Keywords

1.1 Epidemiology

The estimated incidence of adrenocortical carcinoma (ACC) is between 0.5 and 2 new cases per million per year in Western countries. The male/female ratio is 1/1.5 and, according to age, there is a bimodal distribution with two peaks in childhood and between the fourth and fifth decade [1].

1.2 Clinical Presentation

The majority of ACCs, about 50–60%, are functioning at presentation. Cushing’s syndrome (hypercortisolism) or mixed Cushing’s and virilizing syndromes are observed in 50–80% of hormone-secreting ACCs. Instead, pure androgen excess is less frequent and estrogen or mineralocorticoid excesses are very rare [2, 3]. The simultaneous presence of multiple secretions is typical of malignant diseases. Possible syndromes, sign and symptoms are summarized in Table 1.1.

Table 1.1 Syndromes, signs and symptoms related to hormone-secreting adrenocortical carcinomas
Full size table

Many ACC are discovered incidentally. With the use of modern imaging techniques, the so-called adrenal incidentalomas are increasingly detected and, based on the published literature, the frequencies of the different underlying tumor types are adrenocortical adenomas in 80%, ACC in 8%, pheochromocytomas in 7%, and metastatic tumors in 5% [4]. The proportion of malignancy is about 2% if adrenal incidentalomas are less than 4 cm, whereas it increases to 6% if they are 4–6 cm in size and up to 25% among those >6 cm [3, 4].

Moreover, mass symptoms, including abdominal discomfort, nausea, vomiting, abdominal fullness or back pain, are present in about 30–40% of ACC patients at diagnosis [3].

The initial evaluation of a patient with ACC should include physical examination, patient history collection and imaging assessment. In particular, in cases of suspected ACC, an extensive steroid hormone work-up is recommended, assessing gluco-, mineralo-, sex-, and precursor-steroids [3]. In addition, for all adrenal masses, plasma-free or urinary-fractionated metanephrines should be measured in order to exclude pheochromocytoma. The aims of hormonal evaluation are multiple: a differential diagnosis with orientation to the nature of the adrenal mass (pheochromocytoma versus adrenocortical carcinoma); the identification of cases with massive steroid excess requiring specific treatments; the selection of patients with negative prognostic biomarkers (i.e., cortisol hypersecretion) [3].

1.3 Staging and Risk Assessment

The most important prognostic factors in early ACC are disease stage, margin-free resection, age, proliferation marker Ki67, and glucocorticoid excess [5].

Tumor staging is an independent predictor of disease recurrence. Specifically, the presence of metastases is by far the strongest indicator of poor prognosis [6]. In the assessment of disease stage, guidelines recommend the tumor, node, metastasis (TNM) classification proposed by the European Network for the Study of Adrenal Tumors (ENSAT) because this system seems to be superior to others and is adopted by the Union for International Cancer Control (UICC) and World Health Organization (WHO) [3]. According to the ENSAT staging system (Table 1.2), ACC can be classified into 4 groups: stage I (≤5 cm) and stage II (>5 cm) tumors are confined to the adrenal gland; stage III tumors are extended into surrounding tissues (i.e., para-adrenal adipose tissue or adjacent organs) or locoregional lymph nodes; stage IV means that distant metastases are present [3, 6]. In a study from the German ACC registry including 416 ACC patients, the 5-year disease-specific survival rate was 82% for patients with stage I, 58% for stage II, 55% for stage III, and 18% for stage IV ACC patients [7, 8].

Table 1.2 European Network for the Study of Adrenal Tumors (ENSAT) staging system for adrenocortical carcinomas
Full size table

Moreover, to better prognosticate patients with advanced disease, a modified ENSAT stage has been proposed to subclassify patients (mENSAT) [9]. In particular, in this modified classification, stage III includes tumors with invasion of surrounding tissues/organs or the renal/cava vein and stage IV is divided on the basis of number of metastatic organs into IVa, IVb, IVc (2, 3 or >3 metastatic organs, including N, respectively). Libé et al. demonstrated the prognostic value of this subclassification with a 5-year overall survival (OS) of 50%, 15%, 14% and 2% for stages III, IVa, IVb, and IVc, respectively [9].

Margin-free resections (R0) correlate with longer OS and recurrence-free survival (RFS) compared to patients with positive margins (R1) [5]. In the case of R0 resection, in fact, 50.4% of patients are still alive 5 years after surgery whereas the survival rate drops to 23.2% and 10.8% in R1 (not microscopically radical) or R2 (not macroscopically radical), respectively [10]. These data underline the importance of carrying out a radical surgical treatment, with en bloc removal of the tumor with clear margins in all those patients for whom the surgical option is indicated [3, 6].

Age is another independent prognostic factor; older adults usually have a poorer prognosis. This is likely multifactorial and related to increased comorbidities and reduced tolerance to systemic therapy. It is unknown if age by itself is associated with a more aggressive tumor [8]. A retrospective study that included 1579 patients found that the mortality relative risk was 1.51 in patients >55 years old compared with younger ones (95% CI 1.34–1.70) [11].

The Ki67 proliferation index is among the most important prognostic markers in ACC. The largest study, from 2015, looked at 319 German patients and 240 patients from three other European countries and showed that the hazard ratio (HR) of the RFS increased sequentially with the Ki67 index, with 10% and 20% percentage scores correlating with HRs of 1.94 (P = 0.0034) and 2.58 (P = 0.001), respectively [12]. The Ki67 index also correlates with median OS: percentage scores less than 10%, 10–19%, and ≥20% were associated with a median OS of 180.5 months, 113.5 months, and 42 months, respectively [8, 13].

Signs of cortisol excess are prognostically relevant either in terms of RFS or in terms of OS [14]. A meta-analysis of 19 studies found that the relative risk of mortality was 1.54 in hormonally functional tumors compared with hormonally nonfunctional tumors (95% CI 1.28–1.85) and 1.71 in cortisol-secreting tumors compared with non-cortisol secreting tumors (95% CI 1.18–2.47) [8, 15]. Among all types of hormone-secreting tumors, glucocorticoid tumors have the poorest prognosis likely due to the immunosuppressive nature and systemic effects of glucocorticoids. In a recent study, Landwehr et al. found an inverse correlation between excess glucocorticoids and tumor-infiltrating lymphocytes (TILs). The study concluded that patients with excess glucocorticoids and low numbers of TILs had a particularly poor median OS of 27 months, whereas those with sufficient numbers of TILs and no excess glucocorticoids had a median OS of 121 months [8, 16].

At the time of recurrence, the prognostic impact of disease-free interval as well as R0 status was reported in several studies [5].

In patients with metastatic disease the prognosis is generally poor but it is more heterogeneous than previously believed and long-term survivors exist. High tumor burden, high tumor grade, high Ki67 index, and uncontrolled symptoms are major factors associated with worse prognosis in these patients [6].

Different multiparametric scores have been studied in order to differentiate ACC cases on a prognostic basis and, in particular, the GRAS score has been developed. GRAS components are: grading (G, Weiss score >6 and/or Ki67 ≥20%); resection status (R), age (A) and tumor or hormone-related symptoms (S). Its prognostic value was demonstrated first in 444 patients with advanced ACC, defined as stage III or synchronous stage IV disease [9]. In particular, this study confirmed the prognostic impact of the different factors. Moreover, when the GRAS parameters were combined with the mENSAT classification, they were found to best stratify the prognosis of patients with advanced ACC. For example, 5-year OS of stage III patients ranged between 60% and 70% in patients <50 years old with an incidentally discovered ACC or with an R0 status and favorable tumor grading but they dropped to 22% when the tumor grade and the R status were both found to be unfavorable. Five-year OS for patients with stage IVa disease was 15% but ranged from 0% to 55% in patients with favorable or unfavorable GRAS parameters [9].

More recently, a modified form of the GRAS classification, termed mGRAS was proposed, which includes the ENSAT stage, focuses on Ki67 for grading and scores each parameter (Table 1.3). This modified score allows better stratification than individual clinical/histopathological characteristics identifying four subgroups with different clinical outcomes, from a more favorable prognosis (median progression-free survival [PFS] of 54 months) to a worse one (median PFS of 3 months) [17]. These data were confirmed in a large multicenter study which included 942 ACC patients who underwent surgical treatment [18]. However, further studies are needed to confirm the prognostic value of mGRAS in patients with advanced tumors.

Table 1.3 mGRAS score
Full size table

Multiparametric scores are important as they could improve the management of ACC, personalizing the frequency of radiological surveillance, rationalizing the use of adjuvant mitotane after radical surgery and creating tailored strategies for each patient.