Phenotypic Susceptibility to Antiretrovirals Among Clades C,F,and B/F Recombinant Antiretroviral-Naive HIV Type 1 Strains

Abstract

To evaluate antiretroviral phenotypic susceptibility of wild-type HIV-1 strains circulating in Brazil, samples from antiretroviral-naive individuals infected with subtypes C (n=16), F (n=9), or B/F (n=7), where reverse transcriptase is B and protease is F, were phenotyped using the Antivirogram Assay (Virco, Mechelen, Belgium). Reduced susceptibility to protease inhibitors (PIs) was observed in one C and three F isolates. None of these samples had any known PI resistance mutations. The phenotypic fold change to one PI was above the biological cut-off in three of 96 (3.1%) clade F phenotypic determinations and in one of 96 (1.0%) clade C. Phenotypic resistance to at least one nucleoside reverse transcriptase inhibitor (NRTI) was found for two B/F, four C, and three F isolates. The phenotypic fold change in susceptibility to NRTIs was above the cut-off value in nine of 111 (8.1%) clade C determinations, as compared to three of 63 (4.8%) for clade F and two of 49 (4.1%) for clade B. The phenotypic fold change to non-NRTI (NNRTI) was above the cut-off in seven of 32 (21.9%) of C isolates determinations, whereas none of the F isolates had a decrease of susceptibility. Only two of the 16 C samples had a known NNRTI resistance mutation. The NNRTI fold change was above the cut-off value in three of 14 (21.4%) phenotypic determinations of Brazilian B/F recombinants, representing clade B reverse transcriptase. NNRTI susceptibility should be better investigated in clade C and B/F recombinants.

Introduction

D rug resistance testing has become an important tool in the management of HIV-infected individuals undergoing antiretroviral therapy.¹ Genotyping and phenotyping methods appear to be equally useful for determining the susceptibility of HIV-1 to antiretroviral drugs.² Genotyping identifies mutations in the HIV genome associated with antiretroviral resistance, whereas phenotyping examines the relative susceptibility of viruses to different antiretroviral concentrations in vitro. It is conceivable that the genetic diversity of HIV-1 influences the susceptibility to antiretroviral drugs. Drug susceptibility tests have been designed for subtype B strains and are performed mainly on those strains. However, the increasing global prevalence of non-B subtypes creates the need to determine the performance of commercial drug resistance assays in testing these HIV-1 subtypes. Furthermore, it has been demonstrated that specific mutations or pathways of antiretroviral resistance differ between HIV-1 clades.^3,4 Some retrospective analysis suggests that individuals infected with clade F HIV-1 respond more poorly to antiretroviral therapy than clade B-infected individuals.⁵ It has also been suggested that the L89M polymorphism, which is highly prevalent in the clade F protease,⁶ plays a role in the antiretroviral response because this polymorphism decreases the susceptibility of viruses to most protease inhibitors (PIs).⁷ In contrast, it has been also speculated that the I93L polymorphism increases the susceptibility of clade C samples to lopinavir.⁸ Although the I93L polymorphism is also common in clade F viruses, the effect in the F subtype has not been determined. In Brazil, the HIV-1 clades B, C, and F cocirculate, and B/F recombinants account for a significant proportion of all circulating clades.^9,10 In this study, we evaluated the mutation profile and in vitro antiretroviral response of samples from antiretroviral-naive Brazilian patients infected with subtypes C, F, or B/F. We evaluated the phenotypic susceptibility and genotypic correlates of resistance in antiretroviral-naive strains to assess the natural susceptibility of the C and F HIV subtypes to PIs as well as nucleoside and nonnucleoside reverse transcriptase inhibitors (NRTIs and NNRTIs).

Materials and Methods

Isolates from 32 antiretroviral-naive Brazilian patients were evaluated. There were 16 subtype C, nine subtype F1, and seven subtype B/F recombinant isolates. All B/F recombinant samples carried an F1 protease and a B reverse transcriptase. Viruses from plasma samples were genotyped at the Federal Universities of São Paulo and Rio de Janeiro, Brazil, using the ViroSeq System (Celera Diagnostics, Alameda, CA). Sequences were analyzed using an ABI 3100 sequencer (Applied Biosystems, Foster City, CA). All genotyping analysis was conducted on a portion of the pol gene that spans the reverse transcriptase and protease regions. Samples identified as recombinants were analyzed using SimPlot version 3.5.1¹¹ and confirmed by bootscanning analysis. The basic principle of bootscanning is that mosaicism is suggested by high levels of phylogenetic relatedness between a query sequence and more than one subtype reference sequence in different genomic regions.¹² Sample sets were subjected to antiretroviral phenotypic analysis using the Antivirogram Assay (Virco, Mechelen, Belgium). This phenotyping assay uses HIV-1 genomes generated through the recombination of polymerase chain reaction (PCR)-amplified pol products derived from patient viruses and a subtype B proviral clone with a deletion of the protease and reverse transcriptase regions. Recombinant viral production and detection methods were conducted as previously described.^13,14 For clade C, F, and B/F recombinant isolates obtained from treatment-naive subjects, the mean, median, and interquartile range of fold change values for each drug, as determined using the Antivirogram Assay, were compared to the same parameters for clade B isolates in the Virco database. The mutation patterns associated with reduced susceptibility to one or more drugs in clade C, F, and B/F isolates were evaluated and compared to the genotypic correlates of drug resistance described for clade B isolates.

Results

The following are shown in Tables 1A and 1B: sample identification; protease and reverse transcriptase subtypes; polymorphisms/mutations at codons associated with resistance; phenotypic fold change in susceptibility to PIs, NRTIs, and NNRTIs; and mean fold change by subtype.¹⁵

Table 1A.

Results of the Phenotypic Susceptibility of HIV-1 Brazilian Clade F and C Samples to Protease Inhibitors Using the Antivirogram Assay and Protease Genotypic Profile

		FC in PI susceptibility vs. wild-type
		APV	ATV	IDV	LPV	NFV	SQV
		Virco biological cut-off value ^a
Sample ID	Protease subtype	(1.8)	(2.0)	(2.1)	(1.6)	(2.3)	(1.7)	Protease polymorphisms/mutations
BR29	F	1.1	0.7	0.5	1.6	1.4	0.8	V3I	R41K	R57K	Q61N	K70R	L89M
BR35	F	1	1	0.7	0.6	0.8	0.4	V3I	L10I	I13V	L33V	E35D	M36I	S37N	R41K	R57K	D60E	Q61N	G68E	L89M
BR36	F	0.7	1.1	1	1.6	0.7	1.5	V3I	G17D	K20R	E35D	M36I	S37N	R41K	K43R	R57K	Q61N	I72R
BR37	F	0.8	1.1	1	0.6	0.9	0.5	V3I	T12I	I13V	M36I	S37N	R41K	R57K	Q61H/N	L89M
BR38	F	1.4	0.9	0.7	1.2	1.7	1.8	V3I	I15V	E35D	M36I	S37N	R41K	R57K	D60E	Q61E	L63P	K70T	V77I
BR39	F	0.3	0.2	1.4	0.7	0.6	0.3	V3I	T12K	I15V	L19I	E35D	M36I	S37N	R41K	R57K	Q61N	E65D	I72T
BR40	F	0.6	0.5	0.9	0.9	1	1	V3I	L10I	I13V	G17E	K20R	M36I	S37E	R57K	Q61N	L63AN	K70R
BR31	F	0.9	1	0.8	0.6	0.8	0.5	V3I	I13I/V	I15I/L	M36I	S37N	R41K	R57K	Q61N	H69Y	L89M
BR43	F	1.2	1.3	0.7	1.1	1.2	2.1	V3I	L10I	I13V	I15V	G16E	K20M	E35D	M36I	S37N	R41K	R57K	Y59H	Q61N	I72R	L89M
BR44	F	0.6	0.4	0.6	2.3	1.4	0.6	V3I	I15V	G17E	K20R	E35D	M36I	S37N	R57K	Q61N	L63T	E65D	I72V
BR46	F	1.1	1.5	0.5	1.3	1.4	0.9	V3I	I15V	K20R	E35D	M36I	S37N	P39Q	R41K	R57K	D60E	Q61D	L63P/S	L89M
BR48	F	0.3	0.1	0.2	0.7	1.1	0.2	V3I	L10V	I15V	S37E	R57K	Q61N	L63A	I64L	I72V	V77I	I93L
BR50	F	1.3	0.7	0.7	1.1	0.5	0.5	V3I	L10I	T12S	I13V	K14R	G16E	K20R	E35D	M36I	S37N	R41K	R57K	Q61N	L63V	I72T
BR51	F	0.3	0.6	1.5	0.9	1	0.9	V3I	I15V	E35D	M36I	S37N	P39Q	R41K	R57K	Q61N	V77I	L89M
BR52	F	0.3	0.4	0.3	0.5	0.3	0.4	V3I	I15V	K20R	M36I/L	S37K	R41N	H69K	V82I	L89M	I93L
BR53	F	0.4	0.3	0.2	0.7	0.8	0.3	V3I	I15V	E35D	M36I	S37N	R41K	R57K	Q61N	L63S	E65D	I72T
Mean		0.8	0.7	0.7	1	0.9	0.8
BR76	C	0.2	0.5	0.3	0.7	0.4	0.5	V3I	I15V	G16A	M36I	S37K	R41N	H69K	L89M	I93L
BR78	C	0.2	0.6	0.3	0.6	0.4	0.5	V3I	I15V	G16E	M36I	S37K	R41N	H69K	L89M	I93L
BR79	C	0.6	0.8	0.5	0.9	1.4	0.5	V3I	I15V	M36I	S37K	R41N	L63T	H69K	L89M	I93L
BR80	C	0.7	0.8	0.4	0.6	0.4	0.9	V3I	I15V	M36I	S37K	R41N	L63T	H69K	V82I	L89M	I93L
BR81	C	0.2	0.1	0.2	0.7	0.7	0	V3I	I15V	M36I	S37K	R41N	L63GN	H69K	L89M	I93L
BR82	C	0.7	0.6	0.2	1.2	1	0.6	V3I	G16E	M36I	S37K	R41N	L63H/UQ/V	I64L	H69K	L89M	I93L
BR83	C	0.5	0.6	0.8	1	1.1	0.9	V3I	I15V	L19I	M36I	S37K	R41N	M46V	H69K	V82I	L89M	I93L
BR84	C	0.4	0.9	0.8	0.7	0.5	0.7	V3I	T12P	I15V	M36T	S37K	R41N	L63T	H69K	L89M	I93L
BR86	C	1.3	1.1	0.8	1.2	1.6	1.8	V3I	I15V	M36I	S37K/N	R41N	L63I/P/S/T	H69K	V82IN	L89M	I93L
BR87	C	0.9	1	1.8	0.6	0.5	1.4	V3I	I15V	L19I	S37K	R411	D60E	H69K	L89M	I93L
BR88	C	0.3	0.3	0.2	0.8	1.3	0.8	V3I	T12N	I15V	M36I	S37N	R41K	H69K	L89M	I93L
BR89	C	0.5	0.6	0.6	0.7	1.1	0.7	V3I	L10I	113V	K14R	M361/T	S37K	R41N	L63P	H69K	L89M	I93L
BR91	C	0.8	1.1	0.3	1	1.3	0.5	V3I	I15V	G16E	K20R	E35D	M36I	S37K	P39S	R41N/S	H69K	L89M	I93L
BR92	C	0.2	0.4	0.2	0.3	0.2	0.3	V3I	T12K	K14R	I15V	M36I	S37K	R41N	H69K	V82I	L89M	I93L
BR94	C	0.6	0.4	0.3	0.6	0.6	0.5	V3I	I15V	S37K	R41N	L63V	H69K	V82I	L89M	I93L
BR99	C	0.8	1.2	1.7	1.4	1.2	0.7	V3I	K14R	I15V	G16E	K20R	M36I	S37K	P39S	R41N	H69K	L89M	I93L
Mean		0.6	0.7	0.6	0.8	0.9	0.7

Values refer to the second generation Virco biological cut-off values.¹³

FC, fold change; PI, protease inhibitor; APV, amprenavir; AZT, zidovudine; IDV, indinavir; LPV, lopinavir; NFV, nelfinavir; SQV, saquinavir. Mean fold changes and fold changes above biological cut-off are in bold.

Table 1B.

Results of Phenotypic Susceptibility of HIV-1 Brazilian Samples to Reverse Transcriptase Inhibitors Using the Antivirogram Assay

	FC susceptibility vs. wild-type
		NRTIs							NNRTIs
		Virco biological cut-off value ^a							Virco biological cut-off
Sample ID	RT subtype	ABC (2.2)	ddI (2.2)	FTC (3.5)	3TC (2.4)	d4T (2.3)	TDF (2.1)	AZT (2.1)	EFV (3.4)	NVP (5.5)	RT resistance mutation codons
BR29	B/F	0.7	0.5	0.4	0.8	0.4	1	2.7	1.6	0.7
BR35	B/F	0.5	0.7	2	0.7	0.6	0.4	1.2	0.9	2.8
BR36	B/F	0.5	1.1	1.2	0.7	1.6	1	0.7	4.1	7.2
BR37	B/F	1.1	1	0.7	1.1	1.1	1	1.2	2.6	1.2
BR38	B/F	0.7	1.9	0.9	1.4	1.8	1.7	0.7	1	1.1
BR39	B/F	0.8	2.1	3.6	0.6	0.4	0.7	0.7	3.8	1.6
BR40	B/F	0.5	0.6	1.1	1.1	0.8	0.4	1.9	0.2	0.9
Mean		0.7	1.1	1.4	0.9	1.0	0.9	1.3	2.0	2.2
BR31	F	0.8	0.9	0.7	0.5	1.3	0.8	1.2	1.5	1.7
BR43	F	0.2	2.3	0.4	1.1	1.5	1	0.6	0.7	1.1	E40D
BR44	F	1.7	0.7	1.3	0.6	1.5	0.6	2.9	1.2	0.8
BR46	F	0.8	1.1	1	0.7	2.2	0.5	3.2	0.7	1.8
BR48	F	0.2	0.2	0.1	0.4	0.2	0.6	0.2	0.5	0.5	150V
BR50	F	1	0.7	1.4	0.9	1.3	1.4	1.2	1.1	0.9
BR51	F	1.6	0.3	0.5	0.6	0.3	0.3	0.3	0.2	0.7	G333D
BR52	F	0.3	0.5	0.9	1.1	0.7	0.4	1.8	0.9	0.2
BR53	F	0.5	0.2	0.3	0.6	0.2	0.5	0.4	0.2	0.2
Mean		0.7	0.9	1.1	0.8	1.0	0.8	1.3	1.4	1.5
BR76	C	0.4	0.5	0.7	1	0.3	1.4	0.8	1.9	20.7	A98G
BR78	C	0.4	0.9	1	0.9	0.4	0.9	0.6	0.2	1.2
BR79	C	0.4	0.8	1.7	1.2	0.5	1.4	0.5	2.2	0.8
BR80	C	0.6	0.4	3.3	0.6	0.4	0.8	1.9	2.1	4.7	A98S
BR81	C	0.2	0.2	0.1	0.7	0.2	0.6	0.2	0.5	0.5
BR82	C	0.6	1.8	1.5	1.6	2.6	1	1.1	2	6.3
BR83	C	0.5	1.2	1.3	0.8	0.6	1.4	0.9	2.2	0.9
BR84	C	0.4	1.3	0.5	1.6	0.3	0.8	0.6	2.3	0.4
BR86	C	2.7		1.4	2	2	1.5	3.4	3.9	0.8	E40D
BR87	C	1.9	1.8	2.2	1.1	2.1	1.4	2.6	18.7	61.6	K103M
BR88	C	0.2	0.3	0.3	0.7	0.3	1.6	0.5	0.5	0.7
BR89	C	0.8	0.5	1.4	0.8	0.4	0.6	0.9	0.6	0.7
BR91	C	1.3	1	0.9	1.2	1	0.5	1.2	0.9	0.5
BR92	C	0.3	0.3	0.6	0.8	0.3	0.2	0.7	0.5	0.3
BR94	C	0.5	0.7	0.1	0.6	0.6	0.8	0.4	0.5	6.1
BR99	C	1.2	2.4	1.1	2.6	4	2.4	2.3	3	7.2
Mean		0.8	0.9	1.1	1.1	1.0	1.1	1.2	2.6	7.1

Values refer to the second generation Virco biological cut-off values.¹³

FC, fold change; NRTIs, nucleoside reverse transcriptase inhibitors; NNRTIs, nonnucleoside reverse transcriptase inhibitors; RT, reverse transcriptase; ABC, abacavir; ddI, didanosine; FTC, entricitabine; 3TC, lamivudine; d4T, stavudine; TDF, tenofovir; AZT, zidovudine; EFV, efavirenz; NVP, nevirapine. Mean fold changes and fold changes above the biological cut-off are in bold.

The Antivirogram HIV-1 fold change values are interpreted using cut-off values to separate viral isolates with high antiretroviral susceptibility from those with low antiretroviral susceptibility. Biological cut-off values are derived from in vitro experiments and mark the upper limit of natural variation in phenotypic susceptibility among wild-type viruses, which includes 97.5% of strains.¹⁵ Although these biological cut-off values were typically calculated using clade B strains, we analyzed the percentage of viruses from clades C, F, and the Brazilian B/F recombinant that remained above the biological cut-off value. There was in total 96 phenotypic determinations for PIs among the clade 16 subtype C isolates as well as 96 phenotypic determinations for PIs among the 16 strains with protease F, nine of which harbored subtype F RT and seven of which harbored subtype B RT. For PI susceptibility, one out of 96 phenotypic determinations (1.0%) of the isolates from clade C and three out of 96 determinations (3.1%) from clade F had fold change values above the upper limit of the biological cut-off values (Fischer exact test of 0.3). For NRTI susceptibility, nine out of 111 phenotypic determinations (8.1%) of fold change values were above the biological cut-off for clade C isolates, versus three out of 63 determinations (4.8%) and two out of 49 determinations (4.1%) for clades F and B/F, respectively (Fischer exact test between groups nonsignificant). The phenotyping results for NNRTI susceptibility showed that seven out of 32 phenotypic determinations (21.9%) revealed fold change values above the biological cut-off value in clade C isolates. None of the clade F isolates showed decreased susceptibility to NNRTIs (Fischer exact test comparing clade C versus F, p=0.03). Most importantly, clade C samples did not have any mutation known to correlate with NRTI resistance, and only two samples had mutations correlating with NNRTI resistance (mutations A98G and K103N in samples BR76 and BR87, respectively, Table 1B). Surprisingly, three out of 14 determinations (21.4%) of the Brazilian clade B/F isolates had fold changes above the biological cut-off value for NNRTI susceptibility, although no statistical differences were detected for NNRTI resistance between clades B/F and F (Fischer exact tests, p=0.07) or B/F versus C (p=0.6).

All but four viral samples were fully susceptible to all six PIs tested (Table 1A). Three subtype F samples (BR38, BR43, and BR44) showed reduced susceptibility to PIs. BR38 had a fold change of 1.8 for saquinavir and carried the following protease polymorphisms/mutations: V3I, I15V, E35D, M36I, S37N, R41K, R57K, D60E, Q61E, L63P, K70T, and V77I. The BR43 isolate also showed decreased susceptibility to saquinavir (fold change of 2.1) and carried the protease polymorphisms/mutations V3I, L10V, I13V, I15V, G16E, K20M, E35D, M36I, S37N, R41K, R57K, Y59H, Q61N, I72T, and L89M. The BR44 isolate had a fold change of 2.3 for lopinavir and contained the amino acid substitutions V3I, I15V, G17E, K20R, E35D, M36I, S37N, R57K, Q61N, L63T, E65D, and I72V. Only one clade C isolate (BR86) had a low level of resistance, with a 0.1 fold change above the biological cut-off value. This isolate contained the following amino acid substitutions: V3I, I15V, M36I, S37K/N, R41N, L63I/P/S/T, H69K, V82I/V, L89M, and I93L. As shown in Table 1A, these four viruses had fold changes discreetly above the biological cut-off values for each respective drug.¹⁵ Since other strains tested here displayed the same mutation profile without a reduced PI susceptibility, direct involvement of those amino acid substitutions with drug resistance could not be speculated. One exception was the Y59H substitution. This substitution was detected only in the BR43 isolate, which showed a loss of susceptibility to saquinavir. All 28 of the remaining isolates (13 of subtype F and 15 of subtype C) were fully susceptible to all PIs tested. The mean phenotypic fold change per subtype sample set was nearly 1.0, which is comparable to that of subtype B wild-type strains.

Accessory amino acid substitutions that correlated with selective pressure from antiretroviral drugs among subtype B, including substitutions at codons 10, 20, 36, 63, 77, and 82, were detected in the protease region of non-B samples. Changes were found at the following prevalences: L10I (12.5% in subtype F, 6.2% in subtype C), L10V (18.7% in subtype F, 6.2% in subtype C), K20R (43.7% in subtype F, 12.5% in subtype C), M36I (87.5% in subtype F, 81.2% in subtype C), M36T (12.5% in subtype C), L63P (12.5% in subtype F and subtype C), L63A/G/H/I/L/Q/S/T/V (31.2% in subtype F, 43.7% in subtype C), V77I (18.7% in subtype F), and V82I (6.2% in subtype F and 31.2% in subtype C). The L89M polymorphism did not have any effect on PI susceptibility among the clade F isolates (fold change ≤1.0). The I93L polymorphism also did not have any effect on PI susceptibility among clade C isolates, as shown by the mean phenotypic fold change of 1.0 in susceptibility to lopinavir. Although previous studies have shown that the L89M polymorphism has an impact on saquinavir, nelfinavir, amprenavir, and lopinavir susceptibility,^7,16 we found no data to support a consistent change in PI susceptibility related to this polymorphism. It has also been previously reported that susceptibility to lopinavir increases with the I93L substitution in clade C isolates, and the poor response to PI therapy is empowered by I93L.^8,17 In contrast, we found I93L in all subtype C samples analyzed. The mean fold change value for the clade C samples was very close to 1.0 (0.8), and the susceptibility to other PIs was not compromised. The remaining polymorphisms are of uncertain significance. As previously described,¹⁸ 100% of the clade F isolates carried V3I and R57K; 81.2% carried S37N, R41K, and Q61N; 62.5% carried E35D; 56.2% carried I15V; and 50% carried L89M. For clade C, 100% of the isolates carried S37K, R41N, H69K, L89M, and I93L; 87.5% carried I15V; and 31.2% carried G16A/E. Notably, subtype F isolates had S37N, R41K, and R57K polymorphisms, whereas subtype C isolates had S37K, R41N, and H69K polymorphisms in all samples. When different substitutions related to PI resistance in clade B (K20I, M36V/L, and L63S/A/T/F) were considered, all but two subtype F isolates (87.5%) carried two to four mutations. Nearly 62% of subtype C samples carried one to three of those mutations.

We also analyzed the susceptibility of clades C and F to reverse transcriptase inhibitors (seven NRTIs and two NNRTIs). The number of Brazilian isolates included from each clade was as follows: seven subtype B/F, 16 subtype C, and nine subtype F (Table 1B). Decreased susceptibility to at least one reverse transcriptase inhibitor was observed in 10 viral isolates. There were two clade B/F (BR36 and BR39), three clade F (BR43, BR44, and BR46), and six clade C (BR76, BR82, BR86, BR87, BR94, and BR99) samples with decreased susceptibility to at least one reverse transcriptase inhibitor (Table 1B). Because clades C and F carried many reverse transcriptase polymorphisms, Table 1B shows only the amino acid substitutions known to be related to drug resistance.

Within the clade F samples, BR43 had decreased susceptibility to didanosine with a fold change of 0.1 above the cut-off value. BR44 and BR46 had decreased susceptibility to zidovudine with fold changes of 0.2 and 0.5 above the cut-off values, respectively. These three clade F samples showed loss of susceptibility without any amino acid substitutions previously described as related to antiretroviral resistance.

Within the clade C samples, BR76 had the A98G mutation and a phenotypic fold change of 20.7 to nevirapine, whereas BR87 had the K103N mutation and phenotypic fold changes of 18.7 and 61.6 to efavirenz and nevirapine, respectively. The BR86 isolate had a decreased susceptibility to abacavir, zidovudine, and efavirenz but only one NRTI resistance-related codon that was previously described. However, this codon had a different amino acid substitution: E40D. BR43 of clade F also had this amino acid change, but no decreased susceptibility was observed. The other four subtype C isolates with decreased susceptibility to NNRTIs had no known resistance mutations. The mean fold change values of clade F samples were comparable to those of clade B for all reverse transcriptase inhibitors tested. The mean fold change values were also near 1.0 for the clade C samples, with the exception of susceptibility to efavirenz (2.6) and nevirapine (7.1).

As seen in Table 1b, two out of seven BFs samples presented fold change above biological cut-offs for NRTIs (zidovudine and emtricitabine); compared to three out of nine F samples (two samples to zidovudine and one to didanosine), and four out of 16 subtype C samples (zidovudine three samples, stavudine two samples, and didanosine, abacavir, and tenofovir one sample each). The study that determined the biological cut-offs for this phenotypic assay evaluated 1,650 samples for zidovudine,¹⁵ and considering the 97.5th percentile profile for fold change, it was established that the prevalence of samples above biological cut-offs was higher in our set of F subtype strains (Fischer's exact test, p=0.02) and C strains (Fischer's exact test, p=0.007). For NRTIs, no other differences were statistically different, with a trend for stavudine and clade C, where 1,638 samples have been originally tested for this drug by Verlinder et al. (Fischer's exact test, p=0.06).¹⁵

Two out of seven BF samples presented fold change above biological cut-offs to NNRTIs (two to efavirenz and one to neveripine), compared to zero F samples and six C samples (five to nevirapine and two to efavirenz). Again here, compared to the original study that defined the biological cut-offs that evaluated 2,333 samples from naive individual to efavirenz and according to the 97.5th percentile profile, the prevalence of samples with fold changes above the biological cut-off among our BF set of samples was higher (Fischer's exact test, p=0.01). Also, when we excluded the two samples that harbored transmitted resistant mutations to NNRTIs in the set of subtype C samples (samples BR76 and BR87, Table 1B), and compared the prevalence of fold change above the biological cut-off to nevirapine with the 2,357 samples investigated by Verlinden et al.,¹⁵ the prevalence was higher in our set of samples (Fischer's exact test, p=0.007). We also found three F samples with fold change above the biological cut-offs to PIs, two to saquinavir, and one to lopinavir, and one subtype C sample to saquinavir (Table 1a).

Discussion

Although genetic polymorphisms at positions associated with PI resistance were found at a higher frequency in the two non-B subtypes isolated from antiretroviral-naive subjects, no phenotypic decrease in PI susceptibility was identified in this sample set. However, a decrease in susceptibility to NRTIs and NNRTIs was found among the non-B HIV-1 samples. Remarkably, only two of the 11 samples with decreased susceptibility to antiretrovirals revealed known resistance mutations related to reverse transcriptase inhibitors (A98G and K103N), suggesting that the genotypic correlates of decreased susceptibility among clade C, F, and B/F recombinants is not completely understood. Of note, the A98G substitution is related to the decrease in susceptibility to etravirine.¹⁹ One recent study demonstrated that the rate of resistance-related mutation acquisition in clade C samples may be inferior to that of clade B samples.²⁰ One reasonable explanation for the lower prevalence of mutations in clade C-infected individuals than in clade B-infected individuals is that some correlates of genotypic resistance in HIV-1 clade C may be unknown.

Another hypothetical explanation for the lack of mutations known to confer resistance to reverse transcriptase inhibitors is that genetic mutations outside of the reverse transcriptase catalytic region may influence susceptibility to certain reverse transcriptase inhibitors. In fact, there is mounting data that mutations in the RT connection and RNase H regions lead to NRTI or NNRTI resistance.^21,22 Resistance mutations outside of the reverse transcriptase region may explain the lack of resistance mutations found because B/F recombinant samples cluster with clade B in the reverse transcriptase region. However, one B/F recombinant sample showed resistance to NRTIs and two samples had resistance to NNRTIs. Sequencing outside of the protease and reverse transcriptase catalytic regions was not performed; therefore, we are unable to assign the outside regions to any particular HIV-1 subtype.

Although all clade C samples tested had the I93L polymorphism in the protease region, we were unable to confirm previous results that suggested in vitro hypersusceptibility to lopinavir because of this natural polymorphism.⁸ Additionally, the I93L substitution has been related to lower in vivo susceptibility to indinavir and nelfinavir among clade B-infected individuals,²³ which also was not confirmed in the present study. We were also unable to confirm that the L89M mutation in clade F samples would compromise the natural susceptibility to PIs, as previously described.⁷ However, it can be argued that the high number of clade-related polymorphisms found in non-B strains acts to decrease the genetic barrier to PI resistance, even in the absence of baseline resistance to PIs, as described in the present study. It has been reported that the polymorphisms found at positions 10 and 36 are the strongest predictors of therapeutic failure, appearing in nearly 40% of antiretroviral-naive subjects who failed PI-based therapy.²⁴ Notably, the M36I mutation was present in 87.5% of the F-subtype and 75.0% of the C-subtype proteases in our study, and the phenotypic assays did not recognize any significant loss in PI susceptibility.

Although previously antiretroviral-naive individuals treated with boosted PIs do not typically have PI mutations or virological resistance,^25

–28 one study revealed a genetic progression in five patients treated with boosted lopinavir monotherapy, two infected with clade B HIV-1 and three infected with CRF02_AG.²⁹ Although no phenotypic resistance was confirmed in these two clade B isolates, it has been confirmed in two CRF02_AG strains with the emergence of L76V and other mutations. In those cases, it can be speculated that the high number of polymorphisms in non-B strains allows genetic evolution and selective pressure of PIs.

Another point of great interest is the relationship between the genetic diversity of HIV and response to treatment. An unrecognized risk of antiretroviral failure may be related to a high number of natural polymorphisms in non-B strains. Although the results of some studies indicate that the level of response is lower among patients infected with non-B HIV,⁵ other studies have shown the opposite.³⁰

We recognize that the small size of the sample in this study is a major limitation and that further studies are needed to confirm our results comparing non-B to B strains. Nevertheless, we propose that until better correlates of genotypic resistance are available, phenotypic tests for non-B clades should be more widely accessible.

Footnotes

Acknowledgments

This work was partially supported by Fundação de Amparo a Pesquisa do Estado de São Paulo (04/15856-9). Antivirogram assays were supported and executed by Virco (Mechelen, Belgium).

Author Disclosure Statement

P.M. is a current employee of LifeTechnologies, Brazil and R.D. has served on advisory boards and spoken at symposia or local events for Virco. Virco had not played any decision-making role in the design of this study.

References

Youree

, D'Aquila

. Antiretroviral resistance testing for clinical management. AIDS Rev, 2002; 4:3–12.

Dunne

, Mitchell

, Coberly

et al. Comparison of genotyping and phenotyping methods for determining susceptibility of HIV-1 to antiretroviral drugs. AIDS, 2001; 15:1471–1475.

Munerato

, Sucupira

, Oliveros

et al. HIV type 1 antiretroviral resistance mutations in subtypes B, C, and F in the City of Sao Paulo, Brazil. AIDS Res Hum Retroviruses, 2010; 26:265–273.

Deforche

, Silander

, Camacho

et al. Analysis of HIV-1 pol sequences using Bayesian networks: Implications for drug resistance. Bioinformatics, 2006; 22:2975–2979.

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.

Abecasis

, Deforche

, Snoeck

et al. Protease mutation M89I/V is linked to therapy failure in patients infected with the HIV-1 non-B subtypes C, F or G. AIDS, 2005; 19:1799–1806.

Calazans

, Brindeiro

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.

Gonzalez

, Brindeiro

, Tarin

et al. In vitro hypersusceptibility of human immunodeficiency virus type 1 subtype C protease to lopinavir. Antimicrob Agents Chemother, 2003; 47:2817–2822.

Soares

, Santos

, Pellegrini

, Sprinz

, Tanuri

, Soares

. Epidemiologic and molecular characterization of human immunodeficiency virus type 1 in southern Brazil. J Acquir Immune Defic Syndr, 2003; 34:520–526.

10.

Sucupira

, Caseiro

, Alves

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.

11.

Lole

, Bollinger

, Paranjape

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.

12.

Salminen

, Carr

, Burke

, McCutchan

. Identification of breakpoints in intergenotypic recombinants of HIV type 1 by bootscanning. AIDS Res Hum Retroviruses, 1995; 11:1423–1425.

13.

Hertogs

, de Bethune

, Miller

et al. A rapid method for simultaneous detection of phenotypic resistance to inhibitors of protease and reverse transcriptase in recombinant human immunodeficiency virus type 1 isolates from patients treated with antiretroviral drugs. Antimicrob Agents Chemother, 1998; 42:269–276.

14.

Petropoulos

, Parkin

, Limoli

et al. A novel phenotypic drug susceptibility assay for human immunodeficiency virus type 1. Antimicrob Agents Chemother, 2000; 44:920–928.

15.

Verlinden

, Vermeiren

, Lecocq

. Assessment of the Antivirogram^® performance over time including a revised definition of biological test cut-off values. XIVth International HIV DRUG RESISTANCE Workshop, Québec City, Québec, Canada, 2005.

16.

Condra

, Holder

, Schleif

et al. Genetic correlates of in vivo viral resistance to indinavir, a human immunodeficiency virus type 1 protease inhibitor. J Virol, 1996; 70:8270–8276.

17.

Servais

, Lambert

, Fontaine

et al. Variant human immunodeficiency virus type 1 proteases and response to combination therapy including a protease inhibitor. Antimicrob Agents Chemother, 2001; 45:893–900.

18.

Tanuri

, Vicente

, Otsuki

et al. Genetic variation and susceptibilities to protease inhibitors among subtype B and F isolates in Brazil. Antimicrob Agents Chemother, 1999; 43:253–258.

19.

Johnson

, Brun-Vezinet

, Clotet

et al. Update of the drug resistance mutations in HIV-1: December 2010. Top HIV Med, 2010; 18:156–163.

20.

Soares

, Santos

, Sousa

et al. Differential drug resistance acquisition in HIV-1 of subtypes B and C. PLoS ONE, 2007; 2:e730.

21.

Santos

, Lengruber

, Soares

et al. Conservation patterns of HIV-1 RT connection and RNase H domains: Identification of new mutations in NRTI-treated patients. PLoS ONE, 2008; 3:e1781.

22.

Yap

, Sheen

, Fahey

et al. N348I in the connection domain of HIV-1 reverse transcriptase confers zidovudine and nevirapine resistance. PLoS Med, 2007; 4:e335.

23.

Kempf

, Isaacson

, King

et al. Identification of genotypic changes in human immunodeficiency virus protease that correlate with reduced susceptibility to the protease inhibitor lopinavir among viral isolates from protease inhibitor-experienced patients. J Virol, 2001; 75:7462–7469.

24.

Hermans

, Schimt

, Kabeya

et al. Virological response to salvage therapy at 6 months in patients with B and non-B subtypes. Proceedings of the Ninth Conference on Retroviruses and Opportunistic Infections. Seattle, WA, 2002.

25.

Ananworanich

, Hirschel

, Sirivichayakul

et al. Absence of resistance mutations in antiretroviral-naive patients treated with ritonavir-boosted saquinavir. Antivir Ther, 2006; 11:631–635.

26.

Eron

Jr , Yeni

, Gathe

et al. The KLEAN study of fosamprenavir-ritonavir versus lopinavir-ritonavir, each in combination with abacavir-lamivudine, for initial treatment of HIV infection over 48 weeks: A randomised non-inferiority trial. Lancet, 2006; 368:476–482.

27.

Kempf

, King

, Bernstein

et al. Incidence of resistance in a double-blind study comparing lopinavir/ritonavir plus stavudine and lamivudine to nelfinavir plus stavudine and lamivudine. J Infect Dis, 2004; 189:51–60.

28.

Swindells

, DiRienzo

, Wilkin

et al. Regimen simplification to atazanavir-ritonavir alone as maintenance antiretroviral therapy after sustained virologic suppression. JAMA, 2006; 296:806–814.

29.

Delaugerre

, Flandre

, Chaix

et al. Protease inhibitor resistance analysis in the MONARK trial comparing first-line lopinavir-ritonavir monotherapy to lopinavir-ritonavir plus zidovudine and lamivudine triple therapy. Antimicrob Agents Chemother, 2009; 53:2934–2939.

30.

Geretti

, Harrison

, Green

, Dunn

. Exploring the effect of HIV-1 subtype on virological and immunological responses to first-line HAART. 6th European HIV Drug Resistance Workshop. Budapest, 2008.