CrossMark
ORIGINAL ARTICLE
Usefulness of Wieneke criteria in assessing morphologic characteristics of adrenocortical tumors in children
Gaurav Chatterjee1 . Shatavisha DasGupta1 . Gautam Mukherjee1 . Moumita Sengupta1 · Paromita Roy2 · Indu Arun2 · Chhanda Datta1 .
Prafulla Kumar Mishra3 · Sugato Banerjee4,5 · Uttara Chatterjee1,5
Accepted: 7 April 2015/Published online: 17 April 2015 @ Springer-Verlag Berlin Heidelberg 2015
Abstract
Purpose Adrenocortical tumors (ACT) occur rarely in pediatric age group. Pediatric ACTs behave differently from their histologically similar adult counterparts and standard adult criteria often cannot accurately predict their clinical behavior. The aim of the present study was to document the clinicopathologic spectrum of pediatric ACTs and to assess the utility of Wieneke scoring system in predicting clinical behavior of these tumors.
Methods This multi-institutional study comprised of 13 cases of pediatric ACTs from January 2005 to May 2014. Clinical features and gross pathologic characteristics were obtained from records. Comprehensive analyses of micro- scopic features were performed. Each tumor was assessed according to criteria proposed by Wieneke et al. and was assigned to benign, intermediate for malignancy or ma- lignant group. The standard adult Weiss criteria were also applied for comparison.
Results There were total 6 cases of adrenocortical ade- nomas and 7 cases of adrenocortical carcinomas. Most of the children (76.9 %) presented with endocrine dysfunc- tion. Lower age of presentation was significantly associated with better prognosis. Applying Wieneke criteria, there were 6 benign and 6 malignant cases and one case was assigned to intermediate for malignancy group. The clinical behavior of all the cases was consistent with Wieneke cri- teria categorization. Applying Weiss criteria, 3 cases with benign clinical behavior were assigned to malignant group. Conclusion Our study validates the reliability of Wieneke scoring system in predicting malignancy in pediatric ACTs. It is simple and easy to use and therefore useful in day-to- day practice.
Keywords Pediatric adrenocortical tumors · Wieneke criteria
☒ Uttara Chatterjee uttarac1@gmail.com
Gaurav Chatterjee gauravchatnobel@gmail.com
Shatavisha DasGupta s_dg2007@yahoo.co.in
Gautam Mukherjee subhatam@gmail.com
Moumita Sengupta moumita.sengupta83@gmail.com
Paromita Roy paromita.roy@tmckolkata.com
Indu Arun rindups@gmail.com
Chhanda Datta drcdatta2012@gmail.com
Prafulla Kumar Mishra mishraprafulla_k@rediffmail.com Sugato Banerjee sugmala@yahoo.co.in
1 Department of Pathology, IPGME&R, Kolkata, India
2 Department of Pathology, Tata Medical Centre, Kolkata, India
3 Department of Pediatric Surgery, IPGME&R, Kolkata, India
4 Department of Pediatric Surgery, KPC Medical College, Kolkata, India
5 Park Clinic, Kolkata, India
Introduction
Adrenocortical tumors (ACT) are rare in pediatric age group, with majority of the tumors being adrenocortical carcinomas (ACC) [1]. They constitute about 0.2 % of all pediatric tumors and 5-6 % of all adrenal tumors, with a reported incidence of 0.2-0.3 new cases per 1 million children per year. There is geographical variation in in- cidence as it occurs remarkably more frequently in Southern Brazil [2-4] (12-18 times more than in United States of America or Europe) probably due to the high prevalence of a specific germ line p53 mutation in that population [5].
There is always an element of uncertainty in distin- guishing adrenocortical adenomas (ACA) from ACC, a feature shared by other endocrine tumors [1]. This is particularly problematic in pediatric cases as it is clearly established that pediatric ACTs behave differently than their histologically similar adult counterparts and the Weiss criteria [6, 7], most widely used in adult ACTs, often overpredict the diagnosis of malignancy in these tumors. It is generally acknowledged that no single pathologic feature itself is diagnostic of malignancy, and definitive set of pathologic criteria separating pediatric ACCs from ACAs remains uncertain [8-11]. This is due to, in part, the rarity of these tumors which precludes large study series. Previous studies have suggested clinical features including age at diagnosis and pre- senting endocrine symptom are important prognostic parameters [12, 13]. Further, there is no universally agreed upon staging system for these tumors and no immunohistochemical marker including Ki-67 has been found to be useful in separating ACCs from ACAs [10, 11]. In one of the largest studies till date, Wieneke et al. [10] proposed a set of 9 macroscopic and microscopic criteria for malignancy in pediatric ACTs. This scoring system categorizes pediatric ACTS into “benign”, “in- termediate for malignancy” and “malignant”. This system is easy to use and therefore useful in daily practice. Nevertheless, the predictive value of the system proposed by Wieneke et al. is yet to be validated as there is only one other pediatric series [14] so far assessing its utility.
The aim of the present study was to document the clinicopathologic spectrum of pediatric ACTs and to evaluate the usefulness of Wieneke scoring system in predicting malignancy of these tumors. The standard Weiss criteria for adults were also applied for comparison.
Materials and methods
This was a multi-institutional, retrospective study com- prising of 13 cases of pediatric ACTs over the last 10 years from January 2005 to May 2014. The study was approved by Institutional Ethical Committee of IPGME&R, Kolkata. Inclusion in the study required the presence of ACTs and excluded adrenal medullary neoplasms (pheochromocy- toma, neuroblastic tumors) and cysts. Pertinent clinical features such as age at presentation, sex, presenting symptoms and signs of endocrine dysfunction, any relevant past history were obtained from records. Radiological findings, operative and gross pathologic findings including tumor weight and size were recorded. Complete follow-up data were available for 11 patients including tumor stage, treatment methods used and current disease status/metas- tasis. The post-surgical staging was performed according to the modified staging system used in International Pediatric Adrenocortical Tumor Registry (IPATR) study [12]: ST I (Tumor completely excised with negative margins, tumor weight ≤200 g and absence of metastasis); ST II (Tumor completely excised with negative margins, tumor weight >200 g or microscopic residual tumor and absence of metastasis); ST III (Macroscopic residual or inoperable tumor); ST IV (Hematogenous metastasis at presentation). Two children were lost to follow-up. Patients with recur- rent/metastatic disease were considered clinically malignant.
A comprehensive analysis of macroscopic and micro- scopic pathological features was done. Slides from all cases were reviewed and features were carefully assessed and recorded according to criteria proposed by Wieneke et al. [10]. Histologic features included in Weiss criteria were also assessed. Features such as tumor size, weight, pe- riadrenal extension, capsular invasion and vena cava in- vasion were obtained from past records, if not present in the reviewed slides. Tumors were categorized as benign or malignant according to both Wieneke and Weiss scoring system (Table 1).
Statistical analysis
Numerical variables were compared between groups by Mann-Whitney U test and categorical variables were compared between groups by Fischer’s exact test. Agree- ment between benign-malignant diagnosis by Wieneke and Weiss criteria was assessed by calculating k coefficient. For individual parameters, the standard diagnostic indices namely sensitivity, specificity, positive predictive value
| Wieneke criteria | Weiss criteria |
|---|---|
| Tumor weight >400 g | High nuclear grade (Führman G3/G4) |
| Tumor size >10.5 cm | >5 mitosis/50 HPF |
| Extension into periadrenal soft tissues and/or adjacent organs | Abnormal mitoses |
| Invasion into vena cava | <25 % clear cells |
| Venous invasion | >33 % diffuse architecture |
| Capsular invasion | Tumor necrosis |
| Presence of tumor necrosis | Venous invasion |
| >15 mitoses per 20 HPF | Sinusoid invasion |
| Presence of atypical mitotic figures | Capsular invasion |
Tumors showing two or less of features included in Wieneke scoring system were classified as benign, presence of four or more criteria indicated malignant, whereas tumor with three features was categorized as intermediate for malignancy. On the other hand, tumors were classified as ACCs according to Weiss system if three or more criteria were present
(PPV) and negative predictive value (NPV) were calcu- lated. Key findings have been presented with 95 % confi- dence interval (CI). All statistical analysis were done using MedCalc version 11.6 (Mariakerke, Belgium: MedCalc Software 2011) and Statistica version 6 (Tulsa, Oklahoma: StatSoft Inc., 2001) and P value <0.05 has been considered as statistically significant.
Results
Clinical findings
Total 13 cases of pediatric ACTs were diagnosed over a period of 10 years with 6 ACAs and 7 ACCs. There were total 9 girls and 4 boys. Out of 9 girls, 5 had features of ACA and 4 of them presented with ACC. The mean age at presentation was 2.9 years with an age range of 4 months to 16 years. Total 6 infants were included in the study. The girls had an older age of presentation than the boys (3.7 years vs 15 months). Children with ACC had a sig- nificantly greater age of presentation than those with ACA (4.8 years vs 9 months, P value = 0.027).
Majority of the patients (76.9 %) presented with signs and symptoms of endocrine dysfunction. Virilization, ei- ther isolated or associated with Cushing’s syndrome, was the most common presentation (61.5 %). Single case of isolated Cushing’s syndrome occurred in a 10-year-old girl having ACC. An infant girl presenting with ACC had features of feminization with vaginal bleeding. Non-func- tional cases presented with abdominal mass or pain. There were no cases associated with Beckwith-Wiedemann or any other named syndrome.
The mean follow-up duration was 26.5 months. A total of 2 and 7 children had ST I and ST II (for tumor weight >200g/microscopic residuals) tumors, respectively. One child had ST III disease and 3 children presented with metastasis (ST IV). Six children achieved complete re- mission (2 of them ST I and 4 of them ST II). Three of seven ST II tumors progressed and 2 of them developed metastasis. Among the 5 children who developed metas- tasis, 3 had pulmonary metastasis and 2 had developed liver metastasis. Surgical approach involved complete re- section of localized tumors. A second-look surgery was performed in one case with gross residual tumor after chemotherapy. In a single case with regional enlarged lymph nodes, berry-picking of the enlarged nodes was performed. Treatment protocol for advanced tumors was not uniform but involved mitotane, cisplatin, doxorubicin and etoposide in various combinations. Two children on miototane are alive with metastatic disease (6 and 12 months after diagnosis). Three children died 3-18 months after diagnosis despite chemotherapy. Two children with histological features of malignancy were lost to follow-up with one of them having a pulmonary metastasis at diagnosis.
Pathologic findings
The tumor weights ranged from 200 to 600 g with a mean of 394.2 g. ACAs had a mean tumor weight of 295.8 g, whereas the ACCs had a mean tumor weight of 478.6 g. As a criterion to determine malignancy, tumor weight had a sensitivity of 71.43 % (95 % confidence interval, CI 29.04-96.33) and specificity of 83.33 % (95 % CI 35.88-99.58 %). The ACTs ranged in size from 4 cm to
13 cm with a mean of 7.65 cm. The ACAs had an average size of 6.25 cm, while the average size for ACCs was 8.86 cm. Tumor size had sensitivity of 28.57 % (95 % CI 3.67-70.96 %) and specificity of 83.33 % (95 % CI 35.88-99.58 %) as a criterion for malignancy. One ACC had macroscopic invasion into inferior vena cava. Extension into periadrenal soft tissue/adjacent organs was identified in 3 ACCs (sensitivity = 42.86 %, 95 % CI 9.92-81.59 % and specificity = 100 %, 95 % CI 54.07-100 %).
Microscopically, most of the tumors were composed of sheets of polygonal cells with eosinophilic cytoplasm ad- mixed with a varied proportion of clear cells. Venous invasion (sensitivity = 71.43 %, 95 % CI 29.04-96.33 and specifici- ty = 100 %, 95 % CI 58.07-100 %) and capsular invasion (sensitivity = 85.71 %, 95 % CI 42.13-99.64 % and speci- ficity = 100 %, 95 % CI 54.07-100 %) were identified in 5 and 6 ACCs, respectively. A total of 5 ACCs and one ACA showed presence of tumor necrosis (sensitivity = 71.43 %, 95 % CI 29.04-96.33 and specificity = 83.33 %, 95 % CI 35.88-99.58 %). Five of the ACCs had >15 mitoses/20 HPF (sensitivity = 71.43 %, 95 % CI 29.04-96.33 and specifici- ty = 100 %, 95 % CI 58.07-100 %) with 3 of them showing presence of atypical mitotic figures (sensitivity = 42.86 %, 95 % CI 9.9-81.59 % and specificity = 100 %, 95 % CI 54.07-100 %).
Applying the scoring system proposed by Wieneke et al. [10] (Table 2), 6 tumors were classified as benign as they had 2 or less unfavorable criteria and 6 tumors with 4 or more of the criteria were categorized as malignant. There was one case with three unfavorable features and was classified as intermediate for malignancy. This pathologic categorization was predictive of the clinical outcome with Cohen’s kappa coefficient of 0.847 (almost perfect agree- ment with standard error = 0.145, 95 % CI 0.562-1.000). All of the 6 cases categorized as benign had complete re- mission whereas 3 out of 6 malignant cases died of metastatic disease, 1 child is alive with recurrent disease and 2 cases were lost to follow-up, one of them had pul- monary metastasis at presentation. The case assigned to intermediate for malignancy group showed a malignant clinical course with development of liver metastasis.
Among other pathologic features, a 7-month-old girl presented with a large (425 g, 15 × 10 x 6 cm) ACA showing extensive myxoid change in the stroma. Two of the 7 ACCs showed focal myxoid change. Adult criteria lacked specificity in that <25 % clear cells (specifici- ty = 50 %, 95 % CI 11.81-88.19 %) were present in all of the 7 ACCs but also in 3 ACAs. Similarly, 6 ACCs and 3 ACAs had >33 % diffuse architecture (specificity = 50 %, 95 % CI 11.81-88.19 %). High nuclear grade (Grade III/
IV) and >5 mitoses/50 HPF were present in 6 ACCs and 2 ACAs (specificity = 66.67 %, 95 % CI 22.28-95.67 %). Applying the Weiss criteria (Table 3), 10 cases were as- signed to malignant group and 3 cases were categorized as benign. However, three of the children grouped as malig- nant enjoyed benign clinical courses with complete re- mission. The Cohen’s kappa coefficient for Weiss system was 0.519 (moderate agreement with standard er- ror = 0.214, 95 % CI 0.0997-0.937).
Discussion
The rarity of ACTs in children has led to very few study series dealing with the clinicopathologic features of these tumors. This has hampered the search for potential prog- nostic morphological parameters as well as therapeutic decision-making protocols [15]. The greatest challenge is the lack of correlation between histopathological features and clinical behavior [1]. Previous studies have empha- sized that various established criteria to distinguish ACCs from ACAs in adults are not helpful in predicting the biological behavior in children. With the advent of phar- macological replacement therapy, pediatric ACTs, despite having ominous histological features, usually have a more benign clinical course than their histologically similar adult counterparts. This has led to the search for new pathologic parameters to accurately predict the clinical nature of these tumors [10, 15-17].
In the IPATR study [12], the largest study till date, a total of 254 children were studied. However, it was ac- knowledged that “Because consistent histologic criteria were not used to classify pediatric adrenocortical tumors as benign (adenoma) or malignant (carcinoma), these data were not examined as prognostic factors”. This and a few other studies [11, 13] have suggested clinical parameters useful in predicting the clinical behavior of these tumors. These include female sex, early (<4 year) age at presen- tation and isolated virilisation. In our study, there was a significant association between early age of presentation and a benign clinical outcome. There seems to be a biphasic age distribution with 11 out of 13 cases presenting at or earlier than 4 years of age, which is also the age for unique pediatric solid tumors. The majority (71.4 %) of the infants in our study had ACAs. Two children presented at 10 and 16 years. There was no case in 5-9 years age group. In keeping with previous studies [10], the 2 older children presented with isolated Cushing’s syndrome and non- functional abdominal mass, respectively. Both of them had ACCs. Pediatric ACTs may actually be a heterogeneous
| Case | Age at presentation | Tumor weight (g) | Tumor size (cm) | Extension into periadrenal soft tissue and/or adjacent organs | Invasion into vena cava | Venous invasion | Capsular invasion | Tumor necrosis | >15 mitoses/20 HPF | Atypical mitotic figures | Wieneke score/category | Follow- up |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 6 months | 300 | 5 | Absent | Absent | Absent | Absent | Present | Absent | Absent | 1/benign | CR |
| 2 | 3 years | 300 | 4.5 | Absent | Absent | Present | Present | Present | Present | Absent | 4/malignant | AWD |
| 3 | 1 year | 500 | 10 | Present | Present | Present | Present | Absent | Present | Present | 7/malignant | DOD |
| 2 months | ||||||||||||
| 4 | 6 months | 425 | 12 | Absent | Absent | Absent | Absent | Absent | Absent | Absent | 2/benign | CR |
| 5 | 5 months | 200 | 4 | Absent | Absent | Absent | Absent | Absent | Absent | Absent | 0/benign | CR |
| 6 | 1 year | 600 | 12 | Present | Absent | Present | Present | Absent | Present | Absent | 6/malignant | DOD |
| 1 month | ||||||||||||
| 7 | 6 months | 450 | 7 | Absent | Absent | Absent | Present | Present | Absent | Absent | 3/intermediate | AWD |
| 8 | 16 years | 600 | 13 | Present | Absent | Present | Present | Present | Absent | Absent | 7/malignant | DOD |
| 9 | 2 years | 350 | 6 | Absent | Absent | Absent | Absent | Absent | Absent | Absent | 0/benign | CR |
| 10 | 2 years | 200 | 7.5 | Absent | Absent | Present | Present | Present | Present | Present | 5/malignant | NA |
| 11 | 10 years | 700 | 8 | Absent | Absent | Absent | Absent | Present | Present | Present | 4/malignant | NA |
| 12 | 9 months | 200 | 5 | Absent | Absent | Absent | Absent | Absent | Absent | Absent | 0/benign | CR |
| 13 | 4 months | 300 | 5.5 | Absent | Absent | Absent | Absent | Absent | Absent | Absent | 0/benign | CR |
CR complete remission, AWD alive with disease, DOD died of disease, NA follow-up data not available
| Case | Age at presentation | Nuclear grade | >5 mitosis/50 HPF | Abnormal mitoses | <25 % clear cells | >33 % diffuse architecture | Tumor necrosis | Venous invasion | Sinusoid invasion | Capsular invasion | Weiss score/category | Follow- up |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 6 months | G3 | Absent | Absent | Present | Present | Present | Absent | Absent | Absent | 4/malignant | CR |
| 2 | 3 years | G4 | Present | Absent | Present | Present | Present | Present | Present | Present | 8/malignant | AWD |
| 3 | 1 year | G4 | Present | Present | Present | Present | Absent | Present | Present | Present | 8/malignant | DOD |
| 2 months | ||||||||||||
| 4 | 6 months | G2 | Absent | Absent | Absent | Absent | Absent | Absent | Absent | Absent | 0/benign | CR |
| 5 | 5 months | G2 | Absent | Absent | Absent | Absent | Absent | Absent | Absent | Absent | 0/benign | CR |
| 6 | 1 year | G3 | Present | Absent | Present | Present | Absent | Present | Present | Present | 7/malignant | DOD |
| 1 month | ||||||||||||
| 7 | 6 months | G2 | Absent | Absent | Present | Present | Present | Absent | Absent | Present | 4/malignant | AWD |
| 8 | 16 years | G3 | Present | Absent | Present | Present | Present | Present | Present | Present | 8/malignant | DOD |
| 9 | 2 years | G2 | Present | Absent | Present | Present | Absent | Absent | Absent | Absent | 3/malignant | CR |
| 10 | 2 years | G3 | Present | Present | Present | Absent | Present | Present | Present | Present | 8/malignant | NA |
| 11 | 10 years | G4 | Present | Present | Present | Present | Present | Absent | Absent | Absent | 6/malignant | NA |
| 12 | 9 months | G2 | Absent | Absent | Absent | Absent | Absent | Absent | Absent | Absent | 0/benign | CR |
| 13 | 4 months | G3 | Present | Absent | Present | Present | Absent | Absent | Absent | Absent | 3/malignant | CR |
Bold values signify discrepancy between clinical outcome and categorization applying Weiss criteria
CR complete remission, AWD alive with disease, DOD died of disease, NA follow-up data not available
group of tumors comprising tumors presenting at or before 4 years and at or after 10 years of age [1, 10]. The 2013 SEER study [13] has also concluded that age at presenta- tion was significantly predictive of poor outcome.
In our study, we had 9 girls with 5 ACAs and 4 ACCs and did not find any significant association between gender and clinical outcome as suggested by Sbragia et al. [11]. In contrast to the adult cases, most of the pediatric ACTs are hormonally functional and this is usually not indicative of a poor outcome. On the contrary, the IPATR study has suggested that isolated virilization has better prognosis [12]. In our study, virilization was the most common mode of presentation (61.5 %) and total 3 ACAs and 2 ACCs presented with isolated virilization.
There is no universally agreed upon staging system for pediatric ACTs [1]. Further, previous studies have shown that immunohistochemical markers including p53, ki-67, c-erb-B2, Bcl-2, MMP2 and HLA class II antigen are not helpful in predicting clinical behavior of these tumors [11, 14]. This has led to a continuing search for reproducible morphological features to accurately distinguish pediatric ACCs from ACAs. In 2003, Wieneke et al. [10] proposed a set of pathologic criteria useful in predicting the clinical behavior of these tumors. This system is easy to use in routine practice. To the best of our knowledge, there has been only one published study so far assessing the re- liability of Wieneke scoring system since it was originally proposed [14]. We have applied the Wieneke system to all of our cases, trying to assess its utility in predicting clinical outcome of pediatric ACTs.
Among the macroscopic criteria, tumor weight and size have been well studied. Tumor size has been noted to be an important factor in predicting survival in Brazilian population [8-11, 17-19]. Tumor weight has been a part of various proposed staging system [1, 12], although different cut-offs have been used. The Wieneke scoring system uses 400 g as a cut-off [10]. In our study, 5 out of 7 ACCs had weight >400 g. However, there was one ACA with myxoid change weighing 425 g. Only 2 out of 7 ACCs had size >10.5 cm. The studies by Wieneke et al. [10] and Magro et al. [14] have suggested that tumor weight and size cannot be used alone as a predictor of malignancy. Other macroscopic criteria such as invasion into vena cava and periadrenal extension were 100 % specific, and were ob- served in 1 and 3 of our ACC cases, respectively.
Microscopically, most of the pediatric ACTs were composed of sheets of cells with abundant eosinophilic cytoplasm and varied proportion of nuclear pleomorphism and atypia. Capsular invasion and venous invasion were present in most of the ACCs (6 and 5 out of 7 ACCs, respectively). Five ACCs and 1 ACA had tumor necrosis. Five ACCs had >15 mitoses/20 HPF and 3 of them showed presence of atypical mitoses. The other microscopic
criteria included in Wieneke system were more or less specific with different sensitivities.
As a constellation of criteria, Wieneke system accu- rately predicted the clinical outcome of the pediatric ACTs in our series (Cohen’s kappa coefficient of 0.847, almost perfect agreement with standard error = 0.145, 95 % CI 0.562-1.000). All the cases assigned to benign group achieved complete remission, whereas 3 out 6 malignant cases died of metastasis, 1 child was alive with disease, 2 cases were lost to follow-up. This is in corroboration with the findings of the study by Magro et al. [14]. Three chil- dren with ST II tumors, malignant according to Wieneke criteria, progressively developed recurrence/metastasis (5-25 months post-surgery). Previous studies have found that the time to tumor recurrence ranged from 1 to 48 months, but only 2 of 15 cases relapsed more than 1 year from initial surgery [8, 11]. Mean duration of fol- low-up of our study was 26.5 months (range 3-98 months). Among the three children who progressed from STII, the single case assigned to intermediate for malignancy group had a malignant clinical course.
Additional pathologic criteria useful in adult ACTs lacked specificity. Less than 25 % clear cells and >33 % diffuse architecture were present in most of the ACCs, but also in half of our ACAs. Similarly, high nuclear grade was seen in 2 ACAs. When the tumors were classified ac- cording to Weiss system, 10 tumors were classified as malignant, 3 of which achieved complete remission. This finding indicated that adult criteria cannot accurately pre- dict the clinical outcome of pediatric ACTs and can lead to overdiagnosis in a fair number of cases.
Among other pathologic features not included in Wie- neke or Weiss system, Papotti et al. [20] have reported the significance of stromal myxoid change in adult ACTs. Magro et al. [14] found that focal myxoid change was present in most of the pediatric ACCs. In our series, 2 out of 7 ACC showed presence of focal myxoid change, whereas an infant girl with ACA had extensive myxoid change. It has been suggested that abundant myxoid change in stroma is associated with small cells with mild atypia, whereas focal myxoid change in stroma is more often as- sociated with malignancy [14, 20].
Regarding the molecular pathogenesis of these tumors, mutations of p53 tumor suppressor gene have been impli- cated in tumor development and progression. Germ line p53 mutation has been found in 50-80 % ACCs in children [21]. In the Italian series by Dall’Igna et al. [22], mutations involving the common hot spots were found in 35 % of tumors (2 benign, 5 malignant). Recently, several groups have performed genomic, transcriptomic, microRNA-based and DNA methylation analysis of ACTs and have come up with promising new strategies to distinguish ACTs from ACAs [23]. Upregulation of IGF2, overexpression of mir-
483 (3p and 5p) and downregulation of mir-195 have been found to be associated with ACCs [24, 25]. However, most of these studies have been performed on adult ACTs. A recent study on childhood ACTs using SNP array profiling identified recurrent alterations in loci comprising well- known oncogenes (MYC, MDM2, PDGFRA, KIT, MCL1, BCL2L1) and tumor suppressor genes (TP53, RB1, RPH3AL) [26]. Doghman et al. [27] found 26 dysregulated miRNAs in a study comprising 25 pediatric ACTs. How- ever, there is a lack of an integrated view incorporating all of these information making it difficult to apply them in clinical practice, especially in the pediatric scenario.
Many authors have speculated hypothesis on the reason for the discrepancy between clinical behavior of pediatric and adult ACTs. It has been suggested that pediatric ACTs presenting <4 years of age may arise from fetal adreno- cortical cells and this may explain the limited malignant potential of these tumors [1, 4, 12, 13]. In support of this view, pediatric ACCs have similar biochemical and steroidogenic characteristics to the fetal adrenocortical cells [28, 29]. It contrast to the adult ACTs, MMP2 ex- pression and loss of HLA class II antigen do not distinguish between pediatric ACCs and ACAs. This finding indicates that pediatric ACTs have distinct molecular features [14]. On the other hand, the ACTs occurring in the adolescent age group are associated with poorer clinical outcome and they may arise from adult adrenocortical cells [1, 10]. In this heterogeneous setting, it is important to apply any pathologic criteria in an individual context. Wieneke et al. [10] have also acknowledged that no single pathologic criterion is itself discriminatory between pediatric ACCs and ACAs and each case should be assessed individually.
There are several limitations to this study. First, since the study is retrospective, it is susceptible to confounding by unmeasured factors. Second, the small sample size due to the rarity of the pediatric ACTs makes it difficult to assess the more subtle associations. Larger observational and prospective studies involving multiple centers and countries over a longer period of time are required to ad- dress these shortcomings.
In conclusion, we have analyzed the clinicopathologic features of pediatric ACTs and assessed the utility of Wieneke scoring system in predicting the clinical outcome of these tumors. Lower age of presentation was sig- nificantly associated with better prognosis. Our study validates the applicability of Wieneke scoring system to predict prognosis in pediatric ACTs. Larger prospective studies are needed to further assess the biological nature and clinical behavior of these rare tumors.
Acknowledgments We are grateful to Prof. Avijit Hazra, Depart- ment of Pharmacology, I.P.G.M.E & R, Kolkata for statistical analysis.
Conflict of interest Authors declare no conflict of interest.
References
1. Dehner L, Hill D (2009) Adrenal cortical neoplasms in children: why so many carcinomas and yet so many survivors? Pediatr Dev Pathol 12:284-291
2. Sandrini R, Ribeiro R, DeLacerd L (1997) Childhood adreno- cortical tumors. J Clin Endocrinol Metab 82:2027-2031
3. Ribeiro R, Figueiredo B (2004) Childhood adrenocortical tu- mours. Eur J Cancer 40:1117-1126
4. Rodriguez-Galindo C, Figueiredo B, Zambetti G, Ribeiro R (2005) Biology, clinical characteristics, and management of adrenocorti- cal tumors in children. Pediatr Blood Cancer 45:265-273
5. Ribeiro R, Sandrini F, Figueiredo B, Zambetti G, Michalkiewicz E, Lafferty A et al (2001) An inherited p53 mutation that con- tributes in a tissue-specific manner to pediatric adrenal cortical carcinoma. Proc Natl Acad Sci 98:9330-9335
6. Weiss L (1984) Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 8:163-170
7. Weiss L, Medeiros L, Vickery A Jr (1989) Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 13:202-206
8. Bugg M, Ribeiro R, Roberson P, Lloyd R, Sandrini R, Silva J, Epelman S, Shapiro D, Parham D (1994) Correlation of patho- logic 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
9. Cagle P, Hough A, Jeffrey T, Page D, Johnson E, Kirkland R, Holcombe J, Hawkins E (1986) Comparison of adrenal cortical tumors in children and adults. Cancer 57:2235-2237
10. Wieneke J, Thompson L, Heffess C (2003) Adrenal cortical neoplasms in the pediatric population: a clinicopathologic and immunophenotypic analysis of 83 patients. Am J Surg Pathol 27:867-881
11. Sbragia L, Oliveira-Filho A, Vassallo J, Pinto G, Guerra-Junior G, Bustorff-Silva J (2005) Adrenocortical tumors in Brazilian children: immunohistochemical markers and prognostic factors. Arch Pathol Lab Med 129:1127-1131
12. Michalkiewicz E, Sandrini R, Figueiredo B, Miranda E, Caran E, Oliveira-Filho A et al (2004) Clinical and outcome characteristics of children with adrenocortical tumors: a report from the Inter- national Pediatric Adrenocortical Tumor Registry. J Clin Oncol 22:838-845
13. McAteer J, Huaco J, Gow K (2013) Predictors of survival in pedi- atric adrenocortical carcinoma: a surveillance, epidemiology, and end results (SEER) program study. J Pediatr Surg 48:1025-1031
14. Magro G, Esposito G, Cecchetto G, Dall’Igna P, Marcato R, Gambini C et al (2012) Pediatric adrenocortical tumors: mor- phological diagnostic criteria and immunohistochemical expres- sion 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 43:31-39
15. Ribeiro R, Pinto E, Zambetti G, Rodriguez-Galindo C (2012) The International Pediatric Adrenocortical Tumor Registry initiative: contributions to clinical, biological, and treatment advances in pediatric adrenocortical tumors. Mol Cell Endocrinol 351:37-43
16. Sabbaga C, Avilla S, Schulz C, Garbers J, Blucher D (1993) Adrenocortical carcinoma in children: clinical aspects and prog- nosis. J Pediatr Surg 28:841-843
17. Ribeiro R, Neto R, Schell M, Lacerda L, Sambaio G, Cat I (1990) Adrenocortical carcinoma in children: a study of 40 cases. J Clin Oncol 8(1):67-74
18. Sturgeon C, Shen W, Clark O, Duh Q, Kebebew E (2006) Risk assessment in 457 adrenal cortical carcinomas: how much does tumor size predict the likelihood of malignancy? J Am Coll Surg 202:423-430
19. Driver C, Birch J, Bruce J (1998) Adrenal cortical tumors in childhood. Pediatr Hematol Oncol 15:527-532
20. Papotti M, Volante M, Duregon E, Delsedime L, Terzolo M, Berruti A, Rosai J (2010) Adrenocortical tumors with myxoid features: a distinct morphologic and phenotypical variant ex- hibiting malignant behavior. Am J Surg Pathol 34:973-983
21. Wasserman J, Zambetti G, Malkin D (2012) Towards an under- standing of the role of p53 in adrenocortical carcinogenesis. Mol Cell Endocrinol 351:101-110
22. Dall’Igna P, Virgone C, De Salvo G, Bertorelle R, Indolfi P, De Paoli A et al (2014) Adrenocortical tumors in Italian children: analysis of clinical characteristics and P53 status. Data from the national registries. J Pediatr Surg 49:1367-1371
23. Papotti M, Duregon E, Volante M, McNicol A (2014) Pathology of the adrenal cortex: a reappraisal of the past 25 years focusing on adrenal cortical tumors. Endocr Pathol 25:35-48
24. Bezerra J, LatronicHo A (2014) MicroRNA Era: the importance for diagnosis and prognosis of adrenocortical tumors. Biomed Res Int 2014:1-6
25. de Reynies A, Assie G, Rickman D, Tissier F, Groussin L, Rene- Corail F et al (2009) Gene expression profiling reveals a new classification of adrenocortical tumors and identifies molecular predictors of malignancy and survival. J Clin Oncol 27:1108- 1115
26. Letouzé E, Rosati R, Komechen H, Doghman M, Marisa L, Flück C et al (2012) SNP array profiling of childhood adrenocortical tumors reveals distinct pathways of tumorigenesis and highlights candidate driver genes. J Clin Endocrinol Metab 97:1284-1293
27. Doghman M, Wakil A, Cardinaud B, Thomas E, Wang J, Zhao W et al (2010) Regulation of insulin-like growth factor-mammalian target of rapamycin signaling by microRNA in childhood adrenocortical tumors. Cancer Res 70:4666-4675
28. Wilkin F, Gagné N, Paquette J, Oligny L, Deal C (2000) Pediatric adrenocortical tumors: molecular events leading to insulin-like growth factor II gene overexpression 1. J Clin Endocrinol Metab 85:2048-2056
29. Boechat G, Stinghen S, Custódio G, Pianovski M, de Oliveira Figueiredo F, Jenkins J, Zambetti G, Ribeiro R, Figueiredo B (2011) Placental alkaline phosphatase in pediatric adrenocortical cancer. J Pediatr Hematol Oncol 33:149-153