Content uploaded by Angel Cebolla
Author content
All content in this area was uploaded by Angel Cebolla on Feb 28, 2018
Content may be subject to copyright.
Determination of gluten consumption in celiac disease patients on a
gluten-free diet
Jack A Syage,1Ciarán P Kelly,2Matthew A Dickason,1Angel Cebolla Ramirez,3Francisco Leon,3Remedios Dominguez,3
and Jennifer A Sealey-Voyksner1
1ImmunogenX, Newport Beach, CA; 2Beth Israel Deaconess Medical Center, Harvard Medical School, Boston MA; and 3Biomedal, Seville, Spain
ABSTRACT
Background: Celiac disease (CD) patients adhering to a gluten-
free diet (GFD) are exposed frequently to low levels of gluten
that contribute to symptoms and persistent intestinal histologic
damage.
Objective: We analyzed prior clinical data to determine how much
gluten is accidentally consumed while on a GFD. The aim was to
understand the range of gluten consumption for a wide distribution
of CD patients.
Design: A meta-analysis was conducted on data from 2 different clin-
ical programs: 1) measurements of gluten in stool and urine in CD
and non-CD populations; and 2) analysis of data from trials for the
investigational therapeutic latiglutenase. The stool and urine studies
included controlled gluten challenges. A calibration factor was ap-
plied that allowed normal ingestion of gluten to be computed from
the urine and stool measurements. From the latiglutenase trial data,
a determination of gluten consumption was made by estimating how
much gluten was eliminated from patients’ diets due to a trial effect
that led to improved histology even in the placebo group.
Results: The average inadvertent exposure to gluten by CD individ-
uals on a GFD was estimated to be ∼150–400 (mean) and ∼100–150
(median) mg/d using the stool test and ∼300–400 (mean) and ∼150
(median) mg/d using the urine test. The analyses of the latiglutenase
data for CD individuals with moderate to severe symptoms indicate
that patients ingested signicantly >200 mg/d of gluten.
Conclusions: These surrogate biomarkers of gluten ingestion indi-
cate that many individuals following a GFD regularly consume suf-
cient gluten to trigger symptoms and perpetuate intestinal histologic
damage. Am J Clin Nutr 2018;107:201–207.
Keywords: celiac disease, gluten exposure, gluten-free diet
INTRODUCTION
Celiac disease (CD) is the most common autoimmune gas-
trointestinal disease, affecting ∼1% of the world population
(1–3). There are currently no US Food and Drug Administration
(FDA)–approved treatments, other than a gluten-free diet (GFD),
which is exceedingly difcult to maintain. The average Western
diet contains ∼5–15 g gluten/d (4). Gluten ingestion as low as
50 mg/d can be harmful to some celiac patients (5). The elim-
ination of 99% of gluten from a diet may still be insufcient
to avoid symptoms and histologic damage. The FDA has estab-
lished a guideline that foods labeled gluten free must contain <20
ppm gluten (6). However, there are difculties with currently ap-
proved analytical methods for the detection and quantication of
gluten in certain foods (e.g., fermented and hydrolyzed foods)
(7–9).
There is an unmet need to protect against unintended gluten
ingestion, particularly since persistent uncontrolled gluten expo-
sure is known to lead to life-long health issues and comorbidities
such as anemia, malnutrition, and lymphoma (10). As such, in-
vestigational drugs in clinical development are generally intended
to be used as an adjunct to a GFD (11–14).
Despite the obvious need to protect CD patients against ex-
posure to gluten consumption, there is surprisingly very little
known about the quantity of gluten that is accidently consumed
episodically and continually for those on a GFD. Much has been
written about GFDs and the complexities, difculties, and chal-
lenges associated with maintaining strict adherence across so-
cial and demographic groups and behaviors (15–17). However,
we are unaware of any studies that attempt to analytically de-
termine the actual quantity of gluten that is consumed while on
aGFD.
In this work, we performed a meta-analysis based on data from
clinical studies that provided key information needed to deter-
mine the amount of gluten that CD patients consume while at-
tempting to follow a GFD.
Supported by ImmunogenX.
ImmunogenX is a clinical-stage company developing the therapeutic drug
latiglutenase for treating celiac disease and also a minimally invasive drug
biomarker and blood test for monitoring the villous health of the small intes-
tine.
Address correspondence to JAS (e-mail: jsyage@immunogenx.com).
Abbreviations: CD, celiac disease; FDA, Food and Drug Administration;
GFD, gluten-free diet; GIP, gluten immunogenic peptide; LOD, limit of de-
tection; Vh:Cd, villous height–to–crypt depth ratio.
Received August 14, 2017. Accepted for publication November 29, 2017.
First published online February 26, 2018; doi: https://doi.org/10.1093/
ajcn/nqx049.
Am J Clin Nutr 2018;107:201–207. Printed in USA. © 2018 American Society for Nutrition. All rights reserved. 201
Downloaded from https://academic.oup.com/ajcn/article-abstract/107/2/201/4911450
by PPD Development LP user
on 27 February 2018
202 SYAGE ET AL.
FIGURE 1 Flow chart outlining the basis for selecting the studies used in the present analysis. CD, celiac disease; GFD, gluten-free diet; GIP, gluten
immunogenic peptide; Vh:Cd, villous height–to–crypt depth ratio.
METHODS
The analysis was conducted combining clinical study re-
sults from the following sources: 1) measurements of gluten
in stool (NCT02711397 and NCT01478867) (18,19) and urine
(NCT02344758) (20) in non-CD and CD populations; and 2)
trial effect and symptom results for the study of latiglutenase
(NCT01255696) (13) and (NCT01917630) (21). These studies
(except for NCT01255696) focused primarily on CD patients
experiencing moderate to severe symptoms. The decision pro-
cess for selecting these studies is delineated by the ow chart in
Figure 1.
The measurements of gluten immunogenic peptides (GIPs) in
stool were performed using the ELISA Sandwich G12/G21 assay
(iVYLISA GIP, Biomedal SL). In urine, the results were deter-
mined by a quantitative lateral ow immunoassay (iVYCHECK
GIP Urine, Biomedal SL) using a lateral ow reader A1/G12
(iVYCHECK Reader, Biomedal SL). These measurements were
accompanied by controlled measurements of gluten exposure uti-
lizing a gluten challenge. A conversion factor (described be-
low) was determined that allowed the ingestion of gluten to be
computed. Gluten challenge measurements in stool showed that
gluten could be detected in stool for ≤4 d after a gluten chal-
lenge, indicating a sufciently long residence time (18,19), such
that a measured value of gluten in stool on any given day could
be a good indication of the peak gluten level and no additional
correction factor was needed. Gluten challenge measurements in
urine indicated a residence time of gluten in urine of about half
aday(20). Therefore, random measurements of gluten in urine
following unintended gluten ingestion would, on average, register
a value of about half that for the peak gluten level. The conver-
sion factor from GIP (expressed as ng/mL urine) to gluten inges-
tion (expressed as mg) was determined by 2 different methods: 1
was based on uncorrelated urine measurements following 42- and
84-mg gluten challenges, and the other was based on correlated
urine measurements to 500 mg gluten challenge. The former
method used a factor of 2 correction for the gluten residence
time. The details of how these conversion factors are computed
are given in the Results section.
In the second analysis we estimated the quantity of gluten that
was removed from the diets of patients in a Phase 2b latiglutenase
study (ALV003-1221) (21). The quantity of gluten eliminated
was determined by relating the improvement in villous height to
crypt depth ratio (Vh:Cd) for placebo patients to a standard t that
related deterioration of Vh:Cd to a continuous gluten intrusion,
as measured in a previous Phase 2a latiglutenase trial (ALV003-
1021) (13). Three cohort groups were given 1.5, 3.0, and 6.0 g of
gluten daily for 6 wk and experienced changes to their mucosa
(࢞Vh:Cd) of 2.8–2.2, 2.6–1.5, and 2.8–1.1, respectively. A poly-
nomial t with a (0,0) intercept led to the following equation:
wg=46.5Vh:Cd2−68.2Vh:Cd (1)
where wgis the aggregate change in gluten weight to the diets
of the patients corresponding to the mucosal change (Vh:Cd).
The placebo as well as latiglutenase arms improved their mucosal
health due to the Hawthorne effect. From this improvement, one
can estimate the amount of gluten removed from their diets using
Equation 1. We further estimated the amount of gluten that re-
mained in their diets during the treatment phase of the trial, using
the trial effect improvement on symptoms (Results section). The
total gluten intake for these symptomatic patients before the trial
began can thus be estimated.
The statistical methods for the data used in this analysis are
published in the referenced manuscripts. The analysis in this
Downloaded from https://academic.oup.com/ajcn/article-abstract/107/2/201/4911450
by PPD Development LP user
on 27 February 2018
GLUTEN CONSUMPTION IN CELIAC DISEASE PATIENTS 203
TABLE 1
Gluten consumption as measured in stool1
GIP concentration (μg/g stool) Gluten daily consumption (mg)
Cohort nMean Median SD Mean Median SD
Healthy non-CD
Adults 73 5.23 7.60 2.95 7802 11,699 4757
CD on GFD
Adults (≥13 y) 74 0.22 0.13 0.40 244 141 488
Children (4–12 y) 79 0.32 0.11 0.88 387 118 1216
Children (0–3 y) 35 0.14 0.10 0.19 155 104 214
1Conversion factor for y=GIP (μg/g stool) to x=gluten daily consumption (mg) is y=0.0649x2+1.0461x. CD, celiac disease; GFD, gluten-free diet;
GIP, gluten immunogenic peptide.
study used standard polynomial ts and mean, median, and SD
calculations. The histogram plots were calculated using the his-
togram function in Excel and using a bin size to provide sufcient
resolution of the gluten distribution.
RESULTS
Measurements of gluten in stool and urine
In a series of studies, an estimate of gluten consumption by CD
patients following a GFD was determined by measuring GIPs in
stool (18,19) and urine (20). Population groups included healthy
non-CD patients and CD patients, each segmented as adults and
children. Measurements were also performed on healthy patients
under controlled gluten challenge conditions.
Table 1 shows results for stool samples. In this study a gluten
challenge was conducted for several days to equilibrate the gluten
content in stool. There was a 1- to 2-d induction period for gluten
to be detected in stool and similarly to be eliminated. The factor
for converting GIP concentration (in micrograms per gram) (x
variable) to gluten daily consumption (in milligrams) (yvariable)
was determined from measured mean values of 6.2 and 14.9 μg/g
in stool for daily gluten challenges of 9 and 30 g (18). Fitting
to a second-order polynomial going through the origin gave the
relation y=0.0649x2+1.0461x+0.0. The computed gluten
daily consumption for healthy non-CD adults was found to be
7.8 g (mean) and 11.7 g (median). Non-CD children were not
included in this study. This value is consistent with 5–15 g for
a typical gluten-containing diet (measured in Denmark) (4). This
analysis utilized the complete set of data [compared with the data
plotted in Figure 2 of Comino et al. (18)] for values above the
quantitation limit.
Both the mean and median values for each population group
are reported in Tabl e 1 due to the asymmetry in the distribution
of gluten ingestion. The computed daily gluten consumption for
CD adults (≥13 y old) on a GFD was 244 mg (mean) and 141 mg
(median), for older children (4–12 y old) it was 387 mg (mean)
and 118 mg (median), and for younger children (0–3 y old) it was
155 mg (mean) and 104 mg (median). These values must be qual-
ied in terms of accuracy as more than half of these individuals
recorded GIP (micrograms per gram) values below the limit of
quantitation (LOQ =0.16 μg GIP/g stool sample), computing to
∼169 mg gluten consumed daily (LOD =0.06 μg/g correspond-
ingto63mg)(19). As an example, for the ≥13 y old patients,
45 of 74 measured below the LOQ, which computes to 39%
averaging >169 mg gluten consumed daily. For children 0–3 and
4–12 y old this gure is 14% and 28%, showing a trend toward
increased gluten consumption with age.
Table 2 summarizes the results of urine samples, an indepen-
dent study from the stool results. Two methods were used to
estimate the conversion factor from GIP concentration in urine
to daily gluten consumption. The rst made use of the origi-
nal data from Moreno et al. (20) where the LOD was 3.5 ng
GIP/mL and gluten challenges of 42 and 84 mg (revised from
originally reported values of 25 and 50 mg) were mostly unde-
tected and detected, respectively. We therefore estimate the LOD
to correspond to the average of these gluten ingestions, giving
63 mg. The collection of the urine was uncorrelated with the
gluten consumption and a factor of 0.5 was introduced, reecting
the ∼0.5-d residence time during which gluten remains in urine
(essentially the full width, half height of the peak concentration)
(20). This led to a conversion factor for gluten daily consump-
tion (in milligrams) per GIP concentration (expressed as ng/mL
urine) measured of 63/(3.5 ×0.5) =36. A subsequent gluten
challenge study gave 18 patients 500 mg of gluten on 2 subse-
quent days (at dinner time) and then collected urine samples at
3 specic times during the following day (to be published). The
LOD was 2.2 ng GIP/mL. These unpublished results (A Cebolla
and R Dominguez) indicate a conversion factor of 50 (similar to
above), except the 0.5 factor was not necessary because the urine
measurement was at a xed time after ingestion. This value is
consistent with the previous factor and is the value that we use
in the current analysis as we believe these are more accurate data
and remove the need for the 0.5 urine residence time assumption.
We found that the computed daily gluten consumption for
healthy non-CD individuals was 5.7 g (mean) and 4.3 g (median)
for adults and 4.4 g (mean) and 0.74 g (median) for children.
This is lower than the 5–15 g of gluten (cited above) for a typical
gluten-containing diet and lower than the measurements in stool
(above). This may reect an upper limit or saturation level for the
ELISA assay. The computed daily gluten consumption for CD
individuals on a GFD was 363 mg (mean) and 158 mg (median)
for adults and 316 mg (mean) and 149 mg (median) for chil-
dren. Of note, similar to the stool results above, more than half of
these individuals recorded GIP (expressed as ng/mL) values be-
low the LOD (<188 mg gluten). Therefore, we assigned values
that linearly decremented from the LOD value to zero. We be-
lieve this is a more realistic assumption than assigning values of
zero as it is more reective of the extrapolation of values above
the LOD. By comparison, if these values are assigned zero, then
Downloaded from https://academic.oup.com/ajcn/article-abstract/107/2/201/4911450
by PPD Development LP user
on 27 February 2018
204 SYAGE ET AL.
FIGURE 2 Histogram plots of the distribution of gluten consumption in the different population groups as measured by GIP in stool. Dotted lines represent
demarcations for patients (in percentages) with greater-than-indicated gluten consumption. CD, celiac disease; GIP, gluten immunogenic peptide.
for adults the mean decreases from 363 to 319 mg, not a marked
difference.
We evaluated the distribution of gluten ingestion for the dif-
ferent CD and non-CD population groups. The stool measure-
ments are presented in Figure 2. For the non-CD adults (healthy
controls), there is a large spike in the distributions for gluten
consumption of >7 g, which corresponds to the upper signal sat-
uration limit. This corresponds to 58% of the population consum-
ing >7 g gluten/d. For children (0–3 and 4–12 y old) and adult
(≥13 y old) CD populations on a GFD, the gluten consumption
is considerably less, but consumption is not insignicant; adult
consumption of gluten of >300 mg/d occurs 18% of the time as
seen in Figure 2.
Figure 3 shows histogram plots for the distribution of test
subjects for the urine measurements. These results show that
the large majority of non-CD subjects consume <10 g gluten/d
TABLE 2
Gluten consumption as measured in urine1
GIP concentration (ng/mL urine) Gluten daily consumption (mg)
Cohort nMean Median SD Mean Median SD
Healthy non-CD
Adults 42 113.2 85.3 112.1 5658 4264 5605
Children 34 87.9 14.8 117.0 4395 740 5848
CD on GFD
Adults 27 7.3 3.2 12.7 363 158 634
Children 31 6.3 3.0 7.1 316 149 354
1Conversion factor for GIP (ng/mL urine) to gluten daily consumption (mg): 50. CD, celiac disease; GFD, gluten-free diet; GIP, gluten immunogenic
peptide.
Downloaded from https://academic.oup.com/ajcn/article-abstract/107/2/201/4911450
by PPD Development LP user
on 27 February 2018
GLUTEN CONSUMPTION IN CELIAC DISEASE PATIENTS 205
FIGURE 3 Histogram plots of the distribution of gluten consumption in the different population groups as measured by GIP in urine. Dotted lines represent
demarcations for patients (in percentages) with greater-than-indicated gluten consumption. CD, celiac disease; GIP, gluten immunogenic peptide.
(a low value for a gluten-containing diet). The CD subjects obvi-
ously consume considerably less gluten on a daily basis, and these
data show 30% of adults and 32% of children consume >300 mg
gluten/d which is greater than, but consistent with, the results in
Figure 2 for the stool analysis.
Calculations based on the latiglutenase clinical trials
The Phase 2b ALV003-1221 trial was a real-world study where
patients were instructed to continue their GFD, but not to change
their normal dietary behavior. Over the 12-wk trial period, pa-
tients, on average, improved their GFD and mucosal health as
measured by their Vh:Cd. Figure 4 shows that the improvement
was similar for seropositive and seronegative subjects (mean
Vh:Cd 0.28) and for a small group of patients who contin-
ued on for a total of 24 wk the improvement increased further
(mean Vh:Cd 0.41). If we input 0.28 for Vh:Cd in Equation
1,weobtainwgof –15.4 g over the 12-wk treatment period,
which computes to a mean of 184 mg/d of gluten removed from
individual’s normal GFDs. Of note, Equation 1, which was de-
rived from added gluten to the diet, is not symmetric for positive
and negative values of wg.Ifwemakeitsymmetric,thenwe
obtain wgof –22.7 g or 271 mg/d. The former calculation as-
sumes that mucosal damage occurs more quickly or more exten-
sively than recovery; the latter calculation assumes they are the
same. There is no conclusive understanding of these phenomena,
and therefore we treat them as boundary conditions and use the
average of 228 mg/d for gluten elimination. This determination is
limited by the assumption that cumulative change in Vh:Cd from
total exposures over 6 wk of gluten challenge is similar to the
same exposure (or removal of gluten) over 12 wk.
The above conditional calculation is based on objective mea-
sures that clearly show that trial-bound patients signicantly re-
duced gluten from their diets and sets a lower limit on the amount
of gluten typically consumed in their normal GFDs. More dif-
cult is an estimate of the gluten consumption that remained in
their diets during the treatment period. There was clearly contin-
ued gluten intake as the trial revealed a statistically and clinically
signicant symptom improvement with the drug from baseline
relative to placebo of the order of 30–50% (for seropositive pa-
tients) for abdominal pain, bloating, tiredness, and constipation
(14). Any further effort to estimate the total gluten consumption
in this population of moderately to severely symptomatic patients
would invoke unjustiable assumptions so we instead leave this
Downloaded from https://academic.oup.com/ajcn/article-abstract/107/2/201/4911450
by PPD Development LP user
on 27 February 2018
206 SYAGE ET AL.
FIGURE 4 Vh:Cd for the placebo patients evidencing the trial effect in the Phase 2b trial. Patient numbers are: seropositive: n=54 (12 wk), n=12
(24 wk); seronegative: n=68 (12 wk), n=18 (24 wk). Seropositive patients are dened as patients registering a positive titer for either of anti-TG2 IgA, DGP
IgA, or DGP IgG. Delta, change in villous height–to–crypt depth for patients; DGP, deamidated gliaden peptide; PT, patient; TG2, tissue transglutaminase 2;
Vh:Cd, villous height–to–crypt depth; , change.
analysis with the knowledge that the amount is >228 mg/d by a
factor of 1/(1 – f)wherefis the fraction of gluten that was elimi-
nated from the diet due to the trial effect. By way of example, if
f=50%, then the implied pretrial gluten consumption would be
456 mg/d.
DISCUSSION
Gluten is ubiquitous, and continued exposure leads to persis-
tent histologic injury and episodic symptom distress in CD pa-
tients following a GFD. The GFD is vulnerable to gluten expo-
sure and it is hoped that these results will inform CD patients of
the need to re-evaluate their diets under the guidance of a clini-
cian or dietician and provide a guide for drug development for
this autoimmune disease, for which no effective drug therapy
exists.
Adult CD patients evidently consume, on average, potentially
unsafe levels of gluten while on a GFD (Tables 1 and 2). Mean
daily consumptions for adults were determined to be 244 mg
(stool analysis), 363 mg (urine analysis) and >228 mg (ALV003-
1221 trial analysis), with the latter value likely to be greater than
the former values. There is a general consistency in these analy-
ses and the potentially higher value for the latter case may reect
that this trial enrolled a more symptomatic CD population. The
fact that the placebo patients improved histologically in a 12-wk
trial due to a trial effect substantiates that they were consuming
signicant amounts of gluten before the trial. The stool and urine
analyses were also conducted on a population of children, and al-
though they generally consumed less than adults, the mean gluten
consumption may still be regarded as above the recommended
level of gluten ingestion for patients with CD (Tables 1 and 2and
Figures 2 and 3).
It should be noted that there is considerable variation in GIP
concentration in stool and urine that may impact the accuracy of
these analyses. A single ingestion of 0.5 g gluten showed a GIP
concentration in the rst urine in the morning that varied from
undetectable (only 1 out of 18 patients) to 41.9 ng/mL. Differ-
ences in GIP concentration in stool of volunteers ingesting the
same amount of gluten (9 or 30 g) were found (18). Variation of
excreted GIP from identical gluten intake could be due to the in-
teraction with other ingested food, the time from gluten ingestion,
the glutenase activity of the microbiome, the intestinal motility,
differential amount of digestive juices and enzymes, etc. In urine,
the variation may also be affected by the amount of ingested wa-
ter and the leakiness of the intestine. The differential modication
of the GIP by deamidation by transglutaminase might also con-
tribute to the variability of the results for single gluten ingestion.
There is, however, signicant correlation between the amount of
gluten ingested and the gluten excreted by either stool or urine,
and the sample sizes are sufciently large to statistically average
over these variabilities. However, it should be recognized that the
stool and urine tests are relatively new and the methods continue
to be improved.
Distribution plots for the frequency of occurrence at various
gluten ingestion levels showed that although many, if not most,
CD patients are able to control their gluten intake reasonably well,
others cannot. These analyses were not able to distinguish be-
tween individuals who are diligent or not diligent at maintaining
their GFD. For example, a low value may reect a diligent per-
son who consistently consumes a little gluten, or a less diligent
person who happened to consume a little gluten on the day of
the measurement. Regardless, the frequent measurement of high
gluten consumption indicates that a reasonable fraction of the
CD population have difculty controlling their GFD. Data from
Figures 2 and 3show that depending on the subclass of CD pa-
tients, anywhere from 3% to 19% of patients consume >600 mg
gluten on a daily basis. The ALV003-1221 trial data suggests that
individuals on a GFD cannot avoid accidental gluten intrusions
and these small amounts are sufcient to trigger severe symp-
tomatic responses and may contribute to histologic damage.
We thank Chaitan Khosla for encouraging us and reviewing this work and
Philip Lavin for valuable assistance on the statistical analysis.
The authors’ responsibilities were as follows: JAS and ACR: designed the
research; ACR, RD, FL, and JAS: conducted the research and provided es-
sential materials; JAS, CPK, and JAS-V: analyzed the data; JAS, CPK, and
MAD: wrote the paper: JAS, CPK, and ACR: had primary responsibility for
nal content; and all authors: read and approved the nal manuscript. JAS,
MAD, and JAS-V own stock in ImmunogenX; CPK has received research
support from Aptalis and has served on advisory boards of ImmunogenX,
Celimmune, Cour Pharmaceuticals, Innovate, and Takeda Pharmaceuticals;
ACR and FL own stock in Biomedal, Seville, Spain and Celimmune LLC.
The remaining author had no conicts of interest to declare.
Downloaded from https://academic.oup.com/ajcn/article-abstract/107/2/201/4911450
by PPD Development LP user
on 27 February 2018
GLUTEN CONSUMPTION IN CELIAC DISEASE PATIENTS 207
REFERENCES
1. Rubio-Tapia A, Murray JA. Celiac disease. Curr Opin Gastroenterol
2010;26:116–22.
2. Schuppan D, Junker Y, Barisani D. Celiac disease: from pathogenesis
to novel therapies. Gastroenterology 2009;137:1912–33.
3. Al-Bawardy B, Codipilly DC, Rubio-Tapia A, Bruining DH, Hansel
SL, Murray JA. Celiac disease: a clinical review. Abdom Radiol
2017;42:351–60.
4. Hoppe C, Gøbel R, Kristensen M, Lind MV, Matthiessen J, Christensen
T, Trolle E, Fagt S, Madsen ML, Husby S. Intake and sources of gluten
in 20- to 75-year-old Danish adults: a national dietary survey. Eur J Nutr
2017;56:107–17.
5. Catassi C, Fabiani E, Iacono G, D’Agate C, Francavilla R, Biagi F, Volta
U, Accomando S, Picarelli A, De Vitis I, et al. A prospective, double-
blind, placebo-controlled trial to establish a safe gluten threshold for
patients with celiac disease. Am J Clin Nutr 2007;85:160–6.
6. US Food and Drug Administration Report, “Guidance for Industry;
Gluten-Free Labeling of Foods,” June 2014.
7. Diaz-Amigo C, Popping B. Gluten and gluten-free: issues and
considerations of labeling regulations, detection methods, and assay
validation. J AOAC Int 2012;95:337–48.
8. Sealey-Voyksner JA, Khosla C, Voyksner RD, Jorgenson JW. Novel
aspects of quantitation of immunogenic wheat gluten peptides by liquid
chromatography-mass spectrometry/mass spectrometry. J. Chromatogr
A 2010;1217:4167–83.
9. Scherf KA, Poms RE. Recent developments in analytical method for
tracing gluten. J Cereal Sci 2016;67:112–22.
10. Bethune MT, Khosla C. Parallels between pathogens and gluten
peptides in celiac sprue. PLoS Pathog 4:e34.
11. Lefer DA, Kelly CP, Green PH, Fedorak RN, DiMarino A, Perrow W,
Rasmussen H, Wang C, Bercik P, Bachir NM, Murray JA. Larazotide
acetate for persistent symptoms of celiac disease despite a gluten-free
diet: a randomized controlled trial. Gastroenterology 2015;148:1311–9.
12. Petersen J, Montserrat V, Mujico JR, Loh KL, Beringer DX, van
Lummel M, Thompson A, Mearin ML, Schweizer J, Kooy-Winkelaar
Y, et al. T-cell receptor recognition of HLA-DQ2-gliadin complexes
associated with celiac disease. Nature Struct Mol Bio 2014;21:
480–8.
13. Lähdeaho ML, Kaukinen K, Laurila K, Vuotikka P, Koivurova OP,
Kärjä-Lahdensuu T, Marcantonio A, Adelman DC, Mäki M. Glutenase
ALV003 attenuates gluten-induced mucosal injury in patients with
celiac disease. Gastroenterology 2014;146:1649–58.
14. Syage JA, Murray JA, Green PHR, Khosla C. Latiglutenase improves
symptoms in seropositive celiac disease patients while on a gluten-free
diet. Dig Dis Sci 2017;62:2428–32.
15. See JA, Kaukinen K, Makharia GK, Gibson PR, Murray JA.
Practical insights into gluten-free diets. Nat Rev Gastroenterol Hepatol
2015;12:580–91.
16. Hall NJ, Rubin GP, Charnock A. Intentional and inadvertent non-
adherence in aduct celiac disease. A cross-sectional survey. Appetite
2013;68:56–62.
17. Ford S, Howard R, Oyebode J. Psychosocial aspects of coeliac disease:
a cross-sectional survey of a UK population. Br J Health Psychol
2012;17:743–57.
18. Comino I, Real A, Vivas S, MÁ Síglez, Caminero A, Nistal E, Casqueiro
J, Rodríguez-Herrera A, Cebolla A, Sousa C. Monitoring of gluten-
free diet compliance in celiac patients by assessment of gliadin 33-mer
equivalent epitopes in stool. Am J Clin Nutr 2012;95:670–7.
19. Comino I, Fernandez-Banares F, Esteve M, Ortigosa L, Castillejo
G, Fambuena B, Ribes-Koninckx C, Sierra C, Rodriguez-Herrara A,
Salazar JC, et al. Fecal gluten peptides reveal limitations of serological
tests and food questionnaires for monitoring gluten-free diet in celiac
disease patients. Am J Gastroenterol 2016;111:1456–65.
20. Moreno ML, Cebolla A, Munoz-Suano A, Carrillo-Carrion C, Comino
I, Pizarro A, Leon F, Rodriguez-Herrera A, Sousa C. Detection of gluten
immunogenic peptides in the urine of patients with coeliac disease
reveals transgressions in the gluten-free diet and incomplete mucosal
healing. Gut 2017;66:250–7.
21. Murray JA, Kelly CP, Green PH, Marcantonio A, Wu TT, Mäki
M, Adelman DC. No difference between latiglutenase and placebo
in reducing villous atrophy or improving symptoms in patients with
symptomatic celiac disease. Gastroenterology 2017;152:787–98.
Downloaded from https://academic.oup.com/ajcn/article-abstract/107/2/201/4911450
by PPD Development LP user
on 27 February 2018
