In my last glioblastoma grant application, I made the following statement:
survival has barely moved in years, indeed the FDA’s approval of current standard treatment was based on improvement in median survival from 12.1 months (radiotherapy alone) to just 14.6 months (radiotherapy plus temozolomide)
I thought this was a great thing to say (and to put in italics) because it emphasizes the problem: This disease is terrible, and it’s obvious that our progress has been pathetic if an improvement of just a couple of months is the basis for current clinical practice. So, if I do anything at all to change mortality for the better, there should be enthusiasm for my grant—right? (As it happens, in this case enthusiasm was muted—the grant got a score of 30-odd, which in NIH-speak means “we don’t have anything to criticize about this proposal, but are giving you a score that will guarantee your grant doesn’t get funded.”)
We often tend to focus our efforts on the areas of cancer research that clearly need the most work. If we could improve outcomes for a disease like glioblastoma it would have a big impact on people’s lives. And, it seems to make sense that what we should do is try to understand why treatments don’t work, or try to find treatments that do work for these disease.
But, what if this isn’t the most effective way forward?
What if instead of focusing on the things we do badly we instead tried to understand what’s working when things go well?
Take testicular cancer, for example, where progress has been spectacular. Most people know that Lance Armstrong had advanced testicular cancer that had spread to other organs, and yet he was treated so effectively that he came back afterwards to repeatedly win the Tour de France. When talking about this to laypeople, I try to remind them that while Armstrong is an exceptional cyclist, he is, in fact, a typical testicular cancer patient–we appear to be able to cure almost all such men.
So, if 90% of men with metastatic testicular cancer have this outcome, why can’t it be the same for women with ovarian cancer, even when we try the same kinds of drugs?
Many researchers are trying to better understand ovarian cancer to improve women’s outcomes. Would it be better if, instead, we understood what’s going right in testicular cancer patients and applied that information to ovarian cancer patients?
Lets take this idea a little further and consider two types of leukemia where, in my mind, two clear success stories stand out (and they are the ones we always talk about).
First is the success story of chronic myelogenous leukemia (CML), which used to be another awful disease. I show my medical and pharmacology graduate students a figure that charts CML incidence and mortality every year since 1997.
- In 1997, if you were diagnosed with CML it was more than likely you’d die from the disease: 60% of the people diagnosed did.
- In 2002, CML incidence hadn’t changed, but mortality started to drop.
- In 2007 mortality was 10% of incidence and today it’s less than 5% of incidence.
That means that about 4,893 of the 5,150 or so Americans estimated to be diagnosed with CML this year will be cured, versus 2,060 who would have been cured 14 years ago.
In biomedical research, five years is almost no time at all, so this is an amazingly fast change in mortality. In fact, the rapidity of improvement shows that as soon as we knew what to do to save lives, the main hurdle was getting it implemented across the country.
Second is the success story of pediatric acute lymphocytic leukemia (ALL), the commonest leukemia in children. Dr. Steve Hunger, who leads the cancer center at Children’s Hospital Colorado, often shows a slide demonstrating survival of ALL patients over the past few decades. In the 1960s survival was dismal. Almost all the kids died of the disease.
Over the years as pediatric oncologists tweaked treatment regimens (with nasty old fashioned kinds of drugs) and kept at it, and at it. Very gradually, survival improved. Steve’s slide traces cohorts of patients ever few years, and you can see things always were just a little bit better as time went on. Nowadays most children with ALL are successfully treated and cured so that about 90 percent of the kids will survive.
The difference between the ALL success story and the CML success story is important: For CML, improvement came in essentially one single spectacular jump. We went from dismal to great in the space of just five years or so. Steve can correct me if I’m wrong, but for ALL there was no great leap forward, just a series of continuous baby steps in the right direction over decades.
The CML success was due to the fact that we understood that it is driven by a particular defect—activation of a kinase called Bcr-Abl—and because Brian Druker and others realized that targeting this defect with a drug—the famous drug Gleevec—should be enough to make a big difference.
The CML-Gleevec story is the poster child for rational, targeted therapy, which is of course why I talk about it when I’m teaching. Indeed, I’d go further and say that the CML-Gleevec story is the basis of much of our current effort in cancer therapy. We’re trying to find the equivalents of Bcr-Abl for subtypes of other cancers (CML is rather unusual because it is almost entirely driven by the same thing going wrong in almost every patient with the disease). The CML success makes good scientific success and serves as a model for what we are doing elsewhere.
But how about the successes with ALL or testicular cancer? We didn’t find a fantastic targeted agent that hit the specific defect that drives the cancer. In the case of ALL, we gradually learned to combine nasty toxic drugs so they worked better. Without any major single improvement we made spectacular progress.
Do we understand the science behind why today’s combination treatments for ALL or testicular cancer work? I’d say no. We know what steps were taken, what worked and what seemed not to work. We know how the pediatric oncologists were able to make the logistics come together to make progress. But I don’t think we understand why traditional cytotoxic drugs that you’d imagine do pretty much the same things in all cancer cells (unlike Gleevec, which only works in cancer cells with the right kind of kinase like Bcr-Abl) can be successfully combined to treat pediatric ALL but not other cancers.
So, what if we focused more on understanding the real underlying biological mechanisms that made the “baby step” success story possible? Would we gain insights into other situations where, despite our best efforts, we haven’t been able to keep those baby steps going?
Baby steps certainly aren’t as sexy as a near overnight success, such as that of Gleevec and CML. But you can win baseball games with lots of singles as well as a few home runs, can’t you? And winning the game counts for more than how you win it, especially if you wind up losing it!