icon-folder.gif   Conference Reports for NATAP  
  XVII International HIV Drug Resistance Workshop
June 10-14, 2008
Sitges, Spain
Back grey_arrow_rt.gif
Report on the XV11 International Drug Resistance Workshop, June 10-14, 2008, Sitges, Spain
  Mark A Wainberg
McGill University AIDS Centre
Montreal, Canada
1. Integrase Inhibitors (resistance)
2. New drugs- NNRTI inhibitors: IDX899; Bevirimat:
HIV maturation inhibitor
3. etravirine resistance: pre-existing nucleoside resistance-associated mutations were able to hypersensitize to Etravirine; clinical cutoffs
4. How extensive is NNRTI mutation K103N in Africa:
Abstract 120 presented data from South Africa in which it was reported that approximately 10% of women who had been involved in studies for prevention of mother-to-child transmission of HIV-1 had developed K103N mutations, irrespective of any prior exposure to single dose nevirapine.
5. intensification with efavirenz or kaletrs.
6. low-level mutations associated with drug resistance may, in fact, have the potential to compromise subsequent clinical responsiveness
7. Mutation/Resistance patterns in Africa: certain mutations may be more predominant in certain viral subtypes......Abstract 51 demonstrated that subtype C viruses may be more prone toward the development of the K65R substitution than are viruses of other subtypes. In Abstract 108, use of ultrasensitive testing revealed the presence of a number of minor mutations that were present as sub-populations (in South Africa in patients failing first and second-line regimens). One of the most important findings presented was that 12 of 14 viral samples from patients with subtype C viruses contained the K65R mutation at levels between 0.5% to 4.4% of the total population.....Were K65R to occur at higher frequency in countries in which subtype C viruses are endemic and to be easily transmitted, there could be potential complications for therapeutic interventions in such settings
The scientific program of this conference was interesting and contained new information on HIV drug resistance in a number of areas. These are summarized as follows:
1. Integrase Inhibitors
This was a topic of considerable attention. One of the major reasons for this is that we now have an approved anti-Integrase compound for treatment of HIV disease, ie Raltegravir. This drug specifically blocks the strand transfer event that is associated with the HIV Integrase enzyme that also possesses other enzymatic activities that are essential for viral replication. This means that it may also someday be possible to develop drugs that target those other enzymatic steps as well.
In view of the fact that Raltegravir is the most recently approved drug in the anti-HIV armamentarium, it is not surprising that our understanding of resistance mutations in regard to the use of this compound and other inhibitors of HIV integrase is still evolving. It is significant that a number of abstracts provided mutually reinforcing information in regard to this topic.
For example, Abstract 6 provided a more detailed characterization of mutations associated with resistance against Raltagravir than what has previously been known. Notably, work presented at previous conferences has described that two major pathways are involved in raltegravir resistance that initially include mutations within the Integrase gene at either position N155H or a series of three different mutations at position Q148, i.e. Q148R/H/K, meaning that the wild-type amino acid Q normally seen at this position can mutate to become either R,H, or K. In this abstract, the authors presented a longitudinal analysis from the Merck phase 2 clinical trial 005, in which they monitored resistance over several months in treatment-experienced patients who had previously developed resistance against other classes of anti-retroviral drugs. The results showed that there was a tendency over time for Q148H to become the pre-dominant mutation in patients whose viruses became resistant to raltegravir, i.e., Q148H was present in approximately 75% of total cases. This result is important because it also revealed a tendency for the other signature mutations associated with raltegravir resistance, ie N155H, Q148R, Q148K, to be supplanted over time. The implication, still to be confirmed by additional studies, is that Q148H, in combination with other secondary resistance-associated mutations, may cause HIV to have greater replicative fitness than other combinations of resistance-conferring mutations and pathways.
What might underlie this tendency for raltegravir-resistant viruses to evove preferentially along a Q148H pathway? Several alternative hypotheses are that Q148H mutations might develop independently and continuously over time and become predominant, as discussed above, for reasons that primarily relate to the increased replicative fitness of viruses possessing this mutation in comparison to the others that are associated with resistance against integrase inhibitors. This would be similar to the emergence over time of the M184V mutation in reverse transcriptase in place of M184I. Even though both these mutations are selected by the NRTIs 3TC and FTC, and both confer high level resistance to these drugs, it is M184V that eventually becomes predominant because it is associated with higher viral replicative fitness than M184I.
Alternatively, it is possible that the other mutations at position Q148, including R and K, might themselves now mutate so as to change to Q148H. This topic will now demand considerable attention if we are to better understand the origins of drug resistance against Raltagravir, Elvitegravir, and other Integrase strand transfer inhibitors that will be developed in the future.
In Abstract 7, S. Fransen et al analyzed patients who had failed therapy in the Merck-sponsored phase 3 Benchmrk studies and also reported a significant presence of primary mutations at position 143 in the integrase gene. It is conceivable that these mutations may be relatively late to arise in the context of resistance against Raltagravir; this remains to be confirmed but is also suggested by a study from France (see below). In many of the samples studied, the E92Q mutation was also found and was frequently on the same clone in the integrase(IN) gene as N155H but not in the same clone as the mutations that were identified at position 148. In some instances, mixtures of mutations were found, suggesting the possibility that resistance might have developed simultaneously along different pathways.. Consistent with data from Abstract 6, it appeared that the Q148H mutation may have become predominant, again, as hypothesized above, for reasons of increased viral fitness. Results consistent with those reported above were also obtained in data presented in Abstracts 9 and 10, in which a relative fitness disadvantage of N155H- containing viruses relative to those containing the Q148 series of mutations was reported. In Abstract 12, substitutions were reported to occur at position Y143, involving possible substitutions involving three different amino acid substitutions, i.e. C/H/R, in the context of compensatory mutations that might have contributed to an overall increase in viral replication capacity.
Novel information was presented in Abstract 13 to confirm work presented at earlier conferences that the E92Q substitution seems to be more closely associated with resistance against Elvitagravir than raltegravir in patients enrolled in a Gilead-sponsored phase 2 trial and that this substitution was often associated with a compensatory mutation at position L68V that resulted in an increase in viral replicative capacity. It is also interesting that L68V was never seen in association with either N155H or Q148H/R/K. This may mean that secondary mutations associated with resistance to elvitegravir may turn out to be different than those found in patients who develop resistance against raltegravir. Unfortunately, it nonetheless appears that complete cross-resistance is likely to occur between both of these two INSTI compounds and that sequencing of these two drugs in therapy will not be possible. This underlines the need for continuing drug development to continue in this area.
Of all the above findings, that which is likely to have the greatest clinical significance in the short term is the apparent preponderance of Q148H as a signature mutation that assumes increasing ascendency in longitudinal studies. Clearly, further clinical studies in association with basic science research will be necessary to shed further light on these findings. It is also interesting that Abstract 17 contains important new information on how certain of the mutations associated with resistance against INSTIs also compromised the biochemical efficiency associated with various steps of the integrase-mediated reactions. These results help to explain how mutations associated with resistance against INSTIs may also result in reduced viral replicative capacity, since impairment of viral enzyme function should logically be expected to compromise the ability of HIV to replicate. This situation is analogous, for example, to the diminished fitness conferred to HIV by the M184V mutation in reverse transcriptase, associated with high-level resistance to 3TC and FTC.
The subject of clinical cut-offs for raltegravir and other members of the integrase family of compounds is important. However, it is probably too early to state with definitiveness what the answer to this issue will be in terms of fold-resistance. What is clear is that the presence of only a single signature mutation is unable to confer high-level resistance and that an accumulation of secondary mutations (at least one or two) is needed in the presence of a primary mutation for high-level resistance to be present. In general, it is clear that raltegravir is not a low genetic barrier drug for resistance such as some of the NNRTIs. Neither does it have the very high genetic barrier of some of the boosted PIs. Precise characterization of the character of raltegravir as possessing an intermediate genetic barrier for development of resistance will require further study.
2. New drugs
a. NNRTI inhibitors. Many observers have articulated the clear need to develop novel NNRTI compounds that will not be fully compromised by the mutations associated with resistance against either Nevirapine or Efavirenz. Such a molecule could help to provide patients who possess such mutations as K103N or Y181C with renewed hope for the future, in terms of ability to benefit from the NNRTI family of compounds. One such drug, that was recently granted approval by the FDA, is Etravirine. The development of a second such compound has now been reported in Abstract 5, in which scientists from Idenix Pharmaceuticals, in collaboration with clinical investigators, documented that a novel compound termed IDX899, with potential to be used on a once-daily basis, resulted in significant reductions in viral load when administered as the equivalent of monotherapy, i.e, as an add-on drug given in addition to an ongoing background regimen, in treatment experienced patients. Tissue culture analysis revealed that IDX899 required a large number of passages and accumulation of multiple mutations in order for low level resistance to develop, and that such resistance, when present, was responsible for lower overall levels of cross-resistance with other NNRTIs than either Efavirenz or TMC125 (Etraverine). The phase 2 clinical study, reported in this abstract, showed that median changes in plasma viral RNA from baseline over 8 days were -1.95 log copies of viral RNA. These results are accompanied by data showing that the selection of mutations at positions E138K and Y181C had occurred and that these may represent independent pathways for selection of resistance against IDX899. However, an accumulation of multiple other mutations seems to also be required in order for resistance against this compound to develop. The fact that IDX899 apparently has a very high genetic barrier for development of resistance, coupled with its potential to be dosed once-daily, establishes strong rationale for its continuing development and potential for co-formulation with other drugs, eg, nucleoside RTinhibitors, toward the efficient inhibition of HIV replication. b. Bevirimat. This compound has been reported on at many previous conferences, and is now reported on again in Abstract 8. Bevirimat is a first of class compound that acts as an inhibitor of HIV maturation. New findings in regard to Bevirimat (PA-457) relate to its use in a phase 2 clinical trial, ie, study 203, in which it was administered to patients who had failed their most recent regimens over 14 days as an add-on drug, in addition to background, as the possible equivalent of functional monotherapy (n=46). The results indicated that approximately 45% of patients had a viral load RNA drop >0.5 logs during this time, while some who received higher doses of Bevirimat had viral log drops that were 1.26 log. Of non-responders, 79% had low plasma trough levels of the compound. However, individuals who had the highest plasma levels did not necessarily have the best virologic responses.
The investigators also explored polymorphisms in the gag protein between positions 357-380 as possible contributors to eventual resistance against Beviramat. The results showed that superior virologic responses were obtained in patients who did not possess polymorphisms at any of positions 369, 370, and 371. Moreover, patients who did not possess any of these polymorphisms had an average viral load drop of 1.26 logs vs 0.6 logs for individuals who did have the polymorphisms in question. Although further research is necessary to address this and related topics, it is worth pointing out that a source of continuing concern in regard to Beviramat has been the potential for HIV to have generated cleavage site mutations and polymorphisms in the Gag gene as a result of long term use of protease inhibitors against which resistance might have developed. This is because of strong evidence that resistance to PIs can be due to Gag cleavage site mutations as well as to better understood mutations within the HIV protease(PR) gene. It is unknown how such substitutions might impact on the durability of treatment response to Bevirimat and clinical trials will be necessary to resolve these and related issues. In the meantime, the field looks forward to the continuing development of Beviramat as an anti-viral drug.
3. Recently Approved Drugs: Novel Results
Abstracts 121 and 122 presentes novel and mutually confirmative results in regard to newly analyzed data from the DUET-1 and DUET-2 trials. It will be recalled that these phase 3 clinical trials studied the combination of Etravirine together with boosted darunavir in patients who had failed earlier therapies and who possessed reistance mutations associated with each of NRTIs, NNRTIs, and PIs. In both of these studies, it has now been shown that a large number of pre-existing nucleoside resistance-associated mutations were able to hypersensitize to Etravirine. These novel findings help to explain the success of the DUET studies, since the patients in the trials all had nucleoside-associated mutations.
The above results have also been corroborated in tissue culture studies that have also shown that the list of mutations that were apparently able to hypersensitize to Etravirine is large. Indeed, the list includes many of the most common substitutions associated with NRTI failure. These findings collectively imply that the best use of Etravirine may continue to be in second-line, third-line, or even salvage therapy, in part because it may act best in individuals who have previous resistance against nucleoside compounds. This point may not apply to the same extent to other compounds that are still being developed for potential use in first-line therapy e.g. TMC-278 among others. It will also be interesting to know to what extent the novel compound being developed by Idenix, i.e. IDX899 (see above) might also involve significant hypersensitization by nucleoside analogue mutations. Conceivably, however, one could argue that it might not be advantageous for a first-line NNRTI to depend, in part, on the presence of such substitutions for its activity. It could also be argued that these results underline the relative importance of Etravirine in the success that was achieved in the DUET studies. This is not to suggest, however, that Darunavir was not also key toward success of the DUET regimens. Indeed, there are multiple other studies that point to the success of Darunavir in clinical trials in a context in which Etravirine was not employed as part of the therapeutic regimens employed, e.g. TITAN, POWER 1, 2, 3. The issue of NNRTI hypersensitization by NRTI mutations will require further analysis in the context of clinical trials in order to establish clinical significance.
The findings in Abstracts 121 and 122 on hypersensitization by NRTI mutations suggests that discussions of clinical cut-offs for Etravirine might be more complicated than expected. This subject was dealt with in a general sense in Abstracts 24, 110, and 130. The data suggest that the clinical cut-off for resistance against Etravirine is likely to be >5.0 in terms of fold-resistance. However, it is also clear that most of the work performed to date on this topic is based on an understanding of which NNRTI mutations were present and did not factor in a potential role for hypersensitizing NRTI mutations. This is a major reason why this subject should still be considered as in flux.
Interestingly, in Abstract 129, results were presented to suggest that prior treatment with Nevirapine (NVP) may result in an accumulation of mutations that compromise future responsiveness to Etravirine to a greater extent than occurs after failure on an Efaverenz (EFV)-based regimen.
4. Issues of resistance in Developing Countries
The relative availability of ARVs in developing countries today as compared with the situation that existed only 5 years ago has brought hope to millions of people infected by HIV. However, there is growing recognition that problems pertaining to HIV drug resistance will be inevitable in developing country settings. As in western countries, there is now debate on the clinical significance of minority viral species that contain drug resistance mutations and on whether or not the frequency of mutations associated with HIV drug resistance will be important. As an example, Abstract 120 presented data from South Africa in which it was reported that approximately 10% of women who had been involved in studies for prevention of mother-to-child transmission of HIV-1 had developed K103N mutations, irrespective of any prior exposure to single dose nevirapine. An obvious question is now the extent to which transmission of the K103N mutation may have occurred to these women and the extent to which transmitted resistance may now be a problem in South Africa and other developing country settings.. Of course, the issue of whether or not the existence of mutations such as K103N may potentially pre-dispose to subsequent virologic failure is key.
The conference contained a number of abstracts also suggesting that the presence of low-level mutations associated with drug resistance may, in fact, have the potential to compromise subsequent clinical responsiveness. The latter topic has important relevance to both short-term and long-term treatment responsiveness in developed as well as in developing country settings. In this context, a major issue is the amount of time that may be required in order for relevant ARVs to amplify a particular mutated variant of HIV from the viral quasispecies under drug selective pressure. The answer to this question will depend on what other drugs are part of the patient's ARV regimen as well as the potency of those other drugs. If they can successfully keep viral replication in check over long periods, then a durable treatment response might be expected. Of course, adherence to the ARV regimen will be key in this situation. Without adherence, it is likely that outgrowth of drug-resistant mutated varieties that are present as detectable minority species will occur even faster than what would take place if such minority resistant variants were not present.
Abstract 72 contained data in regard to a heroic effort to potentially reduce viral burden to levels below those that are normally achievable through the use of ARV regimens. Notably, efforts were made to intensify otherwise successful first-line treatment regimens with either Efavirenz or LPV/r. Unfortunately, the results demonstrated that such intensification did not impact to significant extent on overall treatment response. It is interesting that current efforts are now underway to determine whether the use of Raltagravir as an intensifying agent might make a more significant difference. Of course, the rationale for use of Raltagravir in this context is that earlier studies had shown that this drug is associated with more rapid drops in viral load after use in first-line therapy than have been observed with other classes of antiretroviral compounds. The data obtained must be regarded as disappointing in regard to therapy in both developed and developing countries.
It is also important to note that data continue to accumulate on the likelihood that certain mutations may be more predominant in certain viral subtypes. This topic is of obvious importance for developing countries in which non-B HIV subtypes predominate. Research presented in Abstract 51 demonstrated that subtype C viruses may be more prone toward the development of the K65R substitution than are viruses of other subtypes. In Abstract 108, use of ultrasensitive testing revealed the presence of a number of minor mutations that were present as sub-populations (in South Africa in patients failing first and second-line regimens). One of the most important findings presented was that 12 of 14 viral samples from patients with subtype C viruses contained the K65R mutation at levels between 0.5% to 4.4% of the total population. These results reinforce earlier data from both this series of conferences and the published literature that subtype C viruses may have a greater tendency than those of other subtypes to develop K65R. Validation of this result might have important implications in regard to both treatment as well as transmitted resistance since K65R, in general, is responsible for significant cross-resistance to multiple members of the nucleoside family of drugs. In this context, prior use of ZDV in treatment strategies may have mitigated against subsequent selection of K65R, as a consequence of the well described antagonism between thymidine analogue mutations (TAMs) and K65R; these seem to represent incompatible mutational pathways in regard to development of resistance. In this context, the current trend away from use of Zidovudine-based regimens throughout the world, including developing country settings, may portend toward higher incidence rates of K65 in the future. Were K65R to occur at higher frequency in countries in which subtype C viruses are endemic and to be easily transmitted, there could be potential complications for therapeutic interventions in such settings. It is also important to understand that K65R is not so much a signature mutation for any single compound such as Tenofovir as much as it is a broadly acting mutation responsible for cross-resistance against virtually all N(t)RTI drugs with the exception of ZDV. Hence, efforts should be made to minimize the extent to which K65R might be likely to develop and, in addition, to prevent its transmission.