In this paper we provide a structured literature analysis focused on Deep Learning (DL) models used to support inference in cancer biology with a particular emphasis on multi-omics analysis. The work focuses on how existing models address the need for better dialogue with prior knowledge, biological plausibility and interpretability, fundamental properties in the biomedical domain. We discuss the recent evolutionary arch of DL models in the direction of integrating prior biological relational and network knowledge to support better generalisation (e.g. pathways or Protein-Protein-Interaction networks) and interpretability. This represents a fundamental functional shift towards models which can integrate mechanistic and statistical inference aspects. We discuss representational methodologies for the integration of domain prior knowledge in such models. The paper also provides a critical outlook into contemporary methods for explainability and interpretabiltiy. This analysis points in the direction of a convergence between encoding prior knowledge and improved interpretability.