New Information on FL Survival

The Blood Cancer Journal published a very new article on Follicular Lymphoma this week that has (in my opinion) some very important things to say about FL treatments and survival. It isn't all necessarily brand new information, but it confirms some important things, and has some big implications.

Before we get to the specifics, I want to point out that this article looks at a lot of statistics, and makes statements about how treatments and FLIPI scores affect Overall Survival. So here is your very important reminder: Statistics do not have any impact on you as an individual. Numbers are not destiny. A number that describes the median Overall Survival for a large group does not describe each person in that group. If the median Overall Survival of a group of cancer patients is 10 years, that doesn't mean that any individual in that group will die in 10 years. It means half of them lived less that 10 years, and half lived more than 10 years. And notice the past tense -- those numbers are based on things that happened in the past, and may not account for the effects of newer treatments.

Click here to learn more about (or remind yourself about) how to read FL survival statistics. The important thing to remember is a statistic about a large group is not a prediction about you as an individual.

***************************

On to the article.

It's called "Follicular lymphoma in the modern era: Survival, treatment outcomes, and identification of high-risk subgroups." The study looks back at patients who were treated for FL at Memorial Sloan Kettering Cancer Center in New York between 1998 and 2009. The researchers were interested in a few things -- what the Overall Survival was for the group; the OS after each new treatment; and how FLIPI scores might change over time, and the possible effects of that change.

So here's the first good bit of news -- in this group of 1088 FL patients (all were grade 1 to 3A), the median Overall Survival had not yet been reached. That's always good news. If median OS measures the point at which half of the patients have died, then the longer it takes to get to that point, the better. The researchers point out that there have been "more than 20 years follow-up" with the earliest patients in the study, and the median OS has still not been reached. As I said, that's a very good thing, and it's in line with a lot of what I have seen lately in trying to figure out an OS for FL.

One of the two main lessons the researchers learned from this study has to do with OS and PFS after each line of treatment. By this, they mean a new treatment after the last treatment has stopped working. So someone might get B-R as a first line treatment. After a few years, they need treatment again, and maybe R-squared is the second line treatment. Then maybe RadioImmunoTherapy a few years later as a third-line treatment. And so on.

As they point out, there hasn't been much research on the OS and PFS for different lines of treatments. We look at OS for the whole group, but not how well patients do after each line.

Here's what they found:

They looked at 1088 patients altogether. 164 of them (about 15%) never needed treatment.

922 (85%) had a first line treatment. Again, median OS was not yet reached for this group, and the PFS was 4.73 years.

475 (43%) had a second line treatment. Median OS for this group was 11.7 years, and PFS was 1.5 years.

299 (27%) had a third line treatment. Median OS for this group was 8.8 years, and PFS was 1.1 years.

198 (18%) had a fourth line treatment. OS was 5.3 years and PFS was 0.9 years.

128 (12%) had a fifth line treatment. OS was 3.1 years and PFS was 0.6 years.

81 (7%) had a sixth line treatment.  OS was 1.9 years and PFS was 0.5 years.

It's easy to see that the more treatments a patient has, outcomes get worse. This makes sense -- a more aggressive form of Follicular Lymphoma will require more treatments. Again, I will remind you that numbers don't predict any individual's future. A 5.3 year OS after a fourth treatment means that half of those 198 patients lived longer than 5.3 years.

What I find really interesting about this group of statistics is the fact that fewer and fewer patients need more treatments. In other words, not all 922 patients needed 6 lines of treatment -- only 7% of them did. There are lots of reasons for that (including the age of "typical" FL patients). But for me, it was a nice reminder that what I learned when I was diagnosed wasn't right. I was told that FL gets more aggressive over time -- a first line treatment might last 8 years,  then the second line might last 3 years, then 2 years, etc. That's certainly the case for some patients, but not all. (And certainly not for me.) In my mind, I would be dealing with treatments every few years. For a lot of patients, we get long periods between treatments. This data supports that. And I think it's an important reminder, especially for newly diagnosed patients.

Of course, that shorter OS and PFS from line to line is important to the researchers. Because OS is so hard to measure (since FL patients are living so much longer), then PFS between lines might be an important benchmark for treatments. In other words, when a new treatment is being considered for approval, maybe a good way to think about it would not be "Does this treatment increase OS?" but instead "Does this treatment increase the PFS for patients who have received a fourth line of treatment?" A treatment might help a large group in a certain way, but looking more closely at the statistics, we can find that it helps a smaller group in a more significant way.

****************************

The second big lesson from the study had to do with FLIPI scores. I'm setting this off in a new section because I want to make sure this is explained clearly.

I've written about FLIPI before. It's the Follicular Lymphoma International Prognostic Index. The idea behind it was to look at a few different factors (originally patient age, FL stage, LDH levels, hemoglobin levels, etc). Patients were categorized as low, intermediate, and high risk, with worsening survival rates for each.

FLIPI is problematic for lots of reasons, and there have been lots of attempts to make it better. The important thing here is, like with any FL statistics, a particular FLIPI score does not indicate anything about you as an individual. My own FLIPI made me "intermediate risk" at diagnosis, giving me a 10 year survival of 51%. That was 12 years ago. A group's numbers are not your own destiny. Remember that.

So back to the study.

The researchers looked at the patients who watched-and-waited after diagnosis. Many of them never needed treatment. Some did eventually. The question they asked was, what happened to patients whose FLIPI score changed during the time they were watching and waiting?

A little context for this one. Many patients are able to watch and wait after diagnosis (I was one of them). Instead of getting treatment right away, our FL was stable, or less aggressive. My own oncologist had the philosophy of doing as little as necessary, so he suggested watching and waiting. There is no difference in OS between patients who wait and those who get treatment right away (which this study confirmed).

For those watch-and-wait patients who do need treatment after a while, there is some kind of event that triggers the need. For me, it was swelling in my left leg, caused by lymph nodes that were blocking some things and causing swelling. Other patients might have other events that trigger the need for treatment. There's no objective event that makes a doctor say "It's time to start treatment." Some might be more willing to let the waiting continue, as long as there are no major problems.

This study essentially asks if a change in FLIPI score might be that kind of objective event. If a patient's hemoglobin dropped, or LDH level went up, maybe that's a signal that the FL is getting more aggressive and treatment is necessary.

And that's just what they found. Patients who had a change in FLIPI score had a PFS of about 3 years, while patients who had a stable FLIPI had a PFS of 6 years. Change in staging, number of nodes involved, and LDH were the FLIPI factors that were most important.

The researchers suggest that FLIPI might be used to identify patients who are watching and waiting that might need treatment sooner than a doctor's clinical observation would suggest. In other words, if my doctor had paid attention to my FLIPI, he might have suggested treatment sooner than he did by seeing the potential problems associated with my swollen leg. The researchers suggest that maybe watch and wait isn't as good an idea as it might once have been.

This lesson is interesting to me, given that I am a part of the group that it looked at (though not a part of the study). But it seems kind of obvious to me that certain events (like more nodes being involved, or an increase in LDH evels) would make an oncologist take notice and suggest treatment. I guess the suggestion that something more objective, like a FLIPI score change, should be the measure, is a good one. But I also hope that any doctor who thinks watching and waiting is a good idea is also the kind of doctor who would look carefully at changes in data, and make careful decisions based on those changes.

**************************

So, for me, there are some good things to come out of this, and some other things that are not exactly bad,  but that do confirm some less positive things about FL that we already knew. (How's that for looking on the bright side?)

I do like that the long Overall Survival for FL has been confirmed. I like that the idea that everyone's FL will get more aggressive over time has been questioned. I like that watching and waiting works for lots of people.

The study also brings  anew set of data to researchers who can think about how it changes their views of things. There is a nice section on how this study matches up with some other big studies of FL patients. More knowledge is good.

The researchers also are clear that this study has limitations. It was a look at only one cancer center, so there is a chance that geography plays a role. It is also retrospective, looking back at patient records from 20 years ago, so it doesn't account for newer developments, and doesn't let researchers ask the kinds of questions that might wanted to have asked. They can only use the data that is available.

Still, it's a good opportunity to remind ourselves that statistics can tell researchers a lot, but they can't tell us anything about ourselves as individual patients. Each of our situations will be different. The best we can do is learn as much as we can, and work with an oncologist that we trust to get the best care for ourselves that we can.