Palifermin for the Reduction of Acute GVHD: a Randomized, Double-Blind, Placebo-Controlled Trial

ORIGINAL ARTICLE Palifermin for the reduction of acute GVHD: a randomized, double-blind, placebo-controlled trial

MH Jagasia1, R Abonour2, GD Long3, BJ Bolwell4, GG Laport5, TB Shore6, S Durrant7, J Szer8, M-G Chen9, R Lizambri9 and EK Waller10

This prospective, randomized, double-blind, placebo-controlled study evaluated the efficacy of palifermin to reduce the incidence of severe (grade 3--4) acute GVHD after myeloablation and allo-SCT. Adults who received allo-SCT for hematologic malignancies received placebo or palifermin 60 mg/kg daily on three consecutive days before conditioning and a single dose of 180 mg/kg after conditioning, but often 1 or 2 days before allo-SCT. Subjects received MTX (plus CYA or tacrolimus) on days 1, 3, 6 and 11. Acute GVHD was evaluated once weekly and oral mucositis was evaluated daily. Subjects were randomly assigned to placebo (n ¼ 78) or palifermin (n ¼ 77). Conditioning included TBI in approximately half of the subjects (48% placebo, 51% palifermin). The primary efficacy end point, subject incidence of grade 3--4 acute GVHD, was similar between treatment groups (17% placebo, 16% palifermin). Grade 3--4 oral mucositis (73% placebo, 81% palifermin) and other secondary efficacy end points were similar between treatment groups. The most commonly reported treatment-related adverse events were skin/s.c. events such as rash, pruritus, and erythema. This exploratory study of acute GVHD after myeloablation and allo- SCT did not provide evidence of a treatment effect with this dosing regimen of palifermin.

Bone Marrow Transplantation (2012) 47, 1350--1355; doi:10.1038/bmt.2011.261; published online 13 February 2012 Keywords: hematopoietic cell transplantation; GVHD; palifermin; MTX

INTRODUCTION (60 mg/kg daily) prior to the start of myeloablative and one Allo-SCT is the treatment of choice for a variety of hematological collapsed dose (180 mg/kg) at the end of the myeloablative malignancies.1 Although the use of reduced intensity conditioning regimen, often prior to allo-SCT. regimens has increased, myeloablative regimens continue to be a The primary aim of this randomized, double-blind, placebo- standard of care for young patients. About 35--50% of allo-SCT controlled study was to explore the potential of palifermin to recipients develop grade 2--4 acute GVHD, with a higher incidence reduce the incidence of severe (grade 3--4) acute GVHD in after unrelated donor transplants and mismatched donor trans- patients undergoing allo-SCT from either a related donor or an plants compared with HLA-identical sibling donor transplants.2 HLA-matched, unrelated donor. Oral mucositis with ulceration (World Health Organization (WHO) grade 2--4) occurs in approximately 75% of patients receiving myeloablative regimens.3 Prophylaxis with a MTX- and calcineurin MATERIALS AND METHODS inhibitor-based regimen is effective in reducing GVHD,4 but Subjects worsens mucositis in many patients.5 Adults of at least 18 years of age were eligible for the study if they had a Keratinocyte growth factor appears to modulate the endogen- hematologic malignancy (including myelodysplastic syndromes) and were ous response to epithelial injury. In animal studies, recombinant scheduled to receive allogeneic marrow or PBPC transplantation after one human Keratinocyte growth factor (palifermin) reduced GVHD and out of six myeloablative conditioning regimens: CY plus TBI±etoposide, inhibited rejection of allo-SCT.6--8 Palifermin was shown to TBI plus etoposide, melphalan plus TBI 41100 cGy, BU plus CY, BU plus decrease the incidence and duration of severe oral mucositis in melphalan (fully ablative doses) or ﬂudarabine plus melphalan (fully patients who received TBI-based conditioning therapy and ablative doses). Subjects were required to have Karnofsky Performance autologous SCT.9 The approved dose of palifermin for reduction Status X70% and to have either a related donor or an HLA-matched of severe oral mucositis is 60 mg/kg daily by i.v. bolus injections for unrelated donor identical at 6/6 HLA-A, -B and -DRB1 loci. For unrelated 3 days before and 3 days after an myeloablative regimen and SCT. donors, molecular typing of class I and class II was mandatory. Prior studies evaluated the efﬁcacy and safety of palifermin to Key exclusion criteria included the following: other malignancies, prior reduce GVHD either with this dose, or with a collapsed 180 mg/kg BM or PBPC transplantation, previous use of palifermin, active infection or dose 1 day before the start of the myeloablative regimen and oral mucositis, congestive heart failure (NYHA class III or IV), use of a T-cell 60 mg/kg daily for at least 3 days after the end of the depleted graft for GVHD prophylaxis, inadequate renal, liver or pulmonary myeloablative regimen.10 --13 Because MTX is dosed 1, 3, 6 and function, pregnancy or breast feeding, refusal to use adequate contra- 11 days after allo-SCT and the interaction between palifermin and ception during the study, or participation in another investigational device MTX was not known, this study used a regimen of three doses or drug trial in the previous 30 days.

1Hematology and Stem Cell Transplant, Vanderbilt-Ingram Cancer Center, Nashville, TN, USA; 2Division of Hematology/Oncology, Indiana University, Indianapolis, IN, USA; 3Division of Cellular Therapy, Duke University Medical Center, Durham, NC, USA; 4Hematologic Oncology and Blood Disorders, Cleveland Clinic, Cleveland, OH, USA; 5Blood and Marrow Transplantation, Stanford University Medical Center, Stanford, CA, USA; 6Division of Hematology/Oncology, New York Presbyterian Hospital/Weill Cornell Medical College, New York, NY, USA; 7Division of Oncology, Royal Brisbane & Women’s Hospital, Brisbane, Queensland, Australia; 8Clinical Haematology, Royal Melbourne Hospital, Melbourne, Victoria, Australia; 9Amgen Inc., Thousand Oaks, CA, USA and 10Hematology and Medical Oncology, The Winship Cancer Institute of Emory University, Atlanta, GA, USA. Correspondence: Dr MH Jagasia, Hematology and Stem Cell Transplant, Vanderbilt University Medical Center, Vanderbilt-Ingram Cancer Center, Nashville, TN, USA. E-mail: [email protected] Received 20 September 2011; revised 7 December 2011; accepted 22 December 2011; published online 13 February 2012 Palifermin and acute GVHD in SCT MH Jagasia et al 1351 Study design engraftment, these data were not captured for purposes of the study. Thus, This prospective, stratified, randomized, double-blind, placebo-controlled, conventional criteria of neutrophil and platelet engraftment could not be multicenter phase II study was conducted between December 2005 and followed during the conduct of the study. Previous studies in both November 2008 at 16 sites in the United States and 4 sites in Australia. The autologous and allogeneic transplant did not report any detrimental study followed the principles of the Declaration of Helsinki. Written impact of palifermin on neutrophil and platelet engraftment.9--12 informed consent was obtained for each subject and an institutional review board approved the study at each site. Statistical analysis Subjects were randomly assigned in a 1:1 ratio to receive placebo or Efficacy analyses included all randomized subjects. Safety analyses palifermin by i.v. bolus injections. The randomization scheme was stratified included all randomized subjects who received at least one dose of study on TBI use (yes or no), matched donor type (related or unrelated) and medication. The study was designed to provide an estimate of the source of donor cells (marrow or PBPC). incidence of severe GVHD in an allogeneic transplant setting and was not Day 0 was defined as the day of marrow or PBPC transplantation. The powered to provide definitive proof of efficacy. The planned sample size recommended dosing schedule for conditioning regimens was from day was 200 subjects. Because of slow enrollment, a decision was made by the À11 to day À2. Palifermin 60 mg/kg daily (or matching placebo) was sponsor in consultation with a data-safety monitoring board to terminate administered on 3 consecutive days before the start of the conditioning the study at 155 subjects. This still allowed an acceptable level of precision regimen. A single dose of palifermin 180 mg/kg or placebo was for the estimate of the primary end point in this exploratory study and administered at least 24 h after the last dose of myeloablative allowed the results to be available up to 12 months earlier. Estimates of chemotherapy or radiation, at least 96 h after the last 60 mg/kg dose of treatment differences between palifermin and placebo were provided with study medication and at least 24 h before the first post-transplant dose of 95% confidence intervals (CIs) for the incidence of severe GVHD and other MTX. incidence end points. The Cochran--Mantel--Haenszel weight was used to Subjects received MTX for GVHD prophylaxis on days 1, 3 and 6 combine estimates across randomization strata. (planned), and on day 11 (if toxicity allowed) at doses of 15, 10, 10 and 10 mg/m2, respectively. A calcineurin inhibitor (cyclosporine or tacrolimus) was administered with MTX and dosed per institutional norms. Filgrastim RESULTS was not required but could be administered per the standard practice at Subject disposition each institution. Subjects were evaluated for efficacy once weekly (±3 days) for 8 weeks, Subject disposition is summarized in Figure 1. Of the 187 potential and then at days 70, 84 and 100 (±7 days). The severity of acute GVHD subjects screened, 155 subjects met the eligibility criteria and was evaluated with the modified Keystone criteria assessment scale,14 were randomly assigned to receive placebo (n ¼ 78) or palifermin based on physical exam and laboratory serum values. Biopsies of affected (n ¼ 77). Two of the subjects in the placebo group received a dose organs were obtained when possible. The degree of GVHD in individual of commercial palifermin in error; they were included in the organs was scored by at least two qualified assessors at the site who were placebo group for efficacy analyses and the palifermin group for blinded to study medication assignment. The investigator reviewed any safety analyses. A total of 151 subjects received at least one dose discrepant GVHD scores. Two experts (EKW and MJ) independently of study treatment (75 (96%) placebo, 76 (99%) palifermin), most reviewed GVHD scoring for each enrolled subject. completed study treatment (69 (88%) placebo, 71 (92%) Trained assessors at each site used the WHO criteria15 to assess oral palifermin) and most completed the study (63 (81%) placebo, 55 mucositis severity at screening and then once daily during hospitalization, (71%) palifermin). from the first day of the conditioning regimen until the subject was discharged or day 28, whichever was earlier. Thereafter, if the subject had Baseline demographics and treatment characteristics grade 3--4 oral mucositis, the daily oral mucositis assessments were to Baseline demographics of study participants and the donor source continue until oral mucositis returned to grade p2. As few assessors as and type were generally similar between treatment groups possible were assigned at each site to keep the assessments consistent. (Table 1). The use of study medication and TBI were also similar Adverse events and concomitant medications were recorded at each between treatment groups (Table 2). Approximately half of the assessment. For each adverse event, the investigator recorded details such subjects in each treatment group received TBI (35 (48%) placebo, as the severity, seriousness, relationship to study medication and 40 (51%) palifermin), with a mean±s.d. of 1269.9±149.5 cGy in resolution of the event. A blood sample for serum palifermin Ab assays the placebo group and 1236.1±222.3 cGy in the palifermin group. was obtained on day 28. Antibodies against palifermin were assayed by an Chemotherapy doses were similar between treatment groups electrochemiluminescence-based immunoassay; a positive test was (Table 2). The mean CY doses were similar between the two followed by a cell-based assay for neutralizing antibodies. Information treatment groups (100.3±28.7 vs 100.4±34.4 mg/kg), but more about the use of the total parenteral nutrition was collected up until subjects in the palifermin group received CY (57 (78%) placebo, 64 engraftment or day 28, whichever occurred first. (82%) palifermin), which was given with either TBI or BU (30 (41%) placebo, 33 (42%) palifermin). Compared with the placebo group, Efficacy and safety end points fewer subjects in the palifermin group received fludarabine (10 (14%) placebo, 5 (6%) palifermin). Melphalan and etoposide use The primary efficacy end point was the subject incidence of severe (grade was similar between treatment groups. 3--4), acute GVHD. Secondary efficacy end points included the incidence of Of the 151 subjects who received study medication, 11 (7%), 82 grade 2--4 acute GVHD, the incidence of MTX prophylaxis on day 11, the (55%), and 51 (33%) received the collapsed dose on days À2, À1 incidence of severe oral mucositis (grade 3--4), the duration of severe oral and day 0 (day of transplant), respectively. In all, 5 (3%) subjects mucositis (overall and among subjects with severe oral mucositis), the did not receive a transplant and 2 (1%) others did not receive the incidence of parenteral or transdermal opioid analgesic use and the collapsed dose. incidence of total parenteral nutrition. The incidence and duration of grade 2--4 oral mucositis were exploratory efficacy end points. Safety was assessed by the subject incidence of adverse events, Ab formation, Efficacy neutrophil engraftment (ANC X1.0 Â 109/L for 3 consecutive days or The between-group differences for GVHD outcomes are shown in X10.0 Â 109/L for 1 day) and platelet engraftment (X20 Â 109/L) by day Figure 2. The subject incidence of the primary efficacy end point, 100. Neutrophil and platelet counts were done on a weekly basis to grade 3--4 acute GVHD, was similar between the placebo and coincide with weekly GVHD assessment. Although participating centers palifermin groups (12 (17%) vs 12 (16%); difference (95% CI): 0.5% followed institutional criteria of checking counts more frequently prior to (À11.0%, 12.1%)). The subject incidence of the secondary efficacy

& 2012 Macmillan Publishers Limited Bone Marrow Transplantation (2012) 1350 --1355 Palifermin and acute GVHD in SCT MH Jagasia et al 1352 Screened 187

Randomized 155

Placebo Palifermin Primary analysis set 78 Primary analysis set 77 Safety analysis set* 73 Safety analysis set* 78

Treated 75 (96%) Treated 76 (99%) Completed treatment 69 (88%) Completed treatment 71 (92%) Discontinued study 15 (18%) Discontinued study 22 (29%) Death 5 (6%) Death 8 (10%) Protocol deviation 2 (3%) Consent withdrawn 6 (8%) Adverse event 2 (3%) Disease progression 4 (5%) Consent withdrawn 2 (3%) Administrative decision 2 (3%) Disease progression 2 (3%) Protocol deviation 1 (1%) Ineligibility determined 1 (1%) Adverse event 1 (1%) Other 1 (1%) Completed study 55 (71%) Completed study 63 (81%)

Figure 1. Subject Disposition. * Two subjects in the placebo group received palifermin in error; they were included in the placebo group for the efﬁcacy analysis and the palifermin group for safety analysis.

Table 1. Baseline demographics, patient characteristics, and donor Table 2. Treatments administered (safety analysis set) type and source (primary analysis set) Treatment Placebo Palifermin Characteristic Placebo Palifermin (n ¼ 73) (n ¼ 78) (n ¼ 78) (n ¼ 77) Study medication Sex, male, n (%) 49 (63) 40 (52) Received study medication, 73 (100) 78 (100) n (%)a Race/ethnicity, n (%) No. of doses of study medication, 4.0±0.3 3.8±0.5 White/Caucasian 67 (86) 65 (84) mean±s.d. Hispanic/Latino 4 (5) 5 (6) Black/African American 4 (5) 3 (4) Radiation Asian 2 (3) 3 (4) Received TBI, n (%) 35 (48) 40 (51) Native Hawaiian or other 1 (1) 0 (0) Total TBI dose (cGy), mean±s.d.b 1269.9±149.5 1236.1±222.3 Pacific Islander Other 0 (0) 1 (1) Chemotherapy Age (years), median (range) 44 (18--64) 42 (18--62) Received CY, n (%) 57 (78) 64 (82) CY dose (mg/kg), mean±s.d.b 100.3±28.7 100.4±34.4 Diagnosis, n (%) Received BU, n (%) 30 (41) 33 (42) Leukemia 62 (79) 55 (71) BU dose (mg/kg), mean±s.d.b 11.5±4.0 12.6±2.9 Myelodysplastic syndrome 9 (12) 12 (16) Received melphalan, n (%) 10 (14) 8 (10) Non-Hodgkin’s lymphoma 6 (8) 9 (12) Melphalan dose (mg/m2), 126.5±18.3 122.3±21.5 Multiple myeloma 1 (1) 0 (0) mean±s.d.b Hodgkin’s disease 0 (0) 1 (1) Received fludarabine 10 (14) 5 (6) Fludarabine dose (mg/kg), 3.1±1.5 3.5±1.8 Disease status, n (%) b mean±s.d. Complete remission 46 (59) 42 (55) Received etoposide, n (%) 5 (7) 6 (8) Partial remission 10 (13) 10 (13) Etoposide dose (mg/kg), 49.8±7.8 55.6±5.8 Stable disease 14 (18) 15 (19) mean±s.d.b Disease progression 8 (10) 10 (13) Karnofsky performance score, 91.2±8.1 90.5±7.9 aOf the 75 subjects in the placebo group for efficacy analysis, 2 subjects mean±s.d. received palifermin in error; they were included in the palifermin group for Prior chemotherapy, n (%) 76 (97) 70 (91) safety analysis. bMean (±s.d.) doses were calculated among the subjects Prior radiotherapy, n (%) 3 (4) 4 (5) who received that treatment. Donor type, n (%) HLA-identical, related 46 (59) 45 (58) HLA-matched, unrelated 32 (41) 32 (42)

Donor source, n (%) end point, grade 2--4 acute GVHD, was lower in the placebo group Marrow 17 (22) 18 (23) than in the palifermin group (29 (40%) vs 43 (58%); difference Peripheral blood 61 (78) 59 (77) (95% CI): À17.9% (À33.4%, À2.4%)). The proportion of subjects progenitor cells who received the ﬁnal, optional dose of MTX on day 11 was similar

Bone Marrow Transplantation (2012) 1350 --1355 & 2012 Macmillan Publishers Limited Palifermin and acute GVHD in SCT MH Jagasia et al 1353 Between-Group difference Placebo Palifermin GVHD Outcomes, n (%) (n = 78) (n = 77) (95% confidence interval)

Severe (Grade 3–4) Acute GVHD 12 (17) 12 (16)

Grade 2–4 acute GVHD 29 (40) 43 (58)

Methotrexate use at day 11 56 (72) 60 (78)

Mucositis outcomes, n (%)

Severe (Grade 3–4) oral mucositis 57 (73) 62 (81)

Grade 2–4 oral mucositis 69 (88) 69 (90)

Parenteral/Transdermal opioid use 50 (64) 48 (62)

Total parenteral nutrition 38 (49) 43 (56)

-40% -30% -20% -10% 0% 10% 20% Favors placebo Favors palifermin Figure 2. Between-group differences and 95% CIs for efﬁcacy outcomes (primary analysis set).

Table 3. Duration of oral mucositis (primary analysis set) Table 4. Adverse events that occurred in X20% of palifermin subjects (safety analysis set) Mean±s.d. duration, days Adverse event Placebo (n ¼ 73) Palifermin (n ¼ 78)

Any adverse event 72 (99) 77 (99) Placebo Palifermin Difference Nausea 70 (96) 73 (94) (n ¼ 78) (n ¼ 77) (95% CI)a Diarrhea 65 (89) 64 (82) Severe (Grade 3--4) oral mucositis Vomiting 53 (73) 54 (69) All subjects 8.1±6.8 7.8±6.7 0.5 (À1.5, 2.5) Fatigue 47 (64) 54 (69) Subjects with severe 10.2±6.3 9.4±6.3 0.8 (À1.4, 3.0) Headache 38 (52) 46 (59) oral mucositis Pyrexia 37 (51) 38 (49) Grade 2--4 oral mucositis 12.6±7.7 13.5±8.5 À0.6 (À3.2, 1.9) Abdominal pain 28 (38) 35 (45) Rash 27 (37) 35 (45) aAdjusting for the effects of the stratiﬁcation factors using the Cochran -- Hypertension 25 (34) 29 (37) Mantel --Haenszel weights. Insomnia 29 (40) 28 (36) Peripheral edema 30 (41) 26 (33) Anorexia 26 (36) 26 (33) Cough 22 (30) 26 (33) between the placebo and palifermin groups (56 (72%) vs 60 (78%); Febrile neutropenia 24 (33) 25 (32) difference (95% CI): À6.5% (À19.4%, 6.4%)). GVHD 23 (32) 22 (28) The incidences were similar between the placebo and Tachycardia 17 (23) 21 (27) palifermin groups (Figure 2) for severe (grade 3--4) oral mucositis Pruritus 20 (27) 20 (26) (57 (73%) vs 62 (81%)), grade 2--4 oral mucositis (69 (88%) vs 69 Hypomagnesaemia 25 (34) 18 (23) (90%)), parenteral or transdermal opioid use (50 (64%) vs 48 (62%)) Anxiety 20 (27) 18 (23) and total parenteral nutrition (38 (49%) vs 43 (56%)). The duration Back pain 20 (27) 18 (23) of oral mucositis was also similar between treatment groups Dyspnea 16 (22) 18 (23) Constipation 29 (40) 16 (21) (Table 3), including both the duration of severe oral mucositis Asthenia 18 (25) 16 (21) (overall and among those subjects with severe oral mucositis) and Dyspepsia 12 (16) 16 (21) the duration of grade 2--4 oral mucositis. Hyperglycemia 8 (11) 16 (21)

Safety placebo, 8 (10%) palifermin), and 5 subjects within 100 days after Most subjects reported at least one adverse event (72 (99%) the study (2 (3%) placebo, 3 (3%) palifermin). The investigators did placebo, 77 (99%) palifermin). There were no clinically signiﬁcant not consider any of these deaths to be related to the study differences in the subject incidences of individual adverse events medication. between the treatment groups (Table 4). Documented neutrophil count (ANC X10.0 Â 109/L) occurred in Study drug-related adverse events were reported for 23 (32%) 22 (30%) placebo subjects and 30 (38%) palifermin subjects; the placebo subjects and 31 (40%) palifermin subjects. The most proportion of subjects who achieved a lower ANC target on 3 commonly reported adverse events that the investigator con- consecutive days could not be analyzed because ANC values were sidered related to the study treatment are shown in Table 5. not collected daily. Platelet engraftment (X20 Â 109/L) occurred Adverse events in three subjects (2 (3%) placebo, 1 (1%) in 69 (95%) placebo subjects and 74 (95%) palifermin subjects. palifermin) and disease progression in six subjects (2 (3%) placebo, One (1%) palifermin subject had a seropositive result for anti- 4 (5%) palifermin) were reported as reasons for study discontinua- palifermin antibodies during screening, prior to palifermin tion. Deaths were reported for 13 subjects during the study (5 (6%) administration; this subject was retested with negative results

& 2012 Macmillan Publishers Limited Bone Marrow Transplantation (2012) 1350 --1355 Palifermin and acute GVHD in SCT MH Jagasia et al 1354 use or the total parenteral nutrition. The altered, collapsed dose X Table 5. Study drug-related adverse events in 5% of palifermin and timing for the post-preparative regimen could have been subjects (safety analysis set) contributing factors. The approved dosing schedule for palifermin includes three daily 60 mg/kg doses before conditioning and three Adverse event Placebo Palifermin daily 60 mg/kg doses after transplant. In the current study, subjects (n ¼ 73) (n ¼ 78) received the recommended doses of palifermin prior to condition- Any study drug-related adverse 23 (32) 31 (40) ing, but a collapsed dosing regimen with a single dose of palifermin event 180 mg/kg was used after conditioning. Additionally, the post- preparative dose of palifermin was often administered prior to Gastrointestinal disorders 2 (3) 18 (23) transplant on day À1orÀ2 to try to augment mucositis protection Tongue-coated 1 (1) 6 (8) prior to initiating MTX on day 1 for GVHD prophylaxis. Use of the Tongue disorder 0 (0) 5 (6) collapsed dose of palifermin as the post-dose, combined with the difference in timing compared with the currently approved dose Skin and subcutaneous tissue 15 (21) 19 (24) and schedule of palifermin, may be an important limitation in the disorders extrapolation of these results to the expected effects of the Rash 6 (8) 11 (14) Pruritus 4 (5) 10 (13) approved dosing schedule for palifermin in this setting. Erythema 3 (4) 4 (5) Grade 2--4 acute GVHD was reported in a higher proportion of patients receiving palifermin compared with placebo. The exact reason for this is not clear. Palifermin-induced skin rashes tend to on day 28. Seronegative results were obtained from 65 (89%) occur early after administration of the drug and are transient. placebo subjects and 66 (85%) palifermin subjects. The remaining Although skin rashes post-transplant could be misdiagnosed as subjects were not tested for anti-palifermin antibodies. skin GVHD, the timing of the rash immediately after the conditioning regimen and prior to engraftment, the recommen- dation to pursue a skin biopsy on the study prior to the diagnosis DISCUSSION of skin GVHD and blinded assessment by two physicians of weekly Preclinical studies suggested that palifermin decreased GVHD assessment of acute GVHD data make this an unlikely cause for incidence and lethality without decreasing graft-vs-tumor effect.6--8 the higher incidence of GVHD in the palifermin group. Long-term In patients with hematologic malignancies receiving myelotoxic follow-up of the present study is ongoing at 6 months and at 1, 2, therapy requiring hematopoietic stem cell support, palifermin 3, 4 and 5 years to evaluate chronic GVHD, disease status and treatment was associated with significant decreases in the secondary malignancies. incidence and duration of WHO grade 4 and grade 3--4 mucositis, Neutrophil engraftment rates were similar between treatment along with improvement in patient-reported outcomes, use of groups. Neutrophil engraftment was defined as ANC of opioid analgesics and the total parenteral nutrition.9,16 However, X1.0 Â 109/L for 3 consecutive days or ANC of X10.0 Â 109/L most of those studies evaluated the effects of palifermin among for 1 day, whichever was sooner. The ANC target of X1.0 Â 109/L patients undergoing autologous SCT after an intensive, fractio- for 3 consecutive days could not be used in this study because at nated TBI-based conditioning regimen. no time were ANC results scheduled to be collected on 3 In allo-SCT, mucositis is related not only to the preparative consecutive days. Thus, the lower-than-expected neutrophil regimen but also is compounded by the MTX that is used for engraftment in the current study was attributable to the higher GVHD prophylaxis.5 The exact post-preparative dosing regimen for single-day engraftment criteria. palifermin in allo-SCT is not as clearly defined by clinical data as is The original sample size of the study was 200 subjects, but the case for autologous-SCT. Several investigators have studied because of slow enrollment, the study was stopped after only 155 palifermin in allo-SCT---with the intent of decreasing GVHD and subjects were randomized. The decision to stop the study was sparing graft-vs-tumor effect---using a post-preparative dose of made by the sponsor in consultation with a data-safety 60 mg/kg daily for at least 3 days,10 --13 which is the same post- monitoring board. Stopping the study early enabled the release preparative dose that was used in studies of autologous SCT.9 of the results sooner, while still allowing the objectives of this Using this dosing regimen, a retrospective study concluded that exploratory study to be fulfilled in providing an estimate of severe palifermin decreased the incidence of acute GVHD compared with GVHD. As this was an estimation study, reducing the sample size historical controls and it decreased mucositis, opioid requirements would only affect the precision for the estimate of the primary end and the total parenteral nutrition.10 However, another retro- point. If the results had demonstrated superior efficacy to support spective cohort study showed no effect of palifermin on GVHD in the need for additional trials, stopping this study early would have allo-SCT.11 In a prospective, randomized trial, Blazar et al.12 enabled those trials to begin sooner, perhaps by as much as investigated various dosing regimens of palifermin in allo-SCT and 1 year. Unfortunately, the study results showed a treatment concluded there were no significant differences in acute GVHD difference of only 1% for the primary efficacy end point of grade incidence, acute GVHD severity, survival or day 100 relapse rates 3--4 acute GVHD (17% placebo vs 16% palifermin), which was well between the palifermin and placebo groups. Palifermin was below any efficacy threshold to pursue further studies. associated with reduced incidence and severity of mucositis Clinical trials are often planned based on important findings in (measured three times weekly), but only in patients conditioned animal models of transplantation. However, human studies often with fractionated TBI-based regimens.12 Levine et al.13 reported on do not confirm these findings, making extrapolation from animal the long-term outcomes of this cohort and found no difference models complex and challenging. Palifermin showed clinical between the palifermin and placebo groups in the incidence of benefit as a mucoprotective agent in autologous hematopoietic chronic GVHD or relapse risk. As shown in some of the prior cell transplantation,9 but the hope of palifermin as an agent to trials,11,12 there was no positive impact of palifermin on either prevent GVHD without losing graft-vs-tumor effect in allo-SCT has grade 2--4 or grade 3--4 acute GVHD incidence or severity. not been realized in this study. Based on the findings of this Mucositis was monitored stringently in this study, with once- exploratory study, no additional, larger study is planned to daily assessments by the trained investigators compared with evaluate this dose of palifermin (three daily doses of 60 mg/kg assessments three times weekly in the prior randomized study.12 before allo-SCT and a single, collapsed dose of 180 mg/kg after Although the previous studies reported a decrease in mucositis,10 --12 conditioning therapy but before allo-SCT) to reduce GVHD or oral this study did not show any benefit in mucositis incidence, opioid mucositis in allo-SCT.

Bone Marrow Transplantation (2012) 1350 --1355 & 2012 Macmillan Publishers Limited Palifermin and acute GVHD in SCT MH Jagasia et al 1355 CONFLICT OF INTEREST 6 Krijanovski OI, Hill GR, Cooke KR, Teshima T, Crawford JM, Brinson YS et al. MGC and RL are employees of Amgen Inc. and hold stock in Amgen Inc. MHJ and Keratinocyte growth factor separates graft-versus-leukemia effects from EKW were compensated by Amgen Inc. for consultation during study planning and graft-versus-host disease. Blood 1999; 94: 825 --831. for providing central review of GVHD scores. 7 Panoskaltsis-Mortari A, Lacey DL, Vallera DA, Blazar BR. Keratinocyte growth factor administered before conditioning ameliorates graft-versus-host disease after allogeneic bone marrow transplantation in mice. Blood 1998; 92: 3960 --3967. 8 Panoskaltsis-Mortari A, Taylor PA, Rubin JS, Uren A, Welniak LA, Murphy WJ et al. ACKNOWLEDGEMENTS Keratinocyte growth factor facilitates alloengraftment and ameliorates graft- versus-host disease in mice by a mechanism independent of repair of This study was supported by research funding from Amgen Inc. Jonathan Latham of conditioning-induced tissue injury. Blood 2000; 96: 4350 --4356. PharmaScribe, LLC received funding from Amgen Inc. to provide assistance with the 9 Spielberger R, Stiff P, Bensinger W, Gentile T, Weisdorf D, Kewalramani T et al. preparation of the manuscript. Xuesong Guan of Amgen Inc. provided assistance with Palifermin for oral mucositis after intensive therapy for hematologic cancers. statistical analyses. The NCT00964899 study investigators were as follows (listed N Engl J Med 2004; 351: 2590 --2598. alphabetically): Rafat Abonour, Gorgun Akpek, Carlos Bachier, Brian Bolwell, Simon 10 Nasilowska-Adamska B, Rzepecki P, Manko J, Czyz A, Markiewicz M, Federowicz I Durrant, Steven Goldstein, Richard Herrmann, David Hurd, Madan Jagasia, Ginna et al. The inﬂuence of palifermin (Kepivance) on oral mucositis and acute graft Laport, Ian Lewis, Gwynn Long, Kenneth F Mangan, Philip McCarthy, John M McCarty, versus host disease in patients with hematological diseases undergoing Shin Mineishi, Han Myint, Scott D Rowley, Tsiporah Shore, Ricardo Spielberger, Roger hematopoietic stem cell transplant. Bone Marrow Transplant 2007; 40: 983 --988. Strair, Jeff Szer and Edmund K Waller. 11 Langner S, Staber P, Schub N, Gramatzki M, Grothe W, Behre G et al. Palifermin reduces incidence and severity of oral mucositis in allogeneic stem-cell transplant recipients. Bone Marrow Transplant 2008; 42: 275 --279. 12 Blazar BR, Weisdorf DJ, Defor T, Goldman A, Braun T, Silver S et al. Phase 1/2 REFERENCES randomized, placebo-control trial of palifermin to prevent graft-versus-host 1 Jenq RR, van den Brink MR. Allogeneic haematopoietic stem cell transplantation: disease (GVHD) after allogeneic hematopoietic stem cell transplantation (HSCT). individualized stem cell and immune therapy of cancer. Nat Rev Cancer 2010; 10: Blood 2006; 108: 3216 --3222. 213 --221. 13 Levine JE, Blazar BR, DeFor T, Ferrara JL, Weisdorf DJ. Long-term follow-up of a 2 Jacobsohn DA, Vogelsang GB. Acute graft versus host disease. Orphanet J Rare Dis phase I/II randomized, placebo-controlled trial of palifermin to prevent graft- 2007; 2: 35. versus-host disease (GVHD) after related donor allogeneic hematopoietic cell 3 Vera-Llonch M, Oster G, Ford CM, Lu J, Sonis S. Oral mucositis and outcomes of transplantation (HCT). Biol Blood Marrow Transplant 2008; 14: 1017 --1021. allogeneic hematopoietic stem-cell transplantation in patients with hematologic 14 Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J et al. 1994 malignancies. Support Care Cancer 2007; 15: 491 --496. Consensus Conference on Acute GVHD Grading. Bone Marrow Transplant 1995; 4 Ram R, Gafter-Gvili A, Yeshurun M, Paul M, Raanani P, Shpilberg O. Prophylaxis 15: 825 --828. regimens for GVHD: systematic review and meta-analysis. Bone Marrow Transplant 15 World Health Organization. A Handbook for Reporting Results of Cancer Treatment. 2009; 43: 643 --653. World Health Organization: Geneva, 1979. 5 Cutler C, Li S, Kim HT, Laglenne P, Szeto KC, Hoffmeister L et al. Mucositis after 16 Stiff PJ, Emmanouilides C, Bensinger WI, Gentile T, Blazar B, Shea TC et al. allogeneic hematopoietic stem cell transplantation: a cohort study of methotrex- Palifermin reduces patient-reported mouth and throat soreness and improves ate- and non-methotrexate-containing graft-versus-host disease prophylaxis patient functioning in the hematopoietic stem-cell transplantation setting. J Clin regimens. Biol Blood Marrow Transplant 2005; 11: 383 --388. Oncol 2006; 24: 5186 --5193.

& 2012 Macmillan Publishers Limited Bone Marrow Transplantation (2012) 1350 --1355