A Diagnostic Approach to Adrenocortical Tumors
Check for updates
Anjelica Hodgson, MDa, Sara Pakbaz, MDb, Ozgur Mete, MDC,d, *
KEYWORDS
. Adrenal gland . Cushing syndrome . Primary aldosteronism · Virilism and feminization
· Adrenal cortical adenoma · Adrenal cortical carcinoma · Nodular cortical disease · IGF-2
Key points
· Confirmation of adrenocortical origin, distinction of benign from malignant lesions, providing prog- nostic information in adrenocortical carcinoma, and correlation of laboratory results with clinico- pathologic findings are among the critical responsibilities of pathologists who evaluate adrenocortical lesions.
. Several scoring schemes and algorithms have traditionally been used to distinguish adrenocortical ad- enomas from carcinomas.
· Immunohistochemical biomarkers and molecular diagnostics can distinguish adrenocortical carci- nomas from adrenocortical adenomas and can also provide important prognostic information.
· Advances in molecular biology have resulted in better understanding of the pathogenesis and molec- ular characteristics of adrenocortical tumors.
ABSTRACT
A drenocortical tumors range from primary bilateral micronodular or macronodular forms of adrenocortical disease to conven- tional adrenocortical adenomas and carcinomas. Accurate classification of these neoplasms is crit- ical given the varied pathogenesis, clinical behavior, and outcome of these different lesions. Confirmation of adrenocortical origin, diagnosing malignancy, providing relevant prognostic infor- mation in adrenocortical carcinoma, and correla- tion of laboratory results with clinicopathologic findings are among the important responsibilities of pathologists who evaluate these lesions. This article focuses on a practical approach to the eval- uation of adrenocortical tumors with an emphasis
on clinical and imaging findings, morphologic characteristics, and multifactorial diagnostic schemes and algorithms.
OVERVIEW
Adrenocortical tumors (ACTs), as the name im- plies, originate within the adrenal cortex. The cortex makes up the outer portion of the ad- renal gland and is composed of 3 histologic zones: zona glomerulosa (ZG), zona fasciculata (ZF), and zona reticularis (ZR). Each zone possesses specific histomorphologic and ultrastructural features as well as distinct func- tional capabilities related to hormone production.1
Disclosure: The authors have nothing to disclose.
a Department of Laboratory Medicine and Pathobiology, The University of Toronto, 1 King’s College Circle, Medical Sciences Building, Toronto, Ontario M5S1A8, Canada; b Department of Pathology, University Health Network, The University of Toronto, 200 Elizabeth Street, 11th Floor, Toronto, Ontario M5G2C4, Canada; Department of Pathology, University Health Network, 200 Elizabeth Street, 11th Floor, Toronto, Ontario M5G2C4, Canada; d Department of Laboratory Medicine and Pathobiology, The University of Toronto, Tor- onto, Ontario, Canada
* Corresponding author. E-mail address: ozgur.mete2@uhn.ca
https://doi.org/10.1016/j.path.2019.08.005
Because of the increasing use of diagnostic im- aging, lesions arising in the adrenal cortex, both symptomatic and asymptomatic, are being identified with growing frequency. Once an adre- nocortical nodule is identified, the diagnostic con- siderations that must be considered range from indolent and benign, such as adrenal cortical ade- noma (ACA), to often aggressive and malignant, such as adrenal cortical carcinoma (ACC). In addi- tion to these well-recognized conventional entities, advances in molecular biology and identification of clonality in the setting of bilateral nodular forms of adrenocortical hyperplasia have brought entities such as primary bilateral macronodular adreno- cortical disease (also known as primary bilateral macronodular adrenocortical hyperplasia [PBMAH]) and primary pigmented micronodular adrenocortical disease (PPNAD) into the spectrum of clonal lesions to be considered under the adre- nocortical neoplasia umbrella. Accurate classifica- tion of ACTs is critical given the varied pathogenesis, clinical behavior, and outcome of these lesions.
In this article, an approach to the evaluation of ACTs is discussed, with an emphasis on important clinical information along with macroscopic and microscopic features, diagnostic scoring schemes, molecular biology, and ancillary tests.
CLINICAL, BIOCHEMICAL, AND RADIOLOGICAL FEATURES OF ADRENOCORTICAL TUMORS
Adrenal nodules are common, with an estimated 10% of the population thought to harbor some sort of adrenal cortical lesion.2 These lesions span a proliferative spectrum on which adrenocor- tical hyperplasia (a reversible and genetically sta- ble process), clonal nodules of PBMAH and PPNAD (irreversible processes with distinct ge- netic signatures), ACA, and ACC must be consid- ered. A distinct functional adrenocortical lesion lacking features of malignancy is designated an ACA, whereas incidentally discovered nonfunc- tional adrenocortical nodules less than 1.0 cm are often designated as cortical nodular disease (Fig. 1). There is evidence to suggest that an adre- nocortical nodule as small as 3 to 5 mm may indeed be clonal3,4; therefore, the concept of cortical nodular disease also encompasses small adenomas.
ACAs are more common than ACCs; about 5% of the general population is estimated to harbor an ACA, whereas ACCs are seen in less than 2 per million people.2,5,6 With the exception of famil- ial cases, ACCs typically present in the fourth and fifth decades of life and more commonly in
women, whereas ACAs occur in all age groups and affect men and women equally. Given that primary adrenocortical lesions arise from hormone- producing cells, it is no surprise that these lesions may be hormonally active and cause clinical signs and symptoms related to hormone excess. Based on the presence or absence of hormonal activity, ACTs are classified as functional or nonfunctional,
respectively. Several hormones can be produced by functional ACTs, including mineralocorticoids (aldosterone), glucocorticoids (cortisol), and sex steroids (androgens). Table 1 summarizes the key clinical, biochemical, and clinicopathologic features of functioning ACTs.6-12 Importantly, pri- mary aldosteronism is extremely rare in ACCs, whereas virilization and sex hormone excess are
| Associated Endocrinopathies | |||
|---|---|---|---|
| Mineralocorticoid Excess (Primary Aldosteronism) | Glucocorticoid Excess (Adrenal Cushing Syndrome) | Sex Steroid Excess (Feminization and Virilization Syndrome) | |
| Clinical Features | Third-sixth decade; equal male/female ratio Hypertension with hypokalemia or normokalemia; nonspecific symptoms including muscle weakness, easy fatigability, headache, palpitations, nocturia, polyuria, and polydipsia | Any age; women affected more commonly than men Central obesity, facial rounding, hirsutism, easy bruising, skin striae, poor wound healing, muscle weakness, hypertension, hyperglycemia, osteoporosis in adulthood; weight gain and growth failure in childhood; subclinical Cushing syndrome | Variable clinical findings based on the patient's age, gender, and level of hormone excess; virilization in women (increased muscle mass and facial hair, deep voice and amenorrhea); in prepuberty age (pubic hair growth); feminization in men (gynecomastia and impotence) |
| Biochemical Aspects | Increased plasma aldosterone and high plasma aldosterone/renin ratio following an aldosterone suppression test Preoperative bilateral adrenal venous sampling for aldosterone/cortisol ratio or lateralization index with or without cosyntropin stimulation | Increased cortisol level in at least 2 endocrine assays (ie, baseline morning and evening measurements of 24-h urinary free cortisol, serum free cortisol, or late-night salivary free cortisol), autonomy confirmed by a dexamethasone suppression test. Plasma ACTH is also measured | Increased level of DHEA, DHEA-S, androstenedione, dihydrotestosterone, testosterone, estrogen, hydroxyprogesterone, or estradiol Neonatal screening and biochemical tests for glucocorticoid deficiency and salt wasting crisis are required when CAH is suspected |
| Clinicopathologic Aspects | Bilateral ZG hyperplasia (most common, around 60%) Unilateral ZG hyperplasia (2%) | ACTH-dependent ACH (80%) ACTH-independent ACH (2%) (eg, PBMAHª and PPNADb) ACA (10%) ACC (8%) | Classic and nonclassic CAH ACT (adults often manifest with ACC in the absence of congenital adrenal hyperplasia) |
| ACA (30%-40%) | |||
| ACC (1%) | |||
Abbreviations: ACH, adrenocortical hyperplasia; ACTH, adrenocorticotrophic hormone; CAH, congenital adrenocortical hyperplasia; DHEA, dehydroepiandrosterone; DHEA-S, dehydroepiandrosterone sulfate.
a Primary bilateral macronodular adrenocortical hyperplasia.
b Primary pigmented micronodular adrenocortical disease.
almost always related to malignancy in adults, in the absence of congenital adrenal hyperplasia.9-11 Expression of 2 or more hormones by an ACT is seen more frequently in ACCs, with the most com- mon synchronous combination being sex steroid and cortisol secretion.
Adrenal lesions are often discovered by imaging studies intended to evaluate for some other disease process. When discovered incidentally, these lesions are commonly referred to as incidentalomas.13 Once discovered, abdominal computed tomography (CT) is the general imaging modality of choice to evaluate adrenal cortical le- sions.14 The features that should be assessed include tumor size, appearance (ie, integrity and invasiveness), heterogeneity, lipid content, and the rate of washout after intravenous administra- tion of contrast. In neonates with congenital adre- nal hyperplasia, adrenal ultrasonography is useful.15 Some studies have suggested that chemical-shift MRI may be the preferred imaging modality in young patients or in cases in which a patient with an iodinated contrast allergy is found to have an indeterminate adrenal mass on unen- hanced CT.14,15
Radiological findings of bilateral nodular adre- nocortical disease are variable. In PBMAH, massive bilateral enlargement is a characteristic finding.16 Unlike PBMAH, CT and MRI findings can sometimes underestimate the extent of, and possibly miss, primary bilateral micronodular adrenocortical disease.17 In these scenarios, the use of functional imaging studies (eg, noriodocho- lesterol scintigraphy or PET-CT scans) can be helpful in identifying the bilateral nature of the
disease; functional imaging studies may be of particular value especially in the setting of normal CT scans with abnormal biochemical findings. 17 From a biochemical perspective, patients with PPNAD often show a characteristic paradoxic response to a dexamethasone suppression (Liddle test) with an increase in 17-hydroxycorticosteroid and urinary free cortisol. 18
In the case of biochemically active ACTs, imag- ing studies may be used to locate the hypersecret- ing tumor, although, in some scenarios, locating the lesion is sometimes difficult, especially in some cases of primary aldosteronism. In these cases, adrenal vein sampling (AVS) to assess aldo- sterone/cortisol ratios in both adrenal veins is considered the gold standard in order to lateralize the side of a hyperfunctioning tumor6,9; however, this invasive technique often yields borderline re- sults. In addition, in cases in which the ACT cose- cretes cortisol and aldosterone, the characteristic hormone gradient may be lost. Recent evidence has shown that the use of molecular adrenal imag- ing studies (eg, metomidate PET-CT) along with AVS can be helpful in identifying the source of pri- mary aldosteronism.19,20 In patients with hyper- cortisolism, the diagnosis of adrenal Cushing syndrome is considered only after exclusion of adrenocorticotrophic hormone (ACTH) -depen- dent disease (eg, pituitary or ectopic source of ACTH). Because of this, brain MRI to look for a pi- tuitary tumor, bilateral inferior petrosal sinus sam- pling in nonvisible pituitary lesions, and a chest and abdominal/pelvic CT scan may be required. 14
The common radiological features of ACTs are summarized in Table 2. Note that indeterminate
| Radiological Characteristics | Benign | Malignant |
|---|---|---|
| Tumor size | Often <4-6 cm | Often ≥4-6 cm |
| Tumor appearance | Well delineated; round with regular border | Ill-defined, irregular border; areas of necrosis |
| Tumor heterogeneity | Homogeneous enhancement | Heterogeneous enhancement |
| Lipid content and density appearance on unenhanced CT | Lipid rich Low density on unenhanced CT (≤10 Hounsfield units) | Lipid poor High density on unenhanced CT (>10 Hounsfield units) |
| Contrast enhancement and washout pattern 15 min after intravenous administration of contrastª | Enhanced with rapid washout (≥40%) after 15 min | Enhanced with less washout (15%-25%) after 15 min |
| Chemical-shift MRI | Loss of signal intensity on the out-of-phase image | No loss of signal intensity on the out-of-phase image |
a Hypervascular metastatic tumors such as renal cell carcinoma and hepatocellular carcinoma can have rapid washout mimicking an adrenocortical adenoma.
radiological adrenocortical lesions are commonly reported,14,21,22 although detailed descriptions of those findings are beyond the scope of this article.
MORPHOLOGIC DIAGNOSIS OF ADRENOCORTICAL TUMORS
GROSS EXAMINATION
Complete morphologic assessment of ACTs be- gins with a review of the available clinical, biochemical, and radiological information, fol- lowed by a thorough gross examination of the adrenalectomy resection specimen. It is important to note what type of surgery has been done, because some specimens may be more frag- mented than others.23 An increasing number of surgical centers are performing morcellation pro- cedures at the time of laparoscopic adrenalec- tomy when removing small tumors considered to be ACAs based on preoperative imaging. From a pathologic perspective, this is obviously not ideal. However, care should be taken to evaluate the same parameters as in an intact resection spec- imen whenever possible.24
When evaluating an adrenalectomy specimen, the usual principles of macroscopic examination apply, including specimen painting, margin identi- fication, specimen measurement, and specimen weighing. When looking at the cut surface of the specimen, the tumor as well as the nontumor adre- nal parenchyma must be carefully evaluated. In adults, the normal adrenal cortex has an average thickness of at least 2 mm.10 In patients with a cortisol-secreting lesion, the cortex becomes thin (also known as atrophy of the cortex) because of negative feedback inhibition by the autonomous cortisol secretion on the hypothalamic-pituitary- adrenal axis.10,25 Failure to recognize this finding may have critical consequences because this in- formation can be lifesaving if hormone replace- ment therapy was not provided to the patient postoperatively.25 Among patients with subclinical Cushing syndrome, some may be undetected pre- operatively. Any additional nodularity, their mea- surements, and the presence or absence of associated pigmentation within the cortex besides the main lesion should also be commented on and sampled adequately for microscopic examination. The appearance and distribution of the medulla should also be described. If feasible, photographs of both the intact and serial sections of the spec- imen should be taken; rarely, having gross images to review can be critical in the assessment of a resection specimen.
The tumor should be accurately measured and its relationship to the capsule should be
scrutinized with any gross disruption or invasion into periadrenal tissue noted. Ideally, periadrenal adipose tissue should be dissected away for accu- rate weighing of the tumor although sometimes, this is not always possible. Both benign and malig- nant ACTs frequently present as solitary masses, although multifocal ACTs can also occur. ACAs are more commonly 5 cm or less, whereas ACCs are often significantly larger and often weigh more than 100 g,26 although exceptions have been reported. ACCs may show a vague multinod- ular appearance reflective of the heterogeneity often seen microscopically, which represents a potential pitfall in the examination of these lesions, because undersampling may miss areas of the tu- mor that are dictating biological behavior.
Most commonly, ACAs are well delineated and homogeneous, but they infrequently show hemor- rhage and cystic degeneration (Fig. 2). Most are yellow-gold (frequently in aldosterone-producing ACAs), reflecting their lipid-rich nature (see Fig. 2A), whereas oncocytic ACTs or ACTs with oncocytic change (see Fig. 2B) have a distinct red-brown appearance. So-called black ade- nomas have also been described (see Fig. 2C); these lesions get their distinctive color from lipo- fuscin accumulation. The micronodules (<1 cm) of PPNAD and some forms of primary micronodu- lar adrenocortical disease may also show a vari- able degree of pigmentation.
In contrast, ACCs are more often heterogeneous and may show fibrous bands. In addition, hemor- rhage, necrosis, and calcification are commonly identified. They may be well circumscribed or show an irregular border and infiltration into adja- cent structures. When locally invasive, ACCs tend to invade regional venous structures (Fig. 3), adipose tissue surrounding the adrenal gland, and adjacent organs such as the kidney.27
In addition to sampling the tumor proper for microscopic examination, it is critical to sample the tumor interface with adjacent tissue, because the authors have found this to be the area where invasive growth and angioinvasion are most evident. As for every resection specimen, lymph nodes should be entirely submitted when present.
MICROSCOPIC EXAMINATION
Traditionally, ACTs are thought to arise from the different zones of the adrenal cortex and, as such, they most often recapitulate the cellular morphology characteristic of different adrenocor- tical cells. For example, it is common for ACAs to be composed of corded/nested lipid-rich cells with abundant vacuolated clear cytoplasm and low nuclear/cytoplasmic ratio reminiscent of the
A
C
B
ZF, in addition to more compact or eosinophilic cells reminiscent of the ZR layer.
Depending on the hormonal functionality of the ACT being evaluated, genotypic-phenotypic fea- tures of the tumor and specific changes within the nontumor cortex may be apparent. For example, ACAs producing aldosterone are likely to be the most heterogeneous group among ACAs. Depending on their underlying molecular biology, they can show combinations of different cell types, including lipid-rich (ZF-like) cells, lipid- poor compact (ZR-like) cells, smaller cells
resembling ZG-like cells, and compact cells with overlapping features of ZG-like and ZR-like cells (Figs. 4 and 5). The native ZG is commonly hyper- plastic, so-called paradoxical ZG layer hyperplasia (Fig. 6), and may show micronodular proliferations that are now recognized as aldosterone-producing cell clusters (APCCs) (Fig. 7), a new concept in primary aldosteronism.28 When treated with spironolactone, an antagonist of aldosterone, aldosterone-producing ACAs, in addition to the background parenchyma, show characteristic eosinophilic concentric lamellated intracellular
Liver
ACC
inclusions called spironolactone bodies (see Figs. 5 and 7; Fig. 8). It has been reported that spirono- lactone bodies are not seen when eplerenone is administered (aldosterone receptor antagonist).29
In adrenal Cushing syndrome, there is a spec- trum of cortisol-producing proliferations from bilateral micronodules (<1 cm) and macronodules (>1 cm) with their distinct molecular alterations
representing PPNAD and PBMAH, respec- tively.4,28,30,31 Conceptually, the bilateral micro- nodular or macronodular disease can be regarded as multifocal ACAs, given their clonal nature. ACCs have been described in the context of PPNAD.32,33 Morphologically, PPNAD is composed of bilateral cortical micronodules composed of pigmented compact cell-rich prolif- erations with variable degrees of cortical atrophy in the intervening cortex. Sometimes, isolated nonpigmented or weakly/variably pigmented micronodular proliferations can also occur in the setting of Cushing syndrome. Tumor nodules iden- tified in PBMAH are composed of cortical
proliferations composed mainly of lipid-rich cells and variable amounts of compact cells that repre- sent multiple nodules mostly exceeding 1 cm and irregular enlargement of both adrenal glands.28 Conventional cortisol-secreting ACAs show various morphologic features ranging from pure clear cell tumors to mixed clear and compact cell ACTs. The autonomous neoplastic cortisol secretion from the adrenal gland shuts down the hypothalamic corticotropin-releasing hormone, resulting in resting corticotrophs via Crooke hya- line change in addition to a lack of ACTH- mediated trophic changes in both adrenal glands. Consistently, the background adrenal
cortex and/or internodular adrenal cortex become atrophic because of the absence of the ZR layer along with a thinned ZF layer (Fig. 9). Sex hormone-secreting ACTs, in contrast with ACTs secreting other hormones, usually tend to be enriched in compact eosinophilic cells normally seen in the ZF. Nonfunctional ACAs can also show morphologic heterogeneity.
Several morphologic changes can be observed in both ACAs and ACCs, including oncocytic and myxoid changes as well as myelolipomatous change. Tumors are designated as mixed onco- cytic ACTs when the cortical neoplasm shows 50% to 90% oncocytic change, whereas the term pure oncocytic ACT is applied to those tu- mors that show greater than 90% oncocytic
60 um
change.34,35 Some rare changes, especially myx- oid changes, are more common in ACCs. Sarco- matoid areas can be seen in ACCs and, when identified, can cause confusion with other malig- nant tumors in the differential diagnosis, particu- larly primary or secondary (metastatic) sarcomas. Given the range of morphologic heterogeneity, 4 histologic variants of ACC are currently recog- nized: conventional (Fig. 10), oncocytic (Fig. 11), myxoid (Fig. 12), and sarcomatoid.27 ACCs can also show a combination of these variants (Fig. 13).
To aid in establishing a diagnosis of ACC, several multifactorial scoring schemes/algorithms have been described that evaluate several
features that have been associated with poor outcome (Fig. 14). In 1984, Weiss36,37 proposed the first diagnostic scheme, colloquially known as the Weiss criteria. The original Weiss criteria focused on the assessment of 9 features: high nu- clear grade (Fuhrman grading system), mitotic rate more than 5 mitoses per 50 high-power fields (Fig. 15), atypical mitotic figures, less than 25% clear cells, diffuse architecture (defined as pattern- less sheets exceeding 30% of the tumor; nested/ alveolar, columnar, trabecular, or cordlike areas are defined as nondiffuse growth), tumor necrosis (Fig. 16), venous invasion (Fig. 17), sinusoidal in- vasion, and capsular invasion. Of the 43 cases in the original Weiss series, none of the 24 tumors
4 mm
meeting 2 or fewer criteria metastasized or recurred, whereas all but 1 of the remaining 19 cases meeting 4 or more criteria either metasta- sized or recurred.36 The original Weiss criteria have since been modified because of the lack of reproducibility and interpretive difficulties. The modified Weiss criteria (by Aubert and col- leagues26) were thought to be more reproducible and easier to apply because they called for the assessment of 5 instead of 9 criteria.
Because the Weiss criteria include some fea- tures of malignancy that are morphologically inherent to pure oncocytic ACTs (eg, diffuse growth pattern, scarcity of clear cells, and promi- nent nucleoli leading to high-grade nuclear
scoring), the Lin-Weiss-Bisceglia criteria38 were described with an increased focus on invasiveness and mitotic count and less focus on some other morphologic features defined in the Weiss criteria. The Lin-Weiss-Bisceglia criteria38 use a major and minor criteria framework to classify oncocytic ACTs for which the presence of 1 major criteria (mitotic rate >5 mitoses per 50 high-power fields, atypical mitotic figures, venous invasion) indicates malignancy (Fig. 18) and the presence of 1 to 4 minor criteria (tumor size >10 cm and/or weight >200 g, necrosis, sinusoidal invasion, capsular invasion) indicates a tumor of uncertain malignant potential. Oncocytic ACAs should not show any of the major or minor criteria.
Modified Weiss Scoring
Wieneke criteria
Lin-Weiss-Bisceglia criteria
Reticulin Algorithm
Pediatric
Oncocytic
Conventional Oncocytic Myxoid
Weiss Scoring Conventional
Tumor weight >400 g Tumor size >10.5 cm Extra-adrenal extension Venous invasion Capsular invasion Tumor necrosis Mitotic activity >15 per 20 HPF Atypical mitosis Score >3: Malignant
Major criteria (1 major: malignant) Vascular (venous) invasion Mitotic activity >5 per 50 HPF Atypical mitosis
Altered reticulin framework with any of the following parameters Venous invasion Mitotic activity >5 per 50 HPF Necrosis
Weiss Scoring (Score ≥3 out of 9: Malignant) High Fuhrman Nuclear Grade (FNG III or IV) Mitosis >5 per 50 HPF (High Power Fields)
Atypical mitosis Clear cells ≤25%
Minor criteria (1-4 minor: UMP) >10 cm and/or >200 g Necrosis Capsular invasion Sinusoidal invasion
Diffuse architecture >30%
Helsinki Scoring
Necrosis
Venous invasion
Sinusoidal invasion
Conventional Oncocytic Myxoid
Capsular invasion
Modified Weiss (Score ≥3 out of 7: Malignant)
Mitotic rate (>5 per 50HPF): 2
Clear cells ≤25%:2
5 x Necrosis 3x mitotic activity >5/50 HPF numeric value of Ki67% Score >8.5: Malignant
Atypical mitosis: 1
Necrosis: 1
Capsular invasion: 1
Fig. 14. Multifactorial diagnostic schemes to aid in making the diagnosis of adrenocortical carcinoma. Several multifactorial scoring schemes/algorithms have been described to evaluate several features that have been asso- ciated with poor outcome in adrenocortical carcinomas. The Weiss and modified Weiss scoring schemes have been used to distinguish conventional adrenocortical carcinomas identified in adults. Pediatric adrenocortical tu- mors are typically assessed using the Wieneke multifactorial system. Oncocytic adrenocortical tumors are assessed using the Lin-Weiss-Bisceglia scoring scheme. The reticulin algorithm can be applied to conventional, oncocytic, and myxoid adrenocortical tumors identified in adults. The Helsinki scoring scheme can distinguish malignancy in conventional adrenocortical tumors in adults. This approach also provides prognostic information. Although the data on myxoid adrenocortical tumors are limited, recent evidence suggests that the diagnostic performance of the Helsinki scoring system in oncocytic adrenocortical tumors was not as good as the Lin-Weiss-Bisceglia system. However, the Helsinki score was able to predict poor prognostic subgroups of oncocytic adrenocortical carci- nomas. Generated from Dr. Ozgur Mete’s USCAP 2019 Endocrine Pathology Society Companion Meeting Lecture on Challenges in Adrenal Cortical Pathology. HPF, high-power field; UMP, uncertain malignant potential.
The first algorithm to consider the use of a histo- chemical stain as a requirement in the morpho- logic evaluation of ACTs is the reticulin algorithm. This algorithm is gaining significant popularity among diagnosticians because of the objective nature of evaluating the tumor reticulin network combined with the limited number of features needed to define malignancy.39-41 ACAs typically show preserved reticulin framework (Fig. 19). When an altered reticulin network in an ACT is identified (Fig. 20) in combination with 1 or more parameters (vascular invasion, tumor necrosis, or mitotic activity >5 per 50 high-power fields), the diagnosis of ACC can be made.39-41 In addition to conventional forms of ACCs, recent series
have also underscored the usefulness of the retic- ulin algorithm in oncocytic34,35,40-42 and myxoid ACTs.40 Because the distinction of myxoid ACCs can be challenging using the Weiss parameters, 43 the reticulin algorithm may add value to the diag- nostic work-up. Of note, a multicentric validation series of 245 ACTs also included 2 cases of pedi- atric ACT.40 Although the interpretation of the reticulin histochemistry is often an easy task for di- agnosticians, they should be aware of the common pitfalls related to its interpretation (dis- cussed later).
Ki67 labeling index has been shown to be prog- nostically significant in ACCs (discussed later). 44 The Helsinki scoring system, proposed in 201545
and subsequently validated in a later series, 46 has incorporated the role of the Ki67 labeling index into its criteria for malignancy along with mitotic rate (>5 per 50 high-power fields) and presence/ absence of necrosis. The Helsinki system uses a weighted-point system: 5 points are awarded if necrosis is present and 3 points are awarded if more than 5 mitoses are seen in 50 high-power fields. The value of the Ki67 index (percentage positive tumor nuclei from hot spots) is used as the third scoring component. A score greater than 8.5 indicates a malignant lesion. Of note, the Ki67 index should be evaluated in the area of highest proliferative activity; no visual assessment is allowed, and, ideally, the evaluation should be done by an automated image analysis software nuclear algorithm. The Italian validation series included oncocytic (mixed and pure tumors) and myxoid ACCs in addition to conventional ACCs.46 In this validation cohort, oncocytic ACCs (diagnosed based on the Lin-Weiss- Bisceglia criteria) had a Helsinki score ranging from 3 to 79.46 Although there were only a few myxoid tumors from which to draw a reliable conclusion, the Helsinki score of 19 or greater captured a significant proportion of aggressive forms of oncocytic ACCs. 46 In a recent multicenter French cohort, the diagnostic performance of the Helsinki scoring scheme in oncocytic ACTs was not as good as other schemes, but the Helsinki score was more useful in the prediction of poor prognostic subgroups of oncocytic ACCs.35 These findings have pointed out the diagnostic limitations of this scoring system; however, further studies are still needed to expand the diagnostic limita- tions of this scheme.
Although rare, 47 pediatric ACTs often represent a diagnostic conundrum because it has been shown that some histologic features typically associated with malignancy in adult adrenocortical neoplasms do not necessarily correlate with poor outcome in pediatric patients. Although this may partially be related to differences in application of some cardinal features of malignancy among diag- nosticians, the Wieneke system48 was devised to better delineate benign from malignant ACTs in the pediatric setting. The Wieneke system incor- porates some traditional prognostic variables (atypical mitoses, vascular invasion, capsular inva- sion, tumor necrosis, mitotic rate >15 per 20 high- power fields) into its scoring algorithm, in addition to gland size and weight as well as tumor involve- ment of local structures, including vena cava and other extra-adrenal structures. When 4 parame- ters are met, a diagnosis of pediatric ACC can be made. An assignment of uncertain malignant po- tential can be made with a score of 3, whereas a score of 1 or 2 is thought to suggest a benign course. This approach to pediatric ACTs has also been evaluated in subsequent series. 49-51
It has been the observation of our team mem- bers that areas suggestive of adenoma-to- carcinoma tumor progression exist in some ACTs (O.M., author observation). Moreover, ACA-like low-proliferative regions can be seen admixed with nodular areas with high-grade proliferation. These lesions can be particularly challenging for diagnosticians, especially when limited represen- tative sampling was performed from the tumor. Even so, there are very rare ACTs that are in a diagnostic gray zone and, in those cases, some di- agnosticians use the term atypical or borderline
adrenocortical neoplasm, whereas others apply the term ACT of uncertain malignant potential (UMP) for similar presentations. When the rare diagnosis of an ACT-UMP is being considered, additional sections should be submitted of the entire periphery of the tumor along with the central portion, and a diligent search including serial/ deeper sections should be conducted to evaluate for features of malignancy. The diagnostic work- up of an ACT is no longer restricted to conven- tional histomorphology. There are several immu- nohistochemical biomarkers that can be useful in establishing a diagnosis of malignancy.41,52 Simi- larly, the diagnosis and prognostication of ACC is
now possible when applying various molecular biology techniques in the evaluation of an ACT53 (discussed later).
Once a diagnosis of malignancy has been established, ACCs are further categorized as high grade or low grade based on mitotic counts from hot spots, depending on whether up to 20 (low grade) or more than 20 mitoses (high grade) are seen per 50 high-power fields. This approach stems from the original article by Weiss and col- leagues, 37 but this has been adopted in the past decade in several practices given its prognostic significance.52,54-56 Volante and colleagues39 also showed that ACCs can be further
prognosticated combining tumor stage and mitotic activity (≤9 per 50 high-power fields vs >9 per 50 high-power fields). Mete and colleagues also showed that the cutoff of 10 mitoses per 50 high-power fields had a better performance in the correlation of disease-free survival in ACCs.41
In practice, most ACCs are easily separated from ACAs, especially when they are widely inva- sive and highly proliferative. Overall, it has been re- ported that vascular invasion (see Fig. 17), defined as tumor cells invading through a vessel wall and/ or intravascular tumor cells admixed with thrombus, is the strongest diagnostic parameter and predictive of poor outcome,41 and thus it is reasonable to suggest that every case should be thoroughly scrutinized to rule this finding in or out, even in cases that are obviously malignant.
HISTOCHEMISTRY IN THE DIAGNOSIS OF ADRENOCORTICAL TUMORS
The use of histochemistry in benign ACTs is limited. In the setting of primary aldosteronism, Luxol fast blue (see Fig. 8) can be used to identify spironolactone bodies.º Histochemistry has gained popularity with the introduction of the retic- ulin algorithm because the algorithm is now regarded as a simple and reproducible method that can be used to separate ACAs from ACCs (see Figs. 19 and 20); however, clinicians should recognize pitfalls and pearls in the interpretation of reticulin findings. Although validation of the reticulin algorithm in large series of pediatric ACCs is currently lacking, this algorithmic approach has been shown to aid in making the diagnosis of conventional as well as onco- cytic34,40,41 and myxoid variants.40 The backbone of the reticulin algorithm is the reticulin histochemi- cal stain (most commonly Gordon-Sweet-Silver). Assessment of this staining in ACTs evaluates both quantitative (loss of continuity in reticulin fi- bers) and qualitative (abnormal reticulin network with irregular thickness of fibers, frayed appear- ance, or pericellular pattern leading to a meshlike appearance) changes.40,41 Qualitative alterations in some ACTs may be the source of underestima- tion of malignancy because clinicians may encounter ACCs rich in areas of meshlike pericel- lular reticulin staining that can be mistaken for an unaltered reticulin framework. By being aware of this potential pitfall, additional careful examination of the reticulin histochemistry can solve this quan- dary because ACCs tend to show variable degrees of reticulin disruption even in the presence of qual- itative changes.41 Although the quantitative alter- ations are most helpful,41 ACAs with areas of
degeneration (eg, hemorrhage, extensive post- biopsy changes) can also show variable disruption of the reticulin framework. Furthermore, examples of very focal loss of reticulin staining and foci of incomplete or complete pericellular pattern have also been described in some ACAs.41 Therefore, altered reticulin framework should be taken into consideration along with other histologic findings as well as biomarker profiling when evaluating an ACT.
IMMUNOHISTOCHEMISTRY IN THE DIAGNOSIS OF ADRENOCORTICAL TUMORS
Immunohistochemistry has become an integral component of the diagnostic work-up of ACTs. Biomarkers are routinely used to address various clinical needs, including confirmation of the adre- nocortical origin, distinction of functional tumors, supporting the diagnosis of malignancy, providing prognostic and theranostic information in ACCs, and facilitating the screening process for germline pathogenesis.52 Relevant immunohistochemical biomarkers are summarized in Box 1.
Several locoregional (eg, renal cell carcinoma, pheochromocytoma, epithelioid perivascular epithelioid cell neoplasms, sarcoma) and metasta- tic neoplasms (eg, melanoma, hepatocellular car- cinoma) can simulate ACTs. Therefore, the confirmation of cortical origin should be consid- ered in all ACTs, especially in the absence of adre- nocortical hormone excess. Failure to confirm adrenocortical origin resulting in misdiagnoses has clearly been shown as being one of the major lessons from consultations practices.57 To confirm adrenocortical origin, steroidogenic factor-1 (SF- 1) (Fig. 21), a transcription factor characteristic of steroidogenic tissues, is the most specific diag- nostic biomarker.52 Negativity for SF-1 in sarco- matoid components of ACCs58 as well as in ACTs with suboptimal tissue fixation are important pitfalls in the interpretation of SF-1 staining.
Once adrenocortical origin has been estab- lished, several biomarkers can be used to aid in distinguishing ACA from ACC, especially with challenging surgical specimens and particularly in the evaluation of core biopsy material. Mete and colleagues41 reported that juxtanuclear insu- linlike growth factor 2 (IGF-2) (Fig. 22) staining optimized at 1:3000 to 1:6000 dilutions was the most useful diagnostic biomarker of adult ACCs because this pattern of staining was absent in ACAs. The juxtanuclear Golgi pattern is thought to reflect impairment in translation and processing of the IGF-2 molecule in the Golgi apparatus, which results in IGF-2 overexpression. 41,59
Box 1
Summary of practical immunohistochemistry in the assessment of adrenocortical tumors
Adrenal cortical confirmatory biomarkers Steroidogenic factor-1 (most specific), Melan-A, calretinin, alpha-inhibin, synaptophysin Functionality-related biomarkers
Cytochrome P (CYP) 11B2, CYP11B1, HSD3B1 and HSD3B2 (3ß-hydroxysteroid dehydrogenase type 1 and type 2)
Pathogenic biomarkers
Insulinlike growth factor 2 (juxtanuclear staining), p53 (overexpression or global loss), beta-catenin (diffuse nuclear pattern)
Proliferation-related biomarkers Ki67 (MIB1 antibody), phosphohistone-H3
Prognostic biomarkers
Ki67, p53, beta-catenin (diffuse nuclear pattern), DAXX, ATRX
Germline susceptibility screening biomarkers
Menin (multiple endocrine neoplasia type 1 [MEN1] syndrome), p27 (MEN4 syndrome), p53 (Li-Fraumeni syndrome), MMR (mismatch repair) proteins (Lynch syndrome), beta-catenin and APC (familial adeno- matous polyposis [FAP] syndrome), succinate dehydrogenase subunit B (SDHB; familial paraganglioma syndrome caused by SDHx)
Abnormal p53 immunoexpression is a well- known marker of malignancy in several organs. Although pediatric ACCs tend to show more frequent TP53 alterations, 60-63 overexpression (more frequent) (Fig. 23) or global loss as a result of TP53 gene mutation has been seen in about 20% to 25% of adult ACCs.60,62 Aberrant expres- sion has been associated with high-grade prolifer- ative features and tumor aggressiveness.62,63
Diffuse nuclear and cytoplasmic beta-catenin expression (Fig. 24) reflecting the activation of the Wnt pathway can be a feature of ACCs. 41 This finding alone should not warrant a diagnosis of malignancy because ACAs can also harbor CTNNB1 mutations and show activation of the Wnt pathway.28 In addition, patients with familial
adenomatous polyposis (FAP) can also manifest with ACTs, including ACAs. In contrast, there is a general consensus that ACCs with diffuse cyto- plasmic and nuclear beta-catenin reactivity are more frequently associated with a poor prognosis. 63-65
As proliferation-driven neoplasms, an accurate assessment of cell proliferation biomarkers in ACTs is a crucial clinical task for pathologists assessing these tumors. Several biomarkers have been used to do this, such as Ki67, phosphohistone-H3 (PHH3), p53, BUB1B, HURP, and NEK2.41,52 PHH3 (Fig. 25) can be considered to distinguish mitotic figures from apoptotic or cells with crush artifact; thus, it can assist mitotic count66 to enable accurate mitotic tumor grading.
Ki67 is one of the most important biomarkers and is readily available in almost all pathology labora- tories. Most ACCs show a Ki67 proliferation exceeding 5%41 (Fig. 26). The MIB1 antibody (anti-human Ki67 monoclonal antibody) is consid- ered the gold standard for this assessment.63 The management of patients with ACCs requires the knowledge of an accurate Ki67 labeling index along with other tumor characteristics.67 Several pitfalls and pearls exist in the assessment of the Ki67 proliferation index. Because ACCs are well known to display intratumoral proliferative hetero- geneity (see Fig. 26), the first step in obtaining an accurate and meaningful Ki67 labeling index is the selection of the right tumor block based on high mitotic density seen in hematoxylin-eosin- stained sections. Tumor blocks with poor tissue fixation can result in impaired detection of nuclear antigen. If there is any concern regarding tissue fixation, multiple blocks should be assessed. Although the antibody and staining methods used can vary from one laboratory to another, vi- sual assessment is no longer an acceptable option for the analysis of the Ki67 in ACCs.68,69 Manual counting or automated image analysis nuclear al- gorithms from hot spots of nuclear labeling (prefer- ably 1000-2000 tumor cells) should be assessed in ACCs (Fig. 27). Some studies have classified ACCs into 3 groups: low-risk (grade 1), intermediate-risk (grade 2), and high-risk (grade 3) categories based on the Ki-67 index with different cutoffs (<10%, 10%-19%, ≥20%; or <20%, 20%-50%, >50%).66,70 Overall, Ki67 label- ing index is considered an important factor in
prognosticating ACCs66,70,71 as well as in deter- mining the need for adjuvant therapies.52,67 A recent French series of pediatric ACCs also high- lighted the impact of Ki67 because pediatric ACCs with poor prognosis had 2 of the following parameters: Ki67 greater than 15%, mitotic activ- ity greater than 15 per 20 high-power fields, vascular invasion, tumor necrosis, and adrenal capsule invasion. 49
Regarding other prognostic biomarkers of ACCs, ATRX and DAXX (proteins that regulate telomere elongation) have been previously stud- ied.41,72 Global loss of DAXX and ATRX are commonly seen in ACCs and loss of DAXX is more frequent in disease-free patients.41 Some biomarkers associated with DNA damage repair (eg, PBK) as well as the phosphatidylinositol 3 ki- nase (PI3K) signaling pathway (eg, PTEN and phospho-mTOR [mammalian target of rapamycin]) have also been investigated for diagnostic and prognostic purposes in ACCs; other biomarkers have been reported to predict response to mito- tane or other chemotherapy regimens; however, further studies are needed to expand the use of these biomarkers in routine practice.73
Antibodies against key enzymes involved in the steroidogenic pathway can be used to assess the functionality of ACTs28,74 (Fig. 28). Among these, monoclonal antibodies against HSD3B1/2 (3ß-hydroxysteroid dehydrogenase type 1 and type 2) and CYP11B1/2 (cytochrome P450 family 11 subfamily B member 1 and mem- ber 2) have shown promise. CYP11B1 is typically expressed in the ZF layer, whereas CYP11B2
(aldosterone synthase) expression is exclusive to the ZG layer. In primary aldosteronism, the distinction of functional sites can be challenging and the use of CYP11B2 can assist in the functional assessment.6,28,52,74 ZG hyperplasia and sites of APCCs can be located using CYP11B2, especially in the absence of an adre- nal mass.75
In general, family history, early onset, bilateral- ity, and multifocality are features that suggest germline susceptibility, and exclusion of an un- derlying disorder is required. Most ACTs
occurring in adults are sporadic, whereas pediat- ric tumors are more frequently associated with germline disease.6,52 Menin, p27, p53, beta- catenin, MMR (mismatch repair) proteins, and succinate dehydrogenase subunit B (SDHB) immunohistochemistry can assist screening of adrenal manifestations of multiple endocrine neoplasia type 1 (MEN1),6,52,76 MEN4,6,52 Li-Frau- meni,61,77 FAP,78 Lynch,79 and familial paragan- glioma syndromes.80 Other syndromes, including Beckwith-Wiedemann syndrome (IGF-2, H19 at the 11p15 locus), Carney complex (PRKAR1A),
3
Summary
!!!
8
-
Layer Attributes
Percent Positive Nuclei
48.363
Intensity Score
3
[3+] Percent Nuclei
9 3825
[2+] Percent Nuclei
11.1894
[1+) Percent Nuclei
7.79113
F
[0+] Percent Nuclei
51.637
Average Positive Intensity
156.918
Average Negative Intensity
236.422
2
[3+) Nuclei
709
[2+] Nuclei
270
[1+] Nuclei
188
[[+] Nuclei
1246
Total Nuclei
2413
Average Nuclear RGB Intensity
148.094
Average Nuclear Size [Pboels)
214.309
Average Nuclear Size [um”2)
53.513
Area of Analysis (Pixels)
1923458.
Area of Analysis (mm”2)
0.48028763571122002
=** Algorithm Inputs ****
== Algorithm Inputs ***
Algorithm
Nuclear v9
Version
9.1
·
Layer Regions
+-XO.
Region
Length [um) Area (um2) |Text
Percent Positive Nuclei
Intensity Sco
1
1394
124508
47.1591
3
1
2
3014
354984
48.859
3
.
+
Zona Glomerulosa
Zona Fasciculata
Zona Reticularis
Cholesterol
Cholesterol
Cholesterol
CYP11A1 and StAR
CYP11A1 and StAR
CYP11A1 and StAR
Pregnenolone
Pregnenolone
Pregnenolone
CYP17
CYP17
17 OH-Pregnenolone
17 OH-Pregnenolone
3ß-HSD
3B-HSD
17, 21 lyase
SULT2A1 sulfotransferase
Progesterone
17 OH-Progesterone
DHEA
DHEAS
CYP21A2 (21-hydroxylase)
CYP21A2 (21-hydroxylase)
11-Deoxycorticosterone
11-deoxycortisol
3ß-HSD
17ß-HSD
Androstenediole
CYP11B2
CYP11B1
3B-HSD
Corticosterone
Cortisol
Androstenedione
17ß-HSD
Testosterone
CYP11B2
Aromatase
Aromatase
18 OH-Corticosterone
Estrone
17ß-HSD
Estradiol
CYP11B2
Aldosterone
and neurofibromatosis type 1 (NF1), may be asso- ciated with ACCs.32,81,82
ULTRASTRUCTURAL EXAMINATION IN THE DIAGNOSIS OF ADRENOCORTICAL TUMORS
Ultrastructural examination has limited value in the modern surgical pathology of the adrenal cortex. Despite its limited utility, ultrastructural examina- tion may be useful in identifying aldosterone- producing proliferations that show platelike or lamellar mitochondrial cristae, in contrast with nonfunctional and cortisol-producing cells that show tubulovesicular cristae.6
MOLECULAR BIOLOGY IN ADRENOCORTICAL NEOPLASMS
The past decade has seen significant progress in the understanding of adrenocortical tumorigenesis as well as the cellular mechanisms implicated in primary aldosteronism and adrenal Cushing syn- drome. Most aldosterone-producing ACAs and APCCs are associated with increased
transcription of CYP11B2 because of molecular al- terations leading to aberrant activation of the cal- cium/calmodulin kinase pathway.28,83,84 Among these, KCNJ5 mutations are the most common al- terations, followed by ATP1A1, ATP2B3, and CACNA1D mutations.28,85-89 Unlike primary aldo- steronism, most cortisol-producing ACAs harbor molecular alterations (most common PRKACA mutations, followed by PRKAR1A, GNAS, PDE8B, PDE11A, PRKACB, and MC2R mutations) implicated in the cyclic AMP (cAMP)/protein ki- nase A (PKA) pathway.10,28,90,91 As seen in the example of Carney complex-related PPNAD, mul- tiple microscopic ACAs identified in bilateral pri- mary micronodular adrenocortical disease arise in the background of germline mutations (most commonly inactivating PRKAR1A mutations) that result in aberrant signaling of the cAMP/PKA pathway.10,28,92,93 In contrast, around 55% of PBMAH manifests with germline ARMC5 muta- tions.94 Alterations in MEN1, APC, FH, PDE8B, and PDE11A have also been reported in PBMAH.93,95-99 In contrast, CTNNB1 mutations are more frequently implicated in nonfunctioning ACAs; these also occur in primary aldosteronism
Detailed gross examination and submission of the entire periphery for microscopic evaluation
Workup for other possible primary and secondary tumors
NO
Confirmation of adrenocortical origin
ACA
In the absence of >5 mitoses/50 HPF, necrosis, vascular invasion or suspicious features
YES
NO
Altered reticulin framework
YES
NO
Any of the following:
NO
Mitotic count of >5/50HPF, necrosis, vascular invasion
Suspicious features
YES
Consider immunohistochemical screening and germline testing in setting of possible hereditary disease
ACC
YES
IGF-2 immunoexpression with a juxtanuclear staining pattern
NO
Mitosis-based grading
Any of these features: Ki67 labeling index >5%
≤20/50HPF Low grade
>20/50HPF High grade
p53 overexpression or global loss Diffuse nuclear beta-catenin
AND
Ki67 labeling index
YES
NO
AND
p53 and beta-catenin prognostication
Methylome signature Transcriptome profiling
ACT-UMP vs ACC Molecular testing
ACA vs ACT-UMP Molecular testing
or Follow-up
or adrenal Cushing syndrome.28 Advances have generated several genotype-phenotype correla- tions in functioning ACAs.28
Gene expression profiles based on transcrip- tome (gene expression) profiling, exome or whole-genome sequencing and single nucleotide polymorphism arrays, miRnome (microRNA expression) analysis, chromosomal alterations, and methylome (DNA-methylation) signatures have expanded the molecular landscape of ACTs by highlighting distinct diagnostic and prognostic signatures of ACCs. Transcriptome studies have shown that ACCs can be distinguished from ACAs based on differential gene expression pro- files.64,65 Among these, upregulation of IGF-2 stands out as one of the most important diagnostic characteristics of ACCs,64,65 irrespective of tumor cytomorphology, proliferative grade, and prog- nosis. The absence of IGF-2 alterations in adult ACAs makes IGF-2 a reliable candidate to be used in the distinction of ACCs. This finding is translated in paranuclear IGF-2 immunoreactivity in adult ACCs, as discussed earlier. Loss of het- erozygosity (11p15) involving the /GF-2 locus has also been noted in 91% of pediatric ACTs, under- scoring its role in tumorigenesis. 100
Transcriptome profiling has also shed light on the prognostic heterogeneity of ACCs by defining the bad and the good prognostic gene expression clusters.64,65 CTNNB1 and TP53 alterations, along with overexpression of cell cycle-related genes, have been specifically seen in the bad prognostic transcriptome clusters.
Subsequent studies have identified common drivers, chromosomal alterations, and miRnome profiles implicated in the pathogenesis of ACCs.53,101,102 The Wnt/beta-catenin signaling pathway (eg, ZNRF3, CTNNB1) has been found to be the most frequently altered pathway in ACCs.53,101,102 Other common driver genes have been implicated in the PKA pathway (eg, PRKAR1A), cell cycle regulation (eg, TP53, RB1, CDKN2A, CDK4, MDM2) as well as the chromatin remodeling (eg, MEN1, DAXX, ATRX) and chromo- some maintenance (eg, TERT, TERF2).53,101,102 CTNNB1 mutations are almost mutually exclusive from TP53 mutations. When comparing primary tumor tissue with metastatic clones, a 2.8-fold higher mutation rate was identified in addition to increased mutational heterogeneity among meta- static sites. 103
The Cancer Genome Atlas (TCGA) study identi- fied whole-genome doubling as a sign of tumor progression in ACC.53 The integrated genomics from ENSAT (the European Network for the Study of Adrenal Tumors)101 and TCGA53 series have further refined the spectrum of prognostic
molecular clusters of transcriptome data series. Transcriptome clusters have been correlated with ENSAT molecular prognostic subgroups based on methylome and miRnome signatures.101 The TCGA series defined 3 distinct molecular prog- nostic clusters that could be predicted using a 68-CpG probe DNA-methylation signature.53 ACCs with high-CpG island methylator phenotype (CIMP) status were enriched in the worst prog- nostic subgroups in both ENSAT and TCGA se- ries.53,101 Similarly, tumors within the high proliferative category and steroid phenotype were enriched in the worst prognostic TCGA clus- ter.53 A recent series identified the G0S2 (G0/G1 Switch 2) methylation as a hallmark of biologically aggressive and rapidly recurrent ACCs with high- CIMP status, irrespective of the tumor grade. 104 The same series underscored the role of BUB1B- PINK1 score (based on expression of BUB1B and PINK1), which also showed prognostic impact in earlier studies64,105 in the prognostication and treatment rationalization of ACCs with unmethy- lated G0S2 status.104
Heterogeneity and diagnostic challenges in pe- diatric ACCs have currently limited the character- ization of these lesions, although it has been shown that synchronous TP53 and ATRX muta- tions and associated genomics changes predict poor prognosis. 100
MAKING THE RIGHT DIAGNOSIS: USING AN ALGORITHMIC APPROACH
Practicing pathologists who are not exposed frequently to ACTs may be uncomfortable during the handling of such specimens. A systematic approach is often helpful in rendering the correct diagnosis. Although the use of ancillary tools can be helpful, detailed examination of morphologic findings is still the basis of any evaluation. This article proposes a practical integrated algorithmic approach to ACTs to tackle some of these com- mon diagnostic challenges (Fig. 29).
REFERENCES
1. Val P, Martinez A. Editorial: adrenal cortex: from physiology to disease. Front Endocrinol 2016;7: 1-2.
2. Young WF. Clinical practice. The incidentally discovered adrenal mass. N Engl J Med 2007; 356:601-10.
3. van Nederveen FH, de Krijger RR. Precursor le- sions of the adrenal gland. Pathobiology 2007;74: 285-90.
4. Mete O, Asa SL. Precursor lesions of endocrine system neoplasms. Pathology 2013;45:316-30.
5. Erickson LA, Rivera M, Zhang J. Adrenocortical carcinoma: review and update. Adv Anat Pathol 2014;21:151-9.
6. Duan K, Giordano TJ, Mete O. Adrenal cortical pro- liferations. In: Mete O, Asa SL, editors. Endocrine pathology. United Kingdom: Cambridge University Press; 2016. p. 602-27.
7. Funder JW, Carey RM, Fardella C, et al. Case detection, diagnosis, and treatment of patients with primary aldosteronism: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 2008;93:3266-81.
8. Stowasser M. Update in primary aldosteronism. J Clin Endocrinol Metab 2015; 100:1-10.
9. Duan K, Mete O. Clinicopathologic correlates of pri- mary aldosteronism. Arch Pathol Lab Med 2015; 139:948-54.
10. Duan K, Gomez Hernandez K, Mete O. Clinico- pathological correlates of adrenal Cushing’s syn- drome. J Clin Pathol 2015;68:175-86.
11. El-Maouche D, Arlt W, Merke DP. Congenital adre- nal hyperplasia. Lancet 2017;390:2194-210.
12. Sasano H. The adrenal cortex. In: Stefaneanu L, Sasano H, Kovacs K, editors. Molecular and Cellular Endocrine Pathology. London: Arnold; 2000. p. 221-52.
13. Grumbach MM, Biller BMK, Braunstein GD, et al. Management of the clinically inapparent adrenal mass (‘incidentaloma’). Ann Intern Med 2003;138: 424-9.
14. Francis IR. Distinguishing benign from malignant adrenal masses. Cancer Imaging 2003;3:102-10.
15. Daneman D, Daneman A. Diagnostic imaging of the thyroid and adrenal glands in childhood. Endo- crinol Metab Clin North Am 2005;34:745-68.
16. Doppman JL, Nieman LK, Travis WD, et al. CT and MR imaging of massive macronodular adrenocor- tical disease: a rare cause of autonomous primary adrenal hypercortisolism. J Comput Assist Tomogr 1991;15:773-9.
17. Vezzosi D, Tenenbaum F, Cazabat L, et al. Hormon- al, radiological, NP-59 scintigraphy, and patholog- ical correlations in patients with cushing’s syndrome due to primary pigmented nodular adre- nocortical disease (PPNAD). J Clin Endocrinol Metab 2015;100:4332-8.
18. Louiset E, Stratakis CA, Perraudin V, et al. The par- adoxical increase in cortisol secretion induced by dexamethasone in primary pigmented nodular adrenocortical disease involves a glucocorticoid receptor-mediated effect of dexamethasone on protein kinase A catalytic subunits. J Clin Endocri- nol Metab 2009;94:2406-13.
19. Mendichovszky IA, Powlson AS, Manavaki R, et al. Targeted molecular imaging in adrenal disease-an emerging role for metomidate PET-CT. Diagnostics (Basel) 2016;6, [pii:E42].
20. O’Shea PM, O’Donoghue D, Bashari W, et al. 11 C- Metomidate PET/CT is a useful adjunct for laterali- sation of primary aldosteronism in routine clinical practice. Clin Endocrinol (Oxf) 2019. https://doi. org/10.1111/cen.13942.
21. Dunnick NR, Korobkin M. Imaging of adrenal inci- dentalomas: current status. AJR Am J Roentgenol 2002; 179:559-68.
22. Francis IR, Mayo-Smith WW. Adrenal imaging. In: Hodler J, Kubik-Huch RA, von Schulthess GK, ed- itors. Diseases of the abdomen and pelvis 2018- 2021. Cham (Switzerland): Springer International Publishing; 2018. p. 85-90.
23. Zografos GN, Vasiliadis G, Farfaras AN, et al. Laparoscopic surgery for malignant adrenal tu- mors. JSLS 2009;13:196-202.
24. McNicol AM. A diagnostic approach to adrenal cortical lesions. Endocr Pathol 2008; 19:241-51.
25. Mete O, Asa SL. Morphological distinction of cortisol-producing and aldosterone-producing ad- renal cortical adenomas: not only possible but a critical clinical responsibility. Histopathology 2012; 60:1015-6, [author reply: 1016-7].
26. Aubert S, Wacrenier A, Leroy X, et al. Weiss system revisited: a clinicopathologic and immunohisto- chemical study of 49 adrenocortical tumors. Am J Surg Pathol 2002;26:1612-9.
27. Lloyd RV, Osamura RY, Kloppel G, et al. WHO clas- sification of tumours of endocrine organs. 4th edi- tion. Lyon (France): WHO Press; 2017.
28. Mete O, Duan K. The many faces of primary aldo- steronism and cushing syndrome: a reflection of adrenocortical tumor heterogeneity. Front Med 2018;5:54.
29. Patel KA, Calomeni EP, Nadasdy T, et al. Adrenal gland inclusions in patients treated with aldoste- rone antagonists (Spironolactone/Eplerenone): incidence, morphology, and ultrastructural find- ings. Diagn Pathol 2014;9:147.
30. Lowe KM, Young WF, Lyssikatos C, et al. Cushing syndrome in carney complex: clinical, pathologic, and molecular genetic findings in the 17 affected mayo clinic patients. Am J Surg Pathol 2017;41: 171-81.
31. Stratakis CA, Boikos SA. Genetics of adrenal tu- mors associated with Cushing’s syndrome: a new classification for bilateral adrenocortical hyperpla- sias. Nat Clin Pract Endocrinol Metab 2007;3: 748-57.
32. Morin E, Mete O, Wasserman JD, et al. Carney complex with adrenal cortical carcinoma. J Clin En- docrinol Metab 2012;97:E202-6.
33. Anselmo J, Medeiros S, Carneiro V, et al. A large family with Carney complex caused by the S147G PRKAR1A mutation shows a unique spectrum of disease including adrenocortical cancer. J Clin En- docrinol Metab 2012;97:351-9.
34. Duregon E, Volante M, Cappia S, et al. Oncocytic adrenocortical tumors: diagnostic algorithm and mitochondrial DNA profile in 27 cases. Am J Surg Pathol 2011;35:1882-93.
35. Renaudin K, Smati S, Wargny M, et al. Clinicopath- ological description of 43 oncocytic adrenocortical tumors: importance of Ki-67 in histoprognostic evaluation. Mod Pathol 2018;31:1708-16.
36. Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocor- tical tumors. Am J Surg Pathol 1984;8:163-9.
37. Weiss LM, Medeiros LJ, Vickery AL. Pathologic fea- tures of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989;13:202-6.
38. Bisceglia M, Ludovico O, Di Mattia A, et al. Adreno- cortical oncocytic tumors: report of 10 cases and review of the literature. Int J Surg Pathol 2004; 12: 231-43.
39. Volante M, Bollito E, Sperone P, et al. Clinicopatho- logical study of a series of 92 adrenocortical carci- nomas: from a proposal of simplified diagnostic algorithm to prognostic stratification. Histopatholo- gy 2009;55:535-43.
40. Duregon E, Fassina A, Volante M, et al. The reticulin algorithm for adrenocortical tumor diagnosis: a multicentric validation study on 245 unpublished cases. Am J Surg Pathol 2013;37:1433-40.
41. Mete O, Gucer H, Kefeli M, et al. Diagnostic and prognostic biomarkers of adrenal cortical carci- noma. Am J Surg Pathol 2018;42:201-13.
42. Fonseca D, Murthy SS, Tagore KR, et al. Diagnosis of adrenocortical tumors by reticulin algorithm. In- dian J Endocrinol Metab 2017;21:734-7.
43. Papotti M, Volante M, Duregon E, et al. Adrenocor- tical tumors with myxoid features: a distinct morphologic and phenotypical variant exhibiting malignant behavior. Am J Surg Pathol 2010;34: 973-83.
44. Morimoto R, Satoh F, Murakami O, et al. Immuno- histochemistry of a proliferation marker Ki67/MIB1 in adrenocortical carcinomas: Ki67/MIB1 labeling index is a predictor for recurrence of adrenocortical carcinomas. Endocr J 2008;55:49-55.
45. Pennanen M, Heiskanen I, Sane T, et al. Helsinki score-a novel model for prediction of metastases in adrenocortical carcinomas. Hum Pathol 2015; 46:404-10.
46. Duregon E, Cappellesso R, Maffeis V, et al. Valida- tion of the prognostic role of the ‘Helsinki Score’ in 225 cases of adrenocortical carcinoma. Hum Pathol 2017;62:1-7.
47. Lalli E, Figueiredo BC. Pediatric adrenocortical tu- mors: what they can tell us on adrenal development and comparison with adult adrenal tumors. Front Endocrinol 2015;6:1-9.
48. Wieneke JA, Thompson LDR, Heffess CS. Adrenal cortical neoplasms in the pediatric population: a
clinicopathologic and immunophenotypic analysis of 83 patients. Am J Surg Pathol 2003;27:867-81.
49. Picard C, Orbach D, Carton M, et al. Revisiting the role of the pathological grading in pediatric adrenal cortical tumors: results from a national cohort study with pathological review. Mod Pathol 2018. https:// doi.org/10.1038/s41379-018-0174-8.
50. Das S, Sengupta M, Islam N, et al. Weineke criteria, Ki-67 index and p53 status to study pediatric adre- nocortical tumors: Is there a correlation? J Pediatr Surg 2016;51:1795-800.
51. Chatterjee G, DasGupta S, Mukherjee G, et al. Use- fulness of Wieneke criteria in assessing morpho- logic characteristics of adrenocortical tumors in children. Pediatr Surg Int 2015;31:563-71.
52. Mete O, Asa SL, Giordano TJ, et al. Immunohisto- chemical biomarkers of adrenal cortical neo- plasms. Endocr Pathol 2018;29:137-49.
53. Zheng S, Cherniack AD, Dewal N, et al. Comprehensive pan-genomic characterization of adrenocortical carcinoma. Cancer Cell 2016;29: 723-36.
54. Miller BS, Gauger PG, Hammer GD, et al. Proposal for modification of the ENSAT staging system for adrenocortical carcinoma using tumor grade. Lan- genbecks Arch Surg 2010;395:955-61.
55. Assié G, Antoni G, Tissier F, et al. Prognostic pa- rameters of metastatic adrenocortical carcinoma. J Clin Endocrinol Metab 2007;92:148-54.
56. Giordano TJ. The argument for mitotic rate-based grading for the prognostication of adrenocortical carcinoma. Am J Surg Pathol 2011;35:471-3.
57. Duregon E, Volante M, Bollito E, et al. Pitfalls in the diagnosis of adrenocortical tumors: a lesson from 300 consultation cases. Hum Pathol 2015;46: 1799-807.
58. Papathomas TG, Duregon E, Korpershoek E, et al. Sarcomatoid adrenocortical carcinoma: a compre- hensive pathological, immunohistochemical, and targeted next-generation sequencing analysis. Hum Pathol 2016;58:113-22.
59. Schmitt A, Saremaslani P, Schmid S, et al. IGFII and MIB1 immunohistochemistry is helpful for the differentiation of benign from malignant adrenocor- tical tumours. Histopathology 2006;49:298-307.
60. Reincke M, Karl M, Travis WH, et al. p53 mutations in human adrenocortical neoplasms: immunohisto- chemical and molecular studies. J Clin Endocrinol Metab 1994;78:790-4.
61. Wasserman JD, Novokmet A, Eichler-Jonsson C, et al. Prevalence and functional consequence of TP53 mutations in pediatric adrenocortical carci- noma: a children’s oncology group study. J Clin Oncol 2015;33:602-9.
62. Waldmann J, Patsalis N, Fendrich V, et al. Clinical impact of TP53 alterations in adrenocortical carci- nomas. Langenbecks Arch Surg 2012;397:209-16.
63. Jouinot A, Bertherat J. Management of endocrine disease: adrenocortical carcinoma: differentiating the good from the poor prognosis tumors. Eur J En- docrinol 2018;178:R215-30.
64. de Reyniès A, Assié G, Rickman DS, et al. Gene expression profiling reveals a new classification of adrenocortical tumors and identifies molecular pre- dictors of malignancy and survival. J Clin Oncol 2009;27:1108-15.
65. Giordano TJ, Kuick R, Else T, et al. Molecular clas- sification and prognostication of adrenocortical tu- mors by transcriptome profiling. Clin Cancer Res 2009;15:668-76.
66. Duregon E, Molinaro L, Volante M, et al. Compara- tive diagnostic and prognostic performances of the hematoxylin-eosin and phospho-histone H3 mitotic count and Ki-67 index in adrenocortical carcinoma. Mod Pathol 2014;27:1246-54.
67. Fassnacht M, Dekkers O, Else T, et al. European Society of Endocrinology Clinical Practice Guide- lines on the management of adrenocortical carci- noma in adults, in collaboration with the European Network for the Study of Adrenal Tumors. Eur J En- docrinol 2018;179:G1-46.
68. Papathomas TG, Pucci E, Giordano TJ, et al. An in- ternational Ki67 reproducibility study in adrenal cortical carcinoma. Am J Surg Pathol 2016;40: 569-76.
69. Lu H, Papathomas TG, van Zessen D, et al. Auto- mated Selection of Hotspots (ASH): enhanced automated segmentation and adaptive step finding for Ki67 hotspot detection in adrenal cortical can- cer. Diagn Pathol 2014;9:216.
70. Beuschlein F, Weigel J, Saeger W, et al. Major prog- nostic role of Ki67 in localized adrenocortical carci- noma after complete resection. J Clin Endocrinol Metab 2015;100:841-9.
71. Yamazaki Y, Nakamura Y, Shibahara Y, et al. Com- parison of the methods for measuring the Ki-67 la- beling index in adrenocortical carcinoma: manual versus digital image analysis. Hum Pathol 2016; 53:41-50.
72. Heaphy CM, de Wilde RF, Jiao Y, et al. Altered telo- meres in tumors with ATRX and DAXX mutations. Science 2011;333:425.
73. Ross JS, Wang K, Rand JV, et al. Next-generation sequencing of adrenocortical carcinoma reveals new routes to targeted therapies. J Clin Pathol 2014;67:968-73.
74. Gomez-Sanchez CE, Gomez-Sanchez EP. Immu- nohistochemistry of the adrenal in primary aldoste- ronism. Curr Opin Endocrinol Diabetes Obes 2016; 23:242-8.
75. Yamazaki Y, Nakamura Y, Omata K, et al. Histopathological classification of cross-sectional image-negative hyperaldosteronism. J Clin Endo- crinol Metab 2017; 102:1182-92.
76. Skogseid B, Rastad J, Gobl A, et al. Adrenal lesion in multiple endocrine neoplasia type 1. Surgery 1995;118:1077-82.
77. Wagner J, Portwine C, Rabin K, et al. High fre- quency of germline p53 mutations in childhood adrenocortical cancer. J Natl Cancer Inst 1994; 86:1707-10.
78. Smith TG, Clark SK, Katz DE, et al. Adrenal masses are associated with familial adenomatous polypo- sis. Dis Colon Rectum 2000;43:1739-42.
79. Raymond VM, Everett JN, Furtado LV, et al. Adreno- cortical carcinoma is a lynch syndrome-associated cancer. J Clin Oncol 2013;31:3012-8.
80. Else T, Lerario AM, Everett J, et al. Adrenocortical carcinoma and succinate dehydrogenase gene mutations: an observational case series. Eur J En- docrinol 2017;177:439-44.
81. Menon RK, Ferrau F, Kurzawinski TR, et al. Adrenal cancer in neurofibromatosis type 1: case report and DNA analysis. Endocrinol Diabetes Metab Case Rep 2014;2014:140074.
82. Henry I, Jeanpierre M, Couillin P, et al. Molecular definition of the 11p15.5 region involved in Beckwith-Wiedemann syndrome and probably in predisposition to adrenocortical carcinoma. Hum Genet 1989;81:273-7.
83. Omata K, Anand SK, Hovelson DH, et al. Aldo- sterone-producing cell clusters frequently harbor somatic mutations and accumulate with age in normal adrenals. J Endocr Soc 2017;1: 787-99.
84. Omata K, Satoh F, Morimoto R, et al. Cellular and genetic causes of idiopathic hyperaldosteronism. Hypertension 2018;72:874-80.
85. Monticone S, Else T, Mulatero P, et al. Understand- ing primary aldosteronism: impact of next genera- tion sequencing and expression profiling. Mol Cell Endocrinol 2015;399:311-20.
86. Seidel E, Scholl UI. Intracellular molecular differ- ences in aldosterone- compared to cortisol- secreting adrenal cortical adenomas. Front Endo- crinol 2016;7:75.
87. Åkerström T, Crona J, Delgado Verdugo A, et al. Comprehensive re-sequencing of adrenal aldoste- rone producing lesions reveal three somatic muta- tions near the KCNJ5 potassium channel selectivity filter. PLoS One 2012;7:e41926.
88. Tan GC, Negro G, Pinggera A, et al. Aldosterone- producing adenomas: histopathology-genotype correlation and identification of a novel CACNA1D mutation. Hypertension 2017;70:129-36.
89. Nanba K, Chen AX, Omata K, et al. Molecular het- erogeneity in aldosterone-producing adenomas. J Clin Endocrinol Metab 2016;101:999-1007.
90. Lodish M, Stratakis CA. A genetic and molecular update on adrenocortical causes of Cushing syn- drome. Nat Rev Endocrinol 2016; 12:255-62.
91. Espiard S, Knape MJ, Bathon K, et al. Activating PRKACB somatic mutation in cortisol-producing adenomas. JCI Insight 2018;3.
92. Kamilaris CDC, Faucz FR, Voutetakis A, et al. Carney complex. Exp Clin Endocrinol Diabetes 2019;127:156-64.
93. Hannah-Shmouni F, Faucz FR, Stratakis CA. Alter- ations of Phosphodiesterases in Adrenocortical Tu- mors. Front Endocrinol 2016;7:111.
94. Assié G, Libé R, Espiard S, et al. ARMC5 mutations in macronodular adrenal hyperplasia with Cush- ing’s syndrome. N Engl J Med 2013;369:2105-14.
95. Yoshida M, Hiroi M, Imai T, et al. A case of ACTH- independent macronodular adrenal hyperplasia associated with multiple endocrine neoplasia type 1. Endocr J 2011;58:269-77.
96. Libé R, Fratticci A, Coste J, et al. Phosphodiesterase 11A (PDE11A) and genetic predisposition to adreno- cortical tumors. Clin Cancer Res 2008; 14:4016-24.
97. Matyakhina L, Freedman RJ, Bourdeau I, et al. He- reditary leiomyomatosis associated with bilateral, massive, macronodular adrenocortical disease and atypical cushing syndrome: a clinical and mo- lecular genetic investigation. J Clin Endocrinol Metab 2005;90:3773-9.
98. Hsiao H-P, Kirschner LS, Bourdeau I, et al. Clinical and genetic heterogeneity, overlap with other tumor syndromes, and atypical glucocorticoid hormone secretion in adrenocorticotropin-independent mac- ronodular adrenal hyperplasia compared with other adrenocortical tumors. J Clin Endocrinol Metab 2009;94:2930-7.
99. Gaujoux S, Pinson S, Gimenez-Roqueplo A-P, et al. Inactivation of the APC gene is constant in adreno- cortical tumors from patients with familial adenoma- tous polyposis but not frequent in sporadic adrenocortical cancers. Clin Cancer Res 2010;16: 5133-41.
100. Pinto EM, Chen X, Easton J, et al. Genomic land- scape of paediatric adrenocortical tumours. Nat Commun 2015;6:6302.
101. Assié G, Letouzé E, Fassnacht M, et al. Integrated genomic characterization of adrenocortical carci- noma. Nat Genet 2014;46:607-12.
102. Juhlin CC, Goh G, Healy JM, et al. Whole-exome sequencing characterizes the landscape of so- matic mutations and copy number alterations in adrenocortical carcinoma. J Clin Endocrinol Metab 2015;100:E493-502.
103. Gara SK, Lack J, Zhang L, et al. Metastatic adreno- cortical carcinoma displays higher mutation rate and tumor heterogeneity than primary tumors. Nat Commun 2018;9:4172.
104. Mohan DR, Lerario AM, Else T, et al. Targeted assessment of G0S2 methylation identifies a rapidly recurrent, routinely fatal molecular subtype of adrenocortical carcinoma. Clin Cancer Res 2019;25(11):3276-88.
105. Fragoso MCBV, Almeida MQ, Mazzuco TL, et al. Combined expression of BUB1B, DLGAP5, and PINK1 as predictors of poor outcome in adrenocor- tical tumors: validation in a Brazilian cohort of adult and pediatric patients. Eur J Endocrinol 2012; 166: 61-7.