Experience using MedDRA for global events coding in HIV clinical trials.

OBJECTIVES: To assess the feasibility of coding with MedDRA, the Medical Dictionary for Regulatory Activities, not only serious adverse events required for notification but also all other events usually reported in HIV clinical trials. To develop an approach for MedDRA implementation within an institutional research unit that contributes to an efficient, concise and reproducible event coding. To evaluate the impact of the maintenance and the versioning of this new medical terminology. METHODS: MedDRA versions 3.0 and 5.0 were used for coding hundreds of events reported through two HIV clinical trials. The implementation of MedDRA consisted in the setup of a training program, guidelines to clinical investigators, coding rules and medical validation process. MedDRA version 6.1 was applied to the coding made with the MedDRA version 5.0 in order to identify the assignments affected by the new version and to determine the impact of versioning. RESULTS: Coding with MedDRA all types of events in HIV clinical trials was feasible even though coders experienced some difficulties due mainly to the lack of precision in the investigator verbatim and the high specificity and sensitivity of MedDRA. The addition of appropriate tools to support the use of MedDRA improved significantly the coding of all types of events in HIV clinical trials. The impact of MedDRA versioning was limited and did not result in significant issues. The global implementation process of MedDRA required important resources in terms of qualified personnel, organisation and maintenance. CONCLUSIONS: Guidelines for investigators, coding rules and medical validation appeared to be mandatory for a successful implementation of MedDRA. The use of MedDRA, with the addition of the mentioned support tools, should ensure coding consistency and facilitate the clinical and tolerance analyses and meta-analyses in clinical trials.

Effect of zidovudine resistance mutations on virologic response to treatment with zidovudine or stavudine, each in combination with lamivudine and indinavir.

The authors studied the effect of zidovudine (ZDV) resistance mutation on virologic response to treatment with ZDV or stavudine (d4T) each in combination with lamivudine and indinavir. Viral genotyping was performed on plasma HIV-1 RNA at study entry and concerned 155 patients previously treated with ZDV, didanosine, or zalcitabine and enrolled in the NOVAVIR (Agence National de Recherche sur le SIDA [ANRS] 073) trial. Three virologic responses were investigated: early response (<50 copies/mL at week 24), late response (<500 copies/mL at week 80), and virologic failure (two HIV-1 RNA >5000 copies/mL). Patients were classified as resistant or susceptible to ZDV according to the ANRS algorithm. Plasma viral RNA from 123 of 155 patients had two or more ZDV resistance mutations. The number of ZDV resistance mutations was positively correlated with the duration of prior antiviral therapy (p <.001). At week 24, 74% and 77% of patients with virus classified as resistant were responders in the d4T and ZDV arm, respectively. Similar results were found at week 80. Virologic failure was reached in 7 of 24 patients with virus classified as susceptible and in 26 of 131 patients with resistant virus (p =.29). In the ZDV arm, patients classified as resistant had longer times to virologic failure than those classified as susceptible (p =.003). In conclusion, sustained virologic response despite presence of ZDV resistance mutations implies that these mutations do not preclude an early and durable response to treatment with a potent three-drug regimen in these patients. Patients susceptible to ZDV had lower median mean corpuscular volumes and lower random indinavir levels, suggesting that adherence was the main reason for failure.

Increased risk of lipoatrophy under stavudine in HIV-1-infected patients: results of a substudy from a comparative trial.

OBJECTIVES: To compare the incidence of clinical lipodystrophy in HIV-1-infected patients receiving zidovudine or stavudine, in combination with indinavir and lamivudine, in a randomized trial. METHODS: NOVAVIR was a randomized multicentre trial comparing stavudine/lamivudine/indinavir and zidovudine/lamivudine/indinavir in 170 patients pretreated with zidovudine, didanosine or zalcitabine (> 6 months), but naive for lamivudine, stavudine and protease inhibitors. The incidence of clinical lipodystrophy and metabolic abnormalities was assessed in a subgroup of 101 patients after 30 months of follow-up. RESULTS: The incidence of lipoatrophy was increased in the stavudine arm versus the zidovudine arm, as followed: facial atrophy: 48 versus 22% of patients, P = 0.011, lower limb atrophy: 49 versus 22% of patients, P = 0.006, buttock atrophy: 47 versus 20% of patients, P = 0.009, venomegaly: 57 versus 24% of patients, P = 0.001. There was no significant difference in the incidence of clinical signs of central fat accumulation nor in fasting metabolic parameters at month 30 between the two arms. In multivariate analyses, the stavudine arm, previous therapy with didanosine, and a lower CD4 cell count at study entry were associated with an increased risk of lipoatrophy, whereas older patients and women had an increased risk of lipohypertrophy. CONCLUSION: Patients receiving stavudine/lamivudine/indinavir had a greater rate of clinical lipodystrophy, mainly lipoatrophy, than those treated with zidovudine/lamivudine/indinavir.

Final analysis of the Trilège induction-maintenance trial: results at 18 months.

BACKGROUND: First results of Trilège demonstrated that the strategy of less intensive antiviral therapy is less effective than continuation of triple-drug therapy. OBJECTIVE: To compare the final number of failures at month 18 and to study viral dynamics in patients experiencing a virological failure. DESIGN: Longitudinal follow-up from a randomized controlled trial. SETTING: Forty-three AIDS clinical-trial units. PATIENTS: A total of 279 HIV-1 infected adults randomized in Trilège. MEASUREMENTS: Analysis of recurrent values of HIV RNA > 500 copies/ml beyond time to virologic failure. RESULTS: A total of 83 patients experienced virological failure by month 18; 10 in the zidovudine (ZDV) + lamivudine (3TC) + indinavir (IDV) arm, 46 in the ZDV + 3TC arm, and 27 in the ZDV + IDV arm, confirming previous results. Whatever the treatment ultimately received, 87% of patients had an HIV RNA < 500 copies/ml at month 18 with no statistical difference between randomized arms. Patients experiencing a failure in the triple-drug regimen had a greater tendency to maintain HIV RNA > 500 copies/ml beyond the time of virological failure than patients in both less intensive treatment groups who experienced failure. Lower levels of HIV RNA at failure and reinitiating of either the original triple-drug regimen or a new combination of nucleoside analogue reverse transcriptase inhibitors and protease inhibitors were associated with lower hazard ratios for developing recurrent HIV RNA > 500 copies/ml. CONCLUSION: Results confirmed the failure of a less intensive regimen to maintain patients with a viral suppression (HIV RNA < 500 copies/ml). Although there is a lower incidence of failure in the triple-drug regimen, randomization to a less intensive regimen of ZDV + 3TC or ZDV + IDV was not detrimental, as treatment modification, either to the original triple regimen, or a different regimen was successful.