Clinical Overview of HIV Disease
Section 7: Laboratory Testing
HIV Antibody Testing
HIV infection is usually diagnosed by testing serum for antibodies to HIV
using a commercially available enzyme-linked immunosorbent assay (ELISA or EIA).
Because the ELISA test is not entirely specific, positive results are confirmed
with a Western blot assay, which identifies antibodies to specific components of
HIV.( 184,185 ). The 2-step process may mean that a patient must wait for a week
or more to receive test results.
ELISA is quite sensitive in chronic HIV infection (although decline in antibody
responses have been reported in advanced AIDS), but because antibody production
does not occur immediately upon infection, an infected individual may test ELISA
negative during a "window period" that varies in length from a few weeks to a
few months after infection, depending on the individual case and assay used.
Despite negative antibody testing during this window period, an individual may
have high viral load and be at high risk of transmitting infection.
Newer methodologies allow antibody testing on saliva ( 186,187 ) and urine (
187,188 ) specimens, although positive results should be confirmed with
serologic testing. Home testing methods are also available.( 189 ) Rapid HIV
serum testing, with results available in 3-30 minutes, has shown 99-100%
sensitivity and specificity compared to ELISA when tested in clinical settings,(
190 ) including in resource-poor settings ( 191-193 ) and in pooled specimens.(
194 ) In recent years, with the availability of rapid tests such as OraQuick
(Abbott Laboratories, Abbott Park, IL) and Reveal (MedMira, Halifax, Nova
Scotia, Canada), rapid testing protocols are being implemented in many
countries, and will likely become commonplace.
"Detuned" Antibody Testing
By decreasing the sensitivity of ELISA assays, relatively recent infection
(in which antibodies are present in lower concentrations and bind to HIV less
effectively) can be distinguished from established infection (in which
antibodies reach stable levels and have been selected for more avid binding to
HIV). Soon after infection (but after the window period) an individual will test
positive on the standard ELISA, but negative on the less-sensitive ("detuned")
test. After maturation of the antibody response, both tests will give a positive
result. Such "sensitive/less sensitive" or detuned ELISA testing strategies can
be used to identify individuals who are in the early months of HIV infection and
can help to identify incident infections in epidemiologic studies.( 195,196 )
CD4 Testing
The CD4 cell count in blood correlates with the risk of OIs in HIV disease,(
197,198 ) and is therefore a useful marker for HIV disease staging. CD4 count is
the main criterion for clinical decision making in guidelines developed in the
United States for the prophylaxis of OIs ( 27,199 ) and for HIV treatment.( 200
)
The CDC recommends CD4 testing every 3-6 months in all HIV-infected persons,(
201 ) but different intervals may be appropriate to the individual case. More
than 1.6 million CD4 cell measurements are performed annually by approximately
600 testing laboratories in the United States.( 202 )
Because CD4 cells are a subset of all T lymphocytes, which are in turn a subset
of all white blood cells, variations in CD4 count can occur in response to a
variety of variables including concurrent infection, medications, stress,
malnutrition, vitamin deficiencies, and normal diurnal variation. Often, these
variables affect many subsets of lymphocytes and not exclusively CD4 cells;
thus, the percentage of T lymphocytes that are CD4 positive will remain
relatively stable. On the contrary, the depletion of T lymphocytes in HIV
disease primarily affects CD4 cells, causing a relative CD4 cytopenia and a drop
in the CD4 cell percentage. Additionally, an inversion of the normal CD4/CD8
cell ratio, which is usually >1 in non-HIV-infected individuals, may be seen
with progressive CD4 cell depletion due to HIV. Thus, the CD4 percent and the
CD4/CD8 ratio may help the clinician determine if a change in absolute CD4 count
is due to the effects of HIV disease or to some other factor.
Until recently, most absolute CD4 cell counts were determined using 2
instruments, a hematology analyzer and a flow cytometer (dual-platform
technology [DPT]). The CD4 count produced from DPT is the product of 3
laboratory measurements: the white blood cell count, the percentage of white
blood cells that are lymphocytes (differential), and the percentage of
lymphocytes that are CD positive (determined by flow cytometry). Single-platform
technology (SPT) is designed to enable determinations of both absolute and
percentage lymphocyte subset values using a single tube.( 203 ) SPT, introduced
for clinical application in 1996 is becoming the preferred method of CD4 count
determination in a number of laboratories.( 204 )
Both SPT and DPT flow cytometry technology for CD4 count determination require
specialized equipment and technician training. In resource-limited settings
where CD4 count may be unavailable, the total lymphocyte count (TLC), which can
be determined simply and cheaply, may be used as a surrogate for CD4 in
determining stage of HIV infection.( 205-207 ) For example, in a cohort of
HIV-positive people in south India, a TLC of <1,400 cells/µL has been shown to
be a good predictor of a CD4 count <200 cells/µL and thus an appropriate
surrogate marker for initiating cotrimoxazole prophylaxis.( 207 ) TLC may also
have applications in monitoring response to antiretroviral therapy in place of
or in conjunction with CD4 count. In an analysis of patients initiating a triple
antiretroviral drug regimen, an increase in TLC was associated with an increase
in CD4 count and a decrease in plasma viral load.( 208 )
HIV Viral Load Testing
Three technologies exist to measure HIV viral load in serum: reverse
transcription polymerase chain reaction (RT-PCR), branched DNA (bDNA), and
nucleic acid sequence-based amplification assay (NASBA). The basic principles
underlying these assays are similar--HIV is detected using DNA sequences that
bind specifically to those in the virus--but results may vary between tests.
Whereas early versions of the bDNA and RT-PCR techniques showed 2- to 2.5-fold
differences in results, the version 3.0 bDNA assay and version 1.5 RT-PCR test
yield values that are highly correlated ( r = 0.96) and in good agreement
(92.7%).( 315,316 ) Testing of non-clade B HIV-1 viruses using these methods may
not yield such highly correlated results.( 317 ) Thus, it is advised that
clinicians consistently use the same test when possible to compare results over
time.
Because viral load may vary by orders of magnitude, results of viral load
testing are often expressed in log units, where each increase of 1 log
corresponds to a factor of 10. Thus, a viral load of 1,000 would be 3 log units,
and the difference between a viral load of 1,000 and 10,000 would be 1 log unit.
A change in viral load of >0.5 log copies/mL (approximately 3-fold) exceeds
assay and diurnal variations, and may be considered to represent a true
biological event, whereas a change of <0.5 log copies/mL cannot be distinguished
from random variability. Diurnal variation in stable HIV viral loads is
approximately 0.4 log copies/mL.( 212 ) Acute intercurrent infection ( 213 ) or
immunization ( 214,215 ) may also transiently increase viral load.
HIV Antigen Testing
Assays for HIV antigens, notably the p24 antigen encoded by the gag gene, can be used to screen donated blood products.( 247 ) Measurement of p24 antigen may also be a less expensive yet effective alternative to HIV RNA testing in monitoring response to treatment.( 248 ) Testing for p24 may also be used to diagnose early HIV infection, because this viral antigen can be detected in the blood of infected individuals prior to the development of antibodies (seroconversion) detectable by ELISA or Western blot tests. In identifying primary HIV infection, p24 is more specific (99% vs. 95-97%) but less sensitive (79% vs. 100%) than HIV RNA determinations (either PCR or bDNA).( 249 )
HIV Resistance Testing
Resistance to antiretroviral drugs is unfortunately common in treated
populations. Resistance testing can be useful in determining which drugs not to
use in a treatment-experienced patient whose viral load is increasing despite
therapy, or in a previously untreated individual who may have been infected with
HIV resistant to one or more drugs.
Two types of HIV resistance testing are available. Genotypic assays detect
genetic mutations in the coding regions of the protease and reverse
transcriptase enzymes in HIV isolated from the patient. Using the results,
standardized algorithms are applied to predict resistance to various
antiretrovirals. Phenotypic assays are more similar to standard bacteriologic
sensitivity assays in that they are performed by culturing a fixed inoculum of
HIV genetic material isolated from the patient with serial dilutions of
individual antiretroviral drugs.
Prospective trials of genotypic ( 216-218 ) and phenotypic ( 219 ) assays have
shown benefit with each of these assays in achieving virologic control in
patients failing antiretroviral regimens. Additionally, with increasing
incidence of drug resistance in individuals recently infected with HIV,( 220 )
resistance testing during acute or early HIV infection may have important
long-term clinical relevance. Resistance testing in chronically HIV-infected
individuals provides information only on resistance to the medications being
taken at the time of the test. In individuals who have changed or interrupted
antiretroviral treatment, HIV harboring resistance mutations selected by prior
treatment may be "archived" as proviral DNA in long-lived resting lymphocytes or
macrophages, and may not be detected by resistance tests. Archived strains,
however, can be expected to return to dominance if selected by the drugs to
which they possess resistance. Results of resistance testing are therefore not a
substitute for the patient's clinical antiretroviral history, which must also be
taken into account. Further, because many resistance mutations tend to become
outgrown by wild-type virus when the drug in question is no longer present to
select for the resistance mutation, resistance testing in chronically infected
individuals who are not on ART at the time of testing is unlikely to be of use
and may provide misleading information.
In general, genotypic assays are more easily available, cheaper, and more
rapidly performed than phenotypic assays. A genotype result is more likely than
a phenotype to detect resistance from a minor variant or population mixture.
Genotype testing identifies only the dominant strains representing >10-20% of
virus circulating in blood at the time of testing.( 221,222 ). Genotypic assays
require an HIV viral load of >1,000 copies/mL to be reliable. Different
laboratories use different algorithms for determining drug resistance from a
given genotype result, and those different algorithms often produce discordant
results, particularly for NRTIs.( 223 )
A phenotypic resistance test has the advantage that results are generated in a
minimal inhibitory concentration (MIC) format that is more familiar to many
clinicians and also emphasizes the gradation of resistance that often exists;
however, the phenotypic threshold value that correlates with clinical resistance
is still debated for certain antiretrovirals. A phenotype gives results for one
drug at a time and is unable to predict the effects of combinations of drugs.
Phenotypic resistance assays may be particularly useful when combined with
monitoring of drug levels in treating individuals with highly resistant virus.
The "virtual phenotype" approach combines databases of matched genotypes and
phenotypes from the same viral isolates to predict the phenotypic susceptibility
of viruses with known genotypic sequences.
Current guidelines in the United States recommend resistance testing in cases of
acute or recent HIV infection, for certain patients who have been infected as
long as 2 years or more prior to initiating therapy, in cases of antiretroviral
failure, and during pregnancy.( 318 )
It is worth reemphasizing that, given the limitations of all currently available
resistance assays to detect archived mutations and minor variants, results of
resistance testing must always be interpreted in the context of prior
antiretroviral history and previous resistance testing.
Therapeutic Drug Monitoring
The measurement of antiretroviral drug concentrations in patient serum may be
used to predict toxicity,( 236 ) maximize efficacy,( 237 ) assess effects of
drug-drug interactions,( 238,239 ) and provide evidence regarding medication
adherence.(235)
Therapeutic drug monitoring (TDM) requires proper timing of sampling relative to
dosing and meals, and sampling the appropriate body compartment. For PIs and
NNRTIs, drug concentrations are measured in the plasma compartment, whereas for
nucleoside and nucleotide reverse transcriptase inhibitors, measurement of
intracellular metabolites is necessary.
In general, data on the efficacy of TDM in clinical practice are mixed. In
certain circumstances, such as in pregnant or pediatric patients, TDM may
provide data on drug concentrations that have not otherwise been well
characterized, but data from large studies are lacking to support its routine
use in clinical care.( 240 )
Other HIV Testing Techniques
ELISA or HIV viral load testing of fluids other than blood (seminal and
vaginal fluid, cerebrospinal fluid, urine, and saliva) are currently available
or under investigation. The clinical applications of some of these methods are
well proven (ie, diagnosis of HIV infection with saliva or urine ELISA testing [
241 ]), whereas for others it is less so (ie, serum viral load as a predictor of
infectivity [ 242 ] or testing of semen for use in in vitro fertilization [ 243
]).
Various culture techniques are available to isolate HIV from patient specimens.(
244 ) HIV can be quantitated by determination of proviral DNA in peripheral
blood mononuclear cells.( 245 ) Proviral DNA has been used to test babies for
HIV infection who were born to HIV-positive mothers.( 246 )
Assays of HIV antigens, notably p24 antigen, can be used to screen donated blood
products.( 247 ) It may also be an effective, inexpensive alternative to HIV RNA
testing in monitoring response to treatment.( 248 ) Additionally, because it is
a direct viral antigen, p24 can be detected in the blood of infected individuals
prior to the seroconversion detected by ELISA or Western blot tests. When
comparing p24 antigenemia to HIV RNA determinations (either PCR or bDNA) in
identifying primary HIV infection, p24 is equally specific (99% vs. 95-97%) but
less sensitive (79% vs. 100%).( 249 )
| « Previous | Table of Contents | Next » |
Medical information: Cure HIV, Inc. does not intend to provide specific medical advice or treatment. Cure HIV, Inc. intends to provide the website visitors with documents and information to better understand HIV / AIDS and its prevention and treatment.