OPINION:
Billions of dollars are spent annually on researching and improving cancer treatments, such as chemotherapy, gene therapy and immunotherapy.
There has been an improvement in cancer survival rates, but much of the improvement is skewed by an early-detection bias. This makes survival times look longer without changing when a patient dies.
At any rate, there will still be more than 600,000 cancer deaths in the U.S. this year. At the same time, biochemical profiles remain incomplete due to the thousands of unidentified biochemicals collectively known as the dark metabolome. This is an area of research that could uncover clues leading to more cost-effective treatments and cures — not only for many forms of cancer, but for other intractable diseases too.
Advances in machine learning and artificial intelligence have made it possible to find patterns, trends and correlations in a huge amount of data — even when each data point consists of hundreds of variables.
Think of each person studied as a data point and the level of each biochemical as one of hundreds of variables associated with it. If analytical chemists were given the resources to develop methods to separate, identify and quantify as many of these unknown biochemicals as possible, then data scientists could uncover patterns that could have been hidden up till now.
If this is a worthwhile goal, is enough work being done to achieve it?
Philadelphia, Pennsylvania
Please read our comment policy before commenting.