ADRENOCORTICAL CARCINOMA IN CHILDREN
Review and Recent Innovations
Louis S. Liou, MD, PhD, and Robert Kay, MD
Pediatric adrenocortical carcinoma is a rare and potentially fatal disease. In contrast to tumor in adults, most pediatric adrenocortical carcinomas are functional. Despite the dra- matic clinical presentation, most patients have a delay in diagnosis that may be attrib- uted to the physician’s unfamiliarity with the disease entity. Although the biochemical and histologic diagnosis have remained largely unchanged, radiographic evaluation has im- proved. In addition, molecular research has revealed new information about the cancer biology of adrenocortical carcinomas, and novel therapeutic approaches could result from these molecular data. Unfortunately, de- spite these advances, treatment of the disease has largely remained unchanged. Total surgi- cal resection of the tumor is still the mainstay of treatment, and new surgical techniques and perioperative management have dimin- ished the morbidity and mortality associated with this procedure. Finally, a proportion of patients present with metastatic disease and while mitotane remains the cornerstone of treatment, a more efficacious and less toxic modality is needed.
This article is an update of the article writ- ten by Chudler and Kay20 which appeared in the August 1989 issue of Urologic Clinics of North America. This article reviews molecular biology, minimally invasive surgery, func- tional imaging, and new chemotherapeutic agents not discussed in the original work. It
is hoped that the information herein will give the reader increased familiarity with this un- usual tumor, leading to earlier recognition, prompt intervention, and improved survival of these patients.
INCIDENCE AND EPIDEMIOLOGY
Adrenocortical carcinoma is a rare tumor with an annual incidence ranging from 0.5 to 2 cases per million population. In contrast, adenomas are found in as many as 2% of all autopsies and are among the most common tumors in humans. There is a bimodal occur- rence, with one peak occurring at age less than 5 years and the other peak in the fourth to fifth decade of life.13 In their review of 87 series of adrenocortical carcinoma encom- passing 1891 adult patients, Wooten and King119 found a slight female predominance (4:3), a slightly higher left-sided incidence (52.8%), a 2.4% occurrence of bilateral tumors, and an overall 59.3% incidence of functional tumors. In contrast, pediatric adrenocortical tumors comprise only 0.2% of all childhood malignancies. Neuroblastomas, ganglioneu- roblastomas, and pheochromocytomas ac- count for the majority of pediatric adrenal gland tumors.122
A review of the literature provides two large groups of pediatric adrenocortical tu- mor populations. The first series comes from
southern Brazil (Curitiba) where the high in- cidence of adrenocortical carcinoma rivals the area’s rates of Wilms’ tumor, neuroblastoma, and non-Hodgkin’s lymphoma.78, 97 The sec- ond group of patients come from the rest of the world. Although there are slight differ- ences, both groups had an occurrence at a median/mean age of less than 5 years (range, birth to 16 years) and a female-to-male ratio of at least 2:1. Hormonally functional tumors were found in 80% to 100% of the patients. Review of the pediatric series revealed ap- proximately a 50:50 right:left laterality, with no series reporting bilateral lesions (Table 1). Other case series have shown either a left- sided21, 50 or right-sided prevalence.49, 105 Bilat- erality has been reported in 2% to 10% of the cases.49, 86 These discrepancies may represent sampling errors secondary to the small num- ber of cases within each series. Cacciari and co-workers17 reported a male predominance (1.5:1) in their series of nine patients, whereas Teinturier and co-workers109 reported an equal ratio of male-to-female patients (27:27) with two age peaks of maximal occurrence at 0 to 4 and 12 to 14 years. Cacciari’s report may suffer from sampling error secondary to the small numbers, whereas Teinturier’s se- ries included a larger number of hereditary- associated disease patients (three hemihyper- trophy and eight Li-Fraumeni cases), which could explain the abnormal sex ratio and the overall reported lower survival.
ETIOLOGY AND MOLECULAR BIOLOGY
The cause of adrenocortical carcinoma is unknown. One theory is that the cancer may arise from preexisting adrenal pathology. In support of this theory are several reports of adrenocortical tumors arising 3 to 36 years after an initial diagnosis of congenital adrenal hyperplasia,5, 62, 87, 89 suggesting that neoplastic transformation of adrenocortical tissue is caused by chronic adrenocorticotropic hor- mone (ACTH) stimulation.87, 89 In contrast, the demonstration of small histologically malig- nant-appearing tumors does not support the malignant transformation of a preexisting ad- enoma but, rather, suggests a second theory that adrenocortical carcinomas arise de novo.
Recent evidence demonstrates that adreno- cortical carcinomas do indeed arise de novo, and that such tumors do not result from pre- existing adrenal pathology. Multiple studies have confirmed that 100% of adrenal carcino-
mas are monoclonal, arising from a single progenitor cell. In contrast, adenomas are found to be monoclonal in 62% of patients, polyclonal in 21%, and ambiguous in 17%.93 If the cause was chronic ACTH stimulation of the adrenal cells, a predominantly polyclonal population of cells would be expected in car- cinomas. Few adenomas and even fewer car- cinomas are found to contain either ACTH receptor mutations or deletions. In addition, the proteins (Gas, Gai) regulating cAMP, the second messenger of the ACTH receptor, are found to be predominantly normal in adeno- mas and carcinomas.93 Unlike hyperfunction- ing thyroid adenomas, 30% of which contain point mutation in the thyroid-stimulating hormone (TSH), adrenocortical carcinomas have a normal cell signaling pathway.91
Because aberrations in the ACTH signaling pathway are not the cause of adrenocortical carcinoma, other candidate genes and pro- teins have been studied, including IGFII, p57, H19, p53, RAS, RB, Ki-67, and DAX-1. IGFII encodes an adrenal growth factor that is over- expressed at the mRNA and protein level in 84% of adrenal carcinomas compared with 6% of adenomas.54 Both receptors, IGF-I and IGF- II, are also present in adrenocortical tumors, allowing the complete cycle of autocrine or paracrine stimulation by IGFII to occur. The IGFII gene locus is at chromosome 11p15 and found to be maternally imprinted (sole ex- pression of the paternal allele). Nearly 80% of adrenal carcinomas show loss of constitutive heterozygosity of the 11p15 region36; however, this change alone is not sufficient to explain the heterogeneous nature of IGFII hypersecre- tion. In addition, IGFII increases are paral- leled by decreases of H19, the putative IGF repressor, and p57KIP2, another genomically imprinted putative tumor suppressor also lo- cated at 11p15. This pattern of expression is also found in androgen-secreting adenomas, whereas normal levels of IGFII, H19, and p57KIP2 are found in cortisol- and aldosterone- secreting adenomas.71 Other growth factors found to be overexpressed in adrenocortical carcinoma are TGF alpha and epidermal growth factor.55, 99
The tumor suppressor gene p53 is most commonly mutated in human malignancy. Such mutation is generally a late event in tumorigenesis, is associated with a malignant phenotype, and occurs within exons 5 to 8 90% of the time. In one study, 28.5% of carci- nomas were found to have p53 mutations, whereas all adenomas had the wild-type gene.76 An inherited germline p53 mutation,
| Study | Number of Patients | Median Age (range) | Functional (%) | Time Delay Prior To Diagnosis (mean) | M:F Ratio (R:L%) | Metastasis at Diagnosis (%) | Percent Overall Survival (follow-up) |
|---|---|---|---|---|---|---|---|
| Hayles et al,47 | 12 | 3 y | 100 | 3 mo-5 y | 1:11 | 8 | 92 |
| USA, 1966 | (1.4-9.5 y) | (NR) | |||||
| Kenney et al,57 | 8 | 3.5 y | 100 | 1-27 mo | 2:6 | 50 | 38 |
| USA, 1968 | (1 d-12 y) | (10.6 mo) | (57:43) | (5 y) | |||
| Burrington and | 8 | 2.3 y | 100 | 2-30 mo | 1:7 | 13 | 38 |
| Stephens,16 | (0.3-4.5 y) | (12 mo) | (43:57) | (2-18 y) | |||
| Canada, 1969 | |||||||
| Stewart et al,105 | 5 | 1.5 y | 100 | 2-18 mo | 1:4 | 20 | 80 |
| England, 1974 | (0.7-11 y) | (median, 8 mo) | (60:40) | (1-12 y) | |||
| Zaitoon and | 7 | 3 y | 86 | 2-24 mo | 1:6 | 29 | 29 |
| Mackie,123 | (1.3-5 y) | (12 mo) | (NR) | ||||
| Canada, 1978 | |||||||
| Kay et al,56 USA, | 5 | 3.5 y | 80 | - | 2:3 | 60 | 40 |
| 1983 | (1.5-16 y) | (60:40) | (2 mo-20 y) | ||||
| Lee et al,65 USA, | 8 | 3.8 y | 100 | 1-24 mo | 3:5 | 13 | 88 |
| 1985 | (2.3-4.9 y) | (11 mo) | (50:50) | (1-6 y) | |||
| Cagle et al,18 | 23 | mean, 5.5 y | 96 | - | 4:19 | 26 | 74 |
| USA, 1986 | (NR) | (2-25 y) | |||||
| Cacciari et al,17 | 9 | 5 y | 100 | 1-36 mo | 6:3 | 0 | 100 |
| Italy, 1986 | (4-9.8 y) | (median, 2 mo) | (44:56) | (2 mo 10 y) | |||
| Hartley et al,45 | 14 | 3.8 y | - | - | 2:12 | - | 57 |
| England, 1987 | (0.7-12.6 y) | (43:57) | (10 mo-19 y) | ||||
| Ribeiro et al,95 | 40 | 3.9 y | 98 | 3 d-60 mo | 12:28 | 13 | 43 |
| Brazil, 1990 | (17 mo) | (NR) | |||||
| Moore et al,84 | 6 | mean, 2.2 y | 83 | 4> 2 y ** | 3:3 | 0 | 83 |
| Australia, 1991 | (0.5-6 y) | (NR) | (1-15 y) | ||||
| Lack et al,63 | 30 | 5 y | 87 | 1-44 mo | 12:18 | 53 | 47 |
| USA, 1992 | (0.5-19 y) | (9 mo) | (60:40) | (1.5-18 y) | |||
| Sabbaga et al,97 | 55 | #41<5 y | 96 | - | 13:42 | 18 | 46 |
| Brazil, 1993 | #31<2 y | (57:43) | |||||
| Federici et al,33 | 12 | mean, 5 y | 100 | - | 5:7 | 8 | 83 |
| Italy, 1994 | (0.5-13 y) | (58:42) | (6.4 y) | ||||
| Mendonca et al,81 | 18 | 2.4 y | 100 | 16 mo | 5:13 | 11 | 89 |
| Brazil, 1995 | (0.7-14.6 y) | (6 mo-7 y)) | (44:56) | (24-114 mo) | |||
| Bergada et al,8 | 20 | mean, 7 y | 100 | 1-30 mo | 5:15 | 5 | 85 |
| Argentina, | (0.4-15.6 y) | (10 mo) | (NR) | ||||
| 1996 | 10<5 y ** | ||||||
| 8>10 y ** | |||||||
| Mayer et al,80 | 11 | 7 y | 100 | 2-60 mo | 3:8 | 9 | 91 |
| Canada, 1997 | (0.7-16y) | (18 mo) | (45:55) | (9 mo-13 y) | |||
| Sandrini et al,98 | 58 | 4.3 y | 93 | 3 d-61 mo | 17:41 | 14 | 50 |
| Brazil, 1997 | (10 mo) | (50:50) | (1-214 mo) | ||||
| Driver et al,28 | 18 | mean, 5.6 y | 89 | 1 d-24 mo | 3:15 | 64 | 33 |
| England, 1998 | (0.7-15 y) | (7 mo) | (56:44) | (114 mo) | |||
| Teinturier et al,109 | 54 | 4 y | 81 | 1-66 mo | 27:27 | 13 | 49 |
| France, 1999 | (0.4-16 y) | (6 mo) | (NR) | (5 y) | |||
| Total/ Average | 412* | 4.1 yt | 94.5 | 10.6 moţ | 128:293 | 21.4 | 63.5 |
| (51.9:48.1) |
NR = not reported.
*Although it is likely that some series contain overlapping patients, confirmation was not possible as patient identity was kept confidential in all series.
** Number of patients less years old.
tIn cases where the median was not reported or was unable to be calculated, the mean was substituted to arrive at the total average. #In cases where the mean was not reported or was unable to be calculated, the median was substituted to arrive at the total average.
termed Li-Fraumeni syndrome, results in multiple primary tumors that primarily con- sist of adrenocortical carcinoma, sarcoma, leu- kemia, lung and laryngeal cancer, and breast and brain tumors (SBLA).77 Vargas and col- leagues114 examined the prognostic role of
p53, Ki-67, and retinoblastoma in adrenal le- sions (hyperplasia, adenomas, adrenocortical carcinoma, and adrenocortical carcinoma me- tastasis). Retinoblastoma was not present in all lesions and was not useful as a marker. Ki-67 (human nuclear proliferation-associated
antigen) and p53 were predictive of malig- nant behavior. None of the benign lesions had p53 mutations and a tumor-proliferating fraction (Ki-67/1000 tumor cells) greater than 80.114 Ras, an oncogene, was not mutated in any adrenal tumors,85 whereas N-ras was found in a small percentage of adenomas and carcinomas.120
The DAX-1 gene encodes an orphan nu- clear hormone receptor essential for normal fetal development of the adrenal cortex. It binds DNA and acts as a transcriptional re- pressor of other nuclear hormone receptors that have a prominent activation role in ste- roidogenesis. DAX-1 levels were found to be high in nonfunctional adenomas and low in aldosterone-producing adenomas and carci- nomas. Intermediate levels of expression were found in cortisol-producing adenomas.94 Although the ACTH signaling pathway, ras oncogene, and RB (retinoblastoma) tumor suppressor do not seem to have a prominent role in adrenocortical carcinoma, other prom- ising targets for diagnosis, prognosis, and therapeutics include IGFII, p57, H19, p53, Ki- 67, and DAX-1.
Instead of using the candidate gene ap- proach, other researchers have examined ab- normalities at a broader chromosomal level. Comparative genomic hybridization studies in adrenocortical carcinomas have revealed complex findings. Some studies report a predominance of gains (increased genetic material or oncogenes),68 whereas others re- port mostly losses (decreased genetic material or loss of a tumor suppressor).60 Most losses are found on 9p, 2, 11q, and 17p (p53 locus), whereas most gains involve 4,5,16p, 20q,7,9q,12q, and 14q. These gains and losses are less frequent in large adenomas (>4.0 cm) and absent in small adenomas (<4.0 cm). These studies demonstrate that adrenal carci- noma involves a multitude of genetic aberra- tions, some of which are not found in other tumors and may be adrenal specific.93
All of the studies described previously have been performed on adult adrenocortical carcinoma, and the literature on molecular characterization of pediatric adrenocortical carcinoma is scarce. A greater incidence of germline p53 mutations has been found in children with adrenocortical carcinoma, but there appears to be no associated Li-Fraumeni family members. Because these children may be the proband, it is appropriate to screen genetically family members of a child diag- nosed with adrenocortical carcinoma.40, 115 Comparative genomic hybridization revealed
a complex hyperdiploid karyotype in one pe- diatric study of adrenocortical carcinoma.82 Another study of nine Brazilian nonfamilial adrenocortical tumors (six carcinomas, three adenomas) revealed a mainly hyperdiploid karyotype with a striking consistent gain of chromosome 9q34 in eight of the nine tumors. In addition, gain of 12q was found in seven of the nine tumors, whereas the main losses involved chromosome 4. Surprisingly, no tu- mors had a loss of chromosome 17 (p53 gene locus).34 These abnormalities may be specific to the Brazilian adrenocortical carcinoma population and may not be similar to the defects found sporadically throughout the world. Additional comparative genomic hy- bridization data on pediatric adrenocortical carcinoma must be accumulated before a final conclusion can be reached.
The molecular information obtained to date and the clinical associations of pediatric adre- nocortical carcinoma increase the possibility of an inherited predisposition. Adrenocortical carcinoma has been reported in immediate family members and first-degree relatives.57, 75 Furthermore, a high incidence of other malig- nancies has been noted in families and the relatives of children with adrenal tumors, and evidence of a familial tendency has been found in Li-Fraumeni, multiple endocrine neoplasia type 1 (MEN-1), Carney’s complex, and Beckwith-Wiedemann syndromes (Table 2). In the series reviewed in Table 1, Li- Fraumeni syndrome was reported in 23 pa- tients, Beckwith-Wiedemann syndrome in 1 patient, and hemihypertrophy in 7 patients.
A relationship exists between hemihyper- trophy and the development of adrenal tu- mors (Fig. 1).56 Hemihypertrophy is found in 1 of 86,000 persons, with a mean age of 4.6 years. Four percent of these patients have associated abdominal tumors or urogenital malformation such as Wilms’ tumor, primary hepatocellular carcinoma, hepatic focal nodu- lar hyperplasia, medullary sponge kidney, and hypospadias. No relationship has been found between the extent and side of hyper- trophy and the severity or the side of tumor, respectively.35, 90 Tumors may develop later in life in patients with congenital hemihypertro- phy, and these patients need close surveil- lance into adulthood.79
Secondary tumors have also been reported in children with adrenocortical tumors. Lev- ine67 demonstrated the late development of other tumors in two patients previously cured of adrenocortical carcinoma, and a review of the literature reveals six cases of adrenocorti-
| Syndrome | Clinical Characteristics | Gene | Chromosome |
|---|---|---|---|
| Li-Fraumeni | Familial susceptibility to cancer (SBLA), sarcoma, leukemia, lung, laryngeal cancer, breast, and brain tumors | p53 | 17p13 |
| Beckwith-Wiedemann | Neonatal macrosomia, macroglossia, omphalocele | IGFII, 11q15, H19, p57KIP2 | 11p15 |
| Carney's complex | Primary pigmented nodular adrenocortical disease (atrial myxomas, swannomas, lentigines, blue nevi of the skin or mucosa) | Carney's locus | 2p16 17q23-24 |
| Multiple endocrine neoplasia type 1 | Hyperparathyroidism, neuroendocrine gut tumors, pituitary adenomas | Menin | 11q13 |
| Congenital adrenal hyperplasia | Female/ male pseudohermaphroditism, cortisol deficiency, mineralocorticoid deficiency or excess | CYP21B CYP11B1 | 6p21 8q24-tel |
Adapted from Reincke M: Mutations in adrenocortical tumors. Horm Metab Res 30:447-455, 1998.
cal carcinoma and secondary tumors, four of which were of central nervous system origin. The coexistence of a second tumor (ganglio- neuroma) with adrenocortical carcinoma has been reported.102 In addition, patients with Li- Fraumeni syndrome (germline p53 mutation)
Rights were not granted to include this figure in electronic media. Please refer to the printed journal.
have a strong personal or family history of neoplasia. Because children with adrenocorti- cal tumors may have a genetic predisposition for additional neoplasms, there should be strict surveillance in addition to thorough family screening.56, 108
Exogenous sources of carcinogenesis have been postulated. Hartley and associates45 ex- amined the epidemiologic data on 14 cases of pediatric adrenocortical tumors in the Man- chester Children’s Tumour Registry from 1954 to 1985. They found that four mothers and seven fathers had been exposed to various dusts and chemicals before or during the in- dex pregnancy. These substances included solvents, tar, chemicals, insulation materials, dyes, cotton and wool dust, and possibly fer- tilizer. Four parents were employed in the rubber or chemical industry.45 An exogenous source has also been postulated in the south Brazilian population. Because the hereditary role of adrenocortical carcinoma in these pa- tients has been discounted and because the geography is highly agricultural in nature, a chemical fertilizer or pesticide origin is specu- lated to be a causative factor.98 Ironically, the major current adjuvant therapy for metastatic adrenocortical carcinoma is mitotane ([o,p’- DDD: 1,1-dichloro-2-(o-chlorophenyl)-2-(p- chlorophenyl)-ethane]), a derivative of an in- dustrial pesticide.58
PATHOLOGY
Differentiation of benign and malignant adrenocortical neoplasms solely on the basis of histopathologic findings is difficult. The
clinicopathologic studies of Hough and Weiss50, 117, 118 have established criteria to dis- tinguish the two adrenal entities in adults; however, many adrenocortical tumors display benign and malignant morphologic character- istics, often precluding an unequivocal classi- fication. Some patients in whom tumors ex- hibited histologically benign features have had late metastases or local recurrence,64, 107, 123 whereas others in whom tumors had a micro- scopic appearance typical of malignancy have survived for years.18, 107, 113 The inconsistency of tumor behavior with histologic findings is even more pronounced in the pediatric popu- lation.84 Despite the variability of pathologic findings, some generalizations can be made with regards to adenoma and carcinoma. First, at diagnosis, adrenocortical carcinomas are larger than their benign counterparts. Ad- enomas tend to be less than 100 g, whereas carcinomas frequently weigh more than 500 g with few less than 200 g.46 Because a solid mass of 100 g has a diameter of at least 6 cm, this value has been used as a guide for intervention in adult nonfunctional adrenal tumors.11 Second, grossly, the cut surface of an adrenal carcinoma is lobulated and varie- gated in color, ranging from tan-yellow to reddish-brown. The tumor may or may not be encapsulated, with zones of necrosis and hemorrhage frequently seen. Areas of calcifi- cation may also be seen (Fig. 2). In contrast, adenomas are typically well encapsulated and more uniform in color, usually yellowish, but less commonly reddish-brown or brown.107 The cut surface is uniform, and zones of hemor- rhage, necrosis, or calcification are usually ab- sent.
Microscopically, adrenocortical carcinoma may closely resemble normal adrenal tissue under low power.105 Under higher magnifica- tion, more typical histologic features are seen, including architectural disarray, frequent mi- toses including bizarre mitotic features, broad fibrous bands, marked cellular pleomor- phism, nuclear atypia, and hyperchromasia.107 Other distinguishing features include micro- scopic foci of necrosis, hemorrhage, and calci- fication, and capsular or vascular invasion. In contrast, the microscopic appearance of ade- nomas includes a well-preserved architecture; less variation in cell size and less nuclear pleomorphism; minimal mitotic activity; the rare presence of necrosis, hemorrhage, or cal- cification; and no evidence of invasion of the tumor capsule or blood vessels. Lack and col- leagues63 concluded that mitotic activity, vas-
A
B
cular invasion, and confluent necrosis were the most reliable microscopic differentiating features between adenomas and carcinomas in pediatric adrenocortical tumors. In their study of the Brazilian population, Bugg and co-workers15 found that histology was the sin- gle most significant prognostic factor, with the proliferative index and ploidy analysis of borderline value. Using almost the same patient population Ribeiro and co-workers95 found that, in addition to histology, size of the tumor was prognostic. Bergada,8 Zai- toon,23 and Federici and their colleagues33 concurred with the prognostic value of tumor size in their pediatric series. Other studies have concluded that these pathologic findings may be misleading, with no correlation of DNA content with histology or clinical fea- tures.25
Although the pathologic distinction be- tween an adrenal adenoma and a carcinoma is possible by applying morphologic criteria, the correlation with clinical outcome is poor59; therefore, it seems prudent to integrate patho- logic, molecular, and clinical data to arrive at a final diagnosis. Based on the pathologic criteria alone, pediatric adrenal neoplasms will be overdiagnosed as carcinomas.
CLINICAL PRESENTATION
Most adrenocortical tumors in children are hormonally active, whereas less than 50% are overtly functional in adults.% In reviewed se- ries, the rate of functional tumors ranges from 81% to 100% (see Table 1). The frequently dramatic clinical presentation caused by ste- roid overproduction makes adrenocortical carcinoma a unique diagnostic entity. The ad- renal cortex produces three major types of steroids: glucocorticoids, mineralocorticoids, and sex steroids (androgens and, to a lesser extent, estrogens). Adrenocortical tumors may autonomously secrete an excess of each type of adrenocorticosteroid, each associated with a particular clinical syndrome. In most instances, a mixed hormonal syndrome is manifested; however, depending on the quan- tity and quality of the steroids produced, one clinical presentation usually predominates.47
Virilization is the most commonly ex- pressed functional syndrome seen in children with adrenocortical tumors.81 In a review of 222 patients in the literature, 66% presented with virilization, whereas the remainder had Cushing’s symptoms.47 The series reviewed in Table 1 report a range of 22% to 100%. Features of androgen excess include axillary and pubic hair; deepening of the voice; acne; a rapid acceleration of height, muscle mass, and bone age; hirsutism; enlargement of the penis or clitoromegaly; and the development of body odor (Fig. 3). Virilization is secondary to hypersecretion of adrenal androgens, in- cluding dehydroepiandrosterone (DHEA), its sulfate (DHEA-S), and androstenedione, which may be converted to testosterone.65
The stigmata of Cushing’s syndrome are present in most of the remaining children with adrenocortical tumors, although many of these patients also display features of viril- ism. Sixty to seventy percent of Cushing’s syndrome cases in adults are the result of endogenous hypercortisolism secondary to pituitary-dependent ACTH overproduction, whereas 80% or more of the cases in children are caused by adrenocortical tumors.105 The classic features of cortisol excess include obe- sity, moon facies, striae, retarded growth rate, muscle wasting, plethora, and generalized hirsutism. Hypertension and impaired glu- cose metabolism may also be found. Truncal obesity may be seen in older children, whereas infants tend to exhibit generalized obesity.49 Linear growth may occur in children with concomitant virilism. Sustained cortisol
hypersecretion with a loss of the normal diur- nal pattern characterizes the pathophysiology. The series reviewed in Table 1 reported Cush- ing’s syndrome symptoms in 5% to 80% of patients, hypertension in 33% to 80%, and advanced bone age or growth parameters in 29% to 100%.
Feminization as a pure functional syn- drome is a rare manifestation of adrenocorti- cal neoplasms, particularly in children. In the normal adrenal gland, small or negligible amounts of estrogens may be secreted. In con- trast, in adrenocortical tumors, overproduc- tion of estrogens, particularly estradiol, may occur.22 Isolated cases of pure feminization have been reported.105 When present in a fe- male, precocious isosexual development oc- curs, with early breast enlargement, acceler- ated growth, and advanced bone age. Vaginal bleeding may also be seen. In the male, bilat- eral gynecomastia is the rule. Accelerated growth rate and delayed pubertal develop- ment may result. Galactorrhea may arise in either sex.29 More commonly, mixed syn- dromes have been reported, exhibiting fea- tures of estrogen and androgen excess.49 Fem-
inization was found in 2% to 25% of the series reviewed in Table 1. In contrast to the omi- nous connotations when isolated feminization is found in adults, the limited pediatric expe- rience shows that 66% of children with femi- nization survive.63
Nonfunctional tumors are infrequent in children. Only about 5% of childhood adreno- cortical tumors produce no clinical evidence of endocrine hyperactivity. Because no symp- toms are exhibited, these patients tend to present with abdominal pain or fullness only after the tumor has attained substantial size or metastasized. A palpable abdominal mass was present in 4% to 79% of the patients reviewed in Table 1. Unlike most other pedi- atric cancers in which a mass is frequently the first or even only indication of malignancy, an abdominal mass is an uncommon presenting complaint of adrenocortical tumor.56 Non- functional tumors occur more commonly in males and, in general, portend a grave prog- nosis owing to the advanced tumor stage at diagnosis.86
Primary hyperaldosteronism in children is usually secondary to adrenocortical hyperpla- sia, and only a few cases of pure primary aldosteronism caused by adrenocortical tu- mor have been reported. Almost all of these lesions were adenomas.116
Symptoms related to metastatic disease rarely lead the family to seek medical atten- tion. Initial metastases have been found in 5% to 64% of patients in separate published studies (Table 1). Metastases most frequently involve the lungs and liver and less com- monly the regional lymph nodes, inferior vena cava, brain, diaphragm, or bone.28, 63
BIOCHEMICAL DIAGNOSIS
The adrenal gland normally secretes mod- est amounts of DHEA, DHEA-S, and andro- stenedione and minimal testosterone. In con- trast, in pathologic conditions such as adrenocortical neoplasia, testosterone and its precursors may be produced in significant amounts, resulting in virilization. Evaluation should therefore be directed toward detection of elevated androgens. This screening should include measurement of plasma testosterone, urinary and plasma DHEA and its sulfate, and urinary 17-ketosteroids, a nonspecific measure of androgenic metabolites. Normally, two thirds of the 17-ketosteroids originate from the adrenal gland and one third from
the gonad. Although the most specific assess- ment of adrenal androgen production is plasma DHEA-S, elevation of urinary 17-keto- steroids almost always indicates adrenal hy- peractivity. Moreover, 17-ketosteroids are usually markedly elevated in the case of ma- lignancy and less so in benign adrenal le- sions.20, 49, 86, 96
The clinical presentation of Cushing’s syn- drome is confirmed by the demonstration of hypercortisolism and the loss of diurnal vari- ation. Cortisol excess may be evaluated by assays for plasma cortisol, urinary 17-hydro- xycorticosteroids, and urinary free cortisol. Urinary 17-hydroxycorticosteroids, which measure metabolites of cortisol, are usually elevated in patients with increased cortisol secretion and in patients with adrenocortical carcinoma in whom virilization is the pre- dominant clinical feature. As is true for 17- ketosteroids, adrenal malignancies generally cause far greater elevations of 17-hydroxycor- ticosteroids and plasma cortisol than seen in benign disease. More recently, measurement of urinary free cortisol (which reflects the un- bound biologically active plasma cortisol fil- tered in the urine) has been considered a more sensitive and specific measure of corti- sol excess. In addition, plasma ACTH and provocative tests using dexamethasone and metyrapone may help to discern the causes of Cushing’s syndrome. Although they are useful to distinguish adrenal from pituitary causes and adrenal tumors from nodular hyperplasia, too often these tests cannot differentiate adrenal adenomas from carci- nomas.20, 49, 86, 105
Although nonfunctioning adrenocortical tumors do not produce excessive active hor- mones, steroid precursors may be produced in abundance. 17-ketosteroid and 17-hydroxy- corticosteroid levels are normal; however, if hormonally inactive precursors are specifi- cally sought, such as metabolites of pregneno- lone, they are often found to be elevated.96 In such cases, these urinary metabolites may serve as markers for tumor recurrence.
RADIOLOGIC STUDIES
Although biochemical confirmation of adrenocortical overactivity is important, it cannot reliably distinguish benign lesions from malignant neoplasms; therefore, endo- crinologic evaluation should proceed swiftly concurrently with localization and staging of
the tumor so that surgical intervention is not delayed.
Daneman and co-workers24 reviewed the radiologic work-up of 17 patients with adre- nocortical tumors and found that a plain ab- dominal radiograph revealed a soft-tissue mass in 47% and calcification in the area of the adrenal gland in 18%. Other studies have shown that as many as 30% of adrenocortical adenomas may have calcifications.30 In addi- tion, plain radiographs can detect pulmonary or bone metastases. Daneman and co-workers also reported the findings of intravenous uro- grams; 53% showed either upper pole flat- tening or displacement, 6% showed a non- functional kidney, and 29% showed a normal kidney on the side of the adrenal tumor.24 Distortion of the caliceal system or nonfunc- tion of the kidney secondary to invasion by an adrenal carcinoma must be differentiated from a primary renal lesion, and nephroto- mography may aid in this regard.96
In the evaluation of adrenal tumors in chil- dren, ultrasound offers the advantages of be- ing noninvasive and radiation free. Lesions 3 cm or larger are readily delineated, and smaller tumors may be identified with real- time scanning. Prando and co-workers92 eval- uated 14 patients with adrenocortical tumors using ultrasonography and found that the smallest lesions were homogenous and smooth with no clear pattern of hyperecho- genecity or hypoechogenecity. All of the 10 large tumors were found to contain complex echogenic patterns, with eight showing the “scar sign.” These radiating linear echoes rep- resent the interphase between separate areas of necrosis, hemorrhage, and neoplasm, which is suggestive of adrenocortical carci- noma.92 Ultrasound is an ideal modality for screening the adrenal region and for assess- ment of recurrence postoperatively; however, it cannot reliably identify smaller lesions as accurately as CT.1 Moreover, it is technically difficult, requires special positioning, and is notoriously operator dependent.
With improved technology, CT has demon- strated unequivocal diagnostic superiority over ultrasound in imaging the adrenal gland. Tumors as small as 0.5 cm have been detected, and lesions larger than 1 cm can be delineated reliably. Detection in the pediatric population may be slightly more difficult ow- ing to the lack of retroperitoneal fat in chil- dren. The presence of tumor thrombus in the adrenal or renal vein or inferior vena cava can be assessed, and one can ensure contralateral
renal function should ipsilateral nephrectomy be necessary. The contralateral adrenal gland can also be imaged (Fig. 4A). Despite the vari- ous radiologic signs in the multimodality practice of radiology, only the presence of regional invasion or distant metastases (liver, lung, or brain) can reliably differentiate be- nign lesions from malignant tumors. Such in- vasion can be ascertained at the time of the initial CT scan to stage the patient ade- quately1, 24, 30 (Table 3). A hepatic lesion found in a pediatric patient with a small adrenal cortical tumor (<5 cm in diameter with weight <50 g) may not necessarily indicate metastatic disease. Cameron and co-workers19 reported two cases in which the adrenal tu- mors were adenomas with either hepatic granulomatous lesions from a parasitic infec- tion or focal nodular hyperplasia, which may be associated with hemihypertrophy.
The body of information concerning the role of MR imaging in the evaluation of adre- nal lesions is continually expanding. Reports indicate that MR imaging has an accuracy comparable with that of CT in detecting adre- nal masses larger than 1 to 2 cm38, 101 (Fig. 4B). MR imaging has the distinct advantage of producing coronal slices, which can identify a 1-cm adrenal mass not demonstrated by a sector CT scan.124 Other studies suggest that the ability of MR imaging to characterize ad- renal tissue may extend to differentiating nonfunctioning adenomas from malignant adrenal lesions. Boraschi and colleagues12, 38 have been able to diagnose adrenal adenomas using the central spot of high-intensity hyper- intense rim sign and homogeneous isointen- sity to liver on gadolinium-enhanced, fat-sup- pressed, spin-echo MR images. MR imaging is also useful for patients with intravenous contrast allergies and for the assessment of venous invasion.30, 39 Although the experience in children is limited, Hanson and co-work- ers44 report that pediatric adrenocortical carci- nomas and adenomas have intermediate en- hancement on T1 MR images and high enhancement on T2 images. In contrast, in adults, adrenocortical adenoma has an intermediate/low intensity in both T1- and T2-weighted images.44 Hanson and associates compared MR imaging with CT and ultra- sound in the evaluation of five pediatric pa- tients with functional adrenocortical tumors. MR imaging readily detected the lesions (1.0- 7.5 cm), which all showed high intensity on T2-weighted sequences. CT also detected all of the lesions, with all but one tumor enhanc-
A
2
TR=
EN-
5Q=
B
SN- 1
88
0
ing after intravenous contrast. Although the multiplanar MR images allowed a better dis- tinction from adjacent structures, both MR imaging and CT demonstrated the contralat- eral normal adrenal. Ultrasound could not visualize the small 1.0-cm lesions or the nor- mal contralateral adrenal.44 MR imaging has an important role in the evaluation of adreno- cortical tumor and may eventually replace CT as the diagnostic choice for localizing adrenal lesions, assessing the extent of disease, and potentially distinguishing benign lesions from malignant neoplasms.
Although they are considered invaluable tools for differentiating hyperplasia from car- cinoma, arteriography and adrenal venogra- phy are generally unnecessary and danger- ously invasive in children. On occasion, adrenal masses may be difficult to distinguish from upper pole renal tumors, and, in such cases, selective arteriography may help. Selec- tive angiography may be indicated for angi- oembolization of a tumor prior to surgery or to halt hemorrhage temporarily. This indica- tion is rare because most of these tumors have an extensive arterial supply. If a suspected
| Designation | Criteria |
|---|---|
| T1 | Tumor 5 cm or smaller, no invasion |
| T2 | Tumor larger than 5 cm, no invasion |
| T3 | Tumor any size, locally invading but not involving adjacent organs |
| T4 | Tumor any size, locally invading adjacent organs |
| NO | No regional positive nodes |
| N1 | Positive regional nodes |
| M0 | No distant metastasis |
| M1 | Distant metastasis present |
| Stage | |
| I | T1 NO MO |
| II | T2 NO MO |
| III | T1 or T2 N1 M0; T3 NO MO |
| IV | Any T, any N, M1; T3 N1; T4 |
Adapted from Contan RM: An approach to handling pediatric thyroid and adrenal tumors excluding neuroblastoma. Am J Clin Pathol 109:S73-81, 1998.
functional tumor cannot be delineated by CT or MR imaging, adrenal venography may re- veal lesions smaller than 1 cm.30 Although this procedure has the potential advantage of obtaining venous blood samples, it has been associated with morbidity, including adrenal insufficiency or hemorrhage.% Inferior vena- cavography may be indicated if the CT find- ings suggest tumor thrombus. In general, these invasive techniques should be reserved for the rare instance in which CT or MR im- aging cannot supply needed information.24, 49
Adrenal scintigraphy using various iodo- cholesterol-labeled analogues has shown promise in diagnosing and differentiating functional adrenal lesions.7, 41, 61, 112 Differential uptake makes it possible to distinguish hy- perplasia, adenomas, and carcinoma because carcinomas do not concentrate the radionu- clide as well as normal tissue. With docu- mented glucocorticoid excess, bilateral sym- metric images indicate adrenal hyperplasia. Unilateral uptake suggests an adenoma, whereas bilateral nonvisualization has been suggestive of carcinoma.112 In cases of aldoste- ronism or virilism, useful results have been limited to the differentiation of hyperplasia from adenoma, with no clear role for evalua- tion of carcinoma.53 The experience remains limited in children, in whom widespread use has been delayed by concerns about radiation exposure.% In addition, imaging is accom- plished over 4 to 7 days, an unacceptably long delay. Isolated case reports have shown that Ga-67 scintigraphy is efficacious in detec-
tion of recurrence51 and therapy response, whereas Tc-99m has been shown to accumu- late in a nonfunctional adrenocortical tu- mor.121 The use of NP-59 (131]-60-iodomethyln- orcholesterol) in adrenal scintigraphy has shown a sensitivity of 71% and a specificity of 100%. The functional information depicted by scintigraphy complements the morpho- logic evaluation by CT, and, in the absence of hormonal dysfunction, the presence of con- cordant CT and NP-59 scans is indicative of a functioning, but not hypersecretory, benign adrenocortical adenoma. In contrast, a discor- dant CT and NP-59 scan is suggestive of a nonfunctioning, space-occupying adrenal le- sion.41 Although adrenal scintigraphy offers promise as a noninvasive diagnostic modality, further study is needed.
Other functional imaging modalities can help differentiate benign lesions from malig- nant adrenal tumor. The results of positron emission tomography (PET) with FDG (2- [fluorine-18]-fluoro-2-deoxy-D-glucose) on 24 adult adrenal masses (1.5-10.0 cm) were cor- related with the findings at CT, surgery, or percutaneous biopsy. PET was helpful in di- agnosing benign versus malignant lesions in all patients. A statistically significant differ- ence was seen in mean uptake of FDG in malignant versus benign tumor.10 Further study is warranted before PET can become a standard imaging modality for the adrenal gland.
Ultrasonography, CT, and MR imaging have largely replaced other modalities as the imaging techniques of choice in the identifi- cation of adrenal masses. MR imaging is at least equal to CT in the detection of adreno- cortical tumors and has some possible advan- tages. Both techniques are superior to ultra- sound. With increased use of pediatric noninvasive imaging (ultrasound) in high- risk patients, it is hoped that adrenal masses will be detected earlier.
DIFFERENTIAL DIAGNOSIS
The differential diagnosis of virilization in children is relatively limited. Congenital adre- nal hyperplasia is the most common cause in either sex when the condition is seen at birth or during early infancy.16, 123 In any of these situations, physical findings are evident at, or shortly after, birth, including labial fusion and clitoromegaly in the female; however, late- onset congenital adrenal hyperplasia in the
male or milder forms in the female may go unrecognized until later in childhood. When symptoms present after a normal infancy, other diagnoses must be entertained. In the male, these diagnoses include Leydig cell tumor of the testis, true isosexual precoc- ity, human chorionic gonadotropin-secreting tumors, and adrenal pathology. Physical ex- amination may help distinguish these possi- bilities.49 Both testes remain small when viril- ization is extratesticular in origin, whereas the testes are enlarged in true sexual precocity or with tumors producing chorionic gonado- tropin. A unilaterally enlarged testis indicates the presence of a testicular tumor. In the pre- pubertal female, an accelerated growth rate associated with virilism in the absence of la- bial fusion strongly suggests an ovarian or adrenal tumor. These conditions should be easily distinguishable radiographically via ul- trasound or CT scan. Although ovarian tu- mors can produce a clinical presentation iden- tical to that of adrenal virilism, they rarely produce features of cortisol excess.% More- over, unlike in patients with adrenal tumors, urinary 17-ketosteroids are usually normal or minimally elevated in patients with ovarian tumors. In addition to adrenocortical tumors, the differential diagnosis of feminization in a male child, which usually presents as gyneco- mastia, should include testicular or extra- gonadal neoplasms such as pulmonary or retroperitoneal tumors producing chorionic gonadotropin or estrogens. Two benign con- ditions that produce gynecomastia in boys must also be considered. In idiopathic prepu- bertal gynecomastia, no accelerated growth rate, advanced bone age, or other signs of virilism are present. Although urinary estro- gen may be increased, 17-ketosteroid excre- tion is normal.125 The second condition, pu- bertal gynecomastia, occurs frequently in adolescent boys and in the absence of devel- opmental abnormalities should not prompt further evaluation.
In girls, other diagnostic considerations in- clude premature thelarche and isosexual pre- cocious puberty. Premature thelarche is benign early breast enlargement not accom- panied by accelerated growth rate, bone age, maturation, or vaginal bleeding.125 True iso- sexual precocious puberty, produced by pitu- itary hyperactivity stimulating the ovaries, must be differentiated from that caused by an ectopic chorionic gonadotropin-producing tumor. Although primary aldosteronism in children is usually a result of adrenocortical
hyperplasia, congenital adrenal hyperplasia must also be considered when mineralocorti- coid excess presents in combination with am- biguous genitalia. Laboratory studies should help distinguish congenital hyperplasia, Cushing’s syndrome, secondary hyperaldo- steronism, and adrenal neoplasia. Radiologic localization is invaluable in confirming the presence of an aldosterone-secreting adrenal tumor. The differential diagnosis of a non- functioning adrenal carcinoma presenting as an abdominal mass should include neuro- blastoma, Wilms’ tumor, hydronephrotic and cystic diseases of the kidney, and adrenal cyst. Rarely, primary mesenchymal tumors have been reported to arise in the adrenal cortex. Radiologic imaging studies will help in differ- entiating these entities.20
SURGICAL TREATMENT
Surgery is the cornerstone of successful treatment of adrenocortical carcinoma. Radi- cal excision with en bloc resection of any local invasion offers the best chance for cure. The authors advocate an anterior transabdominal approach via a bilateral subcostal incision for optimal exposure. Moreover, this approach affords the best opportunity to assess the ex- tent of local invasion, lymph node involve- ment, and the presence of metastatic disease, and to examine the contralateral adrenal gland.105, 123 If extensive local disease is pres- ent, wide en bloc resection of the tumor, nodes, and involved contiguous organs should be performed.21 A thoracoabdominal approach may be used when resecting a par- ticularly large mass. Adrenal tumors with ei- ther level three or four inferior vena cava tumor thrombus (at the intrahepatic veins or intracardiac) necessitate a combined urologic and cardiac team approach.111
In the current era of minimally invasive surgery, laparoscopy has all but replaced open surgery for adult tumors of the adre- nals.13 There is extensive experience with adult adrenal neoplasms. The clinical efficacy of laparoscopic adrenalectomy is comparable with that of open surgery, but the time of hospitalization and convalescence is mark- edly shortened.2 With progressive miniatur- ization, the standard 10/12 mm laparoscopic ports have shrunken to “needlescopic” sizes of 2 and 5 mm ports and 2 mm instruments. These ports are ideal in the pediatric popula- tion, and orchiopexy with or without excision
of remnant testis has been performed exclu- sively with needlescopic instruments.103 Gill and co-workers37 have reported their experi- ence with needlescopic versus laparoscopic adrenalectomy. The needlescopic approach was associated with a shorter surgical time (169 versus 220 minutes), hospital stay (1.1 versus 2.7 days), and convalescence (2.1 ver- sus 3.1 weeks) with similar clinical efficacy. Because of the rarity of the disease, no pediat- ric laparoscopic adrenalectomies for adreno- cortical carcinoma have been reported in the literature. Nevertheless, this minimally inva- sive surgical technique seems to be a promis- ing option for the treatment of pediatric adre- nocortical carcinoma.
Adjuvant therapy has been disappointing thus far, and surgical intervention offers the best chance for long-term disease-free sur- vival, even in the face of extensive disease.21, 96 In the review by Cohn and associates,21 pa- tients with tumors confined to the adrenal gland had a mean survival of 5 years after surgery, whereas patients with disease be- yond the gland had a mean survival of 2.3 years. In nine patients who underwent reop- eration for abdominal recurrence or pulmo- nary metastasis, the mean duration of sur- vival was 3.5 years.21 Two other groups have reported excision of the primary tumor with incomplete resection of tumor thrombus in- volving the vena cava105 and gross tumor em- boli invading the liver capsule.56 Each patient received adjuvant therapy and survived 1 and 12 years postoperatively, respectively. Driver and co-workers28 reported no benefit of adju- vant mitotane in their pediatric series, and Sabbaga and co-workers97 reported 1 survivor among 10 children treated with mitotane in combination with other options. The only sur- vivor also had total excision of the tumor, whereas mitotane alone or in conjunction with irradiation, partial excision, or biopsy yielded no survivors.97 In a report on the nat- ural history of untreated adrenocortical carci- noma, MacFarlane74 found a mean survival of 2.9 months; therefore, the benefits of adjuvant mitotane must be weighed against the risks and side effects. Dickstein and co-workers26 have proposed low-dose adjuvant mitotane (1.5-2.0 g/d) to diminish side effects and pos- sibly prolong survival. Others are opposed to this therapy. Dickstein and colleagues pointed out that their series of four patients was too small to draw a final conclusion4 and sug- gested the need for larger prospective ran- domized trials. The authors believe that ag-
gressive surgical approaches are justified to offer these children the best opportunity for long-term survival.
CHEMOTHERAPY
In the literature, as many as 66% of adult patients diagnosed with adrenocortical carci- noma are stage III and IV,48, 58 whereas one pediatric series reported that 22% of the pa- tients presented with nonlocalized disease.110 The reviewed pediatric series report a range of 5% to 64% of patients with metastases at diagnosis (stage IV) (Table 1). The alarming incidence of advanced disease at the time of diagnosis has prompted investigation into the role of adjuvant therapy. In 1949 Nelson and Woodward discovered that the insecticide chlorophenothane (DDT) led to selective ad- renal cortical necrosis in dogs. In 1960 Bergen- stal and associates9 demonstrated the utility of an isomer of DDT, mitotane (o,p’-DDD), in inducing regression of metastatic adrenocorti- cal carcinoma. This chemical blocks the adre- nal cortical enzyme 11-beta-hydroxylase, with a subsequent decrease in steroidogenesis.30 Mitotane has since been used extensively in adults with adrenocortical carcinoma, but its efficacy in children is not as well known ow- ing to the disease paucity in the pediatric population.
The published results regarding the utility of mitotane are controversial. Adult response rates range from 10% to 60% and depend on the definition of response in each series. Hut- ter and Kayhoe52 reviewed its use in 138 pa- tients, describing measurable tumor regres- sion in 34%; however, the mean duration of this response was only 10.2 months, and a significant increase in survival was not dem- onstrated. Lubitz and associates72 noted objec- tive tumor regression in 61% of 75 patients; however, no patients were cured. Wooten and King119 published an extensive review of all mitotane chemotherapeutic trials in nonoper- ative patients with adrenocortical carcinoma. Fifty-two studies encompassing 551 patients revealed a disappointing overall 35% mito- tane response rate (complete and partial) with a duration ranging from 1 to 205 months.119 Haak and associates42 provided a possible ex- planation of the enormous variability of mito- tane’s efficacy. Among 96 patients with adre- nocortical carcinoma, metastatic disease was diagnosed in 36 (37.5%). Eighty-four patients underwent surgical resection (total and subto-
tal), and 62 patients were eventually treated with mitotane. Thirty of the patients treated attained a mitotane blood serum level of greater than 14 mg/L, whereas 32 had a level less than 14 mg/L. The high blood serum levels had a significant impact on patient sur- vival based on univariate and multivariate analysis. Surprisingly, recurrent disease de- veloped in more than half of the patients who underwent total tumor resection, and the patients who were treated adjuvantly did not fare better than the patients who underwent surgery alone.42 In the pediatric literature, two series have reported a 30% and 38% tumor response rate to mitotane.80, 110
Although anecdotal, there are several docu- mented cases of “cure” with mitotane therapy in patients with metastatic disease. In a re- view of the world literature, Boven and co- workers14 found eight cases of complete re- sponse. Becker and Schumacher6 described two patients with metastatic disease who were alive 4.5 and 8 years after receiving ad- juvant mitotane following incomplete surgi- cal resection. Cagle and associates18 reported the use of o,p’-DDD at the time of initial resection in a child who did well for 11 months but in whom pulmonary metastases appeared 3 months after therapy was with- drawn. Ostuni and Roginsky88 treated a pa- tient with documented functioning metasta- ses with mitotane and 5-fluorouracil for 5 months. Therapy was discontinued because of adrenal insufficiency, and the patient died 10 years later of unrelated causes with au- topsy proven absence of metastatic disease. When aggressive surgery was combined with early use of o,p’-DDD in 17 patients, Schteingart and colleagues100 reported a mean survival of 46.6 months. Longer survival, up to 74 months, was obtained in patients receiv- ing adjuvant mitotane before metastases were evident.100
Despite some degree of mitotane efficacy, opponents point to the disabling side effects. Hutter and Kayhoe52 reported that only 11% of patients did not have toxicity. Gastrointes- tinal distress was seen in 83%, neurotoxicity in 41%, and skin rash in 12%. Luton and associates73 reported significantly improved tolerance when mitotane was combined with cellulose acetylphthalate, which prevents gas- tric absorption of o,p’-DDD. Monitoring plasma levels of the drug has also been sug- gested to maximize its effectiveness while limiting its side effects.73 Although convinc- ing evidence of the ability of mitotane to pro-
long survival is lacking, its role in ameliorat- ing the unpleasant symptoms of steroid excess is more established.86, 105, 123 As many as 75% of patients treated with mitotane can have a resolution of symptoms.119 Antihor- monal therapy can be used to palliate symp- toms of hypercortisolism. Aminoglutethi- mide, a desmolase inhibitor, and metyrapone, an inhibitor of 11-beta-hydroxylase, serve as direct enzyme antagonists and may help sup- press steroid production.106
Alternative adjuvant chemotherapeutic agents have been less well studied. Vincris- tine, vinblastine, and carmustine have used in the terminal stages of the disease without effect.56 Other drugs reported in the litera- ture include cyclophosphamide, actinomycin D, 5-fluorouracil, doxorubicin, carmustine (BCNU), and methotrexate, all usually with poor results. The combination of 5-fluoroura- cil and mitotane cured one patient with meta- static disease,88 and mitotane combined with streptozocin has been successful in isolated cases.31
In a South Western Oncology Group study, cisplatin in combination with mitotane pro- duced objective responses in one third of adult patients with metastatic adrenocortical carcinoma, although all eventually died of their disease.27 Combination cis-platinum and VP-16 has shown some efficacy in pediatric patients with adrenocortical carcinoma but with cis-platinum dose-limiting toxicity95; therefore, Ayass and co-workers3 instituted a high-dose carboplatinum and VP-16 regimen as adjuvant therapy for a 17-month-old boy with a left adrenal mass and metastasis to the chest and brain. After a left adrenalectomy showed adrenocortical carcinoma, the patient was treated with eight cycles of a progres- sively increasing dose of VP-16 (250 mg/m2/ day) and carboplatinum (470 mg/m2/day). The metastases disappeared after the sixth cycle, and the patient was disease free for 10 months.3
Taxol has been shown to produce apoptosis in the steroid-secreting adrenocortical carci- noma cell line, NCI-H295. Suramin and mito- tane are the only other agents known to pro- duce cytotoxicity in this cell line.32 Although no published data on human trials are avail- able, taxol represents another chemothera- peutic opportunity.
Progress in the chemotherapeutic approach to adrenocortical carcinoma has been slow. The gold standard chemotherapeutic agent is still mitotane, but combination chemotherapy
remains promising. No published human data on taxol or immunotherapeutic regimens are known, creating fertile areas for further research.
RADIATION
The role of radiotherapy in children with adrenocortical carcinoma has not been well established. It is generally believed that adre- nal carcinoma in adults is radioresistant; therefore, radiation has not been regarded as an effective form of treatment.43, 69, 74 Stewart and co-workers105 administered 1500 to 3000 rads to four children with adrenocortical car- cinoma, including three with metastatic dis- ease, two of whom survived. LeFevre and colleagues66 described the use of preoperative radiation to shrink an “unresectable” tumor, which they subsequently excised with excel- lent long-term results. Cohn and co-workers21 described the successful use of radiotherapy for palliation of pain from bone metastases but could not demonstrate improved survival when radiation was administered to the ab- dominal field.
Of concern, there are reports of a second primary malignancy developing in the field of radiation administered years earlier for adrenocortical carcinoma. Daneman and co- workers24 reported on one patient who died 19 years after curative resection of adrenocor- tical carcinoma of a malignant neurilemmoma of the cauda equina. Driver and colleagues28 reported that in their series of 14 patients with adrenocortical carcinoma, 5 were long- term survivors, with 3 of the 5 patients dying of a second tumor. All of the lesions were sarcomas and presumed to be caused by the prior irradiation.28 Squire and co-workers104 documented the development of a fatal left breast sarcoma in a 14-year-old girl 13 years after successful combined surgical and radia- tion therapy for a left adrenocortical carci- noma. These cases illustrate that the poor ef- ficacy of radiotherapy in surgically curative adrenal adrenocortical carcinoma may not be worth the risk of a subsequent sarcoma 10 years post therapy. Patients previously treated with irradiation for adrenocortical car- cinoma should undergo long-term (>10 year) follow-up.
PROGNOSIS
The natural history of untreated adrenocor- tical carcinoma was noted by MacFarlane
who stated that the mean survival of 35 pa- tients was 2.9 months.74 In adults, the tumor is highly lethal, with a mortality rate ap- proaching 90%.106 In one series, 50% of pa- tients died within 2 years after the onset of symptoms, and the 3-year survival rate was less than 25%.70 In a review of 222 children with functional adrenocortical tumors, there were only 23 2-year survivors.47 Nonfunction- ing tumors bear an even graver prognosis.86, 105, 116 In addition to the tumor’s inherently aggressive nature, poor survival has been at- tributed to a delay in diagnosis, even though these tumors generally have a phenotypical manifestation. Weatherby and co-workers116 noted that the duration of symptoms before diagnosis ranged from 2 months to 10 years (mean, 18 months). In the series reviewed in Table 1, the mean time ranged from 6 to 36 months, whereas the range was from birth to 5.5 years. With this delay, the patient will ultimately be treated at a higher stage and older age, prognostic indicators of poor out- comes.23 The rarity of the disease further com- promises attempts to establish accurate sur- vival figures in children.116
Some investigators hypothesize that the prognosis for children is not as grim as pre- viously thought.18, 49, 116 The poor postopera- tive survival rates reported in the older pedi- atric literature have been attributed, in part, to a lack of recognition of the need for periop- erative steroid replacement because of the suppression of the contralateral adrenal gland by the hypersecretory neoplasm, and it has been suggested that this problem gave early investigators a false impression of the true malignant potential of adrenocortical tu- mors.18 The data collected by Weatherby and Carney116 support this supposition, with five of nine children who underwent surgical exci- sion of adrenal carcinoma alive without recur- rence or metastases from 3 to 34 years later. Pediatric series published in the 1990s have revealed overall survival rates ranging from 43% to 91% (Table 1). Michalkiewicz and as- sociates83 reported that of 20 pediatric pa- tients with small (<200 cm3 or <100 g) adre- nocortical tumors that were surgically resected, 18 patients were alive at median follow-up of 2.3 years. This overall survival rate of 90% was independent of tumor histol- ogy and correlated with total tumor excision. As more data accumulate, the true prognosis for children with adrenocortical carcinoma will undoubtedly become clearer.
SUMMARY
Adrenocortical carcinoma in childhood is a rare potentially fatal disease. Despite its often dramatic presentation, there typically has been a distressingly long delay between the onset of symptoms and the time of diagnosis. This delay undoubtedly has contributed to the historically poor prognosis in these chil- dren by permitting the disease to reach an advanced stage before treatment is started. It is imperative that the physician recognizes the endocrine manifestations of these tumors early and has a high index of suspicion. Al- though biochemical and histologic evalua- tions are helpful, they often cannot differenti- ate benign lesions from malignant neoplasms and should not unduly delay intervention. Aggressive complete surgical resection con- tinues to be the mainstay of treatment and is the best prognosticator of overall survival. The role of adjuvant therapy and chemother- apy continues to evolve. Molecular studies have increased understanding of cancer biol- ogy and may provide possible novel thera- peutic approaches in the future. It is hoped that increased familiarity with this unusual tumor will result in earlier detection, prompt intervention, and improved survival for chil- dren with adrenocortical carcinoma.
References
1. Abrams HL, Siegelman SS, Adams DF, et al: Com- puted tomography versus ultrasound of the adrenal gland: A prospective study. Radiology 143:121-128, 1982
2. Aldrighetti L, Giacomelli M, Calori G, et al: Impact of minimally invasive surgery on adrenalectomy for incidental tumors: Comparison with laparotomic technique. Int Surg 82:160-164, 1997
3. Ayass M, Gross S, Harper J: High-dose carboplati- num and VP-16 in treatment of metastatic adrenal carcinoma. Am J Pediatr Hematol Oncol 13:470- 472, 1991
4. Barzon L: Comment-is there a role for low doses of mitotane (o,p’-DDD) as adjuvant therapy in adre- nocortical carcinoma? J Clin Endocrinol Metab 84:1488-1489, 1999
5. Bauman A, Bauman CG: Virilizing adrenocortical carcinoma: Development in a patient with salt-los- ing congenital adrenal hyperplasia. JAMA 248:3140- 3141, 1982
6. Becker D, Schumacher OP: O,P’DDD therapy in invasive adrenocortical carcinoma. Ann Intern Med 82:677-679, 1975
7. Beierwaltes WH, Wieland DM, Yu T, et al: Adrenal imaging agents: Rationale, synthesis, formulation, and metabolism. Semin Nucl Med 8:5-21, 1978
8. Bergada I, Venara M, Maglio S, et al: Functional adrenal cortical tumors in pediatric patients: A clini-
copathologic and immunohistochemical study of a long term follow-up series. Cancer 77:771-777, 1996
9. Bergenstal DM, Hertz R, Lipsett MB: Chemotherapy of adrenocortical cancer with o,p’DDD. Ann Intern Med 53:672-682, 1960
10. Boland GW, Goldberg MA, Lee MJ, et al: Indetermi- nate adrenal mass in patients with cancer: Evalua- tion at PET with 2-[F-18]-fluoro-2-deoxy-D-glucose. Radiology 194:131-134, 1995
11. Boothroyd AE, Dicks-Mireaux C, Malone M: Adre- nal cortical tumours in children. Eur J Radiol 18:199-204, 1994
12. Boraschi P, Braccini G, Gigoni R, et al: Diagnosis of adrenal adenoma: Value of central spot of high- intensity hyperintense rim sign and homogeneous isointensity to liver on gadolinium-enhanced fat- suppressed spin-echo MR images. J Magn Reson Imaging 9:304-310, 1999
13. Bornstein SR, Stratakis CA, Chrousos GP: Adreno- cortical tumors: Recent advances in basic concepts and clinical management. Ann Intern Med 130:759- 771, 1999
14. Boven E, Vermorken JB, van Slooten H, et al: Com- plete response of metastasized adrenal cortical carci- noma with o,p’-DDD: Case report and literature review. Cancer 53:26-29, 1984
15. Bugg MF, Ribeiro RC, Roberson PK, et al: Correla- tion of pathologic features with clinical outcome in pediatric adrenocortical neoplasia: A study of a Brazilian population. Brazilian Group for Treatment of Childhood Adrenocortical Tumors. Am J Clin Pathol 101:625-629, 1994
16. Burrington JD, Stephens CA: Virilizing tumors of the adrenal gland in childhood: Report of eight cases. J Pediatr Surg 4:291-302, 1969
17. Cacciari E, Cicognani A, Pirazzoli P, et al: Adreno- cortical tumours in children: Our experience with nine cases. Acta Endocrinol Suppl 279:264-274, 1986
18. Cagle PT, Hough AJ, Pysher TJ, et al: Comparison of adrenal cortical tumors in children and adults. Cancer 57:2235-2237, 1986
19. Cameron FJ, Sohaib SA, Scheimberg I, et al: The significance of hepatic lesions associated with small adrenocortical tumours in childhood. Br J Radiol 70:852-855, 1997
20. Chudler RM, Kay R: Adrenocortical carcinoma in children. Urol Clin North Amer 16:469-479, 1989
21. Cohn K, Gottesman L, Brennan M: Adrenocortical carcinoma. Surgery 100:1170-1177, 1986
22. Comite F, Schiebinger RJ, Albertson BD, et al: Iso- sexual precocious pseudopuberty secondary to a feminizing adrenal tumor. J Clin Endocrinol Metab 58:435-440, 1984
23. Conran RM: An approach to handling pediatric thy- roid and adrenal tumors excluding neuroblastoma. Am J Clin Pathol 109:S73-81, 1998
24. Daneman A, Chan HS, Martin J: Adrenal carcinoma and adenoma in children: A review of 17 patients. Pediatr Radiol 13:11-18, 1983
25. Dehner LP: Neoplasms of the adrenal cortex: Preoc- cupation bordering on obsession. Am J Clin Pathol 101:557-558, 1994
26. Dickstein G, Shechner C, Arad E, et al: Is there a role for low doses of mitotane (o,p’-DDD) as adjuvant therapy in adrenocortical carcinoma? J Clin Endocri- nol Metab 83:3100-3103, 1998
27. Dogliotti L, Berruti A, Pia A, et al: Cytotoxic chemo- therapy for adrenocortical carcinoma. Minerva En- docrinol 20:105-109, 1995
28. Driver CP, Birch J, Gough DC, et al: Adrenal cortical tumors in childhood. Pediatr Hematol Oncol 15:527- 532, 1998
29. Drop SLS, Bruining GJ, Visser HKA, et al: Prolonged galactorrhoea in a 6-year-old girl with isosexual pre- cocious puberty due to a feminizing adrenal tu- mour. Clin Endocrinol 15:37-43, 1981
30. Dunnick NR: Adrenal carcinoma. Radiol Clin North Am 32:99-108, 1994
31. Eriksson B, Oberg K, Curstedt T, et al: Treatment of hormone-producing adrenocortical cancer with o,p’DDD and streptozocin. Cancer 59:1398-1403, 1987
32. Fallo F, Pilon C, Barzon L, et al: Effects of taxol on the human NCI-H295 adrenocortical carcinoma cell line. Endocr Res 22:709-715, 1996
33. Federici S, Galli G, Ceccarelli PL, et al: Adrenocorti- cal tumors in children: A report of 12 cases. Eur J Pediatr Surg 4:21-25, 1994
34. Figueiredo BC, Stratakis CA, Sandrini R, et al: Com- parative genomic hybridization analysis of adreno- cortical tumors of childhood. J Clin Endocrinol Metab 84:1116-1121, 1999
35. Fraumeni JFJ, Miller RW: Adrenocortical neoplasms with hemihypertrophy, brain tumors, and other dis- orders. J Pediatr 70:129-138, 1967
36. Gicquel C, Bertagna X, Le Bouc Y: Recent advances in the pathogenesis of adrenocortical tumours. Eur J Endocrinol 133:133-144, 1995
37. Gill IS, Soble JJ, Sung GT, et al: Needlescopic adrenalectomy-the initial series: Comparison with conventional laparoscopic adrenalectomy. Urology 52:180-186, 1998
38. Glazer GM: MR imaging of the liver, kidneys, and adrenal glands. Radiology 166:303-312, 1988
39. Godine LB, Berdon WE, Brasch RC, et al: Adreno- cortical carcinoma with extension into inferior vena cava and right atrium: Report of 3 cases in children. Pediatr Radiol 20:166-168; discussion, 169, 1990
40. Grayson GH, Moore S, Schneider BG, et al: Novel germline mutation of the p53 tumor suppressor gene in a child with incidentally discovered adrenal cortical carcinoma. Am J Pediatr Hematol Oncol 16:341-347, 1994
41. Gross MD, Shapiro B, Francis IR, et al: Scintigraphic evaluation of clinically silent adrenal masses. J Nucl Med 35:1145-1152, 1994
42. Haak HR, Hermans J, van de Velde CJ, et al: Opti- mal treatment of adrenocortical carcinoma with mi- totane: Results in a consecutive series of 96 patients. Br J Cancer 69:947-951, 1994
43. Hajjar RA, Hickey RC, Samaan NA: Adrenal cortical carcinoma: A study of 32 patients. Cancer 35:549- 554, 1975
44. Hanson JA, Weber A, Reznek RH, et al: Magnetic resonance imaging of adrenocortical adenomas in childhood: Correlation with computed tomography and ultrasound. Pediatr Radiol 26:794-799, 1996
45. Hartley AL, Birch JM, Marsden HB, et al: Adrenal cortical tumours: Epidemiological and familial as- pects. Arch Dis Child 62:683-689, 1987
46. Hawkins EP, Cagle PT: Adrenal cortical neoplasms in children. Am J Clin Pathol 98:382-383, 1992
47. Hayles AB, Hahn HBJ, Sprague RG: Hormone-secre- ting tumors of the adrenal cortex in children. Pediat- rics 37:19-25, 1966
48. Henley DJ, van Heerden JA, Grant CS, et al: Adrenal cortical carcinoma-a continuing challenge. Surgery 94:926-931, 1983
49. Holcombe JH, Pysher TJ, Kirkland RT: Functioning adrenocortical tumors in childhood. In Humphrey GB, Grindey GB, Dehner LP, (eds): Adrenal and Endocrine Tumors in Children, Boston, Martinus Nijhoff, 1983
50. Hough AJ, Hollifield JW, Page DL, et al: Prognostic factors in adrenal cortical tumors: A mathematical analysis of clinical and morphologie data. Am J Clin Pathol 72:390-399, 1979
51. Howman-Giles R, Dalla Pozza L, Uren R: Ga-67 scintigraphy in a child with adrenocortical carci- noma. Clin Nucl Med 18:642-645, 1993
52. Hutter AM, Kayhoe DE: Adrenal cortical carcinoma: Results of treatment with o,p’DDD in 138 patients. Am J Med 41:581-592, 1966
53. Hwang I, Balingit AG, Georgitis WJ, et al: Adreno- cortical SPECT using iodine-131 NP-59. J Nucl Med 39:1460-1463, 1998
54. Ilvesmaki V, Kahri AI, Miettinen PJ, et al: Insulin- like growth factors (IGFs) and their receptors in adrenal tumors: High IGF-II expression in func- tional adrenocortical carcinomas. J Clin Endocrinol Metab 77:852-858, 1993
55. Kamio T, Shigematsu K, Sou H, et al: Immunohisto- chemical expression of epidermal growth factor re- ceptors in human adrenocortical carcinoma. Hum Pathol 21:277-282, 1990
56. Kay R, Schumacher OP, Tank ES: Adrenocortical carcinoma in children. J Urol 130:1130-1132, 1983
57. Kenny FM, Hashida Y, Askari A, et al: Virilizing tumors of the adrenal cortex. Am J Dis Child 115:445-458, 1968
58. Khorram-Manesh A, Ahlman H, Jansson S, et al: Adrenocortical carcinoma: Surgery and mitotane for treatment and steroid profiles for follow-up. World J Surg 22:605-611; discussion, 611-612, 1998
59. King DR, Lack EE: Adrenal cortical carcinoma: A clinical and pathologic study of 49 cases. Cancer 44:239-244, 1979
60. Kjellman M, Kallioniemi OP, Karhu R, et al: Genetic aberrations in adrenocortical tumors detected using comparative genomic hybridization correlate with tumor size and malignancy. Cancer Res 56:4219- 4223, 1996
61. Kloos RT, Gross MD, Shapiro B, et al: Diagnostic dilemma of small incidentally discovered adrenal masses: Role for 131I-6beta-iodomethyl-norcholest- erol scintigraphy. World J Surg 21:36-40, 1997
62. Korth-Schutz S, Levine LS, Roth JA, et al: Virilizing adrenal tumor in a child suppressed with dexameth- asone for three years: Effect of o,p’-DDD on serum and urinary androgens. J Clin Endocrinol Metab 44:433-439, 1977
63. Lack EE, Mulvihill JJ, Travis WD, et al: Adrenal cortical neoplasms in the pediatric and adolescent age group: Clinicopathologic study of 30 cases with emphasis on epidemiological and prognostic fac- tors. Pathology Annual 27(pt 1):1-53, 1992
64. Lanes R, Gonzalez S, Obregon O: Adrenal cortical carcinoma in a 4-year-old child: Temporal changes in steroid secretion. Clin Pediatr 21:164-166, 1982
65. Lee PD, Winter RJ, Green OC: Virilizing adrenocor- tical tumors in childhood: Eight cases and a review of the literature. Pediatrics 76:437-444, 1985
66. LeFevre M, Gerard-Marchant R, Gubler JP: Adrenal cortical carcinoma in children: 42 Patients treated 1958 to 1980 at Villejuif. In Humphrey GB, Grindey GB, Dehner LP, (eds): Adrenal and Endocrine Tu- mors in Children, Boston, Martinus Nijhoff, 1983
67. Levine GW: Adrenocortical carcinoma in two chil- dren with subsequent primary tumors. Am J Dis Child 132:238-240, 1978
68. Limon J, Dal Cin P, Kakati S, et al: Cytogenetic findings in a primary adrenocortical carcinoma. Cancer Genet Cytogenet 26:271-277, 1987
69. Lipsett MB: Treatment of adrenal carcinoma. Mod- ern Treatment 3:1377-1388, 1966
70. Lipsett MB, Hertz R, Ross GT: Clinical and patho- logical aspects of adrenocortical carcinoma. Am J Med 35:374-383, 1963
71. Liu J, Kahri AI, Heikkila P, et al: Ribonucleic acid expression of the clustered imprinted genes, p57KIP2, IGF-II, and H19, in adrenal tumors and cultured adrenal cells. J Clin Endocrinol Metab 82:1766-1771, 1997
72. Lubitz JA, Freeman L, Okun R: Mitotane use in inoperable adrenal cortical carcinoma. JAMA 223:1109-1112, 1973
73. Luton JP, Mahoudeau JA, Bouchard P, et al: Treat- ment of Cushing’s disease by o,p’DDD: Survey of 62 cases. N Engl J Med 300:459-464, 1979
74. MacFarlane DA: Cancer of the adrenal cortex: The natural history, prognosis and treatment in a study of fifty-five cases. Ann R Coll Surg Engl 23:155- 165, 1958
75. Mahloudji M, Ronaghy H, Dutz W: Virilizing adre- nal carcinoma in two sibs. J Med Genet 8:160-163, 1971
76. Malkin D: Germline p53 mutations and heritable cancer. Annu Rev Genet 28:443-465, 1994
77. Malkin D: p53 and the Li-Fraumeni syndrome. Bio- chim Biophys Acta 1198:197-213, 1994
78. Marigo C, Muller H, Davies JN: Survey of cancer in children admitted to a Brazilian charity hospital. J Natl Cancer Inst 43:1231-1240, 1969
79. Mark S, Clark OH, Kaplan RA: A virilized patient with congenital hemihypertrophy. Postgrad Med J 70:752-755, 1994
80. Mayer SK, Oligny LL, Deal C, et al: Childhood adrenocortical tumors: Case series and reevaluation of prognosis-a 24-year experience. J Pediatr Surg 32:911-915, 1997
81. Mendonca BB, Lucon AM, Menezes CA, et al: Clini- cal, hormonal and pathological findings in a com- parative study of adrenocortical neoplasms in child- hood and adulthood. J Urol 154:2004-2009, 1995
82. Mertens F, Kullendorff CM, Moell C, et al: Complex karyotype in a childhood adrenocortical carcinoma. Cancer Genet Cytogenet 105:190-192, 1998
83. Michalkiewicz EL, Sandrini R, Bugg MF, et al: Clini- cal characteristics of small functioning adrenocorti- cal tumors in children. Med Pediatr Oncol 28:175- 178, 1997
84. Moore L, Barker AP, Byard RW, et al: Adrenal corti- cal tumors in childhood-clinicopathological fea- tures of six cases. Pathology 23:94-97, 1991
85. Moul JW, Bishoff JT, Theune SM, et al: Absent ras gene mutations in human adrenal cortical neo- plasms and pheochromocytomas. J Urol 149:1389- 1394, 1993
86. Neblett WW, Frexes-Steed M, Scott HWJ: Experience with adrenocortical neoplasms in childhood. Am Surg 53:117-125, 1987
87. Nogeire C, Fukushima DK, Hellman L, et al: Viriliz- ing adrenal cortical carcinoma. Cancer 40:307-313, 1977
88. Ostuni JA, Roginsky MS: Metastatic adrenal cortical carcinoma: Documented cure with combined che- motherapy. Arch Intern Medic 135:1257-1258, 1975
89. Pang S, Becker D, Cotelingam J, et al: Adrenocortical tumor in a patient with congenital adrenal hyper- plasia due to 21-hydroxylase deficiency. Pediatrics 68:242-246, 1981
90. Parker DA, Skalko RG: Congenital asymmetry: Re- port of 10 cases with associated developmental ab- normalities. Pediatrics 44:584-589, 1969
91. Parma J, Duprez L, Van Sande J, et al: Somatic mutations in the thyrotropin receptor gene cause hyperfunctioning thyroid adenomas. Nature 365:649-651, 1993
92. Prando A, Wallace S, Marins JL, et al: Sonographic findings of adrenal cortical carcinomas in children. Pediatr Radiol 20:163-165; discussion, 169, 1990
93. Reincke M: Mutations in adrenocortical tumors. Horm Metab Res 30:447-455, 1998
94. Reincke M, Beuschlein F, Lalli E, et al: DAX-1 ex- pression in human adrenocortical neoplasms: Impli- cations for steroidogenesis. J Clin Endocrinol Metab 83:2597-2600, 1998
95. Ribeiro RC, Sandrini Neto RS, Schell MJ, et al: Adre- nocortical carcinoma in children: A study of 40 cases. J Clin Oncol 8:67-74, 1990
96. Richie JP, Gittes RF: Carcinoma of the adrenal cor- tex. Cancer 45:1957-1964, 1980
97. Sabbaga CC, Avilla SG, Schulz C, et al: Adrenocorti- cal carcinoma in children: Clinical aspects and prog- nosis. J Pediatr Surg 28:841-843, 1993
98. Sandrini R, Ribeiro RC, DeLacerda L: Extensive per- sonal experience childhood adrenocortical tumors. J Clin Endocrinol Metab 82:2027-2031, 1997
99. Sasano H, Suzuki T, Shizawa S, et al: Transforming growth factor alpha, epidermal growth factor, and epidermal growth factor receptor expression in nor- mal and diseased human adrenal cortex by immu- nohistochemistry and in situ hybridization. Mod Pathol 7:741-746, 1994
100. Schteingart DE, Motazedi A, Noonan RA, et al: Treatment of adrenal carcinomas. Arch Surg 117:1142-1146, 1982
101. Schultz CL, Haaga JR, Fletcher BD, et al: Magnetic resonance imaging of the adrenal glands: A compar- ison with computed tomography. AJR Am J Roent- genol 143:1235-1240, 1984
102. Smets A, Mortele K, De Wever N, et al: Coexistence of an adrenocortical carcinoma with an abdominal ganglioneuroma in a child. Pediatr Radiol 28:329- 331, 1998
103. Soble JJ, Gill IS: Needlescopic urology: Incorporat- ing 2-mm instruments in laparoscopic surgery. Urol- ogy 52:187-194, 1998
104. Squire R, Bianchi A, Jakate SM: Radiation-induced sarcoma of the breast in a female adolescent: Case report with histologic and therapeutic considera- tions. Cancer 61:2444-2447, 1988
105. Stewart DR, Jones PH, Jolleys A: Carcinoma of the adrenal gland in children. J Pediatr Surg 9:59-67, 1974
106. Sullivan M, Boileau M, Hodges CV: Adrenal cortical carcinoma. J Urol 120:660-665, 1978
107. Tang CK, Gray GF: Adrenocortical neoplasms: Prog- nosis and morphology. Urology 5:691-695, 1975
108. Tank ES, Kay R: Neoplasms associated with hemi- hypertrophy, Beckwith-Wiedemann syndrome and aniridia. J Urol 124:266-268, 1980
109. Teinturier C, Brugieres L, Lemerle J, et al: [Adreno- cortical carcinoma in children: Retrospective study of 54 cases]. Arch Pediatr 3:235-240, 1996
110. Teinturier C, Pauchard MS, Brugieres L, et al: Clini-
cal and prognostic aspects of adrenocortical neo- plasms in childhood. Med Pediatr Oncol 32:106- 111, 1999
111. Theman T, Williams WG, Simpson JS, et al: Tumor invasion of the upper inferior vena cava: The use of profound hypothermia and circulation arrest as a surgical adjunct. J Pediatr Surg 13:331-334, 1978
112. Thrall JH, Freitas JE, Beierwaltes WH: Adrenal scin- tigraphy. Semin Nucl Med 8:23-41, 1978
113. van Slooten H, Schaberg A, Smeenk D, et al: Mor- phologic characteristics of benign and malignant adrenocortical tumors. Cancer 55:766-773, 1985
114. Vargas MP, Vargas HI, Kleiner DE, et al: Adrenocor- tical neoplasms: Role of prognostic markers MIB-1, p53 and RB. Am J Surg Pathol 21:556-562, 1997
115. Wagner J, Portwine C, Rabin K, et al: High fre- quency of germline p53 mutations in childhood adrenocortical cancer. J Natl Cancer Inst 86:1707- 1710, 1994
116. Weatherby RP, Carney JA: Pathologic features of childhood adrenocortical tumors. In Humphrey GB, Grindey GB, Dehner LP (eds): Adrenal and Endo- crine Tumors in Children, Boston, Martinus Nij- hoff, 1983
117. Weiss LM: Comparative histologic study of 43 me-
tastasizing and nonmetastasizing adrenocortical tu- mors. Am J Surg Pathol 8:163-169, 1984
118. Weiss LM, Medeiros LJ, Vickery ALJ: Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 13:202-206, 1989
119. Wooten MD, King DK: Adrenal cortical carcinoma: Epidemiology and treatment with mitotane and a review of the literature. Cancer 72:3145-3155, 1993
120. Yashiro T, Hara H, Fulton NC, et al: Point mutations of ras genes in human adrenal cortical tumors: Ab- sence in adrenocortical hyperplasia. World J Surg 18:455-460; discussion, 460-461, 1994
121. Yen T, Yeh S: Nonfunctioning adrenocortical tumor showing concentration of Tc-99m MDP in a child. Clin Nucl Med 21:579-580, 1996
122. Young JLJ, Miller RW: Incidence of malignant tu- mors in US children. J Pediatr 86:254-258, 1975
123. Zaitoon MM, Mackie GG: Adrenal cortical tumors in children. Urology 12:645-649, 1978
124. Ziegelbaum MM, Kay R, Rothner AD, et al: The association of neuroblastoma with myoclonic en- cephalopathy of infants: The use of magnetic reso- nance as an imaging modality. J Urol 139:81-82, 1988
125. Zimmerman D: Adrenal cortex. In Kelias PP, King LR, Belman AB (eds): Urol Clin North Am, 1985
Address reprint requests to Robert Kay, MD Section of Pediatric Urology The Cleveland Clinic Foundation 9500 Euclid Ave. H18 Cleveland, OH 44195