Analysis of telomere length as predictive marker in psoriasis for comorbidities

The last decade, psoriasis has shifted from exclusively a skin disease to a rather systemic disease with associated comorbidities. Patients may present, next to the red scaly lesions, with obesity, non-alcoholic fatty liver disease, diabetes and cardiovascular disease (CVD) 1, 2. Moreover, psoriasis (PsO), an immune-mediated inflammatory disease (IMID), may coexist with other IMIDs such as psoriatic arthritis (PsA), Crohn's disease, rheumatoid arthritis and ankylosing spondylitis 3, 4. In addition, we showed that the risk for metabolic syndrome (MetS) is higher in patients with PsO compared with patients with PsA 5. Timely detection of these comorbidities is of crucial importance as they can greatly reduce the life expectancy of these patients. It has been postulated that dermatologists are in a unique position to screen PsO and PsA patients for these comorbidities. As blood is regularly drawn from patients during dermatological visits, ideally a blood-derived biomarker should be able to predict the patient's risk for these concomitant diseases, such as CVDs and MetS. Several studies have associated telomere(definition) shortening with CVD and MetS, suggesting that measuring the relative telomere length (RTL) can be used as a biomarker 6, 7. To elucidate the clinical value of RTL in the dermatological practice, we investigated the RTL in whole blood from PsO and PsA patients. The main purpose was to examine whether telomere length of PsO and PsA patients could be associated with MetS or CVD. Between January 2011 and May 2012, patients presenting with psoriasis or psoriatic arthritis were invited to enrol in the study and were recruited from various sites: the dermatology and rheumatology clinic of the Ghent University Hospital (n = 43 and 41, respectively), and patient advocacy groups (n = 39). Patients were found eligible if they were older than 18 years, diagnosed with PsA (according to the CASPAR classification criteria) by a rheumatologist or diagnosed with psoriasis by a dermatologist, and able to give informed consent. The study was approved by the ethics committee of the Ghent University Hospital. Using a standard data collection protocol, all patients were questioned and blood was drawn for analysis. Presence of MetS was based on the definition of International Diabetes Foundation in 2004. The prevalence of CVD was reported by patients through a questionnaire. DNA was purified from whole blood with DNeasy Blood and Tissue Kit (Qiagen, Belgium) according to the manufacturer's instructions. RTL analysis was performed as previously described 8. Data analysis was executed with SPSS 22.0 (SPSS Inc., Chicago, IL, USA). Chi-squared test and Mann–Whitney U-test were used to compare groups, and regression binary logistics were performed with the independent variable age to determine its influence on the outcome. A P-value <0.05 was considered statistically significant. After screening for eligible patients, 49 patients with PsO and 55 patients with PsA were enrolled. Table 1 lists the characteristics of the study population, which showed no significant difference in gender, age, disease duration and age of onset, and cardiovascular profile. The study population had previously been described where we found that the prevalence of MetS was significantly higher in the PsO group, supposedly due to significant difference in abdominal obesity between both groups and remained significant in a multivariate-adjusted model including age, disease severity, disease duration, systemic therapy or systemic inflammation (P = 0.002) 5. The prevalence of CVD was not significantly different between PsO and PsA patients. Still, we wondered whether the difference in MetS could be observed in telomere length attrition as well. RTL was calculated as the ratio of telomere repeat copy number (T), normalized against a single copy gene number (S). Table 2 lists the relevant RTLs in the population. First, RTL was compared between PsO and PsA. The RTL in the first group was 1.137641 (±0.183562) and in the second 1.106324 (±0.182521), but the difference did not reach statistical significance (P = 0.214). Similarly, comparison of RTL between patients with and without MetS did not yield a detectable difference (P = 0.997), which is remarkable as Révész and colleagues were able to associate the presence and severity of MetS with shorter telomere lengths 6. Next, we studied several risk factors that may affect telomere attrition. None of the variables listed in Table 2 were significantly linked to RTL. Smoking has been repeatedly shown to affect telomere shortening, yet it could not be linked to RTL in this study. Although the number of patients with CVD was low (n = 9), we found that RTLs were shorter in these patients, but this was due to the age and thus not statistically significant (data not shown). Telomere lengths add predictive power to chronological age for cardiovascular ageing and therefore have been proposed as a biomarker, regardless of its causality in the pathogenesis. Our preliminary results do not support RTL as biomarker to predict MetS or CVD in patients with psoriasis. However, our study has several limitations: (i) small population size, (ii) lack of healthy control group, (iii) low number of patients with CVD and a lack of non-psoriatic group with CVD, (iv) lack of telomerase activity (v) and lack of RTL data over longer time (birth, before disease onset, etc). Nevertheless, our findings do not confirm the hypothesis to use PCR-measured RTLs in patients affected by PsO or PsA to predict their risk for MetS. Due to the low number of affected patients in our cohort, we cannot draw any conclusions for CVD. It is possible that patients with psoriasis represent an ‘exclusive’ population in which the predictive power of RTL is lost as the psoriatic population is too sick. Such theory has been suggested for patients with RA 9. Perhaps composite markers represent a more elegant way to predict complicated diseases such as