Personalization? Not Yet

Some troubling (or, potentially troubling) news from the world of cancer research, as reported in USA Today and other outlets: personalized treatments based on genetic profiling might not be as easy as we think.

The idea behind this kind of treatment is that cancer cells have a genetic footprint. If we can find the genetic problem with a particular type of cancer, then maybe we can base a treatment on it and find a cure. Follicular NHL, for example, has two chromosomes that are out of order (14 and 18, I think. I forget -- I can't see them, anyway, so who cares). Something made them switch positions. So researchers can look for treatments that can switch them back (if that's even possible), or target cells that have those genes switched.

This approach assumes that there is a genetic basis for cancer; this is the dominant assumption about cancer these days. Every type of cancer has its own genetic footprint.

So here's the problem: researchers found that, on some tumors, there are different cancer cells, with completely different genetic profiles, and if the cancer shows up someplace else, those cells can have different profiles as well. So a Follicular NHL  in one patient might have that 14, 18 switch going on in some cells, but 3, 16 in others, and a 9, 12 in still others.  So what happens is, a treatment is developed that targets one genetic problem, but it misses the others, because they weren't known about. So a bunch of cancer cells survive.

The articles I've read on this study present this as the death of personalized medicine.

Hardly.

It's a set back, to be sure. As least a potential set back. But here are a few things to remember:

First, this study involved on a very small number of patients, with one specific kind of cancer. What we know for sure is that these patients have several different genetic profiles in their tumors. But not everyone with their cancer will have the same genetic issues, let alone everyone with every type of cancer. That's the curse of genetic profiling -- it's shown us just how complex cancer is.

Of course, that makes it even more of a blessing, too -- we're getting a  better idea of just how complex cancer is. Isn't is just as likely that, now that we know about these multiple genetic abnormalities, we can start devising tests that look for them, and then corresponding treatments? I see this this as more blessing than curse. We're actually a step closer to understanding cancer. How can that be bad news?

Look, it's been 40 years since we declared war on cancer, and the fight still goes on. Anyone naive enough to believe that personalization would mean an immediate end to cancer should appreciate the wake up call. For the rest of us, we take our good news incrementally, and this is one more small piece to celebrate.

So start celebrating.