Immunoassays

HISTORY OF DEVELOPMENT OF IMMUNOASSAYS

Immunoassays have changed significantly over time, with improvements in the types of

antibodies and antigens available as well as in detection systems (10, 25, 27, 30, 33, 54).

With immunoassays, any analyte can be measured if an antibody can be raised to it or if an

antigenic form is available. The first immunoassays available measured milligram to

microgram quantities of antibodies and relied primarily upon precipitation reactions between

antigen and antibody. In the 1960s, the advent of radioimmunoassay (RIA) heralded

techniques with greater sensitivity and greatly expanded the repertoire of analytes available

for testing. By use of RIA, previously undetectable analytes were now easily available for

testing in the clinical laboratory. The discovery of monoclonal antibodies led to assays with

greater specificities and further expanded the repertoire of analytes available for

measurement. Concerns about utilization of radioactivity and the desire for greater

sensitivities led to the development of chemiluminescence (CL) immunoassays and use of the

avidin-biotin detection system. Increasing automation of laboratory testing has expanded

into the area of immunoassays, with many tests requiring only limited technologist input. As

immunoassays have evolved, there has been increased utilization of various solid-phase

matrices for adherence of either antigens or antibodies. Initially, polypropylene test tubes

were used. This has evolved to microtiter plates, and with the increased use of automated

systems, smaller solid phases such as tiny disks or spheres are being increasingly used.

Thus, immunoassays have significantly advanced in both the level of sensitivity detected and

the breadth of their utilization, so they are now some of the most popular and most widely

used of all laboratory tests.

DEFINITION OF TERMS Back to top

The array of terms used for immunoassays can be a confusing alphabet soup. This chapter

discusses some widely used conventions in terminology; however, the reader may find some

references in which the terms are used differently. Overall, most assays utilize “immuno”

coupled with a second term which describes the type of assay or label used. For example,

immunoprecipitation is an immunoassay utilizing a precipitation reaction. RIA is an

immunoassay that utilizes radioactivity as the label. “Enzyme immunoassay” (EIA) is a more

general term that can be applied to any immunoassay which uses an enzyme label, although

often EIA is used to refer to reagent-limited competitive-type assays. The term enzymelinked

immunosorbent assay (ELISA) can also be used as a general term for any assay

utilizing an enzyme label; however, it is most often used to refer to assays in which the

antigen or antibody is adsorbed to a solid-phase matrix, often then employing a second

enzyme-labeled antibody, the so-called “sandwich” assay format. “Immunometric” is an

additional term used and generally refers to any reagent excess assay. For the purposes of

this chapter, the term EIA is used to refer to any assay using an enzyme, while the term

ELISA refers only to solid-phase sandwich-type assays.

GENERAL CONCEPTS OF ASSAY DESIGN Back to top

There are a number of ways to characterize immunoassays. One useful classification scheme

looks at the amounts of label and reagent available (21). There are three major groups of

immunoassays: label free, reagent excess, and reagent limited. The assays which are label

free rely upon the ability of antigen and antibodies to bind and form detectable agglutination

or precipitation. There are many classic agglutination assays used in the diagnosis of

infectious diseases, such as the Widal test for typhoid fever. The reagent excess methods

require an excess of labeled antigen or antibody, use either one or two sites, and include

immunoblotting and solid-phase ELISA. These are commonly employed immunoassays in

microbiology today. Reagent-limited assays are competitive tests and employ a limited

amount of either antigen or antibody and either require separation or are separation free.

These include classic RIA and EIA and are less often used in diagnosis of infectious diseases.

Another commonly used classification scheme looks at immunoassays as either

heterogeneous (solid phase) or homogeneous (free-solution assays) (10, 27).

Heterogeneous assays are ones in which the bound and free components must be separated,

whereas homogeneous assays do not require a separation step. In addition, heterogeneous

assays involve some type of solid phase to which the immunoreactants are attached.

Homogeneous assays generally are free-solution methods. While this is a useful and

commonly employed classification scheme, there are assays that do not strictly fit into this

classification scheme. For example, agglutination assays and particle-enhanced lightscattering

methods are considered homogeneous assays; however, the antibody is bound to

a solid phase, and there is no required separation of the bound from the free components.

ASSAY INTERPRETATION Back to top

When choosing an assay for the laboratory and in-patient diagnosis, it is critical to

understand the concepts of sensitivity, specificity, and predictive values (54). Sensitivity is

the proportion of individuals with a disease that are correctly identified with a particular test.

Sensitivity defines the true positives (TP), which are the patients with disease identified by

the assay. Conversely, false negatives (FN) are the patients with disease who are not

identified by the test. The formula for sensitivity is as follows: sensitivity = [TP/(TP + FN)] ×

100. Specificity is the proportion of those without the disease that are correctly classified.

Specificity is a measure of the true negatives (TN), which are the patients without disease

not identified by the assay. False positives (FP) are the patients without disease who test

positive. The formula for specificity is as follows: specificity = [TN/(TN + FP)] × 100. With a

highly sensitive test, the majority of diseased individuals are picked up, and thus the number

of false-negative results is very low. In contrast, with a highly specific test, the majority of

individuals without the disease test negative, so the number of false-positive results is very

low. When an assay is developed, the diagnostic cutoffs can be modified to alter both the

sensitivity and the specificity. For example, if one moves the cutoff to a lower level, the

assay sensitivity is increased, with a resulting decrease in specificity. The optimal balance of

these two components must be evaluated for each laboratory test and depends on multiple

factors, such as the utility of the test and the prevalence of the disease in the population.

The probability of having the disease, given the results of a test, is called the predictive

value of the test. Positive predictive value (PPV) determines the percentage of patients with

positive results who are diseased: PPV = [TP/(TP + FP)] × 100. The negative predictive

value (NPV) calculates the percentage of patients with negative test results who do not have

the disease: NPV = [TN/(TN + FN)] × 100. The predictive value of a test combines the

prevalence of disease in a particular population with the sensitivity and specificity. Positive

and negative predictive values are important because they assess the ability of a test to

predict the presence or absence of disease in a patient from a particular population. In this

context, the disease prevalence is a critical component. Prevalence is the proportion of the

population with the disease in question. If a disease state has a low prevalence in the target

population, a positive result will most likely be a false-positive result, whereas the opposite is

true in a high-prevalence population. A potential use of a high negative predictive value is

that a negative test can exclude disease. In addition to the values discussed above, there are

a variety of other statistical methods that can be used to evaluate laboratory tests, such as

odds ratio, receiver-operator curve analysis, and likelihood ratios. It is beyond the scope of

this chapter to discuss these, and the reader is referred to other sources for a more complete

discussion (49).

SCREENING VERSUS DIAGNOSTIC ASSAYS Back to top

An important component of assay design is based upon the ultimate use of the test, i.e.,

whether it will be used as a screening or diagnostic test (40, 51). Screening tests are

designed to pick up disease in asymptomatic individuals who may have early disease or

precursors of disease, whereas diagnostic tests are performed for persons with specific

indications of possible disease. However, the screening procedure itself does not diagnose

the illness; those individuals with a positive result from the screening test need further

evaluation with additional diagnostic tests. If the individual has a previous positive screening

test, the diagnostic test acts as a confirmatory test. The ideal screening test should be both

highly specific and sensitive; however, this may be difficult to achieve. As there is such a

variety of screening tests available, there is not a particular sensitivity target value which is

suggested; nevertheless, the sensitivity should be as high as possible without sacrificing

specificity. It is not advisable to use a test with low specificity as a screening test, since

many people without the disease will screen positive and potentially receive unnecessary

diagnostic procedures. Moreover, for an effective screening test, the prevalence of disease in

the population should be high; if the prevalence of the disease is low, then a positive test will

most likely be a false positive, leading to further unnecessary testing. Other considerations

with regard to screening tests include weighing the cost of the test versus the impact of early

detection. Overall, good screening tests should be easy to perform, inexpensive, and

performed in high-disease-prevalence populations. In addition, early detection of disease

should have a measurable impact on patient outcomes.

SEROLOGIC ASSAYS Back to top

Traditionally, serologic assays referred to the use of serum or plasma samples for the

detection of antibodies to a variety of antigens. This concept has been broadened to refer to

a variety of patient samples, such as cerebrospinal fluid, urine, and other body fluids. In

addition, it refers to the detection of both antibodies and antigen. There are a variety of

clinical scenarios in which serologic assessment is the test of choice. For the identification of

organisms for which culture is difficult or requires prolonged incubation, the determination of

antibody titers or antigen detection can often give a quick answer. Although molecular

techniques have sometimes supplanted serology in these situations, often cost issues make

serology a more viable technique. There are frequent clinical situations where it is

unnecessary to perform a culture if an antigen test is positive. One of the most common

situations is the diagnosis of group A beta-hemolytic streptococcal throat infection. A rapid

immunoassay for the detection of the group A streptococcal antigen is performed. If this is

positive, then treatment can be instituted. Only when this quick test is negative is it

necessary to perform a culture.

Basic Immunologic Reactions

In order to facilitate understanding of antibody titers, a brief review of basic immunologic

reactions is provided (1, 2). Upon initial exposure to an infectious disease (primary antibody

response), there are four phases in the response: an initial lag (or window) phase, when

there is no antibody detected; a log phase, when the antibody titer increases in a logarithmic

fashion; a plateau phase, in which the amount of antibody stabilizes; and a decline phase,

during which the antibody is cleared or catabolized. The actual time course and ultimate

maximum antibody titer depend on the antigen and the host. In the primary response, the

initial antibody response is the production of immunoglobulin M (IgM), which usually appears

after 10 days. The period after initial exposure, but before antibody is produced or is at

sufficient levels to be detected, is called the window period. This can vary, depending on the

infectious agent, from as short as 10 days to as long as 6 months. IgG antibody production

usually begins 10 days after exposure but is much less than the IgM response. As the IgM

antibody level decreases, the IgG level increases, so that usually by the end of the first

month, only IgG antibody is detectable. If there is a repeat infection with the same infectious

agent, the kinetics of the response are different, with a lag phase of only 1 to 3 days, and

IgG antibody is the primary isotype produced. In the months following antigen exposure, the

IgG level reaches a plateau, and the antibody may remain detectable for life, even if there is

no further exposure to the antigen. B lymphocytes utilize membrane-bound antibodies to

recognize a wide array of antigens. In the case of infectious disease agents, the antigens are

often expressed on the microbial surfaces. The particular parts of the expressed antigens

that are bound by antibodies are referred to as epitopes; the strength of the binding of one

epitope to one antibody is called the affinity. Upon repeated infection with a microbe, there is

an increase in the strength of the antigen-antibody binding, a phenomenon called affinity

maturation. However, depending upon the immunoglobulin molecule present, there is more

than one antigen-binding site on each immunoglobulin (IgG, 2 sites; polymerized IgA, 4

sites; and IgM, 10 sites); therefore, the total strength of the antigen-antibody binding is

much greater than the affinity of a single interaction. This is called the avidity. Just as with

affinity maturation, there is an increase in avidity with additional exposure to an antigen.

Upon initial exposure to an antigen, the avidity of the IgG is low; upon secondary exposure,

there is an increase in IgG avidity. Although there is exquisite specificity in the antigenantibody

reaction, there can be a spectrum of antibodies produced in response to a particular

antigen; they can have differing affinities and avidities with a particular antigen and thus can

be responsible for cross-reactions. Recognizing cross-reacting antibodies can be critical to

assay specificity. Often, initial screening assays are set up with crude antigen preparations in

which false-positive reactions can occur. Secondary confirmatory or diagnostic assays utilize

purified and more expensive antigen preparations which confer greater specificity.

Recognition of cross-reactivity is critical in tests used in infectious disease, because often

organisms within the same genus and species share multiple antigenic determinants, making

cross-reactivity a common problem. Although assays are designed to obviate these

problems, laboratorians and clinicians should just be cognizant of the potential for crossreactions.

Caveats in Serologic Interpretation

In general, a positive IgG titer means only that an individual has been exposed to a

particular infectious agent and thus is “immune.” For each infectious agent, the laboratory

result is usually set up so that a positive result is the minimum amount of IgG antibody

present which makes the individual immune. For purposes of this discussion, the term

immune is used; however, this does not necessarily mean that the level of antibody is

protective against reinfection. The actual amount of antigen-specific antibody present in the

serum of a particular individual is host determined and is controlled by immune response

genes which are part of the human histocompatibility system. The IgG titer from a single

serum sample to a particular infectious agent may not be used to determine if the infection is

recent or remote. For example, person A may be a high responder to certain antigens and a

low responder to others. Therefore, if a high titer of IgG is obtained for an individual, it may

be tempting to think that this may represent a more recent exposure; however, this may

indicate only that the individual is a high responder to that particular antigen. Therefore, a

positive IgG titer establishes only that the individual has been exposed to a particular agent

at some time in the past and has detectable IgG. In addition, the nature of the antigen is

important, as some antigens are more effective than others in stimulating the immune

system. Moreover, the ability of the immune system to respond to antigens can be affected

by a variety of factors, such as age. For example, the very young may be unable to respond

to certain types of antigens (e.g., carbohydrates) (1, 2).

Using serologic methods, there are several ways to determine if the infection is recent. The

most useful and frequently used method is assessment of IgM antibody to a particular

infectious disease agent. In general, a positive IgM titer to a particular organism is evidence

of an active (i.e., recently acquired) infection with that agent. However, there are several

considerations to keep in mind in the interpretation of this test. First of all, a positive IgM

titer does not always mean that the infection is recent. There have been reports of persistent

elevations of IgM antibody for a year or more. This has been seen with multiple organisms,

including cytomegalovirus, Mycoplasma pneumoniae, hepatitis A virus, and Toxoplasma

gondii, among others (11, 36, 51). Conversely, a negative IgM titer does not exclude a

recent exposure. The amount of IgM may have been small and resolved quickly; thus, it was

not detected at the time of the assay. The second way to establish a primary infection is to

determine acute- and convalescent-phase titers. This requires drawing two sets of samples

for antibody titer determination: one set early in the exposure to the infectious agent and a

second set 2 to 3 weeks later. Evidence of an acute infection can be confirmed if there is a

fourfold increase in antibody titer between the first and second sets.

While this can sometimes provide information on the pathogenesis of a disease state, it

requires at least 2 to 3 weeks for definitive results, thus obviating its use for early clinical

management. Also, a false-negative reaction is not uncommon due to the fact that it

requires drawing the specimen at a point low enough on the log part of the antibody

response curve to obtain the required fourfold increase in antibody titer. Therefore, the lack

of a fourfold increase does not rule out a primary infection.

At present, many assays report results as absorbance values or international units, making it

problematic to apply the concept of a fourfold increase in titer. While it is possible to develop

an approximate equivalency between titers and absorbance values, this has to be individually

developed for each assay. One method to do this involves collecting multiple pairs of acuteand

convalescent-phase sera to use as reference sera. They must demonstrate a fourfold

increase in titer on traditional assays. They are then run on the EIA for the agent in question,

and reference ranges are reported in optical density units. Assuming that the EIA provides

distinctly different absorbance ranges to adequately separate acute- and convalescent-phase

sera, pairs of test sera can then be run on the EIA, and the values can be compared to the

reference ranges. While this can theoretically provide an adequate way to evaluate acuteand

convalescent-phase titers by using the newer assay formats, it can have multiple

problems. First of all, establishment of the reference range has to be performed separately

on assays for each infectious agent, which can be expensive and time-consuming. Secondly,

the lab has to have multiple sets of positive acute- and convalescent-phase paired sera for

each organism tested, which can be quite difficult to obtain. In addition, depending on the

EIA used and the range of the standard curve, titers may not be easily converted into

equivalent and meaningful absorbance values. Considering the cost and difficulties

associated with this type of analysis, it is not generally recommended. Instead, it is

preferable to test for the presence of an acute infection by using an IgM assay or an IgG

avidity test (see below) or to directly test for the organism by using one of the increasingly

available molecular techniques.

An IgG avidity test (31, 36, 41) can address some of the concerns with IgM testing and

acute- and convalescent-phase titers. IgG antibodies produced early in infection have a low

avidity, but a much greater avidity is seen with a secondary exposure. Using an avidity

assay, in conjunction with the assessment of levels of IgG and IgM antibody to a pathogen,

can provide a much clearer indication of acute infection. Avidity tests are performed with a

modification of the standard IgG EIA, in which IgG antibodies are exposed to a dissociating

agent (usually high concentrations of urea). The serum IgG avidity is estimated by

comparing the treated sample with one left untreated. While this test can be quite useful,

there are several caveats with its use. First of all, low avidity does not always mean that the

infection is recent, because low-avidity antibodies can persist for months to years. In

addition, there can be quality control issues in this testing, with variability in test results

related to the type of assay plates used, the antigens employed, and the type of dissociating

agent used. This test has special utility in testing for some of the pathogens associated with

pre- and perinatal infections (toxoplasmosis, rubella, and cytomegalovirus infection)

(31, 36). For example, one algorithm suggested for prenatal toxoplasmosis testing follows all

positive IgG antibody assays with an avidity test. If the avidity test is high, an acute infection

is ruled out; however, if the avidity test is low, an IgM test is then performed. If the IgM test

is positive, then a recent IgM infection is highly suspected. However, considering the

implications for pregnancy, the FDA recommends that sera with positive IgM results obtained

at a non-reference laboratory then be sent to a toxoplasma reference laboratory for

confirmatory testing.

SPECIFIC IMMUNOASSAYS Back to top

The spectrum of immunologic assays is discussed in detail in the following sections. Table

1 lists selected assays and provides approximate levels of detection.

Back to top

TABLE 1 - Sensitivities of immunoassays a

Technique Approximate sensitivity (per ml)

Precipitation, tube 100 mg

Immunodiffusion 1–3 mg

Agglutination 1 μg

CF 1 μg

Hemagglutination, passive 50 ng

Particle immunoassay 30–50 ng

EIA <1 ng

a Data from references 24 and 46.

Precipitation Reactions

When soluble antigens and antibodies are in equimolar concentrations, they bind and form

insoluble antigen- antibody complexes which form a visible precipitate (27, 32). There are a

number of laboratory tests available that utilize this reaction. Immunodiffusion is the

simplest of the precipitation assays and involves putting the immunoreactants in an inert

semisolid material and then viewing the visible precipitation line. There are several variants

of immunodiffusion. Radial immunodiffusion is designed to provide protein quantitation,

whereas double immunodiffusion (Ouchterlony analysis) allows characterization of the

relationship between different antigens. Overall, immunodiffusion reactions are simple to

perform, easy to evaluate, and inexpensive and can be adapted to a variety of health care

settings. The drawbacks include low sensitivity, as the level of detection is microgram

quantities of antibody or antigen; requirements for relatively large amounts of antigens and

antibody; and long assay times. Immunodiffusion is also routinely used for determination of

titers of antibody to a variety of agents, most commonly antifungal antibodies (Coccidioides,

Aspergillus, andHistoplasma).

Agglutination Reactions

Agglutination reactions require a particulate antigen and its antibody, with the resultant

visible clumping as evidence of a positive reaction (24, 27, 32, 33). A test involving the

particulate antigen which agglutinates the antibody present in the patient sample is termed a

direct agglutination assay. To enhance the visibility of the agglutination reaction, an indirect

assay format can be used, in which the antigen is coupled with a variety of particles that

serve as an inert matrix. Various materials which have been employed include gelatin, latex,

erythrocytes (RBC), polypeptides, and silicates. In addition, soluble antigen can be detected

in a patient sample by absorption of a specific antibody to a particle; this is termed reverse

agglutination. Due to the large IgM molecule with its pentameric structure, IgM antibodies

are several hundred times more efficient at agglutination than IgG and thus give more

consistent and stable agglutination reactions. If the immune response involves primarily IgG

antibody, the reaction may require some type of chemical enhancement or an antiglobulin

reagent. Flocculation assays are another variant of agglutination assays, in which the

particles are suspended. The most frequently used assays are the VDRL and rapid plasma

reagin tests for syphilis.

Many agglutination assays, called hemagglutination assays, employ RBC and use either a

direct or an indirect assay format. Direct agglutination of RBC is commonly used in the blood

bank for ABO typing. For infectious disease diagnosis, one of the most frequently ordered

direct hemagglutination assays is the Monospot test. This test detects the presence of a

heterophile antibody which is produced in infectious mononucleosis and happens to

spontaneously agglutinate equine RBC. The indirect hemagglutination assay is a commonly

used format in which antigen is adsorbed to RBC, thus testing for the presence of specific

antibody in the patient serum. Alternatively, the assay can be modified to test for antigen, in

which case it is called a reverse agglutination assay. For infectious disease testing,

hemagglutination (especially indirect) is a popular assay format, as it is sensitive and simple

to perform and does not require sophisticated equipment. For these reasons, it has been

used in many developing countries for testing of a variety of infectious disease agents, such

as human immunodeficiency virus (HIV), hepatitis viruses (A, B, and C), and Treponema

pallidum. There is a unique type of hemagglutination assay format used primarily in viral

serology called hemagglutination inhibition. It is most commonly used for detection and

quantitation of anti-influenza virus antibodies. It is based on the principle that some viruses

have surface proteins that will agglutinate RBC, so the assay uses the ability of antiviral

antibodies in the patient sample to inhibit the spontaneous agglutination of the test RBC. The

titer of antiviral antibodies is reported as the last dilution of the patient serum still able to

inhibit the agglutination reaction.

Specialized types of agglutination assays that require optical counting are called particle

immunoassays (10,20, 29). They involve primarily the measurement of scattered light which

occurs upon the antigen-antibody reaction, and this is measured by either turbidimetry or

nephelometry. They can be used for testing a wide range of proteins and analytes. Particle

immunoassays are 3 orders of magnitude more sensitive than standard agglutination. One

additional assay is the particle-counting immunoassay, which is used for quantitating

haptens, antigens, and antibody. It is also available in a fully automated immunoassay

format. In this assay, optical cell counting is employed, and there is an assessment of the

decrease in agglutination following the immunoreaction. These assays are sensitive to a level

of nanograms per milliliter. The patient sample is mixed with latex beads coated with

antibody. As the antigen-antibody reaction occurs, the antigen particles are no longer

dispersed in the solution; therefore, the antigen concentration is inversely proportional to the

amount of antigen particles remaining in solution. Antibody can also be quantitated in this

assay. The use of a particle-counting immunoassay has been reported for quantitation of

hepatitis B virus surface antigen, along with quantitation of antibodies to hepatitis C virus, T.

pallidum, and T. gondii (20, 23).

Overall, basic agglutination assays are easy to perform and inexpensive and can be done in a

variety of clinical settings, such as the doctor’s office, the emergency room, and the hospital

bedside and in the field. They are performed either on a card, in tubes, or in microtiter

plates. Often they provide only qualitative results, although an antibody or antigen titer can

be obtained through serial dilutions of the sample. Direct assays continue to be performed

for rare pathogens such as Francisella and Brucella. They utilize an inactivated source of the

whole organism mixed with the patient sera. Although agglutination assays suffer from both

limited sensitivity and limited specificity, they continue to be utilized because they are easy

to perform and relatively inexpensive. If a more quantitative assay is needed, the assay can

be adapted to light-scattering equipment such as a nephelometer to provide more

quantitative and sensitive results. Overall, the major drawback to direct agglutination assays

is their limited sensitivity. They detect only to a level of microgram to milligram quantities of

analytes per milliliter. However, greater sensitivities can be achieved with many of the

variants of direct agglutination. For example, microtiter passive hemagglutination assays for

infectious agents can achieve a sensitivity equivalent to that of a conventional EIA. The more

sensitive hemagglutination assays for measuring antigen can measure as low as 15 to 30

μg/ml. If an agglutination assay is read visually, it is reported as a titer. While these are

fairly sensitive assays, titers are always plus or minus one tube dilution, so a titer of 16 could

actually represent a titer of either 8 or 32. With latex-enhanced nephelometry or

turbidimetry, sensitivity ranges in the area of 30 to 50 ng/ml.

There are several problems that can affect both sensitivity and specificity. The first problem

affecting sensitivity is called the prozone effect (10, 27, 32). This refers to a lack of

agglutination due to an excessive amount of antibodies in the patient sample. The high

concentration of antibody inhibits agglutination, giving a false-negative result. This can be

easily overcome by simply diluting the sample. With regard to specificity, the major concern

is false-positive reactions from IgM rheumatoid factor (RF) (10, 17, 27, 32, 35). This occurs

most commonly in assays in which the latex beads are coated with IgG antibody. This has

been commonly reported for the latex agglutination test for cryptococcal antigen (27, 52).

IgM RF, which is specific for the Fc portion of the IgG molecule, binds and gives a falsepositive

reaction. RF has also been reported to bind to other serum proteins nonspecifically,

also giving a false-positive reaction. It is crucial that the clinician notify the laboratory if the

patient has a known RF. There are several measures that could be taken. First of all, if a

false-positive reaction is suspected, the sample result can be compared to the reaction using

control particles coated with normal IgG. If this indicates that there is a false-positive

reaction, the sample can be treated with a reducing agent such as 2-mercaptoethanol or it

can be treated with pronase. Both of these treatments have been shown to reduce the

majority of false-positive reactions due to IgM RF. Alternatively, the sample could be

pretreated with aggregated IgG to remove the IgM RF; however, this can result in loss of

antigen or specific antibodies and give a false-negative result. Also, an alternate test method

could be used for assessment of the ordered analyte. Most importantly, communication of

pertinent clinical information to the laboratory is critical to ensure the most accurate

diagnostic information.

CF Test

Another traditional immunoassay is the complement fixation (CF) test, which is based upon

the interaction of immune complexes with complement (32). As antigen- antibody complexes

form, the complement cascade is activated and complement components are “fixed” or

consumed. Conversely, if there is no antigen-antibody complex formation, there will be no

activation of the complement cascade. This two-step test is primarily used to determine the

titer of antibodies to specific antigens. For example, to set up a CF test for anti-Mycoplasma

pneumoniae antibodies, patient serum would be incubated with M. pneumoniae antigen and

a defined quantity of guinea pig complement. If the patient sample contains M.

pneumoniae antibody, immune complexes will form and fix the complement. RBC coated

with anti-RBC antibodies are then added to the tube. The final readout for the assay is the

release of hemoglobin from any lysed RBC. If the patient sample is positive for M.

pneumoniae, then there will be no complement remaining, so there will be no release of

hemoglobin. The opposite will occur if the patient sample is negative for anti-M.

pneumoniae antibody. Although CF assays are relatively sensitive and inexpensive, they can

be technically demanding and time-consuming. Therefore, many of these assays have been

converted to ELISA formats. However, a number of laboratories still use them as a

confirmatory test for the presence of antibodies to a variety of infectious agents, such

as Coccidioides, Histoplasma capsulatum, adenovirus, herpesvirus, influenza virus, M.

pneumoniae, and rickettsias.

Neutralization Assays

Neutralization assays are traditional laboratory tests used to determine if an antibody which

can neutralize the infectivity of a particular virus is present (32). The classic assay involves

mixing patient serum antibody samples with virus and then using this mixture to inoculate

either a cell line or a preparation of peripheral blood mononuclear cells. The readout involves

either the assessment of viral cytopathic effect in the cell line or some other measure of viral

replication, such as that obtained by a classic immunoassay of viral protein. For example, in

the case of HIV testing, one can perform a p24 antigen test or reverse transcriptase assay

and look for lower values. Evidence of decreased viral replication confirms the presence of

neutralizing antibody. Although these assays are relatively simple, they can be expensive

and can take days to complete. In addition, they can be difficult to standardize, especially

when comparing results from different laboratories. To decrease the assay length, the

quantitation of viral products can be assessed using PCR; however, this technique can be

expensive and also difficult to standardize between laboratories. A blocking ELISA can also

be performed, in which viral antigen and serum are mixed, after which a standard ELISA for

virus is performed and the decrease in the amount of antibody detected is assessed. An

additional traditional neutralization assay is reverse passive hemagglutination, as previously

discussed. In the field of HIV vaccine development, there is interest in developing new and

better assays for neutralization, since it is crucial in the assessment of vaccine efficacy to

demonstrate that a putative virus can initiate antibody production to prevent infection (37).

Immunofluorescence Assays

The immunofluorescence assay (IFA) uses a histochemical technique to detect either antigen

or serum antibody, utilizing a fluorescent-compound-labeled detector antibody (27, 32, 33).

There are two types of IFA, direct and indirect (Fig. 1). Direct assays are used to detect the

presence of antigens in tissue or body fluids. For example, to detect the presence of

influenza virus in a nasal wash specimen, it is applied to a slide, and then it is overlaid with a

fluorescent-compound-labeled anti-influenza virus antibody. If there is influenza virus

antigen present, there is emission of fluorescent light, which is evaluated with a fluorescence

microscope. Indirect assays are two-step tests used primarily for the detection of antibodies

in serum or a body fluid, although they can also be used for detection of viral antigens, such

as cytomegalovirus. The patient sample is applied to a slide containing the target antigen;

this is allowed to incubate, and specific antibody in the patient sample forms immune

complexes with the antigen present on the slide. Any unbound reagent is then washed away,

and the slide is overlaid with a fluorescent-compound-labeled anti-immunoglobulin. Positive

staining is the emission of fluorescent light. Overall, IFAs are useful tests that are relatively

easy to perform and inexpensive. In addition, they allow the localization of the antibody to a

specific antigen location in the tissue. For example, the IFA for antibody to Epstein-Barr virus

early antigen allows visualization of a specific pattern of staining of the virus-infected cell

line. The disadvantages of this assay are that it is relatively time-consuming and requires

both the purchase of an expensive fluorescence microscope and the presence of trained and

experienced personnel for interpretation. However, there are now available automated

analyzers which perform IFA slide processing, thus freeing up hours of preparatory time

(20).

Back to top

FIGURE 1 Direct and indirect IFAs.

Enzyme Immunoassays

EIAs are taking on increasingly more prominent roles in laboratory medicine

(9, 10, 25, 27, 33, 53, 54). They are found in all areas of the clinical laboratory, in

physicians’ offices, and in at-home testing and are being increasingly used in molecular

pathology laboratories. EIAs have taken the place of RIAs in the majority of laboratories, as

they offer comparable sensitivity without the problems of disposal and the short half-life

associated with radioactive materials. They are also replacing a variety of other techniques in

the laboratory, such as immunofluorescence and agglutination, because EIAs provide greater

objectivity, the potential to automate, and the ability to process large numbers of samples

with less hands-on technician time. As a single unit of enzyme label can amplify a reaction

product severalfold, many EIAs are optimized for detection at the pico- or attomole level.

EIAs can be broadly classified as either homogeneous or heterogeneous assays. In

homogeneous assays, the enzyme activity is altered as part of the immunologic reaction

itself. In these assays, there is no requirement to separate the bound from the free

immunoreactants. Although this technique is especially suited for the measurement of drugs

and haptens, homogeneous assays have not achieved widespread use in microbiology

laboratories. In contrast, heterogeneous immunoassays are widely used in microbiology. In

these assays, the enzyme activity of the labeled immunoreactant is not directly involved in

the immunologic reaction itself.

The basic principle of the heterogeneous EIA is the use of an antibody or antigen conjugated

with an enzyme which, upon reacting with its substrate, forms a measurable reaction

product. Often a color reaction product is produced. The color change is monitored visually or

with the use of a spectrophotometer to determine the proportionality between the amount of

color and the amount of analyte present. An essential component of these assays is the

separation of the bound enzyme-labeled component from the free labeled reagent. Assays

can be competitive or noncompetitive and can be used to measure antigens or antibodies.

The presence of all antibody isotypes can be quantitated depending on the specificity of the

antibodies used. Whenever antibody or antigen is absorbed to the solid phase, the assay is

referred to as an ELISA and also as a sandwich assay.

EIAs can be set up primarily as competitive or noncompetitive (9). Competitive assays most

commonly measure antigens and are set up with either antibody or antigen on the solid

phase. They are often termed limited-reagent methods because the antigens and antibodies

are used in measured and limited amounts. When the assay design uses specific antibody

with which the solid phase is coated, the patient sample containing the putative antigen and

the labeled antigen are added simultaneously and compete for binding to this matrix (Fig. 2).

It is critical that the avidity of the antibody for both the labeled and unlabeled antigens be

the same. In addition, a separate reaction is set up using enzyme-labeled antigen and buffer

alone, which are added to the antibody-adsorbed solid phase. The substrate for the enzyme

is added, and the color reaction is assessed. If the patient sample contains the antigen in

question, it will effectively compete for binding to the solid phase, thus preventing any

enzyme-labeled antigen from binding, giving no or minimal color. This reaction is compared

to a reaction well to which buffer alone is substituted for the patient sample. The separation

of the bound reactant from the free reactant is achieved through the washing steps. As is

true with all competitive assays, the amount of labeled immunoreactant detected through

the enzymatic reaction is inversely proportional to the amount of antigen present in the

sample.