Proopiomelanocortin interference in the measurement of adrenocorticotrophic hormone: a United Kingdom National External Quality Assessment Service study

OBJECTIVE

It is recognized that measurement of ACTH-precursor peptides including proopiomelanocortin (POMC) has clinical utility in identifying the aetiology of Cushing's syndrome. Recent data have also demonstrated cross-reactivity of POMC in ACTH immunoassays used in clinical laboratories. The aim of this study was to assess the cross-reactivity of POMC in the main commercial immunoassays for ACTH and to survey the awareness of laboratory professionals to this potential interference.

METHOD

To assess cross-reactivity, specimens containing ACTH and/or POMC were prepared by the UK National External Quality Assessment Service (UK NEQAS) [Edinburgh]. A separate interpretative exercise was also sent to participating laboratories.

RESULTS

Eighty-seven laboratories measured 'total' ACTH (i.e. ACTH and/or POMC) in their assays. Cross-reactivity of POMC varied from a mean of 1·6-4·7% (reflected in a large percentage increase in measured ACTH of up to 261% due to POMC cross-reactivity) depending on the manufacturer. Major differences in the clinical interpretation of test results were observed in returned responses to the interpretative exercise.

CONCLUSION

An appraisal of POMC cross-reactivity in currently available ACTH immunoassays has been achieved. Cross-reactivity was sufficient to detect ACTH precursors at concentrations that could be found in patients with ectopic ACTH syndrome. These data will assist laboratories in interpreting results when assessing the hypothalamic-pituitary-adrenal axis. Endocrinologists and laboratory professionals should be aware of the degree of cross-reactivity in ACTH immunoassay in order to minimize the risk of misinterpretation of results and/or potentially delayed treatment.

Routine measurement of plasma chromogranin B has limited clinical utility in the management of patients with neuroendocrine tumours

OBJECTIVE

Chromogranin A (CgA) and B (CgB) are markers for monitoring disease status in patients with gastroenteropancreatic neuroendocrine tumours (NETs). These are specialized diagnostic tests often necessitating referral of specimens to a supraregional assay service (SAS) laboratory for analysis. The aim of this audit was to assess whether measurement of either plasma CgA or CgB alone provides sufficient clinical information in comparison with the current practice of measuring both markers together.

DESIGN

A retrospective analysis was undertaken for all chromogranin tests requested for patients with a known NET diagnosis. Results were categorized based on whether plasma concentrations were elevated for one or both CgA and CgB.

RESULTS

A total of 325 sequential patients with a NET diagnosis had plasma chromogranin levels measured during the period of review. Baseline CgA was elevated in 60·9% of patients. Isolated elevations in CgA (with normal CgB) were found in 44·9% of patients, whilst combined elevations in both CgA and CgB were found in 16% of patients. Combined CgA and CgB concentrations within the normal range were observed for 38·5% of patients. Only two patients (0·6%) had an isolated elevation in CgB at baseline. Both patients had a diagnosis of pancreatic NET and were radiologically stable. Plasma CgA and CgB corresponded with disease stage (localized vs metastatic). CgB in addition to CgA did not provide any significant improvement in diagnostic performance for identification of metastatic disease compared to CgA alone.

CONCLUSIONS

Based on this NET population and specific assay performance characteristics, CgA alone provides sufficient information for the management of NET patients; the routine estimation of CgB in all patients is not informative in clinical practice.

False positive carbohydrate antigen 19-9 (CA19-9) results due to a low-molecular weight interference in an apparently healthy male

BACKGROUND

We investigated the presence of interference in a patient who had an elevated CA19-9 concentration using the ADVIA Centaur but results within reference limits with ROCHE Modular Analytics E170 and Brahms KRYPTOR analysers.

METHODS

We performed repeat analyses using the same (ADVIA Centaur) and alternate immunossays (Roche Modular Analytics E170 and Brahms KRYPTOR) on the patient's sample and investigated for known interferences. To determine the nature of the interference, we measured CA19-9 on the ADVIA Centaur after dilution experiments and after incubation with non-immune animal sera and in heterophilic blocking tubes (HBT). We also undertook polyethylene glycol precipitation, lectin inhibition experiments and gel filtration chromatography.

RESULTS

A curvilinear response to dilution was observed with the ADVIA Centaur. Other known interferences were excluded. Treatment with HBT or non-immune animal sera did not give clinically different results from untreated samples. There was only 0.59% recovery after PEG precipitation in the sample from the case patient. Lectin reduced the assay signal in four patient samples (recovery=1.9-14.1%) but not in the case patient (recovery=106.2%). Gel filtration studies suggested the presence of a low molecular weight (approximately 100 kDa) interference in the case patient's serum.

CONCLUSIONS

We report a novel mode of interference and show a non-CA19-9, low molecular-weight interference affecting the ADVIA Centaur CA19-9 immunoassay.