Specific biomarkers of angiogenesis may be clinically useful for risk stratification in systemic sclerosis (SSc), according to research results published in Arthritis and Rheumatology.
Researchers conducted a prospective cohort study with data from patients with prevalent SSc (n=300) to examine the clinical usefulness of vascular biomarker measurements over time as laboratory tests to guide clinical decision making.
Patients included in the study had a diagnosis of prevalent SSc with no evidence of pulmonary hypertension or active ischemic digital lesions. Follow-up included clinical evaluations for signs and symptoms of pulmonary hypertension and digital lesion development; plasma samples were collected at 6- to 12-month intervals.
A majority of the study cohort included white women (mean age, 52±12 years), 27% of whom were centromere positive, 23% RNA polymerase 3 positive, and 23% topoisomerase positive. A total of 58% of patients had limited cutaneous disease. In terms of treatment, 47%, 29%, 26%, and 18% of patients were receiving calcium channel blockers, angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers, aspirin, and statins, respectively. The most prevalent comorbidities were hypertension, coronary artery disease, diabetes, cerebrovascular disease, and peripheral artery disease.
Researchers indicated that 15% of patients (n=46) were diagnosed with pulmonary hypertension and 23% (n=69) developed digital lesions. Average time from enrollment to development of pulmonary hypertension and digital lesions was 3.0±2.4 years and 3.5±2.4 years, respectively.
Researchers examined the relationship between baseline biomarker levels and demographic and disease characteristics, and found no substantial correlation between any biomarker and disease duration, age at enrollment, comorbidities, or age at diagnosis. A mild positive correlation was noted between platelet-derived growth factor (PIGF) and soluble fms-like tyrosine kinase 1 and PIGF and hepatocyte growth factor. Patients with RNA polymerase 3 autoantibodies had significantly higher PIGF at baseline, while those who tested positive for centromere or topoisomerase had significantly lower PIGF values. PIGF was also significantly elevated in patients with diffuse cutaneous skin.
Investigators analyzed 5 biomarkers in time-to-event analyses for pulmonary hypertension and digital lesion outcomes. Median time from the first biomarker measurement to the development of pulmonary hypertension was 3.5 years (interquartile range, 1.6-4.5 years); 3 biomarkers measured at study entry (hepatocyte growth factor, soluble Fit-1 [sFIt1], and PIGF) were significantly associated with the development of pulmonary hypertension. None of the biomarkers collected at baseline were significantly associated with digital lesion development, but biomarkers endoglin, sFIt1, and PIGF collected 2.8±2.5 years before digital lesion development (timepoint 2) were significantly associated with distal lesion development.
Because of the significant associations between baseline PIGF and sFIt1 measurements in patients with pulmonary hypertension, the researchers also sought to evaluate the trend of these individual biomarkers over time. They found that patients who developed pulmonary hypertension had persistently high PIGF and sFIt1 levels throughout follow-up. Compared with patients who did not develop digital lesions, those who developed digital lesions did not have significantly different levels of sFIt1, but had persistently higher levels of PIGF.
Finally, the investigators identified the optimal cutpoint for both PIGF and sFIt1 biomarkers as predictors of pulmonary hypertension. For PIGF, the cutpoint was 9.89 pg/mL, with 82% sensitivity, 56% specificity, and an area under the curve of 0.69. For sFIt1, the optimal cutpoint was 93.8 pg/mL, with 71% specificity, 51% sensitivity, and an area under the curve of 0.61.
Study limitations included lack of a strict referral protocol for patients and exclusion criteria that could have identified and eliminated some patients with pulmonary hypertension from the study, the use of a prevalence cohort, and the potential for missed digital lesion events in between clinic visits.
“These data suggest that molecules involved in angiogenesis reflect vascular perturbation,” the researchers concluded, “and [that] elevations at first encounter may be used [for risk stratification of] patients and aid in clinical phenotyping.”
Mecoli CA, Shah AA, Boin F, Wigley FM, Hummers LK. The utility of plasma vascular biomarkers in systemic sclerosis: A prospective longitudinal analysis [published online March 21, 2020]. Arthritis Rheumatol. doi:10.1002/art.41265