How Netflix-style algorithm may help in personalising cancer treatments

London: Just as the Netflix algorithm generates data to determine which shows to binge-watch next, doctors will soon be able to predict how tumors may behave and offer personalized cancer treatments.

A typical Netflix-like algorithm analyzes data about how often a type of show or movie is watched and whether it’s fun or not. Based on the data, new movies and TV series will be recommended next time.

An international team of researchers from University College London (UCL) in the UK and the University of California, San Diego (UC San Diego) in the US have built a similar algorithm that will one day allow doctors to look at the fully sequenced tumor of adjust a patient and their treatment, the Independent reported.

MS Education Academy

The study, published in the journal Nature, showed that the algorithm used artificial intelligence (AI) to track changes in DNA from when cancer starts to how it grows. The team believes the insights into the genetic traits the tumor picks up as it grows will help doctors understand how a cancer is likely to behave.

“Just as Netflix can predict which shows you’ll binge watch next, we think we can predict how your cancer is likely to behave, based on the changes the genome has gone through before,” Dr. Ludmil Alexandrov, UC San Diego, was quoted as saying.

“Armed with that information, we believe that in the future doctors can provide better and more personalized cancer treatments,” he added.

The new algorithm can search through thousands of lines of genomic data and discover common patterns in how the chromosomes organize and arrange. It can then categorize the patterns that emerge and help scientists identify the types of errors that can occur in cancer, the team explained.

Using the algorithm, the team looked for patterns in the fully sequenced genomes of 9,873 patients with 33 different cancers and identified 21 common errors.

These will now be used to create a blueprint that researchers can use to assess how aggressive the cancer will be, find the weak spots and design new treatments for it.

Of the 21 signatures identified by the algorithm, the scientists found that tumors in which the chromosomes were shattered and reshaped were associated with the worst survival outcomes.

“To stay one step ahead of cancer, we need to anticipate how it adapts and changes. We missed the bigger picture of how huge chunks of genes can be copied, moved or deleted without catastrophic impact on the tumor,” said Dr Nischalan Pillay of UCL.

Leave a Comment

Your email address will not be published.