SSc Subtypes Used to Create Predictive and Descriptive Models for SSc Patients

Predicting mortality and morbidity for scleroderma patients, either of the limited or diffusive cutaneous systemic sclerosis subtypes (lcSSc and dcSSc, respectively) is important to both clinicians and the affected individuals to help determine how patients may respond to certain types of treatments. To assign predictions to patients, researchers involved in the field must first find factors similar among patients to create subgroups within the disease.

One research team from McGill University and Jewish General Hospital in Montreal, led by Dr. Murray Baron and Dr. Hebah Alhajeri, used three different approaches to categorize patients. As described by “Comparison of Systemic Sclerosis Subsets As Predictors of Mortality and Morbidity,” the team used skin involvement, serological subsets, and ACR-EULAR 2013 classification criteria to segment patients. Patients were described as “morbid” if they had a forced vital capacity (FVC) less than 70% predicted, interstitial lung disease (ILD), pulmonary hypertension (PH), or another impaired health-related quality of life. Kaplan Meier survival curves for 805 patients placed in different subsets were also generated and analyzed for relevant trends.

The research team could predict mortality based on autoantibodies detected in serum samples and cluster category derived from ACR-EULAR scores, but not dcSSc vs. lcSSc (Leroy classification). Results for morbidity analysis were less clear. All categories predicted FVC <70% and development of ILD, but none could predict the time to developing PH. Time to an impaired health-related quality of life could be predicted by Leroy classification and detected serum autoantibodies.

In most cases, dcSSc showed worse outcomes than lcSSc. Accordingly, it would be even more useful for clinicians to be able to predict the risk of 5-year mortality for patients diagnosed early with dcSSc. A team from University of Pittsburgh developed such a tool that would be easy to use for such patients. Their model requires only knowledge of the patient’s history, a physical exam, and bloodwork.

“Development and External Validation of a Five-Year Mortality Risk Stratification Tool for Early Diffuse Systemic Sclerosis Patients” described the model’s development. The team used data from 760 adult early dcSSc patients examined between 1980 and 2009 to conduct a regression analysis. Their final model used age, gender, tendon friction rubs, gastrointestinal involvement, RNA polymerase III antibody, and anemia as 5-year mortality predictors.

Adding to this work, another team from University of Michigan in Ann Arbor developed a composite response index to be used during clinical trials treating dcSSc patients seeking to lower their risk of 5-year mortality. The work in “Development of a Composite Index for Clinical Trials in Early Diffuse Cutaneous Systemic Sclerosis–the Combined Response Index in Systemic Sclerosis” found that 16 out of 31 proposed variables could be used to profile patients. For example, if patients developed new renal crisis, new decline in FVC % predicted by 15%, and new onset left ventricular failure or PH, they would not be considered as having improved during the trial.

These studies illustrate how data from patients today and previous can be used to develop predictors and descriptors of disease status in patients diagnosed with SSc in the future. All abstracts were presented at the 2014 American College of Rheumatology Meeting held in November.