How is AI Revolutionizing the Future of Cancer Care?
Exactly how and to what extent AI is set to transform the lives of cancer patients is still a mystery to many.
Johan's medical journey is a fictional story inspired by AI's extraordinary progress in hematological cancers, an area where AI-assisted studies are already revolutionizing medical practice.
This story is intended to raise public awareness of the major role AI will play in medicine as current advances continue to be validated and commercialized.
The fictional case of Johan
Johan is a 65-year-old man with no pre-existing medical conditions, including hypertension and hypercholesterolemia, and who jogs 4 miles every day. He is admitted to hospital with an acute myocardial infarction on September 24, 2028. He received standard treatment, including thrombolytics which helped preserve his cardiac function following the infarction. Further investigations were carried out to determine why this apparently healthy patient suffered a myocardial infarction and revealed the presence of Clonal Hematopoiesis of Indeterminate Potential (CHIP).
What is CHIP?
Clonal Hematopoiesis of Indeterminate Potential (CHIP) is a precancerous condition that was first described in 2014. CHIP is associated with the development of hematological cancers in 4% of patients carrying specific mutated genes.
It has also been associated with a higher incidence of myocardial infarction (2-4 times higher than in patients without CHIP), thrombosis and, potentially, other diseases such as diabetes mellitus. In Johan's case, myocardial infarction was his mode of entry into the disease.
How has AI helped Johan?
Since Johan was diagnosed with CHIP in 2028, his blood is routinely checked for 19 classical parameters at 3-month intervals. AI algorithms are able to identify patterns that may indicate hematological malignancies by integrating large datasets from multiple sources, such as clinical symptoms, laboratory tests, imaging studies or genetic information (AI surveillance). As soon as an abnormality is detected, a notification is sent to Johan and his hematologist, who will organize an immediate follow-up visit.
In February 2034, a blood sample reveals that Johan has 2 abnormalities that his GP would have considered insignificant in standard blood tests. Johan returns to his hematologist for further investigations. He is then diagnosed with a very early form of Acute Promyelocytic Leukemia, considered low risk because it was caught early. AI enables early diagnoses by analyzing genetic mutations, biomarkers and other molecular characteristics of tumors.
Already today, machine learning (ML) models can identify certain genetic mutations associated with specific hematological malignancies. What is still missing, however, is a systematic inventory of all the mutations capable of triggering these different malignancies. The performance of AI models clearly surpasses the classifications established by experts on the basis of their practical experience of the characteristics of these genes. In the near future, all the genes involved in the appearance of these pathologies will undoubtedly be identified (AI enables precision medicine). Since all this data will have been catalogued and stored globally for these conditions, they can be diagnosed much earlier, with the prospect of a much better prognosis.
Johan started treatment with trans-retinoid acid (TRTA) and arsenic trioxide (ASO), to which an additional drug was added after the discovery of a specific mutation in a particular gene. AI facilitates the development of personalized treatment plans by analyzing patient-specific data, including genetic profiles, treatment history, and response to prior treatments.
ML models are able to determine the most appropriate treatment for certain combinations of mutations, a task that would be too complex and time-consuming for human researchers. This translates into better results with fewer side effects. What's more, AI research is also making great strides towards identifying sub-groups of patients requiring more aggressive treatment (part of precision medicine).
In February 2039, Johan is now 76 years old. For 6 years now, he has enjoyed a complete response to his treatment, with no relapses. He continues to enjoy his retirement, taking daily walks and playing golf every other day.
Final Thoughts
This fictional story of Johan is not science fiction. It's already happening. Soon, these techniques will be further validated and commercialized.
All along his medical journey, Johan benefited from advances in AI, especially in the field of hematological malignancies. It goes from early diagnosis (CHIP and Acute Promyelocytic Leukemia) to the identification of high-risk genes, to screening and surveillance, which caught his leukemia very early on, and lastly to personalized treatment.
The outcome: 11 years later, Johan is still alive, with a high quality of life, enjoying his retirement. This story would have been very different if it had happened just 15 years earlier.
