This is inspired by Annalee's post on the main page about the young girl who temporarily beat her rare disease. Find it here:…

Remember, I'm an oncologist. So any non-cancer topic I discuss is out of my field, and definitely (at least to some extent) talking out of my ass. Also, sorry, there will be no snark or sarcasm here. This is too close to my heart.

For some diseases, there are a very few mutations and if you can target them then maybe you can control them indefinitely which is indistinguishable from a cure. It would be like insulin for diabetes - maybe not perfect but extending life by decades. Huntington's Chorea comes to mind. We can identify the mutation and tell people they have a good chance of dying a miserable death. If we had a therapy that could correct whatever problem that mutation caused, well, it would be a true miracle for those people.

I'm not sure of the details of the young girl's case, but it seems almost like Flowers for Algernon with the universe offering her a few weeks of normal life before laughing at her cruelly.

Moving on to something I know something about. Molecular medicine has fantastic promise. Diseases like chronic myelogenous leukemia (CML) can be controlled near to the point indistinguishable from cure because they almost all have the same basic mutation. A translocation between chromosome 9 and 22 puts the Break Cluster Region (bcr) gene next to the Abelson Murine Leukemia (abl) gene and the protein product of that gene gives the cell a competitive advantage, causing leukemia. This mutation occurs in the huge majority of patients with the disease and can be controlled with new drugs like imatinib that defeat the mutation and re-establish normalcy. Some people have been on imatinib for over a decade after suffering on less effective, more toxic treatments for years before that. They have literally been taking it longer than it has been available since they started the drug while it was still in trials. This is the miracle every cancer patient hopes for when he agrees to participate in a trial. The drug is so good that the manufacturers of newer, potentially slightly better drugs are disappointed in their sales. So many people are doing so well on imatinib, why would they switch?


But CML is what George Sledge, past president of the American Society of Clinical Oncology and one of the smartest people I ever met, called a "dumb cancer." It's the same mutation over and over again so it is a perfect target for molecular biologists who still have training wheels on their test tubes. It's stable, it's targetable. It makes sense for us to cut our teeth on that sort of disease. There are a few variants and they can be problematic, but it feels so good to actually change the natural history of this disease, the disappointment of not being able to help every single person is secondary.

Unfortunately, most cancers are "smarter" than that.

Most cancers have hundreds or even thousands of mutations, not just one stable one. I've heard it said that people with lung cancer have one mutation for every 5 cigarettes they have ever smoked. Four per pack, 1,400 per year for a pack a day smoker, tens of thousands for a lifetime habit. Many of these may be insignificant, but which ones are important? Which ones can be targeted? Figuring that out is only the first step.


So you found a mutation that can be targeted. Awesome! Get in line for your Nobel prize. Maybe you found that women who over-express the her-2/neu protein benefit from trastuzumab. We can help those women to live longer even if the cancer spreads and help to improve the cure rate if we find the cancer early. One of the greatest discoveries of all time in oncology. Maybe you found that the 5% of people with lung cancer with an ALK mutation will live longer and experience fewer side effects if they are treated with crizotinib as opposed to chemotherapy? Thank you so much! you have helped 10,000 people every year! But we still struggle to help people to live for two years . The problem is that there are so many other mutations in these cancers and the diseases find a way around the treatments. People still die and I still can't change that.

The problem comes up when you start to find multiple drugs. Which mutation should we target? Which drug works better? Does the order you give the drugs matter? The way traditional trials are done is you choose a treatment plan and compare it to another and see which does better in a specific population. You check for both efficacy and side effects. That is not so hard in lung cancer where you have 200,000 new cases a year in the US alone. But what if you are dealing with rare mutations? What if you have two mutations that occur in 5% of the population and you want to know what happens to people with both of them? That may be less than a thousand patients a year, even in a common disease. And only 3% of people in the US with cancer enroll in clinical trials. That's 30 people a year. A pitifully small sample size to test the side effects and efficacy of a new drug cocktail. How do you design that trial? How can we learn anything?

I am only scratching the surface here, but this is both the promise and the problem of the next wave of medical research. The more we know, the more we learn that the old cliche that I was taught in medical school - every patient is an individual - is truer than anybody ever meant it to be. We need new paradigms to conduct better research. We need physicians to be dedicated to enrolling patients in trials and we need patients willing to enroll in those trials so that the next generation of physicians will be able to give better advice to their children.


We have come so far, so fast, but it is so easy to get discouraged since every new answer raises more questions.