Early management of hemophagocytic lymphohistiocytosis/macrophage activation syndrome: a systematic literature review

Abstract

Objectives

A systematic literature review was undertaken to summarise and analyse the available evidence on the initial hemophagocytic lymphohistiocytosis and macrophage activation syndrome (HLH/MAS) treatment according to the distinctive pathologic background and/or triggering factors.

Methods

An electronic search in the main medical databases (PubMed, EMBASE, and Cochrane Library) was conducted, in order to select the studies providing information related to the HLH/MAS therapy. To further maximise the evidence derived from this systematic review, we focused our attention on clinical studies that included and described >10 HLH/MAS cases.

Results

This literature search led to identify 18,020 articles. After the screening by title, removal of duplicated records, and full-text assessment of the relevant articles, the final search output consisted of 167 articles that provided enough information on the medical management of patients with HLH/MAS In total, 77 clinical studies (including >10 patients) describing the early treatment of HLH/MAS have been included in the present analysis.

Conclusions

Primary HLH is typically managed according to the HLH-94 or HLH-2004 protocols, with the main aim of bringing the patient to a hematopoietic stem cell transplant (HSCT). Secondary forms of HLH/MAS can arise in 3 main clinical contexts, namely rheumatic disorders, oncological diseases, and previously healthy patients developing specific and/or severe infections. Accordingly, different and variable therapeutic approaches have been applied to these 3 categories of secondary HLH/MAS. Despite the survival improvements achieved in the last decades in patients with HLH/MAS, there is a strong need for prospective and controlled studies focused on each specific group of HLH/MAS.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• •Hemophagocytic lymphohistiocytosis (HLH) and macrophage activation syndrome (MAS) can progress rapidly; thus, their early identification and therapeutic management are critical for preventing organ failure and mortality.
• •The therapeutic approach to HLH/MAS greatly varies according to the different clinical contexts and/or triggering factors.

WHAT THIS STUDY ADDS

• •Primary hemophagocytic lymphohistiocytosis (HLH) is typically managed according to the HLH-94 or HLH-2004 protocols, with the aim of bringing the patient to hematopoietic stem cell transplant as soon as possible after remission.
• •Secondary forms of HLH/macrophage activation syndrome (MAS) can arise in 3 main clinical contexts, namely rheumatic disorders, oncological diagnoses, and previously healthy patients developing specific and/or severe infections.
• •Empiric and general immunomodulatory therapies in patients with HLH/MAS could include glucocorticoids, anakinra, and/or intravenous immunoglobulin.
• •Malignancy-related HLH/MAS can be variably treated through intensification of conventional chemotherapy, use/addition of etoposide, or the administration of treatments according to the HLH-94 or HLH-2004 protocols, which in some cases is integrated with tumour-specific therapies.
• •In addition to the conventional HLH protocols, anti–interferon-γ therapy (emapalumab) has emerged as a ground-breaking option, especially in children with refractory, recurrent, or progressive forms of primary HLH, or in patients with intolerance to conventional therapies.
• •Antimicrobial therapies are the mainstay of the therapeutic management of HLH/MAS triggered by infections in patients without any underlying chronic disorders.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY

• •Despite the overall survival improvements achieved in the last decades in patients with hemophagocytic lymphohistiocytosis and macrophage activation syndrome (HLH/MAS), there is a strong need for prospective and controlled studies focused on each specific group of HLH/MAS.
• •Specific and standardised therapeutic protocols for the early treatment of secondary HLH/MAS should be developed according to the specific pathologic context (malignancies, rheumatic diseases, and previously healthy patients with specific infections).

Alt-text: Unlabelled box dummy alt text

INTRODUCTION

In 2020, the European Alliance of Associations for Rheumatology (EULAR) and the American College of Rheumatology (ACR) commissioned provisional points to consider for the diagnosis, management, and monitoring of the early stages of hemophagocytic lymphohistiocytosis/macrophage activation syndrome (HLH/MAS) [1,2]. This goal was pursued by convening an international Task Force. As part of this work, a systematic literature review was completed focusing on the early medical management and treatment of HLH/MAS in both paediatric and adult populations.

HLH was first defined as ‘Histiocytic medullary reticulosis’ by Scott and Robb-Smith in the 1930s. They described a series of patients with ‘fever, wasting and generalised lymphadenopathy’, which were ‘associated with splenic and hepatic enlargement and in the final stages jaundice, purpura and anaemia with profound leukopenia’; the postmortem examination showed ‘generalized hyperplasia of histiocytes actively engaged in phagocytosis of erythrocytes’, referred to as hemophagocytosis [3].

However, the term ‘HLH’ was introduced later in the 1950s [4] and has been extensively used since the early 1980s, in order to describe a process of pathologic hyperinflammation leading to uncontrolled activation of macrophages, T cells, and other immune cells, which could potentially result in a fatal multiorgan failure. It mainly occurred in patients affected with malignancies and rheumatic disorders and children affected by familial (‘primary’) HLH due to specific genetic variants [5]. The term ‘MAS’ was coined in the 1990s with specific reference to acquired (‘secondary’) forms of HLH occurring in patients with rheumatologic disorders [[5], [6], [7]]. Such a persistent and uncontrolled activation of the CD8+ T-cell/macrophage immune loop results in systemic hyperinflammation, which is characterised by the massive production of several proinflammatory cytokines (‘cytokine storm’), leading to dysregulated hemophagocytosis in multiple organs [8,9].

The heterogeneous clinical spectrum of HLH/MAS has prompted several efforts aimed to develop reliable diagnostic criteria to facilitate early diagnosis and timely treatment, which are essential to improve the outcome of this potentially fatal disease [1,2].

Historically, the first diagnostic guidelines for HLH were proposed by Henter et al. [10] on behalf of the HLH Study Group to guide the inclusion of patients in the HLH-94 treatment study conducted by the Histiocyte Society. Briefly, this treatment protocol consisted of dexamethasone (DEXA), etoposide (or VP-16), cyclosporin A (CsA), and the addition of intrathecal methotrexate (MTX) for patients with central nervous system disease [11]. The subsequent experience in the treatment of HLH led this study group by the Histiocyte Society to develop a new treatment protocol (HLH-2004), along with a revised set of diagnostic criteria. Even though minor changes were introduced in this protocol in terms of pharmacologic regimen, it clearly highlighted the need for different therapeutic approaches in primary and secondary forms of HLH [12].

In the rheumatologic setting, HLH/MAS occurs most often in the setting of systemic juvenile idiopathic arthritis (sJIA), which adds further challenges to promptly differentiate between these hyperinflammatory conditions, which share several clinical similarities. Therefore, in 2006, Ravelli et al. [13] developed a set of preliminary diagnostic guidelines specific to MAS complicating sJIA. In 2016, the EULAR/ACR/Paediatric Rheumatology International Trials Organisation Collaborative Initiative (PRINTO) promulgated a set of classification criteria for MAS complicating sJIA based on an international consensus process. In order to simplify the diagnosis of MAS in this specific rheumatologic context, the presence of persistent fever in patients with juvenile idiopathic arthritis (JIA) should indicate the diagnosis of MAS when the values of ferritin and 2 other laboratory parameters (platelet count, aspartate aminotransferase, triglycerides, and fibrinogen) are above or below certain thresholds [14]. However, the onset, presentation, and clinical course of HLH/MAS can be heterogeneous across different underlying pathologic backgrounds and within individual patients [[15], [16], [17], [18]]. Therefore, other broader diagnostic approaches have been proposed, such as the H-score (focused on the wide group of ‘reactive’ HLH linked to infections, malignancies, or autoimmune diseases), guidelines specific to malignancy-associated HLH, and HLH in critically ill patients [[19], [20], [21]].

Although well-established diagnostic criteria are useful to promptly recognise HLH/MAS in different clinical settings, the therapeutic approach should also be differentiated depending on the specific pathologic context. However, most studies and clinical trials have included patients with mixed types of HLH/MAS.

Our systematic literature review was aimed at summarising and analysing the evidence on the initial HLH/MAS treatment according to the distinctive pathologic background and/or triggering factors. In detail, we analysed the different therapeutic approaches in primary and secondary forms of HLH, and inside the latter category, we made a subanalysis of HLH/MAS in rheumatic, oncological, and previously healthy patients developing infections.

METHODS

In this systematic review, the main research question was the assessment of early therapy in different forms of HLH/MAS and the related main outcomes. This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [22] and included all the original articles (such as experimental studies, randomised control trials, case–control studies, cohort studies, observational studies, and case series) providing information on these aforementioned aspects.

Search strategy

Three databases (namely, PubMed, EMBASE, and Cochrane Library) were scrutinised by using relevant keywords related to ‘HLH’ and ‘MAS’, as described in Fig. All relevant literature up to November 2020 was extracted. Only English-language articles were considered.

Figure.

Flow chart of the study selection and inclusion process. HLH, hemophagocytic lymphohistiocytosis; MAS, macrophage activation syndrome.

Study selection

The review process was performed by 5 readers (BS, AVG, MW, MR, and DPi). First, the duplicates from different databases were removed. Then, title and abstract screening was performed by at least 2 readers, who independently indicated all the relevant articles to be retained for full-text review. After this exercise, the relevant articles were selected for data extraction and synthesis. Discrepancies in articles selection were resolved by consensus with an additional reader (SC, ED, or FD).

Data extraction and synthesis

Data extractions were performed by 5 readers (BS, AVG, MW, MR, and DPi) by using a predefined data extraction form. Information on each study’s characteristics (including first author, year of publication, study period, study design, HLH-MAS diagnostic criteria, patients’ number, gender ratio, and age [median or mean, range], underlying diseases, genetic analysis [for primary HLH/MAS forms], infectious triggers, steroidal treatment, nonsteroidal treatment, intravenous immunoglobulin (IVIG) usage, antimicrobial therapies, follow-up time, outcome, additional, and specific information, if relevant).

RESULTS

The results of each step of the literature search according to the PRISMA guidelines are schematically summarised in Fig. In total, the literature search led to identify 18,020 articles: 7560 from PubMed, 199 from the Cochrane Library, and 10,261 from Embase. After the screening by title, the duplicated records (n = 5414) were removed, and 3778 articles were assessed for eligibility based on the abstracts: of those, 258 full-text original articles were considered eligible according to the aim of this systematic review. The final search output consisted of 167 articles that provided enough information on the medical management of patients with HLH/MAS.

To further maximise the evidence derived from this systematic review, we focused our attention on clinical studies that included and described >10 HLH/MAS cases. In total, 77 articles were included in this systematic review. Table 1 [[23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39]] summarises the clinical studies reporting primary HLH or familial HLH (FHL) treatment and survival rates (in >10 patients) [[23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39]], and Table 2 [29,[39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98], [99]] lists the clinical studies reporting treatment and survival rates of patients with secondary HLH/MAS (in >10 patients) [[40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98]].

Table 1.

Summary of all the clinical studies describing FHL treatment and survival rates (in articles including >10 patients among those selected in the present systematic review).

No	Author, year	Country	Study period	Study design	HLH-MAS diagnostic criteria	Patients N M:F	Age M/Me [range]	Disease type	Genetic defect	Infectious trigger	Steroids	Nonsteroidal treatment	IVIG	HSCT	Follow-up M/Me [range]	Outcome (survival)	Notes
1	Henter et al. 2002 [29]	International	1994-1998	Pro	HLH-94	25 13:12	M 13 mo [<6 y]	FHL	? (positive family history)	n/a	Yes (25)	HLH-94	-	n = 20	n/a	A (13) - after HSCT (13) D (12) -pre-HSCT (5) -post-HSCT (7)	—-
2	Ouachee-Chardin et al. 2006 [37]	France	1982-2004	Retro	HLH-94	48 30:18	Me 3 mo [1 d-18 y]	FHL (33)	PRF1 (7) UNC13D (11) STX11 (1) None (8) Incomplete study (21)		Yes (48)	ETO+CsA+ itMTX (15) CsA±ATG (33)	-	n = 48	Me 5.8 y [up to 20 y]	D (20) A (28)	—-
3	Mahlaoui et al. 2007 [33]	France	1991-2005	Retro	HLH-94	38 24:14	Me 2.3 mo [2 d-9.3 y]	FHL	PRF1 (14) UNC13D (11) None (4) Incomplete study (9)	n/a	Yes (38)	See notes	-	n = 16	n/a	A (29) D (9) (pre-HSCT)	- Assessment of the use of ATG in association with corticosteroids (MPL) and CsA, combined with intrathecal MTX. - The total dosage of ATG was 50 mg/kg or 25 mg/kg according to the severity of the disease, over 5 consecutive days.
4	Meeths et al. 2010 [34]	International	?	Retro	HLH-94 HLH-2004	11 4:7	M 4.2 y [0.5-17 y]	FHL (5)	STXBP2 (11)	EBV (4)	Yes (9)	HLH-94 (2) HLH-2004 (6) RTX (1)	-	n = 6	M 2.9 y [1 mo-7 y]	A (8) D (3)	—-
5	Ohga et al. 2010 [36]	Japan	1995-2005	Retro	HLH-1994	43 23:20	Me 0.5 y [6 d-12 y]	FHL	PRF1 (12) UNC13D (11) ?? (20)	n/a	n/a	HLH-1994 (25) ‘CHOP-VP16’ (8) ?? (5)	-	n = 43	n/a	A (29) D (7) ?? (5)	- This study included only primary HLH undergone HSCT.
6	Fisgin et al. 2010 [28]	Turkey	1995-2005	Retro	HLH-94 HLH-04	22 12:10	Me 16.5 mo [0.6-79 mo]	FHL	n/a	n/a	Yes (22)	CSS+CsA+ IVIG (14) HLH-94 (4) HLH-2004 (3) Other (1)	IVIG (14)	n = 1	n/a	A (4) D (18)	- All patients had positive familial history.
7	Trottestam et al., 2011 [39]	International	1994-2003	Pro	HLH-94	60 35:25	Me 2 mo [0-145]	FLH	n/a	n/a	Dexa (249)	HLH-94 (249)	-	n = 60	Me 6.2 y. (0.4-13.7)	A (29) - post-HSCT (29) D (31) - post-HSCT (15) - pre-HSCT (16)	—-
8	Jovanovic et al., 2014 [30]	Serbia	1997-2010	Retro	HLH-2004	17 9:8	Me 7 mo [1-192]	FHL (n = 10) IA-HLH (n = 3) No disease (n = 4)	XLA (1) Munc13-4 (2) Unspecified (7)	CMV (n = 2) Adenovirus (n = 1)	Dexa (n = 17)	HLH-94 (8) HLH-2004 (9)	-	n/a	n/a	A (7) - post-HSCT (4) - pre-HSCT (3) D (10) - pre-HSCT (10)	- In this study, only 17 HLH patients with central nervous system Involvement were described (out of 30 patients mentioned in the results).
9	Bode et al. 2015 [25]	International	n/a	Retro	HLH-2004	28 n/a	M 1.5 y [0.13-34]	PID	SCID (8) CID (10) CGD (8) Others (2)	TB (1) EBV (6) Adenovirus (3) CMV (4) Leishmania spp. (3) Others (11)	Any (16)	HLH-2004 (2) CsA (2)	IVIG (11)	n/a	n/a	A (10) D (16) ? (2)	- 8 patients received no specific HLH therapy.
10	Bergsten et al. 2017 [24]	International	2004-2012	Pro	HLH-2004	168 77:91	Me 3.4 mo	FHL (verified)	PRF1 (?) UNC13D (?) STX11 (?) STXBP2 (?)	Infection (32) EBV-CMV (16)	Yes (168)	HLH-2004	-	n = 133	Me 5.2 y	A (95) - post-HSCT (93) - no-HSCT (7) D (68) - pre-HSCT (28) - post-HSCT (40)	- In total, 369 children were reported in this article (including 201 children without verified FHL). - ‘Patients with mutations in other HLH-associated genes were studied separately. These 29 patients (XLP:16; GS2:11, CHS:1, HPS:1) had a 5-year survival of 59% (43%-81%)’.
11	Beken et al. 2018 [23]	Turkey	1994-2012	Retro	HLH-94 HLH-2004	57 25:32	Me 18 mo	FHL	PRF1 (10) UNC13D (11) STX11 (3) None (13) Incomplete study (20)	n/a	Yes (49)	HLH 94 (17) HLH-2004 (32) Supportive only (8)	-	n = 18	n/a	A (11) - HSCT (9) -no HSCT (2) D (46) - HSCT (9) -no HSCT (37)	- ‘Patients aged 2 y and under, those with a ferritin level more than 1500 mg/dL at the time of diagnosis, and who have spinal cord or hepatic involvement, should be considered as having a poor prognosis’.
12	Messina et al. 2018 [35]	Italy	2000-2014	Retro	Not specified	109 65:44	Me 1 y. [27 d-18 y]	FHL Other PIDs	PRF1 (31) UNC13D (32) GS (6) SH2DIA (5) Others (7) None (15) Incomplete study (13)	n/a	n/a	HLH 94 (9) HLH-2004 (41) Euro-HIT-HLH (3) Others (8) Unknown (48)	-	n = 109	5.2 y [0.9-14.9]	A (78) D (31)	- This study included only primary HLH undergone HSCT.
13	Chinn et al. 2018 [27]	USA	1999-2016	Retro	HLH-2004	122 62:60	Me 6.07 y [0.04-18.50]	FHL (19) Other genetic defects (27) No genetic defect (76)	n/a	Infection (46)	Any (18)	‘Immuno-Chemotherapy’(54)	-	n = 38	18.6 mo [0.04-201.4]	A (n = 76) D (n = 41) ? (n = 5)	- 32 patients were diagnosed with an autoimmune disease, 14 with malignancy. No evident trigger in 32 patients.
14	Locatelli, et al., 2020 [31]	International	n/a (until July 2017)	Pro (Open-Label, Single-Group)	HLH-2004	34 16:18	Me 1 y [0.1-13]	FHL Other PIDs	PRF1 (7) UNC13D (8) STX11 (1) STXBP2 (2) GS (5) SH2DIA (1) BIRC (1) Others (2) Unknown (9)	n/a	Dexa (34)	Emapalumab (34) + CsA (?)	-	n = 26	1-year (after HSCT)	A (n = 24) - post-HSCT (24) D (n = 10) - pre-HSCT (8) - post-HSCT (2)	- Emapalumab: initially, 1 mg/kg every 3 days; then, subsequent doses could be increased to 3, 6, and up to 10 mg per kg. - DEXA was started 1 day before emapalumab at a dose of 5-10 mg/m2/d. - CsA could be continued if the patient was receiving, or could not be used at all after starting emapalumab. - 27 patients were previously treated with HLH protocols and 5 patients were previously untreated.
15	Ma et al. 2020 [32]	China	2015-2019	Retro	HLH-2004	38 23:15	Me 2.5 [0.1-13.7 y]	primary	Genetic analysis available	n/a	Dexa (36) mPSL (2)	HLH-1994 (25) HLH-2004 (11) Doxo-ETO (2)	-	n = 15	36 mo	A (30) D (8)
16	Cetinkaya 2020 [26]	Turkey	2013-2017	Retro	HLH-2004	28 20:8	Me 23 mo	PID	CMV (8) EBV (4)	n/a	Any (28)	CSA (21) ETO (18) RTX+PF (4)	IVIG (28)	n = 5	n/a	A (22) D (9)
17	Shabrish et al. 2021 [38]	India	2010-2020	Retro	HLH-2004	101 68:33	Me 6 mo [0.25-384] Me 6 mo [6-396] Me 18 mo [3-184] Me 48 mo [1-168]	FHL2 FHL3 FHL4 FHL5	PRF1 (50) UNC13D (35) STX11 (9) STXBP2 (9)	n/a	Any (94)	HLH-94 (?) HLH-2004 (?) Total (94)	-	n = 18	n/a	A (25) - before HSCT (12) - after HSCT (13) D (63) - no HLH (7) - after HLH (51) - after HSCT (5) ?? (13)

A, alive; ADM, adriamycin; ALTZ, alemtuzumab; AML, acute myeloid leukaemia; ANCA-V, ANCA-associated vasculitis; ANK, anakinra; AOSD, adult-onset Still disease; ARA-C, cytarabine; ATG, antithymocyte globulin; AZA, azathioprine; CHT, chemotherapy (disease-specific); CMV, cytomegalovirus; CPM, cyclophosphamide; CsA, cyclosporin A; CSS, corticosteroid; CTD, connective tissue disease; D, death; DEXA, dexamethasone; Doxo, doxorubicin; EBV, Epstein–Barr virus; ETN, etanercept; ETO, etoposide; FHL, familial hemophagocytic lymphohistiocytosis; FLU, fluorouracil; HCQ, hydroxychloroquine; HF, hemofiltration; HHV, human herpes virus; HLH, hemophagocytic lymphohistiocytosis; HSCT, hematopoietic stem cell transplantation; HSV, herpes-simplex virus; IA, immune disease–associated; itMTX, intrathecal methotrexate; IVIG, Intravenous immunoglobulin; JIA, juvenile idiopathic arthritis; KD, Kawasaki disease; LA, lymphoma-associated; LCH, Langerhans cell histiocytosis; LCH, Langerhans cell histiocytosis; MA, malignancy-associated; MCTD, mix connective tissue disease; MMF, mycophenolate mofetil; MP, mercaptopurine; MPL, methylprednisolone; MRSA, methicillin-resistant staff aureus; PF, plasmapheresis; PID, primary immunodeficiency; PSL, prednisolone; R-DED, R-DED regimen (ruxolitinib, doxorubicin, etoposide, and dexamethasone); RSV, respiratory syncytial virus, SBEGA, Streptococcus beta-hemolyticus group A; RTX, rituximab; SLE, systemic lupus erythematosus; TB, tuberculosis; TCZ, tocilizumab, IFX, infliximab; TPO, thrombopoietin receptor against; VBL, vinblastine; VCR, vincristine; VZV, varicella zoster virus; XLP, X-linked lymphoproliferative disease.

Table 2.

Summary of clinical studies describing treatment and survival rates of secondary HLH/MAS (in articles including >10 patients among those selected in the present systematic review).

Ref	Author, year	Country	Study period	Study design	HLH-MAS diagnostic criteria	Patients N M:F	AGE M or Me [range]	Disease type	Infectious agent(s)	Steroidal treatment	Nonsteroidal treatment	IVIG	Antimicrobial therapy	Follow-up	Outcome	Notes
1	Risdall et al., 1979 [73]	USA	1970-1977	Retro	n/a	5 4:1 14 7:7	Me 1 y. [1 mo-16 y] Me 28 y. [3-61 y]	No disease Underlying disease: - SLE (1) - renal transplant (13)	EBV (1) Adenovirus (1) Unknown (3) CMV (10) HSV (1) EBV (1) VZV (1) Unknown (1)	Yes (3) Reduced	VBL (1) no	-	Broad-spectrum antibiotics	Me 32 mo [5.87 mo]	A (3) D (2) A (9) D (5)	- All patients showed hemophagocytosis in the bone marrow.
2	Wong et al., 1992 [88]	Hong Kong	1987 -1991	Retro	n/a	40 24:16	M 46.5 y [1-84 y]	LA (16) IA (16) SLE (2) Others (6)	S. aureus (2) Streptococcus spp. (3) S. typhi (1) E. coli (1) Acinetobacter spp. (1) R. tsutsugamushi (1) C. psyttaci (1) HBV (1) Adenovirus (1) Dengue (1) ‘presumed’ (1)	Yes (2) [SLE]	‘chemotherapy’ [malignancies]	-	Yes [IA]	n/a	A (22) D (18)	- All patients showed hemophagocytosis in the bone marrow.
3	Aricò et al., 1996 [41]	International	1989-1994	Retro	Variable	122 61:61	Me 2.9 mo [0-70 mo]	FHL (59) Non-FHL (63)	Infections (50) Viral (23) Bacterial (3) Toxoplasma (1) ‘presumed’ (23)	No	ETO (86) TPO (8) VBL+VCR (7) MTX (1) IT-MTX (71) Cranial RXT (9)	-	n/a	Me 44 mo	A (27) D (95)	- No treatment was administered in 20 because the diagnosis was postmortem. - 102 patients received ‘HLH-directed chemotherapy’ - 29 patients underwent HCST.
4	Imashuku et al., 1997 [57]	Japan	1986-1995	Retro	HLH-94	82 34:48	n/a [1 mo-23 y]	n/a	EBV (15)	PSL or DEX (69)	Yes (69) ETO (40) DOXO (?) VCR (?) CPM (?) MTX(?)	IVIG (n/a)	n/a	4 y.	A (47) D (35)	- 13 patients received supportive therapy alone.
5	Chen et al., 1998 [48]	Taiwan	1990-1996	Retro	HLH-94	22 9:13	Me 4.4 y [0.5-17]	Unknown (MA, PID, FLH excluded)	EBV (n = 13)	PSL (22)	ETO ‘VP1 regimen’ (n = 13) ‘VP2 regimen’ (n = 9)	IVIG2 (n = 16) IVIG5 (n = 1)	ACY or GAN (3)	Me 5.1 mo [0.5-86]	A (13) D (9)	- ETO was given at 150 mg/m2/d over 3 consecutive days, repeated every 3–4 weeks (VP1) or twice weekly for 2 weeks, and then either biweekly or weekly (VP2).
6	Takahashi et. al., 2001 [82]	Japan	1990-1999	Retro	n/a	27	n/a [16-78 y]	NK-T LA	n/a	No	‘LA standard chemotherapy’	n/a	n/a	Me 46 d [2-239 d]	A (0) D (20)	- All patients showed hemophagocytosis in the bone marrow.
7	Imashuku et al., 2001 [59]	Japan	1992-1999	Retro	HLH-94	21 10:11 26 10:16	Me 5.2 y [0.8-31 y] Me 4.7 y. [0.6-17 y]	FHL (1) XLP (1) Unknown (45)	EBV (47)	Yes (47?)	CCS±IVIG±CsA (21), followed by HLH-94 (17) ETO (26), including HLH-94 (19)	IVIG (10)	n/a	2 mo	A (17) D (4) A (23) D (3)	- ‘Early introduction of an etoposide-based regimen is a critical factor in securing long-term survival in patients with EBV-HLH…’.
8	Veerakul et al., 2002 [85]	Thailand	1989-1998	Retro	HLH-94	52 33:19	M 4.3 y [0.5-12]	IA (15) MA (25) Unknown (12)	Salmonella spp. (3) Staphylococcus spp. (3) Enterobacter spp. (2) Dengue (3) Malaria (2) EBV (1) S. marcescens. (n = 1) Penicillium spp. (n = 1)	Any (2)	‘Chemotherapy’ (23)	IVIG (26)	Empirical antibiotics	n/a	A (36) D (16)	—
9	Henter et al., 2002 [29]	International	1994-1998	Retro	HLH-94	113 61:52	< 15 years	MA (4) JIA (1) Metabolic (1) FHL (25)	n/a	Dexa (113)	HLH-94 (113) (HSCT = 65)	-	n/a	38 mo [15-69]	A (63) D (50)	- A (63): Post-HSCT (40) & Pre-HSCT (23) - D (50): Post-HSCT (25) & Pre-HSCT (25)
10	Imashuku et al., 2003 [58]	Japan	1990-2001	Retro	HLH-94	11 3:8	Me 23 y [17-33]	LA (2) Unspecified (9)	EBV (11)	PSL (4) mPSL (4) Dexa (1) None (2)	ETO+CSA+CTX (6) ETO only (2) Other regimens (3)	-	n/a	n/a [max. 15 mo, related to alive patients]	A (2) D (9)	- 2 patients received HSCT from HLA-matched sibling donors (one was alive at 15 mo, one died).
11	Imashuku et al., 2004 [60]	Japan	1992-2001	Retro	HLH-94	78 29:49	< 2 years to > 15 years	LA (?)	EBV (78)	Any (78)	CSS+IVIG (37), then added ETO (25) HLH-94 (41)	IVIG (37)	n/a	43 mo [1-122 mo]	A (59) D (19)	HSCT (12): A (9), D (3)
12	Buyse et al., 2010 [45]	France	1998-2009	Retro	HLH-2004	56 43:13	Me 49 [n/a]	MA (43) HIV (18) AUTO (7) Unknown (4)	Viral (10) Nonviral (13)	Any (31)	ETO (45) Chemotherapy (32)	IVIG (3)	Antibacterial (46) Antifungal (17) Antiviral (10) Anti-TB (9)	n/a	A (27) D (29)	——-
13	Trottestam et al., 2011 [39]	International	1994-2003	Pro	HLH-94	249 137:112	M 2 y [0-15.3]	FHL (60) Others (189)	n/a (history of infection, n = 109)	Dexa (249)	HLH-94 (249) ‘major changes’ (25)	-	n/a	Me 6.2 y. (0.4-13.7)	A (135) D (114)	- Excluding patients with major treatment changes, 5-year OS is 56%.
14	Miettunen et al., 2011 [68]	Canada	2006-2009	Retro	sJIA-MAS 2006	12 7:5	[0.5-15.5]	AUTO (12): -sJIA (8) -ARF (1) -KD (1) -Others (2)	n/a	mPSL (12)	CsA (10) ETN (1) ETO (2), then ANK (12)	IVIG (10)	n/a	n/a [2-40 mo]	A (12) D (0)	- These patients were treated with ANK after conventional therapy (2 mg/kg/d s.c.; maximum dose 100 mg/d).
15	Hu et al., 2012 [56]	China	2007-2011	Retro	HLH-2004	15 5:10	Me 41 y [14-73]	AUTO (2) LA (2) IA (7) Unknown (4)	EBV (1) CMV (2) MRSA (2) ‘bacteria’ (2)	PDN	VCR+CPM (15)	IVIG (15)	Broad-spectrum antibiotics	M 72.5 w [2-204]	A (10) D (5)	—-
16	Switala et al., 2012 [81]	South Africa	1991-2010	Retro	HLH-94 HLH-2004	15	[9-225 mo]	MA (4) KD (1) SLE (1) Endocarditis (1) No disease (8)	EBV (3) CMV (1) RSV (1) S. pneumoniae (1)	Dexa (10) PDN (2) Dexa+ PDN (1)	ETO (6) CsA (2)	-	n/a	n/a	A (10) D (4) ? (1)	—-
17	Wang et al., 2013 [87]	China	2010-2011	Pro	HLH-2004	40 25:15	Me 31 y. [17-59]	IA (17) LA (10) AUTO (8) No disease (5)	EBV (17)	Dexa (17) mPSL (10) mPSL (8) Dexa (5)	HLH-94 (17) ECHOP (10) FLU (8) HLH-94 (5)	-	rhTPO (20)*	Me 258 d	A (26) D (14)	* This study is an RCT to assess the use of rhTPO in HLH/MAS. The total OS was 65% with no difference according to the use of rhTPO.
18	Lecronier et al., 2013 [64]	France	2002-2011	Retro	HLH-2004	17 12:5	Me 42 y [5-68]	No disease	C. Burnetii (8) R. conorii (9)	Any (1)	—-	IVIG (2)	Doxycycline (17) Others (6)	n/a	A (17) D (0)	—-
19	Marsh et al., 2013 [67]	USA	2004-2010	Retro	HLH-2004	22 n/a	n/a	Secondary (n/a)	n/a	Dexa (22) mPSL (2)	ETO (17) CsA (8) IT-MTX (5) ARA-C (1) RTX (3)	IVIG (?)	Antimicrobial therapy (?)	n/a	A (17) Post-HSCT (16) Pre-HSCT (1) D (5) Pre-HSCT (4) Post-HSCT (1)	- The study’s aim was to assess alemtuzumab as a salvage therapy of refractory HLH after conventional therapy, which seems to be effective for survival until HSCT.
20	Delavigne et al., 2014 [50]	France	2006-2010	Retro	Not specified	32 19:13	Me 62 y	AML (32)	Bacterial (11) Viral (9) Fungal (11) Mucormycosis (1)	Any (13)	ETO (2) CsA (1)	IVIG (22)	Specific antimicrobial therapy	3 mo	A (30) D (2)	- HLH can be encountered in up to 10% of AML patients receiving intensive chemotherapy.
21	Rajajee et al., 2014 [72]	India	2008-2011	Retro	HLH-1994	40 22:18	M 4.3 y [0.1-11]	Secondary (n/a)	n/a	G1: Dexa (22) G2: Dexa (12) G3: Dexa (6)	G1: - (22) G2: ETO+CSA (12) G3: ETO+CSA (6)	G1: IVIG2 (22) G2: - (12) G3: IVIG2 (6)	n/a	12 mo	‘good’ (33) ‘bad’ (7)	- G1: Dexa + IVIG2 (‘good’ outcome: n = 19) - G2: HLH-2004 Prot. (‘good’ outcome: n = 10) - G3: IVIG2 followed by HLH-2004 Prot. (‘good’ outcome: n = 4).
22	Bode et al., 2014 [42]	Germany	1999-2012	Retro	HLH-2004	13 6:7	n/a [0.5-8.2 y]	No disease	Leishmaniasis (13)	Any (7)	HLH-04 (4) CsA (1) MP (1) AZA (1)	IVIG (1)	AmphB (12) Sodium stibogluconate (1)	n/a	A (13)	- Liposomal amphotericin B was successful in visceral leishmaniasis complicated with HLH - The usual dose was 3 mg/kg/d at days 1-5, and 10.
23	Dao et al., 2014 [49]	Vietnam	2010-2012	Pro	HLH-2004	91 53:36	< 15 years [2 mo-14.2 y]	n/a	EBV (40) CMV (6) EBV+CMV (43)	Deaxa (87)	Csa+ETO (50) Csa (7)	-	n/a	n/a	A (64) D (25) ? (2)	—–
24	Kaufman et al., 2014 [63]	USA	?	Retro	sJIA-MAS 2006 HLH-2004	14 7:7	n/a [0.9-16 y]	sJIA	n/a	Any (14)	CsA (10) RTX (1)	IVIG (1) ATG (1)	n/a	n/a	A (12) D (2)	—–
25	Li et al., 2014 [66]	China	2005-2014	Retro	HLH-2004	16 12:4	n/a [15-63]	LA (16)	EBV (7)	Dexa (4)	ECHOP (5) LCHOP (1) ETO (4) CsA (1)	-	n/a	n/a	A (1) D (15)	—–
26	Riviere et al., 2014 [74]	France	2006-2011	Retro	Not specified	162 109:53	Me 48 [35-62]	MA (103) IA (40) AUTO (5) - SLE (3) - AOSD (2) No disease (14)	Bacteria (9) Mycobacteria (13) Viruses (10) -CMV (6) - EBV (2) - Others (2) Parasites (6) Fungi (2)	Any (100)	ETO (81)	IVIG (1)	n/a	Me 4.3 mo	A (94) D (68)	- 61 patients received no specific treatment, but were treated only for their underlying disease.
27	Minoia et al., 2014 [69]	Inter national	> 2002	Retro	Not specified	362 154:208	Me 5.3	sJIA	n/a	Any 341	CsA (213) Biologics (52) - ANK (33) - Others (19) ETO (40) Other drugs (24) Plasmapheresis (14)	IVIG (125)	n/a	n/a	A (319) D (28) ? (15)	- ‘…patients who received etoposide had more severe disease, manifestations, and were more likely to be admitted to the ICU than those who did not. ’
28	Townsend et al., 2015 [83]	USA	2003-2013	Retro	HLH-2004	11 9:2	M 43.9	HIV (9) Transplant (1) No disease (1)	Histoplasmosis (11)	PDN (3) PSL (1)	See notes	IVIG (2)	AmphB (9) Itra (6) Others (2)	30 d	A (6) D (5)	- In total, 5 patients received a treatment with steroids and/or IVIG; 6 patients received no specific immunomodulatory treatment.
29	Yokota et al., 2015 [92]	Japan	2008-2015	Pro	sJIA-MAS 2006	23 10:13	Me 10 y [2-18]	sJIA	n/a	Dexa (14) mPSL (14) TCZ???	CsA (16) Plasmapheresis (5)	-	n/a	17 d (6-45)	A (19) D (2) ? (2)	- All sJIA patients were on treatment with TCZ at the moment of MAS diagnosis.
30	Valade et al., 2015 [84]	France	2007-2014	Retro	HLH-2004	117 83:34	Me 46	HIV (44) MA (85) AUTO (6) Unknown (3)	TB (9) Fungi/parasites (8) Viruses (5) Bacteria (1)	Any (78)	ETO (90)	IVIG (12)	n/a	n/a	A (66) D (51)	Coagulation disorders are common in ICU patients with severe HLH. Fibrinogen <2 g/l and low PT were associated with higher mortality.
31	Shakoory et al., 2016 [76]	International	n/a	Pro	Not specified	43 23:20	M 52.3 y. [18-75]	Sepsis-related MAS	n/a	No	Anakinra (26) Placebo (17)	n/a	n/a	28 d	A (23) D (20)	- Phase III RCT in severe sepsis: anakinra (2.0 mg/kg/h for 72 h) (IV continuously). - 28-day mortality: anakinra vs. placebo (34.6% vs. 64.7%, p = 0.0006).
32	Hernandez-Jimenez et al., 2016 [55]	Spain	2010-2015	Retro	HLH-2004	18 13:18	M 51 y [22-88]	MA (8) IA (6) AOSD (1) No disease (3)	EBV (1) Leishmania (4) CMV (1)	mPSL (6) Dexa (10)	CsA (2) ETO (1) FK506 (1)	IVIG (3)	n/a	n/a	A (10) D (8)	—–
33	Grom et al., 2016 [53]	International	Until 2014	Pro	HLH-2004 MAS-JIA 2006	31 11:20	n/a [2-18 y]	sJIA (21)	EBV (3) CMV (1) HSV (1) Adenovirus (1) Others (2)	Any (30)	ETO (2) CsA (10)	IVIG (6)	GCV (1)	n/a	A (29) D (2)	- This study assessed the occurrence of MAS in patients included in a phase 2/3 trial for the treatment of sJIA with CNK. - 21 and 10 patients categorised as probable or possible MAS, respectively.
34	Yang et al., 2016 [90]	China	2009-2014	Retro		34 21:13	n/a [19-78 y]	LA (34)	EBV (15)	Any (34)	HLH-2004 (2) R-EPOCH (10) EPOCH (10) ECHOP (4) Others (7)	-	n/a	14-694 d	A (12) D (22)	—–
35	Ellis et al., 2016 [51]	Puerto Rico	2008-2013	Retro	HLH-2004	22 12:10	Me 1 y [0.2-17]	No disease (22)	Dengue (22)	Any (16)	ETO (8) Other drugs (8)	IVIG (13)	No	n/a	A (21) D (1)	—–
36	Ahn et al., 2017 [98]	South Korea	2005-2016	Retro	2016 EULAR/ACR/PRINTO MAS criteria	36 2:32	Me 47.5 y. [n/a]	AOSD	n/a	Any (36)	ETO (6) CsA (13) MTX (4) CPM (1) TCZ (3) IFX (1) ANK (1)	IVIG (5)	No	n/a	A (24) D (12	—–
37	Ahn et al., 2017 [40]	Korea	2005-2016	Retro	2016 EULAR/ ACR/PRINTO MAS criteria	54 8:46	M 36.5 y	SLE	n/a	Any (53)	CPM (8) MMF (7) HCQ (14) AZA (1) MTX (1)	-	No	n/a	A (35) D (19)	——
38	Leow et al., 2017 [65]	Singapore	2000-2015	Retro	HLH-94 HLH-2004	14 8:6	Me 8.2 y	n/a	EBV (8)	mPSL (7)	HLH-2004 (9) CsA (2)	IVIG (11)	n/a	n/a	A (6) D (8)	- The study population is represented by critically ill patients admitted to PICU with HLH. - 2 patients were finally diagnosed with FHL - 1 patients underwent HSCT.
39	Haytoglu et al., 2017 [54]	Turkey	2009-2015	Retro	HLH-94	12 5:7	n/a [2.2-7.7 y]	secondary	Leishmania (4) CMV (2) EBV (3) VZV (1) Influenza (1) Acinetobacter spp. + Candida spp. (1) Unknown (1)	Any (12)	HLH 2004 (2) Plasmapheresis (3)	IVIG (7)	AmphB (4) Broad-spectrum antibiotic (1) Antifungal (1) Antiviral (2)	n/a	A (10) D (2)	- The study population is represented by critically ill patients admitted to PICU with HLH.
40	Buda et al., 2018 [44]	Poland	2005-2013	Retro	HLH-2004	24 7:17	M 8.4 y. [1-17]	AUTO (14) - sJIA (12) - SLE (1) -KD (1) IA (4) No disease (6)	‘Myco-bacterium’ (1) ‘Borrelia’ (1) EBV (1) Parvovirus B19 (1)	HD-GCs (16) SD-GCs (6) None (2)	CSA (15) IFX (2) ETO (2) HF (2)	IVIG2 (14)	Broad-spectrum antibiotic (24)	1-8 y	A (24) D (0)	- The authors concluded that treatment of HLH/MAS based on HD-GCs, CSA, and IVIG is an effective therapy in children.
41	Zhang et al., 2018 [94]	China	2010-2015	Retro	HLH-2004	174 111:63	Me 51 y [17-90]	MA (92) AUTO (6) No disease (76)	EBV (44) CMV (11) Others (2)	Any (122)	CsA (1) Chemotherapy (21)	IVIG (31)	Antiviral (?)	n/a	A (72) D (102)	- 5-year OS was 41.8%, and 4-year HLH-free survival was 30.1%.
42	Ezequiel Borgia et al., 2018 [99]	Canada	2002-2012	Retro	Not specified	38 8:30	M 13.7 y (children)	jSLE	Infection (6) - HSV1 (1) - SBEGA (1) - EBV (1) - E. coli (1) - Candida spp. (1) - TB (1)	mPSL (26) PDN (7) Dexa (6)	CsA (11) FK506 (2) ETO (5) ANK (2) ALTZ (1) MTX (1) MMF (19) CPM (11) AZA (11)	IVIG (22)	n/a	M 3.5 y.	A (36) D (2)	—-
43	Zhao et al., 2018 [95]	China	2010-2015	Retro	HLH-2004	179 99:80	Me 2 y. [0.5-9.6]	FHL (19) Secondary (160)	EBV (97)	Dexa (179)	HLH-2004	-	n/a	3 y	A (117) D (45) ? (17)	- HSCT performed in 8 patients (7 alive).
44	Jin et al., 2018 [62]	China	2008-2016	Retro	HLH-2004	42 30:12	Me 34.5 y. [14-78]	LA (42)	EBV (25)	Any (5)	HLH-1994 (9) HLH-04 (5) Chemotherapy (17)	-		n/a	A (15) D (26) ? (1)	- 9 patients underwent HCST - 1-year OS was 15.4%.
45	Yanagisawa et al., 2019 [89]	Japan	2007-2011	Pro	HLH-2004	73 34:39	Me 2.6 y. [0-18]	IA-H (41) FLH (9) Unknown (23)	EBV (41)	Dexa (74)	HLH-2004 (74)	-	n/a	n/a	See notes	- 17 patients (resistant to therapy or with relapse or FLH) underwent HSCT (3-year OS: 64.7%). - 3-year OS: 85.3% (EBV-HLH); 66.7% (FLH); 56.2% (Other HLH).
46	Sato et al., 2019 [75]	Japan	2004-2016	Retro	Preliminary MAS-jSLE 2009	11 2:9	Me 13.0 y [7-17]	jSLE (11)	n/a	Any (11)	MMF (5) CsA (2) AZA (1)	IVIG (2)		n/a	A (11) D (0)	- MAS was observed during the initial stage of jSLE.
47	Jin et al., 2019 [61]	China		Retro	HLH-2004	16 6:10	n/a [1-12 y]	IA (16)	O. tsutsugamushi (16)	Dexa (2)	-	IVIG2 (5)	See notes	n/a	A (10) D (1)	Patients were treated with IV azithromycin for 5 days and, in case of nonresponse, chloramphenicol was used.
48	Chellapan-dian et al., 2019 [47]	Canada	2000-2015	Retro	HLH-2004	32 11:21	Me 1.1 y [0.05-20.6]	LCH (32)	EBV (8) CMV (1) EBV+CMV (1) HHV6+CMV (1) Adeno+EBV (1) Rhino+EBV (1) Candida (1)	Not specified	LCH therapy (13) HLH protocol (16) Other therapies (8)	IVIG (8)	n/a	n/a	A (27) D (5)	- 5 patients underwent HSCT.
49	Song et al., 2019a [78]	China	2017	Retro	HLH-2004	93 54:39	Me 26 y. [1-70]	IA	EBV (93)	Not specified	ETO (52) non-ETO (41)	-	n/a	n/a	See notes	- OS in the ETO group was significantly better than non-ETO patients. - In patients <18 y, the OS in the ETO group was not significantly better than non-ETO patients, in contrast to patients ≥18 years.
50	Song et al., 2019b [79]	China	2011-2018	Retro	HLH-2004	13 (pregnant)	Me 26 y. [20-36]	IA (4) MA (1) AUTO (2) No disease (6)	EBV (2) S. epidermidis (1) M. tuberculosis (1)	Any (13)	ETO (6) Non-ETO (7) +CsA (2)	IVIG (?)	n/a	Me 29 mo (0.4-93)	A (10) D (2) ? (1)	—-
51	Oguz et al., 2019 [70]	Turkey	2011-2016	Retro	HLH-2004	24 13:11	n/a [2-202 mo]	IA (12) AUTO (9) - JIA (5) - KD (1) - DRESS (3) MA (2) No disease (1)	Adenovirus (2) Brucella (2) Salmonella (1) Leishmania (1) EBV (2) EBV+CMV (2) Listeria (1) Parvo (1)		HLH -2004 (8) Plasmapheresis (11)	-	Empiric antibiotic therapy (24)	n/a	A (19) D (5)	—–
52	Bubik et al., 2020 [43]	USA	2004-2018	Retro	HLH-2004	31 19:12	Me 46 y. [18-74]	MA (9) AUTO (8) - AOSD (6) - SLE (1) - Other (1) IA (8) No disease (8)	EBV (5) ‘viral’ (1) Histoplasma (1) Blastomyces (1)	Dexa (31)	HLH-2004 [‘with modifications’] (31)	-	n/a	Me 71 d	A (11) D (20)	- 1-year OS was 35%.
53	Eloseily et al., 2020 [52]	USA	2008-2016	Retro	mainly sJIA-MAS criteria (42)	44 19:25	M 10.3 y. [1-19]	AUTO (27) -sJIA (13) -SLE (5) -MCTD (3) -Others (6) MA (3) Other disease (4) No disease (10)	HSV (2) HHV-6 (4) EBV (3) CMV (1) Mycoplasma (1) Others (8)	Any (28)	Anakinra (44) CsA (10) ETO (4) Other biologics (3)	IVIG (3)	n/a	n/a	A (32) D (12)	- The earlier start of anakinra (≤5 days) was associated with reduced mortality (p = 0.046).
54	Zhou et al., 2020 [96]	China	2017-2019	Retro	HLH-2004	70 45:25	HLH* M 42.2 y [n/a] R-DED* M 44.7 y [n/a]	LA (70)	n/a	Dexa (70)	HLH-1994 (34) R-DED* ETO + ruxolitinib + doxorubicin (36)	-	n/a	HLH-1994 Me 0.2 mo [0.1-12] R-DED* Me 3.3 mo [0.2–12]	n/a	- OS was significantly longer in the R-DED group than in the HLH-94 group (p = 0.003).
55	Wang et al., 2020 [86]	China	2012-2017	Retro	HLH-2004 HLH-2009	55 17:38	32.9-13.7	AOSD	n/a	mPSL (6) PDN (49)	CsA/FK506 (14) ETO (10) TCZ (5) MTX (2) Others (20)	IVIG (31)	n/a	15.6-28.7 mo	A (43) D (12)	——
56	Zou et al., 2020 [97]	China	2012-2018	Retro	sJIA-MAS	80 41:39	Me 6.0 y. Me 1.9 y. Me 11.1 y.	sJIA (53) KD (10) CTD (17)	n/a	Any (79) mPSL (64)	CsA (43) TCZ (8) IFX (2) Others (16) Plasmapheresis (7)	IVIG (44)	n/a	n/a	A (74) D (6)	- 1 patient underwent mesenchimal stem cell transplant.
57	Pal et al., 2020 [71]	India	2008-2018	Retro	HLH-2004 JIA-MAS 2016	31 17:14	Me 5.3 y	sJIA (26) SLE (4) KD (1)	n/a	mPSL (31)	CsA (17)	IVIG (7)	n/a	n/a	A (21) D (10)	——
58	Yin et al., 2020 [91]	China	2015-2019	Retro	HLH-2004	111 57:54	Me 51 y. (39-61)	IA (23) MA (67) AUTO (9) No disease (12)	EBV (52)	Any (111)	HLH-94 (19) Nothing (15) HLH-94 + CHOP-like (40) CHOP-like (27)	IVIG (8)	n/a	83 d	A (38) D (73)	- Low fT3 is frequently found in HLH and is associated with worse outcomes in these patients.
59	Simon et al., 2020 [77]	India	2012-2016	Pro (2-y) + Retro (5-y)	HLH-2004	53 29:24	Me 2 y.	FHL (8) IA (34) AUTO (5) No disease (10)	Bacterial (14) Viral (10) Fungal (2) Parasitic (2) Rickettsia (6)	Any (37)	ETO (19) CsA (21) MTX (4) Nothing (15)	IVIG (15)	n/a	n/a	A (20) D (22) ? (1)	—–
60	Carvalho et al., 2020 [46]	Brazil	2012-2017	Retro	HLH-2004	39 21:18	2.86-2.08 y [0.5-9.1]	IA (39)	Leishmania (39)	Dexa (15)	Immuno-suppressive drugs (6)	-	Antibiotic (31) N-m-Glu (18) AmphB (21)	1-year	A (39) D (0)	—-
61	Suarez Hormiga et al., 2020 [80]	Spain	1998-2018	Retro	HLH-2004	15 11:4	M 42 y [20-67]	HIV Patients (15)	HHV (5) P. jiroveci (2) EBV (1) CMV (2) Micobacteria (2) Blastocystis (1) Unknown (2)	Any (15)	ETO (3) CsaA (1) ADM (1) DXR (1) CPM (1) No drugs (6)	IVIG (2)	Anti-retroviral therapy (15)	30 days	A (4) D (11)	—-
62	Yoon et al., 2020 [93]	Korea	2017-2021	Pro	HLH-2004	55 35:20	n/a	MA (40) IA (12) SLE (1) Nothing (2)	n/a	Dexa (55)	HLH-94 (55) + CHT (40)	-	n/a	Me 14.5 mo	A (31) D (24)	—-

A, alive; ADM, adriamycin; ALTZ, alemtuzumab; AML, acute myeloid leukaemia; ANCA-V, ANCA-associated vasculitis; ANK, anakinra; AOSD, adult-onset Still disease; ARA-C, Cytarabine; ATG, anti-thymocyte globulin; AZA, azathioprine; CHT, chemotherapy (disease-specific); CMV, Cytomegalovirus; CPM, cyclophosphamide; CsA, cyclosporin A; CSS, corticosteroid; CTD, connective tissue disease; D, death; DEXA, dexamethasone; Doxo, doxorubicin; EBV, Epstein–Barr virus; ECHOP, etoposide, cyclophosphamide, vincristine, prednisone, doxorubicin hydrochloride; ETN, etanercept; ETO, etoposide; FHL, familial hemophagocytic lymphohistiocytosis; FLU, Fluorouracil; HCQ, hydroxychloroquine; HF, hemofiltration; HHV, human herpes virus; HLH, hemophagocytic lymphohistiocytosis; HSCT, haematopoietic stem cell transplantation; HSV, herpes-simplex virus; IA, immune disease–associated; itMTX, intrathecal methotrexate; IVIG, Intravenous immunoglobulin; JIA, juvenile idiopathic arthritis; KD, Kawasaki disease; LA, lymphoma-associated; LCH, Langerhans cell histiocytosis; LCH, Langerhans cell histiocytosis; MA, malignancy-associated; MAS, macrophage activation syndrome; MCTD, mix connective tissue disease; MMF, mycophenolate mofetil; MP, mercaptopurine; MRSA, methicillin-resistant staff aureus; PF, plasmapheresis; PICU, pediatric intensive care unit; PID, primary immunodeficiency; PSL, prednisolone; RCT, randomized controlled trial; R-DED, R-DED regimen (ruxolitinib, doxorubicin, etoposide, and dexamethasone); R-EPOCH, rituximab-etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin hydrochloride; RSV, respiratory syncytial virus, SBEGA, Streptococcus beta-hemolyticus group A; RTX, rituximab; SLE, systemic lupus erythematosus; TB, tuberculosis; TCZ, tocilizumab, IFX, infliximab; TPO, thrombopoietin receptor against; VBL, vinblastine; VCR, Vincristine; VZV, varicella zoster virus; XLP, X-linked lymphoproliferative disease.

DISCUSSION

Early therapeutic management of primary HLH

As already mentioned, primary HLH diseases are the forms associated with a genetic cause. Historically, they were also defined as FHL, but these diseases often appear as sporadic cases without a clear family history. However, they may also be triggered by an infection [8].

Genetic diseases that cause impairment in the cytolytic functions of cytotoxic T cells and/or Natural killer (NK) cells are part of the spectrum of primary HLH. The causative mutations are usually inherited with an autosomal recessive pattern and cause early-onset diseases that manifest in the paediatric age and are often elicited by viral (especially Epstein–Barr virus [EBV]) infections. PRF1 (protein: perforin 1; disease: FHL2), UNC13D (protein: Munc 13-4; disease: FHL3), STX11 (protein: syntaxin 11; disease: FHL4), and STXBP2 (protein: Munc 18-2; disease: FHL5) are the genes predominantly involved in primary HLH forms not associated with hypopigmentation. Indeed, some primary immunodeficiencies with partial albinism represent the second main group included in the primary HLH group. These are Griscelli syndrome (GS, gene: RAB27A), Chediak–Higashi syndrome (CH, gene: LYST), and Hermansky–Pudlak syndrome 2 (HP, gene: AP3B1). Additional X-linked disorders, such as XLP (X-linked lymphoproliferative disease) XLP-1 (gene: SH2DIA), XLP-2 (gene: XIAP/BIRC), and XMEN (X-linked immunodeficiency with magnesium defect, Epstein-Barr virus infection, and neoplasia (gene: MAGT1)) syndromes and autosomal recessive disorders, such as interleukin 2-inducible T-cell kinase deficiency and CD27 deficiency among others, are also associated with HLH occurrence, through more complicated mechanisms of predisposition to this disease [14,16]. Autosomal dominant gain-of-function pathogenic variants in the NLRC4 gene (coding for a protein involved in the inflammasome activation) can also cause recurrent HLH/MAS episodes, in addition to early-onset enterocolitis [8,11,12,15,17,18,[100], [101], [102], [103], [104], [105], [106]].

The therapy for primary HLH should aim at 3 main principles: (i) ‘correction’ of the underlying genetic defect, whenever this is possible; (ii) treatment of infectious trigger(s), if any are identified; and (iii) immunosuppression. As highlighted in the HLH-2004 protocol, many patients with genetic HLH require HSCT for long-term survival [16,19,101].

As mentioned, Table 1 summarises the articles (including >10 patients) focused (or at least provide enough information) on the pre-HSCT treatment and outcome of primary HLH. The majority of patients have been managed according to the HLH-94 or HLH-2004 protocols (based on the administration of DEXA, CsA, and intrathecal MTX), which enable at least 65% of children affected with primary HLH to undergo HSCT. However, this success rate is likely higher, considering that some studies included different therapeutic protocols and the waiting time before HSCT is often not specified.

According to the overall results of the clinical trials based on the HLH-2004 protocol, pre-HSCT mortality has been reduced by around 20% over the last decades, which is partly due to the reduction in the time to receive HSCT. Notably, the early introduction of CsA in the HLH-2004 protocol did not result in a statistically significant reduction of pre-HSCT mortality, but a risk-group stratification of children with confirmed FHL (in addition to earlier HSCT) may further improve the outcome for these patients, whose 5-year overall survival post-HSCT was 71% [24,107]. Recently, Locatelli et al. [31] described the results of the NI-0501-04 trial, a phase 2-3, open-label, single-group study aimed at assessing the efficacy of emapalumab in patients affected with FHL. Emapalumab is an anti–interferon-γ monoclonal antibody, which can bind both free and receptor-bound interferon-γ and, thus, neutralise its actions. Emapalumab was found to be effective in patients with primary HLH and displayed a low level of toxicity, even in conjunction with other immunosuppressive therapies. Emapalumab can be indicated to treat patients with refractory, recurrent, or progressive forms of primary HLH or in case of intolerance to conventional therapies [108]. An interesting real-world analysis of the emapalumab effectiveness in patients with primary HLH showed results consistent with the aforementioned pivotal trial [109].

In addition to primary immune deficiencies (PIDs), HLH/MAS have been reported in patients with other types of inherited disorders, especially metabolic diseases. Inborn errors of metabolism may cause HLH/MAS presumably due to the intracellular accumulation of nondegraded substrates in the lysosomes and other organelles of the macrophages, which may overstimulate the inflammasome pathway, leading to disproportionate inflammatory response and HLH/MAS [8]. Although there are reports of different types of inherited metabolic conditions presenting with HLH/MAS, lysinuric protein intolerance may be the disease most frequently reported in association with HLH/MAS, probably due to mutations in the SLC7A7 gene [110]. However, the range of reported diseases is broad and spans from lysosomal storage diseases, such as Gaucher disease [111] and Wolman disease [112], to biotinidase deficiency [113], organic acidemias [114], galactosemia [115], Pearson syndrome, and glycogen storage diseases [116]. However, no specific information on the specific therapeutic approach came out from the present systematic literature review in these diseases because they are extremely rare and all the available articles in this regard were case reports [8,100,[107], [108], [109],117,118].

Early therapy in secondary HLH/MAS

In acquired (secondary) groups of HLH/MAS, the overwhelming immune activation leading to systemic cytokine storm often develops in the presence of an underlying disease, which is usually represented by an autoimmune (most often rheumatologic) disorder or a malignancy. However, other conditions resulting in secondary immunosuppression (as it occurs in critically ill patients or those affected with HIV infection) can be observed as predisposing conditions. Moreover, as previously mentioned, an ongoing infection (viral in most cases) can act as a trigger factor for secondary HLH/MAS. Nevertheless, in some patients with HLH/MAS, no underlying condition and/or infection can be identified [[119], [120], [121]].

Patients affected with oncological disorders

Among malignancies, lymphomas (especially NK and T forms) represent the most frequent pathologic background of HLH, which is frequently triggered by different types of infections, with herpes viruses being the most frequently implicated agents [122].

Whenever HSV and CMV are detected, acyclovir and ganciclovir are usually prescribed, respectively. However, EBV infection is detected in most cases of HLH developing in oncological patients EBV infection, and no specific antiviral conventional therapy is available. IVIGs have been used as supportive therapy in the pharmacologic approach to HLH in these immunocompromised patients, in addition to establishing a broad-spectrum empirical and, whenever possible, targeted antimicrobial therapies [123], as shown in large part of the articles included in this systematic review. Among 60 articles included in Table 2, IVIGs have been variably employed in 38 studies, which included different types of secondary HLH/MAS [42,44,45,47,48,[50], [51], [52], [53], [54], [55], [56], [57],[59], [60], [61],[63], [64], [65],[67], [68], [69],71,72,74,75,77,79,80,[83], [84], [85], [86],91,94,97,98].

Most of the selected studies included different types of patients with secondary HLH and a few ones provided information about a specific pathologic background. Overall, most studies including oncological HLH/MAS cases reported a therapeutic approach based on chemotherapy regimens or their intensification specific to the type of malignancy [66,85,87,88,94]. For instance, in the study by Buyse et al. [45], 29 patients (out of 56; 51.8%) died of HLH and, among them, 19 (65.5%) were affected by neoplastic lymphoproliferative disorders or other malignancies. In addition to the specific anticancer chemotherapy, in this study, 45 patients (out of a total of 56 patients, who were not all oncological) received etoposide and some also received corticosteroids (n = 31). In the studies by Valade et al. [84] the majority of the study participants (85 of 117) were patients with malignancy-related HLH: in most cases, the treatment was represented by etoposide (90 of 117) and/or steroids (78 of 117). A similar approach has been reported by Riviere et al. [74]. In other cases, HLH-94 or HLH-2004 represented the main treatment [47,91], which was integrated with tumour-specific therapies in some cases [93].

Most studies related to malignancy-associated HLH focused on lymphomas, as already mentioned. Some studies provided a detailed description of the therapeutic approach in these patients. For instance, the study by Wang et al. [87] despite including different types of HLH, described the therapeutic regimens applied to 10 patients with lymphoma-associated HLH, namely the ECHOP (recombinant human endostatin (E) in combination with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP)) like protocols that consisted of etoposide (100 mg/m²) and methylprednisolone (5 mg/kg) for the first 3 days, in addition cyclophosphamide (750 mg/m²), vincristine (1.4 mg/m²), and epirubicin (75/m²) administered only on day 1. These medications were gradually reduced and then stopped within 3 weeks. The survival rate in these patients was 70%. Notably, the main aim of this prospective, randomised, and controlled trial was to assess the efficacy and safety of the recombinant human thrombopoietin in this pathologic setting, which reduced the frequency of severe bleeding and the need for platelet transfusion. The study by Li et al. [66] retrospectively analysed 16 patients with lymphoma diagnosed with HLH. Several patients (n = 7) received chemotherapy regimens based on etoposide or L-asparaginase combined with glucocorticoids, vincristine, cyclophosphamide, and adriamycin, whereas other 6 patients variably received etoposide and/or glucocorticoid and/or cyclosporine A. Those diagnosed with a concomitant EBV active infection (n = 7) were treated with ganciclovir. Three patients died before starting any treatment. In total, only one patient survived. Similarly, Yang et al. [90] focused on patients with lymphoma-associated HLH (n = 34). Most patients were treated with polychemotherapy regimens usually used for these malignancies, such as ECHOP, R-CHOP (rituximab plus CHOP), and R-EPOCH (rittuximab-etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin hydrochloride), in addition to prednisone and, in most of the cases, etoposide. EBV infection was detected in as many as 15 patients, and this was a negative prognostic factor for the outcome. The overall survival rate was equal to 35.3%. Jin et al. [61] described 42 patients affected with NK–T-cell lymphoma who developed HLH. In this study, this complication was treated with HLH-94 or HLH-04 protocols in 14 cases, whereas 17 patients were treated with tumour-specific chemotherapy and others received steroids only or other therapies. The authors also reported that patients with relapsing or poorly responsive HLH (n = 15) received a salvage treatment, which was mainly represented by a regimen, containing doxorubicin, etoposide, and methylprednisolone (DEP). The overall survival was around 35% as well. More recently, Zhou et al. described a larger case series of lymphoma-associated HLH (n = 70). Almost 50% of these patients (n = 34) received the HLH-94 protocols, whereas the other 36 patients were treated according to the so-called ruxolitinib, doxorubicin, etoposide, and dexamethasone protocol, which included etoposide, doxorubicin, and ruxolitinib, and resulted in a longer survival compared to the former treatment [96]. A recent study by Yin et al. [91] included 67 patients with lymphoma-related HLH; they were treated with CHOP-like chemotherapy (n = 67), which was integrated with the HLH-94 protocol in some cases (n = 40). The overall survival was around 35% again, but not all study participants (n = 111) were diagnosed with lymphoma. The HLH-94 protocol was administered to all patients with malignancy-related HLH (n = 40) described by Yoon et al [93]; here, it was integrated with tumour-specific chemotherapy.

There are also a couple of studies focused on specific malignancies other than lymphomas. The one by Delavigne et al. [50] addressed HLH in patients affected with acute myeloid leukaemia. Among the 32 patients fulfilling the diagnostic criteria for HLH, most patients developed this complication during the first induction therapy. The vast majority (>90%) received treatment including corticosteroids and/or IVIG; only 3 patients also received etoposide and/or cyclosporin A; however, in all patients, one bacterial, viral, or fungal infection was identified and properly treated. Here, the survival rate was around 94%. In the study by Chellapandian et al. [47] (including patients with Langerhans histiocytosis developing HLH), the HLH protocols were applied to 50%, whereas the remaining patients were treated by targeted tumour protocols or other chemotherapy regimens. The survival rate resulted to be higher because it was >80%.

In summary, etoposide/steroids-based regimens resulted in the most frequently employed treatments in the setting of malignancy-associated HLH. Most of these patients had lymphomas, and despite the treatment, the overall survival was <50% in several studies.

Patients affected with rheumatologic disorders

The approach to HLH/MAS in patients with rheumatic diseases is different from that observed in secondary forms of oncological patients. Overall, the HLH-94/04 protocols and etoposide are used much less frequently to treat HLH/MAS in patients with an underlying rheumatic disorder, no matter whether they are adults or children [43,44,52,53,63,68,69,75,87,91,92,94,98].

In many articles included in this systematic review, rheumatologic HLH/MAS is described in aggregation with other types of secondary HLH. However, there are several studies that were focused exclusively on patients with rheumatic developing HLH/MAS. Some provided an aggregated analysis of patients affected with different rheumatic diseases [44,52,[68], [69], [70], [71]].

Among biologics, anakinra gained a particular interest. In the study by Eloseily et al. [52] all 27 rheumatic children (sJIA, n = 13; pSLE, n = 5; other diseases, n = 9) received anakinra (at a dosage ranging between 2 and 11 mg/kg/day) in addition to variable regimens of conventional therapy. Notably, an earlier start of anakinra was associated with a significant reduction of the mortality rate in this study. Overall, the survival rate in patients with rheumatic diseases was 86%, and in detail, it was 100% and 70% in patients with systemic JIA and paediatric-onset systemic lupus erythematosus (SLE), respectively. This study also included 17 patients with nonrheumatic conditions (including some with cancer), whose overall survival rate was around 50%. The use of anakinra (at a dosage of 2 mg/kg/day) after conventional HLH therapy was previously reported by Miettunen et al. in 12 paediatric patients (most of them with sJIA); here, all the patients survived [68]. Anakinra was also administered to 33 of the 362 patients with sJIA-associated HLH/MAS described by Minoia et al. [69]; it was the biological drug most frequently used in this study. There were 40 patients (11.8% of the total) treated with etoposide and that was associated with a greater rate of complications and admission to the ICU. Other smaller case series also supported the use of anakinra in the initial medical management of secondary HLH, which may also reduce the use of chemotherapy (etoposide), as reported by Bami et al. [124] for instance.

Some studies focused on specific or individual rheumatic diseases. Among these, most investigated patients with sJIA and adult-onset Still disease (AOSD) developing HLH/MAS [53,63,69,86,92,98]. Kaufman et al. [63] reported 14 patients treated with corticosteroids, 10 of whom were also given cyclosporin A: 85% of them survived. The aforementioned study by Minoia et al. [69] identified corticosteroids and cyclosporin A as the main medications used for HLH in patients with sJIA, with an overall survival rate of at least 88%. Similarly, Yokota et al. [92] found a survival rate of >80% in their case series of 23 patients with sJIA-related HLH/MAS. Regarding the adult population, Ahn et al. [98] described 36 patients with AOSD-related HLH/MAS, who achieved an overall survival rate of 66%. All patients received corticosteroids, and 13 of them cyclosporin A as well; in general, the therapeutic approach was more heterogeneous and variably included etoposide, methotrexate, cyclophosphamide, and some biological drugs. A 78% overall survival rate was reported by Wang et al. [86] in a cohort of 55 patients with AOSD-related HLH/MAS. Although a steroidal treatment was administered to all patients, the associated therapy variably included calcineurin inhibitors, etoposide, methotrexate, and other conventional and biological drugs. Other smaller case series, such as the one by Hot et al. [125] reported HLH occurrence in patients with AOSD with a positive outcome, in general.

A few studies were also focused on SLE-related HLH/MAS [75,94,98]. Ahn et al. [98] described their experience with 54 adult patients, who were variably treated with cyclophosphamide, mycophenolate mofetil (MMF), hydroxychloroquine, or other drugs, in addition to steroids; here, the overall survival was around 65%. Eventually, Sato et al. [75] described 11 children with paediatric-onset SLE-related HLH/MAS with a 100% survival after treatment with steroids, in variable conjunction with MMF, cyclosporin A, or azathioprine in some cases (n = 8).

With regard to the use of IVIG, by pooling the results of studies focused on patients with rheumatic HLH/MAS, around 40% of them had received this treatment. However, in almost all cases, the specific dosage was not specified. In particular, it is therefore not clear whether a 2-day scheme (1 g/kg/day) or a 5-day regimen (0.4-1 g/kg/day) has been used in individual studies. Anyway, the efficacy of IVIG in rheumatologic patients with MAS has not been demonstrated because more large-scale studies are needed [[126], [127], [128]].

The initial therapeutic options for patients with rheumatic developing HLH/MAS that resulted from our systematic review are summarised in Table 3.

Table 3.

Summary of the initial therapeutic options in patients with suspected HLH/MAS

Pediatrics		Adult
Supportive Care: Based on individual needs, supportive care may include IV access, empiric antibiotics for suspected or confirmed infections. Prophylactic antimicrobial and antiviral Tx, cardiopulmonary support, GI and DVT prophylaxis. A multi-disciplinary approach is encouraged when feasible
Initial Treatment Dosing
Administer systemic glucocorticoids, and add* either or both (IVIg, anakinra) depending on individual indications, rate of progression, and local availability: 1-Systemic corticosteroids: a.po prednisone 1-2mg/kg/day b.IV/PO dexamethasone 10mg/m2 c.IV glucocorticoid pulse therapy: methylprednisolone 10-30 mg/kg/day, maximum of 1000 mg per day for 1-3 days followed by oral prednisone/prednisolone 1-2 mg/kg/day (single dose or divided) 2-IV immunoglobulins 1g/kg/day x 2 days 3-Anakinra 5-10 mg/kg/day iv (preferred) or sq		Administer systemic glucocorticoids, and add either or both (IVIg, anakinra) depending on individual indications, rate of progression, and local availability: 1-Systemic corticosteroids: a.po prednisone 1-1.5mg/kg/day (should we synchronize?) b.IV/PO dexamethasone 10mg/m2 c.IV glucocorticoid pulse therapy: methylprednisolone 250-1000 mg/kg/day, for 1-3 days followed by oral prednisone/prednisolone 1 mg/kg/day (single dose or divided) 2-IV immunoglobulins 400 mg - 1g/kg/day x 2-5 days 3-Anakinra 5-10 mg/kg/day iv (preferred) or sq
Treatment Escalation
Indication: -Partial or no response to initial treatment -A definite diagnosis necessitates context specific therapy -New or worsening organ failure Options: -Maximize the dose of the current treatments. -Consider transfer to ICU -Add a course of IV pulse methylprednisolone 10-30mg/kg/day, maximum of 1000 mg per day for 1-3 days -Consider etoposide 100-150mg/m2, especially if triggered by EBV, malignancy onset/relapse, or FHL is suspected based on history.		Indication: -Partial or no response to initial treatment -A definite diagnosis necessitates context specific therapy -New or worsening organ failure Options: -Maximize the dose of the current treatments. -Consider transfer to ICU -Add a course of IV pulse methylprednisolone 250-1000 mg/day for 1-3 days -Add etoposide 100-150mg/m2, especially if EBV or malignancy is suspected based on history.
Other Treatment Options (in Consultation with Experts)
Cyclosporine	CsA 3-5 mg/kg/day IV or PO CSA (aiming at 200 mg/L through value)	CsA 2-7 mg/kg/day IV or PO (aiming at 200 mg/L through value)
Rituximab	Rituximab: 375mg/m2 infusion	Rituximab: 375mg/m2 infusion
Emapalumab	Emapalumab: Starting dose 1 mg/kg every 3 days than advance per protocol up to 10 mg/kg every 3 days	No current dosing available in adults.
Ruxolitinib	No dosing in children with HLH/MAS	Ruxolutinib: 0.3 mg/kg/d on days 1-14.

IV, intra venous; PO, per oral; SQ, sub-cutaneous; CsA, cyclosporin A; EBV, Epstein–Barr virus; HLH/MAS, hemophagocytic lymphohistiocytosis and macrophage activation syndrome.

Patients affected with specific infectious diseases

Secondary HLH can be observed in patients without any underlying and/or chronic medical conditions. In some of these patients, an infectious trigger can be detected, and a large number of infectious agents have been implicated in HLH, as previously stated [1,2]. However, some particular infections have been associated with the occurrence of HLH in previously healthy patients. These cases of secondary HLH can be successfully treated with the appropriate antimicrobial drugs, which are usually able to resolve this complication when their administration is timely.

Visceral leishmaniasis is one of the infections that may lead to secondary HLH. The differentiation of HLH-MAS caused by leishmaniasis from HLH-MAS primarily related to a rheumatic disease can be challenging, and the clinical and laboratory features can be indistinguishable. Bode et al. [42] described 13 patients of these cases, who were all successfully treated with liposomal amphotericin B at the usual dosage of 3 mg/kg/day for 5 days, with one additional dose at day 10. However, some patients were also given immunosuppressive drugs (including MMF, cyclosporin A, or azathioprine) and some (n = 4) were even treated according to the HLH-04 protocol. Additional Leishmania spp.-related cases were described in studies including different types of HLH, but no outcome data are mentioned [54,55,70]. Carvalho et al. [46] reported 39 cases of HLH in patients with visceral leishmaniasis. Many of these patients (n = 21) were treated with amphotericin B deoxycholate (1 mg/kg/day for 14 days), but many others (n = 18) received N-methylglucamine antimony (20 mg/kg/day for 20 consecutive days). Dexamethasone was prescribed in 15 patients, and 6 patients also received some immunosuppressive therapy.

Another interesting example of infection potentially associated with secondary HLH is represented by scrub typhus, usually caused by Orientia tsutsugamushi. There are a couple of reports describing this clinical scenario [61,88]. Jin et al. [61] published a case series including 16 patients: all of them survived, except for one. All these patients received appropriate antibiotic treatment, consisting of intravenous azithromycin (10 mg/kg/day for 5 consecutive days) in 6 patients, and also intravenous chloramphenicol (25-50 mg/kg/day for 7-10 days) in those 10 patients who did not respond to azithromycin. Patients developing more severe HLH also received dexamethasone and/or IVIG (500 mg/kg/day for 2 consecutive days). It is also important to mention cases of HLH associated with human monocytic ehrlichiosis, although no papers fulfilled the inclusion criteria of this systematic literature review. Cabler et al. [129] described 8 cases of HLH from a case series of 49 patients who developed this infection. These patients with HLH all survived except one; only 3 of them were treated with chemotherapy in addition to doxycycline, whereas 5 successfully received antibiotic therapy only.

Another pathologic setting of HLH-like disease related to infections can be represented by patients with sepsis. Recently, Shakoory et al. [76] analysed data from a phase III trial and observed a statistically significant and consistent reduction of the 28-day mortality (34.6% vs 64.7%, respectively) in HLH/MAS septic patients who received anakinra (2 mg/kg/h IV for 72 hours) compared to controls. This study highlights the potential role of IL-1 inhibition in the treatment of patients with sepsis, who show signs of HLH-MAS. Other smaller case series described positive outcomes with anakinra in critically ill children developing secondary HLH in the clinical setting of sepsis with multiple organ dysfunction [130].

CONCLUSION

HLH/MAS is a clinically and etiologically heterogeneous condition, wherein the initial therapeutic approach varies according to the specific pathologic setting.

Primary HLH (or FHL) is typically managed according to the HLH-94 or HLH-2004 protocols, with the aim of bringing the patient to HSCT as soon as possible after remission. Recently, anti–interferon-γ (emapalumab) opened new perspectives, especially in children with refractory, recurrent, or progressive forms of primary HLH, or in patients with intolerance to conventional therapies.

Secondary forms of HLH/MAS can arise in 3 main clinical contexts, namely in patients affected with rheumatic disorders, oncological patients (especially those affected with lymphomas), and previously healthy people developing specific (eg, leishmaniasis, scrub typhus) and/or severe (eg, sepsis) infections.

Overall, malignancy-related HLH/MAS can be variably treated through intensification of conventional chemotherapy, use/addition of etoposide, or the administration of treatments according to the HLH-94 or HLH-2004 protocols, which in some cases is integrated with tumour-specific therapies. In patients with rheumatic HLH/MAS, the therapeutic approach mainly consists of steroids, especially methylprednisolone, which could be variably combined with cyclosporin A and/or biological drugs, rather than etoposide and HLH protocols. HLH/MAS in previously healthy patients is mainly treated with specific antimicrobial therapy without using immunosuppressive drugs, except for steroids in some cases. Intravenous immunoglobulin can find variable application in all these 3 HLH/MAS categories as both immunomodulatory tool and anti-infectious prophylaxis.

Despite the survival improvements achieved in the last decades in patients with HLH/MAS in general, there is a need for prospective and controlled studies focused on each specific group of HLH/MAS.

Editor disclosure

The peer review process did not involve Editorial Board Member Angelo Ravelli and Associate Editor Raphaela Goldbach-Mansky, and the editorial decision-making was led by editors who were not involved in the creation of this manuscript.

Competing interests

SWC received research support from Novartis and AB2Bio, has performed consulting for Simcha and Apollo therapeutics, and received travel support from Sobi. RG-M has received research support from IFM, Lilly, Sobi, and Regeneron, and participates on a DSMB for AstraZeneca. MH has received research support from Incyte. RAM has received honoraria for PracticePoint Communications and participates on the DSMB for Horizon. AR has received honoraria from AbbVie, Alexion, Novartis, Pfizer, Reckitt, Benckiser, and Sobi.