Dissertation committee:
Αθανάσιος Πρωτογέρου, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Αθανάσιος Τζιούφας, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Παναγιώτης Βλαχογιαννόπουλος, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Θεόδωρος Παπαϊωάννου, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Αναστάσιος Κόλλιας, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Κλειώ Μαυραγάνη, Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Ευσταθία Καψογεώργου, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Summary:
Systemic vasculitides constitute an heterogenous group of rare, chronic and often recurrent systemic autoimmune/autoinflammatory diseases, characterized by multi-level heterogeneity in terms of clinical phenotype, histologic patterns, pathogenetic mechanisms and treatment selection strategies. The vascular damage implicates various disease-specific mechanisms including natural and specific immunity, immune complex formation and presence ANCAs. The acceleration of three classical types of arterial damage (inappropriate arterial remodeling, atheromatosis and arteriosclerosis), affecting both the micro- and macro-circulation, has also been proposed as an additional mechanism of vascular injury at least in some PSV types. This is attributed to the interplay between tissue and systemic inflammation, immunosuppressive therapy, and common CVD risk factors and may contribute to the increased CVD morbidity and mortality of PSV patients. The identification of the variable clinical phototypes even within the same type of SV, the identification of clinically relevant biomarkers, concerning both the inflammatory process and comorbidities as well as the molecular stratification of patients towards precise targeted treatments are still unmet needs.
The present study aims to:
I. Explore the presence and potential reversibility of subclinical vascular dysfunction and/or damage both in the micro- and macro-circulation in SV by evaluating four main vascular pathologies (atheromatosis, arterial stiffening, arterial remodeling/hypertrophy and pressure wave reflection impairment) in four different vascular beds (carotid, aortic, femoral and retinal) using gold-standard in clinical practice, non-invasive vascular biomarkers. The monitoring of subclinical vascular damage might provide insights on the development of these vascular pathologies in SV as well as guide the management of these patients, in similar ways as in individuals with CVD risk factors.
II. To identify and investigate the specificity of already existing biomarkers such as pANCAs as well as the identification of novel biomarkers within the metabolic profile of patients related to disease activity.
III. To describe the clinical spectrum of cryoglobulinemic vasculitis in primary Sjögren’s syndrome, identify the biomarkers associated with future lymphoma development and depict the differences with hepatitis C virus related CV.
Patients and Methods
A. To evaluate the micro- and macro-circulation in SV, 73 PSV patients, matched at 1:1 according to age/sex/CVD risk factors with non-inflammatory controls-(NIC) and rheumatoid arthritis-(RA) controls were studied. In the case of GCA, patients with PMR without underlying vasculitis served as a second disease control group. Atheromatosis-(carotid/femoral plaques), arterial stiffening-(cfPWV), pressure wave reflections-(AIx and AIx75) and arterial remodeling-(cIMT) and retinal vessel calibers (CRAE, CRVE) were evaluated in both active and inactive disease state.
B. To explore the metabolic profile in sequential sera of GCA and PMR patients 110 serum samples from 50 patients (33-GCA and 17 PMR) at 3 time points, 0-(V1: active disease), 1 and 6 months-(V2 and V3: remission) of treatment with glucocorticosteroids, were subjected to NMR-based metabolomic analysis. Multi- and univariate statistical analyses were utilized to unveil metabolome alterations following treatment. Moreover, we investigated the occurrence of ANCA-related antigenic specificities in 82 P-ANCA-positive sera by multiplex ELISA, as well as their association with other autoantibodies. The P-ANCA-positive sera corresponded to patients with SV (n=24), SLE (n=28), APS (n=5), SS (n=7), RA (n=3), SSCL (n=1), sarcoidosis (n=1) and Hashimoto thyroiditis (n= 13).
C. From 1083 pSS patients we identified 71 with cryoglobulinemic vasculitis. pSS-CV patients were matched with pSS patients without cryoglobulins (1:2) and HCV-CV patients (1:1). Clinical, laboratory and outcome features were analyzed. A data driven logistic regression model was applied for pSS-CV patients and their pSS cryoglobulin negative controls to identify independent features associated with lymphoma.
Results
I. Aortic PWV in PSV was higher by 0.7 m/sec compared to NIC, and by 1.3 m/sec to RA-controls-(p=0.003) and was more pronounced in LVV/MVV (p=0.08 and p=0.001 respectively). AIx was decreased in all PSV (p=0.03) and predominantly in SVV compared to NIC-(p=0.04) and RA-controls (p=0.09 all, p=0.07 active disease). Atherosclerotic plaque formation prevailed in all vascular beds at diagnosis being more enhanced in LVV/MVV (p=0.007, p=0.03, p=0.004), compared to NIC and only both carotid/femoral-(p=0.02) to RA-controls. Carotid IMT was higher in LVV/MVV irrespectively to disease state and control group and was the most sensitive to change biomarker between activity-inactivity. Active GCA was associated with increased PWV/plaques/cIMT compared to matched-NIC, RA and PMR-controls. Venular and arteriolar retinal dilatation was observed in all active disease groups, while inactivity was associated with irreversible dilatation of the arterioles compared to NIC-(p=0.029) and RA-controls-(p=0.008). In brief we found that non-invasive vascular biomarkers allow early identification of vascular pathology at disease diagnosis suggesting disease specific rather than treatment related CVD association, suggesting that early management of CVD risk factors as proposed by biomarkers assessing subclinical atheromatosis and arteriosclerosis could prevent long term CVD events in PSV patients.
II. Distinct metabolic profiles were identified between activity and remission, independently to disease type. N-acetylglycoproteins and cholines of bound phospholipids, emerged as predictive markers of disease activity. Altered levels of 4 out of the 21 small molecules were also observed, including increased levels of phenylalanine, and decreased of glutamine, alanine, and creatinine in active disease. Metabolic fingerprinting discriminated GCA from PMR in remission. GCA and PMR patients exhibited characteristic lipid alterations as a response and/or adverse effect of GCs treatment. Correlation analysis showed that several identified biomarkers were further associated with acute phase reactants, C-Reactive Protein and Erythrocyte Sedimentation Rate. The NMR profile of serum metabolome could identify and propose sensitive biomarkers of inflammation. Metabolome alterations, following GCs treatment, could provide predictors for future steroid-induced side effects.
III. In most P-ANCA-positive patients studied (51/82, 62.3%), these autoantibodies occurred in high titers (>1:160). The analysis of P-ANCA-positive sera revealed reactivity to MPO in only 50% of patients with vasculitides, whereas it was infrequent in the other disease groups studied. Reactivity to other P-ANCA-related autoantigens was also rarely detected. Our findings support that high P-ANCA titers occur in SARD. The P-ANCA-positive staining pattern is associated with MPO specificity in vasculitides, while in other autoimmune diseases, it mostly involves unknown autoantigens.
IV. pSS-CV patients had higher frequency of extraglandular manifestations and lymphoma (OR=9.87, 95% CI: 4.7-20.9) compared to pSS patients without cryoglobulins. Purpura was the commonest vasculitic manifestation (90%), presenting at disease onset in 39% of patients. One third of pSS-CV patients developed B-cell lymphoma within the first 5 years of CV course, with cryoglobulinemia being the strongest independent lymphoma associated feature. Compared to HCV-CV patients, pSS-CV individuals displayed more frequently lymphadenopathy, type II IgMk cryoglobulins and lymphoma (OR = 6.12, 95% CI: 2.7-14.4) and less frequently C4 hypocomplementemia and peripheral neuropathy. In conclusion, pSS-CV has a severe clinical course, overshadowing the typical clinical manifestations of pSS and higher risk for early lymphoma development compared to HCV related CV. Though infrequent, pSS-CV constitutes a distinct severe clinical phenotype of pSS.
Future studies with long term follow-up and stratification according to immunosuppressive treatment will facilitate our better understanding of the link between vascular pathology and inflammation, yet providing new diagnostic, prevention and response treatment biomarkers.
