| |
Predictors of selection of K65R: tenofovir use and lack of thymidine analogue mutations
|
| |
| |
AIDS: Volume 18(15) 21 October 2004
RESEARCH LETTER
Valer, Luisa; Martín-Carbonero, Luz; de Mendoza, Carmen; Corral, Angélica; Soriano, Vincent
Service of Infectious Diseases, Hospital Carlos III, Madrid, Spain.
Sponsorship: This work was partly funded by grants from the Asociación Investigación y Educación en SIDA (AIES), Comunidad Autónoma de Madrid (CAM), RIS (project 173) and Fondo de Investigación Sanitaria (FIS).
Over the past 5 years, 1846 HIV-infected patients underwent drug resistance testing at our institution. None out of 216 drug-naive subjects showed K65R. However, it was recognized in 53 out of 1630 antiretroviral-experienced patients (3.3%), of whom 10 had never been exposed to tenofovir. The rate of K65R increased from 0.6% in 1999 to 11.5% in 2004. The recognition of K65R correlated negatively with the presence of thymidine analogue mutations but positively with Q151M.
High rates of early virological failure associated with the emergence of the K65R mutation at the HIV-1 reverse transcriptase gene have recently been reported among HIV-infected patients on tenofovir-containing triple nucleos(t)ide regimens. Data from large genotypic databases have shown that K65R is quite uncommon, although its rate might be increasing. K65R is selected in vitro by tenofovir, zalcitabine, didanosine, stavudine and abacavir. In vivo, tenofovir, abacavir and didanosine have been shown to select the K65R mutation. Although K65R reduces the susceptibility to tenofovir and to a lesser extent affects the activity of other nucleoside analogues such as abacavir, didanosine and lamivudine, it retains the activity of zidovudine. Therefore, viruses carrying K65R show broad cross-resistance to nucleos(t)ide reverse transcriptase inhibitors.
Data about the rate of K65R and its association with other nucleoside analogue resistance mutations are scarce. Herein, we describe the rate of K65R in a large database of genotypic drug resistance reports in a reference hospital in Madrid, Spain. Its association with other reverse transcriptase resistance mutations is further described.
All plasma samples referred to our reference laboratory for drug resistance testing over the past 5 years were examined. A total of 1846 specimens with genotypic reports were analysed. They were drawn from 216 drug-naive and 1630 treatment-experienced HIV-infected patients.
A total of 53 specimens showed K65R, providing an overall rate of 2.9%. None of drug-naive individuals showed K65R. The rate of K65R significantly increased over time among treatment-experienced patients, from 0.6% in 1999 to 11.5% in the first trimester of 2004.
Although K65R was identified mainly in individuals failing tenofovir-based combinations, it was also recognized in 10 patients never exposed to tenofovir. The latter were receiving stavudine plus didanosine (four), abacavir plus lamivudine (two), stavudine plus lamivudine (two), stavudine plus abacavir (one) and zalcitabine (one) at the time of resistance testing. The remaining 43 subjects carrying K65R were failing different tenofovir-based combinations: tenofovir plus didanosine (32), tenofovir plus abacavir (six), tenofovir plus lamivudine (four) and tenofovir alone (one).
The rate of thymidine analogue mutations (TAM) was significantly lower among patients harbouring K65R with respect to the rest. Whereas TAM appeared in 60.1% of plasma specimens collected from treatment-experienced patients lacking K65R, they were present in only 12 out of 53 with K65R (22.6%; P < 0.0001). Moreover, only K70R or the uncommon K219E were seen along with K65R. No patients with K65R viruses harboured more than these two TAM.
In the univariate analysis, the presence of K65R was associated with the absence of M41L (P < 0.0001), E44D/A (P = 0.0013), D67N (P < 0.0001), L210W (P < 0.0001) and T215Y/F (P < 0.0001). In contrast, K65R was associated with the presence of V75I (P < 0.0001), Y115F (P < 0.0001) and the Q151 complex (P < 0.0001). In the multivariate analysis, the presence of K65R was inversely associated with the number of TAM [odds ratio (OR) 0.54; 95% confidence interval (CI) 0.24-0.54] and with the presence of T215Y/F (OR 0.09; 95% CI 0.01-0.9), whereas it was positively associated with the presence of Q151M (OR 4.82; 95% CI 1.76-13.22) and the total number of other nucleoside analogue resistance mutations distinct of TAM (OR 1.59; 95% CI 1.25-2.02). Of note was the fact that the M184V mutation accompanied K65R in 24 cases (45%). However, no association was found between the presence of K65R and M184V, although a trend towards its concomitant appearance was noted (P = 0.1).
As other groups have shown, the selection of K65R is significantly associated with the use of tenofovir. The incidence of this mutation has increased in recent times, presumably as a result of the increasing use of tenofovir in clinical practice. However, as was shown in our study, other nucleoside combinations may also favour the selection of K65R, although it appears to occur more rarely. In our analysis, K65R was particularly frequent among patients failing tenofovir/didanosine-based combinations, but this may indirectly reflect the fact that a large group of patients in our institution have been exposed to this combination during the past 2 years.
We found a reciprocal exclusion between TAM and mutation K65R. They may represent divergent and antagonistic pathways driving to nucleoside analogue resistance. Moreover, only two TAM (K70R and K219E) were seen along with K65R. Codon 219 changes seem to be the only TAM tolerated by K65R, whereas a clear negative interaction seems to exist with codon 41 and 215 mutations. Of note was the fact that all our cases with codon 219 changes showed a specific amino acid substitution (K219E), whereas other changes (K219Q and K219N), which are frequently seen with other TAM, never appeared with K65R.
Previous studies have claimed that the Q151M complex might be associated with K65R. This association may be clinically relevant, because tenofovir seems to preserve its activity in the presence of Q151M alone but not when K65R is also present. Viruses harbouring Q151M along with the K65R mutation are highly resistant to all nucleoside analogues. In our study, 10 out of 53 patients with K65R had Q151M.
The frequent selection of M184V along with K65R results in a novel multi-nucleoside-resistant genotype, although it may provide in vitro an increased susceptibility to zidovudine and perhaps to stavudine. Clinical data are needed to clarify to what extent the K65R+M184V genotype compromises the activity of nucleoside analogues. This information may be particularly relevant given that tenofovir is now widely used with lamivudine, and single pills containing tenofovir and emtricitabine are expected to be available soon.
In conclusion, the rate of K65R has increased significantly over the past 5 years among treatment-experienced patients, and currently is above 10%. This fact is directly related to the wide use of tenofovir in clinical practice. Given the antagonism between K65R and TAM, further clinical studies assessing the benefit of using combination nucleoside analogues driving these different resistance pathways should be conducted.
|
|
| |
| |
|
|
|
