The article outlines a new explanation for cancer's mysterious genesis and proliferation throughout the human body. Its a similar article to the book review Carl wrote about epigenetics. I don't recall his book review focused specifically on cancer as a disease- it was more of a discussion about how gene therapy and epigenetics could combat a variety of conditions (I would need to reread it to be sure).
The article in The Economist mentions that genes are switched on and off by plastering the DNA or surrounding proteins with chromosomes containing clusters of acetyl and methyl groups. This would explain why Carl mentioned methylation as carbon, hydrogen and oxygen. He merely forgot to mention that the methylation can also sometimes be acetylation. Acetylation does, in fact, include oxygen. So- instead of getting the definition of methyl group wrong perhaps he forgot to include the additional functional group (acetyl). Then, carbon, hydrogen and oxygen would all be involved, as correctly noted in the article.
Glad I got this ironed out. Didn't think Carl was prone to errors.