Detection of Clonal T Cells in the Circulation of Patients With Nephrogenic Systemic Fibrosis

Background  Nephrogenic systemic fibrosis is a sclerodermalike disease in patients with acute or chronic renal insuffiency related to administration of gadolinium-containing contrast agents. Previous studies have demonstrated clonal T-cell populations in the blood of patients with systemic sclerosis, suggesting that these cells may be involved in the pathogenesis of the disease. Facing the clinical similarities of both diseases, we hypothesized that clonal expansion of T cells could be present in nephrogenic systemic fibrosis as well.

Observations  Findings from polymerase chain reaction and high-resolution capillary electrophoresis for T-cell receptor γ gene rearrangement analysis showed that all 6 prospectively evaluated patients (100%) with nephrogenic systemic fibrosis had detectable clonal T cells in their peripheral blood. In contrast, only 4 of the 15 control patients (27%) with chronic renal failure and none of the 12 healthy individuals analyzed in this study had evidence for T-cell clonality using the same type of examination. Clonal T-cell–positive patients with systemic sclerosis have previously been reported to better respond to extracorporeal photopheresis. However, this was not the case in 2 of our patients with nephrogenic systemic fibrosis.

Conclusion  As in systemic sclerosis, clonally expanded T-cell populations could play a critical role in the pathogenesis of nephrogenic systemic fibrosis, probably as an in vivo–activated inflammatory response to gadolinium exposure.

Nephrogenic systemic fibrosis (NSF), first reported by Cowper et al,¹ is a sclerodermalike fibrosing disorder that can occur in the setting of acute or chronic renal failure. There is accumulating evidence that NSF is caused by exposure to gadolinium-containing contrast agents.² Nevertheless, additional cofactors such as systemic inflammation or tissue and vascular injury might be necessary for the development of NSF at the time of gadolinium application.³ Clinically, NSF presents as firm papules and occasionally as nodules that coalesce into large indurated, sclerotic plaques. Skin lesions commonly develop on the lower extremities but also affect the trunk, upper extremities, and even internal organs (Figure 1). The clinical course of NSF is progressive in most cases, leading to considerable morbidity and mortality.⁴

An important differential diagnosis of NSF is systemic sclerosis (SSc). While both NSF and SSc at their late stage frequently result in widespread skin sclerosis, other characteristic features of SSc such as Raynaud phenomenon, sclerodactyly, digital pitting scars, and facial involvement (eg, masklike skin, telangiectasias, shrunken nose, radial furrowing around the mouth, and microcheilia) are absent in NSF. Besides their clinical similarities, both conditions are characterized by excessive collagen deposition and expression of transforming growth factor β (TGF-β). Some years ago, high proportions of expanded clonal T-cell populations were detected in the peripheral blood of patients with SSc, suggesting a pathogenetic role in the disease.^5,6 Interestingly, one study showed that patients with SSc with circulating clonal T cells had a better chance of response to treatment with extracorporeal photopheresis,⁵ which has recently been reported to have mild beneficial effects in NSF as well.⁷ The purpose of this pilot study was to search for clonally expanded T-cell populations in the peripheral blood of patients with NSF. Moreover, we thought to determine whether the presence of clonal T cells is associated with a response to photopheresis.

Peripheral blood mononuclear cells (PBMCs) of patients with NSF and control subjects (patients with chronic renal failure without NSF and healthy individuals) were obtained for analysis of T-cell receptor (TCR) γ gene rearrangement using polymerase chain reaction (PCR) and high-resolution capillary electrophoresis. Diagnosis of NSF was based on the following criteria: (1) presence of renal insufficiency at the time of disease onset, (2) exposure to gadolinium-containing contrast agents before the beginning of skin sclerosis, (3) characteristic histopathologic findings (CD34⁺ spindle cells, dermal histiocytes, accumulation of dermal mucin, and thickened collagen bundles), and (4) characteristic clinical presentation (progressive skin sclerosis with primarily affection of the distal extremities and trunk with sparing of the head and neck). As previously reported, the following 3 differential diagnoses were excluded: scleromyxedema (absence of paraproteinemia), SSc (absence of antinuclear antibodies, anticentromere antibodies, anti-Scl 70 antibodies, and Raynaud phenomenon), and sclerodermoid graft-vs-host disease (absence of allogeneic stem cell transplantation).⁷ The protocol of this study was approved by the ethics review board of the Ruhr-University Bochum, Bochum, Germany, and the study was conducted according to Declaration of Helsinki principles. Informed consent was obtained from every patient included.

The presence of a dominant T-cell clone in the peripheral blood was assessed by fragment length analysis of TCRγ PCR products, verified by high-resolution capillary electrophoresis. Genomic DNA was prepared from PBMCs and PCR was used to amplify specific regions of the TCRγ gene (Gen Bank AF159056) by using 2 primer mixes (InVivoScribe Technologies, San Diego, California) to cover all possible Vγ and Jγ combinations according to the BIOMED-2 Concerted Action.⁸ In detail, the first primer mix targets the Vγ1 to 8 + Vγ10 genes and Jγ1.1, Jγ1.3, Jγ2.1, and Jγ2.3 genes. The second mix contains primers that target the Vγ9 and Vγ11 genes and the same joining γ primers as in the first mix. All joining primers were fluorescence labeled. After PCR, the products were denatured, and fluorescent fragments were separated by capillary electrophoresis (CEQ 8000 Genetic Analysis System; Beckman-Coulter, Krefeld, Germany). To determine the sizes of DNA fragments, a size standard up to 420 nucleotides was used for every sample. Amplification resulted into 2 ranges for product sizes of nucleotides: Vγ1 to 8 + Vγ10 from 145 to 255 base pairs (bp) and for Vγ9 and Vγ11 in the range from 80 to 220 bp. The results from the electrophoretic profiles were calculated as follows: the presence of a dominant T-cell clone (1 or 2 predominant peaks with or without a polyclonal background) was interpreted as positive if peaks were oriented within the valid size range and were at least 3 times the amplitude of the third largest peak in a polyclonal background. The sensitivity of detecting a monoclonal population of T cells within a polyclonal background in the blood ranges from 1% to 5% using the previously described method.⁵

Statistical analysis was performed using the MedCalc statistical software (MedCalc Software; Mariakerke, Belgium). The Fisher exact test and the independent t test were used for data analysis. P < .05 was considered statistically significant.

A total of 6 consecutive patients with NSF (3 men and 3 women) treated at our institution in the year 2008 were included in this study. Moreover, 2 control groups consisting of 15 dialysis patients with chronic renal failure (without NSF) as well as 12 healthy individuals were analyzed for the presence of circulating clonal T-cell populations. The mean age of the patients with NSF, patients with chronic renal failure, and age and sex-matched healthy controls was 60.8 years (range, 48-79 years), 66.8 years (range, 28-86 years), and 56.3 years (range, 57-82 years), respectively. All relevant clinical characteristics of the 6 patients with NSF including underlying type of renal disease and comorbidities are depicted in Table 1. All of them had a history of 1 or more exposures to linear gadolinium-containing contrast agents for magnetic resonance imaging. Linear agents such as gadopentetate dimeglumine (Magnevist; Bayer HealthCare Pharmaceuticals Inc, Wayne, New Jersey) and gadodiamide (Omniscan; GE Healthcare, Waukesha, Wisconsin) are the most strongly associated contrasts with NSF. At the time of gadolinium exposure, all 6 patients received hemodialysis and had a glomerular filtration rate lower than 15 mL/min/1.73 m², consistent with stage 5 chronic kidney disease according to the classification of the National Kidney Foundation. The clinical course of 2 of these patients (patients 1 and 2) has been previously reported.⁹

All patients (n = 6 [100%]) with NSF had clonal T-cell populations in their peripheral blood, presenting as single dominant peaks in the high-resolution capillary electrophoresis. Interestingly, the size and localization of peaks was very similar in 4 of the patients (173, 178, 182, and 186 bp; Figure 2). In contrast, only 4 of the 15 control patients (27%) with chronic renal failure were positive for circulating clonal T cells (NSF vs renal controls; P = .003). None of the 12 healthy individuals had evidence for circulating clonal T cells (NSF vs healthy controls; P = .003). There was no significant difference of clonal T-cell positivity in both control groups (renal vs healthy control patients; P = .10). Complete medical workup in patients with NSF for concomitant autoimmune diseases revealed rheumatoid arthritis in 1 patient (patient 1). Moreover, 2 patients (patients 1 and 3) had a history of neoplastic disease (pituitary adenoma and colon carcinoma) without current evidence of active or metastatic disease. None of the patients with NSF had notable abnormalities in their differential blood, peripheral blood T- and B-cell subcounts, and blood smears, which was considered to exclude the presence of malignant neoplasms of the T-cell lineage such as T-cell leukemia or T-cell lymphoma. The results of flow cytometric immunophenotyping of the 6 patients with NSF are given in Table 2.

Two patients (patients 1 and 2) were treated with extracorporeal photopheresis, and photopheresis was initiated 11 and 16 months after their first diagnosis of NSF, respectively. Patient 1 was treated monthly for a total of 6 months, and patient 2 was treated bimonthly for a total of 15 months. Clinical examination demonstrated both a deterioration of skin sclerosis and joint contractures during therapy with photopheresis. Both patients refused other treatment options for NSF (eg, imatinib mesylate or kidney transplantation) and only received physical therapy after finishing photopheresis. Patients 3 and 4, who had several comorbidities, disclaimed NSF treatment right from the beginning and demanded physical therapy only. In contrast, 2 other patients (patients 5 and 6) experienced a marked decrease of skin induration following successful kidney transplantation.

Within the field of dermatology, T-cell clonality is usually found in cutaneous T-cell lymphoma. A recent large study showed that clonal T cells are detectable in 83.5% of mycosis fungoides and Sézary syndrome cases, but only in 2.3% of benign inflammatory skin disease cases.¹⁰ Nevertheless, dominant T-cell clones have been demonstrated in some benign acute and chronic skin diseases as well, eg, in pityriasis lichenoides et varioliformis acuta, pityriasis lichenoides chronica, or follicular mucinosis.^11-13 Several methods for the determination of clonal TCRγ gene rearrangements have been reported in the literature including denaturing or temperature gradient gel electrophoresis, polyacrylamide gel electrophoresis, heteroduplex analysis, and high-resolution capillary electrophoresis.¹⁴ The latter technique, which was used in this study, seems to have the highest qualitative and analytical sensitivity, though high instrumental costs surely limit its use in daily routine.¹⁰

There is accumulating evidence that in several T-cell mediated autoimmune diseases, autoreactive T cells can undergo clonal activation and expansion.¹⁵ For example, T-cell clonality was detected in the synovial fluid and blood of patients with rheumatoid arthritis and in a subgroup of patients with multiple sclerosis, respectively.^16,17 Moreover, pilot studies have shown the presence of clonal T cells in the skin and blood of patients with SSc, and a hitherto unknown antigen-driven process has been postulated.^18-20

Two studies on TCRγ gene rearrangement analysis in SSc published in the Archives some years ago added important additional informations.^5,6 French et al⁵ screened 13 patients with diffuse cutaneous SSc for TCRγ gene rearrangements and found that 6 (46%) of them had circulating clonal T cells. Marie et al⁶ evaluated 38 consecutive patients with SSc and found clonal T cells in 13 of them (34%). When comparing patients with diffuse and limited cutaneous SSc, those with limited disease exhibited significantly more clonal T-cell populations in their blood than those with diffuse cutaneous disease (12 of 28 patients [34%] vs 1 of 10 patients [10%], respectively). Interestingly, 4 of 6 clone-positive patients (67%) in the study by French et al⁵ experienced a clinically significant response to extracorporeal photopheresis, whereas only 1 of 5 clone-negative patients (20%) responded to this kind of treatment. In the present study, the 2 patients with NSF treated with photopheresis had no clinical response. However, it appears that treatments for NSF initiated after 1 year of disease progression have little chance of ever improving, probably owing to advanced contractures and deposition of excessive collagen.

The results described herein demonstrate for the first time to our knowledge the presence of clonal T-cell populations in the circulation of patients with NSF. A current pathogenetic model for NSF suggests that bone marrow–derived monocytes that display characteristics of T cells (CD45RO⁺/major histocompatibility complex class II⁺/CD34⁺) are recruited, activated, and proliferated from the circulation to the dermis.^3,21 Once these cells have arrived in the skin, they synthesize large amounts of collagen and TGF-β, which subsequently stimulates fibroblasts to produce elevated levels of glycosaminoglycans.^22,23 Enough evidence exists to conclude that the causing trigger of NSF is gadolinium, a substance widely used in contrast agents for magnetic resonance imaging. As gadolinium is eliminated almost entirely by the kidneys, its elimination half-life is considerably prolonged in patients with advanced renal impairment. Transmetalation and liberation of free toxic gadolinium ions from basically linear gadolinium chelate complexes seems to contribute to the initiation of tissue fibrosis. In line with this, gadolinium deposition has been detected in lesional skin of patients with NSF.^24,25 A recent in vivo study using a rat model (Wistar rats) for NSF demonstrated that exposure to gadolinium also results in an overactivation of several proinflammatory cytokines, progressing toward overt skin effects.²⁶ So far, the role of clonal T cells observed in our patients with NSF remains unclear. It is tempting to speculate that clonal T cells in NSF occur as an in vivo activated inflammatory response to gadolinium exposure. Moreover, the similar size and localization of clonal peaks in 4 of our 6 patients with NSF could indicate an expansion in response to the same stimulus (eg, gadolinium). Based on our findings, it remains unclear if the presence of clonal T cells affects the clinical course and prognosis of NSF, and if these cells disappear with successful treatment of NSF. Future investigations are now warranted to better elucidate the role of clonal T cells in the pathogenesis of NSF.

Interestingly, 4 of our 15 control patients (27%) with chronic renal failure showed a dominant T-cell clone as well. Previous studies have demonstrated T-cell subset restrictions in certain forms of renal disease, eg, IgA nephropathy or autoimmune glomerulonephritis.^27,28 Moreover, chronic hemodialysis treatment alone has been reported to contribute to expanded clonal TCR subsets.²⁹ Finally, TCR clonality also exists in a proportion of healthy adults, and its frequency increases with older age.^30,31 Although the pathogenetic role of clonal TCRs in our 4 control patients with chronic renal failure remains unknown, at least 1 of these factors could be causally involved. Nevertheless, clonal T-cell positivity was significantly higher in patients with NSF compared with the dialysis control patients.

Our results should be interpreted in light of the limitation of the study. Given the rarity of NSF, only a small number of patients (with several comorbidities in some of them) were included in this investigation, and analysis of the TCRγ gene rearrangement was only performed at a single time point. Although lymphocytic infiltrates are usually scant within the lesional skin of NSF, longitudinal studies including tissue specimens from different time points at different body sites are necessary to disclose if identical clonal T cells persist over time in an individual patient.

In conclusion, this study demonstrated clonal T-cell populations in the circulation of patients with NSF. Although it is likely that these cells play a role in NSF, the identity of the putative antigen or stimulus eliciting these responses remains unknown. Identification of such an antigen or stimulus could significantly improve our understanding of the pathogenesis of NSF and could help in the development of new treatment strategies for this disease.

Correspondence: Alexander Kreuter, MD, Department of Dermatology and Allergology, Ruhr-University Bochum, Gudrunstrasse 56, D-44791 Bochum, Germany (a.kreuter@derma.de).

Accepted for Publication: May 7, 2009.

Author Contributions: Dr Kreuter had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Kreuter and Schieren. Acquisition of data: Höxtermann, Tigges, and Hahn. Analysis and interpretation of data: Kreuter, Gambichler, and Schieren. Drafting of the manuscript: Kreuter and Schieren. Critical revision of the manuscript for important intellectual content: Höxtermann, Gambichler, Tigges, and Hahn. Statistical analysis: Gambichler. Administrative, technical, and material support: Höxtermann, Tigges, Hahn, and Schieren. Study supervision: Kreuter and Schieren.

Financial Disclosure: None reported.

Additional Contributions: Tanja Jaskowiak and Britta Redeker provided excellent technical assistance.