association of vedolizumab level, anti-drug antibodies ... · bruce & ruth rappaport faculty of...
TRANSCRIPT
Accepted Manuscript
Association of Vedolizumab Level, Anti-drug Antibodies, and α4β7 Occupancy WithResponse in Patients With Inflammatory Bowel Diseases
Bella Ungar, Uri Kopylov, Miri Yavzori, Ella Fudim, Orit Picard, Adi Lahat, DanielCoscas, Matti Waterman, Ola Haj-Natour, Noam Orbach-Zingboim, Ren Mao,Minhu Chen, Yehuda Chowers, Rami Eliakim, Shomron Ben-Horin
PII: S1542-3565(17)31423-4DOI: 10.1016/j.cgh.2017.11.050Reference: YJCGH 55584
To appear in: Clinical Gastroenterology and HepatologyAccepted Date: 17 November 2017
Please cite this article as: Ungar B, Kopylov U, Yavzori M, Fudim E, Picard O, Lahat A, Coscas D,Waterman M, Haj-Natour O, Orbach-Zingboim N, Mao R, Chen M, Chowers Y, Eliakim R, Ben-HorinS, Association of Vedolizumab Level, Anti-drug Antibodies, and α4β7 Occupancy With Response inPatients With Inflammatory Bowel Diseases, Clinical Gastroenterology and Hepatology (2018), doi:10.1016/j.cgh.2017.11.050.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
1
Association of Vedolizumab Level, Anti-drug Antibodies, and α4β7 Occupancy With
Response in Patients With Inflammatory Bowel Diseases
Bella Ungar1, Uri Kopylov1, Miri Yavzori1, Ella Fudim1, Orit Picard1, Adi Lahat1,
Daniel Coscas1, Matti Waterman2, Ola Haj-Natour1, Noam Orbach-Zingboim1, Ren
Mao3, Minhu Chen3, Yehuda Chowers2, Rami Eliakim1, Shomron Ben-Horin1,3
Department of Gastroenterology, Sheba Medical Center Tel Hashomer, Sackler
School of Medicine, Tel-Aviv University, Tel Aviv1 ,Rambam Health Care Campus,
Bruce & Ruth Rappaport Faculty of Medicine, Technion - Israel Institute of
Technology, Haifa2, Israel, and The First Affiliated Hospital of Sun-Yatsen
University, Guangzhou, China3
Short title: Pharmacokinetics & pharmacodynamics of response to vedolizumab therapy
Abbreviations: AVA - anti-vedolizumab-antibodies; CD - Crohn's disease; CRP - C-
reactive protein; HBI - Harvey-Bradshaw index; IBD - inflammatory bowel disease;
SCCAI - Simple Clinical Colitis Activity Index; UC - ulcerative colitis.
Keywords: Vedolizumab; IBD; immunogenicity; clinical outcome.
Corresponding authors:
Bella Ungar, Department of Gastroenterology, Sheba Medical Center, Derech Sheba
2, Ramat-Gan, Israel, 5262100, [email protected], Tel. 972-3-530-2060, Fax
972-3-530-3160
Shomron Ben-Horin, Department of Gastroenterology, Sheba Medical Center, Derech
Sheba 2, Ramat-Gan, Israel, 5262100, [email protected], Tel. 972-3-
530-2060, Fax 972-3-530-3160
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
2
Financial support: This work was supported in part by the 7th Dr. Pinchas
Borenstein Talpiot Medical Leadership Program, The Chaim Sheba Medical Center
(to BU), the 'Talpiot' Medical Leadership Program, The Chaim Sheba Medical Center
(to SBH), the Legacy heritage foundation, Rambam Health Care Center (to YC), and
ECCO-AOCC Visiting Grant (to MR), The Leona M. and Harry B. Helmsley
Charitable Trust (to SBH, RE and YC)
Potential competing interests: Shomron Ben-Horin received consulting and advisory
board fees and/or research support from AbbVie, MSD, Janssen, Takeda and
CellTrion. Uri Kopylov received speaker fees from Abbvie, Janssen and Takeda,
research support from Takeda and Janssen and consulting fees from Takeda and CTS.
Yehuda Chowers received grant support, speaker & consultant fees from AbbVie,
speaker & consultant fees from Janssen, grant support, speaker and consultant fees
from Takeda, speaker fees from Ferring and consultant fees from Medtronics. Rami
Eliakim received consultant and speaker fees from Janssen, Abbvie, Takeda and
Medtronic. Min-hu Chen received speaker fees from Janssen, Falk, Takeda and Ipson.
Matti Waterman received consultant and speaker fees from Janssen, Abbvie, Takeda
and Medtronic. Bella Ungar received consultation fees from Janssen and Abbvie.
None of the other authors have any conflicts to declare. Sheba Medical Center and
Rambam Health Care Campus have filed intellectual property requests on the assays
for vedolizumab and AVA levels.
Author contributions: SBH conceived the study and drafted the manuscript; BU was
involved in study conception, analysis and interpretation of data and manuscript
drafting; UK, MW, DC, AL, OHN and NOZ participated in acquisition of data; EF,
MY and OP took part in data analysis; RE, YC, MR and MC participated in data
interpretation and in critical revision of the manuscript for important intellectual
property. All authors have approved the final draft submitted.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
3
Background & Aims: There are few data available on the real-life pharmacokinetic
and pharmacodynamics features of vedolizumab, a monoclonal antibody against
integrin α4β7. We performed a prospective study of patients with inflammatory bowel
diseases (IBD) treated with vedolizumab to determine serum drug concentrations,
formation of anti-vedolizumab antibodies (AVAs), and integrin α4β7 saturation.
Methods: We performed a prospective study of 106 patients with IBD (67 with
Crohn’s disease and 39 with ulcerative colitis) treated with vedolizumab from
September 2014 through March 2017 at 2 tertiary medical centers in Israel. Clinical
data and serum samples were collected before and during induction and maintenance
therapy. Clinical remission was defined as Harvey-Bradshaw index scores below 5 or
as Simple Clinical Colitis Activity Index scores of 3 or less. We measured serum
levels of vedolizumab, AVAs, and markers of inflammation. Peripheral blood
mononuclear cells were obtained from some patients at designated trough timepoints
and CD3+ CD45RO+ T cells were isolated from 36 samples. Cells were incubated
with fluorescent-conjugated vedolizumab and flow cytometry was used to quantify
α4β7 integrin saturation. We also performed flow cytometry analyses of CD3+
CD45RO+ lamina propria T cells isolated from intestinal mucosa of patients without
IBD (non-IBD controls, n=6), patients with IBD not treated with vedolizumab
(untreated IBD controls, n=8) and patients with IBD treated with vedolizumab (n=15).
Results: Clinical remission was achieved by 48/106 patients (45%) by week 6 and
50/106 patients (48%) by week 14 of treatment. The median level of vedolizumab at
week 6 was higher in patients in clinical remission (40.2 µg/mL) than in patients with
active disease (29.7 µg/mL; P=.05). The median serum level of vedolizumab was
significantly higher in patients with normal level of c-reactive protein (21.8 µg/mL
vedolizumab) vs the level in those with a high level of c-reactive protein (11.9 µg/mL
vedolizumab) during maintenance treatment (P=.0006). The other clinical outcomes
measured were not associated with median serum level of vedolizumab at any time
point examined. AVAs were detected in 17% of patients during induction therapy and
3% of patients during maintenance therapy, but did not correlate with clinical
outcomes. Flow-cytometry analysis of peripheral blood memory T cells (n=36)
showed near-complete occupancy of α4β7 integrin at weeks 2 and 14 and during
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
4
maintenance phase, regardless of response status or drug levels. Most intestinal
CD3+CD45RO+ memory T cells of healthy and IBD controls expressed α4β7 (72%;
inter-quartile range, 56%–81%). In contrast, free α4β7 was detectable on only 5.6% of
intestinal memory cells (inter-quartile range, 4.4%–11.2%) (P<.0001) from
vedolizumab-treated patients, regardless of response.
Conclusion: In a prospective study of real-life patients with IBD, we associated
vedolizumab drug levels with remission and level of a marker of inflammation.
Integrin α4β7 was blocked in almost all T cells from patients treated with
vedolizumab, regardless of serum level of the drug or response to treatment. These
findings indicate a need to explore alternative mechanisms that prevent response to
vedolizumab.
KEY WORDS: trough level; CD; UC; immunogenicity; clinical response
Introduction
The leukocyte α4β7 integrin facilitates leukocyte homing to the gut via binding of
MADCAM1 on intestinal endothelium1-3. Vedolizumab is a monoclonal antibody
(mAb) against α4β7 which is efficacious for patients with moderate to severe CD and
UC4, 5. Whereas higher levels of anti-TNFs, both infliximab and adalimumab, were
associated with clinical remission and mucosal healing in UC and CD6-9, data
regarding pharmacokinetics and pharmacodynamics of vedolizumab are still scarce.
The GEMINI trials have demonstrated that vedolizumab drug levels were positively
associated with clinical response at week 6. However, drug levels have not been
reported for later time-points1011, and it is still unknown whether therapeutic drug-
monitoring (TDM) of vedolizumab could improve patient management. The
regulatory submission of vedolizumab asserts its mechanism of action is mediated by
blockade of α4β7 on target cells, predominantly memory T-cells, thereby inhibiting
these cells from binding to MADCAM1 and migrating into inflamed intestinal
tissues12. In the GEMINI program, integrin receptor saturation on target effector cells
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
5
was demonstrated with high (6mg/kg) and low (2mg/kg) dosing regimens of
vedolizumab, but correlation of target occupancy with individual patients’ response
status was not shown13. Moreover, there are hitherto no data on α4β7 receptor
occupancy in the gut. The present study therefore aimed to assess, in a large
prospective real-life cohort, the relationship between vedolizumab levels, clinico-
biologic end-points and α4β7 target occupancy in the blood and the intestine.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
6
Methods
Patient population
This was a prospective observational study of IBD patients receiving scheduled
vedolizumab therapy between September 2014 and March 2017 at two tertiary
medical centers in Israel: Sheba medical center and Rambam Health Care Campus.
All patients who started scheduled vedolizumab therapy during the study period were
enrolled after providing an informed consent. Patients' clinical characteristics were
recorded before drug initiation. Drug and AVA serum levels, as well as inflammatory
markers, were measured in pre-infusion sera collected prospectively throughout
induction and maintenance therapy, and clinical scores were determined before each
infusion. Baseline colonoscopy was performed by physician discretion and was not
universally employed before vedolizumab initiation. In a subset of patients, peripheral
blood mononuclear cells (PBMC), obtained at the designated trough timepoints, and
lamina propria lymphocytes (LPLs), obtained during lower endoscopy, were also
collected. The study was approved by the medical centers' ethics committees and all
patients gave a written informed consent.
Clinical Scores
Clinical status was determined before each infusion prospectively by HBI (Harvey-
Bradshaw index) for CD and by SCCAI (Simple Clinical Colitis Activity Index) for
UC patients15,16. Clinical remission was defined as HBI <5 or as SCCAI≤ 3,
respectively. Clinical response was defined as drop of ≥3 points of the HBI and
SCCAI scores17. Primary non-response was defined as cessation of vedolizumab
therapy by week 14, due to lack of clinical response as defined above18. Secondary
loss of response was defined as clinical worsening necessitating drug-discontinuation
or additional intervention during maintenance period31.
Development of assay for measurement of vedolizumab concentration
Integrin alfa4-beta7 (2µg/ml, R&D, MN, USA) was added to pre-plated anti-His-tag
(4µg/ml, R&D, MN, USA) wells of ELISA plates (Nunc, Roskilde, Denmark). After
drying, 100µl of 1:1000 diluted serum was added and incubated for 60 min at room
temperature. Plates were then washed and goat anti-human κ chain HRP-labeled
antibody (Serotec, Oxford, UK) was added at a concentration of 66 ng/ml for 40 min.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
7
The results were read by an ELISA reader EL-800 (Biotek Instruments, Winooski,
USA) and expressed as µg/ml. The Level of detection (LoD) of the assay was 3ng/ml
and level of quantitation (LoQ) was 3μg/ml. For assay validation, graded
vedolizumab concentrations were added to vedolizumab-negative sera, demonstrating
significant correlation of signal optical density (OD) with the concentrations added
(rho=1, p<0.0001, Supplementary figure 1).
Determination of anti-vedolizumab antibodies (AVA) concentration
A drug-tolerant ELISA assay using an anti-human λ chain-conjugated detector
antibody was developed. Briefly, 100µl of 1:500 diluted serum was added to pre-
plated 5µg/ml vedolizumab (Takeda Pharma, Denmark) and incubated for 60 min.
Following washing, horseradish peroxidase (HRP) labelled goat anti-human λ-chain
antibody (MP Biomedical, Solon, Ohio, USA) at a concentration of 33 ng/ml was
added for 40 min and reacted with tetramethylbenzidine (TMB) substrate. The results
were read by an ELISA reader EL-800 (Biotek Instruments, Winooski, USA) and
expressed as µg/ml-equivalent (µg/ml-eq) after normalization versus graded
concentrations of 9-1200ng/ml of goat anti-human F(ab′)2 fragment antibody (MP
Biomedicals, USA). AVA levels above 30μg/ml-eq cut-off were considered positive,
as determined by three standard deviations above the mean of 30 unexposed controls.
Fluorescence-activated cell sorting (FACS) analysis of α4β7 occupancy on T-cells
Freshly obtained PBMC and lamina propria (LP) lymphocytes were isolated and
prepared as previously described19. Vedolizumab was biotinylated (EZ-LinkTM
NHS-PEO solid Phase biotinylation kit, Thermo Scientific, USA). Cells were co-
stained with CD3 T-cell marker, CD45RO as a marker of previously-activated cells
(surrogate for memory cells) and with 30mcg/ml of biotinylated-vedolizumab-
strepavidine-PE or a biotinylated-IgG control antibody. The expression of α4β7
(vedolizumab-positive cells) on gated PB or LP CD3+CD45RO+ memory T-cells was
determined by FACS analysis (Navios Flow cytometer, Beckman Coulter, USA) and
results analyzed by Kaluza software (Beckman Coulter, USA). To examine the drug
concentration that completely block all free membranal α4β7, graded concentrations
of vedolizumab were added to PBMC, cultured 5 days in media alone or after
polyclonal stimulation with 5mcg/ml Muromonab-CD3 (OKT3).
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
8
Statistical analysis
Continuous variables were compared by Mann-Whitney test and Fischer's exact test
was used for categorical data. Wilcoxon test was used to compare paired samples. A
receiver-operating characteristic (ROC) analysis was performed for analysis of
vedolizumab levels discriminatory accuracy for categorical variables. Correlations
were analyzed by the Spearman rank test. One-way analysis of variance (ANOVA)
and Cochran's Q test were used to test the differences between quartiles. Multivariable
analysis was performed using backward logistic regression. All P-values were 2-
sided, and a P-value <0.05 was considered statistically significant. All statistics were
performed with MedCalc software (version 12.2.1.0, Mariakerke, Belgium).
Results
Demography and clinical outcomes
124 patients started vedolizumab treatment during the study period; 18 patients were
excluded due to missing sera, non-consent, pregnancy, or cessation of vedolizumab
before the third infusion due to reasons unrelated to clinical non-response. Thus, the
study cohort consisted of 106 IBD (67 CD, 39 UC) patients. The median follow-up
for the study cohort was 30 weeks (IQR 14-54 weeks). The patients' clinical and
demographic characteristics are depicted in table 1. Clinical remission by weeks 6 and
14 were achieved by 48/106 patients (45%) and 50/106 (48%), respectively. Nine
patients (8.4%) experienced primary non-response to induction and 17 (18%) lost
response and discontinued vedolizumab therapy during maintenance; 12 patients
(11.3%) underwent interval shortening during induction (week 10) and 11 patients
(11.5%) during maintenance. Fifty-four patients completed one year of therapy, 17 of
whom (32%) were in clinical remission.
Baseline factors association with pharmacokinetics
Baseline albumin and CRP values significantly correlated with weeks 2, 6 and 14
vedolizumab trough levels (supplementary table 1) and albumin level quartiles were
associated with increased week 14 vedolizumab trough levels (p<0.001, Figure 1). In
addition, female gender, lower baseline weight and younger age were associated with
higher levels at week 2. Female gender was significantly associated with week 6
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
9
levels as well. On multivariate analysis, only baseline albumin remained significantly
correlated with both week 6 and 14 levels (supplementary table 1).
Association of induction period drug levels and clinical and biomarker response
Median week 6 vedolizumab levels were higher in week 6 clinical remission versus
non-remitters patients (40.2µg/mL, IQR 18.3-57.2, versus 29.7µg/mL, IQR 17.5-41.8,
respectively, p=0.05, figure 2a). On quartile analysis, a statistically significant
difference in remission rate was noted between week 6 drug-level-quartiles 2, 3 and 2,
4 (p=0.02, 0.006, figure 2b). However, drug levels were not associated with remission
status at other time-points (supplementary table 2). Week 6 drug levels in partially-
responding patients requiring interval-shortening at week 10 (45.4µg/mL, IQR 19.6-
84.2, n=12) were similar to levels in patients whose intervals were not shortened
(33.9µg/mL, IQR 18.3-48, p=0.28). A sub-analysis including only patients who did
not receive steroids during induction (n=71) similarly showed significant association
of vedolizumab levels with clinical remission only for week 6 (median 41.3µg/mL,
IQR 19.8-40.5, versus 29.8µg/mL, IQR 24.6-53.4, p=0.04).
In an analysis of biomarker response, week 2 vedolizumab levels demonstrated a
significant, albeit weak, negative correlation with week 14 CRP values (rho=-0.25,
p=0.02) and week 6 drug levels were weakly associated with week 6 CRP values
(rho=-0.24, p=0.04, figure 2c). No association was detected between drug levels and
CRP normalization for other time-points (supplementary table 3), and clinical
remission was not predicted by any of the other clinical/demographic factors
(supplementary table 5). A separate sub-group analysis for PK-response correlation
for CD (n=67) and UC patients (n=39) showed similar findings to those of the entire
cohort (supplementary document 1).
Maintenance period drug levels association with clinical-biomarker response
Analysis of 140 serial sera obtained from 60 consecutive patients during vedolizumab
maintenance therapy (week 22 onward) exhibited significantly higher median
vedolizumab levels at time-points with normal versus elevated CRP (21.8µg/mL, IQR
12.9-40.5, versus 11.9µg/mL, IQR 5-26.5, respectively, p=0.0006, figure 3). In
contrast, median levels in clinical responders did not differ from levels in non-
responders (15.9µg/mL, IQR 7.7-25 versus 14µg/mL, IQR 5.3-31.9, respectively,
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
10
p=0.57). Drug levels in patients before interval-shortening were similar to levels in
patients not requiring this intervention, matched for the same time-point (27.8µg/mL,
IQR 11.4-45, versus 15.4µg/mL, IQR 7.1-26.2, p=0.09), and intra-individual temporal
variations in vedolizumab levels appeared similar in responders versus those with
loss-of-response (supplementary figures 3a, 3b). Furthermore, week 6 levels among
patients who went on to have sustained response during maintenance were not
different than levels in those with subsequent loss-of-response (35.7µg/mL, IQR 19-
48.5 vs. 31.5µg/mL, IQR 16.5-53, respectively, p=0.98).
Immunogenicity
Anti-vedolizumab-antibodies (AVA) were measured in 180 sera. Of these, 120 were
consecutive (week 2-6-14) samples of 41 patients during induction (21 with week 14
clinical remission , 20 clinically active). Additional 60 sera obtained from 60
consecutive patients during maintenance therapy were also analyzed (30 of whom
were clinically active at the time of sampling). Positive antibodies at weeks 2 or 6
were detected in 7/41 patients (17%): 3/21 (14.3%) responders and 4/20 (20%) of the
non-responders (p=0.63, OR=1.5, 95% CI 0.3-7.7). AVA were detected with zero
drug levels in one patient and with measurable drug levels (i.e. drug-positive/AVA-
positive) in the other six, with the use of the described drug-tolerant assay. Only 2/60
(3%) patients during maintenance therapy were found AVA positive (One AVA-
positive/drug-positive in a patient in remission, one AVA-positive/drug-negative in a
patient with loss of response).
Intestinal and peripheral blood α4β7 integrin target occupancy in relation to
drug levels and clinical outcome
Peripheral blood lymphocytes (PBL) were prospectively collected at weeks 0, 2 and14
in 13 patients starting vedolizumab therapy (six responders, seven non-responders to
induction). The percentage of free (unbound) α4β7 on CD3+CD45RO+ memory T-
cells was determined by FACS analysis. Median free α4β7 (positive α4β7 CD45RO+
cells) at week 0 was 27% (IQR 18%-31%), and sharply declined to near-complete
target occupancy at week 2 (0.09%, IQR 0.03-0.59%) and week 14 (2%, IQR 0.77%-
3.4%, P<0.0001 for both comparisons, figures 4a, 4b). Near-complete target
occupancy at weeks 2 and 14 was universal and unrelated to response/non-response to
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
11
induction or to drug levels at week 14 (Figure 4b). The percentage of positive α4β7 T-
cells before vedolizumab initiation (day 0), did not predict response to induction in
the present cohort (n=13), nor in an additional exploratory cohort (n=26, data not
shown). Similar to the findings during induction, near-complete α4β7 occupancy was
also found on T-cells obtained during maintenance phase (n=23), regardless of
whether patients were in clinical response (n=12) or loss-of-response (n=11, figure
4b). Similar near-complete blockade of α4β7 was also found for naïve T-cells
(CD3+CD45RO-, data not shown). Given these results, the minimal concentration of
vedolizumab able to block PB T-cells α4β7, in vitro, was explored (see Methods
section). A near-complete blockade of α4β7 (<2% free α4β7) was apparent already
with concentrations of 3μg/ml of vedolizumab in unstimulated resting cells, whereas
stimulated cells had higher baseline expression of α4β7 and required somewhat
higher vedolizumab concentrations to achieve comparable target occupancy (figure
4c).
Finally, mucosal samples were obtained during colonoscopies in control healthy
individuals, IBD controls (not treated with vedolizumab) and IBD patients treated
with vedolizumab (n=29), and LP T-cells were isolated and stained as above.
CD3+CD45RO+ LP-T cells in both healthy and IBD control individuals were
predominantly α4β7positive, regardless of controls’ disease activity state (median
71%, IQR 55.6-81%, n=14). In contrast, in vedolizumab-treated patients, free α4β7 on
intestinal memory cells was significantly lower (median CD3+CD45RO+α4β7+
5.6%, IQR 4.4-11.2%, n=15, p<0.0001, figures 4d, 4e). Intestinal α4β7 occupancy in
vedolizumab-treated patients was similar in inflamed and non-inflamed mucosa
(5.6%, IQR 4.4-11.2, n=15, versus 4.8%, IQR 3.8-11.8, n=9, respectively, p=0.7,
supplementary table 6). There was also no difference in the rate of free α4β7+ on T-
cells among responders and non-responders to therapy (data not shown).
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
12
Discussion
Vedolizumab is efficacious for moderate-to-severe CD and UC4, 20-22. Nevertheless,
scarce data exist pertaining to its pharmacokinetics/pharmacodynamics and data on its
target occupancy in the gut are completely absent. This study showed a modest
association of vedolizumab drug levels with some outcomes of clinical and biomarker
response to therapy. It further showed a nearly-full target α4β7 receptor occupancy on
both peripheral blood and intestinal memory T-cells, within a wide range of drug
concentrations and regardless of response to therapy or not.
Similar to the GEMINI trials23, we also found body weight and albumin to correlate
with induction drug concentration and identified female gender as additional predictor
– possibly due to the lower average weight of females. However, the only predictor
retained on multi-variable analysis was albumin, whose correlation with drug levels
was not only found for extreme albumin values, unlike the GEMINI analysis. This
may be related to differences in FcRn salvage24 and/or in fecal loss of albumin/IgG,
which may be different for the present population compared to the GEMINI trial
population.
In the GEMINI 1 & 2 studies, vedolizumab drug level quartiles at week 6 were
positively associated with week 6 clinical response10 11. In the present cohort, week 6
levels were similarly higher in week 6 responders compared to non-responders,
regardless of concomitant steroid therapy. However, clinical outcomes at other time-
points were not predicted by week 2 or 6 levels. This is echoed in a recently published
analysis of GEMINI trial, in which no definite vedolizumab threshold level could be
determined for discriminating remitters from non-remitters after consideration of co-
variates25. This could suggest some detected drug-response associations may arise
from multiple-testing, implying further corroborating studies are imperative.
Interestingly, a recent real-life study similarly did not find different vedolizumab
levels between responders and non-responders, although week 6 levels<19µg/mL
modestly associated with subsequent need for interval-shortening26. We could not
reproduce such a correlation, possibly because of different cohort characteristics or
size (106 patients in the present cohort versus 47 in the previously reported one), or
due to variations in assay techniques or outcome definitions.
Vedolizumab levels were associated with biomarker response (CRP normalization),
especially during maintenance treatment. Data on maintenance-phase drug levels and
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
13
outcomes are still scant, but the present observations are in line with recent data from
another real-world cohort, presented in an abstract form27. Immunogenicity rates of
vedolizumab were 17% during induction and 3% during maintenance, and were in
agreement with reported low incidence of AVA and with their limited clinical
impact27, 28.
As concluded by FDA and EMA, the mechanism of action of vedolizumab is
blockade of α4β7, predominantly on memory T-cells, to inhibit their MADCAM1-
binding mediated migration into intestinal tissue12. Earlier work showed that even
2mg/kg vedolizumab dose provided near-complete saturation of the integrin-
receptor29, thereby suggesting the need to elucidate the pharmacodynamics impact of
fluctuations of vedolizumab exposure within the current dosing scheme. Recently,
higher baseline mRNA expression of integrin αΕβ7, and its decline after therapy,
were correlated with response to etrolizumab, an anti-β7 integrin antibody30.
However, the rate of α4β7 occupancy with this agent was not reported. The present
findings show target membranal α4β7-integrin on peripheral blood memory T-cells to
be nearly-fully occupied during both induction and maintenance with vedolizumab
and regardless of clinical response status. Whether these results indicate blocking of
α4β7 by vedolizumab or post-binding internalization of α4β7-vedolizumab complex is
still unclear, but even the later implies a near-complete binding of α4β7 by
vedolizumab before putative internalization. Similar to findings in blood, near-full
receptor saturation was demonstrated - for the first time - also on activated T-cells in
the intestinal mucosa, indicating in-situ blockade of T-cells by vedolizumab and/or
the effective prevention of these cells from migration to the gut. Full receptor
occupancy was evident throughout a wide-range of drug concentrations and was
unrelated to response status, thereby arguing against sub-dosed vedolizumab (being
unable to block its target in full) as the cause for suboptimal response in non-
responders. Conversely, however, these data may hint towards other mechanisms
underlying the biologic effect of vedolizumab or its lack-of, such as receptor
occupancy on other T-cell subsets32, which may hypothetically require higher drug
concentration for optimized effect.
There are several limitations to our study. Vedolizumab levels were measured using
an in-house novel validated ELISA-based assay; Thus, corroborating studies using
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
14
other assays are pertinent. Nonetheless, the vedolizumab concentrations herein are in
agreement with those reported in the GEMINI trials, and a high accuracy for detecting
exogenous vedolizumab was shown (supplementary figure 1), thus supporting the
validity of the observations. Secondly, various clinical-biomarker end-points in
relation to drug levels were assessed, but endoscopic outcomes were not evaluated.
In conclusion, in this real-life prospective pharmacokinetic cohort of vedolizumab-
treated IBD patients, drug levels were associated with some – albeit not all - clinical
outcomes of induction and with CRP normalization during maintenance. Peripheral
blood and intestinal target α4β7-integrin is nearly-fully occupied by vedolizumab
throughout wide-range of drug concentrations and regardless of therapy. These
pharmacokinetic-pharmacodynamic data suggest that within current dosing schemes,
drug levels in most non-responders are still sufficient to block the α4β7 target
molecule, indicating the need to explore additional modes-of-action of vedolizumab
and to elucidate other mechanisms underlying non-response to the drug.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
15
Table 1 - Patients' Demographic and Clinical Characteristics
N 106 Age, years (median, IQR) 37 (29.3-46.8) Disease duration, years (median, IQR) 10.5 (6-17) Male / Female ratio 1.2 Smoking at induction, n (%) 15 (14) Jewish Ashkenazi ethnicity, n (%) 41 (39) Previous surgery, n(%) 35 (33) Concomitant medical condition, n (%) 23 (21) BMI (median, IQR) 22 (19.6-25.3) Extra-intestinal manifestations, n (%) 36 (34) Concomitant immunomodulator therapy, n (%) 15 (14) Concomitant steroids at baseline 33 (31) CD, n (%) 67 (63) UC, n (%) 39 (37) CD behavior Inflammatory (%) 27 (45)
Stricturing (%) 18 (30) Penetrating (%) 15 (25)
CD location Ileal (%) 34 (50) Ileo-colonic (%) 22 (33) Colonic (%) 11 (17)
Perianal CD (%) 20 (33.3)
UC location* Left sided colitis (%) 18 (47)
Pancolitis (%) 20 (52) Failure of anti-TNFα therapy, n (%) 91 (86) Baseline albumin level (g/dl, median, IQR) 3.9 (3.6- 4.3) Baseline CRP level (mg/l, median, IQR) 9 (4.3 - 23.7) Vedolizumab week 2 trough serum level, µg/mL (median, IQR)
42.7 (28.9 - 60.7)
Vedolizumab week 6 trough serum level, µg/mL (median, IQR)
33.8 (17.9 - 48)
Vedolizumab week 14 trough serum level, µg/mL (median, IQR)
11.4 (4.6 - 20.4)
Vedolizumab week 22 trough serum level, µg/mL (median, IQR)
16.9 (7.5-26.4)
Vedolizumab week 54 trough serum level, µg/mL (median, IQR)
15 (8.4-26.2)
IBD -Inflammatory bowel disease, CD - Crohn's disease, UC - ulcerative colitis, TNF - Tumor necrosis factor, IQR - interquartile range, CRP - C-reactive protein * One UC patient received vedolizumab induction therapy after proctocolectomy and resultant pouchitis.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
16
Figure Legends
Figure 1
Association between baseline albumin level quartiles and week 14 vedolizumab trough levels (n=106).
Figure 2a
Week 6 vedolizumab levels among patients in clinical remission versus clinically active patients at week 6.
Figure 2b
Vedolizumab trough level quartiles at week 6 in relation to clinical remission rates at the same time-point.
Figure 2c
Correlation between week 6 vedolizumab levels and CRP values. CRP - C-reactive protein
Figure 3
Maintenance-phase vdolizumab levels among 60 patients with normal versus elevated CRP values. CRP - C-reactive protein
Figure 4a
FACS analysis of free α4β7-integrin on peripheral blood memory T-cells
(α4β7+CD45RO+CD3+ cells, out of total CD45RO+CD3+ T-cells), detected by staining with conjugated-vedolizumab at weeks 0-2-14. Exemplary four patients are depicted with their respective induction’s clinical outcome and drug levels. FACS - flow cytometry and fluorescence-activated cell sorting
Figure 4b
Graphic summary of the percentage of free α4β7 on memory T-cells (α4β7+ CD45RO+CD3+) in peripheral blood of vedolizumab-treated patients (separately presented for responders and non-responders). Induction and maintenance period observations are depicted. Each point denotes a single patient, except during induction when two points connected with a line represent a single patient at the two designated timepoints. The color of the connecting line denotes if a patient’s week 14 level vedolizumab was low (‘Low vedo’ - defined as level below the median value at week 14 of entire 106 patients’ cohort, red line) or high (‘high vedo’ - above median, blue line). vedo – vedolizumab. W- week.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
17
Figure 4c
Results of experiments (n=3) of drug concentration- α4β7 blockade titration curve. Unstimulated peripheral blood T-cells were cultured with graded concentrations of un-conjugated vedolizumab in media alone or after polyclonal stimulation by OKT3. Staining with conjugated vedolizumab was performed after five days in culture, to delineate the drug concentration that confers complete receptor occupancy. FACS - flow cytometry and fluorescence-activated cell sorting
Figure 4d
FACS analysis of intestinal memory LP T-cells free α4β7 integrin
(α4β7+CD45RO+CD3+ cells, out of total CD45RO+CD3+ T-cells), determined by staining with conjugated-vedolizumab. Two controls were employed: An IgG-isotype control and a pre-incubation with un-conjugated 50mcg/ml vedolizumab. Exemplary three experiments are shown.
Figure 4e
Box-and-Whisker plot depicting the percentage of α4β7+ on LP-memory T-cells
(α4β7+CD45RO+CD3+ cells, out of total CD45RO+CD3+ T-cells) in intestinal biopsies of control (n=14) or vedolizumab-treated patients (n=15).
Supplementary figure 1
Correlation between graded concentrations of exogenously-added vedolizumab to unexposed serum and ELISA reading by optical density (OD).
Supplementary figure 2a
ROC analysis of vedolizumab levels among CD patients, with clinical remission as a classification variable. CD - Crohn's disease, ROC - receiver-operating characteristic
Supplementary figure 2b
ROC analysis of vedolizumab levels among UC patients, with clinical remission as a classification variable. UC - ulcerative colitis, ROC - receiver-operating characteristic
Supplementary figure 3a
Intra-individual vedolizumab levels during maintenance therapy among responders.
Supplementary figure 3b
Intra-individual vedolizumab levels among patients who lost response to maintenance therapy.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
18
References
1. Selby WS, Janossy G, Bofill M, et al. Intestinal lymphocyte subpopulations in
inflammatory bowel disease: an analysis by immunohistological and cell isolation
techniques. Gut 1984;25:32-40.
2. Thomas S, Baumgart DC. Targeting leukocyte migration and adhesion in Crohn's
disease and ulcerative colitis. Inflammopharmacology 2012;20:1-18.
3. Streeter PR, Berg EL, Rouse BT, et al. A tissue-specific endothelial cell molecule
involved in lymphocyte homing. Nature 1988;331:41-6.
4. Shelton E, Allegretti JR, Stevens B, et al. Efficacy of Vedolizumab as Induction
Therapy in Refractory IBD Patients: A Multicenter Cohort. Inflamm Bowel Dis
2015;21:2879-85.
5. Dulai PS, Singh S, Jiang X, et al. The Real-World Effectiveness and Safety of
Vedolizumab for Moderate-Severe Crohn's Disease: Results From the US VICTORY
Consortium. Am J Gastroenterol 2016.
6. Kopylov U, Seidman E. Predicting durable response or resistance to antitumor
necrosis factor therapy in inflammatory bowel disease. Therap Adv Gastroenterol
2016;9:513-26.
7. Yanai H, Lichtenstein L, Assa A, et al. Levels of drug and antidrug antibodies are
associated with outcome of interventions after loss of response to infliximab or
adalimumab. Clin Gastroenterol Hepatol 2015;13:522-530 e2.
8. Ungar B, Levy I, Yavne Y, et al. Optimizing Anti-TNF-alpha Therapy: Serum Levels of
Infliximab and Adalimumab Are Associated With Mucosal Healing in Patients With
Inflammatory Bowel Diseases. Clin Gastroenterol Hepatol 2016;14:550-557 e2.
9. Paul S, Del Tedesco E, Marotte H, et al. Therapeutic drug monitoring of infliximab
and mucosal healing in inflammatory bowel disease: a prospective study. Inflamm
Bowel Dis 2013;19:2568-76.
10. Sandborn WJ, Feagan BG, Rutgeerts P, et al. Vedolizumab as induction and
maintenance therapy for Crohn's disease. N Engl J Med 2013;369:711-21.
11. Feagan BG, Rutgeerts P, Sands BE, et al. Vedolizumab as induction and maintenance
therapy for ulcerative colitis. N Engl J Med 2013;369:699-710.
12. Agency EM. Assessment report - Entyvio, 2014.
13. Rosario M, Wyant T, Leach T, et al. Vedolizumab Pharmacokinetics,
Pharmacodynamics, Safety, and Tolerability Following Administration of a Single,
Ascending, Intravenous Dose to Healthy Volunteers. Clin Drug Investig 2016;36:913-
923.
14. Ungar B, Chowers Y, Yavzori M, et al. The temporal evolution of antidrug antibodies
in patients with inflammatory bowel disease treated with infliximab. Gut
2014;63:1258-64.
15. Higgins PD, Schwartz M, Mapili J, et al. Patient defined dichotomous end points for
remission and clinical improvement in ulcerative colitis. Gut 2005;54:782-8.
16. Harvey RF, Bradshaw JM. A simple index of Crohn's-disease activity. Lancet
1980;1:514.
17. Vermeire S, Schreiber S, Sandborn WJ, et al. Correlation between the Crohn's
disease activity and Harvey-Bradshaw indices in assessing Crohn's disease severity.
Clin Gastroenterol Hepatol 2010;8:357-63.
18. Papamichael K, Rivals-Lerebours O, Billiet T, et al. Long-Term Outcome of Patients
with Ulcerative Colitis and Primary Non-response to Infliximab. J Crohns Colitis
2016;10:1015-23.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
19
19. Lim KJ, Lee SJ, Kim S, Lee SY et al. Comparable Immune Function Inhibition by the
Infliximab Biosimilar CT-P13: Implications for Treatment of Inflammatory Bowel
Disease. J Crohns Colitis. 2017 May 1;11(5):593-602
20. Amiot A, Grimaud JC, Peyrin-Biroulet L, et al. Effectiveness and Safety of
Vedolizumab Induction Therapy for Patients With Inflammatory Bowel Disease. Clin
Gastroenterol Hepatol 2016.
21. Baumgart DC, Bokemeyer B, Drabik A, et al. Vedolizumab induction therapy for
inflammatory bowel disease in clinical practice--a nationwide consecutive German
cohort study. Aliment Pharmacol Ther 2016;43:1090-102.
22. Dulai PS, Singh S, Jiang X, et al. The Real-World Effectiveness and Safety of
Vedolizumab for Moderate-Severe Crohn's Disease: Results From the US VICTORY
Consortium. Am J Gastroenterol 2016;111:1147-55.
23. Rosario M, Dirks NL, Gastonguay MR, et al. Population pharmacokinetics-
pharmacodynamics of vedolizumab in patients with ulcerative colitis and Crohn's
disease. Aliment Pharmacol Ther 2015;42:188-202.
24. Pyzik M, Rath T, Lencer WI, Baker K, Blumberg RS. FcRn: The Architect Behind the
Immune and Nonimmune Functions of IgG and Albumin. J Immunol. 2015 May
15;194(10):4595-603.
25. Rosario M, French JL, Dirks NL, et al. Exposure-Efficacy Relationships for
Vedolizumab Induction Therapy in Patients with Ulcerative Colitis or Crohn's
Disease. J Crohns Colitis 2017.
26. Williet N, Boschetti G, Fovet M, et al. Association Between Low Trough Levels of
Vedolizumab During Induction Therapy for Inflammatory Bowel DiseasesWith Need
for Additional Doses Within 6 months. Clin Gastroenterol Hepatol 2016.
27. Severine Vermeire (1) AG, Erwin Dreesen (2), Griet Compernolle (2), Miet Peeters
(2), Els Brouwers (2), Vera Ballet (3), Maja Noman (3), Marc Ferrante (3), Gert Van
Assche (3) VEDOLIZUMAB EXPOSURE CORRELATES WITH CLINICAL, BIOLOGICAL AND
ENDOSCOPIC OUTCOME IN PATIENTS WITH INFLAMMATORY BOWEL DISEASE.
United European Gastroenterology Journal 2016;2 (Supplement 1).
28. Kopylov U, Mazor Y, Yavzori M, et al. Clinical utility of antihuman lambda chain-
based enzyme-linked immunosorbent assay (ELISA) versus double antigen ELISA for
the detection of anti-infliximab antibodies. Inflamm Bowel Dis 2012;18:1628-33.
29. Wyant T, Estevam J, Yang L, et al. Development and validation of receptor occupancy
pharmacodynamic assays used in the clinical development of the monoclonal
antibody vedolizumab. Cytometry B Clin Cytom 2016;90:168-76.
30. Tew GW, Hackney JA, Gibbons D, et al. Association Between Response to
Etrolizumab and Expression of Integrin alphaE and Granzyme A in Colon Biopsies of
Patients With Ulcerative Colitis. Gastroenterology 2016;150:477-87 e9.
31. Ben-Horin S, Chowers Y. Review article: loss of response to anti-TNF treatments in
Crohn's disease. Aliment Pharmacol Ther. 2011 May;33(9):987-95
32. Boden EK, Shows D, Chiorean MV, Lord JD, Integrin α4β7 expression preceding and
saturation during vedolizumab therapy correlate with treatment response in
inflammatory bowel disease, Gastroenterology April 2017, Volume 152, Issue 5,
Supplement 1, Page S39
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Supplementary document 1 - Induction period drug levels and clinical and
biomarker response- subanalyses among CD and UC patients
Among CD patients (n=67), similar to the general study population, week 6
vedolizumab levels associated with clinical remission at week 6 (median levels 47.7
µg/mL, IQR 18.4-61 versus 33 µg/mL, IQR 21.2-42.5, p=0.037 in those in remission
versus clinically active patients, respectively). However, ROC analysis demonstrated
only modest discriminatory accuracy (AUC=0.66, 73% sensitivity, 70.3% specificity
for drug level>37.8 µg/mL, p=0.055. Odds-ratio 5.5 for remission with levels over
37.8 µg/mL, p<0.01, supplementary figure 2a). All other associations were
insignificant among CD patients (supplementary table 7). A separate analysis of week
2, 6 levels in relation to HBI values yielded no significant association either (data not
shown).
Among UC patients (n=39), week 2 median vedolizumab levels were significantly
higher - 55.8 µg/mL (IQR 35.4-80.8) in patients who went on to have clinical
remission at week 6, compared to 29.9 µg/mL (IQR 24-47.7) in patients without
clinical remission at week 6 (p=0.02). ROC analysis demonstrated that vedolizumab
week 2 levels above 32.7 µg/mL had a 90% sensitivity and a 61.1% specificity for
clinical remission at week 6 among UC patients (AUC=0.73, p=0.01, odds-ratio=8 for
remission with a drug level>32.7mcg/ml, p=0.019, supplementary figure 2b).
Nevertheless, sub-analyses of vedolizumab levels among UC patients at week 6 vis-a-
vis clinical remission and clinical score (SCCAI) at weeks 6 or 14 did not yield
statistically significant correlations (supplementary table 8).
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Supplementary table 7 - Association between week 2 and 6 vedolizumab trough levels and clinical remission by the end of induction among CD patients (n=67)
Week of trough level measurement
Patients in remission - week 6
Patients clinically active -week 6
P value
Patients in remission - week 14
Patients clinically active -week 14
P value
Week 2 drug level (median, IQR)
41.2, 28.2-60.5
46.8, 34.3-54.5
0.53 44.4, 31-60.3
46.5, 30.7-58.8
0.87
Week 6 drug level (median, IQR)
47.7, 18.4-61
33, 21.2-42.5
0.037 40.2, 25.5-57.2
33.2, 20.7-47.5
0.23
CD - Crohn's disease, IQR - interquartile range
Supplementary table 8 - Association between week 2 and 6 vedolizumab trough levels and clinical remission by the end of induction among UC patients (n=39)
Week of trough level measurement
Patients in remission - week 6
Patients clinically active -week 6
P value
Patients in remission - week 14
Patients clinically active -week 14
P value
Week 2 drug level (median, IQR)
55.8, 35.4-80.8
29.9, 24-47.7
0.02 38.4, 28.4-61.9
40.2, 24.6-72.5
0.89
Week 6 drug level (median, IQR)
35.7, 18.1-48.5
23, 12-42.8
0.31 24.4, 16.4-37.6
36.6, 13.5-54.3
0.2
UC - ulcerative colitis, IQR - interquartile range
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Supplementary Table 1 - Analysis of pharmacokinetic parameters associated with week 2, 6, 14 vedolizumab levels (n=98)
CRP - C reactive protein, CD - Crohn's disease
Supplementary Table 2 - Comparison of clinical remission rates between quartiles of vedolizumab trough levels
Quartiles compared P value Odd Ratio Confidence interval 1, 2 0.09 3.2 0.8-12.1 1, 3 0.4 0.6 0.2-2 1, 4 0.2 0.5 0.15-1.5 2, 4 0.006 0.15 0.04-0.6 2, 3 0.02 0.2 0.05-0.7 3, 4 0.7 0.8 0.25-25
Variable Week 2 univariate
analysis P Value
Week 2 multivariate
analysis P Value
Week 6 univariate
analysis P Value
Week 6 multivariate
analysis P Value
Week 14 univariate
analysis P value
Week 14 multivariate
analysis P Value
Baseline albumin (g/dl)
0.009 0.34 0.002 0.05 <0.0001 0.004
Baseline CRP (mg/l)
0.02 0.53 0.007 0.25 0.01 0.77
Baseline weight (kg)
0.007 0.02 0.28 0.25
Gender 0.02 0.07 0.05 0.99 Concomitant steroidal therapy
0.24 0.83 0.55
Concomitant immunomodulator therapy
0.67 0.28 0.81
IBD type (CD) 0.67 0.17 0.92 Age 0.02 0.74 0.37 0.25 Prior anti-TNFα therapy
0.53 0.9 0.5
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Supplementary Table 3 - Association between week 2, 6 vedolizumab trough levels and CRP normalization by the end of induction (weeks 6, 14, n=95)
Week of trough level measurement
Patients with elevated CRP - week 6
Patients whose CRP normalized - week 6
P value
Patients with elevated CRP - week 14
Patients whose CRP normalized - week 14
P value
Week 2, µg/mL (median, IQR)
40, 31.4-48
35.4, 28-50.6
0.96 37 24.7-47.7
46.4, 28.7-50.9
0.3
Week 6, µg/mL (median, IQR)
22.4, 13.1-36.9
29.4, 17.4-50.3
0.19 24.9, 15-37.2
29.4, 17-45
0.6
CRP - C-reactive protein, IQR - interquartile range
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Supplementary Table 4 - Univariable analysis of pharmacokinetic parameters associated with week 6 and 14 clinical remission (n=106)
CRP - C-reactive protein, CD - Crohn's disease, OR - odds ratio, CI - confidence interval
Supplementary Table 5 - Multivariable analysis of pharmacokinetic parameters associated with week 6 clinical remission
Variable Week 6 remission, p value
Week 6 vedolizumab level, µg/mL 0.34
Baseline CRP, mg/l 0.35 Prior anti-TNFα therapy 0.37 CRP - C-reactive protein. *Only parameters with significant/borderline association with week 6 remission were incorporated into the analysis. Only one parameter was borderline for week 14 remission, hence multivariate analysis was not performed.
Variable Week 6 remission
Week 14 remission
Median - patients in
clinical remission
Median - clinically
active patients
P value
Median - patients in
clinical remission
Median - clinically
active patients
P value
Week 2 vedolizumab level, µg/mL
47.7 43.5 0.3 41.2 46.5 0.88
Week 6 vedolizumab level, µg/mL
40.2 29.7 0.05 35.6 33.1 0.96
Baseline albumin, g/dl 3.9 3.9 0.89 3.9 3.9 0.97 Baseline CRP, mg/l 7.5 12.4 0.09 9.5 8.7 0.8 Baseline weight, kg 60.5 65.4 0.67 70 60.5 0.09 Age, years 37 37 0.44 37 37 0.96 Disease duration, years 10.5 11 0.94 12.5 10 0.63
Variable OR CI P value
OR CI P value
Concomitant immunomodulator therapy
0.5 0.16-1.5 0.22 1.1 0.4-3.3 0.85
Concomitant steroidal therapy
0.54 0.24-1.3 0.15 1.5 0.6-3.4 0.36
IBD type (CD) 1.5 0.7-3.24 0.35 1.6 0.7-3.5 0.25 Prior anti-TNFα therapy 2.5 0.8-8 0.09 1.5 0.5-4.8 0.45 Gender (female) 0.6 0.3-1.3 0.2 1.5 0.7-3.2 0.32
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Supplementary Table 6 - Clinical, endoscopic and histological characteristics of 15 patients whose lamina propria biopsies were analyzed
Patient IBD type
Endoscopy Histology Biopsy location
Days since last
infusion 1 UC Hyperemic mucosa in left
colon, MAYO 1 Mild chronice active
inflammation sigmoid colon
14
2 UC Inflammation with exudates and hemorrhage, MAYO 3
Moderately to severe chronic active colitis with ulceration and granulation
recto-sigmoid
9
3 UC Ulcerated hyperemic mucosa MAYO 1-2
Moderate chronic active colitis with crypt abscesses
transverse - sigmoid
colon
22
4 UC Severely inflamed colonic mucosa
Chronic colitis sigmoid colon
14
5 UC Moderate inflammation with ulcerations
Mild to moderate chronic active colitis
sigmoid colon
14
6 CD Edematous inflamed ulcerated mucosa
Chronic colitis with moderate activity including
cryptitis and granulation tissue formation.
sigmoid colon
12
7 UC Moderate inflammation Colonic mucosa with mild to moderate glandular
architectural distortion and reactive lymphoid
hyperplasia
transverse colon
14
8 Severe inflammation with ulcerations MAYO 3
Colonic mucosa with severe active chronic
colitis.
transverse - sigmoid
colon
21
9 CD Severe inflammation with spontaneous bleeding
Active chronic colitis with ulceration and granulation and regenerative epithelial
changes
sigmoid colon
14
10 CD Hyperemic mucosa at anastomosis
Minimally active colitis with hyperplastic crypts
ileo- colonic
anastomosis
20
11 CD Inflammation with ulceration at anastomosis -
Rutgreets 2
Minimally active colitis with lymphoid follicle formation
sigmoid colon
5
12 UC Inflamed mucosa MAYO 2 Mild to moderate chronic active colitis
sigmoid colon
6
13 UC Mild - moderate inflammation
Active chronic colitis sigmoid colon
18
14 UC Severe inflammation with exudates MAYO 2-3
Moderate to severe active chronic colitis with crypt abscess and granulation
tissue
recto-sigmoid
14
15 UC Erythematous mucosa mildly active chronic colitis with focal cryptitis
rectum 0
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT