Background/Purpose:

Our recent global blood transcript studies show that systemic sclerosis (SSc) patients have a prominent neutrophil signature.  Mechanistic studies with neutrophils are hampered by their short half-life in-vitro, and the fact that neutrophils are not the predominant leukocyte in mice, underscoring the importance of direct human data for their relevance.  Herein, we investigated whether higher neutrophil counts were associated/predictive of more severe disease and worse mortality in SSc.

Methods:

Neutrophil counts were prospectively obtained as part of complete blood count examination in the GENISOS cohort. All patients had a disease duration < 5 years. Mixed effect linear regression analysis was used to examine the relationship between neutrophil count and longitudinal FVC% / modified Rodnan Skin Score (mRSS) measurements.

Results:

At the time of analysis, 444 patients with SSc were enrolled from whom 392 had a baseline neutrophil count available (88.2%).  The mean disease duration (SD) was 2.4 (1.5) years, 59.6% of patients had diffuse SSc. For the longitudinal analysis of serially obtained neutrophil counts, the number of concomitantly obtained mRSS and FVCs were 1676 and 1030, respectively.

At the baseline visit, higher neutrophil count was associated with male gender (p<0.001), diffuse disease type (p<0.001), higher mRSS (p<0.001), lower FVC% (p=0.013), and anti-topoisomerase positivity (p=0.015).  There were no significant associations with age, disease duration, race, and RNA polymerase positivity.

Higher longitudinal neutrophil counts were associated with higher serially obtained mRSS and lower FVC%s (p<0.001 for both).

Next, the neutrophil count was dichotomized based on the top quartile value at the baseline study visit (6100 cells/ul).  Patients with a positive neutrophil count at the baseline visit had on average 6.4 points higher serially obtained mRSS  (p<0.001) and had on average 6.9% lower longitudinal FVC%s ( p=0.007). Moreover, positive neutrophil count was predictive of higher mortality in the univariable model (p<0.001 – Figure), as well as after adjustment for age, gender, topoisomerase status, diffuse disease type and prednisone use (Table).     Specifically, patients with a positive neutrophil count had 1.9 times higher mortality during the follow-up time.

Conclusion:

We show for the first time higher neutrophil count is predictive of more extensive skin/lung involvement and higher mortality in SSc.  Neutrophils might play a role in SSc pathogenesis and their baseline counts can serve as an easily obtainable prognostic biomarker for disease severity.