|Home | About | Journals | Submit | Contact Us | Français|
In contrast to HIV-1 infection, little is known about the natural disease course of HIV-2 and response to ARVs. We describe a cohort of HIV-2 infected patients, focusing on method of diagnosis, ARV treatment and complications.
Through a retrospective review of medical records at our center, we identified 12 cases of HIV-2 infection in our clinic population (1400 active patients) who received care between 2002 and 2011. We summarized clinical characteristics, ARV treatment and outcomes.
Seven cases were male and five were female. All patients were born in West African countries. The mode of transmission was heterosexual intercourse in eleven patients, and intravenous drug use in one patient. The median CD4 count at time of diagnosis was 668/ml (range23/ml to 1,546/ml). HIV-2 quantitative viral load measurements were not uniformly available to clinicians. Four patients were treated with protease inhibitor-based regimens, with a mean increase in CD4 count of 183/ml (range 43/ml to 341/ml). The other eight patients have been observed off ARVs. Two patients experienced complications from HIV, one patient had HIV encephalopathy and molluscom contagiosum, and another had microsporidiosis infection in the setting of AIDS.
Our results support previous studies that HIV-2 has a more indolent disease course than HIV-1, with a spectrum of disease ranging from asymptomatic to AIDS. Development of a reliable quantitative HIV-2 viral load assay to guide management is needed. Further research studies are needed to establish the best time to start ARV treatment in HIV-2 infected patients.
While treatment of human immunodeficiency virus (HIV) type 1 is well characterized in the medical literature, there is far less experience treating persons infected with the HIV-2 virus. HIV-2 is endemic to western Africa, but cases have been described throughout the developed world, including North America. A Portuguese study found that 30% of HIV-2 transmission occurred within Portugal; however most of these infections were thought to have occurred in socially isolated west-African communities within the country1. HIV-2 in North American populations has only recently been described2, 3. HIV-2 progresses to symptomatic disease at a much slower rate than does HIV-14; therefore, it is not clear when to initiate antiretroviral treatment (ART) with the goal of achieving immune restoration while minimizing toxicity5.
HIV-2 shares 50-60% nucleotide homology with HIV-1, but differs in key sequences that confer natural resistance to some antiretrovirals designed to treat HIV-1 infection6. HIV-2 resistance has been described with all classes of antiretroviral medications. In vitro studies show that resistance is mediated by naturally occurring polymorphisms or by selective pressure via exposure to antiretroviral medications6. Nucleoside reverse transcriptase inhibitors (NRTI) and protease inhibitors (PI) appear to be effective at lowering HIV-2 viral load, and have become an integral part of most treatment regimens5, 7, 8, 9. Non-nucleoside reverse transcriptase inhibitors (NNRTI) are not effective against HIV-2 due to resistance conferred by the Y188L polymorphism, which is naturally occurring in all known strains of HIV-2. Data is not available on newer classes of ART. Only one case report of successful treatment of HIV-2 with maraviroc in combination with raltegravir has been reported10. Entry and fusion inhibitors such as maraviroc and enfuvirtide, pose a theoretical risk of resistance because HIV-2 uses a wider variety of co-receptors for entry than does HIV-16. Foscarnet, more commonly used in the treatment of herpes viruses, has been used as salvage therapy for resistant HIV-2 infection11.
Providence, Rhode Island is home to several west-African immigrant communities. Our HIV Center at the Miriam Hospital has provided longitudinal treatment for a number of patients with HIV-2 infection. To summarize our experiences with treating HIV-2 infected patients, we performed a retrospective case series, with particular focus on method of diagnosis, ART treatment, and complications including opportunistic infections.
Through a retrospective review of medical records, we identified all cases of HIV-2 infection in our clinic population (approximately 1400 active patients in 2011) who received care between 2002 and 2011. Patients were eligible if they received care at our Center and met the CDC definition of HIV-2 infection. To meet the CDC diagnostic criteria for HIV-2 infection, patients must have met one of three criteria: 1) HIV-1/HIV-2 type-differentiating antibody immunoassay positive for HIV-2 but not for HIV-1; 2) a positive HIV-2 nucleic acid test; or 3) a positive HIV-2 western blot and negative or indeterminate HIV-1 western blot. For all patients identified as having HIV-2 infection, demographic and clinical data were extracted from the paper and electronic medical records. We summarized clinical characteristics, use of ART and outcomes. To calculate average increase in CD4 count, the mean of all CD4 values after ART initiation was calculated and compared with the CD4 count immediately prior to starting ART treatment. The institutional review board at the Miriam Hospital approved this study.
Twelve cases of HIV-2 infection were identified in our study population and were reviewed (Table 1). Seven cases were male and five were female. The median age at diagnosis was 46 years (range 29-59 years). All patients emigrated from West African countries, specifically Cape Verde (n = 5), Ivory Coast (n = 3), Guinea-Bissau (n = 2), Ghana (n = 1), and Senegal (n = 1). The reported mode of transmission was heterosexual intercourse in eleven patients, and intravenous drug use in one patient. Eight cases were likely contracted in Western Africa (patients had known exposures to HIV-2 within the country of origin), one case was acquired in the U.S., and the location of infection was not known in three cases (no known exposure history to HIV-2). One case of HIV-2 was acquired within the United States as this patient lived in the U.S. since childhood and had a negative HIV test one year prior to testing positive for HIV-2. At the time of diagnosis, she was in a relationship with another patient infected with HIV-2.
In all cases, an HIV-1/2 EIA antibody test was positive and HIV-1 western blot was negative or indeterminate. Confirmatory testing with HIV-2 western blot was positive in all patients. Six patients had HIV-2 quantitative viral load measurements through a research lab or an HIV-2 qualitative viral load assay completed, but use of these tests was not uniform. The median CD4 count at time of diagnosis was 668/ml (range 23/ml to 1,546/ml). At initial presentation, two patients experienced complications from HIV. One patient had HIV encephalopathy and molluscom contagiosum, and another had microsporidiosis infection in the setting of AIDS. One patient experienced end-stage renal disease as a consequence of focal segmental glomerulosclerosis.
To date, four patients have received ART with protease inhibitor (PI)-based regimens. The mean increase in CD4 count after ART initiation was 183/ml (range 43/ml – 341/ml). Eight patients have been observed off ART and have not experienced any complications of HIV and their CD4 counts have remained stable. Among the eight patients observed off ART, there has been a mean decrease in CD4 count of 31/ml over a mean 34 months of follow-up. There were a total of three pregnancies among two female patients during the study period. Zidovudine was used as prophylaxis for one patient during both pregnancies and no prophylaxis was used during the other patient’s pregnancy because it was not deemed necessary by the patient’s provider. Among the three pregnancies, there was no maternal-fetal transmission of HIV-2. Over the 10-year study period, two patients were lost to follow-up or moved. The remaining 10 patients continue to be followed at our Center.
The patients in this study were epidemiologically similar to the New York City and CDC cohorts of HIV-2 patients published by Torian et al2, 3. Most patients in these reports contracted HIV-2 from heterosexual intercourse (91%) and acquired the virus in Western Africa (66-96%). In our cohort, 100% of the patients with HIV-2 were from sub-Saharan Africa and all but one reported contracting HIV-2 through heterosexual sex. One patient acquired HIV-2 while residing in the U.S. Similar to other HIV-2 cohorts, this study population demonstrates a variety of clinical manifestations including asymptomatic, HIV/AIDS and HIV with metabolic and renal disease similar to those seen in patients with HIV-1.
Four patients received treatment for HIV-2 infection. All received PI-based regimens, with increased CD4 counts and no adverse clinical outcomes from the treatment regimen. In 2010, two separate consensus statements on treatment of HIV-2 were released. World Health Organization guidelines recommended that triple NRTI regimens should be first line12, while British guidelines released later in the year recommend PI-based regimens as first line. The difference between these recommendations likely reflects the interval publication of a study comparing triple-NRTI regimens to PI-based regimens. This study showed greater increase in CD4 counts in the PI group14. It is still unknown whether early treatment (CD4 > 500) portends better outcomes in patients with HIV-2 infection. Among the eight patients in our population who were monitored off antiretroviral treatment, none have shown complications of HIV infection and only 2 patients (25%) have shown decline in CD4 count.
Currently, there is no standardized quantitative viral load assay commercially available for HIV-2. Because of this, providers are not able to accurately assess response to ART. When viral load measurements were made among our patients with either research or qualitative assays, results varied. In two patients, undetectable HIV-2 viral loads were present in the absence of antiretroviral treatment and these patients had prior detectable HIV-2 viral loads. Previous studies have described the lower viral burden associated with HIV-215, 16. Unfortunately, it remains unclear if the variability of viral load in these patients was a function of better viral control or an unreliable assay. Our findings support previous reports which emphasized that the absence of detectable HIV-2 viral load should not be used to rule out HIV-2 infection2, 15. The lack of a standardized and reliable quantitative assay for viral load remains an ongoing barrier for clinicians who currently manage HIV-2 infection and the development of such an assay should be a research priority. For now, clinicians need to rely on the CD4 count as a surrogate marker of viral control.
Two patients had uncomplicated pregnancies during the study period. The first patient received prophylaxis with AZT during both of her pregnancies, while the second patient did not receive antiretroviral prophylaxis during her pregnancy. Maternal-fetal transmission of HIV-2 is much less common than with HIV-1, and is directly proportional to viral load15. If viral load is < 50 copies, British guidelines do not recommend antiretroviral prophylaxis, however this recommendation is based on expert opinion (level IV evidence)16. There were no documented cases of maternal-fetal transmission of HIV-2 in the 29 female patients of the New York cohort, although the authors did not comment on the prophylaxis approach in individual pregnancies or the total number of pregnancies2.
This study was limited to patients in whom HIV-2 diagnosis was suspected and specific testing for HIV-2 was sought by the provider. Patients with HIV-1 and HIV-2 co-infection would most likely not have been identified under these circumstances. It is conceivable that HIV-2 diagnosis would be made in a co-infected patient of West African descent who failed to respond to NNRTI-based treatment thus prompting further workup, however no such cases have occurred to date in our clinic.
Our results are consistent with previous studies suggesting that HIV-2 has a more indolent disease course than HIV-1. Diagnosis of HIV-2 in the U.S. relies upon clinical suspicion and understanding the epidemiology of HIV-2. Thus, physicians who treat HIV infection must be educated and aware of the instances that necessitate further investigation into HIV-2. Development of a reliable quantitative HIV-2 viral load assay to confirm diagnosis and guide management is needed. Further investigation should also focus on treatment efficacy and resistance mechanisms of newer antiretroviral agents, including fusion inhibitors, integrase inhibitors and entry inhibitors. Finally, further research studies are needed to establish the best time to start ART treatment in HIV-2-infected patients.
Findings from this study were presented in part at the Annual Meeting of the Infectious Diseases Society of America, Boston, MA, October 21st, 2011.
Conflict of Interest and source of funding: C.B. has received funding by the Lifespan/Tufts/Brown Center for AIDS Research (P30AI42853), an NIH funded program. Neither C.B. nor B.H. has a conflict of interest to report.