I don't know about you, but about this time every couple of years I suddenly become very interested in soccer. The occasion, of course, is the staging of the FIFA World Cup -- women's or men's -- when even armchair enthusiasts feel qualified to pontificate on every pass, penalty and goal.
Fortunately, similarly obsessed friends of mine don't have to rely on my bumbling interpretations of the game to guess at the outcome. Instead there are actual qualified sports professionals experienced in integrating all the nuances of a match into a running win-loss predictions for each team. These types of statistical calculations are called in-game probabilities, and they've been used for decades to predict the likely winners (and losers) of sports matches and elections.
It might not seem at first that a technique used by bookies and pundits could be useful in a medical clinic. But Stanford oncologists Ash Alizadeh, MD, PhD and Maximilian Diehn, MD, PhD, together with instructor of medicine David Kurtz, MD, PhD, and postdoctoral scholars Mohammad Esfahani, PhD, and Florian Scherer, MD, recently devised a way to leverage a similar statistical technique to integrate multiple data points gathered during a cancer patient's course of treatment. They published their results today in Cell.
From our release:
Now researchers at the Stanford University School of Medicine have taken a page from this playbook to generate more accurate prognoses for cancer patients. They've done so by designing a computer algorithm that can integrate many different types of predictive data -- including a tumor's response to treatment and the amount of cancer DNA circulating in a patient's blood during therapy -- to generate a single, dynamic risk assessment at any point in time during a patient's course of treatment.
The researchers coined the name CIRI, for continuous individualized risk index, to emphasize the dynamic nature of the output. And in their retrospective study it appeared to more accurately predict a patient's prognosis than traditional methods.
As Kurtz explained:
Our standard methods of predicting prognoses in these patients are not that accurate. Using standard baseline variables it becomes almost a crystal ball exercise. If a perfectly accurate test has a score of 1, and a test that assigns patients randomly to one of two groups has a score of 0.5 -- essentially a coin toss -- our current methods score at about 0.6. But CIRI's score was around 0.8. Not perfect, but markedly better than we've done in the past.
The researchers first focused their analysis on people with diffuse large B-cell lymphoma, which is the most common blood cancer in the United States. But further study indicated that CIRI is also likely to be useful in predicting outcomes for patients with breast cancer and those with a common type of leukemia. Importantly, it might help clinicians quickly identify subsets of patients who may need more intensive therapy.
Although CIRI must still be verified in studies of recently diagnosed patients, the researchers are hopeful that the tool could one day provide life-changing information to patients.
As Alizadeh said:
When we care for our patients, we are walking on eggshells for a profound period of time while we try to determine whether the cancer is truly gone, or if it is likely to return. And patients are wondering, 'Should I be planning to attend my child's wedding next summer, or should I prioritize making my will?' We are trying to come up with a better way to predict at any point during a patient's course of treatment what their outcome is likely to be.
Photo by Jazmin Oteo