©2024 - Stefano Forte
Decision Support Systems in Cancer Genomics - Tumor Mutational Burden
In recent years, the use of Next-Generation Sequencing (NGS) in oncology has revolutionized our understanding of cancer at the molecular level, providing clinicians with detailed genetic information that can guide diagnosis and treatment. However, the sheer volume and complexity of NGS data present significant challenges. Clinicians and researchers must sift through vast datasets to identify clinically relevant mutations, interpret their impact on cancer progression, and understand their implications for therapeutic options. This complexity underscores the urgent need for a decision support system (DSS) that can effectively analyze and annotate NGS data, helping healthcare providers make informed decisions that enhance patient outcomes. Our research focuses on developing software platforms designed to simplify, refine, and enhance the information obtained from molecular analysis of tumor samples using NGS sequencing. These platforms integrate data from various third-party databases to create a comprehensive and detailed overview of pathology, therapies, and contraindications. The outcome of the analysis is a thorough and easy-to-understand report on the genetic profile of both the patient and the tumor, assisting oncologists or geneticists in selecting the best approach to address the disease.
Decision Support System (DSS) In Oncology
Improving Immuno-oncology
In the field of precision medicine, one of the key challenges in oncology is identifying biomarkers that can predict which patients are most likely to respond positively to immunotherapy. This is crucial for personalizing treatment approaches, as immunological agents—while effective for some patients—may not benefit others, leading to unnecessary side effects and costs. Developing reliable predictive factors to guide patient selection is, therefore, an essential step toward optimizing cancer treatment outcomes. In our study, we conducted a comprehensive pan-cancer analysis of Tumor Mutational Burden (TMB), a biomarker that has shown promise in predicting responses to certain immunotherapies. To support this, we developed an innovative computational pipeline called TMBcalc, designed to accurately calculate TMB. TMBcalc is unique in its ability to identify compact, reliable gene signatures that can be used to estimate TMB, even when using custom targeted-sequencing panels rather than whole-exome sequencing (WES). Our pipeline was rigorously trained on data from 17 different cancer types within The Cancer Genome Atlas (TCGA), ensuring its applicability across a broad spectrum of cancers. Our results indicate a strong correlation between TMB calculated using the gene signatures identified by TMBcalc and TMB obtained from traditional WES methods. This suggests that our approach is both efficient and effective, offering an alternative means of calculating TMB that could be more accessible and feasible in clinical settings where WES may not be practical. We validated the robustness of TMBcalc through extensive testing on several independent datasets, providing a comprehensive assessment of its accuracy and reliability. Specifically, we tested our method on (i) 126 samples from the TCGA database and (ii) multiple external WES datasets focused on colon, breast, and liver cancers, with data sourced from the European Genome-phenome Archive (EGA) and the International Cancer Genome Consortium (ICGC) Data Portal. This rigorous evaluation demonstrated that TMBcalc can produce reliable TMB estimates across diverse cancer types, highlighting its potential for use in clinical practice. Our findings underscore the practicality of TMBcalc as a tool to facilitate more personalized treatment plans, paving the way for more precise, data-driven approaches in immuno-oncology.
©2024 - Stefano Forte