Nick Carne
Medical researchers in the US are developing an algorithm they hope can generate more accurate prognoses for cancer patients by taking a leaf out of the sports playbook – or, perhaps more accurately, the sports betting world.
As a game unfolds, a team’s chances of winning change, and with it the coach’s tactics. Similarly, a patient’s chances of recovery change over time – and in an ideal world that information would be available to inform the approaches that are considered.
With this in mind, the Continuous Individualised Risk Index (CIRI) created by a team from Stanford University’s School of Medicine uses the technique of calculating “in-game probability”; that is, incorporating a variety of continuously generated information.
In the case of a cancer patient, that involves integrating different types of predictive data – such as a tumour’s response to treatment and the amount of cancer DNA circulating in their blood during therapy – to generate a single, dynamic risk assessment.
“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,” says oncologist Ash Alizadeh.
The team has even found that CIRI can help doctors identify people who might benefit from early, more aggressive treatments rather than standard methods.
Their findings are published in a paper published in the journal Cell.
Alizadeh and colleagues began their study by looking at people previously diagnosed with diffuse large B-cell lymphoma (DLDCL), the most common blood cancer in the US.
They gathered data on more than 2500 DLBCL patients from 11 published studies and used this to train a computer algorithm to recognise patterns and combinations likely to affect whether a patient lived for at least 24 months after seemingly successful treatment without experiencing a recurrence of their disease.
They also included information from 132 patients for whom data about circulating tumour DNA levels were available prior to and after the first and second rounds of treatment.
When a DLBCL patient is diagnosed, clinicians assess the initial symptoms, the cell type from which the cancer originated and the size and location of the tumour after the first imaging scan to generate an initial prognosis.
More recently, they also have been able to assess the amount of tumour DNA circulating in a patient’s blood after the first one or two rounds of therapy to determine how the tumour is responding and estimate a patient’s overall risk of succumbing to their disease.
But each of these situations gives a risk based on a snapshot in time rather than aggregating all the data available to generate a single, dynamic risk assessment that can be updated throughout the course of a patient’s treatment.
The researchers wanted to work out whether it’s best to look at the latest information available about a patient, or the earliest information gathered, or an aggregate of all data over many time points.”
“Our standard methods of predicting prognoses in these patients are not that accurate,” says instructor of medicine David Kurtz. “Using standard baseline variables, it becomes almost a crystal ball exercise.
“If a perfectly accurate test has a score of one, 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 next plan to test CIRI’s capabilities with people recently diagnosed with aggressive lymphoma.
