Current treatments in macrophage activation syndrome - METAPHOR projec
Current treatments in macrophage activation syndrome - METAPHOR projec
Current treatments in macrophage activation syndrome - METAPHOR projec
@rashmi3 months ago
Current treatment in macrophage activation syndrome worldwide: a systematic literature review to inform the METAPHOR project
Francesco Baldo 1,2 *, Remco G.A. Erkens 3 , Mao Mizuta 4 , Greta Rogani 3 , Federica Lucioni 1 , Claudia Bracaglia 5 , Dirk Foell 6 , Marco Gattorno 7 , Marija Jelusic 8 , Jordi Anton 9 , Paul Brogan 10,11 , Scott Canna 12 , Shanmuganathan Chandrakasan 13 , Randy Q. Cron 14 , Fabrizio De Benedetti 5 , Alexei Grom 15 , Merav Heshin-Bekenstein 16 , AnnaCarin Horne 17,18 , Raju Khubchandani 19 , Seza Ozen 20 , Pierre Quartier 21,22 , Angelo Ravelli 23 , Masaki Shimizu 24 , Grant Schulert 15 , Christiaan Scott 25 , Rashmi Sinha 26 , Nicolino Ruperto 27 , Joost F Swart 3 , Sebastiaan Vastert 3 and Francesca Minoia 1 , on behalf of the PReS MAS/sJIA Working Party and Paediatric Rheumatology International Trial Organization.
1 Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy; 2 ASST Gaetano Pini, Milan, Italy; 3 Department of Pediatric Rheumatology and Immunology, University Medical Center Utrecht, the Netherlands; 4 Department of Pediatric Rheumatology, Hyogo Prefectural Kobe Children's Hospital, Kobe, Japan ; 5 Division of Rheumatology, IRCCS Ospedale Pediatrico Bambino Gesù, Rome, Italy; 6 University Hospital Muenster, Muenster, Germany; 7 IRCCS Istituto Giannina Gaslini, Reumatologia e Malattie Autoinfiammatorie, Genoa, Italy; 8 University Hospital Centre Zagreb, University School of Medicine, Zagreb, Croatia; 9 Hospital Sant Joan de Déu, Universitat de Barcelona, Barcelona, Spain; 10 Great Ormond Street Hospital for Children, London, UK; 11 University College London Institute of Child Health, London, UK; 12 Children's Hospital of Philadelphia, Philadelphia, PA, USA; 13 Aflac Cancer and Blood Disorders Center Children's Healthcare of Atlanta, Emory University School of Medicine, Atlanta, GA, USA; 14 University of Alabama at Birmingham, Birmingham, AL, USA; 15 Cincinnati Children's Hospital, Cincinnati, OH, USA; 16 Dana Dwek Children's Hospital, Tel Aviv Medical Center, Tel Aviv University, Tel Aviv, Israel; 17 Department of
1
© The Author(s) 2024. Published by Oxford University Press on behalf of the British Society for Rheumatology.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium,
provided the original work is properly cited.
- · High-dose GCs together with IL-1 and IFN ï§ inhibitors have shown efficacy in sJIA-associated MAS.
- · Current level of evidence on MAS treatment, especially in condition other than sJIA, is still poor.
- · MAS treatment is still extremely variable, with potential significant discrepancies across different centres and countries.
- 1. Ravelli A, Minoia F, Davì S, Horne A, Bovis F, Pistorio A, et al. 2016 Classification Criteria for Macrophage Activation Syndrome Complicating Systemic Juvenile Idiopathic Arthritis: A European League Against Rheumatism/American College of Rheumatology/Paediatric Rheumatology International Trials Organisation Collaborative Initiative. Ann Rheum Dis. 2016;75:481-9.
- 2. Minoia F, Bovis F, Davì S, Horne A, Fischbach M, Frosch M, et al. Development and initial validation of the MS score for diagnosis of macrophage activation syndrome in systemic juvenile idiopathic arthritis. Ann Rheum Dis. 2019t;78:1357-1362.
- 3. Parodi A, Davì S, Pringe AB, Pistorio A, Ruperto N, Magni-Manzoni S, et al. Macrophage activation syndrome in juvenile systemic lupus erythematosus: a multinational multicenter study of thirty-eight patients. Arthritis Rheum. 2009;60:3388-99.
- 4. Gerstein M, Borgia RE, Dominguez D, Feldman BM, Liao F, Levy DM, et al. Predicting Macrophage Activation Syndrome in Childhood-onset Systemic Lupus Erythematosus Patients at Diagnosis. J Rheumatol. 2021;48:1450-1457
- 5. Fardet L, Galicier L, Lambotte O, Marzac C, Aumont C, Chahwan D, et al. Development and validation of the HScore, a score for the diagnosis of reactive hemophagocytic syndrome. Arthritis Rheumatol. 2014;66:2613-20
- 6. Eloseily EMA, Minoia F, Crayne CB, Beukelman T, Ravelli A, Cron RQ. Ferritin to Erythrocyte Sedimentation Rate Ratio: Simple Measure to Identify Macrophage Activation Syndrome in Systemic Juvenile Idiopathic Arthritis. ACR Open Rheumatol. 2019;1:345-349
- 7. Bracaglia C, de Graaf K, Pires Marafon D, Guilhot F, Ferlin W, Prencipe G, et al. Elevated circulating levels of interferon-γ and interferon-γ-induced chemokines characterise
- 8. Weiss ES, Girard-Guyonvarc'h C, Holzinger D, de Jesus AA, Tariq Z, Picarsic J et al. Interleukin-18 diagnostically distinguishes and pathogenically promotes human and murine macrophage activation syndrome. Blood. 2018;131:1442-1455.
- 9. Kessel C, Fall N, Grom A, de Jager W, Vastert S, Strippoli R, et al. Definition and validation of serum biomarkers for optimal differentiation of hyperferritinaemic cytokine storm conditions in children: a retrospective cohort study. Lancet Rheumatol. 2021;3:e563-e573
- 10. Minoia F, Davì S, Horne A, Bovis F, Demirkaya E, Akikusa J, et al. Dissecting the heterogeneity of macrophage activation syndrome complicating systemic juvenile idiopathic arthritis. J Rheumatol. 2015;42:994-1001
- 11. Shakoory B, Geerlinks A, Wilejto M, Kernan K, Hines M, Romano M, et al. The 2022 EULAR/ACR points to consider at the early stages of diagnosis and management of suspected haemophagocytic lymphohistiocytosis/macrophage activation syndrome (HLH/MAS). Ann Rheum Dis. 2023;82:1271-1285
- 12. van der Heijde D, Aletaha D, Carmona L, Edwards CJ, Kvien TK, Kouloumas M, et al. 2014 Update of the EULAR standardised operating procedures for EULAR-endorsed recommendations. Ann Rheum Dis. 2015;74:8-13
- 13. Joanna Briggs Institute Critical Appraisal Tools [website]. Available from https://jbi.global/critical-appraisal-tools
- 14. Wei A, Ma H, Li Z, Zhang L, Zhang Q, Wang D, et al. Short-term effectiveness of ruxolitinib in the treatment of recurrent or refractory hemophagocytic lymphohistiocytosis in children. Int J Hematol. 2020;112:568-576
Pediatrics, Karolinska University Hospital, Stockholm, Sweden; 18 Department of Women's and Children's Health, Karolinska Institute, Stockholm, Sweden Karolinska Institute, Solna, Sweden; 19 SRCC Childrens Hospital, Mumbai, India; 20 Department of Pediatrics, Hacettepe University, Ankara, Turkey; 21 Université Paris-Cité, Paris, France; 22 RAISE Reference Centre, Pediatric Immunology-Hematology and Rheumatology Unit, Necker-Enfants Malades Hospital, Paris, France; 23 IRCCS Istituto Giannina Gaslini, Direzione Scientifica, Genoa, Italy; 24 Department of Pediatrics and Developmental Biology, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, Tokyo, Japan; 25 University of Ottawa, Ottawa, Canada; 26 Systemic JIA Foundation, Cincinnati, OH, USA; 27 IRCCS Istituto Giannina Gaslini, Gaslini Trial Centre/Servizio Sperimentazioni Cliniche Pediatriche, PRINTO, Genoa, Italy.
Corresponding author : Francesca Minoia, MD; Pediatric Immuno-Rheumatology Unit - Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, via della Commenda 9, 20122 Milan, Italy E-mail: francesca.minoia@policlinico.mi.it
Short running title : Current treatment of MAS
ABSTRACT
Objective . To assess current treatment in macrophage activation syndrome (MAS) worldwide and to highlight any areas of major heterogeneity of practice.
Methods. A systematic literature search was performed in both Embase and PubMed databases. Paper screening was done by two independent teams based on agreed criteria. Data extraction was standardized following the PICO framework. A panel of experts assessed paper validity, using the Joanna Briggs Institute appraisal tools and category of evidence (CoE) according to EULAR procedure.
Results. Fifty-seven papers were finally included (80% retrospective case-series), describing 1148 patients with MAS: 889 systemic juvenile idiopathic arthritis (sJIA), 137 systemic lupus erythematosus (SLE), 69 Kawasaki disease (KD) and 53 other rheumatologic conditions. Fourteen and 11 studies specified data on MAS associated to SLE and KD, respectively. All papers mentioned glucocorticoids (GCs), mostly methylprednisolone and prednisolone (90%); dexamethasone was used in 7% of patients. Ciclosporin was reported in a wide range of patients according to different cohorts. Anakinra was used in 179 MAS patients, with a favourable outcome in 83% of sJIA-MAS. Etoposide was described by 11 studies, mainly as part of HLH-94/04 protocol. Emapalumab was the only medication tested in a clinical trial in 14 sJIA-MAS, with 93% of MAS remission. Ruxolitinib was the most reported JAK-inhibitor in MAS.
Conclusion. High-dose GCs together with IL-1 and IFN ï§ inhibitors have shown efficacy in MAS, especially in sJIA-associated MAS. However, global level of evidence on MAS treatment, especially in other conditions, is still poor and requires standardized studies to be confirmed.
Keywords : macrophage activation syndrome, haemophagocytic syndromes, haemophagocytic lymphohistiocytosis, treatment
Key messages:
INTRODUCTION
Macrophage activation syndrome (MAS) is an hyperinflammatory life-threatening condition, part of the wide spectrum of hemophagocytic lymphohistiocytosis (HLH). The term MAS refers to a secondary form of HLH that complicates the course of rheumatological conditions. MAS is characterized by a marked hyperferritinemia, cytopenia, liver insufficiency with coagulopathy, neurological manifestations, and a high risk of rapid progression to multiorgan failure. Despite great improvement in diagnosis and management 1-9 , MAS still represents a major challenge in clinical practice.
MAS treatment remains largely empiric and based on expert consensus. Although promising data are emerging, results from large cohorts and standardized trials are still required for most medications used to treat MAS. Multinational data on sJIA-associated MAS highlighted several disparities in its management in relation to geographic location of the treating centre and subspecialty of the caring physicians 10 . Recently, the first international recommendations for the early-stage management of HLH/MAS have been published 11 . Despite their milestone relevance, these guidelines focus on the initial management of the spectrum of haemophagocytic syndromes, and did not specifically address the treatment of MAS. Furthermore, there is a particular lack of evidence on therapeutic approach to MAS associated with rheumatologic conditions other than sJIA. It is thus conceivable that a wide heterogeneity in the management of MAS exists, due to differences in treatment strategies, access to medications and involvement of different specialists.
The METAPHOR project was conceived to provide an overview of current real-life therapeutic approaches to MAS in different clinical settings worldwide by means of a web-survey involving the paediatric rheumatology community part of the Pediatric Rheumatology European Society (PReS) and the Pediatric Rheumatology International Trial Organization (PRINTO) and the
paediatric haematologists from the Histiocyte Society. In this context, a systematic literature review (SLR) to explore available data on MAS treatment was performed.
METHODS
The SLR was conducted following the EULAR standardised operating procedures 12 . A multinational panel of experts in the field of MAS was involved. The PICO (Patient-InterventionComparison-Outcome) framework was adopted to structure the research (see Supplementary Data S1 and Supplementary Table S1, available at Rheumatology online ). Acknowledging the concomitant international effort of the EULAR/PRES task force for sJIA and adult-onset Still disease, which includes a SLR on the treatment of sJIA-associated MAS (De Matteis et al, submitted ), we decided to particularly address MAS in conditions other than sJIA. On June 30 th , 2022 the literature search was performed both in PubMed and Embase databases, and then updated on June 30 th , 2023. Search strings were designed under the supervision of an expert librarian (see Supplementary text ). Main inclusion criteria were: original articles, English language, studies reporting data regarding treatment of patients with MAS, population's age <18-years-old and papers with more than 3 cases reported. Exclusion criteria are detailed in Figure 1 . In light of the scarcity of available data on specific conditions or medication, and only after discussion in our core team, we did exceptionally include a case-report, if this was deemed relevant for the analysis. Papers were checked for duplicates and then screened, using Rayyan software (Cambridge, USA). A first title and abstract screening was performed, and then selected papers were evaluated
through a full-text read.
To establish the quality and the category of evidence of included papers, two members of the Expert Panel evaluated each manuscript independently. The Joanna Briggs Institute critical appraisal tools were used to assess the validity score 13 , identifying three validity levels (low-
moderate-high), and the category of evidence (CoE) was attributed as per EULAR standardized operating procedures 12 .
RESULTS
A total of 6588 papers were identified through the first search. After the deletion of duplicates and the title/abstract selection, 560 articles underwent full text screening and finally 57 studies fulfilled the eligibility criteria ( Figure 1 ). Twenty-three papers reported sJIA cohorts, 4 SLE cohorts, 8 KD cohorts, while in 22 studies the described population was mixed. Thirty-six were single-centre retrospective case series, 10 multicentre retrospective case series, 2 single-centre retrospective cohorts, 1 multicentre prospective cohort, only 1 was a standardized single arm open label clinical trial; 7 case reports were included for the relevancy of the medication or the condition reported. Three additional studies about JAK-inhibitors (JAK-i) 14-16 were considered, despite reporting data about mixed HLH cohorts; data from those studies only contributed to the JAK-i evidence review. Most papers (84%) were found to have low or moderate validity, and almost all (96%) were classified with a CoE of 3 or 4. Supplementary Table S2, available at Rheumatology online, reports all the information available on papers included in the SLR. Data from a total of 1148 patients with MAS were finally evaluated: 889 sJIA, 137 SLE, 69 KD and 53 other rheumatological conditions, including 8 juvenile dermatomyositis, 7 mixed connective tissue disease, 6 vasculitis, 2 antiphospholipid syndrome, 2 spondyloarthritis, 2 undefined connective tissue disease, 2 polyarticular JIA, 1 undefined arthritis, 1 rheumatic fever, 1 enthesitisarthritis (ERA), 1 Kikuchi disease, 1 Sjogren disease, 1 sarcoidosis, 1 cryopyrin associated periodic syndrome, 1 mevalonate-kinase deficiency (MKD), 1 Crohn disease, and 15 unspecified rheumatic
disorders.
Glucocorticoids
All studies mentioned the use of GCs and information were available for 1054 MAS patients (829 sJIA, 91 SLE, 66 KD, 68 other rheumatologic conditions). Among the 300 patients in which this information was assessable, most patients (86%, 258/300) received GCs as a co-medication, while 42/300 (14%) were successfully treated with GCs as monotherapy. Methylprednisolone (MPN) or prednisolone were the mostly used GC (90%), followed by dexamethasone (DEX, 7%). DEX was used in 15%, 10%, and 6% of patients with MAS in the contect of KD, SLE, and sJIA, respectively.
MPN dose ranged from 2 mg/kg/day to 30 mg/kg/day, with high-dose MPN pulses (10-30 mg/kg/day) reported in almost 60% of studies. Interestingly, a tapering regimen of MPN pulses was suggested by Loganathan et al. for severe MAS complicating sJIA in a resource limited setting 17 . DEX dose ranged from 4mg/m 2 /day to 10-15 mg/m 2 /day. Two Japanese studies ,18,19 reported the successful use of dexamethasone palmitate (DEX-P), a liposomal incorporated formulation, in 24 sJIA-MAS patients (17 naïve and 7 refractory to MPN/prednisolone +/- CsA).
Ciclosporin
Fifty studies mentioned the use of CsA in 611 MAS patients (483 sJIA, 34 SLE, 10 KD, 84 other rheumatological diseases). In the largest multinational cohort of sJIA-MAS 20 , CsA was the medication most frequently prescribed besides GCs (61% of patients). Only 10 studies reported details about the route and the dose of administration: CsA was given intravenously (iv) in 29 patients and orally in 12, with dose ranging from 0.8 to 8 mg/kg/day. Trough levels were mentioned only in 3 studies 21-23 and ranged between 78 and 480 ng/ml.
Globally, outcome in patients treated with CsA was assessable for 186 patients (138 sJIA, 9 SLE, 8 KD, 31 other rheumatic diseases): in 6 patients (3%) a poor outcome (4 deaths, 2 severe neurological adverse events) was reported. Posterior reversible encephalopathy syndrome (PRES) was mentioned in 1 sJIA-MAS patient, who was receiving co-treatment with GCs, IVIG and
etoposide 24 . Five sJIA-MAS patients were successfully treated with CsA without modification of the background GC therapy 22,25 .
Etoposide
Details on etoposide were available from 11 studies, for a total of 120 patients (78 SJIA, 14 SLE, 14 KD, 14 other rheumatic diseases); outcome data were available for 17 sJIA, 7 SLE, 14 KD and 4 other rheumatic diseases. Seven patients (17%) died. Neutropenia was the main adverse event reported; in 3 patients, severe bone marrow suppression with sepsis was reported.
Dose of etoposide ranged from 50 to 150 mg/m 2 weekly-biweekly. Of note, two studies reported the use of low dose etoposide (50-100 mg/m 2 /week for 4-11 weeks) 26,27 , in 7 patients with MAS (5 sJIA and 2 SLE). All sJIA patients were refractory to high-dose GCs and CsA, 3/5 also to anakinra (2.7-15 mg/kg/day), and all achieved MAS remission after etoposide. The two patients with SLE had failed oral prednisone: both survived with MAS remission, but one developed longterm CNS sequela.
Anakinra
A total of 179 patients received anakinra for MAS (147 sJIA, 12 SLE, 1 KD, 19 other rheumatologic disorders), reported in 19 studies all published after 2011. Outcome data were available for 82 sJIA, 10 SLE, 1 KD, 12 other rheumatological conditions, and for 3 sHLH treated with iv anakinra continuous infusion ( Table 1 ). A complete response was reported in 68 patients with sJIA-MAS (83%); 8 patients presented an incomplete (10%) and 3 (4%) a lack of response to anakinra, 2 had a recurrency of MAS, and 2 (2%) died. Patients with SLE-MAS treated with anakinra had a favourable outcome in 6/10 (60%), with 4 reported deaths (40%).
In the included studies, anakinra was used with a wide dosing range (2 - 48 mg/kg/day). The highest dose was used as continuous iv infusion in 2 patients: one patient with MAS secondary
to SLE/MCTD was treated for 72 hours without any other medication, but eventually died from multiorgan failure 28 . The second patient was a 9 year-old girl with severe sHLH and neurological involvement without a known trigger, refractory to MPN pulss and IVIG and anakinra (12 mg/kg/day); given her worsening conditions, anakinra was steeply increased to 2 mg/kg/hr (48 mg/kg/day) with a positive outcome 29 . The use of high-dose anakinra (at least 5 mg/kg/day) was specified in 6 studies 26,28-32 for 27 patients, and 93% of them were reported after 2020.
Concomitant medications in patients treated with anakinra were assessable only for 67 episodes of MAS. High-dose anakinra was reported mainly together with GCs and CsA (85% and 37%, respectively), followed by etoposide (15%). Anakinra was used as monotherapy in 6 patients (5 sJIA and 1 SLE/MTCD) 28 : all patients with sJIA achieved MAS remission (dosing range of 2.9 - 6.2 mg/kg/day), while the patient with SLE/MTCD died despite being treated with high-doses (48 mg/kg/day iv). Data on MAS patients treated with anakinra as single medication on the background of GCs were available from two studies 28,30 reporting 15 episodes of MAS: all the 10 episodes with assessable outcome data achieved MAS remission.
Emapalumab
The first and only clinical trial in MAS assessed the role of emapalumab (anti-IFN ï§ monoclonal antibody) on sJIA-associated MAS refractory to high-dose GCs 31 . In this single-arm, open label trial, 14 sJIA-MAS were included: 8 were refractory also to CsA and 7 to anakinra. By week 8, MAS remission was achieved in 13/14 patients (93%), with a median time to remission of 25 days. In all patients, emapalumab led to a rapid regression of all MAS parameters and to a significant steroid-sparing effect. No deaths or serious adverse events related to emapalumab were reported. Viral infection/seropositivity was the most frequent side effect (mainly CMV; of note, all patients received acyclovir prophylaxis). Interestingly, the combination of emapalumab with anakinra (up to 4 mg/kg/day) seemed to reduce the occurrence of sJIA flare without
increasing serious events and infection rate. In the trial 1 patient received emapalumab together with high-dose anakinra (7.5 mg/kg/day), with good tolerability and without the mention of specific adverse events.
Other biologics
The use of other biologics in the treatment of MAS was reported in 22 studies: canakinumab and tocilizumab were the most commonly reported biologic agents for sJIA-MAS, while infliximab was mainly used in patients with KD-MAS (7 patients treated with a dose range 310 mg/kg/day and a positive outcome).
Thirty-five patients 33-37 received tocilizumab, and in 26 of them outcome data were available: 22 patients (85%) had MAS remission, in 1 tocilizumab was discontinued for lack of response (4%), and in 3 (12%) for an allergic reaction. Of note, in the two main cohorts of sJIAMAS patients successfully treated with TCZ 33,36 , none of them previously received an IL-1 inhibitor.
Canakinumab was used in 16 patients 37-40 , with a positive response in 14 of them (88%). In particular, Kostik et al. 37 described 8 sJIA-MAS patients all treated with canakinumab: 7 achieved MAS remission, and 1 required the addition of tofacitinib to control MAS recurrency. In 3 patients, canakinumab was successfully used as 1 st line biologic treatment. Interestingly, 3 patients developed severe MAS despite canakinumab standard treatment, and responded to an increase of canakinumab dose, up to 12 mg/kg.
In a cohort of MAS associated to thrombotic microangiopathy (TMA) 41 , 9 patients received complement inhibition (eculizumab) in addition to MAS-target treatment: 7 patients achieved regression of both MAS and TMA, and 2 died.
JAK-inhibitors
In our SLR only one study reporting JAK-i was specifically focused on MAS 38 . In this paper, authors described 10 refractory sJIA, 3 of whom with severe MAS resistant to high-dose GCs and tocilizumab (1 also to etoposide). All of them were treated with ruxolitinib (2.5-5 mg x 2/day) with a rapid regression of MAS without adverse events. Notably, none received IL-1 inhibitors or CsA before Jak-i introduction, and all required the further addition of canakinumab to control underlying sJIA.
Three other studies 14-16 reported the use of ruxolitinib in mixed cohorts of sHLH patients. In a retrospective case series of 9 patients (5 EBV-HLH, 2 fHLH, 1 MAS, 1 unspecified) refractory to the HLH94 protocol, 3 patients (1 MAS) achieved MAS remission, while others required the association with DEX-P 14 . In a case-control study 15 , 11 patients (including 2 sJIA-MAS and 1 KDMAS) were successfully treated with ruxolitinib (7 refractory to HLH04 protocol, 4 naïve). In a pilot, open-label, single arm trial 16 12 sHLH patients (8 EBV-HLH, 2 MAS, 2 unspecified) received ruxolitinib as 1 st line treatment with a positive response in 10 of them.
The only other JAK-i mentioned as a treatment for sJIA-MAS was tofacitinib in 2 patients: in one case tofacitinib was ineffective and was switched to ruxolitinib 38 , while in the other it contributed to control MAS recurrency together with canakinumab 37 .
HSCT
Six studies reported data about HSCT in patients with refractory MAS ,23,34,35,42-44 . In a case series Silva et al. 35 described 5 patients with refractory sJIA-MAS treated with allogeneic HSCT: 1 patient died from pulmonary haemorrhage 85 days after HSCT, 3 developed graft versus host disease, and 5/5 had severe infections following HSCT. All but one patient developed 100% chimerism, and all patients who survived achieved disease remission after HSCT. Chellapandian et al. 44 described a 4-year-old child with sJIA, recurrent MAS and LD, refractory to GCs, anakinra,
methotrexate, tocilizumab and canakinumab, who was successfully treated with emapalumab as bridge therapy to a matched sibling donor allogenic HSCT. HSCT was further mentioned in 4 MAS and 4 sHLH 23,34,42,43 : outcome data were available for 2 MAS, who survived without disease reactivation, and for sHLH patients, of whom one died.
Other treatments
Use of IVIG was reported in 280 sJIA, 46 SLE, 37 KD and 48 other rheumatic diseases, from 41 studies. However, specific data on IVIG efficacy are extremely hard to be extracted, as IVIG was almost always used as part of a combined regimen and no studies focused on IVIG efficacy were found. In 15 studies, plasma-exchange (PE) was mentioned as additional treatment for MAS. Overall, 48 patients with sJIA, 9 with SLE, and 6 with other rheumatic diseases received PE for MAS. In particular, PE was used as part of a combination therapy in 17 patients to control MASassociated TMA 41 .
Treatment of MAS in other rheumatologic diseases other than sJIA
Fourteen papers presented detailed data about SLE-MAS, for a total of 105 patients, with an overall mortality of 7% ( Table 2 ). Bennett et al. 42 compared the differences in MAS treatment between SLE and sJIA in a cohort of 102 sJIA and 19 SLE. SLE patients were more frequently given DEX (32% vs 14%, p = 0.05), cyclophosphamide (21% vs 3%, p = 0.01), and MMF (32% vs 2%, p < 0.001); only children with underlying sJIA received IL-1 antagonists. Similarly, in the cohort by Aytac et al. 45 , all patients with sJIA seen after 2011 received anakinra, while patients with SLE were treated more frequently with IVIG (68% vs 33%) and etoposide (50% vs 32%), and received IL-1 blockade in 30% of cases. In the large cohort of SLE-MAS described by Borgia et al. 46 , only 2 patients were treated with anakinra: both patients were refractory to several treatments, including PE and in one case alemtuzumab and intrathecal methotrexate, and eventually died.
Eleven studies reported detailed information about KD-related MAS in 58 patients (Table 3 ). Treatment of MAS included GCs (85%), IVIG (73%), CsA (19%), and infliximab (12%). Fifteen patients (26%) received etoposide (11 within HLH protocol). Two KD-MAS patients were successfully treated with IVIG alone 47,48 . In our SLR, only one patient received anakinra, with rapid remission 49 . Three patients died (5%, all treated with HLH protocol), and only 1 had persistent coronary artery ectasia.
Differences between paediatric sub-specialties and geographic areas.
Treatments of the cohort of 362 sJIA-MAS described by Minoia et al. 10,20 were stratified, both according to the geographic area of the referral centre and to the subspecialty of the treating physician. Patients followed in North America (NA) more frequently received IVIG and biologics than patients treated in Europe or in other continents (IVIG: NA 54%, Europe 26%, other continents 43%; biologics: NA 34%, Europe 16%, other continents 7%). No significant differences were observed in the percentage of patinets treated with GCs, CsA and etoposide. Paediatric haemato-oncologists more frequently used biologic agents (24% vs 3%, p = 0.02) and etoposide (18% vs 10%, p = 0.04), whereas paediatric rheumatologists more frequently prescribed CsA (67% vs 40%, p < 0.0001).
DISCUSSION
MAS represents a life-threatening condition that requires prompt effective treatment to avoid potentially fatal outcome; however, the therapeutic approach to MAS is still a challenge for clinicians worldwide. Recently, international collaborative efforts have strived for a common standardized approach 11 . In this context, the METAPHOR project is aimed to capture the real-life therapeutic strategies in MAS in different clinical settings, and, in particular, the current
SLR had the main purpose of uncovering areas in which evidence regarding MAS treatment is still lacking, leading to major discrepancies among practitioners.
Despite the sizable amount of data regarding MAS patients reported in literature, the global level of evidence on treatment outcome is still poor, with a scarcity of comparative data across papers, mainly due to the heterogeneous nature of most studies, the lack of standardized outcome measures, and the high risk of bias in attributing effectiveness or safety to a specific medication or condition. Indeed, outcome data on the concomitant use of different therapies are really difficult to extract, as the timing of start of drugs is rarely specified. Furthermore, although MAS is a unique syndrome, the heterogeneity of the underlying rheumatologic backgrounds may differently affect its course and influence the treatment used.
Although not based on any formal clinical trial, high-dose GCs are confirmed as the mainstay of treatment of MAS in all rheumatologic backgrounds across the literature, and GC were used in almost all patients. Together, MPN and prednisolone accounted for 90% of MAS patients, while DEX was mainly used in the context of HLH protocol and in patients with a potential higher risk of CNS involvement 42 . GCs were mostly used as co-medications, and only 14% of MAS were treated with GCs as monotherapy. Interstingly, this data is in line with what we observed in the cohort of 362 sJIA-MAS, where only 19% of patients survived with GCs alone 20 (unpublished data, courtesy dr. F. Minoia and dr. A. Ravelli). Despite difficulties in assessing their specific efficacy, due to the heterogeneity of conditions reported and co-medications used, the role of GCs in MAS is life-saving especially in low-income countries; of note, a tapering scheme of MPN pulses was proposed for severe MAS in resource limited settings 17 . Furthermore, despite limited numbers, DEX-P was successfully used in MAS refractory to MPN pulses and CsA in Japan 19 .
Data on CsA in MAS come only from retrospective cohort studies in which it was mainly used together with several other agents, with variable dosages and routes of administration,
making a reliable evaluation of its efficacy highly biased. However, CsA was confirmed as the most frequently used medication besides GCs, with a global positive efficacy and safety profile. CsA is widely accessible at affordable costs and might play a key role in the treatment of MAS refractory to high-dose GCs, especially in low-income countries or in those centres in which biologic medications are not accessible in a timely manner.
Anakinra is by far the most used biologic treatment for MAS, especially for sJIA-MAS. Despite the fact that no (randomised) controlled clinical trial tested the efficacy of anakinra in MAS, more than 80% of patients with sJIA-MAS treated with anakinra reported a complete regression of MAS, with a high safety profile. An unbiased evaluation of its efficacy and best therapeutic scheme is impossible to make, given the heterogeneity of the studies included. However, data collected strongly support the use of anakinra in patients with sJIA-associated MAS. Evidence of anakinra role in other subtypes of MAS is less robust; however, its safety profile and short half-life make it a valuable option for all sHLH, especially in critical care settings 50 . Data regarding other biologics in MAS are limited. Although no specific biologic used at the indicated regular dose seems to provide full protection against MAS 24,51,52 , small case-series showed positive results of canakinumab and tocilizumab in sJIA-MAS, raising the possibility of a therapeutic alternative in countries where anakinra is not available; however, further data are needed to confirm this preliminary observation.
Emapalumab is the only medication tested in a clinical trial in MAS and showed extremely positive results in high-dose GCs refractory sJIA-MAS with more than 90% of remission 31 . Given its specific target effect on IFN ï§ , emapalumab has a highly promising role for all subtypes of MAS, although these preliminary results need to be confirmed in larger cohorts and in patients with other rheumatologic backgrounds. Notably, emapalumab is still not accessible in most countries worldwide. Given their effect on the IFN ï§ pathway, JAK-i could potentially play an important role
in MAS treatment; however, so far, evidence on MAS is limited to case reports and to mixed sHLH cohorts. For sJIA-MAS, it should be noted that neither IL-1 nor IL-18 receptors signal through JAKs. IL-18 blockade might also represent a promising approach 53 , and an ongoing international trial with a biclonal anti-IL-1 ï¢ /IL-18 antibody is exploring its effect in monogenic diseases associated with inflammatory MAS (NCT04641442)
Since etoposide is a key medication in HLH protocols, its use in severe MAS was extensively reported, albeit associated with a significant toxicity and mortality. In the 362-cohort of sJIA-MAS described by Minoia et al. 20 , etoposide was used in almost 12% of cases and was most frequently prescribed by haemato-oncologists 10 . Interestingly, a low-dose etoposide protocol was successfully used in a small-cohort of highly refractory MAS patients, with a positive outcome 26 , and its role, especially in countries without access to targeted medications, needs to be better explored.
Data reflecting different therapeutic approaches according to geographic areas or subspecialty of the treating physician were assessable only from one cohort of sJIA-MAS. 10,20 . In a recent survey 54 , not included in the SLR due to publication type, GCs were confirmed as the 1 st -line medication for MAS across all the subspecialties; notably, haemato-oncologists preferred DEX over MPN. IL-1 inhibitors were chosen as 1st-line therapy in MAS more frequently by rheumatologists compared to haemato-oncologists, while etoposide was more frequently the 2 nd -line choice of haemato-oncologists.
In conclusion, data regarding MAS treatment are progressively increasing, especially for sJIA-associated MAS, with highly promising results for IL-1 and IFN ï§ inhibitors. However, global level of evidence on MAS treatment, especially in other rheumatologic conditions, is still poor with high biases and scarce reliability in attributing efficacy to a specific medication, due to the retrospective nature and heterogeneity of most studies and the lack of agreed outcome measures.
As a consequence, therapeutic approaches to MAS are still extremely variable, with potential significant discrepancies across different centres and countries. An international effort is needed to optimize therapeutic strategies, reduce gaps in access to medications and harmonize MAS treatment worldwide.
ACKNOWLEDGMENTS
The authors thank Paulien H. Wiersma (University Medical Center Utrecht, the Netherlands) for the guidance in the literature search. Furthermore, authors are profoundly grateful to Elisa Patrone, Marco Garrone, Federico Serra, and Victoria Morozan from PRINTO, and to Luciana Peixoto from the systemic JIA Foundation for their invaluable support throughout the METAPHOR project. The authors also acknowledge the PReS MAS/sJIA Working Party and Paediatric Rheumatology International Trial Organization.
Funding . This study was awarded within the Pediatric Rheumatology European Society (PReS)/ Pediatric Rheumatology International Trial Organization (PRINTO) annual Call for Grants (https://www.printo.it/projects/pres) and partially funded by a grant to IRCCS Policlinico of the Italian Ministry of Health.
Conflict of interest statement . M.G. received speaker or consultancy fees from Novartis, SOBI, Boehringer, Zydus, Fresinius Kabi e Kinisa; S.C. received consultancy fees from SOBI, Pharming, X4 and Electra Therapeutics; A.G. received consultancy fees and research grants from Novartis and SOBI; P.Q. received consultancy fees from AbbVie, Amgen, Bristol-Myers Squibb, Chugai-Roche,
Lilly, Novartis, Novimmune, Pfizer, Sanofi and SOBI; F.M. received consultancy fees from SOBI and Novartis. The remaining authors have declared no conflicts of interest.
Data availability statement . All data relevant to the study are included in the article. Data are available upon request from Dr. Francesca Minoia (francesca.minoia@policlinico.mi.it)
Contributorship . We confirm that all authors have contributed in the study by participating in design and conduct, validity evaluation, data analysis, manuscript preparation
REFERENCES
patients with macrophage activation syndrome complicating systemic juvenile idiopathic arthritis. Ann Rheum Dis. 2017;76:166-172