6 Diagnosis of a malignant adrenal mass: URRENT PINION the role of urinary steroid metabolite profiling
Irina Bancosª and Wiebke Arltb,c
Purpose of review
Adrenal masses are highly prevalent, found in 5% of the population. Differentiation of benign adrenocortical adenoma from adrenocortical carcinoma is currently hampered by the poor specificity and limited evidence base of imaging tests. This review summarizes the results of studies published to date on urine steroid metabolite profiling for distinguishing benign from malignant adrenal masses.
Recent findings
Three studies have described cohorts of at least 100 patients with adrenal tumors showing significant differences between urinary steroid metabolite excretions according to the nature of the underlying lesion, suggesting significant value of steroid metabolite profiling as a highly accurate diagnostic test.
Summary
Steroid profiling is emerging as a powerful novel diagnostic tool with a significant potential for improving the management for patients with adrenal tumors. Although the current studies use gas chromatography- mass spectrometry for proof of concept, widespread use of the method in routine clinical care will depend on transferring the approach to high-throughput tandem mass spectrometry platforms. The use of computational data analysis in conjunction with urine steroid metabolite profiling, that is, steroid metabolomics, adds accuracy and precision.
Keywords
adrenal tumor, adrenocortical carcinoma, diagnostic test, mass spectrometry, steroid metabolomics, steroid profiling
INTRODUCTION
Adrenal masses are common, with a reported overall prevalence of at least 5% in the population [1,2]. Prevalence increases with age, ranging between less than 0.5% in children and around 10% in 70-year- old patients [3,4]. Most patients are diagnosed inci- dentally on cross-sectional abdominal imaging per- formed for indications other than suspected adrenal disease. In the United States alone, the number of computed tomography (CT) imaging performed quadrupled from 21 million per year in 1995 to more than 80 million per year in 2014 [5]. Increase in widespread use coupled with enhancement in the quality of cross-sectional imaging explains the growing numbers of patients newly diagnosed with an adrenal tumor requiring further diagnostic assessment.
Management of patients with a newly detected adrenal mass is dictated by whether there is evidence of associated adrenal hormone excess and whether the adrenal mass is malignant or benign. However, because of the limited accuracy of currently applied diagnostic tools, reaching a confident conclusion in
regards to these two key questions is challenging for many patients with adrenal tumors. Patients fre- quently require additional immediate and/or longi- tudinal follow-up tests and visits, and invasive procedures to ascertain the precise diagnosis, such as adrenal biopsy or unilateral adrenalectomy, add to the burden of disease associated with adrenal incidentalomas.
In this review, we will discuss the evidence on currently available tests to diagnose adrenal malig- nancy and review the data on urinary steroid
ªDivision of Endocrinology, Metabolism and Nutrition, Mayo Clinic, Rochester, Minnesota, USA, bInstitute of Metabolism and Systems Research (IMSR), University of Birmingham and “Centre for Endocrin- ology, Diabetes and Metabolism (CEDAM), Birmingham Health Partners, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK
Correspondence to Irina Bancos, Division of Endocrinology, Metabolism and Nutrition, Mayo Clinic, Rochester, MN, 55902, USA. Tel: +1 507 776 1387; e-mail: Bancos.Irina@mayo.edu
Curr Opin Endocrinol Diabetes Obes 2017, 24:200-207 DOI:10.1097/MED.0000000000000333
KEY POINTS
· Noninvasive diagnosis of malignant adrenal mass is challenging because of suboptimal accuracy of currently employed imaging modalities. In patients with benign adrenal tumors with indeterminate imaging characteristics, current standard of care employs additional imaging tests, occasional adrenal biopsy, and even unnecessary adrenalectomy.
. 24-h urinary steroid profiling reflects steroidogenic output of the adrenal glands and includes measurements of androgen, glucocorticoid and mineralocorticoid precursors, and metabolites. Significant differences in urinary steroids of patients with ACC in comparison with other adrenal tumors have been described.
· Based on several retrospective studies, the performance of steroid profiling as measured by GC-MS was reported at more than 90% sensitivity and specificity for the diagnosis of ACC.
· Despite a significant potential of steroid profiling as a powerful noninvasive diagnostic tool in patients with adrenal tumors, widespread use in routine clinical care will depend on transferring the approach to high- throughput tandem mass spectrometry platforms and on appropriate validation of the method in a large prospective cohort of patients with adrenal tumors.
profiling as a potential noninvasive diagnostic test to detect adrenocortical carcinoma (ACC) in patients with adrenal masses.
DIAGNOSIS OF A MALIGNANT ADRENAL MASS
Most adrenal tumors are discovered incidentally on cross-sectional imaging study performed for a reason other than suspected adrenal disease, for example, unspecific abdominal pain. Imaging characteristics can be helpful in clarifying the nature of the adrenal mass. Pheochromocytomas can be taken out of the equation by highly sensitive and specific biochemical screening with measurement of plasma metanephr- ines. Doing this first helps avoid difficulties in inter- preting imaging findings as there is a significant overlap between pheochromocytomas and ACCs.
Following biochemical exclusion of a catechol- amine-producing pheochromocytoma, imaging is the next step to detect or exclude adrenocortical malignancy. Although a large tumor diameter has limited sensitivity and specificity for detecting an ACC [6,7], additional features such as irregular bor- ders, inhomogeneity of the mass, and signs of local invasion are pointing to underlying adrenocortical malignancy. Tumor density on unenhanced,
noncontrast CT scans is helpful if a homogeneous density value of less than 10 Hounsfield units is measured in the tumor area, which invariably indicates a fat-containing benign adrenal tumor. Conversely, although Hounsfield units values more than 10 are suspicious of malignancy, the majority of such masses eventually turn out to be benign. Similarly, on MRI scans, the loss of signal intensity in the so-called chemical shift analysis also corre- lates with the amount of fat contained in the adre- nal tumor, with a large drop in signal intensity between in-phase and out-of-phase spin-lattice longitudinal relaxation time sequences suggestive of a benign adrenal tumor. Fluorodeoxyglucose PET has been proposed as a better test to diagnose a malignant adrenal mass. However, surprisingly little evidence exists on the diagnostic accuracy of the imaging characteristics of the commonly employed imaging studies, mainly because of the fact that few studies provide an optimal reference standard for comparison [8”]. Although evidence for the accurate diagnostic performance of noncontrast CT Houns- field units cutoff of 10 is more robust, confidence intervals of estimates for other imaging character- istics are wide, based on the very small sample size of studied cohorts (Table 1), and therefore their use, though regularly employed, is not based on high-quality evidence[8”,9”].
Because of the limited specificity of imaging tests in making the diagnosis of adrenocortical malig- nancy in the setting of a high prevalence of benign adrenal tumors, the current diagnostic approach, which almost exclusively employs imaging tests, has a significant impact on the economic burden of disease. Many patients undergo either immediate additional imaging tests to better characterize the adrenal mass or are advised to return for imaging studies to assess tumor growth after an interval of 3- 12 months [10""]. This exposes patients to radiation and increases healthcare costs, without evidence that performing multiple imaging modalities improves the diagnostic performance with regard to differen- tial diagnosis between benign and malignant adrenal masses. Some guidelines recommend annual imag- ing follow for 5 years after the initial detection of an adrenal mass [11,12]. However, following this strategy in patients with a conclusively benign pres- entation on initial imaging, for example, Hounsfield units less than 10 on unenhanced CT, may result in significant radiation exposure and healthcare costs for no good reason. Therefore, the recently published joint clinical guidelines of the European Society for Endocrinology (ESE) and the European Network for the Study of Adrenal Tumors (ENSAT) have recom- mended that such patients should not undergo repeat imaging[10""].
| Diagnostic test | Studies“ (n) | Patientsb (n) | Sensitivity | Specificity |
|---|---|---|---|---|
| Noncontrast CT (HU >10)" | 2 | 102 | 100% (91-100%) | 72% (60-82%) |
| CT contrast-enhanced washout at 10 minc | 1 | 25 | 93% (68-100%) | 100% (69-100%) |
| MRI, loss of signal intensity“ | 2 | 75 | 86% (31-99%) | 85% (73-93%) |
| PET ALR SUV max“ | 2 | 64 | 100% (78-100%) | 96% (57-100%) |
| PET SUV max® | 2 | 64 | 93% (65-99%) | 73% (59-84%) |
| Adrenal biopsy (any malignancy)d | 7 | 217 | 87% (78-93%) | 100% (76-100%) |
| Adrenal biopsy (for ACC)d | 4 | 107 | 70% (42-88%) | 98% (89-98%) |
ACC, adrenocortical carcinoma; HU, Hounsfield units.
“Number of studies included in systematic review.
bNumber of patients included in systematic review.
“Inclusion criteria included original studies reporting on patients with incidentally discovered adrenal tumors and a reference standard for at least 50% of population (reference standard included imaging follow-up (for benign adrenal tumors) and histology after surgery or biopsy (for benign and malignant adrenal tumors). Adapted with permission [8”] dInclusion criteria included original studies on patients undergoing adrenal biopsy and a reference standard of at least 50% population (reference standard included histology from adrenalectomy or autopsy, follow-up imaging at 3-12 months, or clinical follow-up of 2 years). Adapted with permission [9”].
In a proportion of patients, image-guided adre- nal biopsy or adrenalectomy are performed because of concerns about the potentially malignant nature of the incidentally detected adrenal mass. Adrenal biopsy is an expensive procedure with a reported rate of nondiagnostic biopsies of 8.7% and a com- plication rate of 2.5% [9”]. In addition, adrenal biopsy has poor accuracy in making the diagnosis of ACC (Table 1), because of inherent difficulties for pathologists to making an accurate diagnosis based on very limited amounts of tissue, and therefore is not recommended for patients with suspected ACC [10"",13]. In principle, adrenal biopsy is mainly considered in patients with extraadrenal malig- nancy and only, as stated by the European Society for Endocrinology (ESE)-ENSAT guidelines, if ‘all of the following criteria are fulfilled: the lesion is hor- monally inactive (in particular, a pheochromocy- toma has been excluded), the lesion has not been conclusively characterized as benign by imaging, and management would be altered by knowledge of the histology’ [10""].
As adrenalectomy series suggest, one-third to half of all patients undergo adrenalectomy unnecessarily as lesion is consistent with a benign nonfunctioning adrenal tumor [7,10""]. This is likely because of underlying concern for malignancy, possibly related to larger tumor diameters. Indeed, despite the poor accuracy of tumor size in predicting malignancy, several published guidelines recom- mend adrenalectomy at various cutoffs of tumor size of 4-6 cm [11,12,14]. Recent guidelines on man- agement of adrenal tumors acknowledge the lack of evidence on the natural history of large apparently benign adrenal tumors to suggest a certain adrenal
tumor cutoff for adrenalectomy but allow for a highly individualized consideration of surgery in patients with tumors greater than4 cm [10""]. Given the shortcomings of the currently employed imag- ing tests, a noninvasive, accurate, and inexpensive test to distinguish ACC from other adrenal tumors is urgently needed, especially in patients with large adrenal tumors and in patients with adrenal masses with indeterminate imaging characteristics.
STEROID METABOLITE PROFILING
Serum and urinary steroid analysis traditionally play an important role in the diagnosis of adrenal hor- mone excess and disorders of steroidogenesis. Around 15% of adrenal tumors present with either clinically overt adrenal cortex hormonal excess, including overt Cushing syndrome, primary hyper- aldosteronism, and, uncommonly, hyperandrogen- ism [7,15]. In addition, up to 30-50% of patients with adrenal tumors present with evidence of mild autonomous cortisol excess, previously termed sub- clinical Cushing syndrome; the diagnosis of these patients is challenging and management is usually delayed until cortisol-induced comorbidities occur [16”]. Although serum steroid measurements are susceptible to differences occurring because of the diurnal variation in adrenal steroid secretion, 24-h urine steroid excretion represents a more accurate estimate of net total adrenal hormone production [17,18] and thus also holds diagnostic potential for the assessment of adrenal hormonal excess.
Quantitatively, the majority of steroids excreted in the urine of adult individuals are androgens and glucocorticoids with a significantly lower output for
mineralocorticoids and steroid precursors [18-20]. Adrenal steroid production and metabolism are affected by sex, age, and BMI [18,21,22]. Glucocor- ticoid metabolites are higher in men than in women and in obese than in lean persons [21,22]. Looking at ratios of steroids that are substrates and products, respectively, of a distinct steroidogenic enzyme activity allows for representative assessment of elements of the steroidogenic pathway, which is particularly informative in the diagnosis of inborn steroidogenic disorders, including congenital adre- nal hyperplasia. However, by reflecting steroido- genic output of the adrenal glands, urinary steroid profiling inherently has great potential in diagnos- ing adrenal hormone excess.
STEROID METABOLITE PROFILING IN DIAGNOSIS OF ADRENOCORTICAL CARCINOMA
Most ACCs are large tumors at the time of initial diagnosis, and a significant number is found to produce adrenal hormones in excess, though this is not always clinically apparent. In general, the concurrent excess production of multiple steroids, in particular when including androgens, is con- sidered as indicative of adrenocortical malignancy. The majority of ACCs are classified as nonfunction- ing based on conventional serum steroid analysis, which mostly assesses end products of steroidogen- esis. However, ACC cells can be considered to represent a more immature, dedifferentiated cell type as is typical for cancer cells. Hence, it is likely that the steroidogenic pattern produced by adrenal cancer cells is characterized by large amounts of steroid precursors rather than end products of mature and complete steroidogenesis. Consistent with this assumption, a significant number of patients with ACC show increased production of the glucocorticoid precursor 17-hydroxyprogester- one (17OHP), up to half of ACC patients reported by one series [18]. However, serum 17OHP only pro- vides limited sensitivity for detecting enhanced steroid precursor production in ACC patients.
Decades ago, small case series demonstrated differences in urine steroid metabolite excretion in some patients with ACC [23,24]. However, the first large study systematically investigating steroid metabolite excretion in 147 patients with confirmed adrenocortical tumors, 45 patients with ACC, and 102 patients with adrenocortical adenoma (ACA), or benign ACC, was published in 2011 by Arlt et al. [18]. In this proof-of-concept study, the authors described steroid profiling by gas chromatog- raphy-mass spectrometry (GC-MS) combined with computational data analysis utilizing a machine
learning-based approach [18]. The employed GC- MS steroid profiling included measurement of 32 adrenal steroids, including metabolites of glucocor- ticoid, mineralocorticoid, and androgen precursors, providing comprehensive coverage of adrenal steroid output (Fig. 1). This study revealed a distinct steroid pattern, a ‘malignant steroid fingerprint’, that indicated the presence of an ACC with 90% sensitivity and specificity. Three steroid metabolites were most informative in distinguishing ACC from ACA: the glucocorticoid precursor metabolite tetrahydro-11-deoxycortisol (THS) derived from 11- deoxycortisol and the adrenal androgen precursor metabolites pregnenediol and pregnenetriol derived from pregnenolone and 17-hydroxypregnenolone, respectively. In addition, six further steroids were identified on the machine learning algorithm, based on their relevance toward discriminating ACC from ACA: the progesterone metabolite pregnane- diol, the 17OHP metabolite pregnanetriol, the 11-deoxycorticosterone metabolite tetrahydro-11- deoxycorticosterone, the corticosterone metabolite 5a-tetrahydrocortisol, the cortisol metabolite 5x- tetrahydrocortisol (THF), and the androgen metab- olite etiocholanolone. Receiver operating character- istic curve analysis demonstrated the diagnostic information of all 32 steroids resulted in 90% sen- sitivity and specificity, whereas the nine most infor- mative and the top three biomarker steroids provided only slightly lower diagnostic accuracies of 87.7% and 87.2%, respectively. Ten of the 45 ACC patients in this study had metastatic disease at the time of 24-h urine collection; however, their steroid profile did not differ from ACC patients who har- bored the primary tumor only. Steroid output did not correlate with tumor size.
In 2015, Kerkhofs et al. [19] presented their data on an independent cohort of patients with 27 ACCs and 125 ACAs. They performed GC-MS analysis of 22 steroid metabolites but did not employ compu- tational analysis to refine the diagnostic perform- ance of the test. They identified 15 individual steroid markers with a sensitivity of 90% or above in detecting ACC, with the 11-deoxycortisol metab- olite THS again representing the most informative markers. Other informative markers identified by that study included pregnanediol, pregnanetriol, etiocholanolone, the major androgen metabolite androsterone, and the major glucocorticoid metab- olites THF and tetrahydrocortisone. However, spe- cificities varied widely (2-83% for individual steroids). They defined a cutoff of 2.35 pmol/24 h for THS excretion as 100% sensitive and 99% specific for the detection of ACC. However, their ACC cohort was considerably smaller and 67% of their ACC patients showed evidence of cortisol excess,
HO
Glucocorticoid precursors
Mineralocorticoid precursors
Mineralocorticoids
Cholesterol
CYP11A1
STAR
THDOC, 5aTHDOC
THA, 5a THA, THB, 5a THB
5-PD
PD
THAIdo
Preg- nenolone
HSD3B2
Pro- gesterone
CYP21A2
11-deoxy- cortico- sterone
CYP11B2
Cortico- sterone
CYP11B2
18-hydroxy- cortico- sterone
CYP11B2
CYP11B1
Aldosterone
CYP17A1
CYP17A1
THE 5aTHF
5-PT
PT, 17HP
THS
THE
17-hydroxy- pregnenolone
HSD3B2
CYP21A2
HSD11B2
17-hydroxy- progesterone
11-Deoxy- cortisol
CYP11B1
Cortisol
Cortisone
HSD11B1
CYP17A1
CYP17A1
Glucocorticoids
DHEA, 160-
OH-DHEA
An, Etio
An, Etio
An
Dehydro-
HSD3B2
Andro- stenedione
HSD17B3
Testo- sterone
SRD5A1/2
5a-dihydro- testosterone
epiandrosterone
AKR1C3
Androgen precursors
Androgens
whereas this is found in only 18% of their ACA patients. (Table 2). This selection bias might explain why they found the glucocorticoid metabolites, tetrahydrocortisone and THF to be apparently dis- tinguishing between ACC and ACA.
In a recent study by Velikanova et al. [20], carry- ing out GC-MS analysis of 32 steroid metabolites in 24-h urine samples from 31 patients with ACC and 96 patients with ACA, the authors confirmed the findings reported by Arlt et al. [18]. They demon- strated that patients with ACC have significantly higher THS, pregnanediol, 5-pregnanetriol, and 5- pregnanediol than patients with ACA. However, increased urinary excretion of THS was demon- strated only in 74% of their patients with ACC and the authors suggested consideration of additional steroid parameters and ratios [20]. Similar to Kerkhofs et al. [19] they did not employ unbiased
computational data analysis of the steroid excretion data.
All three studies identified THS as the most informative steroid marker indicative of ACC. THS is the metabolite of 11-deoxycortisol, which is con- verted to cortisol by the adrenal-specific steroido- genic enzyme cytochrome P450 family 11 subfamily B member 1 (CYP11B1). Measurement of 11-deox- ycortisol is not routinely used clinically, though recently serum results were reported in a cohort of benign adrenal tumors [25]. Abundance of THS in patients with ACC suggests a relative deficiency of CYP11B1, and may be because of dedifferentiation and mutational changes occurring in ACC. Similarly, the accumulation of the androgen precur- sor steroids 5-pregnanetriol and 5-pregnanediol indicate a relative inefficiency of cytochrome P450 family 17 subfamily A member 1 (17,20 lyase)
| Author, year | Period of data collection | Type of study | Sample size (n) | Women (n, %) | Age of diagnosis (years) | Tumor size (cm) | ACC | ACA | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| n | Hormone excess | n | Hormone excess | ||||||||
| Arlt, | 2003-2006 | Retrospective | 147 | 84 (57%) | ACA (median, | ACA (median, | 45 | None: 12 (27%) | 102 | None: 69 (68%) | |
| 2011 [18] | 6 centers | ranges): 60 | ranges): 2.6 | Cortisol: 11 (24%) | Cortisol: 14 (14%) | ||||||
| (19-84) | (0.9-7.8) | Androgen: 7 (16%) | Aldosterone: 13 (13%) | ||||||||
| ACC (median, | ACC (median, ranges): 9 (1.4-23) | Cortisol + androgen: | Cortisol + aldosterone: | ||||||||
| ranges): 55 | 12 (27%; 2 also | 4 (4%) | |||||||||
| (20-80) | with aldosterone) Estrogen: 3 (7%; in combination with cortisol in 2 and androgen in 1 patient) | Androgen: 2 (2%) | |||||||||
| Kerkhofs, 2015 [19] | 2000-2011 | Retrospective 2 centers | 152 (includes 18 noncortical tumors) | 100 (66%) | All adrenal tumors | All adrenal tumors (median, ranges): 3.5 (0.8-17) | 27 | None: 7 (26%) | 107 | None: 85 (79%) Cortisol: 19 (18%) | |
| (mean, SD): | Hormone excess in | ||||||||||
| 56 (13) | 20 (74%) - single or in combination including: Cortisol: 18 (67%) Androgen in 14 (52%) Estrogen in 1 (4%) | Aldosterone: 3 (3%) | |||||||||
| Velikanova, 2016 [20] | 2014-2015 | Retrospective 3 centers | 139 | 83 (60%) | Nonfunctioning | Nonfunctioning | 31 | Hormone excess | 108 | None: 52 (48%) | |
| ACA (median, | ACA (median, | reported for | Cortisol: 44 (41%) | ||||||||
| ranges): | ranges): | cortisol only in 13 | Aldosterone: | ||||||||
| 55 (50-61) | 3.3 (2.3-4.5) | (42%) patients; | 12 (11%) | ||||||||
| Cortisol-secreting | Cortisol-secreting ACA (median, ranges): 3 (2.5-4.2) | other hormonal | |||||||||
| ACA (median, | excess was | ||||||||||
| ranges): 48 (21-54) | not reported | ||||||||||
| ACC (median, ranges): 43 (33-57) | ACC (median, ranges): 9.1 (7.2-11) | ||||||||||
ACA, adrenocortical adenoma; ACC, adrenocortical carcinoma.
activity in converting 17-hydroxypregnenolone to the major adrenal androgen dehydroepiandroster- one. Based on the available studies, it is evident that the excretion of steroid precursor metabolites is significantly higher in ACCs than in ACAs and we propose that this is because of the relative steroido- genic immaturity of the ACC cells. Steroid precursor metabolite excretion was equally observed both in patients with clinically evident adrenal hormone excess as well as in patients with clinically ‘endo- crine inactive’ ACC. Interestingly, steroid profiling revealed higher glucocorticoid metabolite excretion in ACA than in healthy volunteers, including those classified as nonfunctioning [17,18,20]. At this time, studies systematically exploring potential diagnos- tic implications of urine steroid profiling in hormo- nal excess are currently lacking.
CHALLENGES AND FUTURE DIRECTIONS
Steroid profiling has particular promise as a valuable tool for differential diagnosis in patients with adre- nal tumors presenting with indeterminate imaging characteristics. These patients proceed to have additional studies, most commonly repeat CT imag- ing which add to the lifetime exposure to radiation, but also expose affected patients to cost, inconven- ience, and anxiety of waiting another 6-12 months. Steroid profiling is an attractive noninvasive alternative which could help rule out ACC much earlier in the diagnostic pathway. Earlier diagnosis would allow for adrenalectomy without delay in patients with suspected ACC and avoid additional procedures and tests in patients in whom ACC is conclusively excluded.
Although the above initial results suggest impressive differences between 24-h urine steroid metabolite excretion in patients with ACC when compared with ACAs, at this time there are some barriers which need to be overcome before wide- spread use of steroid profiling as a noninvasive diagnostic tool in patients with adrenal tumors. All initial results are based on retrospective studies with still relatively small numbers of patients total- ing less than 500 patients. Prior to implementation in routine clinical practice, urinary steroid metab- olome profiling as a diagnostic tool will need to be prospectively validated in a much larger cohort of prospectively enrolled patients and compared with a reference standard comprising results of histopa- thology and clinical and imaging follow-up inves- tigations. Recently, a large prospective international multicenter test validation study carried out with support of the ENSAT has completed after successfully recruiting more than 2000 patients with newly diagnosed adrenal mass. Results from this
study, Evaluation of URINE steroid metabolomics in the differential diagnosis of AdrenoCortical Tumors (EURINE-ACT), will conclusively determine the diagnostic accuracy and performance of the steroid profiling in a nonselected cohort of patients in a ‘real-life’ setting.
Even after appropriate validation is completed, there are certain challenges associated with univer- sal implementation of steroid profiling into the clinical laboratory. All available current evidence on steroid profiling relies on measurements per- formed by GC-MS. However, although GC-MS allows the concurrent profiling of the entire steroid metabolome, the method requires special expertise and is only offered by a small number of institutions and laboratories. The method is relatively laborious and hence not cheap. However, these obstacles could be overcome by transfer of the steroid profil- ing method to the high-throughput liquid chroma- tography-tandem mass spectrometry platform and these approaches have now been developed and are currently being investigated for their diagnostic performance. For widespread use of this diagnostic test, it will also be critical to provide a straightfor- ward diagnostic algorithm and this is likely to be achieved by the machine learning-based algorithm described by Arlt et al. [18].
CONCLUSION
In conclusion, mass spectrometry-based 24-h urine steroid metabolome profiling is a highly promising noninvasive diagnostic tool in patients with adrenal tumors, with a diagnostic sensitivity and specificity that exceeds that of diagnostic imaging. Although proof-of-concept and subsequent studies were car- ried out in retrospective cohorts, the outcome of a large prospective test validation study will be avail- able soon. Results will determine whether steroid profiling can become part of widely implemented routine care in the diagnostic evaluation of patients with adrenal masses.
Acknowledgements
None.
Financial support and sponsorship
The work was supported by the Mayo Clinic Foundation for Medical Education and Research (Traveling Scholar- ship, to I.B.), the Medical Research Council UK (Strategic Biomarker Grant G0801473, to W.A.), and the Euro- pean Union under the Seventh Framework Program (FP7/ 2007-2013, Grant Agreement 259735, ENSAT-CAN- CER, to W.A.).
W.A. holds a patent on the use of steroid metab- olomics for the differential diagnosis of adrenal tumors.
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
of special interest
of outstanding interest
1. Bovio S, Cataldi A, Reimondo G, et al. Prevalence of adrenal incidentaloma in a contemporary computerized tomography series. J Endocrinol Invest 2006; 29:298-302.
2. Song JH, Chaudhry FS, Mayo-Smith WW. The incidental adrenal mass on CT: prevalence of adrenal disease in 1049 consecutive adrenal masses in patients with no known malignancy. AJR Am J Roentgenol 2008; 190:1163-1168.
3. Ciftci AO, Senocak ME, Tanyel FC, Buyukpamukcu N. Adrenocortical tumors in children. J Pediatr Surg 2001; 36:549-554.
4. Terzolo M, Stigliano A, Chiodini I, et al., Italian Association of Clinical Endocrinologists. AME position statement on adrenal incidentaloma. Eur J Endocrinol 2011; 164:851-870.
5. OECD. Computed tomography (CT) exams (indicator). 2017
6. Hamrahian AH, Ioachimescu AG, Remer EM, et al. Clinical utility of noncon- trast computed tomography attenuation value (hounsfield units) to differenti- ate adrenal adenomas/hyperplasias from nonadenomas: Cleveland Clinic experience. J Clin Endocrinol Metab 2005; 90:871-877.
7. Mantero F, Terzolo M, Arnaldi G, et al. A survey on adrenal incidentaloma in Italy. Study Group on Adrenal Tumors of the Italian Society of Endocrinology. J Clin Endocrinol Metab 2000; 85:637-644.
8. Dinnes J, Bancos I, Ferrante di Ruffano L, et al. Management of endocrine
disease: imaging for the diagnosis of malignancy in incidentally discovered adrenal masses: a systematic review and meta-analysis. Eur J Endocrinol 2016; 175:R51-R64.
The review includes only studies providing a stringent reference standard and provides diagnostic accuracy estimates of several commonly employed imaging tests.
9. Bancos I, Tamhane S, Shah M, et al. Diagnosis of endocrine disease: the
diagnostic performance of adrenal biopsy: a systematic review and meta- analysis. Eur J Endocrinol 2016; 175:R65-R80.
The review provides rates of nondiagnostic biopsies, complications, and diagnostic accuracy of adrenal biopsy.
10. Fassnacht M, Arlt W, Bancos I, et al. Management of adrenal incidentalomas:
European Society of Endocrinology Clinical Practice Guideline in collabora- tion with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol 2016; 175:G1-G34.
In this work, authors provide guidelines on diagnostic approach and management of patients with adrenal tumors. The guidelines address several clinical dilemmas, including imaging follow-up, surgical approach, and evaluation and management of patients with mild hypercortisolism.
11. Kapoor A, Morris T, Rebello R. Guidelines for the management of the incidentally discovered adrenal mass. Can Urol Assoc J 2011; 5:241- 247.
12. Zeiger MA, Thompson GB, Duh QY, et al., American Association of Clinical Endocrinologists; American Association of Endocrine Surgeons. American Association of Clinical Endocrinologists and American Association of Endo- crine Surgeons Medical Guidelines for the Management of Adrenal Inciden- talomas: executive summary of recommendations. Endocr Pract 2009; 15:450-453.
13. Delivanis DA, Erickson D, Atwell TD, et al. Procedural and clinical outcomes of percutaneous adrenal biopsy in a high-risk population for adrenal malignancy. Clin Endocrinol (Oxf) 2016; 85:710-716.
14. NIH state-of-the-science statement on management of the clinically inappar- ent adrenal mass (‘incidentaloma’). NIH Consens State Sci Statements 2002; 19:1-25.
15. Barzon L, Sonino N, Fallo F, et al. Prevalence and natural history of adrenal incidentalomas. Eur J Endocrinol 2003; 149:273-285.
16. Bancos I, Alahdab F, Crowley RK, et al. Therapy of endocrine disease:
improvement of cardiovascular risk factors after adrenalectomy in patients with adrenal tumors and subclinical Cushing’s syndrome: a systematic review and meta-analysis. Eur J Endocrinol 2016; 175:R283-R295.
The review summarizes low to moderate quality evidence on the cardiovascular benefits of adrenalectomy in patients with mild hypercortisolism.
17. Kotlowska A, Malinski E, Sworczak K, et al. The urinary steroid profile in patients diagnosed with adrenal incidentaloma. Clin Biochem 2009; 42:448-454.
18. Arlt W, Biehl M, Taylor AE, et al. Urine steroid metabolomics as a biomarker tool for detecting malignancy in adrenal tumors. J Clin Endocrinol Metab 2011; 96:3775-3784.
19. Kerkhofs TM, Kerstens MN, Kema IP, et al. Diagnostic value of urinary steroid profiling in the evaluation of adrenal tumors. Horm Cancer 2015; 6:168-175.
20. Velikanova LI, Shafigullina ZR, Lisitsin AA, et al. Different types of urinary steroid profiling obtained by high-performance liquid chromatography and gas chromatography-mass spectrometry in patients with adrenocortical car- cinoma. Horm Cancer 2016; 7:327-335.
21. Crowley RK, Hughes B, Gray J, et al. Longitudinal changes in glucocorticoid metabolism are associated with later development of adverse metabolic phenotype. Eur J Endocrinol 2014; 171:433-442.
22. Tomlinson JW, Finney J, Hughes BA, et al. Reduced glucocorticoid produc- tion rate, decreased 5alpha-reductase activity, and adipose tissue insulin sensitization after weight loss. Diabetes 2008; 57:1536-1543.
23. Finkelstein M, Shoenberger J. Urinary steroids in 10 cases of adrenal carcinoma, with special reference to pregnane-3 alpha, 17 alpha,20 alpha- triol-11-one. J Clin Endocrinol Metab 1959; 19:608-612.
24. Grondal S, Eriksson B, Hagenas L, et al. Steroid profile in urine: a useful tool in the diagnosis and follow up of adrenocortical carcinoma. Acta Endocrinol (Copenh) 1990; 122:656-663.
25. Di Dalmazi G, Fanelli F, Mezzullo M, et al. Steroid profiling by LC-MS/MS in nonsecreting and subclinical cortisol-secreting adrenocortical adenomas. Journal of Clin Endocrinol Metab 2015; 100:3529-3538.