With a reported incidence of 5.86 patients/100 000 per year, sarcoma forms one of the largest groups within the rare cancers [1]. However, over 70 histological and morphological subtypes have been identified, making it the most heterogeneous type of rare cancer [2]. This combination of rarity and heterogeneity poses a therapeutic problem. Surgical resection is the cornerstone of sarcoma treatment, but additional treatment in a neo-adjuvant or adjuvant setting is commonly necessary to improve patient outcomes [3,4]. The systemic treatment guidelines for sarcoma are however far from ideal, which is mainly caused by a scarcity of clinical trials [3,4]. Molecular approaches are emerging to identify actionable mutations but have yet to be validated [5]. This is why the need for models is maintained for the development of novel drugs, drug repurposing, functional validation of molecular findings and even precision medicine [6,7]. The use of novel models for personalised medicine such as patient-derived tumour fragments and organoids is increasing, and therefore the necessity for live tumour tissue rises [[8], [9], [10], [11]]. For the development and validation of representable models or personalised treatment, a large collection of viable samples with well-annotated patient data is required [12,13]. As many biobanks are based on formalin-fixed, paraffin-embedded tissue, obtaining viable tissue is challenging. It is important to standardise the procedures for both collecting and biobanking viable tissue to ensure sample quality and the possibility of generating high-quality models [[13], [14], [15]]. This bedside-to-bench step can be challenging since it often relies upon the connection between researchers and clinicians.

We developed a bedside-to-bench workflow in cooperation with surgeons, pathologists and researchers to facilitate living tissue biobanking of a multitude of sarcoma subtypes. This surgeon-led approach increases the quality of the biobanked tissue. All clinical data are available to the research team in pseudonymised fashion. By correlating preclinical research to clinical data, research quality is improved.