New Model Improves Prognostication for MDS and CMML Patients

Molecular data enhance traditional scoring systems

Myelodysplastic syndromes (MDS) describe a diverse group of hematopoietic disorders which can have distinctly different outcomes, and which may progress to acute myeloid leukemia (AML). About 100,000 people in the U.S. suffer from MDS, with a yearly incidence of 4.9 per 100,000.

Advertising Policy

Cleveland Clinic is a non-profit academic medical center. Advertising on our site helps support our mission. We do not endorse non-Cleveland Clinic products or services Policy

While prognostic models help physicians stratify patients by risk – and are used as default staging systems for these diseases – the models’ accuracy and clinical applicability have been questioned because of the wide variety of treatment sequences patients undergo. Of the four models now in use, most were developed in treatment-naïve patients with primary disease, which does not reflect the clinical reality.

Researchers propose including genetic mutation data

Cleveland Clinic researchers recently proposed a dynamic modification of one of the principal scoring systems. They presented their study in abstract to the 57th American Society of Hematology Annual Meeting & Exposition in Orlando, Fla., in December.

Referred to as IPSS-Rm, the proposed new system incorporates genetic mutational data into the existing revised International Prognostic Scoring System (IPSS-R). 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, the researchers say.

The new scoring system was developed and validated in a study of 508 patients with MDS who were treated at Cleveland Clinic between January 2000 and January 2012.

Older prognostic models applied mainly to newly diagnosed patients

“When we see MDS patients in clinic, we are often hamstrung by prognostic systems that really don’t apply to most of our patients most of the time,” says the study’s corresponding author, Mikkael A. Sekeres, MD, Director of the Leukemia Program at Cleveland Clinic’s Taussig Cancer Institute. “But with the IPSS-Rm, for the first time we can ‘stage’ patients receiving any therapy, and at any time point in their disease course.”

Advertising Policy

There are currently four prognostic models, each developed to risk-stratify patients with MDS:

• International Prognostic Scoring System (IPSS);
• Revised IPSS (IPSS-R);
• World Health Organization (WHO) classification-based Prognostic Scoring System (WPSS); and
• MD Anderson Prognostic Scoring System (MDAPSS).

Each system uses clinical variables as prognostic parameters, and all fall short in clinical practice for various reasons, Dr. Sekeres notes. All systems except MDAPSS, were developed in untreated patients with newly diagnosed primary MDS. The typical patient in the U.S. has received multiple types of treatment in different sequences throughout the course of his or her disease. Certain therapies may modify the disease and improve the patient’s overall survival, but this is not necessarily reflected in prognostic scores.

While there is a growing awareness of the role of somatic mutations in myeloid malignancies, study authors say that the best way to incorporate them into clinical prognostic tools has not yet been defined.

Proposed modification incorporates mutational data into predictive model

The study used tissue samples from patients with MDS and CMML who were treated at Taussig Cancer Institute. The samples were separated into two cohorts: a training cohort (333 patients) used to identify significant variables and mutations that were analyzed to build the new model, and a validation cohort (175 patients) in whom the newly designed model was subsequently applied (N=508). Another set of 53 patients who had samples available from different times in their disease courses was used to assess whether the model could be applied at different time points in the course of the disease.

Advertising Policy

DNA was sequenced and analyzed for mutations.

The median age of the 508 patients was 63 years; 64 percent had lower-risk MDS, 19 percent had intermediate-risk and 26 percent had higher-risk MDS. The median number of lines of treatment for the entire group was two (range, 0-7).

Researchers validate model’s efficacy throughout disease course

Study authors evaluated the prognostic ability of the IPSS-R scoring system in patients with de novo and secondary MDS and CMML who had received a median of two lines of therapy. The authors then incorporated the most significant common mutations into the IPSS-R scoring system to enhance its predictive ability. They applied this new model to a separate patient cohort whose clinical and molecular data were collected after the model was developed, and validated the model’s utility in this new population. They further validated the model in a population of patients in whom serial molecular data had been obtained. All of this was done to demonstrate the utility of the prognostic system in individual patients over the time course of their disease, Dr. Sekeres says.

“What makes this improvement in the IPSS-R special is that unlike many other prognostic systems, we developed it in a large population of patients, then validated in a separate large patient population, and then validated it again in patients for whom we had molecular data available at multiple times during their disease course, as they evolved toward acute myeloid leukemia,” Dr. Sekeres explains. “And we found that the IPSS-Rm worked well in all of these settings.”

More accurate prognostic data should mean more effective therapeutic interventions, and ultimately, enhanced overall survival for patients with MDS and CMML.