William R. Bishai, M.D., Ph.D. and Richard E. Chaisson, M.D.
The Hopkins HIV Report - March 1999
|
While the focus of the Chicago conference heavily emphasized new HIV therapies and their complications, opportunistic infections continued to receive attention. The conference emphasized the dramatic reduction of most opportunistic infections in the HAART era, and mounting evidence was presented to support the cessation of OI prophylaxis in patients responding to HAART. Palella and colleagues in the HIV Outpatient Study chronicled the change in OI incidence from 33 per 100 person-years in 1994 to 5 per 100 p-y in the first quarter of 1998 as the use of antiretroviral therapy increased from 15% to greater than 95% over the same interval [Abstract 689]. Interestingly, high usage of prophylaxis against PCP (greater than 93%) and M. avium (54-58%) remained in place through the first quarter of 1998 [Abstract 691]. Richard Moore and colleagues examined the subpopulations who are benefiting from the decrease in OI incidence by evaluating data from the Johns Hopkins HIV clinic between 1995 and 1998. Among AIDS patients who acquired HIV through homosexual contact there was an 81% reduction in OI incidence (107 to 20 events per 100 person years); among heterosexually transmitted HIV cases there was a 60% reduction (79 to 28 cases per 100 person years), and among injection drug users there was a 45% reduction (55 to 30 cases per 100 person years). Hence, gay and bisexual men have benefited the most by the HAART-induced reduction in OI's, while injection drug users have benefited the least. Differences in adherence were thought to explain some of this variation. In a related study, Sullivan and colleagues found a 60% decrease in the incidence of bacterial pneumonia between 1992-98, with protease inhibitor therapy being the strongest predictor of decreasing risk [Abstract 693].
Two interesting studies raised the question of whether steroids are really beneficial in the treatment of moderate to severe Pneumocystis carinii pneumonia (PCP). A study from Northwestern University evaluating PCP cases from 66 hospitals across seven states found that among the 735 patients who met CDC guidelines for adjunctive steroid use, there was three fold higher mortality among steroid recipients (18% vs. 6%, p=0.02) [McIlraith, et al., Abstract 697]. Steroids were less frequently used in New York, Miami, and Los Angeles and more frequently used in Chicago and Seattle. A related study from five cities and fifty-five hospitals found that drug users had a lower PCP mortality (8.3%) compared with patients with sexually acquired HIV (13% PCP mortality, p <0.001) [Nwadiaro, et al., Abstract 696]. Importantly, drug users were also less likely to receive steroids. While there are concerns about these data (drug users were also more likely to leave therapy against medical advice), these studies suggest a disparity between the benefits of steroids in day to day practice and that seen in controlled clinical trials. It is likely, however, that steroids were given to the sickest patients and withheld from less sick patients.
The question of whether PCP prophylaxis can be safely discontinued was addressed by a study of 323 Spanish patients whose mean CD4 count increased from 106 to 368 cells/mm3 on HAART [Lopez, et al., Abstract LB7]. Over a mean follow-up of 6.6 months there were no episodes of PCP among the 171 patients who stopped PCP prophylaxis. These data support the withdrawal of PCP prophylaxis in patients responding to HAART with CD4 increases surpassing 200 cells/mm3.
The possibility of drug-resistant P. carinii was raised in a study by Huang and colleagues. They examined P. carinii from patients with PCP and amplified sequences from the organism's dihydropteroate sythetase (DHPS) gene which encodes an enzyme that is the site of sulfa and dapsone activity. Overall, more than half of patients had DHPS polymorphisms, though their relationship to resistance is unknown. Increasing drug exposurewas associated with an increasing prevalence of DHPS polymorphisms, and there was a weak trend towards poorer outcomes in patients whose P. carinii lacked wild-type DHPS. The study raises many interesting questions about possible resistance but does not affect current treatment or prophylaxis guidelines [Huang, et al., Abstract 244].
Constance Benson summarized the results of ACTG 223, a phase 2/3 open-label study comparing the efficacy of three-drug MAC treatment with clarithromycin (C)/ethambutol (E)/rifabutin (R) to two-drug combinations, (C + E and C + R). 160 patients were followed for 48 weeks [Abstract 249]. Using microbiologic clearance or clinical and microbiologic improvement as endpoints, there were no significant differences between C + E, C + R, or the three-drug C + E + R regimen. However, there was a statistically significant survival benefit when C + E + R was compared with the two 2-drug regimens. A significant incidence of uveitis occurred in patients taking C + R because of the elevation of rifabutin levels by clarithromycin. In spite of the survival benefit of three-drug therapy, the adverse drug reactions and the difficulties of using rifabutin with protease inhibitors and NNRTIs may lead many clinicians to continue the use of the popular 2-drug clarithromycin-ethambutol combination for the primary treatment of MAC disease.
Cohn and colleagues evaluated patients in ACTG 362 for their adherence to azithromycin 1200 mg weekly for MAC prevention [Abstract 444]. Nine percent of patients missed doses after 8 weeks of therapy, and 12% of patients missed doses at 24 weeks into treatment. Hence, even the simplicity of once weekly preventive regimens does not guarantee strict adherence. Adherence was associated with health beliefs, and patients on methadone had poorer adherence.
Gordin and colleagues reported on the CPCRA/ATCG/PAHO study of tuberculosis prevention among 1583 HIV-infected, tuberculin-positive patients and evaluated risk factors for the development of active tuberculosis [Abstracts 248 & 448]. Among this group there were 45 cases of active tuberculosis over a mean 37 month follow-up time. Risk factors for active disease were a CD4 count less than 200 cells/mm3, a PPD reaction greater than 15 mm in diameter, and the use of antiretroviral therapy (which is presumably a marker of advanced HIV disease). These data suggest increased vigilance is warranted in HIV-infected patients with large tuberculin skin test reactions. Diane Havlir presented data on immune responses to MAC in patients with HIV [Abstract 248]. Compared to healthy controls, patients with HIV had much weaker cellular responses to MAC. After starting HAART, however, proliferative and other responses improved significantly.
Studies at the conference presented further evidence that cessation of maintenance therapy for CMV retinitis is safe in patients responding to HAART. A Spanish study of 16 CMV patients who initiated HAART therapy and sustained CD4 cell increases to greater than 150 showed that at 25 months of follow-up all patients who stopped preventive therapy (n=7) failed to develop recurrent CMV [Tural, et al., Abstract 455]. A French study involving 48 patients with previous CMV retinitis on HAART therapy who then discontinued CMV maintenance therapy found one relapse of CMV retinitis and one relapse of CMV associated neuropathy after 38 weeks of follow-up [Jouan, et al., Abstract 456]. The patients who relapsed were negative for CMV viremia and had stable CD4 cell counts. Torriani reported on 17 patients from San Diego with CD4 cell counts greater than 70 on antiretroviral therapy who were followed for 25 months off CMV maintenance therapy [Abstract 250]. In this group there were five relapses of CMV disease, all in patients with CD4 counts <50 cells/mm3. The investigators were able to show that patients with relapses had reduced lymphoproliferative assay (LPA) responses to CMV antigens. Hence, loss of LPA responsiveness may serve as an early marker for risk of CMV reactivation in a failing HAART regimen.
A number of investigators reported on the pathogenesis of (KS) and its association with HHV8. Two research tests are currently available for identifying HHV8 infection: the LANA serologic test (for the HHV8-encoded latency associated nuclear antigen) and HHV8-based PCR (detecting HHV8 viremia). Fitzpatrick and colleagues from Denver compared LANA to PCR among 216 HIV-infected patients in a cross-sectional analysis [Abstract 247]. They found that HHV8 viremia is uncommon in patients coinfected with HIV and HHV8, and that the HHV8 PCR offered no advantages over the LANA serologic test which has been shown to correlate well with risk of KS. Robert Yarchoan from the NIH summarized the current therapeutic options for HHV8-associated KS [Abstract S17]. The current standard of care has changed significantly from 1994. The use of liposome encapsulated anthracyclines (Doxil and Daunosome) has shown significant advantages over standard ABV (adriamycin, bleomycin, vincristine) chemotherapy. Paclitaxel (Taxol) has also proven to be efficacious in KS. A number of new agents are also being evaluated against HHV8-associated KS. These include topical 9-cis-retinoic acid, anti-angiogenesis agents such as thalidomide and IL-12, and a factor associated with certain preparations of human chorionic gonadotropin (hCG). Therapeutic options for the treatment of KS are likely to expand further in the coming years. It is interesting to note that antiviral therapy with foscarnet, cidofovir or gancyclovir, all active in vitro against HHV-8, was not mentioned [Abstract S17].
Opportunistic Infections in Developing Countries
Alison Grant from the London School of Hygiene and Tropical Medicine discussed opportunistic infection prevention in developing countries [Abstract S31]. Tuberculosis preventive regimens have been proved effective in HIV-seropositive patients in developing countries, and the WHO is promoting broader implementation of screening and prophylaxis. She reviewed data indicating that trimethoprim-sulfamethoxazole (TMP/SMX) has been shown to reduce morbidity and mortality in West African patients with either newly discovered HIV infection or HIV-related tuberculosis. However, it remains to be seen whether widespread use of TMP/SMX will alter antibiotic resistance patterns among common bacterial pathogens. Twenty four-valent pneumococcal polysaccharide vaccines were recently shown to have no protective effect among HIV-infected Ugandans. Fungal infections are rampant in developing countries, but prophylactic antifungal drugs remain unaffordable at present. She concluded that tuberculosis and bacterial infection prophylaxis offer the greatest hope for prolonging life in patients with HIV in developing countries given the expense of antiretroviral therapies.
In general, the quality of the research presented in the Opportunistic Infections sessions was superior, with an array of pathogenetically oriented clinical investigations. The insights gained from these studies will be of great value in developing further management guidelines for opportunistic infections in patients on HAART, as well as in dealing with these complications in patients who lack access to, or who are not responding to, potent antiretroviral therapy.
990301
JH990303
©1999. The Johns Hopkins University AIDS Service, Division of Infectious Diseases. Permission to use and reproduce portions of this newsletter is hereby granted provided that author and publication are fully credited and both copyright and permission notice appear with reprinted material. Inquiries may be directed to Sharon McAvinue, Managing Editor. Website: Johns Hopkins AIDS Service.
The original of this article can be found at http://www.hopkins-aids.edu/publications/report/mar99_3.html
This information is designed to support, not replace, the relationship that exists between you and your doctor.