Adrenal cortical carcinoma: Paediatric aspects – literature review

Introduction

Adrenocortical carcinoma (ACC) is a rare malignant tumour. In children, it accounts for 0.6% of tumours and 0.2% of malignant neoplasms. It accounts for 5% of all adrenocortical tumours (ACTs). There are 2 peaks of incidence: in the first decade of life and between 40 and 50 years of age [1–3]. Based on an analysis of the literature on adrenocortical carcinoma, this article presents the diagnosis, treatment modalities, genetic background, and prognostic factors of ACC in children.

Epidemiology

Only about 20% of adrenocortical tumours in children (ACTs) are classified as adrenocortical adenomas (ACAs), and they have an excellent prognosis [4]. Adrenocortical carcinomas account for 0.2% of malignant tumours in children. 0.2–0.3 cases/million/year in children are described; in adults, the incidence is 3–7 times higher (i.e. 0.7–2 cases/million/year). Most cases of ACC in children occur between the ages of 0 and 4 years [1, 2, 3, 5]. ACC can be familial or sporadic. It is diagnosed in 10% of patients with Li-Fraumeni syndrome, usually caused by a mutation in the TP53 gene. This is the most common hereditary cause of ACC. In adults, most ACCs are sporadic [6].

According to the International Paediatric Adrenocortical Tumour Registry (IPACTR), the average age at diagnosis of ACT was 3.2 years. Less than 15% of patients were older than 13 years at diagnosis. The prevalence of ACT was higher among girls than boys, with an overall female : male ratio of 1.6 : 1. However, the ratio ranged from 1.7 : 1 in the 0–3-year-old group to 6.2 : 1 in the adolescent (≥ 13-year-old) group [3]. Based on data from the SEER study involving paediatric ACC patients (registry of 85 patients), 52% were TNM stage I at the time of diagnosis [7]. Approximately 76% were 4 years of age and younger, compared to 31% of those older than 4 years. In addition, the same study showed a significantly larger tumour size in older patients. Metastases are found in about 25% of children with ACC [8].

Symptoms

Paediatric ACCs are almost always functional (approximately 95% of tumours show hormone production), compared to less than 50% of adult ACCs. Consequently, diagnosis in children is usually made several months after the presentation of clinical symptoms [5, 8].

According to IPACTR data, the most common manifestation of ACC in children is virilisation, which occurs in 60–80% of cases. They are manifested by the following: premature pubarche, acne, clitoral/penile enlargement, voice change, hirsutism, muscle hypertrophy, accelerated growth, and bone age. Virilisation may be seen alone or accompanied by clinical signs of overproduction of other adrenal cortical hormones, including glucocorticoids (15–40%), aldosterone (1–4%), or oestrogen (7%). Overproduction of glucocorticoids alone (Cushing’s syndrome) occurs in 5–8% of cases. It is relatively rare in young children and is more common in adolescents. Primary hyperaldosteronism (Conn’s syndrome) and pure feminisation are very rare, occurring in less than 1% of patients. Symptoms of a tumour mass such as abdominal pain, anorexia, nausea, spasms, and a palpable tumour occur in 5–8% of patients. Less than 10% of children with ACC have non-functional tumours; these are rare among young children and usually occur in adolescents. Acute abdomen due to spontaneous tumour rupture is a rare clinical manifestation of ACT [3, 8]. The most common symptoms in a given tumour type are shown in Table I.

Diagnostics

Criteria for diagnosis

It is recommended that all patients with suspected ACC be discussed at a meeting of a multidisciplinary expert team that should include an endocrinologist, oncologist, pathomorphologist, radiologist, and surgeon [6].

The diagnosis of ACC is made on the basis of histological examination of surgically obtained tissue. Adrenal cortical carcinoma is a malignant epithelial tumour originating from adrenal cortical cells. The diagnosis of adrenal cortical carcinoma is based on invasive growth, increased proliferative activity, and a combination of cytological and architectural features. Mitotic count > 20 mitoses/50 high-power fields (HPF) defines “high-grade ACC”, which has the worst prognosis [5]. Multiple systems have been proposed for pathologic distinction between benign and malignant adrenal cortical tumours (Lawrence Weiss, modified Weiss scale, Wieneke index, and van Slooten index). The modified Weiss scale and Wieneke index are the most commonly used in paediatric practice (Table IV) [13, 14]. A study at the Mayo Clinic between 1950 and 2017 analysed 41 children with ACT and found that the Weiss criteria and modified Weiss criteria were less accurate than the Wieneke index in younger paediatric patients (< 12 years old) [5]. This index classifies ACT into 3 prognostic groups: benign (< 3 points), malignant (> 3 points), and of indeterminate malignant potential (3 points), based on the number of pathological criteria present [4, 14] (Table IV). Tumour stage is most often determined using the ENSAT (European Network for the Study of Adrenal Tumours) classification (Table III). The ACC staging system proposed by Sandrini et al. (1997) is also used in children, later modified by the Children’s Oncology Group (2021) based on clinical data from IPACTR (Table III). The postoperative staging system is based on the extent of resection, tumour size, regional lymph node involvement, and the presence of metastases. It attempts to divide patients by prognosis to identify those who should receive more intensive therapy. However, the proposed system does not take into account age, which is known to be an important prognostic factor in ACC [4, 5]. A pathology report with suspected ACC should include at least the following information: Weiss score/Wieneke index, Ki67 index, resection status, tumour stage (infiltration/no infiltration of the capsule and/or surrounding tissues and organs), and lymph node status [6].

Immunohistochemical examination is also important. Studies assessing Ki67 immunohistochemistry in adrenal cortical carcinomas have shown that most carcinomas have a Ki67 labelling index > 5%. Recent evidence suggests that a Ki67 > 15% is a reliable biomarker to predict poor outcome in adults and paediatric tumours. Ki67 can also help rationalise the need for adjuvant mitotane therapy. Ki67 has been shown to be unevenly distributed in the tumour. Therefore, whole tumour labelling should be performed [6]. To differentiate between a tumour originating from the adrenal cortex and those originating from other tissues, immunohistochemistry for steroidogenic factor-1 (SF1) has the highest sensitivity and specificity – respectively: 98% and 100% [6]. The greatest use in differentiating between tumours of cortical and medullary origin is chromogranin A, which is almost always negative in cortical tumours and shows positivity in 100% of pheochromocytoma tumours. Other neuroendocrine markers (synaptophysin, NSE) are less applicable; although they almost always show a positive reaction in tumours of spinal origin, they are also positive in a significant proportion of cortical tumours (e.g. synaptophysin gives a positive reaction in 67% of cortical tumours). Markers typical of adrenal cortical tumours can show a positive reaction in a small group of pheochromocytoma tumours, for example, inhibin and melan A in 6%, and calretinin in 14% of pheochromocytoma tumours.

Treatment

Treatment of ACT in children is based on adult guidelines. The mainstay of treatment is surgery. Open surgery and complete en bloc resection is the standard surgical approach for a confirmed or highly suspected ACC tumour. In cases of renal invasion, partial nephrectomy should be considered. ACC surgery requires experience in both adrenal and oncological surgery. Several studies have shown that patients operated on by surgeons who perform more than 6–7 such operations per year have shorter hospital stays and fewer complications. The goal of surgery should be to achieve a negative margin, i.e. resection of the R0 tumour [5, 6].

The results of risk stratification in 77 patients showed that patients with complete resection have an excellent prognosis for stage I ACC (86.2% 5-year event-free survival). Due to the fragility of the tumour, there is a high risk of tumour rupture, this occurred in 21% of patients with one resection and in 43% with resection for locoregional recurrence. Hence, laparotomy is recommended, because laparoscopic resections carry a high risk of tumour rupture [8]. Consequently, tumour biopsy should also be avoided, although it does not seem to affect survival in adults [4].

According to ESMO-EURACAN (European Reference Network for Rare Adult Solid Cancers), open surgery is also the standard treatment, but in small tumours (< 6 cm) without features of local invasion; laparoscopic adrenalectomy (respecting the principles of oncological surgery) can be considered if the surgeon has sufficient experience in this type of surgery. Routine locoregional lymph node removal is recommended for highly suspicious ACC tumours. This should include (as a minimum) the adrenal and renal lymph nodes and all suspicious or enlarged lymph nodes identified on preoperative or intraoperative imaging. To determine the absence of lymph node metastases, at least 4 lymph nodes should be collected [6]. In patients with hypercortisolaemia, surgery should be performed under hydrocortisone cover. The decision about follow-up treatment is made on a case-by-case basis. Based on the results of a study presented by the German GPOH-MET 97 (German Society for Paediatric Oncology and Haematology Malignant Endocrine Tumours Study), which used 2 different treatment regimens in combination with mitotane, it was decided that the recommended first-line regimen is either cisplatin, etoposide and doxorubicin (CED) or vincristine, ifosfamide and doxorubicin, or carboplatin and etoposide [4]. The aforementioned chemotherapy is also used as first-line treatment in initially inoperable cases [3].

Mitotane (1,1-dichloro-2-[ochlorophenyl]-2-[p-chlorophenyl]ethane or o, p’-DDD) is an insecticide derivative that causes necrosis of the adrenal cortex. It is currently the only targeted therapy for ACC. With the use of mitotane alone, an objective response has been achieved in approximately 20–30% of patients [4]. The routine use of cytotoxic drugs as an adjuvant treatment is not recommended. However, they should be considered in selected patients with a very high risk of recurrence (e.g. Ki67 > 30%, large tumour thrombus in the vena cava, stage IV or R1 resection) [6].

Once the decision has been made to treat mitotane, it is recommended that treatment be started as soon as possible. It is not recommended that treatment be started later than 3 months after surgery. However, no data are available to demonstrate the superiority of early treatment start or lack of efficacy if treatment is started later than 3 months after surgery. In adult patients without recurrence, who tolerate mitotane acceptably, it is recommended to give it for at least 1–2 years, but no longer than 5 years [6]. However, little information is available on the use of mitotane in children. Response rates appear to be the same as in adults. In the non-randomised, single-arm GPOH-MET 97 study, paediatric patients treated with mitotane for longer than 6 months with serum mitotane levels greater than 14 mg/l had significantly better survival [5]. The 2021 EXPeRT/PARTNER paediatric recommendations suggest treatment with mitotane for 2 years, during the period of highest risk of recurrence. However, it should be noted that only periods with adequate therapeutic plasma concentrations should be considered an effective treatment duration [4]. Responses to mitotane are achieved in 15–60% of treated adult patients. Megerle and Herrmann analysed a group of 127 adult patients and achieved an objective response to treatment in 26 (20.5%), 3 of whom had a complete remission of the disease [15]. In the paediatric population, ACC has been shown to improve outcomes in grades III and IV, although it is poorly tolerated. In a review of 11 children with advanced ACC treated with mitotane and cisplatin, 7 patients showed a measurable response, suggesting that it may be considered in patients in whom complete surgical resection is not possible [8]. Terzolo et al. conducted a retrospective analysis of 177 adult patients with stage ACC I and II, who initially had complete tumour resection and then received complementary mitotane therapy. They showed a significant increase in recurrence-free survival (EFS 50% vs. about 15%) [4]. In 2022, the results of the ADIUVO trial were published. This is a randomised controlled trial evaluating the efficacy of adjuvant mitotane treatment in adult patients with low intermediate risk of ACC recurrence (ACC stage I-III, R0 surgery and Ki67 ≤ 10%). The results of the ADIUVO trial do not support the routine use of adjuvant mitotane in this subgroup of patients, allowing them to avoid potentially toxic treatment [16]. Mitotane in adults is started at a low dose (1 g/day, gradually increased to 2–3 g/day) or high dose with a median maximum dose of 7.5 g/day [15, 17]. Treatment in children should be started at a dose of 1.5 g/m²/day and then gradually increased to 4 g/m²/day. The plasma concentration of mitotane should initially be monitored every 2 weeks or so until a therapeutic concentration is reached, then monthly. The dose of the drug is determined under the control of blood concentration. Its monitoring (therapeutic reference values are 14–20 mg/l) is extremely important because a good response to treatment depends on the level of the drug concentration in the plasma. Such a concentration is usually achieved within 3 to 5 months of treatment. Food and fats increase the absorption of mitotane. The drug penetrates most tissues in the body but is stored to the greatest extent in adipose tissue. Therefore, the dose of the drug and the time it takes to reach a therapeutic concentration is influenced by the body’s fat content. After cessation of administration, the drug continues to be released from storage in adipose tissue for several more months. Its plasma half-life is between 18 and 159 days [15, 17, 18]. Adverse effects usually occur at toxic levels of the drug in the blood (above 20 mg/l). These include the following: nervous system toxicity (lethargy, somnolence, dizziness, vertigo polyneuropathy, neuropsychological delay, developmental regression), gastro and hepatic disturbances (anorexia, vomiting, diarrhoea, hepatitis, increased serum cholesterol, or liver enzyme elevation), and hypertension, or growth retardation. It may also cause secondary hypothyroidism requiring the administration of levothyroxine preparations, probably by direct action on the pituitary gland or by altering thyroid hormone metabolism. Therefore, it is recommended that thyroid hormones be monitored every 3 months [6, 19, 20]. Features of premature central maturation are also observed. Another important disadvantage of mitotane treatment is that it significantly alters steroid hormone metabolism. Testing of serum androgens and their metabolites in the ORD cannot then be used as markers of tumour recurrence. Mitotane treatment also induces a state of functional adrenal insufficiency, which requires steroid hormone replacement. Usually, twice the standard replacement dose is required, but some patients require up to 4–5-fold higher dose. Some patients develop symptoms of mineralocorticoid deficiency (hyperkalaemia, hyponatraemia, hypotension, depressed mood) despite a full hydrocortisone replacement dose. In these patients, fludrocortisone should be added. Before deciding to start treatment, it is recommended that the following be determined: electrolytes and renin [6]. The use of radiotherapy in children with ACC has not been widely studied. Because most paediatric ACC cases have TP53 mutations, which predispose to other cancers, radiotherapy may increase the likelihood of secondary tumour development [8].

New treatments are under investigation: pembrolizumab (a molecular therapy targeting the programmed death receptor 1 [PD-1] on lymphocytes), cabozantinib (a tyrosine kinase inhibitor), cabozantinib-s-malate, and autologous T-cell chimeric antigen receptor-specific B7-H3 (CAR-T) [8].

Genetic factors

Predisposing genetic factors are found in most children and adolescents with ACT. The most common is a germline mutation of the TP53 gene. It is observed along with accumulation of p53 protein, which can be detected by immunohistochemistry. ACC-associated changes in the TP53 gene occur within the following exons: 5–8 (about 80–90%) and 4 and 10 (< 10% combined) [21, 22]. Most of them are sense change mutations (almost 70%). The c.1010G > A (p.R337H) mutation in exon 10 of the TP53 gene occurs in 13% of residents of southern Brazil (mainly the state of Paraná) with a family history of multiple cancers. This represents about 0.27% of the southern Brazilian population and is due to the occurrence of the founder effect. Consequently, the incidence of ACC in children there is 10–15-fold higher than in other parts of the world [3]. It has been found in more than 90% of paediatric patients with ACT in southern and southeastern Brazil. In southern Brazil, there is a newborn screening program for the R337H mutation, which has been successful in the early detection of ACC in this population. In addition, the p.R337H mutation also predisposes to adult ACC in this population [5, 8]. Mutations in the TP53 gene are present in most patients with Li-Fraumeni syndrome (LFS) (about 80%) or Li-Fraumeni-like (LFL, Li-Fraumeni-like) (about 40%) [21, 23]. The presence of mutations in the TP53 gene correlates with larger tumour size, more metastases, and shorter disease-free survival.

Li-Fraumeni syndrome is a rare autosomal dominant cancer predisposition syndrome. The syndrome is characterised by a very high risk (up to more than 85%) of having one or more different types of cancer, in about half of the cases already before the age of 30 years (Table V). ACC develops in less than 10% of LFS patients, mostly in children. 50% to 60% of young children with ACT have TP53 mutations. Pinto, analysing a group of patients with ACC and a confirmed TP53 gene mutation, found that 90% of them developed adrenocortical cancer before the age of 5 years (peak incidence was 1–3 years). After that, the risk of developing ACC decreases and remains low throughout life. In addition, more than 90% of the children studied had virilising or mixed (androgen-secreting and cortisol-secreting) tumours. In comparison, in patients without TP53 gene mutations, these types of secretion occurred in 55% of patients [24]. Up to 5% of adult ACC patients have a known mutation in the TP53 gene and about 3% have a primary diagnosis of Lynch syndrome [6]. These observations recommend that children with ACC should be screened for germline TP53 mutations. Other syndromes with a higher than expected incidence of ACC are shown in Table V.

Prognostic factors

The most important favourable prognostic factor is complete tumour resection. According to the IPACTR data of the 57 patients who had distant or local, residual tumour after surgery, only 8 remained disease-free. Conversely, the long-term survival rate is about 75% for children with completely resected tumours [3]. Children who do not have metastases at the time of diagnosis, i.e. ENSAT stage I and II, also have a better prognosis. Rupture of the tumour capsule during surgery was associated with a poor prognosis, even in patients whose tumours were completely resected. Data from the IPACTR registry also showed that among 192 patients with tumours weighing more than 200 g, the event-free survival (EFS) was 39%, compared to 87% for those with smaller tumours [3]. Two large ENSAT studies have shown that the degree of proliferation assessed by the Ki67 marker is the most significant prognostic marker for both adrenal limited and invasive stages of ACC [25, 26]. Significantly higher Ki67 expression (> 10%) positively correlates with a worse prognosis. According to data presented by Pinto, in children the 3-year progression-free survival (PFS) for 27 patients with Ki67 > 15% was 48.5% compared to 96.2% for 29 patients with Ki-67 < 15% [24]. Molecular alterations are also associated with a worse prognosis. These are much more common in ACC than in ACA. Children, unlike adults, have been shown to have similar levels of IGF2 (insulin-like growth factor 2) in ACC and ACA. IGF1R (insulin-like growth factor 1 [IGF-1] receptor) gene expression, on the other hand, appears to be the primary hallmark of malignant adrenocortical tumours in children. IGF 1R is a trans-membrane receptor that is activated by IGF-1 and IGF-2. In the differential diagnosis of adrenocortical adenoma and adrenocortical carcinoma, the best results were obtained for a combination of IGF 2 and Ki67, with 96% sensitivity and 100% specificity [18, 21]. The Wnt/b-catenin signalling pathway can lead to adrenal cancer through activating mutations of the b-catenin gene (CTNNB1). It occurs almost exclusively in ACC with a poor prognosis [21]. SF-1 overexpression is much more significant in childhood cancers than in adults. In most childhood cancers, the SF-1 copy number is increased [21, 27–29]. Factors associated with poor prognosis are summarised in Table VI. In contrast, increased expression of MHC class II genes, particularly HLA-DPA1, is associated with a better prognosis in paediatric ACT, regardless of histopathological diagnosis [5].

Follow-up examinations

To control patients, laboratory tests will be performed: diurnal rhythm of cortisol, ACTH, androgen levels, steroid profile in the ORD, and imaging studies: abdominal, pelvic (CT/RM) and chest (CT). Patients with unfavourable clinical and/or histological risk factors (advanced stages, age > 4 years and/or unfavourable histological findings) should undergo clinical, hormonal, and imaging evaluation every 3 months for the first 2 years, then every 4 months in year 3, every 6 months in year 4, and annually in year 5. Patients without adverse clinical and histological risk factors (low stages, favourable pathology) should undergo clinical, imaging, and hormonal testing every 4 months in years 1 and 2, and every 6 months in years 3 and 4 [4]. Patients diagnosed with a mutation in the TP53 gene are also required to follow a surveillance protocol for carriers of the gene, including an annual head MRI.

Prognosis

According to IPACTR data, the survival rate after 2 years and 5 months of follow-up was 61.8%, and the 5-year estimated survival time was 54.7% [3]. Other papers report similar values, i.e. 2- and 5-year survival rates of 61% and 46%, respectively [7, 30–33]. According to data presented in the SEER (Surveillance, Epidemiology, and End Results) program established by the National Cancer Institute in the US, the overall 5-year survival rate for patients less than 4 years old was 91.1%. This rate decreases significantly for the group of patients aged 5–19 years and is 29.8% [3, 7, 30]. According to data presented by Pinto, the 3-year progression-free survival (PFS) for children with a confirmed TP53 gene mutation was 38.9%, compared to 79.4% for those with a negative test for the aforementioned mutation [25]. According to data presented by Ilanchezhian, paediatric patients with ACC have a 5-year survival rate of 30% to 70%. Outcomes in patients with metastatic disease are much worse, with a 5-year survival rate of less than 20% [8]. In another series of patients, Lefevre et al. analysed the clinical features and treatment outcomes of 42 children who were treated in French hospitals under the age of 22 years. Long-term survival was ~50%, and tumour size was the most important prognostic factor [5]. In adults, 5-year survival is 60–80% for tumours confined to the adrenal glands, 35–50% for locally advanced disease, and much lower for metastatic disease (0– 28%) [6]. The overall probability of 5-year survival for children with ACT is greater than 80% if they are younger than 4 years and have completely resected tumours weighing less than 200 g without metastasis [5].

Conclusions

All patients with suspected ACC should be discussed at a meeting of a multidisciplinary expert team, which should include an endocrinologist, oncologist, pathomorphologist, radiologist, and surgeon. The most important favourable prognostic factor in ACC is early detection of the disease, at stage I/II, and radical surgical treatment. Surgery for a tumour suspected to be ACC should only be performed by surgeons experienced in adrenal and oncologic surgery with the goal of complete en bloc resection. Rapidly progressive virilisation indicates the need for urgent diagnosis of ACC in children. Genetic testing for TP53 gene mutations from blood and tumour tissue is indicated in patients with ACC. Current treatment algorithms for paediatric ACC are based on studies in adult populations. ACC in children and adults differ in clinical manifestations, histopathology, and molecular alterations, suggesting that tumourigenesis is different in paediatric and adult patients. Due to its rare occurrence and the age of patients at the time of diagnosis, there are limited opportunities to study new treatments in this group of patients. Further prospective studies are needed to clarify the exact pathogenesis of ACT to improve patient outcomes.