“OFF-LABEL” USE.

“OFF-LABEL” USE. I argued here that testing of new cancer drugs should make greater use of multiple regressions (in addition to controlled experiments). The article I linked to yesterday by Gina Kolata about Avastin cites an estimate that 75% of cancer drug use is “off-label.” (The use of a drug is “off-label” when a drug that has been approved by the FDA for one use is prescribed for a different use.) The article includes an estimate that up 75% of cancer drug use is “off-label.” This extensive “off-label” use must be developing large amounts of information about the effectiveness of drugs against tumors with different characteristics. The article refers to some “studies of a type generally considered less than definitive”, but I think it would be good if these kinds of studies were done systematically on a large scale.