Many abstracts on opportunistic infections presented at this conference were "traditional" reports concerning new therapeutic data. There was also information presented on the natural history of opportunistic infections in the current era of highly potent antiretroviral therapy as well as the impact of opportunistic infections on viral load. The selections below highlight presentations from these three areas.
Therapy of Selected Opportunistic Infections
ACTG 268 [abstract 2] evaluated the gradual initiation of trimethoprim-sulfamethoxazole (TMP/SMX) for PCP prophylaxis. In patients who began TMP/SMX with 2 weeks of gradually increasing doses (up to a daily dose equivalent to a double-strength tablet), there were fewer toxicities than in patients who began therapy immediately with double strength tablets. All patients then continued therapy with double strength tablets. At week 12, only 66% of those who began therapy at full dose were still on treatment, compared with 84% of those who began therapy with gradual initiation. The implications of this study for clinical practice are not entirely clear for several reasons:
1. the added complexity of dose escalation may be more difficult outside of a controlled study.
2. longer term differences were not evaluated (e.g. is there still a difference in the percentage of patients on therapy at 6 months?)
3. it cannot be determined from this study whether dose escalation would still cause less short-term toxicity compared with the lower doses of TMP/SMX frequently used (one double strength tablet 3 times a week, or one single-strength tablet per day). The dose escalation schedule used in this study was TMP/SMX suspension 1 ml x 3 days, then 2 ml x 3 days, 5 ml x 3 days, 10 ml x 3 days, and 20 ml until the follow-up visit at week two.
ACTG 192 [abstract 4] was a double-blind, placebo-controlled trial of paromomycin for the treatment of cryptosporidiosis. This study found no difference in the rates of partial or complete responses in the group treated with paromomycin compared with the placebo group. There was also no difference between groups for any other clinical parameter (e.g. weight change, use of agents for symptomatic relief of diarrhea). The authors concluded that treatment of cryptosporidiosis with paromomycin results in no clinical benefit. Research continues into other potential therapies for this infection. Nitazoxanide is currently being evaluated in ACTG 336 and is available through an open-label program (see Product Information).
A study from the California Collaborative Treatment Group [abstract 5] evaluated the use of high dose fluconazole with or without 5FC for the treatment of cryptococcal meningitis. Response rates improved with higher doses of fluconazole. At 10 weeks, 62% of patients receiving 1,600 or 2,000 mg/d responded compared with 11% among those receiving 800 mg/d. At each fluconazole dose there was also an improved response rate in patients who received concomitant therapy with 5FC. The clinical response rate was 87% in patients receiving 2,000 mg/d of fluconazole and 150 mg/kg/d of 5FC. Most observed toxicities were those expected with these medications: nausea and vomiting with fluconazole, neutropenia and thrombo-cytopenia with 5FC. The results of this trial may lead to a comparative study of high-dose fluconazole + 5FC with current standard therapy, amphotericin B + 5FC. The mortality in this study (15%), however, was higher than is seen with amphotericin B (3-7%).
Viral Load Changes During Acute Opportunistic Infections
- Bush et al [abstract 1] evaluated changes in HIV RNA during episodes of PCP. In a small study of 10 patients, all of whom were on consistent antiretroviral therapy throughout their illness, HIV RNA increased in all patients compared to baseline. HIV RNA levels fell toward pre-diagnosis levels after resolution of PCP, but CD4 counts showed an overall decline through the study period. Preliminary data were also presented from patients with bacterial pneumonia. All thirteen patients who were evaluated had increases in HIV RNA associated with the onset of bacterial pneumonia; 12 of these patients had decreases in RNA after treatment of pneumonia. No data on long-term changes in viral load were available from either study. Similar data were presented by Havlir et al [abstract 360] who described increases in HIV RNA immediately following the diagnosis of disseminated MAC. Although standard clinical practice no longer involves the discontinuation of antiretroviral therapy in the setting of acute illness, these data emphasize the importance of continuing therapy in the face of acute infectious complications.
Opportunistic Infections in the Era of Highly Active Antiretroviral Therapy
Kaspar and DuBois [abstract 358] compared the clinical course of severely immunocompromised patients with an effective CD4 response to combination therapy (stable CD4 increase to >50 cells/mm3) to that of patients who were refractory to such therapy (CD4 count remained below 50 cells/mm3). The patients who were refractory to therapy developed significantly more new opportunistic infections and had higher rates of hospitalization and death. These observations are relevant to current questions regarding the competence of re-expanded CD4 cells; these data suggest that some effective immune restoration does occur even in patients with severe CD4 cell depletion when beginning antiretroviral therapy. There were also several posters describing the resolution of opportunistic complications in patients taking combination/protease inhibitor therapy.
Benhamou et al [abstract 357] described patients with chronic cryptosporidiosis and microsporidiosis who showed clearance of these infections and clinical improvement in patients treated with triple therapy.
Tashima et al [abstract 355] and Henry et al [abstract 356] described the improvement and/or resolution of a variety of opportunistic complications in patients treated with combinations including protease inhibitors.
In contrast, there were also reports suggesting a lack of restoration of immune competence with highly active antiretroviral therapy even in the setting of marked increases in CD4 count.
Gilquin et al [abstract 354] reported the occurrence of CMV infections in patients (n=5) on combination therapy despite high CD4 counts (median 150 cells/mm3).
Jacobsen et al [abstract 353] described the development of CMV retinitis in patients with CD4 counts >50 cells/mm3 (a finding not commonly seen prior to the advent of combination/protease inhibitor therapy). This observation suggests that the immunologic benefits of such therapy may not be as great as is indicated by the CD4 increases observed. The authors also highlight the fact that some of the patients who developed retinitis with relatively high absolute CD4 counts never showed a similar increase in CD4 percentage. This finding raises the question of whether CD4 percentage may be more useful as a clinical marker in this setting.
Other evidence suggesting a lack of immune reconstitution with protease inhibitor therapy, with implications for the development of opportunistic infections, was discussed by Connors et al [abstract 369]. This group showed that patients with CD4 counts <50 cells/mm3 had a preferential decrease in naive cells compared with memory cells. Indinavir treatment led to an increase in naive CD4 cells only if they were present prior to the initiation of therapy. Again, this finding suggests that CD4 count increases on potent antiretroviral therapy may not be associated with as much protection from opportunistic infections as would be expected. This question is being addressed with the development of clinical trials in which patients with high CD4 counts in response to antiretroviral therapy will have prophylaxis discontinued.