Abstract
Objective. To analyze the clinical efficacy of anti-tumor necrosis factor-α (TNF-α) therapy in treatment of synovitis, acne, pustulosis, hyperostosis, osteitis (SAPHO) syndrome, we describe cases of refractory SAPHO syndrome and review cases treated with anti-TNF-α reported in the literature.
Methods. We describe 6 cases of patients with SAPHO syndrome treated with anti-TNF-α between 2004 and 2008. Therapeutic response was evaluated according to improvement in pain score, amelioration of disease activity, and improvement in function. The efficacy of treatment was considered to be reduced need for analgesics and/or antiinflammatory therapy.
Results. In our series, 4 patients received infliximab, 1 etanercept, and 1 adalimumab. These treatments brought clinical response in 4 patients (66.6%): response was sustained with infliximab in 1 case for 7 months; with adalimumab in another case for 22 months; and with etanercept in 2 cases for 1 and 42 months, respectively. In contrast, 2 other patients showed no response to infliximab. Improvement was initially temporary after infusions 1 and 2, then pain recurred at Week 14. Skin lesions were healed in 3 of 4 cases, but recurred or worsened in 2 cases, after infusion 2 of infliximab. Treatment was generally well tolerated. Paradoxical psoriasis was noted in 2 cases and urticaria in 1.
Conclusion. Given our results and those from the literature, TNF-α blockers should be considered in the therapeutic strategy of refractory cases of SAPHO syndrome, despite their effect seeming less impressive than in other spondyloarthropathies.
SAPHO syndrome (synovitis, acne, pustulosis, hyperostosis, osteitis) was defined as a clinicoradiological entity combining skin, bone, and joint manifestations. The denotation was first proposed in 19871 after investigation of 85 cases. The main target is the anterior chest wall, where common findings include osteosclerosis and enlargement of the internal portion of the clavicles and first ribs. The axial skeleton (spine and sacroiliac joints) and peripheral bones can be involved. Skin diseases associated with this type of osteitis include palmoplantar pustulosis (PPP), severe acne, and various patterns of psoriasis.
Converging arguments indicate that SAPHO syndrome can be classified with the inflammatory spondyloarthropathies, which typically affect the spine. The therapeutic strategy was largely inspired by that for spondyloarthropathies. Because of the low incidence and different patterns of disease expression, most reports describe treatment responses from anecdotal cases and small series of patients. Thus, the therapeutic strategy of SAPHO syndrome remains unclear. In general terms, treatment consists of nonsteroidal antiinflammatory drugs (NSAID) for osteoarticular symptoms and topical treatment for skin lesions. Efficacy is considered satisfactory in about two-thirds of cases2.
A recent interventional study of patients with SAPHO syndrome showed positive bacteriological cultures for Propionibacterium acnes in 14 of 21 (67%) patients who had undergone a needle biopsy of osteitis lesions3. This is a significant addition to publications showing an association of SAPHO with P. acnes. Thus, different protocols of antibiotic therapy were tried: at least 6 uncontrolled studies showed efficacy of antibiotic therapy (azithromycin, doxycycline, sulfamethoxazole/trimethoprim), but decreased disease activity might be related to the natural remitting course of the disease and not to the efficacy of the antibiotic therapy4. Refractory forms have led to the use of disease-modifying antirheumatic drugs. Therapy with methotrexate (MTX) and sulfasalazine was attempted, but with unsatisfactory results2. Bisphosphonates were proposed, but their efficacy is difficult to assess in the absence of a randomized controlled study and requires more evidence. Nevertheless, in most reports, this treatment is considered effective for chronic active osteitis, usually with rapid and prolonged pain relief5. Refractory cases of SAPHO require intensive care. With analogy to spondyloarthropathy therapy, tumor necrosis factor-α (TNF-α) blockers have been proposed as third-line therapy.
We describe 6 cases of SAPHO syndrome treated with anti-TNF-α agents between 2004 and 2008 and review cases in the literature. Patients were selected according to 2 inclusion criteria: (1) diagnosis of SAPHO syndrome by clinical and radiologic criteria [on bone radiography, computerized axial tomography (CT), and scintigraphy]; and (2) a painful clinical progression despite therapy with several NSAID associated with at least one of the following treatments: MTX, pamidronate, antibiotics.
The therapeutic response was evaluated according to improvement in pain score on a visual analog scale (VAS, 0–100 mm), amelioration of disease activity [Bath Ankylosing Sponylitis Disease Activity Index (BASDAI)], and improvement in function [Bath Ankylosing Spondylitis Functional Index (BASFI)]. These latter 2 indexes, designed for measuring disease activity and function in primary ankylosing spondylitis, have been shown to have good internal consistency in peripheral and axial psoriatic arthritis6. For this reason, we adopted these scoring methods in our study.
The efficacy of treatment was considered to be reduced need for analgesics and/or antiinflammatory therapy. Treatment safety was evaluated.
Case report 1
A 58-year-old woman presented with a history of pain in the anterior chest wall with inflammatory dorsalgia for the previous 6 years. Erythrocyte sedimentation rate (ESR) was 4 mm/h and C-reactive protein 0.4 mg/l. Spinal magnetic resonance imaging (MRI) showed osteosclerosis of the T8 vertebra. Sacroiliitis was not observed on pelvic plain radiography. Biopsy of the T8 lesion demonstrated chronic inflammation but no tumor. Some months later, PPP developed. A diagnosis of SAPHO syndrome was made. She initially received NSAID, then several monthly infusions of pamidronate (60 mg over 2 yrs). However, right sternoclavicular pain persisted, as well as inflammatory talalgia. In light of the refractory SAPHO syndrome, treatment with infliximab (5 mg/kg at Weeks 0, 2, and 6 and then every 8 weeks) was started. The patient showed clinical response, with gradual amelioration of disease activity over 7 months. The BASDAI score decreased, from 6.2 before infusion 1, to 2.4 before infusion 5; the BASFI decreased from 5.5 to 2.2 and the VAS score from 65 to 50. Morning stiffness was alleviated, from 4 h before infusion 1 to 1.5 h before infusion 5. NSAID were stopped. Similarly, PPP was relieved. Tolerance to infliximab treatment was good.
Case report 2
A 36-year-old woman with a history of Hashimoto thyroiditis presented over a period of 7 years with recurrent episodes of swelling in the right knee. Knee arthroscopy in 2004 revealed active hyperplastic synovitis. Two years later, inflammatory, left sternoclavicular pain appeared, with PPP. ESR was 46 mm/h and CRP 7 mg/l. CT scan of the chest wall revealed marked erosion and narrowing in the left sternoclavicular and first left sternocostal joints, with irregularity in and sclerosis of the side rib and hypertrophy of the sternal side. MRI of the spine and sacroiliac joints gave normal results. She was given a diagnosis of SAPHO syndrome. NSAID and MTX were ineffective. The left sternoclavicular joint remained inflamed, with cervicodorsal rachialgia and oligoarthritis of the knees. Accordingly, treatment with infliximab (5 mg/kg at Week 0, 2, and 6 and then every 8 weeks) was started. Before infusion 2, clinical response was observed: the BASDAI decreased from 4.7 to 2.6 and the BASFI from 2.0 to 1.1. Morning stiffness decreased and ESR returned to the normal range. However, at Week 14, pain recurred in the anterior chest wall, and BASDAI increased to 3.4. The patient showed paradoxical psoriasis and substantial exacerbation of PPP confirmed on skin biopsy. A switch to etanercept was attempted. However, after 3 months of treatment, disease remained active, with BASDAI 4.6 and BASFI 3.2.
Case report 3
A 45-year-old woman presented with a rash of PPP and subsequent inflammatory pain of the left sternoclavicular joint present for 6 years. CT scan revealed osteosclerosis of the sternum, first left rib, and iliac side of the right sacroiliac joint. She was given a diagnosis of SAPHO syndrome. Symptoms were resistant to treatment (NSAID, antibiotics, and monthly infusion of pamidronate 60 mg). BASDAI was 5 and VAS score 85 mm. Treatment with infliximab 5 mg/kg was started. At Week 2 and before infusion 2, the BASDAI decreased to 1.0; the patient had no morning stiffness. NSAID were stopped; skin lesions had healed. By contrast, at Week 14 and before infliximab infusion 3, pain recurred in the anterior chest wall. The BASDAI increased to 6.4; NSAID were restarted. During infusion 5, due to an allergic urticaria reaction on the trunk and thighs, therapy was interrupted. A switch to adalimumab was attempted and led to partial improvement within 1 month; BASDAI was 4.6 and VAS score 60 mm.
Case report 4
A 61-year-old woman complained of inflammatory pain in the anterior chest wall; she had no skin lesions. A persistent biologic inflammatory syndrome (ESR 64 mm/h and CRP 26 mg/l) was noted. Bone scintigraphy revealed hyperfixation of the manubrium. CT scan revealed sclerosis of sternoclavicular joints without infiltration or involvement of soft tissue. Plain radiography of the pelvis revealed marked hyperostosis and osteosclerosis of the left sacroiliac joint. The patient was given a diagnosis of SAPHO syndrome, and treatment with several NSAID and pamidronate (2 infusions). However, disease remained active. Treatment with adalimumab 40 mg every 2 weeks was then introduced. Clinical remission occurred after injection 2 and during 2-year followup. BASDAI decreased from 7.9 before treatment to 1.1 after 22 months of therapy; BASFI decreased from 6.2 to 0.4 and VAS score from 86 to 8 mm. ESR decreased from 55 to 9 mm/h after 2 years and CRP from 38 to 5 mg/l. Finally, NSAID were stopped after 14 months of treatment.
Case report 5
A 53-year-old woman presented with swelling of the right upper limb. Doppler venous ultra-sonography revealed stenosis of the humeral, axillary, and right subclavian veins. She also complained of right clavicular and sternocostal pain. CT scan revealed hyperostosis and osteosclerosis of the right clavicle, first 3 ribs, and sternal manubrium. Bone scintigraphy revealed hyperfixation of these lesions. Biopsy of the clavicle showed a chronic aseptic inflammatory lesion without osteolytic reaction. She was given a diagnosis of SAPHO syndrome complicated by venous thrombosis. She showed no skin manifestations. Symptoms were resistant to different NSAID, i.e., several infusions of pamidronate 60 mg monthly during 18 months and MTX. She continued to have stiffness for 2 h in the morning and the VAS score was 90 mm; ESR was 51 mm/h and CRP 76 mg/l. Treatment with a TNF-α blocker (etanercept) was started. At Week 4, she reported no pain, with VAS score 0. She did not wake at night and had no morning stiffness; ESR was 19 mm/h and CRP 11 mg/l.
Case report 6
A 29-year-old woman had had rachialgia associated with inflammatory pain of the right sternoclavicular joint for 4 years. ESR was 20 mm/h and CRP 6 mg/l. Bone scintigraphy revealed hyperfixation of T5, T6 and T7, manubrium, and right sternoclavicular joint. CT scan revealed osteosclerosis of T6 and T7 suggesting inflammatory spondylitis, which was confirmed by spinal MRI. She was given a diagnosis of SAPHO syndrome and treatment was started with several NSAID and monthly infusions of pamidronate 60 mg, which initially relieved pain. Symptoms recurred, with PPP. She had morning stiffness for 2 h; ESR was 34 mm/h and CRP 10 mg/l. Infusions were started with infiximab 5 mg/kg at Weeks 0, 2, and 6 and then every 8 weeks. Clinical response occurred at Week 2, with no more morning stiffness. ESR (27 mm/h) and CRP (7 mg/l) were not significantly changed. Three weeks after infusion 1, a psoriatic rash appeared on the thighs, trunk, and face and was confirmed by skin biopsy. A switch to etanercept led to complete relief of pain and the inflammatory syndrome. No psoriasis or PPP was found. The current followup is 42 months.
DISCUSSION
In general terms, NSAID and analgesic treatments help control osteoarticular symptoms of SAPHO syndrome. In cases that do not respond, second-line therapies are tried. To date, SAPHO syndrome is commonly refractory to glucocorticoids and disease-modifying antirheumatic drugs including MTX and sulfasalazine. In some patients with refractory disease, anti-TNF-α therapy has been proposed as a third-line therapy. Its use in SAPHO syndrome is mostly based on the difficulties encountered when treating SAPHO syndrome. Indeed, as the pathogenesis of SAPHO syndrome is unknown and due to its low incidence and different patterns of disease expression, there is no well established therapy7. Most reports describe treatment responses based on anecdotal cases and small series of patients. Further, the use of anti-TNF-α therapy is also based on the efficacy of TNF blockers in spondyloarthopathies. Most investigators8,9 have supported a link between SAPHO syndrome and spondyloarthopathies because of frequent axial involvement and association with psoriasis and inflammatory bowel diseases in 13% of cases10, relatively high frequency of positive HLA-B27 in 15% of cases11, and efficacy of NSAID. Some have also suggested that SAPHO syndrome could be an atypical form of psoriatic arthritis12.
Proinflammatory cytokines such as TNF-α may be involved in generating or perpetuating the rheumatic manifestations of SAPHO syndrome. A group in Germany documented TNF-α overexpression in mandibular osteitis13. Further support for the involvement of TNF-α comes from the promising results obtained with TNF-α antagonists in patients with SAPHO syndrome.
In our cases of some refractory forms of SAPHO syndrome, TNF-α blockers were prescribed. Clinical response was achieved in 4 cases (Patients 1, 4, 5, and 6): response was sustained with infliximab in Case 1 for 7 months; with adalimumab in Case 4 for 22 months; and with etanercept in Case 5 for 1 month and in Case 6 for 42 months. Clinical response occurred after infusion 4 of infliximab and after injection 2 of etanercept and adalimumab. By contrast, 2 other patients (Cases 2 and 3) showed no significant response to infliximab.
Only 19 reports of adult cases of SAPHO syndrome treated with anti-TNF-α have been published; characteristics of these cases are given in Table 113–22. These cases were in general refractory to alternative treatment such as bisphosphonates, MTX, sulfasalazine, and even cyclosporine. The only 2 anti-TNF-α agents used were infliximab and etanercept. Infliximab was prescribed in all cases except 2. Therefore, our Case 4 was the first reported case of SAPHO syndrome treated with adalimumab. In the 19 previous cases, the clinical response was rapid after infusion 1 in 13 patients (68.4%), after infusion 2 in 4 (21%), and after infusion 3 in 2 (10.5%). Clinical response was maintained in all patients. Osteoarticular pain did not recur; this result, often described as remission, was maintained with a followup of 8 to 18 months during treatment. The maximal followup was 42 months in our series. Sometimes the same result is maintained when anti-TNF-α therapy is stopped14.
In parallel, we observed healed skin lesions after the first infusion in 3 of 4 of our cases (Cases 1, 2, and 3). These results support reports in the literature. Indeed, among the 15 cases of skin manifestations in the literature, healing was observed mainly for acne — in 12 (80%) of these cases. Of note, the experience of the Italian investigators suggested lower efficacy of infliximab for PPP than for osteoarticular manifestations17 detected in 4 cases of SAPHO syndrome. In our series, skin lesions recurred or worsened after infusion 2 of infliximab in 3 cases (Cases 2, 3, and 6). The effect of anti-TNF-α on skin lesions can be paradoxical. In a recent review of the literature concerning skin complications from anti-TNF-α therapy, among 120 patients (with rheumatoid arthritis, ankylosing spondylitis, SAPHO syndrome, psoriatic arthritis, and other diagnoses), 37 cases of recurring PPP were noted23. A possible reason for deterioration of skin pustulosis could be activation of P. acnes with anti-TNF-α. Thus, combined therapy including anti-TNF medication and an antibiotic may be a reasonable solution4.
Treatment was generally well tolerated in our cases, with no severe side effects. Nevertheless, we observed drug-induced dermatitis of the lower limbs in 1 case and paradoxical psoriasis in 2 cases treated with infliximab (Cases 2 and 6). Various skin reactions have been reported with infliximab therapy, often several months after initiation of treatment; reactions include erythema multiforme, skin vasculitis, lichenoid eruption, and annular granuloma, as well as eczematous rashes.
Our series suggests the efficacy of anti-TNF-α therapy in accord with cases described in the literature. In all these cases, the effectiveness was quick, with longterm maintenance. Five of our patients had already received bisphosphonates, without pain relief. Four of these cases responded to anti-TNF-α therapy. Similarly, we observed ineffectiveness of bisphosphonates in 6 cases in the literature, which suggests anti-TNF-α therapy as having a positive response. Bisphosphonates have been given to patients with SAPHO syndrome because of the features of the bone lesions, with new bone formation a characteristic pathologic feature, and some positive but transient results obtained with these drugs in some patients with spondyloarthropathies but not in controlled studies24. Thus, anti-TNF-α therapy seems effective even for disease resistant to bisphosphonates. However, this efficacy seems to be less impressive than that usually reported for other spondyloarthropathies.
Because of difficulties of treatment of SAPHO syndrome, anti-TNF-α therapy was proposed for refractory forms as third-line therapy. In our series of 6 cases, treatment seemed effective in 4 (66.7%). Reports of 19 cases of SAPHO syndrome treated with anti-TNF-α therapy in the literature show a positive response in most. These data suggest a real effectiveness of these biotherapies for SAPHO syndrome. We also report the first case of SAPHO syndrome treated effectively with adalimumab.
Given our results and reports in the literature, anti-TNF-α therapies are an efficient treatment for patients with refractory SAPHO syndrome, with an early response. Because randomized controlled trials are difficult to perform for this rare disease, confirmation of these findings in a study with a larger number of patients is required.
- Accepted for publication March 18, 2010.