Decoding Cancer’s Past to Predict its Future: A New Approach to Tumor Evolution
Researchers at IDIBAPS in Barcelona are pioneering a new method to understand cancer development by deciphering patterns within DNA methylation – chemical modifications to DNA that don’t change the sequence itself, but can alter gene expression. Initially dismissed as “background noise,” these patterns are now being revealed as a rich source of information about a tumor’s history.
The team, led by researcher Martín-Subero, has developed a methodology utilizing artificial intelligence to interpret these methylation patterns. By applying mathematical models, they’ve been able to reconstruct the evolutionary trajectory of tumors from a study of 2,000 samples. This allows them to determine when a tumor began to grow, its growth rate, and the degree of diversity within its cells.
“What we discarded before, now we have found that it is indeed a gold mine, gives us information that was hidden into the eyes of the whole world,” Martín-Subero explains.
This “secret history” revealed by methylation patterns has potential clinical implications. The researchers have created a tool to anticipate how a cancer will evolve, potentially predicting future aggressiveness and the timing of necessary treatment, even for slower-progressing cancers like chronic lymphocytic leukemia.
While promising, the research is still in its early stages.Manel Esteller, a research professor at the josep carreras Leukemia Institute, emphasizes the need for further experimental validation before clinical implementation. He notes that existing DNA methylation analysis techniques already combine biological verification with advanced algorithms to achieve similar results, especially in leukemia.
Though, Alejo Rodríguez Fraticelli, an ICREA researcher at the Institut de Recerca Biomèdica de Barcelona (IRB), highlights the low cost of the newly developed technique as a meaningful advantage. He believes the ability to use epigenetic information as “molecular barcodes” to track disease progression at a low cost is a true innovation.
Martín-Subero acknowledges that the methodology isn’t yet ready for widespread clinical use, requiring a company to translate the research into a practical request. Still, he stresses its cost-effectiveness and its potential to fundamentally improve our understanding of cancer biology. ”If we know the past of cancer, we can advance to your future and make a better management of clinical resources for a better treatment and better patient prognosis estimate,” he concludes.
