I seem to remember hearing a skeptic podccast that dug into this, and found the increase in rates to be well within the standard deviation for such cancers. No source though, so grain of salt, ect.