Role of Artificial Intelligence in genomics for advance cancer care

We are living in an amazing era of discovery in the field of oncology that continues to significantly influence cancer treatment, with a potential practice-changing impact. However, we are still in a nascent stage when it comes to a complete understanding of the biology of cancer.

Many major drugs have initial response, but the non-response rates vary between 30 percent and 70 percent. Patients suffering from the same type of cancer might respond differently to different drugs. Therefore, we have realized that the standard of care treatment, which is more organ-specific, might not be successful in all cases and cancer cannot be treated in the same way.

The success of genomic medicine can already be seen in clinical practice and the generalized treatment is shifting toward genome-based personalized treatment. Information of the comprehensive genomic profile of a patient is enabling the oncologist to break the cycle of trial-and-error medicine and adopt a tailored-action and evidence-based therapy/treatment plan. Targeted therapy, immunotherapy, and personalised radiation therapy are being offered in a variety of cancers with significant improvement in survival.

The power of genomics and personalized medicine is to offer the right diagnosis at the right time and choose the right treatment. It is to tell the patient that this particular treatment is not for all but only for the particular patient, based on his/her genomic signature.

We must understand that genomics is a big data field – something that requires computational approaches to interrogate the enormous volume of data generated by sequencing technologies, and to marry it in meaningful ways with other biological and clinical data. Artificial Intelligence (AI) and machine learning (ML) tools have great potential in recognizing patterns in large volumes of data, extracting relationships between complex features and characteristics in data (including images) that cannot be perceived by the human brain.

AI and ML tools will transform the analysis of the datasets, generated from genome profiling of cancer patients, and provide new biological insights in the following ways:

• Understanding accurately the complexity of the disease and predicting who will respond better and who will face the side effects of a particular treatment;
• Detecting the primary tumor of origin from liquid biopsy;
• Predicting progression and metastasis in cancer;
• Identifying and distinguishing disease-causing genomic variants from the benign variants; and
• Accurately identifying genetic disorders.

It is now a proven fact that cancer is essentially a disease of genome, which is caused by acquired or inherited genetic alteration. Genetic alternation that is harmful may increase a person’s chance or risk of developing a disease, such as cancer. Cancer can sometimes appear to run in families even if it is not caused by an inherited mutation. For example, a shared environment or lifestyle, such as tobacco use, can cause similar cancers to develop among family members. However, certain patterns – such as the types of cancer that develop, other non-cancer conditions that are seen, and the ages at which cancer typically develops – may suggest the presence of a hereditary cancer syndrome. The genetic mutations that cause many of the known hereditary cancer syndromes have been identified, and genetic testing can confirm whether a condition is indeed the result of an inherited syndrome.

Using AI and ML tools, new risk scores are being developed to identify younger adults who are most likely to develop cancer. Also, different algorithms are being generated for effective cancer screening by combining cancer biomarkers and molecular imaging. The early screening of such individuals and their family members who are at risk will immensely help in stringent monitoring and surveillance and prompt them to adopt simple lifestyle changes, including nutritious diet, maintaining a healthy weight, exercising, limiting alcoholic beverages, and avoiding tobacco products.

Therefore, the integration of AI technology in cancer care could improve the precision and ease of diagnosis and early detection, and help in right clinical decision-making the first time and better management. This will lead to better health outcome, reduce health disparities, and dramatically improve patients’ quality of life.