Laboratory Diagnosis of HIV Infection
Click CC to turn on closed captioning.
Published: October 2009Print Record of Viewing
Dr. Yao will review the current recommendations for serologic and molecular testing for diagnosis and follow-up of HIV infection.
Presenter: Dr. Joseph Yao, Director of the Hepatitis/HIV Molecular Laboratory and the Hepatitis/HIV Serology Laboratory, in the Division of Clinical Microbiology at Mayo Clinic
Questions and Feedback
Welcome to Mayo Medical Laboratories' Hot Topics. These presentations provide short discussion of current topics and may be helpful to you in your practice.
Our presenter for this program is Dr. Joseph Yao, Director of the Hepatitis/HIV Serology and Molecular Laboratories in the Division of Clinical Microbiology, Department of Laboratory Medicine and Pathology at Mayo Clinic. Dr. Yao will review the current recommendations for serologic and molecular testing for diagnosis and follow-up of HIV infection.
HIV Markers During Early Infection
First, let’s review the virologic and serologic course of an individual who is infected with HIV. Here, from this graph, you see that the first marker that becomes detectable after infection is the HIV RNA, indicated by the green line. This is detectable by current molecular methods at about 11 days from the time of infection or exposure to HIV.
The second marker that becomes detectable in the laboratory is the HIV p24 antigen, indicated in the purple line. This is detectable by day 16 from exposure. And finally, the HIV antibodies that are detectable by current commercial assays occur at about day 22 from the date of infection. So the most widely used serologic tests, which are HIV antibody screening tests, are actually the least sensitive in picking up HIV infection compared to the other 2 markers.
Assay-Defined HIV Detection Windows and Infection Periods
Before I discuss further the various tests available commercially, let’s also look at the various detection windows and infection periods that are commonly indicated in the literature, because it has caused some confusion as to what are the definitions.
You can see from this timeline, indicated in the blue line and the various time points indicated by the down arrows, that there are basically 2 windows of HIV infection for detection. The first is the seroconversion window, which starts from time of infection, indicated by the first arrow on the left-hand side of the timeline, to the time point where antibody becomes detectable. So, this seroconversion window period actually includes the eclipse period and the acute infection period. The eclipse period is the period at which time that only molecular tests can detect the presence of HIV RNA. The acute infection is the period between viral infection detectable by molecular tests and a serologic response, which is detectable by serologic assays. Now the incidence window is the period from the time of antibody detection first from the infected individual until a specific time point where the serologic assay can determine recent infection. So that particular window period is also known as the recent infection. And, after this assay-specific detection point for recent infection, we see long-standing HIV infection. This timeline has been proposed by the World Health Organization Technical Working Group on HIV Incidence Assays.
Serologic Tests for HIV Infection in US
Currently in the United States, here are the serologic tests that are commercially available for detection of HIV infection. The most widely used are the ELISA assays, the enzyme-linked immunosorbant assays, which can come in the form of enzyme immunoassay or chemoluminescent immunoassay. They can detect either HIV-1 antibodies, or HIV-2 antibodies, or HIV-1 p24 antigen, or a combination of HIV-1 and -2; and then lastly the fourth generation serologic test is a combination of antibody and p24 antigen.
These are actually in clinical trials in the United States, but they are commercially available in Europe and other parts of the world. Then, the other serologic tests that are also used, particularly in resource-poor countries, as well as some urgent point-of-care testing facilities in the United States, are the rapid HIV antibody tests.
And, basically, there are 2 methods, the immunochromatography method, as well as the membrane immunoconcentration method. These rapid test devices are available to detect either HIV-1 antibodies alone, HIV-2 antibodies alone, or a combination of HIV-1 and -2 antibodies. Then the final group of serologic tests are the so-called supplemental tests, also known as confirmatory tests for HIV-1 and -2 antibodies. And there are essentially 2 methods that are commercially available: one is the Western blot and the other is the Immunoblot. Please note that the HIV-2 antibody tests by Western blot or Immunoblot are currently not yet FDA approved for diagnostic use.
Principle of Future Combined HIV Antigen-Antibody ELISAs
In this slide, I’ve illustrated the chemistry principle of the HIV antigen and antibody combination enzyme immunoassays that are undergoing clinical trials in the United States, but are commercially available in Europe and the rest of the world. You can see in this schematic diagram that both the antigens of HIV-1 and -2, indicated by the blue and purple diamonds, are actually attached in the reaction well and the antibodies for capturing HIV-1 p24 antigen are indicated in blue and those are also attached to the stationary well. So the target that this assay is trying to capture are the p24 antigen represented by the orange diamond and the orange-colored antibodies that are represented as an inverted Y.
Virology Tests for HIV Infection
Virologic tests are also available for the diagnosis of HIV infection. Early in the epidemic of HIV infection, the first tests that were available for diagnosis or detection of HIV are viral cultures using CD4 cells; these are the human helper T cells that are infected by HIV viruses. And, using these viral cultures, one is able to detect production of viral p24 antigens in the supernatant of cell cultures from CD4 cell lines.
Later on, when molecular tests are more refined, we have polymerase chain reaction or PCR assays for qualitative and quantitative detection of HIV-1 and -2. For qualitative PCR assays, one could detect HIV-1 proviral DNA. This is the DNA that is incorporated into the CD4 whole cells, DNA that belong to HIV-1 viral genome. One could also detect a combination of HIV proviral DNA and RNA and also laboratory-developed assays, particularly in certain research investigator laboratories, one could also design a qualitative detection of HIV-2 RNA. For quantitative assays, there are commercially available and FDA- approved assays for quantifying HIV-1 RNA, and then there are laboratory-developed assays for quantifying HIV-2 RNA.
Two other commercial laboratory tests available utilize transcription-mediated amplification for qualitative detection of HIV-1 RNA, such as the Gen-Probe Aptima assay for diagnostic purposes, and the branched DNA method from Siemens Healthcare Diagnostics, which utilizes signal amplification for quantitation of HIV-1 RNA.
HIV Serologic Screening Algorithm
Next, I will go through the HIV Serologic Screening and Supplemental Testing Algorithm. You can see in the top of this flowchart, that the Mayo Medical Laboratories test #9333 is a screening evaluation for HIV-1 and -2 antibodies. If that is reactive, it is automatically reflexed to HIV-1 antibody supplemental test by Western blot. From that test, one could see possibly 4 test results. Either positive, which means confirmed HIV-1 infection, or negative, which means that it’s truly negative or could be an HIV-2 infection giving the initial HIV-1 and -2 antibody screen results. Or, it could be an indeterminate or uninterpretable Western blot result, indicated in the gray box in the middle.
If the HIV-1 antibody confirmation by Western blot is negative in this testing algorithm, at Mayo Medical Laboratories, we automatically reflex to HIV-2 antibody-only screen. And if that is negative, then we recommend repeating serologic HIV testing starting from the HIV-1 and -2 antibody screen algorithm in 1 to 3 months for high-risk individuals. If the HIV-2 antibody screen is positive, then it’s recommended that we proceed to HIV-2 antibody supplemental test by immunoblot and that is done automatically on antibody screen-reactive specimen and, if that immunoblot assay result is negative, again we recommend HIV-2 antibody screen in 1 to 3 months, if clinically indicated.
For Western blot results that are indeterminate or uninterpretable, 2 tests are performed automatically. One is the HIV-2 antibody screen indicated by #86702, and also the #81758 HIV-1 antibody supplemental test by immunofluorescence. The reason for immunoflorescence test is to give provider additional assurance that we’re not missing an early HIV-1 infection that may be missed by the Western blot, since the immunofluorescence is also an FDA-approved supplemental test for HIV-1 antibody detection.
Based on the immunofluorescence test result, one could see a negative or a positive result. There’s also an additional option that the laboratory or provider could seek further confidence in an indeterminate or uninterpretable test result for Western blot and that is to proceed to HIV-1 proviral DNA qualitative detection by PCR Mayo Medical Laboratories #88635. And that is a research use only test and that can be positive or negative and then, depending on whether the individual is at high risk or low risk, the recommendation is indicated on the right-hand side of this testing algorithm.
HIV Rapid Serology Follow-up Algorithm
For individuals who had a reactive or presumptive positive rapid antibody test, here is the recommendation for a follow-up laboratory evaluation. And let’s concentrate just on the top one-third of this algorithm. For positive or reactive rapid HIV-1 or rapid HIV-1 and -2 antibody tests, we recommend a combination of HIV-1 and -2 confirmatory evaluation listed there, which is to perform a Western blot, or to actually perform Western blot confirmatory test as well as an antibody screen, which will include an HIV-1 and -2 antibody screening test in the laboratory using enzyme immunoassays or chemiluminescence immunoassays.
Then the rest of the algorithm below is the same as the previous slide. The reason for including an enzyme immunoassay screen after a rapid antibody test result is in case if one does decide to do a follow-up evaluation at a later date, that would start with an HIV-1 and -2 antibody. So this is mainly for later comparisons.
Indeterminate HIV-1 Antibody Western Blot Results
There are some well-known causes of indeterminate HIV-1 antibody Western blot results. I’ve listed them here. Most frequently, we see are due to nonspecific reactions, so there are some proteins or antibodies in the patient serum, particularly those with autoimmune disorders such as SLE rheumatoid arthritis that can cross react with the antigen on the Western blot strips. Another major group that we’ve seen indeterminate Western blot in a false-positive screening antibody test is the pregnant women, because they produce allo-antibodies during pregnancy and these antibodies tend to cross-react with the HIV-1 and -2 antibody screening tests as well as the Western blot results.
So the approach with individuals with such indeterminate Western blot results is to repeat antibody testing in 2 to 4 weeks in high-risk individuals. For low-risk individuals, repeat testing is recommended in 3 to 6 months. In pregnant women, they are likely to have a false-positive antibody screening test and indeterminate Western blot results until 1 month after delivery, where the allo-antibodies are cleared from their system. The other possibility, instead of waiting 2 to 4 weeks in high-risk individuals, is to proceed to detection of HIV-1 RNA or proviral DNA using molecular tests.
Rapid HIV Antibody Tests Licensed in US, 2009
There are currently 6 rapid HIV antibody tests that are licensed in the United States for screening of HIV infection. Those listed with an asterisk - there are 5 of these 6 - are also CLIA waived for point-of-care testing. Of note, many of these are available and FDA approved for testing saliva, plasma, serum, as well as whole blood, and they utilize a variety of chemistry methods, especially lateral flow immunochromatography or a flow-through membrane immunoconcentration.
Predictive Value: Single Screening Test
I do wish to stress that if one uses the rapid screening antibody test in a population with low prevalence of HIV infection, such as those with HIV prevalence of <5%, the positive and negative predictive value is related to the prevalence of a particular disease. So, for example, if HIV prevalence is 1% in a population that the laboratory serves, then despite having a good sensitivity and specificity of the test, the positive predictive value (meaning that a positive result indicates true positive) is only 83% indicated in this table. And, in many communities in the United States where the prevalence is 0.1% or less (meaning less than 1 in 1000 has HIV infection), the positive predictive value is only 33%, so it's actually less accurate than a flip of a coin.
Rapid HIV Antibody Tests: Clinical Applications
Currently, the standard of practice in the United States with use of rapid HIV antibody tests involves using this assay for detecting HIV infection in source patients in the event of occupational exposure, such as needle stick injury or surgical knife accidental cut in the operating room, where we need to know if the source patient is HIV positive if the HIV-infection status is unknown, so that we could properly implement prophylaxis for the health care workers who are exposed.
The other major group where this rapid antibody test is very helpful is determining the HIV-serologic status of pregnant women who are actively in labor, but did not have prenatal HIV screening tests. Here in these individuals, who are infected but did not have adequate prenatal care, anti-HIV therapy can be implemented during labor to prevent transmission of HIV infection to the newborn during the birthing process.
Thirdly, high-risk individuals who present to the hospital emergency department with suggestion of infection, but with unknown HIV-serologic status, and here in these individuals, the association with HIV infection have important implications for management of the individuals.
Follow-Up Testing of Reactive Rapid HIV-1 Antibody Tests Results
I also want to remind the listener that follow-up testing of reactive rapid antibody tests include supplemental testing with Western blot or the indirect immunofluorescence assays, even if the laboratory immunoassays such as EIA or CIA are nonreactive. And, if the confirmatory tests yielded negative or indeterminate results, follow-up testing should be done in 4 weeks after the initial rapid antibody test result was obtained.
Diagnostic HIV Testing in Infants
The last group of individuals that I wish to discuss testing approaches is infants, infants born to HIV-infected mothers. The US Working Group on Antiretroviral Therapy and Medical Management of HIV-Infected Children recommended that 2 positive HIV virologic tests on blood samples taken on 2 separate dates are necessary to make a diagnosis of HIV infection in such newborn infants. And the testing that can be used includes the qualitative HIV-1 proviral DNA by PCR, the quantitative HIV-1 RNA by PCR, or an HIV-1 RNA by transcription-mediated amplification.
The testing should be done at birth, preferably within 48 hours of birth, at 14 days, at 1 to 2 months, and finally at 3 to 6 months. So, if any 2 of these time-point testings yield a positive HIV-test results, then an HIV diagnosis can be made. Please note that the HIV-1 proviral DNA test and the HIV-1 RNA quantitative tests, both by PCR, are not FDA approved for such indication, but they are nevertheless recommended as a testing option by the US Working Group on Antiretroviral Therapy and Medical Management of HIV-Infected Children.
Therefore, HIV infection can be reasonably excluded if there are more than 2 negative HIV-virologic tests, preferably taken at 1 month or more in the individual newborn and again at 4 months after birth. Alternatively, if molecular tests or virologic tests are not available, then 2 negative HIV-serologic tests, at more than 6 months after age, is also possible to exclude HIV infection in such infants born to HIV-infected mothers.
Awareness of Serostatus Among People with HIV and Estimates of Transmission in US
The last section I wish to discuss is indications for HIV testing, and the Center for Disease Control and Prevention in 2008 published their survey finding of knowledge of HIV-serologic status among patients in the United States and correlate their awareness with estimates of transmission of HIV infection in the United States. You can see that with about 1 million individuals living with HIV infection or AIDS in the US, about a quarter of them are unaware of their infection because they don't think they were at risk or they've never been tested even if they are at risk. However, these 25% of the infected individuals accounted for up to 50% of the new infections occurring every year, which is around 40,000 cases. So in the effort to decrease or reduce new infections of HIV, it is important that we identify those 25% of individuals who are infected but unaware of the infection.
Reasons for Testing: Late vs. Early Testers
When the CDC looked at the reason that these individuals come for HIV testing, we see this graph where most of them actually present very late in their HIV infection for testing because they became ill. Whereas, if we looked at those who came for early diagnostic testing, you can see that most of them came because their partners are at risk and were found to have HIV infection, and early diagnostic testing is defined as those who were tested more than 5 years before they developed AIDS. Whereas those who actually presented for late or early testing, very few of them were tested because of routine check-up, as indicated by the 2 bars in the red circle here. So, in the effort to reduce HIV transmission, we should increase our effort by promoting HIV testing in doing a routine check-up.
2006 US CDC Recommendations: Adults and Adolescents
It is because of these findings that the CDC recommended in 2006 that all adults and adolescents should have a routine voluntary HIV-antibody screening test if they fall into the age group of 13 to 64 and not a screen only because of risk factors for prevalence of HIV infection. Those with known risks for HIV should also have HIV screening at least once a year.
Thirdly, the approach for HIV testing should be an opt-out approach, meaning that patient should consult and notified that they're being tested unless they object. So this is in contrast to the previous approach of opt-in testing, where the patient must give consent before they're being tested. With the opt-out approach, it reduced the barrier and obstacles for HIV screening, and one of the recommendations is that whenever a patient sign a consent for medical care, that is understood that HIV testing will be included as part of the general care for that patient unless he or she specifically objects to HIV testing.
Counseling for testing is not required. It should be just be part of general medical care in every patient encounter in the health care setting. The CDC also recommended that state and local regulations requiring informed consent be reviewed and revised to lower the barrier and obstacles for routine voluntary screening. The CDC recommendation also has provision that such routine screening tests can be pulled back if there is low prevalence of HIV infection, such as <0.1 %, so that continuous screening may not be cost effective in these populations.
2006 US CDC Recommendations: Pregnant Women
Finally, for pregnant women, the CDC also recommended universal opt-out screening tests for prenatal care of these pregnant women. During their consent for prenatal care, they should be informed that HIV testing is part of their routine prenatal care. And those pregnant women who are known to be at risk for HIV infection, they should be tested again in the third trimester if their initial testing were negative for HIV.
And in the labor and delivery setting, women who have not had previous HIV-serologic tests, they also should undergo rapid antibody testing in the opt-out approach so that appropriate antiretroviral prophylaxis can be initiated during labor and delivery. Also, newborns who are born to HIV-infected mothers or mothers with unknown HIV status, should also be screened and tested in the opt-out approach.
In summary, many HIV serologic and virologic laboratory test methods are available. However, the diagnostic laboratories should choose an approach of initial and supplemental testing that is most suitable for the specific clinical practice population that it serves, the laboratory resources, and patients' affordability. It's also important to remember the HIV infection is transmitted mainly from those who are unaware of their infection and, therefore, all individuals of all ages, especially those sexually active, should be tested at least once in their lifetime and annually for those individuals who are at risk for HIV infection.
I wish to disclose that I have received research grants from Abbott Molecular, Roche Diagnostics, and Siemens Healthcare Diagnostics; and I also served on the advisory boards for the 3 commercial companies mentioned who manufacture molecular diagnostic tests for HIV.