Are laboratories reporting serum quantitative hCG results correctly?
There are > 100 different commercial immunoassay test kits for measuring hCG. Most procedures detect intact hCG (nicked + nonnicked hCG) or hCG + hCG[beta], and a few immunoassays detect free hCG[beta] alone or with other forms (4). In clinical laboratories in the US, fewer than 20 immunoassay methods are commonly used. These assays vary in immunospecificity toward different forms of hCG, and results of measurements may have distinct clinical applications. It is important that laboratorians, who perform the test and consult with physicians about test procedures and results, be aware of the heterogeneity of hCG and the specificity of the method in service.
The Wadsworth Center of the New York State Department of Health conducts CLIA'88-approved proficiency testing programs, and CLIA'88 regulations define acceptable performance criteria for qualitative and quantitative test results for hCG (14). Based on information from manufacturers' product inserts for the packages of test kits commonly used by clinical laboratories in the US, the method principles, and the occurrence of test kit names in peer-reviewed articles, we request participant laboratories to report results as "Intact hCG" and/or "Total [beta] hCG" for quantitative results of hCG as appropriate.
In the proficiency test event of January 2006, we prepared 1 of 5 specimens by addition of intact hCG and free hCG[beta] to final concentrations of 50 IU/L and 20 IU/L, respectively, as well as other non-hCG-related hormones, into human serum matrix that had essentially no endogenous hCG. Intact hCG, WHO standard (code 75/589), was purchased from the National Institute for Biological Standards and Control, and free hCG[beta] was from AspenBio Pharma. The addition of both intact hCG and free hCG[beta] was intended to mimic pathophysiological conditions and to challenge the specificity of various immunoassays. Along with 4 other proficiency test specimens prepared by the sole addition of intact hCG, the samples were delivered overnight on ice to participant laboratories. Results of total [beta] hCG were reported by 266 laboratories. (Complete details on results from the proficiency test events discussed in this paper are available at http://www.wadsworth.org/chemheme/chem/endo/ptframes.htm.)
[FIGURE 1 OMITTED]
Total [beta] hCG results for specimens with both intact hCG and hCG[beta] showed higher overall variability (16% CV) than the samples spiked with only intact hCG (9.6% CV). We also noticed a bimodal distribution of results reported by users of the Roche Elecsys system for total [beta] hCG, with the specimen containing both forms of hCG: 2 medians were centered at approximately 47 IU/L and 67 IU/L (Fig. 1). In contrast, we found normal distributions of results for the other 4 specimens with this method. Further investigation revealed that 2 different reagent systems are provided for the Roche Elecsys instrument, reagent "HCG STAT" for measuring intact hCG and reagent "HCG + [beta]" for measuring total [beta] hCG (hCG + hCG[beta]). Those laboratories that reported results of approximately 47 IU/L used the reagent HCG STAT, and those that reported values of approximately 67 IU/L used the reagent HCG + [beta]. Unfortunately, many laboratories that used the HCG STAT procedure reported their results as total [beta] hCG.
We carried out further studies in the proficiency test event of May 2007. We prepared sample E03 by the sole addition of intact hCG at approximately 45 IU/L; sample E04 by the sole addition of free hCG[beta] at approximately 30 IU/L; and sample E05 by spiking with both intact hCG (approximately 40 IU/L) and free hCG[beta] (approximately 20 IU/L). These preparations, plus other non-hCG related analytes, were added to an essentially hCG-free human serum matrix. In sample E04, the intact hCG values should be undetectable, since no intact hCG was added, and total [beta] hCG values should be close to the calculated concentration of 30 IU/L. Results for total [beta] hCG were reported by 235 participant laboratories using 12 FDA-approved immunoassay methods and ranged from < 1 to 49.6 IU/L for sample E04 (Table 1). In contrast, results ranged from 33.3 to 53.9 IU/L for sample E03 and 40 to 83 IU/L for sample E05. Results for intact hCG were reported by 61 laboratories using 7 immunoassay instruments and ranged from 1 to 50 IU/L for sample E04, in comparison to results ranging from 34 to 53.3 IU/L for sample E03 and 35 to 77.2 IU/L for sample E05 (Table 1). Investigation of the sources of the discrepancy revealed that the Dade Behring Dimension Flex reagent cartridge hCG detects only the intact hCG molecule, and therefore the hCG values measured on this instrument were close to zero in sample E04, close to 45 IU/L in sample E03, and about 40 IU/L in sample E05. However, 20 of 65 users of this method reported results as total [beta] hCG, rather than the correct intact hCG. Errors were observed with the following 3 automated immunoassay instruments, which detect both intact hCG and free hCG[beta] (total [beta] hCG or hCG + hCG[beta]): Bayer Advia Centaur (Total hCG), Vitros ECi/ECiQ (Total [beta]-hCG), and Siemens Immulite series (Immulite 1000 HCG and Immulite 2000 HCG). For these instruments, however, a total of 8 laboratories reported results as intact hCG. Two Elecsys HCG STAT users also reported intact hCG results as total [beta] hCG (Table 1).
In addition, the recovery of total [beta] hCG differed among these immunoassay methods (Table 1). The variation is likely due to the difference in immunoreactivity of these immunoassays toward hCG[beta] and/or differences in calibration.
CLIA requires that proficiency test specimens be treated in the same manner as patient specimens (14). Erroneous reporting of proficiency test results will thus reflect errors in reporting of patient results that could cause misdiagnosis and mismanagement of trophoblastic diseases, although such errors may not affect the diagnosis of normal pregnancy, in which both intact hCG and total [beta] hCG increase dramatically (4, 5). Given the overall reporting error rate of 13.1% (8 of 61) for intact hCG and 9.3% (22 of 235) for total [beta] hCG, the occurrence of such errors is likely to be a significant problem worldwide. Such errors also occur in the literature; for instance, publications have reported "total hCG" determined on the Dade Dimension instrument and "intact hCG" measured on the Immulite system (15, 16).
The utility of monitoring hCG and its various forms reaches beyond the detection of pregnancy to various disease conditions, e.g., ectopic pregnancy, choriocarcinoma, testicular cancer, trophoblastic diseases, and nontrophoblastic tumors (2, 3, 5, 7, 9-12, 17). National Academy of Clinical Biochemistry (NACB) and European Group on Tumor Markers (EGTM) guidelines emphasize that both intact hCG and free hCG[beta] be detected by assay kits for tumor markers (18, 19).
To provide reliable services, laboratorians and clinicians must know which forms of hCG the method in service detects. We conclude that at least 4 factors have contributed to the incorrect reporting of hCG results by laboratories: (a) the complexity of the hCG molecule and confusion of nomenclature on the various forms of hCG; (b) laboratory personnel's lack of awareness of the distinctions of the forms of hCG and failure to recognize the specificity of assays for their measurement; (c) manufacturers' lack of clarity and uniformity in reagent labeling; and (d) most product inserts' lack of information on the specificity of each method to the various forms of hCG.
Solving this problem will require involvement of multiple parties in reaching a consensus on nomenclature and clarity of information provided in package inserts. The availability of new WHO reference materials for hCG and its 5 other forms makes this task feasible (20). An urgent need is for manufacturers to clearly state what forms of hCG are detected by each assay. Appropriate labeling and information on assay specificity should then be provided to the user, since current labeling is both inadequate and confusing (4). Furthermore, it has been noted that when hCG[beta] is measured by assays detecting hCG and hCG[beta] together, the results are erroneous as they are based on the IU for hCG; it has been recommended that assay procedures report results in molar concentrations (5).
Grant/Funding Support: None declared.
Financial Disclosures: None declared.
(1.) Pierce JG, Parsons TF. Glycoprotein hormones: structure and function. Ann Rev Biochem 1981; 50:465-95.
(2.) Young RC. Gynecologic malignancies. In: Kasper DL, Braunwald E, Fauci AS, Hauser SL, Longo DL, Jameson JL, Isselbacher KJ, eds. Harrison's Principles of Internal Medicine. 16th ed. New York: McGraw-Hill, 2007;553-8.
(3.) Motzer RJ, Bosl GJ. Testicular cancer. In: Kasper DL, Braunwald E, Fauci AS, Hauser SL, Longo DL, Jameson JL, Isselbacher KJ, eds. Harrison's Principles of Internal Medicine. 16th ed. New York: McGraw-Hill, 2007;550-3.
(4.) Cole LA. Immunoassay of human chorionic gonadotropin, its free subunits, and metabolites. Clin Chem 1997;43:2233-43.
(5.) Stenman UH, Tiitinen A, Alfthan H, Valmu L. The classification, functions and clinical use of different isoforms of HCG. Hum Reprod Update 2006; 12:769-84.
(6.) Thomas CMG, Reijnders FJL, Segers MFG, Doesburg WH, Rolland R. Human choriogonadotropin (hCG): comparisons between determinations of intact hCG, free hCG [beta]-subunit, and "total" hCG + [beta] in serum during the first half of high-risk pregnancy. Clin Chem 1990;36:651-5.
(7.) Fan C, Goto S, Furuhashi Y, Tomoda Y. Radioimmunoassay of the serum free beta-subunit of human chorionic gonadotropin in trophoblastic disease. J Clin Endocrinol Metab 1987;64:313- 8.
(8.) Rinker AD, Tietz NW. [beta]-hCG vs intact hCG assays in the detection of trophoblastic disease. Clin Chem 1989;35:1799-800.
(9.) Cole LA, Shahabi S, Butler SA, Mitchell H, Newlands ES, Behrman HR, Verrill HL. Utility of commonly used commercial human chorionic gonadotropin immunoassays in the diagnosis and management of trophoblastic diseases. Clin Chem 2001;47:308-15.
(10.) Cole LA, Khanlian SA, Muller CY, Giddings A, Kohorn E, Berkowitz R. Gestational trophoblastic diseases: 3. Human chorionic gonadotropin-free [beta]-subunit, a reliable marker of placental site trophoblastic tumors. Gynecol Oncol 2006;102: 160-4.
(11.) Sailer B, Clara R, Spottl G, Siddle K, Mann K. Testicular cancer secretes intact human choriogonadotropin (hCG) and its free [beta]-subunit: evidence that hCG (+ hCG- [beta]) assays are the most reliable in diagnosis and follow-up. Clin Chem 1990;36: 234-9.
(12.) Cole LA, Sutton JM. Selecting an appropriate hCG test for managing gestational trophoblastic disease and cancer. J Reprod Med 2004;49:545-53.
(13.) Palomaki GE, Lambert-Messerlian GM, Canick JA. A summary analysis of Down syndrome markers in the late first trimester. Adv Clin Chem 2007; 43:177-210.
(14.) US Department of Health and Human Services. Medicare, Medicaid and CLIA programs: regulations implementing the Clinical Laboratory Improvement Amendments of 1988 (CLIA), final rule. Fed Regist 1992;57:7002-186.
(15.) Massart C, Lucas C, Rioux-Leclercq N, Fergelot P, Pouvreau-Quillien V, Volatron AC. Importance of the detection method for intact dimeric human chorionic gonadotropin without interference with the free human chorionic gonadotropin [beta] subunit for pregnancy exclusion before liver transplantation in a woman with cholangiocarcinoma. Clin Chem Lab Med 2005;43:886-9.
(16.) Hallahan T, Krantz D, Orlandi F, Rossi C, Curcio P, Macri S, Larsen J, Buchanan P, Macri J. First trimester biochemical screening for Down syndrome: free beta hCG versus intact hCG. Prenat Diagn 2000;20:785-9.
(17.) van Trommel NE, Sweep FC, Schijf CP, Massuger LF, Thomas CM. Diagnosis of hydatidiform mole and persistent trophoblastic disease: diagnostic accuracy of total human chorionic gonadotropin (hCG), free hCG [alpha]- and [beta]-subunits, and their ratios. Eur J Endocrinol 2005;153:565-75.
(18.) Lamerz R, Albrecht W, Bialk P, Dati F, Duffy MJ, Gerl A, et al. Tumour markers in germ cell cancer: EGTM recommendations. Anticancer Res 1999; 19:2785- 820.
(19.) Sturgeon C. Practice guidelines for tumor marker use in the clinic. Clin Chem 2002;48:1151-9.
(20.) Birken S, Berger P, Bidart JM, Weber M, Bristow A, Norman R, Sturgeon C, Stenman UH. Preparation and characterization of new WHO reference reagents for human chorionic gonadotropin and metabolites. Clin Chem 2003;49:144 -54.
Zhimin (Tim) Cao,  * Robert Rej [1,2]
 Division of Molecular Medicine, Wadsworth Center for Laboratories and Research, New York State Department of Health, Albany, NY; 2 Department of Biomedical Sciences, School of Public Health, University at Albany, Albany, NY; * address correspondence to this author at: Division of Molecular Medicine, Wadsworth Center for Laboratories and Research, New York State Department of Health, Albany, NY 12201-0509. Fax 518-474-9185; e-mail email@example.com.
Table 1. Summary of laboratory responses for quantitative hCG measurements. Sample E03 (only intact hCG added) Instrument, label Intact HCG, Total [beta] HCG, of reagent name IU/L IU/L % Abbott Architect, total 35.9 (2.4) 80 [beta]-hCG (n = 7) Abbott AxSym, total 46.8 (2.7) 104 [beta]-hCG (n = 4) Bayer ADVIA:Centaur, 42, 44 (n = 2) 42.0 (2.5) 93 Total hCG (n = 60) Bayer ACS:180, total hCG 45.0 (1.0) 100 (ThCG) (n = 8) Beckman Coulter UniCel 34.6 (1.6) 77 DxI 800, total [beta]hCG (n = 17) Beckman Coulter Access, 33.3 (1.8) 74 total [beta]hCG (n = 47) Diagnostic Systems 43 (n = 1) Laboratories, intact hCG ELISA Dade Behring Dimension, 46.5 (1.7) 46.3 (1.9) Flex reagent cartridge (n = 45) (n = 20) [beta]hCG Dade Behring Stratus CS, 42.3 (1.4) 94 [beta]HCG TestPak (n = 3) Vitros ECi/ECiQ, total 43 (n = 1) 43.6 (2.5) 97 [beta]-hCG (n = 12) Roche Elecsys, Cobas and 42.6 (3.1) 41, 43 Modular, HCG STAT (n = 5) (n = 2) Roche Elecsys, Cobas and 41.3 (2.7) 92 Modular, HCG + [beta] (n = 24) Siemens IMMULITE series, 53.3 (3.7) 53.9 (4.3) 120 HCG (n = 5) (n = 31) BioMerieux Vidas, VIDAS 34, 41 HCG (n = 2) Target, based on 45 45 100 additions Sample E04 (only intact hCG[beta] added) Instrument, label Intact HCG, Total [beta] HCG, of reagent name IU/L IU/L % Abbott Architect, total 18.6 (2.1) 62 [beta]-hCG (n = 7) Abbott AxSym, total 36.5 (1.2) 122 [beta]-hCG (n = 4) Bayer ADVIA:Centaur, 23, 24 (n = 2) 21.5 (1.6) 72 Total hCG (n = 60) Bayer ACS:180, total hCG 22.3 (0.9) 74 (ThCG) (n = 8) Beckman Coulter UniCel 36.8 (1.5) 123 DxI 800, total [beta]hCG (n = 17) Beckman Coulter Access, 35.7 (2.2) 119 total [beta]hCG (n = 47) Diagnostic Systems 1 (n = 1) Laboratories, intact hCG ELISA Dade Behring Dimension, 1 (0.0) 1.5 (0.7) Flex reagent cartridge (n = 45) (n = 20) [beta]hCG Dade Behring Stratus CS, 29.0 (0.9) 97 [beta]HCG TestPak (n = 3) Vitros ECi/ECiQ, total 50 (n = 1) 48.8 (2.9) 160 [beta]-hCG (n = 12) Roche Elecsys, Cobas and <3.3 (n = 5) <1, 1.0 Modular, HCG STAT (n = 2) Roche Elecsys, Cobas and 30.7 (1.9) 102 Modular, HCG + [beta] (n = 24) Siemens IMMULITE series, 48.9 (3.0) 49.6 (6.4) 165 HCG (n = 5) (n = 31) BioMerieux Vidas, VIDAS 2, 2 (n = 2) HCG Target, based on 1 30 100 additions Sample E05 (only intact hCG and hCG[beta] added) Instrument, label Intact HCG, Total [beta] HCG, of reagent name IU/L IU/L % Abbott Architect, total 48.0 (2.2) 80 [beta]-hCG (n = 7) Abbott AxSym, total 67.2 (4.9) 112 [beta]-hCG (n = 4) Bayer ADVIA:Centaur, 55, 56 (n = 2) 53.8 (2.8) 90 Total hCG (n = 60) Bayer ACS:180, total hCG 56.7 (2.8) 95 (ThCG) (n = 8) Beckman Coulter UniCel 59.2 (3.7) 99 DxI 800, total [beta]hCG (n = 17) Beckman Coulter Access, 58.6 (2.9) 98 total [beta]hCG (n = 47) Diagnostic Systems 35 (n = 1) Laboratories, intact hCG ELISA Dade Behring Dimension, 40.8 (1.6) 40.8 (1.7) Flex reagent cartridge (n = 45) (n = 20) [beta]hCG Dade Behring Stratus CS, 59.7 (1.4) 100 [beta]HCG TestPak (n = 3) Vitros ECi/ECiQ, total 71 (n = 1) 68.7 (4.9) 115 [beta]-hCG (n = 12) Roche Elecsys, Cobas and 42.8 (2.3) 40, 43 (n = 2) Modular, HCG STAT (n = 5) Roche Elecsys, Cobas and 59.6 (3.6) 99 Modular, HCG + [beta] (n = 24) Siemens IMMULITE series, 77.2 (4.9) 83.0 (9.5) 138 HCG (n = 5) (n = 31) BioMerieux Vidas, VIDAS 40, 44 (n = 2) HCG Target, based on 40 60 100 additions Results are expressed as mean (SD) or each result when n < 3. Percentages of calculated target values are based on addition of known quantities of hCG. Results of intact hCG and that of erroneously reported as total [beta] hCG are not calculated for % recovery. Complete details on results from the proficiency test events discussed in this paper are available at http://www.wadsworth.org/ chemheme/chem/endo/ptframes.htm.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Brief Communications|
|Author:||Cao, Zhimin "Tim"; Rej, Robert|
|Date:||Apr 1, 2008|
|Previous Article:||Multiplexed assays for detection of mutations in PIK3CA.|
|Next Article:||Evaluation of a prolonged prothrombin time.|