Alfredo Guillén-del Castillo

Good outcome of interstitial lung disease in patients with scleroderma associated to anti-PM/Scl antibody.

OBJECTIVE The objective of this article was to establish the clinical course of interstitial lung disease (ILD) in scleroderma related to the presence of anti-PM/Scl antibody compared with anti-Scl-70 in a Spanish cohort. Furthermore, no study has thoroughly investigated the outcome of pulmonary function test in the first group of patients. METHODS A total of 63 Spanish patients with scleroderma and ILD were selected in a retrospective observational study. Among them, 14 were positive for anti-PM/Scl antibodies and 49 for anti-Scl-70. Clinical assessments, including pulmonary function test, were collected. Variations equal or greater than 10% in forced vital capacity (FVC) were considered significant. Progression-free survival of disease was defined as the period of stable illness since pulmonary fibrosis diagnosis. RESULTS Anti-Scl-70 patients had a higher frequency of diffuse SSc subset, peripheral vasculopathy, and gastrointestinal involvement. Inflammatory myopathy was associated to anti-PM/Scl antibody. Anti-PM/Scl patients presented more improvement in FVC during follow-up, 30.8% compared to a 7.1% in Scl-70 group (P = 0.04), with less worsening of this parameter (15.4% vs 52.4% in Scl-70 patients, P = 0.01), and secondary less frequency of severe restrictive pattern (FVC < 50%) (7.7% compared to 42.9% in the other group, P = 0.02). Regarding treatment, more anticalcineurinics were used in anti-PM/Scl patients, while cyclophosphamide and mycophenolate were mainly used in anti-Scl-70 patients. The progression-free survival of disease was higher in anti-PM/Scl patients, with 76% at 10 years from diagnosis of ILD against a 29% in the Scl-70 group. CONCLUSIONS Several features and prognosis of ILD in SSc may be modified depending on the identified immunological profile.

Novel risk factors related to cancer in scleroderma

OBJECTIVE Emerging data have shown an increased risk of malignancy among patients diagnosed with systemic sclerosis (SSc) so identification of risk factors linking both disorders might have prognostic implications. The aim of this study was to assess the clinical and treatment-related risk factors for cancer in a single-center cohort of patients with SSc. METHODS Demographic, clinical, capillaroscopic, immunological and treatment-related data from 432 consecutive SSc patients were retrospectively analyzed. Variables that reached significant association in the univariate analysis were entered into a logistic regression in order to identify independent risk factors for cancer. RESULTS Malignancy was diagnosed in 53 patients (12.2%). Fifty-eight neoplasms were identified, among which breast (n=15), lung (n=10) and hematologic (n=9) malignancies were the most prevalent. In 19 patients the diagnosis of both scleroderma and tumour was made in <3years apart. Cancer significantly decreased the probability of survival (OR=2.61; 95%CI 1.46-4.69; p=0.001). No association with age, sex, smoking, cutaneous subset or RNA polymerase-III antibodies was found. However, risk of cancer was directly associated with the presence of anti-PM/Scl antibodies (OR=3.90; 95%CI 1.31-11.61; p=0.014), and inversely related to aspirin use (OR=0.33; 95%CI 0.12-0.90; p=0.031), which remained as independent risk factors for cancer on multivariate analysis. CONCLUSIONS PM/Scl antibodies seem to be associated with a higher risk of cancer in scleroderma. In contrast, the use of aspirin is related to a lower risk of cancer in our series. More studies are needed to ascertain the role of anti PM/Scl antibodies and aspirin in the development of malignancy among patients with SSc.

Digital ulcers and cutaneous subsets of systemic sclerosis: Clinical, immunological, nailfold capillaroscopy, and survival differences in the Spanish RESCLE Registry

OBJECTIVE Digital ulcers (DU) are the most common vascular complication of systemic sclerosis (SSc). We compared the characteristics between patients with prior or current DU with those never affected and evaluated whether a history of DU may be a predictor of vascular, organ involvement, and/or death in patients with SSc. METHODS Data from SSc patients with or without prior or current DU were collected by 19 referral centers in an ongoing registry of Spanish SSc patients, named Registro de ESCLErodermia (RESCLE). Demographics, organ involvement, autoimmunity features, nailfold capillary pattern, survival time, and causes of death were analyzed to identify DU related characteristics and survival of the entire series and according to the following cutaneous subsets-diffuse cutaneous SSc (dcSSc), limited cutaneous SSc (lcSSc), and SSc sine scleroderma (ssSSc). RESULTS Out of 1326, 552 patients enrolled in the RESCLE registry had prior or current DU, 88% were women, the mean age was 50 ± 16 years, and the mean disease duration from first SSc symptom was 7.6 ± 9.6 years. Many significant differences were observed in the univariate analysis between patients with and without prior/current DU. Multivariate analysis identified that history of prior/current DU in patients with SSc was independently associated to younger age at SSc diagnosis, diffuse cutaneous SSc, peripheral vascular manifestations such Raynauds phenomenon, telangiectasia, and acro-osteolysis but no other vascular features such as pulmonary arterial hypertension or scleroderma renal crisis. DU was also associated to calcinosis cutis, interstitial lung disease, as well as worse survival. Multivariate analysis performed in the cutaneous subsets showed that prior/current DU were independently associated: (1) in dcSSc, to younger age at SSc diagnosis, presence of telangiectasia and calcinosis and rarely a non-SSc pattern on nailfold capillaroscopy; (2) in lcSSc, to younger age at SSc diagnosis, presence of Raynauds phenomenon as well as calcinosis cutis, interstitial lung disease, and higher incidence of death from all causes; and (3) in ssSSc, to younger age at first SSc symptom and greater incidence of death from all causes. CONCLUSIONS Digital ulcers develop in patients with SSc younger at diagnosis, mainly in patients with dcSSc and lcSSc, and they are associated to other peripheral vascular manifestations such as Raynauds phenomenon, telangiectasia, and acro-osteolysis but also to calcinosis, and interstitial lung disease. History of DU in SSc leads to worse survival, also noticeable for lcSSc and ssSSc subsets but not for dcSSc patients.

Lung transplantation in systemic sclerosis: A single center cohort study

OBJECTIVE Lung transplantation (LT) has been proposed as a treatment for advanced interstitial lung disease (ILD) and/or pulmonary hypertension (PH) associated to systemic sclerosis (SSc) but few studies have been reported. The aim of this study was to describe the clinical features, complications and survival of a single-center cohort of patients with SSc that underwent LT and to compare their survival with a group of non-SSc transplanted patients. METHODS Fifteen patients with SSc were transplanted between May 2005 and April 2015. Standard international criteria were used to determine eligibility for LT. The severity of gastroesophageal involvement was not considered as a major contraindication if symptoms were under control. RESULTS Eight (53.3%) patients had diffuse cutaneous SSc. Eleven (73%) underwent bilateral LT. The main indication for LT was ILD, with or associated PH in 4 cases. Acute cellular rejection and infections were the most frequent complications. Functional lung tests tended to keep stable after transplantation. Median survival was 2.4 years (Q1-Q3: 0.7-3.7 years). We did not find differences in survival between patients transplanted with SSc versus those transplanted due to non-SSc ILD or PH. SSc complications were scarce with no patient developing PH after LT. CONCLUSIONS LT was an effective treatment for advanced ILD and/or PH associated to SSc in our study. Gastroesophageal reflux was not a limitation for LT in SSc in this study. Complications and survival did not differ from non-SSc patients undergoing LT.

Hepatobiliary involvement in systemic sclerosis and the cutaneous subsets. Characteristics and survival of patients from the spanish rescle registry

OBJECTIVE To assess the prevalence and causes of hepatobiliary involvement (HBI) in systemic sclerosis (SSc), to investigate the clinical characteristics and prognosis of SSc patients with HBI (SSc-HBI) and without HBI (SSc-non-HBI), and to compare both groups according to the cutaneous SSc subsets. METHODS In all, 1572 SSc patients were collected in the RESCLE registry up to January 2015, and all hepatobiliary disturbances were recorded. We investigated the HBI-related characteristics and survival from the entire SSc cohort and according to the following cutaneous subsets: diffuse cutaneous SSc (dcSSc), limited cutaneous SSc (lcSSc), and SSc sine scleroderma (ssSSc). RESULTS Out of 1572, 118 (7.5%) patients had HBI. Primary biliary cholangitis (PBC) was largely the main cause (n = 67, 4.3%), followed by autoimmune hepatitis (n = 19, 1.2%), and anti-mitochondrial negative PBC (n = 6, 0.4%). Other causes of HBI were as follows: secondary liver diseases (n = 11, 0.7%), SSc-related HBI (n = 7, 0.4%), nodular regenerative hyperplasia (n = 3, 0.2%), liver cirrhosis (n = 3, 0.2%), and HBI of unknown origin (n = 2, 0.1%). In multivariate analysis, HBI was independently associated to lesser risk of dcSSc (5.1% vs. 24.4%), and higher frequency of calcinosis (26% vs. 18%), left ventricular diastolic dysfunction (46% vs. 27%), sicca syndrome (51% vs. 29%), and anti-centromere antibodies (ACA, 73% vs. 44%). According to the cutaneous subsets, HBI was associated (1) in lcSSc, to longer time from SSc onset to diagnosis (10.8 ± 12.5 vs. 7.2 ± 9.3 years), sicca syndrome (54% vs. 33%), and ACA (80% vs. 56%); (2) in ssSSc, to sicca syndrome (44% vs. 19%), and (3) in dcSSc, no associations were found. HBI was the cause of death in 2.3% patients but the cumulative survival according to the presence or absence of HBI showed no differences. CONCLUSIONS HBI prevalence in SSc is 7.5% and dcSSc is the least involved subset. PBC is the main cause of HBI. Patients with SSc-HBI exhibited specific clinical and immunologic profile. Survival is similar for SSc patients with HBI.

Valor pronóstico del pH en el condensado de aire exhalado y de la fracción exhalada de óxido nítrico en la enfermedad pulmonar intersticial asociada a esclerosis sistémica

INTRODUCTION Interstitial lung disease (ILD) is one of the major causes of death in systemic sclerosis (SSc). This study investigated exhaled breath (EB) and exhaled breath condensate (EBC) biomarkers in patients with SSc and analyzed their role as a prognostic tool in SSc-related ILD. METHODS Fraction exhaled nitric oxide (FeNO) and exhaled carbon monoxide (eCO) measured in EB, together with pH, nitrite, nitrate and interleukin-6 levels measured in EBC were prospectively analyzed in 35 patients with SSc. Twelve patients had established ILD by chest high-resolution computed tomography (HRCT), and 23 patients showed no evidence of ILD. EB and EBC biomarkers were determined at inclusion, and pulmonary function tests were annually performed during 4 years of follow-up. RESULTS No differences at baseline biomarkers levels were found between groups. In all patients studied, low EBC pH levels were associated with a decreased diffusing capacity for carbon monoxide (DLCO) during follow-up. Low FeNO levels were correlated with lower forced vital capacity (FVC) at baseline, 4years of follow-up and with a decrease in FVC and DLCO during monitoring. Among ILD patients, high eCO levels were correlated with lower baseline FVC. In the global cohort, a worse progression-free survival was identified in patients with EBC pH values lower than 7.88 and FeNO levels lower than 10.75ppb (Log Rank P=.03 and P<.01, respectively). CONCLUSIONS EB and EBC could help to detect patients likely to present a deterioration on lung function during follow up.

Correction to: First clinical symptom as a prognostic factor in systemic sclerosis: results of a retrospective nationwide cohort study

When first published, this article inadvertently listed the RESCLE investigators individually within the author list. The names should instead have been listed within the Acknowledgements section only. The corrected author list and the updated Acknowledgements section are presented in this Correction.

Very early and early systemic sclerosis in the Spanish scleroderma Registry (RESCLE) cohort

OBJECTIVES According to the existence of subclinical organ involvement pre-scleroderma should be divided into two subsets: very early and early disease. Pre-scleroderma patients included in the Spanish Scleroderma Registry (RESCLE) Cohort were reclassified into subsets. Differences were evaluated and the risk of progression to definite systemic sclerosis was estimated. METHODS The characteristics of very early and early SSc patients were compared. A logistic regression model was used to determine the risk factors of progression. RESULTS 1632 patients were included, 36 (2.2%) in the very early subset and 111 (6.8%) in the early subset. There were no differences in sex, age at disease onset, duration of Raynauds phenomenon, antinuclear antibodies or capillaroscopic findings. Three (8.3%) very early SSc patients evolved to definite SSc, 2 (5.6%) of them meeting the ACR/EULAR 2013 criteria, unlike 31 (28%) early SSc patients, 20 (24%) of them meeting the criteria (p=0.034). Digestive involvement was an independent risk factor of progression (OR 17; 95% CI, 6.1-47.2). CONCLUSIONS The classification of early forms of scleroderma identifies patients with different prognostic risk of progression. The evolution to definite SSc is more frequent in early than in very early SSc patients. Digestive involvement is a risk factor of progression. An active assessment of organ damage in preclinical stages allows a correct classification and risk stratification, with implications for monitoring and treatment.

Understanding the Role of Antibodies as Markers of Cancer-Associated Systemic Sclerosis: Comment on the Article by Shah et al

To the Editor: I read with interest the article by Ben-Zvi et al on the effectiveness of anakinra in colchicine-resistant familial Mediterranean fever (FMF) (Ben-Zvi I, Kukuy O, Giat E, Pras E, Feld O, Kivity S, et al. Anakinra for colchicine-resistant familial Mediterranean fever: a randomized, double-blind, placebocontrolled trial. Arthritis Rheumatol 2017;69:854–62). Two important issues, however, should be noted concerning their study. First, although the authors acknowledge that the sample size used was too small, they neglect to note that the size makes their conclusions prone to both Type I error (claiming significance where there is none) and Type II error (claiming lack of significance where significance exists) (Halsey LG, CurranEverett D, Vowler SL, Drummund GB. The fickle P value generates irreproducible results. Nat Methods 2015;12:179–85). The conclusion that anakinra is useful only for arthritis in FMF (possibly a Type I error) and not for peritonitis or pleuritis (possibly Type II errors) is, I suggest, premature. The second issue is that the statistics related to the survival analysis based on the time to reach an end point of 4 attacks are inaccurate. In Figure 2 of their article, the authors compare 2 KaplanMeier curves, one representing the anakinra group, and the other representing the placebo group. The time unit that Ben-Zvi and colleagues use for this comparison is mean days, and they come up with a significant P value without indicating the test used. There are 2 problems here. First, the usual way to compare 2 Kaplan-Meier curves is by log rank test, and in the Methods section the authors state that a log rank test was indeed used for this analysis. However, it is difficult to see how this test could have been used since it is nonparametric and compares proportions of events per unit time and not the means of a time unit. The second problem is that the authors use days as the time unit in their analysis, a data assessment interval that they had not used during the study. It should be noted that choosing shorter time intervals, such as hours, for the same analysis would provide an even more statistically significant P value. Overall, I am concerned that the authors’ conclusions regarding both significant and nonsignificant results cannot be considered valid based on the statistical methods used.