The Association Between Primary Antiretroviral Resistance and HAART Virologic Failure in a Developing Set

Abstract

Santos is a Brazilian port city with high HIV incidence, high primary antiretroviral resistance levels, high HIV-1 BF recombinants prevalence, and high rates of antiretroviral virologic failure. We evaluated factors related to virologic failure after 48 weeks of HAART in this population. We compared demographic and HIV profiles among 43 individuals with virologic failure (group 1) and 37 with virologic success (group 2) after 48 weeks of HAART initiation. The overall primary antiretroviral resistance prevalence was 31.2%; 46.5% in group 1 and 13.5% in group 2 (p < 0.005). Nine patients from group 1 and seven from group 2 were infected by F or BF strains. Fifteen individuals presented with NRTI mutations, 13 with NNRTI mutations, three with PI mutations, and five with NRTI and NNRTI mutations. No significant differences were observed in baseline viral load, CD4, clade assignment, antiretrovirals used, or demographics among groups or patients harboring resistant versus wild-type viruses. In this region, there was a high prevalence of antiretroviral resistance among long standing infected patients whose disease had progressed. This finding supports the concept that resistance testing prior to ART initiation is cost effective. The association between primary antiretroviral resistance and virologic failure may suggest that primary resistance greatly impairs antiretroviral activity.

Introduction

B razil is a developing country that has successfully implemented antiretroviral treatment and monitoring tests through its public health system to all HIV-infected individuals according to local guidelines (www.aids.gov.br). Almost all FDA-licensed antiretroviral drugs are available, and many of these drugs are generic formulations. Therefore, one of the major concerns in HIV therapy is primary and secondary antiretroviral resistance. In fact, 44.8% of individuals who fail antiretroviral treatment in Brazil will present resistance to two classes of antiretroviral drugs, and 34% of these individuals will present resistance to all three antiretroviral drug classes. Cross-resistance to multiple drug classes is extensive.¹

On the other hand, the frequency of primary drug resistance in Brazil is considered to be low according to several isolated surveys conducted in drug-naive populations, and in these surveys the frequency of primary drug resistance ranged from 1.4% to 8.3%.^2

–8 These surveys were conducted in different and larger Brazilian cities. Interestingly, the prevalence of primary drug resistance in chronically infected individuals generally tends to be lower when compared to the prevalence found in recently infected subjects. This is probably because chronically infected subjects acquired HIV infection during a period in which the individuals were less likely to be exposed to antiretroviral drugs. Indeed, a number of recent studies have confirmed that the majority of transmitted antiretroviral resistance mutations may persist for long periods.^{4

–12}

Multiple HIV-1 subtypes circulate in Brazil. The most prevalent circulating HIV subtype is subtype B, followed by subtype F, and then subtype C.¹³ There are also a number of circulating recombinant forms (CRFs) that occur between subtypes B and F^14,15 or subtypes B and C.¹⁶ There are also a large number of unique BF recombinant forms.¹⁷ Interestingly, CRF_28 and CRF_29, which are both BF recombinants and were the first CRFs described in Brazil, present with exponential growth rates that are higher than the exponential growth rates of the parental subtype B or subtype F.¹⁸ Both CRF_28 and CRF_29 were reported in the city of Santos, State of São Paulo, Brazil, which is a port city with a high incidence of HIV infection (www.aids.gov.br). There is a high prevalence (36%) of primary antiretroviral resistance in this city, and approximately 50% of the recently infected individuals were reported to be infected by the B/F recombinant forms of HIV.^6,19 Virologic failure to highly active antiretroviral treatment (HAART) has also been recognized as being exceedingly high in this city.²⁰ Interestingly, CRF_28 and CRF_29 carry an F subtype protease. This protease has been associated with natural resistance, in vitro, to a number of protease inhibitors due to the L89M polymorphism.²¹

Therefore, we designed a retrospective case-control analysis to evaluate factors that could be related to antiretroviral virologic failure among individuals in this geographic region.

Materials and Methods

Study population

We compared the demographic and HIV baseline profiles between individuals with virologic success and individuals presenting virologic failure. All patients were followed at the outpatient clinics of CRAIDS in the city of Santos. We selected 43 individuals who were failing HAART (group 1) and compared them to 37 individuals whose viral loads were below the level of detection 48 weeks after the initiation of treatment (group 2). All individuals initiated HAART in 2002, and they were all considered to be adherent to their antiretroviral drug regime. Adherence to therapy was based on the frequency with which the patients obtained their prescribed medication from a centralized pharmacy (no missing prescriptions by patients) and according to their physicians' impression as recorded in medical charts. Patients were chosen for group 1 if they were maintained on their original, prescribed, antiretroviral treatment for a period of at least 24 weeks, and patients were chosen for group 2 if they had maintained their treatments for 48 weeks. All individuals were using antiretroviral treatments during the 48 week duration of this study. Due to virologic failure, four individuals from group 1 substituted their antiretroviral therapy during the follow-up period. Available follow-up samples from week 48 from individuals who were identified as harboring HIV-1 with primary antiretroviral resistance were also accessed for genotyping. This study was approved by the Institutional Review Board of the Federal University of São Paulo, Brazil.

Laboratory analysis and statistics

To enter this study, individuals were chosen if they allowed us to perform retrospective assessments of viral load (Roche COBAS AMPLICOR HIV-1 Monitor, version 1.5) and CD4/CD8⁺ T cell counts immediately before antiretroviral initiation (baseline) and every 3 months after antiretroviral initiation for up to 48 weeks.

We accessed plasma samples that were collected immediately prior to the initiation of antiretroviral therapy and had been stored, undisturbed, in a −80°C freezer. All samples collected in 2002 were sequenced using ViroSeq v2.0 (Celera Diagnostics, Alameda, CA) in 2006, and therefore patients and physicians were unaware of the results by the time empirical antiretroviral treatment was initiated in 2002. Mutations (excluding common polymorphisms) that were associated with resistance to nucleoside reverse transcriptase inhibitors (NRTIs), nonnucleoside reverse transcriptase inhibitors (NNRTIs), and protease inhibitors (PIs) were evaluated according to published guidelines.²² The 215 revertant viruses, which harbor known mutations in the reverse transcriptase regions of the pol gene, were also considered to be resistant in our analysis.²² The protease and reverse transcriptase regions of the pol gene were assigned to subtypes by analyzing phylogenetic hierarchies with the PHYLIP program package, version 3.57 (GenBank accession numbers pending). The recombinant pattern analysis of each genomic region was performed with the Recombination Identification Program (RIP) from the Los Alamos database (http://hiv-web.lanl.gov). Each sequence that was suspected of being a recombinant sequence was aligned with a putative reference sequence from the parental strain.

Statistical analyses were performed using Chi-square and Fisher's exact tests. Phylogenetic analyses were performed using the PHYLIP program package, version 3.57. Analysis of recombination points was performed using SimPlot 2.5 (SimPlot, Baltimore, MD).

Results

The demographic characteristics of the patients are shown in Table 1. These data revealed that there were no statistical differences between the two experimental groups that were tested. The mean viral loads at baseline were 140,316 copies/ml of HIV-1 RNA (5.1 log₁₀) in group 1 (median 60,000; 4.8 log₁₀) compared to 375,268 copies/ml of HIV-1 RNA (5.6 log₁₀) in group 2 (median 89,000; 4.9 log₁₀). The mean CD4⁺ and CD8⁺ T cell counts at baseline for group 1 were 331 and 889 (median 291 and 781), respectively. The mean CD4⁺ and CD8⁺ T cell counts at baseline for group 2 were 360 and 969 (median 281 and 895), respectively. All patients from group 2 presented with a viral load that was below 50 copies/ml at 24 weeks, a mean viral load that was below 62 copies/ml of HIV-1 RNA at week 36, and a mean viral load that was below 61 copies/ml at week 48. Viral loads that were above 500 copies/ml were not detected in any group 2 patients at weeks 36 and 48.

Table 1.

Demographic Characteristics of the Study Population

	Group 1 (%)	Group 2 (%)
Gender
Male	27 (62.7%)	24 (64.8%)
Age
18–30	12 (27.9%)	8 (21.6%)
31–50	30 (69.7%)	28 (75.6%)
More than 50	1 (2.3%)	1 (2.7%)
Risk factor for HIV
Heterosexual	22 (51.1%)	22 (59.4%)
Bisexual	11 (25.5%)	9 (24.3%)
IDU	4 (9.3%)	4 (10.8%)
MSM	6 (13.9%)	2 (7.4%)
Total	43	37

The prevalence of non-clade B viruses was 20.9% in group 1 and 18.9% in group 2. Among the 16 non-clade B strains, six viruses presented with clade F protease and reverse transcriptase, nine viruses presented with protease F and reverse transcriptase B (resembling CRF_28 or CRF_29), and one virus presented with protease B and reverse transcriptase F. The mean basal viral loads that were assessed among individual infected with clade B viruses were 282,444 copies/ml (5.5 log₁₀) compared to 115,131 copies/ml (5.1 log₁₀) in patients infected with non-clade B viruses. The mean pretreatment CD4⁺ T cell counts and CD8⁺ T cell counts were 355 and 922, respectively, for individuals infected with clade B viruses. The mean pretreatment CD4⁺ T cell counts and CD8⁺ T cell counts were 302 and 944, respectively, for individuals infected with non-clade B viruses. No statistically significant differences were found between the two patient groups for basal viral loads, CD4⁺ or CD8⁺ T cell counts before the initiation of treatment, or for the prevalence/profile of non-B strains.

We detected 13 individuals with viruses that carried NRTI-related resistance mutations, 13 individuals with viruses that carried NNRTI resistance, and three individuals with viruses that carried PI mutations. Four individuals presented with viruses that carried resistance-related mutations to two antiretroviral classes (Table 2). Of the individuals whose viruses carried primary drug resistance, 20 individuals (46.5%) were in group 1 and only 5 (13.5%) were in group 2 (p = 0.001). All resistance mutations in group 1 related to the prescribed antiretroviral therapy; however, two out of three resistance mutations in group 2 were related to the prescribed antiretroviral therapy. The mean basal viral loads among individuals who were harboring resistant HIV were 126,996 copies/ml (5.10 log₁₀) compared to 302,200 copies/ml (5.48 log₁₀) among individuals who were harboring sensitive viruses (p = 0.2285). The mean pretreatment CD4⁺ T cell counts and CD8⁺ T cell counts among individuals with resistant viruses were 334 and 966, respectively. The mean pretreatment CD4⁺ T cell counts and CD8⁺ T cell counts among individuals with sensitive viruses were 349 and 908, respectively.

Table 2.

Resistance Profile of Individuals in Group 1 (Virologic Failures) and Group 2 (Virologic Success) ^a

	NRTI	NNRTI	PI	Initial Rx	Basal viral load log₁₀ copies/ml	Subtype (protease/RT)
Group 1
2	M184V	—	—	ZDV, 3TC, IDV	5.3	B/B
5	K70R, M184V	K103N, P225H	—	d4T, 3TC, EFZ	3.8	B/B
6	T215DE	—	—	d4T, 3TC, IDV-r	5.0	B/B
9*	—	K103N	—	d4T, 3TC, EFZ	5.2	B/B
12	M184M/I	—	—	ZDV, 3TC, NFV	4.6	B/B
13	—	K103N	—	ZDV, 3TC, NVP	4.5	B/B
17*	—	K103N	—	d4T, ddI, EFV	4.6	F/B
21*	V75M	—	—	ZDV, 3TC, NFV	5.2	B/B
24*	M184V		—	ZDV, 3TC, EFZ	4.4	B/B
25*	M184V	—	—	ZDV, 3TC, EFZ	4.9	B/B
26*	—	M230L	—	d4T, 3TC, EFZ	5.8	F/F
30*	M184I	—	—	ZDV, 3TC, NVP	4.5	B/B
32*	—	K103N, M230L/M	—	d4T, 3TC, NVP	4.4	B/F
34*	—	—	M46L	d4T, 3TC, NFV	4.9	B/B
35*		—	Y181C	d4T, 3TC, NFV,	5.1	B/B
36*	T215S	Y181V	—	ZDV, 3TC, IDV	4.8	B/B
40	M184V	—	Y181C	ZDV, 3TC, NVP	3.6	B/B
41*	M184V	—	—	ZDV, 3TC, EFZ	5.3
42*	—	K103N	—	d4T, 3TC, EFZ	4.4	B/B
43*	—	K103N, P225H	—	ZDV, 3TC, EFZ	3.9	F/B
Group 2						B/B
08	—	—	V82F	ZDV, 3TC, NVP	5.7	B/B
09	M184V	Y188L	—	d4T, 3TC, EFV	5.1	B/B
13	—	—	M46L,A71V	ZDV, 3TC, IDV-r	5.7	B/B
27	—	K103N, Y181C	—	ZDV, 3TC, NVP	3.7	B/B
28	K219N	—	—	ZDV, 3TC, NVP	4.8	F/B

Initial Rx refers to the antiretroviral treatment prescribed for each individual patient, and viral load relates to the basal viral load of each individual. Note that patient 8 from group 2 presented a PI-related mutation but was not treated with a PI-containing regimen. Follow-up samples collected from individuals who were genotyped after 48 weeks of antiretroviral exposure are denoted by an asterisk (*). Only patients 30 and 43 acquired new mutations in follow-up samples (T69D and M184V, respectively). EFV, efavirenz; NVP, nevirapine; NFP, nelfinavir; IDV, indinavir; IDV-r, ritonavir-boosted indinavir.

Among the individuals in group 2 who carried viruses with primary drug resistance, patient 08 presented with a virus that carried the protease mutation V82F. However, this patient was treated with zidovudine, 3TC, and nevirapin, and this treatment regime achieved viral loads that were below the level of detection in the first measurement that was conducted 3 months after the initiation of treatment. Patient 13 presented with a virus that carried two protease mutations, M46L and A71V, and was treated with a boosted indinavir. This treatment strategy also achieved a viral load that was below the level of detection 3 months after the initiation of treatment. Patient 27 presented with a virus that carried 2 NNRTI mutations, K103N and Y181C, and was treated with zidovudine, 3TC, and nevirapine. This patient's basal viral load was 5100 copies/ml. The viral load dropped to 380 copies/ml after 3 months of treatment, and the viral load was below the level of detection 6 months after the initiation of treatment. Interestingly, patient 09 presented with a virus that carried the M184V and Y188L mutations, had a basal viral load of 120,000 copies/ml, was treated with d4T, 3TC, and efavirenz, and the patient's viral loads were below the level of detection 3 months after the initiation of treatment. Patient 28 had a basal viral load of 4.8 copies/ml and carried a K219N baseline mutation. This patient was treated with combivir plus nevirapine. This treatment regime achieved a viral load of 3.7 copies/ml 3 months after the initiation of treatment, and the viral load was below the level of detection 6 months after the initiation of treatment.

Follow-up samples were available for 14 out of 19 individuals from group 1 who received 48 weeks of treatment and presented with viruses that were carriers for primary drug resistance. Among these individuals, all baseline mutations were present after 48 weeks of antiretroviral failure. In addition to the baseline mutations, we observed an emergence of a virus carrying the mutation M184V in patient 43, who originally presented a K103N and P225H at baseline. Likewise, we observed a virus that carried a T69D mutation in patient 30, who already had a M184 mutation at baseline. The mutation acquisition pace, which was based on the antiretroviral exposure among each individual with primary drug resistance, was 0.14 new NRTI mutations/year (two new mutations in 14 individuals exposed to NRTIs) and 0 new NNRTI mutations/year among 13 individuals who were continuously exposed to NNRTIs. No new polymorphic or secondary mutations developed among patients infected with non-clade B strains.

Discussion

Although this study was not designed to fully investigate the general prevalence of primary drug resistance in HIV, the results presented here provide information about the exceedingly high transmission of antiretroviral-resistant viruses in this geographic area of Brazil. The individuals assessed in this study presented with HIV disease, which had progressed, at least to some extent, to the point of requiring antiretroviral treatment. Furthermore, the only variable that was investigated that was related to the virologic failure group was primary drug resistance.

Primary drug resistance has, in general, been associated with an increase in the time period required to achieve viral loads that extend below the level of detection following antiretroviral treatment.²³ Nonetheless, we clearly demonstrated that primary drug resistance was strongly associated with a failure to acquire full viral suppression 1 year after the initiation of treatment. To reach this conclusion, a case-control study was designed that maximized the patient's adherence to treatment by implementing the following steps: patients were consistently obtaining their medication at the pharmacy, patients routinely arrived for their scheduled appointments, patients performed all of the requested laboratory evaluations, and patients were considered adherent according to their physicians' personal impression.

In Brazil, it has been reported that individuals infected with HIV-1, subtype F, may present increased virologic failure when treated with a regimen containing AZT, 3TC, and unboosted indinavir.²⁴ It has also been hypothesized that virologic failure among individuals who are infected with clade F may be related to a natural resistance to PIs,²¹ which may be overcome by the use of boosted PIs.²⁵ Nonetheless, we have yet to find any evidence in our evaluations that either clade F or B/F HIV-infected individuals actually present with higher virologic failure while undergoing antiretroviral treatment. With the exception of primary drug resistance, none of the demographic or virologic variables investigated in this cohort of patients was associated with virologic failure.

Of particular note, in the groups that we assessed in this study, which had been infected with HIV for an extended period of time, we still detected a high prevalence of HIV harboring primary drug resistance mutations. In fact, there is growing evidence suggesting that transmitted resistant mutations persist over time in the absence of the selective pressure of antiretroviral therapies,^9
–11 which is in contrast to what happens with secondary resistance. HIV infection will be seeded by only one viral strain in the majority of cases,²⁶ which will result in a reduction in the number of fit wild-type viruses that would otherwise compete with the original drug-resistant viruses. Viruses carrying the M184V mutation, which according to some reports may revert over time, are an exception to the persistence of the transmitted drug resistance mutations.¹² In the case of the M184V mutation, the reversion over time may be the result of back mutation. We can speculate that the hypermutation process (APOBEC3G/F mediated G-to-A substitutions) that converts the nucleoside triplet GTG (valine) to ATG (methionine) may help the virus to evolve, which may explain the relatively low prevalence of the M184V mutation in primary drug resistance strains of HIV compared to the mutations that are seen in secondary drug resistance. Nonetheless, and perhaps unexpectedly, we found a relatively high prevalence of the M184V (12.5%) mutation in our cohort of patients with a long standing history of HIV infection.

We also observed that the baseline viral loads, which were assessed immediately before the initiation of treatment, did not differ between individuals who harbored resistant viruses and individuals who harbored naive viruses. This may suggest that the drug-resistant viruses were fit enough to be transmitted and also fit enough to significantly replicate. It has been advocated that drug-resistant strains of HIV-1 have an impaired capacity to replicate.²⁷ However, there may be a threshold at which the number of mutations acquired by the virus may restore viral fitness in secondary resistance.¹ As suggested by data presented here, this process may result in drug-resistant viruses that are selected during transmission and may retain a reasonable capacity to replicate

Interestingly, three group B individuals with primary drug resistance achieved viral loads that were below the level of detection. In general, patients were either treated with effective antiretrovirals (patient 08,Table 2), or the baseline viral loads were low (patient 27), or the synergistic effect/residual activity of used antiretrovirals may have played a role in the efficacy of the antiretroviral drugs (patient 9).

We also investigated the acquisition of new mutations in the patients from group 1 (14 out of 19) who had samples that were still available, even after 48 weeks of treatment and virologic failure. Unexpectedly, only a single new mutation emerged (M184V in patient 43) that gave us an NRTI mutation acquisition rate of 0.14 mutation/year, which was negligible. Furthermore, none of the new NNRTI mutations emerged from the group 1 individuals who failed nevirapine or efavirenz, which may be good news in this group of patients at least with regards to of the potential for cross-resistance to etravirine.

According to local guidelines, resistance tests are not recommended prior to antiretroviral initiation in Brazil due to the low general prevalence of primary drug resistance (www.aids.gov.br). Although we recognize that the relatively small number of individuals and the retrospective nature of our study may preclude more definitive results, we believe that our study emphasizes that indication of resistance tests should be based on geographic trends of primary drug resistance. Furthermore, due to the high prevalence of resistance to NNRTI (27.5%) that was associated with the absence of residual activity of efavirenz or nevirapine, the use of boosted PIs as an initial treatment option may be a wiser empirical choice in this geographic region.

Footnotes

Acknowledgments

This study was supported by the Função de Amparo a Pesquisa do Estado de São Paulo (FAPESP), Brazil, Grants 06/52722–6. This work has been previously presented at the XVIII International HIV Drug Resistance Workshop, 2009, Fort Myers, Florida and the 7th European HIV Drug Resistance Workshop, 2009, Stockholm, Sweden.

Author Disclosure Statement

Dr. Diaz and Dr Caseiro have spoken at symposia or local events for Abbott Laboratories, Glaxo SmithKline, Roche, Boehringer Ingelheim, Merck, Pfizer, Gilead, Bristol-Myers Squibb, and Schering-Plough.

References

Munerato

, Sucupira

, Oliveros

, Souza

, Pereira

, Janini

et al. HIV-1 Antiretroviral resistance mutations in subtypes B, C and F in the city of São Paulo, Brazil. AIDS Res Hum Retroviruses, 2010; 26,3:265–273.

Inocencio

, Pereira

, Sucupira

, Fernandez

, Jorge

, Souza

et al. Brazilian Network for HIV Drug Resistance Surveillance: A survey of individuals recently diagnosed with HIV. J Int AIDS Soc, 2009; 12:20.

Brindeiro

, Diaz

, Sabino

, Morgado

, Pires

, Brigido

et al. Brazilian Network for HIV Drug Resistance Surveillance (HIV-BResNet): A survey of chronically infected individuals. AIDS, 2003; 17:1063–1069.

Pedroso

, Queiroz

, Alcantara

, Drexler

, Diaz

, Weyll

, Brites

. High prevalence of primary antiretroviral resistance among HIV-1-infected adults and children in Bahia, a northeast state of Brazil. J Acquir Immune Defic Syndr, 2007; 45:251–253.

Varella

, Ferreira

, de Castro

, Zalis

, Tavares

. Human immunodeficiency virus type 1 protease and reverse transcriptase mutation patterns among treatment-naive patients in different stages of infection in Rio de Janeiro, Brazil. J Med Virol, 2007; 79:1033–1039.

Sucupira

, Caseiro

, Alves

, Tescarollo

, Janini

, Sabino

et al. High levels of primary antiretroviral resistance genotypic mutations and B/F recombinants in Santos, Brazil. AIDS Patient Care STDS, 2007; 21:116–128.

de Medeiros

, Lacerda

, Cavalcanti

, de Albuquerque Mde

. Primary resistance of human immunodeficiency virus type 1 in a reference center in Recife, Pernambuco, Brazil. Mem Inst Oswaldo Cruz, 2006; 101:845–849.

Diaz

, De Souza

, Sucupira

, Cavalieri

, Ribeiro

, Mondini

et al. Lower prevalence of primary antiretroviral resistance among clade C infected individuals as compared to clade B in South Brazil's emerging clade C HIV-1 epidemic. XVII International AIDS Conference, Mexico City, Mexico, August 3–8, 2008.

Little

, Frost

, Wong

, Smith

, Pond

, Ignacio

et al. Persistence of transmitted drug resistance among subjects with primary human immunodeficiency virus infection. J Virol, 2008; 82:5510–5518.

10.

Barbour

, Hecht

, Wrin

, Liegler

, Ramstead

, Busch

et al. Persistence of primary drug resistance among recently HIV-1 infected adults. AIDS, 2004; 18:1683–1689.

11.

Delaugerre

, Morand-Joubert

, Chaix

, Picard

, Marcelin

, Schneider

et al. Persistence of multidrug-resistant HIV-1 without antiretroviral treatment 2 years after sexual transmission. Antivir Ther, 2004; 9:415–421.

12.

Jain

, Sucupira

, Deeks

, Liegler

, Pilcher

, Grant

et al. Persistence of transmitted drug-resistance mutations in patients with acute HIV. 16th Conference on Retroviruses and Opportunistic Infections, Montreal, Canada, February 8–11, 2009.

13.

Soares

, De Oliveira

, Brindeiro

, Diaz

, Sabino

, Brigido

et al. A specific subtype C of human immunodeficiency virus type 1 circulates in Brazil. AIDS, 2003; 17:11–21.

14.

De Sa Filho

, Sucupira

, Caseiro

, Sabino

, Diaz

, Janini

. Identification of two HIV type 1 circulating recombinant forms in Brazil. AIDS Res Hum Retroviruses, 2006; 22:1–13.

15.

Guimaraes

, Eyer-Silva

, Couto-Fernandez

, Morgado

. Identification of two new CRF_BF in Rio de Janeiro State, Brazil. AIDS, 2008; 22:433–435.

16.

Santos

, Sousa

, Soares

, Sanabani

, Martinez

, Sprinz

et al. Characterization of a new circulating recombinant form comprising HIV-1 subtypes C and B in southern Brazil. AIDS, 2006; 20:2011–2019.

17.

Sa Filho

, Sanabani

, Diaz

, Munerato

, Brunstein

, Fusuma

et al. Analysis of full-length human immunodeficiency virus type 1 genome reveals a variable spectrum of subtypes B and F recombinants in São Paulo, Brazil. AIDS Res Hum Retroviruses, 2005; 21:145–151.

18.

Leal

E ML

, Janini

, Diaz

. Evolutionary dynamics of HIV-1 BF and CB recombinants and its parental counterparts in South America. Retrovirol Res Treat, 2008; 1:1–14.

19.

de Sa-Filho

, Soares Mda

, Candido

, Gagliani

, Cavaliere

, Diaz

, Caseiro

. HIV type 1 pol gene diversity and antiretroviral drug resistance mutations in Santos, Brazil. AIDS Res Hum Retroviruses, 2008; 24:347–353.

20.

Caseiro

, Golega

, Etzel

, Diaz

. Characterization of virologic failure after an initially successful 48-week course of antiretroviral therapy in HIV/AIDS outpatients treated in Santos, Brazil. Braz J Infect Dis, 2008; 12:162–166.

21.

Calazans

, Brindeiro

, Verli

, Arruda

, Gonzalez

et al. Low accumulation of L90M in protease from subtype F HIV-1 with resistance to protease inhibitors is caused by the L89M polymorphism. J Infect Dis, 2005; 191:1961–1970.

22.

Bennett

, Camacho

, Otelea

, Kuritzkes

, Fleury

, Kiuchi

et al. Drug resistance mutations for surveillance of transmitted HIV-1 drug-resistance: 2009 update. PLoS One, 2009; 4:e4724.

23.

Little

, Daar

, D'Aquila

, Keiser

, Connick

, Whitcomb

et al. Reduced antiretroviral drug susceptibility among patients with primary HIV infection. JAMA, 1999; 282:1142–1149.

24.

Accetturi

, Pardini

, Novaes Pinto

, Turcato

Jr , Lewi

, Diaz

. Effects of CCR5 genetic polymorphism and HIV-1 subtype in antiretroviral response in Brazilian HIV-1-infected patients. J Acquir Immune Defic Syndr, 2000; 24:399–400.

25.

Diaz

, Vasconcelos

, Hayden

, Tenore

, Turcato

Jr , Palacios

, Sucupira

. Similar efficacy of lopinavir/ritonavir-containing regimens among clades B and F HIV-1-Infected individuals in Brazil. J Acquir Immune Defic Syndr, 2008; 47:399–401.

26.

Keele

, Giorgi

, Salazar-Gonzalez

, Decker

, Pham

, Salazar

et al. Identification and characterization of transmitted and early founder virus envelopes in primary HIV-1 infection. Proc Natl Acad Sci USA, 2008; 105:7552–7557.

27.

Weber

, Henry

, Arts

, Quinones-Mateu

. Viral fitness: Relation to drug resistance mutations and mechanisms involved: Nucleoside reverse transcriptase inhibitor mutations. Curr Opin HIV AIDS, 2007; 2:81–87.