Treatment Patterns and Clinical Outcomes of Patients with Moderate to Severe Acute Graft-Versus-Host Disease: A Multicenter Chart Review Study

Acute graft-versus-host disease (aGVHD) remains a barrier to successful allogeneic hematopoietic stem cell transplantation (HSCT) outcomes. Contemporary comprehensive analyses of real-world clinical outcomes among patients who develop aGVHD post-HSCT are needed to better understand the unmet needs of this patient population. This multicenter, retrospective chart review describes treatment patterns and clinical outcomes among patients (≥18 years old) from Finland, Sweden, and France who developed grades II–IV aGVHD after their first HSCT (January 2016–June 2017). From 13 participating centers, 151 patients were included. The median (Q1, Q3) age at HSCT was 56 (45, 62) years old. One line of aGVHD treatment was received by 47.7%, and the most common first-line treatment was methylprednisolone (alone or in a combination regimen, 74.2%; monotherapy, 25.8%). Among patients treated with methylprednisolone, 79.5% achieved a complete or partial response. The median (Q1, Q3) number of treatment lines was 2.0 (1.0, 3.0). The median (Q1, Q3) time to obtain an aGVHD diagnosis from transplant was 29.5 (21.0, 44.0) days, and 14.5 (7.0, 34.0) days to achieve the best response for 110 evaluable patients. At 6 and 12 months, 53.6% and 49.0%, respectively, achieved a complete response. Chronic GVHD occurred in 37.7% of patients, and aGVHD reoccurred in 26.5%. Following aGVHD diagnosis, mortality rates were 30.0% at 6 months and 37.3% at 12 months. Findings from this study demonstrate a continuing unmet need for new therapies that control aGVHD and improve mortality.


Introduction
Allogeneic hematopoietic stem cell transplantation (HSCT) is the only potentially curative therapy for hematologic cancers and other diseases of the hematopoietic system [1][2][3].However, 30% to 60% of allogeneic HSCT recipients develop acute graft-versus-host disease (aGVHD) [4][5][6], which primarily manifests in the skin, liver, and gastrointestinal (GI) tract [7].aGVHD occurs when alloreactive donor T-cells attack healthy tissue rather than tumor cells [8].Risk factors for the development of aGVHD include the extent of HLA mismatch (unrelated donor or HLA-mismatched donor), stem cell source (e.g., peripheral blood, bone marrow), sex disparity between donor and recipient, higher-intensity conditioning regimen, and the type of GVHD prophylaxis [8].
Acute GVHD is a potentially life-threatening complication that represents a major cause of non-relapse mortality [9,10].Furthermore, approximately 35% to 60% of patients do not respond to, or become refractory to, recommended first-line treatment with corticosteroids [4,11,12], with considerably higher mortality rates observed among patients with steroid-resistant disease than in those with steroid-sensitive disease [12].
In a prior analysis from the present study, that included only patients from Finland and Sweden with grades II-IV aGVHD, nearly 90% of patients required hospitalization, with a median length of stay of 26 days and over 10 outpatient or emergency department visits per year on average [18].Patients who received multiple treatment lines had longer hospital stays than those who received only one line of treatment.The aim of this analysis was to describe real-world treatment patterns and clinical outcomes among patients who developed moderate or severe aGVHD following allogeneic HSCT in Finland, Sweden, and France.

Study Design and Patients
This retrospective observational chart review study was originally planned in six European countries (Germany, Italy, Sweden, Finland, United Kingdom, and France).Due to the sponsor's decision, the study was terminated early and was not initiated in Italy or Germany; an independent study was run in the United Kingdom according to a local protocol and case report form, but the limited data collected were not suitable for inclusion in this analysis.Thus, the study was ultimately conducted in Finland and Sweden (the AGHOS study) and in France (the GRAFITE study).The two studies had a common protocol and study design, with minor differences noted below.The present analysis describes data collected across 13 study centers (two centers in Finland, one in Sweden, and 10 in France).
Eligible patients were ≥18 years old and received allogeneic HSCT from any donor source using bone marrow, peripheral blood stem cells, or umbilical cord blood between 1 January 2016 and 30 June 2017.At least 12 months before data collection, patients were diagnosed with grades II-IV aGVHD per Mount Sinai Acute GVHD International Consortium (MAGIC) criteria [19] or, alternatively, grades II-IV according to the Glucksberg Severity Index or Keystone criteria [20], or grades B-D based on International Bone Marrow Transplant Registry [IBMTR] criteria [21].In France, only modified Glucksberg criteria were allowed for grading at diagnosis.Comparability across the different grading systems was ensured by assessing aGVHD severity based on the extent of skin, liver, and GI involvement.Grades II-IV aGVHD were defined as skin stage ≥ 3 and/or liver ≥ 1 and/or GI ≥ 1 per MAGIC and modified Glucksberg/Keystone criteria; grades B-D were defined as skin stage ≥ 2 and/or liver ≥ 1 and/or GI ≥ 1 per IBMTR criteria (Supplementary Figure S1) [22].Patients who were graded based on MAGIC criteria were compared with those graded on alternative scales by using mapping rules derived from grade and organ score definitions for each respective scale (Supplementary Figures S2 and S3) [18].
Patients were required to have medical records containing clinical details of the original disease that led to an HSCT, as well as clinical information on aGVHD presentation and treatment.Patients who developed aGVHD were included retrospectively, beginning with those who received HSCT on 30 June 2017, then working backwards until 1 January 2016.In France, all consecutive patients were identified, and the final cohort for inclusion was randomly selected.In Finland and Sweden, patients were included up until 1 January 2016, or until the target sample size was reached (whichever occurred first).A sample size of 135 patients was predefined to describe continuous and categorical variables with a predefined precision level of 0.06; 13% was the maximum percentage of patients with nonavailable data due to the retrospective data.The coefficient of variation (standard deviation [SD]/mean) considered was 0.5, with 95% CI for continuous variables corresponding to 0.128.For categorical variables, an occurrence rate of 50% was considered with a precision level of 6%.

Data Collection
Patient medical records were reviewed from the index date (date of allogeneic HSCT) until the day of data collection or until death or lost to follow-up (whichever occurred first).Detailed patient demographics, transplant characteristics, HSCT characteristics and clinical presentation, treatments, and outcomes of aGVHD were abstracted from medical records retrospectively and entered into electronic case report forms.Disease diagnoses (e.g., comorbidities) and medical procedures were coded per the Medical Dictionary for Regulatory Activities.

Ethical Considerations
The study was conducted in accordance with the study protocol, the Declaration of Helsinki, and all relevant regulatory requirements.Independent ethics committees/ institutional review boards approved the study protocol prior to patient enrollment.The GRAFITE study was conducted in compliance with MR-004 in France.Personal data were processed in accordance with the EU General Data Protection Regulation.

Statistical Analyses
All statistical analyses were conducted using SAS ® version 9.2 or higher (SAS Institute, Cary, NC, USA).Data were summarized using descriptive statistics and reported for the overall population and stratified by country.Continuous variables were reported as mean (SD) or median (Q1, Q3); categorical values were summarized as frequency and percentage of the total study population, and by subgroups where appropriate.A prespecified subgroup analysis was conducted, based on the prophylactic regimen received, to evaluate clinical outcomes (e.g., time from HSCT to aGVHD diagnosis, best overall response, cGVHD development, recurrence of GVHD).

Subgroup Analysis
A subgroup analysis assessed clinical outcomes stratified by the prophylaxis regimen received (Table 4).Among patients who received the most common prophylactic regimen of cyclosporine and mycophenolate (n = 40), the median (Q1, Q3) time from transplant to aGVHD diagnosis was 28.0 ( 17 ).Additionally, of the 40 patients who received prophylactic cyclosporine plus mycophenolate, 15 (37.5%) had aGVHD recurrence, which was the highest percentage across the prophylactic regimen subgroups.Lastly, of the 15 patients who received in vivo T-cell depletion, cyclosporine, and methotrexate, aGVHD recurrence was observed in two patients (13.3%), which was the lowest rate of aGVHD recurrence.

Discussion
This real-world study provides a comprehensive overview of the clinical characteristics, treatment patterns, and clinical outcomes of patients who developed aGVHD following allogeneic HSCT in real-world settings in three European countries.Most patients received first-line treatment for aGVHD with methylprednisolone-containing regimens, although variations were seen in the treatment approaches across countries.The overall response rate to these agents in the first-line setting was nearly 80%.However, the patients in this study population were ultimately heavily treated for aGVHD, with over one-quarter receiving three or more lines of treatment.More than one-third of patients developed cGVHD following aGVHD, and nearly one-third of patients in the study died from any cause within 6 months of their aGVHD diagnosis.There may be many reasons for the apparently lower mortality rate among patients from Finland vs. Sweden or France.Since the data reflect all-cause mortality, one possible explanation is that there could have been more patients with mild aGVHD in the Finnish cohort and more patients with severe aGVHD in the French cohort, with excess deaths owing to the procedure rather than disease relapse.Further, it is the authors' understanding that treatment-related mortality is generally relatively low in Finnish centers, which may reflect pretreatment practices (e.g., the potentially greater utilization of treosulfan there than in other centers).
Findings from this study support the limited previous literature showing poor outcomes among patients who develop aGVHD post-HSCT, including several retrospective claims analyses conducted in the US [13][14][15]23].These previous studies demonstrated that patients who developed aGVHD following HSCT experienced an increased incidence of serious organ system conditions and infections, higher healthcare resource utilization and costs, and greater mortality than those who did not develop aGVHD.While the most common first-line treatment with methylprednisolone in the current study was similar to that utilized in a 2014-2016 US-based chart review (corticosteroids, 97.9%) [23], aGVHD prophylaxis received in the current study (cyclosporine 84.8%; mycophenolate 55.0%) differed from that administered in US-based studies, which showed that the use of tacrolimus as aGVHD prophylaxis has grown since 1999-2002, virtually replacing cyclosporine during the 2006-2012 period (27% vs. 72% and 80% vs. 17%, respectively) [16].Additionally, in a US-based chart review, the majority of patients who received allogeneic HSCT

Discussion
This real-world study provides a comprehensive overview of the clinical characteristics, treatment patterns, and clinical outcomes of patients who developed aGVHD following allogeneic HSCT in real-world settings in three European countries.Most patients received first-line treatment for aGVHD with methylprednisolone-containing regimens, although variations were seen in the treatment approaches across countries.The overall response rate to these agents in the first-line setting was nearly 80%.However, the patients in this study population were ultimately heavily treated for aGVHD, with over one-quarter receiving three or more lines of treatment.More than one-third of patients developed cGVHD following aGVHD, and nearly one-third of patients in the study died from any cause within 6 months of their aGVHD diagnosis.There may be many reasons for the apparently lower mortality rate among patients from Finland vs. Sweden or France.Since the data reflect allcause mortality, one possible explanation is that there could have been more patients with mild aGVHD in the Finnish cohort and more patients with severe aGVHD in the French cohort, with excess deaths owing to the procedure rather than disease relapse.Further, it is the authors' understanding that treatment-related mortality is generally relatively low in Finnish centers, which may reflect pretreatment practices (e.g., the potentially greater utilization of treosulfan there than in other centers).
Findings from this study support the limited previous literature showing poor outcomes among patients who develop aGVHD post-HSCT, including several retrospective claims analyses conducted in the US [13][14][15]23].These previous studies demonstrated that patients who developed aGVHD following HSCT experienced an increased incidence of serious organ system conditions and infections, higher healthcare resource utilization and costs, and greater mortality than those who did not develop aGVHD.While the most common first-line treatment with methylprednisolone in the current study was similar to that utilized in a 2014-2016 US-based chart review (corticosteroids, 97.9%) [23], aGVHD prophylaxis received in the current study (cyclosporine 84.8%; mycophenolate 55.0%) differed from that administered in US-based studies, which showed that the use of tacrolimus as aGVHD prophylaxis has grown since 1999-2002, virtually replacing cyclosporine during the 2006-2012 period (27% vs. 72% and 80% vs. 17%, respectively) [16].Additionally, in a US-based chart review, the majority of patients who received allogeneic HSCT between 2014 and 2016 received tacrolimus-based GVHD prophylaxis (56.4%) rather than cyclosporine or mycophenolate (26.1% and 44.0%, respectively) [23].
Although findings were not compared against a non-GVHD cohort in the current study, substantial all-cause mortality was observed among patients who developed aGVHD after HSCT.In the current analysis, at the end of the first year after aGVHD diagnosis, 63% of patients were alive, which fell within the range reported in a prior CIBMTR study (40-70% 1-year survival, depending on aGVHD grade) [16].Mortality rates in the current study were also similar to 2006-2012 1-year mortality rates in a US-based study (Grade II aGVHD: 30%; Grade III-IV aGVHD: 60%) [16], and in the 2014-2016 US-based chart review (Grade II aGVHD: 31.8%;Grade III-IV aGVHD: 41.7%) [23].
Our group has previously shown that moderate to severe aGVHD is associated with high rates of hospitalizations and outpatient visits among patients receiving allogeneic HSCT at transplant centers in Finland and Sweden [18].Taken together, findings from this study and previous reports demonstrate the substantial burden of aGVHD among allogeneic HSCT recipients and highlight the need for more effective treatment strategies that can improve clinical outcomes.
Several limitations should be noted.First, this was a retrospective chart review study with potential for inaccurate or missing data entries.Additionally, the study was characterized by a small sample size, particularly in the Nordic populations.Finally, study data are heterogeneous due to several reasons, such as differences between centers in aGVHD diagnosis, treatment of aGVHD, and numbers of recruited patients.Variability existed in patient selection across countries, which may have influenced clinical outcomes.The differences in staging and grading systems among sites (i.e., MAGIC vs. modified Glucksberg) may have affected the results, although an alignment process was performed to ensure compatibility across grading systems (Supplementary Figures S1-S3).The alignment process grouped and mapped modified Glucksberg cases into MAGIC (instead of doing the opposite, i.e., mapping or naively comparing patients graded in MAGIC with patients graded in Glucksberg) [18].

Conclusions
In conclusion, moderate to severe aGVHD post-HSCT is associated with poor clinical outcomes, including the need for multiple lines of treatment, the development of cGVHD, and high mortality rates.Additional prospective studies covering a greater geographic domain are needed to better understand the widespread burden of aGVHD among patients receiving allogeneic HSCT for hematologic malignancies and disorders.

Figure 1 .
Figure 1.First-line Treatment of aGVHD: (A) Percentage of Patients Receiving Methylprednisolonebased Treatment Regimens (Monotherapy or Combination Therapy), (B) Methylprednisolone-containing Treatment Regimens, 1 and (C) Response to First-Line Treatment with Methylprednisolone. 2 aGVHD, acute graft-versus-host disease; CR, complete response; CS, corticosteroids; MP, methylprednisolone; PR, partial response; VGPR, very good partial response. 1Percentages of patients who received each type of regimen 2 as monotherapy or combined with any other treatments.

Figure 1 .
Figure 1.First-line Treatment of aGVHD: (A) Percentage of Patients Receiving Methylprednisolonebased Treatment Regimens (Monotherapy or Combination Therapy), (B) Methylprednisolonecontaining Treatment Regimens, 1 and (C) Response to First-Line Treatment with Methylprednisolone. 2 aGVHD, acute graft-versus-host disease; CR, complete response; CS, corticosteroids; MP, methylprednisolone; PR, partial response; VGPR, very good partial response. 1Percentages of patients who received each type of regimen 2 as monotherapy or combined with any other treatments.

Table 4 .
Subgroup Analysis of Clinical Outcomes by Prophylactic Treatment Regimen.1

Figure 2 .
Figure 2. All-Cause Mortality at 6 and 12 Months and End of Follow-Up Since aGVHD Development.aGVHD, acute graft-versus-host disease.

Figure 2 .
Figure 2. All-Cause Mortality at 6 and 12 Months and End of Follow-Up Since aGVHD Development.aGVHD, acute graft-versus-host disease.

Table 2 .
Patient Demographics and Clinical Characteristics.