Sentinel node procedures (SNP) are performed with the use of tracer-agents, mainly radio-colloid and/or blue dye. Fluorescent agents have emerged as a new tracer-agent to identify the SLN intra-operatively with near-infrared imaging. Our aim is to compare the detection rate of fluorescent agents to current "golden standards" (blue dye and/or radio-colloid) for the SNP by means of a systematic review and meta-analysis without any restrictions based on tumor type. A systematic search in PubMed, Embase and The Cochrane Library was performed. Articles that compared the detection rates of fluorescent agents with radio-colloid and/or blue dye were included. Meta-analyses were performed for breast, gynecological and dermatological cancer using a random effects model. In total 6195 articles were screened which resulted in a final inclusion of 55 articles. All studies used indocyanine green (ICG) as fluorescent agent. Meta-analyses comparing ICG with blue dye showed a significant and clinically relevant difference in detection rate in favor of ICG, for both breast, dermatological and gynecological cancer. Meta-analyses comparing ICG with radio-colloid did not show any significant differences, with the exception of ICG versus radio-colloid + blue dye for the bilateral SLN detection in gynecological cancer. Near-infrared fluorescence imaging using ICG provides a higher detection rate compared to blue dye for the SNP in a range of different tumor types. SLN detection rates of ICG are comparable to radio-colloid. Due to their complementary characteristics in terms of spatial resolution and transdermal sensitivity, we suggest to use a combination of both ICG and a radio-colloid.

Central venous access device (CVAD)-related complications are associated with high morbidity rates. This study was performed to underline the importance of CVAD-complication prevention and treatment. An audit of practice of CVAD-related complications in pediatric oncology patients receiving a CVAD between January 2015 and June 2017 was performed. CVADs included were totally implantable venous access ports (TIVAPs), Hickman-Broviac® (HB), nontunneled, and peripherally inserted CVADs. A total of 201 children, with 307 CVADs, were analyzed. The incidence rates per 1000 CVAD-days for the most common complications were 1.66 for malfunctions, and 1.51 for central line-associated bloodstream infections (CLABSIs). Of all CVADs inserted, 37.1% were removed owing to complications, of which 45.6% were owing to CLABSIs. In 42% of the CLABSIs, the CLABSI could be successfully cured with systemic antibiotic treatment only. Of all included patients, 5.0% were admitted to the intensive care unit owing to CLABSI. The HB-CVAD compared to the TIVAP was a risk factor for CVAD-related complications, CLABSIs and dislocations in particular. The incidence of CVAD-related complications is high. Research on the prevention and treatment of CVAD-related complications in pediatric oncology patients should be a high priority for all health care professionals. Prognosis study (retrospective). Level II.

Despite druggable events to be present in 80 % of neuroblastomapatients within the Princess Máxima Center precision medicine program 'iTHER', clinical uptake of treatment recommendations has been low, and the clinical impact for individual patients remains hard to predict. This stresses the need for a method integrating genomics and transcriptomics with functional approaches into therapeutic decision making. We aimed to launch an online repository integrating genomics and transcriptomics with high-throughput drug screening (HTS) of nineteen commonly used neuroblastoma cell lines and fifteen neuroblastoma patient-derived organoids (NBL-PDOs). Cell lines, NBL-PDOs and their parental tumors were characterized utilizing (lc)WGS, WES and RNAseq. Cells were exposed to ∼200 compounds. Results were transferred to the R2 visualization platform. A powerful reference set of cell lines is available, reflecting distinct known pharmacologic vulnerabilities. HTS identified additional therapeutic vulnerabilities, such as a striking correlation between a positive mesenchymal signature and sensitivity to BCL2-inhibitor venetoclax. Finally, we explored personalized drug sensitivities within iTHER, demonstrating HTS can support genomic and transcriptomic results, thereby strengthening the rationale for clinical uptake. We established a dynamic publicly available dataset with detailed genomic, transcriptomic, and pharmacological annotation of classical neuroblastoma cell lines as well as novel sharable NBL-PDOs, representing the heterogeneous landscape of neuroblastoma. We anticipate that in vitro drug screening will be complementary to genomic-guided precision medicine by supporting clinical decision making, thereby improving prognosis for all neuroblastoma patients in the future.

Purpose. Due to the size and localization of Wilms tumor (WT), nephron-sparing surgery (NSS) is only possible in a limited number of cases. When NSS is considered, the surgeon preoperatively requires a thorough understanding of the patient-specific anatomy to prevent positive surgical margins and other complications. Through a collaboration between the radiology and pediatric surgery departments and 3D imaging specialists, a 3D visualization workflow was developed to improve preoperative planning of NSS for WT patients. Methods. The 3D visualization workflow combines a MRA sequence, a segmentation protocol, and augmented reality (AR) visualization, additional to in-house 3D printing. A noncontrast-enhanced MRA scan was added to the MRI protocol. MRI sequences were segmented with a segmentation protocol in an open-source software package. The resulting 3D models were visualized in AR with a HoloLens and 3D print. Results. In a pilot study, five WT patients eligible for NSS were preoperatively planned through the 3D visualization workflow. AR visualization software was fast and free to use and allowed adequate handling of the 3D holograms. The 3D printed models were considered convenient and practical for intraoperative orientation. The patient-friendly, fast, and low-cost 3D visualization workflow was easily implemented and appeared to be valuable for the preparation of NSS. Conclusion. This pilot study demonstrates how a strong collaboration between the pediatric surgery and radiology departments and 3D imaging specialists will help to shape the future of pediatric oncological surgery. This 3D visualization workflow aims to prepare pediatric oncological surgeons for nephron-sparing surgery in patients with Wilms tumors.

Wilms tumor is a common pediatric solid tumor. To evaluate tumor response to chemotherapy and decide whether nephron-sparing surgery is possible, tumor volume measurements based on magnetic resonance imaging (MRI) are important. Currently, radiological volume measurements are based on measuring tumor dimensions in three directions. Manual segmentation-based volume measurements might be more accurate, but this process is time-consuming and user-dependent. The aim of this study was to investigate whether manual segmentation-based volume measurements are more accurate and to explore whether these segmentations can be automated using deep learning. We included the MRI images of 45 Wilms tumor patients (age 0–18 years). First, we compared radiological tumor volumes with manual segmentation-based tumor volume measurements. Next, we created an automated segmentation method by training a nnU-Net in a five-fold cross-validation. Segmentation quality was validated by comparing the automated segmentation with the manually created ground truth segmentations, using Dice scores and the 95th percentile of the Hausdorff distances (HD95). On average, manual tumor segmentations result in larger tumor volumes. For automated segmentation, the median dice was 0.90. The median HD95 was 7.2 mm. We showed that radiological volume measurements underestimated tumor volume by about 10% when compared to manual segmentation-based volume measurements. Deep learning can potentially be used to replace manual segmentation to benefit from accurate volume measurements without time and observer constraints.