People died of cancer long before doctors even knew there was such a thing. In the 1840s, examinations by a man named Rudolf Virchow produced a series of diagrams of uncontrolled cell growth. These sketches became the basis for much of the way we diagnose cancer today.
Even with 21st-century technology, not much has changed. Though our powers of imaging have never been stronger, doctors are still making diagnoses based on positively medieval standards. It's as much divination as it is a science, according to Otis Brawley, the American Cancer Society's chief medical officer and executive VP.
"The pathologist says, "This fits the profile of what Virchow called cancer,'" Brawley told delegates at TEDMED, a three-day conference on health and medicine. "They're not really saying, 'This is cancer'; they're saying this is what Virchow called cancer in the 1840s."
Think about that. We're still diagnosing cancer today, over 170 years after it was given a name, exactly the same way that Virchow did: by comparing one set of images to another.
As we've discovered over the years, not all tumors are lethal. Some are benign. So, one of the major challenges in cancer diagnosis is determining which kind of tumor a patient has. Because of the uncertainty associated with imaging interpretation, we actually end up overtreating cancer. That's both more costly and more dangerous than would be optimal.
What we need, said Brawley, is to overhaul cancer screenings, whose usefulness is vastly overrated. Mammograms account for only a fraction of the 30 percent decline since 1991 in breast cancer mortality rates. In fact, most of the gains have come from awareness and early detection.
"We actually need to have a genomic definition of cancer," said Brawley, "a modern day definition so we can say that this thing that looks like cancer is inherently programmed to either stay at one centimeter -- never move, never metastasize, never cause harm -- or it's genomically programmed to metastasize. The cancer we need to watch versus the cancer we need to treat."
We want to hear what you think. Submit a letter to the editor or write to firstname.lastname@example.org.