Secondary Logo

Journal Logo

CLINICAL SCIENCE

Increase in transmitted drug resistance in migrants from sub-Saharan Africa diagnosed with HIV-1 in Sweden

Andersson, Emmia,b; Nordquist, Agnesc; Esbjörnsson, Joakimd,e; Flamholc, Leof; Gisslén, Magnusg; Hejdeman, Boh; Marrone, Gaetanoi; Norrgren, Hansj; Svedhem, Veronicai,k; Wendahl, Suzannel; Albert, Janb,c; Sönnerborg, Andersa,b,i,k

Author Information
doi: 10.1097/QAD.0000000000001763

Abstract

Introduction

Transmitted drug resistance (TDR) in HIV type 1 (HIV-1) infection was first described in 1993 [1–3]. During the first decade of antiretroviral therapy (ART) the prevalence of TDR rose to 10–15% in Europe and USA [4]. Today, HIV healthcare and first-line treatments are robust, and viral suppression is the rule in many high-income countries [5]. This has contributed to stable or declining prevalence of TDR around 8% in Europe [6,7].

In low-income and middle-income countries (LMIC), first-line ART has so far been based on nucleoside reverse transcriptase inhibitors (NRTI) and nonnucleoside reverse transcriptase inhibitors (NNRTI). A meta-regression-analysis of TDR surveillance studies in LMIC until 2010 showed a significant increase of primarily NNRTI TDR, over time in sub-Saharan Africa (SSA). The increase was steepest in east Africa reaching 7.4% prevalence 8–9 after rollout of ART. A recent update in 2016 including individuals with both TDR and other reasons to pretreatment resistance (PDR), confirmed this trend and the predictions of the prevalence of NNRTI PDR for 2016 were 15.5% in East Africa, 11% in Southern Africa and 7.2% in West/Central Africa [8].

In Sweden, a TDR prevalence of 5.6% was found in 2003–2010, with a positive association between TDR and MSM and subtype B, respectively [9]. Also, a study investigating the presence of the NNRTI mutations Y181C and K103N with a sensitive allele-specific PCR in 55 ART-naïve migrants from East Africa 2002–2013, found only two patients (3.6%) [10]. However, because of an increasing and high proportion of patients originating from SSA and diagnosed in Sweden, we hypothesized that the TDR situation in Sweden could have changed. We, therefore, investigated the prevalence and trends of TDR in patients newly diagnosed with HIV-1 2010–2016.

Methods

InfCareHIV database

The Swedish InfCareHIV database includes demographic and clinical information as well as genotyping results and the viral pol sequences obtained in routine clinical care from more than 99% of Swedish residents with known HIV infection [11]. As of March 2017, the time of the latest data extraction, 10 858 patients were registered in the database of whom 7151 were still followed up at the Swedish HIV clinics.

Study population

In the database, 2406 HIV-1 patients aged at least 18 years and with a first positive serology during 2010–2016 were recorded. Of these patients, 622 (26%) were excluded because of the lack of a pol sequence within 180 days after diagnosis and 71 (3.0%) because of self-reported ART-experience at inclusion. The study population, thus, consisted of 1713 (71%) patients (Table 1). The coverage over the study years was: 2010: 64%; 2011: 70%; 2012: 66%, 2013: 71%; 2014: 70%; 2015: 81%; 2016: 82%. The study was approved by the Stockholm Ethical Review Board (2014/928–31/2). All patient information was treated confidentially, including coded identification.

T1-6
Table 1:
Characteristics of patients with and without transmitted drug resistance.

Sequencing of HIV-1 pol

The pol sequences were generated by Sanger sequencing and base identification at Karolinska University Hospital, Stockholm, Sahlgrenska University Hospital, Gothenburg or the Swedish Institute for Infectious Disease Control, Stockholm per commercial methods (ViroSeq HIV-1 genotyping kit vs 2; Celera Diagnostics, Alameda, California, USA and ABI PRISM 3100 genetic analyzer; Applied Biosystems, Foster City, California, USA), or in-house protocols [9].

Historical sequences for extended phylogenetic analysis

To further investigate the epidemiology of TDR in Sweden over time, 1379 pol sequences from 2003 to 2010 [9] and 201 from MSM diagnosed 1992–2002 in Stockholm [12], were included in a phylogenetic analysis.

Surveillance drug resistance mutation analysis

The Stanford CPR v 6.0 tool was used to screen the pol sequences for surveillance drug resistance mutations (SDRMs), according to the WHO 2009 SDRM list [13]. In patients with detected SDRMs (n = 121), the Stanford HIVdb tool v 8.3 was used to assess their clinical impact, scored as susceptible (S), intermediately or possibly resistant (I) or fully resistant (R) [14,15]. Detection of any SDRM was classified as TDR. Of the 1713 pol sequences, 1699 had reverse transcriptase and 100 protease sequences meeting the quality demands for SDRM analysis as assessed by the Stanford CPR v 6.0 tool quality control. One thousand, six hundred and eighty-seven pol sequences had both reverse transcriptase and protease sequences of sufficient quality. All 1713 sequences were analyzed by the tool, producing quality-controlled results for reverse transcriptase, protease or both.

Subtype determination and cluster identification

Pol sequences were aligned using the Clustal algorithm and manually edited in BioEdit [16] and MEGA 6.0 [17]. The final sequence alignment length was 1056 bp. The alignment was aligned with the Los Alamos Sequence Database Subtype Reference Alignment [18] and subtypes were determined by a maximum-likelihood phylogeny reconstructed in PhyML [19] with aLRT-SH branch support [20,21]. All sequences from the study population, except 14 that were too short, were included (n = 1699). Additional maximum-likelihood phylogenies were then constructed for subtype/recombinant groups (A1, B, C, D, CRF01_AE, CRF02_AG) and analyzed for clusters. Cluster analysis was performed using Cluster Picker [22]. A cut-off for branch support of 0.90 and a distance threshold of 6% were chosen after exploratory analyses [23]. In this study, we describe only clusters of HIV-1 strains with TDR from one or more study patients.

CD4+ T-cell decline trajectory analysis for estimation of time of infection

A CD4+ T-cell decline trajectory model was used to estimate the year of infection for the study patients [24]. The model extrapolates the CD4+ T-cell decline backwards from the CD4+ count at diagnosis to estimate a probable date of infection, while accounting for patient age and region of birth (Europe, Africa or other).

Statistics

Statistical analysis was carried out with StataSE 12 [25]. Chi-square test and logistic regression were applied as adequate. Ordered logistic regression was used to assess trends over the study period. A confidence level of 95% was chosen, with P values less than 0.05 indicating statistical significance.

Results

Patient characteristics

Of 1713 patients (Table 1), a majority (64%) was male and the median age was 37 years. Median CD4+ count was 282 cells/μl (range: 0–1670). The number of patients per year varied from 209 to 270. The most common routes of infection were heterosexual 53%, MSM 34%, other/unknown 8.8%. The most common regions of infection were SSA 30%, Sweden 28%, South-East Asia 15%. HIV-1 subtype B (HIV-1B) was most common (27%) followed by HIV-1C (21%) and HIV-1CRF01_AE (19%). HIV-1B was associated with MSM and infection in Sweden (P < 0.001, chi-square test). HIV-1C was associated to origin in SSA, infection in SSA, and a heterosexual route (P < 0.001, chi-square test).

Increased transmitted drug resistance to nonnucleoside reverse transcriptase inhibitors

During the whole study period, one or more SDRMs were found in 121 of the 1713 patients [7.1% (95% confidence interval, CI 5.8–8.3%)]. TDR increased from 5.2% (95% CI 2.5–7.8%) in 2010 to 8.1% (CI 95% 4.4–12%) in 2016 (P = 0.084, ordered logistic regression), with the highest prevalence in 2014; 12% (CI 95% 8.1–17%; Fig. 1).

F1-6
Fig. 1:
Prevalence of transmitted drug resistance per drug class and year.n = absolute number of patients with any TDR per year. TDR, transmitted drug resistance.

The overall prevalences of NRTI, NNRTI and protease inhibitor TDR were 3.8% (95% CI 2.9–4.7%), 3.7% (95% CI 2.8–4.6%) and 1.2% (95% CI 0.7–1.8%), respectively. NNRTI TDR increased significantly from 1.5% (95% CI 0–2.9%) in 2010 to 6.2% (95% CI 2.9–9.5%) in 2016 (P = 0.006, ordered logistic regression). Neither NRTI nor protease inhibitor TDR showed any significant trend (Fig. 1). The prevalence of multiclass TDR was 1.5% (95% CI 0.9–2.0%) with no increase over time (data not shown).

A majority (69%) of the TDR pol sequences (n = 83) harbored one SDRM (NRTI: 30; NNRTI: 38; protease inhibitor: 15; Table 2) and 31% (n = 38) had more than one SDRM (Table 3). The number of patients with single-class SDRMs for NRTIs, NNRTIs, or protease inhibitors was n = 40, n = 41 and n = 15, respectively. SDRMs for more than one drug class were found in 21% (n = 25) of individuals with TDR.

T2-6
Table 2:
Singleton surveillance drug resistance mutations in the study population.
T3-6
Table 3:
Multiple surveillance drug resistance mutations in the study population.

M41L and K103N were the most common singleton mutations and found in 43% (n = 13) and 74% (n = 28) of the patients with single NRTI and NNRTI SDRM, respectively (Table 2). Of all MSM diagnosed with HIV-1 during the study period (n = 574), 301 (52%) were reported to be infected in Sweden and of those nine (3.0%) had a virus with M41L.

Increase of transmitted nonnucleoside reverse transcriptase inhibitor in patients from sub-Saharan Africa

Whenever all DRMs were combined, no difference in the prevalence of any TDR was seen between those infected in or outside Sweden (Table 1). The presence of any TDR and NRTI TDR, respectively, was significantly associated to HIV-1B (P < 0.001, chi-square test). A significant association (P = 0.03, chi-square test) for NRTI mutations only and infection in Sweden was found. NNRTI TDR was strongly associated to infection in SSA, and to origin in SSA (P = 0.003 and P = 0.002, respectively, chi-square test). NNRTI TDR was more common in women, 5.3% (95% CI 3.5–7%) than in men, 2.8% (95% CI 1.8–3.8%) (P = 0.010, chi-square test) and associated with HIV-1C (only K103N, n = 9; only V106M n = 3; only Y181C, n = 2; only M230L n = 1; more than one NNRTI SDRM, n = 5) (P = 0.04, chi-square test). TDR affecting more than one drug class (MDR) was associated with infection in SSA and female sex (P = 0.035 and P = 0.033, respectively, chi-square test). In separate analysis of migrants from SSA, there was, however, no association between NNRTI TDR or MDR and sex.

The prevalence of TDR in patients infected in SSA increased from 1.2% in 2010 (95% CI −1.0–3.5%) to 16% in 2016 (95% CI 5.0–27%; P = 0.003, ordered logistic regression). This was caused by a rise in NNRTI TDR from no cases in 2010 to 16% (95% CI 5.0–27%) in 2016 (P = 0.002, ordered logistic regression). TDR trends in the 522 patients who were reported to be infected in SSA are shown in Fig. 2. Of 121 patients with TDR, 46 (38%) were born in SSA and 30 of them (25%) originated from countries in East Africa.

F2-6
Fig. 2:
Prevalence of transmitted drug resistance per drug class in patients infected in sub-Saharan Africa.n = absolute number of patients with any TDR per year. TDR, transmitted drug resistance.

Transmitted drug resistance clusters in HIV-1B persist among Swedish MSM

Study patients with TDR were involved in 24 clusters, including dyads (clusters of two sequences; Supplementary Table 1, https://links.lww.com/QAD/B225), mainly among HIV-1B strains (n = 14), but also in HIV-1A1 (n = 3), HIV-1C (n = 2), and HIV-1CRF01_AE (n = 5). In HIV-1B, one dyad, three triads and nine larger clusters were found. The largest cluster consisted of 26 sequences with the M41L SDRM and involved 10 study patients and 16 older Swedish sequences. All patients in the cluster were reported to be MSM, 23 were reported to be infected in Sweden, 1 in Europe and 2 in Asia. The M41L cluster dates back to 1994 and the latest patient was diagnosed in 2015. A cluster with the K103N SDRM (n = 7) involved five study patients, all MSM, diagnosed between 2008 and 2014, of whom three were reported to be infected in Sweden. In other HIV-1B-clusters only one or few patients had TDR. In HIV-1C and HIV-1CRF01_AE only dyads were seen. In HIV-1A1, two larger clusters were seen but with only one and two sequences with TDR, respectively.

Migrants from sub-Saharan Africa were infected with transmitted drug resistant HIV-1 before arrival to Sweden

The CD4+-trajectory model produced valid estimates for year of infection for 1615 (94%) patients. Fifty-six patients with record of primary HIV infection (PHI) and 31 with record of suspected PHI were excluded from analysis, 10 patients had missing data on region of birth, and one patient had a missing CD4+ cell count. The median of estimated year for infection was 2008 (range 1991–2016) for all patients (n = 1,615), and 2006 (1991–2016) for patients infected in SSA (n = 518). In patients with TDR (n = 119), the median year of infection was also 2008 (1994–2016). No trends in TDR related to the estimated year of infection were found in the whole population or in the SSA migrant group. In individuals with TDR, born in SSA, with records of year of arrival to Sweden and a valid estimate for year of infection (n = 40), our estimates supported infection before arrival to Sweden in all cases except two. Of a total of 46 patients from SSA with TDR, 38 self-reported that they had been infected in a SSA country, two in Sweden, one in Europe and five had unknown/no recorded country of infection. The three patients who were reported or estimated to be infected after arrival to Sweden were women who had the SDRMs K103N, M184V and M46I, respectively. One of the strains clustered with another strain without TDR. The two other strains were not part of any phylogenetic cluster.

Discussion

In this study of patients newly diagnosed with HIV-1 in Sweden 2010–2016, the prevalence of TDR was 7.1% (95% CI 5.8–8.3%), which is higher than the 5.6% (95% CI 4.5–6.9%) estimate from 2003 to 2010 [9]. This was mostly related to clinically relevant NNRTI-resistance in migrants from SSA, whereas TDR was less related to MSM than previously. Our phylogenetic analyses and CD4+-trajectory model supported that the TDR had occurred before arrival to Sweden.

The majority of newly HIV-1-diagnosed patients in Sweden are migrants, and therefore, the viral characteristics of the Swedish HIV-1 epidemic mirror those in LMIC, including a diverse subtype pattern [26,27] and an increasing prevalence of TDR as recently reported in SSA by the WHO [8,28]. A similar increase of NNRTI TDR in migrants from SSA has been reported from Spain, during 1998–2010. [29] However, a low (2.6%) prevalence of NNRTI SDRMs and no association to country of infection was found in Germany during 2010–2014. [30] This difference and the recent report of an increasing rate of primary NNRTI resistance globally [8] underlines the importance of solid surveillance systems in European countries wherever there is a substantial migration of patients from LMIC, such as done in the SPREAD programme [4,6]. However, there are no published accumulated European data after 2010 and there is a need of more up-to-date reports in view of the emerging increase of primary drug resistance.

It shall be noted that SDRMs may disappear from the major viral population if the selective pressure of ART is not present. Thus, monitoring with more sensitive techniques, such as next generation sequencing, is likely to reveal higher prevalences of TDR [10]. A recent study from our group based on a CD4+ T-cell decline trajectory model, suggests that postimmigration acquisition of HIV-1 infection in migrants to Sweden is more common than reported [31]. An increasing prevalence of TDR in this context might increase the risk of TDR propagation in Sweden. However, in the present study only three out of 46 SSA migrants with TDR were either reported or estimated to have been infected after arrival to Sweden.

The estimate of TDR in MSM was similar (8.5%) as in our earlier study (9.5%) [9]. However, this was to a large extent because of the persistence of the earlier described cluster with singleton M41L [9,12], which including 2010–2016 data consists of 26 isolates diagnosed over the years. The origin can be traced back to a patient diagnosed in 1994 with M41L and T215N. The clustering pattern indicate that the 41L strain is mainly propagated among ART-naive patients, most probably not aware of their HIV status. TDR transmission from treatment-naive patients has also been described elsewhere in Europe [32,33]. Although M41L alone does not give clinically relevant resistance to antiretroviral drugs currently used in high-income countries, other clinically significant TDR patterns were present in MSM, including a cluster of sequences with K103N from patients diagnosed 2008–2014. This cluster is worrying as it has a more pronounced effect on the therapeutic options including high-level resistance to efavirenz, which is still recommended as an alternative in the Swedish and European treatment guidelines [34,35].

The presence of NRTI SDRMs has impact on several strategies for the use of antiretrovirals. There is currently an ongoing discussion whether dual therapy, for example, with lamivudine/emtricitabine (3TC/FTC) and dolutegravir, could be used in switching or induction-maintenance approaches [36]. Our finding of the M184V mutation in several patients underscores the importance of considering retrospective resistance testing in patients wherever such a dual therapy is considered. The presence of M184V and/or thymidine-analog mutations (TAM) also impacts the efficacy of preexposure prophylaxis (PrEP) with tenofovir (TDF) in combination with FTC. PrEP has been shown to be highly efficient in high-risk individuals [37,38] and any transmitted and acquired resistance to these drugs may have negative consequences for HIV prevention [39]. In our present study, three patients from SSA had the K65R. Of all our strains with SDRMs, 25 (21%) exhibited SDRMs, which cause decreased sensitivity to TDF (n = 8) or 3TC/FTC (n = 8) or both (n = 9).

Among MSM, two strains were classified as R and five as I to TDF by Stanford HIVdb because of presence of TAM. Among migrants from SSA, three strains were classified as R and four strains as I to TDF, including three patients with K65R in combination with other mutations. Also, 3TC/FTC resistance was found in 15 patients (three singleton M184V and 12 strains carrying other SDRMs). Two MSM had strains classified as R to both TDF and 3TC/FTC (Table 3). The prevalence of TDF resistance in newly diagnosed patients is thus still low but in view of the emerging K65R in viral strains of patients failing TDF-containing regimen globally and the relatively high rate of M184V [40], an increased use of PrEP using TDF/FTC must be carefully followed by surveillance of TDR to TDF and other NRTIs [39].

Our study has some limitations; SDRMs in newly diagnosed patients is not analogous with clinically relevant resistance to antiretroviral drugs. We cannot exclude the possibility that some patients with SDRMs might have been exposed to nonregistered ART, such as prophylaxis against mother-to-child transmission or earlier nonrevealed use of ART. The small number of patients who displayed advanced combinations of SDRMs might indicate such an acquired resistance. However, none of the patients had ongoing treatment at diagnosis, according to the InfCare HIV database. Also, in this register-based study, we analyzed mutations in the protease and reverse transcriptase, and did not investigate integrase mutations or TDR in minority viral variants. Our study has the advantage of minimizing selection bias by including all Swedish patients with complete data [5,41]. The Swedish InfCareHIV database has well developed quality control, and data can be extracted and analyzed in real-time to produce up-to-date estimates useful both in research and clinic.

In conclusion, our results indicate a shift in the Swedish HIV-1 epidemic. TDR in newly diagnosed MSM is relatively unchanged or slightly declining, whilst NNRTI TDR in patients born and/or infected in SSA has increased significantly during 2010–2016. The most common TDR pattern in Swedish MSM was a clonal spread of an old M41L harboring strain, whereas migrants from SSA displayed clinically relevant NNRTI mutations. Also, several cases with predicted resistance to TDF and/or FTC/3TC were documented, which is of relevance for use of PrEP.

Continued monitoring of pretreatment drug resistance in newly diagnosed HIV-1 in Sweden and other European countries is highly relevant. In depth-studies using deep sequencing techniques to detect resistant minor viral variants in migrants from SSA would contribute further to the knowledge on pretreatment resistance in this group.

Acknowledgements

Authors’ contributions: E.A., J.A. and A.S. planned the project. E.A. extracted the data, carried out the TDR analysis and was responsible for writing the manuscript. J.A. and A.S. supervised this process. E.A., A.N., J.E. and J.A. carried out phylogenetic analyses and subtyping. G.M. performed the CD4+-trajectory analysis. L.F., M.G., B.H., H.N., V.S. and S.W. contributed data to the analysis. All authors took part in writing and revising the manuscript.

Sources of support: FMM (Swedish Society for Medical Microbiology) grant 50 000 SEK, Swedish Research Council, Stockholm County Council, the Swedish Society of Medical Research.

Conflicts of interest

There are no conflicts of interest.

Data partly presented previously at HIV & Hepatitis Nordic Conference 2015, Stockholm, Sweden.

References

1. Sonnerborg A, Johansson B, Ayehunie S, Julander I. Transmission of zidovudine-resistant HIV-1. AIDS 1993; 7:1684–1685.
2. Erice A, Mayers DL, Strike DG, Sannerud KJ, McCutchan FE, Henry K, et al. Brief report: primary infection with zidovudine-resistant human immunodeficiency virus type 1. N Engl J Med 1993; 328:1163–1165.
3. Wahlberg J, Fiore J, Angarano G, Uhlen M, Albert J. Apparent selection against transmission of zidovudine-resistant human immunodeficiency virus type 1 variants. J Infect Dis 1994; 169:611–614.
4. Frentz D, Van de Vijver DA, Abecasis AB, Albert J, Hamouda O, Jorgensen LB, et al. Increase in transmitted resistance to nonnucleoside reverse transcriptase inhibitors among newly diagnosed HIV-1 infections in Europe. BMC Infect Dis 2014; 14:407.
5. Gisslen M, Svedhem V, Lindborg L, Flamholc L, Norrgren H, Wendahl S, et al. Sweden, the first country to achieve the Joint United Nations Programme on HIV/AIDS (UNAIDS)/World Health Organization (WHO) 90-90-90 continuum of HIV care targets. HIV Med 2017; 18:305–307.
6. Hofstra LM, Sauvageot N, Albert J, Alexiev I, Garcia F, Struck D, et al. Transmission of HIV drug resistance and the predicted effect on current first-line regimens in Europe. Clin Infect Dis 2016; 62:655–663.
7. Frange P, Assoumou L, Descamps D, Cheret A, Goujard C, Tran L, et al. French ANRS CO 6 PRIMO Cohort, the ANRS 147 OPTIPRIM Clinical Trial and the AC11 Resistance Study Groups; French ANRS CO 6 PRIMO Cohort the ANRS 147 OPTIPRIM Clinical Trial and the AC11 Resistance Study Groups. HIV-1 subtype B-infected MSM may have driven the spread of transmitted resistant strains in France in 2007-12: impact on susceptibility to first-line strategies. J Antimicrob Chemother 2015; 70:2084–2089.
8. World Health Organization. HIV Drug Resistance Report 2017. Available at: http://www.who.int/hiv/pub/drugresistance/hivdr-report-2017/en/: World Health Organization; 2017.
9. Karlsson A, Bjorkman P, Bratt G, Ekvall H, Gisslen M, Sonnerborg A, et al. Low prevalence of transmitted drug resistance in patients newly diagnosed with HIV-1 infection in Sweden 2003–2010. PloS One 2012; 7:e33484.
10. Ekici H, Amogne W, Aderaye G, Lindquist L, Sonnerborg A, Abdurahman S. Minority drug-resistant HIV-1 variants in treatment naive East-African and Caucasian patients detected by allele-specific real-time PCR. PloS One 2014; 9:e111042.
11. Dalgaard LS, Sogaard OS, Jensen-Fangel S, Larsen CS, Sonnerborg A, Ostergaard L. Use of InfCare HIV to identify and characterize suboptimally treated HIV patients at a Danish HIV clinic: a cross-sectional cohort study. Scand J Infect Dis 2012; 44:108–114.
12. Lindstrom A, Ohlis A, Huigen M, Nijhuis M, Berglund T, Bratt G, et al. HIV-1 transmission cluster with M41L 'singleton’ mutation and decreased transmission of resistance in newly diagnosed Swedish homosexual men. Antivir Ther 2006; 11:1031–1039.
13. Gifford RJ, Liu TF, Rhee SY, Kiuchi M, Hue S, Pillay D, et al. The calibrated population resistance tool: standardized genotypic estimation of transmitted HIV-1 drug resistance. Bioinformatics 2009; 25:1197–1198.
14. Liu TF, Shafer RW. Web resources for HIV type 1 genotypic-resistance test interpretation. Clin Infect Dis 2006; 42:1608–1618.
15. Stanford University. HIVdb Program, Version 8.3, HIV Drug Resistance Database. Available at: https://stanford.edu; 2017.
16. Hall T. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl Acids Symp Ser 1999; 41:95–98.
17. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: molecular evolutionary genetics analysis version 6.0. Mol Biol Evol 2013; 30:2725–2729.
18. Los Alamos Sequence Database. Available at: http://www.hiv.lanl.gov.
19. Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, Gascuel O. New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst Biol 2010; 59:307–321.
20. Hordijk W, Gascuel O. Improving the efficiency of SPR moves in phylogenetic tree search methods based on maximum likelihood. Bioinformatics 2005; 21:4338–4347.
21. Anisimova M, Gascuel O. Approximate likelihood-ratio test for branches: a fast, accurate, and powerful alternative. Syst Biol 2006; 55:539–552.
22. Ragonnet-Cronin M, Hodcroft E, Hue S, Fearnhill E, Delpech V, Brown AJ, et al. Automated analysis of phylogenetic clusters. BMC Bioinformatics 2013; 14:317.
23. Hassan AS, Pybus OG, Sanders EJ, Albert J, Esbjornsson J. Defining HIV-1 transmission clusters based on sequence data: a systematic review and perspectives. AIDS 2017; 31:1211–1222.
24. Brannstrom J, Sonnerborg A, Svedhem V, Neogi U, Marrone G. A high rate of HIV-1 acquisition post immigration among migrants in Sweden determined by a CD4 T-cell decline trajectory model. HIV Med 2017; 18:677–684.
25. Stata Statistical Software. Stata Statistical Software: release 2012. College Station, TX: StataCorp LP; 2011.
26. Neogi U, Haggblom A, Santacatterina M, Bratt G, Gisslen M, Albert J, et al. Temporal trends in the Swedish HIV-1 epidemic: increase in non-B subtypes and recombinant forms over three decades. PloS One 2014; 9:e99390.
27. Esbjornsson J, Mild M, Audelin A, Fonager J, Skar H, Bruun Jorgensen L, et al. HIV-1 transmission between MSM and heterosexuals, and increasing proportions of circulating recombinant forms in the Nordic Countries. Virus Evol 2016; 2:vew010.
28. Gupta RK, Jordan MR, Sultan BJ, Hill A, Davis DH, Gregson J, et al. Global trends in antiretroviral resistance in treatment-naive individuals with HIV after rollout of antiretroviral treatment in resource-limited settings: a global collaborative study and meta-regression analysis. Lancet 2012; 380:1250–1258.
29. Yebra G, de Mulder M, Perez-Elias MJ, Perez-Molina JA, Galan JC, Llenas-Garcia J, et al. Increase of transmitted drug resistance among HIV-infected sub-Saharan Africans residing in Spain in contrast to the native population. PloS One 2011; 6:e26757.
30. Hauser A, Hofmann A, Hanke K, Bremer V, Bartmeyer B, Kuecherer C, et al. National molecular surveillance of recently acquired HIV infections in Germany, 2013 to 2014. Euro Surveill 2017; 22: pii: 30436.
31. Brännström J, Sönnerborg A, Svedhem V, Neogi U, Marrone G. A high rate of HIV-1 acquisition post immigration among migrants in Sweden determined by a CD4 T-cell decline trajectory model. HIV Med 2017; 18:677–684.
32. Mbisa JL, Fearnhill E, Dunn DT, Pillay D, Asboe D, Cane PA. Evidence of self-sustaining drug resistant HIV-1 lineages among untreated patients in the United Kingdom. Clin Infect Dis 2015; 61:829–836.
33. Drescher SM, von Wyl V, Yang WL, Boni J, Yerly S, Shah C, et al. Treatment-naive individuals are the major source of transmitted HIV-1 drug resistance in men who have sex with men in the Swiss HIV Cohort Study. Clin Infect Dis 2014; 58:285–294.
34. EACS. EACS Guidelines version 8.2. January 2017.
35. RAV. Antiretroviral behandling av hiv infektion 2016– Behandlingsrekommendation. Available at: http://www.sls.se/globalassets/rav/rekommendationer/rav_hiv_2016_160524.pdf; 2016.
36. Cahn P, Rolon MJ, Figueroa MI, Gun A, Patterson P, Sued O. Dolutegravir-lamivudine as initial therapy in HIV-1 infected, ARV-naive patients, 48-week results of the PADDLE (Pilot Antiretroviral Design with Dolutegravir LamivudinE) study. J Int AIDS Soc 2017; 20:1–7.
37. World Health Organization. WHO Guidelines approved by the Guidelines Review Committee. Guideline on when to start antiretroviral therapy and on pre-exposure prophylaxis for HIV. Geneva, Switzerland: WHO; 2015.
38. Cairns G, McCormack S, Molina JM. The European preexposure prophylaxis revolution. Curr Opin HIV AIDS 2016; 11:74–79.
39. van Tienen C, van de Vijver D, Noori T, Sonnerborg A, Boucher C. Letter to the editor: preexposure prophylaxis for HIV in Europe: the need for resistance surveillance. Euro Surveill 2017; 22: pii: 30483.
40. Hosseinipour MC, Gupta RK, Van Zyl G, Eron JJ, Nachega JB. Emergence of HIV drug resistance during first- and second-line antiretroviral therapy in resource-limited settings. J Infect Dis 2013; 207 (suppl 2):S49–S56.
41. Haggblom A, Svedhem V, Singh K, Sonnerborg A, Neogi U. Virological failure in patients with HIV-1 subtype C receiving antiretroviral therapy: an analysis of a prospective national cohort in Sweden. Lancet HIV 2016; 3:e166–e174.
Keywords:

antiretroviral drugs; drug resistance; HIV; sub-Saharan Africa; Sweden; transients and migrants; transmission

Supplemental Digital Content

Copyright © 2018 Wolters Kluwer Health, Inc.