-
2nd International Workshop on Drug Resistance & Treatment Strategies 24 - 27 June 1998, Lake Maggiore, Italy
-
Antiretroviral therapy failure may be related to target cell availability - a further role for hydroxyurea? \
-
A further role for hydroxyurea?
-
Newer drugs and combinations with salvage potential
-
Resistance to newer antiretrovirals
-
Relationship Between Genotypic Resistance, Phenotypic Resistance, and Clinical Response
-
Does a Protease Inhibitor have additional effects, even with VL <25 copies/mL?
-
Proactive switching of Antiretroviral combinations - a computer simulation
|
2 nd International Workshop on Drug Resistance |
Lake Maggiore in the week preceding the 12 th World AIDS Conference in Geneva was the setting for the annual HIV Drug Resi stance and Treatment Strategies workshop. Leading clinicians, virologists, researchers, and a handful of treatment activists met to discuss and exchange new data and current concepts in HIV therapy. An expanded range of retrovirals and wider access to resistance testing technology have extended the opportunity to understand the mechanisms of failure of antiretroviral therapy and to plan salvage strategies.
AIDS Treatment Projects Raffi Babakhanian and Doctor Fax Medical advisor, Dr Graeme Moyle were both delegates at this workshop. The reports below represent a selection of some of the most immediately relevant presentations for clinical management of HIV-infected patients.
The next issue of ATP's Doctor Fax is scheduled for distribution next Friday and will consist of reports from the 12 th World AIDS Conference in Geneva which took place from June 28 - July3, 1998.
Author: Paul Blanchard.
|
Antiviral Response, therapy failure and Salvage
|
Antiretroviral therapy failure may be related to target cell availability
The opening plenary session of the meeting considered the theoretical role of target cell (primarily CD4+ lymphocytes) availability in anti-HIV treatment failure. Angela McLean of the Institute for Animal Health in the U.K. explored the hypothesis that dynamic changes in both CD4+ cell populations and viral load under the influence of antiretroviral therapy may interact to influence the probability of the emergence of drug resistant virus
Although viral loads fall during treatment, target cells for possible viral replication increase. The chance of generating a drug resistant mutant also decreases with decreasing viral load. The competition for target cells amongst HIV quasi-species (including drug-resistant), however, decreases as CD4+ lymphocyte counts rise. The model for generation and extinction of strains of HIV with different drug sensitivities developed by Dr McLean makes three predictions which could have a major impact on treatment strategies.
-
Inadequate virological suppression in the absence of resistance mutations
Treatment may fail to control viral replication and HIV re-emerge even in the absence of drug resistance mutations. In the maintenance therapy study ACTG343 patients randomised to stop ZDV/3TC and continue on IDV monotherapy experienced rebound of virus (n=9) which was wild type. Patient history did not suggest poor adherence [abs 74]. Similar data were observed in the SPICE study, with persons with detectable virus at week 16 commonly showing no typical PI resistance associated mutations [abs 88]. This rebound in the absence of resistance may not simply relate to insufficient potency of the therapy but might reflect incomplete prevention of release of infectious virus from all productively infected cell lines in the presence of an expanded population of new target cells. Interestingly, both the ACTG 343 study mentioned above [abs 74], and a study of viral breakthrough in naïve patients receiving triple or quadruple therapy including 3TC [abs 83], found that isolates from patients with a rebound in viral load invariably displayed the M184V mutation associated with resistance to 3TC. Thus, early failure to maintain viral suppression may be due to the failure of a single component of the combination regimen (most often, and most readily 3TC). This can occur without mutations in the protease region associated with resistance.
Could it be that compartmentation of both cell types and anatomical locations are occurring in terms of exposure of HIV to the components of a combination? If 3TC is the only drug active in certain cell types or locations, then these reservoirs of HIV are effectively receiving monotherapy with that compound leading to rapid resistance. M184V mutated virus would then rapidly seed the increasing numbers of target cells leading to a rebound in plasma HIV viral load. On analysis, differences were found in the RT gene [abs 128] and protease gene [abs 112] between blood and seminal fluid of HIV-infected individuals receiving antiretrovirals, indicating compartmental variance in resistance. This supports the hypothesis that the replication and evolution of distinct populations of virus can occur in an individual. Similar findings have previously been reported for CSF compared to blood and in untreated as well as treated patients.
Patients failed on multiple therapies, however, frequently have mutations consistent with resistance to both RTI's and PI's [abs 85]. It has now been shown that such multidrug resistant virus is fully capable of being transmitted. Indeed, a San Francisco based study identified one individual recently infected with HIV harbouring multiple genetic markers of drug resistance (RT: D67N, M184V, T215Y, K219Q; PR: M36I, M46I, L63P, A71T, N88D, V82I, L90M). Unsurprisingly, response to triple therapy (ZDV, 3TC, IND) was blunted.
Transmission of resistant virus now appears to be a relatively common event. In the Swiss cohort of 67 primary infection patients collected in Jan 1996-Dec 1997, ZDV resistance was present in 7.5%, and major PI mutations (82, 90 +/- other compensatory polymorphisms) i n 5%. Resistance mutations characteristic of other NAs were infrequent [abs 107]. Similar observations were reported in therapy naïve patients presenting for care in Belgium [abs 113], Iowa [abs 114], Spain [abs 118] and the UK [abs 117]. Both European groups and US reported individuals seroconverting with HIV resistant to PIs, NAs SYMBOL 177 "Symbol" 10 NNRTIs [abs 107, 108] and with the Q151M mutinucleoside resistance pattern [abs 123].
-
First few weeks of treatment initiation may be crucial to success
The first few weeks of antiviral therapy, when both viral load and target cell numbers are high, may play a crucial role in the generation of partially resistant mutants. If clinically relevant, this observation would produce some testable hypotheses. Firstly, rate of viral decay during the first few days after treatment initiation could be expected to correlate to subsequent likelihood of virological failure. The more rapid the decay (the steeper the initial decline, and shorter the time to below detection), the less likely or less rapidly would drug resistance be expected to evolve. Secondly, those who initiate antiretrovirals with lower initial CD4+ lymphocyte counts may be less likely to generate resistant HIV mutants. However, this theoretical advantage may well be offset by a greater number of naturally occurring viral quasispecies (due to period and extent of prior HIV replication needed to reach lower CD4 counts) and greater likelihood of intolerability of medication in more advanced populations. Thirdly, artificially lowering CD4+ lym phocyte counts at the initiation of antiretroviral therapy (by cytostatics, whole body irradiation or anti-CD4 monoclonal antibodies for example) may reduce the risk of generation of anti-HIV drug resistant mutants.
-
Rapid, large increases in CD4 cell counts in response to combination therapy may not be all good
Large increases in target cell numbers as a consequence of immune reconstitution, while reducing risk of opportunistic infections, also increases the probability of treatment failure. As Dr McClean explains "Since HIV is an obligate parasite, it has no intrinsic growth rate. The viral reproduction rate is strongly dependent on the number of target cells available in which it can reproduce and therefore increases as a result of immune reconstitution. " This hypothesis was supported by the observation, again in study ACTG 343, that virological failure was, indeed, more likely in those patients experiencing a large CD4 cell rise [abs 74]. Improvement in HIV specific CTL in concert with rises in CD4+ lymphocyte populations may well offset this effect, but has yet to be reported with PI-containing regimens in chronic infection (although this was reported in hydroxyurea containing regimens - see below). Specific induction of anti-HIV CTL responses may be needed to give additional suppression of HIV in the face of increasing target cell availability. A late breaker presentation at Geneva reported the preliminary efficacy of immune stimulation with a killed gp-120 depleted whole HIV preparation (Remune) as immunotherapy in patients receiving combination therapy [Geneva abs.31227]. This study contained a control arm of adjuvant only, and at week 20 the Remune receiving arm demonstrated strong lymphocyte proliferative responses to a variety of HIV antigens which were not seen in the control arm (more details in the Geneva reports starting in the next issue of ATP's DocFax).
A further role for hydroxyurea?
Hydroxyurea (HU) represents a novel approach to the suppression of HIV replication, which is now being consi dered with increasing interest particularly in salvage or eradication strategies. The mechanism of action of is thought to be cellular. Hydroxyurea lowers the levels of the cellular competitor of ddI (dATP) thus favouring the incorporation of ddI. Although the combination of ddI and hydroxyurea is unable to prevent the emergence of mutants conferring resistance to ddI the resistant virus remains sensitive to ddI in the presence of hydroxyurea. It may also act favourably on the intracellular kinetics of other nucleoside analogues besides ddI, and may also have a role in the reversal of impaired phosphorylation, which may be a result of long-term NA use. ATP s DocFax Issues 16, 30 32, 33, 41, 42 & 46 contain previous reports on hydroxyurea. Since cellular proteins are less prone to mutations than viral ones, resistance to HU is not expected.
Palmer, of Stanford University Medical Centre presented further in vitro data showing that hydroxyurea at clinically achievable and tolerated concentrations potentiates the activity of didanosine and the novel nucleotide analogues adefovir dipivoxil (Preveon TM ) and PMPA. Further, hydroxyurea also increased the activity of these drugs against clinical isolates of HIV displaying resistance to these compounds as well as a multidrug resistant and a multinucleoside resistant isolate [abs 3]
Blunting of CD4 cell response due to HU related lymphotoxicity/antiproliferative effects may theoretically improve the chance of durable response to therapy by limiting new target cell availability (see article above). However, underlying populations of naï ve cells and reduction in CD8 activation appears similar to patients on other HAART regimens. HU may also reduce the size if the pool of HIV-infected latent CD4 cells. Patients treated with long-term HU-ddI therapy were also found to have improvement in HIV-specific CTL. Improvement in HIV specific CTL has not been reported with PI-containing regimens. These data may provide a possible explanation for failure to observe viral rebound in a small number of patients stopping HU-containing regimens [abs 149].
Author: Paul Blanchard.
Newer drugs and combinations with salvage potential
Adefovir
Several nucleotides are in clinical development including adefovir and the more clinically potent PMPA. Activity of adefovir in persons mostly with a mean 3 years prior NA therapy was examined in 442 patients with 150 in virology subset. Of these, 142 had matched baseline and WK 24 samples. 76% had 184V and 53% a mean of 3 ZDV related mutations at baseline. WK24 change in VL was 0.53 log. Results were similar in persons with M184V or dual ZDV/3TC resistance associated mutations in background. Mutations associated with adefovir in vitro (K70E, K65R) were not observed to arise by wk 24 (abs 4).
In vitro testing of two-drug combinations of adefovir with either ddI, foscarnet, nevirapine or efavirenz was performed against both wild-type HIV, a multidrug-resistant isolate (MDR) containing six RT mutations (41L, 67N, 184V, 210W, 215Y, 219N) and six PI mutations (10I, 48V, 54T 63Q, 71V, 82A), and a second multinucleoside-resistant (MNR) isolate containing four RT inhibitor mutations (75I, 77L, 116Y, 151M). IC 50 values for adefovir dipivoxil were 3.3, 5.6 and 9.7 SYMBOL 109 "Symbol" 10 m, respectively for the wild-type, MDR and MNR isolates. The 2-drug combinations showed the following:
| Wild-type | MDR | MNR | |
| ADV + ddI | Add | Add | Varied |
| ADV + foscarnet | Syn | Syn | Varied |
| ADV + NVP | Syn | Syn | Syn |
| ADV + EFV | Add* | Syn* | Syn |
Add = additive, Syn = synergistic, * = at higher concentrations.
These in vitro studies have led the group from Stanford University Medical centre to move forward with a study of adefovir/efavirenz + hydroxyurea in patients with resistant viral strains [abs 12]. Interestingly patients with 184V may also show increased susceptibility in the range of 3-4 -fold to the nucleotides PMPA and adefovir [abs 34].
Nelfinavir, saquinavir-SGC, abacavir with and without nevirapine in patients with failure of a prior indinavir combination
Nelfinavir 1250mg bd (NFV), saquinavir-SGC 1200mg bd (SQV-sgc), abacavir 300mg bd (ABC) with or without nevirapine 200mg bd (NVP) was administered to 20 patients with evidence of failure of a prior indinavir (IND)-containing regimen (VL >2500 copies/mL after >24 weeks of IND therapy)
Median baseline CD4 count and VL were similar between the two groups at 290 cells/mL and 4.4log copies/mL respectively. Genotypic evidence for indinavir resistance was present at baseline in 17/19 evaluable samples. Four patients discontinued therapy due to intolerance.
At 16 weeks of follow up the median change in viral load was -1.2log (range -0.14 to -2.05) in the 3-drug group and -2.33log (range -.01 to -3.21) in the 4-drug group. Using an ultrasensitive assay (50 copies/mL), 0/7 and 4/9 patients in each group had an undetectable VL at week 16.
Author: Paul Blanchard.
Resistance to newer antiretrovirals
By now, it's well known that cross-resistance within the three antiretroviral drug classes limits the number of drug combinations available to patients, despite the enormous number of potential drug combinations available. Unfortunately, cross-resistance extends to most of the newer investigational agents closest to approval, making them unlikely candidates for effective salvage therapy.
Adefovir Resistance
Adefovir dipivoxil (Preveon) is the first nucleotide reverse transcriptase inhibitors. As such, it might have been assumed that cross-resistance to available drugs wouldn t have been an important issue. That has not turned out to be the case, however. Adefovir is active against wild type virus and virus with 3TC resistance or dual ZDV/3TC resistance [abs 4]. However, adefovir is not active against virus with high-level ZDV resistance. Interestingly, 3TC-resistant virus (virus with the M184V mutation) is more susceptible to adefovir than 3TC-sensitive virus, and the 184 mutation appeared to increase susceptibility to adefovir of virus with high-level ZDV resistance. Antiviral efficacy to adefovir based on genotypic resistance patterns is summarised below:
| Genotypic profile | Log RNA change |
| no mutation | -0.65 |
| low-level ZDV resistance | -0.65 |
| high-level ZDV resistance | -0.05 |
| M184V | -0.94 |
| M184V + low-level ZDV resistance | -0.75 |
| M184V + high-level ZDV resistance | -0.51 > |
Abacavir Resistance
The guanasine analogue abacavir has demonstrated potent clinical activity. Mutations selected in vitro include at codons 65, 74, 115 and 184, single mutations alone causing 2-3 fold changes in sensitivity to abacavir. Dual ZDV/3TC resistance was associated with 70% >8fold and 30% >4-fold resistant to abacavir. Most mutants with 3TC or ZDV resistance alone were phenoypically sensitive to abacavir (<4-fold changes) in vitro. The Q151M resistance complex showed >8-fold resistance to abacavir. In vivo patients with only 74V (5 isolates) or 184V (>40 isolates) responded similarly to wild type (>40 isolates) to abacavir. Presence of multiple ( > 3) mutations to ZDV + 184V had attenuated responses to abacavir at week 12-16. Phenotypic resistance (> 8-fold) to abacavir was associated with poor antiviral responses [abs 52]. A second study reported 11/17 patients with M184V at baseline on ZDV/3TC achieved a VL reduction of >0.5log with the addition of abacavir [abs 100] . In patients failed on multiple prior therapies and CD4 <100, VL >30K, use of abacavir + at least one new agent was associated with no change in mean VL, however 23% of patients had >0.5log VL reduction. No association with number of RT mutations and response was observed in this study. [abs 101] . Additionally, some patients with multiple mutations selected for by abacavir monotherapy may achieve undetectable VL with the addition of ZDV/3TC over 16 weeks treatment [abs 99].
Abacavir resistance is seen in virus with multi-drug resistance due to the 151 complex or due to the 69 S-S-S mutation [abs 15] along with ZDV resistance mutations [abs 52].
Protease Inhibitor Resistance
Kinetic data on PI binding to HIV protease may represent a better guide to resistance and cross-resistance than phenotype as patients may experience viral rebound on PI therapy with a single mutation but < 4-fold phenotype c hanges. With saquinavir mutations Ki values are 13.5, 3 and 419 fold higher with 48V, L90M and dual mutants respectively. (abst 38). Gains in amprenavir binding and reductions in saquinavir binding also appear to explain the in vitro observation of re-sensitisation to amprenavir when amprenavir resistant virus (mutant at codons 46, 47 and 50) is passaged in the presence of saquinavir (abs24). The 82T mutant increased Ki to RTV 17 fold, with a triple mutant (at codons 36+54+82) increased Ki 411-fold. Additional mutations provided no further increase in Ki but lead to substantially improved viral catalytic capacity, often to above wild type [abs 110] . Additionally, PI resistance conferring mutations negatively impact viral fitness. The D30N mutation selected by NFV reduced fitness 29% relative to wild type, the SQV L90M mutation reducing fitness 11% relative to wild type. The common polymorphism L63P was noted to partially compensate for this loss of fitness, similar to that reported with the IDV/RTV key codon 82 mutations [abs 130] . Patients experiencing early virologic rebound during amprenavir combination with ZDV/3TC (n=7) showed no mutations in some cases (3/7), 3TC resistance alone (3/7) and both NA and PI mutations (1/7, mutants in protease at codons 5 0, 54, 84 each as minorities). In amprenavir monotherapy (n=19) patients with rebounding viral load, 4/19 had the I50V mutation, as well as 16/19 having other mutations typical of PIs (including at codons 10, 20, 46, 82 and 84) [abs 71 & 86].
Whilst loss of suppression with triple therapy may occur with 3TC resistance, presence of ZDV resistance at baseline did not prevent durable viral suppression with ZDV/3TC/RTV [abs 92].
PI combination therapy is an increasingly used strategy. Preliminary data from a practice-based study suggest RTV-SQV+2NA may result in more patients achieving <400 copies/ml than triple therapy with either IDV or RTV [abs 72]. In vitro models of PI combinations containing RTV suggest both components of the combination of RTV/SQV contribute equally to activity of this combination at the 400/400mg bid dosing level. In comparison with 400/400mg bid of RTV/IDV the model suggested that IDV activity would drive the potency of this combination [abs 74].
In maintenance therapy study ACTG 343 patients randomised to stop ZDV/3TC and continue on IDV monotherapy, patients experienced rebound of virus (n=9) which was wild type. Patient history did not suggest poor adherence [abs 74]. CD4 remains important to risk of clinical events, with elevation of CD4 persisting over >1 year in persons with detectable virus on therapy having a similar risk of events to a matched population persons with <400 viral loads [abs 75].
Amprenavir (AMP) resistance data from ACTG 347 [abs 71] was presented. These PI and 3TC naï ve patients received either ZDV/3TC/AMP or AMP monotherapy. 4/19 AMP monotherapy recipients had the I50V mutation, while 16/19 had mutations associated with cross-resistance to other protease inhibitors at L10, M46, K20, V82, and I84. Four of eight monotherapy failures PBMC isolates show cross resistance to other PI's consistent with genotypes. As with other triple therapy failure including a PI, triple therapy recipient s who experienced a rebound did not often display AMP resistance, but more likely had 3TC resistance. This is an early event, and AMP resistance would be expected to follow. These data do not support the earlier contention that AMP might not display cross-resistance with all other available protease inhibitors of this class due to a unique resistance pattern.
NNRTI resistance
With increasing concern about long-term toxicity with some PIs there is renewed interest in NNRTIs.
With efavirenz patients failing on monotherapy or during combination with IDV, K103N was observed in 31/33 treatment failures. Similarly this mutation was found in 25 of 30 patients failing after adding EFV to ZDV/3TC. Additional observed mutations included V108I, P225H and occasional L100I or at codons 190 or 188. A novel mutation, possibly compensatory was also observed at codon 135 associated with K103N. Additionally some mutations such as at 108 or 225 substantially reduced in vitro sensitivity to EFV in the presence of K103N [abstract 19].
Mutations or clusters of mutations were noted to be associated with response to combination therapy in ACTG241 (ZDV/ddI/Nev in ZDV treated). Wild-type to ZDV, particularly at codon 215 responded best to therapy than those with ZDV mutations, regardless trial arm (adding ddI, Nev or both) [abs 50].
D4T and d4T/ddI resistance more clearly defined
Mechanisms of stavudine failure have been poorly understood, in part due to a paucity of data. Analysis of virus with phenotypic resistance to stavudine by Antivirogram derived from multiple NA treated patients. In 50 samples with >4-fold changes, 6 had Q151M (multi-NA resistant pattern), V75T in 2 (previously reported in vitro ) and other novel codon 75 mutations in a further 9 samples. The most common pattern was ZDV mutations at codons 41/67/210/215 plus 69D or N (previously reported with ddC monotherapy) SYMBOL 177 "Symbol" 10 184V (n=23) or a novel codon 69S mutation with a double insertion SYMBOL 177 "Symbol" 10 ZDV mutations or 184V (n=10). [abs 15]. This novel insertion was also reported in 3 of 5 isolates analysed by another group [abst7]. The increasingly used combination of d4T/ddI in vitro selects for novel mutations at codons 57 and 98 associated with an 11.5 fold resistance to ddI and 4.5 fold change to d4T. [abs 31] . Some physicians have suggested this combination may best keep future NA options open. However, one in vivo study from Frankfurt found resistance to ddI or d4T was associated commonly with the Q151M multi-NA resistant virus- 9/10 demonstrating >10-fold change in ddI sensitivity and 12/15 with >10-fold d4T resistance. [abst 97] . A US group, however, observed no consistent pattern of mutations although the Q151M mutation was observed in several isolates and a mutation at codon 333, previously associated with dual ZDV/3TC resistance was also seen in one sample. [abstract 96] . Clearly, more data are required to answer this question; however, these represent the first data to raise resistance concerns with this combination.
Increasing numbers of ZDV associated mutations may be associated with a poorer response to subsequent d4T based therapy [abs 94] although changes in d4T phosphorylation have been previously suggested to influence response to second line d4T.
Author: Paul Blanchard.
Relationship between Genotypic Resistance, Phenotypic Resistance, and Clinical Response
There is still a great deal of scepticism about the current clinical utility of phenotypic and genotypic resistance assays. Most experts would agree, however, that as these assays improve they will become an increasingly important clinical tool Rapid resistance assays will be commercially available soon, and there are now promising data emerging from genotype/phenotype relational databases.
Baseline resistance patterns have been shown to predict response to abacavir, ritonavir/saquinavir, nelfinavir, and various 4-drug combinations. In a ritonavir/saquinavir trial involving 84 patients, a retrospective analysis found that baseline phenotypic and genotypic resistance patterns were independent predictors of drug failure. Furthermore, genotypic findings usually correlate well with phenotypic resistance or sensitivity. However, findings of resistance using these studies are much more predictive of response than findings of drug sensitivity, perhaps due to the relative insensitivity of these assays to minority strains. In other words, the tests are much better at giving bad news than good news, and in many cases the bad news can be inferred from the drug history. Retrospective correlation with drug failure is strong, but prospective studies are needed. Despite their limitations, Dr. Mellors concluded that the presentations from Lago Maggiore contributed a great deal toward the validation of these controversial assays [Geneva oral session 227].
Zolopa presented data from a retrospective cohort study from Stanford looking at genotypic predictors of response to ritonavir/saquinavir salvage therapy [Geneva abs 32287]. The study involved 54 patients failing a protease inhibitor-containing regimen. All were switched to RTV/SQV plus nucleoside analogues (NRTIs), although only 23% received new NRTIs at the time of switch. Complete response was defined as HIV RNA < 500 copies/mL; partial response was defined as detectable virus with >0.5 log decrease; and non-response was defined as a change in RNA <0.5 log. In a multivariate analysis, genotypic data was found to add predictive value to information that could be obtained from a clinical and drug history:
| Multivariate Prediction Models | Explained variance (R2) |
| Clinical/Drug History | 0.45 |
| Protease Mutations | 0.59 |
| Protease + RT Mutations | 0.68 |
| Comprehensive (history and mutations) | 0.72 |
Thus, the HIV genotype provides information that helps to determine the likelihood of response to salvage therapy. In fact, genotypic data has stronger predictive value, which is independent of information that can be derived from the clinical and drug history. However, predicting failure is a far cry from knowing how to treat it.
Source: www.healthcg.com
Author: Joel Gallant, M.D.
Does a Protease Inhibitor have additional effects, even with VL <25 copies/mL?
Two presentations suggest that double nucleoside analogue therapy, even when it achieves viral suppression to <25 copies/mL, does not prevent resistance or allow viral clearance as efficiently as a protease inhibitor containing combination [abs 146 & 152].
Clearance of lymph node HIV RNA and maintenance of normal germinal centre morphology was achieved more reliably by a protease inhibitor containing regimen compared to a NA regimen although both groups had long term suppression (<25 copies/ml) of plasma HIV RNA (abs 146). Additionally, germinal centre recovery was noted as early as one month into PI based HAART therapy. However, cell populations in germinal centres at this early time point appeared to differ from health controls, raising concern that improvements in LN gross architecture may not fully reflect restoration of germinal centre function [abs 141] . Despite maintenance of plasma VL <20 copies/ml, genetic evolution or appearance of resistance conferring mutations may appear, including in patients on PI-containing regimens [abs 146, 147, 152] . These data suggest that patients achieving the best measurable virologic responses to therapy may eventually fail therapy due to persistent low level viral replication. Additionally, maintenance of latently infected cells with recoverable/viable virus does not appear to require on-going HIV replication, even in patients receiving PI's [abs 148].
Author: Paul Blanchard.
Proactive switching of Antiretroviral combinations - a computer simulation
In an attempt to forestall the development of resistance, it would seem logical that switching therapies before viral rebound might be a proactive method to delay genotypic change associated with resistance. Such an approach would lead to cycling of HAART regimens according to a predetermined time plan, rather than as a response to virological failure.
Computer modelling of viral load monitoring and switching strategies determined that a proactive policy, which periodically changes combination regimens prior to detecting evidence of drug failure, may significantly reduce the duration of viral replication under selective pressure of a failing regimen. Clinical trials would be needed to establish the validity of this hypothesis.
Author: Paul Blanchard.