Single-cell atlas of developing murine adrenal gland reveals relation of Schwann cell precursor signature to neuroblastoma phenotype

Neuroblastoma is the most common extracranial solid tumor and accounts for ∼10% of pediatric cancer-related deaths. The exact cell of origin has yet to be elucidated, but it is generally accepted that neuroblastoma derives from the neural crest and should thus be considered an embryonal malignancy. About 50% of primary neuroblastoma tumors arise in the adrenal gland. Here, we present an atlas of the developing mouse adrenal gland at a single-cell level. Five main cell cluster groups (medulla, cortex, endothelial, stroma, and immune) make up the mouse adrenal gland during fetal development. The medulla group, which is of neural crest origin, is further divided into seven clusters. Of interest is the Schwann cell precursor ("SCP") and the "neuroblast" cluster, a highly cycling cluster that shares markers with sympathoblasts. The signature of the medullary SCP cluster differentiates neuroblastoma patients based on disease phenotype: The SCP signature score anticorrelates with ALK and MYCN expression, two indicators of poor prognosis. Furthermore, a high SCP signature score is associated with better overall survival rates. This study provides an insight into the developing adrenal gland and introduces the SCP gene signature as being of interest for further research in understanding neuroblastoma phenotype.

Laparoscopic radical 'no-touch' left pancreatosplenectomy for pancreatic ductal adenocarcinoma: technique and results

Background

Laparoscopic left pancreatectomy has been well described for benign pancreatic lesions, but its role in pancreatic adenocarcinoma remains open to debate. We report our results adopting a laparoscopic technique that obeys established oncologic principles of open distal pancreatosplenectomy.

Methods

This is a post hoc analysis of a prospectively kept database of 135 consecutive patients undergoing laparoscopic left pancreatectomy, performed across two sites in the UK and the Netherlands (07/2007–07/2015 Southampton and 10/2013–07/2015 Amsterdam). Primary outcomes were resection margin and lymph node retrieval. Secondary endpoints were other perioperative outcomes, including post-operative pancreatic fistula. Definition of radical resection was distance tumour to resection margin >1 mm. All patients underwent ‘laparoscopic radical left pancreatosplenectomy’ (LRLP) which involves ‘hanging’ the pancreas including Gerota’s fascia, followed by clockwise dissection, including formal lymphadenectomy.

Results

LRLP for pancreatic adenocarcinoma was performed in 25 patients. Seven of the 25 patients (28 %) had extended resections, including the adrenal gland (n = 3), duodenojejunal flexure (n = 2) or transverse mesocolon (n = 3). Mean age was 68 years (54–81). Conversion rate was 0 %, mean operative time 240 min and mean blood loss 340 ml. Median intensive/high care and hospital stay were 1 and 5 days, respectively. Clavien–Dindo score 3+ complication rate was 12 % and ISGPF grade B/C pancreatic fistula rate 28 %; 90-day (or in-hospital) mortality was 0 %. The pancreatic resection margin was clear in all patients, and the posterior margin was involved (<1 mm) in 6 patients, meaning an overall R0 resection rate of 76 %. No resection margin was microscopically involved. Median nodal sample was 15 nodes (3–26). With an average follow-up of 17.2 months, 1-year survival was 88 %.

Conclusions

A standardised laparoscopic approach to pancreatic adenocarcinoma in the left pancreas can be adopted safely. Our study shows that these results can be reproduced across multiple sites using the same technique.