Two studies presented at the 2015 ASH Annual Meeting highlighted efforts by researchers to move beyond current risk stratification models for patients with myelodysplastic syndromes (MDS), toward revised models that include somatic mutations.
Incorporating mutations creates risk models that more closely reflect real-world patients with MDS than do currently available models. Existing prognostic models, such as the Revised International Prognostic Scoring System (IPSS-R), were created using treatment-naïve patients with primary disease who do not receive therapy subsequently and, therefore, do not reflect typical MDS patients.
Mutations Predict Overall Survival Independent of IPSS-R
In the first study, Rafael Bejar, MD, PhD, and colleagues from the Moores Cancer Center at the University of California, San Diego, identified mutations in MDS-associated genes and clinical outcomes to create a prognostic model that could provide a deeper level of risk stratification, regardless of MDS subtype or prior therapies.1
While some mutations have been recognized as important for predicting survival, they have not been widely accepted among the medical community, the authors noted. “There still isn’t consensus on how best to use this information, or even how best to test for it,” Dr. Bejar told ASH Clinical News. “Our group is trying to amass data from around the globe to see if we can identify a best understanding of how these mutations affect prognosis and how we might come to a consensus on how to use them in the future.”
Dr. Bejar and colleagues pooled data from 3,392 patients with MDS across multiple centers to determine the relationship between mutations in MDS-associated genes and clinical measures, such as overall survival (OS).
“We have been able to identify many genes that have independent significance of the IPPS-R and to estimate what that independent risk is,” said Dr. Bejar.
With a median follow-up of 3.7 years, median survival among the patient cohort was 2.88 years. Mutations in 12 different genes were associated with reduced OS, including: ASXL1, CBL, EZH2, IDH2, NF1, NRAS, PTPN11, RUNX1, SRSF2, STAG2, TP53, and U2AF1. Mutations in one gene, SF3B1, were found to prolong survival.
“The large size of the cohort allowed for more precise estimates of survival in less frequently mutated genes,” Dr. Bejar and colleagues explained. “For example, mutations of IDH2 (present in 3.4% of cases; n=103) were associated with shorter OS (HR=1.61; 95% CI 1.26-2.05; p=0.0001) whereas IDH1 mutations (present in 2.4% of cases; n=77) were only marginal (HR=1.29; 95% CI 0.97-1.72; p=0.082), demonstrating the distinct impact of mutations in these highly related genes.”
The following mutated genes had independent prognostic significance:
- TP53 (hazard ratio [HR] = 2.37; 95% CI 1.94-2.90)
- CBL (HR=1.57; 95% CI 1.22-2.03)
- EZH2 (HR=1.55; 95% CI 1.22-2.03)
- RUNX1 (HR=1.50; 95% CI 1.24-1.83)
Notably, the 58 percent of patients without mutations in any of the “adverse” genes listed above had a longer median survival than those with the adverse mutations (4.8 years vs. 1.6 years respectively; p<0.0001), even after correction for IPSS-R risk groups.
In the future, the authors wrote, a multivariable analysis of this dataset will examine the combined contribution of mutated genes to prognosis. “Mutations in several genes retain their prognostic significance after adjustment for IPSS-R risk groups, indicating that these select abnormalities could refine the prediction of prognosis when incorporated into a clinical scoring system such as the IPSS-RM,” Dr. Bejar and co-authors concluded. “The results of this analysis will serve as the template with which to build an integrated molecular risk model for MDS.”
The Revised “Molecular” International Prognostic Scoring System
In another study presented at the annual meeting, Aziz Nazha, MD, from the Leukemia Program at Cleveland Clinic’s Taussig Cancer Institute, reported results of a newly developed prognostic model. This model, referred to as the IPSS-Rm, incorporates genetic mutational data into the IPSS-R and was created using clinical and mutational data gathered from patients with MDS, most of whom were treated with a variety of therapies, from 2000 to 2012.2
The additional data enhance the predictive ability of the scoring system at any time in the course of the disease and without regard to initial or subsequent treatments, Dr. Nazha and colleagues explained.
The model was created and evaluated in 508 patients with de novo and secondary MDS or chronic myelomonocytic leukemia:
- A training cohort, consisting of 333 patients whose data would be used to help build the model
- A validation cohort, consisting of 175 patients who would be used to validate the model that was created
The median age among all patients was 63 years; 64 percent had lower-risk MDS, 19 percent had intermediate-risk MDS, and 26 percent had higher-risk MDS. The median number of lines of treatment for the entire group was two (range = 0-7).
Within the training cohort, the researchers identified 24 gene mutations that were present in at least 10 patients and then used a Cox proportional hazard analysis to identify independent clinical and molecular prognostic factors for OS. These included age, IPSS-R, and the genes EZH2, SF3B1, and TP53.
Using these factors, the investigators developed a weighted algorithm to identify four prognostic groups for OS (p<0.0001), which were then substantiated in the validation cohort: low (median OS=37.4 months), intermediate-1 (23.2 months), intermediate-2 (19.9 months), and high (12.2 months).
To further validate whether the new model could be applied at any time during disease course, Dr. Nazha and colleagues sequenced paired samples from another set of 53 MDS patients at different timepoints in their disease (diagnosis, after treatment failure, and at the time of progression to AML), and demonstrated that the four prognostic groups remained distinct across sample collection times.
“This model is dynamic and valid at varying time points of a patient’s disease course,” Dr. Nazha and researchers concluded. Overall, “the modified IPSS-R scoring system incorporates mutational data and enhances its predictive ability in patients with MDS, regardless of initial or subsequent treatments.”
- Bejar R, Papaemmanuil E, Haferlach T, et al. Somatic mutations in MDS patients are associated with clinical features and predict prognosis independent of the IPSS-R: Analysis of combined datasets from the International Working Group for prognosis MDS-molecular committee. Abstract #907. Presented at the American Society of Hematology Annual Meeting, December 7, 2015; Orlando, FL.
- Nazha A, Narkhede MS, Radiovoyevitch T, et al. The revised International Prognostic Scoring System “Molecular” (IPSS-Rm), a validated and dynamic model in treated patients with myelodysplastic syndromes (MDS). Abstract #607. Presented at the American Society of Hematology Annual Meeting, December 7, 2015; Orlando, FL.