Happy Statistics (Or, There's Always a Bright Side)

As a scientist, I've always felt that information is of central importance. You find out something--you may love it, or hate it--but a fact is a fact. This is reality.

I originally had a very small, early-stage breast tumor back in 1996. The treatment was local--simple, limited surgery ("lumpectomy"--a nice combination of vernacular and technical), and 6 weeks of radiation. Then, in 2007, I developed breathing trouble, and we found that the cancer had spread, or metastasized, to my lung sac, chest wall, and several bones. Long story short, I had several months of pretty brutal chemotherapy, and have been receiving a comparatively benign drug regimen ever since.

As this story unfolded, I knew that my prospects for long-term survival were pretty terrible, but I wanted to find out as much as I could about my particular disease state. Early on, I stumbled across some interesting graphs, which I circulated to my family with a textual explanation that combines the numerical facts with a dose of possibly unjustifiable optimisim--I call this "informed denial." I've been updating this narrative as time passes....These two graphs are from the National Cancer Institute / Surveillance Epidemiology and End Results (SEER) web site and are based on a large amount of data reported for cancer patients during the years indicated in the titles.

(Title: Relative Survival Rates by Stage at Diagnosis For Breast Cancer, All Races, All Ages, Females; SEER 9 Registries for 1988-2003)

OK, with my diagnosis ("distant"=metastatic), I am the green diamonds on the rather sadly plummeting bottom line. The median survival after my diagnosis is 2 years—pretty grim. But that’s only 50% of my group who are doomed at 2 years—and as it’s been longer than that for me, I’m obviously not in that bunch. Some people have a bad disease at diagnosis—multiple tumor sites in lungs, liver, brain—and they don’t live long. I had only a very few sites…minimal visceral involvement, and no brain/spinal cord metastasis.

And, although the survivor line plunges down at first, it begins to level off after a while, suggesting the presence of a base group of people who live for a loooooonnnnnnnnggggggg time. Why, at 10 years after diagnosis, more than 10% of my group is actually still alive! Is there any reason why I shouldn’t be in that 10%? I think not!

Now, for something similarly interesting—conditional survival rates. This means, if you live X years, what is your chance of living another 5 years? Or more precisely, for a group that survives X years, what percent of them will live another 5 years? The next graph shows these patterns for the different groups.

(Title: Conditional 5-Year Relative Survival Rates by Stage at Diagnosis, For Breast Cancer, All Races, Females, All Ages; SEER 9 Registries for 1975-2003)


Isn’t it nice to see the curves pointing up? This simply means that the longer you live with this disease, the better the chance that you will continue to survive. Seems logical to me—and this pattern is especially noticeable for the “distant” (green diamond) group members, who have the lowest survival to begin with.

Since we’re among friends (please forget for a moment what I said about statistics in the opening blog), we can loosely interpret these group survival percentages as probabilities that an individual person in the group who survives the “conditional period” (X years) will survive another 5 years.

That is, having now lived more than 5 years after my diagnosis of metastasis, I have about a 50-50 chance of making it to 10 years. And if I make it 10 years, the chance is almost 75% that I’ll be around for another 5. Sounds good to me!

Again, this graph suggests a steady core of metastatic cancer, “distant” (green diamond) survivors, who are going to make it for a long, long time. Whatever the odds, I intend to be in that core.