Dear Editor,

We would like to thank van der Maas and colleagues for their interest in and useful feedback on our review [1]. They raise several important points. First, regarding the interpretation of the studies included in the review, we focused on studies examining discontinuation of biologic disease modifying antirheumatic drugs (DMARDs) that examined patient outcomes. Thus, we had difficulty in fitting a tapering and discontinuation study [2] in the framework of our review since some patients fail during the tapering phase and never fully discontinue drug.

Nevertheless, we extracted outcome (maintenance of discontinuation) information from Figure 2D of their paper [2]. We interpreted this figure as the "proportion free of failure" after discontinuation among those who underwent discontinuation (n = 12). Thus, it was interpreted as all patients failed by 7 months (around 200 days) after discontinuation. The "follow up duration" we defined in our review is the time since biologic discontinuation to failure (i.e., resumption of biologic DMARDs or disease flare) or study end. Thus, the follow-up duration we reported may have differed from the original study report because of the different framework of our review.

Second, the letter inquires about the thresholds for biologic discontinuation failure. Most studies focused on an absolute threshold either at remission (DAS28-ESR 2.6) or low disease activity (DAS28-ESR 3.2). Thus, we felt the relative threshold corresponding to this range of disease activity (i.e., change in DAS28-ESR > 0.6) was more relevant to our review. However, we agree that full description of the criteria is important.

Finally, we appreciate the suggestion of another study that could be included in this review. In a rapidly evolving scientific area with no standardized search terms, compiling a complete list of studies is challenging.

References

1. Yoshida K, Sung Y-K, Kavanaugh A, et al. Biologic discontinuation studies: a systematic review of methods. Ann Rheum Dis 2013 May 30.

2. Van der Maas A, Kievit W, van den Bemt BJF, et al. Down-titration and discontinuation of infliximab in rheumatoid arthritis patients with stable low disease activity and stable treatment: an observational cohort study. Ann Rheum Dis 2012 ;71(11):1849-54.

Conflict of Interest:

None declared

Dear Editor,

The work carried out by the authors on calcium and the cardiovascular risk is of primary importance. We thank the authors for questions and comments on the SEKOIA study, safety being a primary concern for us.
The number of emergent adverse events reported in SEKOIA study was similar in the 3 treatments groups: 85.8%, 87.9% and 86.5% in the SrRan 1g, SrRan 2g and placebo groups as well as the number of serious emergent adverse events: 17.0%, 16.5% and 17.4%, respectively. The adverse reactions considered by the investigators as related to the study drug were similar to those reported in osteoporotic patients: diarrhea (3.3%, 6.6% and 2.7%, respectively), nausea (2.0%, 2.7% and 1.8% respectively), and headache (1.6 in each SrRAn group and 0.7% in the placebo group). Regarding cutaneous safety, 16.3% of the patients reported a skin disorders in the SrRan 2g group compared to 12.4% and 12.2% in the SrRan 1g group and in the placebo group, respectively which is in line with the incidence of skin reaction in osteoporotic patients. No cases of DRESS, SJS or TEN were reported.
When specifically focusing on cardiac safety, a similar incidence of cardiac events was reported in the 3 treatment groups: 5.5%, 5.7% and 5.8%, respectively. Five cases of serious myocardial infarction were reported in the 2g group while 1 case was reported in the SrRan 1g group and in the placebo group. All events occurred in patients with major cardiovascular risk factors or comorbidities. All patients were suffering from hypertension from several years and all but one were overweight or obese. Among the strontium-ranelate treated patients, all presented additional risks factors such as hypercholesterolemia, diabetes, or previous or family myocardial ischaemia. Based on this small number of events and on the presence of numerous comorbidities, it seemed difficult at the time of the results to formally conclude to a higher risk of MI with SrRan. These results were then submitted to the European agency and were part of the routine benefit-risk assessment of the strontium ranelate. When applying at posteriori, the newly contraindication of the European agency to the SEKOIA population (i.e. when excluding patients with pathology at baseline corresponding to the newly defined contraindications: uncontrolled hypertension, thromboembolic events including deep vein thrombosis and pulmonary embolism, ischaemic heart disease, peripheral arterial disease and/or cerebrovascular disease), the number of MI become comparable between the 3 treatment groups: 1 event in the SrRan 1g group, 2 in the SrRan 2g group and 1 in the placebo group, demonstrating that the contraindications are effective to minimize the cardiac risk in this population of OA patients.

Conflict of Interest:

None declared

Response to the eLetter by Maksymowych WP, entitled ?Evidence in Support of the Validity of the TNF Brake Hypothesis"

Dear Editor,

with great interest we read the Letter by our colleague W.Maksymowych, entitled ?Evidence in Support of the Validity of the TNF Brake Hypothesis", which commented on our paper "Continuous increase in the rate of new bone formation in patients with ankylosing spondylitis (AS)" [1]. He argues that our interpretation of the results of our study is misleading in relation to the TNF brake hypothesis proposed by him. Since there have been several versions of this hypothesis [2-4] we have decided not to go into much detail and semantics but rather discuss the recent progress in the field. One of the major open questions related to the pathophysiology of AS is the nature of the link between inflammation and ankylosis. Several years ago we have reported that syndesmophyte formation after 2 years is more probable if, at baseline, spinal inflammatory lesions as detected by magnetic resonance imaging (MRI) using STIR sequences are present [5], and this finding was confirmed later [2]. However, it was clear from the beginning that this couldn't be the only influencing factor, since the majority of syndesmophytes appeared to have grown from vertebral edges without any bone marrow edema at baseline [2, 5], and that finding was also confirmed later [6]. In our most recent paper [7] on imaging results of the EASIC cohort related to the course of radiographic progression under TNF-blocker treatment over 5 years, MRI examinations of AS patients at baseline and after 2 years were included.
In this larger study, the regression of inflammation alone was not predictive of new bone formation. More importantly, another MRI finding has attracted increasing interest in the last years, and that is characterized by fat signals detected in T1 sequences [7, 8]. Indeed, both our groups have shown that there are different types of spinal lesions that can be differentiated by MRI techniques in AS and that may play different role in the important sequence of events from inflammation to new bone formation [4, 7]. There are basically four types of MRI findings in the spine that may precede syndesmophyte formation: (i) signs of inflammation without any other pathologic finding in parallel, (ii) signs of inflam?mation with a concomitant fat signal, (iii) a fat signal without signs of inflammation and (iv) no lesions at all [7].

There are several studies that suggest that anti-TNF therapy does not inhibit new bone formation in AS [9-11], but, in addition to the paper discussed here [1], another very recent one also reported some reduction of syndesmophyte formation [12]. However, both studies were clearly not performed with patients in an early stage of disease. From several long- term studies we know that there definitely is some progression in AS patients treated with TNF blockers and that there is no major difference between different dosages of the anti-TNF compound [13]. Factors that predict radiographic progression such as gender, prevalent syndesmophytes and smoking have been described [14] - and these should be controlled for in well-powered analyses with sufficient patient numbers. In the paper here under discussion we report on radiographic outcomes in a small number of AS patients treated for 8 years with infliximab [1]. The comparison to a historical cohort suggests that syndesmophyte formation may decelerate over longer periods of time in patients on anti-TNF therapy. Since patients were treated with TNF blockers almost continuously over a time period of 8 years in this study, the number of spinal inflammatory lesions is likely to be considerably reduced in most cases already after some months of treatment. However, in another paper we showed that about 20% of the spinal inflammation detected at baseline is still present after 2 years [15]. Furthermore, it is noteworthy that within this time period of 8 years there was a short period of treatment discontinuation of about 4-6 months after 3 years of continuous therapy [16]. This may have caused some worsening of spinal inflammation in that time period. However, this was probably again suppressed when anti-TNF therapy was re-administered but that was not investigated.
The basis of our concluding statement in that paper was indeed that the TNF-brake hypothesis implies that the use of anti-TNF agents may accelerate the development of new bone since TNF upregulates dickkopf-1 which, in turn, downregulates Wingless (Wnt) pathway signaling for new bone formation [17]. The background here is that proinflammatory cytokines such as TNF have been shown to stimulate expression of bone forming factors, such bone morphogenetic proteins (BMPs) and Wnt proteins. Thus, by antagonising TNF, dickkopf-1 expression will also decrease and allow signalling for new bone formation through the Wnt pathway.
A recent study on serum levels of dickkopf-1 and sclerostin in patients with AS has confirmed that this could indeed play a role [18]. The TNF brake hypothesis has initially been put forward to explain the observation that new syndesmophytes are more likely to develop at sites where inflammation has resolved (low TNF, low dickkopf-1, high Wnt) as opposed to sites of persistent inflammation (high TNF, high dickkopf-1, low Wnt) [13]. The finding that resolved but not persistent chronic inflammatory lesions (CILs) are associated with new syndesmophytes [3] led to the assumption that, once inflammation resolves, either spontaneously or through pharmacologic suppression of TNF, this allows signalling through Wnt to promote new bone formation. In that scenario, in an established inflammatory lesion, TNF may act primarily as a brake on new bone formation through Dickkopf-1. Resolution of the CIL by anti-TNF therapy may allow tissue repair to become manifest as bone, while persistence of the CIL may preclude syndesmophyte formation. [3]. However, on the other hand, our recent data suggest [7], much as proposed by W.Maksymowych in his letter, that early inflammatory lesions may well resolve without sequelae (new bone formation), if effective anti-TNF therapy is instituted and inflammation does resolve prior to activation of bone formation. Whether an 'average' AS patient does really present with a mixture of different inflammatory lesions has not yet been shown to date. It seems likely that, next to the activity of the disease, age and disease duration will also have an influence on that. In this regard, it may be more important to perform analyses on the level of vertebral edges rather than on the patients? level, to be able to also study intercorrelations between vertebral edges within patients. In conclusion, our understanding of the mechanisms responsible for the process of new bone formation in patients with axial SpA with and without treatment with TNF blockers is still limited. More prospective studies including the performance of MRIs in short intervals in combination with biomarkers, clinical findings and radiographs or even more sophisticated outcome parameters providing imaging results and new scoring methods with a high sensitivity to change will hopefully shed more light and give better answers to this complicated scenario of axial spondyloarthritis.

References

1. Baraliakos X, Haibel H, Listing J, et al. Continuous long-term anti-TNF therapy does not lead to an increase in the rate of new bone formation over 8 years in patients with ankylosing spondylitis. Ann Rheum Dis. 2013 Mar 27.

2. Maksymowych WP, Chiowchanwisawakit P, Clare T, et al. Inflammatory lesions of the spine on magnetic resonance imaging predict the development of new syndesmophytes in ankylosing spondylitis: evidence of a relationship between inflammation and new bone formation. Arthritis Rheum 2009; 60(1):93-102.

3. Pedersen SJ, Chiowchanwisawakit P, Lambert RG, et al. Resolution of inflammation following treatment of ankylosing spondylitis is associated with new bone formation. J Rheumatol 2011; 38(7):1349-1354.

4. Maksymowych WP, Morency N, Conner-Spady B, et al. Suppression of inflammation and effects on new bone formation in ankylosing spondylitis: evidence for a window of opportunity in disease modification. Ann Rheum Dis 2012 May 5.

5. Baraliakos X, Listing J, Rudwaleit M, et al. The relationship between inflammation and new bone formation in patients with ankylosing spondylitis. Arthritis Res Ther 2008;10(5):R104.

6. van der Heijde D, Machado P, Braun J, et al. MRI inflammation at the vertebral unit only marginally predicts new syndesmophyte formation: a multilevel analysis in patients with ankylosing spondylitis. Ann Rheum Dis 2012; 71(3):369-373.

7. Baraliakos X, Heldmann F, Callhoff J, et al. Which spinal lesions are associated with new bone formation in patients with ankylosing spondylitis treated with anti-TNF agents? - a long-term observational study using magnetic resonance imaging and conventional radiography. Ann Rheum Dis 2013; accepted for publication.

8. Chiowchanwisawakit P, Lambert RG, Conner-Spady B, et al. Focal fat lesions at vertebral corners on magnetic resonance imaging predict the development of new syndesmophytes in ankylosing spondylitis. Arthritis Rheum 2011; 63(8):2215-2225.

9. van der Heijde D, Burmester G, Melo-Gomes J, et al. Inhibition of radiographic progression with combination etanercept and methotrexate in patients with moderately active rheumatoid arthritis previously treated with monotherapy. Ann Rheum Dis 2009;68(7):1113-1118.

10. van der Heijde D, Landewe R, Baraliakos X, et al. Radiographic findings following two years of infliximab therapy in patients with ankylosing spondylitis. Arthritis Rheum 2008;58(10):3063-3070.

11. van der Heijde D, Salonen D, Weissman BN, et al. Assessment of radiographic progression in the spines of patients with ankylosing spondylitis treated with adalimumab for up to 2 years. Arthritis Res Ther 2009;11(4):R127.

12. Haroon N, Inman RD, Learch TJ, et al. The Impact of TNF-inhibitors on radiographic progression in Ankylosing Spondylitis. Arthritis Rheum 2013 Jul 1.

13. Braun J, Baraliakos X, Hermann KG, et al. The effect of two golimumab doses on radiographic progression in ankylosing spondylitis: results through 4 years of the GO-RAISE trial. Ann Rheum Dis 2013 May 3.

14. Poddubnyy D, Haibel H, Listing J, et al. Baseline radiographic damage, elevated acute-phase reactant levels, and cigarette smoking status predict spinal radiographic progression in early axial spondylarthritis. Arthritis Rheum 2012;64(5):1388-1398.

15. Baraliakos X, Listing J, Haibel H, et al. The significance of vertebral erosions associated with spinal inflammation in patients with ankylosing spondylitis as identified by magnetic resonance imaging and the changes observed after 2 years of anti-TNF therapy J Rheumatol 2013; accepted for publication.

16. Baraliakos X, Listing J, Brandt J, et al. Clinical response to discontinuation of anti-TNF therapy in patients with ankylosing spondylitis after 3 years of continuous treatment with infliximab. Arthritis Res Ther 2005;7(3):R439-444.

17. Diarra D, Stolina M, Polzer K, et al. Dickkopf-1 is a master regulator of joint remodeling. Nat Med 2007;13(2):156-163.

18. Appel H, Ruiz-Heiland G, Listing J, et al. Altered skeletal expression of sclerostin and its link to radiographic progression in ankylosing spondylitis. Arthritis Rheum 2009;60(11):3257-3262.

Conflict of Interest:

None declared

Dear Editor,

We were pleased to learn of Moulis et al.'s update on their experience with abatacept in three RP patients, particularly since their disease manifestations reflected the same manifestations suggested in our study to be of greatest interest for abatacept: chondritis and peripheral arthritis. Interestingly one patient achieved a complete, while another achieved only a partial, corticosteroid-sparing response, and the third patient could not taper corticosteroids at all. We note that the patient who achieved a complete response was given abatacept in combination with dapsone and methotrexate, while the other two appear to have been given abatacept as only monotherapy. In that context, it is notable that the one subject in our study who developed an excellent long-term response to abatacept was taking mycophenolate mofetil concomitantly (subject 002), while another with a less durable response was taking only prednisone without a concomitant immunosuppressant (subject 003) (1), suggesting that combination therapy involving abatacept and a conventional immunomodulatory drug may provide greater efficacy, as described with abatacept in rheumatoid arthritis [2].
At the same time, though, the two subjects in our study who developed severe relapses warranting study discontinuation were in fact already taking concomitant immunosuppressive drugs (subjects 001 and 004; leflunomide and mycophenolate mofetil, respectively) - but as previously recognized these patients had more severe disease, and/or might have benefitted more from an intravenous instead of subcutaneous abatacept dosing regimen as well as bridging therapy with corticosteroids. These observations further reinforce an importance relevance of costimulation to the pathogenesis of at least part of the chondritis and arthritis of RP, but at the same time remind us of our incomplete understanding of its disease and therapeutic context. We hope that ongoing observations and study of abatacept in this challenging disease will continue to provide further insights in this regard.

References

1. Peng SL, Rodriguez D. Abatacept in relapsing polychondritis. Ann Rheum Dis 2013;72:1427-9. 2. Buch MH, Vital EM, Emery P. Abatacept in the treatment of rheumatoid arthritis. Arthritis Res Ther 2008;10:S5.

Conflict of Interest:

None declared