Current treatments for colorectal cancer (CRC) patients show disappointing therapeutic outcomes. Therefore, there is a compelling need for better therapy choices. The newest approaches using genomic analysis for precision medicine enable the identification of an actionable cancer driver gene present in the tumor for further targeting (eg, KRAS, BRAF, etc). However, cancer may be promoted by the contribution of several genes rather than the alteration of one or two genes alone. Consequently, current targeted therapeutic strategies are not improving long-term outcomes or progression-free survival. Hence, comprehensive genetic models are required to offer more accurate diagnosing aimed to identify bona personalized therapeutic strategies. We employed a method developed at the Icahn School of Medicine at Mount Sinai (NY), which enables simultaneous targeting of multiple mutations driving tumorigenesis. We first identified the whole genomic landscape associated with the patient’s tumor. Next, we reconstructed this genetic complexity, including up to 20 cancer-associated altered genes present in the patient's tumor, in the last portion of the intestine of the fruit fly Drosophila melanogaster. Thus, this fly developed a CR tumor genetically similar to that of the patient, creating a most complete avatar model. Subsequently, fly avatars were expanded to up to half a million per patient and were then used to screen the full FDA/EMA drug libraries. Finally, effective drug cocktails identified were presented to the patients and oncologists. We present here a unique methodology to identify personalized CRC drug treatments based on individual patients' entire tumor genomes. This technology has already demonstrated improved progression-free survival in terminal CRC patients. The result is a fully customized treatment program, comprising on- and off-label oncology drugs and non-cancer drugs. Our platform allows the design of N-of-1 clinical studies aimed to identify the specific genetic factors that are necessary for tumor growth in an individual patient, and the best drug combination to tackle it. This novel platform can make possible a more precise diagnosis, and together with avatar modeling and in vivo drug screening, may make it possible to identify fully tailored therapeutics. By addressing the patient tumor genomic complexity, this personalized practice-changing approach may provide an alternative and more efficient treatment option for individual CRC patients.