Interobserver variability between experienced and inexperienced observers in the histopathological analysis of Wilms tumors: a pilot study for future algorithmic approach

Background Histopathological classification of Wilms tumors determines treatment regimen. Machine learning has been shown to contribute to histopathological classification in various malignancies but requires large numbers of manually annotated images and thus specific pathological knowledge. This study aimed to assess whether trained, inexperienced observers could contribute to reliable annotation of Wilms tumor components for classification performed by machine learning. Methods Four inexperienced observers (medical students) were trained in histopathology of normal kidneys and Wilms tumors by an experienced observer (pediatric pathologist). Twenty randomly selected scanned Wilms tumor-slides (from n = 1472 slides) were annotated, and annotations were independently classified by both the inexperienced observers and two experienced pediatric pathologists. Agreement between the six observers and for each tissue element was measured using kappa statistics (κ). Results Pairwise interobserver agreement between all inexperienced and experienced observers was high (range: 0.845–0.950). The interobserver variability for the different histological elements, including all vital tumor components and therapy-related effects, showed high values for all κ-coefficients (> 0.827). Conclusions Inexperienced observers can be trained to recognize specific histopathological tumor and tissue elements with high interobserver agreement with experienced observers. Nevertheless, supervision by experienced pathologists remains necessary. Results of this study can be used to facilitate more rapid progress for supervised machine learning-based algorithm development in pediatric pathology and beyond.

LAG-3 overexpression in pediatric Hodgkin lymphoma.

10531 Background: The role of the PD-1/PD-L1 axis in Hodgkin Lymphoma (HL) has led to FDA approval for use of inhibitors in chemotherapy-refractory HL. Numerous additional immune checkpoints may be useful targets, but have not yet been evaluated in HL. LAG-3, a related checkpoint, has been demonstrated to be overexpressed on tumor infiltrating lymphocytes (TILs) of a variety of cancers with associated poor outcomes. LAG-3 is known to inhibit T cell proliferation and activation, representing a possible therapeutic pathway for anti-tumor immunity. However, LAG-3 has yet to be evaluated in pediatric cancers. The purpose of this study is to characterize the expression pattern and clinical significance of LAG-3 in pediatric HL using immunohistochemistry. Methods: Patient tumor samples from prior Children’s Oncology Group clinical trials (AHOD0031) with matched patient outcome data containing 200 patient samples were obtained. 95% confidence intervals were calculated based on a range of observed prevalence of immune checkpoint expression. Using immunohistochemistry, paraffin embedded samples were tested for the expression of LAG-3 and PD-L1. Samples were stained for CD30 to better delineate Reed Sternberg cells from the remainder of the tumor microenvironment. Immune checkpoint staining was compared to positive controls of normal tonsil tissue, and negative controls of 3T3 cells. Expression intensity was scored by a Pediatric Pathologist. Results: 115 unique HL patient cases with evaluable HL tissue and correlating clinical outcome data were analyzed. Samples from 73/115 patients (63%) demonstrated positive LAG-3 staining, defined as over 10% of TILs expressing LAG-3. No demographic data including gender, race/ethnicity, age, or stage were significantly associated with LAG-3 expression. While not statistically significant there was a numerical difference in event free survival (EFS) and patients with LAG-3 expression demonstrated worse EFS. In terms of degree of LAG-3 expression, patients with the lowest positive expression were found to have the worst EFS, and those with highest expression demonstrated the best EFS. 97% of patient cases were found to be PD-L1 positive. 71/73 (97%) of patients who expressed LAG3 were also PDL1+, and 71/106 (67%) of PDL1+ cases were also LAG3+. LAG-3 and PD-L1 were found to be independent (p = 0.09). Conclusions: This project is innovative in its characterization of LAG-3 as an immune checkpoint target in pediatric HL. We hope that the information from this project will be used to support new clinical trials for pediatric patients with Hodgkin lymphoma.

Jean W Keeling

This is a short appreciation of Jean W Keeling as a pediatric pathologist.

Distinguishing Encephaloclastic Lesions Resulting From Primary or Secondary Pyruvate Dehydrogenase Deficiency From Other Neonatal or Infantile Cavitary Brain Lesions

The central nervous system (CNS) is a highly complex and energy-dependent organ that is subject to a wide variety of metabolic, hypoxic-ischemic, and infectious insults that result in cystic changes. Diagnosis of metabolic defects causing extensive cystic changes is particularly challenging for the pediatric pathologist, due to the rarity of these conditions. Pyruvate dehydrogenase (PDH) deficiency is one of the most common etiologies of congenital lactic acidosis, caused by mutations in subunits of the large mitochondrial matrix complex, and characterized by periventricular cysts, although few detailed reports focusing on neuropathologic findings exist. In addition, rare defects in other mitochondrial enzymes such as short-chain enoyl-CoA hydratase (SCEH, encoded by ECHS1 gene) can cause secondary PDH deficiency and present with neonatal lactic acidosis, but neuropathological findings have never been reported. Nonmetabolic conditions can also produce CNS cystic lesions, primarily in newborns. The pathologist must therefore distinguish between these etiologically disparate conditions which can produce CNS cavitary lesions. Here, we compare and contrast the gross and microscopic findings of cysts associated with cases of PDH and SCEH deficiencies with other neonatal cystic brain diseases including periventricular leukomalacia, neonatal Alexander disease, Canavan disease, and a case of cysts associated with a vascular abnormality. Our studies show that PDH and SCEH deficiencies are not grossly or histologically distinguishable from each other and both are associated with smooth-walled cysts largely limited to the telencephalic germinal matrix. Both show an absence of prominent hemosiderin deposits, Rosenthal fibers, vacuolization of the white matter, and gliosis or axonal damage in the surrounding parenchyma. These features can help distinguish PDH/SCEH deficiency from other pediatric/neonatal cystic CNS disorders, especially those produced by hypoxic ischemic conditions. Cysts, usually bilateral, confined to the telencephalic germinal matrix should elicit metabolic and genetic testing to appropriately diagnose PDH and SCEH and distinguish them from each other.

Histopathologic Analysis in Sudden Infant and Child Deaths

The forensic pathologist responsible for sudden unexpected death (SUD) investigation in the pediatric setting faces many challenges. It usually takes many years to obtain reasonable experience and exposure to the wide variety of diseases that may present as SUD in a pediatric context, and to appreciate the differences in the etiology and clinical context between the pediatric and adult SUD setting. In pediatric SUD, it is necessary to conduct a systematic, pediatric-focused autopsy investigation including extensive histopathological assessment and ancillary testing. Postmortem histologic findings in the context of SUD in the pediatric population are often subtle and distinctly different from those seen in the adult population. The pathologist must have an understanding of both developmental and pathological processes in order to correctly interpret the findings during a pediatric autopsy. A system-based, histopathology-focused review of common entities, normal variants, and incidental findings that can prove challenging will be discussed. For the forensic pathologist tasked with pediatric SUD autopsies, development of a strong collaborative relationship with a pediatric pathologist and/or neuropathologist to assist with histopathological analysis is strongly endorsed.

Pioneer in Pediatric Pathology: William A (Bill) Newton Jr (1923–)

William A (Bill) Newton Jr practiced pediatric pathology and hematology/oncology at Children’s Hospital of Columbus, Ohio, for over 40 years starting in 1952. Newton was an original member of the Pediatric Pathology Club, which preceded the Society for Pediatric Pathology, and was its president from 1968 to 1969. He published important independent observations in pediatric pathology, helped establish systematic cooperative pediatric tumor pathology review by experts, became an acclaimed expert on the diagnosis of rhabdomyosarcoma, was a critical contributor to many pediatric oncology clinical trials, made important early contributions to tumor banking in pediatrics, and trained numerous pediatric pathology and pediatric oncology fellows. Finally, he concluded his career as a humanitarian, leading important volunteer work aimed at improving pediatric cancer care in China. This most interesting pediatric pathologist was simultaneously a Brigadier General in the U.S. Army. Bill Newton’s life and career, which is reviewed in detail here, should be of immense interest and an inspiration to the Pediatric & Developmental Pathology readership.

Association of Retroplacental Blood With Basal Plate Myofibers

Objectives Diagnosed clinical abruption showing blood clot should be signed out in the pathology report as retroplacental hemorrhage with or without parenchymal indentation, and submitted clot separate from the placenta should be weighed. In our experience, some cases sent as clinical abruptions have been cases of morbid adherence. This study was undertaken to evaluate the association of retroplacental blood with basal plate myofibers (BPMF). Methods One hundred fifty-six placentas reviewed by a board-certified pediatric pathologist at a community hospital were evaluated for significant retroplacental blood. Basal plates were reviewed for deviations from normal. Results Of the 156 placentas, 33 (21%) had significant retroplacental blood; 21/156 (13%) had a separate clot, of which 11/21 (52%) had BPMF. Eleven BPMF-associated separate clots ranged from 10.5 to 60 g (average 23), while the clots of 10 cases with no demonstrated BPMF ranged from 19 to 440 g (average 82), tending to be larger ( p < .03). Basal plate damage prior to delivery was noted in both sets of placentas. BPMF placentas could have myometrial damage prior to delivery. Conclusions Since BPMF may confer a risk for accreta in a subsequent pregnancy, submission of a separate clot with the placenta should lead the pathologist to evaluate for basal plate myofibers on H&E and consider if there is an evidence-based indication to do an actin stain, before presuming a diagnosis of abruption.

Severe Fat Accumulation in Multiple Organs in Pediatric Autopsies

Background The observation of fat accumulation in different organs at pediatric autopsy may help determine the cause of death. However, a comprehensive study on fat accumulation and related etiologies is still lacking. Aim To investigate the incidence of fat accumulation in different organs in pediatric autopsies and their relevance to the cause of death. Methods From February 2004 until March 2015, fat accumulation was assessed in 561 routinely performed pediatric autopsies in our center. Age at death ranged from 14 weeks of gestation for fetuses to 16 years. Samples of liver, heart, kidney, and muscle were stained with Oil-Red-O and scored for fat accumulation by one pediatric pathologist. Results Fat accumulation in ≥1 organ(s) was present in 132 (39.8%) of 332 prenatal cases. Only 3 (0.9%) had fat accumulation in all 4 organs. For postnatal cases, fat accumulation in ≥1 organ(s) was present in 106 (46.3%) of 229 postnatal cases. Only 12 (5.2%) had fat accumulation in all 4 organs. Fat accumulation was mostly seen in liver, both in prenatal and postnatal cases. Fatty acid oxidation disorders were the only cause of death associated with severe fat accumulation in all 4 organs. No other etiologies could be linked directly to distribution or severity of fat accumulation. Conclusion Severe fat accumulation in at least 4 organs is a rare finding in pediatric autopsies, whereas fat accumulation in only the liver is relatively common. Severe fat accumulation in both liver, muscle, kidney, and heart was only seen in cases with fatty acid oxidation disorders.