Safety and Efficacy of Teduglutide After 52 Weeks of Treatment in Patients With Short Bowel Intestinal Failure

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2013;11: Safety and Efficacy of Teduglutide After 52 Weeks of Treatment in Patients With Short Bowel Intestinal Failure STEPHEN J. D. O KEEFE,* PALLE B. JEPPESEN, RICHARD GILROY, MAREK PERTKIEWICZ, JOHANE P. ALLARD, and BERNARD MESSING # *Division of Gastroenterology, University of Pittsburgh, Pittsburgh, Pennsylvania; Department of Medical Gastroenterology, Rigshospitalet, Copenhagen, Denmark; Department of Gastroenterology, University of Kansas Medical Center, Kansas City, Kansas; Department of General Surgery and Clinical Nutrition, Medical University of Warsaw, Warsaw, Poland; Division of Gastroenterology, University of Toronto, Toronto, Ontario, Canada; # Hopital Beaujon Service de Gastroenterologie et Assistance Nutritive, Clichy, France BACKGROUND & AIMS: METHODS: RESULTS: CONCLUSIONS: Although home parenteral nutrition (PN) can save the lives of patients with massive bowel loss that results in short-bowel syndrome and intestinal failure, quality of life is impaired by PN and its complications. We examined the 12-month tolerability and efficacy of teduglutide to reduce PN dependency. Patients who received teduglutide (0.05 or 0.10 mg/kg/d) for 24 weeks in a randomized controlled trial were eligible for a 28-week double-blind extension study; 52 patients were given 52 weeks of the same doses of teduglutide. We investigated the safety, tolerability, and clinical efficacy (defined as a clinically meaningful >20% reduction in weekly PN volume from baseline) at week 52. The most common adverse events reported included headache (35%), nausea (31%), and abdominal pain (25%); 7 patients withdrew because of adverse events (gastrointestinal disorders in 4). Both groups had progressive reduction in PN. At week 52, 68% of the 0.05-mg/kg/d and 52% of the 0.10-mg/kg/d dose group had a >20% reduction in PN, with a reduction of 1 or more days of PN dependency in 68% and 37%, respectively. Four patients achieved complete independence from PN. For patients with short-bowel syndrome intestinal failure, the efficacy of teduglutide was maintained over 52 weeks and the safety profile was sufficient for it to be considered for long-term use. Further studies are needed to determine whether these effects will translate into improved quality of life and reduced PN complications. ClinicalTrials.gov number, NCT Keywords: Short Bowel; Intestinal Failure; GLP-2. Watch this article s video abstract and others at tiny.cc/bz9jv. Scan the quick response (QR) code to the left with your mobile device to watch this article s video abstract and others. Don t have a QR code reader? Get one by searching QR Scanner in your mobile device s app store. See editorial on page 824. Human beings have evolved to survive in a potentially hostile environment. Vital organ function is preserved by either the occurrence of duplicate organs, in which case loss of one is covered by the function of the other, or, as in the case of single organs, such as the liver, pancreas, and gut, by an increase in function by the remaining cellular mass a process termed adaptation. With regard to the gut, loss of the whole colon is compatible with life because the small intestine adapts with time, reducing ileostomy fluid losses. Similarly, the loss of 50% of the small intestine is well tolerated in time by the remaining bowel increasing its reabsorption of secretions. Intestinal failure, defined as the inability to maintain protein-energy, fluid, electrolyte, or micronutrient balances when on a conventionally accepted, normal diet, 1 generally only occurs when more than 70% of the small intestine and colon, or more than 80% of the small intestine alone, has been lost. 1 Recent research has explored the use of peptide hormones that might be involved in the natural adaptation process to see Abbreviations used in this paper: ECL, electrochemiluminescent; ECP, Escherichia coli protein; GH, growth hormone; GLP-2, glucagonlike peptide-2; IV, intravenous; PN, parenteral nutrition; RCT, randomized controlled trial; SAE, serious adverse event; SBS-IF, intestinal failure associated with short-bowel syndrome by the AGA Institute /$

2 816 O KEEFE ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 7 Figure 1. Flow chart for the 52- week efficacy and safety analysis. Patients who received placebo treatment in the initial 6-month RCT are not included in this report. whether adaptation can be enhanced. The process of adaptation leads to hypertrophy of the remaining gut and hyperplasia of the mucosal surface in the form of villous hyperplasia. The biological half-life of these peptides is short; analogs have been synthesized by recombinant DNA techniques to make them more resistant to proteases, extending the half-life to hours rather than minutes so that they can be administered as single subcutaneous daily injections. First, growth hormone (GH) was manipulated to form Zorbtive (somatropin [recombinant DNA origin] for injection; EMD Serono, Rockland, MA), which clinical trials have shown to have the capacity to increase energy absorption up to 400 kcal/d. 2 Unfortunately, GH is not specific to the gut but affects tissues throughout the body and, in excess, is responsible for acromegaly. Second, because of these concerns, attention moved to glucagon-like peptide-2 (GLP-2), a peptide secreted by enteroendocrine L-cells in the distal bowel that is gut-specific. GLP-2 mimics many of the intestinal adaptation responses; namely, it produces villous hypertrophy, retards gastric secretion and emptying, increases mucosal blood flow and absorption, and is therefore a natural next candidate to investigate. 3,4 Initial human studies showed the potential of the native peptide to reduce fecal weight and improve fluid balance, 3 but because of its short half-life, 4 the peptide had to be injected 3 times per day. An analog, teduglutide, has been synthesized that is resistant to degradation by endogenous dipeptidyl peptidase-4, enabling it to be delivered as a single daily subcutaneous injection. 5,6 After the publication of encouraging uncontrolled studies in patients with intestinal failure associated with short-bowel syndrome (SBS-IF), 5,6 a 24-week, multicenter, multinational study was conducted in 83 patients with SBS-IF, randomized to 2 dose levels of teduglutide or placebo. 7 Findings showed that teduglutide 0.05 mg/kg/d was more effective than teduglutide 0.10 mg/kg/d in producing clinically significant reductions (ie, 20% 100%) in parenteral nutrition (PN). In addition, villus height, fasting plasma citrulline (a measure of enterocyte mass), and lean body mass also were increased without any increase in serious adverse event (SAEs). 7 The current analysis assessed the long-term safety, tolerability, and clinical efficacy of teduglutide in patients with SBS-IF during a 52-week treatment period. Methods Study Design Patients who completed a 24-week, randomized, double-blind, placebo-controlled, parallel group, multicenter, multi-international study of teduglutide (0.05 or 0.10 mg/kg/ d) 7 were eligible to enroll in this 28-week double-blind extension study, for a total treatment period of 52 weeks. Patients who were randomized to teduglutide in the initial randomized controlled trial (RCT) continued with the same teduglutide dose in the extension study (Figure 1). Patients who received placebo in the initial 24-week RCT also were eligible for randomized treatment in the extension study but they were excluded from this analysis because they did not receive teduglutide for the same length of time (ie, 52 weeks overall) as the patients who received teduglutide during the initial 24-week RCT. Patient Selection All patients who had satisfied the published inclusion and exclusion criteria 5 and completed the initial 24-week RCT were offered participation in the extension study. After receiving approval from local institutional review boards and medical ethics committees, 32 centers in the United States, Canada, Denmark, France, Poland, Germany, The Netherlands, the United Kingdom, and Belgium screened patients of both sexes who were at least 18 years of age and who had a history of SBS-IF caused by intestinal resection that resulted in dependency on parenteral energy or fluid and electrolytes at least 3 times per week for a period of at least 12 continuous months before the start of the study. Supplementary Table 1 provides the proportional recruitment from all the multicenter multinational centers who participated in the study. Exclusion criteria included a history of cancer or clinically significant lymphoproliferative disease with fewer than 5 years of a documented disease-free state, pregnancy, lactation, active Crohn s disease, radiation enteritis, scleroderma, celiac disease, refractory or tropical sprue, alcohol or drug abuse within 1 year, previous use of or allergies to teduglutide, or any hospitalization within 1 month before screening. A 12-week washout period was required for infliximab, GH or growth factors such

3 July 2013 TEDUGLUTIDE IN SHORT BOWEL SYNDROME 817 as native GLP-2, or other biological therapy, and a 30-day washout was required for systemic corticosteroids, methotrexate, cyclosporine, tacrolimus, sirolimus, octreotide, intravenous (IV) glutamine, or any investigational drug. The use of antimotility and antidiarrheal agents (eg, loperamide, diphenoxylate, codeine, and other opiates), histamine-2 receptor antagonists, proton pump inhibitors, bile acid binding agents, oral glutamine, diuretics, and oral rehydration solutions was required to be stable for 4 or more weeks before baseline evaluations and remain stable during the study. Protocol for Parenteral Nutrition Adjustment and Maintenance At the commencement of the initial RCT, eligible patients signed informed consent forms and entered a PN optimization and stabilization period in which IV fluid infusion volumes were adjusted until urine outputs were between 1 and 2 L/d and contained more than 20 meq of sodium and blood creatinine concentrations were normal. If successfully stabilized, patients were randomized to drug or placebo injected subcutaneously once daily for 24 weeks into 1 of the 4 quadrants of the abdomen or either thigh, with sites rotated daily. If the patient missed a dose, it was to be administered as soon as possible, with consecutive doses separated by at least 12 hours. The efficacy of the drug was assessed by its ability to reduce IV fluid requirements defined by the IV volumes required to maintain constant urine secretions of 1 to 2 L/d. PN was reduced at 4-week intervals if 48-hour urine volumes increased by 10% or more. At the end of the 24-week RCT, willing patients continued for an additional 28 weeks of active therapy for a total of 52 weeks as described earlier and the same protocol was applied to maintain fluid balance. Details of the weaning algorithm are provided in the Supplementary Table 2. The study data were gathered by the investigators and by a contract research organization. The data were held and analyzed by NPS Pharmaceuticals in collaboration with the principal investigators on the writing team, who had full access to the data. Collection and Analysis of Samples Details regarding the collection and analysis of data have been published previously. 7 Antibodies to teduglutide and Escherichia coli protein (ECP) antigens as control were measured using 2 different assays: electrochemiluminescent (ECL) (Tandem Laboratories, Biotechnology Services, West Trenton, NJ and BioVeris Corp, Gaithersburg, MD) and Meso Scale Discovery (MSD; Rockville, MD). ECL was used at baseline, at the end of the 24-week RCT, and again at the end of treatment, using similar methods for both teduglutide and ECP antibodies. The method is a direct antibody assay using biotinylated teduglutide for capture and ruthenylated protein A/G for detection; it is standardized using either an affinity-purified rabbit antiteduglutide antibody IgG or an affinity-purified rabbit anti-ecp antibody IgG for teduglutide or ECP antigens, respectively. In addition, a decision was made in accordance with the newly revised 2009 Guidance for Industry: BioAnalytical Method Validation 8 to repeat the immunogenicity assays for antibodies to teduglutide from samples taken during the 24-week RCT using the more sensitive MSD methods. A functional assay measuring cyclic adenosine monophosphate accumulation was used to characterize the neutralizing potential of any antibodies against teduglutide. Study Objectives The objectives of this RCT with extension study were to evaluate the long-term safety, tolerability, and efficacy of daily teduglutide administration for 52 weeks. As in the initial RCT, a clinically significant response was defined as a reduction of 20% or more in weekly PN volume at week 52 compared with baseline (beginning of the 52-week treatment period). Twenty percent was chosen because it would translate to the average home parenteral nutrition patient being able to eliminate 1 day of PN per week. Assessments of safety included monitoring for AEs, laboratory tests (hematology, serum chemistries, and urinalysis with microscopic analysis), and clinical evaluations (vital signs, physical examination, and electrocardiograms). Safety assessments for all patients also included body weight, 48-hour oral fluid intake and urine output, IV catheter complications, colonoscopy (if colon was present), and antibodies to teduglutide and ECP. Efficacy evaluations included PN volume and plasma citrulline levels at week 52. Most safety and efficacy evaluations and procedures were performed at 4-week intervals throughout the RCT and for the first 2 visits of the extension study; thereafter, they were performed at 4- to 6-week intervals through the end of treatment, and at follow-up evaluation 4 weeks after the last dose. Assessments were to be scheduled as close to noon as possible ( 2 h) to allow optimum time between parenteral administrations and to provide correlation between patients and visits. Secondary end points included fasting plasma citrulline levels to assess the absorptive enterocyte mass. 9 Data Analysis The primary patient population for all analyses was the intent-to-treat population, defined as all patients who entered the study and received the study drug. The safety population consisted of all subjects who received at least one dose of study drug. Only observed data were analyzed; missing data were not imputed (eg, last observation was not carried forward). Baseline and demographic characteristics, efficacy variables overall and for each treatment group, drug exposure for each treatment group, treatment-emergent AEs, and laboratory test results were summarized by descriptive statistics. Treatment groups at baseline were assessed for continuous variables by 1-way analysis of variance using 2-sided tests of significance and for discrete variables by the chi-squared test. Significance of the change from baseline within treatment groups was assessed by paired t tests. The average weekly parenteral support was calculated at 2-week intervals for patients who had at least 9 of the 14 daily parenteral volume data points. All authors had access to the study data and reviewed and approved the final manuscript. Results Patients Of the 56 patients who successfully completed the initial 24-week RCT in the teduglutide treatment arms, 52 agreed to continue on the 28-week extension (25 patients on 0.05 mg/kg/d and 27 patients on 0.10 mg/kg/d), for a combined 52-week treatment period (Figure 1). These 52 patients formed

4 818 O KEEFE ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 7 Table 1. Demographic Characteristics of Intent-to-Treat Population Teduglutide treatment group, mg/kg/d Variable 0.05/0.05 a (n 25) 0.10/0.10 a (n 27) Total (N 52) Age, y Mean (SD) 46.7 (13.48) 49.4 (14.04) 48.1 (13.76) Median (range) 49.0 (21 67) 53.0 (20 80) 51.0 (20 80) Sex, n (%) Female 10 (40.0) 14 (51.9) 24 (46.2) Male 15 (60.0) 13 (48.1) 28 (53.8) Race, n (%) Black 3 (12.0) 2 (7.4) 5 (10.0) White 22 (88.0) 25 (92.6) 47 (90.4) BMI, kg/m 2 Mean (SD) 21.5 (3.20) 22.1 (2.21) 21.8 (2.71) Median (range) 21.2 (16 28) 21.0 (19 26) 21.1 (16 28) Colon, n (%) Yes 19 (76.0) 18 (66.7) 37 (71.2) Etiology, n (%) Crohn s disease 7 (28.0) 11 (40.7) 18 (34.6) Vascular disease 11 (44.0) 7 (25.9) 18 (34.6) Injury 3 (12.0) 2 (7.4) 5 (10.0) Volvulus 2 (8.0) 4 (14.8) 6 (11.5) Other 2 (8.0) 3 (11.1) 5 (10.0) Remaining small intestinal length, cm Mean (SD) 63.4 (46.10) 57.6 (28.88) 60.6 (38.34) Median (range) 50.0 (10 200) 60.0 (10 135) 60.0 (10 200) Not known, n (%) 1 (4.0) 4 (14.8) 5 (9.6) Prescribed PN volume at study entry, L/wk Mean (SD) 9.8 (4.37) 13.1 (6.28) (5.65) Median (range) 8.0 (4 18) 11.4 (4 33) 10.6 (4 33) PN consumption level at study entry, n (%) Level 1: IV fluids only 4 (16.0) 2 (7.4) 6 (11.5) Level 2: PN 3 5 wk 15 (60.0) 15 (55.6) 30 (57.7) Level 3: PN 6 7 wk 6 (24.0) 10 (37.0) 16 (30.8) Treated for IV line infection, past 6 mo, n (%) Yes 10 (40.0) 5 (18.5) 15 (28.8) a Patients enrolled in the extension study were continued on teduglutide at the same dose as they were given during the RCT. the subject of the current analysis. Table 1 shows that the etiology of the intestinal failure was chiefly surgical resection for Crohn s disease or ischemic bowel (35% each). The overall estimated average remaining length of small bowel was 60.6 cm; approximately two thirds of patients had some colon-in-continuity. The majority of patients (58%) needed PN 3 to 5 days per week and 31% needed PN daily; PN consisted of IV fluids and electrolytes alone in 12%. Before the initiation of teduglutide treatment, the average time on PN was approximately 7 years, ranging from 1 to 24 years. Of the 52 patients who entered the extension study, 43 completed the 52-week treatment period. Patient compliance (defined as 80% of actual doses taken) was observed in both treatment groups for all dosing weeks, with no noticeable difference between treatment groups. The overall treatment compliance was 85% or greater for all dosing weeks. Safety and Tolerability No clinically meaningful differences were seen in vital signs, electrocardiogram, body weight, and physical examination during the study. There were no deaths. There were no significant changes in hemoglobin level, white cell count, platelets, differential counts, urea and electrolyte levels, liver function tests, and C-reactive protein concentrations over the 52- week treatment period (Supplementary Table 3). Overall, 96% (50 of 52) of patients reported at least one treatment-emergent AE during the 52-week treatment period. The most commonly reported treatment-emergent AEs were headache (35%), nausea (31%), abdominal pain (25%), nasopharyngitis (25%), vomiting (17%), catheter sepsis (17%), and urinary tract infection (17%) (Table 2). A total of 27 patients (52%) reported treatment-emergent AEs that were considered to be related to teduglutide: these events included 19 reports of gastrointestinal disturbance, chiefly abdominal pain; 26 injection site complaints (including bruising, erythema, hemorrhage, induration, pain, pruritus, reaction and/or swelling); and 6 complaints related to stomal hypertrophy and injury or stoma site reaction (some patients reported 1 AE). A total of 7 (13%) patients discontinued from the study because of an AE; of those, 4 cases were considered treatmentrelated, involving gastrointestinal disorders of moderate severity. All 4 patients had a history of Crohn s disease; in 1 patient, the symptoms resolved after discontinuation but were ongoing

5 July 2013 TEDUGLUTIDE IN SHORT BOWEL SYNDROME 819 Table 2. Summary of Treatment-Emergent Adverse Events Teduglutide 0.05/0.05 mg/kg/d (n 25) Teduglutide 0.10/0.10 mg/kg/d (n 27) Total N 52 Any AEs, n (%) a 23 (92) 27 (100) 50 (96) Cardiac disorders 3 (12) 2 (7) 5 (10) Gastrointestinal disorders 17 (68) 17 (63) 34 (65) Abdominal discomfort 1 (4) 1 (4) 2 (4) Abdominal distension 4 (16) 4 (15) 8 (15) Abdominal pain 7 (28) 6 (22) 13 (25) Diarrhea 1 (4) 3 (11) 4 (8) Flatulence 1 (4) 3 (11) 4 (8) Nausea 5 (20) 11 (41) 16 (31) Vomiting 2 (8) 7 (26) 9 (17) General disorders and administration site conditions 13 (52) 19 (70) 32 (62) Asthenia 2 (8) 3 (11) 5 (10) Injection site bruising 2 (8) 8 (30) 10 (19) Injection site erythema 1 (4) 7 (26) 8 (15) Hepatobiliary disorders 3 (12) 2 (7) 5 (10) Infections and infestations 21 (84) 20 (74) 41 (79) Catheter bacteremia 1 (4) 2 (7) 3 (6) Catheter sepsis 5 (20) 4 (15) 9 (17) Nasopharyngitis 6 (24) 7 (26) 13 (25) Sinusitis 2 (8) 2 (7) 4 (8) Urinary tract infection 5 (20) 4 (15) 9 (17) Injury, poisoning, and procedural complications b 7 (28) 9 (33) 16 (31) Contusion 1 (4) 4 (15) 5 (10) Intestinal stoma complication 3 (12) 3 (11) 6 (12) Investigations c 10 (40) 8 (30) 18 (35) Metabolism and nutrition disorders 6 (24) 5 (19) 11 (21) Dehydration 2 (8) 2 (7) 4 (8) Hypokalemia 2 (8) 0 2 (4) Musculoskeletal and connective tissue disorders 9 (36) 11 (41) 20 (39) Arthralgia 2 (8) 3 (11) 5 (10) Back pain 1 (4) 3 (11) 4 (8) Muscle cramp 2 (8) 2 (7) 4 (8) Peripheral edema 1 (4) 3 (11) 4 (8) Nervous system disorders 10 (40) 14 (52) 24 (46) Dizziness 2 (8) 3 (11) 5 (10) Headache 7 (28) 11 (41) 18 (35) Psychiatric disorders 4 (16) 2 (7) 6 (12) Renal and urinary disorders 3 (12) 4 (15) 7 (14) Reproductive system and breast disorders Respiratory, thoracic, and mediastinal disorders 8 (32) 5 (19) 13 (25) Respiratory tract congestion 0 2 (7) 2 (4) Skin and subcutaneous tissue disorders 5 (20) 8 (30) 13 (25) Social circumstances (family stress) 0 1 (4) 1 (2) Surgical and medical procedures 3 (12) 0 3 (6) Tooth extraction 3 (12) 0 3 (6) Vascular disorders 3 (12) 0 3 (6) a More than one event could be reported in a single patient. b Category also includes reports of device failure; excoriation; joint sprain; limb injury; medical device complication; muscle strain; postprocedural complication, hemorrhage, or pain; road traffic accident; skin injury; skin laceration; stoma site reaction; superficial injury of eye; and wound. c Includes abnormalities noted on blood or urine tests and decreased weight. in the other 3 patients. These cases are described in additional detail in Supplementary Table 4. A total of 27 patients (52%) reported SAEs; 5 patients experienced SAEs that were considered drug-related. Of the 27 patients reporting SAEs, 2 patients discontinued treatment; neither case was considered drug-related. One of these 2 cases was a 22-year-old woman (in the teduglutide 0.05-mg/kg/d group) who withdrew because of the finding of a small (2 mm) colon polyp that was removed by biopsy and shown to be a hyperplastic (not neoplastic) lesion; the other was a 63-year-old patient (in the teduglutide 0.10-mg/kg/d group) who experienced a cerebrovascular accident (stroke). In addition, 1 patient (in the teduglutide 0.05-mg/kg/d group) experienced an SAE (abdominal distress) that the investigator considered related to study drug, however, this SAE did not lead to study discontinuation. This patient was a 66-year-old white

6 820 O KEEFE ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 7 Figure 2. Mean reduction in parenteral support over 52 weeks for the 2 dose levels of teduglutide. Patients enrolled in the extension study were continued on teduglutide at the same dose as they were given during the RCT. woman with a history of small intestinal resection caused by vascular disease. She completed the first 24 weeks without complications but she was admitted to the hospital through the emergency room during the 28-week extension study with cramping, abdominal pain, and nausea, and was diagnosed with severe intestinal (partial small bowel) obstruction. She received nothing by mouth for 2 days and then advanced to clear liquids. The event resolved without surgery and she was discharged within 5 days. Teduglutide was not discontinued. No cancers were found on colonoscopy at the end of the study. No adenomatous polyps were identified; as noted earlier, one polyp was removed by biopsy and shown to be a hyperplastic lesion. There were no deaths during the study. Efficacy There were no significant changes in 7-day oral intakes and urine outputs over the 52 weeks. There was a progressive reduction in PN fluid infusions in the teduglutide-treated groups throughout the 52-week study period, with a greater average reduction in the teduglutide 0.05-mg/kg/d group (Figure 2). Patients on teduglutide 0.05 mg/kg/d decreased their PN volume by 4.9 L/wk (52%), and patients on teduglutide 0.10 mg/kg/d decreased their PN volume by 3.3 L/wk (26%) at week 52. At week 52, 68% of the patients in the teduglutide 0.05-mg/ kg/d group and 52% of the patients in the teduglutide mg/kg/d group were responders, defined as achieving the end point of a clinically significant response, with reductions of 20% or greater of baseline PN volume (ie, start of initial RCT). In comparison, 46% of the patients in the 0.05-mg/kg/d group and 25% of the 0.10-mg/kg/d group were responders at 24 weeks (ie, end of initial RCT), as previously reported. As shown in Table 3, 4 of the 24 responders at 24 weeks became nonresponders (defined by 20% reduction in PN) and 11 of the 19 nonresponders at 24 weeks became responders at 52 weeks. Of the 18 responders in the 24-week study who remained responders at 52 weeks, 12 of 16 (75%) were in the teduglutide 0.05-mg/kg/d dose group and 6 of 8 (75%) were in the teduglutide 0.10-mg/ kg/d dose group. Four subjects were completely weaned from parenteral support. Three patients in the teduglutide 0.05-mg/ kg/d treatment group became completely independent of PN after 25, 2, and 6.5 years on this treatment, receiving 5.4, 3.5, and 12.0 L/d parenteral support per week at baseline, respectively. Another patient receiving the teduglutide 0.10-mg/kg/d dose had been receiving parenteral support for 3.7 years and received 4.5 L/d parenteral support at baseline. Reductions in PN volume translated into a reduction of 1 or more days of PN dependency by 52 weeks in 68% of those who received treatment with a teduglutide 0.05-mg/kg/d dose and 37% of those with a teduglutide 0.10-mg/kg/d dose; 4 patients achieved complete independence from PN (3 during the 24-wk RCT and 1 during the 28-wk extension study). Table 3. Proportion of Responders ( 20% Reduction in PN Volume) Stratified by Response Outcome in the Initial RCT and Extension Study Teduglutide treatment group Initial RCT response outcome 0.05 mg/kg/d Responder Nonresponder 0.10 mg/kg/d Responder Nonresponder Total Responder Nonresponder Extension study, n (%) of patients Responder 12 (85.7) 5 (83.3) 8 (80.0) 6 (46.2) 20 (83.3) 11 (57.9) Nonresponder 2 (14.3) 1 (16.7) 2 (20.0) 7 (53.8) 4 (16.7) 8 (42.1)

7 July 2013 TEDUGLUTIDE IN SHORT BOWEL SYNDROME 821 Although the primary focus was on reductions in PN volume, a reduction in volume generally was associated with a reduction in IV nutrients. Among the 43 patients who received 52 weeks of teduglutide therapy, those on teduglutide 0.05 mg/kg/d had an IV energy intake reduction of kcal/wk and patients on teduglutide 0.10 mg/kg/d had an IV energy intake reduction of kcal/wk at week 52. Effect of Discontinuation of Drug As expected, 4 weeks after stopping drug there was a need to increase the PN volume in the teduglutide 0.05-mg/ kg/d group but not the teduglutide 0.10-mg/kg/d group. PN requirements, compared with end of study, changed from 4.0 ( 3.4) to 5.5 ( 4.4) L/wk in the teduglutide 0.05-mg/kg/d group and from 8.5 ( 5.1) to 7.9 ( 3.7) L/wk in the teduglutide 0.10-mg/kg/d group. Changes in Plasma Citrulline Level Evidence of increased enterocyte mass was shown by a 68% increase (P.0001 vs baseline) in fasting plasma citrulline level in the teduglutide 0.05-mg/kg/d group and an 86% increase (P.0001 vs baseline) in the teduglutide 0.10-mg/kg/d group. By 4 weeks after discontinuation after the end of the study, citrulline concentrations decreased by 20% in the 0.05-mg/kg/d group and by 32% in the 0.10-mg/ kg/d group from levels at week 52. These levels remained higher, at 24.4 M ( 11.51) and 19.8 M ( 7.99), respectively, than the baseline values of 19.5 M ( 9.91) and 16.6 M ( 8.80) before teduglutide therapy (ie, at the start of the RCT). Antibodies to Teduglutide and Escherichia coli Protein The ECL assay detected no antibodies specific to teduglutide during the 52-week treatment period, although antibodies specific to ECP were observed in 14 teduglutide-treated patients at the end of the 24-week RCT and in 16 patients at the end of treatment. Antibodies to teduglutide were detected by the MSD assay in 1 patient at baseline and in 14 patients at the end of the 24-week RCT. No neutralizing antibodies were detected by functional assays measuring cyclic adenosine monophosphate accumulation. Discussion This study represents a large body of evidence to date on the long-term efficacy and safety of teduglutide in the management of patients with SBS-IF. The results showed that the efficacy of the drug not only was maintained, but also increased after 52 weeks of treatment, and that the adverse effect profile was satisfactory for it to be considered for longterm use. The proportion of patients achieving a clinically meaningful reduction in PN volume (ie, 20% from baseline, see Study Objectives section) increased from 46% at 28 weeks to 68% at 52 weeks in the teduglutide 0.05-mg/kg/d group, which was approximately equal to an average reduction of 5 L of PN per week. This was associated with reductions in the number of days of PN infusions per week, a reduction in the need for catheter access, reductions in the parenteral energy requirements, and an increase in fasting plasma citrulline concentrations, which are biomarkers of enterocyte mass. In addition, 4 patients were able to discontinue PN fluids completely over the 52 weeks: 3 patients during the first 6 months and 1 patient during the second month. It is noteworthy that 11 of the 19 nonresponders at 24 weeks became responders by the end of the study. In contrast, 4 of the initial 24 responders became nonresponders during the second 6 months, but probably for reasons unrelated to the drug activity. One patient lost her diary, which invalidated the PN calculations. Another patient withdrew from the study after the detection of a colon adenoma during a routine endoscopy, which later was reclassified as a hyperplastic polyp on biopsy. A third patient developed an acute catheter-associated infection requiring hospitalization for IV antibiotic therapy accompanied by additional IV fluid infusions. Two patients continued to experience a decrease in their PN requirements compared with baseline, but at the 10% to 20% level. From a physiological point of view, it is difficult to understand why there was a linear decrease in PN needs over the 12-month period because, experimentally, the hormone has maximal effect on gut morphology within 4 weeks. Interestingly, the placebo-treated patients in the initial 6-month RCT also showed a progressive decrease, but to a limited degree. 7 This makes it likely that, with tight fluid balance control, the adaptation process continues far longer than we currently believe. If this is true, we may conclude that exogenous GLP-2 has the power to hyperadapt. Experimental evidence for this includes, first, the recognized increase in natural GLP-2 in the bloodstream after intestinal resection and its correlation with degree of resection 10 ; second, the amplification by exogenous GLP-2 treatment of the postresection increase in intestinal sucrose activity and mucosal hyperplasia in rats 11 ; and third, the increase in GLP-2 receptor expression after exogenous GLP-2 treatment of resected rats. 12 Clearly, further study on this is required, and it remains to be seen whether hyperadaptation persists after discontinuation of the drug. From our limited measurements on plasma citrulline in the current study, it seems unlikely. Our results clearly confirm the potency of this GLP-2 analog. The observed changes in IV fluid and nutritional requirements can be explained by the known complex actions of the endogenous gut peptide hormone GLP-2 on the small intestine, namely increased mucosal growth and function, increased mucosal blood flow, reduced gastric emptying, increased gut transit time, and reduced acid secretion. 7 The simple substitution of glycine for alanine has extended the half-life to 2.3 hours, which, taken together with the high bioavailability (87%) from subcutaneous injection, 4,6 enables the drug to be given as a convenient single daily dose. 5 Unfortunately, potent drugs have potent adverse effects, and although the adverse effect profile observed during 52 weeks of observation is acceptable, there is concern that the proliferative properties of the peptide may increase the risk of subacute obstructions in patients with adhesions and that it might enhance the risk of neoplastic progression in epithelial tissues. Animal models have suggested that GLP-2, in supraphysiologic concentrations, may enhance rather than induce tumor growth. 12,13 It was reassuring that no new neoplastic lesions were detected in any of the patients included in this 12-month study but, given the age of most

8 822 O KEEFE ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 7 patients with SBS-IF, careful colonoscopic surveillance will be mandatory in future longer-term studies. There are several therapeutic agents available to clinical practice that are effective in improving reabsorption of secretions and reducing fluid losses in patients with SBS-IF, but the attraction of teduglutide is its specificity of action in enhancing the natural adaptation process. For example, simple antimotility agents such as loperamide keep the digesta in contact longer with the remaining mucosal surface and increase colonic fluid absorption. 13,14 Acid-suppressant therapy and, in particular, proton pump inhibitors suppress endogenous secretions and reduce fluid losses. 15 Somatostatin derivatives such as octreotide are even more effective in suppressing secretory losses, from not only the stomach, but also the pancreas and biliary tree. 16,17 In addition, they inhibit motility, facilitating reabsorption. These beneficial effects can be dramatic in patients with insatiable thirst and high end-jejunostomy losses leading to serious fluid overload. 17 Unfortunately, long-term use is not advised because of concerns, based both on experimental 18 and human short-bowel studies, 16 about counteradaptation effects in suppressing pancreatic endocrine and exocrine function, and inhibiting mucosal growth and turnover. Recombinant GH therapy may offer additional advantages because it potentiates energy absorption, but unfortunately its actions are pleiotropic, affecting visceral and skeletal function accounting for the increased AEs and its restriction to short-term use (Cochran meta-analysis). 19 The AE profile of teduglutide was consistent with the known characteristics of the SBS patient population, the pharmacologic properties of the agent, or the mode of administration. The most common AEs in this study were headache, abdominal pain, nausea, nasopharyngitis, vomiting, catheter sepsis, and urinary tract infection. Of more concern are the data on the 4 patients, described in more detail in Supplementary Table 4, who were withdrawn from the study because of abdominal distress thought to be related to the drug. In 3 of these patients, the symptoms recurred after discontinuation of the drug, suggesting that the underlying bowel condition was responsible for the symptoms, but in 1 patient the acute intestinal obstruction appeared to be related to drug therapy and resolved after discontinuation. From what we know of the effects of GLP-2 on the intestine, it is plausible that the trophic effects of teduglutide could exacerbate obstruction in patients with pre-existing subacute obstruction. We know from personal experience that many patients treated with teduglutide show clinically obvious stomal enlargement. Another important point to note is that the use of teduglutide also is likely to increase the absorption of drugs, leading to toxicity. The symptoms of most patients with severe SBS are managed with high doses of antidiarrheal drugs (eg, loperamide and codeine). Improved absorption may result in constipation, which may precipitate symptoms of intestinal obstruction. Drugs with a narrow therapeutic index, such as benzodiazepine and digoxin, should be monitored carefully to avoid overdosing. In addition, increased absorption of fluid may lead to fluid retention and the exacerbation of congestive heart failure in susceptible individuals. With these concerns in mind, close monitoring at the commencement of drug administration is recommended and individual dose adjustment may be necessary in selected patients. It is assumed that the average reduction in PN needs by 5 L/wk or a reduction of 1 infusion day per week would translate into an improvement in quality of life, a reduction in potential long-term PN complications, and a reduction in overall costs, but all these objectives will need to be studied when the drug becomes available in clinical practice. This study was not sufficiently powered to answer these questions, and clinical trial conditions are not the same as clinical practice conditions (eg, clinical trial conditions include extra measurements and extra hospital visits). Furthermore, some subgroups of patients, such as those within 1 year of resection surgery or with radiation enteritis, were excluded from this study, and it remains to be seen what the overall applicability of this product will be in clinical practice. Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at and at doi.org/ /j.cgh References 1. O Keefe SJ, Buchman AL, Fishbein TM, et al. Short bowel syndrome and intestinal failure: consensus definitions and overview. Clin Gastroenterol Hepatol 2006;4: Seguy D, Vahedi K, Kapel N, et al. Low-dose growth hormone in adult home parenteral nutrition-dependent short bowel syndrome patients: a positive study. Gastroenterology 2003;124: Jeppesen PB, Hartmann B, Thulesen J, et al. Glucagon-like peptide 2 improves nutrient absorption and nutritional status in short-bowel patients with no colon. Gastroenterology 2001;120: Vipperla K, O Keefe SJ. Teduglutide for the treatment of short bowel syndrome. Expert Rev Gastroenterol Hepatol 2011;5: Jeppesen PB, Sanguinetti EL, Buchman A, et al. Teduglutide (ALX-0600), a dipeptidyl peptidase IV resistant glucagon-like peptide 2 analogue, improves intestinal function in short bowel syndrome patients. Gut 2005;54: Marier JF, Beliveau M, Mouksassi MS, et al. Pharmacokinetics, safety, and tolerability of teduglutide, a glucagon-like peptide-2 (GLP-2) analog, following multiple ascending subcutaneous administrations in healthy subjects. J Clin Pharmacol 2008;48: Jeppesen PB, Gilroy R, Pertkiewicz M, et al. Randomised placebocontrolled trial of teduglutide in reducing parenteral nutrition and/or intravenous fluid requirements in patients with short bowel syndrome. Gut 2011;60: Guidance for industry: bioanalytical method validation. Rockville, MD: US Department of Health and Human Services, US Food and Drug Administration, Center for Drug Evaluation and Research, Center for Veterinary Medicine; Crenn P, Coudray-Lucas C, Thuillier F, et al. Postabsorptive plasma citrulline concentration is a marker of absorptive enterocyte mass and intestinal failure in humans. Gastroenterology 2000;119: Martin GR, Wallace LE, Hartmann B, et al. Nutrient-stimulated GLP-2 release and crypt cell proliferation in experimental short bowel syndrome. Am J Physiol Gastrointest Liver Physiol 2005; 288:G431 G Scott RB, Kirk D, MacNaughton WK, et al. GLP-2 augments the adaptive response to massive intestinal resection in rat. Am J Physiol 1998;275:G911 G921.

9 July 2013 TEDUGLUTIDE IN SHORT BOWEL SYNDROME Koopmann MC, Nelson DW, Murali SG, et al. Exogenous glucagon-like peptide-2 (GLP-2) augments GLP-2 receptor mrna and maintains proglucagon mrna levels in resected rats. JPEN J Parenter Enteral Nutr 2008;32: Yazbeck R, Howarth GS, Abbott CA. Growth factor based therapies and intestinal disease: is glucagon-like peptide-2 the new way forward? Cytokine Growth Factor Rev 2009;20: Remington M, Fleming CR, Malagelada JR. Inhibition of postprandial pancreatic and biliary secretion by loperamide in patients with short bowel syndrome. Gut 1982;23: Buchman AL, Scolapio J, Fryer J. AGA technical review on short bowel syndrome and intestinal transplantation. Gastroenterology 2003;124: O Keefe SJ, Haymond MW, Bennet WM, et al. Long-acting somatostatin analogue therapy and protein metabolism in patients with jejunostomies. Gastroenterology 1994;107: O Keefe SJ, Peterson ME, Fleming CR. Octreotide as an adjunct to home parenteral nutrition in the management of permanent end-jejunostomy syndrome. JPEN J Parenter Enteral Nutr 1994; 18: Vanderhoof JA, Kollman KA. Lack of inhibitory effect of octreotide on intestinal adaptation in short bowel syndrome in the rat. J Pediatr Gastroenterol Nutr 1998;26: Wales PW, Nasr A, de Silva N, et al. Human growth hormone and glutamine for patients with short bowel syndrome. Cochrane Database Syst Rev 2010;6:2010CD Reprint requests Address requests for reprints to: Stephen J. D. O Keefe, MD, MSc, Division of Gastroenterology, University of Pittsburgh, Pittsburgh, Pennsylvania sjokeefe@pitt.edu; fax: (412) Acknowledgments The authors thank Peter A. Rittenhouse, PhD, and Maryann Travaglini, PharmD, of Complete Healthcare Communications, Inc (Chadds Ford, PA), for assisting the authors in editing the manuscript. Conflicts of interest The authors disclose the following: Stephen O Keefe, Palle Jeppesen, Richard Gilroy, Bernard Messing, and Johane Allard have served on the advisory board for NPS Pharmaceuticals and have been site investigators for clinical trials conducted by NPS Pharmaceuticals; and Marek Pertkiewicz has served as a consultant for NPS Pharmaceuticals and has been a site investigator for clinical trials conducted by NPS Pharmaceuticals. Funding This study was supported by NPS Pharmaceuticals.

10 July 2013 TEDUGLUTIDE IN SHORT BOWEL SYNDROME 823.e1 Supplementary Materials and Methods Protocol for Reduction of Parenteral Nutrition Volume At each study visit (every 4 6 weeks throughout the 28-week extension study), patients were assessed for PN adjustment according to the same weaning algorithm as used in the initial RCT. Measurements of 48-hour oral fluid intake and urinary output measurements included 1 day on and 1 day off PN, unless the PN was infused daily. To ensure safety of PN reduction, all patients were to be evaluated for clinical signs and symptoms of dehydration at 3 to 4 and at 6 to 7 days after each reduction in PN. If there was evidence of dehydration, the PN was to be restored to the previous level. Supplementary Table 1. Summary of Patient Recruitment: All Patients Screened Center Screened (N 139) Placebo (n 16) 0.05 mg/kg/d (n 35) 0.10 mg/kg/d (n 33) Total randomized (N 84) (6.5%) 0 (0.0%) 2 (5.7%) 3 (9.1%) 5 (6.0%) (5.8%) 3 (18.8%) 2 (5.7%) 1 (3.0%) 6 (7.1%) (1.4%) 0 (0.0%) 0 (0.0%) 1 (3.0%) 1 (1.2%) (5.0%) 2 (12.5%) 3 (8.6%) 1 (3.0%) 6 (7.1%) (2.2%) 1 (6.3%) 0 (0.0%) 1 (3.0%) 2 (2.4%) (0.7%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) (0.7%) 0 (0.0%) 1 (2.9%) 0 (0.0%) 1 (1.2%) (5.0%) 1 (6.3%) 0 (0.0%) 1 (3.0%) 2 (2.4%) (2.2%) 0 (0.0%) 1 (2.9%) 1 (3.0%) 2 (2.4%) (0.7%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) (2.9%) 0 (0.0%) 0 (0.0%) 2 (6.1%) 2 (2.4%) (1.4%) 0 (0.0%) 1 (2.9%) 1 (3.0%) 2 (2.4%) (0.7%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) (2.2%) 0 (0.0%) 1 (2.9%) 1 (3.0%) 2 (2.4%) (0.7%) 1 (6.3%) 0 (0.0%) 0 (0.0%) 1 (1.2%) (1.4%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) (3.6%) 0 (0.0%) 1 (2.9%) 0 (0.0%) 1 (1.2%) (2.2%) 0 (0.0%) 1 (2.9%) 2 (6.1%) 3 (3.6%) (6.5%) 0 (0.0%) 3 (8.6%) 2 (6.1%) 5 (6.0%) (2.2%) 0 (0.0%) 1 (2.9%) 0 (0.0%) 1 (1.2%) (6.5%) 1 (6.3%) 4 (11.4%) 2 (6.1%) 7 (8.3%) (7.2%) 2 (12.5%) 3 (8.6%) 3 (9.1%) 8 (9.5%) (8.6%) 1 (6.3%) 4 (11.4%) 4 (12.1%) 9 (10.7%) (6.5%) 0 (0.0%) 4 (11.4%) 1 (3.0%) 5 (6.0%) (2.2%) 1 (6.3%) 0 (0.0%) 0 (0.0%) 1 (1.2%) (1.4%) 1 (6.3%) 1 (2.9%) 0 (0.0%) 2 (2.4%) (0.7%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) (1.4%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) (0.7%) 1 (6.3%) 0 (0.0%) 2 (6.1%) 3 (3.6%) (2.2%) 0 (0.0%) 0 (0.0%) 1 (3.0%) 1 (1.2%) (5.0%) 1 (6.3%) 2 (5.7%) 2 (6.1%) 5 (6.0%) (0.7%) 0 (0.0%) 0 (0.0%) 1 (3.0%) 1 (1.2%) Total 139 (100%) 16 (100%) 35 (100%) 33 (100%) 84 (100%) NOTE. Table shows a summary of subject recruitment; all subjects were screened. The study was conducted at a total of 32 centers: 15 in the United States, 4 in Canada, and 13 in European countries.

11 823.e2 O KEEFE ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 7 Supplementary Table 2. Algorithm for PN Adjustments During Teduglutide Treatment Daily urine output, 48-h average PN adjustment 1.0 L/d 1.0 L/d and baseline a Baseline and 10% increase over baseline a 10% increase over baseline but 2.0 L/d 10% increase over baseline and 2.0 L/d Increase PN volume to level of last adjustment If patient is dehydrated or inadequately nourished, increase PN volume; if not, maintain PN volume at current level Maintain PN volume at current level Reduce PN volume by 10% of stabilized baseline level b Reduce PN volume by 10% of stabilized baseline level b up to a clinically appropriate amount a Baseline urine output is the urine volume obtained during the stabilization period before entering the initial RCT. b Baseline PN volume is the prescribed volume after patients showed urine output volume stability for 4 weeks before randomization/ entry into the initial RCT. Supplementary Table 3. Changes in Laboratory Assessments at 52 Weeks Teduglutide 0.05/0.05 mg/kg/d Teduglutide 0.10/0.10 mg/kg/d Analyte Baseline (n 25) Change from baseline Baseline (n 27) Change from baseline Alanine transaminase level, U/L Aspartate aminotransferase level, U/L Glutamyl transpeptidase level, U/L Total bilirubin level, mol/l Alkaline phosphatase level, U/L C-reactive protein level, mg/l Creatinine (plasma) level, mol/l Urine sodium level, mmol/l Blood urea nitrogen level, mmol/l Hemoglobin level, g/l NOTE. Laboratory assessments conducted at baseline and at the end of treatment showed nonsignificant reductions in liver function test values and increases in C-reactive protein concentrations with teduglutide therapy.

12 July 2013 TEDUGLUTIDE IN SHORT BOWEL SYNDROME 823.e3 Supplementary Table 4. Narratives of Patients Discontinuing the Study Because of an Adverse Event Considered Related to Treatment Subject 1: a 53-year-old white woman with a history of Crohn s disease and short bowel completed the first 50 weeks on teduglutide 0.10 mg/kg/d successfully but then complained of the non SAEs of abdominal pain and vomiting of moderate severity during the extension study. The investigator considered these events to be related to the study drug and withdrew the patient from the study. Symptoms of abdominal pain and vomiting were ongoing as of the last report. Subject 2: a 30-year-old white man with a history of Crohn s disease developed nausea and vomiting at 32 weeks after having been randomized to teduglutide 0.10 mg/kg/d. He also complained of a mild cough. No chest radiograph was performed. He had been treated for anemia (hemoglobin level, 6.4 mmol/l) and transfused earlier in the study. The investigator considered the nausea and vomiting related to the study drug and withdrew the patient from study. The AEs of nausea, vomiting, and cough resolved. Subject 3: a 62-year-old white woman with a history of Crohn s disease and recurrent abdominal pain completed 26 weeks of teduglutide 0.10 mg/kg/d without adverse events, before developing a recurrence of moderately severe abdominal pain. The investigator considered the abdominal pain related to the study drug and withdrew the patient from the study. The AE of abdominal pain was ongoing as of the last report. Subject 4: a 53-year-old white woman with a history of Crohn s disease and upper abdominal pain experienced an increase in weight as the only relevant change in her medical history after completing the first 24 weeks of teduglutide 0.05 mg/kg/d. After 44 weeks of treatment she developed a recurrent acute attack of Crohn s disease of moderate severity. The investigator was concerned that the recurrence was a consequence of the drug and withdrew her from the study. Follow-up evaluation has shown a persistence of Crohn s disease activity.