Treatment of adrenocortical carcinoma

INTRODUCTION — Adrenocortical carcinomas (ACCs) are rare and frequently aggressive tumors that may be functional (hormone-secreting) and cause Cushing's syndrome and/or virilization, or nonfunctional and present as an abdominal mass or as an incidental finding.

The management of ACC will be discussed here. The clinical presentation and diagnostic evaluation of patients with adrenal masses, the staging workup for ACC, and the management of benign adrenal adenomas are reviewed separately. (See "Clinical presentation and evaluation of adrenocortical tumors" and "Evaluation and management of the adrenal incidentaloma".)

PRIMARY TREATMENT — Our approach to management is largely consistent with the clinical practice guidelines on the management of adrenocortical cancer published by the European Society of Endocrinology in collaboration with the European Network for the Study of Adrenal Tumors (ENSAT) [1].

Initial surgery — Complete surgical resection is the only potentially curative treatment for adrenocortical carcinoma (ACC) [1,2]. For patients with potentially resectable stage I to III disease who are surgical candidates, we recommend complete surgical resection as initial therapy.

Before proceeding to a surgical excision, all patients must undergo a complete hormonal assessment to determine the secretory activity of the tumor. It is particularly important to identify those with cortisol-producing tumors. These patients, even those with mild hypercortisolism, have some degree of hypothalamic-pituitary-adrenal (HPA) axis suppression and require glucocorticoid coverage to prevent postoperative adrenal insufficiency [1,3]. (See "Clinical presentation and evaluation of adrenocortical tumors", section on 'Hormonal evaluation'.)

The surgery should be performed in a specialized referral center by teams of surgeons with specific expertise to avoid tumor spillage and incomplete resection, which is associated with a poor prognosis [1,4,5]. (See 'Stage and margin status' below.)

The preoperative preparation for and techniques of surgical adrenalectomy are discussed elsewhere. (See "Adrenalectomy techniques".)

For potentially resectable tumors invading adjacent organs, surgery often needs to be extensive, with en bloc resection of involved organs such as kidney, liver, spleen, pancreas, stomach, and colon [6]. Intracaval extension or tumor thrombus is not a contraindication to surgery; resection may be facilitated by cardiopulmonary bypass [7].

Suspicious lymph nodes should be resected, but the benefit of routine lymphadenectomy has not yet been established. ACCs often spread via lymphatic drainage. A benefit for routine lymphadenectomy during adrenalectomy was suggested in a report from the German ACC Study Group of 283 patients with completely resected ACC [8]. There was a significantly reduced risk for tumor recurrence and disease-related death (hazard ratio [HR] 0.54, 95% CI 0.29-0.99) in patients who underwent lymphadenectomy versus those who did not. In a second study, peritumoral lymph node resection was associated with improved overall survival (OS) [9].

Metastatic lymphatic spread may be more extensive than previously thought. In a study of 56 patients with ACC, lymphatic recurrence was detected in a number of areas [10].

Left-sided ACC (n=36):

●Left renal hilum (50 percent)

●Perirenal fat tissue cranial to the renal hilum (47 to 55 percent in ventral and dorsal areas)

●Para-aortic (47 percent)

●Interaortocaval (22 percent)

●In the perirenal fat tissue caudal to the renal hilum (17 to 20 percent)

Right-sided ACC (n = 20):

●Perirenal fat tissue cranial to the renal hilum (45 to 55 percent in ventral and dorsal areas)

●Interaortocaval (35 percent)

●In the area of the right renal artery (10 percent)

●Paracaval (15 percent)

●Left para-aortic lymph node (10 percent)

Although resection is technically possible for most patients with stage I to III disease (table 1), it is not curative for many, presumably because occult micrometastases are present at the time of initial presentation, even with stage I disease [11-15]. As an example, in a single center report of 202 consecutive cases of ACC, 40 percent of patients with stage I to III disease (table 1) had developed distant metastasis two years after diagnosis (27, 46, and 63 percent of patients with stage I, II, and III disease, respectively) [16].

However, there is some evidence that outcomes are improving over time. (See 'Prognosis' below.)

Even if the tumor cannot be removed entirely, some clinicians advocate maximal debulking as a means of improving survival [11,17-21], although others disagree as to the survival benefit of this strategy [22,23]. Randomized trials to support routine debulking of nonresectable tumors are lacking, and decision making must be individualized, taking into account the underlying tumor biology, rate of progression, and the histologic grade [1,4]. For patients with advanced functioning tumors, debulking may help to control hormone hypersecretion and increase the efficacy of further therapies [2,5,24]. Overall, however, patients with unresectable disease have a poor prognosis, particularly with high-grade disease, often surviving only three to nine months, and they are often better palliated with medical management [18,25]. (See 'Recurrent or advanced adrenocortical cancer' below.)

The role of neoadjuvant systemic therapy prior to surgery for patients with locally advanced disease is not defined, and this is not considered a standard approach. In adults, one retrospective study suggested that patients with locoregionally advanced tumors with borderline resectable tumors might benefit from cisplatin-based neoadjuvant chemotherapy [26]. However, definitive evidence is still lacking, and therapeutic decisions should be individualized.

Laparoscopic versus open resection — We suggest open rather than laparoscopic adrenalectomy for patients with ACC. Laparoscopic surgery has largely replaced the open technique for management of benign adrenal pathology, and laparoscopic removal of even large adrenal tumors can be performed safely by experienced surgeons [27-29]. (See "Adrenalectomy techniques".)

However, the role of laparoscopic resection for ACCs is controversial, and open surgery remains the standard approach at least in our centers, despite the European clinical practice guideline on adrenal incidentaloma, which recommends laparoscopic resection for tumors less than 6 cm even when ACC is suspected in the absence of local tumor invasion [30]. Clinical practice guidelines recommend surgery for all tumors with radiologic findings suspicious of malignancy and evidence for local invasion [1]. However, for tumors <6 cm without any evidence of local invasion (unknown if benign or malignant), laparoscopic adrenalectomy is reasonable if the surgeon has sufficient experience. Several retrospective studies have shown more frequent or earlier recurrences and a shorter disease-free survival when this technique is used for management of ACC [4,31-33]. This is not a universal finding, however, and others have shown comparable outcomes from laparoscopic and open adrenalectomy, particularly for tumors up to 10 cm in size in specialized centers [34-38]. Nonetheless, this is not a widely accepted approach, and we and others [34] suggest open rather than laparoscopic resection for known or highly suspected ACC, regardless of size [4,31,39,40].

Prognostic factors

Stage and margin status — The most important clinical factors that determine prognosis of ACC are disease stage and completeness of resection [1,12,16,17,25,41-45] (see "Clinical presentation and evaluation of adrenocortical tumors", section on 'Staging'). In one observational series of 253 patients from the French Association of Endocrine Surgeons Study Group, five-year OS was 66, 58, 24, and 0 percent for stage I, II, III, and IV (metastatic) disease, respectively [41].

A modification has been proposed for the ENSAT staging system that incorporates histologic grade of differentiation (table 2) [46]. In one report, incorporating patient age (>55 years) into available staging systems appeared to better predict OS in patients with stage I and II ACC [47].

However, survival differs widely for any given tumor stage, and many other factors influence outcomes. Regardless of stage, incomplete resection is associated with a poor prognosis (median survival generally less than one year) [4,5]. The influence of margin status on prognosis was shown in a report from the National Cancer Database, in which five-year OS rates for patients with ACC and uninvolved, microscopically involved, and macroscopically involved margins were 46, 21, and 10 percent, respectively [48].

Histology — In addition to stage and completeness of resection, the biologic behavior of ACCs is also influenced by pathologic/morphologic factors. Histologically, the appearance of ACCs ranges from mild atypia to wildly anaplastic tumors composed of monstrous giant cells. The widely applied multifactorial scoring system of Weiss is based upon nine histopathologic features (nuclear grade, mitotic rate, atypical mitoses, clear cell component, diffuse architecture, tumor necrosis, invasion of venous or sinus structures, or tumor capsule); tumors with less than three features are usually considered as benign [49]. (See "Clinical presentation and evaluation of adrenocortical tumors", section on 'Fine-needle aspiration biopsy'.)

These criteria have been validated as prognostic factors in more contemporary series both in adults and children [12,43,44]. As an example, in one study of 124 patients with ACC, significant predictors of disease progression included distant metastasis at presentation; tumor invasion of vessels, tumor capsule, or adjacent organs; tumor necrosis; a high mitotic rate; the presence of atypical mitosis; and overexpression of mdm-2 [12]. Five-year disease-free survival was significantly different for patients with one or two, three or four, or more than four adverse features (84, 37, and 9 percent, respectively).

Several studies confirm the prognostic value of markers of proliferation, including mitotic rate, and Ki-67 expression (as detected by a monoclonal antibody against Ki-67, MIB-1) [12,50-54]. The importance of proliferative rate in prognostication was shown in an analysis of 124 patients with ACC, in whom the five-year disease-specific survival rates for individuals with tumor mitotic rates of ≤5, 6 to 10, 21 to 50, and >50 per 50 high-power fields [HPF] were 63, 50, 25, and 0 percent, respectively [12].

The combined utility of tumor stage and mitotic rate to assess prognosis was addressed in a study of 92 patients with malignant ACC [51]. Three risk groups were identified (low [stage I/II, mitoses ≤9/50 HPF], intermediate [stage I/II plus >9 mitoses/50 HPF or stage III/IV plus mitotic rate ≤9/50 HPF], and high [stage III/IV and mitotic rate >9/50 HPF]) that had significantly different rates of mean disease-free survival (62, 17, and 12 months, respectively) and OS (not attained, 66, and 26 months, respectively).

The Ki-67 labeling index (LI) has been used to select patients for adjuvant chemotherapy in addition to mitotane. While the absolute value of Ki-67 LI that permits the differentiation of a very high-risk ACC from low- or intermediate-risk ACC is not established [3,55], a high-grade ACC, which is defined in part by a high mitotic rate (and/or Ki-67 score ≥20 percent), is at a higher risk of recurrence than a low-grade ACC. (See 'Patients with very high recurrence risk' below.)

Other factors — Other clinical factors, immunohistochemistry, and molecular markers are associated with poor survival. Among clinical factors, older age at diagnosis and hypersecretion of cortisol have been recognized as adverse factors [16].

Molecular predictors of malignant potential and survival are emerging, but none are ready for clinical use [56-61].

Some reports suggest the value of mutated TP53, mutated b-catenin, ERCC1 (excision-repair, complementing defective in Chinese Hamster 1 gene), IGF2, SF1 (splicing factor-1), the glucose transporter GLUT1, low expression of the SGK1 (serum/glucocorticoid-regulated kinase 1) gene, and G0S2 (G0/G1 Switch 2 gene) hypermethylation are predictors of poor prognosis [12,62-69]. Studies with global gene expression profiling suggest that specific patterns of gene expression are strongly associated with poor outcomes, independently of traditional histologic predictors such as tumor stage and Weiss score [56,57,59,61,70].

TREATMENT AFTER INITIAL SURGERY

Approach to adjuvant therapy — For patients with resected adrenocortical carcinoma, the decision to offer adjuvant therapy is primarily based upon the risk of disease recurrence, which is influenced by three major prognostic factors: tumor stage, completeness of resection, and proliferation rate (as determined by the mitotic rate, MIB-1 staining, or Ki-67 expression) [49,51,52,55,71]. (See 'Prognostic factors' above.)

Patients at lower risk of recurrence typically have a good prognosis and may be candidates for surveillance. In contrast, those at higher recurrence risk may be candidates for adjuvant therapy, such as systemic therapy containing mitotane or radiation therapy. The use of molecular markers to better define individual risk for recurrence and tailor decisions about therapy remains investigational [56,61]. (See 'Other factors' above.)

Patients with low recurrence risk — For most patients at low risk of recurrence after complete resection (stage I to III disease, microscopically complete [R0] resection, Ki-67 ≤10 percent), we suggest observation rather than adjuvant mitotane. In a randomized clinical trial, the use of adjuvant mitotane did not significantly improve recurrence-free survival (RFS) or overall survival (OS) over observation in this low-risk population [72]. These recommendations are consistent with guidelines from the European Society of Endocrinology/European Network for the Study of Adrenal Tumours (ENSAT) and endorsed by A5 (American Australian Asian Adrenal Alliance) as outlined by an international consensus panel on the treatment of ACC [52].

Alternatively, adjuvant mitotane may be appropriate for select patients who meet criteria for low-risk disease but also harbor some features that suggest a potential risk for recurrence. Examples include patients with stage III low-grade disease, or those with a large (>20 cm) stage II tumor with a borderline mitotic rate or Ki-67 score. However, if such patients are treated, a shorter duration would be appropriate (eg, two years), with low threshold to discontinue mitotane if side effects are difficult to manage. Further details on the optimal duration of adjuvant mitotane are discussed below. (See 'Duration' below.)

The approach to adjuvant therapy in patients with low-risk, resected ACC disease is evolving as survival rates have improved over time [73]. Although initial observational data supported the use of adjuvant therapy after resection in most patients with ACC due to high recurrence rates [12,74,75], subsequent studies have suggested that patients with low-risk disease have favorable five-year RFS rates of approximately 75 percent [72,73]. Further data on the efficacy of adjuvant therapy in resected ACC are discussed below. (See 'Efficacy' below.)

These observational studies led to a randomized clinical trial (ADIUVO) in 91 patients at low-risk for ACC [72]. Low-risk features included stage I to III ACC, completed (R0) resection, and Ki-67 ≤10 percent. In this study, patients were randomly assigned to either a minimum of two years of adjuvant mitotane or observation. For patients receiving mitotane, the median duration of therapy was 21 months. Adjuvant mitotane, when compared with observation, failed to demonstrate statistically significant improvements in RFS at median follow-up of 48 months (five-year RFS 79 versus 75 percent, HR 0.74, 95% CI 0.3-1.85) or OS at median follow-up of 55 months (five-year OS 95 versus 86 percent, HR 0.46, 95% CI 0.08-1.92). A concurrent observational (ADIUVO OBSERVATIONAL) study demonstrated similar results, with a recurrence rate of 21 percent among patients with low-risk disease. The trial was closed early due to low accrual rates; additionally, despite recruitment at centers of excellence, few patients met the inclusion criteria for mitotic rate (Ki-67 ≤10 percent).

Data from the German ACC registry suggest that observation alone is also appropriate if patients also have a tumor size <8 cm and no microscopic evidence of invasion of blood vessels or the tumor capsule, in addition to the criteria for low-risk disease [76].

Patients with high recurrence risk — For patients at high risk of recurrence, we suggest the use of adjuvant mitotane use alone (without chemotherapy) rather than observation alone. Clinical features that indicate high recurrence risk include any one of the following:

●Histologically high-grade disease (Ki-67 staining of >10 percent and <20 percent of tumor cells [77] or >20 mitotic figures per 50 HPF regardless of tumor size [46,78]).

●Incompletely resected disease [52].

●Intraoperative tumor spillage from tumor fracture or capsule rupture.

●Large, low-grade tumors with vascular or capsular invasion.

This approach is consistent with clinical guidelines from the European Society of Endocrinology as outlined by ENSAT and endorsed by A5 in international consensus panel guidelines [52]. Further details on the efficacy, toxicity, and administration of adjuvant mitotane are discussed below. (See 'Adjuvant mitotane' below.)

Patients with very high recurrence risk — Although high-quality data are limited, patients at very high risk for an early recurrence (eg, very high Ki-67 staining [≥20 percent]) and extensive vascular invasion/vena cava thrombus) may be offered an adjuvant regimen that includes the addition of cisplatin-based chemotherapy to mitotane. However, it is not known if cytotoxic chemotherapy alone or in combination with mitotane is more effective than adjuvant mitotane alone [3].

Data on the efficacy of cytotoxic chemotherapy in the adjuvant setting are as follows:

●In a retrospective cohort study, 31 patients aged 4 to 59 years with resected ACC and very high risk disease (median Ki-67 of 30 percent) received adjuvant platinum-based chemotherapy [79]. Most patients received a combination of either cisplatin or carboplatin plus etoposide and a majority (28 patients) were also treated with adjuvant mitotane. Compared with matched controls, the addition of adjuvant platinum-based chemotherapy to mitotane was associated with improved RFS (HR 0.19) and OS (HR 0.26) three months after surgery.

●In a separate case series, the combination of cisplatin and etoposide was explored in the adjuvant setting among five individuals aged 1.2 to 21 years [80]. All received etoposide 165 mg/m² and cisplatin 90 mg/m² every three to four weeks for at least six cycles, beginning shortly after surgical resection, and all remained in remission 29 to 109 months later. However, ACC in children has a different natural history than that arising in adults. (See 'Pediatric patients' below.)

●Other adjuvant therapies have been evaluated, such as the combination of mitotane plus streptozotocin [81], but these approaches remain investigational.

The use of Ki-67 LI to select patients for adjuvant chemotherapy in addition to mitotane has been suggested [71,76]. A potential prognostic role for Ki-67 LI following complete resection of ACC was shown in an analysis of 319 patients with resected ENSAT stage I to III ACC [55]. On multivariate analysis, age, tumor size, venous tumor thrombus, and Ki-67 LI were all significantly correlated with RFS and OS; however, Ki-67 LI was the most important prognostic factor. The median RFS durations in patients with a Ki-67 LI <10, 10 to 19, and ≥20 percent were 53, 32, and 9 months, respectively; the corresponding values for median OS were 181, 114, and 42 months, respectively. Importantly, the predictive value of Ki-67 LI (ie, its ability to predict benefit from cytotoxic chemotherapy) was not addressed.

While the absolute value of Ki-67 LI that permits the differentiation of a "high-risk" from low- or intermediate-risk ACC is not established [3], a high-grade ACC, which is defined in part by a high mitotic rate (and/or Ki-67 score ≥20 percent), is at a higher risk of recurrence than low-grade ACC.

Adjuvant mitotane — Mitotane (o,p'-DDD, a congener of the pesticide dichlorodiphenyltrichloroethane [DDT]) is an adrenocorticolytic drug that has efficacy in patients with ACC [11,82-84]. It has been used in the adjuvant setting for primary therapy of unresectable disease and for the treatment of disease recurrence, either alone or in combination with other cytotoxic agents. (See 'Mitotane monotherapy' below and "Medical therapy of hypercortisolism (Cushing's syndrome)", section on 'Mitotane'.)

Efficacy — Data demonstrating the benefit of routine postoperative (adjuvant) administration of mitotane have been mixed, in large part due to the rarity of ACC and the lack of prospective, randomized clinical trials. While a number of uncontrolled reports suggest that adjuvant mitotane may delay or prevent recurrence in patients undergoing complete primary resection for nonmetastatic disease [4,14,73,75,85-91], others have failed to show any benefit in terms of disease-free or OS [5,17,23,41,50,74,92-97]. Among the potential reasons for these disparate results are insufficient dosing in some studies and variability in an individual tumor's ability to metabolize mitotane (which is required for therapeutic action).

The best available evidence that adjuvant mitotane improves outcomes for patients with resected stage I to III ACC comes from a retrospective analysis of 177 patients who had undergone macroscopically complete radical surgery at eight Italian and 47 German centers between 1985 and 2005 [75]. In the Italian cohort, 47 received adjuvant mitotane while 55 patients did not (Italian control group). None of the 75 patients in the German cohort received adjuvant mitotane (German control group).

The groups were well matched, with the exception that the German control patients were more likely to have stage I or II disease (84 versus 67 percent of both Italian groups). With a median follow-up that ranged from 43 to 68 months in the three groups, the following results were seen:

●In the mitotane group, 27 patients received 1 to 3 grams daily, and the remainder received 3 to 5 grams daily. Serum levels were not routinely measured. The median duration of treatment was 29 months for both doses.

●Mitotane treatment was associated with a significantly longer RFS as compared with either control group (median RFS 42 versus 10 and 25 months in the Italian and German control groups, respectively).

●There were also fewer deaths from ACC in the mitotane group than in either control group (25 versus 55 and 41 percent, respectively). Median OS durations in the three groups were 110 months in the mitotane group compared with 52 months in the Italian controls and 67 months in the German control group.

●Mitotane was reasonably well-tolerated. Grade 3 gastrointestinal (nausea, vomiting, or elevated serum gamma-glutamyl transpeptidase [GGT]) or neurologic events (confusion, ataxia, vertigo) were observed in 15 and 20 percent of the patients who received the higher dose mitotane regimen; neither of these problems occurred with lower doses.

These data support the view that adjuvant mitotane therapy, even at relatively low doses, improves RFS and OS after macroscopically complete radical resection for stage I, II, or III ACC. The strengths of this retrospective series include its large size, involvement of multiple institutions, well-matched control cohort, long duration, systematic nature of follow-up, and carefully conducted statistical analysis [91]. A report of this cohort after nine additional years of follow-up suggests that the benefit of adjuvant mitotane on disease-free survival is maintained [98].

Benefit was also supported by three other studies [73,99,100]. In one report of 149 German ACC registry patients with stage II ACC, the subgroup of 35 patients who received adjuvant mitotane displayed a significantly better five-year survival than the 114 patients not receiving adjuvant mitotane (87 versus 53 percent, respectively). In a retrospective study of 100 patients treated with adjuvant mitotane and 52 who did not receive mitotane, adjuvant therapy lowered the risk of recurrence but did not improve OS [100]. However, a survival benefit was seen in a subset of mitotane-treated patients (those with stage III ACC or an elevated Ki-67 index).

Duration

●High-risk ACC – For patients with resected high-risk ACC, independent of stage, we aim to continue adjuvant mitotane for three to five years, depending on tolerability and the length of time patient has achieved target therapeutic mitotane levels. The rationale for this duration is that it is very atypical for a patient with ACC to recur after four to five years. (See 'Patients with high recurrence risk' above.)

●Low-risk ACC – Observation is preferred for most patients with resected low-risk ACC. However, for those with indications for adjuvant therapy, we aim for a minimum of two years of adjuvant mitotane [1,71], based on the duration of administration used in the ADIUVO trial [72]. For patients with low-risk disease and tumor spillage, we will, at times, continue adjuvant therapy for up to five years if the drug is well tolerated [4]. In various studies, the median duration of adjuvant mitotane ranges between 21 and 29 months [72,75]. (See 'Efficacy' above and 'Patients with low recurrence risk' above.)

Suggested regimen — While benefit of mitotane in the adjuvant setting has been reported using relatively low doses (1 to 3 grams daily) [75], current treatment protocols are based upon achieving therapeutic mitotane serum levels. We suggest that all patients receiving mitotane (including those being treated for advanced disease) undergo therapeutic monitoring with plasma mitotane levels every four to six weeks.

The following represents our approach to managing therapy with mitotane:

●We initiate adjuvant treatment as soon as possible after surgery (within three months).

●Mitotane is initiated at 0.5 g twice per day and increased to 6 g/day over 4 to 12 weeks as tolerated, with monitoring of mitotane level every two to three weeks.

●Because of the long serum half-life and accumulation in adipose tissue, several weeks may be necessary to raise serum mitotane concentrations to the target level of 14 to 20 mcg/mL [96,101].

●When serum mitotane monitoring is not available, we suggest trying to push dosing to toxicity (up to 6 to 8 g per day) for patients with high-risk disease and adjust according to tolerability. Some patients may not tolerate more than 2 g per day.

●At some centers, mitotane is administered to all patients using rapid escalation of high doses (4 g/day rapidly escalating to therapeutic doses within two weeks) [102]. This regimen may shorten the time needed to reach therapeutic levels of mitotane, but it requires closer follow-up and combined clinical and mitotane level monitoring and may be more frequently associated with side effects.

●Nausea should be treated using metoclopramide or a serotonin (5-HT3) receptor antagonist such as ondansetron. (See "Prevention of chemotherapy-induced nausea and vomiting in adults".)

Our approach of measuring serum mitotane levels, targeting a range of 14 to 20 mcg/mL, has been validated in several studies as correlating with therapeutic response while minimizing toxicity [6,93,96,101,103-106]. As examples:

●A retrospective series included 91 patients receiving mitotane for unresectable or metastatic ACC who underwent assay of serum mitotane levels within three months of attaining the best response to mitotane with or without chemotherapy [104]. A significantly higher number of responders (11 of 17 responders) had serum mitotane levels >14 mcg/mL, and of the six responders who had lower levels, all were receiving concurrent chemotherapy. Survival was also significantly longer among patients with levels >14 mcg/mL (median survival 24 versus 18 months, hazard ratio [HR] for death 0.52, 95% CI 0.28-0.97).

●In a prospective study, therapeutic monitoring achieved better tolerance of adjuvant mitotane therapy with fewer side effects [103].

●Furthermore, in the adjuvant setting, improved outcomes for patients with therapeutic mitotane levels were shown in a report of 43 consecutive patients undergoing surgical treatment for ACC followed by monitored mitotane; 29 (67 percent) had stage III or IV tumors, and 33 underwent potentially curative surgery [6]. Four patients treated prior to 1990 (when mitotane monitoring was not available) received a low dose of the drug (1 g twice daily), while 13 others did not tolerate the drug and were kept at lower levels; 24 patients (75 percent of whom had stage III or IV tumors) were maintained at mitotane levels exceeding 14 mcg/mL. Despite the preponderance of stage III and IV tumors, the five-year disease-specific survival rate was high (64 percent). When the HR for death from ACC was compared over time between the patients with high mitotane levels versus the patients with lower doses or without mitotane treatment, there was a positive survival effect that persisted up to four years after surgery.

Every patient metabolizes mitotane to different degrees and, hence, requires unique dosing to achieve therapeutic blood levels. Therapeutic serum levels can at times be achieved with low doses of mitotane. As an example, in one series of eight patients receiving 3 g daily (of the non-micronized formulation), therapeutic concentrations (14 to 20 mcg/mL) were reached in three to five months (after a cumulative dose of approximately 360 g), and five patients maintained clinical benefit (stable disease or disease free) with minimal toxicity for 8 to 40 months with as little as 1 to 2 g daily [107].

Inherited polymorphisms in drug-metabolizing enzymes have been described that may predict the individual response to adjuvant mitotane [108-110]. However, at present, there is no defined clinical role for genotyping prior to the initiation of mitotane therapy in this or any other setting.

In the United States and Canada, plasma levels of mitotane can be measured in commercial laboratories and some academic centers; in Europe, they are available through HRA Pharma, which distributes mitotane as an orphan drug. We recommend monitoring plasma mitotane levels in all patients, if possible, as outlined below.

Monitoring response

Imaging — The efficacy of adjuvant mitotane can be assessed periodically during therapy by the following techniques:

●Surveillance for recurrence of disease should include contrast-enhanced computed tomography (CT) scans (or magnetic resonance imaging [MRI]) of the chest, abdomen, and pelvis every three months for two to three years, then every four to six months for five years. (See "Radiation-related risks of imaging".)

●Although the role of fluorodeoxyglucose (FDG)-positron emission tomography (PET) in post-treatment monitoring is not yet established, some centers add integrated FDG-PET/CT at six-month intervals in the post-treatment follow-up strategy [34]. Limited studies comparing the usefulness of integrated PET/CT scans (which are typically done without intravenous [IV] contrast) versus diagnostic (ie, contrast-enhanced) CT performed at the same time have shown that integrated PET/CT detects more lesions than does PET or CT alone, and that PET should be considered a complementary examination to contrast-enhanced CT or MRI [111-113]. In one study, PET was more sensitive than CT in detecting local recurrence, while CT was more sensitive in detecting small lung or peritoneal metastases [113].

Nevertheless, the utility of PET in the setting of primary treatment remains controversial. At the University of Michigan, PET is not performed routinely, either prior to primary surgery for a suspected ACC or as a component of the post-treatment surveillance strategy. However, other centers (including that of one of the authors [AL]) routinely perform PET prior to surgery for a suspected ACC and add integrated FDG-PET/CT at six-month intervals in the post-treatment follow-up strategy for patients with high-risk disease who have been shown to have FDG-avid disease preoperatively (see 'Local therapy' below). In more advanced disease, PET/CT may be preferred for assessment of chemotherapeutic response; in a small proportion of patients in one report, PET/CT predicted response before anatomic changes were detected on CT [114].

Biochemical — For patients with a completely resected, steroid-producing ACC, some groups monitor for recurrent hormone excess every three months for two years with steroid tumor markers such as cortisol (measured in the morning before hydrocortisone dose), dehydroepiandrosterone sulfate (DHEAS), androstenedione, testosterone, estradiol, or mineralocorticoid based upon the steroid profile in the initial tumor. Other groups do not routinely screen for recurrent hormone excess unless or until there is evidence of recurrent disease.

Mitotane, which increases serum concentrations of corticosteroid-binding globulin (CBG), may result in serum cortisol values that are artifactually elevated [115] without significant changes in corticotropin (ACTH). Despite alterations in cortisol metabolism induced by mitotane, 24-hour urinary free cortisol excretion remains the best index of cortisol production currently available (table 3) [115]. (See 'Adrenal insufficiency' below.)

Use of liquid chromatography-mass spectroscopy assays for urine cortisol and its metabolites should eventually offer better tools to assess steroid hormone production and requirements in ACC patients receiving mitotane [116].

Toxicities

Side effects — The toxicity profile of mitotane limits tolerability, particularly at doses above 6 grams per day. The most common side effects are fatigue, nausea, vomiting, and anorexia, but skin rash, diarrhea, lethargy, sedation, confusion, dizziness, ataxia, gynecomastia, arthralgias, leukopenia, prolonged bleeding time, hematuria, and reversible growth arrest in children also occur [15,117].

Other adverse effects include frequent hyperlipidemia [118] (elevated low-density lipoprotein [LDL] cholesterol, high-density lipoprotein [HDL] cholesterol, triglycerides), hypouricemia, and hepatotoxicity (table 3). If statins are used for the dyslipidemia, pravastatin and rosuvastatin should be considered as they have less drug interaction with mitotane.

While increased concentrations of serum GGT and alkaline phosphatase are almost invariably observed, a significant rise in transaminases or bilirubin is infrequent.

Adrenal insufficiency — Mitotane is a steroidogenesis inhibitor and adrenolytic drug. Glucocorticoid replacement therapy is necessary for patients treated with mitotane (unless the ACC is an advanced, glucocorticoid-producing tumor presenting as Cushing's syndrome). Independent of the pharmacologic effects of mitotane, for patients with cortisol-secreting ACC, suppression of ACTH-adrenal axis will require replacement with glucocorticoids following complete tumor resection. (See 'Initial surgery' above.)

Very long-term use of mitotane usually induces atrophy and/or steroidogenic inhibition of the normal adrenal glands, thereby causing cortisol deficiency. The zona glomerulosa is more resistant to the adrenolytic effect of mitotane, and aldosterone deficiency may not occur until after several months of therapy. (See "Medical therapy of hypercortisolism (Cushing's syndrome)".)

Our approach to managing adrenal insufficiency includes the following:

●Initiate glucocorticoid therapy – We suggest starting replacement glucocorticoid therapy when mitotane treatment is initiated because one cannot predict when a patient will become hypocortisolemic.

For all patients who are receiving mitotane, we initiate replacement with hydrocortisone (30 to 40 mg daily in divided doses, two to three times per day). Of note, most patients eventually require a two- to threefold increase in hydrocortisone. Patients undergoing mitotane treatment should wear a medical alert bracelet and be instructed to increase their doses of hydrocortisone during stress.

In patients with residual ACC and persistent hypercortisolemia, glucocorticoid replacement should not be initiated until hypercortisolism is controlled with mitotane and steroid enzyme inhibitors. (See 'Medical treatment of hormone excess' below.)

Several times the usual maintenance doses of glucocorticoids are sometimes needed as the induction of hepatic cytochrome P450 enzymes by mitotane increases the rate of metabolism of cortisol, dexamethasone [119,120], and fludrocortisone. Inadequately treated adrenal insufficiency enhances mitotane-induced side effects and reduces drug tolerance [15,90,103]. A brief trial of a higher glucocorticoid dose may mitigate some side effects that could be attributable to inadequate cortisol replacement and allow continuation of mitotane at the same dose.

●Monitoring – Close monitoring of blood sodium, potassium, creatinine, ACTH, and 24-hour urine free urinary cortisol levels is necessary to avoid adrenal insufficiency and acute hyperkalemia in patients treated with mitotane, whether in the adjuvant setting or for advanced disease (table 3).

Measurement of serum mitotane levels together with 24-hour urine free cortisol with an assessment of side effects should guide mitotane and hydrocortisone dosing. If side effects do not abate with glucocorticoid therapy, temporary discontinuation of mitotane is indicated, followed by reinstitution at a lower dose. Measuring cortisol levels in hair may eventually become useful to assess chronic hydrocortisone replacement in ACC [121].

●Aldosterone deficiency – Mitotane can also eventually cause aldosterone deficiency. Unlike glucocorticoid replacement, we do not start mineralocorticoid replacement right away. Instead, we suggest monitoring blood pressure at each visit, serum potassium at every three months, and plasma renin every six months (table 3). Mineralocorticoid deficiency should be suspected if the patient develops postural hypotension, hyponatremia, or hyperkalemia and has an elevation in plasma renin activity.

When clinical and biochemical signs of aldosterone deficiency become present, we suggest starting fludrocortisone (0.1 to 0.3 mg daily) to restore normal clinical and biochemical parameters (table 3). As noted, the metabolism of fludrocortisone is increased by mitotane.

●Stopping mitotane and monitoring for recovery – After stopping mitotane therapy, hydrocortisone therapy should be continued but progressively decreased to more physiologic levels (as mitotane acceleration of glucocorticoid metabolism returns to normal) until the patient has no clinical or biochemical evidence of adrenal insufficiency.

Patients should be evaluated for potential recovery of adrenal insufficiency every six months by measuring morning levels of serum cortisol before intake of hydrocortisone replacement. Although adrenal insufficiency may become permanent in some patients, depending upon the duration of mitotane therapy, it may recover in others. As an example, in a study of 23 patients with ACC treated with adjuvant mitotane for a minimum of two years, 18 of 23 (78 percent) patients achieved a complete hypothalamic-pituitary-adrenal (HPA) axis recovery within a mean interval of 2.7 years after discontinuing the adjuvant mitotane [122].

Reproductive issues — In men receiving mitotane therapy, hypogonadism is common, often requiring replacement testosterone therapy. In addition to its effect on CBG, mitotane increases serum concentrations of other binding globulins, including sex hormone-binding globulin (SHBG) [15,103]. In addition to potential decreases in free testosterone levels, mitotane induces a strong inhibition of systemic 5-alpha-reductase activity, which may explain the relative inefficiency of testosterone replacement in mitotane-treated men [120]. Serum testosterone should be measured routinely in men receiving mitotane (table 3). (See "Clinical features and diagnosis of male hypogonadism" and "Testosterone treatment of male hypogonadism".)

The impact of mitotane on reproductive function in women is less clear. Increased levels of both luteinizing hormone (LH) and follicle-stimulating hormone (FSH) have been observed, which in some premenopausal women has resulted in the development of large ovarian cysts [123]. These cysts may be associated with lower abdomen discomfort and pain; cases of ovarian torsion requiring surgery have also been observed. (See "Ovarian and fallopian tube torsion".)

There are concerns that pregnancy could lead to an increased likelihood of relapse, although data are minimal [40]. In addition, the safety of mitotane during pregnancy is unknown. Therefore, for women of reproductive age scheduled to receive adjuvant mitotane therapy after ACC resection, we suggest contraception during and for at least one year after mitotane as the drug has a long half-life and is deposited in adipose tissues. We suggest barrier methods of contraception as mitotane accelerates steroid metabolism and may decrease the contraceptive efficacy of hormonal contraception. (See "Contraception: Counseling and selection".)

Hypothyroidism — In addition to its effect on CBG, mitotane increases serum concentrations of other binding globulins, including thyroid hormone-binding globulin (TBG). Therefore, serum concentrations of free thyroxine (T4) may be decreased [15,103]. In addition, newer studies indicate a direct inhibitory effect of mitotane on secretion of thyroid-stimulating hormone (TSH) [103]. As patients on mitotane therapy often present with fatigue, thyroid function tests should be monitored periodically (table 3). Hypothyroidism with low TSH is frequent, and free T4 should be monitored. Thyroid hormone replacement is suggested in patients with clinical symptoms of hypothyroidism and low free T4 values.

Drug interactions — Mitotane is a potent inducer of CYP3A4 metabolism. Limited data suggest that this can result in important drug-drug interactions [124]. In a report of two patients with ACC who were receiving mitotane in combination with sunitinib (a drug metabolized by CYP3A4), serum sunitinib concentrations were decreased by approximately 80 percent of expected values [125]. Although clinical data are currently unavailable, mitotane administration could reduce the efficacy of other drugs relevant to the management of patients with ACC, such as steroid hormone replacement, benzodiazepines, macrolide antibiotics, some opioids and statins, and dihydropyridine type calcium channel antagonists [124].

Specific interactions of mitotane with other medications may be determined using Lexicomp drug interactions.

Adjuvant radiation therapy — Data on the usefulness of adjuvant radiation therapy following surgical resection has been limited by the retrospective nature of studies with small number of patients. The European Society of Endocrinology clinical practice guidelines panel suggest against the routine use of radiation therapy in patients with stage I to II and R0 resection. However, they suggest considering radiation in addition to mitotane therapy in patients with R1 or Rx resection or in stage III [1].

We suggest the addition of postoperative radiation therapy (RT) for all patients with incompletely resected ACC, stage III disease with other characteristics of high recurrence risk, those who have tumor spillage at the time of resection, and for all patients with high-grade ACC (>20 mitotic figures per 50 HPF). The benefit of adjuvant RT is limited to improved local control; no clear survival benefit has been shown except in one study. We suggest starting RT as soon as possible following surgery (ideally within 12 weeks).

In the past, ACC was thought to be a relatively radioresistant tumor. However, this is likely not true with modern RT techniques [126]. While there are no randomized trials testing the efficacy of adjuvant RT in patients with resected ACC, some, but not all, studies support a benefit in certain patients who are at high risk for a local recurrence [126-132].

●One retrospective report addressing the benefit of adjuvant RT is from the German ACC registry [127]. Outcomes of 14 patients (stage I, II, and III (table 1) in seven, three, and three patients, respectively) without macroscopic residual disease who received postoperative RT were compared with 14 others who did not receive RT and who were matched for resection status, use of adjuvant mitotane, stage, and tumor size. Local recurrence developed in 2 of 14 irradiated patients compared with 11 of 14 who did not receive RT. However, despite this apparent benefit, neither disease-free nor OS were significantly better in irradiated patients.

A later analysis of patterns of failure in these patients suggested that many of the recurrences developed in the inter-aortocaval region [126]. This observation emphasizes the importance of covering this area in the radiation portal, as well as the ipsilateral lymphatic drainage bed, if adjuvant RT is planned.

●A retrospective study from the University of Michigan compared outcomes of 20 patients with localized ACC undergoing an R0/R1 resection followed by postoperative RT (median dose 55 Gy) with those of 20 patients treated contemporaneously with surgery alone, matched for stage, surgical margin status, tumor grade, and use of adjuvant mitotane [129]. At a median follow-up of 34 months, local recurrence was significantly more frequent in those not receiving RT, with local recurrence in 12 patients versus one (60 versus 5 percent, HR 12.59, 95% CI 1.62-97.88). As was seen in the German ACC registry analysis, this benefit did not translate into an improvement in RFS or OS.

●In contrast, in a retrospective study of 16 patients with ACC who had undergone either surgery followed by adjuvant RT (n = 16) or surgery only (n = 32), the rate of local recurrence was not significantly reduced by adjuvant RT (7 of 16 [44 percent] and 10 of 32 [31.3 percent] in the RT and control groups, respectively) [133].

●In the largest single institution study (including 39 patients), gross surgical resection of ACC improved local RFS, all RFS, and OS in a propensity-matched analysis. Therefore, we suggest that adjuvant RT should be considered a part of multidisciplinary management for patients with ACC [134].

Larger, prospective, randomized studies are needed to better define the role of adjuvant RT after resection of ACC, in particular to confirm if the decrease in local recurrence rate provides any OS benefit.

Our approach is similar to that of the investigators from the German ACC registry, who recommended adjuvant RT for all patients with microscopically incomplete (R1 or R2) or uncertain (Rx) margin status and for those with stage III disease (according to ENSAT criteria (table 4)) even if resection has been complete [2,126]. They suggest that adjuvant RT be considered for patients who have had a complete (R0) resection of a tumor >8 cm in size with tumor invasion of the blood vessels (but not large tumor thrombus in the vena cava) and a Ki-67 proliferative index of >10 percent, and for patients who have intraoperative violation of the tumor capsule, tumor spillage, or dissemination of "necrotic" fluid. Because the risk of a local recurrence is highest in the first two years, they recommend starting RT no later than three months after surgery.

There is disagreement as to the role of adjuvant RT in patients who have undergone laparoscopic rather than open resection, and there are no published data to help in resolving this point. There are different opinions regarding the safety and efficacy of laparoscopic resection of primary ACC. Thus, it is understandable that there are different opinions regarding the use of adjuvant RT in the setting of laparoscopic resection for a resected ACC that does not otherwise meet the criteria described above for adjuvant RT (tumor spillage at the time of resection, incompletely resected, high-grade lesion). The German ACC group does not recommend adjuvant RT in this situation.

The Adrenal Cancer Program at the University of Michigan, as reflected in the opinion of one of the authors (GH), does not recommend laparoscopic adrenalectomy for ACC and hence is more aggressive in recommending adjuvant RT following a laparoscopic resection of ACC if there is any concern that a sufficient oncologic resection was not performed. The other author (AL) does not pursue adjuvant RT after a laparoscopic resection (ACC at pathology, low ACC risk by preoperative investigation and German ACC criteria described above) as long as the laparoscopic surgery was performed by an expert surgeon (ie, one who specializes in and is trained in minimally invasive endocrine or oncology surgery, who practices in a large academic referral center, and who does adrenal surgery weekly).

RECURRENT OR ADVANCED ADRENOCORTICAL CANCER — There is no curative therapy for metastatic or recurrent adrenocortical carcinoma (ACC). However, the symptoms of steroid excess can be controlled by medical therapy. In most cases, metastatic disease is fatal within one year, although there are rare long-term responses ascribed to chemotherapy with or without mitotane [1,117,135,136].

Local therapy

Surgery — For patients with isolated, locally recurrent ACC that is surgically accessible, we suggest complete surgical resection followed by mitotane therapy.

Where complete removal is feasible, aggressive surgical resection of locally recurrent disease, with the aim of achieving negative surgical margins, should be undertaken [75,86,137-139]. The best candidates are those who have potentially resectable disease with a disease-free interval of at least one year after initial treatment.

Resection may also be considered in the rare patient who presents with synchronous limited, potentially resectable hepatic or pulmonary metastases [140]. Resection of locally recurrent disease may also be indicated for patients in whom surgery will be able to remove a majority of tumor burden or decrease severe hypercortisolism that is otherwise difficult to control; however, recovery from surgery may be slow, delaying administration of systemic therapy [4]. Thus, this approach should be limited to selected patients with uncontrollable symptomatic hormone excess or who are in imminent danger from organ invasion or compression.

Aggressive resection of locally recurrent or distant disease may prolong survival in some patients [4,5,137-139,141,142], and at least some reports suggest that long-term survival rates are higher among patients who undergo resection compared with those who do not [74,137,140,143,144]:

●In a series of 47 patients with ACC undergoing resection for locally recurrent or distant metastatic disease at Memorial Sloan-Kettering Cancer Center (MSKCC), five-year survival rates for patients with completely and incompletely resected locally recurrent disease were 57 and 0 percent, respectively [137].

●Similar outcomes were reported in a series of 57 patients undergoing 116 procedures (23 for liver metastases, 48 for pulmonary metastases, 13 for metastases at other sites, 22 for abdominal disease including local recurrences) for recurrent or metastatic ACC [144]. Five-year survival was 41 percent, and median survival was significantly longer for those with a disease-free interval >12 months (6.6 versus 1.7 years). The use of chemotherapy and mitotane had no effect on survival.

●In another report of 28 patients undergoing resection for liver metastases from ACC (25 isolated, three with extrahepatic metastases), the five-year survival rate was 39 percent [142].

Response to neoadjuvant chemotherapy may be useful in defining which patients may benefit from surgical intervention. Radiofrequency ablation (RFA) is an alternative approach for these patients. (See 'Radiofrequency ablation' below.)

Radiation therapy — For palliation of symptoms from locally advanced or distant metastatic disease (eg, a bone metastasis), we use palliative radiation therapy (RT). The available data support the palliative benefit of RT for unresectable local tumor that is causing local symptoms or for distant symptomatic metastases, such as in bone [4,86,126,145-147]. In a survey of published reports totaling 91 patients receiving palliative RT for advanced ACC from the German ACC registry, benefit (ie, pain relief, reduction in paresthesia or paralysis) was observed in 57 percent [126]. (See "Radiation therapy for the management of painful bone metastases".)

Stereotactic radiosurgery may be beneficial for patients who have a good performance status and limited metastases to the brain, lung, or liver. (See "Stereotactic cranial radiosurgery" and "Radiation therapy techniques in cancer treatment", section on 'Stereotactic radiation therapy techniques'.)

Radiofrequency ablation — Percutaneous RFA may provide short-term local control of an unresectable primary tumor, particularly for those <5 cm in diameter [4,148,149]. The long-term impact on survival is unknown. RFA has also been used to treat small liver metastases [148,150,151].

Systemic therapy

Mitotane monotherapy — Primary treatment with mitotane may be indicated for patients who have histologically proven ACC in whom surgery is incomplete, not feasible, or contraindicated.

The quality of the available literature on mitotane monotherapy is poor, and the results are highly variable. Most studies were conducted in early years without monitoring of tumor response with adequate imaging. Moreover, mitotane has often been given in a suboptimal way; much of the variability in outcomes may be attributable to subtherapeutic serum concentrations (see 'Suggested regimen' above). No data are available on survival or drug tolerance among patients with unresectable disease who are treated with an approach that includes monitoring of plasma mitotane concentrations (see 'Suggested regimen' above and 'Toxicities' above). Determinants of mitotane efficacy are unknown; CYP2W1, which is highly expressed in ACC, may represent a new predictive marker for the response to mitotane treatment [108].

Treatment benefits are generally short lived [17,18], and survival is not consistently prolonged. While there are isolated case reports of long-term disease control and rare cases of prolonged complete remission in patients with inoperable or metastatic disease [117,152-154], these are almost always patients with low-grade disease. In several studies of primary mitotane therapy, the median survival was only 6.5 months [83,155], no different from that expected in untreated patients [18,94,156].

The main benefit of mitotane for patients with unresectable advanced disease is usually reduction in symptoms of hypercortisolism (weakness, myopathy, diabetes, immunosuppression, insomnia). While a decrease in excess hormone production is reported in up to 75 percent of patients and tumor size is reduced in as many as one-third of cases [83,92,93,155], mitotane therapy alone is usually not sufficient to significantly decrease cortisol levels or have long-lasting effects on tumor growth. In a review of nine prospective series totaling 246 patients with advanced ACC (not restricted to primary therapy), the estimated mean objective regression rate with mitotane was only 26 percent [15]. Furthermore, some patients who have an objective response with mitotane are incapacitated by the side effects of the drug [157]. (See 'Toxicities' above.)

We reserve mitotane monotherapy for the few patients who have a minimal burden (low number of tumor-involved organs) of low-grade (ie, low mitotic rate) disease and have recurred relatively late (two to three years) after surgery. These are the patients who are likely to have a prolonged survival [54]. In the setting of extensive (multiple tumor-involved organs), rapidly progressive, high-grade disease, mitotane is almost always administered in combination with cytotoxic chemotherapy because of a generally accepted view that response rates are higher. (See 'Chemotherapy plus mitotane' below.)

For patients with active disease and visible lesions, mitotane is typically continued lifelong or until disease progression. However, several protocols for new targeted agents for ACC specifically require discontinuation of mitotane because of potentially detrimental drug interactions. Specific issues related to dosing, monitoring of serum levels during therapy, side effects, and drug interactions are addressed above. (See 'Suggested regimen' above and 'Drug interactions' above and 'Toxicities' above.)

For patients whose disease progresses while receiving adjuvant mitotane, or whose disease is high grade or rapidly progressing (on or off mitotane), our usual practice is to initiate cytotoxic chemotherapy, oftentimes while continuing mitotane, given the higher response rates.

Cytotoxic chemotherapy — Many cytotoxic drugs have been studied in patients with advanced ACC, either alone or in combination with mitotane. Few non-mitotane-containing chemotherapy regimens have been tested in ACC, largely due to the rarity of the disease. Progress in therapy of advanced ACC has been limited by the rarity of the disease and the relatively small number of patients in each study.

●Although they are frequently used, single agents such as cisplatin or doxorubicin are associated with response rates that are generally less than 30 percent and of short duration [158-160].

●Results are disappointing for chemotherapy combinations when not combined with mitotane [161-166].

Chemotherapy plus mitotane — It is generally accepted, though not proven, that chemotherapy plus mitotane produces better outcomes than does mitotane alone. Therefore, in the setting of extensive (multiple tumor-involved organs), rapidly progressive, high-grade disease, mitotane is almost always administered in combination with cytotoxic chemotherapy. For patients receiving combined therapy, we suggest mitotane in combination with etoposide, doxorubicin, and cisplatin (EDP) as front-line therapy.

Mitotane increases the cytotoxic activity of other chemotherapeutic drugs on human adrenal carcinoma cells in vitro, possibly by acting as an antagonist of the multidrug resistance (MDR) protein, which is found in high concentrations in ACC, and functions as a drug efflux pump [167,168]. These data provide a rationale for combination therapy.

A few prospective trials have explored combination regimens that contain mitotane. Although no chemotherapy regimen has been shown to improve overall survival (OS) in patients with advanced ACC, some of the more encouraging results have been with the combination of EDP plus mitotane:

●An early study of EDP plus mitotane in 72 ACC patients described an overall response rate of 49 percent [169].

●The largest trial of advanced ACC to date, the First International Randomized trial in Locally Advanced and Metastatic Adrenocortical Carcinoma Treatment (FIRM-ACT), randomly assigned 304 patients with advanced ACC not amenable to radical surgery to mitotane plus either EDP or streptozotocin [170]. Rates of objective tumor response (23 versus 9 percent) and median progression-free survival (5 versus 2.1 months) were both significantly better in the EDP-mitotane (EDP-M) group, although these benefits did not translate into a significantly prolonged survival (median 14.8 versus 12 months). This lack of significant difference in OS, despite better progression-free survival for EDP-M, is possibly due to the crossover design of the study since EDP-M was superior as a second-line treatment as well. Rates of serious adverse events did not differ significantly between treatments. The efficacy of both regimens as second-line therapy was similar to their efficacy as first-line therapy.

●Results with other chemotherapy drugs combined with mitotane seem to be less promising [81,171-174].

Newer and investigational approaches — Several novel approaches are under study for treatment of advanced ACC, many of which represent molecularly targeted therapies. These therapies have only been evaluated in the context of clinical trials, often after progression of disease on standard therapies (eg, mitotane or EDP-mitotane).

Immunotherapy — Programmed cell death ligand (PD-L1) is expressed in some adrenocortical carcinomas (ACCs) and their associated tumor infiltrating lymphocytes, leading to interest in the use of checkpoint inhibitor immunotherapy in these patients. Further studies are needed to determine predictors of immunotherapy response in those with ACC as treatment appears effective regardless of microsatellite instability-high (MSI-H) status [175].

Pembrolizumab has shown a good safety profile and response rates ranging from 14 to 50 percent in phase II studies conducted in both adult and pediatric populations [175-177]:

●In an open-label, single-arm phase II study, 39 adult patients with advanced ACC were treated with pembrolizumab [175]. A majority had received prior systemic therapy (either mitotane or platinum-based chemotherapy). At a median follow-up of approximately 18 months, the objective response rate was 23 percent (9 patients); median progression-free and OS were 2 and 24 months respectively. Additionally, responses were seen in two of six patients with microsatellite instability high/mismatch repair deficient (MSI-H/MMR-D) tumors. Rates of grade ≥3 toxicity were low (13 percent).

●Similar results were seen in another open-label, single-arm phase II trial. In this study, 16 adult patients with ACC refractory to previous therapies (approximately one-half with cortisol-producing tumors) were treated with the programmed death 1 (PD-1) inhibitor pembrolizumab [176]. At approximately seven-month follow-up, objective responses were seen in two patients (14 percent) and stable disease in seven patients (50 percent). Rates of grade ≥3 toxicities were low (13 percent), with only one patient discontinuing therapy due to pulmonary toxicity.

●In a phase I-II open label trial of pembrolizumab (KEYNOTE-051), which enrolled 154 pediatric patients with various relapsed solid tumors, two of four patients with adrenocortical carcinoma achieved a partial response [177].

Other approaches

●IGF1R inhibitors – Approximately 80 percent of ACCs overexpress insulin-like growth factor (IGF) type 2 (IGF-2), which is known to signal predominantly through the IGF-1 receptor (IGF1R).

While preclinical and early phase studies of monoclonal antibodies or small molecules inhibiting IFG1R were initially promising [178-181], a subsequent phase I-II trial of the anti-IGF1R antibody cixutumumab demonstrated only limited overall efficacy [182]. Additionally, a placebo-controlled phase III trial of linsitinib, an oral small molecule of both the IGF1R and insulin receptor, did not demonstrate an improvement in disease-free or OS in patients with advanced ACC [183]. However, partial responses and disease stabilization in a subset of patients were observed with these drugs as single agents [182-184] and in combination with other drugs (such as temsirolimus) [185].

Although we do not offer IGF1R inhibitors in patient with advanced ACC, further studies may determine molecular markers that can predict which patients may benefit from this approach.

●VEGF inhibitors – Vascular endothelial growth factor (VEGF) is upregulated in ACC tumor tissue, and some studies have found that circulating VEGF levels are significantly greater in patients with ACC as compared with those with adrenal adenomas.

While case reports suggested activity of the antiangiogenic agents thalidomide [186,187], sorafenib, and sunitinib [186-189], subsequent phase II trials have demonstrated limited efficacy of VEGF inhibitors both as single agents (eg, axitinib [190], sunitinib [191]), or in combination with chemotherapy (eg, bevacizumab plus capecitabine [192]; sorafenib plus weekly paclitaxel [193]).

●EGFR inhibitors – The finding that over 80 percent of ACCs express the epidermal growth factor receptor (EGFR) [194,195] provides a rationale for the study of agents that target the EGFR. Unfortunately, salvage therapy with the small molecular EGFR inhibitor erlotinib in combination with gemcitabine was of little benefit in a trial of 10 assessable patients with advanced ACC who had failed at least two other systemic chemotherapy regimens [196]. Only one patient experienced a minor response, while eight patients experienced disease progression.

●Radionuclide therapy – ACC can also be targeted with therapeutic radionuclides. Examples of active radionuclides include iodine-131-metomidate [197] and agents that target 11-beta-hydroxylase such as (R)-1-[1-(4-[131-I]iodophenyl)ethyl]-1H-imidazole-5-carboxylic acid azetidinyl ([iodine-131]-IMAZA) [198].

Medical treatment of hormone excess

Hypercortisolism — Adrenal function should be closely monitored in patients with ACC because they can have either adrenal insufficiency (caused by surgery or mitotane) or excess cortisol secretion (caused by persistent or recurrent tumor). In patients with hypercortisolism (either those who are not taking mitotane or those whose hypercortisolism is not controlled by mitotane), addition of a more specific adrenal enzyme inhibitor is often required. Control of hypercortisolism is important as patients can die prematurely of infections related to immunosuppression induced by their hypercortisolism and chemotherapy rather than by tumor burden.

While there have been no published large series to determine which available steroidogenic enzyme inhibitors result in better control, we consider metyrapone as the first drug of choice in patients with ACC and hypercortisolism. While ketoconazole is effective in benign adrenal disease, it is our experience at the University of Michigan that it is rarely able to control the hypercortisolism in ACC.

Metyrapone, which is now more easily available in North America (new distributor HRA Pharma, Paris, France, via its specialty pharmacy Direct Success Inc. with order form available on the web [199] or by phone at 1-855-674-7663), can be very effective in ACC, achieving eucortisolemia within three to seven days. If eucortisolism is not achieved with metyrapone, combination therapy with ketoconazole and mitotane can be utilized. The choice of therapy has been influenced by different drug availability in various countries.

Medical therapy of hypercortisolism is reviewed in detail separately, but summarized briefly below (see "Medical therapy of hypercortisolism (Cushing's syndrome)"):

●For patients with ACC, we suggest metyrapone starting at 250 mg four times daily and increasing up to 6 g per day.

●If metyrapone alone is insufficient or not tolerated, ketoconazole is started at 200 mg three times/day, increasing daily as needed to 400 mg three times/day. Higher doses are seldom more effective.

●Their effect can be assessed within a few days by measuring 24-hour urine cortisol on a frequent basis initially. Combination of both drugs and mitotane may be necessary to achieve adequate control.

●In severe uncontrolled cases, addition of mifepristone, a glucocorticoid receptor antagonist, can be beneficial [200].

●In acute situations with patients who are unable to take drugs orally, intravenous etomidate (which blocks 11-beta-hydroxylase) can be used in a low, nonhypnotic dose of 0.3 mg/kg per hour. However, this drug is difficult to use and requires inpatient monitoring. (See "Medical therapy of hypercortisolism (Cushing's syndrome)".)

Adrenal insufficiency is managed as described above using replacement mainly with hydrocortisone. Aldosterone deficiency is replaced with addition of fludrocortisone (0.1 to 0.3 mg daily) and adjusted to restore normal blood pressure and serum levels of potassium and renin. (See 'Adrenal insufficiency' above.)

For patients who have increased hypercortisolism and are being treated with metyrapone, salt retention and hypertension can occur because of increased production of the mineralocorticoid deoxycorticosterone. We routinely use a mineralocorticoid receptor inhibitor (spironolactone or eplerenone) or amiloride in this situation, adding diuretics and additional antihypertensive drugs if necessary.

When potassium-sparing diuretics are used in association with potassium supplements, frequent serum potassium monitoring is necessary as acute hyperkalemia can occur when hypercortisolism becomes controlled or when renal function is impaired. Adequate control of diabetes and high blood pressure are also important.

Other — A small percentage of adult ACC and a high percentage of pediatric ACC often presents with virilization with/without hypercortisolism. Virilization is best treated with specific androgen blockage with androgen receptor inhibitors (bicalutamide at 50 mg per day) or 5-alpha-reductase inhibition (finasteride at 5 mg per day).

While spironolactone is often used to control androgen effects in women with androgen excess in the setting of benign disease, it is often ineffective in the setting of ACC with very high serum androgen concentrations.

Rare estrogen-producing ACCs are treated with any of the antiestrogen therapies (such as tamoxifen). (See "Management of gynecomastia", section on 'Pharmacologic therapy'.)

PROGNOSIS — Overall, survival is poor for adrenocortical carcinoma (ACC) [16,17,41,48,50,75,86,87,137,201,202]. Five-year survival is approximately 45 to 60 percent for early stage disease and 10 to 25 percent for advanced stage disease [1].

Past results suggested a poor prognosis even for patients with early stage disease [11,17,25]. However, some contemporary series suggest that outcomes are improving and that patients with these tumors are living longer [41,50,87,137], as illustrated by the following data:

●Among 139 adults treated for ACC over a 20-year period at MD Anderson Cancer Center, the five-year survival rate was 60 percent, despite the fact that 47 (34 percent) had distant metastases at diagnosis [50].

●In a second series of 113 patients treated at Memorial Sloan-Kettering Cancer Center (MSKCC), those who underwent complete resection of primary (not recurrent) disease (n = 68) had a five-year survival of 55 percent (median 74 months) [137].

●Patients followed closely after surgery for stage II ACC in specialized centers and who received adjuvant mitotane had a much better prognosis than previously reported for this stage [73]. (See 'Adjuvant mitotane' above.)

●The improved prognosis of treated patients in contemporary series can be illustrated by the previously described French Association of Endocrine Surgery series of 253 patients (age range 5 to 81 years, mean 47 years) with ACC treated between 1978 and 1997 [41]. The distribution by stage was as follows: stage I, 16; stage II, 126; stage III (locoregionally advanced disease), 57; stage IV (distant metastases only), 54 patients. Potentially curative surgery was performed in 182 (72 percent), and postoperative mitotane was administered to 135 (54 percent) patients.

Five-year survival rates were 38 percent overall and 50 percent in the curatively treated group. Stratified by disease stage, five-year survival rates were 66, 58, 24, and 0 percent for stage I, II, III, and IV (metastatic) disease, respectively. Despite no significant difference in the numbers of patients undergoing potentially curative surgery, outcomes were significantly better among those treated after 1988.

The reasons for the improved outcomes from treatment of ACC over time are not clear. Although the use of mitotane increased over the 20-year time period, this had a significant impact on prognosis in multivariate analysis only for those patients who did not undergo potentially curative resection.

SPECIAL POPULATIONS

Pediatric patients — Pediatric patients with adrenocortical carcinoma (ACC) differ from adults in clinicopathologic characteristics, prognosis, and management.

Clinicopathologic characteristics

●Familial cancer syndromes – Most children with adrenocortical tumors have associated familial cancer syndromes (mainly Beckwith-Wiedemann syndrome and Li-Fraumeni syndrome) [203]. (See "Clinical presentation and evaluation of adrenocortical tumors", section on 'Hereditary cancer syndromes' and "Li-Fraumeni syndrome" and "Beckwith-Wiedemann syndrome".)

●Histopathology – Unlike adults, severe histopathologic features in children, such as high Weiss score, are not reliable predictors of poor prognosis [62]. (See "Clinical presentation and evaluation of adrenocortical tumors", section on 'Fine-needle aspiration biopsy'.)

●Germline mutations in TP53 – Approximately 50 percent or more of pediatric patients with ACCs in North America have germline mutations in the TP53 gene (which encodes the p53 tumor protein) [204-206]. In contrast, approximately 8 percent of adults have these mutations [207]. Loss of heterozygosity for 17p13 (the cytogenetic location of the TP53 gene) occurs in 25 to 70 percent of sporadic ACCs [207-210].

Prognosis — The prognosis in children who have ACC appears to be better than that of adults, especially those for those with early-stage disease [42,202,211-214].

●In an observational series of 228 children and adolescents under age 20 with ACC, the estimated five-year event-free survival (EFS) rate was 54 percent overall; and for those with stage I or II disease, 91 and 53 percent, respectively [42]. Stage I disease (completely excised nonmetastatic tumors ≤200 g), presenting signs of virilization alone, and age less than three years were all associated with significantly better survival. In contrast, the prognosis was poor for patients with metastatic or residual disease or larger resected tumors (weighing more than 200 grams).

●Similar results were seen in an analysis of 85 patients with ACC under age 20 years reported to the National Cancer Institute (NCI) Surveillance, Epidemiology, and End Results (SEER) database between 1973 and 2008 [215]. Younger patients (≤4 years old) were more likely to have favorable features than older patients (4 to 19 years), including local disease (76 versus 31 percent), tumor size <10 cm (69 versus 31 percent), and better five-year survival (91 versus 30 percent). In multivariate analysis, the most significant independent predictors of disease-specific death were age greater than four years and distant disease. After accounting for tumor size, the only significant predictor of cancer-specific death was age 5 to 19 years.

Management — Because the clinicopathologic features and prognosis of ACC in pediatric patients differs from that of adults, they should be managed at a center with multidisciplinary expertise in pediatric ACC when possible [205]. An International Pediatric Adrenocortical Tumor Registry has been established to facilitate data collection, study, and treatment of these rare pediatric tumors [212].

There are limited prospective studies evaluating management strategies for pediatric patients with ACC.

A prospective, single-arm study from the Children's Oncology Group (ARAR0332) reported the survival outcomes of 78 pediatric patients with ACC who received different treatment strategies based on disease stage as follows [206]:

●Stage I disease – Adrenalectomy alone

●Stage II disease – Adrenalectomy and retroperitoneal lymph node dissection

●Stage III or IV (metastatic) disease – Mitotane plus chemotherapy (etoposide, doxorubicin, and cisplatin [EDP]), followed by surgery of the primary tumor and metastases as clinically indicated

At median follow-up of 60 months, the five-year EFS and overall survival (OS) results were:

●All patients – 63 and 77 percent

●Stage I – 86 and 95 percent

●Stage II – 53 and 79 percent

●Stage III – 81 and 95 percent

●Stage IV – 7 and 16 percent

Approximately one-third (32 percent) of patients receiving mitotane plus chemotherapy experienced significant treatment-related toxicity and could not complete scheduled therapy. Thus, although patients without metastatic disease who received chemotherapy (stage III disease) demonstrated improved survival compared with those treated with adrenalectomy and lymph node dissection alone (stage II disease), further treatment modifications may be necessary to improve chemotherapy tolerance. This difference in survival also suggests that lymph node dissection may not have been adequate to remove all micrometastases. Other possible explanations include increased efficacy of chemotherapy for treating micrometastases in stage III tumors, or more aggressive biology in stage II tumors.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Adrenal cancer" and "Society guideline links: Adrenal incidentaloma".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Adrenal cancer (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Clinical presentation – Adrenocortical carcinoma (ACC) is a rare but aggressive malignancy that can cause Cushing's syndrome, virilization, hyperaldosteronism, feminization, an abdominal mass, or no symptoms and be discovered incidentally. (See "Clinical presentation and evaluation of adrenocortical tumors".)

●Prognosis – Past results suggest a poor prognosis even for patients with early stage adrenal cancer. However, some contemporary series suggest that patients with ACC are living longer. The reasons for the possible improvement in prognosis are unclear, but adjuvant mitotane therapy appears to be beneficial. (See 'Prognosis' above.)

●Initial treatment, potentially resectable disease – The only potentially curative treatment for stage I to III ACC is surgical resection, and it is the initial approach, even if a complete resection cannot be achieved. (See 'Initial surgery' above.)

•For patients with stage I to III adrenocortical cancer who are surgical candidates, we suggest open rather than laparoscopic surgical resection, regardless of tumor size (Grade 2C).

•Suspicious lymph nodes should be resected, but the benefit of routine lymphadenectomy has not been fully established yet.

●Approach to adjuvant systemic therapy – Although resection is technically possible in most patients with stage I to III disease, it is not curative for many, presumably because occult micrometastases are present at the time of initial presentation. For all patients with completely resected ACC, our approach to adjuvant therapy is based upon risk of disease recurrence:

•Low recurrence risk – For most patients at low risk of recurrence after complete resection (stage I to III, microscopically complete [R0] resection, Ki-67 ≤10 percent), we suggest observation rather than adjuvant mitotane (Grade 2C) as the addition of adjuvant mitotane failed to demonstrate statistically significant improvements in recurrence-free survival (RFS) or overall survival (OS) in a randomized trial. (See 'Patients with low recurrence risk' above.)

We do offer adjuvant mitotane to select patients who meet criteria for low-risk disease but also harbor some features that suggest a potential risk for recurrence. Examples include patients with stage III low-grade disease, or those with a large (>20 cm) stage II tumor with a borderline mitotic rate or Ki-67 score.

•High recurrence risk – For those at high risk of disease recurrence after complete resection, we suggest adjuvant mitotane therapy alone (without chemotherapy) rather than observation (Grade 2C). Clinical features that indicate a high risk of disease recurrence include high-grade disease (Ki-67 >10 percent and <20 percent or mitotic rate greater than 20 per 50 high-power fields [HPF]); incompletely resected disease; intraoperative tumor spillage or fracture; and large, low-grade tumors with vascular or capsular invasion. (See 'Patients with high recurrence risk' above and 'Adjuvant mitotane' above.)

•Very high recurrence risk – For select patients at very high risk for an early recurrence (eg, very high Ki-67 staining (≥20 percent) and extensive vascular invasion/vena cava thrombus), we typically suggest the addition of cisplatin-based chemotherapy to mitotane (Grade 2C), although high-quality data are limited. (See 'Patients with very high recurrence risk' above.)

•Indications for adjuvant radiation therapy – We suggest the addition of adjuvant radiation therapy (RT) for all patients with incompletely resected ACC, stage III disease with other characteristics of high recurrence risk, those who have tumor spillage at the time of resection, and for all patients with high-grade ACC (>20 mitotic figures per 50 HPF) (Grade 2C). (See 'Adjuvant radiation therapy' above.)

There is disagreement as to the role of adjuvant RT in patients who have undergone laparoscopic resection for an ACC that otherwise does not meet criteria for postoperative RT. (See 'Radiation therapy' above.)

●Duration of adjuvant mitotane – The optimal duration of adjuvant mitotane therapy is not established and is based on the risk of disease recurrence, tolerance, and the maximal length of time the patient has achieved target therapeutic mitotane levels. (See 'Duration' above.)

•For most patients with resected low-risk ACC, observation is preferred. For those with indications for adjuvant therapy, we aim for a minimum of two years of mitotane.

•For patients with resected high-risk ACC, we treat with three to five years of adjuvant therapy.

●Monitoring for mitotane toxicities

•Target mitotane levels – If available, serum monitoring during mitotane therapy should be used to achieve therapeutic levels and avoid toxicity. Mitotane levels should be maintained between 14 and 20 mcg/mL. (See 'Suggested regimen' above.)

•Glucocorticoid replacement for adrenal insufficiency – Glucocorticoid replacement is necessary following resection of cortisol-secreting ACC. It should also be started in patients with non-cortisol-secreting tumors when adjuvant mitotane therapy is initiated because the drug's adrenolytic activity will result in adrenal insufficiency in most patients. We use hydrocortisone rather than dexamethasone for glucocorticoid coverage. Supraphysiologic doses become indicated because mitotane progressively accelerates cortisol metabolism by induction of the P450 system. (See 'Adrenal insufficiency' above.)

•Treatment of aldosterone deficiency – Mitotane can also eventually cause aldosterone deficiency. We suggest monitoring blood pressure at each visit, serum potassium at every three months, and plasma renin every six months (table 3). When clinical and biochemical signs of aldosterone deficiency become present, addition of fludrocortisone (0.1 to 0.3 mg daily) is initiated and adjusted to restore normal clinical and biochemical parameters. (See 'Adrenal insufficiency' above.)

●Post-treatment surveillance

•Imaging – Post-treatment surveillance includes computed tomography (CT) scans of the chest and abdomen every three months for two years, then every six months for five years. The benefit of fluorodeoxyglucose (FDG)-positron emission tomography (PET) in the post-treatment surveillance strategy is controversial, and practice is variable. (See 'Imaging' above.)

•Biochemical monitoring – For patients with a completely resected, steroid-producing ACC, some groups monitor patients every three months for two years with steroid tumor markers such as cortisol (measured in the morning before hydrocortisone dose), dehydroepiandrosterone sulfate (DHEAS), androstenedione, testosterone, estradiol, or mineralocorticoid based on the steroid profile in the initial tumor. Other groups do not routinely screen for recurrent hormone excess unless or until there is evidence of recurrent disease. (See 'Biochemical' above.)

●Unresectable, recurrent, or advanced disease – For patients with isolated, locally recurrent ACC that is surgically accessible, we suggest complete surgical resection followed by mitotane therapy (Grade 2C) (see 'Local therapy' above). Resection may also be considered in the rare patient who presents with limited, potentially resectable hepatic or pulmonary metastases and for selected patients with uncontrollable symptomatic hormone excess. (See 'Local therapy' above.)

•For most patients with unresectable disease, we suggest combined mitotane plus chemotherapy rather than mitotane or chemotherapy alone (Grade 2C). However, mitotane monotherapy is a reasonable option for patients with a limited burden of low-grade, slowly progressive disease. (See 'Mitotane monotherapy' above.)

For patients receiving combined therapy, we suggest mitotane in combination with etoposide, doxorubicin, and cisplatin (EDP) rather than streptozotocin (Grade 2C). If possible, patients should be referred to centers conducting randomized clinical trials. (See 'Chemotherapy plus mitotane' above.)

●Indications for radiofrequency ablation – Percutaneous radiofrequency ablation (RFA) may provide short-term local control of an unresectable primary tumor, particularly for those <5 cm in diameter. (See 'Radiofrequency ablation' above.)

●Medical treatment of hormone excess – In patients with hypercortisolism (who are either not taking mitotane or whose hypercortisolism is not controlled by mitotane), we suggest metyrapone or, if necessary, its combined use with ketoconazole and mitotane to achieve control of hypercortisolism. Appropriate control of glucose, potassium levels, blood pressure, and infections are also important. (See 'Medical treatment of hormone excess' above.)

●Indications for palliative RT – For palliation of symptoms from locally advanced or distant metastatic disease (eg, a bone metastasis), we use palliative RT. (See 'Radiation therapy' above.)

●Investigational approaches – Progress in treating advanced ACC has been severely limited by the rarity of the disease. However, multicenter clinical trials for recurrent and advanced disease, several of which are studying molecularly targeted therapy, are in progress. (See 'Newer and investigational approaches' above.)

DISCLOSURE — The views expressed in this topic are those of the author(s) and do not reflect the official views or policy of the United States Government or its components.