Milea Timbergen

335 Chapter 4 examines the genome-wide DNA methylation profiles of S45F and T41A mutated DTF. MeD-seq analysis was performed on 29 primary DTF tumours including 15 T41A samples and 14 S45F samples. In total, 365 regions were identified as differentially methylated regions (DMRs). Clustering analysis yielded no clear separation was observed between the two mutation groups. Only few DMRs had a fold change of ≥ 1.5 implicating no difference in methylation between mutation types. No differences in expression of the corresponding genes of those DMRs between the two mutation types were confirmed in an independent mRNA Affymetrix dataset. Immunoprecipitation did not reveal an association of wild-type β-catenin or mutant variants with DNMT1. Part II - Diagnosis and Treatment Chapter 5 describes the use of a new technology: radiomics that related large amounts of quantitative imaging features to clinical outcome. In this chapter we used radiomics in an effort to simplify the diagnosis of DTF by creating a model that could distinguish extremity DTF from other extremity soft tissue sarcoma (STS) on a pre-treatment T1-weighted (T1w) magnetic resonance imaging (MRI) scan. Furthermore a prediction model was created for determining the CTNNB1 mutation type. The MRI data encompassed tumours from 203 patients and included 72 patients with DTF and 131 patients with STS. Tumours were semi-automatically annotated on the T1w images, from which 424 features were extracted. The T1w radiomics model showed a mean area under the curve (AUC) of 0.79 on the full dataset. Addition of T2w and post-contrast sequences did not improve the performance. On the location matched cohort, the T1w model had a mean AUC of 0.88 while the radiologists had an mean AUC of 0.80 and 0.88, respectively. For the prediction of the mutations (S45F, T41A and wild-type), the T1w model showed an mean AUC of 0.61, 0.56, and 0.74. From this research we have to conclude that the radiomics model was able to distinguish DTF from STS with high accuracy similar to two radiologists, but was not able to predict the CTNNB1 mutation status. In Chapter 6 we performed a systematic literature search to systematically evaluate the results of retrospective studies describing the active surveillance approach. A total of 24 articles were included describing a total of 3541 patients of which 40.7% (n = 1404) received active surveillance. The majority were females and the majority had a primary tumour. About 20% of patients had progressive disease, 58% had stable disease and 20% of patients had a partial response. About 30% of patients needed shift to an active form of 13