`
`research-article2015
`
`Original Communication
`
`Teduglutide for Safe Reduction of Parenteral Nutrient and/
`or Fluid Requirements in Adults: A Systematic Review
`
`Jane K. Naberhuis, BS1; and Kelly A. Tappenden, PhD, RD, FASPEN1,2
`
`Journal of Parenteral and Enteral
`Nutrition
`Volume XX Number X
`Month 201X 1 –11
`© 2015 American Society
`for Parenteral and Enteral Nutrition
`DOI: 10.1177/0148607115582063
`jpen.sagepub.com
`hosted at
`online.sagepub.com
`
`Abstract
`Background: Teduglutide (Gattex; NPS Pharma, Bedminster, NJ), a recombinant analogue of human glucagon-like peptide 2 (GLP-2),
`is the first long-term medical therapy approved for the treatment of adults dependent on parenteral nutrition (PN). Objective: To assess
`the efficacy and safety of teduglutide in reducing PN (parenteral nutrient and/or fluid) requirements in PN-dependent adults. Methods:
`Studies were identified using predefined search criteria and multiple databases, including Medline and Embase. The search was completed
`to November 30, 2014, in the absence of date or study design restrictions. Citation inclusion criteria and methodological quality were
`assessed by 2 independent reviewers. Outcomes of interest were changes in parenteral nutrient or fluid requirements and adverse event
`incidence. From 2693 unique citations, 76 abstracts were reviewed. Fourteen reports met the inclusion criteria, including data from 2 phase
`III, double-blind, placebo-controlled clinical trials and their respective extension studies. Data extraction was performed by 2 reviewers
`using a standardized form. Results: Teduglutide reduced PN requirements compared with placebo, whereas adverse event incidence was
`similar. Limitations: Number of subjects studied and length of follow-up. Conclusions: Teduglutide appears to be a safe and well-tolerated
`means to reduce PN dependence in adults, regardless of PN dependence duration. (JPEN J Parenter Enteral Nutr. XXXX;xx:xx-xx)
`
`Keywords
`teduglutide; parenteral nutrition; systematic review; intestinal failure
`
`Clinical Relevancy Statement
`
`Patients with intestinal failure (IF) are dependent on parenteral
`nutrition (PN) for nutrients and/or fluid, and prolonged PN
`dependence is associated with decreased quality of life and
`numerous complications. Teduglutide is the first long-term
`pharmacologic treatment indicated for adult patients with short
`bowel syndrome who are dependent on parenteral support.
`This systematic review demonstrates that teduglutide is effica-
`cious for minimizing PN dependence in adults regardless of
`PN dependence duration, with a therapeutic gain assessed from
`32.6%–39.4% compared with placebo in reducing PN volume
`requirements by ≥20%. Furthermore, longer teduglutide treat-
`ment duration is associated with increased clinical gains, and
`adverse event incidence on teduglutide is similar to that
`observed with placebo and is consistent with underlying IF.
`
`Introduction
`
`Intestinal failure (IF), caused by disease, congenital defect, or
`surgical resection, is characterized by the inability to maintain
`protein, energy, fluid, electrolyte, or micronutrient balance.1
`Parenteral nutrition (PN) is often required in IF to maintain body
`weight as well as fluid, nutrient, and electrolyte balance. While
`life-saving, long-term or permanent dependence on PN is asso-
`ciated with decreased quality of life1–7 and numerous complica-
`tions, including catheter-related bloodstream infections and
`sepsis, which are the primary cause of morbidity and hospital
`
`readmission in these patients.8 The risk of PN-related mortality
`rises with increasing PN dependence duration,9 but with proper
`care, PN complications are rarely lethal,10,11 and most deaths of
`patients receiving long-term PN are attributable to the underly-
`ing disease rather than to the administration of PN.12
`The goal of IF treatment is to promote enteral autonomy by
`maximizing the functional capacity of the remnant intestine,
`which is capable of increasing its absorptive capacity through
`mucosal surface area expansion and enhancement of absorptive
`efficiency per unit surface area.13–16 Capacity for this functional
`adaptation is maximal in the first 2 years following intestinal
`failure onset,17 and if enteral autonomy is not achieved during this
`
`From the 1Division of Nutritional Sciences and 2Department of Food
`Science and Human Nutrition, University of Illinois, Urbana, Illinois.
`
`Financial disclosure: None declared.
`
`Conflict of interest: Kelly A. Tappenden provided advisory board and
`educational services to NPS Pharma.
`
`Received for publication February 2, 2015; accepted for publication
`March 23, 2015.
`
`Supplemental material is available for this article at http://pen.sagepub.
`com/supplemental.
`
`Corresponding Author:
`Kelly A. Tappenden, PhD, RD, FASPEN, Kraft Foods Human Nutrition
`Endowed Professor, Department of Food Science and Human Nutrition,
`Division of Nutritional Sciences, 443 Bevier Hall, 905 South Goodwin
`Ave, Urbana, IL 61801, USA.
`Email: tappende@illinois.edu
`
`
`
`Downloaded from at UNIV CALIFORNIA BERKELEY LIB on December 10, 2015pen.sagepub.com
`
`
`
`Page 1
`
`
`
`2
`
`Journal of Parenteral and Enteral Nutrition XX(X)
`
`period, the likelihood of permanent IF and PN dependence is
`95%.18,19 However, enteral autonomy can be achieved beyond this
`initial 2-year period if effective long-term strategies are employed
`to maximize intestinal adaptation following resection.20,21
`Adaptation of the remnant intestine can be stimulated
`through a variety of interventions, including both dietary and
`pharmacologic strategies.22 Until recently, pharmacological
`treatments have focused largely on antisecretory, antimotility,
`and antidiarrheal medications. One promising pharmacologic
`intervention is the provision of exogenous glucagon-like pep-
`tide 2 (GLP-2). GLP-2 is a 33–amino acid peptide secreted
`from the enteroendocrine L cells of the distal intestine in
`response to luminal nutrients. First reported to stimulate
`enterocyte proliferation in 1996,23 GLP-2 has gained wide-
`spread support as an intestinotrophic mediator capable of
`increasing absorptive surface area, preventing mucosal atro-
`phy, and increasing DNA, RNA, and protein concentrations in
`intestinal cells of animals sustained on PN.24–26 Furthermore,
`GLP-2 enhances nutrient and fluid absorption,27 increases
`intestinal barrier function,28 and inhibits gastric emptying and
`stimulates intestinal blood flow.29–31 In a proof-of-concept
`study, GLP-2 increased intestinal wet weight absorption and
`decreased diarrhea in patients with short bowel syndrome
`(SBS).32
`GLP-2 has demonstrated consistent therapeutic promise for
`IF treatment. However, the half-life of GLP-2 is extremely
`short due to rapid degradation by dipeptidyl peptidase IV.
`Thus, teduglutide, a GLP-2 analogue that substitutes glycine
`for alanine in the second N-terminus position, was created,
`which extends the half-life from 7 minutes to 1.3–2 hours.33–35
`The US Food and Drug Administration (FDA) granted tedu-
`glutide orphan drug designation in 2000 and approved it for
`marketing for treatment of PN-dependent adult patients with
`SBS in December 2012.35 Teduglutide has also been approved
`for marketing in Europe under the trade name Revestive.36
`Given the complications and decreased quality of life asso-
`ciated with prolonged PN dependence, the potential for dupli-
`cate publication bias, and that extension study data are only yet
`available in abstract form, which may change substantially or
`never reach publication, the objective of this systematic review
`is 2-fold: (1) to distill the available data on teduglutide safety
`and efficacy in reducing PN requirements to original results
`and (2) to measure the impact of teduglutide via calculation of
`summary measures, including the number needed to treat to
`benefit (NNTB) or harm (NNTH), the odds ratio (OR), and
`therapeutic gain so that treatment decisions can be evidence
`based and well informed, taking into consideration both bene-
`fits and potential harms of teduglutide treatment.
`
`Methods
`
`This study was conducted according to the procedures outlined
`by the Cochrane Collaboration for systematic reviews37 to
`
`assess the safety and efficacy of teduglutide in reducing PN
`requirements in PN-dependent adults. A standard protocol for
`study identification, inclusion, and data abstraction was devel-
`oped and followed after establishment of the following study
`(population, intervention, comparison, and outcome [PICO])
`question: “In PN-dependent adult humans, would adding tedu-
`glutide to standard intestinal rehabilitation therapies safely
`result in reduced PN requirements when compared with stan-
`dard intestinal rehabilitation therapies alone?” These standard
`rehabilitation strategies include individualized treatments
`based on patients’ residual anatomy and SBS status and may
`include optimization of PN and/or conventional medications
`such as antisecretory agents or antidiarrheals.
`Multiple databases (Suppl. Table S1), clinical trial and
`adverse event registries, and pharmaceutical industry data-
`bases were searched from database
`inception
`through
`November 30, 2014, in the absence of date or study design
`restrictions using the following search terms: alx-0600, gattex,
`gly(2)-GLP-2, (gly2)GLP-2, revestive, and teduglutide. Results
`were restricted to English-language studies that enrolled
`PN-dependent adult humans and employed teduglutide, alone
`or in combination with additional therapies, to investigate the
`efficacy and/or safety of teduglutide in reducing PN require-
`ments. References from identified citations were cross-refer-
`enced for completeness. The outcomes of interest were changes
`in PN requirements and adverse event (AE) incidence. No
`restrictions were applied to the ways in which changes in PN
`requirements were expressed in study results. Hits were
`assessed for inclusion criteria and methodological quality by
`the 2 authors, including multiples domains of selection, perfor-
`mance, detection, attrition, reporting, and other biases. In the
`event where a risk of bias was unclear, attempts were made to
`clarify by contacting the senior study authors. Methodological
`quality of studies was graded per the Cochrane Collaboration,
`and discrepancies in trial bias assessments between reviewers
`were resolved by consensus. A data extraction form was devel-
`oped and piloted jointly by the authors using a representative
`sample of the studies to be reviewed, after which both authors
`performed data extraction. Qualitative data synthesis, rather
`than meta-analysis, was performed due to variations in length,
`timing, and dosing strategies of the included trials. Summary
`statistics, including NNTB (NNTB = 1/[teduglutide responder
`rate – placebo responder rate], rounded up to the next whole
`number), NNTH (NNTH = 1/[teduglutide event rate – placebo
`event rate], rounded up to the next whole number), OR (OR =
`[number of teduglutide-treated subjects experiencing event/
`number of event-free teduglutide treated subjects]/[number of
`placebo-treated subjects experiencing event/number of event-
`free placebo-treated subjects]), and therapeutic gain (teduglu-
`tide responder rate – placebo responder rate), were calculated
`as described by The Cochrane Collaboration37 to directly com-
`pare the safety and clinical efficacy of teduglutide with that of
`placebo.
`
`
`
`Downloaded from at UNIV CALIFORNIA BERKELEY LIB on December 10, 2015pen.sagepub.com
`
`
`
`Page 2
`
`
`
`Naberhuis and Tappenden
`
`3
`
`Figure 1. Study flowchart. GLP-2, glucagon-like peptide 2.
`
`Results
`
`Included Studies
`
`A total of 2693 citations were identified, and 1402 unique
`results remained after removal of duplicates. Potentially rel-
`evant citations were evaluated for inclusion after cross-refer-
`encing index terms and titles. Seventy-six abstracts were
`reviewed, after which the remaining 58 full-text articles and
`meeting abstracts were assessed for inclusion. Fourteen met
`the inclusion criteria (Figure 1). Reasons for article exclusion
`included duplicate data, review articles or articles that pro-
`vided interim findings when final results were available, use
`of native rather than analogue GLP-2, and enrollment of sub-
`jects who were not PN dependent. Five of the included cita-
`tions are full-text articles, and 9 are meeting abstracts. These
`citations describe 3 trials as well as their respective extension
`
`and substudies. Characteristics of included studies, including
`study durations, populations, and outcomes of interest, are
`found in Table 1.
`
`Risk of Bias in Included Studies
`
`All included studies had a low risk of bias in the following
`domains: (1) random-sequence generation (selection bias),
`(2) incomplete outcome data (attrition bias), (3) selective
`reporting (reporting bias), and (4) other bias (Figure 2). Risk
`of allocation concealment (selection) bias in the Gilroy et al38
`study, risks of blinding of participants and personnel (perfor-
`mance) and blinding of outcome assessment (detection) bias
`in the Jeppesen et al39–41 studies, and risks of blinding of out-
`come assessment (detection bias) in the Jeppesen et al,42,43
`Iyer et al,44 and Fujioka et al45 studies were determined to be
`
`
`
`Downloaded from at UNIV CALIFORNIA BERKELEY LIB on December 10, 2015pen.sagepub.com
`
`
`
`Page 3
`
`
`
`4
`
`Journal of Parenteral and Enteral Nutrition XX(X)
`
`Table 1. Characteristics of Included Studies.
`
`Studya
`
`Jeppesen et al48
`
`Jeppesen et al39,40,41,c
`
`Tappenden et al52
`
`O’Keefe et al49
`Gilroy et al38,c
`
`Compher et al46
`
`Jeppesen et al50
`(STEPS)
`
`Jeppesen et al,42,c
`Jeppesen et al,43,c
`Fujioka et al45,c
`(STEPS-2)
`
`Iyer et al44,c
`(STEPS-3)
`
`Ukleja et al47,c
`
`Description
`
`Phase III clinical trial. SBS males and females ≥18 y receiving
`PN ≥3 d/wk for ≥12 mo. Randomized to teduglutide 0.05 mg/
`kg/d (n = 35), teduglutide 0.10 mg/kg/d (n = 32), or placebo
`(n = 16) for 6 mo.
`
`Subset of Jeppesen et al48 subjects. Teduglutide 0.05 mg/kg/d
`(n = 10), teduglutide 0.10 mg/kg/d (n = 7), or placebo (n = 4)
`for 6 mo.
`Subset of Jeppesen et al48 subjects. Teduglutide 0.05 mg/kg/d
`(n = 32), teduglutide 0.10 mg/kg/d (n = 30), or placebo (n =
`15) for 6 mo.
`Extension of Jeppesen et al.48 Subjects previously on
`teduglutide 0.05 mg/kg/d (n = 25) or teduglutide 0.10
`mg/kg/d (n = 27) received 7 additional mo of same
`dose. Previously placebo-treated subjects randomized to
`teduglutide 0.05 mg/kg/d (n = 6) or 0.10 mg/kg/d (n = 7) for
`7 mo.
`Extension of Jeppesen et al.48 Subjects with stable (n = 15) or
`decreased (n = 7) PN requirement by 12 mo off teduglutide
`compared with those with increased PN requirement (n = 15).
`Phase III clinical trial. SBS males and females ≥18 y on PN ≥3
`d/wk for ≥12 mo. Previously teduglutide-treated subjects not
`eligible. Subjects randomized to receive teduglutide 0.05 mg/
`kg/d (n = 43) or placebo (n = 43) for 6 mo.
`
`Open-label extension of STEPS. Treatment in STEPS/
`STEPS-2: teduglutide/teduglutide 0.05 mg/kg/d (n =
`30), placebo/teduglutide 0.05 mg/kg/d (n = 29), or not
`randomized/teduglutide 0.05 mg/kg/d (n = 6) for 18–24 mo.
`
`Open-label extension of STEPS/STEPS-2. Teduglutide 0.05
`mg/kg/d (n = 14), with STEPS/STEPS-3 treatment of
`teduglutide/teduglutide (n = 5; treatment duration ≤42 mo),
`placebo/teduglutide (n = 6; treatment duration ≤36 mo), or
`not treated/teduglutide (n = 3; treatment duration ≤36 mo).
`
`Retrospective chart review of patients with SBS (n = 6)
`following FDA approval of teduglutide. Teduglutide 0.05
`mg/kg/d administered for 1–12 mo.
`
`Outcomes
`
`1. PN volume
`2. Responder rateb
`3. Complete PN weaning
`4. Intestinal adaptation
`5. Safety
`
`1. Intestinal adaptation
`
`1. Safety
`
`1. PN infusion frequencyd
`2. Responder rate
`3. Complete PN weaning
`4. Safety
`
`1. PN volume
`2. Complete PN weaning
`3. Safety
`
`1. PN volume
`2. PN infusion frequency
`3. Responder rate
`4. Safety
`
`1. PN volume
`2. PN infusion frequency
`3. Responder rate
`4. Complete PN weaning
`5. Safety
`1. PN volume
`2. PN infusion frequency
`3. Complete PN weaning
`4. Safety
`
`1. PN volume
`2. Responder rate
`3. Complete PN weaning
`4. Safety
`
`FDA, Food and Drug Administration; PN, parenteral nutrition; SBS, short bowel syndrome; STEPS, Study of Teduglutide Effectiveness in Parenteral
`Nutrition–Dependent SBS Subjects.
`aParent studies are in gray. Associated/extension studies are directly below each parent study.
`bResponder rate refers to subjects that achieved ≥20% volume reduction in PN requirement.
`cDenotes meeting abstract.
`dPN infusion frequency expressed as d/wk PN required.
`
`unclear since these domains were not specifically addressed
`in these citations. High risk of allocation concealment
`(selection) and blinding of participants and personnel (per-
`formance) bias were noted in the open-label Jeppesen
`et al,42,43 Iyer et al,44 and Fujioka et al45 studies. Risk of
`blinding of outcome assessment (detection) bias was also
`high in the Gilroy et al,38 Compher et al,46 and Ukleja et al47
`studies since the treatments were known by the outcome
`assessors.
`
`Outcomes of Interest
`
`Efficacy
`Responder rate. Table 2 shows the proportion of subjects
`classified as responders across studies, achieving ≥20% reduc-
`tion by volume in weekly PN requirements. In Jeppesen et al,48
`response rate at 20 and maintained at 24 weeks of treatment
`was higher (P = .005) in teduglutide 0.05 mg/kg/d (0.05 group)
`vs placebo subjects (NNTB = 3, OR = 12.63, therapeutic gain
`
`
`
`Downloaded from at UNIV CALIFORNIA BERKELEY LIB on December 10, 2015pen.sagepub.com
`
`
`
`Page 4
`
`
`
`Naberhuis and Tappenden
`
`5
`
`Low risk of bias Unclear risk of bias High risk of bias
`
`Random sequence generaƟon (selecƟon bias)
`
`AllocaƟon concealment (selecƟon bias)
`Blinding of parƟcipants & personnel (performance bias)
`Blinding of outcome assessment (detecƟon bias)
`
`Incomplete outcome data (aƩriƟon bias)
`
`SelecƟve reporƟng (reporƟng bias)
`
`Other bias
`
`0%
`
`25%
`
`50%
`
`75%
`
`100%
`
`Figure 2. Risk of bias assessment. Results of each bias domain
`are presented as percentages across all included studies.
`
`= 39.4%). Response rate of teduglutide 0.10 mg/kg/d (0.10
`group) subjects did not differ (P = .17) from placebo (NNTB
`= 6, OR = 5.00, therapeutic gain = 18.7%) Seventeen of 25
`(68.4%) 0.05 subjects and 14 of 27 (52.2%) 0.01 subjects were
`responders by 52 weeks of treatment.49 Compared with week
`24, by week 52, 4 of the 24 responders become nonresponders,
`and 11 of 19 nonresponders became responders. Twelve of the
`18 subjects who became responders by week 24 and remained
`so through week 52 were treated with teduglutide 0.05 and 6
`with teduglutide 0.10.49 Of subjects receiving placebo in the
`initial study48 but teduglutide in the extension study,49 6 of 6
`(100.0%) and 2 of 7 (28.6%) responded to teduglutide 0.05 and
`0.10, respectively.38
`Similarly, in the Jeppesen et al50 (Study of Teduglutide
`Effectiveness
`in Parenteral Nutrition–Dependent SBS
`Subjects [STEPS]) study, more (P = .002) teduglutide 0.05 vs
`placebo subjects were responders at week 24 (Table 2; NNTB
`= 4, OR = 3.89, therapeutic gain = 32.6%). In the extension
`study in which all subjects received teduglutide 0.05 (STEPS-
`2),43,45 subjects previously treated with teduglutide 0.05, pla-
`cebo, or not randomized achieved responder rates of 28 of 30
`(93.3%), 16 of 29 (55.2%), and 4 of 6 (66.7%), respectively.
`Teduglutide response was observed regardless of subject
`characteristics (age, remnant anatomy, baseline PN require-
`ments, or disease etiology).43 Importantly, teduglutide effi-
`cacy was demonstrated in responder rate ORs of >1 in both
`phase III trials.48,50 In the Ukleja et al47 study, all 6 patients
`(100.0%) experienced >20% reduction in PN volume while
`taking teduglutide.
`
`Changes in PN volume requirements. Using a strict paren-
`teral weaning algorithm that allowed for reductions in PN vol-
`umes of ≤10% at 4-week intervals, both the teduglutide 0.05
`and teduglutide 0.10 groups in the Jeppesen et al48 trial had
`reduced PN volume requirements compared with baseline at
`weeks 8, 12, 16, 20, and 24 (all P < .05). The placebo group
`also achieved significant reductions at weeks 12 and 24 (P =
`.02 and 0.03, respectively). At week 24, both teduglutide dose
`groups achieved mean PN volume requirement reductions
`of 2.5 L/wk, while the placebo group achieved a 0.91-L/wk
`
`reduction (P = .08). At week 24, the teduglutide 0.05, tedu-
`glutide 0.10, and placebo groups also achieved reductions
`(P = .001, P = .03, and P = .056, respectively) in parenteral
`energy intake compared with baseline, but reductions in either
`teduglutide-treated group did not differ (P = .11) from placebo.
`By 52 weeks of treatment,49 the teduglutide 0.05 and teduglu-
`tide 0.10 groups decreased their PN volume requirements by
`4.9 L/wk (52%) and 3.3 L/wk (26%), respectively, compared
`with baseline. However, 4 weeks after stopping treatment, PN
`requirements of both the teduglutide 0.05 and 0.10 groups
`increased compared with study end (from 4.0 ± 3.4 to 5.5 ± 4.4
`L/wk and 8.5 ± 5.1 to 7.9 ± 3.7 L/wk, respectively). There were
`no significant changes in 7-day urine outputs or oral intakes
`over the 52-week study period.
`Subjects with increased (INC) PN requirements by 12
`months after stopping teduglutide46 had a greater (P = .04)
`PN volume reduction while on drug compared with those
`with stable (STABLE) or decreased (DEC) requirements at
`12 months off drug (−4.7 vs −1.9 L/wk, respectively). INC
`had increased (P < .001) PN requirements at 3, 6, and 12
`months off drug vs study end while STABLE/DEC require-
`ments did not change. Furthermore, INC PN requirements
`were higher (P = .001) than STABLE/DEC (11.9 vs 5.7 L/
`wk) at 12 months off drug. Similar trends were observed in
`the subset of drug responders, in that INC had increased (P <
`.001) PN volume requirements at 3, 6, and 12 months com-
`pared with study end, while STABLE/DEC PN requirements
`did not change, and INC requirements were greater (P =
`.003) than those of STABLE/DEC subjects at 12 months off
`drug.
`In STEPS,50 using a weaning algorithm that allowed for
`10%–30% PN volume reductions of baseline PN levels at
`4-week intervals, teduglutide 0.05 and placebo subjects
`achieved mean L/wk reductions in PN volume requirements
`of 4.4 ± 3.8 (baseline 12.9 ± 7.8) and 2.3 ± 2.7 (baseline 13.2
`± 7.4), respectively, after 24 weeks of treatment. The differ-
`ence in absolute change in PN volume requirements between
`these groups was significant by week 8 (P < .01) and remained
`so through week 24 (P < .001). Similarly, the difference in
`percentage reduction in PN volume from baseline to week 24
`between groups became significant (P < .03) at week 12 and
`remained significant (P < .03) through week 24. By STEPS50/
`STEPS-243,45 treatment, the mean PN volume requirement
`reduction from baseline was 7.6 (66%), 3.1 (28%), and 4.0
`(39%) L/wk in the groups treated with teduglutide/teduglu-
`tide, placebo/teduglutide, and not randomized/teduglutide,
`respectively. By STEPS50/STEPS-344 treatment, teduglutide/
`teduglutide, placebo/teduglutide, and not-treated/teduglutide
`subjects reduced their PN requirements from baseline by 9.8
`(50%), 3.3 (35%), and 5.2 (73%) L/wk, respectively. In
`Ukleja et al,47 6 of 6 (100.0%) subjects experienced >20%
`reduction in PN volume requirements from baseline require-
`ments of 1–8 L/wk.
`
`
`
`Downloaded from at UNIV CALIFORNIA BERKELEY LIB on December 10, 2015pen.sagepub.com
`
`
`
`Page 5
`
`
`
`6
`
`Journal of Parenteral and Enteral Nutrition XX(X)
`
`Table 2. Efficacy of Teduglutide: Responder Rate.
`
`Studya
`
`Jeppesen et al48
` NNTBc/ORd/therapeutic gaine vs placebo (%)
`O’Keefe et al49
`Gilroy et al38
`STEPS: Jeppesen et al50
`
` NNTB/OR/therapeutic gain vs placebo (%)
`STEPS-2: Jeppesen et al,43 Fujioka et al45
`Ukleja et al47
`
`Teduglutide
`0.05 mg/kg/d
`
`45.7
`
`3/12.63/39.4
`68.4
`100.0
`
`62.8
`
`4/3.89/32.6
`73.8
`
`100.0
`
`Responder Rate,b %
`
`Teduglutide
`0.10 mg/kg/d
`
`25.0
`
`6/5.00/18.7
`52.2
`28.6
`
`—
`
`—
`—
`
`—
`
`Placebo
`
`6.3
`
`—
`—
`—
`
`30.2
`
`—
`—
`
`—
`
`NNTB, number needed to treat for an additional beneficial outcome; OR, odds ratio; PN, parenteral nutrition; SBS, short bowel syndrome; STEPS, Study
`of Teduglutide Effectiveness in Parenteral Nutrition–Dependent SBS Subjects; —, study did not include this treatment group.
`a Parent studies are in gray. Associated/extension studies are directly below each parent study.
`b Responder rate = (number of subjects achieving ≥20% volume reduction in PN requirement/total number of subjects in group) × 100.
`c NNTB = 1/(teduglutide responder rate – placebo responder rate), rounded up to next whole number.
`dOR = (number of teduglutide-treated responders/number of teduglutide-treated nonresponders)/(number placebo-treated responders/number placebo-
`treated nonresponders).
`e Therapeutic gain = (teduglutide responder rate – placebo responder rate).
`
`PN infusion frequency. Neither teduglutide-treated group
`experienced a significant reduction in the number of days per
`week that PN was required in the Jeppesen et al48 trial, but by
`52 weeks of treatment,49 17 of 25 (68%) teduglutide 0.05 and
`10 of 27 (37%) teduglutide 0.10 subjects achieved a ≥1 addi-
`tional d/wk reduction.
`In STEPS,50 which employed a more aggressive weaning
`algorithm than the 2011 trial,48 more (P = .005) teduglutide 0.05
`than placebo subjects (21 of 39 [54%] vs 9 of 39 [23%], respec-
`tively) achieved ≥1 additional d/wk off of PN by 24 weeks of
`treatment. In STEPS-2,43,45 38 of 65 (58.5%) subjects achieved
`≥1 additional d/wk reduction in PN requirements (by STEPS/
`STEPS-2 treatment: teduglutide/teduglutide, 21/30 [70.0%];
`placebo/teduglutide, 14/29 [48.3%]; not treated/teduglutide, 3/6
`[50.0%]), and 25 of 65 (38.5%) achieved ≥3 additional d/wk
`reduction in PN requirements. Of these 25, 18 (72%), 5 (20%),
`and 2 (8%) were previously treated with teduglutide, with pla-
`cebo, or not randomized, respectively.50 In STEPS-3,44 mean
`weekly PN infusion was reduced by 3.0, 1.7, and 2.8 d/wk in
`groups with STEPS/STEPS-3 treatment of teduglutide/teduglu-
`tide, placebo/teduglutide, and not treated/teduglutide.
`
`Complete weaning. Three subjects completely weaned
`from PN by 24 weeks of treatment in the Jeppesen et al trial48
`and remained off of PN 12 months later.46 Two of these sub-
`jects received teduglutide 0.05, and 1 received teduglutide
`0.10. By 52 weeks of treatment,49 1 additional teduglutide 0.05
`subject weaned from PN.
`In STEPS,50 no subjects were completely weaned from PN
`by 24 weeks of treatment, but in STEPS-2,43,45 13 of 88 (15%)
`achieved independence from PN, 10 of whom received tedu-
`glutide in the original randomized controlled trial.50 In STEPS
`3,44 2 subjects achieved independence from PN after 126 and
`
`130 weeks of teduglutide treatment. In the Ukleja et al47 study,
`4 of 6 subjects (66.7%) were able to wean from PN.
`
`Measures of intestinal adaptation. In Jeppesen et al,48
`teduglutide 0.05 subjects produced more (P < .05) urine at
`all time points vs baseline, despite constant oral fluid intake.
`Teduglutide 0.10 subjects had increased (P = .04) urine pro-
`duction at week 4, after which urine production returned to
`baseline, likely due to the reduced (P < .05) oral fluid intake at
`weeks 4, 8, 12, 16, 20, and 24. After 52 weeks of treatment,49
`fasting plasma citrulline, a biomarker of gut function and
`mucosal mass,51 increased 68% (P < .001) in teduglutide 0.05
`and 86% (P < .001) in teduglutide 0.10 subjects compared with
`baseline. These levels decreased after 4 weeks off drug by 20%
`and 32%, respectively, but remained higher than at study start.
`Seventy-two-hour balance studies were conducted in a subset
`of subjects from the Jeppesen et al48 study. At week 24, the pooled
`teduglutide groups (n = 11) demonstrated reduced fecal energy
`excretion41 and wet weight39 (P = .03 and P = .01, respectively),
`and in those whose dietary intake differed <10% from baseline,
`intestinal absorption significantly increased (P < .05) at weeks 8
`and 24.41 Teduglutide treatment also decreased fecal sodium (P <
`.001) and potassium39 (P = .003) excretion and increased plasma
`citrulline from baseline to week 24 (P = .001).40
`In STEPS,50 oral fluid intake of placebo-treated subjects
`exceeded (P < .05) that of teduglutide-treated subjects at weeks
`12, 20, and 24, and the reduction in fluid composite effect,
`defined as a summation of the increase in urine output (L/wk),
`reduction in PN/intravenous volume (L/wk), and reduction in
`oral fluid intake (L/wk), was greater (P ≤ 0.05) in the teduglu-
`tide vs placebo group at all time points. After 24 weeks of treat-
`ment teduglutide, but not placebo, increased (P < .001) plasma
`citrulline over baseline.
`
`
`
`Downloaded from at UNIV CALIFORNIA BERKELEY LIB on December 10, 2015pen.sagepub.com
`
`
`
`Page 6
`
`
`
`Naberhuis and Tappenden
`
`Table 3. Safety of Teduglutide: Adverse Event Incidence.
`
`Studya
`
`Jeppesen et al48
`
` AE
` SAE
` Dropout
` Laboratory findingsc
` Death
`Tappenden et al52
` Pathology (small and large bowel)d
` New secondary diagnosise
`O’Keefe et al49
` TE-AE
` TE-AE related to teduglutide
` SAE
` SAE related to teduglutide
` Dropout
` Pathology (colon only)
` Laboratory findings
` Death
`STEPS: Jeppesen et al50
` ≥1 TE-AE
` TE-SAE
` Dropout due to TE-AE
` Death
` Laboratory findings
`STEPS-2: Jeppesen et al,42,43 Fujioka et al45
` TE-AE
` SAE
` Dropout
` Laboratory findings
`STEPS-3: Iyer et al44
` TE-AE
` Dropout due to AE
` Malignancy, GI obstruction, death
`Ukleja et al47
` Complications
` Teduglutide discontinuation
` Teduglutide dose reduction
`
`Teduglutide
`0.05 mg/kg/d
`
`94.3
`37.1
`17.1
`
`NS
` 0.0
`
`0.0
`12.5
`
`92.0
` 51.9
` 51.9
` 9.6
`20.0
`0.0
`
`NS
` 0.0
`
`83.3
`35.7
`4.8
`0.0
` NS
`
`95.5
`63.6
`26.1
` NS
`
`100.0
`0.0
`0.0
`
`50.0
`16.7
`16.7
`
`Event Rate,b %
`
`Teduglutide
`0.10 mg/kg/d
`
`96.9
`34.4
`6.3
`
`0.0
`10.0
`
`100.0
`
`14.8
`0.0
`
`—
`—
`—
`—
`—
`
`—
`—
`—
`—
`
`—
`—
`—
`
`—
`—
`—
`
`7
`
`Placebo
`
`93.8
`31.3
`6.3
`
`0.0
`60.0
`
`—
`—
`—
`—
`—
`—
`—
`—
`
`79.1
`27.9
`6.9
`0.0
` NS
`
`—
`—
`—
`—
`
`—
`—
`—
`
`—
`—
`—
`
`AE, adverse event; GI, gastrointestinal; NS, not significant; SAE, serious adverse event; SBS, short bowel syndrome; STEPS, Study of Teduglutide
`Effectiveness in Parenteral Nutrition–Dependent SBS Subjects; TE-AE, treatment-emergent adverse event; TE-SAE, treatment-emergent serious adverse
`event; —, study did not include this treatment group.
`a Parent studies are in gray. Associated/extension studies are directly below each parent study.
`b Event rate = (number of subjects experiencing event/total number of subjects in group) × 100. More than 1 event could occur in a single subject.
`cLaboratory findings include the following values: vital signs, electrocardiogram, hemoglobin, platelets, differential white blood cells, urea, electrolytes,
`liver function tests, and C-reactive protein.
`d Pathology refers to dysplastic transformation, including adenomatous polyps.
`eNew secondary diagnoses include colitis (ulcerative, acute nonspecific, collagenous, eosinophilic, lymphocytic), Crohn’s disease, nonspecific increased
`mucosal inflammation, sarcoidosis, or villus abnormality (decreased villus height and/or volume).
`
`AE incidence. Table 3 illustrates the AEs, treatment-emergent
`AEs (TE-AE), serious AEs (SAE), treatment-emergent serious
`AEs (TE-SAE), and dropout rates of included studies, as well as
`information on laboratory findings, pathology, and death. In
`
`Jeppesen et al,48 79 of 83 (95.2%) subjects experienced at least 1
`AE, most commonly abdominal pain, headache, and nausea.
`The most common SAEs were catheter-related complications
`and infections, small intestinal obstruction, and fever. Compared
`
`
`
`Downloaded from at UNIV CALIFORNIA BERKELEY LIB on December 10, 2015pen.sagepub.com
`
`
`
`Page 7
`
`
`
`8
`
`Journal of Parenteral and Enteral Nutrition XX(X)
`
`Table 4. Safety of Teduglutide: Number Needed to Treat to
`Harm and Adverse Event Odds Ratio.
`
`Teduglutide 0.05
`mg/kg/d vs Placebo
`
`Teduglutide 0.10
`mg/kg/d vs Placebo
`
`Study
`
`NNTHa
`
`ORb
`
`NNTH
`
`OR
`
`Jeppesen et al48
`
`187
` AE
`17
` SAE
`10
` Dropouts
`STEPS: Jeppesen et al50
` ≥1 TE-AE
`24
` TE-SAE
`13
` Dropouts
`47
`
`1.10
`1.30
`3.10
`
`1.32
`1.44
`0.75
`
`32
`32
`∞
`
`—
`—
`—
`
`2.07
`1.15
`1.00
`
`—
`—
`—
`
`AE, adverse event; NNTH, number needed to treat for one additional
`harmful outcome; OR, odds ratio; SAE, serious adverse event; SBS,
`short bowel syndrome; STEPS, Study of Teduglutide Effectiveness
`in Parenteral Nutrition–Dependent SBS Subjects; TE-AE, treatment-
`emergent adverse event; TE-SAE, treatment-emergent se