Time to resolution of veno-occlusive disease/sinusoidal obstruction syndrome post-HCT in defibrotide-treated adults from two European registries Authors: Mohamad Mohty, Régis Peffault de Latour, Myriam Labopin, Vian Amber, Amélie Petitprez, Deborah Gutierrez, Nalina Dronamraju, Hélène Labussière-Wallet, Didier Blaise Published: 2026 ABSTRACT We report a pooled analysis of real-world data of adult patients from two European registries of defibrotide use (DEFIFrance and EBMT Post-Authorization Safety Study [EBMT PASS]) to provide insights into the duration of defibrotide therapy to achieve resolution of veno-occlusive disease/sinusoidal obstruction syndrome (VOD/SOS) symptoms by severity. Overall, 275 defibrotide-treated adults with VOD/SOS post-hematopoietic cell transplantation (HCT) were included (severe VOD/SOS, n = 215; nonsevere VOD/SOS, n = 60). Median time from the start of defibrotide treatment to VOD/SOS resolution was 18.5 and 16.5 days in patients with severe and nonsevere VOD/SOS, respectively. Among patients with VOD/SOS resolution, symptoms resolved after >21 days in 31% (n = 40/128) and 32% (n = 14/44) with severe and nonsevere VOD/SOS, respectively. In all patients with VOD/SOS resolution independent of severity (n = 174), the Kaplan-Meier–estimated survival at Day 100 post-HCT was 82% (95% confidence interval [CI]: 75%, 87%); this was 78% (95% CI: 70%, 84%) and 91% (95% CI: 78%, 96%) in patients with severe and nonsevere VOD/SOS, respectively. Serious treatment-emergent adverse events of interest occurred in 31% of patients with VOD/SOS resolution. This pooled analysis supports the utility of defibrotide treatment for VOD/SOS and highlights that treatment may often need to be continued beyond 21 days to achieve resolution. Introduction Veno-occlusive disease/sinusoidal obstruction syndrome (VOD/SOS) is a potentially life-threatening complication of hematopoietic cell transplantation (HCT), which occurs due to activation and damage of the sinusoidal endothelium [1]. The incidence of VOD/SOS is estimated to be approximately 10–15% post-HCT with myeloablative conditioning (MAC), but recent references have reported a lower incidence (<5%) that could be attributed to changes in the management of patients and/or the increased use of reduced intensity conditioning (RIC) regimens [2–5]. Risk factors for VOD/SOS include MAC regimens, HCT with an unrelated donor, having a second HCT, older age, Karnofsky score <90%, previous liver disease, and prior treatment with gemtuzumab ozogamicin or inotuzumab ozogamicin [1, 6–8]. Clinical presentation of VOD/SOS includes rapid weight gain, hyperbilirubinemia, hepatomegaly, and ascites [1]. Severe VOD/SOS is associated with a mortality rate of >80% if left untreated [2, 3]. Defibrotide has been shown to protect endothelial cells and restore the thrombotic-fibrinolytic balance [9, 10]. It is approved in the European Union for the treatment of severe VOD/SOS post-HCT in patients aged >1 month [11] and in the US for the treatment of VOD/SOS in patients with renal or pulmonary dysfunction post-HCT [12]. Clinical guidelines recommend the prompt initiation of defibrotide for at least 21 days after the diagnosis of VOD/SOS or until resolution [3]. Two registry studies (DEFIFrance and EBMT Post-Authorization Safety Study [EBMT PASS]) were established in France and Europe, respectively, to collect real-world data on the safety and efficacy of defibrotide [13, 14]. The primary efficacy and safety results from DEFIFrance [13] and EBMT PASS [14] have been previously reported. Here, we report a pooled analysis of adult patients in the DEFIFrance and EBMT PASS registries to provide further insights into the defibrotide treatment duration and time to VOD/SOS resolution in the real-world setting. Since VOD/SOS incidence, clinical presentation, and outcomes of treatment in children are known to differ from those reported in adults, a pooled analysis of the pediatric patients will be reported separately in adherence to the EBMT diagnosis and severity grading criteria in pediatric patients [15]. Methods Registry study designs DEFIFrance was an observational, post-marketing study that collected retrospective and prospective data on patients receiving defibrotide at 53 French transplant centers from July 2014 to March 2020 [13]. VOD/SOS severity for patients aged ≥18 years was categorized using adult EBMT criteria [16] and was graded as severe/very severe or mild/moderate disease [13]. EBMT PASS was a European, multinational, prospective, observational study that enrolled defibrotide-treated patients from April 2015 to July 2018 [14]. VOD/SOS severity was graded based on the clinical expertise of the investigators as severe or nonsevere. In both studies, most patients received 25 mg/kg of defibrotide per day, consistent with the defibrotide recommended dose and schedule per the label [11–14]. Both registries collected data on defibrotide administration and VOD/SOS resolution, defined as complete response ([CR] total serum bilirubin <2 mg/dL and multiorgan failure [MOF] resolution per investigator’s assessment) of VOD/SOS at Day 100 post-HCT in DEFIFrance [13], and defined as the overall rate of resolution of both VOD/SOS and multiorgan dysfunction/failure ([MOD/MOF] as determined by the investigator) at Day 100 post-HCT in EBMT PASS [14]. VOD/SOS diagnosis in DEFIFrance and EBMT PASS was determined by the investigator and based on classical/standard criteria (including, but not limited to, hyperbilirubinemia, hepatomegaly, ascites, and weight gain >5%). Pooled analysis: patients and endpoints All adult patients (aged ≥18 years) who received defibrotide for the treatment of post-HCT VOD/SOS from DEFIFrance and EBMT PASS were included in this pooled analysis; patients who were included in both registries were identified and only counted once. Patients were categorized as having severe or nonsevere VOD/SOS in this pooled analysis. VOD/SOS grading was performed differently across the two registries; therefore, for this pooled analysis, patients graded as severe/very severe in DEFIFrance were categorized as severe, while those graded as mild/moderate in DEFIFrance were categorized as nonsevere, harmonized with EBMT PASS classifications [13, 14]. In DEFIFrance, the primary endpoints were survival and CR (total serum bilirubin <2 mg/dL and MOF resolution per investigator’s assessment) at Day 100 post-HCT in patients with severe/very severe VOD/SOS. Key secondary endpoints included the evaluation of treatment-emergent serious adverse events (AEs) of interest (irrespective of relationship to defibrotide treatment); overall mortality and VOD/SOS-related mortality; prognostic factors with an impact on CR or survival; and the rate of graft-versus-host disease post-HCT. In EBMT PASS, the primary endpoint was the incidence rate of serious AEs of interest up to 12 months post-HCT in patients with severe VOD/SOS. Secondary endpoints included patient demographics, Day 100 survival post-HCT, mortality due to VOD/SOS and overall rate of VOD/SOS (and MOD/MOF, if present) resolution (as determined by the investigator). In this pooled analysis, the evaluated endpoints were time to VOD/SOS diagnosis post-HCT, defibrotide treatment duration, rates of VOD/SOS resolution and time to resolution, Kaplan-Meier (KM)-estimated survival at Day 100 post-HCT, serious treatment-emergent adverse events (TEAEs) of interest (bleeding, coagulopathy, embolism, hemorrhage, hypotension, infection, septicemia, and thrombosis), causes of death at Day 100 post-HCT, and the impact of conditioning intensity (RIC vs. MAC) on VOD/SOS severity, outcomes, and safety. Statistical analysis Summary statistics are presented for baseline data, time to VOD/SOS diagnosis, defibrotide treatment duration, time to resolution of VOD/SOS, and safety variables. Survival at Day 100 post-HCT is estimated using the KM method. Patients who withdrew from the study were censored at the time of withdrawal. Analyses were conducted in all defibrotide-treated patients (N = 279), in patients with known severity (n = 275), and in patients with VOD/SOS resolution regardless of severity (n = 174), as indicated in the respective tables. Results Patient demographics and clinical characteristics Overall, there were 279 defibrotide-treated adult patients with VOD/SOS post-HCT in the pooled dataset. Of these patients, severity of VOD/SOS was missing for four patients; therefore, 275 patients were included in this analysis, of whom 215 (78%) patients had severe VOD/SOS and 60 (22%) patients had nonsevere VOD/SOS (Fig. 1). The median age was similar for patients with severe VOD/SOS (51.3 years [min–max: 18–74]) vs. nonsevere VOD/SOS (53.5 years [min–max: 19–69]; Table 1). Regardless of disease severity, most patients (≥85%) received allograft transplants. Among patients who had resolution of symptoms ≤21 days (n = 118) or >21 days (n = 54) after start of defibrotide treatment, the demographic and clinical characteristics were comparable with the overall population, regardless of VOD/SOS severity (Table S1). In 275 patients who had data available on conditioning regimen, 138 (50%) received RIC and 137 (50%) received MAC. The percentage of patients who received MAC was numerically higher among patients with nonsevere VOD/SOS (62%) compared with patients with severe VOD/SOS (46%; Table 1).Fig. 1Flow diagram of patients enrolled in the pooled analysis of DEFIFrance and EBMT PASS.aIncluded 336 adult and pediatric patients; 251 patients with severe/very severe VOD/SOS (defined in this analysis as severe VOD/SOS) and 81 patients with mild/moderate VOD/SOS (defined in this analysis as nonsevere VOD/SOS); data on severity of VOD/SOS were missing from DEFIFrance for two adult patients with resolution of VOD/SOS symptoms and two adult patients without resolution of VOD/SOS symptoms. bIncluded 104 adult and pediatric patients; 62 patients with severe VOD/SOS and 42 patients with nonsevere VOD/SOS. cNumber of adult patients with known severity of VOD/SOS. EBMT PASS EBMT Post-Authorization Safety Study, HCT hematopoietic cell transplantation, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease.Table 1Baseline patient demographics and clinical characteristics.Severe VOD/SOSaNonsevere VOD/SOSaCR (n = 128)No CR (n = 87)Overall (n = 215)CR (n = 44)No CR (n = 16)Overall (n = 60)Median age (min–max), years51.1 (18–74)51.4 (20–70)51.3 (18–74)52.3 (19–68)62.3 (22–69)53.5 (19–69)Male, n (%)77 (60)50 (57)127 (59)28 (64)12 (75)40 (67)Number of previous transplants,b n (%) 124 (19)14 (16)38 (18)9 (20)2 (12)11 (18) 2+4 (3)7 (8)11 (5)2 (4)0 (0.0)2 (3)MAC, n (%)59 (46)41 (47)100 (46)30 (68)7 (44)37 (62)Type of transplant,c n (%) Autograft10 (8)4 (5)14 (6)6 (14)2 (12)8 (13) Allograft118 (92)82 (94)200 (93)38 (86)14 (88)52 (87)Donor type (allograft only),d n (%) HLA similar to the siblings34 (29)22 (27)56 (28)9 (24)6 (43)15 (29) Non-parent (unrelated)48 (41)33 (40)81 (40)15 (39)6 (43)21 (40) Matchede39 (81)23 (70)62 (77)10 (67)2 (33)12 (57) Mismatchede9 (19)9 (27)18 (22)5 (33)4 (67)9 (43) Missing01 (3)1 (1)000 Haplo-identical25 (21)14 (17)39 (20)6 (16)2 (14)8 (15) Umbilical cord blood cells03 (4)3 (2)1 (3)01 (2) Missing11 (9)10 (12)21 (10)7 (18)07 (13)Primary disease (DEFIFrance),f n (%) AML37 (29)24 (28)61 (28)11 (25)7 (44)18 (30) Lymphoma26 (20)16 (18)42 (20)6 (14)2 (12)8 (13) ALL19 (15)12 (14)31 (14)9 (20)1 (6)10 (17) MDS16 (12)12 (14)28 (13)6 (14)4 (25)10 (17) Myelofibrosis8 (6)7 (8)15 (7)2 (4)2 (12)4 (7)Primary disease (EBMT PASS),g n (%) AML and related precursor neoplasms4 (3)2 (2)6 (3)2 (4)02 (3) Precursor lymphoid neoplasms1 (0.8)3 (3)4 (2)1 (2)01 (2) MDS1 (0.8)2 (2)3 (1)1 (2)01 (2) Non-Hodgkin lymphoma1 (0.8)1 (1)2 (1)1 (2)01 (2) Chronic myeloid leukemia1 (0.8)01 (0.5)1 (2)01 (2) Myeloproliferative neoplasia1 (0.8)01 (0.5)2 (4)02 (3)Percentages were rounded to the nearest whole number.ALL acute lymphocytic leukemia, AML acute myeloid leukemia, CR complete response, EBMT PASS EBMT Post-Authorization Safety Study, HLA human leukocyte antigen, MAC myeloablative conditioning, MDS myelodysplastic syndromes, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease.aData on severity of VOD/SOS were missing from DEFIFrance for two patients with resolution of VOD/SOS symptoms and two patients without resolution of VOD/SOS symptoms, therefore, these four patients were not included. Severe VOD/SOS consists of patients who were graded as having severe/very severe disease in DEFIFrance and severe disease in EBMT PASS; nonsevere VOD/SOS consists of patients who were graded as having mild/moderate disease from DEFIFrance and nonsevere disease in EBMT PASS.bFor patients with known number of previous transplants.cType of transplant was missing for one patient with severe VOD/SOS who did not achieve a CR.dPercentages were calculated from all patients who received allograft transplant.ePercentages were calculated from all patients with a non-parent (unrelated) donor.fReported in ≥10% of patients with known primary disease in severe or nonsevere group.gReported in ≥2% of patients with known primary disease in severe or nonsevere group.Patients who received reduced intensity conditioning were not pooled by VOD/SOS severity and CR status. Flow diagram of patients enrolled in the pooled analysis of DEFIFrance and EBMT PASS. aIncluded 336 adult and pediatric patients; 251 patients with severe/very severe VOD/SOS (defined in this analysis as severe VOD/SOS) and 81 patients with mild/moderate VOD/SOS (defined in this analysis as nonsevere VOD/SOS); data on severity of VOD/SOS were missing from DEFIFrance for two adult patients with resolution of VOD/SOS symptoms and two adult patients without resolution of VOD/SOS symptoms. bIncluded 104 adult and pediatric patients; 62 patients with severe VOD/SOS and 42 patients with nonsevere VOD/SOS. cNumber of adult patients with known severity of VOD/SOS. EBMT PASS EBMT Post-Authorization Safety Study, HCT hematopoietic cell transplantation, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease. Table 1Baseline patient demographics and clinical characteristics.Severe VOD/SOSaNonsevere VOD/SOSaCR (n = 128)No CR (n = 87)Overall (n = 215)CR (n = 44)No CR (n = 16)Overall (n = 60)Median age (min–max), years51.1 (18–74)51.4 (20–70)51.3 (18–74)52.3 (19–68)62.3 (22–69)53.5 (19–69)Male, n (%)77 (60)50 (57)127 (59)28 (64)12 (75)40 (67)Number of previous transplants,b n (%) 124 (19)14 (16)38 (18)9 (20)2 (12)11 (18) 2+4 (3)7 (8)11 (5)2 (4)0 (0.0)2 (3)MAC, n (%)59 (46)41 (47)100 (46)30 (68)7 (44)37 (62)Type of transplant,c n (%) Autograft10 (8)4 (5)14 (6)6 (14)2 (12)8 (13) Allograft118 (92)82 (94)200 (93)38 (86)14 (88)52 (87)Donor type (allograft only),d n (%) HLA similar to the siblings34 (29)22 (27)56 (28)9 (24)6 (43)15 (29) Non-parent (unrelated)48 (41)33 (40)81 (40)15 (39)6 (43)21 (40) Matchede39 (81)23 (70)62 (77)10 (67)2 (33)12 (57) Mismatchede9 (19)9 (27)18 (22)5 (33)4 (67)9 (43) Missing01 (3)1 (1)000 Haplo-identical25 (21)14 (17)39 (20)6 (16)2 (14)8 (15) Umbilical cord blood cells03 (4)3 (2)1 (3)01 (2) Missing11 (9)10 (12)21 (10)7 (18)07 (13)Primary disease (DEFIFrance),f n (%) AML37 (29)24 (28)61 (28)11 (25)7 (44)18 (30) Lymphoma26 (20)16 (18)42 (20)6 (14)2 (12)8 (13) ALL19 (15)12 (14)31 (14)9 (20)1 (6)10 (17) MDS16 (12)12 (14)28 (13)6 (14)4 (25)10 (17) Myelofibrosis8 (6)7 (8)15 (7)2 (4)2 (12)4 (7)Primary disease (EBMT PASS),g n (%) AML and related precursor neoplasms4 (3)2 (2)6 (3)2 (4)02 (3) Precursor lymphoid neoplasms1 (0.8)3 (3)4 (2)1 (2)01 (2) MDS1 (0.8)2 (2)3 (1)1 (2)01 (2) Non-Hodgkin lymphoma1 (0.8)1 (1)2 (1)1 (2)01 (2) Chronic myeloid leukemia1 (0.8)01 (0.5)1 (2)01 (2) Myeloproliferative neoplasia1 (0.8)01 (0.5)2 (4)02 (3)Percentages were rounded to the nearest whole number.ALL acute lymphocytic leukemia, AML acute myeloid leukemia, CR complete response, EBMT PASS EBMT Post-Authorization Safety Study, HLA human leukocyte antigen, MAC myeloablative conditioning, MDS myelodysplastic syndromes, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease.aData on severity of VOD/SOS were missing from DEFIFrance for two patients with resolution of VOD/SOS symptoms and two patients without resolution of VOD/SOS symptoms, therefore, these four patients were not included. Severe VOD/SOS consists of patients who were graded as having severe/very severe disease in DEFIFrance and severe disease in EBMT PASS; nonsevere VOD/SOS consists of patients who were graded as having mild/moderate disease from DEFIFrance and nonsevere disease in EBMT PASS.bFor patients with known number of previous transplants.cType of transplant was missing for one patient with severe VOD/SOS who did not achieve a CR.dPercentages were calculated from all patients who received allograft transplant.ePercentages were calculated from all patients with a non-parent (unrelated) donor.fReported in ≥10% of patients with known primary disease in severe or nonsevere group.gReported in ≥2% of patients with known primary disease in severe or nonsevere group.Patients who received reduced intensity conditioning were not pooled by VOD/SOS severity and CR status. Baseline patient demographics and clinical characteristics. Percentages were rounded to the nearest whole number. ALL acute lymphocytic leukemia, AML acute myeloid leukemia, CR complete response, EBMT PASS EBMT Post-Authorization Safety Study, HLA human leukocyte antigen, MAC myeloablative conditioning, MDS myelodysplastic syndromes, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease. aData on severity of VOD/SOS were missing from DEFIFrance for two patients with resolution of VOD/SOS symptoms and two patients without resolution of VOD/SOS symptoms, therefore, these four patients were not included. Severe VOD/SOS consists of patients who were graded as having severe/very severe disease in DEFIFrance and severe disease in EBMT PASS; nonsevere VOD/SOS consists of patients who were graded as having mild/moderate disease from DEFIFrance and nonsevere disease in EBMT PASS. bFor patients with known number of previous transplants. cType of transplant was missing for one patient with severe VOD/SOS who did not achieve a CR. dPercentages were calculated from all patients who received allograft transplant. ePercentages were calculated from all patients with a non-parent (unrelated) donor. fReported in ≥10% of patients with known primary disease in severe or nonsevere group. gReported in ≥2% of patients with known primary disease in severe or nonsevere group. Patients who received reduced intensity conditioning were not pooled by VOD/SOS severity and CR status. Timing of VOD/SOS diagnosis and duration of defibrotide treatment In the overall population with known severity, median time from HCT to VOD/SOS diagnosis was 14 (quartile 1 [Q1]–quartile 3 [Q3]: 8–23; min–max: -7–495) and 14 (Q1–Q3: 8–28.5; min–max: 1–319) days in severe and nonsevere VOD/SOS groups, respectively (Fig. 2A). Median time from VOD/SOS diagnosis to the start of defibrotide treatment was 0 (Q1–Q3: 0–0; min–max: -1–24) and 0 (Q1–Q3: 0–1; min–max: 0–17) days in patients with severe and nonsevere VOD/SOS, respectively. In the overall population, median duration of defibrotide treatment was 16 (Q1–Q3: 9–22; min–max: 1–258) and 15 (Q1–Q3: 12–22; min–max: 3–55) days in patients with severe and nonsevere VOD/SOS, respectively (Fig. 2B). In patients with known severity who achieved a CR (severe VOD/SOS: 128/215 [60%]; nonsevere VOD/SOS: 44/60 [73%]), median treatment duration was longer in patients with severe VOD/SOS compared with nonsevere VOD/SOS (18.5 [Q1–Q3: 13.5–22; min–max: 1–258] vs. 15.5 [Q1–Q3: 13–22; min–max: 4–55] days; Fig. 2C).Fig. 2Time from HCT to VOD/SOS diagnosis and defibrotide treatment duration.aOne patient had missing data. CR complete response, HCT hematopoietic cell transplantation, Q1 quartile 1, Q3 quartile 3, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease. Time from HCT to VOD/SOS diagnosis and defibrotide treatment duration. aOne patient had missing data. CR complete response, HCT hematopoietic cell transplantation, Q1 quartile 1, Q3 quartile 3, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease. In an analysis of patients who received RIC (n = 138) vs. MAC (n = 137), median time from HCT to VOD/SOS diagnosis was 12 (Q1–Q3: 7–21; min–max: –7–495) vs. 15 (Q1–Q3: 10–26; min–max: 0–336) days, respectively (Table S1). Median time from VOD/SOS diagnosis to the start of defibrotide treatment was 0 (Q1–Q3: 0–0; min–max: 0–16) and 0 (Q1–Q3: 0–1; min–max: -1–24) days in patients receiving RIC vs. MAC, respectively (Table S1). Median treatment duration was 15 (Q1–Q3: 10–22; min–max: 1–60) and 16 (Q1–Q3: 11–22; min–max: 1–258) days in patients receiving RIC vs. MAC, respectively (Table S2). Time to resolution of VOD/SOS Overall, VOD/SOS resolution was achieved in 172/275 (63%) patients with known severity (severe group: 128/215 [60%]; nonsevere group: 44/60 [73%]) (Fig. 1). Median time from the start of defibrotide treatment to resolution was 18.5 days (Q1–Q3: 13–23; min–max: 4–109) in patients with severe VOD/SOS and 16.5 days (Q1–Q3: 12–25; min–max: 1–114) in patients with nonsevere VOD/SOS. Among patients with resolution, 31% with severe VOD/SOS and 32% with nonsevere VOD/SOS had resolution of symptoms after >21 days after start of defibrotide treatment, respectively (Fig. 3).Fig. 3Time to resolution of VOD/SOS in patients who achieved a CR.Percentages were rounded to the nearest whole number. CR complete response, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease. Time to resolution of VOD/SOS in patients who achieved a CR. Percentages were rounded to the nearest whole number. CR complete response, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease. In patients who received RIC vs. MAC, resolution occurred in 83/138 (60%) vs. 89/137 (65%) patients, respectively. Median time to resolution from VOD/SOS diagnosis was 20 (Q1–Q3: 13–25; min–max: 5–62) days in the RIC-conditioned group and 18 (Q1–Q3: 14–24; min–max: 4–115) days in the MAC-conditioned group (Table S2). Among patients with resolution, symptoms resolved after >21 days from VOD/SOS diagnosis in 40% vs. 37% of patients who received RIC vs. MAC, respectively (Table S2). KM-estimated survival rate at Day 100 post-HCT in patients with VOD/SOS resolution In the population of all patients with VOD/SOS resolution regardless of VOD/SOS severity (174/279 [62%]), the KM-estimated survival rate at Day 100 post-HCT independent of severity was 82% (95% confidence interval [CI]: 75%, 87%; Fig. 4). In patients with known severity of VOD/SOS, the KM-estimated survival rate at Day 100 post-HCT in those with severe VOD/SOS (n = 128) and nonsevere VOD/SOS (n = 44) were 78% (95% CI: 70%, 84%) and 91% (95% CI: 78%, 96%), respectively (Fig. 4).Fig. 4KM-estimated survival rate at Day 100 post-HCT in all patients with VOD/SOS resolution and by severity.CI confidence interval, HCT hematopoietic cell transplantation, KM Kaplan-Meier, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease. KM-estimated survival rate at Day 100 post-HCT in all patients with VOD/SOS resolution and by severity. CI confidence interval, HCT hematopoietic cell transplantation, KM Kaplan-Meier, SOS sinusoidal obstruction syndrome, VOD veno-occlusive disease. Among patients who received RIC (n = 138) vs. MAC (n = 137), the KM-estimated survival rate at Day 100 post-HCT was 49% (95% CI: 41%, 57%) and 69% (95% CI: 60%, 76%), respectively (Fig. S1). Serious TEAEs of interest and causes of death at Day 100 In total, 32% (90/279) of adult patients experienced serious TEAEs of interest. Serious TEAEs of interest occurred in 31% of patients with VOD/SOS resolution (Table 2). Rates of serious TEAEs of interest were similar regardless of treatment duration (≤21 days: 29%; >21 days: 34%). The most common serious TEAEs of interest by category were infection (≤21 days: 16%; >21 days: 21%) and hemorrhage (≤21 days: 15%; >21 days: 17%; Table 2). When stratified by severity and regardless of achievement of a CR, rates of serious TEAEs of interest were similar in patients with severe VOD/SOS (32%) and nonsevere VOD/SOS (35%; Table S3). Patients receiving RIC or MAC experienced similar rates of serious TEAEs of interest (33% [45/138] and 33% [45/137], respectively).Table 2Serious TEAEs of interest in patients who achieved a CR.Serious TEAEs of interestDefibrotide ≤21 days (n = 116)Defibrotide >21 days (n = 58)Overall (n = 174)Any serious TEAE of interest, n (%)34 (29)20 (34)54 (31)Serious TEAEs of interest by category, n (%) Infection19 (16)12 (21)31 (18) Hemorrhage17 (15)10 (17)27 (16) Hypotension5 (4)05 (3) Bleeding2 (2)2 (3)4 (2) Coagulopathy3 (3)1 (2)4 (2) Septicemia02 (3)2 (1) Embolism1 (0.9)01 (0.6) Thrombosis1 (0.9)01 (0.6)Serious TEAEs of interest in ≥2% of patients with CR, n (%) Hemorrhagic cystitis5 (4)3 (5)8 (5) Cytomegalovirus infection4 (3)2 (3)6 (3) BK virus infection2 (2)3 (5)5 (3) Cytomegalovirus infection reactivation1 (0.9)4 (7)5 (3) Aspergillus infection4 (3)04 (2) Coagulopathy3 (3)1 (2)4 (2) Hemorrhage3 (3)1 (2)4 (2) Hypotension4 (3)04 (2) Viral hemorrhagic cystitis2 (2)2 (3)4 (2) Fungemia02 (3)2 (1)Percentages were rounded to the nearest whole number. CR complete response, TEAE treatment-emergent adverse event. Table 2Serious TEAEs of interest in patients who achieved a CR.Serious TEAEs of interestDefibrotide ≤21 days (n = 116)Defibrotide >21 days (n = 58)Overall (n = 174)Any serious TEAE of interest, n (%)34 (29)20 (34)54 (31)Serious TEAEs of interest by category, n (%) Infection19 (16)12 (21)31 (18) Hemorrhage17 (15)10 (17)27 (16) Hypotension5 (4)05 (3) Bleeding2 (2)2 (3)4 (2) Coagulopathy3 (3)1 (2)4 (2) Septicemia02 (3)2 (1) Embolism1 (0.9)01 (0.6) Thrombosis1 (0.9)01 (0.6)Serious TEAEs of interest in ≥2% of patients with CR, n (%) Hemorrhagic cystitis5 (4)3 (5)8 (5) Cytomegalovirus infection4 (3)2 (3)6 (3) BK virus infection2 (2)3 (5)5 (3) Cytomegalovirus infection reactivation1 (0.9)4 (7)5 (3) Aspergillus infection4 (3)04 (2) Coagulopathy3 (3)1 (2)4 (2) Hemorrhage3 (3)1 (2)4 (2) Hypotension4 (3)04 (2) Viral hemorrhagic cystitis2 (2)2 (3)4 (2) Fungemia02 (3)2 (1)Percentages were rounded to the nearest whole number. CR complete response, TEAE treatment-emergent adverse event. Serious TEAEs of interest in patients who achieved a CR. Percentages were rounded to the nearest whole number. CR complete response, TEAE treatment-emergent adverse event. Causes of death at Day 100 post-HCT according to VOD/SOS severity are detailed in Table S4. Among patients with resolution, the most common causes of death related to HCT were infection (severe VOD/SOS: 13%; nonsevere VOD/SOS: 4%), graft-versus-host disease (severe VOD/SOS: 7%; nonsevere VOD/SOS: 2%), and pulmonary toxicity (severe: 6%; nonsevere: 4%; Table S4). The most common causes of death related to HCT for patients without resolution were VOD/SOS (severe VOD/SOS: 46%; nonsevere VOD/SOS: 25%) and infection (severe VOD/SOS: 38%; nonsevere VOD/SOS: 44%; Table S4). In patients receiving RIC or MAC, the most common causes of death related to HCT were infection (RIC: 30%; MAC: 13%) and VOD/SOS (RIC: 16%; MAC: 17%); graft-versus-host disease was more common in RIC (14%) vs MAC (4%; Table S5). Discussion This pooled analysis of data from DEFIFrance and EBMT PASS provides a large dataset to further explore the duration of defibrotide treatment relative to the resolution of VOD/SOS in adults in a real-world setting. Overall, 62% of patients achieved VOD/SOS resolution, with a KM-estimated survival rate at Day 100 post-HCT of 82% in those who achieved CR. Furthermore, as may be expected, patients with nonsevere VOD/SOS had shorter time to resolution (median: 16.5 days and 18.5 days, respectively) and a higher rate of survival (KM-estimated survival rate at Day 100 post-HCT: 91% and 78%, respectively) than those with severe VOD/SOS [17]. Time from HCT to VOD/SOS diagnosis (median: 14 days for both nonsevere and severe VOD/SOS) was consistent with previous reports [17, 18]. Recommended defibrotide treatment duration is at least 21 days until the symptoms and signs of severe VOD/SOS resolve, as recommended in the European Union, or until resolution of VOD/SOS up to a maximum of 60 days according to the US package insert [11, 12]. In clinical practice, a shorter treatment duration of defibrotide has been reported [19, 20]. In this pooled analysis, depending on VOD/SOS severity, the median treatment duration was 15–16 days. In patients who achieved a CR, the median treatment duration was slightly longer, at 15.5–18.5 days, consistent with the median time from start of defibrotide treatment to resolution of VOD/SOS symptoms (16.5–18.5 days), indicating that some patients who achieved a CR also stopped defibrotide treatment after achieving resolution within 21 days. Our findings reinforce that stopping defibrotide at 21 days may be premature in a substantial proportion of patients, with 31% requiring >21 days of defibrotide treatment to achieve resolution. This observation is consistent with previous studies and suggests that 21 days may not be enough to achieve resolution in some patients [17, 18, 21]. While our study provides valuable real-world data on the duration of defibrotide treatment for VOD/SOS, the ability to predict which patients will need longer treatment remains an unmet need. A scoring system incorporating pre-VOD/SOS clinical factors (e.g., baseline ferritin and other inflammatory markers, renal and hepatic function, time from HCT to VOD/SOS onset, and disease severity) could be valuable for individualized treatment planning, particularly in resource-constrained settings where access to defibrotide may be limited. However, the present registry-based datasets did not include several of these key variables, and the heterogeneity of data collection across centers limits the development of a robust predictive model within the current analysis. Future prospective studies with standardized, granular baseline assessments are warranted to identify predictors of treatment duration and to support the development of such a scoring system. Several recent studies have investigated biomarker-based prognostic models [22–25]. For example, the Hokkaido ultrasonography-based system (HokUS-10) uses risk factors to assess SOS-related mortality and showed that ultrasonography findings after treatment can predict treatment outcomes for SOS [24]. Such novel scoring systems could be further advanced to identify patients most likely to benefit from treatments such as defibrotide and predict their time to resolution, which can support optimization of resource allocation, particularly given treatment costs with defibrotide. Nevertheless, prolonged treatment, tailored to clinical response, appears critical to maximize survival, particularly in patients with severe VOD/SOS. Persevering with defibrotide treatment to achieve VOD/SOS resolution is important as it could directly impact survival outcomes for patients with this potentially life-threatening condition. Regardless of severity, Day 100 survival was high (82%) in patients with VOD/SOS resolution, similar to previous studies [21] and highlights the importance of continuing treatment until symptom resolution. Among patients who achieved resolution, patients with severe VOD/SOS had lower KM-estimated survival rates than patients with nonsevere VOD/SOS, highlighting the importance of prompt diagnosis and treatment before patients advance to a more severe stage of disease; this observation also aligns with the previous observation that earlier initiation of defibrotide improves Day 100 survival [26]. A substantial proportion of patients achieved resolution of VOD/SOS symptoms, regardless of whether they received RIC or MAC (60% and 65%, respectively). However, patients receiving RIC vs. MAC had a lower KM-estimated survival rate at Day 100 post-HCT (49% vs. 69%). The survival gap may reflect differences in baseline patient characteristics, such as frailty and prior HCT in patients who received RIC, with a higher proportion of RIC- vs MAC-treated patients’ cause of death being infection (30% vs 13%) and graft-versus-host disease (14% vs 6%), and thus, not necessarily a result of the regimen intensity itself. These results highlight the need for vigilance in recognizing VOD/SOS symptoms independent of conditioning chemotherapy; early recognition and rapid initiation of treatment for VOD/SOS are crucial to enhancing patient outcomes and reducing the risk of severe complications. The safety profile of defibrotide in the real-world setting was consistent with reports from previous studies [18, 26, 27], supporting the utility of defibrotide treatment per the prescribing information. As registry studies, there are some limitations to this pooled analysis from DEFIFrance and EBMT PASS. These studies did not have a control arm of patients with VOD/SOS post-HCT who did not receive defibrotide treatment, limiting the evaluation of the effectiveness of defibrotide at resolving VOD/SOS symptoms and reducing mortality. However, given prior results with defibrotide treatment for VOD/SOS, this was considered an ethically sound approach as a placebo-controlled arm would be incompatible with clinical equipoise in life-threatening diseases such as VOD/SOS that lack other effective therapies [18]. Furthermore, given the myriad of reasons that a patient may have stopped defibrotide therapy (e.g., CR, adverse event, early death post-HCT), analysis of outcomes by treatment duration is not feasible from this real-world dataset. Additionally, the methods used for VOD/SOS diagnosis and severity grading were not uniform, as criteria were not consistent across DEFIFrance and EBMT PASS, but rather dependent on the physician and institutional practice. In 2023, EBMT updated the diagnosis and severity criteria for adult patients (refined EBMT criteria 2023), which have since been widely adopted and may facilitate earlier diagnosis of VOD/SOS and defibrotide treatment initiation [1]. The absence of uniform application of the refined 2023 EBMT adult criteria across both registries limits comparability; future analyses under these standardized definitions will be essential. The recurrence of VOD/SOS was not collected, precluding investigation into its incidence or potential correlation with defibrotide treatment duration. Such an analysis should be considered for future studies to further the understanding of long-term outcomes with defibrotide. In conclusion, this pooled analysis of adult data from DEFIFrance and EBMT PASS supports the utility of defibrotide treatment for HCT-associated VOD/SOS and highlights that treatment may often need to be continued beyond 21 days to achieve resolution to improve the likelihood of survival. Clinical data on defibrotide in VOD/SOS diagnosed with the refined EBMT criteria 2023 and prospective studies that evaluate outcomes of different lengths of defibrotide treatment will be beneficial to guide clinical practices. Supplementary information Supplementary Information Supplementary Information