Liquid biopsies in pediatric solid tumors

Cancer remains a leading cause of childhood death in high-income countries, including the Netherlands. Solid tumors, in particular, continue to have poor survival rates despite advances in multimodal treatment. The main reason is the high recurrence rate, which current diagnostics fail to predict adequately. Standard techniques are not sensitive enough to detect minimal residual disease (MRD) – small numbers of tumor cells that persist during or after treatment. This thesis explores the use of liquid biopsies as a more sensitive and less invasive way to monitor disease, guide treatment decisions, and improve outcomes.

Part I focuses on MRD detection in children with neuroblastoma, the most common extracranial solid tumor in childhood. Neuroblastoma has a highly variable clinical course, with half of patients presenting with high-risk disease and survival below 50%. One of the strongest predictors of outcome is metastatic spread to the bone marrow, which is also the most frequent site of relapse.

We conducted a large multicenter prospective study analyzing bone marrow samples using mRNA marker panels. These molecular tests proved more sensitive and prognostically relevant than the current standard, immunocytology. Importantly, many patients thought to be in remission still harbored detectable tumor cells, which correlated with poor outcomes. We also developed new marker panels targeting mesenchymal tumor cells that escape conventional detection, and a multiplex assay combining multiple markers to optimize the use of small patient samples. These advances bring molecular MRD diagnostics closer to clinical application.

Part II investigates circulating tumor DNA (ctDNA) as a biomarker across several pediatric solid tumors, including neuroblastoma, nephroblastoma, rhabdomyosarcoma, and Ewing sarcoma. Tumor cells release DNA fragments into the bloodstream, which can be detected with advanced technologies. We showed that hypermethylated RASSF1A is a powerful biomarker, measurable in plasma with droplet digital PCR (ddPCR). This method enables precise quantification of ctDNA, reflecting tumor burden at diagnosis, during treatment, and at relapse.

Further studies extended ctDNA applications: shallow whole-genome sequencing allowed detection of chromosomal copy number variations; methylation profiling identified rhabdomyosarcoma subtypes; and combining ctDNA with microRNA biomarkers improved diagnosis of testicular germ cell tumors, even outperforming current clinical markers. Finally, we developed patient-specific ddPCR assays to track tumor-specific DNA breakpoints in cfDNA, demonstrating that ctDNA can provide individualized monitoring throughout treatment.

In conclusion, this thesis demonstrates that liquid biopsies—through both bone marrow mRNA analysis and ctDNA detection—can significantly improve the sensitivity of MRD detection, offer better prognostic information, and reduce the burden of invasive diagnostics for children with cancer. These methods hold great promise to refine risk stratification, guide therapy, and ultimately improve survival rates in pediatric solid