Research Article: Risk of HIV Infection and Lethality Are Decreased in CCR5del32  Heterozygotes: Focus Nosocomial Infection Study and Meta-analysis

Date Published: , 2012

Publisher: A.I. Gordeyev

Author(s): S.A. Borinskaya, Zh.M. Kozhekbaeva, A.V. Zalesov, E.V. Olseeva, A.R. Maksimov, S.I. Kutsev, M.M.  Garaev, A.V. Rubanovich, N.K. Yankovsky.



CCR5del32 Homozygous deletion in the chemokine receptor R5 gene
provides almost complete protection to individuals against HIV infection.
However, data relating to the protective effect forCCR5del32
heterozygous individuals have been contradictory. The frequency of
theCCR5del32allele in population control cohorts was
compared with that of a group of children (27 Kalmyks and 50 Russians) infected
by G-subtype HIV-1 in a nosocomial outbreak. The frequency of
theCCR5del32allele was shown to be lower among the infected
children in comparison with that of the control group; however, the difference
was small and statistically insignificant. Similar results were obtained in a
number of earlier studies. The insignificance of the small differences could be
a result of one of two reasons. (i) The fact that there is no protective effect
of the heterozygous state, and that the phenomenon depends only on the
fluctuation of allele frequencies. In this case, there would be no differences
even if the infected cohort is enlarged. (ii)The protective effect of the
heterozygous state is real; however, the size of the studied cohort is
insufficient to demonstrate it. In order to discern between these two reasons, a
meta-analysis of data from 25 published articles (a total of 5,963 HIV-infected
individuals and 5,048 individuals in the control group, including the
authors’ own data) was undertaken. A conclusion was drawn from the
meta-analysis that theCCR5del32 allele protects individuals
against the HIV infection even in a heterozygous state
(OR=1.22, 95%CI=1.10–1.36). The risk of
HIV infection forCCR5 wt/del32 heterozygotes was lower by at
least 13% as compared to that for wild typeCCR5
wt/wthomozygotes. Prior to this study, no data of
the type or any conclusions had been published for Caucasians. The mortality
rate in the 15 years following the infection was found to be approximately 40%
lower forCCR5del32 heterozygotes in comparison with that for
the wild type homozygotes in the studied group. The size of the studied group
was insufficient to claim difference validity
(OR=2.0;p= 0.705), even though the effect
quantitatively matched the published data. The features of the meta-analysis
influencing the threshold level and the statistical validity of the effects are
being discussed. The level of theCCR5del32 protective effect on
the chances to be infected with HIV and on the outcome of the HIV infection was
assessed for various ethnic groups.

Partial Text

Having started with a single case detected in 1981, the AIDS epidemic is now one of
the most important health care issues both in Russia and the rest of the world
[1]. The evolution of the epidemic in
Russia was characterized by the formation of clusters of nosocomial infection that
took place in 1988–1989. The outbreak of the infection began with a
HIV-infected child at a children hospital in Elista. Antiepidemic measures were not
taken, which resulted in the spread of the epidemic throughout medical institutions
in Kalmykia, the Rostov and Volgograd districts, as well as Stavropol Territory. A
single focus (focus of infection) was responsible for infecting more than 260
children and their mothers [2, 3]. Many of them have long passed away (
Fig. 1 ).

Blood samples from the collection of the Biotechnology Laboratory (Ivanovsky
Institute of Virology, Russian Academy of Medical Sciences) were used in this study.
The samples were obtained as a result of planned medical examinations of individuals
with nosocomial HIV infection, during the period spanning 1991–2007. Consent
letters were obtained from the parents of each of the examined children giving
permission to use some of the samples obtained for research purposes. This sample of
HIV-infected patients is unique, since there is no variability of infection
development associated with differences in viral strains. All of the patients were
infected with the same viral strain (HIV-1 subtype G) originating from the original
child that had been infected (focus nosocomial infection) [22, 23]. Furthermore,
most patients belonged to two ethnic groups (Russians and Kalmyks), thereby reducing
the possible influence of genetic heterogeneity in each cohort. Anonymous data on
patients’ birth dates and death dates in cases of fatal outcomes were obtained
for 107 HIV-infected patients in the Rostov district (all Russians) and 60
HIV-infected patients from Elista (47 Kalmyks and 13 Russians). Blood samples of
HIV-infected children (50 Russians and 27 Kalmyks, age varied from less than 1 year
to 16 years; median age 2.7 years) were used for the study. Blood samples taken from
healthy volunteers were used as control samples. The first control group consisted
of students of the Rostov State Medical University (the majority of whom were born
in 1986–1990). According to the results of the survey, they were
second-generation Russians and were born in the Rostov district. The second control
group consisted of Kalmyks living in Elista (ethnicity was established based on the
survey results). Blood samples were collected in full compliance with the
informed-consent procedure. The genetic study project obtained approval from the
Ethics Committee of the Institute of the Institute of General Genetics, Russian
Academy of Sciences.

Allele and genotype frequencies in HIV-infected patients and in the control

The effect of the heterozygous bearing of the CCR5del32 deletion
allele on the risk of HIV infection in populations of European origin (without
allowance for the route of infection transmission, viral serotype, and differences
in antiretroviral therapy) was first assessed based on the meta-analysis of
published data. The protective effect observed was rather small (
OR  = 1.22) but was statistically significant and corresponded
to a decrease of at least 13% in the risk of infection in the individuals with the
CCR5wt/del32 genotype, according to the calculations. The small
OR value explains why the differences between the frequencies
of genotypes and/or CCR5del32 allele between the groups of HIV-infected individuals
and the control group detected in most articles are statistically insignificant.