This article focuses on the use of deep learning synthetic data generation methods to assess the risk of future treatments and medication for preventing venous thromboembolism (VTE) in cancer patients, based on a small dataset of genetic and clinical variables. The study employs CopulaGANs to generate synthetic tabular data, which is then used to train a Deep Learning-based classifier using domain adaptation techniques. The trained model is fine-tuned using real data and performs better than current state-of-the-art medical scores in assessing VTE risk. Additionally, the resulting Precision-Recall curve offers flexibility in selecting different and better operational points for VTE risk assessment.