NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Geretti AM, editor. Antiretroviral Resistance in Clinical Practice. London: Mediscript; 2006.

Cover of Antiretroviral Resistance in Clinical Practice

Antiretroviral Resistance in Clinical Practice.

Show details

Appendix 16Case studies

1Development of resistance on first-line therapy
2Evolution of resistance during therapy with ritonavir-boosted protease inhibitors
3Reluctance to take antiretroviral therapy
4A child with uncertain treatment history and extensive drug resistance
5Unrecognised transmitted resistance
6A case of persistent low-level viraemia
7A young black African woman presenting for HIV testing

Case study 1: Development of resistance on first-line therapy

Presented by Ben Killingley
Discussion by Anna Maria Geretti

Patient background

A 26-year-old homosexual man was diagnosed HIV-1 positive when he presented with fever, rash, myalgia and lymphadenopathy. The test results for HIV-1 antibody/antigen were consistent with acute seroconversion. The CD4 count was 138 cells/μl (11%). Serology tests for hepatitis B, hepatitis C and syphilis were negative.

Would you request a baseline resistance test?

There is extensive evidence for the transmission of drug-resistant HIV-1, although the prevalence of transmitted (or primary) resistance varies considerably across cohorts. Current guidelines recommend that a baseline resistance test should be carried out in newly diagnosed patients presenting with a recent infection. Knowledge of viral susceptibilities allows the construction of an appropriate antiretroviral regimen at the time when therapy becomes indicated.

Would you request a genotypic or a phenotypic resistance test?

A genotypic test would be indicated in this setting. Genotypic testing may detect mutations that signal the presence of resistance without having significant impact on the virus phenotype. One example is that of mutation T215S in the reverse transcriptase gene, where the wild-type threonine (encoded by the triplet ACC) is replaced by serine (encoded by TCC). This represents a transition mutation (or revertant) between the wild-type and the resistant mutant T215Y, where threonine is replaced by tyrosine (encoded by TAC). The T215S does not confer significant phenotypic resistance to zidovudine (ZDV) or other nucleoside reverse transcriptase inhibitors (NRTIs). However, detection of T215S signals the presence of the resistant mutant even though it has not been detected, and this has been associated with a nearly 3-fold increase in the risk of virological failure in individuals receiving thymidine analogues as part of their first antiretroviral regimen [1].

A genotypic resistance test showed the following results:

  • Reverse transcriptase (RT ): wild-type
  • Protease (PR): L63T

The L63T mutation in the protease gene is a polymorphism that does not confer significant resistance to the protease inhibitors (PIs) in the absence of other mutations. Treatment was started 3 weeks after diagnosis with tenofovir (TDF; 245 mg once daily), lamivudine (3TC; 300 mg once daily) and ritonavir-boosted atazanavir (ATV/r; 300/100 mg once daily). The plasma HIV-1 RNA load at the start of therapy was 83,700 copies/ml.

Is antiretroviral therapy recommended in patients with primary HIV-1 infection?

To date, there is no evidence of long-term clinical benefit to support the routine use of antiretroviral therapy in primary infection. It has been proposed that early treatment may be important in preserving HIV-1-specific immune responses. Individuals who present with severe or prolonged symptoms may benefit from therapy, although the optimal duration of therapy remains to be determined. The potential benefits of early treatment should be balanced against the possible problems of drug toxicity, compliance and emergence of resistance. Trials are currently under way to assess treatment of early HIV-1 infection and, whenever possible, patients should be offered the opportunity to enter these trials.

After 12 weeks of treatment, the CD4 cell count had risen to 662 cells/μl (26%). The viral load was 9744 copies/ml.

Would you consider this response satisfactory?

The improvement in CD4 cell count is anticipated after a fall during seroconversion. After 12 weeks of therapy, the majority of patients would be expected to have achieved a viral load <50 copies/ml. In this case, the viral load declined by less than 1 log10 (4.9 to 4.0 log10 copies/ml). This suboptimal response may signal problems with compliance, pharmacokinetics or transmitted resistance at levels below the detection limit of the genotypic resistance assay.

What action is required in this case?

The recommended actions include reviewing adherence and tolerability, excluding unfavourable drug–drug interactions, measuring plasma ATV levels, and repeating the viral load measurement and the genotypic resistance test.

Adherence was reviewed and encouraged. The bilirubin levels had increased from 12 to 39 μmol/l since the start of treatment, consistent with ATV dosing. The patient was not taking other prescribed or over-the-counter medications, and the ATV plasma trough levels were satisfactory. The patient continued on TDF/3TC/ATV/r. After a further 8 weeks of therapy (20 weeks of therapy in total), the viral load was 1305 copies/ml. A genotypic resistance test showed the following results:

  • RT: M184I/V/M
  • PR: L63T

What is the significance of the M184I/V/M mutations?

The resistance test has detected a mixed population of M184 wild-type virus (M184M), the intermediate mutant M184I and the mutant M184V. These changes signal emerging drug resistance, affecting 3TC first as a result of the low genetic barrier to resistance of this drug. M184I/V mutations confer resistance to 3TC and emtricitabine (FTC). They also confer low-level resistance to abacavir (ABC) and didanosine (ddI), but in isolation do not appear to compromise clinical responses to these drugs [2]. The PI resistance profile was unchanged.

The antiretroviral regimen was changed to ABC (600 mg once daily), TDF (245 mg once daily) and ritonavir-boosted lopinavir (LPV/r 400/100 twice daily). After 8 weeks, the viral load was 1396 copies/ml. A further genotypic resistance test showed the following results:

  • RT: M184M/V, K65K/R, L74L/V
  • PR: M36I, L63P

What is the clinical evidence in support of TDF/ABC as an NRTI backbone?

Concerns about the use of TDF/ABC were expressed following the high rates of virological failure with the triple NRTI combination of TDF/ABC/3TC in first-line therapy [3]. Pharmacokinetic studies have not detected an interaction between TDF and ABC. In addition, data from observational and cohort studies indicated good outcomes with TDF/ABC as the NRTI backbone in combination with non-nucleoside reverse transcriptase inhibitors (NNRTIs) and PIs [4,5], although controlled prospective data in first-line therapy are lacking.

How do you interpret the results of the genotypic resistance test?

The detection of mixed populations of wild-type and resistant virus at multiple sites is consistent with a virus population evolving under suboptimal virological suppression. M184V is maintained by ABC, although ABC is less effective than 3TC in maintaining selective pressure on the mutation. K65R was selected by TDF and ABC, and L74V was selected by ABC. Emergence of secondary protease mutations, such as M36I, in the absence of major protease resistance mutations, has been observed in patients with detectable viral load while receiving therapy with ritonavir-boosted PIs [68]. These changes signal virus evolution under drug pressure, but do not appear to confer significant shifts in phenotypic susceptibility to the PIs in routine phenotypic assays. The significance of the emerging mutations is currently uncertain.

What is the impact of the resistance mutations identified?

K65R confers significant NRTI resistance affecting TDF, ABC, ddI, 3TC, FTC and possibly stavudine (d4T) [9]. L74V in isolation confers modest phenotypic resistance to some NRTIs, especially ddI, but its resistance effects become more significant in the presence of M184V and K65R. The combination of M184V, K65R and L74V is likely to result in significant resistance to 3TC, FTC, TDF, ddI and ABC. Resistance effects for d4T are also possible. At least in vitro, K65R, L74V and M184V confer hypersusceptibility to ZDV.

Therapeutic drug monitoring (TDM) for LPV showed the following results:

  • Pre-dose level (Ctrough) = 7978 ng/ml
  • 2-hour peak level (Cmax) = 9115 ng/ml
  • Expected concentrations 5500±4000 (trough) and 9500±4400 ng/ml (peak)

The LPV levels were within the range required to suppress wild-type virus. The antiretroviral regimen was intensified with the addition of saquinavir (SQV; 1000 mg twice daily) and ABC was replaced by ZDV (250 mg twice daily).

What factors can affect drug plasma levels?

There are many factors that may affect drug levels, including genetic determinants of drug metabolism, effects of food on drug absorption, renal or hepatic impairment, drug interactions including interaction with herbal and other supplements, and other host-related factors such as age, body weight and pregnancy.

What was the rationale for maintaining TDF in the new regimen despite the presence of resistance?

The presence of K65R is likely to confer significant resistance to TDF. In single-drug intensification studies of patients with detectable viral load while on combination therapy, the addition of TDF in six patients with K65R did not produce significant changes in viral load [10]. The rationale for maintaining TDF in the regimen was to preserve selective pressure on K65R, thereby increasing susceptibility to ZDV and antagonising the emergence of ZDV resistance. There is limited evidence that these in vitro effects translate into significant clinical benefit. Case reports have described patients with K65R failing TDF-containing highly active antiretroviral therapy who showed virological suppression after single-drug intensification with ZDV [11].

The patient received counselling to support his adherence as it had become apparent that there had been problems with his compliance during previous therapy. A plan for discontinuation of therapy after 12 months was discussed. After a further 6 weeks, the viral load was <50 copies/ml and the CD4 cell count was 649 cells/μl (26%).

References

1.
Violin M, Cozzi-Lepri A, Velleca R. et al. Risk of failure in patients with 215 HIV-1 revertants starting their first thymidine analog-containing highly active antiretroviral therapy. AIDS. 2004;18:227–235. [PubMed: 15075540]
2.
Winters M, Bosch RJ, Albrecht MA, Katzenstein DA. Clinical impact of the M184V mutation on switching to didanosine or maintaining lamivudine treatment in nucleoside reverse transcriptase inhibitor experienced patients. J Infect Dis. 2003;188:537–540. [PubMed: 12898440]
3.
Gallant J, Rodriguez AE, Weinberg W et al. Early non-response to tenofovir DF (TDF) + abacavir (ABC) and lamivudine (3TC) in a randomised trial compared to efavirenz (EFV) + ABC + 3TC: ESS30009. 43rd Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, 2003, Abstr. 1722a.
4.
Ward D, Curtin J and Owne C. Combination therapy with tenofovir and abacavir in clinical practice. Seventh International Congress on Drug Therapy in HIV Infection, Glasgow, 2004, Abstr. P309.
5.
Perez M, Terron J, Antela A et al. Virological outcome of TDF plus ABC-based regimens in previously HIV suppressed patients (24wk preliminary results from recover study). Seventh International Congress on Drug Therapy in HIV Infection, Glasgow, 2004, Abstr. P308.
6.
Gulick RM, da Silva B, McMillan F et al. Lopinavir/ritonavir (LPV/r)-based therapy in antiretroviral naive HIV-infected patients: 6-year follow-up of study 720. Seventh International Congress on Drug Therapy in HIV Infection, Glasgow, 2004, Abstr. P28.
7.
Kempf DJ, Isaacson JD, King MS. 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. [PMC free article: PMC114981] [PubMed: 11462018]
8.
Molina JM, Gathe J, Lim PL et al. Comprehensive resistance testing in antiretroviral naive patients treated with once-daily lopinavir/ritonavir plus tenofovir and emtricitabine: 48-week results from study 418. 15th International AIDS Conference, Bangkok, 2004, Abstr. WePeB5701.
9.
Garcia-Lerma JG, Macinnes H, Bennett D. et al. A novel genetic pathway of human immunodeficiency virus type 1 resistance to stavudine mediated by the K65R mutation. J Virol. 2003;77:5685–5693. [PMC free article: PMC154026] [PubMed: 12719561]
10.
Miller MD, Margot N, Lu B. et al. Genotypic and phenotypic predictors of the magnitude of response to tenofovir disoproxil fumarate treatment in antiretroviral-experienced patients. J Infect Dis. 2004;189:837–846. [PubMed: 14976601]
11.
Staszewski S, Dauer B, Stuermer M et al. Intensification of a failing regimen with AZT may cause sustained virologic suppression in the presence of the K65R mutation. Third European HIV Drug Resistance Workshop, Athens, 2005, Abstr. 89.

Case study 2: Evolution of resistance during therapy with ritonavir-boosted protease inhibitors

Presented by Gaia Nebbia
Discussion by Anna Maria Geretti

Patient background

A 35-year-old homosexual man was diagnosed HIV-1 positive in November 1999, when he presented with pneumonia, oesophageal candidiasis, anaemia (haemoglobin level, 9.7 g/dl) and weight loss. The CD4 count was 174 cells/μl (8%) and the plasma viral load was >750,000 copies/ml. Antiretroviral treatment was started soon after the diagnosis with Combivir [zidovudine (ZDV)/lamivudine (3TC), one tablet twice daily), indinavir (IDV, 600 mg twice daily) and ritonavir (RTV, 200 mg twice daily]. After 6 weeks, the patient's anaemia became more severe (haemoglobin level, 4.7 g/dl), possibly because of the ZDV. The nucleoside reverse transcriptase inhibitor (NRTI) backbone was changed to stavudine (d4T, 40 mg twice daily) and 3TC (150 mg twice daily), while IDV and RTV were continued. In September 2000, the CD4 count had increased to 827 cells/μl (19%) and the viral load was undetectable (<50 copies/ml). In April 2001, the viral load rebounded to 545 copies/ml.

What should the next step be?

Problems with adherence are the most likely explanation for the viral load rebound and should be addressed immediately. It is also important to review tolerability, repeat the viral load measurement and exclude pharmacokinetic interactions that may reduce drug exposure.

The patient reported occasionally missing doses over the previous 3 months and received counselling to support adherence. No significant problems with tolerability were reported and unfavourable drug interactions were excluded. The repeat viral load test showed 1250 copies/ml. A genotypic resistance test showed the following results:

  • Reverse transcriptase (RT): M184V
  • Protease (PR): L10V, M46I

What is the significance of these protease resistance mutations?

Both protease resistance mutations are traditionally classified as secondary protease mutations. When present with other mutations, M46I reduces susceptibility to IDV, RTV, nelfinavir (NFV), lopinavir (LPV), amprenavir (APV) and atazanavir (ATV). L10V also contributes resistance to all available protease inhibitors (PIs) when present with other mutations. The combination of L10V and M46I would be expected to reduce susceptibility to ATV and NFV, with some additional impact on susceptibility to IDV, RTV, LPV and APV.

In June 2001, the antiretroviral regimen was changed to abacavir (ABC, 300 mg twice daily), d4T (40 mg twice daily) and ritonavir-boosted LPV (LPV/r, three capsules twice daily). After 5 weeks, the viral load was <50 copies/ml and the CD4 count was 1281 cells/μl (23%). In December 2002, the viral load was again detectable at 9970 copies/ml. Therapeutic drug monitoring (TDM) showed that the trough concentration (Ctrough) for LPV was 8.7 times the estimated level required to suppress wild-type virus. A genotypic resistance test showed the following results:

  • RT: M184V
  • PR: L10V, V32I, M46I

How do you interpret these findings?

The genotypic test results indicate protease evolution under drug pressure, with the emergence of one additional mutation, V32I, a substrate cleft mutation that contributes resistance to LPV as well as to IDV, RTV and APV. Both V32I and M46I have been shown to emerge in patients failing LPV/r, clearly indicating a role in resistance to the drug [1]. In PI-experienced patients, resistance to LPV has been associated with a large number of mutations, including those at codons 10, 16, 20, 24, 32, 33, 34, 36, 43, 46 , 47, 48, 50, 53, 54, 58 , 63, 71, 73, 74 , 82, 84, 89 and 90. The number of mutations resulting in high-level resistance to LPV varies in different studies, but the accumulation of multiple mutations is expected to be required to confer significant levels of resistance to the drug [2]. The combination of the three protease mutations is expected to confer low-level resistance to LPV/r, as well as to ATV, IDV, RTV and possibly saquinavir (SQV), with more significant resistance to APV and NFV.

The TDM result showed that the plasma levels of LPV were above the minimum concentration required to inhibit wild-type virus and probably above the concentration required to inhibit virus with a modest degree of LPV resistance. Taken together, these data suggest a problem of suboptimal adherence driving viral evolution towards increasing levels of drug resistance.

The viral load remained detectable at 1320 copies/ml and in, March 2003, the antiretroviral regimen of ABC/d4T/LPV/r was intensified with the addition of TDF (245 mg once daily).

Is there evidence that intensification with TDF is effective in patients experiencing virological failure?

Two studies (Gilead Sciences 902 and 907) have shown that, in drug-experienced patients with low-level viraemia, the addition of TDF to a failing regimen resulted in an average viral load reduction of −0.6 log10 copies/ml by week 24. In patients with M184V, the drop in viral load was −0.96 log10 copies/ml compared with −0.12 log10 copies/ml observed in patients who received placebo [3].

After 6 weeks, the viral load was <50 copies/ml. The CD4 count was 998 cells/μl (24%). In December 2003, despite good reported adherence, the viral load rebounded to 1510 copies/ml. A repeat TDM test showed that the Ctrough for LPV was 9.8 times the estimated level required to suppress wild-type virus. A genotypic resistance test showed the following results:

  • RT: D67N, K70R, K219Q
  • PR: L10V, V32I, M46I

What is the significance of these mutations?

The resistance results show three thymidine analogue mutations (TAMs), reflecting selective pressure with d4T. The combination of D67N, K70R and K219Q is expected to confer significant resistance to ZDV, whereas D67N contributes modest resistance to d4T, ddI, ABC and TDF. The M184V mutation is no longer detected because of lack of selective pressure with 3TC and the lower degree of selective pressure exerted by ABC on the mutation. No further evolution of protease resistance was observed.

In January 2004, the antiretroviral regimen was changed to TDF (245 mg once daily), LPV/r (four capsules twice daily) and efavirenz (EFV, 600 mg once daily).

Why was the dose of LPV/r increased?

The TDM results give an indication of the drug concentration relative to the minimum estimated level required to suppress wild-type virus. Given the difficulty in establishing the optimal level required to inhibit drug-resistant virus, increasing the dose of LPV might help to ensure that drug exposure is adequate for suppressing partially resistant virus. This strategy must be balanced against the risk of increasing drug toxicity and reducing adherence through an increased pill burden.

After 4 months, the patient remained on TDF/LPV/r/EFV. The viral load was 9900 copies/ml and the CD4 count was 912 cells/μl (22%). TDM showed a Ctrough for LPV of 8.7 times the estimated level required to suppress wild-type virus. The EFV concentration was within the target range. A further genotypic test showed the following results:

  • RT: D67N, K70R, K101E, K103N, K219Q
  • PR: L10V, V32I, M46I, I47A

How do you interpret this resistance profile?

The genotypic test showed two further resistance mutations, K101E and K103N, selected by EFV and together conferring high-level resistance to EFV and nevirapine (NVP). An additional mutation is noted in the protease. I47A has been previously observed to emerge during failure of LPV/r [1]. Although traditionally classified as a secondary protease mutation, I47A is expected to confer significant resistance to LPV. Phenotypic testing may help in assessing the impact of the novel resistance mutation pattern of L10V, V32I, M46I and I47A on resistance to LPV and other PIs.

In November 2004, a virtual phenotype (vircoTYPE, reproduced, with permission, from Virco BVBA) was requested with the following results:

Image casestudy2fu1.jpg

How do you interpret this report?

In agreement with the genotypic resistance test result, significant resistance effects are predicted for ZDV and available NNRTIs. A virtual phenotype is available only for APV and LPV. For other PIs, the interpretation of resistance is rule-based because of insufficient matches in the database, which can occur with novel resistance patterns. The predicted concentration of LPV required to inhibit viral replication by 50% (IC50) relative to the IC50 of a reference virus (i.e. the fold change for LPV) is 71.6, which is well above the upper clinical cut-off of 40 determined within study 957 [4].

A phenotypic test (Antivirogram, Virco) was requested, with the following results:

DrugFold change
Indinavir10.0
Nelfinavir7.8
Saquinavir0.2
Amprenavir5.8
Lopinavir63.8
Atazanavir2.1
Tipranavir0.7

What is the significance of these findings?

Using the clinical cut-offs proposed by Virco [5], the fold changes indicate: high-level resistance to LPV/r and NFV, with at least 80% loss of virological response; intermediate resistance to IDV/r and APV/r, with between 20% and 80% loss of response; susceptibility to ATV/r; and hypersusceptibility to SQV/r. There is also susceptibility to ritonavir-boosted tipranavir (TPV/r).

Salvage therapy was started with TDF, ABC, SQV/r (1000/100 mg twice daily) and enfuvirtide. After 4 weeks, the viral load was <50 copies/ml. After 6 months, the viral load remained suppressed and the patient discontinued enfuvirtide without virological rebound.

What other options were available for this patient?

The options considered included:

  • Structured treatment interruption (STI). This strategy would help to avoid the development of further resistance, reduce drug exposure, and possibly favour improved adherence on resumption of therapy. However, it poses a risk of falling CD4 counts and rising viral load, with the potential for clinical events. Although the patient has a good CD4 count, a decline in count may occur rapidly in patients with nadir CD4 counts <200 cells/μl. The patient would therefore require close monitoring. Maintenance therapy with 3TC alone may be an alternative to a complete STI.
  • Continuing the failing regimen. Despite repeated episodes of virological failure, the patient has achieved and maintained a high CD4 count. Continuing the current regimen would be the least favoured option because of the risk of accumulating further resistance, which may compromise future drug options.
  • Entering a clinical trial. This would provide access to novel treatment options and strategies.

References

1.
Friend J, Parkin N, Liegler T. et al. Isolated lopinavir resistance after virological rebound of a ritonavir/lopinavir-based regimen. AIDS. 2004;18:1965–1966. [PubMed: 15353986]
2.
Kempf DJ, Isaacson JD, King MS. 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. [PMC free article: PMC114981] [PubMed: 11462018]
3.
Miller MD, Margot N, Lu B. et al. Genotypic and phenotypic predictors of the magnitude of response to tenofovir disoproxil fumarate treatment in antiretroviral-experienced patients. J Infect Dis. 2004;189:837–846. [PubMed: 14976601]
4.
Kempf DJ, Isaacson JD, King MS. et al. Analysis of the virological response with respect to baseline viral phenotype and genotype in protease inhibitor-experienced HIV-1-infected patients receiving lopinavir/ritonavir therapy. Antivir Ther. 2002;7:165–174. [PubMed: 12487383]
5.
Bacheler L, Winters B, Harrigan PR et al. Estimation of clinically relevant cutoff for phenotypic resistance data. Seventh International Congress on Drug Therapy in HIV Infection, Glasgow, 2004, Abstr. P93.

Case study 3: Reluctance to take antiretroviral therapy

Presented by Thomas Lutz
Discussion by Schlomo Staszewski and Anna Maria Geretti

Patient background

A 62-year-old woman was diagnosed with HIV-1 infection in July 2002 during investigations for thoracic herpes zoster infection. Her baseline CD4 count was 204 cells/μl (14%) and the plasma HIV-1 RNA load was 297,000 copies/ml. A resistance test showed a wild-type genotype. Despite recurring oral candidiasis, fatigue, lymph node swelling, night sweats and mild peripheral neuropathy, the patient at first refused antiretroviral therapy. In March 2003, when her CD4 count was 189 cells/μl (12%), it was agreed that therapy was necessary, despite significant reluctance on the part of the patient. A regimen with lamivudine (3TC, 150 mg twice daily), tenofovir (TDF, 245 mg once daily) and nevirapine (200 mg twice daily) was selected. On day 10 of therapy, a pronounced and intensely itchy exanthema developed in the absence of an elevation in serum transaminase levels. She was treated with antihistamines and systemic steroids, and therapy was switched to 3TC (150 mg twice daily), stavudine (d4T, 20 mg twice daily) and TDF (245 mg once daily). The patient refused a ritonavir-boosted protease inhibitor (PI/r) because she was worried about pill burden and side effects, and she also refused to switch to efavirenz (EFV) because she feared another outbreak of rash and central nervous system symptoms such as dizziness.

Why was a low dose of d4T given?

The normal recommended dose of d4T is 40 mg twice daily for >60 kg body weight and 30 mg twice daily for a lower body weight (the patient's body weight was 55 kg). It has been shown that reducing the dose of d4T (i.e. 30 mg twice daily for body weight >60 kg) reduces the risk of side effects, including fat loss and blood dyslipidaemia, while preserving the antiviral activity of the drug [1]. In view of this patient's history of peripheral neuropathy, a reduced dose of d4T was used.

The patient tolerated the new regimen well and showed a durable suppression to <50 copies/ml without evidence of virological blips or sustained viral load rebound. The CD4 count increased to 360 cells/μl (19%). In July 2005, after 27 months of successful therapy, the viral load rebounded to 3910 copies/ml. A genotype resistance test showed the following results:

  • Reverse transcriptase: K65R, M184V
  • Protease: G16E, M36I

How do you interpret these findings?

The changes detected in the protease gene are polymorphisms not associated with resistance in the absence of other mutations. In the reverse transcriptase, the K65R mutation was selected by TDF and possibly by d4T in the regimen, whereas M184V was selected by 3TC. No thymidine analogue mutations (TAMs) were observed, despite selective pressure with the thymidine analogue d4T. The two mutations K65R and M184V together have been reported to confer the median fold change in drug susceptibility by phenotypic testing (Phenosense, Virologic) [2] as shown in Table 1.

Table 1. Fold change in drug susceptibility.

Table 1

Fold change in drug susceptibility.

These results indicate high-level resistance to 3TC (and emtricitabine). The level of phenotypic resistance to abacavir (ABC) and didanosine (ddI) is low, but nonetheless near or above the proposed clinical cut-offs that predict some or complete loss of response to the drugs. No phenotypic resistance is observed for d4T, although in vitro data suggest an impact of K65R on d4T susceptibility that may not be captured by the phenotypic test [2]. Furthermore, emergence of K65R has been reported in patients receiving d4T/3TC/EFV [3]. It is interesting that the median level of phenotypic resistance to TDF is low despite the presence of K65R, a mutation that is known to severely impair responses to TDF in vivo [4]. This finding is probably the result of the phenotypic re-sensitisation effects of M184V. Finally, hypersusceptibility to zidovudine (ZDV) is observed, reflecting the sensitisation effects of both K65R and M184V on ZDV.

In July 2005, based on the genotypic test results, d4T was replaced by ZDV (250 mg twice daily), while TDF and 3TC were maintained. Four weeks later, the virus load had dropped to 64 copies/ml. The patient continues on the current regimen with good adherence and tolerability, and with a viral load of <50 copies/ml.

What is the current understanding of triple nucleoside reverse transcriptase inhibitor (NRTI) regimens?

Triple NRTI therapy is not generally recommended. Regimens of ZDV/3TC/ABC and d4T/3TC/ABC have been shown to be less efficacious than standard non-nucleoside reverse transcriptase inhibitor (NNRTI)- or PI-based therapy, whereas unacceptably high rates of virological failure have been observed with the combinations ddI/d4T/3TC, ddI/d4T/ABC, ABC/TDF/3TC and TDF/ddI/3TC. The high rate of virological failure observed with ABC/TDF/3TC and TDF/ddI/3TC has been related to the strong selective pressure on the K65R mutation, which confers significant resistance to all drugs in these regimens. Evidence suggests that inclusion of ZDV may improve the performance of triple NRTI regimens, reflecting favourable antagonistic interactions between the mutational pathways induced by ZDV (TAMs) and those induced by ABC (L74V and K65R) or TDF (K65R). The hypersusceptibility to ZDV conferred by both K65R and M184V may play a role in improving responses [5]. Reciprocal interactions between mutations include: (1) the antagonistic effects of K65R and M184V on TAM-mediated excision of the incorporated NRTI; and (2) the antagonistic effects of TAMs on K65R-mediated substrate discriminations. A significant impact of both K65R and M184V on the viral replication capacity is also likely [6]. Both ZDV and d4T are regarded as thymidine analogues but there seem to be differences in the selection of resistance mutations with regard to their effect on the hypersusceptibility to the K65R-containing virus. d4T may have the potential to select K65R [7] especially in combination with ddI and ABC [8]. Our case study suggests that the effect of hypersusceptibility of ZDV in the presence of K65R is greater than that of d4T. Quadruple NRTI regimens with ZDV/3TC/ABC/TDF are currently being evaluated as a possible option where standard NNRTI-or PI-based therapy cannot be administered.

References

1.
Milinkovic A, Lopez S, Vidal S et al. A randomized open study comparing the effect of reducing stavudine dose vs. switching to tenofovir on mitochondrial function, metabolic parameters, and subcutaneous fat in HIV-infected patients receiving antiretroviral therapy containing stavudine. 12th Conference on Retroviruses and Opportunistic Infections, Boston, 2005, Abstr. 857.
2.
Underwood M, St Clair M, Ross L et al. Cross-resistance of clinical samples with K65R, L74V, and M184V mutations. 12th Conference on Retroviruses and Opportunistic Infections, Boston, 2005, Abstr. 714.
3.
Garcia-Lerma JG, Macinnes H, Bennett D. et al. A novel genetic pathway of human immunodeficiency virus type 1 resistance to stavudine mediated by the K65R mutation. J Virol. 2003;77:5685–5693. [PMC free article: PMC154026] [PubMed: 12719561]
4.
McColl DJ, Margot NA, Cheng AK et al. Baseline predictors of resistance development in Study 903: a three year, randomized comparison of tenofovir disoproxil fumarate (TDF) or stavudine (d4T) in combination with lamivudine (3TC) and efavirenz (EFV) in treatment-naive HIV-1 patients. Second European Drug Resistance Workshop, Rome, 2004.
5.
Miller MD, Margot N, Lu B. et al. Genotypic and phenotypic predictors of the magnitude of response to tenofovir disoproxil fumarate treatment in antiretroviral-experienced patients. J Infect Dis. 2004;189:837–846. [PubMed: 14976601]
6.
White KL, Margot NA, Wrin T. et al. Molecular mechanisms of resistance to human immunodeficiency virus type 1 with reverse transcriptase mutations K65R and K65R+M184V and their effects on enzyme function and viral replication capacity. Antimicrob Agents Chemother. 2002;46:3437–3446. [PMC free article: PMC128721] [PubMed: 12384348]
7.
Gallant JE, Staszewski S, Pozniak AL. et al. 903 Study Group. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA. 2004;292:191–201. [PubMed: 15249568]
8.
Roge BT, Barfod TS, Kirk O. et al. Resistance profiles and adherence at primary virological failure in three different highly active antiretroviral therapy regimens: analysis of failure rates in a randomized study. HIV Med. 2004;5:344–351. [PubMed: 15369509]

Case study 4: A child with uncertain treatment history and extensive drug resistance

Presented by Caroline Foster
Discussion by Hermione Lyall and Anna Maria Geretti

Patient background

A 12-year-old boy with vertically acquired HIV-1 infection was visiting the UK from South America when he was admitted to hospital with fever, abdominal pain, blanching maculopapular rash and mild conjunctivitis. He had no previous AIDS-defining illness and had not been told of his HIV-1 status. Limited information on previous treatment history was available. He had been intermittently on antiretroviral therapy since the age of 2 years, including periods of mono and dual therapy with nucleoside reverse transcriptase inhibitors (NRTIs). Three weeks previously he had started lamivudine (3TC, 100 mg twice daily), stavudine (d4T, 30 mg twice daily) and the non-nucleoside reverse transcriptase inhibitor (NNRTI) nevirapine (NVP, 300 mg once daily). His CD4 count was 780 cells/μl (45%) and plasma HIV-1 RNA load was 3112 copies/ml. No previous CD4 counts or viral load measurements were available. The alanine aminotransferase (ALT) level was 86 IU/l (normal range, <40 IU/l) and the lactate dehydrogenase (LDH) level was 668 U/l (normal range, 50–450 U/l). Bacteriological and virological cultures of blood, stool, urine and throat swab yielded no growth. Serological tests for hepatitis A, hepatitis B, hepatitis C, Epstein-Barr virus and cytomegalovirus were negative.

In view of the presenting features, would you consider stopping NVP?

The toxicity profile of NVP in children is similar to that described in adults. The typical side effects are most commonly seen within 6 weeks of starting therapy. Whereas a mild rash is common and may be self-limiting, this patient has evidence of systemic involvement and conjunctivitis. This suggests the potential development of Stevens–Johnson syndrome; hence, NVP should be stopped. Fulminant fatal hepatitis associated with NVP has been reported in adults, with the risk of severe hepatic toxicity being greatest in women and those with higher CD4 counts (above 250 cells/μl).

His symptoms resolved and liver function normalised following the withdrawal of NVP, which was replaced by ritonavir-boosted lopinavir (LPV/r, three capsules twice daily). A week later he returned to his country of origin. There he was unable to access LPV/r and was switched to Combivir [zidovudine (ZDV)/3TC, one tablet once daily] and efavirenz (EFV, 300 mg once daily). These doses were below those recommended for his weight of 38.5 kg. After 6 months, he moved permanently to the UK to live with an aunt following the death of his grandmother, who had been his main carer. At this time, his height and weight were average for his age and he showed no signs of fat re-distribution.

What investigations are required at this point?

This patient is treatment-experienced and is currently receiving inadequate drug doses. Accurate dosing in children can be difficult using standard formulations, and can easily result in either under-dosing, leading to risk of virological failure, or over-dosing, leading to risk of toxicity. In this case, there is a substantial risk of suboptimal virological suppression and emergence of drug resistance. Investigations required include CD4 count, viral load measurement and a genotypic resistance assay.

His combination therapy was continued with Combivir (one tablet twice daily) and EFV (400 mg once daily) pending results of the above investigations. The CD4 count was 900 cells/μl (31%) and the viral load was 2214 copies/ml. The genotypic resistance test showed the following mutations:

  • Reverse transcriptase (RT ): M41L, E44D, D67N, T69N, K103N, V108I, M184V, L210W, T215Y
  • Protease (PR): L10I, L63P, V77I, V82I

What is the effect of these mutations on susceptibility to the NRTIs?

There is extensive resistance to the NRTIs. The presence of pathway 1 thymidine analogue mutations (TAMs; including those at codons 41, 210 and 215), with D67N, significantly reduces susceptibility to ZDV, d4T, didanosine (ddI), abacavir (ABC) and tenofovir (TDF). The presence of M184V confers high-level resistance to 3TC and emtricitabine (FTC), while also increasing resistance to ABC. In vitro, M184V also increases resistance to ddI, although this effect is not apparent in vivo. Finally, in the presence of TAMs, M184V reduces the level of resistance to ZDV, d4T and TDF. In this case, there may be residual antiviral activity with TDF and possibly ddI and, if M184V is maintained, additional antiviral activity may be gained by exploiting the re-sensitisation effects of M184V on ZDV, d4T and TDF, and the negative impact of this mutation on viral fitness.

What is the effect of these mutations on susceptibility to the NNRTIs?

There is complete resistance to NVP and EFV, with no significant residual antiviral activity or negative impact on viral fitness.

What is the effect of these mutations on susceptibility to the protease inhibitors?

The mutations detected are polymorphisms that do not confer significant resistance to protease inhibitors (PIs). It should be noted that the V82I mutation behaves differently from other codon changes at the same position (i.e. V82A/S/T/F/M). Whereas the latter reduce susceptibility to multiple PIs, V82I occurs commonly in drug-naive patients, especially in those infected with non-B HIV-1 subtypes, and does not have significant resistance effects.

What are the management options?

Three possible options were considered:

  • Continue the current regimen of Combivir and EFV. The patient has maintained a high CD4 count despite virological failure. However, little, if any, benefit is to be gained from continuing EFV after emergence of K103N. Furthermore, ongoing viral replication may lead to the acquisition of further NNRTI-resistance mutations, possibly co-existing on the same viral genome as double and triple mutants, which may reduce his chances of responding to the new NNRTIs currently in development.
  • Stop therapy. A structured treatment interruption (STI) may be considered, preferably within a clinical trial. Ongoing trials are evaluating the option of stopping therapy in children and adolescents with a CD4 count >30% and viral load <50 copies/ml, and recommencing therapy once the CD4 count falls to 20% [1]. This child has survived to 12 years of age with relatively few symptoms despite suboptimal therapy, but his nadir CD4 count is unknown. The risk associated with stopping therapy is an increment in viral load and a rapid fall in CD4 count, which may lead to clinical events. As an alternative to a complete STI, a period of monotherapy with 3TC may be considered with close monitoring. This is based on evidence derived from studies in adults showing that in patients with extensive resistance, including M184V, 3TC monotherapy leads to a slower CD4 count decline compared with stopping therapy completely [2].
  • Change therapy. The patient has extensive NRTI resistance and complete resistance to available NNRTIs. A salvage regimen will need to be based on the use of a ritonavir-boosted PI. The NRTI backbone may include TDF, ZDV and 3TC, in the hope of exploiting the potential beneficial effects of maintaining selective pressure on the M184V mutation. The combination of TDF and ZDV may also be beneficial, given the antagonism between ZDV-induced TAMs and TDF-induced K65R. Thus a treatment option may include Combivir, TDF and LPV/r, with the possibility of intensifying therapy with an additional PI if the response is suboptimal. The fusion inhibitor enfuvertide may be considered, although experience in children is currently limited.

Therapy was stopped. After 3 weeks the patient still had detectable plasma levels of EFV, and 4 months later he remains asymptomatic with a CD4 count of 910 cells/μl (33%).

Why does he have detectable plasma levels of EFV 3 weeks after discontinuing the drug?

The rates of metabolism of the NNRTIs are extremely variable between patients, with up to 20% showing detectable levels 3 weeks after last dosing [3]. This is of particular importance for patients stopping a successful NNRTI-based regimen because it can result in the selection of NNRTI-resistant virus. To reduce this risk, it is generally recommended that when interrupting an NNRTI-based regimen, the NRTI backbone should be continued for 2 weeks after stopping the EFV, preferably with the addition of a ritonavir-boosted PI.

What are the advantages of temporarily discontinuing therapy for this patient?

This child is facing major upheaval. He is approaching adolescence and has just moved to a new country. His first language is Spanish and he is due to start secondary school. Of most importance, he is unaware of his diagnosis and will require help and support to fully understand his HIV-1 status over the coming months. Adherence to antiretroviral therapy in adolescence can be particularly poor and this is an opportunity to improve his understanding and allow him to adjust to his new circumstances prior to restarting treatment. Thus, managing the psychosocial issues surrounding the case is the best strategy to prevent further accumulation of resistance while avoiding drug toxicity.

References

1.
Paediatric European Network for Treatment of AIDS (PENTA). http://www​.ctu.mrc.ac​.uk/penta/trials.htm (accessed 19 January 2006).
2.
Gianotti E, Menzo S, Danise A et al. E-184V study: immunological and virological correlates of HIV-1 replicative capacity. 14th HIV Drug Resistance Workshop, Quebec City, 2005, Abstr. 160.
3.
Taylor S, Allen S, Fidler S et al. Stop study: after discontinuation of efavirenz, plasma concentrations may persist for two weeks or longer. 11th Conference on Retroviruses and Opportunistic Infections, San Francisco, 2004, Abstr. 131.

Case study 5: Unrecognised transmitted resistance

Presented by Andrew Benzie
Discussion by Nicola Mackie and Anna Maria Geretti

Patient background

A 35-year-old homosexual man was diagnosed HIV-1-antibody positive in 1999 after presenting with recurrent episodes of oral candidiasis. At the time of diagnosis, the CD4 count was 20 cells/μl (3%) and the plasma viral load was greater than 500,000 copies/ml. Antiretroviral therapy was started 2 weeks after diagnosis with Combivir [zidovudine (ZDV)/lamivudine (3TC), one tablet twice daily] and nelfinavir (NFV, 750 mg twice daily). Viral load suppression to <50 copies/ml was achieved, accompanied by a rise in CD4 count to 230 cells/μl (15%). The viral load was <50 copies/ml after 60 weeks of treatment; rebound to 3650 copies/ml was observed at week 68. On week 77 of treatment, persistent viraemia, despite good reported adherence, led to change of therapy to didanosine (ddI, 400 mg once daily), stavudine (d4T, 40 mg twice daily) and the non-nucleoside reverse transcriptase inhibitor (NNRTI) nevirapine (NVP, 200 mg twice daily).

Based on current knowledge, would the management have been different?

Current UK guidelines would recommend obtaining a baseline resistance test before starting therapy. In the UK, the prevalence of resistance was 11% in drug-naive patients tested in 1999 [1]. In addition, all current guidelines would recommend resistance testing at the time of failure to inform the choice of the second-line regimen. The major concern would be whether thymidine analogue mutations (TAMs) have emerged during the time of detectable viral replication (at least 9 weeks) prior to therapy change. The emergence of TAMs would significantly diminish susceptibility to d4T and ddI, making the NNRTI-based regimen less robust. Finally, current guidelines would recommend against the use of d4T and ddI in combination because of a significant risk of toxicity.

Which drugs select for TAMs?

TAMS are nucleoside reverse transcriptase inhibitor (NRTI) resistance mutations occurring at reverse transcriptase (RT ) codons 41, 67, 70, 210, 215 and 219, and derive from treatment with the thymidine analogues ZDV and d4T. Two main pathways are recognised: pathway 1 includes the mutations M41L, L210W and T215Y; and pathway 2 includes D67N, K70R and K219Q/E. TAM pathway 1 is associated with greater levels of resistance to ZDV and d4T and cross-resistance to abacavir (ABC), ddI and tenofovir (TDF).

How quickly are TAMs selected in first-line therapy with ZDV and lamivudine (3TC)?

In patients receiving the NRTIs ZDV and 3TC in combination, accumulation of TAMS occurs slowly over time and in a step-wise manner. A direct antagonistic effect of M184V on the emergence of TAMs has been proposed [2].

The patient showed a suboptimal virological response to ddI/d4T/NVP, with persistent detectable low-level viraemia at levels between 833 and 21,298 copies/ml. The CD4 count remained stable at 340 cells/μl (19%). A retrospective genotypic resistance test was performed on a stored sample collected prior to the start of the current ddI/d4T/NVP regimen, while the patient was still receiving Combivir/NFV. The retrospective sample showed the following results:

  • Reverse transcriptase (RT ): K103N, M184V
  • Protease (PR): L63P

How can these results be explained?

The M184V mutation could have been selected by the 3TC in Combivir. K103N is selected by exposure to NNRTIs and was unexpected given that the sample had been taken before the start of ddI/d4T/NVP. It was therefore concluded that the mutation probably represented transmitted resistance. The L63P mutation is a common polymorphism that is not associated with resistance to the protease inhibitors (PIs) in the absence of other protease mutations. These findings indicated that the patient had been essentially receiving dual therapy with the two NRTIs (ddI and d4T) while on ddI/d4T/NVP.

A further genotypic resistance test was performed on a current sample, with the following results:

  • RT: D67N, K70R, K103N, K219Q
  • PR: L63P

How do you interpret these findings?

Ongoing viral replication while on ddI/d4T has led to the emergence of three TAMs from pathway 2: D67N, K70R and K219E. D67N is associated with resistance to ZDV, d4T, ddI, ABC and TDF. K70R and K219Q increase the level of resistance to ZDV. Together, these mutations would be expected to confer significant resistance to ZDV, but have a low impact on resistance to other NRTIs. K103N is being maintained by NVP in the regimen. Conversely, the M184V has become undetectable by routine testing due to lack of selective drug pressure. No changes have occurred in the protease gene.

At the request of the patient, treatment was stopped. After 12 months, the CD4 count was 100 cells/μl (14%) and the viral load was 186,305 copies/ml. In 2003, therapy was restarted with ABC (300 mg twice daily), ddI (250 mg once daily) and TDF (245 mg once daily). There was no virological or immunological response to ABC/ddI/TDF. A repeat genotypic resistance test showed the following results:

  • RT: K65R, K103N
  • PR: L63P

How is this pattern of mutations interpreted?

K65R was selected by TDF, ddI and ABC. The mutation reduces susceptibility to ddI and ABC and probably abrogates the activity of TDF. It may also reduce susceptibility to d4T.

Persistence of K103N long after NNRTI discontinuation is observed commonly and is probably related to the small overall impact of the mutation on viral fitness. No changes were observed in the protease gene.

The regimen was changed to Combivir (one tablet twice daily), ddI (400 mg once daily) and ritonavir-boosted lopinavir (LPV/r, 400/100 mg twice daily). Despite a reduction in viral load of greater than 2 log10 copies/ml within the first 4 weeks of therapy, the viral load was never completely suppressed. After 24 weeks, the viral load was 1787 copies/ml and the CD4 count was 310 cells/μl (17%). A genotypic resistance test showed the following results:

  • RT: D67N, K70R, K103N, M184V, K219Q
  • PR: L63P

How do you interpret the disappearance of the K65R mutation and the reappearance of TAMs and M184V?

HIV-1-infected individuals who develop drug resistance harbour a mixed population of viruses known as quasispecies. At any given time, the dominant virus population reflects a balancing act between escape from drug pressure and preserved viral fitness. Outgrowth of a population that is advantageous to the virus may cause the apparent disappearance of another population. Furthermore, antagonistic interactions may be at play. Antagonistic interactions occur between TAMs, especially T215Y and K65R: these resistance mutations do not commonly occur together and rarely coexist on the same viral genome [3,4]. It should be noted that mutations selected by drug pressure are likely to persist as minority virus variants and archived resistance. Routine resistance tests may only pick up populations of viral variants at a level of approximately 20%. As a result, virus populations present at low frequency escape detection by routine testing. Nonetheless, these mutants remain detectable by ultrasensitive methods, and can impact on virological responses [5,6]. Therefore, resistance should be regarded as permanent.

In this patient, the drug resistance mutations accumulated over time indicate a significant level of resistance to all available NRTIs and NNRTIs. The patient is now at risk of developing PI resistance mutations.

Due to significant diarrhoea and incomplete virological response, therapy was changed to 3TC (300 mg once daily), saquinavir (SQV, 1000 mg twice daily), fosamprenavir (FPV, 700 mg twice daily) and ritonavir (RTV, 100 mg twice daily). Two months later, the viral load was <50 copies/ml and the CD4 count was 360 cells/μl (20%). The patient reported good adherence. Therapeutic drug monitoring (TDM) showed a trough concentration (Ctrough) for SQV of 278 ng/ml (estimated Ctrough for wild-type virus, 100 ng/ml) and Ctrough for amprenavir of 1179 ng/ml (estimated minimum Ctrough for wild-type virus 400 ng/ml).

How do you interpret these TDM results?

The co-administration of FPV with SQV and RTV can result in a statistically significant decrease in SQV concentrations. In this case, the Ctrough levels for both SQV and FPV were above the recommended thresholds. Given good tolerability and virological response, the regimen was continued unchanged.

References

1.
UK Group on Transmitted HIV Drug Resistance. Time trends in primary resistance to HIV drugs in the United Kingdom: multicentre observational study. BMJ. 2005;331:1368–1371. [PMC free article: PMC1309643] [PubMed: 16299012]
2.
Ait-Khaled M, Stone C, Amphlett G. on behalf of the CNA3002 International Study Team. et al. M184V is associated with a low incidence of thymidine analogue mutations and low phenotypic resistance to zidovudine and stavudine. AIDS. 2002;16:1686–1689. [PubMed: 12172093]
3.
Parikh U, Barnas D, Bixby C et al. K65R and T215Y are not present on the same viral genome in plasma samples with both mutations detected by population sequencing. 12th Conference on Retroviruses and Opportunistic Infections, Boston, 2005, Abstr. 98.
4.
Wirden M, Malet I, Derrache A et al. Clonal analysis of HIV quasi-species in patients harbouring plasma genotype with K65R mutation associated with thymidine analogue mutations or L74V substitution. 12th Conference on Retroviruses and Opportunistic Infections, Boston, 2005, Abstr. 702. [PubMed: 15802984]
5.
Mellors J, Palmer S, Nissley D et al. Low frequency NNRTI-resistant variants contribute to failure of efavirenz-containing regimens. 11th Conference on Retroviruses and Opportunistic Infections, San Francisco, 2004, Abstr. 39.
6.
Bae AS, Waters JM, Margot NA et al. Pre-existing L74V is a risk factor for virological non-response and development of K65R in patients taking tenofovir DF (TDF). XIII International HIV Drug Resistance Workshop, Tenerife, 2004, Abstr. 158.

Case study 6: A case of persistent low-level viraemia

Presented by Iain Reeves
Discussion by Martin Fisher and Anna Maria Geretti

Patient background

A 63-year-old man presented in July 2003 having transferred his care from another hospital. He described having been diagnosed HIV-1 positive in May 2000, after presenting with cutaneous Kaposi's sarcoma. At the time of diagnosis his CD4 count was 50 cells/μl (5%) and the plasma HIV-1 RNA load was >500,000 copies/ml. He had started therapy with Combivir [zidovudine (ZDV)/lamivudine (3TC), one tablet twice daily] and nevirapine (NVP, 200 mg twice daily) in January 2001 and his viral load had been well controlled. At that time, the viral load was 122 copies/ml (15%) and the CD4 count was 232 cells/μl (15%).

What action should be taken?

Confirmation of clinical details, treatment history and previous CD4 count and viral load results should be urgently sought from the hospital where he was previously treated, together with details of any baseline or previous resistance test results. Adherence and use of possible interacting medicines and supplements should be carefully reviewed. Early repeat viral load testing should be undertaken.

Data from his previous hospital showed that the viral load had been fluctuating between 50 and 400 copies/ml for the last 6 months. No baseline or previous resistance test was available. The repeat viral load was 330 copies/ml. A switch in therapy was planned.

Should a resistance test be performed before switching therapy?

Conventionally, a viral load greater than 1000 copies/ml is required for successful resistance testing. Some laboratories may be able to provide resistance testing at lower viral load levels, by using larger volumes of blood, ultracentrifugation steps and nested PCR methodologies.

What combination regimen would be most likely to achieve virological suppression?

The patient is likely to have the M184V mutation conferring resistance to 3TC, and one or more NVP-associated mutations (e.g. Y181C). Thymidine analogue mutations (TAMs) selected by ZDV may also have begun to emerge. In the absence of resistance data, the preferred strategy is to change as many drugs as possible, with the selection of agents most likely to have activity against the mutations that are predicted to be present. In this case, the new regimen should be based on a ritonavir-boosted protease inhibitor (PI/r). The choice of nucleoside reverse transcriptase inhibitors (NRTIs) may include tenofovir (TDF), didanosine (ddI) and abacavir (ABC). Without resistance data, it is not possible to predict with certainty the degree of susceptibility to these drugs, or indeed whether there is preserved susceptibility to ZDV and stavudine (d4T).

Resistance testing was not attempted. Following a discussion with the patient, a new regimen was started with TDF (245 mg once daily), ddI (250 mg once daily) and ritonavir-boosted atazanavir (ATV/r, 300/100 mg once daily).

Is ATV/r a reasonable choice of PI/r in a previously PI-naive patient who may have accumulated some resistance to the NRTI backbone?

There is extensive clinical evidence supporting the use of ritonavir-boosted lopinavir (LPV/r) in both treatment-experienced and treatment-naive individuals. Data on the use of ATV/r in PI-naive patients are not currently available, but evidence indicates that the overall activity is good in patients with limited drug resistance [1]. Alternative options include ritonavir-boosted saquinavir (SQV/r) or fosamprenavir (FPV/r).

The viral load was 174 copies/ml 4 weeks after starting TDF/ddI/ATV/r, and 169 copies/ml after 6 weeks. The patient was not taking other medications or over-the-counter remedies and reported excellent adherence. The dose of ATV/r was increased to 400/100 mg once daily. After 10 weeks, the viral load remained detectable at 364 copies/ml. Therapeutic drug monitoring (TDM) showed ATV concentrations above those required to suppress wild-type virus.

How do you interpret these findings?

The viral load has not responded to an antiretroviral regimen that would be expected to produce complete suppression. Adherence was good and the patient was not taking drugs that can reduce ATV levels, such as acid-suppressive agents [2]. There is an interaction between TDF and ATV, reducing total ATV exposure by 25% [3], but this is not thought to be clinically significant when using ritonavir boosting. In this case, the ATV dose had been raised to ensure adequate levels, without apparent benefit on viral load suppression. Turning attention to the NRTI backbone, there is evidence that the combination of TDF and ddI is less effective than standard NRTI backbones, with early virological failure observed in drug-naive individuals given TDF/ddI/efavirenz [4,5]. Finally, the lack of response could reflect the presence of more substantial NRTI resistance than that predicted at the time of treatment change, or the presence of resistance mutations acquired at the time of infection and affecting susceptibility to the NRTIs or the PIs.

The patient switched therapy from ATV/r to LPV/r. After 4 weeks, the viral load was 3740 copies/ml.

What should be done now?

Adherence and drug interactions should again be discussed. The viral load is now high enough to perform a conventional resistance test and TDM can again be performed to measure LPV levels.

TDM showed LPV trough concentrations (Ctrough) 13 times the estimated Ctrough required to inhibit wild-type virus. A genotypic resistance test showed the following results:

  • Reverse transcriptase (RT ): L210F
  • Protease (PR): wild-type

How do you interpret these results?

L210F is an uncommon mutation and its significance is unknown. The absence of a baseline resistance test makes it difficult to determine whether the mutation emerged as a result of drug pressure, which would support a role in drug resistance. The mutation is not expected to impact significantly on susceptibility to TDF or ddI, but resistance effects on these or other NRTIs cannot be excluded.

The absence of PI-resistance mutations despite several weeks of detectable virus replication on a PI/r is consistent with data from clinical trials investigating the use of PI/r in drug-naive individuals. In these studies, no major protease-resistance mutations were detected in patients with treatment failure while receiving first-line therapy with two NRTIs in combination with LPV/r [610], FPV/r [11] or SQV/r [12]. It is currently debated whether the lack of major resistance mutations or measurable phenotypic resistance truly excludes the presence of other mechanisms of drug resistance.

TDF was replaced by ABC to overcome a possible weakness of the NRTI backbone. After 4 weeks, the viral load was 838 copies/ml. The patient remained on ABC/ddI/LPV/r and over the next 10 months his viral load continued to fluctuate between 500 and 1000 copies/ml, while the CD4 count remained between 200 and 300 cells/μl (15–17%). Two further genotype tests gave identical results to the first. After nearly 18 months, the viral load became persistently undetectable (<50 copies/ml).

What is the significance of these findings?

These findings suggest that the major problem had been one of suboptimal adherence. This phenomenon has been seen in studies of LPV/r where improved adherence resulted in virological suppression [13]. It should be noted that data on the use of ABC/ddI as NRTI backbone are currently lacking. The patient is being monitored closely.

References

1.
Johnson M, Grinsztejn B, Rodriguez C. et al. Atazanavir plus ritonavir or saquinavir, and lopinavir/rotonavir in patients experiencing multiple virological failures. AIDS. 2005;19:153–162. [PubMed: 15668540]
2.
Agarwala S, Gray K, Wang Y and Grasela D. Pharmacokinetic effects of omeprazole on atazanavir coadministration with ritonavir in healthy volunteers. 12th Conference on Retroviruses and Opportunistic Infections, Boston, 2005, Abstr. 658.
3.
Taburet AM, Piketty C, Grard L et al. Pharmacokinetic parameters of atazanavir/ritonavir when combined to tenofovir in HIV infected patients with multiple treatment failures: a sub-study of Puzzle2-ANRS 107 Trial. 10th Conference on Retroviruses and Opportunistic Infections, Boston, 2003, Abstr. 537.
4.
Podzamczer D, Ferrer E, Gatell JM et al. Early virologic failure and occurrence of resistance in naive patients receiving tenofovir, didanosine and efavirenz. XIII International HIV Drug Resistance Workshop, Tenerife, 2004, Abstr. 156.
5.
Moyle G, Maitland D, Hand J et al. Early virologic failure in persons with viral loads >100,000 copies/mL and CD4+ cell counts <200/mcL receiving didanosine/tenofovir/efavirenz as initial therapy: results from a randomized comparative trial. 44th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, DC, 2004, Abstr. H566.
6.
Gulick RM, da Silva B, McMillan F et al. Lopinavir/ritonavir (LPV/r)-based therapy in antiretroviral (arv)-naive, HIV-infected patients: 6-year follow-up of study 720. Seventh Congress on Drug Therapy in HIV Infection, Glasgow, 2004, Abstr. P28.
7.
Kempf DJ, King MS, Bernstein B. 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. [PubMed: 14702153]
8.
Cahn P, Renz C, Saez-Llorens X et al. Treatment of HIV-infected children with Kaletra (lopinavir/ritonavir) oral solution: 60 week follow-up. First International AIDS Society Conference on HIV Pathogenesis and Treatment, Buenos Aires, 2001, Abstr. 779.
9.
Feinberg J, Eron J, Bernstein B et al. Once-daily vs. twice-daily Kaletra (lopinavir/ritonavir) in antiretroviral-naive HIV+ patients: 72 week follow-up. XIV International AIDS Conference, Barcelona, 2002, Abstr. TuPeB4445.
10.
Molina JM, Wilkin A, Domingo P et al. Once-daily vs. twice-daily lopinavir/ritonavir in antiretroviral-naive patients: 96-week results. Third International AIDS Society Conference on HIV Pathogenesis and Treatment, Rio de Janeiro, 2005, Abstr. WePe12.3C12.
11.
MacManus S, Yates PJ, Elston RC. et al. GW433908/ritonavir once daily in antiretroviral therapy-naive HIV-infected patients: absence of protease resistance at 48 weeks. AIDS. 2004;18:651–655. [PubMed: 15090770]
12.
Ananworanich J, Ruxrungtham K, Sirivichayakul S et al. Resistance mutations in ART-naive patients treated with ritonavir-boosed saquinavir. Third International AIDS Society Conference on HIV Pathogenesis and Treatment, Rio de Janeiro, 2005, Abstr. WePe4.4C12.
13.
Campo RE, Lalanne R, Tanner TJ. et al. Lopinavir/ritonavir maintenance monotherapy after successful viral suppression with standard highly active antiretroviral therapy in HIV-1-infected patients. AIDS. 2005;19:447–452. [PubMed: 15750401]

Case study 7: A young black African woman presenting for HIV testing

Presented by Cecelia Theodore
Discussion by Anthony Newell and Anna Maria Geretti

Patient background

A 24-year-old heterosexual woman originally from Uganda presented for HIV antibody testing. She was diagnosed HIV-1 positive, with a baseline CD4 count of 38 cells/μl (4%) and a plasma HIV-1 RNA load of 103,865 copies/ml. Serology for hepatitis C and syphilis were both negative; however, she was found to be infected with hepatitis B virus (HBV) and was surface-antigen and e-antigen positive, with an HBV DNA level of 2.4 billion copies/ml. Liver function tests were normal. Abdominal ultrasonography demonstrated a homogenous echo pattern with no focal abnormalities. She reported that she had been in the UK for 9 months. She denied sexual contact with anyone known to be infected with HIV-1. Her husband had died of tuberculosis 10 months earlier in Uganda. She had no children.

Should a baseline resistance test be requested?

Current UK guidelines recommend that all patients should have a baseline resistance test performed at the time of diagnosis. However, there are limited data on the prevalence of resistance in drug-naive patients originally from Africa and presenting with chronic, established infection. In one study, the prevalence of transmitted resistance in the UK was found not to differ significantly in black Africans (17%) and heterosexuals (male, 15%; female, 17%) relative to white (15%) and homosexual (14%) people [1]. A second study found that, among patients newly diagnosed with an established HIV-1 infection, transmitted resistance was significantly more common in white homosexual males compared with a group of predominantly black African heterosexuals [2]. Current recommendations take into account increasing access to antiretroviral therapy in many African countries, which may lead to growing rates of transmitted resistance. Guidelines also warn of the possibility that apparent transmitted resistance may mask undisclosed antiretroviral exposure [3].

While waiting for the genotype resistance test results, it was decided to start the patient on a combination of zidovudine (ZDV, 250 mg twice daily), tenofovir (TDF, 245 mg once daily) and ritonavir-boosted lopinavir (LPV/r, three capsules twice daily).

Is the use of TDF without lamivudine (3TC) adequate treatment for a patient co-infected with HBV?

The use of 3TC as a single anti-HBV agent is effective in suppressing HBV DNA, but carries a significant risk of HBV resistance, with 20% of patients developing resistance for each year of treatment [4]. The use of TDF as single-agent therapy is considered less likely to lead to the rapid acquisition of HBV resistance in co-infected patients. This area is currently under investigation and studies are ongoing to determine whether co-infected patients are best managed by sequential HBV monotherapy or combination therapy.

The baseline genotypic resistance test showed the following results:

  • Reverse transcriptase (RT ): V75I/V, Y181C/Y, M184M/V
  • Protease (PR ): L33F, M36I
  • Subtype: complex pol gene recombinant

How do you interpret these findings?

The resistance test has detected mixtures of wild-type and drug-resistant viruses, with resistance mutations for the nucleoside and non-nucleoside reverse transcriptase inhibitors (NRTIs and NNRTIs, respectively) and the protease inhibitors (PIs). The RT mutation V75I is usually found in NRTI-experienced patients as part of the multi-NRTI-resistance complex that includes Q151M, F77L and F116Y. In isolation, V75I has uncertain effects on NRTI susceptibility. The M184V mutation is selected by 3TC, emtricitabine (FTC) or abacavir (ABC) and confers high-level resistance to 3TC and FTC and low-level resistance to ABC and didanosine (ddI). In isolation, M184V does not appear to reduce virological responses to ABC or ddI. The RT mutation Y181C is selected by nevirapine (NVP) and causes high-level resistance to NVP and low-level resistance to efavirenz (EFV). In the PR gene, the L33F mutation is usually observed in PI-experienced patients. It is selected by amprenavir, ritonavir and LPV/r, and contributes resistance to these PIs, as well as to atazanavir and tipranavir. The M36I mutation can occur as a polymorphism, especially in non-B HIV-1 subtypes, and in isolation does not confer significant PI resistance. The two PI mutations together are likely to result in very modest resistance effects for ritonavir-boosted PIs.

These results may be taken as evidence of transmitted resistance, with the mutants gradually reverting to minority populations as they are outgrown by wild-type virus. Alternatively, the mutations may represent resistance acquired during recent exposure to antiretroviral therapy.

On repeated questioning, the patient denied any previous exposure to HIV-1 therapies.

Does her initial antiretroviral therapy need to be amended?

The combination of ZDV/TDF/LPV/r would be expected to perform well against virus with this genotypic profile.

Is there a risk of HBV resistance to 3TC?

As there is a concern that this woman may have been exposed to 3TC, it may be prudent to determine whether HBV resistance to the drug has emerged.

HBV genotyping demonstrated a fully drug-sensitive virus. Therapy was switched from ZDV to Combivir (ZDV/3TC, one tablet twice daily) so that dual anti-HBV therapy could be administered. The HIV-1 viral load after 8 weeks was <50 copies/ml, whereas the CD4 count was 62 cells/μl (8%). The HBV DNA load was 77,000 copies/ml.

References

1.
Cane P, Chrystie I, Dunn D. et al. UK Group on Transmitted HIV Drug Resistance. Time trends in primary resistance to HIV drugs in the United Kingdom: multicentre observational study. BMJ. 2005;331:1368. [PMC free article: PMC1309643] [PubMed: 16299012]
2.
Geretti AM, Booth CL, Labbett W. et al. Risk group predicts the prevalence of primary resistance amongst newly diagnosed HIV-infected patients presenting with established infection according to the STARHS algorithm. Antivir Ther. 2005;10:S131.
3.
Natha M, Newell A, Pakianathan M. Non-disclosure of previously known HIV seropositivity in patients `newly' diagnosed with HIV infection. Sex Transm Infect. 2005;81:182–183. [PMC free article: PMC1764674] [PubMed: 15800104]
4.
Benhamou Y, Bochet M, Thibault V. et al. Long-term incidence of hepatitis B virus resistance to lamivudine in human immunodeficiency virus-infected patients. Hepatology. 1999;30:1302–1306. [PubMed: 10534354]
Copyright © 2006, Mediscript.
Bookshelf ID: NBK2248