Is switching protease inhibitor-based effective antiretroviral therapy safe in patients with AIDS-associated Kaposi's sarcoma?

Abstract

The successes of highly active antiretroviral therapy (HAART) based on HIV protease inhibitors (PI) have been followed by concerns caused by the long-term adverse effects of these compounds. The discontinuation or modification of PI-including HAART is reported in up to 59% (median 33%) of patients, mainly because of toxicity, intolerance or poor adherence to complicated regimens, with a subsequent failure to control HIV replication [1]. A growing number of physicians are thus currently oriented towards the use of PI-sparing antiretroviral drug regimens that include combinations of nucleoside reverse transcriptase inhibitors and non-nucleoside reverse transcriptase inhibitors (NNRTI). In patients who have maintained durable HIV suppression while under a PI-based HAART regimen, the substitution of the PI with an NNRTI or abacavir was shown to be associated with the maintenance of adequate control of HIV replication, while leading to improvements in adherence and the reversal of some PI-related toxicities [2]. However, the effect of switching to a PI-sparing regimen in patients with an ongoing or past history of AIDS-associated Kaposi's sarcoma (KS) is still unclear. The regression of KS is preponderantly, but not exclusively, reported in patients treated with PI-including antiretroviral drug combinations [3–6]. This discrepancy may simply be due to the fact that the use of PI has temporally antedated the introduction of NNRTI as part of HAART regimens. However, PI have been found to promote KS regression in vitro by means of direct anti-angiogenic, anti-tumour and anti-KS effects [7]. The recent description by Bani-Sadr et al. [8] of KS clinical relapse in five HIV-infected patients switching from a PI to a NNRTI-based HAART regimen might lend support to the hypothesis that, irrespective of their potency in suppressing HIV replication and thus inducing immune reconstitution in the host, PI are superior to other antiretroviral drug classes in controlling AIDS-associated KS by virtue of their unique in-vitro effects. We present here the long-term outcome of eight patients with AIDS-associated KS who switched to a PI-sparing combination, without any adverse effect in terms of tumour relapse (see Table 1 for details). These individuals were part of a consecutive series of 24 patients who were started on first-line PI-based antiretroviral therapy at the Institute of Infectious Diseases of the University of Milan in the years 1998–2000 after a histologically proved diagnosis of KS.Table 1: Clinical and laboratory features of eight patients with AIDS-associated Kaposi's sarcoma who switched to a protease inhibitor-sparing regimen after achieving tumour regression.The reasons underlying the switch were antiretroviral simplification (patients 1, 2, 3 and 4), toxicity (patients 5 and 6), failure to suppress HIV replication (patient 7) and poor adherence (patient 8). When antiretroviral therapy was changed, in all patients but one (patient 5), KS had been in complete remission (CR) for a median of 18 months (range 9–27 months). The median CD4 T-cell count was 585 cells/mm3 and the plasma viral load was below 50 copies/ml in all subjects but two (patients 7 and 8). After switching, all eight patients maintained stable virological and immunological parameters over a median follow-up of 32.5 months (range 18–42). KS did not recur in the seven individuals who were in CR at the time of HAART simplification, and it improved further to CR in the single patient (patient 5) who had achieved a partial response (PR) while under the preceding PI-containing regimen. Our data on the absence of KS relapse after switching to a PI-sparing regimen contrast with those reported by Bani-Sadr et al. [8]. This discrepancy may be due to the relatively favourable prognosis of our patients, as only one of them (patient 6) suffered from advanced HIV disease (i.e. CD4 cell count < 200 cells/mm3 and concurrent opportunistic diseases) at KS presentation. In addition, six out of eight patients in our series experienced a sustained suppression of HIV replication throughouhout the follow-up. It therefore seems more probable that patients with poor prognosis AIDS-associated KS and a history of virological failure, such as those reported by Bani-Sadr et al. [8] may be more prone to tumour relapse, regardless of the type of antiretroviral regimen. A recent study found both PI and NNRTI-based regimens equally effective in reducing the incidence of AIDS-associated KS in a large cohort of HIV-infected at-risk individuals [9]. Our experience indicates that antiretroviral drug combinations may be safely simplified towards PI-sparing regimens even in patients with a previous history of AIDS-associated KS.