How to Read Survival Statistics

A couple of weeks ago, a reader made a comment about Overall Survival, and I gave a brief response. But I know there are some new readers here who were diagnosed fairly recently, so I think it's a good time for a reminder about "Overall Survival" and what it means.

It's especially important for folks who are newly diagnosed. For most of us, when we heard those words -- "You have cancer" -- our minds immediately went to some variation of "How long do I have?" Survival statistics serve as a kind of easy way to figure out how worried we should be. 

But survival statistics don't tell us anything about our own situations. It's easy to read a statistic that says the survival rate for a cancer is X years, and say "Well, I only have X years left." That's just not true. That's not how statistics work.

I'll use myself as a quick example. When I was diagnosed in 2008, the Median Overall Survival for Follicular Lymphoma was thought to be 8-10 years. Here I am, 16 years later. 

One of the problems with that 8-10 year statistic is that it was probably already outdated when I saw it. At that point, Rituxan had been in use widespread for FL for about 10 years (it was approved by the FDA in 1997). Rituxan is generally credited with increasing the Median Overall Survival, which had probably already done by the time I was diagnosed. But that 8-10 year figure was from the last large study of  FL survival, done before Rituxan, and looking backwards anyway, measuring the survival rate of people who were diagnosed in the 1980s. 

So that's the first danger with looking at survival statistics and thinking they are a measure of your fate -- they're probably outdated, maybe very outdated, and don't take into account any newer treatments.

As for the other problems, let's look at the term "Median Overall Survival," which is the official term for survival statistics. 

First of all, it's important to remember what "Overall Survival" is measuring. This statistic looks at death from any cause, not from the disease being considered. A statistician looks at a bunch of records of cancer patients, marks which have died, and notes when it happened. Those deaths aren't distinguished by cause. So it includes people who died from their cancer. And people who died from complications of their treatment. But it also includes people who died from heart disease, car accidents, getting too close to wildlife at national parks -- any cause of death. "Overall Survival" simply looks at how many people died, and how long they lived after they were diagnosed.

That's especially important for Follicular Lymphoma patients. We are generally diagnosed with FL sometime in our 60's (I haven't seen a solid statistic for this anytime recently). So for a group of people who are diagnosed fairly late in life, we can probably expect a lifespan of 15-20 years. That just makes sense -- anyone at the age of 65 (whether they have FL or not) probably has a life expectancy of about 15-20 years. 

And that is indeed about what researchers put as Median Overall Survival -- about 15-20 years. In other words, as a group, we're pretty much in line with the general population.

The other important word in this formulation is "Median." That's a term from statistics. It's different from "average," because one big number can throw off an average. "Median" means the exact middle. So if you had 999 Follicular Lymphoma patients, and you lined them up based on how long they survived after diagnosis, then number 500 in that line up would be the median. (There would 499 on either side of that patient.)

Since the "median" is the exact middle, that means 499 of those patients would have a shorter survival than #500. But it also means that 499 would have a longer survival.

Think about that. If #500 has a survival rate of 20 years, that means half of us will have a survival rate of greater than 20 years. How much greater? Maybe 20 years and 1 day. Maybe 25 years. Maybe 50. 

That's especially important for us who were diagnosed on the young side. Like me -- I was diagnosed at 40 years old. That doesn't mean I can expect to only live a few more years. The patients on the right side of #500 don't have an upper limit. Statistically, I could live until I'm 90 -- a 50 year survival rate. 

(By the way, that's how long I expect to live -- at least.)

But what about those folks who were to the left of #500? That still doesn't mean anything. It includes people who didn't know they had FL until very late in their lives, maybe in their late 70s, and then had a normal life expectancy. It includes people who didn't get a diagnosis until their disease was very far along. It includes people who walked out of the doctor's office after getting a diagnosis and had a piano fall on them 5 minutes later. 

The point is this -- statistics are not destiny. The numbers that describe a group of thousands of people, especially because they are collected by looking at the past, don't tell us anything about us as individuals. Our lives are all different, and our diseases are different.

I personally stopped looking at statistics a long time ago. When I was first diagnosed, the things that threw me into a depression were almost always statistics.  Numbers seem so definite, so correct, so real. And it's very easy to get sucked up into them and take them as the truth about ourselves. But that;s just not how it works. 

So if you want some numbers, here is a good one. 

A long time ago, a reader told me that her doctor was discussing treatments with her, and how excited he was about there being so many new options in the pipeline. The doctor told her "If we can keep a Follicular Lymphoma patient alive for 5 years, we can keep them alive for 50 years."

And that conversation happened before CAR-T, before bispecifics, before R-squared. And who knows what other treatments that might be coming soon.  

If you feel yourself getting pulled down by numbers, keep that one in mind.

I hope this helps some newer folks to better understand what survival statistics actually mean, and how little significance you should give to them. And I hope those of you who have been around for a while appreciate the reminder. Numbers are not destiny.

Take care.