Antiretroviral therapy (ART) has been highly successful in slowing the progress of immunodeficiency in persons infected with HIV-1 . However, the clinical and immunologic benefit of antiretroviral therapy is compromised by the emergence of drug-resistant viruses often facilitated by incomplete virus suppression and a high mutation rate. Mutations associated with drug resistance in HIV-1 can also be transmitted to persons who are antiretroviral-naive, which can limit first-line treatment options and substantially decrease the effectiveness of subsequent antiretroviral regimens. Transmitted, or ‘primary’, drug resistance has been reported in the United States for many years [2–4]. Previous surveillance projects and studies in North America and Western Europe have estimated prevalence of transmitted drug resistance to range from 4 to 20% with the highest prevalence in countries with long-established use of ART [2,3,5–9].
Data collected from drug resistance testing help inform regional treatment guidelines for therapeutic regimens and identify potential gaps in treatment and/or prevention strategies. Data on circulating drug-resistant viruses in treatment-naive individuals also aid in the development of new drugs and inform antiretroviral treatment guidelines which currently recommend drug resistance testing after diagnosis [10,11] to better assist with the selection of first-line regimens. In addition, transmission of a drug-resistant virus could indicate that some individuals continue to engage in risky behavior while aware of their HIV-positive status. This signifies a failure both in behavioral prevention efforts  and ART. Further study of drug-resistance patterns may help identify these failures and provide opportunities for improvement of prevention practices.
Rapid viral evolution in HIV-1 impacts public health because of the virus' high genetic variability. HIV-1 group M is currently classified into 9 subtypes and 38 circulating recombinant forms . Monitoring viral diversity can provide important information to better understand HIV transmission and ensure newly developed testing assays, ART, and potential vaccine development efforts are effective for all viral subtypes . The epidemic in North America involves primarily a single HIV clade, with subtype B constituting an overwhelming majority of viruses. Previous findings have shown that non-B clades comprise between 3 and 5% of HIV diagnoses in the United States and Canada [6,7,14]. However, some areas in the United States and Canada have higher prevalence of non-B subtypes likely caused by immigration patterns and visiting international travelers [7,15,16].
Many of the previous estimates of the genetic diversity and prevalence of drug-resistant HIV-1 in the United States were performed on small convenience samples and focused on specific high-risk populations or geographic areas. Although these analyses provide useful information, a larger, more diverse group of newly diagnosed individuals would provide a better estimate of the prevalence of transmitted drug resistance and subtype diversity in the United States. The Variant, Atypical, and Resistant HIV Surveillance System (VARHS) was established to produce a population-based method of determining the prevalence of transmitted drug-resistant HIV and the distribution of subtypes among individuals newly diagnosed with HIV and reported to the national HIV case surveillance system in the United States.
Sequence collection and analysis
The Variant, Atypical and Resistant HIV Surveillance System was initially funded and incorporated into routine national HIV surveillance in 2004. The primary objective of VARHS is to collect genetic sequence data from the pol region (protease and at least the first 900 base pairs of reverse transcriptase) of HIV-1 from persons with a new diagnosis of HIV infection reported to the national HIV/AIDS reporting system (HARS). VARHS data are used to: calculate the proportion of persons newly diagnosed with HIV whose viruses have mutations associated with HIV drug resistance; identify resistance patterns; determine risk factors associated with transmission of drug-resistant strains; and gain insight into the genetic diversity of the strains and distribution of B and non-B variants.
Sequences included in this analysis were collected from surveillance areas funded for VARHS and from a companion evaluation project to determine the feasibility of performing PCR amplification on dried fluid spots (blood, serum or plasma) for surveillance purposes. These areas collected HIV-1 genetic sequence data from drug-resistance testing performed within 3 months of HIV diagnosis for antiretroviral-naive individuals reported to HARS and diagnosed at VARHS-participating facilities during 2006. Information was collected in 11 surveillance areas [Colorado, Illinois (including the city of Chicago), Louisiana, Michigan, Minnesota, Mississippi, North Carolina, Pennsylvania (excluding Philadelphia), Seattle/King County (Washington), South Carolina and Virginia] conducting confidential, name-based HIV surveillance during 2006. These surveillance jurisdictions represented approximately 18% of all newly diagnosed AIDS cases reported to HARS in 2006.
There were 2030 sequences collected through VARHS in 2006. Of the sequences collected, 1923 were collected from remnant diagnostic sera sequenced at one of three laboratories: Stanford University Clinical Virology Laboratory, Palo Alto, California; the Michigan Department of Community Health, Viral Molecular Section, Lansing, Michigan; and the Minnesota Department of Health, Molecular Epidemiology Unit, St. Paul, Minnesota. Sequencing was performed using nested polymerase chain reaction (PCR) with in-house group M consensus primers at Stanford and the Minnesota Department of Health and commercial primers  at the Michigan Department of Health. The remaining 107 sequences originated from physician-ordered drug-resistance tests and were collected from private laboratories performing genotype testing and reported to the state or local health department. These sequences were from blood drawn within 3 months of diagnosis from drug-naive persons (per the VARHS protocol described earlier). Nucleic acid sequences were aligned and translated into amino acid sequences using Sierra , a program developed by the Stanford Online HIV Drug Resistance Database (HIVDB) team, which employs the same methods as the Stanford HIVDB .
The derived nucleotide pol sequences were screened for B and non-B subtypes using a similarity method employed by the Stanford HIVDB . All sequences classified to non-B clades, as well as those classified with less than 90% similarity to reference strains, were evaluated using phylogenetic methods. A sample of subtype B sequences was also phylogenetically evaluated to confirm the similarity method classification. Nucleotide sequences were aligned using the Clustal W 1.83 multiple sequence alignment program included in the GeneStudio package . Reference strains of groups M, N, and O were extracted from the Los Alamos Databases . Phylogenetic analysis was performed by the Neighbor Joining method, with the nucleotide distance calculated by Kimura's two parameter method included in the PHYLIP package version 3.5c, with and without bootstrapping . An HIV-2 Rod sequence was used as the out group. Confirmation of subtype assignment was performed separately for each sequence. We used SimPlot software  for recombination analysis of DNA sequences.
Using the infrastructure of national HIV surveillance, each participating surveillance area collected VARHS-related data elements and submitted them monthly to CDC without personal identifiers. Information related to drug resistance and subtype was merged with the national, deduplicated HIV surveillance dataset. In the analysis, we excluded sequences derived from specimens collected more than 3 months after diagnosis and/or from persons with a history of antiretroviral drug use. We examined demographic differences in the population with sequences compared with the total population of newly diagnosed and reported individuals in the 11 VARHS jurisdictions. We also analyzed TDRM prevalence and HIV genetic diversity differences among demographic characteristics. We used X2 and Fischer's exact test when appropriate to determine significance. All statistical calculations were made using SAS version 9.1.3.
List of drug resistance-associated mutations for surveillance in the United States
We based our evaluation of TDRM on the global HIV-1 surveillance mutation list (GSML) recently updated by Bennett et al.  developed for interpreting TDRM for all subtypes. This list includes mutations that have a prevalence of at least 1% in treated persons and omits polymorphic mutations (prevalence in treatment-naïve persons of ≥0.5%) in any subtype. We adapted the GSML to include mutations that were nonpolymorphic in subtype B (i.e. had a prevalence of ≤0.5%, in treatment-naive, subtype B sequences and had a prevalence of ≥1% in treated subtype B sequences) according to the Surveillance Drug Resistance Mutations (SDRM) Worksheet on the Stanford HIVDB website . These additional mutations were included to maximize sensitivity for the primarily subtype B epidemic in the United States. To avoid overestimating the prevalence of TDRM because of mutations that are nonpolymorphic in subtype B but naturally polymorphic in other prevalent subtypes or circulating recombinant forms (CRFs), those mutations that occurred in more than 0.5% of non-B treatment-naïve sequences were excluded from the list when analyzing non-B strains for TDRM. Sequences identified as distinct from the nine pure subtypes and CRF01-AE or CRF02-AG were assigned a subtype according to nucleic acid patterns in the protease and RT genes of the pol region for purposes of evaluating transmitted drug resistance-associated mutations. Unique HIV-1 recombinants with undefined subtype in the pol region were not evaluated for TDRM.
The list of mutations (Table 1) we used for evaluating prevalence of TDRM in the United States added four mutations associated with protease inhibitor resistance at three protease positions and six mutations associated with nucleoside reverse transcriptase inhibitor (NRTI) resistance at three RT positions as compared with the GSML. This mutation list was designed for surveillance purposes and is not appropriate for use in patient clinical management.
Sequences were obtained from 2030 of 10 860 (18.7%) persons newly diagnosed with HIV and reported in the 11 surveillance areas in 2006. Thirty-three sequences were excluded because the full sequences were not available, resulting in 1997 sequences used for evaluation of TDRM. The percentage of persons with newly diagnosed HIV infections from which a sequence was collected ranged from 5.7 to 54.4% by surveillance area, with a median percentage of 20.4%. Among all persons newly diagnosed with HIV-1 reported to HARS in the 11 VARHS-funded surveillance areas during 2006, the proportion of HIV cases with sequences differed by demographic characteristics (P < 0.01) except for female transmission categories and sex (Table 2).
Among persons for whom sequence data were available, TDRM for any of the three evaluated antiretroviral drug classes occurred in 292 (14.6%) persons: 156 (7.8%) had TDRM for NNRTIs, 111 (5.6%) for NRTIs, and 90 (4.5%) for protease inhibitors (Table 3). Additionally, 241 (12.1%) persons had TDRM for a single drug class, 37 (1.9%) for two drug classes, and 14 (0.7%) for three drug classes. Four of the mutations (V11I, Q58E and T74S in protease; T69N in RT) that were included to maximize sensitivity for the primarily subtype B epidemic in the United States were borderline polymorphic in subtype B in that they occur in 0.4% (or range) of untreated persons. If these mutations were excluded from the list, the prevalence of any TDRM would decrease to 12.8%, protease inhibitor-related TDRM would decrease to 3.2% and NRTI-related TDRM would decrease to 5.1%. If only the mutations on the GSML were used to identify TDRM, the prevalence of TDRM would decrease to 12.7%. Likewise the prevalence of protease inhibitor-related TDRM would decrease to 3.0%, and TDRM for NRTI would decrease to 4.4%.
The most prevalent TDRM for NNRTIs was K103N (n = 104), which causes reduced susceptibility to all approved NNRTI drugs and occurred in 66.7% viruses resistant to any NNRTIs (Table 1). The M41L mutation (n = 35), which confers resistance to thymidine analogs, such as zidovudine and stavudine, was the second most common TDRM and the most common NRTI mutation, occurring in 31.5% viruses with TDRM in this drug class. The L90M mutation, selected by saquinavir/ritonovir and nelfinavir, was present in 24 (26.7%) of persons with TDRM for protease inhibitors. Prevalence of TDRM did not differ significantly between men and women, by age, among race/ethnicities, or by population density of the area of residence at diagnosis.
HIV-1 subtype B was the most prevalent subtype, occurring in 1922 (96.2%) of persons for whom full protease and RT sequence data were available. Non-B subtypes occurred in 75 (3.8%) individuals for whom full sequences were available. The most prevalent non-B clade was subtype C, which was found in 25 (1.3%) persons, followed by CRF02-AG, found in 21 (1.1%) persons, and subtype A, found in 9 (0.5%) persons. Additional clades including subtypes D and G, CRF01-AE, CRF10-CD, CRF12-BF, CRF22–01A1 and unique recombinants each were present in less than 0.5% of persons. The presence of TDRM was not significantly different for persons with subtype-B virus compared with persons with non-B subtype viruses (Table 4).
The study presents a standardized approach for surveillance of TDRM and analysis using a mutation list developed for the primarily subtype B epidemic in the United States. The approaches employed here build the foundation for the future analyses of drug resistance surveillance data and TDRM prevalence in the United States. Our results indicate that the prevalence of TDRM is higher than recent estimates from similarly designed datasets in other North American and European countries with long-established use of ART [7,9,14,25] and higher than previous studies in the United States [3,6]. However, interpreting these results as evidence of an increasing trend would be inappropriate  given that areas included in the current analysis differed from those included in previous analyses due to the maturation of VARHS. Further, the current analysis used a mutation list specifically suited for surveillance in the United States. Our results also indicate a higher prevalence of TDRM for NNRTI and a lower prevalence in TDRM for NRTI. This is consistent with global trends and is expected given the increased use of NNRTIs in ART ; however, continued surveillance and further research is required to understand fully these differences.
We found no significant differences in the presence of TDRM among sex, age, and race/ethnicity categories. The lack of disparity in prevalence of TDRM between non-Hispanic whites and non-Hispanic blacks differs from previous findings based on CDC surveillance data which showed non-Hispanic whites tended to have higher prevalence of drug resistance than non-Hispanic blacks [3,6]. In a previous study of 10 U.S. cities conducted by CDC among persons newly diagnosed with HIV from 1997 to 2001 , the prevalence of drug resistance was 5.4% in non-Hispanic blacks, and 13% among non-Hispanic whites. Among chronically infected drug-naive patients in 25 cities enrolled in the Terry Beirn research community program between 1999 and 2000 , non-Hispanic whites were twice as likely as non-Hispanic blacks to be infected with drug-resistant HIV. Our finding may be related to increased access to care, and therefore ARVs, among non-Hispanic blacks, as well as other racial/ethnic groups which have traditionally been underserved. The difference between our findings and previous findings could also be related to data collection procedures. A large proportion of the population with sequence information collected through VARHS was from public health facilities, and therefore may more accurately reflect the prevalence of TDRM in traditionally underserved populations.
The 3.8% prevalence of non-B subtypes and recombinant forms is consistent with previous findings in the United States and Canada [6,7,15,28–30]. The most prevalent non-B clades were C and CRF02-AG, which is also consistent with previous findings. A large majority of persons with non-B subtype among those with sequences collected were identified as immigrating from countries where non-B subtypes are prevalent. This provides a baseline for monitoring changes in the distribution of subtypes over time in the United States, and for identifying opportunities for prevention and assessing the deployment of resources for the purposes of providing patient care.
There are some limitations to consider when interpreting our findings. Our data reflect the proportion of persons newly diagnosed with HIV in the United States classified with TDRM, not the prevalence of TDRM at HIV infection, which may be underestimated. Sequence information from newly diagnosed persons with specimens drawn within 3 months of diagnosis were collected. However, newly diagnosed HIV infections in the United States range from acutely infected to chronically infected persons, the latter group comprising the majority of persons newly surveilled. Mutations that confer drug resistance can revert to wild type in the absence of drug pressure in chronically infected individuals, due to the fitness advantage for viruses without TDRM. Therefore, some of the persons with sequence data who have no detectable TDRM may have been initially infected with TDRM, but, at the time of specimen collection, such mutations were not detectable by standard genotyping methods. These cases would not be identified as having TDRM, even though these individuals may have reduced drug susceptibility once treatment is started. Using ultrasensitive methods that detect low-level resistance mutations, Johnson et al.  demonstrated the preponderance of minority resistance mutations in drug-naïve persons and documented a 40% increase in TDRM in a study of newly diagnosed persons in the United States that had a TDRM prevalence of 20% by conventional sequencing.
The number of persons with TDRM may also be overestimated due to the possibility of the presence of low-level polymorphisms included in the drug mutation list. Bennett et al.  set the exclusion cut-off for mutation prevalence in drug-naïve sequences at 0.5%. A stricter cut-off would reduce the likelihood of erroneously including natural polymorphisms on the TDRM list, but might also eliminate some other mutations truly associated with transmitted resistance. We elected to include the additional mutations to avoid excluding mutations that may be polymorphic in non-B subtypes but indicative of TDRM in subtype B. New information on polymorphic substitutions and mutations associated with new drugs are published regularly and may need to be considered in revising our mutation list in the future. In addition, changes in genetic diversity patterns may require us to adopt a list that will be more sensitive to non-B subtypes if the proportion of persons with non-B subtypes in the U.S. population increases.
The true proportion of non-B subtypes in the U.S. may be higher than our data indicate because none of the five metropolitan statistical areas with the highest numbers of African-born immigrants were among the areas submitting sequences for VARHS . Future studies from VARHS will include data from both New York City and Los Angeles County, which have the country's largest and third largest African-born immigrant populations, respectively . The inclusion of data from these areas will provide a more representative population in which to study genetic diversity in the United States.
Our results may not be representative of TDRM in the United States overall. Sequence data were collected in 10 states and 1 county. which comprised approximately 18% of AIDS cases reported in the United States in 2006 and do not include high morbidity jurisdictions such as California and New York. In addition, sequences were available from only 18.3% of newly diagnosed HIV infections in the participating surveillance areas. Finally, persons diagnosed at STD clinics and Counseling and Testing Centers had a much higher completeness of sequence data than persons diagnosed in other facility types. Given that availability of sequence data was predicated on having successful PCR amplification, the observed population may have systematically excluded persons with low viral load, insufficient quantity of sera caused by the number of other tests being administered, or other reasons.
The current study represents an analysis of the largest population-based dataset containing HIV genetic sequence data to date in the United States. To improve the ability of drug-resistance surveillance to produce a representative estimate of the prevalence of TDRM and subtype distribution in the United States, state and local health departments funded for VARHS are implementing reporting of electronic HIV genotype data from drug-resistance tests ordered as part of medical care of newly diagnosed persons. This information is collected by the health department when a new case of HIV is reported. With increases in electronic laboratory reporting, more states and cities may be inclined to include genetic sequence data among their lists of laboratory results reportable for public health purposes. Our findings support the standard of care recommendations from IAS-USA  and the U.S. Department of Health and Human Services  to include drug-resistance testing at diagnosis. Obtaining this information as early as possible in the course of the infection is important to account for mutations that may become undetectable as they revert to wild type. Future assessments can also include analysis of changes in mutation patterns and differences in identified TDRM by geographic region and infection duration. Advances in sensitive testing technology , which would enable detection of reverted mutations at minority levels, would reduce the risk of underestimating the prevalence of TDRM.
Members of the Variant, Atypical, and Resistant HIV Surveillance group: Jennifer A. Donnelly, BS (Colorado); Donna Peace (Chicago); David Barker, MD, MPH (Illinois); Samuel Ramirez, MPH (Louisiana); Mary Grace Brandt, PhD (Michigan); Luisa Pessoa-Brandao, MS (Minnesota); Sonita Singh, MPH (Mississippi); John Barnhardt, MPH (North Carolina); Godwin Obiri, DrPH, MS (Pennsylvania); Samira Kahn, MSW (South Carolina); Christina Thibault, MPH (Seattle/King County); Dena Benson, MPH (Virginia).
Author contributions: W.H.W. and R.A.Z. had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: W.H.W., R.A.Z., H.Z., D.P., J.P., U.R.B., K.C.M., W.H., J.A.J. and H.I.H. Acquisition of data: W.H.W., R.A.Z., J.P., U.R.B. and K.C.M. Analysis and interpretation of data: W.H.W., R.A.Z., H.Z., D.P., J.P., U.R.B., K.C.M., W.H., J.A.J. and H.I.H. Drafting of the manuscript: W.H.W., R.A.Z. and H.Z. Critical revision of the manuscript: W.H.W., R.A.Z., H.Z., D.P., J.P., U.R.B., K.C.M., W.H., J.A.J. and H.I.H. Administrative, technical or material support: J.P., W.H., J.A.J. and H.I.H.
Funding: The Centers for Disease Control and Prevention (CDC) funds all states and the District of Columbia to conduct HIV/AIDS surveillance and selected areas to perform Variant, Atypical and Resistant HIV Surveillance (VARHS) and provides technical assistance to all funded areas. Participating investigators and contributors from state, county or city health departments were fully or partially supported through CDC funds to states, counties or cities to conduct HIV/AIDS case surveillance and VARHS. All other participating investigators and contributors are CDC employees or contractors.
The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
1. Palella FJ Jr, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al
. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998; 338:853–860.
2. Little SJ, Holte S, Routy JP, Daar ES, Markowitz M, Collier AC, et al
. Antiretroviral-drug resistance
among patients recently infected with HIV. N Engl J Med 2002; 347:385–394.
3. Weinstock HS, Zaidi I, Heneine W, Bennett D, Garcia-Lerma JG, Douglas JM Jr, et al
. The epidemiology of antiretroviral drug resistance
among drug-naive HIV-1
-infected persons in 10 US cities. J Infect Dis 2004; 189:2174–2180.
4. Ross L, Lim ML, Liao Q, Wine B, Rodriguez AE, Weinberg W, et al
. Prevalence of antiretroviral drug resistance
and resistance-associated mutations in antiretroviral therapy-naive HIV-infected individuals from 40 United States
cities. HIV Clin Trials 2007; 8:1–8.
5. Geretti AM. HIV-1
subtypes: epidemiology and significance for HIV management. Curr Opin Infect Dis 2006; 19:1–7.
6. Wheeler W, Mahle K, Bodnar U, Pienaizek D, Kline R, Hall I, et al. Antiretroviral drug-resistance mutations and subtypes in drug-naive persons newly diagnosed with HIV-1 infection, US
, March 2003 to October 2006. In: 14th Conference on Retroviruses, Opportunistic, Infections,
Los Angeles, CA, 2007.
strain and primary drug resistance
in Canada: surveillance
report to March 31, 2005: Surveillance and Risk Assessment Division, Centre for Infectious Disease Prevention and Control,
Public Health Agency of Canada; 2006.
8. Pillay D. Current patterns in the epidemiology of primary HIV drug resistance
in North America and Europe. Antivir Ther 2004; 9:695–702.
9. Transmission of drug-resistant HIV-1 in Europe remains limited to single classes
10. Hirsch MS, Gunthard HF, Schapiro JM, Brun-Vezinet F, Clotet B, Hammer SM, et al
. Antiretroviral drug resistance
testing in adult HIV-1
infection: 2008 recommendations of an International AIDS Society-USA panel. Clin Infect Dis 2008; 47:266–285.
11. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services
. November 3, 208; 1–139. Available at http://www.aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf
. [Accessed 23 July 2009]
12. Grant RM, Hecht FM, Warmerdam M, Liu L, Liegler T, Petropoulos CJ, et al
. Time trends in primary HIV-1 drug resistance
among recently infected persons. JAMA 2002; 288:181–188.
13. Leitner T, Korber B, Daniels M, Calef C, Foley B. HIV-1 subtype and circulating recombinant form (CRF) reference sequences, 2005.
In Leitner T, Foley B, Hahn B, Marx P, McCutchan F, Mellors J, et al.
, editors. HIV Sequence Compendium 2005
. Los Alamos, NM: Theoretical Biology and Biophysics Group, Los Alamos National Laboratory; 2005. pp. 41–48.
14. Jayaraman GC, Archibald CP, Kim J, Rekart ML, Singh AE, Harmen S, et al
. A population-based approach to determine the prevalence of transmitted drug-resistant HIV among recent versus established HIV infections: results from the Canadian HIV strain and drug resistance surveillance
program. J Acquir Immune Defic Syndr 2006; 42:86–90.
15. Sides TL, Akinsete O, Henry K, Wotton JT, Carr PW, Bartkus J. HIV-1 subtype
diversity in Minnesota. J Infect Dis 2005; 192:37–45.
16. Lin HH, Gaschen BK, Collie M, El-Fishaway M, Chen Z, Korber BT, et al
. Genetic characterization of diverse HIV-1
strains in an immigrant population living in New York City. J Acquir Immune Defic Syndr 2006; 41:399–404.
19. Rhee SY, Gonzales MJ, Kantor R, Betts BJ, Ravela J, Shafer RW. Human immunodeficiency virus reverse transcriptase and protease sequence database. Nucleic Acids Res 2003; 31:298–303.
22. Lole KS, Bollinger RC, Paranjape RS, Gadkari D, Kulkarni SS, Novak NG, et al
. Full-length human immunodeficiency virus type 1 genomes from subtype C-infected seroconverters in India, with evidence of intersubtype recombination. J Virol 1999; 73:152–160.
23. Bennett DE, Camacho RJ, Otelea D, Kuritzkes DR, Fleury H, Kiuchi M, et al
. Drug resistance
mutations for surveillance
of transmitted HIV-1
drug-resistance: 2009 update. PLoS ONE 2009; 4:e4724.
25. Evidence of a decline in transmitted HIV-1 drug resistance in the United Kingdom
26. Green H, Tilston P, Fearnhill E, Pillay D, Dunn DT. The impact of different definitions on the estimated rate of transmitted HIV drug resistance
in the United Kingdom. J Acquir Immune Defic Syndr 2008; 49:196–204.
27. Novak RM, Chen L, MacArthur RD, Baxter JD, Huppler Hullsiek K, Peng G, et al
. Prevalence of antiretroviral drug resistance
mutations in chronically HIV-infected, treatment-naive patients: implications for routine resistance screening before initiation of antiretroviral therapy. Clin Infect Dis 2005; 40:468–474.
28. Weidle PJ, Ganea CE, Irwin KL, Pieniazek D, McGowan JP, Olivo N, et al
. Presence of human immunodeficiency virus (HIV) type 1, group M, non-B subtypes, Bronx, New York: a sentinel site for monitoring HIV genetic diversity in the United States
. J Infect Dis 2000; 181:470–475.
29. Jayaraman GC, Gleeson T, Rekart ML, Cook D, Preiksaitis J, Sidaway F, et al
. Prevalence and determinants of HIV-1
subtypes in Canada: enhancing routinely collected information through the Canadian HIV Strain and Drug Resistance Surveillance
Program. Can Commun Dis Rep 2003; 29:29–36.
30. Bennett D. HIV [corrected] genetic diversity surveillance
in the United States
. J Infect Dis 2005; 192:4–9.
31. Johnson JA, Li JF, Wei X, Lipscomb J, Irlbeck D, Craig C, et al
. Minority HIV-1 drug resistance
mutations are present in antiretroviral treatment-naive populations and associate with reduced treatment efficacy. PLoS Med 2008; 5:e158.
32. Malone N, KBaluja K, Costanzo J, Davis C. The Foreign-Born Population: Census 2000 Brief.
In U.S. Department of Commerce, Economics and Statistics Administration, U.S. Census Bureau; 2003.
33. Johnson VA, Brun-Vezinet F, Clotet B, Gunthard HF, Kuritzkes DR, Pillay D, et al
. Update of the drug resistance
mutations in HIV-1
. Top HIV Med 2008; 16:138–145.