Pathologic Markers Determining Prognosis in Patients With Treated or Healing Giant Cell Arteritis.
Academic Article
Overview
abstract
PURPOSE: To provide quantitative evidence linking the CD68 (cluster of differentiation 68)+ macrophage marker found on temporal artery biopsies (TABs) with disease prognosis. DESIGN: Retrospective, cross-sectional study. METHODS: We examined 42 consecutive patients who had undergone unilateral TABs at a single hospital in 2015. Clinical data, laboratory data, and histopathologic features of TABs were recorded. Inclusion criteria were clinical diagnosis of giant cell arteritis (GCA) with TAB performed at the same center. CD68 immunohistochemistry was used to label macrophages in the TABs. Primary outcome was multiple logistic regression and bivariate comparisons to measure the association between CD68+ cells per histologic section with placement on immunomodulatory therapy (IMT). RESULTS: Twenty seven patients were female (64%), with a mean age of 72 (standard deviation [SD.] ±7.7). Eleven patients (26%) were placed on IMT, 17 (40%) had disease recurrence during steroid taper, and 25 (60%) were referred to rheumatology. Of 42 biopsies, 35 underwent staining with CD68 to confirm active inflammation in suspicious, but not diagnostic, specimens. Patients eventually placed on IMT had increased CD68+ cells per slice compared to those not on IMT (median 5.00 [25th-75th quartile 2.00-7.15] vs 1.21 [0.38-2.57], P = .031, respectively). A receiver operating characteristic (ROC) curve demonstrates that 2.17 CD68+ cells/slice predicts placement on IMT with an odds ratio of 1.54 (95% confidence interval 1.02-2.33, P = .038). CONCLUSIONS: Patients refractory to initial steroid tapers and those eventually placed on IMT had increased CD68 cells per section. CD68+ macrophages and their location on the internal elastic lamina may predict disease severity in patients with presumed GCA. Our results suggest that this marker may expedite patient triaging to alternate treatment to the usual steroid therapy.
