- Email: [email protected]
Tu1934 Assessment of Serum C-Reactive Protein, Fecal Lactoferrin, and Fecal Calprotectin in Patients With Moderate-Severely Active Crohns Disease: Results From the IM UNITI Maintenance Study
AGA Abstracts
Tu1933
and 29.5% on PBO. The proportions of pts with normalized CRP concentrations remained stable through Wk44 in the UST groups, with 19.6% on q8w and 22.8% on q12w, but decreased from 29.5% to 10.5% of the pts randomized to SC PBO. Median fLac levels at maintenance BL were 58.2 µg/g, 32.0, and 70.4 µg/g in the PBO, q8w and q12w groups, respectively(Table). Median change at Wk44 was 0.00 µg/g and -2.99 µg/g in the q8w and q12w groups vs 26.8 µg/g in the PBO group (p=0.001 and p=0.007 respectively). Baseline median fCal levels were: 191.1 µg/g and 493.5 µg/g for the q8w and q12w groups vs 262.1 µg/g for PBO. At Wk44, median fCal did not change from BL in both UST groups, while increasing by 154ug/g on SC PBO (Table 1). Among pts with an induction BL fCal value >250 µg/g, the proportions of pts with values £100 µg/g at Wk44 were significantly greater in both SC UST groups (19.3% &15.4% on q8w & q12w, respectively) compared with the PBO group (5.4%, p=0.008 and p=0.028, respectively). A generally similar pattern of reductions in fLac and in proportions with fCal < 250 at Wk44 was also observed. Conclusion: Reductions in CRP, fCal and fLac attained with UST induction were maintained at Wk44 in both the SC UST 90 mg q12w and q8w groups, providing further support for the clinical efficacy observed in the IM-UNITI maintenance trial through significant reductions in these objective markers of active inflammation.
Fecal Calprotectin Improves the Predictive Power of 3 Practical Indices for Mucosal Healing Among Patients With Crohn's Disease: Results From PREDICT William Sandborn, Remo Panaccione, Jean-Frederic Colombel, Edouard Louis, Anne M. Robinson, Kori Wallace, Mareiki Bereswill, Marisa Shapiro, Naijun Chen, Martha Skup, Song Wang, Roopal B. Thakkar, Jingdong Chao Background: Mucosal healing (MH) is an important treatment goal for Crohn's disease (CD), but requires the use of endoscopy to assess which is expensive and invasive. Methods: PREDICT is a multi-center study to evaluate if biomarkers (e.g., platelet count, fecal calprotectin [FCP]) and patient reported outcomes (e.g., rectal bleeding), alone or in combination, can predict endoscopic MH. Three predictive indices, composed of biomarkers and patient reported outcomes, based upon 3 different endoscopic scoring methods, were created from the prior EXTEND1 trial using the Least Absolute Shrinkage and Selection Operator (LASSO) procedure. In PREDICT, endoscopic MH was assessed by 3 endoscopic scoring methods: 1) centrally reviewed Simple Endoscopic Score for CD (SES-CD); 2) centrally reviewed CD Endoscopic Index of Severity (CDEIS); and 3) site investigator's visual assessment. The cutoffs used for MH (SES-CD£5, CDEIS£4.24) were developed from centrally read endoscopies in EXTEND1. In PREDICT, these 3 MH definitions were used as gold standards to test the validity of the predictive indices. The 3 indices were evaluated using a Receiver Operating Characteristics (ROC) analysis. Area under the ROC curve (AUC) £ 0.75 indicates the index is not clinically meaningful. After dichotomizing the indices at the cut-off value that maximizes the sum of sensitivity and specificity, prediction performance was assessed via positive predictive value (PPV) and negative predictive value (NPV). To determine the clinical relevance of the predictive indices as a surrogate for MH, another cut-off value for the predictive indices was identified as the lowest probability from logistic regression that achieved at least 90% of PPV. Results: ROC analysis showed that the CDEIS index had the highest predictive power (AUC=0.710) followed by the SES-CD index (AUC=0.590) and the visual assessment index (AUC=0.583). After addition of FCP to the model, AUC increased to 0.818 for the CDEIS index, to 0.828 for the SES-CD index, and to 0.764 for the visual index ( Table). ROC analysis showed FCP alone was a powerful predictor of MH but did not achieve 90% PPV. Conclusion: Predictive power of indices for MH by CDEIS, SES-CD, and visual assessment were substantially improved by addition of FCP to the indices. The 3 improved indices can be practically adapted in clinical setting based on the cut-off probability that achieved 90% PPV and may be clinically useful in identifying patients who achieve MH. References: 1Sandborn WJ, et al. J Crohns Colitis. 2012;6(1 Suppl):S3.
Tu1935 The Within-Stool and Within-Day Variability of Fecal Calprotectin in Patients With Inflammatory Bowel Disease Lillian Du, Rae Foshaug, Vivian Huang, Karen Kroeker, Levinus A. Dieleman, Brendan P. Halloran, Karen Wong, Richard N. Fedorak Background: Fecal calprotectin (FC) constitutes approximately 60% of neutrophil cytosolic proteins, and its measured level in feces is proportional to the degree of neutrophil migration into the bowel lumen. The use of FC as a non-invasive stool biomarker for differentiating IBD from IBS has been well validated, and there is a strong correlation between FC and the presence of endoscopic inflammatory lesions. However, recent studies have demonstrated within-day and between-day FC variability in patients with active IBD, possibly limiting the reliability of using a single sample for monitoring disease activity and guiding management. Aim: To assess the within-stool and within-day variability of fecal calprotectin concentrations in patients with Crohn's disease or ulcerative colitis. Methods: This is a prospective observational study evaluating a cross-sectional cohort of IBD patients >18 years presenting for outpatient follow-up or being admitted to hospital in clinical flare based on either a Harvey Bradshaw Index score greater than 4 for Crohn's disease or a partial Mayo score greater than 1 for ulcerative colitis. Eligible patients were instructed to collect three stool samples from their first bowel movement of the day and from up to two additional bowel movements (from the afternoon and from the evening) within a 24-hour period. FC concentrations were measured by a rapid, quantitative point-of-care test using lateral flow technology (Quantum Blue ®). For FC levels in different samples from the same bowel movement, an intraclass correlation coefficient was used to determine the test-retest reliability. One-way ANOVA was calculated to determine the effect of time of day on the level of fecal calprotectin. Results:. The overall within stool correlation coefficient was 0.93, demonstrating excellent test-retest reliability within samples. One-way ANOVA results of samples collected at different times of day ranged from F=21.35 (p=0.007) to F=35.76 (p=0.003), suggesting significantly higher FC levels in morning samples compared to evening samples. Conclusions: The preliminary results of our study suggest that within a single bowel movement, fecal calprotectin levels are robustly reproducible. However, there appears to be significant variation in FC concentrations between different bowel movements within the same day, and this should be considered when interpreting FC results.
a
Cut-off value is the probability for multi-predictors indices and FCP value for FCP alone as the predictor.
Tu1934 Assessment of Serum C-Reactive Protein, Fecal Lactoferrin, and Fecal Calprotectin in Patients With Moderate-Severely Active Crohns Disease: Results From the IM UNITI Maintenance Study William Sandborn, Christopher Gasink, Douglas Jacobstein, Long-Long Gao, Jewel Johanns, Stephan Targan, Subrata Ghosh, Stephen B. Hanauer Objective: Inflammatory markers serve as surrogate markers of disease activity in Crohn's disease (CD). The objective was to evaluate markers of inflammation including serum Creactive protein(CRP), fecal lactoferrin(fLac), and fecal calprotectin (fCal) in a maintenance study of subcutaneous (SC) ustekinumab(UST) in patients(pts) with moderate-severely active CD who have either failed anti-TNF or conventional therapy. Methods: The randomized, double-blind, placebo-controlled, IM-UNITI maintenance study evaluated the safety and efficacy of SC UST in the primary population of pts who achieved clinical response (100 point decrease in CDAI) to IV UST induction therapy in either of the UNITI-1 & 2 induction studies (n=388). These pts were randomized to SC placebo, UST 90mg q8w (q8w) or UST 90mg q12w (q12w) and examined through the primary endpoint of clinical remission 44 weeks later. Evaluations of markers of inflammation (CRP, fLac and fCal) were evaluated as secondary endpoints. Results: At Wk0 baseline (BL) of the maintenance study, CRP levels were similar across the treatment groups, with a median CRP level of 4.66 mg/L. At Wk44, median increase from BL in CRP levels were increases of 0.51 mg/L, and 0.42 mg/ L for the q8w and q12w groups vs a 4.07 mg/L median increase without UST maintenance on SC PBO (p<0.001 and p=0.002 respectively). Proportions of pts with normalized CRP levels (£ 3mg/L) at BL of the maintenance study were: 19.6% on q8w, 21.8% on q12w,
AGA Abstracts
X : 10052$CH01 03-28-16 00:57:28 PDFd : 10052B : e
Tu1936 Natural History of Radiological Transmural Response Identified on Serial Enterography in Small Bowel Crohn's Disease Patients Parakkal Deepak, Joel G. Fletcher, Jeff L. Fidler, John M. Barlow, Shannon P. Sheedy, Amy kolbe, William S. Harmsen, Terry Therneau, Edward V. Loftus, Stephanie L. Hansel, Brenda D. Becker, David H. Bruining Introduction: Radiological transmural response (TR) as a treatment goal at the 1st followup scan in small bowel Crohn's disease (CD) has been shown to be associated with reduced probability of CD-related surgeries, hospitalizations, and rescue corticosteroid usage. We examined the natural history of long-term radiological TR including the durability and the influence of baseline predictors on the maintenance of TR. Methods: A cohort of patient (pts) with small bowel CD who had serial computed tomography enterography (CTE) and/or magnetic resonance enterography (MRE) between 1/1/2002 and 10/31/2014 was constructed. Inclusion required pre-therapy (rx) CTE/MRE with follow-up imaging after 6
S-982
Page 982
Tu1933
and 29.5% on PBO. The proportions of pts with normalized CRP concentrations remained stable through Wk44 in the UST groups, with 19.6% on q8w and 22.8% on q12w, but decreased from 29.5% to 10.5% of the pts randomized to SC PBO. Median fLac levels at maintenance BL were 58.2 µg/g, 32.0, and 70.4 µg/g in the PBO, q8w and q12w groups, respectively(Table). Median change at Wk44 was 0.00 µg/g and -2.99 µg/g in the q8w and q12w groups vs 26.8 µg/g in the PBO group (p=0.001 and p=0.007 respectively). Baseline median fCal levels were: 191.1 µg/g and 493.5 µg/g for the q8w and q12w groups vs 262.1 µg/g for PBO. At Wk44, median fCal did not change from BL in both UST groups, while increasing by 154ug/g on SC PBO (Table 1). Among pts with an induction BL fCal value >250 µg/g, the proportions of pts with values £100 µg/g at Wk44 were significantly greater in both SC UST groups (19.3% &15.4% on q8w & q12w, respectively) compared with the PBO group (5.4%, p=0.008 and p=0.028, respectively). A generally similar pattern of reductions in fLac and in proportions with fCal < 250 at Wk44 was also observed. Conclusion: Reductions in CRP, fCal and fLac attained with UST induction were maintained at Wk44 in both the SC UST 90 mg q12w and q8w groups, providing further support for the clinical efficacy observed in the IM-UNITI maintenance trial through significant reductions in these objective markers of active inflammation.
Fecal Calprotectin Improves the Predictive Power of 3 Practical Indices for Mucosal Healing Among Patients With Crohn's Disease: Results From PREDICT William Sandborn, Remo Panaccione, Jean-Frederic Colombel, Edouard Louis, Anne M. Robinson, Kori Wallace, Mareiki Bereswill, Marisa Shapiro, Naijun Chen, Martha Skup, Song Wang, Roopal B. Thakkar, Jingdong Chao Background: Mucosal healing (MH) is an important treatment goal for Crohn's disease (CD), but requires the use of endoscopy to assess which is expensive and invasive. Methods: PREDICT is a multi-center study to evaluate if biomarkers (e.g., platelet count, fecal calprotectin [FCP]) and patient reported outcomes (e.g., rectal bleeding), alone or in combination, can predict endoscopic MH. Three predictive indices, composed of biomarkers and patient reported outcomes, based upon 3 different endoscopic scoring methods, were created from the prior EXTEND1 trial using the Least Absolute Shrinkage and Selection Operator (LASSO) procedure. In PREDICT, endoscopic MH was assessed by 3 endoscopic scoring methods: 1) centrally reviewed Simple Endoscopic Score for CD (SES-CD); 2) centrally reviewed CD Endoscopic Index of Severity (CDEIS); and 3) site investigator's visual assessment. The cutoffs used for MH (SES-CD£5, CDEIS£4.24) were developed from centrally read endoscopies in EXTEND1. In PREDICT, these 3 MH definitions were used as gold standards to test the validity of the predictive indices. The 3 indices were evaluated using a Receiver Operating Characteristics (ROC) analysis. Area under the ROC curve (AUC) £ 0.75 indicates the index is not clinically meaningful. After dichotomizing the indices at the cut-off value that maximizes the sum of sensitivity and specificity, prediction performance was assessed via positive predictive value (PPV) and negative predictive value (NPV). To determine the clinical relevance of the predictive indices as a surrogate for MH, another cut-off value for the predictive indices was identified as the lowest probability from logistic regression that achieved at least 90% of PPV. Results: ROC analysis showed that the CDEIS index had the highest predictive power (AUC=0.710) followed by the SES-CD index (AUC=0.590) and the visual assessment index (AUC=0.583). After addition of FCP to the model, AUC increased to 0.818 for the CDEIS index, to 0.828 for the SES-CD index, and to 0.764 for the visual index ( Table). ROC analysis showed FCP alone was a powerful predictor of MH but did not achieve 90% PPV. Conclusion: Predictive power of indices for MH by CDEIS, SES-CD, and visual assessment were substantially improved by addition of FCP to the indices. The 3 improved indices can be practically adapted in clinical setting based on the cut-off probability that achieved 90% PPV and may be clinically useful in identifying patients who achieve MH. References: 1Sandborn WJ, et al. J Crohns Colitis. 2012;6(1 Suppl):S3.
Tu1935 The Within-Stool and Within-Day Variability of Fecal Calprotectin in Patients With Inflammatory Bowel Disease Lillian Du, Rae Foshaug, Vivian Huang, Karen Kroeker, Levinus A. Dieleman, Brendan P. Halloran, Karen Wong, Richard N. Fedorak Background: Fecal calprotectin (FC) constitutes approximately 60% of neutrophil cytosolic proteins, and its measured level in feces is proportional to the degree of neutrophil migration into the bowel lumen. The use of FC as a non-invasive stool biomarker for differentiating IBD from IBS has been well validated, and there is a strong correlation between FC and the presence of endoscopic inflammatory lesions. However, recent studies have demonstrated within-day and between-day FC variability in patients with active IBD, possibly limiting the reliability of using a single sample for monitoring disease activity and guiding management. Aim: To assess the within-stool and within-day variability of fecal calprotectin concentrations in patients with Crohn's disease or ulcerative colitis. Methods: This is a prospective observational study evaluating a cross-sectional cohort of IBD patients >18 years presenting for outpatient follow-up or being admitted to hospital in clinical flare based on either a Harvey Bradshaw Index score greater than 4 for Crohn's disease or a partial Mayo score greater than 1 for ulcerative colitis. Eligible patients were instructed to collect three stool samples from their first bowel movement of the day and from up to two additional bowel movements (from the afternoon and from the evening) within a 24-hour period. FC concentrations were measured by a rapid, quantitative point-of-care test using lateral flow technology (Quantum Blue ®). For FC levels in different samples from the same bowel movement, an intraclass correlation coefficient was used to determine the test-retest reliability. One-way ANOVA was calculated to determine the effect of time of day on the level of fecal calprotectin. Results:. The overall within stool correlation coefficient was 0.93, demonstrating excellent test-retest reliability within samples. One-way ANOVA results of samples collected at different times of day ranged from F=21.35 (p=0.007) to F=35.76 (p=0.003), suggesting significantly higher FC levels in morning samples compared to evening samples. Conclusions: The preliminary results of our study suggest that within a single bowel movement, fecal calprotectin levels are robustly reproducible. However, there appears to be significant variation in FC concentrations between different bowel movements within the same day, and this should be considered when interpreting FC results.
a
Cut-off value is the probability for multi-predictors indices and FCP value for FCP alone as the predictor.
Tu1934 Assessment of Serum C-Reactive Protein, Fecal Lactoferrin, and Fecal Calprotectin in Patients With Moderate-Severely Active Crohns Disease: Results From the IM UNITI Maintenance Study William Sandborn, Christopher Gasink, Douglas Jacobstein, Long-Long Gao, Jewel Johanns, Stephan Targan, Subrata Ghosh, Stephen B. Hanauer Objective: Inflammatory markers serve as surrogate markers of disease activity in Crohn's disease (CD). The objective was to evaluate markers of inflammation including serum Creactive protein(CRP), fecal lactoferrin(fLac), and fecal calprotectin (fCal) in a maintenance study of subcutaneous (SC) ustekinumab(UST) in patients(pts) with moderate-severely active CD who have either failed anti-TNF or conventional therapy. Methods: The randomized, double-blind, placebo-controlled, IM-UNITI maintenance study evaluated the safety and efficacy of SC UST in the primary population of pts who achieved clinical response (100 point decrease in CDAI) to IV UST induction therapy in either of the UNITI-1 & 2 induction studies (n=388). These pts were randomized to SC placebo, UST 90mg q8w (q8w) or UST 90mg q12w (q12w) and examined through the primary endpoint of clinical remission 44 weeks later. Evaluations of markers of inflammation (CRP, fLac and fCal) were evaluated as secondary endpoints. Results: At Wk0 baseline (BL) of the maintenance study, CRP levels were similar across the treatment groups, with a median CRP level of 4.66 mg/L. At Wk44, median increase from BL in CRP levels were increases of 0.51 mg/L, and 0.42 mg/ L for the q8w and q12w groups vs a 4.07 mg/L median increase without UST maintenance on SC PBO (p<0.001 and p=0.002 respectively). Proportions of pts with normalized CRP levels (£ 3mg/L) at BL of the maintenance study were: 19.6% on q8w, 21.8% on q12w,
AGA Abstracts
X : 10052$CH01 03-28-16 00:57:28 PDFd : 10052B : e
Tu1936 Natural History of Radiological Transmural Response Identified on Serial Enterography in Small Bowel Crohn's Disease Patients Parakkal Deepak, Joel G. Fletcher, Jeff L. Fidler, John M. Barlow, Shannon P. Sheedy, Amy kolbe, William S. Harmsen, Terry Therneau, Edward V. Loftus, Stephanie L. Hansel, Brenda D. Becker, David H. Bruining Introduction: Radiological transmural response (TR) as a treatment goal at the 1st followup scan in small bowel Crohn's disease (CD) has been shown to be associated with reduced probability of CD-related surgeries, hospitalizations, and rescue corticosteroid usage. We examined the natural history of long-term radiological TR including the durability and the influence of baseline predictors on the maintenance of TR. Methods: A cohort of patient (pts) with small bowel CD who had serial computed tomography enterography (CTE) and/or magnetic resonance enterography (MRE) between 1/1/2002 and 10/31/2014 was constructed. Inclusion required pre-therapy (rx) CTE/MRE with follow-up imaging after 6
S-982
Page 982