By 2032, assuming current levels of diagnosis and retention and ART initiation eligibility criteria are maintained, 3.3 million are projected to be receiving ART, with 35% (1.1 million) on second-line regimens. Changing the threshold at which a person is eligible for treatment is predicted to have minimal impact on the number of people on ART and on the number requiring second-line regimens. However, under a scenario of enhanced diagnosis and retention, between 4.9 and 5.2 million will be on ART, of whom 40–42% (2.0–2.1 millions) on a second-line regimen.
Figure 1b shows the estimated number of people living with NNRTI-resistant virus in 2012 and projected numbers in 20 years’ time. In 2012, 652 000 are estimated to have NNRTI resistance, of whom 42% (approximately 275 000) are people with NRMV with viral load above 500 copies/ml (see footnote of Figure 1b), and therefore have an increased risk transmitting NNRTI drug-resistant virus . If existing policies continue, the number of people living with NNRTI resistance is predicted to be 2.8-fold higher in 2032 (from 652 000 in 2012 to 1 862 000). Of these, 68% are predicted to be on ART with suppressed viral load or to have resistance in minority virus. Therefore, the subset of the population with an increased risk of transmitting NNRTI-resistant virus (viral load >500 copies/ml and NRMV) is approximately 594 000. Modifying the CD4+ cell threshold at which a person is eligible to start ART from 350 to 500 cells/μl or to all people diagnosed with HIV, regardless of CD4+ cell count, without the enhancement in diagnosis and retention, results in a 3.0-fold and a 3.2-fold increase, respectively, in the number of people living with NNRTI resistance, compared with the level in 2012.
Alternatively, with enhanced diagnosis and retention and people with CD4+ cell between 350 and 500 cells/μl additionally becoming eligible to initiate ART, the number of people carrying NNRTI resistance in 2032 is expected to be 4.1-fold higher than in 2012. However, 73% are predicted to be on ART with suppressed viral load or have the resistant virus in minority virus, yielding approximately 719 000 individuals with a viral load above 500 copies/ml and NRMV.
Overall, the percentage of people with resistant virus that can be originally ascribed to TDR is predicted to increase from 33% in 2012 to 38% in 2032 without enhancement of diagnosis and retention, but to remain stable at approximately 33% should an enhancement in diagnosis and retention and a change in the CD4+ cell threshold at which people are eligible to initiate ART to CD4+ cell count less than 500 cells/μl, occur.
In Figure 1c, the average HIV incidence, stratified by whether the virus is drug-sensitive or drug-resistant in 2012 and in 20 years’ time is displayed. Overall HIV incidence is predicted to be 22% lower (95% CI: −23%, −21%) in 20 years’ time under the current scenario. With no enhancement in diagnosis and retention, changing ART eligibility criteria, from 350 to 500 cells/μl or at diagnosis, has only a moderate effect: an additional 3 and 6% reduction in incidence, respectively (up to 28%). The same reduction in HIV incidence (28%) can be achieved if, instead, there is an enhancement in diagnosis and retention and the ART initiation threshold is maintained at CD4+ cell count less than 350 cells/μl. A change in initiation threshold to CD4+ cell count less than 500 cells/μl, in addition to the enhancement in diagnosis and retention, confers 36% reduction in HIV incidence in 20 years’ time and ART initiation at diagnosis leads to a 48% reduction (95% CI: −49 to −47%). Average HIV incidence with TDR is estimated to be 0.16 occurrences per 100 person-years in 2012. Modifying the CD4+ cell count at which a person is eligible to initiate ART (from 350 to 500 cells/μl or at diagnosis regardless of CD4+ cell count) has a negligible impact on future incidence of new infections with TDR, whatever the scenario on diagnosis and retention.
There is great interest in expanding ART by increasing the number of HIV-positive people in care and by changing the initiation threshold. Our results suggest that prevalence of drug resistance and need for more expensive second-line regimens are likely to increase substantially in future years as ART roll-out continues, even if policies are not changed. This is an inevitable consequence of having an increasing number of people on ART. Nevertheless, if current levels of diagnosis, retention, eligibility criteria to initiate ART (CD4+ cell count <350 cells/μl), and viral load monitoring are maintained, in 20 years’ time, over 60% of people with drug resistance to first-line agents are projected to be on a suppressive, second-line regimen. Likely due, largely, to the increase in the number of people on ART with viral replication suppressed, overall HIV incidence is predicted to drop by 22% in 20 years, if current levels of diagnosis and retention are maintained. It is noteworthy that by substantially improving diagnosis and retention, while maintaining as ART initiation policy CD4+ cell count less than 350 cells/μl, it is possible to achieve the same reduction in HIV incidence as a change in the ART initiation to all people diagnosed, with the important difference that the first will avoid ART initiation in persons for whom the individual health benefits remain unproven. However, the feasibility of implementing our scenario of enhanced diagnosis and retention is difficult to assess.
The decrease is predicted to be 36%, if diagnosis and retention are increased substantially and a policy of ART initiation at CD4+ cell count below 500 cells/μl is adopted. Incidence of HIV infections with TDR is predicted to remain stable. However, due to an overall decrease in HIV incidence, the proportion of new infections with TDR is projected to increase substantially (from 14% in 2012 to 30% in 20 years’ time).
It must be recognized that the high levels of increase in diagnosis and retention assumed in these simulations would require major investment, not only to provide treatment to higher numbers of people, but to significantly strengthen health systems to dramatically improve pre-ART and on ART retention. Public health interventions to promote social acceptance and reduce stigma will be necessary to achieve these high levels of diagnosis, linkage to care and retention. Interventions such as home-based counseling and testing , self-testing , and mobile voluntary counseling and testing supported by community mobilization  have demonstrated to be feasible and effective in increasing the number of people tested for HIV. Provision of point-of-care CD4+ cell measurement at the same time and place of HIV testing  and formal pre-ART care services providing counseling, regular review, clinical staging, social and psychological support and prevention and management of opportunistic infections, such as tuberculosis (TB), have been shown to increase, both, the proportion evaluated for ART eligibility and, the second, initiated on ART . Once on ART, allowing patients to visit the clinic less often and providing adherence monitoring with community groups have been found to be effective in minimizing the number of people lost to follow-up .
Data on new infections with TDR come mainly from WHO surveillance studies. In these, a resistance test is conducted among samples from newly diagnosed individuals less than 25 years of age and/or with a CD4+ cell count more than 500 cells/μl (if available) and no previous pregnancy, if female . These criteria increase the likelihood that people surveyed are likely to have been recently infected with HIV and ART-naive. In our work, the modeled proportion of new diagnoses with resistance in 2012 is 5.9%, similar to that obtained when restricted to people with the criteria used in WHO surveys (data not shown). This figure is much lower than the estimated 14% of new infections with TDR. The difference between the proportion with TDR at infection and at diagnosis may be explained by the time-lag between infection and diagnosis, in that those currently diagnosed with HIV may have been infected earlier, when resistance levels were lower. In addition, some drug-resistance mutations may not persist in majority virus after infection .
Other mathematical models find that in resource-limited settings, the prevalence of acquired and transmitted resistance will increase with greater ART availability [32,33]. Blower et al.  predicted that providing ART to 10–50% of an HIV infected population was likely to result in 5.9% of new infections having resistance in 10 years after treatment roll-out . Given the initial plan to roll out ART in Africa to 3 million individuals  corresponding to 5–10% of the HIV-infected population, Blower et al.  estimated that after 10 years, the proportion of new infections with TDR to be below 5%. Recently, Wagner et al. (19) investigated the impact of universal access to treatment compared to a universal ‘test and treat’ strategy in South Africa on HIV incidence and TDR. They predicted that the incidence of TDR would remain below 0.1% and that widespread access to treatment could, in some cases, even reduce transmission due to the increased selection of drug-resistant strains in people on ART, which were assumed to be 50% less transmissible than wild-type strains.
These general trends in drug resistance we have predicted for South Africa may well be relevant for other countries in sub-Saharan Africa, but there are differences worth noting. One main factor is that viral load monitoring is routinely available in South Africa; therefore, people failing first-line regimen may be expected to switch to second-line more quickly after virologic failure than in other settings and we have previously shown that TDR levels are diminished with introduction of viral load monitoring . Furthermore, levels of adherence, virologic suppression, and rates of ART interruption between settings would likely result in differences in resistance , but we suggest that any relatively small difference between countries in these factors would have a modest impact on our main overall predicted trends.
In 20 years’ time, which is the time frame used in this analysis, it is possible that in South Africa, current NNRTI will not be part of the first-line regimen anymore. Integrase inhibitors, such as dolutegravir, could potentially be available soon in fixed-dose combination with a similarly low cost and lower level of toxicity. The new South African guidelines recommend a third-line regimen , but currently the public health service offers two lines of treatment. Given the uncertainty regarding when these drugs will actually become available, it was considered appropriate to assume as antiretroviral regimen those currently in use in South Africa; therefore, those who fail the second-line regimen will remain on a boosted protease inhibitor regimen.
It is plausible as well that, in the future, point-of-care viral load will become available with possibility for more frequent viral load measurements which could slightly reduce TDR. Therefore, our estimates are potentially conservative for predicted NNRTI resistance. In addition, if high levels of TDR emerge, it is possible that WHO would recommend, for example, changing the first-line to a boosted protease inhibitor regimen. This possible policy change has not been included in these simulations and would sharply curb the transmission of resistant virus. Our model projections can be updated over time as it becomes clearer that important changes in ART programmes such as this will occur in future.
In conclusion, our results suggest that whereas increases in prevalence of drug resistance are likely as ART coverage is increased, incidence of resistance is unlikely to significantly rise and concern over resistance development should not, in itself, inhibit increases in ART coverage. Health system strengthening to improve treatment diagnosis and retention in care, to increase community acceptance and to reduce stigma may limit incident infection and mitigate transmitted and acquired drug resistance.
The authors acknowledge the use of the UCL Legion High Performance Computing Facility (Legion@UCL), and associated support services, in the completion of this work, the World Health Organization, and one reviewer who provided very helpful comments that improved the paper. Michael R. Jordan acknowledges funding from The Tufts Center for AIDS Research (CFAR): CFAR P30AI42853.
Conflicts of interest
The modeling work has been funded by the World Health Organization.
No relevant conflicts of interest
1. Johnson LF. Access to antiretroviral treatment in South Africa, 2004–2011
. S Afr J HIV Med
3. Boulle A, Van CG, Hilderbrand K, Cragg C, Abrahams M, Mathee S, et al. Seven-year experience of a primary care antiretroviral treatment programme in Khayelitsha, South Africa
4. Hoffmann CJ, Charalambous S, Sim J, Ledwaba J, Schwikkard G, Chaisson RE, et al. Viremia, resuppression, and time to resistance in human immunodeficiency virus (HIV) subtype C during first-line antiretroviral therapy in South Africa
. Clin Infect Dis
5. Wallis CL, Mellors JW, Venter WD, Sanne I, Stevens W. Varied patterns of HIV-1 drug resistance on failing first-line antiretroviral therapy in South Africa
. J Acquir Immune Defic Syndr
6. WHO HIV
/AIDS Programme. The HIV drug resistance
report – 2012; 2012.
7. Gupta RK, Jordan MR, Sultan BJ, Hill A, Davis DHJ, 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
8. World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV
infection: recommendations for a public health approach; 2013.
9. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, Kumarasamy N, et al. Prevention of HIV-1 infection with early antiretroviral therapy
. N Engl J Med
10. Eaton JW, Menzies NA, Stover J, Cambiano V, Chindelevitch L, Cori A, et al. Health benefits, costs, and cost-effectiveness of earlier eligibility for adult antiretroviral therapy and expanded treatment coverage: a combined analysis of 12 mathematical models
. Lancet Glob Health
11. Phillips AN, Pillay D, Garnett G, Bennett D, Vitoria M, Cambiano V, et al. Effect on transmission of HIV-1 resistance of timing of implementation of viral load monitoring to determine switches from first to second-line antiretroviral regimens in resource-limited settings
12. Phillips AN, Pillay D, Miners AH, Bennett DE, Gilks CF, Lundgren JD. Outcomes from monitoring of patients on antiretroviral therapy in resource-limited settings with viral load, CD4 cell count, or clinical observation alone: a computer simulation model
13. Shisana O, Rehle T, Simbayi LC, Zuma K, Jooste S, Pillay-van-Wyk V, et al.South African national HIV
prevalence, incidence, behaviour and communication survey 2008: a turning tide among teenagers? Cape Town: HSRC Press; 2009.
14. Adam MA, Johnson LF. Estimation of adult antiretroviral treatment coverage in South Africa
. S Afr Med J
15. The World Bank. Population Projection Tables by Country and Group; 2011.
16. South Africa
National Department of Health. National Antiretroviral Treatment
17. South Africa
National Department of Health. The South African antiretroviral treatment
18. Corvasce S, Violin M, Romano L, Razzolini F, Vicenti I, Galli A, et al. Evidence of differential selection of HIV-1 variants carrying drug-resistant mutations in seroconverters
. Antivir Ther
19. Turner D, Brenner B, Routy JP, Moisi D, Rosberger Z, Roger M, et al. Diminished representation of HIV-1 variants containing select drug resistance-conferring mutations in primary HIV-1 infection
. J Acquir Immune Defic Syndr
20. Castro H, Pillay D, Cane P, Asboe D, Cambiano V, Phillips A, et al. Persistence of HIV-1 transmitted drug resistance mutations
. J Infect Dis
2013; 208:1459–1463. doi: 10.1093/infdis/jit345
21. Devereux HL, Youle M, Johnson MA, Loveday C. Rapid decline in detectability of HIV-1 drug resistance mutations after stopping therapy
22. Deeks SG, Grant RM, Wrin T, Paxinos EE, Liegler T, Hoh R, et al. Persistence of drug-resistant HIV-1 after a structured treatment interruption and its impact on treatment response
23. Hollingsworth TD, Anderson RM, Fraser C. HIV-1 transmission, by stage of infection
. J Infect Dis
24. Fox MP, Cutsem GV, Giddy J, Maskew M, Keiser O, Prozesky H, et al. Rates and predictors of failure of first-line antiretroviral therapy and switch to second-line ART in South Africa
. J Acquir Immune Defic Syndr
25. South Africa
National Department of Health. The South African antiretroviral treatment
guidelines 2013; 2013.
26. Sabapathy K, Van den Bergh R, Fidler S, Hayes R, Ford N. Uptake of home-based voluntary HIV testing in sub-Saharan Africa: a systematic review and meta-analysis
. PloS Med
27. MacPherson P, Lalloo D, Choko A, van Oosterhout J, Thindwa D, Webb E, et al.Home assessment and initiation of ART following HIV
self-testing: a cluster-randomised trial to improve linkage to ART in Blantyre, Malawi. In: 20th Conference on Retroviruses and Opportunistic Infections; 2013.
28. Sweat M, Morin S, Celentano D, Mulawa M, Singh B, Mbwambo J, et al. Community-based intervention to increase HIV testing and case detection in people aged 16-32 years in Tanzania, Zimbabwe, and Thailand (NIMH Project Accept, HPTN 043): a randomised study
. Lancet Infect Dis
29. Jani IV, Sitoe NE, Alfai ER, Chongo PL, Quevedo JI, Rocha BM, et al. Effect of point-of-care CD4 cell count tests on retention of patients and rates of antiretroviral therapy initiation in primary health clinics: an observational cohort study
30. Burtle D, Welfare W, Elden S, Mamvura C, Vandelanotte J, Petherick E, et al. Introduction and evaluation of a ‘pre-ART care’ service in Swaziland: an operational research study
. BMJ Open
31. Bennett DE, Myatt M, Bertagnolio S, Sutherland D, Gilks CF. Recommendations for surveillance of transmitted HIV drug resistance in countries scaling up antiretroviral treatment
. Antivir Ther
32. Blower S, Ma L, Farmer P, Koenig S. Predicting the impact of antiretrovirals in resource-poor settings: preventing HIV infections whilst controlling drug resistance
. Curr Drug Targets Infect Disord
33. Baggaley RF, Garnett GP, Ferguson NM. Modelling the impact of antiretroviral use in resource-poor settings
. PLoS Med
34. Tobias RL. The President's Emergency Plan for AIDS Relief: US Five-Year Global HIV
/AIDS Strategy; 2004.
35. Blower S, Bodine E, Kahn J, McFarland W. The antiretroviral rollout and drug-resistant HIV in Africa: insights from empirical data and theoretical models
36. Fox Z, Phillips A, Cohen C, Neuhaus J, Baxter J, Emery S, et al. Viral resuppression and detection of drug resistance following interruption of a suppressive nonnucleoside reverse transcriptase inhibitor-based regimen
antiretroviral therapy for prevention; antiretroviral treatment; drug resistance; HIV; mathematical model; prevention; South Africa; test and treat
Supplemental Digital Content
© 2014 Lippincott Williams & Wilkins, Inc.