Background/Purpose: ROCKs are implicated in the pathogenesis of many vascular diseases.  ROCK activation is associated with Th17 differentiation and production of Th17-associated cytokines, IL-17 and IL-21.  Th17 cells and related cytokines are present in increased levels in the vascular inflammatory infiltrate in active GCA.  ROCK activity in GCA is unknown.   The aim of this study was to assess ROCK activity in temporal artery biopsy (TAB) specimens in GCA versus controls.  

Methods: All TAB performed at a tertiary care center over a 5 year period were identified.  Charts were reviewed for clinical information; included subjects had TAB specimens and clinical information available.  Subjects were categorized according to 1990 ACR criteria and TAB status into 3 groups: GCA with positive TAB, GCA with negative TAB and age-sex-matched controls. 

Paraffin-embedded temporal artery specimens were stained for Phospho-Ezrin/Radixin/Moesin (pERM), a surrogate of ROCK activity, using immunohistochemical (IHC) stain.  IHC stained slides were reviewed by a pathologist blinded to clinical status.  Three separate areas (endothelium, adventitia and vaso vasorum) were scored for intensity of staining on a scale of 0-2 for total possible composite score of 6. Primary outcome was biopsy pERM intensity score in subjects with GCA compared to controls.      

Results: Nineteen subjects with GCA had positive TAB, 17 subjects had GCA with negative TAB and18 age-sex-matched controls were analyzed.  Mean age was 77.9±9.1years and 81.4% were female (Table1).  Biopsy pERM intensity scores ranged from 2-6 with 28.3% scores £4 and 72.2% of scores >4 (high).   79.4% of high scores occurred in GCA subjects compared to controls (p=0.0033).  Compared to controls, GCA subjects with either positive or negative TAB showed statistically significantly higher pERM intensity scores (p=0.035 and p=0.0083 respectively).  Adjusting for comorbid diabetes, hypertension, prednisone use, and statin use, GCA subjects still had significantly higher pERM intensity scores compared to controls (OR 7.3; 95%CI 1.9-25.9, p=0.0046). High score for diagnosis of GCA had sensitivity of 86%, specificity of 55.6%.  Comparing GCA with negative TAB to controls, high score had sensitivity of 90.4% and negative predictive value of 90.9%. 

Conclusion: Subjects with GCA had more intense pERM staining in TAB specimens compared to age and sex-matched controls regardless of whether TAB was positive or negative, independent of prednisone dose, statin use, or vascular comorbidities. In this population, high pERM staining score had a sensitivity comparable to what is reported as sensitivities of TAB routine histopathology itself and suggests it may be a useful adjunctive diagnostic tool especially in patients with suspected GCA and negative TAB.  The ROCK pathway warrants further investigation in GCA as inhibition of this pathway is a potential therapeutic target.  

Table 1