WILEY
Pediatric Blood & Cancer
SOCIÉTÉ INTERNATIONALE D’ONCOLOGIE PÉDIATRIQUE SKOP INTERNATIONAL SOCIETY OF PAEDIATRIC ONCOLOGY
aspho The American Society of Pediatric Hematology/Oncology
Outcome and prognostic factors in high-risk childhood adrenocortical carcinomas: A report from the European Cooperative Study Group on Pediatric Rare Tumors (EXPERT)
Giovanni Cecchetto1 Alba Ganarin1 Ewa Bien2 Peter Vorwerk3
Gianni Bisogno4 ☐ Jan Godzinski5 Patrizia Dall’Igna1 ☐ Yves Reguerre6
Dominik Schneider7 ☒ Laurence Brugières8 Pierre Leblond9 Andrea Ferrari10
Ines Brecht11 Angela De Paoli12 Daniel Orbach13
1 Pediatric Surgery Unit, Department of Woman’s and Children’s Health, University Hospital of Padua, Padua, Italy
2 Department of Pediatrics, Medical University, Gdansk, Poland
3 Pediatric Oncology Department, Otto von Guericke University Children’s Hospital, Magde- burg, Germany
4 Pediatric Hematology-Oncology Unit, Depart- ment of Woman’s and Children’s Health, Univer- sity Hospital of Padua, Padua, Italy
5 Department of Pediatric Surgery, Marciniak Hospital, Wroclaw, Poland
6 Oncology and Hematology Unit, French Pedi- atric Rare Tumor Group (groupe Fracture), CHU de Saint Denis de La Réunion. Saint Denis, France
7 Clinic of Pediatrics, Municipal Hospital Dort- mund, Dortmund, Germany
8 Pediatric Department, Gustave Roussy Cancer Campus, Villejuif, France
9 Pediatric Department, Centre Oscar Lambret, Lille, France
10 Fondazione IRCCS Istituto Nazionale Tumori Milano, Milan, Italy
11 University Children’s Hospital, Erlangen, Germany
12 Clinical Trials and Biostatistics Unit, IRCCS Istituto Oncologico Veneto, Padua, Italy
13 Department of Pediatric, Adolescent, Young Adult Oncology, French Pediatric Rare Tumor group (groupe Fracture), Institut Curie, Paris, France
Correspondence
Alba Ganarin, Pediatric Surgery Unit, Depart- ment of Woman’s and Children’s Health, Univer- sity Hospital of Padua, Padua, Italy. Email: alba.ganarin@gmail.com
Abstract
Objectives: The aim of this retrospective international analysis was to evaluate the role of risk factors in pediatric patients with adrenocortical carcinoma (ACC) observed in European countries (2000-2013) in an attempt to identify factors associated with poor prognosis.
Procedures: Data were retrieved from databases of Germany, France, Poland, and Italy, which form the European Cooperative Study Group on Pediatric Rare Tumors (EXPERT). Patients were less than 18 years old, with at least one of the following tumor-related risk factors: metastases, volume more than 200 cm3, Cushing syndrome, vascular or regional lymph node invasion, initial biopsy, or incomplete excision. Role of patients’ age was also evaluated.
Results: Eighty-two patients were evaluated: 62 with localized disease and 20 with metastases. The 3-year progression-free survival (PFS) and overall survival (OS) were 39% and 55% for the whole population, respectively, and 51% and 73% for localized diseases, respectively. Concerning the whole population, PFS and OS were influenced by distant metastases, tumor volume, lymph node involvement, age, and presence of two or more risk factors. Factors significant only at OS were vascular involvement and incomplete surgery. At multivariable analysis, the main factors at PFS were volume more than 200 cm3 (hazard ratio [HR]: 2.6, 95% confidence interval [CI]: 1.18- 5.70) and presence of distant metastases (HR: 8.26, 95% CI: 3.49-19.51). The OS was significantly influenced by the presence of metastases (P < 0.0001). Concerning patients with localized tumors, the only significant prognostic factor was volume more than 200 cm3 with a HR of 4.38 (95% CI: 1.60-12.00) for PFS and of 3.68 (95% CI: 1.02-13.30) for OS.
Conclusions: Distant metastases and large tumor volume were the main unfavorable prognostic factors. Presence of two or more factors related to ACC was associated with an aggressive behav- ior of disease.
KEYWORDS
adolescents, adrenocortical carcinoma, children, risk factors
1 INTRODUCTION
Adrenocortical tumors are rare in childhood, representing 0.2% of all pediatric tumors with an incidence of 0.2 new cases per 1 million chil- dren per year.1 The distinction between large benign adenoma and adrenocortical carcinoma (ACC) may be difficult, and the optimal treat- ment tricky: a complete surgical resection, which provides the best chance of cure,2,3 is not always possible and the role of the medi- cal therapy is controversial.4 The different staging systems for ACC adopted in the pediatric population do not facilitate comparative stud- ies, and the prognostic factors accepted for adults often lack sensitivity and specificity.5-7 As histologic malignant criteria are not completely shared by pathologists,8-10 a sharp demarcation between benign and malignant lesions, even after surgery, has not yet been identified, mak- ing it difficult to better delineate patients who potentially may need perioperative therapy.11 Since the range of survival of ACC is very large in the literature, the aim of this retrospective multicentric study was to analyze the outcome of patients affected by ACC with risk factors, observed in different European countries in the same period, and to assess the weight of each adverse prognostic factor.
2 PATIENTS AND METHODS
The study has been conducted by the members of the European Cooperative Study Group on Pediatric Rare Tumors (EXPERT group) founded in 2008 in order to achieve scientific and clinical improve- ment in very rare pediatric solid tumors.12 The analysis included only patients with a diagnosis of “high-risk” ACC, less than 18 years of age, and registered between 2000 and 2013 in the national databases of four countries (France, Germany, Poland, and Italy).
To be considered at “high risk,” the tumor had to show at least one of the following characteristics, related to the tumor, based on the lit- erature data2,4,13-17: volume more than 200 cm3, Cushing syndrome, initial biopsy (open or tru-cut), surgical excision with microscopic resid- uals or spillage (R1) or macroscopic residuals (R2), regional lymph node involvement, histologic vascular invasion, and distant metastases at diagnosis. Age at diagnosis was also evaluated as a risk factor.
The information collected from the database of each country included demographic details (age, gender, date of diagnosis), clinical features (associated syndromes/malformations, symptoms, hormone values), tumor characteristics (size of the primary tumor and extent of the disease), initial therapeutic approach, and outcome. Two different types of tumor staging were used: the UICC-TNM staging system,18 similar to that adopted for adults, and the Children’s Oncology Group staging system,19 which, besides the results of primary surgery, also considers the size of the tumor and the postoperative hormone val- ues. Since a single global re-staging was not feasible, tumor stage was not evaluated as a prognostic factor. However, factors determin- ing stage in both systems were included in the analysis. Informed
consent to data collection and analysis was managed according to national guidelines. Pathologic diagnosis had been reviewed at the national level, but no specific review was conducted for this study.
Overall, the therapy strategy was uniform, and surgery was exclu- sively performed at the diagnosis of localized disease if a microscopi- cally complete excision (R0) was feasible; biopsy was obtained in other cases if the hormonal pattern was not characteristic. Patients with initial resection with microscopic (R1) or macroscopic (R2) residu- als and those with inoperable tumors received chemotherapy and/or mitotane with various modalities. Delayed surgery was planned in case of response to chemotherapy. The details on medical treatments and secondary surgery were not specifically analyzed because the regi- mens varied over the years in each country and for each group of patients.
3 STAT IST ICAL ANALYS IS
The Clinical Trials and Biostatistics Unit at Veneto Institute of Oncol- ogy of Padua managed data collection and analysis. The survival time was calculated from the date of diagnosis to the time of last follow- up or event. Relapse or death for any cause was considered for progression-free survival (PFS). The overall survival (OS) was mea- sured from the date of diagnosis to death for any reason. The sur- vival probability was computed by means of the Kaplan-Meier method, and the heterogeneity in survival among strata of selected variables was assessed through the log-rank test. The 1- and 3-year PFS and OS were reported with their 95% confidence intervals (CIs). A P value less than 0.05 was considered statistically significant. Univariate and mul- tivariable analyses were conducted using the Cox proportional haz- ards regression method to identify the factors present at diagnosis that were associated with the risk of progression or death. A stepwise vari- able selection procedure was applied to identify a subset of covariates for the final model, considering all the variables with a P value of at least 0.05 at univariate analysis and evaluating each of the nonsignif- icant variables in a second step. Hazard ratios (HRs) with the 95% CI, calculated according to the Wald method, were reported for significant variables. All data analyses were performed using the SAS statistical package (SAS, release 9.2; SAS Institute Inc., Cary, North Carolina).
4 RESULTS
4.1 Clinical aspects and surgical treatment
The characteristics of 82 evaluable patients (62 with localized disease, 20 with metastases) with tumor features and initial surgery results are summarized in Table 1. The median age at diagnosis was 8.1 years (interquartile [IQ] range: 2.7-14.0): 5.2 years (IQ range: 2.2-13.5) for patients with localized disease and 13.4 years (IQ range: 8.2-15.4) for metastatic patients. Tumor volume more than 200 cm3 and vascular invasion were more frequent in metastatic cases. Cushing syndrome was evident in 25/80 patients at diagnosis (data not available for two patients with localized disease).
| Localized | Metastatic | Total | ||||
|---|---|---|---|---|---|---|
| N = 62 | % | N = 20 | % | N= 82 | % | |
| General features | ||||||
| Country | ||||||
| France | 9 | 15 | 4 | 20 | 13 | 16 |
| Germany | 25 | 40 | 5 | 25 | 30 | 37 |
| Italy | 15 | 24 | 6 | 30 | 21 | 26 |
| Poland | 13 | 21 | 5 | 25 | 18 | 22 |
| Age at diagnosis (years) | ||||||
| ≤4 | 26 | 42 | 2 | 10 | 28 | 34 |
| 4-10 | 14 | 23 | 5 | 25 | 19 | 23 |
| >10 | 22 | 36 | 13 | 65 | 35 | 43 |
| Risk factors related to the tumor | ||||||
| Volume >200 cm3 | ||||||
| Yes | 36 | 58 | 14 | 70 | 50 | 61 |
| No | 26 | 42 | 6 | 30 | 32 | 39 |
| Cushing symptoms | ||||||
| Yes | 18 | 29 | 7 | 35 | 25 | 31 |
| No | 42 | 68 | 13 | 65 | 55 | 67 |
| Unknown | 2 | 3.2 | – | – | 2 | 2 |
| Vascular invasion | ||||||
| Yes | 9 | 15 | 5 | 25 | 14 | 17 |
| No | 53 | 86 | 14 | 70 | 67 | 82 |
| Unknown | – | – | 1 | 5 | 1 | 1 |
| Lymph node involvement | ||||||
| Yes | 5 | 8 | 11 | 55 | 16 | 20 |
| No | 43 | 69 | 6 | 30 | 49 | 60 |
| Unknown | 14 | 23 | 3 | 15 | 17 | 21 |
| Type of initial surgery | ||||||
| R0 | 28 | 45 | 3 | 15 | 31 | 38 |
| R1 | 23 | 37 | 4 | 20 | 27 | 33 |
| R2 | 4 | 6 | 4 | 20 | 8 | 10 |
| Biopsy | 7 | 11 | 9 | 45 | 16 | 20 |
A genetic or malformative syndrome was present in 14% of the patients: Li-Fraumeni syndrome (six), kidney malformations (two), Ullrich-Turner syndrome (one), facial abnormalities (one), and micro- cephaly (one).
Thirty-eight percent of the patients underwent R0 resection at diagnosis: 45% with localized and 15% with metastatic disease. In two cases, the surgical excision was done immediately after a diagnostic biopsy. Four nephrectomies and one splenectomy were also required. Biopsy was the initial surgical approach in 19% of the patients: in 11% with localized and 45% with metastatic disease. The median follow- up for alive patients in the whole cohort was 53.3 months (IQ range: 10.8-81.1). A relapse was observed in 40 patients (28 with localized, 12 with metastatic ACC). The median time to relapse was 8.4 months (9.8 in localized, 4.2 in metastatic disease). Thirty-six patients died due
to disease: 29 after a relapse, one due to a second tumor, six due to unspecified reasons. At the end of the follow-up, 46 patients were alive: 29 in first complete remission (CR), seven in second CR, and 10 with disease. The 3-year PFS and OS for the whole population were 39% (95% CI: 27-50) and 55% (95% CI: 42-66), respectively. The PFS and OS for patients with localized and metastatic disease are shown in Figs. 1A and 1B, respectively.
4.2 Analysis of prognostic factors
4.2.1 Overall population
Table 2 summarizes the survival probability according to risk factors at univariate analysis. Patients less than or equal to 4 years of age had a better outcome than the older ones, both at 3-year PFS (67%) and OS (86%). Presence of metastases was the most important poor prognos- tic factor related to tumor both at 3-year PFS and OS (P < 0.0001) and it was always associated at least with another risk factor. All metastatic patients died within 15 months after diagnosis. The outcome of two other patients who were lost at follow-up at 1.3 and 8.5 months after diagnosis is not known.
Large tumor volume showed a statistically significant role only at PFS (P = 0.0046), while the presence of histologic vascular invasion was related to a poorer OS (P = 0.030). Lymph node involvement, avail- able for 65 patients, resulted significant at PFS (P = 0.0005) and OS (P = 0.0036). Although the presence of a Cushing syndrome was asso- ciated with poorer survival, this prognostic factor did not have a statis- tical impact.
The type of initial surgery was influenced by the initial tumor exten- sion and significant only at OS (P = 0.0026): in case of R0 resection, the 3-year OS was 72%; but if the initial surgery was only a biopsy, it decreases drastically to 14%. Intermediate values were associated with R1 (64%) and R2 (45%) resection.
An analysis of the value of the number of risk factors/patient on outcome considered 71 cases with the following information: tumor volume, Cushing syndrome, vascular invasion, initial biopsy, incomplete excision, and metastases (regional lymph node involve- ment was excluded due to incomplete data). One risk factor was present in 38% of the patients, two factors in 27% and at least three factors in 35%. PFS and OS were statistically different in the three groups (P = 0.0010 and <0.0001, respectively), with an evident better outcome in patients with only one risk factor (Figs. 2A and 2B).
A multivariable Cox regression analysis has been performed consid- ering 64 patients with all the data available (Table 3): only large tumor volume (HR: 2.6, 95% CI: 1.2-5.7, P = 0.017) and presence of metas- tases (HR: 8.3, 95% CI: 3.5-19.5, P < 0.0001) statistically affected PFS. OS was significantly influenced only by the presence of metastases (P <0.0001).
4.2.2 Localized tumors
Considering the severe impact of metastases on survival, a further analysis was performed on the 62 patients with localized disease (Table 4). Age older than 4 years was a significant risk factor either
A
B
1,0
1,0
0,8
Survival Probability
0,8
0,6
Survival Probability
0,6
Localised patients
Localised patients
0,4
All patients
All patients
0,4
Metastatic patients
0,2
Metastatic patients
0,2
0,0
| 0 | 20 | 40 | 60 Time | (months) 80 | 100 | 120 | 140 | 0,0 0 | 20 | 40 | 60 Time | (months) 80 | 100 | 120 | 140 | |||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| All | 82 | 28 | 23 | 15 | 9 | 4 | 3 | All | 82 | 50 | 31 | 23 | 14 | 6 | 3 | |||
| Localised | 62 | 28 | 23 | 15 | 9 | 4 | 3 | Localised | 62 | 48 | 31 | 23 | 14 | 6 | 3 | |||
| Metastatic | 20 | 0 | Metastatic | 20 | 2 | 0 | ||||||||||||
FIGURE 1 Estimated PFS (A) and OS (B) for the whole cohort of patients (39% [95% CI: 27-50] and 55% [95% CI: 42-66] at 3 years); for localized (51% [95% CI: 37-64] and 73% [95% CI: 58-83] at 3 years); and for metastatic patients (6.3% [95% CI: 0.4-25] and 29% [95% CI: 11-50] at 1 year) with high-risk ACC
in terms of PFS or OS. Among the risk factors related to tumor, only large volume showed a statistical influence at PFS (P = 0.0036). The pejorative impact of lymph node involvement seems significant, but it regards only a part of the evaluable patients. Initial surgery results
were associated with different survivals: complete excision had a bet- ter 3-year PFS (56%) than incomplete microscopic excision (47%) or initial biopsy (29%). Complete excision was associated also with a sig- nificant better 3-year OS. All four patients with localized tumors and
| Characteristic | N | No. of events | No. of deaths | 3-year PFS (95% CI) | P value | 3-year OS (95% CI) | P value |
|---|---|---|---|---|---|---|---|
| Age diagnosis (years) | 0.0146 | 0.0018 | |||||
| ≤4 | 28 | 9 | 5 | 67 (44-82) | 86 (62-95) | ||
| 4-10 | 19 | 13 | 9 | 21 (5.2-44) | 44 (18-67) | ||
| >10 | 35 | 25 | 22 | 28 (14-44) | 39 (22-55) | ||
| Volume >200 cm3 | 0.0046 | 0.0778 | |||||
| Yes | 50 | 36 | 27 | 22 (12-35) | 46 (31-60) | ||
| No | 32 | 11 | 9 | 71 (50-84) | 73 (51-86) | ||
| Cushing syndromeª | 0.3072 | 0.3664 | |||||
| Yes | 25 | 18 | 14 | 32 (15-50) | 50 (28-68) | ||
| No | 55 | 28 | 21 | 43 (28-57) | 59 (43-72) | ||
| Vascular invasionª | 0.3112 | 0.0301 | |||||
| Yes | 14 | 10 | 9 | 24 (5.7-48) | 27 (6.8-53) | ||
| No | 67 | 36 | 26 | 43 (30-56) | 62 (47-73) | ||
| Initial surgery result | 0.4582 | 0.0026 | |||||
| R0 | 31 | 15 | 10 | 51 (30-68) | 72 (49-85) | ||
| R1 | 27 | 16 | 10 | 40 (21-58) | 64 (40-80) | ||
| R2 | 8 | 5 | 4 | 47 (12-76) | 45 (11-75) | ||
| Biopsy | 16 | 12 | 12 | 15 (2.4-37) | 14 (2.3-37) | ||
| Metastases | <0.0001 | <0.0001 | |||||
| Yes | 20 | 18 | 18 | 0 | 0 | ||
| No | 62 | 29 | 18 | 51 (37-64) | 73 (58-83) | ||
| Lymph node involvement | 0.0005 | 0.0036 | |||||
| Yes | 16 | 14 | 12 | 0 | 29 (9.2-53) | ||
| No | 49 | 24 | 18 | 51 (35-65) | 62 (45-75) |
ªThe sum does not add up to the total because of missing values. In bold type: P value less than 0.05, considered statistically significant.
A
B
1.0
1.0
0.8
0.8
1 unfavourable factor
Survival Probability
Survival Probability
0.6
1 unfavourable factor
0.6
2 unfavourable factors
0.4
0.4
2 unfavourable factors
3 or more unfavourable factors
0.2
0.2
3 or more unfavourable factors
0.0
0.0
| 0 | 20 | 40 | 60 | 80 | 100 | 120 | 140 | 0 | 20 | 40 | 60 | 80 | 100 | 120 | 140 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Patients at risk: | Time | (months) | Patients at risk: | Time | (months) | |||||||||||
| 1 unfav. factor | 27 | 16 | 14 | 8 | 4 | 2 | 1 unfav. factor | 27 | 22 | 17 | 13 | 7 | 4 | 2 | ||
| 2 unfav. factors | 19 | 4 | 2 | 1 | 1 | 0 | 2 unfav. factors | 19 | 13 | 6 | 4 | 3 | 0 | |||
| 3 unfav. factors | 25 | 4 | 4 | 4 | 2 | 1 | 3 unfav. factors | 25 | 11 | 5 | 4 | 2 | 1 | |||
FIGURE 2 Estimated PFS (A) and OS (B) according to the number of risk factors related to the tumor (71 patients). The 3-year PFS for patients with one, two, or three/more unfavorable factors was, respectively, 65% (95% CI: 42-81), 17% (95% CI: 3.3-39), or 17% (95% CI: 5.2-34). P = 0.0010. The 3-year OS for patients with one or two or three/more unfavorable factors was, respectively, 86% (95% CI: 62-95), 46% (95% CI: 21- 68), or 25% (95%: 9.6-45). P < 0.0001
| PFS | OS | ||||||
|---|---|---|---|---|---|---|---|
| Characteristic | No. | Events | Hazard ratio (95% CI) | P value | Deaths | Hazard ratio (95% CI) | P value |
| Whole cohort | |||||||
| Volume >200 cm3 | 0.017 | ||||||
| Yes | 36 | 28 | 2.6 (1.2-5.7) | ||||
| No | 28 | 9 | 1ª | ||||
| Presence of metastases | <0.0001 | <0.0001 | |||||
| Yes | 17 | 15 | 8.3 (3.5-19.5) | 15 | 36 (11-120) | ||
| No | 47 | 22 | 1ª | 22 | 1ª | ||
| Localized tumours | |||||||
| Volume >200 cm3 | 0.0041 | 0.047 | |||||
| Yes | 24 | 17 | 4.4 (1.6-12) | 11 | 3.7 (1-13) | ||
| No | 23 | 5 | 1ª | 3 | 1ª | ||
a Reference category.
R2 excision survived, but this result is most likely due to a bias related to the low number of patients. At multivariable Cox regression anal- ysis performed on 47 patients with all the data available, tumor vol- ume was the only significant prognostic factor (Table 3). The HR for PFS was 4.4 (95% CI: 1.6-12, P= 0.0041) in patients with tumor more than 200 cm3 and 3.7 (95% CI: 1.0-13, P = 0.047) for OS.
5 DISCUSSION
The prognostic stratification of ACCs is still a challenging issue due to their rarity, variable presentation, and difficulty in histologic definition11,20,21 (Supplementary Table S1). This article represents a report of the EXPERT, a European network dedicated to data collec- tion and study of pediatric rare tumors. As it happened for different
tumors in previous initiatives,22-24 the EXPERT decided to combine the European data on ACC collected by each national group with the aim to evaluate the main prognostic factors in a large series of patients treated recently. Because the range of survival for patients with ACC is very large in literature, we decided to consider only those patients with at least one adverse prognostic factor in order to evaluate the most important ones.
The rarity of these tumors is confirmed also in this analysis; only 82 patients with high-risk ACC could be evaluated from four national database in a period of 14 years. This series of patients allowed to per- form an analysis of different known risk factors.
We underline that the impact of chemotherapy and mitotane was not evaluated due to the different regimens adopted and the lack of comparative evaluable data. The 3-year OS and PFS in this series are quite low as compared with the data of literature,5,20,21,25,26 but this poor outcome is justified by the fact that we selected only cases that
| Characteristic | N | No. of events | No. of deaths | 3-year PFS (95% CI) | P value | 3-year OS (95% CI) | P value |
|---|---|---|---|---|---|---|---|
| Age diagnosis (years) | 0.049 | 0.028 | |||||
| ≤4 | 26 | 7 | 3 | 73 (48-87) | 94 (67-99) | ||
| 4-10 | 14 | 9 | 5 | 28 (6.8-55) | 60 (25-83) | ||
| >10 | 22 | 13 | 10 | 43 (22-63) | 65 (41-82) | ||
| Volume >200 cm3 | 0.0036 | 0.067 | |||||
| Yes | 36 | 23 | 14 | 31 (16-47) | 64 (44-78) | ||
| No | 26 | 6 | 4 | 85 (61-95) | 89 (62-97) | ||
| Cushing syndromeª | 0.49 | 0.38 | |||||
| Yes | 18 | 11 | 7 | 44 (22-65) | 69 (41-86) | ||
| No | 42 | 17 | 10 | 56 (37-71) | 77 (58-89) | ||
| Vascular invasion | 0.71 | 0.11 | |||||
| Yes | 9 | 5 | 4 | 38 (8.7-67) | 78 (62-88) | ||
| No | 53 | 24 | 14 | 54 (38-67) | 44 (10-74) | ||
| Initial surgery result | 0.26 | 0.044 | |||||
| R0 | 28 | 12 | 7 | 56 (34-74) | 81 (57-92) | ||
| R1 | 23 | 12 | 6 | 47 (25-67) | 76 (48-91) | ||
| R2 | 4 | 0 | 0 | 100 | 100 | ||
| Biopsy | 7 | 5 | 5 | 29 (4.1-61) | 29 (4.1-61) | ||
| Lymph node involvementª | 0.030 | 0.78 | |||||
| Yes | 5 | 5 | 3 | 20 (0.8-58) | 80 (20-97) | ||
| No | 43 | 18 | 12 | 58 (40-73) | 73 (54-85) |
ªThe sum does not add up to the total because of missing values. In bold type: P value less than 0.05, considered statistically significant.
had at least one poor risk factor at diagnosis in order to ensure that the tumor was a real ACC rather than a benign adenoma.
Considering all patients, the age less than or equal to 4 and more than 10 years at diagnosis showed a significant difference both at OS and PFS at the univariate analysis. It is widely accepted that chil- dren have a better outcome than adolescents. Even if the cut-off age above which survival is reduced is still being debated, early child- hood or before adolescence,1,27 in our analysis, patients less than or equal to 4 years of age had a clearly better outcome than the older ones.
Concerning factors related to the tumor, metastatic spread is considered a highly unfavorable prognostic factor, also because the response to medical therapies is normally scarce.13 The Interna- tional Pediatric Adrenocortical Tumor Registry reported a 20% OS for patients with advanced disease20; better results using chemother- apy and mitotane (OS 64.8%) were described recently by the GPOH- MET studies.28 In our series, 18/20 patients with metastases died a few months after diagnosis. Metastases were the most important fac- tor related to the tumor: Figures 1A and 1B show the OS and PFS of patients with metastatic and localized disease, respectively, confirming their very poor prognosis despite all therapies.
Tumor size was more than 200 cm3 in almost two thirds of the patients, with a similar rate in metastatic and localized cases. A tumor volume more than 200 cm3 had a worse impact on PFS and OS for localized tumors and on PFS if we consider the whole group. Notably, this risk factor remains significant in the multivariable analysis. In their
analysis of 24 pediatric adrenal tumors, Flynt et al.27 observed that large size and heterogeneous structure often coincide with the malig- nant nature of the tumors. Although tumor size is assessed with dif- ferent units of measures by different authors (volume, diameter, or weight), the concept that huge ACCs have a worse prognosis is shared by most (Supplementary Table S1).1,14,19,20,27,29,30
At diagnosis, most ACCs occurring in children are localized and can be completely resected.14,20 An initial surgical excision in our series was attempted in most cases (66/82), but the complete resection was possible only in 37.8% of the whole population (31/82) and in 45.2% of the patients with localized disease (28/62), confirming the difficulty in excising these tumors.
At univariate analysis, the R0 resection showed a better 3-year OS than other types of surgery. Several patients, mainly among those with localized disease (23/62), had an initial, unsatisfactory R1 resection, a scenario that even skilled surgeons may encounter in the presence of local or regional lymph nodes invasion, not predictable at the initial radiologic work-up. Although the value of surgical completeness disap- peared at multivariable analysis, these results confirm the importance of a watchful analysis of imaging at diagnosis and of a complete excision of the primary mass.
The prognostic factors we evaluated are similar to those analyzed in the adult population. Bilimoria et al.31 found that poor prognosis is correlated to large, locally invasive, and metastatic diseases. Moreover, also in adult patients, the survival was worsened by older age, involved margins, and presence of regional and distant disease.
Concerning other parameters, Cushing symptoms were present in about 30% of the cases, which is a higher rate if compared with other series, where these symptoms are described in only 5% of the patients.5,20 Vascular invasion and regional nodes involvement were more frequent in patients with metastatic disease. Even if data were not available for all patients, lymph nodes involvement showed a signif- icant influence on the outcome, while vascular tumor invasion did not demonstrate an important weight. Importantly, the presence of two or more risk factors in the same patient was associated to a significantly higher probability of a relapse or an unfavorable outcome.
Even if our study has a number of limitations, including the ret- rospective nature of the data, the different staging systems initially adopted, the lack of precise data concerning chemotherapy and other medical therapies, and the lack of a central pathology review, this large series made it possible to analyze in a multivariate setting the influence of adverse factors related to the tumor and the consequences of differ- ent surgical approaches. As compared with previous studies (Supple- mentary Table S1), this is the first analysis that considers a large group of patients with adverse factors, enrolled in different countries during the same period. From our study results, we can prospectively select those patients who absolutely need perioperative medical therapies in order to improve the outcome (large tumors, metastatic spread, at least two risk factors). Moreover, as survival of metastatic patients is really poor, these patients should have the proposition to be included within prospective new drug protocols.32
Finally, this study demonstrates that international cooperation and prospective cooperative protocols may expand our knowledge about these rare tumors and lead to collection of a more accurate data.33
ACKNOWLEDGMENTS
The authors thank all pediatricians, pediatric oncologists, and surgeons who collaborated in sending information about the cases included in the study.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
REFERENCES
1. Dall’Igna P, Virgone C, De Salvo GL, et al. Adrenocortical tumors in Ital- ian children: analysis of clinical characteristics and P53 status. Data from the national registries. J Pediatr Surg. 2014;49(9):1367-1371.
2. Michalkiewicz E, Sandrini R, Figueiredo B, et al. Clinical and outcome characteristics of children with adrenocortical tumors: a report from the international pediatric adrenocortical tumor registry. J Clin Oncol. 2004;22(5):838-845.
3. Rescorla FJ. Malignant adrenal tumors. Semin Pediatr Surg 2006;15(1):48-56.
4. Marques-Pereira R, Delacerda L, Lacerda HM, Michalkiewicz E, San- drini F, Sandrini R. Childhood adrenocortical tumors: a review. Heredit Cancer Clin Pract. 2006;4:81-89.
5. Sandrini R, Ribeiro RC, DeLacerda L. Childhood adrenocortical tumors. J Clin Endocrinol Metab. 1997;82(7):2027-2031.
6. Ribeiro RC, Sandrini Neto RS, Schell MJ, Lacerda L, Sambaio GA, Cat I. Adrenocortical carcinoma in children: a study of 40 cases. J Clin Oncol. 1990;8(1):67-74.
7. Magro G, Esposito G, Cecchetto G, et al. Pediatric adrenocortical tumors: morphological diagnostic criteria and immunohistochemical expression of matrix metalloproteinase type 2 and human leucocyte- associated antigen (HLA) class II antigens. Results from the Italian pediatric rare tumor (TREP) study project. Hum Pathol. 2012;43(1):31- 39.
8. Bugg MF, Ribeiro RC, Roberson PK, et al. Correlation of pathologic features with clinical outcome in pediatric adrenocortical neoplasia. A study of a Brazilian population. Brazilian group for treatment of child- hood adrenocortical tumors. Am J Clin Pathol. 1994;101(5):625-629.
9. Wieneke JA, Thompson LD, Heffess CS. Adrenal cortical neoplasms in the pediatric population: a clinicopathologic and immunophenotypic analysis of 83 patients. Am J Surg Pathol. 2003;27(7):867-881.
10. Weiss LM, Medeiros LJ, Vickery AL Jr. Pathologic features of prog- nostic significance in adrenocortical carcinoma. Am J Surg Pathol. 1989;13(3):202-206.
11. Duregon E, Volante M, Bollito E, et al. Pitfalls in the diagnosis of adrenocortical tumors: a lesson from 300 consultation cases. Hum Pathol. 2015;46(12):1799-1807.
12. Bisogno G, Ferrari A, Bien E, et al. Rare cancers in children-the EXPeRT initiative: a report from the European cooperative study group on pediatric rare tumors. Klin Padiatr 2012;224(6):416-420.
13. McAteer JP, Huaco JA, Gow KW. Predictors of survival in pedi- atric adrenocortical carcinoma: a surveillance, epidemiology, and end results (SEER) program study. J Pediatr Surg. 2013;48(5):1025-1031.
14. Klein JD, Turner CG, Gray FL, et al. Adrenal cortical tumors in children: factors associated with poor outcome. J Pediatr Surg. 2011;46(6):1201-1207.
15. Sbragia L, Oliveira-Filho AG, Vassallo J, Pinto GA, Guerra-Junior G, Bustorff-Silva J. Adrenocortical tumors in Brazilian children: immuno- histochemical markers and prognostic factors. Arch Pathol Lab Med. 2005;129(9):1127-1131.
16. Tucci S Jr, Martins AC, Suaid HJ, Cologna AJ, Reis RB. The impact of tumor stage on prognosis in children with adrenocortical carcinoma. J Urol. 2005;174(6):2338-2342, discussion 2342.
17. Hanna AM, Pham TH, Askegard-Giesmann JR, et al. Outcome of adrenocortical tumors in children. J Pediatr Surg. 2008;43(5):843-849.
18. Lughezzani G, Sun M, Perrotte P, et al. The European network for the study of adrenal tumors staging system is prognostically superior to the international union against cancer-staging system: a North Ameri- can validation. Eur J Cancer. 2010;46(4):713-719.
19. Rodriguez-Galindo C, Figueiredo BC, Zambetti GP, Ribeiro RC. Biol- ogy, clinical characteristics, and management of adrenocortical tumors in children. Pediatr Blood Cancer. 2005;45(3):265-273.
20. Michalkiewicz E, Sandrini R, Figueiredo B, et al. Clinical and outcome characteristics of children with adrenocortical tumors: a report from the international pediatric adrenocortical tumor registry. J Clin Oncol. 2004;22(5):838-845.
21. Wieneke JA, Thompson LD, Heffess CS. Adrenal cortical neoplasms in the pediatric population: a clinicopathologic and immunophenotypic analysis of 83 patients. Am J Surg Pathol. 2003;27(7):867-881.
22. Bien E, Godzinski J, Dall’igna P, et al. Pancreatoblastoma: a report from the European cooperative study group for paediatric rare tumours (EXPERT). Eur J Cancer. 2011;47(15):2347-2352.
23. Stachowicz-Stencel T, Orbach D, Brecht I, et al. Thymoma and thymic carcinoma in children and adolescents: a report from the European
WILEY
cooperative study group for pediatric rare tumors (EXPERT). Eur J Cancer. 2015;51(16):2444-2452.
24. Bisogno G, Brennan B, Orbach D, et al. Treatment and prognostic fac- tors in pleuropulmonary blastoma: an EXPERT report. Eur J Cancer. 2014;50(1):178-184.
25. Ciftci AO, Senocak ME, Tanyel FC, Buyukpamukcu N. Adrenocortical tumors in children. J Pediatr Surg. 2001;36(4):549-554.
26. Federici S, Galli G, Ceccarelli P, et al. Adrenocortical tumors in children: a report of 12 cases. Eur J Pediatr Surg. 1994;4:21-25.
27. Flynt KA, Dillman JR, Davenport MS, et al. Pediatric adrenocortical neoplasms: can imaging reliably discriminate adenomas from carcino- mas? Pediatr Radiol. 2015;45(8):1160-1168.
28. Redlich A, Boxberger N, Strugala D, et al. Systemic treatment of adrenocortical carcinoma in children: data from the German GPOH- MET 97 trial. Klin Padiatr. 2012;224(6):366-371.
29. Ribeiro RC, Pinto EM, Zambetti GP, Rodriguez-Galindo C. The interna- tional pediatric adrenocortical tumor registry initiative: contributions to clinical, biological, and treatment advances in pediatric adrenocorti- cal tumors. Mol Cell Endocrinol. 2012;351(1):37-43.
30. Bulzico D, de Faria P, de Paula M, et al. Recurrence and mortality prog- nostic factors in childhood adrenocortical tumors: analysis from the Brazilian national institute of cancer experience. Pediatr Hematol Oncol. 2016;31:1-11.
31. Bilimoria KY, Shen WT, Elaraj D, et al. Adrenocortical carcinoma in the united states. Cancer. 2008;113(11):3130-3136.
32. Pearson, Andrew DJ, et al. Implementation of mechanism of action biology-driven early drug development for children with cancer. Eur J Cancer. 2016;62:124-131.
33. Pinto E, Chen X, Easton J, et al. Genomic landscape of paediatric adrenocortical tumours. Nat Commun. 2015;6:6302.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the support- ing information tab for this article.
How to cite this article: Cecchetto G, Ganarin A, Bien E, Vorwerk P, Bisogno G, Godzinsky J, Dall’Igna P, Reguerre Y, Schneider D, Laurence B, Leblond P, Ferrari A, Brecht I, DePaoli A, Orbach D (2016). Outcome and prognostic fac- tors in high-risk childhood adrenocortical carcinomas: A report from the European Cooperative Study Group on Pediatric Rare Tumors (EXPERT). Pediatr Blood Cancer, 00, 1-8. DOI: 10.1002/pbc.26368