There is nothing more disheartening to the patient, and quite frankly for the treating oncologist, than have to hear (or say) the words “I’m afraid the treatment isn’t working”. The scientific term is “disease progression”, but the reality is that the cancer is growing despite the treatment and it doesn’t take an expert to know that isn’t good news.
However, next in line in the statements that seem to generate a decided lack of enthusiasm is “Your cancer is stable”. Stable doesn’t sound so bad, does it? The tumors haven’t changed, nothing is growing, there are no new lesions to suggest progression, but all of these reassurances often seem to fall on deaf ears. Sure, we’ll keep going with the treatment, but it MUST be bad that the tumors didn’t shrink, right? I sometimes think my patients feel I’m being dishonest with them when I get enthusiastic about something as ambiguous as disease stability.
Of course, it isn’t hard to understand this feeling. Who wouldn’t want a clear sign that the cancer is dying, that the treatment is working. Patients want their cancer to be gone, and watching it disappear must be very heartening. I also admit that there is little in my job as gratifying as looking over the scan and seeing that the tumors have all shrunk, what is known as an “objective response”. I’ll often pull this type of scan up on the exam room computer to show off my treatment prowess: look at what I did! But is a response really that important?
Let’s look at tumor responses for a minute. If the tumors shrink but don’t disappear this is called a “partial response” (or PR) and if they disappear entirely this is called a “complete response” or (CR). Response rates have been the traditional marker of whether a chemotherapy drug is effective, and historically drugs that did not cause tumors to shrink have been abandoned. Stable disease (SD) was usually reported separately, if at all, and has been regarded by some as the poor man’s response rate. If your trial failed and there were no responses, at least you can report a high SD rate and maybe someone will think your drug is good for something!
But this concept has changed a lot in the modern era. Trials utilizing so-called “targeted agents” have even reported improvements in overall survival with few or any objective tumor responses. One good example of this is the randomized trial of Nexavar (sorafenib) in hepatocellular (liver) cancer. This trial showed a 3 month improvement in overall survival in patients with HCC treated with sorafenib, despite only 2% of patients “responding” to the drug. Trials like this one have raised the question of just how important response rate is to determining if a drug is helpful or not.
But that was for these new, fancy-schmancy drugs like Nexavar or Tarceva. What about ordinary chemotherapy? Well, at the 2009 ASCO meeting in Orlando there was a poster that I found fascinating but I haven’t heard too much buzz about. A group of investigators from Japan, led by Dr. Hirokazu Watanabe, looked at the survival of advanced NSCLC patients treated with 4 different platinum-based chemo regimens as part of a clinical trial and related the survival to whether the patients had a response, stable disease, or progressive disease as their best outcome.
The results were very interesting. Patients who had an objective response to chemo had a median survival of 16.1 months, pretty good for an advanced lung cancer trial. The patients who had progressive disease had a median survival of only 5.5 months, which was (obviously) significantly worse. But what was so interesting is that the patients who had only stable disease had a median survival of 15.2 months, which was not statistically different from the responders. The survival at 2 years was 33% in the SD group compared to 30% in the responders, raising the question of just how important it is to have actual tumor shrinkage. At least in this trial, it didn’t seem to matter at all.
The concept of benefit in patients who have stable disease isn’t really new, of course. In recent years studies have been reporting something called the “Disease Control Rate” which is the response rate + the stable disease rate. But for patients, I think it is harder to grasp that it is really just as good that the cancer is exactly the same as to have it shrink. Lately I have started being very specific before starting treatment that all we are really hoping for is to keep the cancer from growing (something that is actually more common than getting a response), and that this SD is every bit as good as a response. Hopefully this tempers the disappointment just a little when stable disease is the result.
Of course, this won’t stop me from pulling up the scans to show my patients when the tumors do shrink. It makes a good show, it feels good for us both, and everyone wins. Stable disease is kind of like that vegetable you hate (brussel sprouts, anyone?); you know in your head that it’s good for you, but that doesn’t make it easier to swallow.