Purpose: Treating refractory or relapsed neuroblastoma remains challenging. Monitoring body fluids for tumor-derived molecular information indicating minimal residual disease supports more frequent diagnostic surveillance and may have the power to detect resistant subclones before they give rise to relapses. If actionable targets are identified from liquid biopsies, targeted treatment options can be considered earlier. Experimental Design: Droplet digital PCR assays assessing MYCN and ALK copy numbers and allelic frequencies of ALK p.F1174L and ALK p.R1275Q mutations were applied to longitudinally collected liquid biopsies and matched tumor tissue samples from 31 patients with high-risk neuroblastoma. Total cell-free DNA (cfDNA) levels and marker detection were compared with data from routine clinical diagnostics. Results: Total cfDNA concentrations in blood plasma from patients with high-risk neuroblastoma were higher than in healthy controls and consistently correlated with neuron-specific enolase levels and lactate dehydrogenase activity but not with 123I-meta-iodobenzylguanidine scores at relapse diagnosis. Targeted cfDNA diagnostics proved superior for early relapse detection to all current diagnostics in 2 patients. Marker analysis in cfDNA indicated intratumor heterogeneity for cell clones harboring MYCN amplifications and druggable ALK alterations that were not detectable in matched tumor tissue samples in 17 patients from our cohort. Proof of concept is provided for molecular target detection in cerebrospinal fluid from patients with isolated central nervous system relapses. Conclusions: Tumor-specific alterations can be identified and monitored during disease course in liquid biopsies from pediatric patients with high-risk neuroblastoma. This approach to cfDNA surveillance warrants further prospective validation and exploitation for diagnostic purposes and to guide therapeutic decisions.

Optic nerve sheath meningiomas (ONSM) are rare but can lead to irreversible blindness. Hybrid imaging may enhance tumor delineation and diagnostic accuracy via receptor binding. However, relevant clinical data for ONSM are lacking. We evaluated the feasibility of receptor-based hybrid imaging prior to robotic radiosurgery (RRS). We retrospectively analyzed all of our institution’s patients with suspected ONSM who underwent combined positron emission tomography and magnetic resonance imaging (PET/MRI) with gallium-68-labeled (DOTA0-Phe1-Tyr3) octreotide (Ga68-DOTATOC) before RRS between 2018 and 2019. Eight patients with ten suspected ONSM (female = 7; median age, 51.2 years; IQR, 43.0–66.0) were included. Nine out of ten ONSM were deemed PET-positive with a median standard uptake value (SUV) max of 5.6 (IQR, 2.6–7.8). For all nine ONSM that presented 68Ga-DOTATOC uptake, hybrid PET/MRI was used for target volume contouring prior to RSS. At a median follow-up of 11.7 months (IQR, 9.4–16.4), tumor control was achieved in all patients. Radiosurgery resulted in the improvement of visual acuity in two of eight patients, whereas six showed stable vision. Ga68-DOTATOC-PET/MRI can be used for target volume contouring prior to RRS for ONSM as it enables safe treatment planning and improves diagnostic accuracy.

To evaluate the diagnostic performance of magnetic resonance imaging (MRI) alone in comparison to positron emission tomography/ magnetic resonance imaging (PET/MRI) in patients with meningiomas. 57 patients with a total of 112 meningiomas of the brain were included. PET/MRI, including a fully diagnostic contrast enhanced MRI and PET, were acquired. PET/MRI was used as reference standard. The size and location of meningiomas was recorded. Likelihood-ratio chi-square tests were used to calculate p-values within logistic regression in order to compare different models. A multi-level logistic regression was applied to comply the hierarchical data structure. Multi-level regression adjusts for clustering in data was performed. The majority (n = 103) of meningiomas could be identified based on standard MRI sequences compared to PET/MRI. MRI alone achieved a sensitivity of 95% (95% CI 0.78, 0.99) and specificity of 88% (95% CI 0.58, 0.98). Based on intensity of contrast medium uptake, 97 meningiomas could be diagnosed with intense uptake (93.75%). Sensitivity was lowest with 74% for meningiomas < 0.5 cm3, high with 95% for meningiomas > 2cm3 and highest with 100% for meningiomas 0.5–1.0 cm3. Petroclival meningiomas showed lowest sensitivity with 88% compared to sphenoidal meningiomas with 94% and orbital meningiomas with 100%. Specificity of meningioma diagnostic with MRI was high with 100% for sphenoidal and hemispherical-dural meningiomas and meningiomas with 0.5–1.0 and 1.0–2.0 cm3. Overall MRI enables reliable detection of meningiomas compared to PET/MRI. PET/MRI imaging offers highest sensitivity and specificity for small or difficult located meningiomas.