MSK molecular geneticist Elli Papaemmanouil
A new molecular scoring system promises to improve the treatment of a set of blood cancers called myelodysplastic syndromes (MDS), according to an international team of researchers. The findings were announced June 12, 2022 at the European Hematology Association conference in Vienna and published simultaneously in the journal NEJM Evidence.
For more than two decades, doctors have relied on blood counts, cell appearance, and chromosome changes to assess an MDS patient’s risk and the likelihood of progression to an acute leukemia, which happens in up to 30% of cases.
This risk stratification is critical in guiding treatment decisions. For those with low-risk disease, supportive care with less toxic treatments is usually recommended, while patients with high-risk disease need higher intensity treatments to control disease severity and prevent progression to leukemia.
But with the increasing availability of genetic information on MDS, researchers are looking to incorporate this additional information into their risk assessments to better characterize a person’s risk and therefore improve the tailoring of treatment.
“Over the past decade, the scientific community has uncovered the genes that are mutated in MDS, promising a new era for biomarker development and patient-tailored medicine,” says molecular geneticist Elli Papaemmanouil, from Memorial Sloan Kettering Cancer Center, who led the international effort. “However, we are still learning how to use this information to guide diagnosis, plan optimal treatment, and improve patient outcomes.”
Toward that end, Dr. Papaemmanouil and an international team of researchers came together to pool resources, data, and expertise to develop a new scoring system that incorporates molecular information.
They collected a total of 2,957 patient samples from 24 centers from 13 countries. Those samples were then sequenced by the Integrated Genomics Operation at MSK and analyzed in the Papaemmanouil laboratory. Another 754 samples derived from a Japanese MDS consortium were used to confirm the validity of their initial findings.
The result of this international effort is the International Prognostic Scoring System-Molecular (IPSS-M), which incorporates information from 31 gene mutations into the existing framework to deliver a patient-specific risk score.
IPSS-M is a continuous scale, representing the continuum of risk observed across patients with MDS. IPSS-M also produces six risk categories (from very low to very high) as a summary risk metric.
Compared to the previous scoring system, 46% of patients (or roughly 1 in 2) were assigned to a different risk category with IPSS-M. This means that the treatment recommendations for these patients would likely have changed as well.
The researchers say that IPSS-M is set to become the new international standard for risk stratification of MDS patients at diagnosis.
“Leading this project has been a true honor,” Dr. Papaemmanouil says. “Today we deliver one of the first formal clinical algorithms for risk stratification in cancer that produces a patient-tailored risk score.”
Understanding which genes are important in MDS risk prognostication is also important for the design of diagnostic assays. This study highlights a minimum set of 31 genes, and guidelines for how to report these mutations during a patient’s diagnostic work up.
To support wide adoption of the scoring system, the team has developed a web-based calculator, which clinicians can use to enter patients’ data and then receive a personalized prediction of risk, outcomes, and likelihood of leukemia transformation in order to guide treatment decisions. The web calculator will be hosted by the MDS Foundation, which provided funding for the study.
“It has been a privilege to work with expert MDS clinicians, pathologists, and biologists to develop the IPSS-M, aiming to better guide and ultimately improve the care of patients with MDS,” says Elsa Bernard, a postdoctoral researcher in Dr. Papaemmanouil’s group at MSK and the paper’s first author.
She points out that IPSS-M is also available as an R package, which the scientists hope will be useful to the broader community.
