Abstract

BackgroundThe antiviral efficacy of didanosine in patients experiencing virological failure is not well known

MethodsA total of 168 patients (139 men and 29 women) receiving stable antiretroviral therapy with plasma human immunodeficiency virus type 1 (HIV-1) RNA levels of 1000–100,000 copies/mL were randomly assigned to have didanosine (n=111) or placebo (n=57) added to their currently failing regimen for 4 weeks. The primary efficacy end point was the change in HIV-1 RNA level from baseline to week 4

ResultsAt baseline, the median HIV-1 RNA level was 3.8 log10 copies/mL, the median CD4 cell count was 378 cells/mm3, and the median number of nucleoside reverse-transcriptase inhibitor–associated mutations (NAMs) was 4. At week 4, a significant decrease in the median HIV-1 RNA level was observed in the didanosine group, compared with that in the placebo group (−0.56 vs. +0.07 log10 copies/mL, respectively) (P<.0001). A total of 33 patients (31%) in the didanosine group, compared with 3 (6%) in the placebo group, had HIV-1 RNA levels <400 copies/mL (P<.001). Significant antiviral activity of didanosine was observed in patients with up to 5 NAMs at baseline. Diarrhea occurred in 5 patients (5%) in the didanosine group and 2 patients (4%) in the placebo group

ConclusionsIn HIV-1–infected patients experiencing failure of antiretroviral therapy, didanosine retains short-term antiviral activity

Despite the growing number of antiretroviral agents available, therapeutic options for HIV-1–infected patients experiencing failure of antiretroviral therapy remain limited [1, 2]. Indeed, treatment failure is frequently associated with the emergence of drug-resistance mutations, which decrease the activity of antiretroviral agents. Furthermore, these mutations usually confer cross-resistance among drugs within the same class, and physicians face a difficult challenge when selecting antiretroviral agents to optimize therapy [2, 3]. Nucleoside reverse transcriptase inhibitors (NRTIs), which were the first antiretroviral agents available, are still the most frequently used class of drugs. Resistance to NRTIs has been associated with NRTI-associated mutations (NAMs), which can be detected in plasma by use of genotypic resistance tests that sequence the reverse-transcriptase gene [4]. A panel of different NAMs has been characterized, and algorithms have been defined, to try to correlate the presence of these mutations to the predicted in vivo antiviral activity of NRTIs [4, 5]. These algorithms do perform well for drugs such as zidovudine, lamivudine, and abacavir. However, algorithms used to define genotypic resistance to didanosine are based on limited data. Although in vivo and in vitro resistance to didanosine has been associated with the selection of the L74V mutation, this mutation remains infrequent and was mainly selected when didanosine was used as monotherapy [6–8 ]. It is unclear, however, whether didanosine retains antiviral activity in patients with viruses carrying other NAMs

We therefore designed a randomized, placebo-controlled, comparative trial in patients whose current antiretroviral therapy was failing, to assess the short-term antiviral activity and safety of didanosine and to better understand the cumulative effect of NAMs. The present study provides the clinician with the knowledge that the addition of didanosine to a failing regimen, even in patients who have received didanosine in the past, is both safe and effective in controlling viral load

Patients, Materials, and Methods

PatientsEligible patients were HIV-1–infected adults who were receiving stable antiretroviral therapy for at least 3 months, had a CD4 cell count ⩾100 cells/mm3, and had plasma HIV-1 RNA levels of 1000–100,000 copies/mL. Exclusion criteria were pregnancy, current treatment with didanosine or tenofovir (because of the unknown interactions between didanosine and tenofovir at the time the study was initiated), alcohol abuse, ongoing opportunistic infection, uncontrolled diarrhea, and a history of peripheral neuropathy or pancreatitis. Additional exclusion criteria were serum lipase, creatinine, or total bilirubin levels >1.5 times the upper limit of normal and hepatic aminotransferase levels >3 times the upper limit of normal

Study designThe study was a multicenter, randomized, double-blind, placebo-controlled, comparative trial performed at 29 sites in France. The protocol was approved by the Paris Saint-Louis ethics committee. All patients gave written, informed consent. Randomization was performed by a centralized procedure, and patients were randomly assigned, at a ratio of 2:1, to have didanosine or a matching placebo added to their current antiretroviral regimen. Didanosine was provided as 1 enteric-coated capsule (400 mg if body weight was ⩾60 kg and 250 mg if body weight was <60 kg) and placebo as 1 matching capsule. Patients were provided with 1 bottle containing 30 capsules and were instructed to take 1 capsule of didanosine or placebo every day on an empty stomach or at least 2 h before or after a meal

Study monitoringPatients were assessed at baseline, week 2, and week 4. At each visit, clinical data were collected through medical history and physical examination, and blood specimens were obtained. Routine analyses were performed at each site throughout the study period and included a complete blood count, CD4 cell count, and tests of liver, kidney, muscle, and pancreatic function

Safety was assessed through the reporting of adverse events and abnormal laboratory measurements. The severity of adverse events was assessed by use of the World Health Organization toxicity grading scale (available at: http://icssc.org/adverse_event_reporting.htm)

A centralized genotypic resistance test was performed at baseline, for plasma HIV-1 RNA level, in accordance with the consensus method of the Agence Nationale de Recherches sur le SIDA resistance group, by use of an ABI 3100 genetic analyzer (Applied Biosystems) [9]. Sequences were analyzed by use of Sequence Navigator software (version 1.0.1; Applied Biosystems) and reported as amino acid changes with respect to the sequence of the wild-type virus HXB2. NAMs were defined according to the International AIDS Society–USA drug-resistance mutations group definition [5]. Thymidine-associated mutations (TAMs) were defined as mutations at positions M41L, D67N, K70R, L210W, T215Y/F, or K219E/Q

A second round of genotypic resistance testing was performed at week 4, to detect the potential emergence of new mutations. At the end of the study, plasma samples collected at each visit at the different centers were centralized and batched, to measure plasma HIV-1 RNA levels by use of the Amplicor HIV-1 Monitor 1.5 ultrasensitive assay (Roche Molecular Systems), which has a lower limit of quantification of 50 copies/mL

Adherence to study medication was assessed at the end of the study by counting the remaining capsules in bottles returned by the patients. At the week-4 visit, results of baseline genotypic resistance testing were available, and antiretroviral therapy was optimized at the discretion of each physician

Study end pointsThe primary efficacy end point was the change in plasma HIV-1 RNA level from baseline to week 4. Secondary end points included the change in plasma HIV-1 RNA level from baseline to week 2; the proportion of patients with a plasma HIV-1 RNA level <400 and <50 copies/mL; changes in CD4 cell counts; clinical progression to AIDS (defined as the occurrence of any new clinical event included in category C of the 1993 classification of the Centers for Disease Control and Prevention [10]); and the proportion of patients with moderate (grade 2) to severe (grade 4) adverse events and laboratory abnormalities, who discontinued study medication, and who were adherent to study medication. Relationships between changes in plasma HIV-1 RNA levels and baseline genotypic resistance mutations were also assessed

Statistical analysisThe sample size was calculated on the basis of the primary virologic end point, to show superiority of the didanosine group over the placebo group. On the basis of an expected decrease of 0.5 log10 copies/mL in plasma HIV-1 RNA level at week 4 in the didanosine group (SD, 0.9 log10 copies/mL), 156 patients were required for this assessment (104 in the didanosine group and 52 in the placebo group), with a 2-sided α level of 0.05 and a statistical power of 90%. To account for patients lost to follow-up, a total of 168 patients were randomized

The lower limit of the detectable level was assigned to HIV-1 RNA values reported as less than the lower threshold of detectability. The primary efficacy end point was analyzed on an intention-to-treat basis (including all randomized patients, even if they did not start or prematurely discontinue study medication), by use of the Wilcoxon&amp;rank sum test

For secondary end points, χ2 or Fisher’s exact tests were used to compare categorical variables, and the Wilcoxon&amp;rank sum test was used to compare continuous variables. Two-sided P values are reported for each comparison. Statistical analyses were performed by use of SAS software (version 8.2; SAS Institute). Statistical analysis and data management were performed by the study sponsor (Bristol-Myers Squibb, Rueil, France)

Results

Study populationBetween March 2002 and February 2003, 197 patients underwent screening procedures, and 168 were found to be eligible for randomization into the study (figure 1). The baseline characteristics of the patients were well balanced between the groups (table 1). Most patients were men (83%), with a mean age of 44 years, a median CD4 cell count of 378 cells/mm3, and a median plasma HIV-1 RNA level of 3.8 log10 copies/mL. Baseline antiretroviral regimens included NRTIs and protease inhibitors (PIs) in 35%, NRTIs and nonnucleoside reverse-transcriptase inhibitors (NNRTIs) in 38%, and ⩾3 NRTIs alone in 13% of patients. Also, 68% of patients had a history of didanosine therapy (median duration, 17 months) before entering the trial, and the median time between the end of didanosine therapy and randomization was 3.4 years

Figure 1

Profile of patient enrollment and discontinuation of study medication through 4 weeks of treatment

Figure 1

Profile of patient enrollment and discontinuation of study medication through 4 weeks of treatment

Table 1

Baseline characteristics of the patients

Table 1

Baseline characteristics of the patients

Results of baseline genotypic resistance testing were not available for 9 patients (6 in the didanosine group and 3 in the placebo group). According to the treatment experience of the patients, baseline genotypic resistance testing revealed that 99% of plasma isolates were expressing reverse transcriptase mutations associated with resistance to NRTIs. The median number of NAMs was 4, with 91% of the isolates carrying the M184V/I mutation associated with resistance to lamivudine. Fifty-seven percent of the isolates also carried the T215Y/F mutation, and the median number of TAMs was 3. Very few patients had either the L74V mutation (10 in the didanosine group and 4 in the placebo group) or the K65R mutation (2 in the didanosine group). Also, 58% of the isolates expressed primary NNRTI-associated mutations, and 46% expressed PI-associated mutations

Overall, 97% of the patients completed the 4 weeks of treatment receiving their randomized regimen. Only 5 patients (2 in the didanosine group and 3 in the placebo group) prematurely discontinued study treatment or were lost to follow-up (figure 1)

Study end pointsThe primary end point of the study is shown in figure 2A. At week 2, the didanosine group demonstrated a significant change in median plasma HIV-1 RNA level, compared with the placebo group, and this antiviral response was sustained through week 4 (−0.56 log10 copies/mL in the didanosine group vs. +0.07 log10 copies/mL in the placebo group) (P<.0001). Similar results were obtained in the subgroup of patients who were didanosine experienced (median change in plasma HIV-1 RNA level at week 4, −0.48 log10 copies/mL in the didanosine group vs. +0.07 log10 copies/mL in the placebo group) (P<.0001) (figure 2B)

Figure 2

Median change in plasma HIV-1 RNA level from baseline and interquartile range (IQR) in patients receiving didanosine or placebo in addition to their current failing regimen, by intention-to-treat analysis (A) and analysis of the subgroup of patients who were didanosine experienced (B)

Figure 2

Median change in plasma HIV-1 RNA level from baseline and interquartile range (IQR) in patients receiving didanosine or placebo in addition to their current failing regimen, by intention-to-treat analysis (A) and analysis of the subgroup of patients who were didanosine experienced (B)

Also, the proportion of patients with plasma HIV-1 RNA levels <400 copies/mL at week 4 was significantly higher in the didanosine group (31%) than in the placebo group (6%) (P < .001, Fisher’s exact test) (data not shown). Finally, the proportion of patients with plasma HIV-1 RNA levels <50 copies/mL at week 4 was significantly higher in the didanosine group (11%) than in the placebo group (0%) (P<.01, Fisher’s exact test) (data not shown)

We then assessed the median reduction in plasma HIV-1 RNA level in both groups according to the number of NAMs at baseline (table 2). Among patients with the M184V/I mutation, the median reduction in HIV-1 RNA level was significantly higher in the didanosine group (n=94) than in the placebo group (n=45) (−0.61 vs. +0.07 log10 copies/mL, respectively) (P<.0001). Similarly, among patients with the T215Y/F mutation, the median reduction in HIV-1 RNA level was significantly higher in the didanosine group (n=55) than in the placebo group (n=31) (−0.38 vs. +0.01 log10 copies/mL, respectively) (P=.001). On the other hand, the median reductions in HIV-1 RNA level were only −0.06, −0.20, and −0.08 log10 copies/mL at week 4 among patients in the didanosine group with the L74V (n=9), K65R (n=2), or T69D (n=9) mutations, respectively, which are not significantly different from those in the placebo group (table 2). Of note, in patients with the K70R mutation, the median reduction in HIV-1 RNA level was significantly higher in the didanosine group (n=27) than in the placebo group (n=14) (−0.94 vs. +0.07 log10 copies/mL, respectively) (P=.0002)

Table 2

HIV-1 RNA responses at week 4, according to baseline resistance mutations (intention-to-treat analysis)

Table 2

HIV-1 RNA responses at week 4, according to baseline resistance mutations (intention-to-treat analysis)

We also assessed the median change in plasma HIV-1 RNA level from baseline to week 4 in each group according to the number of NAMs (figure 3A) or TAMs (figure 3B) at baseline. We saw significant antiviral activity of didanosine, compared with that of placebo, in patients with up to 5 NAMs at baseline, with a median reduction of HIV-1 RNA level of −0.45 log10 copies/mL in the didanosine group and +0.07 log10 copies/mL in the placebo group (P=.047). Similarly, we observed significant antiviral activity of didanosine, compared with that of placebo, in patients with up to 3 TAMs at baseline, with a median reduction of HIV-1 RNA level of −0.45 log10 copies/mL in the didanosine group and +0.07 log10 copies/mL in the placebo group (P<.001) (figure 3B). There was also a trend toward a better antiviral response in patients with 1 TAM, compared with that in patients with no TAMs, which might be due to the presence of the K70R mutation in 8 of the 17 patients with 1 TAM

Figure 3

Median change in plasma HIV-1 RNA level from baseline to week 4 and interquartile range (IQR), according to the number of baseline thymidine-associated mutations (TAMs) (A) and nucleoside reverse-transcriptase inhibitor–associated mutations (NAMs) (B), by intention-to-treat analysis

Figure 3

Median change in plasma HIV-1 RNA level from baseline to week 4 and interquartile range (IQR), according to the number of baseline thymidine-associated mutations (TAMs) (A) and nucleoside reverse-transcriptase inhibitor–associated mutations (NAMs) (B), by intention-to-treat analysis

Among the 159 patients for whom results of baseline genotypic resistance testing were available, a second genotypic resistance test was performed for 136 patients (86%) at week 4, to detect the emergence of new NAMs. No new L74V or K65R mutations were detected in the didanosine group, and the distribution of mutations remained well balanced between the groups, except that the V118I mutation emerged more frequently in the didanosine group (5 patients [6%]) than in the placebo group (0 patients). The same low percentage (<5%) of new NNRTI- or PI-associated mutations was observed in the didanosine and the placebo groups at week 4 (data not shown)

There were no significant differences between the groups in the median change in CD4 cell counts over time. At week 4, the median changes from baseline were +1.5 and −6.0 CD4 cells/mm3 in the didanosine and placebo groups, respectively (P=.4, Wilcoxon&amp;rank sum test). No patient experienced progression to AIDS or died during the study

Safety and tolerabilityOnly 5 serious adverse events were reported during the study (3 in the didanosine group and 2 in the placebo group), none of which was related to treatment; 2 of them led to discontinuation of study medication. Overall, the incidence of adverse events was similar between the groups (42 patients [38%] in the didanosine group and 20 patients [36%] in the placebo group) (P=.82). Most of the adverse events were gastrointestinal (20%) or affected the nervous system (14%). Only 5 patients (4.5%) in the didanosine group and 2 patients (3.6%) in the placebo group complained of grade 1–2 diarrhea. No grade 3 diarrhea was reported. Also, 1 patient in the didanosine group had a grade 2 elevation of lipase serum level, and 1 patient in the placebo group had a grade 3 elevation of lipase serum level. Finally, using pill counts (available data on 70% of patients), we observed that adherence to study medication through the 4 weeks of the study was high in both groups (91% of patients in the didanosine group and 100% of patients in the placebo group had taken at least 90% of the study drug) (P=.096)

Discussion

The present study was designed to assess the short-term efficacy and safety of didanosine in treatment-experienced patients experiencing virological failure despite ongoing therapy. The present study has shown that the addition of didanosine to a currently failing regimen was able to induce a significant reduction in viral replication. Indeed, although nearly all patients had NAMs at baseline, the median change in plasma HIV-1 RNA level from baseline was −0.56 log10 copies/mL at week 4 in the didanosine group and +0.07 log10 copies/mL in the placebo group (P<.0001). The antiviral efficacy of didanosine was demonstrated as early as 2 weeks after randomization and was associated with a higher proportion of patients who achieved a plasma HIV-1 RNA level <400 or <50 copies/mL at week 4 in the didanosine group than in the placebo group

However, since only a small percentage of patients in the present study achieved an undetectable plasma HIV-1 RNA level, it is not advisable to use this strategy of didanosine intensification to control viral replication in patients experiencing treatment failure. The level of reduction of viral replication seen in the present study is, however, clinically relevant and comparable to that obtained in add-on studies with abacavir or tenofovir in a similar patient population [17, 19, 20]. This result therefore allows the selection of didanosine as a potentially useful agent to optimize therapy in patients expressing NAMs whose current antiretroviral therapy is failing [11]. Furthermore, since 68% of patients in the present study were didanosine experienced, it is interesting to note that a significant antiviral efficacy of didanosine, over placebo, was also observed in this subgroup of patients. Although the reasons why didanosine was discontinued in these patients remain unclear (virologic failure and/or intolerance to the old formulation of didanosine), we should not preclude the use of didanosine in didanosine-experienced patients

The present study also gave us the ability to better define the antiviral activity of didanosine in the presence of NAMs, mutations that confer decreased susceptibility to NRTIs [4, 5, 12]. lnterestingly, didanosine exerted significant antiviral efficacy in patients who had up to 5 NAMs at baseline, suggesting an interest in the use of didanosine in patients experiencing failure with other NRTIs. However, although very few patients carried the L74V, T69D, or K65R mutations, the addition of didanosine was not, as expected, associated with significant antiviral activity in these patients [8]. In contrast, we have demonstrated that didanosine retained significant antiviral activity against isolates expressing the M184V/I mutation selected in patients whose lamivudine-containing regimen had failed, a finding that has been suggested in previous studies [13–15 ]. Also, we studied the antiviral activity of didanosine according to the presence of TAMs, mutations that are selected by zidovudine or stavudine and that are known to decrease the activity of zidovudine, stavudine, abacavir, and tenofovir in vivo [4, 5, 12, 16, 17]. Although the antiviral activity of didanosine was also affected by TAMs, as has been previously suggested [18, 19], the present study has shown that this drug retained significant antiviral activity against isolates expressing up to 3 TAMs at baseline. This level of activity in such resistant isolates compares favorably with results obtained with abacavir or tenofovir in other studies [16, 17]. Further studies are needed, however, to define a relevant genotypic score associated with antiviral response to didanosine, which may improve the current algorithms

One major limitation of the study was the short duration of follow-up. Indeed, the study lasted for only 4 weeks, which is the average time needed for patients experiencing treatment failure to obtain the results of their genotypic resistance test, to allow for optimization of their treatment. We therefore felt that it was not ethical to extend the study beyond 4 weeks. Indeed, our goal was to demonstrate the antiviral activity of didanosine in patients experiencing treatment failure, not to use a strategy of treatment intensification with didanosine in these patients. At week 4 of the study, the results of the genotypic resistance tests performed at baseline were available, and the treatment of the patients was optimized at their physicians discretion

We think, however, that our results are valid, since the antiviral activity of NRTIs in vivo is rapid, with maximal antiviral activity having been observed as soon as 2 weeks in studies with a similar design [17, 20, 21]. Obviously, however, we are not able to predict the long-term antiviral activity of didanosine in this setting

Another potential concern resulting from the present study was the emergence of new mutations in the didanosine group. We therefore performed a second round of genotypic resistance testing in both groups at the end of the study period, and we did not find differences in the incidence of new reverse transcriptase–associated mutations. No L74V or K65R mutations emerged in the didanosine group. However, the short follow-up in the present study limited the chances of selecting new mutations or allowing the reemergence of previously selected mutations, such as L74V, and this issue should be considered if didanosine is to be recycled in the context of salvage therapy

The modest increase in CD4 cell count seen in the present study is consistent with the results of similar trials and with the low proportion of patients who achieved an undetectable HIV-1 RNA level [17, 20, 21]. As would be expected in a population of patients treated for only 4 weeks, very few adverse events were reported, and no serious adverse event was related to treatment. The present study allowed, for the first time, comparison of the tolerance of the new enteric-coated formulation of didanosine with that of placebo. Also, in this short-term trial, only 5% of patients in the didanosine group complained of diarrhea, a proportion that was similar to that seen in the placebo group

In conclusion, we have demonstrated the short-term safety and antiviral efficacy of didanosine when added to a failing regimen. The results of the present study will allow better identification of treatment-experienced patients with virological failure in whom didanosine could retain significant antiviral activity and could be used in combination with other agents to optimize therapy

AI454-176 JAGUAR Study Team Members

The members of the AI454-176 JAGUAR study team were as follows: J.-M. Molina (trial chair); G. Leleu, M. Troccaz, and G. Chapuis (trial coordinator and monitors); N. Schmidely (trial statistician); V. Calvez and A.-G. Marcellin (trial virologists); J.-M. Molina, J. Pavie, G. Leleu, V. Calvez, N. Schmidely, M. Troccaz, F. Clavel, I. Dujardin, and H. Benech (scientific committee); and P. Yéni, F. Ferchal, and T. Grooters (data safety and monitoring board)

Participating centers (all in France) and investigators were as follows: Hôpital Lagny-Marne-La-Vallee, Lagny—F. David; Hôpital Avicenne, Bobigny—M. Bentata; Hôpital Saint-Jacques, Besançon—B. Hoen; Hôpital Saint-Louis, Paris—J.-M. Molina, B. Loze, J. Pavie, N. Colin de Verdiere, S. Fournier, and I. Madelaine; Hôpital Cochin, Paris—D. Salmon; Hôpital Raymond Poincare, Garches—C. Perronne and P. De Truchis; Hôpital Necker, Paris—B. Dupont, O. Lortholary, and L. Roudière; Hôpital Pitie-Salpetriere, Paris—C. Katlama, A. Simon-Coutellier, V. Calvez, A. Marcellin, and M. Kirstetter; Hôpital Bicetre, Kremlin Bicetre—J.-F. Delfraissy; Paris Hôpital Saint-Antoine, Paris—P.-M. Girard; Hôpital Saint-Andre, Bordeaux—P. Morlat; Hôpital Pessac, Bordeaux—J.-L. Pellegrin; Hôpital Edouard Herriot, Lyon—J.-M. Livrozet; Hôpital Hotel-Dieu, Lyon—C. Trepo and L. Cotte; Hôpital Sainte-Marguerite, Marseille—H. Gastaut and I. Poizot-Martin; Hôpital Conception, Marseille—B. Gallais; Hôpital Gui de Chauliac, Montpellier—J. Reynes; Hôpital de lHotel-Dieu, Nantes— F. Raffi; Hôpital de lArchet, Nice—P. Dellamonica; Hôpital Purpan, Toulouse—P. Massip and B. Marchou; Centre Hospitalo-Universitaire de Caen, Caen—C. Bazin; Hôpital Saint-Etienne, St-Etienne—F. Lucht; Hôpital de Toulon, Toulon—A. Lafeuillade; Hôpital d Orleans, Orleans—L. Hocqueloux and T. Prazuc

Acknowledgments

We thank Nathalie Schmidely (Bristol-Myers Squibb, Rueil, France), who performed the statistical analyses

References

1.
Deeks
SG
Treatment of antiretroviral drug–resistant HIV‐1 infection
Lancet
 , 
2003
, vol. 
362
 (pg. 
2002
-
11
)
2.
Yeni
PG
Hammer
SM
Hirsch
M
, et al.  . 
Treatment for adult HIV infection. 2004 recommendations of the International AIDS Society–USA Panel
JAMA
 , 
2004
, vol. 
292
 (pg. 
251
-
65
)
3.
Dybul
M
Fauci
AS
Bartlett
JG
Kaplan
JE
Pau
AK
Panel on Clinical Practices for the Treatment of
HIV
Guidelines for using antiretroviral agents among HIV‐infected adults and adolescents. Recommendations of the Panel on Clinical Practices for Treatment of HIV
MMWR Morb Mortal Wkly Rep
 , 
2002
, vol. 
51
 (pg. 
1
-
57
)
4.
Hirsch
MS
Brun‐Vezinet
F
Clotet
B
, et al.  . 
Antiretroviral drug resistance testing in adults infected with human immunodeficiency virus type 1: 2003 recommendations of an International AIDS Society–USA Panel
Clin Infect Dis
 , 
2003
, vol. 
37
 (pg. 
113
-
28
)
5.
D’Aquila
RT
Schapiro
JM
Brun‐Vézinet
F
, et al.  . 
Drug resistance mutations in HIV‐1
Top HIV Med
 , 
2003
, vol. 
11
 (pg. 
92
-
6
)
6.
St. Clair
MH
Martin
JL
Tudor‐Williams
G
, et al.  . 
Resistance to ddI and sensitivity to AZT induced by a mutation in HIV‐1 reverse transcriptase
Science
 , 
1991
, vol. 
253
 (pg. 
1557
-
9
)
7.
Shafer
RW
Kozal
MJ
Winters
MA
, et al.  . 
Combination therapy with zidovudine and didanosine selects for drug‐resistant human immunodeficiency virus type 1 strains with unique patterns of pol gene mutations
J Infect Dis
 , 
1994
, vol. 
169
 (pg. 
722
-
9
)
8.
Winters
MA
Shafer
RW
Jellinger
RA
Mamtora
G
Gingeras
T
Merigan
TC
Human immunodeficiency virus type 1 reverse transcriptase genotype and drug susceptibility changes in infected individuals receiving dideoxyinosine monotherapy for 1 to 2 years
Antimicrob Agents Chemother
 , 
1997
, vol. 
41
 (pg. 
757
-
62
)
9.
Jung
M
Agut
H
Candotti
D
Ingrand
D
Katlama
C
Huraux
JM
Susceptibility of HIV‐1 isolates to zidovudine: correlation between widely applicable culture test and PCR analysis
J Acquir Immune Defic Syndr
 , 
1992
, vol. 
5
 (pg. 
359
-
64
)
10.
1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults
MMWR Recomm Rep
 , 
1992
, vol. 
41
 (pg. 
1
-
19
)
11.
Marshner
IC
Collier
AC
Coombs
RW
, et al.  . 
Use of changes in plasma levels of human immunodeficiency virus type 1 RNA to assess the clinical benefit of antiretroviral therapy
J Infect Dis
 , 
1998
, vol. 
177
 (pg. 
40
-
7
)
12.
Whitcomb
JM
Parkin
NT
Chappey
C
Hellmann
NS
Petropoulos
CJ
Broad nucleoside reverse‐transcriptase inhibitor cross‐resistance in human immunodeficiency virus type 1 clinical isolates
J Infect Dis
 , 
2003
, vol. 
188
 (pg. 
992
-
1000
)
13.
Winters
MA
Bosch
RJ
Albrecht
MA
Katzenstein
DA
AIDS Clinical Trials Group 364 Study Team
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
, vol. 
188
 (pg. 
537
-
40
)
14.
Eron
J
Bosch
RJ
Petch
L
Fiscus
S
Frank
I
Antiretroviral activity of didanosine in lamivudine‐experienced subjects in comparison to activity in subjects who were lamivudine‐naive
Antivir Ther
 , 
2002
, vol. 
7(Suppl 1)
 pg. 
S135
 
15.
Pozniak
A
Gazzard
BB
Peeters
M
Hoetelmans
R
Graham
NM
Influence of the M184V mutation on virologic outcome of highly active antiretroviral therapy with or without didanosine
Antivir Ther
 , 
2002
, vol. 
7(Suppl 1)
 pg. 
S164
 
16.
Margot
NA
Isaacson
E
McGowan
I
Cheng
AK
Schooley
RT
Miller
MD
Genotypic and phenotypic analyses of HIV‐1 in antiretroviral experienced patients treated with tenofovir DF
AIDS
 , 
2002
, vol. 
16
 (pg. 
1227
-
35
)
17.
Katlama
C
Clotet
B
Plettenberg
A
, et al.  . 
The role of abacavir (ABC, 1592) in antiretroviral therapy–experienced patients: results from a randomized, double‐blind trial. CNA 3002 European Study Team
AIDS
 , 
2000
, vol. 
14
 (pg. 
781
-
9
)
18.
Costagliola
D
Descamps
D
Calvez
V
, et al.  . 
Presence of thymidine‐associated mutations and response to d4T, abacavir and ddI in the control arm of the NARVAL ANRS 088 trial [abstract 450]
Program and abstracts of the 8th Conference on Retroviruses and Opportunistic Infections (Chicago)
 , 
2001
Alexandria, VA
Foundation for Retrovirology and Human Health
19.
Japour
AJ
Welles
S
D’Aquila
RT
, et al.  . 
Prevalence and clinical significance of zidovudine resistance mutations in human immunodeficiency virus isolated from patients after long‐term zidovudine treatment. AIDS Clinical Trials Group 116B/117 Study Team and the Virology Committee Resistance Working Group
J Infect Dis
 , 
1995
, vol. 
171
 (pg. 
1172
-
9
)
20.
Schooley
RT
Ruane
P
Myers
RA
, et al.  . 
Tenofovir DF in antiretroviral experienced patients: results from a 48‐week, randomized, double‐blind study
AIDS
 , 
2002
, vol. 
16
 (pg. 
1257
-
63
)
21.
Squires
K
Pozniak
AL
Pierone
G
, et al.  . 
Tenofovir disoproxil fumarate in nucleoside‐resistant HIV‐1 infection: a randomized trial
Ann Intern Med
 , 
2003
, vol. 
139
 (pg. 
313
-
20
)
Presented in part: 43rd Annual Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, 14–17 September 2003 (abstract H-447)
Financial support: Bristol-Myers Squibb
Potential conflicts of interest: J.-M.M. has received grant support and lecture fees from Abbott, Bristol-Myers Squibb, DuPont Pharmaceuticals, Gilead Sciences, GlaxoSmithKline, Roche, and Triangle Pharmaceuticals; V.C. has received grant support and lecture fees from Bristol-Myers Squibb, Gilead Sciences, and Triangle Pharmaceuticals
The investigators participating in the AI454-176 JAGUAR study are listed after the text