Dense core secretory vesicles, present throughout the neuroendocrine system, store and release chromogranins. Chromogranins are a group of acidic, soluble secretory proteins which are thought to play a role in the formation of vesicles.
Chromogranins (A, B, and C) form a major constituent of these granules and are widely distributed in the neuroendocrine system. As a result, they are excellent tissue and serum markers for neuroendocrine tumours.
Chromogranin A
Chromogranin A (CgA) was the first of the ‘granins’ to be identified and has the widest distribution. It is quantitatively the major constituent of the secretory granules in the NE cells and is expressed in the cells of the anterior pituitary, C cells of the thy roid, chief cells of the parathyroid, islet cells of the pancreas and the chromaffin cells of the adrenal medulla. It is also widely distributed in the neuroendocrine cells of the bronchial and gastropancreatic systems and the skin.
The human form of CgA is an acidic 439 amino acid protein that is preceded by an 18 amino acid signal peptide. Both the N and C terminals are well conserved between species. The CgA molecule contains a number of mono and dibasic amino acid sites, thus implying fairly extensive and varied post- translational processing. This suggests that it may be a precursor to a number of other peptides. Although the function of CgA has not yet been fully elucidated, it is often cosecreted with the neuroendocrine hormones and peptides and therefore thought to play a role in the processing, packaging, and secretion of neuropeptide precursors and hormones.
The sensitivity and specificity of circulating CgA measurement in the biochemical diagnosis of NET have been extensively studied and values calculated range from 56% to 100%. This variation in values can be attributed to the heterogeneous composition of NET subtypes used in the patient groups, the varying number of subjects used in the patient and control groups, the different reference intervals used, the different methods to obtain the reference intervals and the different CgA assays used.
To investigate the difference between CgA assays, a comparison of four kits demonstrated no significant difference in sensitivity between the assays, however, the sensitivity could be increased significantly when a combination of two non- cross reacting assays were used.
A meta- analysis of 13 studies meeting the inclusion criteria out of 78 potential studies revealed a sensitivity and specificity of 73% and 95%, respectively. Furthermore, the area under the curve was 0.896 indicating the usefulness of CgA. Therefore, most NETs are as sociated with increased circulating levels of CgA. Even tumours that produce hormones with no identifiable clinical features or have lost the ability to synthesize peptides, continue to express CgA As a result, CgA is routinely used as a marker for both diagnosis and monitoring of NETs. It is particularly useful (i) when existing cell- specific markers (explained later in the chapter) are either unstable, rapidly fluctuating or inconvenient for clinical use; (ii) to confirm the neuroendocrine origin of a tumour; and (iii) as a general marker of disease when the neoplastic disease involves multiple neuroendocrine tissues (e.g. multiple endocrine neoplasia).
CgA has been shown to correlate with tumour burden and extent of metastases. Increasing concentrations of CgA can also indicate worse prognosis. However, high circulating CgA concentrations have been measured in gastrinoma patients despite the small primary tumour and absence of metastases.
CgA can be used to monitor recurrence of disease. A comparison of CgA, urinary 5HIAA and radiological measurements in patients with a midgut NET found that elevated plasma CgA was the first to indicate disease recurrence.
There are, however, some limitations to using CgA as a tumour biomarker. An increase in circulating CgA levels due to a tumour often goes undetected until it has reached a size capable of producing appreciably increased amounts of CgA. Further, CgA cannot differentiate between different subtypes of NET and it is not equally expressed in all NETs. Some only weakly express CgA (e.g. small cell carcinoma of the lung) Concentrations are only minimally elevated in patients with insulinomas. Most patients with metastatic foregut and midgut carcinoids have increased (tumour burden de pendent) levels of CgA. Gastrinomas may show increased levels of CgA even in patients with very limited disease. This is prob ably because chronic hypergastrinemia causes hyperplasia of the enterochromaffin cells.
The biological variation of circulating CgA levels has been studied extensively and levels can vary by 30% from day to day in healthy controls and patients alike. In addition, there are a number of non- neoplastic conditions are associated with high CgA levels. Patients on proton pump inhibitors (PPI) have raised CgA levels. Renal impairment, and to a lesser extent, liver failure can also result in increased levels.
CgA undergoes post- translational changes before release. Tumours may, therefore, release different molecular forms of CgA. As a result, a number of different forms of CgA are released into circulation. Thus, levels measured are dependent on the antiserum used. Some assays measure only the whole CgA, while others measure the whole CgA plus fragments that contain the specific epitope to which the antisera was developed. Thus reference ranges are assay- specific. A lack of standardization between the various commercially available immunoassays makes comparison difficult. However, sequential measurement with the same assay can reliably be used to monitor disease progression in a patient.
Pancreastatin is a 49- amino acid (CgA 240– 288) peptide produced by dibasic cleavage of CgA. Pancreastatin assays that use antisera raised to the mid- molecule cross- react strongly to CgA and can be used to measure both pancreastatin and CgA. Assays using antisera raised to the N or C terminals of pancreastatin are, how ever, specific for pancreastatin. Pancreastatin concentrations are shown to correlate well with extent of liver involvement. Specific pancreastatin assays may therefore be used to assess and monitor the extent liver involvement in patients with NET. The advantage of assays using antibodies specific for pancreastatin is that they are unaffected by PPIs.
Chromogranin B
Chromogranin B (657 amino acid peptide) (CgB) or GAWK (a partial sequence of CgB 420– 493) coexists with CgA in the secretory granules and bears a strong homology to CgA in the ter minal regions. Like CgA it is an acidic protein. The relative abundance of CgA and CgB is cell specific. CgA is the dominant granin in the pancreatic endocrine tumours and in the serotonin secreting tumours in the ileum and appendix. However, in rectal carcinoids, where CgA is virtually absent, CgB is the most abundant granin. Elevated CgB concentrations have been detected in some NET patients when CgA is not raised. It is, therefore, measured complementary to CgA in some centres, and improves diagnostic sensitivity. CgB has also been shown to be a more sensitive marker of PNETs than CgA. CgB assays are, however, associated with the same problems of standardization as the CgA assays. The combined measurement of CgA and CgB has a sensitivity of 89% for NETs. A polyclonal anti serum with cross reactivity to both CgA and B has been developed by Eriksson et al..
Circulating levels of CgB have only been shown in phaeochromocytoma to correlate with tumour bulk and there is little evidence to suggest that CgB can be used to predict disease recurrence. Unlike CgA, CgB is not affected by PPI treatment. It is, however, like CgA effected by renal impairment.
Chromogranins are best measured in plasma. Fasting samples are not required. Samples must be centrifuged, plasma aliquoted, and stored at – 20°C immediately after collection, to prevent degradation of the peptide. Patients on acid suppressive therapy should be ad vised to come off treatment before CgA is measured.
