Systems proteogenomics for precision oncology

Genomics and particularly mutational profiling have deepened our understanding of cancer pathology and laid the foundation for precision therapies as a major field in modern oncology. Comprehensive genetic profiling has even led to proposals to replace the current histology-based WHO tumor typing with molecular classes based on the rationale that similar molecular alterations have similar clinical relevance across cancers [1-3]. However, targeted drugs against oncogenic mutations shown to be efficacious in one cancer are often ineffective in another [4]. But not only is it often not possible to transfer druggability of a mutation from one histological tumor type to another, response to targeted therapies shows significant variations also within the same tumor type. While some patients show long-term benefit, others quickly relapse or show no response to therapy despite identical actionable mutations. The reasons for this variability in biological behavior and therefore clinical relevance lie in the often complex genetic background in many tumors, but are also related to epigenetic, post-transcriptional and post-translational modifications not captured by mutational profiling alone. Non-small cell lung cancer, for instance, harbors over 800 genetic aberrations [5], on average, and although many of those are believed to be passenger mutations, a substantial number of mutations can be expected to be involved in the modulation of resistance mechanisms. However, the influence of these rare “tail” mutations is difficult to evaluate for two related reasons. First, even if a large number of mutations can be excluded based on prior biological knowledge, identifying drug combinations to overcome resistance is challenging even for a small number of mutations due to the combinatorial complexity (as an example, 780 alternative two-drug combinations exist for 40 targets [6]). Moreover, even if candidate targeted combination therapies are identified, patient recruitment for clinical trials is difficult due to the rarity of the druggable mutations. Therefore, ways beyond genomics have to be found to identify functionally and clinically relevant molecular alterations within the complex mutational landscape of cancer