Abstract
Uncontrolled studies have suggested that bone marrow-derived mesenchymal stem cells (MSCs) may be effective against acute graft-versus-host disease (aGVHD). We conducted a multicenter, randomized study to assess the efficacy of using ex vivo cultured adult human MSC (remestemcel-L) in addition to second-line therapy to treat steroid-refractory aGVHD ( NCT00366145). In total, 260 patients, 6 months to 70 years of age, were enrolled from August 2006 to May 2009 and were randomized 2:1 to receive 8 intravenous infusions of remestemcel-L or placebo, given over 4 weeks, in addition to second-line therapy according to institutional standards. Four additional infusions over 4 weeks were indicated for patients with incomplete response at day 28. Randomization was stratified by aGVHD grade. Efficacy and safety were assessed through 180 days of follow-up, with the primary endpoint being durable complete response (DCR), defined as complete resolution of aGVHD symptoms for any period of at least 28 days after beginning treatment. Remestemcel-L did not meet the primary endpoint of greater DCR in the intent-to-treat population (35% versus 30%; P = .42). In post hoc analyses, patients with liver involvement who received at least 1 infusion of remestemcel-L had a higher DCR, and higher overall complete or partial response rate (OR) than those who received placebo (29% versus 5%; P = .047). Furthermore, pediatric patients had a higher OR with MSCs compared with placebo (64% versus 23%; P = .05). Similar rates of adverse events were observed between treatment groups. Remestemcel-L was safe and well tolerated. Results of this study did not demonstrate superior DCR compared with placebo when added to standard of care. The favorable clinical responses seen in some patient subsets may warrant further investigation.
Keywords: Mesenchymal stem cells, Acute graft-versus-host disease, Steroid, Allogeneic, Hematopoietic cell transplantation
INTRODUCTION
Although allogeneic hematopoietic cell transplantation (HCT) can cure patients with high-risk hematologic malignancies, acute graft-versus-host disease (aGVHD) remains a significant cause of early death [1]. Occurrence of aGVHD after allogeneic transplantation is common, with approximately 40% to 60% of patients experiencing moderate or severe grade II to IV aGVHD after T-replete allogeneic HCT [2,3].
Immunocompetent donor T cells, contained in the hematopoietic progenitor cell graft, cause aGVHD by reacting against histocompatibility antigens in the immunocompromised host, leading to an inflammatory response that injures target organs [4].
Administration of corticosteroids remains the standard first line of treatment for aGVHD. A systematic review of publications describing the results of corticosteroid treatment for patients with previously untreated grade II to IV aGVHD demonstrated a complete response (CR) rate of 48%; an overall response (OR) rate, defined as CR and partial response (PR), of 64%; and a weighted 6-month survival estimate of 66% [5]. However, patients with high-risk disease, generally defined as visceral and/or multiorgan aGVHD, respond to corticosteroids at significantly lower rates than those with standard-risk disease [6]. Temcell (JCR Pharmaceuticals Co. Ltd, Japan) (cryopreserved, allogeneic mesenchymal stem cells [MSCs]) was recently approved in Japan for the treatment of aGVHD in children and adults [7]. In the United States, the Food and Drug Administration recently approved ruxolitinib for the treatment of steroid-refractory (SR) aGVHD, but there is no agreed-upon optimal standard of care for these patients. Currently available approaches provide a 58% OR rate and 6-month survival at just 49% [5]; hence, more effective therapies are urgently needed for patients with SR aGVHD.
Results from in vitro studies [8–10] and animal models [11–15] have suggested that bone marrow-derived MSCs have immunosuppressive functions. Results from initial clinical studies have indicated a wide range of generally favorable response rates when MSCs were used to treat aGVHD [16–20]. Remestemcel-L ((Prochymal) Mesoblast (USA) bought Prochymal (Osiris, USA)), ex vivo cultured adult human MSCs derived from bone marrow, showed encouraging results in patients with de novo aGVHD [21]. Patients with grade II to IV aGVHD were randomized to receive 2 doses of either 2 or 8 million MSCs/kg added to conventional corticosteroids. The OR rate was 94%, with notably no difference in safety or efficacy between the high and low MSC dose [21].
The primary objective of this randomized, phase 3 study was to evaluate the efficacy of remestemcel-L versus placebo when added to second-line therapy per institutional standard of care in patients with moderate to severe SR aGVHD ( NCT00366145).
METHODS
Study Design
The study was conducted at 72 centers in 7 countries (United States, Canada, Australia, Germany, Italy, Spain, and the United Kingdom). The study protocol, all study protocol amendments, the written study patient information, written informed consent forms, and all other appropriate study-related information were reviewed and approved by an independent ethics committee (IEC) or institutional review board (IRB) at each study site.
Remestemcel-L, consisting of human culture-expanded MSCs from bone marrow aspirates of unrelated, HLA-unmatched donors, is manufactured as previously described [21,22]. Cell products were harvested at passage 5 from 4 male donors and 1 female donor, 19 to 27 years of age, with each infusion product derived from a single donor. MSCs were tested for chromosomal aberrations before use. Some patients who received multiple infusions did receive cell product derived from multiple donors.
Remestemcel-L was formulated in an excipient of Plasma-Lyte physiologic electrolyte solution (Baxter International, Inc. USA), containing 50 g/L human serum albumin and 10% v/v dimethyl sulfoxide, and was cryopreserved until use. Placebo consisted of excipient, without MSCs. At time of administration, the product was thawed and immediately infused into the patient without any additional priming.
Participants
A total of 260 male and female patients, 6 months to 70 years of age, with SR grade B to D aGVHD [23], were enrolled between August 2006 and May 2009. The protocol was amended in October 2007 (after 95 patients had been enrolled) to specifically define SR as failure to improve by at least 1 grade after at least 3 days and no greater than 2 weeks of at least 1 mg/kg/d methylprednisolone or equivalent. Patients may have increased their steroid dose before enrollment. Patients were to be treated within 4 days of randomization, and all patients were to receive second-line treatment for SR aGVHD during the study per institutional standard of care. In urgent situations, second-line therapy may have been started 24 hours before randomization, and remestemcel-L or placebo was to be initiated within the following 3 days. Patients must have had renal creatinine clearance of >30 mL/min and a minimum Karnofsky/Lansky performance of ≥30 at study entry.
Patients were excluded if they had received primary treatment for aGVHD with agents other than steroids or if they had started treatment with another second-line therapy more than 24 hours before randomization (added via October 2007 protocol amendment).
Randomization and Masking
All patients who signed an informed consent form and were determined to be eligible for study participation were randomized in a blinded fashion using a central Interactive Voice Response System (IVRS) on a 2:1 basis to either the active treatment group (remestemcel-L) or the placebo control group, blocked, and stratified by baseline aGVHD grade B versus grade C/D. Only designated technicians in the central processing laboratory at study sites were unblinded to treatment, as they prepared remestemcel-L and placebo infusions in identical infusion bags with identical investigational drug labels. Infusion bags were covered with an amber-colored blinding bag and infusion tubing was covered with an amber-colored tubing sheath. The designated technician in the central processing laboratory (or designee) was responsible for maintaining the investigational agent records, including the randomization treatment assignments by patient identification.
Procedures
Patients were pretreated with hydrocortisone and diphenhydramine (25 to 50 mg each for adults, 0.5 to 1 mg/kg up to 50 mg each for pediatric patient) administered intravenously 30 minutes before infusion. Remestemcel-L, at a dose of 2 × 106 hMSC/kg actual body weight or an equivalent volume of placebo, was infused at 4 to 6 mL/min, twice weekly for 4 consecutive weeks, with infusions at least 3 days apart. The schedule prescribed a total of 8 infusions by day 28. Remestemcel-L or placebo was given in addition to second-line therapy according to a prespecified institutional standard of care. Steroid treatment was to be continued and, if improvement in graft-versus-host disease (GVHD) was observed for a period of 3 to 5 days, could be tapered according to institutional guidelines. Supportive care consisting of transfusion support and anti-infective prophylaxis per institutional practice was allowed. Concomitant medications not approved for any indication were not allowed within 30 days before randomization or during the study.
Patients were assessed on 32 ± 2 days to determine eligibility for further infusions. Patients who displayed PR (improvement in at least 1 organ without worsening in any other organ) or mixed response (improvement in 1 organ but worsening in another) were eligible to receive further infusions once per week for an additional 4 weeks. Patients who developed grade B to D GVHD flare after CR before day 72 were to receive additional infusions according to initial randomized schedule. Patients remained in the study up to 180 days, unless they withdrew or died.
GVHD assessments were performed weekly on all patients from enrollment through day 99 and at days 130, 160, and 180. Severity of aGVHD for each patient was evaluated using the International Blood and Marrow Transplant Registry grading criteria [23]. Presence or absence of aGVHD of the skin, liver, and gut was determined and graded according to the clinical assessment of each patient.
All untoward medical occurrences after the initial treatment were considered adverse events (AEs). Serious AEs were defined according to International Conference on Harmonization E6 standards [24]. Ectopic tissue formation was an AE of special interest and all patients were assessed at prespecified time points with computed tomography/magnetic resonance imaging. Vital signs, physical examination, and laboratory assessments were performed at screening, weekly through study day 63, and at study days 100, 130, 160, and 180/end of study, and urinalysis assessments at screening and study day 180/end of study.
Outcomes
The primary endpoint was durable complete response (DCR), defined as CR for at least 28 consecutive days within the first 100 days after enrollment. CR was defined as the resolution of aGVHD in all involved organs. Key secondary endpoints were survival at days 100 and 180 and cumulative steroid usage. Additional efficacy variables of interest included time to complete response, incidence of CR or PR for each organ, and durability of organ response.
Safety endpoints for the study included infusion toxicity, formation of ectopic tissue foci, AEs, and overall relapse of underlying disease.
Statistical Analysis
The study enrolled 260 patients at 72 sites to obtain the 240 evaluable patients needed to detect an absolute improvement in DCR of 20% over historical experience with 80% power at the .05 significance level, assuming a DCR rate of 49% in remestemcel-L-treated patients.
Analysis of the primary endpoint, DCR, was performed using the Cochran-Mantel-Haenszel test stratified by baseline GVHD grade B versus C/D. Other analyses are unstratified, and all P values reported below except for the primary endpoint are from the Fisher exact test.
The primary endpoint was evaluated in the intent-to-treat (ITT) population, which included all randomized patients. All other efficacy and safety analyses were performed on the modified ITT (mITT) population, which consisted of all patients who received at least 1 dose of remestemcel-L or placebo (n = 244). Prespecified subgroup analyses included those defined by age, sex, duration of steroid treatment, second-line therapies received, and baseline organ involvement. Exploratory analysis of clinical response and survival in high versus standard risk groups was also conducted.
Since the conclusion of the trial in 2009, consensus in the field has established that OR, defined as CR + PR, at day 28, is a more appropriate endpoint for first-line GVHD treatment studies [25,26]. We therefore conducted exploratory analysis of the OR endpoint, using data collected at week 4 and day 32 assessments, and using the following protocol-specified response criteria:
Response criteria:
CR: resolution of aGVHD in all involved organs
- PR: decrease by at least 1 GVHD stage in any 1 organ
- ○System without any worsening in any other organ system
Durable response: a response (eg, improvement in GVHD grade) lasting for at least 28 days
Mixed response: improvement by at least 1 stage in at least 1 evaluable organ with worsening by at least 1 stage in at least 1 other organ
No response: stable or worsening disease
- Stable disease: the absence of any clinically significant differences
- ○(Improvement or worsening) sufficient to meet minimal criteria for improvement or deterioration in any evaluable organ
- ○Worsening: deterioration in at least 1 organ system by 1 stage or more with no improvements in any other organ
Flare: recurrence of aGVHD after a CR
Recent developments in the field also suggest that risk-scoring aGVHD patients may be useful for establishing treatment protocols and predicting response to treatments. We applied an established risk-scoring methodology to patients enrolled in this study to explore response to treatments based on risk. Minnesota risk scoring defines high risk as either baseline stage 4 skin, stage 3 to 4 lower GI, stage 3 to 4 liver, or stage 3 to 4 skin + stage 3 to 4 liver or lower GI, with all other patients classified as standard risk [6,27].
A Data Safety Monitoring Board (DSMB) composed of physicians with clinical expertise in treatment of GVHD and a biostatistician assessed the progress of the trial and reviewed efficacy and safety data after 25% and 50% of randomized patients completed treatment and made recommendations to continue, modify, or terminate the trial.
Role of Funding Source
The preliminary results from this study were presented in abstract form in 2010 [28]. Prespecified analysis results reported in this article were produced by the original sponsor (Osiris Therapeutics, Inc. USA) and were reproduced and confirmed by the current sponsor (Mesoblast, Inc. USA). Post hoc analysis results were produced by the current sponsor. All drafts of this article were prepared by the current sponsor and were reviewed and edited by all authors. All authors take responsibility for the accuracy and completeness of the analysis and for the fidelity of this report to the study protocol.
RESULTS
Figure 1 shows a CONSORT diagram for the study. In total, 260 patients were enrolled and randomized from August 2006 to May 2009, with 173 to receive remestemcel-L and 87 to receive placebo. Of the 260 randomized patients, 244 received at least 1 infusion of remestemcel-L (n = 163) or placebo (n = 81), and these patients comprise the mITT and safety populations. A total of 223 completed the study, 151 (87%) from the remestemcel-L group and 72 (83%) from the placebo group. Thirty-eight enrolled patients did not meet protocol inclusion/exclusion criteria, and 32 were granted a waiver to protocol eligibility criteria by the sponsor’s medical monitor. Thirty-five of the 38 patients who did not meet all eligibility criteria were treated and therefore included in the mITT and safety populations.
Figure 1.
Trial profile. A total of 38 patients (28 remestemcel-L, 10 placebo) did not meet inclusion or exclusion criteria, and 32 of these patients were granted a waiver to protocol eligibility criteria by the sponsor’s medical monitor. Thirty-five of these patients are included in the safety and mITT populations (26 remestemcel-L, 9 placebo), having received at least 1 infusion. Entry criteria that were waived include patient age outside of range specified in the protocol (8 remestemcel-L, 3 placebo); patient not considered to have failed to respond to steroid treatment (3 remestemcel-L, 1 placebo); patient not able to be treated with 4 days of randomization (4 remestemcel-L, 1 placebo); patient started second-line therapy before randomization (6 remestemcel-L); patient received experimental therapy not approved by the Food and Drug Administration within 30 days of randomization (2 remestemcel-L); pregnancy, breastfeeding, or contraceptive requirements (1 remestemcel-L, 1 placebo); and patient received primary therapy other than steroids (1 remestemcel-L).
Table 1 shows baseline demographic and disease characteristics for the mITT population. No differences were noted in baseline age, race, and sex between the remestemcel-L and placebo groups, although children were underrepresented in the remestemcel-L group, based on a 2:1 randomization schedule. The proportion of patients with grade D aGVHD was higher in the remestemcel-L group than in the placebo group (27% versus 16%). Single-organ disease was over-represented and was generally more severe in the remestemcel-L group, whereas disease involving GI + skin was over-represented and generally less severe in the placebo group. To further characterize imbalance in disease presentation between treatment groups, we applied previously published and validated aGVHD risk-scoring methods [6,27] to patients in this study and found that 64% of the remestemcel-L group and 53% of the placebo group were identified as having high-risk aGVHD. Table 1 also shows GVHD prophylaxis and second-line therapy usage. All patients received cyclosporine, tacrolimus, or a combination of prophylactic agents. A variety of second-line agents was administered as per institutional standard. Notably, 11% of remestemcel-L patients did not receive any additional second-line therapy, compared with 4% of placebo patients.
Table 1.
Baseline Characteristics for the mITT Population
| Characteristic | Remestemcel-L (n = 163) | Placebo (n = 81) |
|---|---|---|
| Sex | ||
| Male | 91 (56) | 48 (59) |
| Female | 72 (44) | 33 (41) |
| Age (yr) | 43.8 (16.6) | 40.0 (18.1) |
| Age <18yr | 14 (9) | 13 (16) |
| Age ≥18yr | 149 (91) | 68 (84) |
| Karnofsky/Lansky score ≥30 | 159 (98) | 81 (100) |
| Donor compatibility | ||
| Unrelated | 93 (57) | 47 (58) |
| Related | 70 (43) | 34 (42) |
| Type of transplant | ||
| Bone marrow | 20 (12) | 14 (17) |
| PBSC | 127 (78) | 57 (70) |
| Cord blood | 16 (10) | 10 (12) |
| Days on steroids before randomization | ||
| 3 to 14 days | 98 (60) | 57 (70) |
| > 14 days | 61 (37) | 24 (30) |
| GVHD grade at diagnosis | ||
| Grade B | 37 (23) | 21 (26) |
| Grade C | 82 (50) | 46 (57) |
| Grade D | 44 (27) | 14 (17) |
| Organ involvement | ||
| Single organ | 89 (55) | 35 (43) |
| Lower GI | 49 (30) | 22 (27) |
| Skin | 35 (22) | 13 (16) |
| Liver | 5 (3) | 0 (0) |
| Two organs | 57 (35) | 40 (49) |
| Lower GI + skin | 36 (22) | 27 (33) |
| Lower GI + liver | 16 (10) | 7 (9) |
| Liver + skin | 5 (3) | 6 (8) |
| Three organs | 16 (10) | 6 (7) |
| Minnesota risk | ||
| High | 104 (64) | 43 (53) |
| Standard | 59 (36) | 38 (47) |
| GVHD prophylaxis | ||
| Cyclosporine | 71 (44) | 45 (56) |
| Tacrolimus | 86 (53) | 38 (47) |
| Cellcept | 54 (33) | 31 (38) |
| Cyclosporine/Cellcept | 26 (16) | 19 (24) |
| Cyclosporine/methotrexate | 25 (15) | 14 (17) |
| Cyclosporine/tacrolimus/Cellcept | 0 (0) | 4 (5) |
| Cyclosporine/tacrolimus/methotrexate | 0 (0) | 0 (0) |
| Other–methotrexate | 60 (37) | 24 (30) |
| Other–nonmethotrexate | 14 (9) | 7 (9) |
| Second-line therapy | ||
| Antithymocyte globulin | 35 (22) | 17 (21) |
| Mycophenolate mofetil | 27 (17) | 18 (22) |
| Infliximab | 29 (18) | 13 (16) |
| Etanercept | 23 (14) | 11 (14) |
| Daclizumab | 16 (10) | 13 (16) |
| Pentostatin | 4 (3) | 9 (11) |
| Denileukin diftitox | 7 (4) | 2 (2) |
| Extracorporeal photopheresis | 5 (3) | 2 (2) |
| Alemtuzumab | 1 (0.1) | 0 (0) |
| None | 18 (11) | 3 (4) |
Age is calculated from the screening date and presented as mean (SD). All other data are frequency (%). Percentages for organ involvement are based on patients presenting with the specific combination of organs among all patients. Minnesota risk scoring classifies patients as standard or high risk based on organ staging at baseline using methods reported in MacMillan et al. [6,27] and defines high risk as stage 4 skin, stage 3 to 4 lower GI, stage 3 to 4 liver, or skin stage 3 to 4 + stage 3 to 4 liver or lower GI, with all other patients classified as standard risk. A patient could receive more than one GVHD prophylaxis. Patients in the combined categories are counted in the individual categories. A patient could receive more than 1 second-line agent. Patients are counted in each category if they received that agent.
PBSC indicates peripheral blood stem cell.
In the ITT population, durable complete response lasting at least 28 days (DCR) was observed in 35% of patients in the remestemcel-L group compared with 30% in the placebo group (P = .42). Results for DCR and OR at day 28 for the mITT population overall and for prespecified subgroups are shown in Table 2 (frequencies, percentages, Fisher exact P values) and Figure 2 (odds ratios and 95% confidence intervals [CIs]), as are results for post hoc subgroups of standard- versus high-risk patients and patients with single-organ involvement versus multiorgan involvement. DCR was significantly improved in patients with any liver involvement (29% versus 5%; P = .05).
Table 2.
Durable Complete Response ≥28 Days and Overall Response at Day 28–mITT Population
| Durable Complete Response, n (%) [95% CI] | Overall Response, n (%) [95% CI] | |||
|---|---|---|---|---|
| Characteristic | Remestemcel-L | Placebo | Remestemcel-L | Placebo |
| Overall | 60/163 (36.8) | 26/81 (32.1) | 95/163 (58.3) | 44/81 (54.3) |
| [29.4–44.7] | [22.2–43.4] | [50.3–65.9] | [42.9–65.4] | |
| Grade | ||||
| B | 15/37 (40.5) | 7/21 (33.3) | 18/37 (48.6) | 14/21 (66.7) |
| [24.8–57.9] | [14.6–57.0] | [31.9–65.6] | [43.0–85.4] | |
| C | 34/82 (41.5) | 16/46 (34.8) | 56/82 (68.3) | 23/46 (50.0) |
| [30.7–52.9] | [21.4–50.3] | [57.1–78.1] | [34.9–65.1] | |
| D | 11/44 (25.0) | 3/14 (21.4) | 21/44 (47.7) | 7/14 (50.0) |
| [13.2–40.3] | [4.7–50.1] | [32.5–63.3] | [23.0–77.0] | |
| C/D | 45/126 (35.7) | 19/60 (31.7) | 77/126 (61.1) | 30/60 (50.0) |
| [27.4–44.7] | [20.3–45.0] | [52.0–69.7] | [36.8–63.2] | |
| Minnesota risk | ||||
| Standard | 31/59 (52.5) | 17/38 (44.7) | 35/59 (59.3) | 28/38 (73.7) |
| [39.1–65.7] | [28.6–61.7] | [45.8–71.9] | [56.9–86.6] | |
| High | 29/104 (27.9) | 9/43 (20.9) | 60/104 (57.7) | 16/43 (37.2)† |
| [19.5–37.5] | [10.0–36.0] | [47.6–67.3] | [23.0–53.3] | |
| Organ | ||||
| Any liver | 12/42 (28.6) | 1/19 (5.3)† | 23/42 (54.8) | 5/19 (26.3)* |
| [15.7–44.6] | [0.1–26.0] | [38.7–70.2] | [9.2–51.2] | |
| Any lower Gl | 38/115 (33.0) | 18/59 (30.5) | 65/115 (56.5) | 27/59 (45.8) |
| [24.6–42.4] | [19.2–43.9] | [47.0–65.7] | [32.7–59.3] | |
| Any skin | 31/92 (33.7) | 18/52 (34.6) | 53/92 (57.6) | 29/52 (55.8) |
| [24.2–44.3] | [22.0–49.1] | [46.9–67.9] | [41.3–69.5] | |
| Single organ | 40/90 (44.4) | 15/38 (39.5) | 55/90 (61.1) | 27/38 (71.15) |
| [34.0–55.3] | [24.0–56.6] | [50.3–71.2] | [54.1–84.6] | |
| Multiorgan | 20/73 (27.4) | 11/43 (25.6) | 40/73 (54.8) | 17/43 (39.5) |
| [17.6–39.1] | [13.5–41.2] | [42.7–66.5] | [25.0–55.6] | |
| Age | ||||
| <18 | 9/14 (64.3) | 6/13 (46.2) | 9/14 (64.3) | 3/13 (23.1)* |
| [35.1–87.2] | [19.2–74.9] | [35.1–87.2] | [5.0–53.8] | |
| ≥18 | 51/149 (34.2) | 20/68 (29.4) | 86/149 (57.7) | 41/68 (60.3) |
| [26.7–42.4] | [19.0–41.7] | [49.4–65.8] | [47.7–72.0] | |
| Male | 32/91 (35.2) | 11/48 (22.9) | 51/91 (56.0) | 21/48 (43.8) |
| [25.4–45.9] | [12.0–37.3] | [45.3–66.4] | [29.5–58.8] | |
| Female | 28/72 (38.9) | 15/33 (45.5) | 44/72 (61.1) | 23/33 (69.7) |
| [27.6–51.1] | [28.1–63.7] | [48.9–72.4] | [51.3–84.4] | |
| Unrelated donor | 37/93 (39.8) | 17/47 (36.2) | 59/93 (63.4) | 25/47 (53.2) |
| [29.8–50.5] | [22.7–51.5] | [52.8–73.2] | [38.1–67.9] | |
| Related donor | 23/70 (32.9) | 9/34 (26.5) | 36/70 (51.4) | 19/34 (55.9) |
| [22.1–45.1] | [12.9–44.4] | [39.2–63.6] | [37.9–72.8] | |
| 3–14 days on steroids | 33/98 (33.7) | 20/57 (35.1) | 59/98 (60.2) | 30/57 (52.6) |
| [24.2–43.9] | [22.9–48.9] | [49.8–70.0] | [39.0–66.0] | |
| >14 days on steroids | 24/61 (39.3) | 6/24 (25.0) | 33/61 (54.1) | 14/24 (58.3) |
| [27.1–52.7] | [9.8–46.7] | [40.9–66.9] | [36.6–77.9] | |
| Second-line therapy | ||||
| Antithymocyte globulin | 14/34 (41.2) | 4/17 (23.5) | 21/34 (61.8) | 10/17 (58.8) |
| [24.7–59.3] | [6.8–49.9] | [43.6–77.8] | [32.9–81.6] | |
| Mycophenolate mofetil | 13/27 (48.2) | 5/16 (31.3) | 17/27 (63.0) | 8/16 (50.0) |
| [28.7–68.1] | [11.0–58.7] | [42.4–80.6] | [24.7–75.4] | |
| Infliximab | 13/29 (44.8) | 3/13 (23.1) | 18/29 (62.1) | 7/13 (53.9) |
| [26.5–64.3] | [5.0–53.8] | [42.3–79.3] | [25.1–80.8] | |
| Etanercept | 7/21 (33.3) | 5/11 (45.5) | 10/21 (47.6) | 4/11 (36.4) |
| [14.6–57.0] | [16.8–76.6] | [25.7–70.2] | [10.9–69.2] | |
| Daclizumab | 5/15 (33.3) | 6/11 (54.6) | 11/15 (73.3) | 6/11 (54.6) |
| [11.8–61.6] | [23.4–83.3] | [44.9–92.2] | [23.4–83.3] | |
Table displays frequency of response, number of patients assessed, percent response, and 95% confidence interval, computed by the method of Clopper and Pearson (Biometrika 26:404–413, 1934). Prespecified subgroups are GVHD grade at baseline, organ involvement, age category, sex, donor status, duration of steroid use, and use of second-line therapies. Categorization of risk and single- versus multiorgan involvement are post hoc subgroups. Risk scoring classifies patients as standard or high risk based on organ staging at baseline using methods reported in MacMillan et al. [6,27]. Minnesota risk scoring defines high risk as stage 4 skin, stage 3 to 4 lower GI, stage 3 to 4 liver, or skin stage 3 to 4 + stage 3 to 4 liver or lower GI, with all other patients classified as standard risk. Second-line therapies are included if at least 10% of the study population in a treatment arm or overall received them.
P value from Fisher exact test = .05.
P value from Fisher exact test <.05.
Figure 2.
DCR ≥ 28 days (left)/OR at day 28 (right): mITT population. Figure display of odds ratios and 95% confidence intervals, overall and for prespecified and post hoc subgroups, for DCR ≥ 28 days on the left and OR at day 28 on the right. Prespecified subgroups are GVHD grade at baseline, organ involvement, age category, sex, donor status, duration of steroid use, and use of second-line therapies. Categorization of risk and single- versus multiorgan involvement are post hoc subgroups. Risk scoring classifies patients as standard or high risk based on organ staging at baseline using methods reported in Macmillan et al. [6,27]. Minnesota risk scoring defines high-risk as stage 4 skin, stage 3 to 4 lower GI, stage 3 to 4 liver, or skin stage 3 to 4 + stage 3 to 4 liver or lower GI, with all other patients classified as standard risk. Second-line therapies are included if at least 10% of the study population in a treatment arm or overall received them.
The proportions of patients with OR at day 28 were similar in the remestemcel-L group compared with the placebo group (58% versus 54%; P = .59) overall. The proportions of patients with OR at day 28 were more consistent across most prespecified and post hoc subgroups in remestemcel-L patients than in placebo patients (Table 2). In the remestemcel-L group, OR at day 28 was observed in 59% of the patients with standard-risk aGVHD and in 58% of patients with high-risk aGVHD. In the placebo group, OR at day 28 was observed in 74% of the patients with standard-risk aGVHD versus 37% of those with high-risk aGVHD.
Among high-risk patients, remestemcel-L demonstrated significantly higher OR at day 28 than placebo (58% versus 37%; P = .03), whereas among standard-risk patients, remestemcel-L showed no benefit compared with placebo (59% versus 74%; P = .19). Additionally, remestemcel-L demonstrated improved OR at day 28 compared with placebo in pediatric patients (64% versus 23%; P = .05) and in patients with any liver involvement (55% versus 26%; P = .05) (Table 2).
Per protocol, all patients were to receive institutional-standard second-line therapy. No statistically significant results were found when comparing DCR or OR rates among remestemcel-L-treated and placebo-treated patients based on second-line therapy received, for agents received by at least 10% of patients overall, or in either treatment arm (Table 2). Average daily steroid usage by day of therapy and GVHD grade was recorded, without any significant trends noted for remestemcel-L versus placebo (data not shown).
Overall survival at day 180 was 34% (56/163) for remestemcel-L compared with 42% (34/81) for placebo in the mITT population (log-rank test = 0.28, P = .60, survival curve not shown). Figure 3 shows survival curves for standard- and high-risk patients, with no significant differences observed between treatment groups for either high-risk (log-rank P = .86) or standard-risk (log-rank P = .91) patients. However, risk of 6-month mortality was significantly increased for patients with high-risk aGVHD compared with patients with standard-risk disease for both remestemcel-L (RR = 1.5; 95% CI, 1.1 to 1.9) and placebo (Relative Risk (RR) = 1.6; 95% CI, 1.0 to 2.3) groups. The most common causes of death included infection and ensuing sepsis and/or multiorgan failure resulting from profound immunosuppression.
Figure 3.
Survival of standard- and high-risk patients: mITT population.
A summary of safety parameters is shown in Tables 3 to 5 and Supplementary Table S1. In total, 244 patients (163 remestemcel-L- and 81 placebo treated) who received at least 1 infusion of remestemcel-L or placebo were evaluated for safety. The remestemcel-L group was exposed to a slightly greater number of infusions than the placebo group. Remestemcel-L induced a similarly low rate of probable treatment-related AEs compared with placebo (1.2% versus 3.7%, respectively). Rates of serious AEs were similar between remestemcel-L and placebo groups. Infusion-related reactions were reported in 1.8% of remestemcel-L patients versus 2.5% of placebo patients. A trend toward higher rate of deaths associated with infections was observed in the remestemcel-L group versus placebo (42/163 [26%] versus 12/81 [15%]; P = .07) (Table 3). The rates of infection of any kind were similar between the remestemcel-L and placebo groups (88% versus 82%, respectively) (Table 4). Treatment-related AEs occurring in ≥5% of patients by treatment group are listed in Supplementary Table S1 with no significant difference between treatment groups. Peripheral edema (35%), abdominal pain (22%), and thrombocytopenia (22%) were the most common AEs noted; there were no treatment-related AEs related to ectopic tissue formation (Supplementary Table S1). Finally, we observed no difference between treatment groups in relapse rates of the underlying malignancy that led to HCT through the 180 days of follow-up (Table 5).
Table 3.
Safety Summary–Safety Population
| Characteristic | Remestemcel-L(n = 163) | Placebo (n = 81) |
|---|---|---|
| Exposure to investigational agent | ||
| Total number of infusions, n (%)* | ||
| 0–4 | 15 (9.2) | 13 (16.0) |
| 5–8 | 83 (50.9) | 42 (51.9) |
| 9–12 | 59 (36.2) | 25 (30.9) |
| 13–16 | 6 (3.7) | 1 (1.2) |
| Total cumulative dose (× 106 hMSCs/kg)† | ||
| Mean | 17.7 | 16.2 |
| SD | 6.1 | 6.1 |
| Minimum-maximum | 2–32 | 2–32 |
| Total volume (mL)‡ | 7.08 | 6.48 |
| Extent of exposure (days)§ | ||
| Mean | 36.0 | 31.7 |
| SD | 18.6 | 17.5 |
| Minimum-maximum | 1–97 | 1–77 |
| Treatment-related adverse events by relationship to treatment, n (%)‖ | ||
| Probable | 2 (1.2) | 3 (3.7) |
| Possible | 33 (20.2) | 12 (14.8) |
| Patients with ≥1 treatment-related adverse event by severity, n (%)¶,# | ||
| Grade 1 | 2 (1.2) | 3 (3.7) |
| Grade 3 | 38 (23.3) | 26 (32.1) |
| Grade 4 | 11 (6.7) | 5 (6.2) |
| Grade 5 | 104 (63.8) | 44 (54.3) |
| Grade 3 or higher | 153 (93.9) | 75 (92.6) |
| Patients with ≥1 treatment-related adverse event of infusion toxicity, n (%) | ||
| Patients having ≥1 TEAE of infusion toxicity | 2/163 (1.8) | 2/81 (2.5) |
| Serious AEs, n (%) | ||
| Patients having ≥1 serious AE | 146/163 (89.6) | 71/81 (87.7) |
| Serious AEs with outcome death, n (%) | ||
| Patients having ≥1 serious AE with outcome death | 105/163 (64.4) | 44/81 (54.3) |
| Serious AEs with outcome death by system organ class and relationship to treatment, n (%) | ||
| Blood & lymphatic system disorders | 1/163 (0.6) | 18/81 (1.2) |
| Cardiac disorders | 2/163 (1.2) | 4/81 (4.9) |
| GI disorders | 3/163 (1.8) | 3/81 (3.7) |
| General disorders & administration site conditions | 4/163 (2.5) | 5/81 (6.2) |
| Hepatobiliary disorders | 2/163 (1.2) | 3/81 (3.7) |
| Acute or chronic GVHD | 23/163 (14.1) | 5/81 (6.2) |
| Infections | 42/163 (25.8) | 12/81 (14.8) |
Safety population includes all patients who received at least 1 dose of remestemcel-L or placebo.
hMSC, human mesenchymal stem cell; TEAE, treatment emergent adverse event.
Total number of infusions is calculated from the total number of records on the infusion case report form page where total dose administered is greater than 0.
Cumulative dose is the number of infusions multiplied by 2 × 106 hMSCs/kg.
Total volume calculated from the mean total dose (× 106 hMSCs/kg) divided by the composition of reconstituted product, 2.5 × 106 hMSCs/mL.
Extent of exposure is the number of days from the first infusion to the last infusion.
Patients reporting more than 1 treatment-emergent adverse event within a primary system organ class, high-level term, or preferred term are counted only once at maximum relationship.
Patients reporting more than 1 treatment-emergent adverse event within a primary system organ class, high-level term, or preferred term are counted only once at maximum severity.
Severity grade 1 = mild, 2 = moderate, 3 = severe, 4 = life threatening or disabling, 5 = death related.
Table 5.
Relapse of Underlying Disease by Treatment Group–Safety Population
| Characteristic | Remestemcel-L, n(%) | Placebo, n(%) |
|---|---|---|
| Patients having ≥1 relapse of underlying disease | 13/163 (8.0) | 8/81 (9.9) |
| Disease progression | 1/163 (0.6) | 0/81 (0.0) |
| Acute myeloid leukemia | 6/163 (3.7) | 2/81 (2.5) |
| Acute myeloid leukemia recurrent | 1/163 (0.6) | 1/81 (1.2) |
| Leukemia recurrent | 2/163 (1.2) | 1/81 (1.2) |
| Leukemia | 1/163 (0.6) | 0/81 (0.0) |
| Myelodysplastic syndrome | 0/163 (0.0) | 1/81 (1.2) |
| Central nervous system leukemia | 0/163 (0.0) | 1/81 (1.2) |
| Acute lymphocytic leukemia | 0/163 (0.0) | 1/81 (1.2) |
| Chronic myeloid leukemia | 0/163 (0.0) | 1/81 (1.2) |
| Mantle cell lymphoma | 0/163 (0.0) | 1/81 (1.2) |
Safety population includes all patients who received at least 1 dose of remestemcel-L or placebo. Patients reporting more than 1 treatment-emergent adverse event within a primary system organ class, high-level term, or preferred term are counted only once.
Table 4.
Incidence of Infections–Safety Population
| Remestemcel-L | Placebo | |||
|---|---|---|---|---|
| Characteristic | Patients, n (%) | Events, n | Patients, n (%) | Events, n |
| All infections | 144 (88.3) | 521 | 66 (81.5) | 217 |
| Staphylococcal infections | 42 (25.8) | 58 | 12 (14.8) | 16 |
| Staphylococcal bacteremia | 21 (12.9) | 26 | 5 (6.2) | 5 |
| Staphylococcal infection | 16 (9.8) | 23 | 5 (6.2) | 7 |
| Sepsis, bacteremia, viremia, and fungemia NEC | 40 (24.5) | 53 | 19 (23.5) | 27 |
| Sepsis | 19 (11.7) | 19 | 6 (7.4) | 9 |
| Cytomegalovirus infections | 39 (23.9) | 57 | 19 (23.5) | 22 |
| Cytomegalovirus infection | 26 (16) | 33 | 11 (13.6) | 13 |
| Pneumonia cytomegalovirus | 4 (2.5) | 4 | 0 (0.0) | 0 |
| Bacterial infections NEC | 29 (17.8) | 37 | 9 (11.1) | 13 |
| Cellulitis | 10 (6.1) | 11 | 2 (2.5) | 2 |
| Bacterial infection | 6 (3.7) | 7 | 1 (1.2) | 1 |
| Enterococcal infections | 25 (15.3) | 30 | 13 (16.0) | 14 |
| Enterococcus bacteremia | 10 (6.1) | 10 | 1 (1.2) | 1 |
| Lower respiratory tract & lung infections | 22 (13.5) | 25 | 11 (13.6) | 13 |
| Pneumonia | 21 (12.9) | 22 | 6 (7.4) | 6 |
| Candida infections | 20 (12.3) | 26 | 7 (8.6) | 7 |
| Candidiasis | 9 (5.5) | 9 | 2 (2.5) | 2 |
| Oral candidiasis | 9 (5.5) | 11 | 2 (2.5) | 2 |
| Klebsiella infections | 19 (11.7) | 23 | 5 (6.2) | 6 |
| Klebsiella bacteremia | 8 (4.9) | 10 | 2 (2.5) | 2 |
| Fungal infections NEC | 18 (11.0) | 23 | 4 (4.9) | 4 |
| Fungal skin infection | 4 (2.5) | 4 | 0 (0.0) | 0 |
| Polyomavirus infections | 16 (9.8) | 19 | 5 (6.2) | 6 |
| BK virus infection | 15 (9.2) | 18 | 5 (6.2) | 6 |
| Infections NEC | 11 (6.7) | 12 | 3 (3.7) | 4 |
| Escherichia infections | 7 (4.3) | 7 | 1 (1.2) | 1 |
| Streptococcal infections | 7 (4.3) | 7 | 1 (1.2) | 1 |
| Alpha hemolytic streptococcal infection | 4 (2.5) | 4 | 0 (0.0) | 0 |
| Pseudomonal infections | 16 (9.8) | 22 | 2 (2.5) | 3 |
Safety population includes all patients who received at least 1 dose of remestemcel-L or placebo. Patients reporting more than 1 treatment-emergent adverse event within a primary system organ class, high-level term, or preferred term are counted only once.
DISCUSSION
Herein, we describe the results of the first multicenter, randomized, double-blind, placebo-controlled study of remestemcel-L for patients with SR aGVHD. The study did not meet its primary endpoint of durable complete resolution of all clinical signs of GVHD for at least 28 consecutive days. Furthermore, the overall response rate at day 28, currently endorsed as an appropriate endpoint for acute GVHD treatment studies [25,26], was similar between the remestemcel-L group and the placebo group. In post hoc analyses, specific patient subgroups that appear to benefit from remestemcel-L treatment compared with placebo were identified; these include pediatric patients, patients with any liver involvement, and patients with high-risk aGVHD, typically although not always involving multiple or visceral organs (Figure 2, Table 2). Patients treated with remestemcel-L and placebo experienced similar rates of AEs that were considered probably related to treatment (Tables 3 to 5), including relapse of the underlying disease (Table 5); there was a nonsignificant trend for increased infections for the remestemcel-L group, indicating some level of immune suppression related to MSC use, but risk for a particular class of infection was not identified. Further conclusions regarding infectious risk are limited due to the heterogeneity of second-line therapies. Broadly, the findings support the safety and tolerability of remestemcel-L as compared with other immunosuppressive agents for this population.
However, significant questions remain, including the optimal treatment schedule for MSC therapy and the ideal patient population. The early clinical improvement observed in high-risk subset patients treated with remestemcel-L did not produce survival benefit beyond the treatment period of 8 weeks, suggesting the possibility that more intensive treatment or a form of maintenance therapy, particularly for partial and mixed responders, could potentially produce durable clinical and survival outcomes. Future studies should explore alternative treatment schedules among patients with high-risk disease. Six-month survival for patients with standard-risk disease was comparable to what was observed in a meta-analysis of MSC-treated SR aGVHD patients [20].
Great variability in response to MSCs has been noted across studies. In an attempt to elucidate the mechanisms underlying the variability of response, Dazzi and colleagues [29] performed a set of elegant experiments in a murine GVHD model and demonstrated that MSCs need to undergo apoptosis by recipient cytotoxic cells to effect immune suppression. When apoptotic MSCs are engulfed by recipient phagocytes, indoleamine 2,3-dioxygenase is produced and leads to immune suppression. They studied this hypothesis in a series of 16 patients with SR GVHD with variable response to MSC therapy. The peripheral blood mononuclear cells from patients who responded to MSCs demonstrated significantly more cytotoxicity against MSC compared with nonresponders [29]. An ongoing phase 3 study by the Hemato-Oncologie voor Volwassenen Nederland (HOVON) cooperative group from the Netherlands is investigating the benefit of MSCs versus placebo added to patients with newly diagnosed severe aGVHD with gut and/or liver involvement to further investigate the possible benefit of MSCs for this patient group (EudraCT: 2011-003237-33).
Major limitations of the study were primarily the result of design and of operational deviations from the protocol. First, study design characteristics that had the potential to confound study results included the enrollment of a clinically heterogeneous patient population, including both pediatric and adult patients; the requirement that all patients, including those in the remestemcel-L arm, be treated with an active second-line agent that precluded observing a potential direct benefit from remestemcel-L; and the nonstandardized use of second-line agents across sites. Substantial variation in DCR rates observed in this study within and between treatment arms based on second-line therapy received underscores the extent of variability introduced with the use of heterogeneous second-line therapies. To minimize the confounding variable of heterogeneous second-line therapies, another ongoing European multicenter study by the HOVON cooperative group is investigating the benefit of MSCs in the SR setting compared with mycophenolate mofetil as the specified standard of care comparator (EudraCT: 2012-004915-30). Patients with SR GVHD, which includes visceral involvement, will be randomized to receive 2 weekly infusions of 2 million MSCs/kg or placebo in addition to mycophenolate mofetil.
Additionally, we observed an imbalance between treatment arms in extent (organ involvement) and severity (GVHD grade) of disease at baseline, and age category (pediatric versus adult), which may have confounded results. Finally, as per standard practice, the diagnosis of GVHD was clinical, and histology was not required. Thus, there may have been some overlap between drug toxicity and GVHD, especially in patients with GI and liver involvement, and this may have contributed some variability in GVHD diagnosis.
In conclusion, treatment with remestemcel-L, an unmatched allogeneic cellular product, was well tolerated but did not result in significant benefit over placebo when added to institutional-standard, second-line therapy for patients with SR acute GVHD. Much has changed in our understanding of GVHD and MSC biology since the completion of this study a decade ago. Still, the data obtained from over 100 patients treated with multiple infusions of uniformly manufactured product are informative with regards to safety and provide a benchmark for future studies that may investigate different schedules or formulations of MSCs. One area of continued investigation is whether fresh (or cryopreserved and primed) MSCs may demonstrate enhanced functionality in the clinic compared with the cryopreserved product. Additionally, for patients who presented with high-risk aGVHD involving multiple or visceral organs, particularly the liver, favorable responses were observed in post hoc analyses. Further focused studies are needed to evaluate whether remestemcel-L may provide a clinically significant benefit for SR aGVHD patients with visceral and high-risk disease.
Supplementary Material
ACKNOWLEDGMENTS
The authors thank the patients who consented to participate in this randomized controlled trial.
Financial disclosure: Mesoblast, Inc. (previous sponsor Osiris Therapeutics).
Footnotes
Conflict of interest statement:
A.D., J.U., D.I.M., E.K.W., and E.S. have nothing to disclose. R.S. has received consultancy fees from Gilead, Merck, and Astellas and served on board of directors for Kiadis and the DSMB of Juno/Celegene. P.J.M. received consultancy fees from Enlivex Therapeutics, Genentech, and Grafalon and served on the DSMB for Pfizer. P.K. received consultancy fees from Jazz Pharmaceuticals. D.S., E.B., and J.H. are employees of Mesoblast, Inc.
SUPPLEMENTARY MATERIALS
Supplementary material associated with this article can be found in the online version at doi:10.1016/j.bbmt.2019.08.029.
REFERENCES
- 1.Copelan EA. Hematopoietic stem-cell transplantation. N Engl J Med. 2006;354:1813–1826. [DOI] [PubMed] [Google Scholar]
- 2.Jagasia M, Arora M, Flowers ME, et al. Risk factors for acute GVHD and survival after hematopoietic cell transplantation. Blood. 2012;119:296–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gooley TA, Chien JW, Pergam SA, et al. Reduced mortality after allogeneic hematopoietic-cell transplantation. N Engl J Med. 2010;363:2091–2101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ferrara JLM, Levine JE, Reddy P, et al. Graft-versus-host disease. Lancet. 2009;373:1550–1561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Martin PJ, Rizzo JD, Wingard JR, et al. First- and second-line systemic treatment of acute graft-versus-host disease: recommendations of the American Society of Blood and Marrow Transplantation. Biol Blood Marrow Transplant. 2012;18:1150–1163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.MacMillan ML, Robin M, Harris AC, et al. A refined risk score for acute graft-versus-host disease that predicts response to initial therapy, survival, and transplant-related mortality. Biol Blood Marrow Transplant. 2015;21:761–767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Muroi K, Miyamura K, Okada M, et al. Bone marrow-derived mesenchymal stem cells (JR-031) for steroid-refractory grade III or IV acute graft-versus-host disease: a phase II/III study. Int J Hematology. 2016;103:243–250. [DOI] [PubMed] [Google Scholar]
- 8.Di Nicola M, Carlo-Stella C, Magni M, et al. Human bone marrow stromal cells suppress T-lymphocyte proliferation induced by cellular or nonspecific mitogenic stimuli. Blood. 2002;99:3838–3843. [DOI] [PubMed] [Google Scholar]
- 9.Rasmusson I, Ringden O, Sundberg B, Le Blanc K. Mesenchymal stem cells inhibit the formation of cytotoxic T lymphocytes, but not activated cytotoxic T lymphocytes or natural killer cells. Transplantation. 2003;76:1208–1213. [DOI] [PubMed] [Google Scholar]
- 10.Le Blanc K, Tammik L, Sundberg B, Haynesworth SE, Ringden O. Mesenchymal stem cells inhibit and stimulate mixed lymphocyte cultures and mitogenic responses independently of the major histocompatibility complex. Scand J Immunol. 2003;57:11–20. [DOI] [PubMed] [Google Scholar]
- 11.Maitra B, Szekely E, Gjini K, et al. Human mesenchymal stem cells support unrelated donor hematopoietic stem cells and suppress T-cell activation. Bone Marrow Transplant. 2004;33:597–604. [DOI] [PubMed] [Google Scholar]
- 12.Ramasamy R, Fazekasova H, Lam EW, Soeiro I, Lombardi G, Dazzi F. Mesenchymal stem cells inhibit dendritic cell differentiation and function by preventing entry into the cell cycle. Transplantation. 2007;83:71–76. [DOI] [PubMed] [Google Scholar]
- 13.Ramasamy R, Tong CK, Seow HF, Vidyadaran S, Dazzi F. The immunosuppressive effects of human bone marrow-derived mesenchymal stem cells target T cell proliferation but not its effector function. Cell Immunol. 2008;251:131–136. [DOI] [PubMed] [Google Scholar]
- 14.Polchert D, Sobinsky J, Douglas G, et al. IFN-gamma activation of mesenchymal stem cells for treatment and prevention of graft versus host disease. Eur J Immunol. 2008;38:1745–1755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Joo SY, Cho KA, Jung YJ, et al. Mesenchymal stromal cells inhibit graft-versus-host disease of mice in a dose-dependent manner. Cytotherapy. 2010;12:361–370. [DOI] [PubMed] [Google Scholar]
- 16.Prasad VK, Lucas KG, Kleiner GI, et al. Efficacy and safety of ex vivo cultured adult human mesenchymal stem cells (Prochymal) in pediatric patients with severe refractory acute graft-versus-host disease in a compassionate use study. Biol Blood Marrow Transplant. 2011;17:534–541. [DOI] [PubMed] [Google Scholar]
- 17.Lucchini G, Introna M, Dander E, et al. Platelet-lysate-expanded mesenchymal stromal cells as a salvage therapy for severe resistant graft-versus-host disease in a pediatric population. Biol Blood Marrow Transplant. 2010;16:1293–1301. [DOI] [PubMed] [Google Scholar]
- 18.Ringden O, Uzunel M, Rasmusson I, et al. Mesenchymal stem cells for treatment of therapy-resistant graft-versus-host disease. Transplantation. 2006;81:1390–1397. [DOI] [PubMed] [Google Scholar]
- 19.Kurtzberg J, Prockop S, Teira P, et al. Allogeneic human mesenchymal stem cell therapy (Remestemcel-L, Prochymal) as a rescue agent for severe refractory acute graft-versus-host disease in pediatric patients. Biol Blood Marrow Transplant. 2014;20(2):229–235. [DOI] [PubMed] [Google Scholar]
- 20.Hashmi S, Ahmed M, Murad MH, et al. Survival after mesenchymal stromal cell therapy in steroid-refractory acute graft-versus-host disease: systematic review and meta-analysis. Lancet Haematol. 2016;3(1):e45–e52. [DOI] [PubMed] [Google Scholar]
- 21.Kebriaei P, Isola L, Bahceci E, et al. Adult human mesenchymal stem cells added to corticosteroid therapy for the treatment of acute graft-versus-host disease. Biol Blood Marrow Transplant. 2009;15(7):804–811. [DOI] [PubMed] [Google Scholar]
- 22.Pittenger MF, Mackay AM, Beck SC, et al. Multilineage potential of adult human mesenchymal stem cells. Science. 1999;284:143–147. [DOI] [PubMed] [Google Scholar]
- 23.Rowlings PA, Przepiorka D, Klein JP, et al. IBMTR Severity Index for grading acute graft-versus-host disease: retrospective comparison with Glucksberg grade. Br J Haematol. 1997;97:855–864. [DOI] [PubMed] [Google Scholar]
- 24.Food and Drug Administration. Guidance for Industry. E6 Good Clinical Practice: Consolidated Guidance. Available at: http://www.fda.gov/downloads/Drugs/…/Guidances/ucm073122.pdf. Accessed February 2018.
- 25.Pavletic SZ. Response as an end point in treatment trials for acute GVHD. Bone Marrow Transplant. 2012;47:161–163. [DOI] [PubMed] [Google Scholar]
- 26.MacMillan ML, DeFor TE, Weisdorf DJ. The best endpoint for acute GVHD treatment trials. Blood. 2010;115:5412–5417. [DOI] [PubMed] [Google Scholar]
- 27.MacMillan ML, DeFor TE, Weisdorf DJ. What predicts high risk acute GVHD at onset: identification of those at highest risk by a novel acute GVHD risk score? Br J Haematol. 2012;157(6):732–741. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Martin PJ, Uberti JP, Soiffer RJ, et al. Prochymal improves response rates in patients with steroid-refractory acute graft versus host disease (SRGVHD) involving the liver and gut: results of a randomized, placebo-controlled, multicenter phase III trial in GVHD. Biol Blood Marrow Transplant. 2010;16(2):S169–S170. [Google Scholar]
- 29.Galleu A, Riffo-Vasquez Y, Trento C, et al. Apoptosis in mesenchymal stromal cells induces in vivo recipient-mediated immunomodulation. Sci Transl Med. 2017;9(416). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.