Short-term safety and tolerability of didanosine combined with high- versus low-dose tenofovir disproxil fumarate in ambulatory HIV-1-infected persons

Coadministration of didanosine (ddI) and tenofovir (TDF) results in increased ddI serum concentrations, which may lead to increased risk of ddI-associated toxicities. To evaluate the safety and tolerability of ddI/TDF, we performed a retrospective cohort analysis of patients seen in the HIV Outpatient Study, an ongoing dynamic cohort study of HIV-infected persons in clinical care. Study subjects were those who received at least 14 days of combined ddI/TDF before October 2003. Of 260 subjects who received ddI/TDF-based antiretroviral therapy, 155 (60%) received high-dose ddI (400 mg daily dose) and 105 (40%) received low-dose ddI (100-250 mg daily). Forty-two of the high-dose ddI recipients were later switched to low-dose ddI. The median time of observation for those on high-dose ddI only was 5 months, high-dose ddI switched to low-dose ddI was 16 months, and low-dose ddI only was 5 months (p < 0.05). Discontinuations because of toxicity were more frequent on high-dose ddI regimens (34/155, 22%) than on low-dose ddI regimens (9/105, 9%) (unadjusted odds ratio [OR(unadj)] 3.0, 95% confidence interval [95% CI] 1.30-7.09; p = 0.007). Among subjects without preexisting peripheral neuropathy, 12 (12%) of 101 subjects ever on high-dose ddI regimens had treatment-emergent peripheral neuropathy compared to 2 (4%) of 55 subjects on low-dose ddI regimens (OR(unadj) 3.57; 95% CI, 0.72-24.1; p = 0.14). Among patients without a history of pancreatitis, 6 (4%) of 153 subjects developed pancreatitis after starting high-dose ddI regimens, compared to none of the 103 subjects on low-dose ddI regimens (OR(adj) and 95% CIs undefined; p = 0.08). Severe laboratory abnormalities of creatinine, phosphorous, and bicarbonate were not different between the groups. A summary variable for any event--discontinuation for toxicity, treatment- emergent adverse event or abnormal laboratory values--indicated that 44 (28%) of 155 of those on high-dose ddI versus 13 (12%) of 105 on low-dose ddI developed any event (OR(unadj) 2.81; 95% CI, 1.36-5.86; p = 0.004). In conclusion, high-dose ddI/TDF-based therapy was more frequently associated with drug-related toxicity, adverse events, and treatment discontinuation than low-dose ddI/TDF regimens; low-dose ddI with TDF was generally well tolerated in these HIV-infected persons.