Background: Mediastinal non-seminomas (M-NS) have a poor prognosis compared to non-M-NS germ cell tumours (GCT) (i.e. primary gonadal or mediastinal seminomas (PG-S and M-S) and primary gonadal non-seminoma (PG-NS). M-NS tumors, while having pathological and serological similarity to other nonseminomas, are clinically and biological distinct and are considered IGCCC poor risk. In this study we aimed to evaluate the somatic mutation rate and prevalence of actionable/informative mutations in M-NS compared to M-S and primary GCT to inform consideration of targeted/precision therapy in these patients. Methods: Pre-treatment formalin fixed paraffin embedded specimens were collected from eleven patients with mediastinal and primary GCT patients (4 M-NS, 2 M-S, 3 PG-NS and 2 PG-S). High-density tumor areas were selected for DNA extraction and targeted sequencing of 578 established pan-cancer genes was performed (Roche Nimblegen SeqCaP EZ Comprehensive Cancer Design). Results: Consistent with the known landscape of testicular cancer, we identified five tumors with amplification of chromosome 12p, as well as hotspot mutations in KIT (n = 2) and KRAS (n = 1). Three of the 4 M-NS tumors had a non-synonymous mutation rate > 1 per Mb, which is higher than the primary TGCT and M-S sequenced here (range 0-0.5 per Mb) and the established low mutation rate for TGCT of 0.3-0.9 mutations/Mb. Mutations in TP53 (n = 2) and PTEN (n = 1) were only identified in patients with M-NS. The 2 patients with a TP53 mutation had a high non-synonymous mutation rate (1.3/Mb and 7.3/Mb), extensive metastatic disease with liver metastasis and high tumor markers (β-HCG: 1281 IU/L and α-FPT: 8500 ng/ml) at diagnosis. Both died of their disease and were the only patients in the cohort to have died. Conclusions: This study suggests that M-NS GCTs harbour more somatic alterations than non-M-NS GCTs, potentially underlying their poor prognosis. Our data are consistent with previous reports of an association between TP53 mutations and poor outcomes. Confirmation of our findings could lead to multi-center studies of targeted/precision therapies in patients with M-NS.