Parathyroid hormone assays and their clinical use

INTRODUCTION — Parathyroid hormone (PTH) (1-84) is the biologically active hormone initially produced by the parathyroid glands as a 115-amino-acid polypeptide that undergoes two successive cleavages to yield the 84-amino-acid peptide, which is stored in the parathyroid cells. It is secreted in response to hypocalcemia and few other stimuli into the systemic circulation. It exerts its effects through the interaction of its first 34 amino acids with the type 1 PTH/PTH-related peptide (PTHrP) receptor (PTHR1). PTH fragments, containing carboxyl (C)- or amino (N)-terminal portions of the molecule that arise from either intraglandular or peripheral degradation of the hormone are also present in the circulation, but their biological significance remains to be defined.

PTH (1-84) has a plasma half-life of two to four minutes. In comparison, the C-terminal fragments, which are cleared principally by the kidney, have a half-life that is 5 to 10 times longer. As a result, circulating immunoreactive PTH in normocalcemic subjects comprises:

●PTH (1-84) – 5 to 30 percent

●C-terminal fragments – 70 to 95 percent

●N-terminal fragments – A small percentage

Evidence suggests that some of these fragments, particularly the N-terminally truncated fragment PTH (7-84) (also referred to as non-PTH [1-84]), interact with distinct receptors (C-PTH receptor [C-PTHR]). Non-PTH (1-84) may have roles in the regulation of bone resorption and serum calcium concentration [1], and it is therefore conceivable that quantification of different PTH fragments could add to the diagnostic importance of PTH measurements.

PTH assays and their clinical use will be reviewed here. PTH physiology, receptor binding, and actions are reviewed separately. (See "Parathyroid hormone secretion and action".)

PTH ASSAYS — Radioimmunoassays for the detection of parathyroid hormone (PTH) are referred to as "first" generation assays; consequently, immunometric assays (IMAs) are referred to as "second" and "third" generation assays. The traditional second-generation assays (known as intact PTH assays) measure PTH (1-84) and large carboxyl (C)-terminal PTH fragments, whereas the third-generation assays (bioactive or biointact PTH assays) detect only PTH (1-84) and not C-terminal fragments. Some authorities refer to the IMAs as first-generation IMA (intact PTH assays) and second-generation IMA (bioactive or biointact PTH assays). Subsequently, another assay was developed with which oxidized (biologically inactive) and non-oxidized PTH (biologically fully active) can be differentiated [2]. However, this later assay relies on the ex vivo prevention of oxidation, and a recent study has called its clinical utility into question [3].

Radioimmunoassays — PTH was first measured by radioimmunoassay using different polyclonal antibodies directed against epitopes that proved to be predominantly within the mid- or C-terminal portion of the PTH molecule (figure 1) [4,5]. Such first-generation PTH assays therefore detected predominantly PTH fragments that lack an intact amino (N)-terminus, do not activate the PTH/PTH-related peptide (PTHrP) receptor (PTHR1), and thus do not mediate the hormone's calcemic or phosphaturic actions. Because of the crossreactivity with hormonal fragments and insufficient sensitivity, such assays have now been largely replaced by two-site IMAs, also referred to as "sandwich" assays [6].

Immunometric assays for the detection of PTH — IMAs utilize two different antibodies that are directed against distinct epitopes within the PTH molecule [6-8]. One of these antibodies, directed against a region within the C-terminus, captures circulating PTH (1-84) as well as PTH fragments. The second antibody, directed against an N-terminal epitope, serves as the detection antibody, which can be radioiodinated, biotinylated for detection with enzyme-linked avidin, or tagged otherwise (figure 1). Radiolabeled detection antibodies have been replaced in most commercially available assays with a nonradioactively tagged antibody. The capture antibodies in some IMAs are directed against N-terminal epitopes, while the detection antibodies are directed against C-terminal PTH epitopes. IMAs are intrinsically more sensitive and specific than radioimmunoassays [9]. Thus, they provide excellent sensitivity and specificity even for the diagnosis of mild forms of primary hyperparathyroidism. Furthermore, in contrast to the findings with radioimmunoassays, concentrations of PTH measured by IMAs are usually below the normal range in most patients with humoral hypercalcemia of malignancy and with different forms of hypoparathyroidism [5,6], thus providing superior diagnostic utility.

Intact PTH assay — The first widely used IMA two-site sandwich assays, also known as "intact PTH" assays, use a capture antibody against the C-terminal part of the PTH molecule (epitopes 39-84) and a radioiodinated detection antibody directed towards the N-terminal portion of PTH (epitopes 13-34) and, therefore, detect the intact as well as large C-terminal fragments that lack portions of the N-terminus (ie, the first 12 amino acids) and are termed non-PTH (1-84) [10-15]. (See 'Non-parathyroid hormone (1-84)' below.)

The detailed characterization of this intact PTH assay showed that the detection antibody interacts not only with human PTH (1-34) and PTH (2-34), but also with PTH fragments that are truncated further at the N-terminus, such as PTH (7-84) (figure 1) [15]. The latter fragments (or variants thereof) are biologically inactive with regards to calcium- and phosphate-regulating activity, but some appear to inhibit osteoclastic bone resorption and the formation of mature osteoclasts and, thus, may have indirect hypocalcemic properties [16].

Bioactive PTH (1-84) assay — Subsequent two-site IMAs use a similar capture antibody as the intact PTH assays but detection antibodies directed against epitopes at the extreme N-terminal end (epitopes 1-4) of the molecule, and therefore, they were believed to detect exclusively the biologically active PTH (1-84) (whole PTH or bioactive PTH), but not non-PTH (1-84) fragments (figure 1) [15,17-20].

However, in a study comparing a radioimmunoassay with intact and bioactive PTH immunoassays in healthy individuals and those with hyperparathyroidism, the bioactive PTH assay reacted with PTH (1-84) and with another N-terminal PTH fragment, which is not recognized by the intact PTH assays [13]. The biological significance of this fragment, called N-terminal PTH (N-PTH), is uncertain. (See 'N-terminal PTH' below.)

Oxidized PTH — Patients with advanced chronic kidney disease (CKD) are subject to oxidative stress, and plasma proteins (including PTH) are targets for oxidants [21]. In patients with CKD, a considerable but variable fraction (70 to 90 percent) of measured PTH appears to be oxidized [2,22]. Combined oxidation on methionine 8 and 18 impairs interaction with the PTH receptor, while oxidation on methionine 18 alone has less impact on biological activity [23-25].

Intact and bioactive PTH assays detect both oxidized and non-oxidized PTH, resulting in higher PTH levels than when measured with a non-oxidized PTH assay [2]. The clinical utility of measuring non-oxidized PTH remains controversial, particularly because of concerns regarding ex vivo oxidation [3]. (See 'Secondary hyperparathyroidism' below.)

Non-oxidized PTH assay — A variant of the intact PTH assay has been developed to address oxidation of the methionine residues at positions 8 and 18 [2]. This assay uses an immobilized monoclonal antibody raised against the oxidized peptide to extract this PTH variant from plasma. The remaining PTH is then measured by a two-site assay that uses antibodies directed against PTH (26-32) and PTH (55-64) for capture and detection. Ex vivo oxidation of PTH may be difficult to prevent reliably, and one study suggested that the measurement of non-oxidized provides no diagnostic advantage [3]. (See 'Oxidized PTH' above.)

Assay variability — Although second- and third-generation assays provide reliable measurements, thus allowing sound clinical interpretation, several variables have been proposed to affect PTH values [26]:

●Preanalytical factors include the inherent biologic variability in PTH levels due to hormone pulsatility and diurnal variation, choice of tube for sample collection, as well as sample stability at storage temperature [26-29].

The stability of PTH when stored as serum or plasma EDTA samples at -80°C is not indefinite. Depending upon the assay, the maximum storage times with either dry or EDTA plasma ranges from two months to two years [30]. The purported greater stability of PTH when measured with a third-generation assay may be an added advantage, which has prompted some clinical chemistry laboratories to use third-generation assays instead of "intact" assays [29,31]. Although the International Federation of Clinical Chemists (IFCC) PTH working group prefers plasma to serum [32], serum has the added advantage of allowing simultaneous measurements of calcium and PTH levels, an important consideration considering the substantial impact of calcium on PTH levels. PTH stability depends on the type of tube used. A recent study tested PTH stability in five types of collection tubes in 30 participants (10 healthy persons, 10 patients on hemodialysis, and 10 patients with hyperparathyroidism) [33]. PTH appeared most stable in the EDTA-K3 tube, followed by the heparin anticoagulant tube.

●Analytical factors include the presence of PTH fragments that can crossreact with the capture antibodies. Although capture antibodies in second-generation assays are usually in vast excess, accumulation of large PTH C-terminal fragments in patients with CKD can theoretically affect measurement. Furthermore, N-terminal PTH fragments may interfere with intact PTH measurement in third-generation assays and the proportion of oxidized versus non-oxidized fragments can vary. Using a biotin neutralization assay, biotin was recently shown to significantly decrease PTH levels using the Roche Cobas assay but not when using the Beckman Coulter Access2 analyzer [34].

●Postanalytical factors include the lack of a true normative range in vitamin D-replete individuals with normal kidney function, and one that is tailored to age, race, and possibly sex and menopausal status [35-37], which may hamper interpretation and management of hyperparathyroidism [29,35-38].

In addition, there is currently no reference method for PTH measurement and no shared recombinant or synthetic PTH standard for the different PTH assays [32]. This results in large interassay variability in results obtained on the same samples when using PTH assays provided by different manufacturers and even when using same generation PTH assays that are produced by different manufacturers [39,40]. A systematic comparison of second- and third-generation assays revealed that PTH reference intervals are not interchangeable between assays. The normative range was 1.8 to 8.5 pmol/L for the Atellica second-generation assay, 2.4 to 10.9 pmol/L for the Alinity second-generation assay, and lowest at 1.8 to 7.0 pmol/L for the Cobas 8000 third-generation assay [41]. These observations underscore the importance of defining assay-specific normative ranges, particularly in certain populations (eg, CKD), monitoring patients with assays using the same laboratory/manufacturer, and basing decisions regarding treatment upon trends rather than single laboratory values [38,42-44].

IMAs, including the different PTH assays, are vulnerable to interference with heterophilic antibodies, which can be present in up to 11 percent of the population, that can bridge the capture and signal antibodies and cause a false-positive result. Such antibodies are increasingly prevalent because of the rise in the use of monoclonal antibodies in the treatment of inflammatory disorders, cancer, and transplantation. Such antibodies were shown to cause substantial false elevation in PTH levels in two women. The first patient had successfully undergone kidney transplantation, yet her PTH levels remained extremely elevated (3374 pg/mL) despite medical and surgical therapy [45]. The second patient, who was treated for obesity, was shown to have human anti-mouse immunoglobulin G (IgG) antibodies that resulted in the measurement of PTH levels >5000 pg/mL, but only when using assays that employ mouse monoclonal antibodies for detection [46].

PTH assay by LC-MS/MS — More recently, a liquid chromatography-tandem mass spectrometry (LC-MS/MS) assay was developed and validated for PTH (1-84) levels ranging from 5.7 to 872.6 pg/mL [47]. The reported interassay imprecision was between 1.2 and 3.9 percent, and the accuracy ranged from 96.2– to 103.2 percent. This assay was expected to be unaffected by the presence of heterophilic antibodies. Such findings should be validated in independent studies.

PTH fragments — A number of circulating molecular forms of PTH have been identified by radioimmunoassays and IMAs (figure 1). C-terminal fragments (lacking small or large portions of the N-terminus) are most abundant, representing approximately 80 percent of circulating PTH in healthy individuals and approximately 95 percent in patients with CKD [48]. Large C-terminal fragments with a partially preserved N-terminus (referred to as non-PTH [1-84]) are secreted in vitro by parathyroid tissue from patients with primary and secondary hyperparathyroidism [14,49,50].

An N-terminal form of PTH that is distinct from PTH (1-84) but detectable by bioactive PTH assays has also been identified [13]. This form is secreted in vitro by parathyroid tissue from patients with primary and secondary hyperparathyroidism [13,51]. In one report, a biologically active N-terminal PTH fragment (PTH [1-52]) was identified in the parathyroid adenoma of a patient with primary hyperparathyroidism, who had undetectable serum intact PTH [52].

Non-parathyroid hormone (1-84) — Non-PTH (1-84) fragments are a subset of large C-terminal fragments that lack only a small portion of the N-terminus and are therefore measured by the intact PTH assay. They differ from other C-terminal fragments that lack a large portion of the N-terminus and are not detected by the intact PTH assay. In individuals with normal kidney function, non-PTH (1-84) fragments account for 10 percent of all C-terminal PTH fragments and for 20 percent of the PTH detected by the intact PTH assay [11,12]. In patients with CKD, such as in patients on hemodialysis, non-PTH (1-84) may account for as much as 45 percent of the immunoreactivity measured by the intact PTH assay [53].

The predominant non-PTH (1-84) isoform is PTH (7-84). PTH (7-84) does not bind efficiently to the PTHR1. Instead, it presumably exerts some of its biological effects through an as-yet undefined receptor/binding protein predicted to be specific for the C-terminal portion of PTH (C-PTH receptor) [16,54]. Such a receptor/binding protein has been detected in several different tissues and in clonal cell lines, including ROS 17/2.8 cells [55-59]; the most abundant concentrations of this receptor were detected on osteocytic cells [60]. PTH (7-84), but not PTH (3-34) or PTHrP (7-36), inhibits basal 45Ca-release from pre-labeled calvarial bones. In addition, PTH (7-84) inhibits bone resorption induced by a variety of osteoclast-activating agents such as 1,25-dihydroxyvitamin D, prostaglandin E2, and interleukin (IL)-11 [16].

Synthetic PTH (7-84) has been shown to have hypocalcemic properties when tested in vivo, as evidenced by its ability to block the calcemic actions of PTH (1-34) and PTH (1-84) [18,54]. In addition, one study showed that PTH (7-84) decreased PTH-stimulated bone turnover in rats with moderate kidney failure [61]. Thus, PTH (7-84) seems to exert effects that are the opposite of those of PTH (1-84). This may contribute to the skeletal resistance to PTH that is observed in patients with kidney failure [62].

N-terminal PTH — The existence of a modified form of PTH (1-84), referred to as N-terminal PTH (N-PTH), has been reported [13]. It differs from PTH (1-84) in that it is poorly reactive with antibodies directed against epitopes 12-18 used in most intact PTH assays. In contrast, N-PTH is detected with third-generation bioactive PTH assays (detection antibody requires amino acid residues 1 and 2) (figure 1). In individuals with normal kidney function, N-PTH accounts for 4 to 8 percent of PTH detected with a bioactive PTH assay, whereas it accounts for up to 15 percent in patients with kidney failure [13,14]. N-PTH appears to be overproduced in some forms of primary and secondary hyperparathyroidism and in parathyroid cancer [13,51,63]. At the present time, its biological activity is unknown.

CLINICAL USE OF PTH ASSAYS — The realization that different circulating forms of parathyroid hormone (PTH) are detected by the various PTH assays has led to a reassessment of these assays in conditions of increased and decreased PTH secretion. Available evidence suggests that bioactive or whole PTH assays provide little additional information from that provided by intact PTH assays for the diagnosis of primary hyperparathyroidism. However, some data suggest that the newer assays may be superior in patients with kidney failure, for intraoperative PTH monitoring, and for making the initial diagnosis in patients with primary hyperparathyroidism and inappropriately "normal" serum PTH concentrations measured with intact PTH assays [15,17,64]. Furthermore, bioactive PTH assays have been shown to improve diagnostic accuracy in patients with PTH carcinoma [65].

Primary hyperparathyroidism — Intact PTH assays have been predominantly used for establishing the diagnosis of primary hyperparathyroidism. The results obtained with intact and bioactive PTH assays are highly correlated, and therefore, either intact or bioactive PTH assays can be used for diagnosis of hyperparathyroidism. However, the clinician must be aware of the particular assay and normal values associated with its use. (See "Primary hyperparathyroidism: Diagnosis, differential diagnosis, and evaluation", section on 'Serum PTH'.)

Direct comparison of the two immunometric PTH assays, the intact PTH assay for combined measurement of PTH (1-84) and non-PTH (1-84) and the bioactive PTH assay for the measurement of PTH (1-84) and N-terminal PTH, revealed an excellent correlation (r = 0.922) in 165 patients with primary hyperparathyroidism [17]. Similar findings were reported in other studies [64,66-68]. However, when using bioactive assays, overlap exists in the upper end of the normal range between individuals considered to be healthy and patients with surgically proven hyperparathyroidism. Although limited data [64] suggest that PTH is increased in a higher proportion of patients with primary hyperparathyroidism when using the bioactive PTH assay, several other studies have found no improvement in diagnostic utility [67,68].

The performance of bioactive PTH assays in patients with normocalcemic hyperparathyroidism requires further elucidation [69].

Intraoperative PTH assessment — Rapid measurement of parathyroid hormone (PTH) levels is now frequently used intraoperatively to confirm that an adenoma has been successfully removed; this technique is used particularly during less invasive procedures involving limited neck exploration after preoperative localization of the putative adenoma.

Most available rapid intraoperative assays are intact PTH assays measuring both PTH (1-84) and PTH (7-84), but it is conceivable that a third-generation rapid assay measuring bioactive PTH may be more useful for monitoring during parathyroidectomy. (See "Parathyroid exploration for primary hyperparathyroidism", section on 'Focused parathyroidectomy' and "Intraoperative parathyroid hormone assays".)

Measurement of PTH levels in 74 patients with primary hyperparathyroidism, which were performed in samples obtained 5, 10, and 15 minutes after excision of the adenoma, revealed that PTH, as measured with a bioactive assay, decreased faster after excision than when measured with the intact assay [70,71]. At five minutes, PTH levels had dropped to less than 50 percent of the level before parathyroidectomy in all patients using the bioactive assay but remained above that cut-off in six patients (8 percent) using the intact assay [70].

Secondary hyperparathyroidism — In patients with chronic kidney disease (CKD), PTH concentrations are often measured to diagnose and monitor kidney-related bone disease (see "Overview of the management of chronic kidney disease in adults", section on 'Mineral and bone disorders (MBD)'). Although there is an excellent correlation between intact and bioactive PTH assays (r = 0.923 to 0.977) in patients with CKD on maintenance dialysis, mean PTH levels are typically approximately 30 to 50 percent lower when assessed by the bioactive assay compared with the intact assay [15,17,20,72-76]. This discrepancy is due to the large carboxyl (C)-terminal PTH fragments, detected by intact PTH assays, which accumulate in patients with CKD [10-12,17,50]. In some studies, a reduction in glomerular filtration rate (GFR) across the spectrum of CKD is associated with a preferential increase in non-PTH (1-84), rather than PTH (1-84), resulting in a decrease in the ratio of PTH (1-84)/non-PTH (1-84) [77,78].

Because some of these naturally occurring PTH fragments may have biologic activity distinct from that of PTH (1-84) [16,18,54], an independent assessment of PTH (1-84) and non-PTH (1-84) fragments may increase the diagnostic sensitivity to predict the form of bone disease in patients with CKD [79,80]. Although one study suggested that the calculated ratio of PTH (1-84) (measured by the bioactive PTH assay) to non-PTH (1-84) (difference between the results obtained with intact and bioactive PTH assays) may provide a better prediction of bone turnover than each measurement alone [72], similar studies by others are thus far unable to confirm these findings using either bone histomorphometry or biochemical markers of bone remodeling [7,19,20,73,74,81-85].

The National Kidney Foundation Dialysis Outcome Quality and Improving Global Outcomes (K/DOQI and KDIGO) practice guidelines were formulated to help optimally manage secondary hyperparathyroidism and mineral metabolism abnormalities in patients with CKD. These guidelines are reviewed in detail separately. (See "Management of secondary hyperparathyroidism in adult patients on dialysis" and "Management of secondary hyperparathyroidism in adult nondialysis patients with chronic kidney disease".)

Pseudohypoparathyroidism — Pseudohypoparathyroidism (PHP) refers to a group of heterogeneous disorders defined by unresponsiveness to PTH in the proximal renal tubules and characterized biochemically by hypocalcemia and hyperphosphatemia with elevated PTH concentrations. In patients with PHP type 1, there is diminished urinary cyclic adenosine monophosphate (cAMP) excretion in response to exogenous PTH administration due to a deficiency in the proximal renal tubules of the alpha-subunit of the stimulatory G protein (Gs-alpha) that couples to the PTH receptor (PTHR1).

There are several subtypes of PHP type 1 that are caused by heterozygous mutations within or up-stream of GNAS, the gene encoding Gs-alpha. PHP type 1A (PHP1A) is an autosomal-dominant disease caused by maternally inherited loss-of-function mutations involving GNAS exons 1 to 13 that lead to PTH-resistance in the proximal renal tubules, where Gs-alpha is derived predominantly from the maternal allele. Autosomal-dominant PHP type 1B (AD-PHP1B), on the other hand, is caused by microdeletions, duplications, insertions, or inversions within or upstream of GNAS that are associated with the loss of methylation at one or several differentially methylated regions within GNAS. However, most PHP1B patients have a sporadic form of the disorder, which remains in most cases unresolved at the molecular level; in rare instances, it is caused by paternal uniparental isodisomy involving chromosome 20q. Pseudohypoparathyroidism is discussed in detail elsewhere. (See "Etiology of hypocalcemia in infants and children", section on 'End-organ resistance to PTH (pseudohypoparathyroidism)'.)

The ratio of non-PTH (1-84) to PTH (1-84) has been shown to be higher in patients with PHP1A and PHP1B than in 53 healthy controls (0.56±0.07 versus 0.30±0.11) [86], suggesting that non-PTH (1-84) fragments, which may be similar to PTH (7-84), could be sufficiently elevated to reduce the response of a target organ to PTH in this group of patients [86]. Given the putative hypocalcemic properties of PTH (7-84), it is thus conceivable that the accumulation of non-PTH (1-84) fragments contributes to the pathophysiology of PTH-resistant hypocalcemia in PHP. However, these observations require further investigation.

Hypoparathyroidism — Patients with hypoparathyroidism usually have normal or low PTH levels despite hypocalcemia and hyperphosphatemia (see "Hypoparathyroidism"). However, a small family has been described in which the three affected members were shown to have a homozygous R25C pathogenic variant [87,88]. These patients had normal or low PTH levels when measurements were performed with a second-generation assay that uses a detection antibody that was affinity purified with immobilized 13-34; in contrast, very high PTH levels were detected when using a third-generation assay that uses a detection antibody that was affinity purified with immobilized 1-3. Another variant in the secreted PTH (1-84) is pG12E, which is likely to impair biological activity of the peptide [89]. Although pathogenic variants in the secreted PTH are extremely rare, it is conceivable that some patients with hypocalcemia are misdiagnosed with pseudohypoparathyroidism, rather than hypoparathyroidism. Three patients with symptomatic hypocalcemia were reported to have elevated plasma PTH levels by intact PTH immunoassay, but low or undetectable levels in biointact PTH assays. These opposing assay findings variably support the diagnosis of either pseudohypoparathyroidism or hypoparathyroidism, respectively, and are explained by a homozygous nucleotide change that results in replacement of serine with proline at position 1 of mature PTH [90].

SUMMARY AND RECOMMENDATIONS

●Parathyroid hormone assays – Radioimmunoassays for the measurement of parathyroid hormone (PTH) have been largely replaced by two-site immunometric assays (IMAs), which are the preferred method for PTH measurement today. More recently, a liquid chromatography-tandem mass spectrometry (LC-MS/MS) assay was developed and validated for a broad range of PTH (1-84) levels. (See 'PTH assays' above.)

The importance of oxidized forms of PTH remains to be established. (See 'Oxidized PTH' above.)

●Assay variability – PTH values can vary considerably when measured with assays from different manufacturers or different assays from the same manufacturer. Reliable reference ranges should be established in a multicenter and multiethnic study in vitamin D-replete individuals with normal kidney function. Such ranges should probably be assay specific. (See 'Assay variability' above.)

●PTH fragments – Fragment-specific assays may allow insight into the relative contribution of PTH and its fragments to mineral homeostasis in normal and pathophysiological conditions, but they generally remain for research use only (figure 1). (See 'PTH fragments' above.)

●Clinical use – The results obtained with intact and bioactive PTH assays are highly correlated, and both assay types can be used in routine clinical practice for establishing the diagnosis of hyperparathyroidism and for the diagnosis and monitoring of kidney-related bone disease. In patients with chronic kidney disease (CKD), assay-specific target ranges of PTH need to be established for each assay and laboratory before making therapeutic decisions for individual patients. Thus, the clinician must be aware of the particular assay and normal values associated with its use. (See 'Clinical use of PTH assays' above and "Management of secondary hyperparathyroidism in adult patients on dialysis".)