USA HCG REFERENCE SERVICE
HETEROPHILIC ANTIBODIES AND FALSE POSITIVE hCG TEST

  

HUMAN ANTI-ANIMAL AND HETEROPHILIC ANTIBODIES, THE CAUSE OF FALSE POSITIVE OR PHANTOM HCG.

    Heterophilic antibodies are cross-species antibodies. These are human antibodies that bind animal antibodies. All commercial hCG test used today involve rabbit, mouse, goat, horse or sheep antibodies.

   Commonly, an hCG test will involve 2 animal antibodies, one specific for binding one molecular site on hCG, and one specific for a distant site on hCG. Usually one animal antibody will be in a solid phase (bound to a vessel or to a bead) so will bind and immobilize hCG. The second animal antibody commonly has a dye, enzyme, chemiluminescence or other tracer or label attached (see FIGURE 1, below). When this antibody binds the hCG immobilized by the first antibody or solid phase antibody, it adds a tracer or tag to the immobilized complex (see FIGURE 2, below). After washing away unbound second antibody, the label or tag is measured and the amount is read (see FIGURE 3, below). The amount is directly proportion to the amount of complexes or the amount of hCG. In this case, hCG forms the meat of sandwich, between the immobilized antibody and the tracer antibody. The test measures specifically the meat, or hCG.

FIGURE 1. Device with solid phase capture animal antibody to one site on hCG, and liquid phase tracer animal 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.

 

    Unfortunately, hCG is not the only molecule that can form the meat or bridge between the solid phase animal antibody and tracer animal antibody. Human anti-animal immunoglobulin antibodies gained from human exposure to animals, and cross-species or human heterophilic antibodies can form the meat. Antibodies are always bivalent. That means that they can bind 2 things. An anti-animal antibody or heterophilic antibody can bind 2 animal antibodies. As such, they can link together the immobilized antibody and the tracer antibody, just like hCG can (see FIGURE 4, below). After washing away the excess tracer, the amount of label or tracer attached to the solid support is measured (see FIGURE 5, below), in an hCG test, the positive measurement is assumed to be hCG.
    In most instances hCG assays really measure hCG (>99% of cases). In which case the number measured is representative of the amount of hCG in the serum sample. When  heterophilic antibodies are present they can bind the animal antibodies used in the assay, also forming the sandwiches or links. This causes false positive of phantom hCG results.

FIGURE 4. Serum is added to the device containing capture (solid phase) and tracer (liquid phase) antibodies. The serum sample, instead of containing hCG contains small amounts of human anti animal antibodies or heterophilic antibodies. After a short incubation, the  antibodies form sandwiches or bridges between the capture and tracer antibodies.

FIGURE 5. The device is washed and excess tracer antibody is removed. The amount of tracer (label, shown by red stars) is measured. In true positive cases the amount of tracer is representative of the amount of hCG. In false positive or phantom cases it is representative of something less relevant, human anti-animal or heterophilic antibodies.

     Different hCG tests each have different propensities to be effected by human anti-animal antibodies and by heterophilic antibodies. Most tests, for instance, use one mouse monoclonal antibody and one sheep or goat polyclonal antibody. The use of different animal antibodies limits the possible effects of human anti-animal antibodies, such as a human anti-mouse antibody, for instance. This, however does not necessarily limit the effects of cross-species heterophilic antibodies. The effect of heterophilic antibodies can be limited by flooding the reaction mixture with non-specific animal antibodies or animal serum. In this way heterophilic antibodies will bind the excess of non-specific animal antibodies rather than the specific anti-hCG animal antibodies used in the hCG test. In many cases, however, this still is not sufficient. The effect of heterophilic antibodies can also be limited by antibodies against potential heterophilic antibodies. These include preparation like Scantibodies Inc. HBR.   

   False positive hCG results do not only occur in tests using this sandwich principal. They were first observed (Hussa et al 1985, see references below), and first became a problem with the old-style competitive immunoassays used in the 1960s and 1970s, and still used at a few centers today. These tests use a limited amount of one antibody preparation. They measure the competitive binding of an hCG with a tracer (commonly radioactivity) attached, and that in a serum sample. After incubation, the amount of tracer bound to the antibody is measured. The less the tracer, the more hCG in the sample. Human anti-animal antibodies and heterophilic antibodies can link the limited amount of antibody molecules together. This makes less antibody available for the tracer to bind. The lesser amount of bound tracer shows as a false positive result for hCG.

    People with human anti-animal antibodies or heterophilic antibodies in their blood can give false positive results in tests that can persist from months to many years. With few exceptions, most test give a result that is either in the normal positive range, or gives a higher than normal or lower than normal result. The hCG test is one of those few exceptions. It is either wholly positive, as in pregnancy, or wholly negative, as in the absence of pregnancy. As such, false positive hCG results falsely indicate the presence of pregnancy. Once pregnancy is ruled out, and ectopic pregnancy is excluded, cancer, specifically hCG-producing cancer, such as choriocarcinoma, is commonly considered as present. In many cases this is treated based solely on a continuously positive hCG test. It id for this reason, that a false positive hCG test created a major dilemma.

 

THE USA HCG REFERENCE SERVICE EXPERIENCE WITH FALSE POSITIVE OR PHANTOM hCG

Click on link below to show the most recent publications (requires Adode Acrobat) -

Cole LA, Dai D, Leslie KK, Butler SA, Kohorn EI. Gestational trophoblastic diseases: 4. Presentation with Persistent Low Positive Human Chorionic Gonadotropin Test Results

Cole LA, Khanlian SA. Easy Fix for Clinical Laboratories for the false positive defect with the Abbott AxSym total ß-hCG test. Clin Biochem, 2004; 37:344-349.

Cole LA, Khanlian SA Inappropriate Management of Women with Persistent Low hCG Results. J Reprod Med, 49: 423-432, 2004.

Cole LA, Sutton JM: Selecting an Appropriate hCG Test for Management of Gestational Trophoblastic Diseases and Cancer Cases. J Reprod Med, 49:545-553 2004.

     Table 1 summarizes the  USA hCG Reference Service experience with the 71 patients shown to have false positive hCG tests. Among these women, the average persistent hCG result was 102 ± 152 mIU/ml, with a range of 6.1 – 900 mIU/ml. The majority of false positive hCG results were from women having incidental pregnancy tests. Forty seven patients received unnecessary chemotherapy and 12 had needless surgery for what was later shown to be an hCG test problem. Five patients were being monitored after hydatidiform mole or GTN. Of the 71 patients, 49 (69% of cases) were managed by their center’s laboratory using the Abbott Axsym, 13 (18% of cases) using a Bayer analogous assays, 4 (6% of cases) using the Ortho Vitros, 3 (4% of cases) using the Beckman Access and 1 (1% of cases) each using the Dade Dimension RXL and the Tosoh AIA quantitative automated total hCG test. It is noteworthy, that a high proportion of cases observed in the past 2 years have been using the Bayer analogous assays.

Table 1. The 71 false positive hCG patients that consulted with the USA HCG Reference Service.  All hCG data are hCG Reference Service serum results. Therapy abbreviations- methotrexate (Mtx); actinomycin D (ActD); etoposide + methotrexate + actinomycin D, alternating with cyclophosphamide + vincristine (EMA-CO); hysterectomy (Hys); salpingo-oophorectomy (SalOoph); thoracotomy (Tho). GTN is gestational trophoblastic neoplasm. Abbreviations for hCG tests- Abbott AxSym test (Abbott); Beckman Access (Beckman);  three analogous Bayer tests, Bayer ACS180, Bayer ADVIA Centaur and Bayer ACS180 (Bayer); Ortho Vitros (Ortho); Dade Dimension RXL (Dade); Tosoh A1A (Tosoh).

 

PPhysician’s diagnosis	hCG tests	hCG	Previous history	Ineffective chemotherapy
aaccording to records		mIU/ml a	(prior 6 months)	or surgery
GGTN	Abbott b	900	None	Mtx
GGTN	Abbott x 3 c	600	None	Mtx, ActD, Hys, EMA-CO, Tho
GGTN	Abbott	500	None	Mtx
CChoriocarcinoma	Abbott	467	None	Mtx, ActD, EMA-CO, Hys
GGTN	Abbott	350	Parturition	Mtx
GGTN	Abbott	275	None
None	Abbott	271	None
GGTN	Abbott	218	None	Mtx
GGTN	Abbott	205	None
GGTN	Abbott	200	None	Mtx
Nnone	Abbott	174	None
GGTN	Abbott	160	None	Mtx, ActD
GGTN	Beckman	153	None	Mtx, EMA-CO
Nnone	Abbott	151	None	Mtx
GGTN	Abbott	143	None
GGTN	Abbott	142	Miscarriage	Mtx, ActD, Hys
GGTN	Abbott	139	None	Mtx
EEctopic Pregnancy	Abbott	122	None	Mtx
Nnone	Bayer	117	None	Mtx
GGTN	Ortho	100	Miscarriage	Hys
GGTN	Abbott	97	None	Mtx
GGTN	Abbott	83	None	Mtx
GGTN	Abbott	81	None	Mtx
GGTN	Abbott	81	None	Mtx, ActD, Hys, SalOoph
CChoriocarcinoma	Abbott	80	Choriocarcinoma	Etop
GGTN	Abbott	80	None	Mtx, EMA-CO
CChoriocarcinoma	Abbott	78	None	Mtx, Hys, EMA-CO
GGTN	Beckman	78	None
GGTN	Abbott	74	None
GGTN	Abbott	60	None
GGTN	Abbott	60	None
GGTN	Abbott	57	None	Mtx
Nnone	Ortho	55	Parturition
PPersistent Mole	Abbott	53	Complete Mole
Nnone	Bayer	50	None
CChoriocarcinoma	Abbott	47	None	Mtx, ActD, EMA-CO, Hys
GGTN	Abbott	42	None	Mtx, ActD
GGTN	Ortho	41	Complete Mole	Mtx, ActD
GGTN	Abbott	40	None	Mtx
GGTN	Abbott	37	Miscarriage	Mtx,  Mtx
PPersistent Mole	Abbott	35	Complete Mole	Mtx, Hys
GGTN	Abbott	33	None	Mtx
GGTN	Bayer	32	None	Mtx
GGTN	Bayer	30	None	Mtx
GGTN	Abbott	26	None	Mtx
Nnone	Beckman	25	None	Mtx
GGTN	Abbott	24	None	Hys
GGTN	Abbott	23	None	Mtx
GGTN	Dade	23	Miscarriage	Mtx
GGTN	Abbott	22	None	Mtx, ActD
PPersistent Mole	Abbott	21	Partial Mole	Mtx, AcD
GGTN	Bayer	20	None	Mtx
Nnone	Bayer	19	None
GGTN	Abbott	18	None	Mtx
GGTN	Abbott	17	None	Mtx, Hys
Nnone	Abbott	16	None
GGTN	Abbott	14	GGTN
Nnone	Bayer	13	Miscarriage
Nnone	Abbott	12	None	Hys, SalOoph
Nnone	Bayer	12	None
RResidual trophoblast tissue	Abbott	12	Miscarriage
GGTN	Bayer	12	None	Mtx
GGTN	Abbott	11	None	Mtx
GGTN	Bayer	11	None	Mtx
Nnone	Abbott	10	None	Mtx
GGTN	Ortho	9	None
GGTN	Bayer	9	GTN	Mtx
Nnone	Bayer	8.5	None
Persistent Mole	Abbott	8	Complete Mole	Mtx, ActD, EMA-CO, Hys
GGTN	Tosoh	7.1	Choriocarcinoma
Nnone	Bayer	6.1	None

a hCG results at the time of or immediately prior to referral to the USA hCG Reference Service.

b Of the 71 patients, 49 were managed by their laboratory using the Abbott Axsym, 13 using a Bayer analogous assay, 4 using the Ortho Vitros, 3 using the Beckman Access and 1 each using the Dade Dimension RXL and the Tosoh AIA quantitative automated total hCG test.

c Prior to hysterectomy hCG was tested at 3 independent laboratories, all gave the same false positive hCG result since all used the Abbott AxSym test.

Care is needed to ensure treating only patients with true positive hCG results. Find our what test is being run in the laboratory. If it is the Abbott AxSym, Bayer Centaur or Bayer ASC180 test or a radioimmunoassay. One that clearly has a propensity to give an unduly large number of false positive hCG results (as shown in Table 1 and in the publications), it is important that the accuracy of the test result be checked before starting any therapy. To check for false positive hCG results we recommend -
1. Demonstrating with a quantitative hCG assay (sensitivity 2 mIU/ml), the presence of hCG in serum and urine. Normally, human antibodies to not enter the urinary tract, so will not interfere in a urine hCG test.
2.  Showing parallel results when serum in 2- and 10-fold diluted. Interfering antibodies in some incidences may give different responses at different serum dilutions. 
3. Showing that treatment with Scantibodies Inc. HBR heterophilic antibody blocking antibody does not significantly reduce the hCG test result.
4. Showing very similar (within 20%) hCG results in 3 different manufacturer's hCG test.  

A LOT OF WOMEN HAVE BEEN NEEDLESSLY HURT OR LOST THEIR ABILITY TO HAVE CHILDREN BECAUSE OF FALSE POSITIVE hCG TESTS, IT IS IMPORTANT TO TAKE PRECAUTIONS AND AVOID NEEDLESS THERAPY.

 

REFERENCES

1.     Cole LA, Sutton JM: Selecting an Appropriate hCG Test for Management of Gestational Trophoblastic Diseases and Cancer Patients. J Reprod Med, 2004:49:545-553 2004.

2.     Khanlian SA, Smith HO, Cole LA. Persistent low levels of hCG: A pre-malignant gestational trophoblastic disease. Am J Obstet Gynecol, 2003; 188:1254-59.

3.     Cole LA, Khanlian SA Inappropriate management of women with persistent low hCG results. J Reprod Med 2004; 49: 423-432.

4.     Cole, L.A. and Butler S.A., hCG, its Free Subunits and Metabolites in Trophoblastic Diseases. J. Reprod. Med. 47:433-444, 2003.

5.     Cole, LA, Sutton JM. hCG tests in the management of gestational trophoblastic diseases. Clin Obstet Gynecol 2003; 46: 533-540.

13.   Cole LA, Shahabi S, Butler S, Mitchell H, Newlands ES, Behrman HR, Verrill HL. Utility of commonly used commercial hCG immunoassays in the diagnosis and management of trophoblastic diseases. Clin Chem 2001;47:308-315, 2001.

14.   Vladutiu AO, Sulewski JM, Pudlak KA, Stull CG. Heterophilic antibodies interfering with radioimmunoassay. J Am Med Assoc 1982;248:2489-90.

15.  Hussa RO, Rinke ML, Schweitzer PG. Discordant human chorionic gonadotropin results: causes and solutions. Obstet Gyncol 1985;65:211-219.

16.   Garrett PE, Kutz SR, Hurd JK. False positive results for choriogonadotropin in serum Clin Chem 1983;29:1871-73.