Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

Contents lists available at ScienceDirect

Best Practice & Research Clinical Endocrinology & Metabolism journal homepage: www.elsevier.com/locate/beem

4

Determination of human chorionic gonadotropin Ulf-Håkan Stenman, M.D., Ph.D., Professor Emeritus a, *, Henrik Alfthan, Ph.D. b a b

Department of Clinical Chemistry, Helsinki University, Biomedicum, PB 63, FIN-00014 Helsinki, Finland HUSLAB, Helsinki University Central Hospital, PB 600, HUS, FIN-00029 HUS, Finland

Keywords: human chorionic gonadotropin hCG hCGb hCGbcf pregnancy test trophoblastic cancer testicular cancer preeclampsia epitopes

Determination of human chorionic gonadotropin (hCG) is used for diagnosis and monitoring of pregnancy, pregnancy related disorders, for trophoblastic and some nontrophoblastic tumors. In addition, hCG is determined for doping control in males. Assay of hCG is complicated by the occurrence of different molecular forms, which are detected to various degrees by different assays. The main form of hCG in circulation and in patients with trophoblastic tumors is intact heterodimeric hCG. The free b subunit (hCGb) is a minor form in plasma in both conditions, but it may be the major form aggressive trophoblastic cancer. Therefore, assays measuring hCG and hCGb together are mainly used for diagnosis of pregnancy and trophoblastic diseases. When excreted into urine, most of hCG (and hCGb) is broken down to the core fragment of hCGb (hCGbcf), which is the main immunoreactive form of hCG in urine during pregnancy. Specific determination of hCGb is of value in screening for Down’s syndrome and diagnosis of nontrophoblastic cancer. hCGbcf is of limited utility but it is important because it may disturb assay of hCG in pregnancy. Ó 2013 Elsevier Ltd. All rights reserved.

Introduction Human chorionic gonadotropin (hCG) is a placental hormone necessary for the maintenance of pregnancy. Already in early pregnancy, trophoblasts secrete large amounts of hCG and determination of the concentrations in serum and urine is used to detect pregnancy and pregnancy related disorders.

* Corresponding author. E-mail address: ulf-hakan.stenman@helsinki.fi (U.-H. Stenman). 1521-690X/$ – see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.beem.2013.10.005

784

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

hCG and its subunits are also produced by virtually all trophoblastic tumors and most germ cell tumors of the gonads [1]. Determinations of hCG and its b subunit (hCGb) are very important for monitoring of these diseases. In addition, hCGb is also produced at low concentrations by many nontrophoblastic cancers, for which hCG is an independent prognostic marker [2]. hCG is also produced at low concentrations by the pituitary and can be measured in serum of healthy subjects by highly sensitive assays. In women, the serum concentrations increase after the menopause to levels similar to those in very early pregnancy. For diagnosis of pregnancy, hCG is often measured in urine with pregnancy tests [3]. Urine mostly contains an excess of a degradation products, i.e., the core fragment of hCGb (hCGbcf) that can disturb the assay of hCG [4]. hCG structure hCG consists of two noncovalently linked subunits, hCGb and hCGa. The latter is common to all glycoprotein hormones (GPHs), luteinizing hormone (LH), follicle stimulating hormone (FSH) and thyroid stimulating hormone (TSH). Hence it is also called GPHa. The b subunit of hCG and LH are highly homologous and LH and hCG activate the same LH/CG receptor. hCGb contains 145 while LHb contains 120 amino acids (aa). The C-terminal peptide comprising aa 121–145 is unique to hCG while the core regions of LH and hCG comprising aa 1–120 are highly similar, the homology between these being about 85%. Therefore, many antibodies to hCG also recognize LH and vice versa. About 30% of the weight of hCG consists of carbohydrates. hCGb carries two N-linked glycans at Asn13 and Asn30 and four O-linked glycans linked to Ser121, Ser127, Ser132 and Ser138. The a subunit contains 92 amino acids and two N-linked carbohydrate chains on Asn52 and Asn78. The carbohydrate structure of hCG produced by various tissues may vary considerably [1]. hCG produced in early pregnancy and in cancer contains larger and more highly branched and more extensively sialylated glycans than that produced in mid and late pregnancy [5–7]. Forms of hCG in biological fluids Different molecular forms of hCG occur in circulation and in urine, i.e., intact hCG, nicked hCG (hCGn), hCGb and nicked hCGb (hCGbn), the core fragment of hCGb (hCGbcf) and hCGa [1]. Standards for these six variants have been prepared and initially established as International Reference Reagents (IRR) and the hCG preparation as the 5th International Standard (IS) by the WHO (Table 1). These preparations have been calibrated in substance concentration, i.e., mol but conversion factors to IU have been established by immunoassay [8,9]. When determined by immunoassays, the expression of hCG concentrations in units based on bioactivity is problematic. Immunoassays do not measure bioactivity but the concentrations of epitopes recognized by the antibodies used in the immunoassay. Thus, partially degraded forms of hCG, e.g., hCGn, hCGb and hCGbcf, which lack bioactivity are recognized by many immunoassays [10–12]. Thus, when new and more pure international standards (IS) are introduced, the ratio of bioactivity to mass increases. In the 3rd (and identical 4th IS) the ratio of bioactivity to mass is 9200 IU/mg while in the most recent International Research Preparation (IRP 99/ 688) it is 10,300–15,400 [13]. The ratio between immunoreactivity and bioactivity is also changing with new hCG standards and the result is dependent on the immunoassay used. The 1st IRP has recently been adopted as the 5th IS for hCG. A conversion factor of 12,240 IU/mg has been determined by immunoassay.

Table 1 Major variants of hCG their abbreviations and MWs.

Human chorionic gonadotropin b subunit of hCG Core fragment of hCGb hCGa

Abbreviation

MW

pmol/IU

hCG hCGb hCGbcf hCGa

37,500 23,000 13,000 14,000

2.9 42.5 71.4

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

785

Because hCGb lacks biological activity, its units are very different from those of hCG. Thus 1 IU of the 3rd IS for hCGb corresponds to 1 mg, but this standard is seldom used in practice [10]. If hCGb is measured by an assay measuring hCG and hCGb, the results are expressed in IU/L based on the 3rd IS for hCG. However, the response of hCGb in hCG assays is very variable, some methods underestimating while other ones overestimate it [11,12]. Thus results obtained by different assays are seldom comparable. Because of this problem, the IRPs for hCG and related molecules were calibrated in mol/L. When expressed in these units, results for hCG, hCGb and hCGbcf are comparable if the assays have been properly calibrated [8,10]. hCG epitopes The glycoprotein hormone (GPH) subunits, hCGa and hCGb share structural homology with members of the cystine knot growth factor superfamily [14]. Each subunit contains a cysteine knot motif consisting of two disulfide bonds that link adjacent anti-parallel strands of the peptide chain forming a ring that is permeated by a third disulfide bond. The central cystine knot determines the three-dimensional structure of hCGa and hCGb. Two neighboring hairpin-like peptide loops 1 and 3 protrude from one side of the knot. In hCGb the loops are stabilized by a disulfide bond between Cys 23 and Cys72 while loop 2 protrudes from the opposite side of the knot. The subunits are non-covalently linked in an antiparallel fashion, loops 1 & 3 of one subunit being adjacent to loop 2 of the other subunit [14]. On each subunit, loops 1 and 3 of and the cystine knot on hCGb are dominating antigenic regions [15]. Antibodies recognizing epitopes on hCGb are mainly used for design of hCG assays because this ascertains lack of crossreactivity with LH [15]. However, a combination of an antibody recognizing an epitope on hCGb with an antibody to hCGa or the intact heterodimer can also be used [15,16]. Fig. 1 shows the location of the major epitopes on hCGb and Fig. 2 the epitopes specific for intact hCG (Cepitopes), hCGb and hCGa on a 3-D model of intact hCG. The tips of loops 1 & 2 form a dominant antigenic region comprising five epitopes, b2 b6. Of these, b2 and b4 are specific for hCG, hCGb and hCGbcf while antibodies to b3 and b5 also recognize LH. Epitope b6 is specific for free hCGb. The cysteine knot on hCGb is another major antigenic region comprising epitopes of different specificities. Epitope b1 is a “pan hCG” antibody recognizing all the major forms of hCG. Epitope b7, which also is located around the cysteine knot, is specific for hCGb [15]. This epitope may actually be an antigenic region with more than one epitope as some b7 antibodies can form a sandwich with each other (Stenman, unpublished). Two well-defined epitopes on the C-terminal peptide (CTP), b8 and b9 are completely specific for hCG as the CTP is lacking on LH. CTP antibodies are therefore used in many commercial assays [15]. An antibody recognizing the core-2 glycan on Ser132 and surrounding peptides structures (epitope b15) recognizes hCG with especially large carbohydrate chains called hyperglycosylated hCG (hCG-h) [6]. Antibodies recognizing heterodimeric hCG recognize epitopes C1– C4 located around the cysteine knot. Two of these, C3 and C4 are sensitive to nicking of hCG and therefore underestimate hCGn, while C1 and C2 recognize both hCG and hCGn. Five epitopes have been defined on hCGa (a1–a5). Some of these are exposed only the free subunit while other ones are exposed both on the free subunit and the intact heterodimeric hormone. The latter can be used in combination with an hCGb antibody in assays for intact hCG [17]. The other antibodies are used in specific assays for free hCGa. Assay nomenclature Assays for hCG are often described as “beta-hCG assays” irrespective of whether they measure mainly hCG or hCG together with hCGb. This stems from the fact that the first hCG assays also measured LH and the first assays specific for hCG were produced by immunizing a rabbit with hCGb [18]. This antiserum (SB6) was widely distributed and used in radioimmunoassays, which detected hCG and hCGb together. Assays should be described according to what they measure, i.e., only intact hCG, hCG þ hCGb, or these together with or without hCGbcf [1]. The latter two assay types are often ambiguously described as “total” hCG assays. However, different assays defined as “total hCG” assay recognize different forms of hCG to variable degrees [11,12]. Very few assays recognize hCGbcf, and

786

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

Fig. 1. Epitope map showing the major epitopes discerned by monoclonal antibodies on hCGb. Modified from [23] by permission from Springer Verlag.

those that do underestimate this form [11]. This is important to know if urine samples are analyzed because hCGbcf is the major immunoreactive form of hCG in urine both when hCG or hCGb are the major forms in plasma [19]. Assay design and clinical application History of hCG immunoassays The first assay for hCG described in 1927 was a bioassay like all the first hCG assays. These were based on the ability of hCG to activate receptors on gonads of various test animals like mice, toads and rabbits. The history of the development of hCG assays up to 1987 has been comprehensively described by Hussa [20]. Wide and Gemzell developed the first immunoassay for hCG in 1960. It was an agglutination inhibition assay based on an antiserum to LH, which also recognized hCG. The detection limit of the assay was 500 IU/L and it took 2 h to perform [21]. Lower detection limits were obtained by the use of radioimmunoassay, but crossreaction with LH was a problem. Judith Vaitukaitis produced a specific antiserum by immunization with hCGb. This antiserum, SB6, was widely distributed and facilitated establishment of sensitive assays that recognized both hCG and hCGb. The assay had a sensitivity of 5–10 IU/L and proved to be quite useful for diagnosis of pregnancy, pregnancy-related disorders and various types of cancer [18]. An in-house radioimmunoassay of this type is still used in the Gestational Trophoblastic Disease Centre in the UK for monitoring of patients treated for trophoblastic diseases. This type of assay has the advantage of recognizing the major forms of hCG both in serum (i.e., hCG and hCGb) and in urine (hCG, hCGb and hCGbcf) in a fairly equimolar fashion [12].

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

787

Fig. 2. Epitope map showing the major epitopes discerned by monoclonal antibodies on hCGa, hCGb, and intact hCG on a 3-D model of intact hCG. Modified from [23] by permission from Springer Verlag.

However, commercial radioimmunoassays for hCG are not presently available from any major in vitro diagnostics company. Presently, all widely used hCG assays are based on the sandwich principle, i.e., one antibody bound to a solid phase captures hCG and some of its variants and one or several labeled antibodies recognizes another epitope(s) on the captured antigen. Most assays use monoclonal antibodies and with this assay design it is possible to accurately tune assay specificity. The first sandwich assay for hCG used enzymes as labels. An immunoconcentration assay (ICON) for hCG was one of the first rapid and sensitive (25 IU/ L) pregnancy tests [22]. The epitopes of hCG have been extensively characterized in two workshops and. recommendations made regarding epitope specificity of antibodies used for various types of assay [15,23]. Design of hCG assay for different clinical applications The epitopes recognized by the antibodies used in an hCG assay determines assays specificity. Knowledge of epitope specificity of antibodies facilitates design of assays with desired specificity, i.e., recognizing only some or all of the hCG variants described in Table 1. The specificity and detection limit of hCG assays are dependent on the intended application, detection of pregnancy, prenatal screening, or diagnosis and monitoring of cancer patients. Pregnancy diagnosis is mainly performed with dedicated semiquantitative pregnancy tests, which presently nearly exclusively are based on the lateral flow principle. A typical test contains a strip of

788

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

porous membrane containing a zone with a monoclonal antibody coupled to colored latex beads. The sample, which can be urine, serum, plasma or even blood, is applied to one end of the strip, from where it flows through the membrane. When the sample encounters the zone with the antibody beads, hCG in the sample binds to the beads, which flow with the sample to a zone with another antibody bound to the membrane. This antibody catches beads containing bound hCG forming a colored line. The assay typically takes about 5 min and the detection limit is in the range 10–50 IU/L. Most pregnancy tests perform as expected, but some methods are sensitive to excess antigen that may cause false negative results with urine samples [4]. These mostly contain a 2- to 20-fold excess of hCGbcf that may saturate the binding capacity of either tracer or capture antibody. This situation is most likely to occur at 7–12 weeks of pregnancy, when the hCG concentrations peak [1]. This is a problem for some, but not all assays [4]. Most pregnancy tests recognize only hCG but some also measure hCGb but not hCGbcf. The specificity of pregnancy tests is seldom clearly described in kit inserts. Quantitative assays for measurement of hCG and its variants in serum Sandwich assays using 2, or occasionally 3, monoclonal antibodies are presently used to measure hCG in automatic analyzers. This facilitates design of assays with well-defined specificity. The characteristics of assays used to determine various form of hCG depends on the clinical use. Diagnosis and monitoring of pregnancy Assays recognizing either hCG alone or hCG together with hCGb can be used for monitoring of pregnancy. In early pregnancy, the concentration of hCGb can be up to 5% of that of hCG and its proportion decreases gradually to about 1% in the second trimester [1]. Thus both types of assay give comparable results. Some assays for intact hCG utilize C-type antibodies, some of which fail to recognize nicked hCG. However, this is not a problem for serum or plasma samples during pregnancy because in these hCGn and hCGbn do not occur at concentrations noticeably affecting the results [24]. However, hCG preparations purified from pregnancy urine contain variable proportions of hCGn. Use of these for calibration of hCG assays may cause differences in the calibration [25]. In early pregnancy, most of hCG is hyperglycosylated, i.e., hCG-h [26]. In some studies, hCG-h was found not to be recognized by some hCG assays [27], but this was apparently explained by the fact that the original preparation of hCG-h used was 100% nicked [5]. It is possible, that hCG-h is not detected equally by all hCG assays, but all of 16 pregnancy tests that we have studied recognize hCG and hCG-h equally. After assisted reproduction treatment (in vitro fertilization or intracytoplasmic sperm injection), serum hCG is often measured about 12 days after embryo transfer. At this time, a median hCG concentration has been found to be 126 IU/L in viable pregnancies, while a level below 76 IU/L predicts early pregnancy loss. Although there is wide variation in outcome at various hCG levels, this approach facilitates early recognition of pregnancy loss facilitating rapid renewal of the treatment [28]. A low concentration of hCG-h has been reported to predict early pregnancy loss more accurately than hCG [29]. These results need to be confirmed. Reference values for hCG and hCGb in serum of non-pregnant women and men hCG and hCGb occur at low concentrations in serum from nonpregnant women and men. In menstruating women, the upper reference limit is 3 IU/L, but it increase up to 6 IU/L during the menopause [3]. However, due to difference in calibration and sensitivity of various assays, values up to 16 IU/L have been observed with some assays [30]. The upper reference limit for men under 50 years of age is 1 IU/L while it is 2 IU/L in older men. The hCG in nonpregnant subjects is produced by the pituitary and the concentrations of hCG (and LH) increase after administration of GnRH analogs and are suppressed by estrogens [3]. In men with hypogonadism, the hCG concentration increases [31]. Gonadal dysfunction induced by chemotherapy may also lead to an increase in hCG concentration, which is reversible It is important that this condition is not interpreted as a sign of tumor relapse. The condition can be identified on the basis of very high LH and FSH concentrations [1].

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

789

Screening for Down’s syndrome Women carrying a fetus with Down’s syndrome have elevated concentrations of hCG and hCGb in serum. Determination of hCGb is most useful for first trimester screening while either hCG (or hCGb) is used for second trimester screening. For first trimester screening, hCGb is combined with determination of PAPP-A and measurement of nuchal translucency of the fetus by ultrasound. For second trimester screening, hCG is combined with alphafetoprotein (AFP) and unconjugated estriol. The results for these markers are combined with maternal age for calculation of the probability of Down’s syndrome [32]. During first trimester, the concentrations of hCG are much higher than those of hCGb and it is therefore important that hCG does not interfere in the hCGb assay, e.g., by binding to either capture or tracer antibody. This is possible by using two antibodies that are specific for free hCGb, e.g., two b7 antibodies. Also, due to the large difference in concentrations of hCG and hCGb it is essential to minimize the possibility of dissociation of hCG to subunits by e.g., elevated storage temperatures. Assays used for Down screening are tuned to facilitate determination of the high hCGb concentrations occurring in pregnancy without sample dilution. The assays are therefore fairly insensitive and not useful for monitoring of cancer patients. The serum concentrations of hCG become detectable about 7 days after embryo transfer in IVF. During the first 7–9 weeks of pregnancy, the concentrations increase exponentially, doubling on average every 1.5 days. Maximal levels are reached at 8–10 weeks, after which they decrease, reaching a plateau at 15–18 weeks. The concentrations of hCG and hCGb decrease sharply during the end of the first trimester [1,33]. For screening purposes, the hCGb concentrations are therefore expressed as multiples of the median (MoM) concentrations for each pregnancy week determined in a reference population. Because of differences in calibration of different assays, the MoM values are assay-specific. Screening for preeclampsia Preeclampsia occurs in 2–5% of pregnant mothers and it is worldwide the most common cause of maternal and fetal morbidity and mortality. Treatment with acetylsalicylic acid (ASA) starting in second trimester has been shown to reduce the risk of preeclampsia [34]. Disturbances in the balance between various angiogenic and antiangiogenic factors, have been observed especially in patients with established disease. However, the role of these factors in the development of preeclampsia is not clear [35]. Several biomarkers and clinical characteristics have been evaluated as predictors of preeclampsia. Recently the concentrations of hCG-h in first trimester serum were found to be lower in mothers who later develop preeclampsia than in controls. The combination of hCG-h with pregnancy-associated plasma protein A (PAPP-A), maternal mean arterial pressure, and parity were shown to predict early onset preeclampsia with an AUC value exceeding 0.85 [36]. Assay of hCG-h is dependent on the use of MAb B152, an IgG2a antibody that is sensitive to interference by complement, which causes false low results unless eliminated. This can be achieved by using calcium chelators, e.g. citrate EDTA or in the buffer or citrated plasma as the sample. We have developed and in-house assay based on B152 for hCG-h using hCG from JEG-3 culture fluid as calibrator [37]. A commercial assay has been marketed by Nichols Laboratories but is no more available. A new assay for hCG-h has recently been described [38]. Various forms of hCG as tumor markers hCG is an extremely sensitive marker for trophoblastic tumors, a tumor containing 10,000 cells can produce enough hCG to cause elevated serum concentrations [2,39]. Many testicular germ cell tumors also produce hCG or hCGb and these are very useful markers for monitoring of the disease [40]. In addition, 20–70% of nontrophoblastic tumors produce hCGb at low concentrations [2]. Diagnosis of trophoblastic disease Virtually all trophoblastic tumors produce hCG [41]. They also produce hCGb but the concentrations are usually lower than those of hCG. A high proportion of hCGb is associated with aggressive tumors [39,42]. In our experience, a proportion of hCGb exceeding 5%, based on molar concentrations, has always

790

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

been associated with malignant trophoblastic disease while the proportion is lower in benign molar disease and in pregnancy. In some cases, development of treatment resistance of a trophoblastic tumor has been accompanied by an increasing proportion of hCGb [43] (Fig. 1). We therefore determine hCG and hCGb separately with specific assays. However, because sensitive assays for hCGb are not widely available, it is generally recommended that assays to be used for diagnosis of cancer recognize hCG and hCGb together [44]. It should be noted, that commercial assays of this type recognize hCGb to rather variable degrees, some overestimating while other underestimate it [11,12]. In the UK, the Gestational Trophoblastic Disease Centres determine the various forms of hCG together by RIA. These methods recognize the major forms, i.e., hCG, hCGb and hCGbcf in a fairly equimolar fashion [11]. The ability of an assay to recognize hCGbcf is relevant if urine samples are analyzed, which is used for monitoring of patients after treatment of trophoblastic disease in the UK. Advantages of sandwich assays over RIA are more rapid turn around time and lower detection limits, in some assays below 0.5 IU/L. When used for urine, the upper reference limit of the RIA is 25 IU/L reflecting the rather high detection limit of this assay [12]. Monitoring of treatment for trophoblastic cancer relies heavily on the use of hCG determinations. There is therefore a need to improve standardization of hCG tests in order to facilitate use of common treatment protocols. Germ cell tumors of the testis The main types of testicular germ cell tumors are seminomas and non-seminomatous germ cell tumors of the testis (NSGCT). NSGCTs are classified as embryonal carcinomas, choriocarcinomas, yolk sac tumors and mature or immature teratomas. Most NSGCTs are mixed, i.e., they contain several histological components. About 50–70% of NSGTCs produce hCG, alpha fetoprotein (AFP) or both, whereas seminomas have been reported to produce hCG in 15–30% of the cases [40]. Recent results show that increased serum concentrations of hCGb are seen in 50% of the seminoma patients while hCG is elevated in only 16%. In nearly half of the patients, the elevation of the hCGb concentration is fairly small, 2–15 pmol/L. A value of 15 pmol/l corresponds to 5 IU/L of hCG, which is a commonly used cut-off for “total hCG”. Thus nearly half of the hCGb positive cases would have been missed by an assay measuring hCG and hCGb together [45]. In NSGCTs the concentrations of hCGb and hCG correlate strongly, and both markers are elevated in 72% of the patients. In 8% of the cases, only hCGb was elevated and thus, either marker was elevated in 80%. The concentrations of hCG-h correlate strongly with those of hCG and the median proportion of hCG-h (%hCG-h) being 84% (95% central confidence interval 83–93%). Only a few seminoma patients had detectable concentrations of hCG-h in serum. Thus hCGb is a valuable marker for testicular cancer and especially for seminomas but hCG-h does not provide additional value. Nontrophoblastic cancers Early studies on the expression of hCG in cancer by RIA indicated that intact hCG was produced by several nontrophoblastic tumors [46]. However, later studies with specific assays indicate that in virtually all cases, hCGb is the only form that can be detected in serum of these patients. Expression of hCGb is mostly associated with aggressive disease, and elevated serum levels are quite common occurring in 72–86% of patients with pancreatic and biliary cancer, which are characterized by adverse prognosis [19,47]. Elevated serum concentrations of hCGb have been observed in 25–40% of patients with ovarian, renal, bladder, breast and head and neck cancer. In all of these, an elevated serum concentration is associated with short survival [2]. In ovarian cancer, this association is especially strong [48]. The concentrations of hCGb in patients with nontrophoblastic cancer are usually only moderately elevated and thus assays measuring hCG and hCGb together are of limited use. Assays specific for intact hCG are not useful for detection of nontrophoblastic cancers [2]. False positive results False positive hCG may have serious consequences because, if pregnancy can be ruled out, an elevated hCG concentration in serum is mostly caused by cancer. Falsely elevated results are caused by

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

791

heterophilic antibodies, which react with the immunoglobulins, mainly mouse monoclonal antibodies, used in the assay. Assay manufacturers include mouse IgG that blocks the interference by heterophilic antibodies, but their concentrations are occasionally so high that they cause false positive result. The heterophilic antibodies often cause falsely elevated results also in other immunoassays, but these are not easily detected. In the absence of other evidence of cancer, elevated results for hCG in serum or plasma should be confirmed by assay by repeat assay in the presence of a blocking antibody, assay by an alternative method, and assay of hCG in urine. Causes of interference in immunoassays have recently been comprehensively described [49]. Familial hCG Constantly elevated serum and urine concentrations have been observed in about 15 families. The concentrations are usually low but high enough (10–200 IU/L) to cause suspicion of pregnancy or cancer. The levels fluctuate but stay elevated for at least several years and probably for the life-time of the subject. This very rare condition occurs both in men and women. The prevalence has been estimated to be 1:60,000. It is important to recognize this condition, because it may lead to unnecessary treatments [50]. Doping control Athletes may use hCG for stimulation of testicular function after use of anabolic steroids and testosterone. Determination of hCG in urine is used to detect illicit use of hCG. Men with familial hCG have been suspended from competition because positive doping tests with assays measuring hCG and hCGb together (Stenman et al. unpublished findings). However, these men have had elevated concentrations of hCGb, but not of hCG. hCGb lacks bioactivity and it is not a major breakdown product of hCG urine. Thus WADA has decided that a positive doping result for hCG has to be confirmed with an assay for intact hCG. Summary Determination of hCG is demanding because of the occurrence of different molecular forms in circulation and in urine. Most presently available assays detect hCG and hCGb together while other ones are specific for intact hCG. Assay of hCGb is of potential utility as a tumor marker, but if a specific assay is not available, one detecting hCG and hCGb together should be used. The core fragment of hCGb is the major form of hCG immunoreactivity in urine, where its concentrations may be 10-fold that of hCG. It can therefore disturb assay of hCG and cause false negative results with some pregnancy tests. There is a need to improve standardization of hCG tests in order to facilitate use of common protocols for treatment of trophoblastic disease. Practice points  Different assays recognize different forms of hCG to various degrees. Because of this, change of assay during follow-up of cancer patients may cause misleading results.  False positive hCG results are rare but may lead to suspicion of cancer. This has led to unnecessary treatment and serious complications. In the absence of other evidence of cancer, elevated results for hCG in serum or plasma should be confirmed by assay of hCG with another method, by repeating the assay in the presence of a blocking antibody, and assay of hCG in urine.  False negative hCG results may occur when hCG is assayed in urine due to a hook effect when the concentrations of hCGbcf are highest at 7–11 weeks of pregnancy. This problem concerns some pregnancy tests.

792

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

Research agenda  Establishment of reference methods for the various forms of hCG.  Improved standardization of hCG assays in order to facilitate collaboration between centers treating patients with trophoblastic cancers.  The utility of hCGb as a prognostic marker in general and especially for nontrophoblastic cancer needs to be further evaluated.  Methods used in doping control for detection of self-administration of hCG needs to be standardized and reference values for hCG in urine of men established.  The prevalence and characteristics of familial hCG needs to be studied.

References *[1] Stenman UH, Tiitinen A, Alfthan H, et al. The classification, functions and clinical use of different isoforms of hCG. Hum Reprod Update 2006;12:769–84. [2] Stenman UH, Alfthan H, Hotakainen K. Human chorionic gonadotropin in cancer. Clin Biochem 2004;37:549–61. [3] Stenman UH, Alfthan H, Ranta T, et al. Serum levels of human chorionic gonadotropin in nonpregnant women and men are modulated by gonadotropin-releasing hormone and sex steroids. J Clin Endocrinol Metab 1987;64:730–6. [4] Gronowski AM, Cervinski M, Stenman UH, et al. False-negative results in point-of-care qualitative human chorionic gonadotropin (hCG) devices due to excess hcgbeta core fragment. Clin Chem 2009;55:1389–94. *[5] Elliott MM, Kardana A, Lustbader JW, et al. Carbohydrate and peptide structure of the alpha- and beta-subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32. [6] Birken S, Yershova O, Myers RV, et al. Analysis of human choriogonadotropin core 2 o-glycan isoforms. Mol Cell Endocrinol 2003;204:21–30. [7] Valmu L, Alfthan H, Hotakainen K, et al. Site-specific glycan analysis of human chorionic gonadotropin beta-subunit from malignancies and pregnancy by liquid chromatography–electrospray mass spectrometry. Glycobiology 2006;16: 1207–18. *[8] Birken S, Berger P, Bidart JM, et al. Preparation and characterization of new who reference reagents for human chorionic gonadotropin and metabolites. Clin Chem 2003;49:144–54. [9] Bristow A, Berger P, Bidart JM, et al. Establishment, value assignment, and characterization of new who reference reagents for six molecular forms of human chorionic gonadotropin. Clin Chem 2005;51:177–82. *[10] Stenman UH, Bidart JM, Birken S, et al. Standardization of protein immunoprocedures. Choriogonadotropin (CG). Scand J Clin Lab Invest Suppl 1993;216:42–78. [11] Sturgeon CM, Berger P, Bidart JM, et al. Differences in recognition of the 1st who international reference reagents for hCG-related isoforms by diagnostic immunoassays for human chorionic gonadotropin. Clin Chem 2009;55:1484–91. [12] Harvey RA, Mitchell HD, Stenman UH, et al. Differences in total human chorionic gonadotropin immunoassay analytical specificity and ability to measure human chorionic gonadotropin in gestational trophoblastic disease and germ cell tumors. J Reprod Medicine 2010;55:285–95. [13] Birken S, Armstrong EG, Kolks MA, et al. Structure of the human chorionic gonadotropin beta-subunit fragment from pregnancy urine. Endocrinology 1988;123:572–83. *[14] Lapthorn AJ, Harris DC, Littlejohn A, et al. Crystal structure of human chorionic gonadotropin. Nature 1994;369:455–61. [15] Berger P, Sturgeon C, Bidart JM, et al. The ISOBM TD-7 workshop on hCG and related molecules. Towards user-oriented standardization of pregnancy and tumor diagnosis: assignment of epitopes to the three-dimensional structure of diagnostically and commercially relevant monoclonal antibodies directed against human chorionic gonadotropin and derivatives. Tumour Biol 2002;23:1–38. [16] Stenman UH, Alfthan H, Myllynen L, et al. Ultrarapid and highly sensitive time-resolved fluoroimmunometric assay for chorionic gonadotropin. Lancet 1983;2:647–9. [17] Pettersson K, Siitari H, Hemmila I, et al. Time-resolved fluoroimmunoassay of human choriogonadotropin. Clin Chem 1983;29:60–4. *[18] Vaitukaitis JL, Braunstein GD, Ross GT. A radioimmunoassay which specifically measures human chorionic gonadotropin in the presence of human luteinizing hormone. Am J Obstet Gynecol 1972;113:751–8. [19] Alfthan H, Haglund C, Roberts P, et al. Elevation of free beta subunit of human choriogonadotropin and core beta fragment of human choriogonadotropin in the serum and urine of patients with malignant pancreatic and biliary disease. Cancer Res 1992;52:4628–33. [20] Hussa RO. The clinical marker hCG. New York: Praeger; 1987. [21] Wide L, Gemzell CA. An immunological pregnancy test. Acta Endocrinologica 1960;35:261–7. [22] Valkirs GE, Barton R. Immunoconcentration–a new format for solid-phase immunoassays. Clin Chem 1985;31:1427–31. [23] Berger P, Paus E, Sturgeon C, et al. Candidate epitopes for measurement of hcg and related molecules: the 2nd ISOBM TD-7 workshop. Tumour Biol 2013 [in press]. [24] Hoermann R, Berger P, Spoettl G, et al. Immunological recognition and clinical significance of nicked human chorionic gonadotropin in testicular cancer. Clin Chem 1994;40:2306–12. [25] Lund H, Paus E, Berger P, et al. Epitope analysis and detection of human chorionic gonadotropin (hCG) variants by monoclonal antibodies and mass spectrometry. Tumour Biol 2013 Sep 7 [Epub ahead of print].

U.-H. Stenman, H. Alfthan / Best Practice & Research Clinical Endocrinology & Metabolism 27 (2013) 783–793

793

[26] Kovalevskaya G, Birken S, Kakuma T, et al. Differential expression of human chorionic gonadotropin (hCG) glycosylation isoforms in failing and continuing pregnancies: preliminary characterization of the hyperglycosylated hCG epitope. J Endocrinol 2002;172:497–506. [27] Cole LA, Shahabi S, Butler SA, et al. Utility of commonly used commercial human chorionic gonadotropin immunoassays in the diagnosis and management of trophoblastic diseases. Clin Chem 2001;47:308–15. [28] Poikkeus P, Hiilesmaa V, Tiitinen A. Serum hCG 12 days after embryo transfer in predicting pregnancy outcome. Hum Reprod 2002;17:1901–5. [29] Cole LA. Hyperglycosylated hcg and pregnancy failures. J Reprod Immunol 2012;93:119–22. *[30] Cole LA, Sasaki Y, Muller CY. Normal production of human chorionic gonadotropin in menopause. New Engl J Med 2007; 356:1184–6. [31] Lempiainen A, Hotakainen K, Blomqvist C, et al. Increased human chorionic gonadotropin due to hypogonadism after treatment of a testicular seminoma. Clin Chem 2007;53:1560–1. [32] Wald NJ, Hackshaw AK. Advances in antenatal screening for down syndrome. Bailliere’s Best Pract Res Clin Obstet Gynaecol 2000;14:563–80. *[33] Braunstein GD, Rasor J, Danzer H, et al. Serum human chorionic gonadotropin levels throughout normal pregnancy. Am J Obstet Gynecol 1976;126:678–81. [34] Roberge S, Giguere Y, Villa P, et al. Early administration of low-dose aspirin for the prevention of severe and mild preeclampsia: a systematic review and meta-analysis. Am J Perinatol 2012;29:551–6. [35] Vuorela P, Halmesmaki E. Vascular endothelial growth factor, its receptors, and the tie receptors in the placental bed of women with preeclampsia, diabetes, and intrauterine growth retardation. Am J Perinatol 2006;23:255–63. [36] Keikkala E, Vuorela P, Laivuori H, et al. First trimester hyperglycosylated human chorionic gonadotrophin in serum - a marker of early-onset preeclampsia. Placenta 2013;34:1059–65. [37] Stenman UH, Birken S, Lempiainen A, et al. Elimination of complement interference in immunoassay of hyperglycosylated human chorionic gonadotropin. Clin Chem 2011;57:1075–7. [38] Strom CM, Bonilla-Guererro R, Zhang K, et al. The sensitivity and specificity of hyperglycosylated hcg (hhCG) levels to reliably diagnose clinical IVF pregnancies at 6 days following embryo transfer. J Assisted Reprod Genet 2012;29:609–14. [39] Khazaeli MB, Hedayat MM, Hatch KD, et al. Radioimmunoassay of free beta-subunit of human chorionic gonadotropin as a prognostic test for persistent trophoblastic disease in molar pregnancy. Am J Obstet Gynecol 1986;155:320–4. [40] Stenman UH. Testicular cancer: the perfect paradigm for marker combinations. Scand J Clin Lab Invest 2005;65:181–8. [41] Rushworth AG, Orr AH, Bagshawe KD. The concentration of hCG in the plasma and spinal fluid of patients with trophoblastic tumours in the central nervous system. Br J Cancer 1968;22:253–7. [42] Stenman UH, Alfthan H, Halila H. Determination of chorionic gonadotropin in serum of nonpregnant subjects and patients with trophoblastic cancer by time-resolved immunofluorometric assay. Tumour Biol 1985;5:97. *[43] Vartiainen J, Alfthan H, Lehtovirta P, et al. Elevated hCG and a high proportion of hCG beta in serum preceding the diagnosis of trophoblastic disease by seven months. BJOG: An Int J Obstet Gynaecol 2002;109:589–90. [44] Madersbacher S, Klieber R, Mann K, et al. Free alpha-subunit, free beta-subunit of human chorionic gonadotropin (hcg), and intact hcg in sera of healthy individuals and testicular cancer patients. Clin Chem 1992;38:370–6. [45] Lempiainen A, Stenman UH, Blomqvist C, et al. Free beta-subunit of human chorionic gonadotropin in serum is a diagnostically sensitive marker of seminomatous testicular cancer. Clin Chem 2008;54:1840–3. [46] Braunstein GD, Vaitukaitis JL, Carbone PP, et al. Ectopic production of human chorionic gonadotrophin by neoplasms. Ann Intern Med 1973;78:39–45. *[47] Marcillac I, Troalen F, Bidart JM, et al. Free human chorionic gonadotropin beta subunit in gonadal and nongonadal neoplasms. Cancer Res 1992;52:3901–7. [48] Vartiainen J, Lassus H, Lehtovirta P, et al. Combination of serum hCG beta and p53 tissue expression defines distinct subgroups of serous ovarian carcinoma. Int J Cancer 2008;122:2125–9. [49] Bolstad N, Warren DJ, Nustad K. Heterophilic antibody interference in immunometric assays. Best Pract Res Clin Endocrinol 2013;5:647–61. [50] Cole LA. Familial hCG syndrome. J Reprod Immunol 2012;93:52–7.

Determination of human chorionic gonadotropin.

Determination of human chorionic gonadotropin (hCG) is used for diagnosis and monitoring of pregnancy, pregnancy related disorders, for trophoblastic ...
818KB Sizes 0 Downloads 0 Views