Summary Accurate pathological diagnosis is crucial for optimal management of cancer patients. For the ~100 known central nervous system (CNS) tumour entities, standardization of the diagnostic process has been shown to be particularly challenging - with substantial inter-observer variability in the histopathological diagnosis of many tumour types. We herein present the development of a comprehensive approach for DNA methylation-based CNS tumour classification across all entities and age groups, and demonstrate its application in a routine diagnostic setting. We show that availability of this method may have substantial impact on diagnostic precision compared with standard methods, resulting in a change of diagnosis in up to 12% of prospective cases. For broader accessibility we have designed a free online classifier tool (www.molecularneuropathology.org) requiring no additional onsite data processing. Our results provide a blueprint for the generation of machine learning-based tumour classifiers across other cancer entities, with the potential to fundamentally transform tumour pathology.

The recent development of quinoline-based PET tracers that act as fibroblast-activation-protein inhibitors (FAPIs) demonstrated promising preclinical and clinical results. FAP is overexpressed by cancer-associated fibroblasts of several tumor entities. Here, we quantify the tumor uptake on 68 Ga-FAPI PET/CT of various primary and metastatic tumors to identify the most promising indications for future application. Methods: 68 Ga-FAPI PET/CT scans were requested by various referring physicians according to individual clinical indications that were considered insufficiently covered by 18 F-FDG PET/CT or other imaging modalities. All PET/CT was performed 1 h after injection of 122-312 MBq of 68 Ga-FAPI-04. We retrospectively identified 80 patients with histopathologically proven primary tumors or metastases or radiologically unequivocal metastatic lesions of histologically proven primary tumors. Tumor uptake was quantified by SUV max and SUV mean (60% isocontour). Results: Eighty patients with 28 different tumor entities (54 primary tumors and 229 metastases) were evaluated. The highest average SUV max (.12) was found in sarcoma, esophageal, breast, cholangiocarcinoma, and lung cancer. The lowest 68 Ga-FAPI uptake (average SUV max , 6) was observed in pheochromocytoma, renal cell, differentiated thyroid, adenoid cystic, and gastric cancer. The average SUV max of hepatocellular, colorectal, head-neck, ovarian, pancreatic, and prostate cancer was intermediate . SUV varied across and within all tumor entities. Because of low background in muscle and blood pool (SUV max , 2), the tumor-to-background contrast ratios were more than 3-fold in the intermediate and more than 6fold in the high-intensity uptake group. Conclusion: Several highly prevalent cancers presented with remarkably high uptake and image contrast on 68 Ga-FAPI PET/CT. The high and rather selective tumor uptake may open up new applications for noninvasive tumor characterization, staging examinations, or radioligand therapy. ://jnm.snmjournals.org/content/60/6/801 This article and updated information are available at: http://jnm.snmjournals.org/site/subscriptions/online.xhtml Information about subscriptions to JNM can be found at: http://jnm.snmjournals.org/site/misc/permission.xhtml