Precision Medicine and Drug Resistance in Cancer Laboratory aims to understand the underlying mechanisms of drug resistance in cancer. Our goal is to study how cancers evolve and develop resistance to therapies. We are interested in developing quantitative in vitro 2D/3D co-culture systems with stromal fibroblasts. We leverage single-cell barcoding and next-generation sequencing technologies to delineate actionable mechanisms of drug resistance. To better address patient specific drug resistance, we generate ex vivo Patient-Derived Organoids (PDOs) as a preclinical model system. Specifically, Acar laboratory has three main research direction:

• Existing experimental model systems to study treatment resistance have several limitations such as lack of quantitative evaluation of the process, clinically infeasible drug concentrations used in the existing experimental models failing to capture the true dynamics of cancer growth and relapse observed in the clinic. Further, the majority of these model systems fail to account for the tumour microenvironment, which is known to mitigate drug sensitivity. Cancer-associated fibroblasts (CAFs) are one of the key components of the tumour microenvironment and are known to facilitate drug resistance; however they have been largely absent from in vitro modelling. We aim to derive improved, more clinically relevant, experimental model systems to study treatment resistance.

• Recently, Patient-Derived Organoid (PDO) technologies have advanced substantially, and they can now successfully mimic both characteristics of the disease and response to the drug treatment. Living tumours derived from patient fresh tumour samples recapitulate phenotypic characteristics when cultured in ex vivo 3D culture conditions and they have been shown to be genetically and phenotypically identical to parental tumours. With a close collaboration with Hacettepe and Ankara University Hospitals, we have started the establishment of a Patient-Derived Organoid (PDO) biobank for the first time in Turkey. This biobank will permit us to assess drug sensitivities in a patient-specific context and to test the efficacy of treatment options derived in the pre-clinical setting. Further, to stratify response to different second line therapies, we aim to perform next-generation sequencing (NGS) to determine the underlying genetic and phenotypic drivers of treatment resistance.

• Digital pathology is constantly developing and whole-slide image (WSI) based environment empowers the accession and interpretation of pathological information hidden in specimens. Primary diagnosis for cancers primarily depends on histopathological examinations. As a team, we are working on developing and optimising deep learning models for histopathological WSIs to enable the acquisition, management and extraction of the information that are crucial to decision making processes in cancer. These models include image analysis, critical measurements and pattern recognition algorithms to aid for the diagnosis of malignant tissues.