Histopathological criteria on the diagnosis of pediatric adrenal cortical carcinoma: a retrospective review of 3 histopathological knowing scoring systems in a 35 years series.
Victor Pereira Paschoalin, urologist PhD1, Silvio Tucci, Jr., Professor and head of Urology2, Jose de Bessa, Jr., Associate professor of Urology3, Andrey Garcia Estevanato, Associate Physician of Urology2, Ricardo Brianezi Tiraboschi, urologist and PhD student4, Sonir Roberto Ramber Antonini, Full professor of pediatric endocrinology2, Carlos Augusto Fernandes Molina, Assistant professor of Urology2.
1Private Practice, Feira de Santana / Bahia, Brazil, 2Faculdade de Medicina de Ribeirão Preto Universidade de São Paulo, Ribeirão Preto / São Paulo, Brazil, 3Universidade Estadual de Feira de Santana, Feira de Santana / Bahia, Brazil, 4Faculdade de Medicina de Ribeirão Preto, Ribeirão Preto / São Paulo, Brazil.
BACKGROUND: Adrenal tumors, excluding neuroblastoma, are sporadic in childhood and adolescence, reaching an incidence of 1.5/million/year and corresponding to 0,2% of all malignant tumors in the same age group. International Pediatric Adrenocortical Tumor Registry (IPACTR) revision in 2004 reported that the diagnostic of Adrenocortical Carcinoma (ACC) occurs on average five months after the onset of symptoms and 38% die along the first 2.5 years of follow-up, turning in only 54% the overall survival. The ACC in childhood and adolescence still being classified following histopathological Weiss criteria, developed for adults, has no specificity in prognostic issues and is not entirely recommended in the pediatric population. In an attempt to create a classification more particular for the pediatric population, Wieneke established specific criteria modifying Weiss ones and adding morphometrics data building a new algorithm that reach superior specificity on diagnosis and prognosis for the ACC, which was validated by three retrospective studies. By considering the concernment of clinical prognostic based on histopathological diagnostic to establish therapeutic management, this study aimed to compare, retrospectively, the accuracy of 3 histopathological classifications for ACC driven by Weiss, Van Slooten and Wieneke to predict follow-up of this disease in the pediatric population. It is a 35 years series of ACC surgical patients.
METHODS: Data were retrieved from medical records of 57 patients <18 years old with surgical follow-up over 6 months or death due to ACC along this time separated in two groups, "good prognosis" (without recurrence/death due to ACC) and "bad prognosis" (with recurrence/death due to ACC). Two apart expert pathologists reviewed the ACC surgical specimen classifying them following Weiss, Van Slooten and Wieneke criteria. The slide was built from 5mm paraffin block slice of the sample stained with hematoxylin and eosin.
RESULTS: The median time of follow-up was 126[18-225] months in 38 males (66.7%) and 19 females (33.3%) with a median of ages of 3[1-6.5] years. The median of 3[1.5-6.2] years in "good prognosis group" was smaller than 5[2-10] years in "bad prognosis" group. Overall, 17 (29.8%) patients died due to ACC placed inside the "bad prognosis" group. All deaths occurred within the first 50 surgical months of follow-up, and the earliest one was on the fourth follow-up month and the majority of them along 24 months of follow-up.
Histopathological classification considering the prognosis group:
| FOLLOW-UP Histopathological Criteria | "Bad" Prognosis | "Good" prognosis |
| Weiss | ||
| Malign | 16 | 33 |
| Benign | 1 | 7 |
| Van Slooten | ||
| Malign | 17 | 30 |
| Benign | 0 | 10 |
| Wienecke | ||
| Malign | 16 | 12 |
| Benign | 1 | 28 |
Accuracy of each classification system:
| PARAMETERS | CLASSIFICATION | ||
| Weiss | Van Slooten | Wieneke | |
| Sensitivity | 100% | 94% | 94% |
| Specificity | 17% | 25% | 70% |
| Predictive Positive Value | 36% | 32 % | 57% |
| Predictive Negative Value | 100% | 87% | 96% |
| Likelihood Ratio + | 1.33 | 1.14 | 3.13 |
| Likelihood Ratio - | 0 | 0,35 | 0,08 |
| Accuracy | 40% | 47% | 77% |
CONCLUSIONS: The accuracy of Weiss and Van Slooten classification wasn't greater than flip a coin. Wieneke classification was better than the others, but the accuracy was not good enough to establish a reliable prognosis for the ACC on the pediatric population. However, the sensitivity and negative predictive value of Wieneke classification was around 95%.
Back to 2019 Abstracts
