Box 4 | As good as it gets?

Standard combination regimens for treatment-naive patients are capable of producing optimal virologic benefits (plasma HIV RNA maintained at <50 copies per ml) in nearly 90% of patients at or beyond 48 weeks⁴⁶. The most rigorous analyses of such data centre on intention-to-treat, and count as failures patients who drop out or modify treatment. Given the nature of clinical trial participation, it is unlikely that one could design a trial with fewer than 10% drop-outs or regimen modifications during the first year of therapy. It may not be possible then to develop an initial antiretroviral regimen with better than 90% virologic responses after 1 year. As efficacy may have reached its asymptotic maximum, new antiretroviral regimens for treatment-naive patients will be expected to have equivalent efficacy to existing regimens, but better convenience and tolerability.

Response rates in treatment-experienced patients are currently much lower, with optimal virologic benefits obtained in only 40–60% of patients during the first year of a new regimen³. With the availability of additional potent drugs, especially those in new classes such as HIV integrase inhibitors and entry inhibitors, outcomes in treatment-experienced patients should approach those achieved in treatment-naive patients. Regimens unable to reach and sustain such treatment goals are likely to fall out of favour.