential selection of K65R in one particular subtype or another. Two in vitro studies have demonstrated a preferential selection for K65R among HIV-1 subtype C viruses after exposure to didanosine/d4T19 and TDF/3TC/didanosine31; however, some clinical data do not indicate a higher in vivo selection for K65R.11
There were a number of potential limitations to our analysis. First, our cohort of patients with the K65R mutation was small, which makes it difficult to draw conclusions about the significance of this mutation, its impact on future therapies, and whether patterns of coexisting mutations are significantly different compared with those in patients without K65R. Second, we relied on patient recall for details of previous TDF exposure, which may have been inaccurate especially among those who were already on government-provided ART at program enrollment. It is unlikely, however, that any of these patients had received TDF as it was not a recommended antiretroviral at that time, and availability of TDF within Nigeria's private sector was extremely limited. Third, the length of time between the detection of ART failure and when the first genotype was actually performed varied between patients and may have affected the number and type of acquired mutations. It should be noted that in a previous study of resistance mutations associated with similar first-line regimens, higher viral loads at the time of virological failure detection were associated with the K65R mutation.12 Finally, only 6-month follow-up data were available for patients who switched to second-line ART. A longer duration of follow-up is needed to determine the success of these regimens.
In conclusion, a significantly high rate of K65R was observed among patients infected with predominantly CRF02_AG and G HIV-1 subtypes on first-line ART, in this West African setting. Coexisting NRTI mutations were especially prevalent among patients on non-TDF-containing regimens. The presence of K65R in these situations may limit the success of TDF-containing second-line therapies, particularly in settings where routine genotypic testing is not available and the choice of ART is limited. Because K65R does in fact confer significant resistance to TDF-containing regimens, the rate of development of this drug-resistant mutation may have significant implications for ART international guidelines, where TDF is frequently reserved for second-line use. Further study is therefore needed to assess the exact prevalence of K65R among patients on non-TDF-containing ART in resource-limited settings and their impact on future therapies.
We also would like to acknowledge and thank all patients and staff in the APIN Plus Harvard PEPFAR program.
1. Johnson VA, Brun-Vezinet F, Clotet B, et al, and the International AIDS Society-USA Drug Resistance Mutations Group. Update of the drug resistance mutations in HIV-1. Top HIV Med
. 2008;16:138-145. Review.
2. Gupta R, Chrystie I, O'Shea S, et al. K65R
and Y181C are less prevelant in HAART-experienced HIV-1 subtype A patients. AIDS
3. Gallant JE, Dejesus E, Arribas JR, et al. Tenofovir DF, emtricitabine, and efavirenz vs. zidovudine, lamivudine, and efavirenz for HIV. N Engl J Med
4. Eron JJ. Managing antiretroviral therapy: changing regimens, resistance testing, and the risks from structured treatment interruptions. J Infect Dis
. 2008;197(Suppl 3):S261-S271.
5. Gallant JE, Deresinski S. Tenofovir disproxil fumarate. Clin Infect Dis
6. Winston A, Pozniak A, Mandalia S, et al. Which nucleoside and nucleotide backbone combinations select for the K65R
mutation in HIV-1 reverse transcriptase. AIDS
7. Valer L, Martin-Carbonero L, de Mendoza C, et al. Predictors of selection of K65R
: tenofovir use and lack of thymidine analogue mutations. AIDS
8. Gianotti N, Seminari E, Fusetti G, et al. Impact of a treatment including tenofovir plus didanosine on the selection of the 65R mutation in highly drug-experienced HIV-infected patients. AIDS
9. Ferradini L, Jeannin A, Pinoges L, et al. Scaling up of highly active antiretroviral therapy in a rural district of Malawi: an effectiveness assessment. Lancet
10. Brenner B, Oliveria M, Doualla-Bell F, et al. HIV-1 subtype C viruses rapidly develop K65R
resistance to tenofovir in cell culture. AIDS
11. Miller MD, Margot N, McColl D, et al. K65R
development among subtype C HIV-1-infected patients in tenofovir DF clinical trials. AIDS
12. Sungkanuparph S, Manosuthi W, Kiertibuanakul S, et al. Prevalance and risk factors for developing K65R
'mutations among HIV-1 infected patients who fail an initial regimen of fixed-dose combination stavudine, lamivudine and nevirapine. J Clin Virol
13. de Mendoza C, Garrido C, Corral A, et al. Changing rates and patterns of drug resistance mutations in antiretroviral-experienced HIV-infected patients. AIDS Res Hum Retroviruses
14. Geretti AM. Epidemiology of antiretroviral drug resistance in drug-naive persons. Curr Opin Infect Dis
. 2007;20:22-32. Review.
15. Maslin J, Rogier C, Caron M, et al. Antiretroviral drug resistance among drug-naive HIV-1-infected individuals in Djibouti (Horn of Africa). Antivir Ther
16. Garcia-Lerma JG, MacInnes H, Bennett D, et al. A novel genetic pathway of human immunodeficiency virus type 1 resistance to stavudine mediated by the K65R
mutation. J Virol
17. Margot NA, Lu B, Cheng A, et al, and the Study 903 Team. Resistance development over 144 weeks in treatment-naive patients receiving tenofovir disoproxil fumarate or stavudine with lamivudine and efavirenz in Study 903. HIV Med
18. Pillay V, Pillay C, Kantor R, et al. HIV-1 subtype C drug resistance among pediatric and adult South African patients failing antiretroviral therapy. AIDS Res Hum Retroviruses
19. Doualla-Bell F, Avalos A, Brenner B, et al. High prevalence of the K65R
mutation in human immunodeficiency virus type 1 subtype C isolates from infected patients in Botswana treated with didanosine-based regimens. Antimicrob Agents Chemother
20. Roge BT, Katzenstein TL, Obel N, et al. K65R
with and without S68: a new resistance profile in vivo detected in most patients failing abacavir, didanosine and stavudine. Antivir Ther
21. Trotta MP, Bonfigli S, Ceccherini-Silberstein F, et al. Clinical and genotypic correlates of mutation K65R
in HIV-infected patients failing regimens not including tenofovir. Journal of Medical Virology
22. Zolopa AS, Rhee Y, Shin D, et al. The association of the K65R
with Q151M and the relationship to antiretroviral treatment: results from northern California clinic population. [Abstract 83]. Antivir Ther
23. Boucher S, Recordon-Pinson P, Ragnaud JM, et al. HIV-1 reverse transcriptase (RT) genotypic patterns and treatment characteristics associated with the K65R
RT mutation. HIV Med
24. Zaccarelli M, Federico Perno C, Forbici F, et al. Q151M-mediated multinucleoside resistance: prevalence, risk factors, and response to salvage therapy. Clin Infect Dis
25. Villena C, Prado JG, Puertas MC, et al. Relative fitness and replication capacity of a multinucleoside analogue-resistant clinical human immunodeficiency virus type 1 isolate with a deletion of codon 69 in the reverse transcriptase coding region. J Virol
26. Masquelier B, Race E, Tamalet C, et al. Genotypic and phenotypic resistance patterns of human immunodeficiency virus type 1 variants with insertions or deletions in the reverse transcriptase (RT): multicenter study of patients treated with RT inhibitors. Antimicrob Agents Chemother
27. Imamichi T, Sinha T, Imamichi H, et al. High level resistance to 3'-azido-3'-deoxythimidine due to a deletion in the reverse transcriptase gene of human immunodeficiency virus type 1. J Virol
28. Parikh UM, Bacheler L, Koontz D, et al. The K65R
mutation in human immunodeficiency virus type 1 reverse transcriptase exhibits bidirectional phenotypic antagonism with thymidine analog mutations. J Virol
29. Parikh UM, Barnas DC, Faruki H, et al. Antagonism between the HIV-1 reverse-transcriptase mutation K65R
and thymidine-analogue mutations at the genomic level. J Infect Dis
30. Arribas JR, Pulido F, Delgado R, et al. Lopinavir/ritonavir as single-drug therapy for maintenance of HIV-1 viral suppression: 48-week results of a randomized, controlled, open-label, proof-of-concept pilot clinical trial (OK Study). J Acquir Immune Defic Syndr
31. Brenner BG, Oliveira M, Doualla-Bell F, et al. HIV-1 subtype C viruses rapidly develop K65R
resistance to tenofovir in cell culture. AIDS
Keywords:© 2009 Lippincott Williams & Wilkins, Inc.
drug mutations; K65R; non-B HIV-1 subtypes