A new tool could better predict whether individual cancer patients will benefit from immune checkpoint inhibitors — a type of immunotherapy — using only routine blood tests and clinical data.
The artificial intelligence–based model, called SCORPIO, has been shown to be significantly better at predicting outcomes than the two current biomarkers approved by the U.S. Food and Drug Administration (FDA), according to a new study.
“Immune checkpoint inhibitors are a very powerful tool against cancer, but they don’t yet work for most patients,” said study co-senior author Luc Morris, a surgeon and research lab director at Memorial Sloan Kettering Cancer Center (MSK).
“These drugs are expensive, and they can come with serious side effects.”
So the key is patient selection — matching the drugs with patients who are most likely to benefit, Morris said.
“There are some existing tools that predict whether tumours will respond to these drugs, but they tend to rely on advanced genomic testing that is not widely available around the world,” he said.
“We wanted to develop a model that can help guide treatment decisions using widely available data, such as routine blood tests.”
Checkpoint inhibitors target the immune system rather than the cancer itself. These drugs work by taking the brakes off immune cells, allowing them to better fight cancer. MSK clinicians and scientists played a key role in bringing the new class of drugs to patients.
The new study was jointly overseen by Morris and Diego Chowell, an assistant professor of Immunology and Immunotherapy, Oncological Sciences.
Morris said: “There are currently two FDA-approved biomarkers for predicting response to checkpoint inhibitors: tumour mutational burden (the number of mutations in a tumour) and PD-L1 immunohistochemistry (evaluating the expression of the programmed death-ligand 1 protein in tumour samples).
“Both require samples of the tumour to be collected. Meanwhile, genomic testing to assess mutations is expensive and not available everywhere, and there is a lot of variability evaluating PD-L1 expression.
“Instead, our model relies on readily available clinical data, including routine blood tests performed in clinics around the world — the complete blood count and the comprehensive metabolic profile. And we found that our model outperforms the currently used tests in the clinic.
“The simplicity and affordability of this new approach could help ensure more equitable access to care while also reducing costs and helping ensure patients receive treatments most likely to benefit them individually — whether that ends up being a checkpoint inhibitor or some other type of therapy.
Morris explained that SCORPIO was initially developed by collecting data from MSK patients.
“Collaborating with the team at Mount Sinai, we used a type of artificial intelligence called ‘ensemble machine learning,’ which combines several tools to look for patterns in clinical data from blood tests and treatment outcomes,” he said.
“The model was developed using a rich resource of retrospective data from more than 2,000 patients from MSK who had been treated with checkpoint inhibitors, representing 17 different types of cancer. The model was then tested using data from 2,100 additional MSK patients to verify that it was able to predict outcomes with high accuracy.
“Next, we applied the model to nearly 4,500 patients treated with checkpoint inhibitors in 10 different Phase 3 clinical trials from around the world.
“In total, the study includes nearly 10,000 patients across 21 different cancer types — representing the largest dataset in cancer immunotherapy to date.
“We did this extensive testing and validation because our goal was not just to develop a predictive model, but to develop one that would be widely applicable to patients and physicians in different locations.
Morris said the team now plans to collaborate with hospitals and cancer centres around the world to test the model with additional data from a wider variety of clinical settings, and that work is now underway to develop an interface that is readily accessible by clinicians anywhere in the world.