Adjunctive Therapeutic Plasma Exchange in Refractory Adult‐Onset Still's Disease Complicated by Secondary Macrophage Activation Syndrome: A Single‐Center Experience

ABSTRACT

Introduction

Adult‐onset Still's disease (AOSD) complicated by macrophage activation syndrome (MAS) carries substantial mortality. The role of therapeutic plasma exchange (TPE) remains uncertain.

Methods

We retrospectively analyzed patients with AOSD‐MAS treated with TPE at a single‐center.

Results

All five patients had arthralgia, a salmon‐pink rash, extreme hyperferritinemia, and multiorgan dysfunction including liver dysfunction. Prior to TPE, all patients received pulse glucocorticoids; three also received calcineurin inhibitors and one tocilizumab. TPE was initiated at a median of 12.0 days after diagnosis; the patients underwent a median of six sessions without major adverse events. Following TPE, fever and rash resolved; C‐reactive protein (CRP), ferritin, and liver enzyme levels decreased without further corticosteroid escalation during and immediately after TPE. At 1 month, four patients achieved CRP < 1.0 mg/dL, four recovered with tapered immunosuppression. One patient died from cytomegalovirus infection despite MAS improvement.

Conclusions

Adjunctive TPE was associated with rapid clinical and biochemical improvement.

1. Introduction

Adult‐onset Still's disease (AOSD) is a rare systemic autoinflammatory disorder characterized by high‐spiking fever, an evanescent salmon‐colored rash, arthralgia, lymphadenopathy, markedly elevated acute‐phase reactants with leukocytosis, and striking hyperferritinemia [1]. Although its precise etiology remains elusive, the current evidence implicates innate immunity, particularly macrophage activation and natural killer (NK) cell dysfunction, together with a downstream cytokine and chemokine cascade with contributions from adaptive immune pathways [2]. AOSD is strongly associated with macrophage activation syndrome (MAS), one of the most life‐threatening complications, reported in up to 15% of patients [3]. Secondary MAS in patients with AOSD (AOSD‐MAS) is considered to result from aberrant activation of the innate immune system and cytokine storm. The mortality rate of MAS is approximately 10%–20% [4, 5]. High‐dose corticosteroids are the first‐line treatment of AOSD. In refractory disease, conventional immunosuppressants (e.g., methotrexate, cyclosporin, and tacrolimus) and biologic agents such as interleukin (IL)‐1 inhibitors (anakinra and canakinumab) or IL‐6 receptor inhibitors (tocilizumab) are commonly used [6, 7]. Recent guidelines and reviews have summarized advances in AOSD pathogenesis and therapeutic options, including biologics [8, 9, 10]. However, regarding severe AOSD‐MAS, optimal management remains uncertain; moreover, treatment beyond high‐dose steroids and biologic agents is not standardized. Therapeutic plasma exchange (TPE) is not an established standard therapy for AOSD; the existing guidelines have not incorporated TPE into treatment algorithms [8, 9, 10, 11]. Nonetheless, TPE can mechanically remove complement components, immune complexes, and soluble inflammatory mediators. Small case series suggest that adjunctive TPE may improve outcomes in patients with severe or treatment‐resistant AOSD, such as complicated by MAS [12, 13, 14]. Therefore, TPE may represent a promising life‐saving option in patients with refractory AOSD‐MAS driven by neutrophil activation and cytokine storm.

In view of the above and due to the paucity of literature on TPE in the treatment of AOSD‐MAS, we retrospectively reviewed five patients treated at our hospital between 2018 and 2024 with a combination of TPE, corticosteroids, and immunosuppressive agents and analyzed their clinical features. Furthermore, we determined the feasibility of TPE‐based combination therapy including adjunctive immunosuppressive agents in the effective treatment of AOSD‐MAS.

2. Materials and Methods

2.1. Patients and Data Collection

This was a single‐center case series. Five patients with treatment‐resistant AOSD‐MAS who underwent TPE between 2018 and 2024 at Showa Medical University Hospital were retrospectively evaluated. We collected patient characteristics and clinical parameters such as sex, age, body height and weight, white blood cell count (WBC), C‐reactive protein (CRP), liver enzymes, lactate dehydrogenase (LDH), and ferritin levels pre‐ and post‐TPE. This study was performed in accordance with the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of Showa Medical University (approval number: 2025‐0101). Informed consent was obtained from the study participants through an opt‐out process.

2.2. TPE Settings and Procedure

All patients underwent similar TPE procedures every other day. TPE was generally performed as a veno‐venous procedure; an 11‐Fr double‐lumen hemodialysis catheter was used in all patients. Plasma was separated using a membrane plasma separator (Plasmaflo OP‐05D, Asahi Kasei Medical, Japan; or Plasmacure PE‐05, SB‐Kawasumi Laboratories, Japan). Plasma exchange (PE) was performed using the KM‐9000 system (SB‐Kawasumi Laboratories, Japan). The blood and plasma separation flow rates were set at 100–120 mL/min and 30–36 mL/min (PP/BP ratio: 30%), respectively. Nafamostat mesylate (30 mg/h) was used as an anticoagulant during TPE. Replacement was performed with fresh frozen plasma (FFP) provided by the Japan Red Cross Society (Tokyo, Japan) to prevent bleeding complications caused by thrombocytopenia. The target amount of replacement fluid was set to 1.0× the plasma volume (PV) of each patient, with PV calculated using the following equation: PV = body weight [kg]/13 × (100 − hematocrit [%])/100 [L] [15]. Prophylactic intravenous 8.5% calcium gluconate infusion was given through the return line over the duration of TPE to avoid FFP‐induced citrate toxicity symptoms (paresthesia, tingling, muscle cramps, nausea) or ionized calcium levels below the reference range. The median infusion rate of calcium gluconate was 25 mL/h. All TPE procedures were initiated in an inpatient setting. Discontinuation of TPE, which has been reported as a marker for disease activity and treatment response [16, 17, 18], was considered when serum ferritin showed a ≥ 50% reduction from the pre‐TPE peak value, sustained over at least two consecutive measurements, and was accompanied by improvement in systemic inflammation (e.g., decreasing CRP) and clinical symptoms.

2.3. Definition

The diagnosis of AOSD was made based on the Yamaguchi criteria [19], and MAS was defined clinically with reference to the Ravelli's MAS criteria [20, 21] (e.g., hyperferritinemia, cytopenia, elevated liver enzymes). The primary outcome was remission, defined as CRP lower than 1 mg/dL [22] and the resolution of Still's disease‐related symptoms 1 month after TPE completion. The database did not include erythrocyte sedimentation rate. Improvement in liver function was defined as a ≥ 50% reduction in serum aspartate aminotransferase (AST) and/or alanine aminotransferase (ALT) from the pre‐TPE peak value, sustained over at least two consecutive measurements. Pre‐TPE was defined as immediately before the start of TPE; post‐TPE was defined as the end of all TPE sessions. Safety outcomes included allergic reactions, hypocalcaemia, shivering, and ultrafiltration‐related hypotension. Adverse events (AEs) were graded according to previously published criteria as follows: mild/Grade I (no intervention), moderate/Grade II (intervention required; treatment completed), severe/Grade III (procedure interrupted or abandoned), and fatal/Grade IV [23, 24]. Major AEs were defined as severe (Grade III) or fatal (Grade IV) AEs. We adopted these definitions.

2.4. Statistical Analysis

Continuous variables are expressed as medians (interquartile range); categorical data are expressed as numbers (%). The Wilcoxon signed‐rank test was used to assess values before versus after TPE, with a two‐tailed p value < 0.05 considered statistically significant. All analyses were performed using JMP version 18.2 (SAS Institute Inc., Cary, NC, USA).

3. Results

3.1. Study Population

Five female patients (median age: 66.0 years) were included. Four cases met the MAS criteria at the time of AOSD diagnosis, whereas Case 3 met the MAS criteria 2 weeks after AOSD diagnosis. Arthralgia was universal at presentation. High fever (over 38°C) occurred in three, sore throat in three, the characteristic salmon‐pink rash in all five, and lymphadenopathy in two. The median WBC was 24.3 (11.0, 27.6) × 10³/μL, platelet count 18.2 (14.0, 26.4) × 10⁴/μL, and CRP was 25.1 (19.7, 32.7) mg/dL, indicating marked inflammatory activity at AOSD diagnosis. LDH, 552.0 (488.0, 836.5) IU/L, and liver enzymes, such as AST, 62.0 (43.5, 121.5) U/L and ALT, 48.0 (31.0, 80.5) U/L, were elevated. The peak serum ferritin at diagnosis was extremely high (22908.0 [9685.5, 81190.5] ng/mL). Coagulation abnormalities were frequent: three patients (60%) met the criteria for disseminated intravascular coagulation (DIC); However, none developed thrombotic microangiopathy. In summary, all patients exhibited a severe MAS phenotype with extreme hyperferritinemia, heightened inflammatory markers, and multi‐organ involvement. The summary of clinical characteristics of the patients and classification criteria are presented in Tables 1 and S1.

TABLE 1.

Summary of clinical characteristics and outcomes.

Patient	Case 1	Case 2	Case 3	Case 4	Case 5
Sex	Female	Female	Female	Female	Female
Age at diagnosis (years)	48	79	66	81	23
Comorbidities	HBV carrier, ulcerative colitis	Depression, non‐tuberculous mycobacterial infection	Dyslipidemia, osteoporosis	Hypertension, osteoporosis	Hyperthyroidism
Laboratory parameters at diagnosis
WBC (103/μL)	10.2	25.2	24.3	29.9	11.8
Plt (104/μL)	10.4	18.8	34.0	17.5	18.2
CRP (mg/dL)	25.12	25.88	21.08	39.47	18.23
LDH (IU/L)	484	552	492	807	866
AST (U/L)	53	62	34	124	119
ALT (U/L)	21	48	41	97	64
Ferritin (ng/mL)	6934	28 324	22 908	134 057	12 437
TG (mg/dL)	120	228	119	121	443
Fib (mg/dL)	488	604	586	872	132
sIL‐2R (U/mL)	762	3529	1367	2387	6894
Hemophagocytosis on BMB	N/A	No	Yes	No	Yes
Prior immunosuppression before TPE
Steroid pulse	+	+	+	+	+
Glucocorticoids	+	+	+	+	+
Methotrexate	−	−	−	−	−
Calcineurin inhibitor	+	−	+	+	−
IVIG	−	−	−	−	−
Tocilizumab	−	−	+	−	−
TPE course
Time from diagnosis to TPE onset (days)	12	7	22	3	18
Number of TPE procedures	5	6	5	8	11
Setting	Inpatient	Inpatient	Inpatient	Inpatient	Inpatient
TPE‐related adverse events	Skin rash	−	−	−	Skin rash
ICU stay	−	+	+	+	−
Duration of ICU stay (days)	−	5	7	20	−
Follow‐up duration (months)	13	9	36	56	43
Outcome	Alive	Death; CMV infection	Alive	Alive	Alive

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMB, bone marrow biopsy; CMV, cytomegalovirus; CRP, C‐reactive protein; Fib, fibrinogen; HBV, hepatitis B virus; ICU, intensive care unit; IVIG, intravenous immunoglobulin; LDH, lactate dehydrogenase; N/A, not applicable; Plt, platelet count; sIL‐2R, soluble interleukin‐2 receptor; TG, triglycerides; TPE, therapeutic plasma exchange; WBC, white blood cell count.

3.2. Therapeutic Interventions and Clinical Course

All patients were treated with high‐dose glucocorticoid (oral prednisolone, initial dose of 0.8–1 mg/kg/day, and intravenous pulse methylprednisolone [1000 mg/day or 500 mg/day × 3 days]); three patients received calcineurin inhibitors (tacrolimus or cyclosporin); an IL‐6 receptor antagonist (tocilizumab) was administered to one patient prior to TPE. The dosage of glucocorticoid therapy was chosen by the attending physician, depending on the patient. In three cases, tocilizumab was not selected prior to TPE, due to a history of non‐tuberculous mycobacterial infection or concomitant bacterial infection. Prior immunosuppression is summarized in Table 1.

All patients presented with AOSD‐MAS that was refractory to immunosuppressive therapy, with persistent high CRP and ferritin levels. There were no significant differences observed in WBC, platelets count, CRP, LDH, AST, or ferritin levels between diagnosis and pre TPE (Table 2). TPE with FFP replacement was started for removal of humoral mediators such as proinflammatory cytokines. No major AEs or complications occurred during TPE; specifically, no Grade III or IV events were observed. Two patients developed transient mild skin rashes (Grade I), and two others developed mild shivering (Grade I) related to FFP administration. Symptomatic hypocalcaemia was not observed. Moderate hypotensive reactions (Grade II) occurred in one session and were managed by slowing the procedure and an intravenous fluid bolus. All patients nevertheless completed therapy safely. Following a median of 6 (range 5–11) TPE sessions, all patients showed marked improvement in clinical symptoms, including fever and cutaneous rash. TPE was discontinued when liver function and symptoms improved, and ferritin levels decreased. Figure 1 and Table 2 summarize the changes in key laboratory parameters pre‐ and post‐TPE. TPE was associated with substantial attenuation of systemic inflammation and hepatic dysfunction. After TPE, WBC, CRP, and LDH decreased significantly compared with pre‐TPE values and remained significantly lower at 1 month (Table 2). Specifically, WBC decreased from 34.9 (11.1, 43.2) × 10³/μL pre‐TPE to 5.2 (4.5, 12.3) × 10³/μL post‐TPE and remained low at 7.8 (3.8, 11.8) × 10³/μL at 1 month. CRP decreased from 18.8 (5.1, 20.8) mg/dL to 0.08 (0.06, 2.38) mg/dL and remained low at 0.04 (0.04, 1.54) mg/dL at 1 month. LDH decreased from 812.0 (446.5, 4868.0) IU/L to 277.0 (152.0, 433.5) IU/L and remained low at 284.0 (186.5, 496.5) IU/L at 1 month. AST showed a marked decline after TPE and was significantly lower at 1 month (Table 2). Ferritin also showed a substantial decline after TPE; however, the reduction did not reach statistical significance and variability persisted. Platelet counts remained broadly unchanged across time points.

TABLE 2.

Comparison of laboratory findings between pre‐ and post‐TPE.

Variable	Diagnosis	Pre‐TPE	Post‐TPE	1 month
WBC (103/μL)	24.3 (11.0, 27.6)	34.9 (11.1, 43.2)	5.2 (4.5, 12.3) #	7.8 (3.8, 11.8) $
Plt (104/μL)	18.2 (14.0, 26.4)	22.0 (9.7, 31.8)	19.7 (14.3, 26.2)	24.6 (12.9, 26.7)
CRP (mg/dL)	25.1 (19.7, 32.7)	18.8 (5.1, 20.8)	0.08 (0.06, 2.38) #	0.04 (0.04, 1.54) $
LDH (IU/L)	552.0 (488.0, 836.5)	812.0 (446.5, 4868.0)	277.0 (152.0, 433.5) #	284 (186.5, 496.5) $
AST (U/L)	62.0 (43.5, 121.5)	79.0 (39.0, 934.0)	16.0 (15.5, 94.5)	19.0 (14.0, 31.0) $
Ferritin (ng/mL)	22908.0 (9685.5, 81190.5)	505 68.0 (16 296.0, 142 172.0)	11 005.0 (146.5, 12 954.5)	859.0 (38.0, 26 766.0)

Note: Data represent medians (IQR). The Wilcoxon signed‐rank test was used to assess values.

Abbreviations: AST, aspartate aminotransferase; CRP, C‐reactive protein; IQR, interquartile range; LDH, lactate dehydrogenase; Plt, platelet count; TPE, therapeutic plasma exchange; WBC, white blood cell count.

^# p < 0.05 Pre‐ vs. Post‐TPE.

^$ p < 0.05 Pre vs. 1 month after TPE.

FIGURE 1.

Changes in clinical parameters before and after therapeutic plasma exchange (TPE). Following TPE, inflammatory indices normalized, and liver enzyme levels decreased. (A) WBC decreased post‐TPE. (B) Plt counts remained broadly unchanged across time points. (C) CRP decreased post‐TPE. (D) LDH decreased post‐TPE. (E) AST decline post‐TPE. (F) Ferritin levels generally decreased post‐TPE. Transient re‐elevations occurred in Case 2 and Case 4. These increases were associated with intercurrent infection and/or inflammatory relapse and improved with appropriate treatment. AST, aspartate aminotransferase; CRP, C‐reactive protein; LDH, lactate dehydrogenase; Plt, platelet; TPE, therapeutic plasma exchange; WBC, white blood cell count.

Furthermore, organ dysfunction markers improved in parallel. Among the three patients who presented with DIC, improvement in coagulation abnormalities was observed after initiating FFP as a common replacement fluid during TPE, a rescue therapy. Notably, in Case 5, DIC developed despite intravenous methylprednisolone pulse therapy, prompting initiation of TPE. Three patients were admitted to the intensive care unit (ICU), and one required catecholamines; all were subsequently discharged from the ICU. No patient received intravenous immunoglobulin (IVIG) prior to TPE initiation. However, one patient received IVIG for sepsis during the treatment course, wherein TPE was resumed after infection resolution. IVIG was not administered for the purpose of treating AOSD‐MAS in any patient. One month after TPE, symptoms had improved in all cases, and the primary outcome was achieved in four cases (80%). Ferritin levels generally decreased after TPE. However, transient re‐elevations were observed during follow‐up in Case 2 and Case 4. In Case 2, serum ferritin increased approximately 2 months after TPE in association with a mild relapse of AOSD‐MAS concurrent with a urinary tract infection and subsequently improved after intravenous methylprednisolone pulse therapy and antibiotic therapy. In Case 4, ferritin increased in association with sepsis secondary to a thigh subcutaneous abscess and subsequently decreased following antibiotic therapy and drainage. Four patients recovered from impending organ failure and were discharged with tapered immunosuppression, while one patient died after initial MAS improvement due to cytomegalovirus (CMV) infection.

4. Discussion

We retrospectively evaluated five patients with AOSD‐MAS. After AOSD diagnosis, all patients were treated with a combination of high‐dose glucocorticoids. Three patients received calcineurin inhibitors, while tocilizumab was administered to one patient. Four cases met the MAS criteria at the time of AOSD diagnosis, whereas Case 3 met the MAS criteria 2 weeks after AOSD diagnosis. TPE was initiated at a median of 12.0 days after AOSD diagnosis for treatment‐refractory AOSD‐MAS. All completed an FFP‐replacement TPE course with a median of six sessions. In a recent retrospective study, the incidence of AEs was 6.9% overall and 8.5% among patients receiving TPE [23]. In total, most AEs were mild (60%) and moderate (28%) during TPE. No major AEs or complications occurred; specifically, no Grade III or IV events were observed. TPE was followed by rapid clinical improvement: fever and the salmon‐pink rash abated, WBC count and CRP fell and normalized. Organ dysfunction improved in parallel. Among the three patients who presented with DIC, improvement in coagulation abnormalities was observed after initiating FFP as a common replacement fluid during TPE, a rescue therapy. Notably, in Case 5, DIC developed despite intravenous methylprednisolone pulse therapy, prompting TPE initiation. However, FFP replacement might have contributed to hemostatic improvement, making attributing coagulation normalization solely to mediator removal difficult. Liver enzymes fell markedly without further corticosteroid escalation during the TPE course and immediate post‐TPE period. One month after TPE, there was no recurrence of AOSD‐MAS. Four patients recovered from impending organ failure and were discharged with gradual reduction in immunosuppressive agents. One patient died from CMV infection despite initial MAS improvement.

AOSD is a rare systemic autoinflammatory disorder characterized by dysregulated immune activation. MAS complicates roughly 10%–15% of AOSD cases and is life‐threatening, with mortality around 10%–20% [25, 26]. MAS arises from uncontrolled immune responses driven by unchecked macrophage and T‐cell activation, leading to a cytokine storm and hemophagocytosis with multi‐organ damage [26]. AOSD and MAS share common inflammatory pathways: innate immune cells release high levels of IL‐1β, IL‐6, and IL‐18, alongside interferon‐γ (IFN‐γ) and tumor necrosis factor‐α (TNF‐α), among other mediators [25, 27]. Genetic predispositions and triggers (e.g., viral infection) can precipitate MAS by disrupting immune homeostasis and impairing cytotoxic cell function [28]. This permits a self‐amplifying cascade wherein IFN‐γ from T cells hyperactivates macrophages to release IL‐1β, IL‐6, IL‐18, and TNF‐α, driving hyperinflammation [25]. Notably, extreme hyperferritinemia is a hallmark of AOSD‐MAS and provides a critical diagnostic clue [29]. It has also been discussed as part of the ‘hyperferritinemic syndrome’ shared by MAS, Still's disease, and septic shock [30]. Management of AOSD requires immunomodulatory therapy. High‐dose glucocorticoids are first‐line for acute flares. Conventional immunosuppressants (e.g., methotrexate) may help manage chronic articular disease; however, biologic therapies targeting key cytokines have greatly improved outcomes [1, 27, 28]. Biological disease‐modifying anti‐rheumatic drugs are now widely employed. IL‐1 inhibitors (anakinra and canakinumab) and the anti‐IL‐6 agent (tocilizumab) have demonstrated high efficacy with excellent safety profiles. Moreover, in light of the ‘window of opportunity’ concept, new evidence shows that the earlier these treatments are initiated, the sooner clinical inactivity can be achieved [27]. However, in some patients in our case series, escalation to further immunosuppressive or biologic therapy was constrained by concomitant infection or concern about infection. Therefore, TPE was used as an adjunctive, temporizing intervention.

Although clinical improvement followed TPE, we cannot disentangle the relative contributions of TPE versus concurrent immunosuppression in this retrospective case series. The current EULAR/PReS recommendations for Still's disease and its complication MAS prioritize prompt suppression of hyperinflammation with high‐dose glucocorticoids and cytokine‐targeted immunomodulation (e.g., IL‐1 or IL‐6 inhibition, calcineurin inhibition, and IFN‐γ blockade) [31]. By contrast, TPE is a non‐selective extracorporeal intervention that physically removes circulating inflammatory mediators (including cytokines, immune complexes and other plasma factors), potentially enabling a rapid but transient reduction in cytokine burden while definitive immunomodulatory therapy takes effect. Consistent with this adjunctive role and limited evidence base, the American Society for Apheresis (ASFA) guidelines list hemophagocytic lymphohistiocytosis (including MAS) as a Category III indication for TPE with a weak recommendation (Grade 2C), supporting case‐by‐case decision‐making [11].

Published experience of TPE and double filtration plasmapheresis (DFPP) in AOSD‐MAS remains largely limited to case reports and small series (summarized in Table S2) [12, 13, 14, 32, 33, 34, 35, 36, 37]. These reports often describe rapid improvement in systemic inflammation and organ dysfunction, particularly in patients with refractory disease [13, 14]. Beyond single‐case evidence, a real‐world analysis of autoimmune inflammatory rheumatic disease–associated MAS (including AOSD) suggested that adding TPE to background glucocorticoids and immunosuppressants was associated with prompt cytokine reduction and lower mortality [32]. Conversely, in an adult MAS‐HLH series, TPE induced only partial responses and that required escalation of immunosuppression was frequently required, indicating that TPE should be viewed as a supportive adjunct rather than monotherapy [12].

Against that backdrop, our AOSD‐MAS cases received TPE after pulse glucocorticoids (with calcineurin inhibitors and IL‐6 blockade in selected cases), completing a median of six sessions with FFP replacement, with improvements in ferritin levels and systemic symptom and no major procedural complications. Although ferritin levels decreased following TPE, transient re‐elevations occurred in Case 2 and Case 4 during follow‐up. In our series, these increases were associated with intercurrent infection and/or inflammatory relapse (a mild relapse of AOSD‐MAS concurrent with a urinary tract infection in Case 2 and sepsis secondary to a thigh subcutaneous abscess in Case 4) and improved with appropriate treatment. This underscores that ferritin may rise not only with disease activity, but also with concurrent infection, and should be interpreted alongside clinical findings and other inflammatory markers [30]. Prior reports vary widely in timing, TPE used upfront as a bridge in steroid‐refractory cases or as salvage, and in treatment duration, which is often short and rarely protocolized; ASFA explicitly notes that duration remains undefined. Thus, our comparatively longer course and sequencing after initial immunosuppression highlight two key differences from much of the existing literature. Although, in our series, we observed rapid defervescence and biochemical improvement, guidelines still prioritize early targeted immunomodulation, and the contribution of extended TPE courses to survival or organ recovery remains to be determined. Prospective AOSD‐MAS specific studies are needed to clarify indications, optimal timing, and session number.

There are several limitations in this study. First, this was a single‐arm retrospective study without a control group; it was not possible to adjust for baseline differences or other confounding factors. Consequently, potential biases in our results need to be acknowledged. Second, since immunosuppressive therapy protocols varied between patients, evaluating the specific therapeutic contribution of TPE is difficult. In particular, the optimal timing of TPE initiation within the treatment course and its relationship to disease activity biomarkers need to be clarified. Third, several inflammatory molecules such as serum IL‐18 and/or S100 proteins are proposed as diagnostic biomarkers [38]. These biomarkers appear to identify AOSD‐MAS with high sensitivity and specificity [39, 40]. However, in our case series, AOSD‐MAS was diagnosed using diagnostic criteria [19, 20], so these biomarkers were not measured. Fourth, alternative extracorporeal options, particularly selective plasma exchange (sPE) and DFPP, were neither employed nor evaluated. By restricting the study to a single, non‐selective modality, we cannot determine whether sPE or DFPP might have offered superior efficacy, fewer AEs, or different biochemical profiles compared with standard TPE in patients with AOSD‐MAS. Fifth, since FFP was used as the replacement fluid, improvements in coagulation parameters might partly reflect factor replenishment, and thereby cannot be attributed solely to mediator removal. Sixth, the EULAR/PReS defined remission as a period of at least 6 months with absence of AOSD related symptoms and normal erythrocyte sedimentation rate or CRP [31]. In this study, short‐term outcomes were set to evaluate the early effects of TPE.

5. Conclusions

TPE could serve as a useful adjunctive therapy for patients with treatment‐resistant AOSD‐MAS. In this study, the addition of TPE was associated with rapid improvement. Further evidence is needed to validate the efficacy of TPE in this setting and determine the optimal timing and patient selection for this intervention.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Classification criteria for adult‐onset Still's disease and macrophage activation syndrome.

Table S2: Published reports of therapeutic plasma exchange or related apheresis in adult‐onset Still's disease with MAS/HLH.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.