HCG REFERENCE SERVICE
PRINCIPAL OF HCG TEST AND OF DISCORDANT RESULTS

 

A. PRINCIPALS

    Most hCG tests or pregnancy test used today, whether a home urine test, a physician's office urine or blood test, or a clinical laboratory blood test are "sandwich assays". Sandwich assays use at two or more animal antibodies raised against different sites on hCG. Usually a mouse monoclonal antibody against one site on the hCG molecule, and a mouse monoclonal, or a sheep, rabbit or a goat polyclonal antibody against a second distant site on the hCG molecule. One antibody, the capture antibody, is in a solid phase permanently attached to a tube, plate, membrane, or bead. The second antibody, the tracer antibody, is labeled with a dye, with radioactivity, with an enzyme, a lanthanide, chemiluminescence agent or other identifier. This antibody is in the liquid phase (SEE FIGURE 1, BELOW). Blood (serum or plasma) or urine is added to the assay system. After a short incubation period the hCG binds both the solid phase and liquid phase antibodies linking them. In this way it forms a sandwich or bridge between the solid support or capture antibody and the tracer antibody with the attached label. In this way the label becomes immobilized (SEE FIGURE 2, BELOW). After washing away the serum or plasma, and the excess tracer antibody, the amount of label attached to the solid support is measured. This is directly proportional to the amount of hCG (SEE FIGURE 3, BELOW).
    In a home pregnancy test or physician's office test, you have a disposable plastic device. It has an absorbent stem which is placed in flow of urine, or an opening/window in the plastic into which drops of urine are placed. The urine moves through the plastic device. In part of the device shielded by plastic the hCG (if present) attaches to the tracer or liquid phase antibody with a blue or red or gold color label. The hCG - tracer antibody complex continue to move though the absorbent material inside the device (nitrocellulose membrane) to an opening or window in the plastic. At this point the capture or solid phase antibody is anchored. It is translucent, and is immobilized in a circular or line shape. The capture or solid phase antibody then binds the hCG, forming a sandwich or bridge with the tracer antibody and the dye. A blue, red of gold line or circle appears, indicating that the sandwich or bridge is formed and the presence of hCG. This type of assay is qualitative (yes or no) and cannot be used to measure the actual levels of hCG.
    The clinical laboratory may use manual or automated test. In both types of test, multiwell plates, beads, solid supports or membranes or tubes are coated with the capture antibody. The labeled tracer antibody is in a buffer in the liquid phase. The serum or plasma is added, and hCG links the capture and tracer antibodies. The amount of tracer antibody captured (immobilized) is directly proportional to the amount of sandwich or the amount of hCG. The label is then measured using colorimetry, spectrometry, fluorimetry, chemiluminescence, or by radioactive or other detection methods. Results are determined for standard concentrations of hCG. Results for a patient's serum or plasma sample are compared to the standard hCG results, and recorded.

FIGURE 1. Device with solid phase capture antibody to one site on hCG, and liquid phase tracer antibody (label shown by red star) to second or distant site on hCG

FIGURE 2. Serum or urine containing hCG (shown as ab) added to device. The hCG forms a sandwich or bridge between capture and tracer antibody.

FIGURE 3. Excess tracer antibody is washed away. Amount of label or tracer (red star) is measured. This is proportional to amount of hCG.

 

B. DISCORDANT RESULTS WITH THE hCG TEST

    Over fifty different quantitative hCG assays (clinical laboratory blood tests) are sold in the United States. Over eight different antibody binding sites have been identified on different part of the hCG molecule. Each commercial test uses a different combination of capture and tracer antibodies which recognize any two of the eight different binding site. One problem with hCG assays is the heterogeneity of hCG, its different synthetic forms and degradation products (click here to see illustration of the synthesis and degradation of hCG). This problem is compounded by the wide variation in hCG concentrations observed in serum and urine samples throughout pregnancy (click here to see data on hCG levels during normal pregnancy). Assays that use different combinations of antibodies recognize different molecules. An assay, for instance, that uses an antibody against the hCG ab subunit interface as capture antibody and an antibody against the core of the b subunit as tracer antibody, will recognize normal pregnancy hCG (non-nicked hCG), but might not detect nicked or damaged hCG, an important component of hCG immunoreactivity in clearing pregnancies, trophoblast disease and cancer. Six major types of assay are noted. All six types are commonly used in clinical laboratories in the USA (reference 2, below). The first type uses antibodies that recognize non-nicked hCG only (normal pregnancy or non-damaged hCG), the second type recognizes nicked and non-nicked hCG, the third type recognizes non-nicked hCG and free b subunit only, the fourth type recognizes nicked and non-nicked hCG and free b subunit only, the fifth type does not recognize molecules missing the b subunit C-terminal peptide, and the sixth type recognizes all forms of b subunit (nicked and non-nicked hCG, free b subunit  and b subunit core fragment). The different forms of hCG in serum or plasma may be minor components of the total hCG immunoreactivity in most normal pregnancies. They can, however, be major components or the only hCG-related molecule in serum, plasma or urine in different stages of pregnancy, or in trophoblast disease or cancer (click here to see potential sources of hCG in and outside of pregnancy) (references 1-6, below). As such, two different commercial assays can, in certain circumstances give greatly varying hCG results, or an single assay can fail to detect a low level of hCG present in plasma, serum or urine sample.

   Other causes of discordant results include phantom or false positive hCG. Many cases of phantom or false hCG results have been now been identified by the USA hCG Reference Service.  In these case, human antibodies against human antibodies may cross cross-species (heterophilic antibodies) interfering with hCG test results. Similarly, human antibodies against animal antibodies may interfere with the animal antibodies used in hCG tests (click here to see the principal of false positive or phantom hCG tests). False positive or phantom hCG tests have led to the misdiagnosis of ectopic pregnancy, or to the erroneous assumption of post-gestational choriocarcinoma. A relatively large number of patient have been and are still being needlessly treated with chemotherapy and received a hysterectomy or other surgical procedures based solely on false or phantom hCG levels (see references 7-13, below) (click here to see potential sources of hCG in and outside of pregnancy). The USA hCG Reference Service experience with a large number of cases of women with false positive hCG results, and the needless chemotherapy or surgery the received, is now described in an attached article (hCG Reference Service Report 2002).

 

REFERENCES

1. Elliott M, Kardana A, Lustbader J, Cole L. Carbohydrate and peptide structure of the alpha- and beta-subunits of hCG from normal and aberrant pregnancy and choriocarcinoma. Endocrine 7:15-32, 1997.

2. Cole L. Immunoassay of hCG, its Free Subunits and Metabolites. Clin Chem 43:2233-2243, 1997.

3. Cole L. hCG, free beta subunit (free ß), free alpha-subunit (free alpha) and ß-core fragment (ß-core). Diagn. Endocrinol. Metab., 15:199-220, 1997

4. Cole L, Kohorn E, Kim G. Detecting and monitoring trophoblast disease: New perspectives in measuring hCG levels. J Reprod. Med. 39:193-200, 1994.

5. Alfthan H, Stenman U. Pathophysiological importance of various molecular forms of hCG. Mol Cell Endocrinol 125(1-2):107-20, 1996.

6. Birken S, Maydelman Y, Gawinowicz M, Pound A, Liu Y, Hartree A. Isolation and characterization of human pituitary chorionic gonadotropin. Endocrinology 137:1402-1411, 1996.