Extracorporeal photochemotherapy has been proven effective in selected T-cell mediated diseases.
Extracorporeal photochemotherapy has been proven effective in selected T-cell mediated diseases.
To evaluate the safety and efficacy of extracorporeal photochemotherapy in patients with steroid-dependent Crohn’s disease by an open, monocentric trial in three phases of 24 weeks each.
In phase 1 standardized steroid tapering was initiated in patients with a history of steroid-dependent Crohn’s disease. Those with a prospectively evaluated maintenance dose of at least 10 mg/day prednisolone continued steroid-withdrawal under the application of extracorporeal photochemotherapy in phase 2. The duration of remission or response was followed during phase 3. Colonic tissue bioptically obtained before and after extracorporeal photochemotherapy was studied by immunofluorescence microscopy for the presence of photoadduct positive cells.
Out of 24 patients included in phase 1, 10 entered phase 2 for extracorporeal photochemotherapy. Four subjects achieved remission and four others response. Significant reductions in serum C-reactive protein levels and intestinal permeability were measured, as well as increases in quality of life and plasma adrenocorticotropic hormone levels. No major side-effects were observed. Remission remained stable in three out of four patients during phase 3. In three patients, positive nuclear stainings of photoadducts were detected in colonic mononuclear cells after extracorporeal photochemotherapy.
Extracorporeal photochemotherapy represents a safe steroid-sparing approach in patients with Crohn’s disease and is associated with intestinal homing of photopheresed cells.
Crohn’s disease is a relapsing, inflammatory disorder potentially affecting the entire gastrointestinal tract in a focal pattern. Present theories suggest that pathogenic or resident luminal bacteria stimulate responses of mucosal CD4+ T helper 1-type (Th1) lymphocytes, leading to a chronic inflammation on the basis of genetically determined host susceptibility factors.1, 2 Because current therapeutic approaches for Crohn’s disease are not curative, conservative management of patients is aimed at maintaining remission with only minimal drug-related side-effects.
Glucocorticosteroids represent the most effective therapy for active Crohn’s disease, but steroid dependency or resistance manifests in approximately 50% of patients at attempts to taper medication, which can lead to adrenal suppression and long-term toxicity.3–5 The immunomodulatory agents azathioprine, 6-mercaptopurine or methotrexate succeed in reducing the requirements for glucocorticosteroids in patients with steroid-dependent Crohn’s disease. However, the toxicity of these medications is of concern.6–8 The long-term risk profile of the monoclonal antibody cA2 against tumour necrosis factor α, which has recently emerged as a therapeutic alternative in patients with active Crohn’s disease resistant to standard treatment, still needs to be fully evaluated.9 In this context, the introduction and evaluation of experimental treatment modalities remains a major issue in steroid-dependent Crohn’s disease.
Extracorporeal photochemotherapy is an apheresis-based procedure comprising the successive ex vivo exposure of autologous peripheral blood mononuclear cells to 8-methoxypsoralen and long-wavelength ultraviolet A light, as well as the subsequent reinfusion to the patient.10 Photoactivated 8-methoxypsoralen covalently binds to DNA pyrimidine bases, cell-surface molecules, and cytoplasmic components, causing an anti-proliferative and lethal cell defect. Data from animal auto-immune and transplantation studies suggest that photopheresed cells induce an antigen-specific immune response directed to pathogenic T-cell clones by sparing general immunocompetence.11–13 Extracorporeal photochemotherapy has been shown to be safe and promising in the management of several T-cell mediated diseases, including cutaneous T-cell lymphoma, graft vs. host disease, rejection after organ transplantation and selected auto-immune diseases.14–21
Because of the equally critical role of T lymphocytes in the pathogenesis of inflammatory bowel diseases, we performed a prospective pilot study to evaluate the safety and steroid-sparing efficacy of extracorporeal photochemotherapy in patients with steroid-dependent Crohn’s disease.
The study was performed in accordance with the Helsinki Declaration and after approval by the local ethics committee. All patients enrolled to extracorporeal photochemotherapy gave written informed consent.
Eligible patients had to be between 18 and 65 years of age. They had to have a confirmed diagnosis of Crohn’s disease according to the criteria as described by Malchow et al. and a steroid-dependent course, as defined by one or more clinical relapses within the previous 6 months upon reduction of oral prednisolone below a dose of 10 mg/day.3 Clinical relapse was characterized by a score of 200 or more on the Crohn’s disease activity index (CDAI) for two consecutive weeks.22 In all patients a specific disease history of at least 3 years had to be recorded and the extent of bowel involvement documented by radiologic and/or endoscopic investigations within 1 year prior to study inclusion.
Patients were excluded if: Crohn’s disease was restricted to the rectum; they had undergone intestinal resection within the past 12 months; or they had severe disease complications requiring surgery, or ileo- or colostomy. They were also excluded if they had: systemic infection; clinically significant hepatic, renal or cardiovascular disease; history of cancer; haemoglobin less than 105 g/L; platelet count less than 100 × 109/L; prothrombin time below 60%; or partial thromboplastin time longer than 50 s. Pregnant or breast-feeding women were not considered; neither were patients with a known hypersensitivity to heparin and 8-methoxypsoralen. Furthermore, patients who, in the investigator’s opinion, were unlikely to comply with the protocol were also excluded.
The trial was a prospective, open, monocentric study encompassing three phases with a duration of 24 weeks each.
Three weeks prior to initiation of phase 1, potentially eligible patients were instructed on the use of a diary card to score the CDAI and to document the daily steroid dose. Stool samples were obtained to rule out enteric pathogens. A complete blood count, serum levels of C-reactive protein and clinical chemistry tests were performed.
Patients were assigned to phase 1 with the lowest daily prednisolone dose retrospectively documented to prevent clinical relapse. The aim of phase 1 was to gradually taper steroids according to a standardized regimen in 2-week intervals and to evaluate the steroid maintenance dose, defined as the lowest, prospectively determined, daily prednisolone dose sustaining a CDAI of less than 200 for 2 consecutive weeks (Figure 1).
In week 24 of phase 1, CDAI, C-reactive protein and intestinal permeability were determined as baseline values for the consecutive extracorporeal photochemotherapy phase. Quality of life was measured with the Inflammatory Bowel Disease Questionnaire.23 Plasma samples to quantify adrenocorticotropic hormone were drawn between 8.00 AM and 10.00 AM.
Patients requiring a steroid maintenance dose of at least 10 mg prednisolone entered phase 2 with their individually established dose evaluated during phase 1. In phase 2, starting on day 1 of week 25, the effectiveness of extracorporeal photochemotherapy to enable steroid-withdrawal according to the standardized protocol was investigated. Extracorporeal photochemotherapy was performed using the UVAR photopheresis system (Therakos, West Chester, PA). Briefly, 240 mL of buffy coat and 300 mL of plasma were collected during each treatment from peripheral vein catheters by a standard apheresis procedure and diluted with 200 mL of saline solution plus 10 000 units of heparin; 0.2 mg of 8-methoxypsoralen (Gerot, Vienna, Austria) were added to the final enriched lymphocyte solution containing 6 × 109 cells, as described.24 The solution was passed as a film, 1 mm thick, through a disposable plastic device, exposed to a UV-A light source (2 J/cm2/cell) for 90 min, and then returned to the patient. The mean treatment time was 3.5 h. Patients were treated on two consecutive days at 14-day intervals during 24 weeks, for a total of 24 photopheresis procedures.
Patients were monitored every 2 weeks, including physical examination, determination of CDAI, routine laboratory parameters, and C-reactive protein serum levels, as well as documentation of adverse events and steroid intake. In week 48, Inflammatory Bowel Disease Questionnaire, intestinal permeability and adrenocorticotropic hormone were measured as described for phase 1, and steroid maintenance-dose was re-evaluated.
In phase 3 patients were kept on their re-evaluated steroid maintenance dose and prospectively monitored in 2-week intervals for a further 24 weeks, but at least until clinical relapse, to determine the duration of clinical efficacy after extracorporeal photochemotherapy.
The use of investigational drugs within 6 months, and ciclosporine A or methotrexate within 12 weeks before study start, as well as the initiation of such therapies during the study was not permitted. Patients treated with oral or topical 5-aminosalicylates or derivatives within 4 weeks before study entry, had to continue the medication in the same dose during the trial. Azathioprine was allowed as concomitant therapy with the dose remaining stable within 16 weeks before and through the trial. Up to three 2-week courses with metronidazole or quinolones were allowed during each study phase. Patients with a CDAI score above 450 and those who underwent abdominal surgery were withdrawn, as were patients who were unwilling to comply with the protocol for any reason.
Intestinal permeability was analysed by the lactulose–mannitol test kit for use at home as described by Wyatt et al. and expressed as a ratio of the percentage recovery of the ingested dose of lactulose relative to that of mannitol in the 5-h urine pool (%lactulose/%mannitol). The normal reference range is below 0.030.25
Primary steroid-sparing efficacy parameters were remission, defined by the maintenance of a CDAI score of less than 150 points after discontinuation of prednisolone and response, as defined by a 50% reduction of prednisolone maintenance dose at the end of phase 2. Secondary outcomes were: the cumulative prednisolone intake; mean scores of the CDAI; Inflammatory Bowel Disease Questionnaire; serum concentrations of C-reactive protein; plasma levels of adrenocorticotropic hormone; and intestinal permeability at the end of extracorporeal photochemotherapy compared to baseline.
Biopsy samples of macroscopically involved colonic mucosa obtained from three patients by endoscopy before study initiation and between 44 and 48 h after the last extracorporeal photochemotherapy were used to identify homing of photopheresed peripheral mononuclear cells to the intestinal tissue by immunofluorescence staining. As control, colonic samples from patients undergoing surveillance after polypectomy were utilized. The formalin fixed and paraffin embedded sections were de-waxed in xylol, re-hydrated in a descending alcohol series and a microwave based antigen retrieval was performed. Thereafter, sections were treated by an ‘enhanced’ indirect immunofluorescence technique, as previously described.26 The 8-methoxypsoralen-DNA photoadduct specific monoclonal antibody 9D8 diluted 1:10 in 20% FCS in phosphate buffered saline Tween was used as primary antibody, and a fluorescein isothiocyanate conjugated goat antimouse IgG as secondary antibody. The percentage of positivity was evaluated by nuclear staining with 4′,6′-Diamidino-2-Phenylindole (DAPI; Sigma, Munich, FRG) and routine staining with haematoxylin-eosin from serial sections. Slides were examined with an Olympus IM microscope.
Continuous data of clinical and laboratory measures are presented as the mean and standard error of the mean (± s.e.) or standard deviation (± s.d.). Continuous data of steroid maintenance dose, total prednisolone consumption and duration times are given as median and range, or by the 25th to the 75th quartile (IQR). A comparison of measures was performed by the paired t-test or the Wilcoxon Signed Rank test. The association between the dependent variable remission coded as 0 for non-remission and 1 for remission and independent variables was tested by means of an exact logistic regression model in LogXact separately for each variable (LogXact, Cambridge, USA: Cytel Software Corporation, 1996). Only a linear term was included because of the limited sample size (n=9). Goodness of fit was tested by the Hosmer Lemeshow statistic.
Between March 1996 and October 1997, 786 consecutive patients with Crohn’s disease were screened at our out-patient clinic. Of 86 patients who fulfilled the inclusion criteria, 24 consented to phase 1 (14 female [58%], median age 33 years, range: 19–56 years). Six patients non-adherent to the standardized steroid reduction protocol were prematurely withdrawn from phase 1. Two patients refused extracorporeal photochemotherapy for personal reasons. An additional group of six patients did not enter the extracorporeal photochemotherapy phase due to successful tapering of prednisolone below a maintenance dose of 10 mg during phase 1. Of 10 patients eligible for phase 2 (Table 1), two continued azathioprine during the study, five had a history of severe azathioprine toxicity, and three denied immunosuppressive therapy.
After failing prospective steroid tapering, patients enrolled in the extracorporeal photochemotherapy phase with a steroid maintenance dose sustaining a CDAI below 200. All complied with the steroid tapering regimen. One patient did not complete phase 2 due to laparotomic cholecystectomy in week 12. All patients who finished phase 2 were observed during phase 3.
During the extracorporeal photochemotherapy phase, response was observed in eight out of 10 patients (80%) and remission in four out of these eight patients (40%) after a median time of 10 weeks (range 8–19) and 20 weeks (range 19–23), respectively (Figure 2). The subject who terminated extracorporeal photochemotherapy phase in week 12 had attained response in week 9. The patient with the highest maintenance dose (25 mg prednisolone daily) and longest duration of steroid intake (126 months) at baseline, finished the study as a non-responder with a decrease in maintenance dose of at least 30% (17.5 mg prednisolone).
The variables age, sex, disease site and duration, prednisolone dose at baseline, azathioprine intake, smoking status, previous intestinal resection, baseline scores of CDAI and Inflammatory Bowel Disease Questionnaire, as well as serum C-reactive protein levels, were not associated with primary outcome. The duration of previous prednisolone use was significantly shorter in patients entering remission (n=4; median: 23 months, range: 12–38) compared to all other patients completing phase 2 (n=5; 66 months, range 27–126; P=0.032).
During phase 1, prednisolone dose could not be reduced among patients included in the extracorporeal photochemotherapy phase (n=10; P=0.67). However, steroid maintenance dose was significantly lowered from baseline (median 15 mg/day, range 10–25) to the end of phase 2 (median 2.5 mg/day, range 0–17.5; n=9; P=0.004). The steroid-sparing effect of extracorporeal photochemotherapy was further substantiated by a decrease in the patients’ cumulative prednisolone intake from phase 1 (median 2205 mg, interquartile range 1766–2720) to phase 2 (972 mg, interquartile range 832–1146; n=9; P=0.004).
The average of the mean (± s.e.) scores on the CDAI decreased from phase 1 (155 ± 12) to phase 2 (126 ± 14; P=0.004; n=9; Figure 3A). The median number of dose escalations due to increases on the CDAI beyond scores of 200 or more than 60 points, was three (range one to four) in phase 1 against two (range zero to three; P=0.02) in the extracorporeal photochemotherapy phase. During phase 1, in three patients a single elevation of steroid dose to 50 mg was necessary, which was never required in the extracorporeal photochemotherapy phase. The scores of the Inflammatory Bowel Disease Questionnaire increased from baseline (mean ± s.e.=186 ± 9) to the end of the extracorporeal photochemotherapy phase (204 ± 5; n=9; P=0.008; Figure 3B).
The mean concentrations (± s.e.) of C-reactive protein measured at baseline were 28 ± 6 mg/L and declined to 15 ± 5 mg/L at the end of the extracorporeal photochemotherapy phase (n=9; P=0.017; Figure 3C). Intestinal permeability was elevated in patients with ileal and ileocolonic Crohn’s disease at baseline (n=7; mean ± s.e.: 0.043 ± 0.007% lactulose/%mannitol) and declined after extracorporeal photochemotherapy (0.026 ± 0.004%lactulose/%mannitol; P=0.036).
Baseline mean (± s.d.) plasma concentration of adrenocorticotropic hormone was 7.4 ± 2.2 ng/L with values below the normal reference range (7–74 ng/L) in four patients. At the end of extracorporeal photochemotherapy the mean plasma level increased to 22.1 ± 11.9 ng/L (n=9; P=0.0052; Figure 3D) and was within the normal reference range in all patients.
During the course of 228 photopheresis treatments, only minimal adverse effects were observed, including a single episode of asymptomatic, hypovolemic hypotension in one patient, and a difficulty with venous access resulting in a small haematoma at the venipuncture site in another patient. Occasional temporary mild headaches in a single patient within 4 to 8 hours after photopheresis were considered as possibly related to the procedure. In a 55-year-old patient with additional azathioprine use for 31 months, a monoclonal gammopathy of undetermined significance was detected at the end of phase 2.
Serial evaluations of haematologic and biochemical variables showed no abnormalities due to photopheresis. No intestinal haemorrhage, deterioration of Crohn’s disease or opportunistic infection occurred.
After termination of phase 3, three of four patients remained in remission. One patient with disclosure of monoclonal gammopathy of undetermined significance at the end of phase 2 was discontinued from concomitant azathioprine and relapsed 3 weeks later. The median duration of response (n=4) during follow-up phase was 16.5 weeks (range 4–20 weeks).
In normal large bowel tissue (n=2) and colonic sections from patients with Crohn’s disease before extracorporeal photochemotherapy (n=3), staining for 8-methoxypsoralen-DNA photoadducts was negative. In colonic Crohn’s disease samples from the same patients after extracorporeal photochemotherapy (n=3), approximately 10% of infiltrating mononuclear cells displayed a positive nuclear reactivity with the DNA photoadduct specific monoclonal antibody 9D8 (Figure 4).
Our study of extracorporeal photochemotherapy in Crohn’s disease is unique due to the following features. The results are the first to provide information on the use of extracorporeal photochemotherapy in inflammatory bowel disease and indicate a significant steroid-sparing effect in patients after prospective evaluation of steroid-dependency. Despite reductions in steroid intake, patients experienced an improvement in clinical performance and quality of life. The relevance of these changes is emphasized by a decrease in the objective inflammatory parameters C-reactive protein and intestinal permeability. Extracorporeal photochemotherapy was well-tolerated in the absence of major adverse effects. Furthermore, we could demonstrate an intestine-specific homing of photopheresed leucocytes to inflamed colonic mucosal lesions.
In phase 1 of the present study, prednisolone could be successfully reduced according to a standardized slow steroid-tapering regimen in 25% of patients with a 6-month history of steroid-dependent Crohn’s disease. Feagan et al. reported similar data from a placebo-treated group of steroid-dependent patients, pointing to a decreased but still present steroid-sensitivity in some of those subjects.8 The high drop-out rate due to non-compliance during the steroid tapering phase of our protocol might be attributable to the clinical worsening at permanent steroid reductions.
Because during steroid tapering phase a steroid maintenance dose was evaluated, defined as the lowest, prospectively determined, daily prednisolone intake preventing clinical relapse, patients could be enrolled to extracorporeal photochemotherapy with inactive or very mild active disease. Only patients with a maintenance dose of at least 10 mg prednisolone, reflected by suppressed morning plasma adrenocorticotropic hormone levels were eligible for extracorporeal photochemotherapy. Based on our previous experience that extracorporeal photochemotherapy is more effective when administered on two consecutive days at 14-day intervals, we applied this regimen in the present study.10, 15 Extracorporeal photochemotherapy allowed the discontinuation of prednisolone in 40% of patients and in a further 40%, more than half dose reduction of prednisolone could be established. The period of 20 weeks to remission might be partially dependent on the slowly steroid-tapering schedule, but is reminiscent of the activity profile obtained by the immunomodulatory agent azathioprine in Crohn’s disease.7 In patients with chronic graft vs. host disease, a similar steroid-sparing efficacy of extracorporeal photochemotherapy with successful withdrawal after a median of 80 days has been reported.15
The response to extracorporeal photochemotherapy was not associated with any demographic, clinical or laboratory variable excepting the duration of previous steroid application. Whereas patients with shorter prior prednisolone use were more likely to discontinue steroids, former or current treatment with azathioprine had no influence on clinical outcome. Considering that in seven out of 10 patients azathioprine had previously failed to induce clinical response due to toxicity or insufficient therapeutic efficacy, the positive results obtained by extracorporeal photochemotherapy are particularly encouraging. After completion of extracorporeal photochemotherapy, most patients remained in remission for more than 24 weeks and response was stable for a median time of 16.5 weeks.
Trials of extracorporeal photochemotherapy in the treatment and prevention of acute solid organ rejection as well as treatment of graft vs. host disease, systemic sclerosis, pemphigus vulgaris or systemic lupus erythematosus, indicated both a potential role in immunosuppressive drug-resistent diseases and an excellent safety profile.15–20, 27 The extracorporeal application of a liquid 8-methoxypsoralen formulation directly to the leucocyte-plasma concentrate, in contrast to the former oral drug administration, improved the effectiveness of the procedure due to reliable drug level monitoring, and resulted in a marked reduction in toxicity.10 In our study, extracorporeal photochemotherapy was well-tolerated without significant adverse events. The occurrence of cholecystitis and monoclonal gammopathy of undetermined significance have never been reported previously in association with extracorporeal photochemotherapy and are most likely causally unrelated. Long-term immunosuppression with azathioprine and prednisolone, as well as the age of the patient, may certainly have contributed as known risk factors to the evolvement of the gammopathy.
The exact mechanisms by which extracorporeal photochemotherapy leads to responses in Crohn’s disease and other T-cell mediated diseases have not been elucidated. Because less than 10% of peripheral blood mononuclear cells are affected during a single photopheresis session and treatment intervals are long, a pure cytoreductive action is a most unlikely explanation for its therapeutic efficacy. This is supported by the maintenance of normal peripheral lymphocyte counts (data not shown) and the negative results of lymphocyte apheresis in patients with Crohn’s disease, as well as the results obtained with extracorporeal photochemotherapy in other indications.15, 28 In more recent studies, extracorporeal photochemotherapy is hypothesized to induce a donor-specific T-cell-mediated anti-idiotypic response against non-tolerized, pathogenic T-cells after causing apoptosis in exposed lymphocytes.29, 30 We could demonstrate a homing of photopheresed leucocytes to inflamed colonic mucosal lesions giving explicit evidence that treated cells migrate to the origin of the chronic inflammatory stimulus. Therefore, intestinal recirculating mononuclear cells could be a major target of extracorporeal photochemotherapy in Crohn’s disease. Further studies leading to the identification of these cells and their functional status might help to unravel the mechanism of this treatment. A potential shift from the dominating Th1 cytokine profile described for mucosal Crohn’s disease lesions to Th2 might contribute to the beneficial effects of extracorporeal photochemotherapy.1, 31
In spite of our study’s limitations, including the non-randomized, non-double-blinded character and the small number of patients, the design of this prospective trial, in which each patient served as his own control, justifies the relevance of the presented results. Because steroid tapering had failed in all patients before initiation of extracorporeal photochemotherapy, a spontaneous improvement or remission of Crohn’s disease appears to be very unlikely during treatment. The fact that response was noted after a latency period provides further evidence against a placebo effect. Because in our opinion ‘sham’ apheresis is considered not ethically acceptable, the protocol was not designed in a double-blinded fashion. Nevertheless, because of reservations against the photopheresis procedure, many patients who would have been eligible for phase 1 refused to participate in the trial.
The results of our preliminary study suggest that extracorporeal photochemotherapy is a promising steroid-sparing adjunctive for patients with refractory Crohn’s disease. The mildly invasive procedure is well-tolerated and without important side-effects. Therefore, our data warrant controlled randomized, prospective trials to evaluate the impact of extracorporeal photochemotherapy on the course of Crohn’s disease and to prove its benefit in patients unresponsive to immunosuppressive treatment.