- Email: [email protected]
Is there a place for cyclophosphamide in the treatment of giant-cell arteritis? A case series and systematic review
Seminars in Arthritis and Rheumatism 43 (2013) 105–112
Contents lists available at ScienceDirect
Seminars in Arthritis and Rheumatism journal homepage: www.elsevier.com/locate/semarthrit
Vasculitis
Is there a place for cyclophosphamide in the treatment of giant-cell arteritis? A case series and systematic review Hubert de Boysson, MDa, Jonathan Boutemy, MDa, Christian Creveuil, MDb, Yann Ollivier, MDa, Philippe Letellier, MD, PhDa, Christian Pagnoux, MDc, Boris Bienvenu, MD, PhDa,n a
Department of Internal Medicine, Caen University Hospital, France Biostatistics and Clinical Research Unit, Caen University Hospital, Caen, France c Division of Rheumatology, Rebecca McDonald Center for Arthritis and Autoimmune disease, Mount Sinai Hospital and University Health Network, Toronto, Ontario, Canada b
a r tic l e in fo
Keywords: Giant-cell arteritis Cyclophosphamide Glucocorticoid dependence Glucocorticoid iatrogeny
a bs t r a c t Objective: To report on the effectiveness of cyclophosphamide (CYC) to treat glucocorticoid (GC)dependent giant-cell arteritis (GCA) and/or severe GC-related side effects. Methods: Fifteen patients with GCA and treated with CYC were retrieved from the computerized patient-record system. Glucocorticoid dependence was defined as a prednisone dose of 4 20 mg/day for 6 months or 4 10 mg/day for 1 year in order not to relapse. Response to CYC was defined as improved clinical and biological findings. Remission was defined as a sustained absence (4 12 months) of active signs of vasculitis at a daily GC dose of o 7.5 mg. A literature review searched PubMed for all patients diagnosed with GCA and who received CYC. Results: Our 15 patients responded to monthly pulses of CYC, and all experienced a GC-sparing effect, including five patients who discontinued GC long term. At a median follow-up of 43 (range: 14–75) months after CYC, nine (53%) patients were still in remission and six (40%) had relapsed at 6 (3–36) months after the last CYC infusion. Twelve (80%) patients experienced side effects, leading to discontinuation of CYC in two (13%). A literature review retrieved 88 patients who received CYC: 66 for GC-dependent disease, 53 for GC toxicity, and 14 for severe organ involvement. Their median follow-up time was 24 (4–60) months. Among the 88 patients, 74 (84%) were responsive to CYC and 17 (19%) relapsed, although all were receiving a maintenance therapy with immunosuppressive agents (such as methotrexate). Twenty-nine (33%) patients experienced side effects and 11 (12.5%) discontinued treatment. Conclusion: Cyclophosphamide is an interesting option for GCA patients with GC-dependent disease or with severe GC-related side effects, especially when conventional immunosuppressive agents have failed. & 2013 Elsevier Inc. All rights reserved.
Introduction Giant-cell arteritis (GCA) is the most common type of primary vasculitis. It mainly affects large- and medium-sized vessels, including the cephalic arteries (especially the superficial temporal, the vertebral, the ophthalmic, and the posterior ciliary arteries) as well as the aorta with its first-division branches [1]. Glucocorticoids (GC) have been found to be quite effective and remain a mainstay treatment for GCA. The natural course of the disease includes frequent relapses, which occur in 50% of patients, especially when daily doses of GC are tapered to
n Corresponding author. Service de Me´decine Interne, CHU cˆote de Nacre, avenue de la cˆote de Nacre, BP 95182, 14033 Caen Cedex 9, France. E-mail address: [email protected] (B. Bienvenu).
0049-0172/$ - see front matter & 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.semarthrit.2012.12.023
o10 mg [2]. Relapses require GC dose to be raised and some patients need long-term GC therapy at a moderate- to high-dose, which increases the iatrogenic risks for elderly populations [3]. Some authors define these patients as GC dependent as they have to continue GC in order not to suffer a relapse [4]. It has been suggested that a daily dose of GC of 410 mg/day for 46 months is negatively correlated with long-term survival [5,6]. Glucocorticoid-sparing agents have been tried, such as methotrexate (MTX), azathioprine (AZT), dapsone, hydroxychloroquine, or TNF-a blockers, with inconstant results [4,7–11]. To date, MTX, when initiated at disease onset, is the most studied sparing agent and can reduce relapse rate and decrease the cumulative doses of GC [9,11]. However, even though it is considered a potential GC-sparing agent, there are only anecdotal data on its use in GC-dependent GCA [4]. A few reports, including two recent studies, suggest that cyclophosphamide (CYC) could be a good
106
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
GC-sparing agent for those patients [12–15]. In necrotizing vasculitides, such as granulomatosis with polyangiitis or microscopic polyangiitis, CYC is a cornerstone of treatment, allowing faster tapering of GC and improved immunosuppressive action [16]. Thus, in order to discuss and challenge these findings, we report on the treatment of 15 patients with GCA using intravenous pulses of CYC. In addition, a comprehensive systematic literature review of CYC in GCA was conducted.
Methods Patients Patients’ data were retrieved through a search of the computerized patient-record system in our hospital, using the keywords ‘‘giant-cell arteritis’’ and ‘‘cyclophosphamide’’. Each patient’s characteristics were then analyzed to ensure they satisfied the four following criteria: (1) GCA was diagnosed according to the 1990 American College of Rheumatology (ACR) criteria [17]; (2) all patients responded initially to GC (i.e., improved clinical symptoms and inflammatory laboratory features) but then needed to continue GCs for long term at a moderate- to highdose, or they experienced serious side effects, which led to the use of CYC as a steroid-sparing agent; (3) absence of another condition potentially associated with GCA, such as malignancy, infection, or other inflammatory disease; and (4) a follow-up of at least 1 year after the last pulse of CYC (with available data on outcomes and repeated clinical examinations and laboratory tests). We defined the disease as GC dependent when prednisone dose has to remain 4 20 mg/day for 6 months or 4 10 mg/day for 1 year in order not to relapse. However, CYC was proposed earlier if serious side effects secondary to GC occurred (e.g., unstable type-2 diabetes mellitus, severe psychiatric disorders, Cushing’s syndrome, or severe osteoporosis with fractures). All eligible patients had a complete workup, i.e., a temporal artery biopsy (TAB), a whole-body tomodensitometry, or an 18 FDG PET (positron emission tomography) scan. Exhaustive laboratory investigations included a hemogram, ionogram, serum-creatinine level, coagulation tests (activated partial thromboplastin time), erythrocyte-sedimentation rate (ESR), C-reactive protein (CRP) level, liver-function tests, serum-protein electrophoresis, immunological screening [anti-nuclear antibodies (ANA) and anti-DNA antibodies when ANA were positive, antineutrophil cytoplasmic antibodies (ANCA), complement exploration (C3, C4, CH50), and rheumatoid factor], and infectious serologies (including HIV, HBV, HCV, Lyme disease, and syphilis). All patients gave their informed consent to receive treatment. Studied parameters and definitions Demographics and clinical manifestations at onset of GCA and during the follow-up were extracted directly from the original patients’ medical charts plus the results from biological tests, TABs, radiological investigations, treatment regimens (actual and anterior), and the patients’ outcomes. Doses of GC at initiation and later, as well as dose, frequency, and tolerance to CYC infusions, were recorded. Response to CYC was defined as improved clinical and biological findings following the introduction of CYC. Remission was defined as a sustained (412 months) absence of active signs of vasculitis from receiving a daily dose of o 7.5 mg GC. Relapse consisted of reoccurrence of symptoms and/or inflammatory parameters on laboratory findings, attributable to GCA, which required a sustained increase of therapeutics. Each patient was
seen monthly during CYC pulses and continued follow-ups at our department. Literature review We focused our literature review on patients who were given CYC for GC-dependent GCA or for serious GC side effects. To identify relevant articles, we searched PubMed. The search strategy combined the following terms: giant-cell arteritis, temporal arteritis, or Horton’s disease and cyclophosphamide. All relevant articles were retrieved and additional references quoted in these articles were checked. We excluded articles when a doubtful diagnosis persisted or when data were lacking. All eligible articles were analyzed regarding the course of GCA, the therapeutic regimen, the indications for CYC initiation, and the responses based on clinical and laboratory findings. Statistical analyses Categorical variables are expressed as numbers (percentages); quantitative variables are expressed as means 7 SDs [or medians (range)]. Wilcoxon signed rank test was used to compare the doses of GC before and after CYC. Differences were considered significant when P o 0.05. All tests were performed using GraphPad Prism 5.0c.
Results Patients’ characteristics at diagnosis From our computerized patient-record system, we extracted the data of 300 patients whose history included a diagnosis of GCA: of these, 15 patients (male:female ¼ 2:13) had also received CYC, and all 15 fulfilled the entry criteria. The demographic and clinical characteristics of these 15 patients are summarized in Table 1. Median age at diagnosis was 67 (55–83) years. Five patients had a long-term history of high blood pressure that had been stabilized with adequate treatment, two had diabetes mellitus, and two had coronary disease. All patients had a GCA diagnosis that included at least three ACR criteria. Recent headaches were present at onset in all patients concomitant with visual disturbance in five (33%). Laboratory tests showed increased levels of inflammatory parameters at diagnosis in all patients with elevated ESR and CRP levels [median ESR: 95 (31–113) mm; median CRP: 127 (42–268) mg/L]. A temporal artery biopsy was performed in all patients and showed typical features of GCA in eight (53%), whereas the seven others were considered normal. We ensured that a complete workup ruled out all other mimics, such as infectious diseases or malignancies, especially in patients with negative TAB. All patients had a whole-body tomodensitometry and seven had an 18 FDG PET scan, which showed involvement of the aorta and other large vessels in four patients (nos. 6, 13, 14, 15). One patient (no. 11) also had aortitis as assessed by CT angiography. Although patients 6 and 14 had negative TAB, other conditions causing aortitis, such as Behc- et’s disease, syphilis, tuberculosis, connective tissue disorders, or sarcoidosis, were excluded after exhaustive investigations. Treatment regimen Treatment modalities are listed in Table 2. All patients received GC [median initial dose of prednisone: 50 (30–70) mg] and experienced dramatic improvement a few hours or days after starting therapy. However, all but one patient (no. 15) relapsed
Table 1 Main characteristics of 15 patients with glucocorticoid-dependent giant cell arteritis and/or severe glucocorticoid-related side effects. Gender
Age at diagnosis (years)
Comorbidities
ACR criteria
TAB
Clinical manifestations at diagnosis onset
Aorta involvement
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
F F F F F F M F F F F F M F F
75 68 75 64 65 57 76 67 55 83 66 66 63 75 78
– HBP, ICM, diabetes HBP – HBP HBP – – – – – Angor – – HBP, diabetes
5 3 3 3 3 4 5 4 4 5 4 5 5 4 5
þ þ þ þ þ þ þ þ
Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches,
þ þ þ þ þ
Total
13F/2M
67 (55–83)a
4 (3–5)a
8 þ /7
Headaches (15, 100%), weight loss (10, 67%), PMR (10, 67%), fever (10, 67%), jaw claudication (8, 53%), visual disturbance (5, 33%)
blurred vision, PMR, weight loss, jaw claudication jaw claudication, PMR blurred vision, PMR, jaw claudication fever, PMR, weight loss, jaw claudication unilateral blindness, PMR, jaw claudication, weight loss, fever PMR, fever, weight loss weight loss, jaw claudication PMR, fever, jaw claudication fever, weight loss fever, weight loss weight loss, fever blurred vision, fever, PMR, weight loss blurred vision, fever, PMR PMR, jaw claudication, weight loss fever
5 þ /10
F: female; M: male; PMR: polymyalgia rheumatica; HBP: high blood pressure; ICM: ischemic cardiomyopathy. a
median (range).
Table 2 Treatment of 15 patients with glucocorticoid-dependent giant cell arteritis and/or severe glucocorticoid-related side effects. Patient no.
Indication for CYC
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Cd, Tox Cd, Tox Tox Cd, Tox Cd, Tox Tox, Cd, Tox Cd, Tox Cd, Tox Cd Cd Cd, Tox Cd Cd Tox
Total
Cd (13, 87%), Tox (11, 73%)
Other immunosuppressive drugs before CYC
– – – MTX MTX – – – HCQ / MTX – – DAP, HCQ, MTX, THA
Dose of prednisone when CYC was introduced (mg)
Dose of prednisone at M6/M12 after the last pulse of CYC (mg)
Relapse after CYC (n), months (n)
35 15 12 17.5 20 30 30 25 7.5 20 15 17.5 20 20 12
9/5 0/0 6/5 10/10 10/9 15/12.5 3/3 3/0 5/0 7/5 5/5 5/5 5/2 3.5/0 5/0
No No No Yes, Yes, Yes, No Yes, Yes, No No Yes, No No No
20 (7.5–35)a mg
5 (0–15)a/5 (0–12.5)a mg
6 relapses 6.5 (3–36)a mo
3 36 8 5 13
3
Remission?
Total follow-up (months) Before CYC
After CYC
Yes Yes Yes No No No Yes No No Yes Yes No Yes Yes Yes
6 17 3 35 37 4 6 10 100 11 14 121 10 31 8
75 36 71 69 70 49 23 50 49 25 43 14 26 18 27
9 remissions
11 (4–121)a mo
43 (14–75)a mo
a
Median (range).
107
CYC: cyclophosphamide; Cd: corticosteroid-dependence; Tox: toxicity of corticosteroids; mo: months; M6: month 6; M12: month 12; MTX: methotrexate; HCQ: hydroxychloroquine; DAP: dapsone; THA: thalidomide; mg: milligrams; n: number.
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
Patient no.
108
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
with the first-line therapy [median time till first flare-up after GC initiation was 3.65 (1–16) months], and median daily dose of prednisone at relapse was 15 (10–40) mg. Seven patients experienced more than two relapses (range: 2–9). Intake of GC was increased for all patients who relapsed, and four of these patients received at least one alternative unsuccessful treatment: MTX for four, hydroxychloroquine for two, or dapsone and thalidomide for one. In addition, 13 of our 15 patients became GC dependent and relapsed when prednisone doses were tapered. Furthermore, 11 (73%) patients suffered from GC toxicity: weight gain [median: 11 (6–20) kg] in nine patients, psychiatric disorders (agitation, irritability, insomnia) in eight (including patient 15), muscle atrophy in four, osteoporosis with vertebral-compression fractures in three, acute coronary syndrome in one, and severe pneumonia in one. All patients received intravenous pulses of CYC: five patients were given their first three infusions at 2-weekly intervals, and then monthly for the remaining pulses. The 10 other patients received monthly infusions. Each patient received a median of six (3–10) pulses for a median of 5 (3–7) months. Thirteen patients received 500 mg/m2 (total dose per pulse: 700–1000 mg) and the two others had a fixed dose of 500 mg in each pulse (one patient had renal insufficiency, and the other was aged 76 years and in poor health). The mean cumulative dose was 5150 (72000) mg. All patients continued GC: the median daily dose of prednisone at the introduction of CYC was 20 (7.5–35) mg. Three patients (nos. 13, 14, and 15) were given MTX (0.3 mg/kg/week) for maintenance at 1 month after the last pulse of CYC and no other maintenance therapy was given to any patient after CYC therapy was completed. Outcomes and tolerance Outcomes are shown in Table 2. A response was observed in all patients within the first weeks following the CYC infusions. Twelve patients (80%) were able to reduce their GC intakes to o7.5 mg/day at 12 months after CYC completion. Of these, nine (60%) patients remained relapse-free and achieved a durable remission at a follow-up time of 43 (14–75) months after completing CYC therapy. Six (40%) patients experienced a clinical and biological relapse at 6.5 (3–36) months after CYC pulses had ended and while receiving a tapered daily dose of GC [median dose at relapse: 9.5 (0–15) mg]. Patients 4 and 6 experienced two relapses, which were controlled with increased GC. We did not observe any visual loss if the patients experienced a relapse. None of the patients who relapsed were retreated with CYC. Two patients died: patient 7 had a fatal myocardial infarction 23 months after CYC while he was still in remission, and patient 12 died of hepatocellular carcinoma 14 months after CYC while she was being treated with GC for a relapse. Regarding GC intakes, all patients experienced a sparing effect and were able to significantly reduce their GC dose at 6 and 12 months after the last pulse of CYC compared to before CYC [median dose at 6 months: 5 (0–15) mg, P o 0.0001; at 12 months: 5 (0–12.5) mg, P o 0.0001]. At the time of this analysis, five patients, including the three who received a maintenance therapy of MTX after CYC (patients 13, 14, and 15), did not experience any flare-up and were able to discontinue prednisone, which had been taken for 27 (25–42) months. Four of the five patients with large-vessel involvement (patients 11, 13, 14, and 15) achieved remission after CYC therapy, which was confirmed by a new 18FDG PET scan in patients 13, 14, and 15. Patient 6 was responsive to CYC, and a CT angiogram showed total regression of vascular inflammation, but he relapsed 8 months after CYC and did not have further vascular imagery. Patient 14 developed a thoracic aortic
aneurysm, as assessed by a repeat CT angiogram at 1 year after CYC therapy was completed; however, there were no signs of wall inflammation or other complications. Adverse events were observed in 12 (80%) patients and were likely associated with the use of CYC: infections in 11 (four who required hospital admission), hematological complications in three (anemia and/or neutropenia), and an allergic reaction to mesna in one. CYC was discontinued in two patients after three and five pulses (patients 10 and 5, respectively) because of serious infectious and hematological adverse events (repetitive pneumonias and neutropenia for the first, and erysipelas and neutropenia for the second). Notably, patient 10 did not relapse and remained relapse free during the follow-up. Patient 5 was responsive, but did not achieve remission and relapsed at 37 months. Four patients (1, 3, 8, and 12) received reduced doses of CYC in order to limit adverse events. Review of the literature Table 3 describes the published reports on patients with GCA who received cyclophosphamide. Seventeen articles were eligible for analyses, but two were excluded (GCA with necrotizing glomerulonephritis with crescents, and GCA with orbital inflammation without any information on CYC) [18,19]. Thus, 15 articles were analyzed (including two recent cohorts with 415 patients [14,15]): data on 88 patients (excluding ours) were retrieved [12–15,20–30]. All patients satisfied the ACR criteria (excepted for four patients from the study by Loock et al. where systemic involvement was suggestive of GCA [14]) and a temporal artery biopsy was specified as positive in 45 (51%). Sixty-six patients (75%) received CYC as a GC-sparing agent for a GC-dependent disease, whereas 53 (60%) and 14 (16%) were given CYC because of the toxicity of GC and the severity of clinical manifestations, respectively. Forty-nine (56%) patients had unsuccessfully tried another immunosuppressive agent before CYC (particularly MTX in 45 or AZT in 16). When specified, the method of administration seemed to be equally distributed (44 took it orally vs. 39 intravenously). Treatment duration was specified in two series [14,15] and cumulative dosage was specified in only one study [15]. In contrast, in the study by Loock et al., cumulative dose appeared to be more important in patients given oral CYC compared to patients receiving CYC intravenously (mean dose: 14,400 mg 7 7400 vs. 9200 mg 7 5700) [14]. Maintenance therapy continued after the last intake of CYC in 47 (53%) patients (MTX in 31, AZT in 13, leflunomide in 1, rituximab in 1, and intravenous immunoglobulin in 1). Overall, 74 (84%) of the 88 patients were responsive to CYC and 13 (18%) of these were able to discontinue GC. Seventeen (19%) patients experienced a relapse [13–15]. All relapses were controlled by a temporary increase in GC dose or intensification of the GC-sparing treatment (including successful retreatment with CYC in some patients). Thirty-four adverse events were specified in 29 (33%) patients, including a fatal hepatitis [15]. CYC was discontinued for serious side effects in 11 (12.5%) patients, mainly infections or cytopenias. Patients were followed-up for a median time of 24 (1–100) months.
Discussion Glucocorticoids are the therapy of choice for GCA and most patients can rapidly reduce their doses to reach a physiological or near physiological dose (r 7.5 mg/day) [31]. Proven et al. observed that 50% of patients reached this threshold level in 6 months and 75% of patients within 1 year [31]. Relapses were frequent and occurred in nearly half of the patients when GC
Table 3 Characteristics of 88 patients with giant cell arteritis and who were given cyclophosphamide (retrieved from a literature review). Analyzable patients
TAB
Indication of CYC
IV/Oral
Previous IS agents
Maintenance therapy
Outcomes
Adverse events
Follow-up (months)
Quartuccio et al. [15]
19
9 þ /15
Cd in 15, Tox in 13
Oral in 19
MTX in 12
MTX in 14
Response in 15 (79%), reduction of GC doses r5 mg/day in 9 (47%) at the end of follow-up, GC discontinuation in 6 Relapse in 4 (1–48 mo after CYC)
1 fatal liver failure, 3 hematological AEs, 2 infections, 3 cutaneous and/ or mucosal AEs, 1 liver-test disorder
36 (1–100)
At least 1 serious AE (pneumocystis pneumonia)
39 (16–46)
49
Henes et al. [13]
6/10a
NA
Cd in 6
IV in 6
Vascular stenosis in 6
Loock et al. [14]
31/35b
13 þ /35
Cd in all Tox in 25
Sailler et al. [29]
2
2þ
Epidural inflammation in 2
¨ Ruegg et al. [28]
2
1 þ /2
Vertebral artery occlusion in 2
IV in 25 Oral in 10
Oral in 1
MTX in 3
AZT in 4
Response in all
AZT in 1
MTX in 2
Decrease in FDG uptakes in all Relapse in 1 (21 mo after CYC)
MTX in 30 AZT in 14 LEF in 8
MTX in 15 AZT in 8 LEF in 1
Response in 28 (90%), reduction of GC doses o7.5 mg at 12 mo in 25 and in 8 at the end of follow-up, GC discontinuation in 3 Relapse in 12 (20.5 mo after CYC)
6 leucopenias, 6 infections, 2 allergies to mesna, 1 hemorrhagic cystitis
Discontinuation of CYC in 2
None
IvIG in 1
Response in 2 and GC discontinuation in 1 1 lost to follow-up
NA
NA
None
None
Response in 1, but death of 1 and locked-in state in the other
1 severe pneumonia
NA
IV in 1 Oral in 1
Discontinuation of CYC in 5
IV in 1
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
Reference
Discontinuation of CYC in both
Calguneri et al. [23]
1
þ
Visual loss
Oral
None
None
Response, GC dose at 17.5 mg/day at the end of follow-up
NA
24
Serradell et al. [30]
1
þ
Limb and mesenteric claudication
IV
None
None
Response and GC discontinuation
NA
NA
109
110
Table 3 (continued ) Reference
Analyzable patients
TAB
Indication of CYC
IV/Oral
Previous IS agents
Maintenance therapy
Outcomes
Adverse events
Follow-up (months)
de Vita et al. [12]
4
NA
Cd and Tox in 4
IV in 4
None
None
Response in all and reduction of GC doses o5 mg/ day in all at the end of follow-up
None
16 (10–21)
Gouet et al. [25]
6
6þ
Cd in 6
Oral in 6
None
None
Response in 2 (33%), reduction of GC dose at 11 and 7 mg/day
4 infectious and/or hematological AEs
NA
De la Casa Monje et al. [24]
6
6þ
Tox in 6
NA
None
None
Response in 5 (83%)
None
NA
Discontinuation of CYC in 1 ˜ a Sa´nchez de Pen Rivera et al. [26]
4
4þ
Tox in 4
Oral in 4
None
None
Response in all, GC discontinuation in 2
1 leukopenia
NA
Buttner et al. [22]
1
þ
Multiple cerebral infarcts
Oral
None
None
Response and reduction of GC dose
NA
4
Roelcke et al. [27]
1
þ
Meningoradiculitis
Oral
Cytosine arabinoside
AZT
No response
Hemorrhagic cystitis Discontinuation of CYC
18
Bhatia et al. [21]
1
NA
Cd and Tox
IV
AZT
RTX
Response
Severe pneumonia
6
Alba et al. [20]
3
NA
Cd in 3
NA
NA
None
Response in 3
NA
56
Cd: corticosteroid-dependence; IS: immunosuppressant; RTX: rituximab; IV: intravenous; Tox: toxicity of glucocorticoids; TAB: temporal artery biopsy; mo: months; AEs: adverse events; FDG: 18-Fluorodeoxyglucose; NA: not available. Response: improvement of clinical and biological findings; Relapse: reoccurrence of symptoms or inflammatory parameters on laboratory findings attributable to GCA and requiring a sustained increase of GC. a b
The four other patients suffered from Takayasu arteritis and were ruled out. Outcomes from 31 of 35 patients completing cyclophosphamide treatment were available for analysis.
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
Angioplasty of the mesenteric artery
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
doses were reduced to o10 mg/day [2]. Relapses require a prolonged treatment period and some patients have to continue on a low dosage indefinitely. A small subset of GCA patients require continued high doses of GC in order to avoid relapsing. Our 13 GC-dependent patients who received a threshold dose of 20 (7.5–35) mg/day of GC are rare but problematic as they are exposed to serious side effects and are difficult to treat. Achieving remission is a priority in order to reduce GC intake. Alternative treatments are therefore needed and MTX is often proposed. Although MTX has been validated in combination with high doses of GC as a first-line treatment at the onset of disease, its place in GC-dependent disease is unknown and has not been prospectively studied [4,11]. Notably, our systemic review emphasizes that more than half the patients had received MTX unsuccessfully and remained GC-dependent. In other necrotizing vasculitides, especially small- and medium-sized vessel diseases, remission is usually achieved with cyclophosphamide [16]. This unusual subset of GCA patients could be compared to patients who have severe necrotizing vasculitides, as they may require more aggressive treatment to achieve remission. This may explain why they achieved a better response to CYC rather than to MTX. Our case series and a few reports suggest that CYC may be proposed in GC-dependent GCA or for patients suffering from the iatrogenic effects of GC [12–15,20,21,24–26]. Several case reports also highlight that CYC has been successfully used in lifethreatening or severe GCA, e.g., with central nervous system or digestive involvement [22,23,27–30]. Altogether, among the 103 reported patients with GCA who were given CYC, more than 80% were responsive, whereas most had been previously resistant to multiple lines of immunosuppressive agents. Moreover, we report on the long-term remission of 450% of our patients, with significant reduction in daily doses of GC. When the mean cumulative doses of CYC are compared, we observed similar good outcomes with lower-dose intravenous treatment, similar to those of Loock et al. (12,400 mg 7 7600 from Loock et al. vs. 5150 mg 7 2000 in our patients). We also observed that, in accordance with others, CYC could control active GCA with largevessel involvement, resulting in decreased vessel uptake as assessed by 18FDG PET scans [13]. It has been reported that some young patients with GCA (r67 years old) exhibit higher inflammatory parameters and have higher rates of relapse, suggesting they could have more active and/or more severe vasculitis [32]. Notably, although the mean age at disease onset in different cohorts is reported to be 72–74 years, we and others enrolled younger patients (mean age of 69 years in our cohort, 63 years in Henes et al., and 65.3 years in Loock et al.) [13,14]. These younger patients may constitute a subgroup of GCA patients who require more aggressive treatment to control their disease. The method of administrating CYC, its duration of treatment, and the number of pulses (in intravenous uses) were extremely variable in different reports and did not rely on standardized procedures. No guidelines indicated whether maintenance therapy should be started after CYC. Notably, Quartuccio et al. and Henes et al. reported a low rate of relapse (21% and 17%, respectively), and all their patients who achieved remission with CYC received further maintenance therapy of MTX or AZT [13,15]. Another study from Loock et al. reported different outcomes with a higher proportion of relapsing patients (39%), despite that they all had received a maintenance therapy [14]. In our series, compared to Loock et al., we observed a similar rate of relapse (40%), although none of our six relapsed patients received a maintenance therapy. Moreover, we obtained a better long-term remission rate (based on the same definition of remission: 60% in ours vs. 25.8% in Loock et al.). The different outcomes observed in our study and that of Loock et al. were probably because 69% of Loock et al.’s patients
111
had large-vessel involvement vs. only 33% in our series, but also because GC management differed (initial dose, duration, and tapering scheme). We can hypothesize that we would have observed fewer relapses (similarly to Quartuccio or Henes) if all our patients had received a maintenance therapy. Finally, patients refractory to standard therapies have a high tendency to relapse and may require immunosuppressivemaintenance treatment after achieving remission with CYC, as observed in other primary systemic vasculitides [16]. Many side effects are described with CYC. We and the two other series with 4 10 patients report frequent infectious and hematological complications, such as pneumonia or urinary infection, as well as neutropenia or anemia. Regarding our review, more complications seemed to occur in patients with high cumulative doses, especially in patients receiving oral CYC [14,15]. However, most complications were not severe, although hospital admissions may be required and modification or discontinuation of CYC might be necessary. Liver tests should probably be monitored as fatal liver failure was described for a patient receiving CYC [15]. Bladder toxicity of CYC can provoke hemorrhagic cystitis in a few patients [14,27]. However, the diffuse use of mesna, which can cause allergies during CYC treatment, reduces such a risk. No excess bladder cancers were observed during the follow-up. Finally, Loock et al. suggested that the observed death rate in patients treated by CYC was no higher than that of the general population [14]. Herein, we have reported on the good outcomes of a small proportion of patients described as problematic, who achieved long-term remission with a low delivery dose of CYC. However, the retrospective design of our study is a limitation. Variations in the therapy regimens for the different patients before CYC therapy emphasize the absence of a standardized schedule for GC treatment (starting and tapering dosages) when treating GCA. In our systematic review, only two studies had 4 10 patients and most were case reports or small case series, and mainly described patients with a favorable response. Those with poor outcomes are often not published. In addition, the reviewed patients were not homogeneous and analyzed outcomes varied in different reports. The duration of follow-up and the management of GC dosages were rarely specified.
Conclusion Cyclophosphamide may be proposed as a good sparing agent in the rare subset of GCA patients who are GC-dependent or have a complicated disease, especially when conventional immunosuppressive agents, such as methotrexate or azathioprine, have failed. Intravenous use may be preferred as it diminishes the cumulative dose of CYC and may also achieve remission. As in other systemic vasculitides, a maintenance therapy may be discussed after CYC. However, many adverse events, including severe problems, have been described, highlighting the need to frequently monitor treated patients. The limited number of patients who have received such a treatment precludes any further definitive conclusion, and prospective trials are required to standardize optimal therapy strategies to treat refractory giantcell arteritis.
Acknowledgments The authors thank Dr. J.J. Parienti for his comments on the manuscript and Newmed Services for Editing the paper.
112
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
References [1] Wilkinson IM, Russell RW. Arteries of the head and neck in giant cell arteritis. A pathological study to show the pattern of arterial involvement. Arch Neurol 1972;27:378–91. [2] Salvarani C, Cantini F, Boiardi L, Hunder GG. Polymyalgia rheumatica and giant-cell arteritis. N Engl J Med 2002;347:261–71. [3] Nesher G, Sonnenblick M, Friedlander Y. Analysis of steroid related complications and mortality in temporal arteritis: a 15-year survey of 43 patients. J Rheumatol 1994;21:1283–6. [4] Kotter I, Henes JC, Wagner AD, Loock J, Gross WL. Does glucocorticosteroidresistant large-vessel vasculitis (giant cell arteritis and Takayasu arteritis) exist and how can remission be achieved? A critical review of the literature Clin Exp Rheumatol 2012;30:S114–29. [5] Hachulla E, Boivin V, Pasturel-Michon U, Fauchais AL, Bouroz-Joly J, Perez-Cousin M, et al. Prognostic factors and long-term evolution in a cohort of 133 patients with giant cell arteritis. Clin Exp Rheumatol 2001;19:171–6. [6] Uddhammar A, Eriksson AL, Nystrom L, Stenling R, Rantapaa-Dahlqvist S. Increased mortality due to cardiovascular disease in patients with giant cell arteritis in northern Sweden. J Rheumatol 2002;29:737–42. [7] Cantini F, Niccoli L, Salvarani C, Padula A, Olivieri I. Treatment of longstanding active giant cell arteritis with infliximab: report of four cases. Arthritis Rheum 2001;44:2933–5. [8] De Silva M, Hazleman BL. Azathioprine in giant cell arteritis/polymyalgia rheumatica: a double-blind study. Ann Rheum Dis 1986;45:136–8. [9] Jover JA, Hernandez-Garcia C, Morado IC, Vargas E, Banares A, FernandezGutierrez B. Combined treatment of giant-cell arteritis with methotrexate and prednisone. a randomized, double-blind, placebo-controlled trial. Ann Intern Med 2001;134:106–14. [10] Demaziere A. Dapsone in the long-term treatment of temporal arteritis. Am J Med 1989;87:3. [11] Mahr AD, Jover JA, Spiera RF, Hernandez-Garcia C, Fernandez-Gutierrez B, Lavalley MP, et al. Adjunctive methotrexate for treatment of giant cell arteritis: an individual patient data meta-analysis. Arthritis Rheum 2007;56:2789–97. [12] de Vita S, Tavoni A, Jeracitano G, Gemignani G, Dolcher MP, Bombardieri S. Treatment of giant cell arteritis with cyclophosphamide pulses. J Intern Med 1992;232:373–5. [13] Henes JC, Mueller M, Pfannenberg C, Kanz L, Koetter I. Cyclophosphamide for large vessel vasculitis: assessment of response by PET/CT. Clin Exp Rheumatol 2011;29:S43–8. ¨ [14] Loock J, Henes JC, Kotter I, Witte T, Lamprecht P, Schirmer M, et al. Treatment of refractory giant cell arteritis with cyclophosphamide. A retrospective analysis of 35 patients from three tertiary care centers. Clin Exp Rheumatol 2012;30:S70–6. [15] Quartuccio L, Maset M, De Maglio G, Pontarini E, Fabris M, Mansutti E, et al. Role of oral cyclophosphamide in the treatment of giant cell arteritis. Rheumatology (Oxford) 2012;51:1677–86.
[16] Mukhtyar C, Guillevin L, Cid MC, Dasgupta B, K de Groot, Gross W, et al. EULAR recommendations for the management of primary small and medium vessel vasculitis. Ann Rheum Dis 2009;68:310–7. [17] Hunder GG, Bloch DA, Michel BA, Stevens MB, Arend WP, Calabrese LH, et al. The American College of Rheumatology 1990 criteria for the classification of giant cell arteritis. Arthritis Rheum 1990;33:1122–8. [18] Canton CG, Bernis C, Paraiso V, Barril G, Garcia A, Osorio C, et al. Renal failure in temporal arteritis. Am J Nephrol 1992;12:380–3. [19] Puerto B, Noval S, Veiga C, Contreras I, Mateos E. Apical orbital inflammation in giant cell arteritis. Arch Soc Esp Oftalmol 2007;82:47–9. [20] Alba MA, Mena-Madrazo JA, Reyes E, Flores-Suarez LF. Giant cell arteritis in Mexican patients. J Clin Rheumatol 2012;18:1–7. [21] Bhatia A, Ell PJ, Edwards JC. Anti-CD20 monoclonal antibody (rituximab) as an adjunct in the treatment of giant cell arteritis. Ann Rheum Dis 2005;64: 1099–100. [22] Buttner T, Heye N, Przuntek H. Temporal arteritis with cerebral complications: report of four cases. Eur Neurol 1994;34:162–7. [23] Calguneri M, Cobankara V, Ozatli D, Guler G, Apras S, Pay S, et al. Is visual loss due to giant cell arteritis reversible? Yonsei Med J 2003;44:155–8. ˜ a Sa´nchez de Rivera JM, [24] de la Casa Monje R, Barbado Herna´ndez FJ, Pen Domı´nguez Castellano A, Sendino Revuelta A, Gonza´lez-Anglada I, et al. Giant-cell arteritis: the clinico-biological manifestations and the complications secondary to steroid treatment. An Med Interna 1991;8:161–5. [25] Gouet D, Marechaud R, Le Berre D, Alcalay M, Becq-Giraudon B, Boissonnot L, et al. Prognosis of treated temporal arteritis. Retrospective study of 87 cases. Presse Med 1986;15:603–6. ˜ a Sa´nchez de Rivera JM, Barbado Herna´ndez FJ, Redondo Sa´nchez C, [26] Pen Va´zquez Rodrı´guez JJ. Is cyclophosphamide useful in the treatment of giant cell arteritis? Med Clin (Barc) 1986;86:306. [27] Roelcke U, Eschle D, Kappos L, Moschopulos M, Laeng RH, Buettner UW. Meningoradiculitis associated with giant cell arteritis. Neurology 2002;59: 1811–2. [28] Ruegg S, Engelter S, Jeanneret C, Hetzel A, Probst A, Steck AJ, et al. Bilateral vertebral artery occlusion resulting from giant cell arteritis: report of 3 cases and review of the literature. Medicine (Baltimore) 2003;82:1–12. [29] Sailler LJ, Porte L, Ollier SM, Astudillo LM, Couret BG, Catalaa I, et al. Giant cell arteritis and spinal cord compression: an overlap syndrome? Mayo Clin Proc 2006;81:89–91. [30] Serradell S, Laiz A, Castano C, de Llobet JM. Usefulness of angioplasty in a patient with biopsy-proven giant cell arteritis and severe stenosis of the superior mesenteric artery. J Rheumatol 2006;33:1204–5. [31] Proven A, Gabriel SE, Orces C, O’Fallon WM, Hunder GG. Glucocorticoid therapy in giant cell arteritis: duration and adverse outcomes. Arthritis Rheum 2003;49:703–8. [32] Alba MA, Garcia-Martinez A, Espigol-Frigole G, Tavera-Bahillo I, PrietoGonzalez S, Hernandez-Rodriguez J. Differences in clinical presentation and outcome in patients with early versus late onset giant-cell arteritis (GCA): analysis of 94 patients. Ann Rheum Dis 2012;71(Suppl 3):230.
Contents lists available at ScienceDirect
Seminars in Arthritis and Rheumatism journal homepage: www.elsevier.com/locate/semarthrit
Vasculitis
Is there a place for cyclophosphamide in the treatment of giant-cell arteritis? A case series and systematic review Hubert de Boysson, MDa, Jonathan Boutemy, MDa, Christian Creveuil, MDb, Yann Ollivier, MDa, Philippe Letellier, MD, PhDa, Christian Pagnoux, MDc, Boris Bienvenu, MD, PhDa,n a
Department of Internal Medicine, Caen University Hospital, France Biostatistics and Clinical Research Unit, Caen University Hospital, Caen, France c Division of Rheumatology, Rebecca McDonald Center for Arthritis and Autoimmune disease, Mount Sinai Hospital and University Health Network, Toronto, Ontario, Canada b
a r tic l e in fo
Keywords: Giant-cell arteritis Cyclophosphamide Glucocorticoid dependence Glucocorticoid iatrogeny
a bs t r a c t Objective: To report on the effectiveness of cyclophosphamide (CYC) to treat glucocorticoid (GC)dependent giant-cell arteritis (GCA) and/or severe GC-related side effects. Methods: Fifteen patients with GCA and treated with CYC were retrieved from the computerized patient-record system. Glucocorticoid dependence was defined as a prednisone dose of 4 20 mg/day for 6 months or 4 10 mg/day for 1 year in order not to relapse. Response to CYC was defined as improved clinical and biological findings. Remission was defined as a sustained absence (4 12 months) of active signs of vasculitis at a daily GC dose of o 7.5 mg. A literature review searched PubMed for all patients diagnosed with GCA and who received CYC. Results: Our 15 patients responded to monthly pulses of CYC, and all experienced a GC-sparing effect, including five patients who discontinued GC long term. At a median follow-up of 43 (range: 14–75) months after CYC, nine (53%) patients were still in remission and six (40%) had relapsed at 6 (3–36) months after the last CYC infusion. Twelve (80%) patients experienced side effects, leading to discontinuation of CYC in two (13%). A literature review retrieved 88 patients who received CYC: 66 for GC-dependent disease, 53 for GC toxicity, and 14 for severe organ involvement. Their median follow-up time was 24 (4–60) months. Among the 88 patients, 74 (84%) were responsive to CYC and 17 (19%) relapsed, although all were receiving a maintenance therapy with immunosuppressive agents (such as methotrexate). Twenty-nine (33%) patients experienced side effects and 11 (12.5%) discontinued treatment. Conclusion: Cyclophosphamide is an interesting option for GCA patients with GC-dependent disease or with severe GC-related side effects, especially when conventional immunosuppressive agents have failed. & 2013 Elsevier Inc. All rights reserved.
Introduction Giant-cell arteritis (GCA) is the most common type of primary vasculitis. It mainly affects large- and medium-sized vessels, including the cephalic arteries (especially the superficial temporal, the vertebral, the ophthalmic, and the posterior ciliary arteries) as well as the aorta with its first-division branches [1]. Glucocorticoids (GC) have been found to be quite effective and remain a mainstay treatment for GCA. The natural course of the disease includes frequent relapses, which occur in 50% of patients, especially when daily doses of GC are tapered to
n Corresponding author. Service de Me´decine Interne, CHU cˆote de Nacre, avenue de la cˆote de Nacre, BP 95182, 14033 Caen Cedex 9, France. E-mail address: [email protected] (B. Bienvenu).
0049-0172/$ - see front matter & 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.semarthrit.2012.12.023
o10 mg [2]. Relapses require GC dose to be raised and some patients need long-term GC therapy at a moderate- to high-dose, which increases the iatrogenic risks for elderly populations [3]. Some authors define these patients as GC dependent as they have to continue GC in order not to suffer a relapse [4]. It has been suggested that a daily dose of GC of 410 mg/day for 46 months is negatively correlated with long-term survival [5,6]. Glucocorticoid-sparing agents have been tried, such as methotrexate (MTX), azathioprine (AZT), dapsone, hydroxychloroquine, or TNF-a blockers, with inconstant results [4,7–11]. To date, MTX, when initiated at disease onset, is the most studied sparing agent and can reduce relapse rate and decrease the cumulative doses of GC [9,11]. However, even though it is considered a potential GC-sparing agent, there are only anecdotal data on its use in GC-dependent GCA [4]. A few reports, including two recent studies, suggest that cyclophosphamide (CYC) could be a good
106
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
GC-sparing agent for those patients [12–15]. In necrotizing vasculitides, such as granulomatosis with polyangiitis or microscopic polyangiitis, CYC is a cornerstone of treatment, allowing faster tapering of GC and improved immunosuppressive action [16]. Thus, in order to discuss and challenge these findings, we report on the treatment of 15 patients with GCA using intravenous pulses of CYC. In addition, a comprehensive systematic literature review of CYC in GCA was conducted.
Methods Patients Patients’ data were retrieved through a search of the computerized patient-record system in our hospital, using the keywords ‘‘giant-cell arteritis’’ and ‘‘cyclophosphamide’’. Each patient’s characteristics were then analyzed to ensure they satisfied the four following criteria: (1) GCA was diagnosed according to the 1990 American College of Rheumatology (ACR) criteria [17]; (2) all patients responded initially to GC (i.e., improved clinical symptoms and inflammatory laboratory features) but then needed to continue GCs for long term at a moderate- to highdose, or they experienced serious side effects, which led to the use of CYC as a steroid-sparing agent; (3) absence of another condition potentially associated with GCA, such as malignancy, infection, or other inflammatory disease; and (4) a follow-up of at least 1 year after the last pulse of CYC (with available data on outcomes and repeated clinical examinations and laboratory tests). We defined the disease as GC dependent when prednisone dose has to remain 4 20 mg/day for 6 months or 4 10 mg/day for 1 year in order not to relapse. However, CYC was proposed earlier if serious side effects secondary to GC occurred (e.g., unstable type-2 diabetes mellitus, severe psychiatric disorders, Cushing’s syndrome, or severe osteoporosis with fractures). All eligible patients had a complete workup, i.e., a temporal artery biopsy (TAB), a whole-body tomodensitometry, or an 18 FDG PET (positron emission tomography) scan. Exhaustive laboratory investigations included a hemogram, ionogram, serum-creatinine level, coagulation tests (activated partial thromboplastin time), erythrocyte-sedimentation rate (ESR), C-reactive protein (CRP) level, liver-function tests, serum-protein electrophoresis, immunological screening [anti-nuclear antibodies (ANA) and anti-DNA antibodies when ANA were positive, antineutrophil cytoplasmic antibodies (ANCA), complement exploration (C3, C4, CH50), and rheumatoid factor], and infectious serologies (including HIV, HBV, HCV, Lyme disease, and syphilis). All patients gave their informed consent to receive treatment. Studied parameters and definitions Demographics and clinical manifestations at onset of GCA and during the follow-up were extracted directly from the original patients’ medical charts plus the results from biological tests, TABs, radiological investigations, treatment regimens (actual and anterior), and the patients’ outcomes. Doses of GC at initiation and later, as well as dose, frequency, and tolerance to CYC infusions, were recorded. Response to CYC was defined as improved clinical and biological findings following the introduction of CYC. Remission was defined as a sustained (412 months) absence of active signs of vasculitis from receiving a daily dose of o 7.5 mg GC. Relapse consisted of reoccurrence of symptoms and/or inflammatory parameters on laboratory findings, attributable to GCA, which required a sustained increase of therapeutics. Each patient was
seen monthly during CYC pulses and continued follow-ups at our department. Literature review We focused our literature review on patients who were given CYC for GC-dependent GCA or for serious GC side effects. To identify relevant articles, we searched PubMed. The search strategy combined the following terms: giant-cell arteritis, temporal arteritis, or Horton’s disease and cyclophosphamide. All relevant articles were retrieved and additional references quoted in these articles were checked. We excluded articles when a doubtful diagnosis persisted or when data were lacking. All eligible articles were analyzed regarding the course of GCA, the therapeutic regimen, the indications for CYC initiation, and the responses based on clinical and laboratory findings. Statistical analyses Categorical variables are expressed as numbers (percentages); quantitative variables are expressed as means 7 SDs [or medians (range)]. Wilcoxon signed rank test was used to compare the doses of GC before and after CYC. Differences were considered significant when P o 0.05. All tests were performed using GraphPad Prism 5.0c.
Results Patients’ characteristics at diagnosis From our computerized patient-record system, we extracted the data of 300 patients whose history included a diagnosis of GCA: of these, 15 patients (male:female ¼ 2:13) had also received CYC, and all 15 fulfilled the entry criteria. The demographic and clinical characteristics of these 15 patients are summarized in Table 1. Median age at diagnosis was 67 (55–83) years. Five patients had a long-term history of high blood pressure that had been stabilized with adequate treatment, two had diabetes mellitus, and two had coronary disease. All patients had a GCA diagnosis that included at least three ACR criteria. Recent headaches were present at onset in all patients concomitant with visual disturbance in five (33%). Laboratory tests showed increased levels of inflammatory parameters at diagnosis in all patients with elevated ESR and CRP levels [median ESR: 95 (31–113) mm; median CRP: 127 (42–268) mg/L]. A temporal artery biopsy was performed in all patients and showed typical features of GCA in eight (53%), whereas the seven others were considered normal. We ensured that a complete workup ruled out all other mimics, such as infectious diseases or malignancies, especially in patients with negative TAB. All patients had a whole-body tomodensitometry and seven had an 18 FDG PET scan, which showed involvement of the aorta and other large vessels in four patients (nos. 6, 13, 14, 15). One patient (no. 11) also had aortitis as assessed by CT angiography. Although patients 6 and 14 had negative TAB, other conditions causing aortitis, such as Behc- et’s disease, syphilis, tuberculosis, connective tissue disorders, or sarcoidosis, were excluded after exhaustive investigations. Treatment regimen Treatment modalities are listed in Table 2. All patients received GC [median initial dose of prednisone: 50 (30–70) mg] and experienced dramatic improvement a few hours or days after starting therapy. However, all but one patient (no. 15) relapsed
Table 1 Main characteristics of 15 patients with glucocorticoid-dependent giant cell arteritis and/or severe glucocorticoid-related side effects. Gender
Age at diagnosis (years)
Comorbidities
ACR criteria
TAB
Clinical manifestations at diagnosis onset
Aorta involvement
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
F F F F F F M F F F F F M F F
75 68 75 64 65 57 76 67 55 83 66 66 63 75 78
– HBP, ICM, diabetes HBP – HBP HBP – – – – – Angor – – HBP, diabetes
5 3 3 3 3 4 5 4 4 5 4 5 5 4 5
þ þ þ þ þ þ þ þ
Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches, Headaches,
þ þ þ þ þ
Total
13F/2M
67 (55–83)a
4 (3–5)a
8 þ /7
Headaches (15, 100%), weight loss (10, 67%), PMR (10, 67%), fever (10, 67%), jaw claudication (8, 53%), visual disturbance (5, 33%)
blurred vision, PMR, weight loss, jaw claudication jaw claudication, PMR blurred vision, PMR, jaw claudication fever, PMR, weight loss, jaw claudication unilateral blindness, PMR, jaw claudication, weight loss, fever PMR, fever, weight loss weight loss, jaw claudication PMR, fever, jaw claudication fever, weight loss fever, weight loss weight loss, fever blurred vision, fever, PMR, weight loss blurred vision, fever, PMR PMR, jaw claudication, weight loss fever
5 þ /10
F: female; M: male; PMR: polymyalgia rheumatica; HBP: high blood pressure; ICM: ischemic cardiomyopathy. a
median (range).
Table 2 Treatment of 15 patients with glucocorticoid-dependent giant cell arteritis and/or severe glucocorticoid-related side effects. Patient no.
Indication for CYC
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Cd, Tox Cd, Tox Tox Cd, Tox Cd, Tox Tox, Cd, Tox Cd, Tox Cd, Tox Cd Cd Cd, Tox Cd Cd Tox
Total
Cd (13, 87%), Tox (11, 73%)
Other immunosuppressive drugs before CYC
– – – MTX MTX – – – HCQ / MTX – – DAP, HCQ, MTX, THA
Dose of prednisone when CYC was introduced (mg)
Dose of prednisone at M6/M12 after the last pulse of CYC (mg)
Relapse after CYC (n), months (n)
35 15 12 17.5 20 30 30 25 7.5 20 15 17.5 20 20 12
9/5 0/0 6/5 10/10 10/9 15/12.5 3/3 3/0 5/0 7/5 5/5 5/5 5/2 3.5/0 5/0
No No No Yes, Yes, Yes, No Yes, Yes, No No Yes, No No No
20 (7.5–35)a mg
5 (0–15)a/5 (0–12.5)a mg
6 relapses 6.5 (3–36)a mo
3 36 8 5 13
3
Remission?
Total follow-up (months) Before CYC
After CYC
Yes Yes Yes No No No Yes No No Yes Yes No Yes Yes Yes
6 17 3 35 37 4 6 10 100 11 14 121 10 31 8
75 36 71 69 70 49 23 50 49 25 43 14 26 18 27
9 remissions
11 (4–121)a mo
43 (14–75)a mo
a
Median (range).
107
CYC: cyclophosphamide; Cd: corticosteroid-dependence; Tox: toxicity of corticosteroids; mo: months; M6: month 6; M12: month 12; MTX: methotrexate; HCQ: hydroxychloroquine; DAP: dapsone; THA: thalidomide; mg: milligrams; n: number.
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
Patient no.
108
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
with the first-line therapy [median time till first flare-up after GC initiation was 3.65 (1–16) months], and median daily dose of prednisone at relapse was 15 (10–40) mg. Seven patients experienced more than two relapses (range: 2–9). Intake of GC was increased for all patients who relapsed, and four of these patients received at least one alternative unsuccessful treatment: MTX for four, hydroxychloroquine for two, or dapsone and thalidomide for one. In addition, 13 of our 15 patients became GC dependent and relapsed when prednisone doses were tapered. Furthermore, 11 (73%) patients suffered from GC toxicity: weight gain [median: 11 (6–20) kg] in nine patients, psychiatric disorders (agitation, irritability, insomnia) in eight (including patient 15), muscle atrophy in four, osteoporosis with vertebral-compression fractures in three, acute coronary syndrome in one, and severe pneumonia in one. All patients received intravenous pulses of CYC: five patients were given their first three infusions at 2-weekly intervals, and then monthly for the remaining pulses. The 10 other patients received monthly infusions. Each patient received a median of six (3–10) pulses for a median of 5 (3–7) months. Thirteen patients received 500 mg/m2 (total dose per pulse: 700–1000 mg) and the two others had a fixed dose of 500 mg in each pulse (one patient had renal insufficiency, and the other was aged 76 years and in poor health). The mean cumulative dose was 5150 (72000) mg. All patients continued GC: the median daily dose of prednisone at the introduction of CYC was 20 (7.5–35) mg. Three patients (nos. 13, 14, and 15) were given MTX (0.3 mg/kg/week) for maintenance at 1 month after the last pulse of CYC and no other maintenance therapy was given to any patient after CYC therapy was completed. Outcomes and tolerance Outcomes are shown in Table 2. A response was observed in all patients within the first weeks following the CYC infusions. Twelve patients (80%) were able to reduce their GC intakes to o7.5 mg/day at 12 months after CYC completion. Of these, nine (60%) patients remained relapse-free and achieved a durable remission at a follow-up time of 43 (14–75) months after completing CYC therapy. Six (40%) patients experienced a clinical and biological relapse at 6.5 (3–36) months after CYC pulses had ended and while receiving a tapered daily dose of GC [median dose at relapse: 9.5 (0–15) mg]. Patients 4 and 6 experienced two relapses, which were controlled with increased GC. We did not observe any visual loss if the patients experienced a relapse. None of the patients who relapsed were retreated with CYC. Two patients died: patient 7 had a fatal myocardial infarction 23 months after CYC while he was still in remission, and patient 12 died of hepatocellular carcinoma 14 months after CYC while she was being treated with GC for a relapse. Regarding GC intakes, all patients experienced a sparing effect and were able to significantly reduce their GC dose at 6 and 12 months after the last pulse of CYC compared to before CYC [median dose at 6 months: 5 (0–15) mg, P o 0.0001; at 12 months: 5 (0–12.5) mg, P o 0.0001]. At the time of this analysis, five patients, including the three who received a maintenance therapy of MTX after CYC (patients 13, 14, and 15), did not experience any flare-up and were able to discontinue prednisone, which had been taken for 27 (25–42) months. Four of the five patients with large-vessel involvement (patients 11, 13, 14, and 15) achieved remission after CYC therapy, which was confirmed by a new 18FDG PET scan in patients 13, 14, and 15. Patient 6 was responsive to CYC, and a CT angiogram showed total regression of vascular inflammation, but he relapsed 8 months after CYC and did not have further vascular imagery. Patient 14 developed a thoracic aortic
aneurysm, as assessed by a repeat CT angiogram at 1 year after CYC therapy was completed; however, there were no signs of wall inflammation or other complications. Adverse events were observed in 12 (80%) patients and were likely associated with the use of CYC: infections in 11 (four who required hospital admission), hematological complications in three (anemia and/or neutropenia), and an allergic reaction to mesna in one. CYC was discontinued in two patients after three and five pulses (patients 10 and 5, respectively) because of serious infectious and hematological adverse events (repetitive pneumonias and neutropenia for the first, and erysipelas and neutropenia for the second). Notably, patient 10 did not relapse and remained relapse free during the follow-up. Patient 5 was responsive, but did not achieve remission and relapsed at 37 months. Four patients (1, 3, 8, and 12) received reduced doses of CYC in order to limit adverse events. Review of the literature Table 3 describes the published reports on patients with GCA who received cyclophosphamide. Seventeen articles were eligible for analyses, but two were excluded (GCA with necrotizing glomerulonephritis with crescents, and GCA with orbital inflammation without any information on CYC) [18,19]. Thus, 15 articles were analyzed (including two recent cohorts with 415 patients [14,15]): data on 88 patients (excluding ours) were retrieved [12–15,20–30]. All patients satisfied the ACR criteria (excepted for four patients from the study by Loock et al. where systemic involvement was suggestive of GCA [14]) and a temporal artery biopsy was specified as positive in 45 (51%). Sixty-six patients (75%) received CYC as a GC-sparing agent for a GC-dependent disease, whereas 53 (60%) and 14 (16%) were given CYC because of the toxicity of GC and the severity of clinical manifestations, respectively. Forty-nine (56%) patients had unsuccessfully tried another immunosuppressive agent before CYC (particularly MTX in 45 or AZT in 16). When specified, the method of administration seemed to be equally distributed (44 took it orally vs. 39 intravenously). Treatment duration was specified in two series [14,15] and cumulative dosage was specified in only one study [15]. In contrast, in the study by Loock et al., cumulative dose appeared to be more important in patients given oral CYC compared to patients receiving CYC intravenously (mean dose: 14,400 mg 7 7400 vs. 9200 mg 7 5700) [14]. Maintenance therapy continued after the last intake of CYC in 47 (53%) patients (MTX in 31, AZT in 13, leflunomide in 1, rituximab in 1, and intravenous immunoglobulin in 1). Overall, 74 (84%) of the 88 patients were responsive to CYC and 13 (18%) of these were able to discontinue GC. Seventeen (19%) patients experienced a relapse [13–15]. All relapses were controlled by a temporary increase in GC dose or intensification of the GC-sparing treatment (including successful retreatment with CYC in some patients). Thirty-four adverse events were specified in 29 (33%) patients, including a fatal hepatitis [15]. CYC was discontinued for serious side effects in 11 (12.5%) patients, mainly infections or cytopenias. Patients were followed-up for a median time of 24 (1–100) months.
Discussion Glucocorticoids are the therapy of choice for GCA and most patients can rapidly reduce their doses to reach a physiological or near physiological dose (r 7.5 mg/day) [31]. Proven et al. observed that 50% of patients reached this threshold level in 6 months and 75% of patients within 1 year [31]. Relapses were frequent and occurred in nearly half of the patients when GC
Table 3 Characteristics of 88 patients with giant cell arteritis and who were given cyclophosphamide (retrieved from a literature review). Analyzable patients
TAB
Indication of CYC
IV/Oral
Previous IS agents
Maintenance therapy
Outcomes
Adverse events
Follow-up (months)
Quartuccio et al. [15]
19
9 þ /15
Cd in 15, Tox in 13
Oral in 19
MTX in 12
MTX in 14
Response in 15 (79%), reduction of GC doses r5 mg/day in 9 (47%) at the end of follow-up, GC discontinuation in 6 Relapse in 4 (1–48 mo after CYC)
1 fatal liver failure, 3 hematological AEs, 2 infections, 3 cutaneous and/ or mucosal AEs, 1 liver-test disorder
36 (1–100)
At least 1 serious AE (pneumocystis pneumonia)
39 (16–46)
49
Henes et al. [13]
6/10a
NA
Cd in 6
IV in 6
Vascular stenosis in 6
Loock et al. [14]
31/35b
13 þ /35
Cd in all Tox in 25
Sailler et al. [29]
2
2þ
Epidural inflammation in 2
¨ Ruegg et al. [28]
2
1 þ /2
Vertebral artery occlusion in 2
IV in 25 Oral in 10
Oral in 1
MTX in 3
AZT in 4
Response in all
AZT in 1
MTX in 2
Decrease in FDG uptakes in all Relapse in 1 (21 mo after CYC)
MTX in 30 AZT in 14 LEF in 8
MTX in 15 AZT in 8 LEF in 1
Response in 28 (90%), reduction of GC doses o7.5 mg at 12 mo in 25 and in 8 at the end of follow-up, GC discontinuation in 3 Relapse in 12 (20.5 mo after CYC)
6 leucopenias, 6 infections, 2 allergies to mesna, 1 hemorrhagic cystitis
Discontinuation of CYC in 2
None
IvIG in 1
Response in 2 and GC discontinuation in 1 1 lost to follow-up
NA
NA
None
None
Response in 1, but death of 1 and locked-in state in the other
1 severe pneumonia
NA
IV in 1 Oral in 1
Discontinuation of CYC in 5
IV in 1
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
Reference
Discontinuation of CYC in both
Calguneri et al. [23]
1
þ
Visual loss
Oral
None
None
Response, GC dose at 17.5 mg/day at the end of follow-up
NA
24
Serradell et al. [30]
1
þ
Limb and mesenteric claudication
IV
None
None
Response and GC discontinuation
NA
NA
109
110
Table 3 (continued ) Reference
Analyzable patients
TAB
Indication of CYC
IV/Oral
Previous IS agents
Maintenance therapy
Outcomes
Adverse events
Follow-up (months)
de Vita et al. [12]
4
NA
Cd and Tox in 4
IV in 4
None
None
Response in all and reduction of GC doses o5 mg/ day in all at the end of follow-up
None
16 (10–21)
Gouet et al. [25]
6
6þ
Cd in 6
Oral in 6
None
None
Response in 2 (33%), reduction of GC dose at 11 and 7 mg/day
4 infectious and/or hematological AEs
NA
De la Casa Monje et al. [24]
6
6þ
Tox in 6
NA
None
None
Response in 5 (83%)
None
NA
Discontinuation of CYC in 1 ˜ a Sa´nchez de Pen Rivera et al. [26]
4
4þ
Tox in 4
Oral in 4
None
None
Response in all, GC discontinuation in 2
1 leukopenia
NA
Buttner et al. [22]
1
þ
Multiple cerebral infarcts
Oral
None
None
Response and reduction of GC dose
NA
4
Roelcke et al. [27]
1
þ
Meningoradiculitis
Oral
Cytosine arabinoside
AZT
No response
Hemorrhagic cystitis Discontinuation of CYC
18
Bhatia et al. [21]
1
NA
Cd and Tox
IV
AZT
RTX
Response
Severe pneumonia
6
Alba et al. [20]
3
NA
Cd in 3
NA
NA
None
Response in 3
NA
56
Cd: corticosteroid-dependence; IS: immunosuppressant; RTX: rituximab; IV: intravenous; Tox: toxicity of glucocorticoids; TAB: temporal artery biopsy; mo: months; AEs: adverse events; FDG: 18-Fluorodeoxyglucose; NA: not available. Response: improvement of clinical and biological findings; Relapse: reoccurrence of symptoms or inflammatory parameters on laboratory findings attributable to GCA and requiring a sustained increase of GC. a b
The four other patients suffered from Takayasu arteritis and were ruled out. Outcomes from 31 of 35 patients completing cyclophosphamide treatment were available for analysis.
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
Angioplasty of the mesenteric artery
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
doses were reduced to o10 mg/day [2]. Relapses require a prolonged treatment period and some patients have to continue on a low dosage indefinitely. A small subset of GCA patients require continued high doses of GC in order to avoid relapsing. Our 13 GC-dependent patients who received a threshold dose of 20 (7.5–35) mg/day of GC are rare but problematic as they are exposed to serious side effects and are difficult to treat. Achieving remission is a priority in order to reduce GC intake. Alternative treatments are therefore needed and MTX is often proposed. Although MTX has been validated in combination with high doses of GC as a first-line treatment at the onset of disease, its place in GC-dependent disease is unknown and has not been prospectively studied [4,11]. Notably, our systemic review emphasizes that more than half the patients had received MTX unsuccessfully and remained GC-dependent. In other necrotizing vasculitides, especially small- and medium-sized vessel diseases, remission is usually achieved with cyclophosphamide [16]. This unusual subset of GCA patients could be compared to patients who have severe necrotizing vasculitides, as they may require more aggressive treatment to achieve remission. This may explain why they achieved a better response to CYC rather than to MTX. Our case series and a few reports suggest that CYC may be proposed in GC-dependent GCA or for patients suffering from the iatrogenic effects of GC [12–15,20,21,24–26]. Several case reports also highlight that CYC has been successfully used in lifethreatening or severe GCA, e.g., with central nervous system or digestive involvement [22,23,27–30]. Altogether, among the 103 reported patients with GCA who were given CYC, more than 80% were responsive, whereas most had been previously resistant to multiple lines of immunosuppressive agents. Moreover, we report on the long-term remission of 450% of our patients, with significant reduction in daily doses of GC. When the mean cumulative doses of CYC are compared, we observed similar good outcomes with lower-dose intravenous treatment, similar to those of Loock et al. (12,400 mg 7 7600 from Loock et al. vs. 5150 mg 7 2000 in our patients). We also observed that, in accordance with others, CYC could control active GCA with largevessel involvement, resulting in decreased vessel uptake as assessed by 18FDG PET scans [13]. It has been reported that some young patients with GCA (r67 years old) exhibit higher inflammatory parameters and have higher rates of relapse, suggesting they could have more active and/or more severe vasculitis [32]. Notably, although the mean age at disease onset in different cohorts is reported to be 72–74 years, we and others enrolled younger patients (mean age of 69 years in our cohort, 63 years in Henes et al., and 65.3 years in Loock et al.) [13,14]. These younger patients may constitute a subgroup of GCA patients who require more aggressive treatment to control their disease. The method of administrating CYC, its duration of treatment, and the number of pulses (in intravenous uses) were extremely variable in different reports and did not rely on standardized procedures. No guidelines indicated whether maintenance therapy should be started after CYC. Notably, Quartuccio et al. and Henes et al. reported a low rate of relapse (21% and 17%, respectively), and all their patients who achieved remission with CYC received further maintenance therapy of MTX or AZT [13,15]. Another study from Loock et al. reported different outcomes with a higher proportion of relapsing patients (39%), despite that they all had received a maintenance therapy [14]. In our series, compared to Loock et al., we observed a similar rate of relapse (40%), although none of our six relapsed patients received a maintenance therapy. Moreover, we obtained a better long-term remission rate (based on the same definition of remission: 60% in ours vs. 25.8% in Loock et al.). The different outcomes observed in our study and that of Loock et al. were probably because 69% of Loock et al.’s patients
111
had large-vessel involvement vs. only 33% in our series, but also because GC management differed (initial dose, duration, and tapering scheme). We can hypothesize that we would have observed fewer relapses (similarly to Quartuccio or Henes) if all our patients had received a maintenance therapy. Finally, patients refractory to standard therapies have a high tendency to relapse and may require immunosuppressivemaintenance treatment after achieving remission with CYC, as observed in other primary systemic vasculitides [16]. Many side effects are described with CYC. We and the two other series with 4 10 patients report frequent infectious and hematological complications, such as pneumonia or urinary infection, as well as neutropenia or anemia. Regarding our review, more complications seemed to occur in patients with high cumulative doses, especially in patients receiving oral CYC [14,15]. However, most complications were not severe, although hospital admissions may be required and modification or discontinuation of CYC might be necessary. Liver tests should probably be monitored as fatal liver failure was described for a patient receiving CYC [15]. Bladder toxicity of CYC can provoke hemorrhagic cystitis in a few patients [14,27]. However, the diffuse use of mesna, which can cause allergies during CYC treatment, reduces such a risk. No excess bladder cancers were observed during the follow-up. Finally, Loock et al. suggested that the observed death rate in patients treated by CYC was no higher than that of the general population [14]. Herein, we have reported on the good outcomes of a small proportion of patients described as problematic, who achieved long-term remission with a low delivery dose of CYC. However, the retrospective design of our study is a limitation. Variations in the therapy regimens for the different patients before CYC therapy emphasize the absence of a standardized schedule for GC treatment (starting and tapering dosages) when treating GCA. In our systematic review, only two studies had 4 10 patients and most were case reports or small case series, and mainly described patients with a favorable response. Those with poor outcomes are often not published. In addition, the reviewed patients were not homogeneous and analyzed outcomes varied in different reports. The duration of follow-up and the management of GC dosages were rarely specified.
Conclusion Cyclophosphamide may be proposed as a good sparing agent in the rare subset of GCA patients who are GC-dependent or have a complicated disease, especially when conventional immunosuppressive agents, such as methotrexate or azathioprine, have failed. Intravenous use may be preferred as it diminishes the cumulative dose of CYC and may also achieve remission. As in other systemic vasculitides, a maintenance therapy may be discussed after CYC. However, many adverse events, including severe problems, have been described, highlighting the need to frequently monitor treated patients. The limited number of patients who have received such a treatment precludes any further definitive conclusion, and prospective trials are required to standardize optimal therapy strategies to treat refractory giantcell arteritis.
Acknowledgments The authors thank Dr. J.J. Parienti for his comments on the manuscript and Newmed Services for Editing the paper.
112
H. de Boysson et al. / Seminars in Arthritis and Rheumatism 43 (2013) 105–112
References [1] Wilkinson IM, Russell RW. Arteries of the head and neck in giant cell arteritis. A pathological study to show the pattern of arterial involvement. Arch Neurol 1972;27:378–91. [2] Salvarani C, Cantini F, Boiardi L, Hunder GG. Polymyalgia rheumatica and giant-cell arteritis. N Engl J Med 2002;347:261–71. [3] Nesher G, Sonnenblick M, Friedlander Y. Analysis of steroid related complications and mortality in temporal arteritis: a 15-year survey of 43 patients. J Rheumatol 1994;21:1283–6. [4] Kotter I, Henes JC, Wagner AD, Loock J, Gross WL. Does glucocorticosteroidresistant large-vessel vasculitis (giant cell arteritis and Takayasu arteritis) exist and how can remission be achieved? A critical review of the literature Clin Exp Rheumatol 2012;30:S114–29. [5] Hachulla E, Boivin V, Pasturel-Michon U, Fauchais AL, Bouroz-Joly J, Perez-Cousin M, et al. Prognostic factors and long-term evolution in a cohort of 133 patients with giant cell arteritis. Clin Exp Rheumatol 2001;19:171–6. [6] Uddhammar A, Eriksson AL, Nystrom L, Stenling R, Rantapaa-Dahlqvist S. Increased mortality due to cardiovascular disease in patients with giant cell arteritis in northern Sweden. J Rheumatol 2002;29:737–42. [7] Cantini F, Niccoli L, Salvarani C, Padula A, Olivieri I. Treatment of longstanding active giant cell arteritis with infliximab: report of four cases. Arthritis Rheum 2001;44:2933–5. [8] De Silva M, Hazleman BL. Azathioprine in giant cell arteritis/polymyalgia rheumatica: a double-blind study. Ann Rheum Dis 1986;45:136–8. [9] Jover JA, Hernandez-Garcia C, Morado IC, Vargas E, Banares A, FernandezGutierrez B. Combined treatment of giant-cell arteritis with methotrexate and prednisone. a randomized, double-blind, placebo-controlled trial. Ann Intern Med 2001;134:106–14. [10] Demaziere A. Dapsone in the long-term treatment of temporal arteritis. Am J Med 1989;87:3. [11] Mahr AD, Jover JA, Spiera RF, Hernandez-Garcia C, Fernandez-Gutierrez B, Lavalley MP, et al. Adjunctive methotrexate for treatment of giant cell arteritis: an individual patient data meta-analysis. Arthritis Rheum 2007;56:2789–97. [12] de Vita S, Tavoni A, Jeracitano G, Gemignani G, Dolcher MP, Bombardieri S. Treatment of giant cell arteritis with cyclophosphamide pulses. J Intern Med 1992;232:373–5. [13] Henes JC, Mueller M, Pfannenberg C, Kanz L, Koetter I. Cyclophosphamide for large vessel vasculitis: assessment of response by PET/CT. Clin Exp Rheumatol 2011;29:S43–8. ¨ [14] Loock J, Henes JC, Kotter I, Witte T, Lamprecht P, Schirmer M, et al. Treatment of refractory giant cell arteritis with cyclophosphamide. A retrospective analysis of 35 patients from three tertiary care centers. Clin Exp Rheumatol 2012;30:S70–6. [15] Quartuccio L, Maset M, De Maglio G, Pontarini E, Fabris M, Mansutti E, et al. Role of oral cyclophosphamide in the treatment of giant cell arteritis. Rheumatology (Oxford) 2012;51:1677–86.
[16] Mukhtyar C, Guillevin L, Cid MC, Dasgupta B, K de Groot, Gross W, et al. EULAR recommendations for the management of primary small and medium vessel vasculitis. Ann Rheum Dis 2009;68:310–7. [17] Hunder GG, Bloch DA, Michel BA, Stevens MB, Arend WP, Calabrese LH, et al. The American College of Rheumatology 1990 criteria for the classification of giant cell arteritis. Arthritis Rheum 1990;33:1122–8. [18] Canton CG, Bernis C, Paraiso V, Barril G, Garcia A, Osorio C, et al. Renal failure in temporal arteritis. Am J Nephrol 1992;12:380–3. [19] Puerto B, Noval S, Veiga C, Contreras I, Mateos E. Apical orbital inflammation in giant cell arteritis. Arch Soc Esp Oftalmol 2007;82:47–9. [20] Alba MA, Mena-Madrazo JA, Reyes E, Flores-Suarez LF. Giant cell arteritis in Mexican patients. J Clin Rheumatol 2012;18:1–7. [21] Bhatia A, Ell PJ, Edwards JC. Anti-CD20 monoclonal antibody (rituximab) as an adjunct in the treatment of giant cell arteritis. Ann Rheum Dis 2005;64: 1099–100. [22] Buttner T, Heye N, Przuntek H. Temporal arteritis with cerebral complications: report of four cases. Eur Neurol 1994;34:162–7. [23] Calguneri M, Cobankara V, Ozatli D, Guler G, Apras S, Pay S, et al. Is visual loss due to giant cell arteritis reversible? Yonsei Med J 2003;44:155–8. ˜ a Sa´nchez de Rivera JM, [24] de la Casa Monje R, Barbado Herna´ndez FJ, Pen Domı´nguez Castellano A, Sendino Revuelta A, Gonza´lez-Anglada I, et al. Giant-cell arteritis: the clinico-biological manifestations and the complications secondary to steroid treatment. An Med Interna 1991;8:161–5. [25] Gouet D, Marechaud R, Le Berre D, Alcalay M, Becq-Giraudon B, Boissonnot L, et al. Prognosis of treated temporal arteritis. Retrospective study of 87 cases. Presse Med 1986;15:603–6. ˜ a Sa´nchez de Rivera JM, Barbado Herna´ndez FJ, Redondo Sa´nchez C, [26] Pen Va´zquez Rodrı´guez JJ. Is cyclophosphamide useful in the treatment of giant cell arteritis? Med Clin (Barc) 1986;86:306. [27] Roelcke U, Eschle D, Kappos L, Moschopulos M, Laeng RH, Buettner UW. Meningoradiculitis associated with giant cell arteritis. Neurology 2002;59: 1811–2. [28] Ruegg S, Engelter S, Jeanneret C, Hetzel A, Probst A, Steck AJ, et al. Bilateral vertebral artery occlusion resulting from giant cell arteritis: report of 3 cases and review of the literature. Medicine (Baltimore) 2003;82:1–12. [29] Sailler LJ, Porte L, Ollier SM, Astudillo LM, Couret BG, Catalaa I, et al. Giant cell arteritis and spinal cord compression: an overlap syndrome? Mayo Clin Proc 2006;81:89–91. [30] Serradell S, Laiz A, Castano C, de Llobet JM. Usefulness of angioplasty in a patient with biopsy-proven giant cell arteritis and severe stenosis of the superior mesenteric artery. J Rheumatol 2006;33:1204–5. [31] Proven A, Gabriel SE, Orces C, O’Fallon WM, Hunder GG. Glucocorticoid therapy in giant cell arteritis: duration and adverse outcomes. Arthritis Rheum 2003;49:703–8. [32] Alba MA, Garcia-Martinez A, Espigol-Frigole G, Tavera-Bahillo I, PrietoGonzalez S, Hernandez-Rodriguez J. Differences in clinical presentation and outcome in patients with early versus late onset giant-cell arteritis (GCA): analysis of 94 patients. Ann Rheum Dis 2012;71(Suppl 3):230.