Adrenocortical carcinoma (ACC) is a rare malignant endocrine neoplasm with an estimated incidence of 0.5 to 2 cases per 1 million people annually in the United States,1,2 accounting for 0.02% of all cancers reported annually.1 The prognosis for most patients diagnosed with ACC is disappointingly poor because of delays in diagnosis and the absence of effective systemic therapy. Approximately 50% of affected patients do not survive beyond 2 years after diagnosis, and the 5-year mortality rate approaches 80%.3 ACC has a bimodal age distribution with an increased incidence among children younger than 5 years of age and in individuals in their fourth and fifth decades of life. A slightly higher incidence rate is reported for women than for men.4 Approximately 40% of ACCs produce clinically significant excess amounts of steroid hormones, resulting in characteristic signs and symptoms;5 female patients are more likely to have an associated clinical endocrine syndrome. Surgery remains the only effective curative treatment for ACC. In a 1996 study of risk factors, cigarette smoking and the use of oral contraceptives were found to be associated with the development of ACC.6 An association has also been described between ACC and congenital adrenal hyperplasia.7

The etiology of ACC is unknown. A study of adrenocortical tumor clonality reported that whereas most benign adrenocortical lesions were polyclonal, ACCs were monoclonal, thus suggesting that ACC develops through the uncontrolled growth of a single cell.8

ACCs may be sporadic or occur as part of a hereditary tumor syndrome. Investigations of genetic alterations present in adrenocortical tumors have revealed the involvement of multiple chromosomal loci that correlate with regions that are abnormal in familial cancer syndromes. Such loci include those associated with Li-Fraumeni syndrome (LFS; p53 gene on17p13), multiple endocrine neoplasia type I (MEN1; MEN1 gene on 11q13), Beckwith-Wiedemann syndrome (11p15.5, correlated with the overproduction of insulin-like growth factor [IGF] II), and the Carney complex (loss of heterozygosity on 2p16).9

Although a multistep tumor progression model has been suggested in the etiology of sporadic ACC, proof of a hyperplasia-to-adenoma-to-carcinoma sequence is lacking.10,11 Insights into the pathogenesis of sporadic ACC have been gained from the study of familial cancer syndromes that include ACC. For instance, the most frequently inherited p53 mutations associated with LFS are also found in sporadic cases of ACC.12 One of the most common p53 point mutants, Arg 175 to His, fails to bind DNA and results in complete loss of p53 transcriptional activity. Although this mutation presents with a classic LFS cancer spectrum, including ACC, it also accounts for 6% of the missense mutations identified in all human cancers.13 In Brazil, where the rate of pediatric ACC is 10 to 15 times greater than the overall worldwide incidence, the majority of patients have the same germline point mutation of p53 encoding an Arg 337 to His amino acid substitution in exon 10.14 This mutation results in ACC development without the other associated tumor types seen in LFS. A pH-dependent destabilization of the mutant p53 tetramer in this R337H polymorphism allows for a adrenocortical-specific tumor formation.15

Sporadic ACCs, similar to hereditary ACCs associated with Beckwith-Wiedemann syndrome, have been found to overexpress the IGF-II gene. Several studies have identified a greater than 100-fold higher expression of IGF-II in 60% to 90% of sporadic ACCs compared with normal adrenal tissue and adrenal cortical adenomas.16,17 Increased IGF-II is thought to play a role in the etiology of ACC but is most likely in conjunction with concominant changes in expression of other genes such as 11p15LOH.18,19 Squamous cell carcinoma–related oncogene (SCCRO) is a novel gene involved in the hedgehog-signaling pathway, which is important in the development of the adrenal cortex. SCCRO is an “onco-developmental gene” important in both normal cellular function in the regulated state and in carcinogenesis in the dysregulated state; it may play a role in the development of adrenal cortical carcinoma.20

Approximatley 60% of ACCs in adults are hormonally functioning in that they produce measurable adrenal hormone excess; 40% of patients with ACC present with symptoms of hormonal excess, and in an additional 20% of patients, the hormone excess is subclinical and detected only on biochemical evaluation. The most common excess hormones identified are cortisol, aldosterone, and sex steroids; each is associated with a well-defined clinical syndrome (Table 16-1). Most adrenal tumors that secrete excess amounts of multiple steroid hormones are malignant, so this may be the first clue to a diagnosis of ACC.

Cushing’s syndrome (CS) is the most common clinical hormone excess syndrome associated with ACC (present in 30% to 40% of cases). It is caused by excess production of cortisol, and in ACC patients, it is often seen in combination with virilization caused by concomitant androgen hypersecretion. CS is characterized by truncal obesity; a rounded face (moon facies); violaceous striae on the abdominal wall; muscle weakness; and thin, fragile skin. Patients may have any combination of fatigue, proximal muscle weakness, osteoporosis, hypertension, and glucose intolerance. The rapid onset of CS with virilizing features is characterisitic of ACC.

Excess androgen production often goes unoticed in men; however, in women, it produces signs that may include hirsutism, a deepened voice, menstrual irregularity, male pattern baldness, and clitoromegaly. ACC-associated virilization is characteristically more pronounced than that associated with functioning adrenocortical adenomas; the severity of virilization in malignant adrenal tumors is correlated with a characteristically relatively high rate of cosecretion of 17-ketosteroids and dehydroepiandrosterone (DHEA).21

Excess production of estrogen is rare; when it occurs in women, it may result in menstrual irregularity or breast tenderness. Estrogen excess in men may present as feminization, with associated impotence, decreased libido, testicular atrophy, and gynecomastia.22 Production of aldosterone by ACC is relatively uncommon, occurring as an isolated hormone elevation in only approximately 3% of patients and in up to 11% as part of a mixed hormone presentation. When present, aldosterone excess is usually associated with the typical findings of hypertension and hypokalemia (which may be profound).23 Of note, these same symptoms may be caused by severe CS in the absence of excess aldosterone production; in such patients, the overwhelming excess levels of cortisol may stimulate mineralcorticoid receptors.24 In children, ACCs are most commonly functioning tumors (90% of cases); most pediatric ACCs produce androgens, resulting in precocious puberty and virilization.

Patients with nonfunctioning ACC typically present with symptoms related to their large retroperitoneal mass, including abdominal pain, weight loss, nausea and vomiting, or early satiety, and they may have a palpable abdominal mass. Occasionally, a nonfunctioning ACC is discovered incidentally during adominal or thoracic imaging. Unfortunately, patients with nonfunctioning ACCs in particular may not present until they develop signs or symptoms of advanced or metastatic disease (e.g., severe fatigue, jaundice, bone pain, pulmonary embolism). At the time of initial presentation, 70% of cases have extra-adrenal spread, mostly to the lung (45% to 53% of cases), liver (42% to 46% of cases), and lymph nodes (18% to 40% of cases).1,25 In a series of 160 consecutive operative patients evaluated for ACC by Lee et al., 142 (89%) patients had locoregional disease and 18 (11%) patients had distant disease at the time of presentation. Locoregional disease may include involvement of the ipsilateral kidney, retroperitoneal or peritoneal spaces, regional lymph nodes, diaphragm, inferior vena cava (IVC), or liver (on the right) or the pancreas, spleen, or colon (on the left).26

The diagnostic evaluation of a patient with known or suspected ACC should always include a biochemical endocrine analysis and radiographic imaging. Endocrine evaluation is performed to identify or confirm excess hormone production (Table 16-2). Confirmation of excess hormone production is used as a guide to preoperative preparation (e.g., partial control of the effects of cortisol or aldosterone overproduction before laparotomy is done may be clinically very helpful in reducing the operative risk). Hormone levels, if documented as elevated preoperatively, can also be used as a “tumor marker” to monitor outcome after surgery, evaluate response to systemic therapy, assist in the interpretation of postoperative imaging studies, and guide interval follow-up.

Listed in Table 16-2 and described below are options for testing. In patients presenting with an adrenal mass for which ACC (as well as pheochromocytoma) is in the differential diagnosis, we routinely perform biochemical evaluation of the major potential associated syndromes (pheochromocytoma, CS, virilization or feminization, and hyperaldosteronism). The recommendations included here assume that the patient is considered at relatively elevated risk for ACC based on the size of the adrenal tumor (generally at least 4 cm), its imaging characteristics (irregular, inhomogeneous, or locally invasive), or the presence of suggestive signs or symptoms (pain, clinical evidence of steroid hormone excess).

Every patient with a newly diagnosed adrenal mass should be assessed for excess cathecholamine production to rule out pheochromocytoma. Traditionally, this has been accomplished by measurement of fractionated metanephrines, fractionated catecholamines, and urine vanillylmandelic acid in a 24-hour urinary specimen. Alternatively (and increasingly commonly), fractionated plasma-free metanephrines may be substituted for the timed urine collection as a screening test. In comparing the diagnostic efficacy of both tests, Sawka et al.27 found that urinary tests yielded fewer false-positive results. However, the plasma test is somewhat more sensitive as well as somewhat more convenient for the patient than a timed urine collection; plasma screening may also be preferred in patients at risk for an inherited endocrine syndrome. At our institution, we routinely screen for pheochromocytoma by measurement of fractionated plasma-free metanephrines in patients with adrenal masses suspicious for ACC.

Cortisol is the hormone most frequently overexpressed by ACC, either alone or in combination with sex steroids or aldosterone. In patients without signs or symptoms of cortisol excess, an overnight 1-mg dexamethasone suppression test (DST) is a sensitive way to exclude the presence of cortisol overproduction. This test involves the administration of 1 mg of dexamethasone orally at 10 PM and the measurement of a plasma cortisol level the following morning at 8 AM. A normal response is a suppression of the morning cortisol level to 3 μg/dL or less. Failure of dexamethasone to suppress the plasma cortisol level below 3 g/dL suggests that the patient may have autonomous cortisol production. Because false-positive results with the 1-mg overnight DST are relatively common, patients with an elevated cortisol level after overnight testing should undergo confirmatory testing with the combination of paired 8 AM serum cortisol and adrenocorticotropic hormone (ACTH) levels together with a 24-hour urinary-free cortisol level.

Timed urine collection for cortisol determination should also be obtained in patients with obvious signs or symptoms of cortisol excess at presentation; screening via overnight DST is unnecessary in such patients. In equivocal circumstances, a formal 2-day low-dose DST (0.5 mg of dexamethasone every 6 hours for 2 days with pre- and post-DST 24-hour urinary-free cortisol determination) may be used to confirm the presence of subtle overproduction of cortisol by an adrenal tumor. However, in our experience, this test is rarely necessary in the diagnosis, preoperative preparation, or postoperative follow-up of patients with known or suspected ACC.

Evaluation for androgen and estrogen excess should include serum DHEA sulfate, androstenedione, testosterone, and 17-β-estradiol. Evaluation for hyperaldosteronism includes measurement of serum potassium (to identify hypokalemia) and serum level of aldosterone and renin activity (so that an aldosterone:renin ratio can be calculated).

Imaging

Computed tomography (CT) is the initial imaging modality of choice for the diagnosis and characterization of most adrenal tumors, including those possibly representing ACC (Table 16-3). Most ACCs are at least 5 cm in diameter,28 so size is an important criterion when evaluating an adrenal cortical lesion with CT imaging. Only 2% of adrenal lesions 4 cm or smaller are ACCs, 6% of lesions 4.1 to 6.0 cm are ACCs, and 25% of lesions larger than 6 cm are ACCs. It is important to recognize that CT may underestimate the true size of a tumor (as measured from the surgical specimen) by up to 20%.29 In a review by Barnett et al.,29 the mean radiographic size estimate for ACCs was 9.5 cm (1.7 to 30 cm), but the mean pathologic measurement of the resected tumors was 11.7 cm (3.0 to 30 cm). This difference was significant (P = 0.001).

Using a cutoff of more than 30 Hounsfield units (HU), contrast-enhanced CT has been reported to have both a high sensitivity and a high positive predictive value in identifying malignant adrenal tumors.30 The use of intravenous contrast material may also be helpful in characterizing and staging ACCs. On contrast-enhanced images, ACCs typically show peripheral enhancement of the mass with a central non-enhanced area of necrosis, giving the tumor a heterogenous appearance. Additionally, the measurement of contrast washout may also be helpful in distinguishing benign adrenocortical adenomas from ACCs; malignant cortical tumors characteristically demonstrate a slower washout.

ACCs are often irregular with poorly defined margins. Calcifications or cystic degeneration are seen in about 30% of these tumors. CT also provides essential staging information in allowing detection of metastatic disease in the regional lymph nodes, liver, and lungs. Finally, CT assists in surgical planning. Imaging occasionally demonstrates invasion of the upper pole of the kidney or the IVC; associated adjacent nodal metastasis are also occasionally identified.

Magnetic resonance imaging (MRI) is also an excellent alternative imaging modality for the characterization of adrenal tumors, including ACCs. On T1-weighted images, ACCs are typically hypointense relative to the liver, but on T2-weighted images, ACCs are hyperintense relative to the liver. However, it is important to recognize that ACCs may appear heterogeneous on both T1- and T2-weighted images owing to internal hemorrhage and necrosis, and blood products in areas of hemorrhage may cause portions of an ACC to appear bright on T1-weighted images. One advantage of MRI is its ability to demonstrate flow within major blood vessels, which may allow visualization of invasion of an ACC into the renal vein or IVC. Of note, right-sided ACCs in particular have a propensity to form venous tumor emboli, and vascular invasion and tumor thrombus can be seen with flow-sensitive MRI sequences.31 Enhancement after MRI contrast agent administration is usually most pronounced around the periphery of ACCs in non-necrotic areas, and contrast washout is often prolonged.

[18F]fluorodeoxyglucose (FDG) positron emission tomography (PET) is an important imaging modality in the evaluation of many different types of malignant processes. FDG-PET exploits the characteristically high glucose consumption of many tumor cells and can be a sensitive imaging modality in the identification of ACC recurrence and metastasis. However, FDG-PET has some limitations, including the fact that infections and inflammatory conditions (including postoperative inflammation) can result in prominent focal FDG uptake, potentially leading to false-positive findings on PET imaging. Nevertheless, FDG-PET, particularly when combined with non-contrast CT imaging (PET-CT), can be extremely useful in the detection of recurrence or metastases in patients with ACC. In particular, PET-CT can be helpful in excluding distant metastatic disease in patients with recurrent ACC being considered for surgery.

Size

As mentioned previously, perhaps the single most important predictor of malignancy in localized, nonmetastatic adrenocortical tumors is size. Most ACCs are more than 6 cm in diameter at the time of diagnosis,32 but adrenal masses smaller than 4 cm in diameter are generally benign. These data have been incorporated into the National Institutes of Health’s consensus criteria for evaluation of incidental adrenal masses, which include a recommendation that any adrenal mass larger than 6 cm in diameter be resected, regardless of its functional status or imaging characteristics.33 A review of the literature over a 30-year period found that 9% of reported ACCs were smaller than 5 cm in diameter at the time of diagnosis.34 As a result, many surgeons have chosen to resect all tumors larger than 4 to 5 cm in size in patients who are good surgical candidates. Smaller lesions that have imaging characteristics suspicious for primary ACC (irregular borders, heterogenous enhancement) should also be resected, preferably by an open approach (as discussed below).