Adrenocortical Carcinoma in Children: A Study of 40 Cases
By Raul C. Ribeiro, Romolo Sandrini Neto, Michael J. Schell, Luis Lacerda, Gilberto A. Sambaio, and Israil Cat
Adrenocortical carcinoma (ACC), a very rare tumor in children in the United States, is apparently more common among Brazilian children. We reviewed the medical records of 40 children whose disease was diagnosed between 1966 and 1987. There were 12 boys and 28 girls; their median age was 3.9 years (range, 1 day to 15.7 years). Virilization was the most common clinical sign (37 of 40); other signs included abdominal mass, deepened voice, plethora, hyperten- sion, seizures (seven patients) and, rarely, weight loss (two patients). The median time between first signs or symptoms and diagnosis was 1.4 years (range, 3 days to 5 years). Four of 33 tumors were classified as benign according to the Weiss, van Slooten, or Hough systems (tumor tissue was unavailable for seven patients). Tumors were completely resected in 26 of 38 patients; of those, 17 are in continuous complete remission, five relapsed, and four have been lost to follow-up. One patient, who had local recurrence, has been in a third complete remission for 18+ months after tumor resec- tion and chemotherapy (cisplatin and etoposide). Of
A DRENOCORTICAL carcinoma (ACC) oc- curs very rarely in children. The actual incidence of this tumor is unknown, although recently published data suggest an incidence of 75 to 115 new cases each year in the United States (0.5 per million per year).1 For reasons yet to be determined, ACC occurs much more fre- quently in children in southern Brazil.2 Because so few children have been treated for ACC, even at major cancer centers, clinical experience is relatively limited; therefore, optimal therapy has not been established. Surgery continues to be the front-line therapeutic modality, although some patients develop recurrent. disease in spite of “curative” resection. These patients and those with metastases at diagnosis pose a clinical challenge to oncologists treating children with this tumor. To ascertain the clinical and biologic characteristics’ of patients at high risk of treat- ment failure, we reviewed the medical records of 40 children treated for ACC in the Pediatrics Department at the Hospital de Clinicas in Cu- ritiba, Brazil. The results of our investigation suggest that older age, interval of ≥ 6 months between the initial symptoms and diagnosis, increased urinary steroid levels, and a tumor size greater than 200 cm3 are associated with an
the remaining 14 patients, 11 died of progressive disease, the diagnosis was confirmed at autopsy in two, and one has been lost to follow-up. Univariate analysis disclosed that age ≥ 3.5 years at diagnosis, interval of ≥ 6 months between first symptoms and diagnosis, tumor weight greater than 100 g, tumor size greater than 200 cm3, and high levels of urinary 17-ketosteroids (17-KS) and 17-hydroxycorticosteroids (17-OH) were associated with an unfavorable out- come. Multivariate analysis disclosed that only a tumor size greater than 200 cm3 independently identi- fies those patients with an unfavorable prognosis. Among the variables known before surgery, age, and the interval between first symptoms and diagnosis were important predictors of, outcome. Our data suggest that some children with ACC and certain clinical characteristics are at high risk of primary treatment failure and, therefore, are good candidates for investigational adjuvant therapy.
J Clin Oncol 8:67-74. @ 1990 by American Society of Clinical Oncology.
increased risk of treatment failure. Only the last variable showed independent prognostic value.
PATIENTS AND METHODS
During a 21-year period, from 1966 to 1987, 40 children were diagnosed as having ACC either after surgery (38 patients) or confirmed at autopsy (two patients). The age distribution and gender of the patients with ACC are shown in Fig 1. The median age was 3.9 years (range, 1 day to 15.7 years). Consistent with previous reports,3 there was a clear predominance of girls (n = 28), especially among very young children.
The diagnosis was made on the basis of the gross and histologic appearance of tissue obtained at the time of surgery or autopsy according to established criteria.4 For this study, the histopathologic material was reexamined, without prior
From the Department of Hematology-Oncology and Divi- sion of Biostatistics and Information Systems, St Jude Children’s Research Hospital, Memphis, TN; and the Depart- ment of Pediatrics, Federal University of Parana, Curitiba, Brazil.
Submitted January 27, 1989; accepted August 21, 1989. Supported in part by the American Lebanese Syrian Associated Charities.
Address reprint requests to Romolo Sandrini Neto, MD, Departmento de Pedatria, Hospital de Clinicas, UFPr, 80069, Curitiba, PR, Brazil.
@ 1990 by American Society of Clinical Oncology. 0732-183X/90/0801-0009$3.00/0
14
13
☐ Girls
12
☒ Boys
11
Number of Cases
10
9
8
7
6
5
4
3
2
1
0-2
2-4
4-6 6-8 8-10 10-12 12-14 14-16
Age in Years
knowledge of the patients’ outcomes, and classified according to the criteria established by Hough et al,5 van Slooten et al6 and Weiss.7 As suggested by these investigators, several histologic parameters were analyzed. The sum of the numeri- cal values of each parameter indicated whether the tumor should be considered malignant, benign, or of indeterminate histology. The system of Hough et al requires inclusion of clinical data in association with the histologic parameters to arrive at the final score. We used only the histologic parame- ters of malignancy as described previously by Gandour and Grizzle.8
The age and sex of the patients, the signs and symptoms at diagnosis, clinical history, and the volume and weight of each tumor were obtained from the clinical records and pathology reports.
Biochemical evaluations were performed by measuring urinary excretion of 17-hydroxycorticosteroid (17-OH),9 17- ketosteroids (17-KS),10 and dehydroepiandrosterone sulfate (DHA)” before and after dexamethasone administration.12 Normal ranges for these urinary steroids have been estab- lished previously.13-15 In recent years, plasma levels of DHA, as well as testosterone, aldosterone levels, and cortisol have also been determined by using commercially available re- agents (Travenol Laboratories Inc, Cambridge, MA; Diagnos- tic Products Corp, Los Angeles, CA).
A patient was considered to have Cushing’s syndrome if he or she had moon facies, weight gain, centripetal distribution of fat (abdomen and upper dorsal region), plethora, hyperten- sion, striae, and easy bruising.16 A girl was considered to be virilized if she had one or more of the following signs: clitorimegaly, hirsutism, deepened voice, and male muscula- ture. Precocious puberty was the criterion for boys. Patients with clinical features of virilization and Cushing’s syndrome or high 24-hour urinary excretion levels of 17-OH were designated as having “mixed”-type tumors; patients with no clinically evident endocrine syndrome were considered to have nonfunctional tumors.
Various modes of therapy were used for different patients in this series. Surgery was considered initially for every patient, and complete resection was attempted whenever possible. Patients with inoperable tumors or metastatic dis- ease at diagnosis were given a variety of treatments including radiotherapy and chemotherapy that included fluorouracil,
doxorubicin, cyclophosphamide, and O,P’-DDD (1,1 dichloro- 2-[O-chlorophenyl]-2-[p-chlorophenyl]-ethane. One patient received eight cycles of cisplatin and etoposide after complete resection of a recurrent tumor.
Statistical Analysis
For each continuous prognostic variable, patients were divided into two groups (with a minimum of five patients in each group), so that the difference in event-free survival duration would be maximized.17 We define the cut-point separating the two groups as the best discriminatory value. To establish the best cut-point for tumor volume and weight, the eight patients whose tumors were classified as unresect- able were excluded from the univariate analysis. The propor- tional hazards regression model of Cox18 was used to assess the multivariate significance of each prognostic variable. DHA was not included because these data were available for only 21 evaluable patients. (Having selected the cut-point from univariate analysis, the eight patients with unresectable tumors were included in the worse-risk group for tumor volumes.) Event-free survival was defined as the interval from complete tumor resection to relapse. A time of zero was assigned to those patients whose tumors were incompletely resected. Pearson’s r was used to evaluate the correlation among variables.
RESULTS
The diagnosis of ACC was made in most (72.5%) of the patients 6 months or more after appearance of clinical symptoms; it was made in only 11 patients within the first 6 months of symptoms of the disease.
The spectrum and incidence of the major presenting signs and symptoms are listed in Table 1. Virilization, manifested by premature appearance of pubic and axillary hair and clito-
| Clinical Features | Patients (N = 40) | Percent |
|---|---|---|
| Premature pubic and axillary hair | 37 | 92.5 |
| Clitorimegaly | 26* | 92.0 |
| Phallomegaly | 11ț | 91.6 |
| Hirsutism | 25 | 62.5 |
| Palpable abdominal mass | 22 | 55.0 |
| Hypertension | 19 | 47.5 |
| Acne | 19 | 47.5 |
| Plethora | 17 | 42.5 |
| Moon facies | 14 | 35.0 |
| Deepened voice | 13 | 32.5 |
| Seizure | 7 | 17.5 |
| Headache | 5 | 12.5 |
| Weight loss | 2 | 5.0 |
| Polyphagia and polydipsia | 3 | 7.5 |
*Girls only (26 of 28).
+Boys only (11 of 12).
ADRENOCORTICAL CARCINOMA
rimegaly or phallomegaly, was the predominant feature, affecting 37 of the 40 children. Hirsut- ism (62.5% of patients), acne (47.5%), hyperten- sion (47.5%), plethora (42.5%), moon facies (35%), deepened voice (32.5%), and seizure (17.5%) were common clinical features. Abdom- inal mass was palpable in 22 patients (55.5%). Consistent with the biology of this disease, an abdominal mass was not the dominant clinical manifestation; rather, most symptoms could be attributed to steroid production by the tumor. Thirty-nine patients had functional tumors. Of the 37 patients who presented with virilization, at least 22 had clinical or laboratory features of Cushing’s syndrome, two patients presented with signs of Cushing’s syndrome alone, and one had no symptoms of endocrine disturbance. Some patients were so virilized that these clinical features may have masked those of Cushing’s syndrome. The girl with no endocrine distur- bance had an abdominal mass that was detected during the neonatal period.
Of the 30 patients whose bone age was evalu- ated, 25 had a bone age that was disproportion- ately increased relative to their chronologic age. Figure 2 depicts the distribution, by percentiles, of the heights of 38 patients and the weights of 40. Seven patients’ heights and nine patients’ weights were above the 97th percentile; weight loss was uncommon (5%).
Hemoglobin values ranged from 6.2 g/dL to 17.2 g/dL (median, 12 g/dL). Leukocyte counts ranged from 5,200/mm3 to 29,800/mm3 (me- dian, 12,000/mm3). Baseline 24-hour urinary excretion of 17-KS ranged from 2.8 mg/day to
11
☐ Height
10
Weight
9
Number of Cases
8
7
6
5
4
3
2
1
0-3
3-10
10-25
25-50
50-75
75-90
90-97
97-100
Percentile
| Patient n | 17-OH (mg/m2/day) | 17-KS (mg/day) | DHA (mg/day) |
|---|---|---|---|
| 1 | 4.5 | 18.6 | 9.9 |
| 2 | 18.9 | 78 | 75.8 |
| 3 | 13.9 | 49 | NA |
| 4 | 7.1 | 29 | 20.1 |
| 5 | 13.5 | 211 | 1.0 |
| 6 | 17.6 | 125 | 98 |
| 7 | 8.0 | 8.9 | 8.0 |
| 8 | NA | 194 | 96.7 |
| 9 | 82.2 | 109 | 8.9 |
| 10 | 2.7 | 10.8 | NA |
| 11 | 17.1 | 35.9 | NA |
| 12 | 5.2 | 67 | NA |
| 13 | NA | 255 | 60 |
| 14 | 7.9 | 168 | 184 |
| 15 | 9.0 | 83.4 | 7.5 |
| 16 | 14.7 | 41 | NA |
| 17 | NA | 125.7 | NA |
| 18 | NA | NA | NA |
| 19 | 4.3 | 21.4 | 10.6 |
| 20 | 2.3 | 4.6 | 0.6 |
| 21 | 1.3 | 7.7 | 0.02 |
| 22 | 8.0 | 6.1 | NA |
| 23 | NA | NA | NA |
| 24 | 15.5 | 8.3 | 0.03 |
| 25 | 0.9 | 76.7 | 41.6 |
| 26 | 3.2 | 25.2 | 3.3 |
| 27 | 6.2 | 47.2 | 31.8 |
| 28 | NA | NA | NA |
| 29 | 1.8 | 8.0 | NA |
| 30 | 25.2 | 23 | NA |
| 31 | 12.6 | 3.9 | 3.0 |
| 32 | 15.6 | 9.6 | NA |
| 33 | 6.0 | 23.1 | NA |
| 34 | 9.6 | 8.9 | 3.8 |
| 35 | 1.4 | 4.9 | 0.3 |
| 36 | NA | NA | NA |
| 37 | NA | 2.8 | NA |
| 38 | NA | NA | NA |
| 39 | 19.5 | 6.7 | 1.0 |
| 40 | 4 | 73.9 | 46.4 |
Abbreviations: 17-OH, 17-hydroxycorticosteroids; 17-KS, 17- ketosteroids; DHA, dehydroepiandrosterone; NA, not available.
255 mg/day (median, 25.2 mg/day; normal, 0.5 to 18 mg/m2), 17-OH from 0.9 mg/m2/day to 82 mg/m2/day (median, 11.5 mg/m2/day; normal 3.0 ± 1.0 mg/m2/day), and DHA from 0.02 mg/day to 184 mg/day (median, 8.9 mg/day; normal, 0.7 - 3.9 mg/day) (Table 2). The uri- nary excretion of 17-KS, 17-OH, and DHA was elevated in 35 of 35, 21 of 31, and 20 of 23 patients, respectively (Table 2). The administra- tion of dexamethasone had no effect on 17-KS excretion, but 17-OH excretion was suppressed
in three patients. Urinary excretion of 17-OH is thought to be prognostically significant. In our group, 21 of 31 patients had excretion levels above the normal range (Fig 3). Of the 10 patients whose baseline 24-hour excretion of 17-OH was within the normal range, eight (80%) are alive; whereas of the 19 patients with ele- vated levels of 17-OH and adequate follow-up, only six (31.5%) are alive.
The weight, volume, and histology of the tumors is shown in Table 3. Tumor volume ranged from 4.5 cm3 to 1460 cm3 (median, 396 cm3) and tumor weight from 3.0 g to 1320 g (median, 420 g). There was a strong correlation (r = . 97) between tumor volume and weight. Of 33 patients for whom tumor specimens were available, only one (no. 32) had tissue that met criteria for a benign classification in all three systems whereas the tumors from two patients (no. 20 and 39) were classified as benign in two systems and those from two others (no. 3 and 37) were benign in only one system. The remaining tumors without benign histology met the criteria for malignancy in at least two of the three classification systems.
Thirty-eight patients underwent surgery. Tu- mors were considered to be completely resected in 26 and partially resected in six. Six others had biopsies only. Five patients had metastatic dis- ease at diagnosis, four in the lungs and one in the lungs and liver. Of the 26 patients whose tumors were thought to be completely resected, five relapsed and four were lost to follow-up. Of the five patients who relapsed, four died of progres- sive disease and one is alive 18+ months after a
third resection of a recurrent tumor and eight monthly cycles of cisplatin and etoposide.
Of the remaining 14 patients, 12 had unresect- able tumors, 11 died of progressive disease, and one was lost to follow-up. The diagnosis was confirmed at autopsy in two children who died before it could be established if their tumors were resectable.
Because no large series of children with ACC has been described, the clinical and biologic characteristics associated with prognosis have not been firmly established. The data in this series of patients were incomplete; four patients were lost to follow-up after resection of their tumors. For the remaining 36 assessable patients, complete information was not available for all variables included in the statistical analysis. The interval between first signs and diagnosis was lacking in one case; 24-hour urinary excretion levels of 17-KS and 17-OH were not recorded for five and nine patients, respectively; tumor weights and volumes were missing for 12 and seven patients; and tumor material was unavailable for reevaluation in seven cases. By univariate analy- sis, we determined the best discriminatory value of each variable and its influence on the risk of treatment failure: age more than 3.5 years, interval more than 6 months between the first signs and diagnosis, urinary excretion of 17-OH ≥ 4.4 mg/m2/day, 17-KS ≥ 24.0 mg/day, DHA ≥ 46.6 mg/day, tumor volume ≥ 200 cm3, and tumor weight ≥ 80 g. Since determination of the best discriminatory value involved multiple test- ing with different groupings of patients, the significance of these findings must be interpreted
A
80
30
D
17OH (mg/m2/24h)
20
D
A
A
D
A
LTF
D
D
D
D
10
LTF
A
LTF
D
D
D
5
A
D
A
D
D
D
0
?
1
2
Ca
V
5
O
7
9
10
11
12
14
15
16
19
2 12
21
22
24
25
26
27
29
30
31
32
33
34
5
39 40
Patient Number
| Patient Number | Volume (cm3) | Weight (g) | Histologic Criteria* | ||
|---|---|---|---|---|---|
| Hough et al' | Weiss7 | van Slooten et al | |||
| 1 | 4.5 | NA | M | M | M |
| 2 | 825.0 | 800.0 | M | M | M |
| 3 | 396.0 | 60.0 | B | M | 1 |
| 4 | 175.0 | NA | M | M | M |
| 5 | 910.0 | 460.0 | 1 | M | M |
| 6 | 1190.0 | 1040.0 | M | M | M |
| 7 | 441.0 | 175.0 | M | M | M |
| 9 | 800.0 | 420.0 | M | M | M |
| 10 | 337.0 | 220.0 | M | M | M |
| 12 | 200.0 | 210.0 | I | M | M |
| 16 | NA | NA | M | M | M |
| 17 | NA | 1320.0 | M | M | M |
| 18 | 715.0 | 390.0 | M | M | M |
| 19 | 36.0 | NA | M | M | M |
| 20 | 72.0 | 14.0 | I | B | B |
| 21 | NA | NA | 1 | M | M |
| 23 | 30.0 | NA | M | M | M |
| 24 | 56.0 | 40.0 | 1 | M | M |
| 25 | 70.0 | 70.0 | M | M | M |
| 26 | 72.0 | 15.0 | I | M | M |
| 27 | 90.0 | 65.0 | 1 | M | M |
| 28 | NA | NA | M | M | M |
| 29 | NA | NA | M | M | M |
| 30 | 1460.0 | 800.0 | M | M | M |
| 31 | NA | NA | M | M | M |
| 32 | NA | NA | B | B | B |
| 33 | 7.0 | 15.0 | M | M | M |
| 34 | 6.0 | 3.0 | I | M | M |
| 35 | NA | NA | I | M | M |
| 37 | 6.0 | NA | I | B | 1 |
| 38 | NA | NA | M | M | M |
| 39 | 36.0 | 17.0 | B | B | 1 |
| 40 | NA | NA | M | M | M |
*See Materials and Methods.
Abbreviations: M, malignant; B, benign; I, indeterminate; NA, not available.
with caution. However, significant differences between groups were noted in all possible values for age, 17-OH excretion, and tumor volume.
Because many of these variables were interre- lated, a multivariate analysis including all vari- ables except tumor size disclosed that age ≤ 3.5 years, interval between first signs and diagnosis ≥ 6 months, and levels of urinary steroids were each significantly associated with prognosis (Ta- ble 4); however, when tumor volume was in- cluded in the analysis, it was the only factor that retained prognostic significance. Tumor size is highly correlated (r = 0.50, P = 0.006) with later diagnosis and, to a lesser degree (r = 0.37, P = 0.05), with older age indicating that pa- tients diagnosed more than 6 months after first
symptoms and/or older than 3.5 years tend to have tumors greater than 200 cm3. In this in- stance, tumor size alone is sufficient to indicate those patients at greater risk of treatment failure. Figure 4A shows the adverse impact on event- free survival produced by tumors 200 cm3 or larger. Event-free survival for the entire group is presented in Fig 4B.
DISCUSSION
In a recent review of the literature, Humphrey et al19 analyzed the data on 72 children with ACC, reporting that age up to 7 years was associated with a 5-year survival rate of 53% compared with 17% for patients at least 9 years of age. Additionally, among the 40 patients in the younger age group, 74% of patients with tumor weights less than 170 g had a 5-year survival rate compared with 36% in patients with larger tumor weights. Our observations generally agree with theirs. However, we used a data-adaptive proce- dure to divide the patients into two groups, obtaining smaller values for both age (3.5 years) and tumor weight (70 g). We elected to use tumor volume instead of weight in the multivari- ate analysis because of the strong correlation between the two measures (r = . 97) and the fact that volume can be determined easily during routine diagnostic imaging procedures. Addition- ally, we found the interval between first signs and diagnosis to be important in prognosis.
Tumor histology has been considered by many to predict the course of the disease.4-6,8 Several of these investigators have attempted to identify specific histologic features for tumors from pa- tients who developed metastases after so-called curative resection of their tumors. By using three of these classification systems,5-7 we found that tumors from only four (12.5%) patients (no. 20, 32, 37, and 39) did not meet the criteria for classification as malignant in any of the three systems. These findings agree with those reported by Hough et al,5 who classified only two of nine pediatric adrenocortical tumors as benign. Of the four patients in our study whose tumors had no malignant histology, three are surviving disease- free for 8 + years, 5+ years, and 6+ months, and one patient is lost to follow-up. Of the 29 patients whose tumors had malignant histology, 14 (48%) are in complete remission, results similar to those reported by Cagle et al.20
Recently, molecular and cytogenetic tech-
| Variable | Number of Patients | Failure | P Value From Univariate Analysis | P Value From Multivariate Analysis | ||
|---|---|---|---|---|---|---|
| N | Y | |||||
| Age (yr) | ≤ 3.5 | 18 | 12 | 6 | <0.01 | NS |
| > 3.5 | 18 | 5 | 13 | |||
| Sex | ||||||
| Girls | 27 | 12 | 15 | NS | NS | |
| Boys | 9 | 5 | 4 | |||
| Interval (mo) | ≤ 6 | 13 | 10 | 3 | <0.01 | NS |
| > 6 | 22 | 6 | 16 | |||
| 17-KS (mg/day) | ≤ 21.4 | 12 | 8 | 4 | <0.01 | NS |
| > 21.4 | 19 | 6 | 13 | |||
| 17-OH (mg/day) | ≤ 5.2 | 11 | 7 | 4 | <0.01 | NS |
| > 5.2 | 17 | 6 | 11 | |||
| DHA (mg/day) | ≤ 41.6 | 15 | 11 | 4 | <0.01 | NS |
| > 41.6 | 6 | 1 | 5 | |||
| Tumor volume (cm3) | ≤ 176 | 12 | 11 | 1 | <0.01 | 0.01 |
| > 200 | 9 | 2 | 7 | |||
| Tumor weight (g) | ≤70 | 7 | 6 | 1 | <0.01 | NI |
| ≥ 80 | 9 | 2 | 7 | |||
| Histology | ||||||
| Benign | 4 | 3 | 1 | NS | NS | |
| Malignant | 25 | 14 | 11 | |||
Abbreviations: NS, not significant; NI, not included in the model.
niques have been used to assign risk status to newly diagnosed patients with a variety of solid tumors.21-24 The results of flow cytometric DNA analysis of pediatric adrenocortical tumors by Taylor et al suggest that the treatment response potential of these tumors can be predicted.25 Although the prognostic usefulness of these bio- technologies is currently limited, our ability to predict treatment outcome and to improve treat- ment strategies for children with ACC should be markedly improved by increasing our knowledge through using these techniques.
The predominant clinical manifestation of ACC in this group of patients was virilization alone or in association with Cushing’s syndrome. Cushing’s syndrome alone, or nonfunctional tu- mors, was uncommon in our patients. About 40% of adult patients have nonfunctional ACC29-31; moreover, Cushing’s syndrome is the most com- mon endocrine syndrome among the functional tumors. Sex distribution is also peculiar to chil- dren with ACC. In children, and more so among very young children, there is a striking predomi- nance of girls.32,33 In our series, the female:male ratio was 4:1 in children younger than 4 years old. Signs and symptoms preceded the diagnosis of tumor by 6 months or longer in 22 of the children, possibly reflecting the general lack of medical facilities in this region or, more likely,
the tumor’s tendency to grow slowly. Moreover, in contrast to the weight loss associated with most pediatric tumors, ACC increases the body mass and, to a certain extent, increases the healthy appearance of affected children. A per- ceptible abdominal mass is usually noted late in the course of the disease. The pediatrician evalu- ating a child with signs of virilization should be aware that small tumors can have significant clinical manifestations and consider this tumor in the differential diagnosis of abnormal sexual development.
Surgery, with complete tumor resection, is considered curative for most children with ACC. In our series, tumor resection was considered complete in 26 patients, and five of 22 evaluable patients had tumor recurrence. One of these patient’s tumor was resected three times, and, after the last surgery, eight monthly cycles of cisplatin and etoposide were given. He is surviv- ing disease-free 18+ months from the third surgery. The remaining four patients died of progressive disease in spite of radiotherapy and/ or chemotherapy, principally with O,P’DDD. This drug has been used to treat inoperable, recurrent, and metastatic ACC, with response rates ranging from 10% to 60% in different series of patients.31-33 There have been indications that the drug is more effective in children than in
Proportion of Patients Not Failing
1.0
A
L
1
.L
IT
0.8
<200 cm (n=12)
0.6
0.4
0.2
≥200 cm (n=17)
0.0
0
1
2
3
4
5
6
7
8
9
Years
Proportion of Patients Not Failing
1.0
B
0.8
0.6
11
0.4
(n=36)
0.2
0.0
0
1
2
3
4
5
6
7
8
9
19
Years
adults34; moreover, there have been reports of well-documented case histories of ACC possibly having been cured by O,P’DDD therapy.33 Radiotherapy35 and other antineoplastic drugs have been used,36,37 but their efficacy has not been established.
For reasons yet to be determined, the incidence of this tumor seems to be relatively high in southern Brazil.2 ACC is known to be part of a familial syndrome that includes sarcomas, brain tumors, breast cancer, and leukemias.26 The extended pedigrees of many of these affected families suggest that the apparent susceptibility to certain types of cancer is inherited in an autosomal dominant manner.27 The retrospective nature of our study precluded the construction of extended pedigrees of most of the families in this series. One intriguing observation by Hartley et al is that four mothers and seven fathers of the 14 patients with ACC registered in the Manchester Children’s Tumor Registry had been exposed to potentially toxic substances before or during the index pregnancy.28 Southern Brazil is a predomi- nantly agricultural region where chemical fertil-
izers, insecticides, and herbicides are used indis- criminately. Future case-controlled studies are warranted to determine what, if any, genetic or environmental factors are etiologically associ- ated with ACC.
In summary, ACC occurs rarely in children, the vast majority of whom, when affected, present with marked signs of abnormal sexual develop- ment. Early diagnosis increases the likelihood of cure by surgery. Careful evaluation of clinical and biologic characteristics at diagnosis such as age, interval from first signs, and tumor size will permit the discrimination of those children at higher risk of tumor recurrence. For these chil- dren or those with inoperable tumors or metasta- sis at diagnosis, a trial of O,P’DDD is probably warranted. For those children whose tumors fail to respond to O,P’DDD, a combination of cispla- tin and etoposide should also be considered.
ACKNOWLEDGMENT
The authors thank Peggy Vandiveer for typing and Linda Daniels for editorial review of the manuscript.
REFERENCES
1. Brennan MF: The adrenal gland, in DeVita VT Jr., Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology, (ed 2), Philadelphia, PA; Lippincott, 1985; pp 1191-1206
2. Marigo C, Muller H, Davies JNP: Survey of cancer in children admitted to a Brazilian charity hospital. J Natl Cancer Inst 43:1231-1240, 1968.
3. Birch JM, Blair V: Increase in childhood carcinomas in North-West England. Lancet 1:833, 1988 (letter)
4. Schteingart DE, Oberman HA, Friedman BA, et al: Adrenal cortical neoplasms producing Cushing’s syndrome: A clinicopathologic study. Cancer 22:1005-1013, 1968
5. Hough AJ, Hollifield JW, Page DL, et al: Prognostic factors in adrenal cortical tumors. A mathematical analysis of clinical and morphologic data. Am J Clin Pathol 72:390- 399, 1979
6. van Slooten H, Schaberg A, Smeenk D, et al: Morpho- logic characteristics of benign and malignant adrenocortical tumors. Cancer 55:766-773, 1985
7. Weiss LM: Comparative histologic study of 43 metasti- sizing and nonmetastisizing adrenocortical tumors. Am J Surg Pathol 8:163-169, 1984
8. Gandour MJ, Grizzle WW: A small adrenocortical carcinoma with aggressive behavior. An evaluation of criteria for malignancy. Arch Pathol Lab Med 110:1076-1079, 1986
9. Callow NH, Callow RK, Eminens CW: Colorimetric determination of substances containing the groupings CH2CO in urine extracts as an indication of androgen content. Biochem J 32:1312-1331, 1938
10. Silber RH, Porter CC: The determination of 17, 21-dehydroxy-20-ketosteroids in urine and plasma. J Biol Chem 210:923-932, 1954
11. Buster JE, Abraham GE: Radioimmunoassay of plasma dehydroepiandrosterone. Anal Lett 5:203-215, 1972
12. Migeon CJ, Green OC, Eckert JP: Study of adrenocor- tical function in obesity. Metabolism 12:718-739, 1963
13. Tulchinsky D, Simmer HH: Sources of plasma 17 alpha-hydroxyprogesterone in human pregnancy. J Clin En- docrinol Metab 35:799-808, 1972
14. Wilkins L: Diagnostic methods bioassay and chemi- cals, in Wilkins L, Blizzard RM, Migean CJ (eds): The Diagnosis and Treatment of Endocrine Disorders in Child- hood and Adolescence (ed 3). Springfield, IL, Charles C. Thomas Publishers, 1965, pp 45-72
15. Kiorschner MA, Lipset MB, Wilson H: Metabolites of exogenous dehydroepiandrosterone in man. Acta Endocrinol (KBV) 43:387-398, 1963
16. Urbanic RC, George JM: Cushing’s disease: 18 years experience. Medicine 60:14-24, 1981
17. Cox DR, Lewis PAW: The Statistical Analysis of Series of Events, London, Chapman and Hall, 1966
18. Cox DR: Regression models and life-tables (with discussion). J R Stat Soc B 34:187-220, 1972
19. Humphrey GB, Pysher T, Holcombe J, et al: Overview on the management of adrenocortical carcinoma (ACC). Cancer Treat Res 17:349-358, 1983
20. Cagle PT, Hough AJ, Pysher J, et al: Comparison of adrenal cortical tumors in children and adults. Cancer 57:2235-2237, 1986
21. Brodeur GM, Seeger RC, Schwab M, et al: Amplifica- tion of N-myc in untreated human neuroblastomas correlates with advanced disease stage. Science 224:1121-1124, 1984
22. Nau MM, Brooks BJ Jr, Carney DN, et al: Human small-cell lung cancers show amplification and expression of the N-myc gene. Proc Natl Acad Sci 83:1092-1096, 1986
23. Slamon DJ, Clark GM, Wong SG, et al: Human breast cancer: Correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235:177- 182, 1987
24. Riou G, Barrois M, Le MG, et al: C-myc proto- oncogene expression and prognosis in early carcinoma of the uterine cervix. Lancet 1:761-763, 1987
25. Taylor SR, Roederer M, Murphy RF: Flow cytomet- ric DNA analysis of adrenocortical tumors in children. Cancer 59:2059-2063, 1987
26. Kenny FM, Hashida Y, Askari HA, et al: Virilising tumors of the adrenal cortex. Am J Dis Child 115:445-448, 1968
27. Lynch HT, Mulcahy GM, Harris RE, et al: Genetic and pathologic findings in a kindred with hereditary sarcoma, breast cancer, brain tumors, leukemia, lung, laryngeal, and adrenal cortical carcinoma. Cancer 41:2055-2064, 1978
28. Hartley AL, Birch JM, Marsden HB, et al: Adrenal cortical tumours: Epidemiological and familial aspects. Arch Dis Child 62:683-689, 1987
29. Schteingart DE, Motazedi A, Noonan RA, et al: Treatment of adrenal carcinomas. Arch Surg 117:1142-1146, 1982
30. Henley DJ, van Heerden JA, Grant CS, et al: Adrenal cortical carcinoma-a continuing challenge. Surgery 94:926- 931, 1983
31. Bertagna C, Orth DN: Clinical and laboratory find- ings and results of therapy in 58 patients with adrenocortical tumors admitted to a single medical center (1951-1978). Am J Med 71:855-875, 1981
32. Hajjar RA, Hickey RC, Samaan NA: Adrenal corti- cal carcinoma. A study of 32 patients. Cancer 35:549-554, 1975
33. LeFevre M, Gerard-Marchant R, Gubler JP, et al: Adrenal cortical carcinoma in children: 42 patients treated from 1958 to 1980 at Villejuif. Cancer Treat Res 17:265-276, 1983
34. Grieg F, Oberfield SE, Levin LS, et al: Recovery of adrenal function after treatment of adrenocortical carcinoma with O,P’DDD. Clin Endocrinol 20:389-399, 1984
35. Stewart DR, Jones PHM, Jolleys A: Carcinoma of the adrenal gland in children. J Pediatr Surg 9:59-67, 1974
36. Brennan MF: Adrenocortical carcinoma. CA 37:348- 365, 1987
37. Johnson DH, Greco A: Treatment of metastatic adre- nal cortical carcinoma with cisplatin and etoposide (VP-16). Cancer 58:2198-2202, 1986