Hydroxychloroquine is a good second-line treatment for adults with immune thrombocytopenia and positive antinuclear antibodies

Transcription

1 AJH Hydroxychloroquine is a good second-line treatment for adults with immune thrombocytopenia and positive antinuclear antibodies Mehdi Khellaf, 1 * Amèlie Chabrol, 1 Matthieu Mahevas, 1 Françoise Roudot-Thoraval, 2 Nicolas Limal, 1 Laetitia Languille, 1 Philippe Bierling, 3 Marc Michel, 1 and Bertrand Godeau 1 Treatment of patients with lupus-associated thrombocytopenia (SLE-ITP) is not standardized. We report data on efficacy and safety of hydroxychloroquine (HCQ) in this setting and in ITP patients with positive antinuclear antibodies (ANA) without definite SLE. Inclusion criteria were: definite diagnosis of ITP with a platelet count (PLT) < /L, ANA 1/160 on Hep2 cells with or without a definite diagnosis of SLE, and no sustained response to at least one previous treatment-line and treatment with HCQ. Response criteria were Complete Response (CR) for PLT /L and Response (R) for PLT /L and at least twice the initial value. Forty patients (32 females) with a mean age of years and PLT at ITP diagnosis of /L were analyzed. Twelve (30%) patients had a SLE-ITP, 28 patients had only positive ANA. All the patients failed to respond to oral prednisone with a median of two treatment-lines (1 5) before HCQ which was initially given in combination with another ITP treatment in 36 patients. Overall response rate was 60% (24/40) including 18 lasting CR and six lasting R maintained with a median follow-up of 64 months (6 146), in 3 = 4 of cases with only HCQ and no concomitant ITP treatment. The response rate (CR1R) was higher in the SLE group vs ANApositive group (83% vs 50%, P < 0.05). No patient stopped HCQ because of a side-effect. HCQ appears to be a safe and effective second line treatment for patients with SLE-ITP or ITP and high titer of ANA. This trial was registered at as # NCT Am. J. Hematol. 89: , VC 2013 Wiley Periodicals, Inc. Introduction Primary immune thrombocytopenia (ITP) is an acquired immune-mediated disorder characterized by isolated thrombocytopenia defined as peripheral blood platelet count < /L and the absence of an underlying cause of thrombocytopenia [1]. The pathophysiology is complex and the concepts of the mechanisms of thrombocytopenia have shifted from the traditional view of increased platelet destruction, mainly by splenic macrophages and mediated by platelet auto-antibodies, to associated impaired platelet production and T-cell mediated effects. Primary ITP remains a diagnosis of exclusion both from nonimmune causes of thrombocytopenia and immune thrombocytopenia that develops in the context of other disorders (secondary immune thrombocytopenia) [2]. Thrombocytopenia is one of the American College of Rheumatology (ACR) criteria for diagnosis of systemic lupus erythematosus (SLE) and occurs in 5 20% of cases of SLE [3,4]. Mechanisms of immune thrombocytopenia in SLE could differ from those observed in primary ITP and are probably multifactorial, including antiplatelet glycoprotein antibodies such as in ITP but also immune complexes, antiphospholipid antibodies, autoantibodies to the c-mpl receptor and megakaryocytes [2]. Antinuclear antibodies (ANA) can also be present in more than 20% of adults with primary ITP who do not fulfil the ACR criteria for SLE [5 7]. The significance of the presence of ANA in primary ITP is still debated. In this setting, only a few patients show overt SLE, but the presence of ANA seems to be a risk factor of chronic ITP and resistance to classical second-line treatment for ITP [8,9]. SLE-associated thrombocytopenia is generally not severe and does not require specific treatment [10]. No established standard therapy exists for symptomatic and marked thrombocytopenia for SLE-associated thrombocytopenia and follow-up data for these patients are lacking [1,11]. For patients showing failed response to first-line treatments proposed for primary ITP (i.e., steroids and intravenous immunoglobulin), several options [12] have been proposed for SLE-associated thrombocytopenia and include rituximab [13,14], splenectomy [15], danazol [16], or immunosuppressive drugs [17]. Hydroxychloroquine (HCQ), an antimalarial drug, is essential treatment for SLE [18]. So far, only one study suggested that HCQ could be effective for ITP associated with SLE [19]. We report the results of HCQ used to treat ITP in adults with ANA, with or without SLE. 1 Medecine Interne, Centre de reference des cytopenies autoimmunes de l adulte, Centre Hospitalier Universitaire (CHU) Henri-Mondor, Assistance Publique H^opitaux de Paris (APHP), Universite Paris Est Creteil (UPEC), Creteil; 2 Departement de sante publique, Universite Paris Est Creteil, Creteil, France; 3 Etablissement Français du Sang, CHU Henri-Mondor, Creteil, France Conflict of interest: Nothing to report. *Correspondence to: Mehdi Khellaf; Department of Internal Medicine, Centre Hospitalier Universitaire Henri-Mondor, Universite Paris Est Creteil (UPEC), 51, avenue du Marechal-de-Lattre-de-Tassigny, Creteil Cedex, France. mehdi.khellaf@hmn.aphp.fr Received for publication: 21 June 2013; Revised: 9 September 2013; Accepted: 3 October 2013 Am. J. Hematol. 89: , Published online: 28 October 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: /ajh VC 2013 Wiley Periodicals, Inc. 194 American Journal of Hematology, Vol. 89, No. 2, February 2014 doi: /ajh.23609

2 Hydroxychloroquine is a good second-line treatment for adults TABLE I. Characteristics of Patients at Diagnosis of Immune Thrombocytopenia (ITP) Total patients (n 5 40) SLE patients (n 5 12) Non-SLE patients (n 5 28) P value Age (years), mean (SD) 35 (17) 31 (18) 37 (17) ns Sex, no. (male : female) 8 : 32 1 : 11 7 : 21 ns Platelet count (310 9 /L), mean (SD) 14 (13) 11.7 (10) 13.7 (12) ns Hb level (g/dl), mean (SD) 11.7 (1.9) 11.7 (1.3) 11.7 (2.2) ns Associated haemolytic anemia, no ns Bleeding score, median (range) 4.7 (0 19) 4.3 (0 17) 4.9 (0 19) ns ANA titers, median 1/1280 1/1280 1/1280 ns Antiphospholipid antibodies, no. (%) 14 (35) 6 (50) 8 (28) ns LA 8 (20) 4 (33) 4 (14) ACL 14 (35) 6 (50) 8 (28) ACL IgG40 UI 11 (27) 5 (42) 6 (21) ACL IgG<40 UI 3 (7) 1 (8) 2 (7) Antiplatelet antibodies, no. (%) a 14/21 (67) 6/9 (67) 8/12 (67) ns Hb, hemoglobin; ANA, antinuclear antibodies; LA, lupus anticoagulant; ACL, anticardiolipid antibodies; SLE, systemic lupus erythematosus. a Antiplatelet antibodies were not searched in all patients. Patients and Methods We included data for every patient regularly followed for ITP between 1990 and 2008 in the national referral center for study of autoimmune cytopenias in adults, Creteil, France. Patients had marked thrombocytopenia (platelet count < /L on at least two consecutive occasions) associated with ANA biologically assessed (1/160) and received HCQ after no sustained response to at least one previous treatment-line for ITP. ITP was defined by standard criteria [20]. We exluded data for patients with disease other than SLE known to be associated with ITP (HIV or hepatitis C virus infection, lymphoproliferative disorders, thyroid disease, drug-induced ITP). Bleeding severity at the time of diagnosis was evaluated by a standardized bleeding score [21]. This score quantifies the hemorrhagic syndrome of ITP by adding points associated with different kinds of clinical bleeding signs. Unlike the original score, which includes age and bleeding to guide treatment, the tool we used was modified to measure only bleeding, and age was excluded. Positivity for antiplatelet antibodies was assessed by a monoclonal antibody immobilization of platelet antigen assay (MAIPA) [22] and for ANA by the immune fluorescent technique on Hep-2 cell substrate [23]. Only titers 1/160 were considered. Data for other clinical and biological manifestations of SLE were gathered, and patients were classified as having definite SLE (4 ACR criteria) or no SLE (< 4 ACR criteria). Data for SLE patients were included only if thrombocytopenia was the predominant manifestation. Lupus activity was assessed by the SLE Disease Activty Index (SLEDAI) score [24]. HCQ could be prescribed after failure of a previous first- or second-line treatment, including oral prednisone. We excluded data for patients receiving HCQ for an indication other than thrombocytopenia. We also excluded data for patients who received HCQ with steroids as first-line treatment because in case of response, we could not be sure that the response was due to HCQ. Outcome and response to HCQ were assessed retrospectively. According to international working group recommendations [20], response to treatment was evaluated by the platelet count peak during the first 8 weeks of treatment and was classified as complete response (CR) with count /L, response (R) with count 30 to /L and at least twice the initial value, or failure (F) with count < /L or less than twice the initial value or if rescue treatment was required. Statistical analysis. Data are described as number (%), mean (SD) or median (range). A two-sample Student t-test was used to compare quantitative variables and Fisher exact test for categorical variables. P < 0.05 was considered statistically significant. Results In all, 110 patients with ITP and ANA positivity 1/160 were regularly followed in our institution. Among them, 88 patients received HCQ as part of their treatment. We excluded data for 48 patients because of platelet count /L (n 5 22), HCQ introduced as first-line treatment with steroids (n 5 9), HCQ introduced for a reason other than thrombocytopenia (n 5 8), insufficient data (n 5 8), or human hepatitis C infection (n 5 1). The baseline characteristics of the 40 remaining patients are in Table I. Most were Caucasian (n 5 29, 72%), nine were Afro-Caribbean, and two were Asian. Mean platelet count at ITP diagnosis was /L. A total of 12 patients (30%) had definite SLE according to ACR criteria (4 criteria [n 5 8], 5 criteria [n 5 2], and 6 criteria [n 5 2]). Two had clinically associated antiphospholipid syndrome [25]. The mean SLEDAI score at thrombocytopenia diagnosis was 4.1 (range 1 9) and the median maximum SLEDAI score during the observation period was 7.1 (range 1 20). Thrombocytopenia preceded SLE diagnosis for three patients, was simultaneous with SLE diagnosis for 7, and followed SLE diagnosis for 2. These last two patients did not receive HCQ during the 5 years before thrombocytopenia. Overall, 28 patients had thrombocytopenia associated with ANA positivity but did not fulfill the ACR criteria for SLE: 13 with two criteria and 15 with three criteria. HCQ was started at a median of 4 months (range 1 month to 10 years) after the initial diagnosis of ITP. The median platelet count at this time was /L (range /L). All patients showed failed response to oral prednisone; before HCQ, they received a median of two treatments (range 1 5 treatments) (Table II). None of the second-line ITP treatments before HCQ (dapsone, danazol, azathioprine, or splenectomy) was effective. The dose of HCQ was 200 mg per day for three patients, 200 mg twice a day for 33 (83%), and 600 mg per day for 4. HCQ was given in combination with other treatments for 36 (90%), including oral prednisone (n 5 36) with a mean initial dose of 50 mg/day (range mg/day), dapsone (n 5 8), and danazol (n 5 5). The initial overall response rate within the first 8 weeks was 27% (11/40), with 4 CR and 7 R and median platelet count /L (range /L). The overall response rate was 60% (24/40), with 18 CR and 6 R (Fig. 1). The mean time to R was 3.07 months (range 2 6 months) and to CR was 5.6 months (range 2 10 months) (Fig. 2). The median duration of HCQ treatment was 5 years (range years). At a median follow-up of 64 months (range months), all responses (CR and R) were sustained, except for one patient showing relapse 6 months after withdrawal of HCQ but with a new complete response after reintroduction of HCQ. None of the other patients experienced loss of HCQ efficacy. At the end of followup, for the 24 patients (60%) with sustained response, 5 received no treatment, 12 received HCQ as the sole treatment and 7 received HCQ with dapsone (n 5 2), prednisone >10 mg/day (n 5 4) or prednisone <10 mg/day* (n 5 1). On univariate analysis, sustained response to HCQ for ITP was associated with SLE status (10/12, 83%) as compared with ANA positivity without definite SLE (14/28, 50%) (P < 0.05) (Table III). Blood HCQ concentration was checked in two non-responding patients and was below the therapeutic range for both. The overall safety of HCQ doi: /ajh American Journal of Hematology, Vol. 89, No. 2, February

3 Khellaf et al. RESEARCH ARTICLE TABLE II. Treatments received by patients with ITP before Hydroxychloroquine Treatment and Pattern of Transient Response Total patients (n 5 40) SLE patients (n 5 12) Non-SLE patients (n 5 28) Corticosteroids 40 (100%) 12 (100%) 28 (100%) Overall response (CR1R) 35 (87%) 11 (92%) 24 (86%) Intravenous immunoglobulins 15 (37%) 5 (42%) 10 (36%) Overall response (CR1R) 12 (80%) 4 (80%) 8 (80%) Danazol 5 (12%) 3 (25%) 2 (7%) Overall response (CR1R) 2 (40%) 1 (33%) 1 (50%) Dapsone 2 (5%) 1 (8%) 1 (4%) Overall response (CR1R) 1 (50%) 1 (100%) 0 (0%) Rituximab 0 (0%) 0 (0%) 0 (0%) Overall response (CR1R) NA NA NA Azathioprine 1 (2.5%) 1 (8.3%) 0 (0%) Overall response (CR1R) 0 (0%) 0 (0%) NA Splenectomy 2 (5%) 2 (17%) 0 (0%) Overall response (CR1R) 0 (0%) 0 (0%) NA CR, complete response (see text); NA, non applicable; R, response (see text). TABLE III. Factors Associated with Sustained Response (Complete Response Plus Response) to Hydroxychloroquine for Patients with ITP Sustained responders (n 5 24) Nonresponders (n 5 16) P value Age (years), mean (SD) 33 (17) 39 (17) 0.66 Sex (male) (%) 5 (21) 3 (19) ITP with SLE (%) 10 (42) 2 (12.5) Bleeding score at diagnosis, 2.9 (0 12) 4.7 (0 19) 0.24 median (range) No. of treatments before 1.56 (1 5) 1.75 (1 3) 0.46 HCQ, median (range) Duration of ITP before 23 (1 124) 16 (1 121) 0.83 HCQ, months, median (range) SLEDAI score at HCQ onset, 2.5 (1 9) 1.7 (0 5) 0.11 median (range) ANA titers 1/320e (%) 14 (58.3) 9 (56.3) Positive for anti-dna 5 (21) 4 (25) 0.76 antibodies (%) Positive for antiplatelet 9/16 (56) 5/10 (50) 0.89 antibodies (%) Positive for antiphospholipid antibodies (%) a 7 (29) 7 (44) Figure 1. Flow chart of 40 patients with immune thrombocytopenia (ITP) who were positive for antinuclear antibodies (ANA) and received hydroxychloroquine (HCQ) treatment. SLE 5 systemic lupus erythematosus, CR 5 complete response, R 5 response, the first- or second-line treatment was ineffective before HCQ treatment. Figure 2. Response to hydroxychloroquine (HCQ) (platelet count) for 24 patients with ITP. Data are mean 6SD. ANA, antinuclear antibody; DNA, desoxyribonucletodide; HCQ, hydroxychloroquine; ITP, immune thrombocytopenic purpura; SLE, Disease Acitivty Index (SLEDAI). a Positive antiphospholipid antibodies include lupus anticoagulant and/or anticardiolipin antibodies. was good, and no retinal deposits or toxic effects in eyes were observed. No patient stopped treatment because of a side effect. Discussion Treatment of SLE-associated thrombocytopenia is not well codified. Corticosteroids are usually considered first-line treatment, but the long-term response is disappointing and a relapse of the thrombocytopenia is usually observed when the steroid dose is tapered [17]. A previous study suggested that HCQ could be effective for severe SLEassociated thrombocytopenia [19]. Our results confirm a benefit of HCQ associated with corticosteroids for SLE-associated thrombocytopenia; 83% of our patients showed a sustained response, and prednisone could be rapidly tapered or stopped in most. For ITP patients with ANA positivity who do not fulfil the ACR criteria for SLE, a previous case report suggested that HCQ could be useful in this setting [26]. Our results confirm that HCQ could be used as a steroid-sparing agent because 50% of these patients showed a response with a safer 196 American Journal of Hematology, Vol. 89, No. 2, February 2014 doi: /ajh.23609

4 Hydroxychloroquine is a good second-line treatment for adults platelet count and a decrease of the concomitant treatment. The threshold titers of ANA are still debated [27] and the significance of low titers (1/100) is not clear. We selected only patients with ANA titers 1/160 because this seems to be the minimum level required to consider a potential underlying connective tissue disorder [28]. Most patients received a concomitant ITP treatment with HCQ, mainly prednisone. HCQ may not be the main reason leading to disease remission in patients with a sustained platelet response. Nevertheless, HCQ may have been efficacious because all patients showed previous failed response to prednisone alone and most of the responders were able to stop concomitant treatments without relapse. The need for an initial concomitant treatment in ITP patients could be explained by the delayed onset of action of HCQ because most of the patients showed a substantial increase in platelet count after 3 months. According to our previous experience suggesting that HCQ could be a good option for SLE-associated thrombocytopenia [19], our policy is now to introduce HCQ soon after the diagnosis of ITP. HCQ was indeed started after a median of only 4 months after the initial diagnosis, and most of our patients previously received only steroids and intravenous immunoglobulin, which could explain the good outcome we observed. HCQ could be less effective in refractory SLE patients who failed to respond to immunosuppressive agents or splenectomy. The response rate during the first 8 weeks following the start of HCQ was only 27%, with maximal response after more than 5 months of treatment. This important point should be taken into account by patients and physicians who should not stop the treatment too early in case of a disappointing initial response. The immunomodulatory mechanism of action of HCQ in SLE is not well known [29], but HCQ was found to prevent SLE flare in a controlled study [18]. The use of HCQ for ITP patients positive for ANA with or without a definite SLE diagnosis represents an interesting alternative to other second-line ITP treatments such as rituximab, immunosuppressive drugs, danazol or splenectomy [1,11]. Uncontrolled study of a low number of patients suggested that rituximab might be effective in ITP associated with SLE [14], but some concerns about long-term cumulative toxicity are emerging, especially in combination with other immunosuppressive therapy. Cases of fatal, progressive, multifocal leukoencephalopathy have been described with rituximab for SLE that may be linked to an immunosuppressive effect by the disease itself [30,31]. The place of splenectomy in ITP with SLE is also controversial because it has been linked to the development of new autoimmune phenomenon (mainly vasculitis) and alterations in the clinical course of patients with prior autoimmune disease by a progressive redistribution of memory B cells that may influence autoimmune disease activity [32 34]. Another consideration with HCQ is safety. We observed no side effects. Moreover, SLE mainly affects childbearing women. Women with ITP associated with SLE who contemplate pregnancy may opt for HCQ because most options for management of maternal platelet count, including rituximab, may not be safe for the fetus. HCQ appears to be safe for pregnant women, and breast feeding seems possible [35]. One of the issues with HCQ treatment is patient compliance. In our study, only a few patients had undergone HCQ blood testing, which could be an effective tool for patients not responding to HCQ, for detecting non-compliance or HCQ low serum level, shown in SLE to be correlated with greater disease activity [36]. To conclude, although retrospective and concerning a limited number of patients, our study strongly suggests that HCQ is a safe and effective treatment for ITP with SLE in patients showing failed longlasting response to prednisone alone or with second-line treatment. This relatively inexpensive and well-tolerated treatment should be systematically considered in this setting regardless of SLE activity. Moreover, HCQ may be helpful as a corticosteroid-sparing strategy for patients with chronic ITP and ANA positivity without definite SLE. Author contributions MK, BG, AC, MA, and NL performed the research. FRT, BG, AC, and MK designed the study and MK, AC, BG, MM, and PB wrote the article. References 1. Provan D, Stasi R, Newland AC, et al. International consensus report on the investigation and management of primary immune thrombocytopenia. Blood 2010;115: Cines DB, Blanchette VS. Immune thrombocytopenic purpura. N Engl J Med 2002;346: Tan EM, Cohen AS, Fries JF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25: Ward MM, Pyun E, Studenski S. Mortality risks associated with specific clinical manifestations of systemic lupus erythematosus. Arch Intern Med 1996;156: Kurata Y, Miyagawa S, Kosugi S, et al. Hightiter antinuclear antibodies, anti-ssa/ro antibodies and anti-nuclear RNP antibodies in patients with idiopathic thrombocytopenic purpura. Thromb Haemost 1994;71: Altintas A, Ozel A, Okur N, et al. Prevalence and clinical significance of elevated antinuclear antibody test in children and adult patients with idiopathic thrombocytopenic purpura. J Thromb Thrombolysis 2007;24: Vantelon JM, Godeau B, Andre C, et al. Screening for autoimmune markers is unnecessary during follow-up of adults with autoimmune thrombocytopenic purpura and no autoimmune markers at onset. Thromb Haemost 2000;83: Leung AY, Chim CS, Kwong YL, et al. Clinicopathologic and prognostic features of chronic idiopathic thrombocytopenic purpura in adult Chinese patients: An analysis of 220 cases. Ann Hematol 2001;80: Pratt EL, Tarantino MD, Wagner D, et al. Prevalence of elevated antithyroid antibodies and antinuclear antibodies in children with immune thrombocytopenic purpura. Am J Hematol 2005;79: Ziakas PD, Poulou LS, Giannouli S, et al. Thrombocytopenia in lupus: Baseline C3 as an independent risk factor for relapse. Ann Rheum Dis 2007;66: Neunert C, Lim W, Crowther M, et al. The American Society of Hematology 2011 evidencebased practice guideline for immune thrombocytopenia. Blood 2011;117: Hepburn AL, Narat S, Mason JC. The management of peripheral blood cytopenias in systemic lupus erythematosus. Rheumatology (Oxford) 2010;49: Kneitz C, Wilhelm M, Tony HP. Effective B cell depletion with rituximab in the treatment of autoimmune diseases. Immunobiology 2002;206: Terrier B, Amoura Z, Ravaud P, et al. Safety and efficacy of rituximab in systemic lupus erythematosus: Results from 136 patients from the French AutoImmunity and Rituximab registry. Arthritis Rheum 2010;62: Zhou J, Wu Z, Zhou Z, et al. Efficacy and safety of laparoscopic splenectomy in thrombocytopenia secondary to systemic lupus erythematosus. Clin Rheumatol 2013;32: Letchumanan P, Thumboo J. Danazol in the treatment of systemic lupus erythematosus: A qualitative systematic review. Semin Arthritis Rheum 2011;40: Newman K, Owlia MB, El-Hemaidi I, et al. Management of immune cytopenias in patients with systemic lupus erythematosus Old and new. Autoimmun Rev 2013;12: The Canadian Hydroxychloroquine Study Group. A randomized study of the effect of withdrawing hydroxychloroquine sulfate in systemic lupus erythematosus. N Engl J Med 1991;324: Arnal C, Piette JC, Leone J, et al. Treatment of severe immune thrombocytopenia associated with systemic lupus erythematosus: 59 cases. J Rheumatol 2002;29: Rodeghiero F, Stasi R, Gernsheimer T, et al. Standardization of terminology, definitions and outcome criteria in immune thrombocytopenic purpura of adults and children: Report from an international working group. Blood 2009;113: Khellaf M, Michel M, Schaeffer A, et al. Assessment of a therapeutic strategy for adults with severe autoimmune thrombocytopenic purpura based on a bleeding score rather than platelet count. Haematologica 2005;90: Kiefel V, Santoso S, Weisheit M, et al. Monoclonal antibody Specific immobilization of platelet antigens (MAIPA): A new tool for the identification of plateletreactive antibodies. Blood 1987;70: Bayer PM, Bauerfeind S, Bienvenu J, et al. Multicenter evaluation study on a new HEp2 ANA screening enzyme immune assay. J Autoimmun 1999;13: Bombardier C, Gladman DD, Urowitz MB, et al. Derivation of the SLEDAI. A disease activity index for lupus patients. The Committee on Prognosis Studies in SLE. Arthritis Rheum 1992;35: doi: /ajh American Journal of Hematology, Vol. 89, No. 2, February

5 Khellaf et al. RESEARCH ARTICLE 25. Miyakis S, Lockshin MD, Atsumi T, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4: Bockow B, Kaplan TB. Refractory immune thrombocytopenia successfully treated with high-dose vitamin D supplementation and hydroxychloroquine: Two case reports. J Med Case Rep 2013; 7: Abeles AM, Abeles M. The clinical utility of a positive antinuclear antibody test result. Am J Med 2013;126: Adams BB, Mutasim DF. The diagnostic value of anti-nuclear antibody testing. Int J Dermatol 2000;39: van den Borne BE, Dijkmans BA, de Rooij HH, et al. Chloroquine and hydroxychloroquine equally affect tumor necrosis factor-alpha, interleukin 6, and interferon-gamma production by peripheral blood mononuclear cells. J Rheumatol 1997;24: Molloy ES, Calabrese LH. Progressive multifocal leukoencephalopathy: A national estimate of frequency in systemic lupus erythematosus and other rheumatic diseases. Arthritis Rheum 2009; 60: Molloy ES, Calabrese LH. Progressive multifocal leukoencephalopathy associated with immunosuppressive therapy in rheumatic diseases: Evolving role of biologic therapies. Arthritis Rheum 2012;64: Hall S, McCormick JL, Jr., Greipp PR, et al. Splenectomy does not cure the thrombocytopenia of systemic lupus erythematosus. Ann Intern Med 1985;102: Patel S, Kramer N, Rosenstein ED. Evolving connective tissue disease influenced by splenectomy: Beneath the sword of Dameshek. J Clin Rheumatol 2010;16: Rivero SJ, Alger M, Alarcon-Segovia D. Splenectomy for hemocytopenia in systemic lupus erythematosus. A controlled appraisal. Arch Intern Med 1979;139: Costedoat-Chalumeau N, Amoura Z, Huong DL, et al. Safety of hydroxychloroquine in pregnant patients with connective tissue diseases. Review of the literature. Autoimmun Rev 2005;4: Costedoat-Chalumeau N, Amoura Z, Hulot JS, et al. Low blood concentration of hydroxychloroquine is a marker for and predictor of disease exacerbations in patients with systemic lupus erythematosus. Arthritis Rheum 2006;54: American Journal of Hematology, Vol. 89, No. 2, February 2014 doi: /ajh.23609