Virologic Response Associated With a Change in Protease Inhibitor Therapy

In the report by Guardiola and colleagues,¹ a switch in protease inhibitor (PI) therapy for reasons other than therapeutic failure occurred in 74% of patients who were treated with PI-based regimens. The therapeutic strategy in these patients needs careful assessment before PI changes, since the patients may be put at greater risk for therapeutic failure after therapy change.

Figure 1 shows the virologic response in 14 patients who had a switch in PI after having achieved viral RNA suppression (viral RNA load <400 copies per milliliter) compared with 28 patients who continued to receive a successful regimen. Patients from 2 clinics were included in this retrospective chart review if they were treated with an indinavir sulfate regimen for at least 6 months and had 2 viral RNA measures below detection while receiving indinavir-combination therapy.² From this group, patients were further identified for the switch group if their PI therapy had changed from indinavir to nelfinavir mesylate and if they had at least 1 viral RNA measurement after the PI switch (t₀). Patients who continued to receive indinavir regimens constituted the comparison group. Two comparison patients per nelfinavir switch patient were randomly selected and matched for status of PI use prior to indinavir treatment and start date of indinavir treatment. Each comparison patient was assigned a date that allowed a time receiving indinavir equal to the time receiving indinavir of the matched patient who switched to nelfinavir. Patients in both groups had to have evidence of viral RNA suppression in the 3 months prior to t₀. In the nelfinavir switch patients without a documented reason for PI change, the patients also had to have documented undetectable viral RNA (viral RNA load <400 copies per milliliter) within 4 weeks after the switch, indicating that the PI change was not a result of therapeutic failure. In Kaplan-Meier analysis, the last viral RNA measure available for each patient was defined as censored if viral RNA was undetectable and no further observations were available or as indicative of failure if the viral RNA load was greater than 400 copies per milliliter.