P.M. and R.B. share senior authorship.
Phase I trial of sargramostim in pediatric Crohn's disease
Article first published online: 5 JAN 2010
Copyright © 2010 Crohn's & Colitis Foundation of America, Inc.
Inflammatory Bowel Diseases
Volume 16, Issue 7, pages 1203–1208, July 2010
How to Cite
Kelsen, J. R., Rosh, J., Heyman, M., Winter, H. S., Ferry, G., Cohen, S., Mamula, P. and Baldassano, R. N. (2010), Phase I trial of sargramostim in pediatric Crohn's disease. Inflamm Bowel Dis, 16: 1203–1208. doi: 10.1002/ibd.21204
- Issue published online: 11 JUN 2010
- Article first published online: 5 JAN 2010
- Manuscript Accepted: 30 SEP 2009
- Manuscript Received: 8 SEP 2009
- Crohn's disease;
- open-label study
Improving granulocyte function may represent an effective therapy for Crohn's disease (CD). We performed a Phase I-2 trial of sargramostim (SRG) in children with CD.
This was multicenter, open-label study in 6–16-year-old patients with moderate to severely active CD. Patients received either 4 or 6 μg/kg SRG subcutaneously daily for 8 weeks, with and without concomitant corticosteroids (CS). The primary endpoint was identification of a safe and tolerable dose in children. The secondary endpoint was establishment of the pharmacokinetics (PK). Efficacy, a tertiary endpoint, was measured by the Pediatric CD Activity Index (PCDAI). Response was defined as a decrease from baseline of ≥12.5 points and remission as absolute PCDAI of ≤10.
In all, 22 patients were enrolled: 12 and 10 received 4 and 6 mg/kg, respectively; 19 completed the course. Both doses were found to be safe and well tolerated. Mild injection-site reactions occurred in 90% of patients. Three patients required dose reductions due to elevated absolute neutrophil counts. Following 4 μg/kg the mean area under the curve (AUC) was 2.64 and 2.80 ngh/mL for the 6–11- and 12–16-year-old groups, respectively. The mean half-life (t1/2) was 1.22 and 1.59 hours, respectively. Following 6 μg/kg, the mean AUC was 5.01 ngh/mL for the 12–16-year-old group, a 1.8-fold increase. A total of 16/18 patients (88%) achieved remission or response.
Sargramostim at both 4 and 6 mg/kg was well tolerated. PK analysis suggested dose proportionality unaffected by CS exposure. Remission and response data are encouraging, but further trials are needed to assess efficacy. (Inflamm Bowel Dis 2010)
Approximately 25% of new cases of inflammatory bowel disease (IBD) occur in children and adolescents. The management of pediatric Crohn's disease (CD) presents unique challenges, as this population is at risk for multiple complications including malnutrition, bone disease, short stature, and psychosocial issues that can be devastating to the future health of a developing child.1
While researchers have made significant advances in our knowledge of the disease, the pathogenesis of CD is still poorly defined. It is clear, however, that the disease is multifactorial. A leading hypothesis for the development of IBD in susceptible individuals is an aberrant immune response to enteric flora. One model to explain the etiology of this excessive inflammatory response is a compensatory downstream reaction. The primary defect arises from components of the impaired innate immune system, including neutrophils, monocytes, and macrophages, which are critical for maintaining normal mucosal barrier function. Defects in the function of these cells may lead to lack of clearance of antigenic stimuli, chronic activation of the immune system, disruption of normal regulatory controls, and a subsequent inflammatory process that is largely mediated by T lymphocytes.2 The discovery of 2 of the genetic associations with CD provides significant evidence that CD in some individuals may be the result of a defective innate immune system. Mutations in NOD2/CARD15 and ATG16L1, innate immune-related genes associated with protection against bacteria, lead to a selective decrease in alpha-defensin production by Paneth cells3 and altered bacterial cell killing by autophages,4 respectively. Therefore, treatment strategies aimed at augmenting the innate immune response rather than suppressing it may be successful.5
Granulocyte-macrophage colony-stimulating factor (GM-CSF, Sargramostim, SRG) is a pleiotropic cytokine, primarily known as a hematopoietic growth factor that controls production of granulocytes, and enhances neutrophil effector function.6, 7 GM-CSF may help overcome defects in neutrophil and macrophage function related to CD due to its broad range of effects on hematopoietic cells.8–10 GM-CSF is expressed by CD4+ T cells and Paneth cells in the intestinal epithelium and myeloid cells; furthermore, intestinal epithelium cells throughout the gastrointestinal tract express GM-CSF receptors.4, 11 Sainathan et al12 demonstrated that GM-CSF-treated mice had decreased expression of tumor necrosis factor alpha (TNF-α), interleukin (IL) 1-β, and IL1-α. This study also demonstrated that these mice had an increase in type 1 interferons in the ileum through dendritic cells, which contributes both to the innate and adaptive immunity. Therefore, it is plausible that GM-CSF may help support the intestinal innate immune system and may be an effective agent in CD.5, 13
While several studies in adults with CD report a positive outcome following treatment by activating the immune system through the use of SRG,4, 5, 14 no studies have been reported in the pediatric population. The aim of our study was to identify the appropriate doses and to characterize the safety profile of SRG treatment with and without concomitant corticosteroid (CS) therapy in pediatric patients with active CD. The secondary objectives were to characterize the pharmacokinetic and pharmacodynamic properties of sargramostim treatment with concomitant CSs. Efficacy was a tertiary endpoint with the goal to obtain preliminary information in pediatric patients with CD.
MATERIALS AND METHODS
This was a multicenter, open-label, dose escalation study of sargramostim (SRG) treatment in pediatric patients with moderate to severely active CD (PCDAI ≥30 points) who underwent induction using SRG with or without CSs. Patients were grouped by age (6–11 and 12–16 years). Two doses of SRG, 4 μg/kg and 6 μg/kg, administered subcutaneously (SC) once daily for 8 weeks (56 doses), were studied.
The dose of 6 μg/kg SRG SC utilized in this study was based on the dose selected for the Phase 2 and 3 studies of sargramostim in an adult CD population.4 In those studies 6 μg/kg was selected based on dose-dependent effects observed on neutrophil and monocyte counts and the higher incidence of adverse events (AEs), especially bone pain, observed in the 8 μg/kg dose cohort. As sargramostim had not been previously administered to pediatric patients with CD, the first cohort of patients received 4 μg/kg. Once this level was shown to be tolerated (minimal bone pain), a second group of patients was treated at 6 μg/kg.
In the steroid cohort, the first dose of SRG was within 7 days of initiation of CS treatment (corticosteroid dose was weight adjusted by protocol to maximum dose of 60 mg/day prednisone or equivalent either intravenously or orally). When a clinical response was achieved as defined as a decrease from baseline PCDAI (Pediatric Crohn's Disease Activity Index) score of 12.5 points, the CS dose was tapered weekly over a 4-week period, until the medication was discontinued.
Eligible patients were between the ages of 6–16 years with a confirmed diagnosis of CD at time of study entry based on standard radiologic, endoscopic, or histologic criteria.2 The patients were required to have moderate/severely active disease defined by PCDAI scores ≥30 points.15, 16 Patients were excluded if they were pregnant, had positive stool cultures for infection, an ostomy, or any immediate need for gastrointestinal (GI) surgery or had GI surgery 3 months prior to receiving the first dose of SRG. Patients were also excluded if they were on concurrent therapy with any of the following drugs within 4 weeks prior to receiving the first dose of the study drug: immunomodulators, cyclophosphamide, mycophenolate mofetil, tacrolimus, cyclosporine, thalidomide, or anti-TNF therapy within 8 weeks prior to first dose of SRG. Any prior use of natalizumab was an exclusion criteria as was any investigational drug within 4 weeks of the study drug, including prior use of recombinant human GM-CSF (sargramostim or molgramostim) or granulocyte colony-stimulating factor (G-CSF; filgrastim or pegfilgrastim). Furthermore, current use of nutritional therapy that provided >50% of daily caloric intake or current infection excluded patients from the study as well. Patients with severe anemia (hgb <8 g/d:) or significant biochemical profile abnormalities were excluded.
Patients were withdrawn from the study at any time after enrollment if they experienced a significant increase in CD symptoms as determined by the investigator or an increase in PCDAI score of ≥20 points from screening or rescreening/baseline. Additional reasons for study discontinuation included an AE or toxicity requiring discontinuation of treatment, or if a patient missed >2 doses of sargramostim in 1 week unless secondary to absolute neutrophil count (ANC) ≥30,000 cells/μL or persistent injection site reaction or bone pain. Patients were withdrawn if they required more than 2 dose reductions (see below) of the treatment drug.
Dose interruptions were performed as follows: If the severity and/or duration of the ISR (injection site reaction) or bone pain persisted for longer than 1 week despite following protocol, the dose was reduced. For patients with ANC ≥30,000 cells/μL, SRG was interrupted until ANC was <15,000 cells/μL. Blood samples were obtained every 2 days to monitor ANC counts. SRG dosing was reduced as follows:
For patients receiving 4 μg/kg, SRG was restarted at 2 μg/kg and remained at this level for the remainder of the treatment period. No further dose reductions were allowed. If the patient required an additional dose reduction, the patient was withdrawn from the study.
For patients receiving 6 μg/kg, SRG was restarted at 4 μg/kg and remained at this dose for the remainder of the treatment period. If an additional dose reduction was required, the dose was reduced to 2 μg/kg. No further dose reduction was allowed. If the patient required an additional dose reduction, the patient was withdrawn from the study.
The study was conducted at 6 study centers in the United States. All laboratory determinations were performed by a local laboratory at each study center.
Upon meeting entry criteria patients received the first dose of SRG, which was given daily for 8 weeks. The first day of SRG dosing was Week 0 (Day 1) of the treatment phase. Patients were evaluated at the study site at Weeks 1 (Day 8), 2 (Day 15), 4 (Day 29), and 8 (Day 57, end-of-treatment [EOT]). Patients were called at weeks 3, 5, 6, and 7 to assess safety and disease status. Patients returned for an end-of-study (EOS) evaluation 30 days following their last dose of SRG.
The study drug was manufactured and supplied by Bayer HealthCare Pharmaceuticals (Seattle, WA). An informed consent/assent form explaining the procedures of the study including the potential hazards was reviewed and approved by the Institutional Review Boards (IRBs) or Institutional Ethics Committees prior to its use. The study was conducted in accordance with Good Clinical Practice (GCP) as required by 21 Code of Federal Regulations (CFR) Parts 50, 56, and 312, and the standard operating procedures for clinical investigation. This study was registered under a ClinicalTrials.gov Identifier.
For 1 year following the study, patients were followed by phone once a month and had a clinic or office visit once every 3 months. These visits included determinations of disease activity using the PCDAI and PGA (Physician Global Assessment).
The safety of SRG was assessed by laboratory studies (hematology and chemistry profiles; anti-SRG antibody formation), premature discontinuations from the study due to AEs or toxicities; and AEs and serious adverse events (SAEs). Dosing was interrupted for Grade 3 or 4 AEs judged by the investigator to be related to study drug and restarted when the AE resolved to Grade 1 or less. The NCI Common Toxicity Criteria for Adverse Events (CTCAE) v. 3.0 is a descriptive terminology which can be utilized for AE reporting. A 5-point grading (severity) scale is provided for each AE term. The objective of the system is to improve accuracy, precision, and completeness and to standardize reporting.17
Pharmacokinetic and Pharmacodynamic Variables
Primary pharmacokinetic variables evaluated were maximum concentration (Cmax), area under the curve (AUC), and clearance (CL/f). Secondary pharmacokinetic variables include AUC (0 − tlast), tmax, t1/2.
Primary pharmacodynamic parameters included leukocyte and differential cell counts. Serum concentration-time profiles of individual patients were analyzed using a compartment model independent (CMID) method employing a validated computer program WinNonlin Professional v. 5.0 (Pharsight). Predose pharmacokinetic blood sampling was obtained during Weeks 0, 1, and 2, with complete pharmacokinetic studies performed at Week 2 (Day 15).
CD activity was assessed using the PCDAI and PGA. Health-related quality of life (QOL) was assessed in patients ≥10 years of age using the IMPACT-III questionnaire. Response was defined as decrease from baseline PCDAI score of ≥12.5 points, and remission was defined as PCDAI score <10 points.
Twenty-two patients were enrolled and were divided into 2 cohorts based on dose. One cohort was treated with 4 μg/kg and included 12 patients. The other cohort included 10 patients and was treated with the 6 μg/kg dose. In the 4 μg/kg cohort, 6 patients of the 12 were on CS. In the 6 μg/kg cohort, 4 of the 10 patients received CS.
Of the 22 patients enrolled, 19 completed SRG therapy. One patient prematurely discontinued the study secondary to disease exacerbation, per protocol. Two patients were removed secondary to noncompliance.
Characteristics of Study Population
All 22 patients treated in this study ranged in age from 8–16 years. The mean ages for the 4 treatment groups based on CS use and sargramostim dose were comparable. The patients' demographics are depicted in Table 1.
|CS Use||No CS Use|
|4 μg/kg||6 μg/kg||4 μg/kg||6 μg/kg|
|Age (range 8-16)||Mean ± SD||11.5 ± 2.07||12.5 ± 2.26||11.8 ± 2.79||12.3 ± 1.71|
|Baseline PCDAI||Missing (N)||2||3||1||0|
|Mean ± SD||46.9 ± 6.57||35.8 ± 6.29||40.5 ± 6.47||31.9 ± 3.15|
The majority of AEs were grade 1 or 2. Twenty-one of the 22 patients treated with sargramostim experienced at least 1 AE. Three patients had grade 3 AEs (all receiving 6 μg/kg); there were no grade 4 AEs. There was no difference in the incidence of grade 1 or 2 AE by dose level. The most frequently reported drug-related AEs were headache, back pain, vomiting, pyrexia, nausea, white blood cell (WBC) count increase, abdominal pain, and eosinophil count increase (Table 2).
|Adverse Events||CS-4 μg/kg (N=6)||CS-6 μg/kg (N=6)||CS+ 4 μg/kg (N=6)||CS+ 6 μg/kg (N=4)||Total Grade ≥ Grade 3 Toxicity|
|Back pain||1 Gr1||2 Gr1||3 Gr1||4 μg/kg: 0|
|1 Gr2||1 Gr2||1 Gr2||6 μg/kg: 0|
|Vomiting||2 Gr1||2 Gr1||1 Gr1||1 Gr2||4 μg/kg: 0|
|2 Gr2||1 Gr2||6 μg/kg: 0|
|Headache||2 Gr1||3 Gr1||2 Gr1||1 Gr1||4 μg/kg: 0|
|1 Gr2||1 Gr2||6 μg/kg:1:|
|Pyrexia||2 Gr1||2 Gr2||1 G1||1 Gr2||4 μg/kg: 0|
|1 Gr1||1 Gr2||6 μg/kg: 0|
|Nausea||2 Gr1||1 Gr1||1 Gr2||1 Gr2||4 μg/kg: 0|
|2 Gr2||6 μg/kg: 0|
|White blood cell count increased||1 Gr1||1 Gr1||1 Gr1||4 μg/kg: 0|
|1 Gr2||6 μg/kg: 0|
|Abdominal pain||1 Gr1||1 Gr3||1 Gr2||1 Gr3||4 μg/kg: 0|
|6 μg/kg: 2|
|Eosinophil Count increased||1 Gr1||1 Gr1||2 Gr1||4 μg/kg: 0|
|6 μg/kg: 0|
The majority of patients (20 of 22) reported grade 1 or 2 injection site reactions during SRG treatment. No deaths occurred during the study.
As noted above, there were no ≥grade 3 toxicities at the 4-μg/kg level. At 6 μg/kg 3 subjects were listed as grade 3 toxicity. This included 2 subjects (16.7%) in the CS− group and 1 subject (10.0%) in the CS+ group. The AEs were 2 subjects with abdominal pain, which investigators thought were possibly due to the drug, and 1 subject with anemia, which was thought most likely not due to the drug.
Except for the 2 episodes of abdominal pain reported by the subject who discontinued the study, all SAEs were resolved and all the subjects were able to continue treatment.
Nine subjects, 4 receiving 4 μg/kg (2 CS−, 2 CS+), and 5 receiving 6 μg/kg (4 CS−, 1 CS+) had their doses of study drug interrupted due to AEs. These AEs were grade 1 or grade 2 in intensity, and the majority (10 of 15) were unlikely related to the study drug as determined by the investigator.
Expected elevations in the hematology parameters ANC, eosinophils, and WBC were observed during SRG treatment. These increases resolved to around baseline levels within 30 days after the last dose of the study drug. Findings in the serum chemistry parameters were unremarkable and no clinically meaningful changes in the vital signs parameters were observed during the study.
PK data were available for 19 patients. Following 4 μg/kg, the mean AUC was 2.64 and 2.80 ngh/mL for the 6–11- and 12–16-year-old groups, respectively; Cmax was 1.04 and 0.78 ng/mL, tmax was 1.5 and 2.0 h, and mean t1/2 was 1.22 and 1.59 h, respectively. Following 6 μg/kg, the mean AUC was 5.01 ngh/mL for the 12–16-year-old group—a 1.8-fold increase. The percent extrapolated AUC was <20% in most patients. Concomitant steroid use did not appear to affect SRG exposure.
Based on the evaluation of predose PK samples collected on Days 8 and 15, little to no SRG accumulation was observed following 2 weeks of therapy. Given the half-life of SRG of ≈1.5 hours after SC administration, this observation was consistent with the expectation that SRG was completely cleared from circulation with the defined dosing interval of 24 hours (Fig. 1).
Efficacy assessments were performed including PCDAI and PGA evaluations. The mean baseline PCDAI scores for the 7 subjects in the CS− group and the 9 subjects in the CS+ group included in the analyses were 42.1 and 36.7 points, respectively. In all, 16/18 (88%) achieved either remission or response at EOT; 10/18 (55%) patients achieved remission. Rapid improvement in PCDAI was noted after 1 week, with a mean decrease in PCDAI of 30.2 ± 9.4 points in the non-CS group by EOT. This response was sustained for ≥4 weeks post-EOT. After treatment Week 1, the PCDAI response declined in the CS group, most likely due to rapid CS withdrawal attempted prior to EOT.
The number of subjects with improvement as assessed by the PGA was smaller (5 of 22) than the PCDAI; no patient had a worsening PGA (Fig. 2).
Pediatric patients with CD require careful management to ensure attainment of growth potential, proper development, and a normal childhood. Effective treatments may have complications that impede achieving these goals. Increasing evidence indicates that CD may be the result of a defect of the innate immune system and provides an opportunity for novel strategies in the treatment of IBD.4, 18, 19 Several studies have evaluated treatment of SRG in adult CD populations.4 In this pilot study we performed the first pediatric trial of SRG in CD.
We evaluated the safety of 2 doses of SRG in pediatric CD patients, 4 μg/kg and 6 μg/kg. The medication was well tolerated. The most common AEs were injection site reactions. Similar to the studies enrolling adult subjects,4 events in the children were transient, occurred in the initial 2 weeks of therapy, and decreased with subsequent treatment. Both doses studied were proven to be safe and tolerable.
The pharmacokinetic evaluation of SRG demonstrated that maximum serum SRG concentration was reached within 2 hours after SC administration. Higher exposures were achieved following administration of 6 μg/kg/d compared to 4 μg/kg/d for the corresponding age groups. While the dose effect did not appear to be linear, significance of dose proportionality could not be reached due to the limited number of subjects enrolled in the study.
Efficacy with SRG, although not a primary goal, was clearly demonstrated: 16/18 (83%) of patients completing therapy achieved either remission or response at EOT. While 1 group of patients was on concomitant steroids, the non-CS group demonstrated an excellent response, with 5/8 in remission by the EOT and 3/8 demonstrating response by EOT. A rapid improvement in PCDAI was noted after 1 week, with a mean decrease in PCDAI of 30.2 ± 9.4 points in the non-CS group by EOT. This response was sustained for ≥4 weeks post-EOT. This finding is similar to the findings in the adult trials.
The safety and efficacy of SRG in CD has been studied in 2 open-label Phase 1 studies and 1 Phase 2 double-blind randomized study in adults.1, 4 The pilot study of SRG included 15 adult patients with moderate to severe CD and demonstrated a high rate of clinical response and remission with limited toxicity.1 Based on these results a double-blind randomized, placebo-controlled Phase 2 trial was performed in 124 patients with moderate to severely active CD. The SRG group received 6 μg/kg for 57 days and concomitant CSs were not allowed. This study showed clinical efficacy as well. While the primary endpoint of a decrease in CDAI of 70 was not achieved at Day 57, multiple secondary aims were achieved. At all assessment points, patients receiving SRG demonstrated a greater decrease in CDAI of 100 points from baseline and had a decrease of 70 points on days other than 57. A significantly greater proportion of patients who received SRG achieved full remission as compared to placebo (40% versus 19%, P = 0.014) at the EOS treatment. Furthermore, SRG was associated with mucosal healing and improved QOL.4 The authors concluded that SRG was well tolerated and generally safe in patients with CD. One adult Phase III international trial evaluating Gm-CSF showed no clinical efficacy in comparison with placebo. The investigators speculated that the high placebo effect may be secondary to the different patient populations and different practice patterns. The study did demonstrate that the drug was safe.20
In conclusion, the recent advances in genetics, immunology, and microbiology have demonstrated that the CD may be the result of a defect in the innate immune system, resulting in a maladaptive response. Therefore, SRG is an excellent candidate drug for the treatment of pediatric CD in order to augment the innate immune system and epithelial barrier function. In this study, we demonstrated, similar to the adult trials, that SRG 4 and 6 μg/kg are safe and well-tolerated doses. While preliminary results are promising, further prospective studies designed to assess efficacy and optimal dosing of this therapeutic option in children and adults with CD are indicated.
- 2Inflammatory bowel disease in childhood. In: SatsangiJ, SutherlandL, ColombelJF, eds. Inflammatory Bowel Diseases, 1st ed. Oxford, UK: Churchill Livingstone; 2003: 699–715..
- 17Cancer Therapy Evaluation Program Revised March 23, 1998, Common Toxicity Criteria, Version 2.0. DCTD, NCI, NIH, DHHS March 1998.
- 19Sargramostim induces steroid-free remission in corticosteroid-dependent Crohn disease: results of n.o.v.e.l. 2, a Phase II multicenter study. Gastoenterology. 2007; 132S2., , , et al.
- 20Efficacy and safety of sargramostim in moderate to severe Crohn's disease: results of NOVEL 4, a phase III multicenter study. Gastroenterology. 2007; 132S2., , , et al.