ORIGINAL ARTICLE
Fecal Calprotectin as Predictor of Relapse in Patients With Inflammatory Bowel Disease Under Maintenance Infliximab Therapy Rocio Ferreiro-Iglesias, MD,* Manuel Barreiro-de Acosta, MD, PhD,* Manuel Otero Santiago, BSc,w Aurelio Lorenzo Gonzalez, MD,* Carmen Alonso de la Pen˜a, BSc,w Alfonso J. Benitez Estevez, BSc,w and Juan Enrique Dominguez-Mun˜oz, MD, PhD*
Background and Goals: Predicting relapse in Inflammatory Bowel Disease (IBD) could allow for early changes of treatment. Close monitoring of fecal calprotectin (FC) could be useful to predict relapse in IBD. Aim of the study was to evaluate the predictive value of a rapid FC test to predict flares in patients with IBD under maintenance therapy with Infliximab. Study: A prospective observational cohort study was designed. IBD patients in clinical remission under maintenance Infliximab therapy were included. FC was measured using a rapid test on a stool sample obtained within 24 hours before Infliximab infusion. Clinical examination was performed 2 months after that infusion. Results: Fifty-three patients were included (52.8% female). Thirtythree patients (62.3%) had Crohn’s disease and 20 (37.7%) had ulcerative colitis. All patients were in remission at inclusion. After 2 months, 41 patients (77.4%) remained in clinical remission and 12 (22.6%) presented a relapse. FC (mean ± SD) in relapsing and notrelapsing disease was 332 ± 168 and 110 ± 163 mg/g, respectively (P < 0.005). A FC concentration > 160 mg/g had a sensitivity of 91.7%, and specificity of 82.9% to predict relapse. Conclusions: In IBD patients under Infliximab maintenance therapy, high FC levels allow predicting relapse within the following 2 months. Long-term remission is associated with low calprotectin levels. Further studies are required to confirm these results. Key Words: calprotectin, inflammatory bowel disease, Crohn’s disease, ulcerative colitis, Infliximab
(J Clin Gastroenterol 2016;50:147–151)
C
hronic Inflammatory Bowel Diseases (IBD) are characterized by flares and periods of remission. The inflammatory activity in IBD determines the onset of Received for publication September 7, 2014; accepted February 13, 2015. From the Departments of *Gastroenterology and Hepatology; and wLaboratory Medicine, University Hospital of Santiago de Compostela, Santiago de Compostela, Spain. All authors have made substantial contributions to all of the following: R.F.I., M.B.A.: the conception and design of the study; R.F.I., M.O.S.: acquisition of data; R.F.I.; drafting the article; J.E.D.M., M.B.A., M.O.S., A.L.G., A.J.B.E., C.A.P., J.E.D.M.: revising it critically for important intellectual content; and M.B.A., J.E.D.M.: final approval of the version to be submitted. The authors declare that they have nothing to disclose. Reprints: Rocio Ferreiro-Iglesias, MD, Department of Gastroenterology and Hepatology, University Hospital of Santiago de Compostela, c/Choupana, s/n, Santiago de Compostela 15706, Spain (e-mail: [email protected]). Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 50, Number 2, February 2016
symptoms and the development of complications, which usually require modification of treatment. Different clinical and endoscopic indexes, as well as inflammatory biomarkers are used to evaluate the activity of the disease.1–6 The risk of relapse is however unpredictable. The correlation between clinical indexes and the presence of inflammatory activity is far to be optimal, and the sensitivity and specificity of clinical indexes for diagnosing IBD relapse is limited, especially for mild flares.6 Endoscopic indexes could be considered as the gold standard to predict IBD relapse due to their high sensitivity and specificity, but colonoscopy is an invasive, expensive, and not always well-tolerated technique.7,8 Some biomarkers have been proposed to predict IBD relapse, which are more specific than clinical indexes and better tolerated and cheaper than endoscopic indexes. Blood markers are the most commonly used, mainly C-reactive protein (CRP). However, CRP is unspecific and it can be elevated in other inflammatory processes different from IBD.9,10 Fecal markers such as calprotectin are more specific than blood markers, and the main advantage is that of being unaffected by extraintestinal processes. Calprotectin binds to calcium and zinc and constitutes about 30% to 60% of the total cytosolic proteins of neutrophils. It is found in plasma, in some body fluids, and in activated macrophages. In the presence of inflammation of the intestinal mucosa, exudation of these cells into the lumen is produced. The presence of this protein in feces is thus directly related to the migration of neutrophils into the gastrointestinal lumen. Calprotectin is highly stable in feces and only a single small stool sample is required for its quantification.7,8,11–13 Fecal calprotectin (FC) levels allow differentiating accurately organic intestinal diseases from functional disorders such as irritable bowel syndrome. However, it is not specific for IBD and it can also be elevated in other intestinal diseases such as neoplasms, celiac disease, gastrointestinal infections, and use of nonsteroidal anti-inflammatory drugs.14–18 According to studies recruiting patients with IBD in remission, FC could be considered as an appropriate tool for predicting relapse of the disease and thus allowing for early changes in treatment.6,19–25 However, data are scarce regarding the usefulness of close monitoring of FC levels to predict relapse in IBD patients receiving biological therapies.26 The prognostic value of FC varies most probably with different treatments and different severity of the disease. In addition, previous reported data are not solid enough due to the wide heterogeneity of patients included www.jcge.com |
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
147
Ferreiro-Iglesias et al
and the low number of studies including patients receiving biological therapy.6,19–26 As a result, well-defined predictive factors being of help to optimize biological therapy to prevent IBD relapses are needed. The aim of the study was to evaluate the accuracy of a rapid FC test to predict flares in IBD patients in remission under maintenance treatment with Infliximab.
MATERIALS AND METHODS A cross-sectional, observational, prospective cohort study with consecutive patient recruitment was designed and conducted at the IBD Unit of the Department of Gastroenterology of the University Hospital of Santiago de Compostela, Spain. Patients aged 18 years or above with a previous diagnosis of IBD based on clinical, endoscopic, radiologic, and histologic criteria were considered for inclusion in the study. Patients should be in clinical remission for at least 6 months under maintenance Infliximab therapy (5 mg/kg every 8 weeks). Patients receiving different Infliximab regimens such as 10 mg/kg or those who received the drug at 6-week intervals were excluded. Any antibiotic or nonsteroidal anti-inflammatory drug therapy within the 6 months before inclusion, as well as the presence of any relevant cardiorespiratory, liver, hematological, neurological, renal, or serious psychiatric disorders were also considered as exclusion criteria. Changes in concomitant therapies were allowed according to clinical practice. All included patients collected a small sample of feces in a disposable plastic container at inclusion, within the 24 hours before Infliximab infusion. Clinical and demographic characteristics, as well as clinical disease activity indexes were recorded at that time point and at week 8, just before Infliximab infusion. Relapse was defined in Crohn’s disease as a Harvey and Bradshaw index higher than 4, and in ulcerative colitis (UC) as a Partial Mayo index higher than 2. The Harvey and Brashaw index consists of clinical data such as general well-being (0 = very well, 1 = slightly below average, 2 = poor, 3 = very poor, 4 = terrible), abdominal pain (0 = none, 1 = mild, 2 = moderate, 3 = severe), number of liquid stools per day, abdominal mass (0 = none, 1 = dubious, 2 = definite, 3 = tender), and complications (arthralgia, uveitis, erythema nodosum, aphthous ulcers, pyoderma gangrenosum, anal fissure, new fistula, and abscess). The Partial Mayo index consists of clinical data such as rectal bleeding (0 = no blood seen, 1 = streaks of blood with stool less than half of the time, 2 = obvious blood with stool most of the time, 3 = blood alone passed), stool frequency (0 = normal number of stools for this patient, 1 = 1 to 2 stools more than normal, 2 = 3 to 4 stools more than normal, 3 = 5 or more stools more than normal), and physician’s global assessment (0 = normal, 1 = mild disease, 2 = moderate disease, 3 = severe disease). These 3 categories rated from 0 to 3 that are summed to give a total score ranging from 0 to 9. All patients gave their informed consent before inclusion. The study was approved by the local Ethics Committee and it was performed according to the declaration of Helsinki and following the rules for Good Clinical Practice.
FC Quantification Stool samples were processed freshly within the same day they were received by a trained technician who was blinded to any clinical data. Calprotectin extracts were prepared by using the BU¨HLMANN Smart-Prep fecal
J Clin Gastroenterol
Volume 50, Number 2, February 2016
TABLE 1. Clinical and Sociodemographic Data
All Patients (n = 53) Age (median, range) 46 (y) Smoking habit [n (%)] Current 8 Never 26 Former 19 Sex [n (% females)] 28 Disease duration 9 (median, range) (y) Previous IBD 20 surgery [n (%)] Extraintestinal 23 manifestations [n (%)] Extent of disease [n (%)] Left-sided colitis 2 Extensive colitis 18 Ileitis 10 Ileocolitis 18 Colitis 5
Crohn’s Disease (n = 33)
Ulcerative Colitis (n = 20)
(18-68)
41 (18-43)
51 (19-68)
(15.1) (49.1) (35.8) (52.8) (2-24)
7 14 12 20 9
1 12 7 8 9.5
(37.7)
18 (54.5)
2 (10.0)
(43.4)
12 (36.4)
11 (55.0)
(3.7) (34.0) (18.9) (34.0) (9.4)
— — 10 (30.3) 18 (54.5) 5 (15.2)
2 (10.0) 18 (90.0) — —
(21.2) (42.4) (36.4) (60.6) (2-21)
(5.0) (60.0) (35.0) (40.0) (2-24)
IBD indicates inflammatory bowel disease.
sample preparation kit (Bu¨hlmann Laboratories AG, Scho¨nenbuch, Switzerland).27 Fecal extracts were pressed into the bottom cap of a specific device (Roche Diagnostics, West Sussex, United Kingdom). The device was closed and 4 mL extraction buffer was added. The extract was vortexed for 1 minute, diluted 1:16 with extraction buffer, vortexed again, and then centrifugated for 5 minutes at 3000g. Diluted extract (60 mL) was transferred into a test cartridge coated with a monoclonal anticalprotectin antibody. A second monoclonal anticalprotectin conjugated to gold
FIGURE 1. Box graph representation of calprotectin measurements in inflammatory bowel disease patients at inclusion according to the development of relapse during follow-up. The lower and higher limits of the box represent the first and third quartiles, respectively. The black crossbar line in the box represents the median. The top and the bottom black crossbar lines represent the highest and lowest values, respectively. The symbol “J” indicates outliers; “*” extreme values.
148 | www.jcge.com Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved. Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 50, Number 2, February 2016
Fecal Calprotectin as Predictor of Relapse
TABLE 2. Accuracy of Fecal Calprotectin as a Test for Predicting Relapse in Patients With Inflammatory Bowel Disease (IBD Patients, n = 53), Crohn’s Disease (n = 33), and Ulcerative Colitis (n = 20)
Diagnostic Accuracy for Relapse Type of IBD
Calprotectin Cutoff Value (lg/g)
S (%)
Sp (%)
PPV (%)
NPV (%)
Area Under the ROC Curve (95% CI)
P
> 160 > 160 > 198
91.7 87.5 100
82.9 84.0 81.3
68.7 66.9 48.5
96.1 94.8 100
0.888 (0.771-0.958) 0.890 (0.732-0.972) 0.859 (0.632-0.972)
0.005 0.005 0.005
IBD patients Crohn’s disease Ulcerative colitis
CI indicates confidence interval; IBD, inflammatory bowel disease; NPV, negative-predictive value; PPV, positive-predictive value; ROC, receiver-operating characteristic; S, sensitivity; Sp, specificity.
colloid was released into the test cartridge by addition of the diluted extract. After incubation for 12 minutes at room temperature, the result was read using BU¨HLMANN Quantum Blue Reader as indicated by the manufacturer. The lower range Quantum Blue calprotectin has been developed to cover a calprotectin level range of 30 to 300 mg/g and the higher range has been developed to cover a calprotectin level range of 30 to 1800 mg/g. The intra-assay coefficient of variation of FC test is
Fecal Calprotectin as Predictor of Relapse in Patients With Inflammatory Bowel Disease Under Maintenance Infliximab Therapy Rocio Ferreiro-Iglesias, MD,* Manuel Barreiro-de Acosta, MD, PhD,* Manuel Otero Santiago, BSc,w Aurelio Lorenzo Gonzalez, MD,* Carmen Alonso de la Pen˜a, BSc,w Alfonso J. Benitez Estevez, BSc,w and Juan Enrique Dominguez-Mun˜oz, MD, PhD*
Background and Goals: Predicting relapse in Inflammatory Bowel Disease (IBD) could allow for early changes of treatment. Close monitoring of fecal calprotectin (FC) could be useful to predict relapse in IBD. Aim of the study was to evaluate the predictive value of a rapid FC test to predict flares in patients with IBD under maintenance therapy with Infliximab. Study: A prospective observational cohort study was designed. IBD patients in clinical remission under maintenance Infliximab therapy were included. FC was measured using a rapid test on a stool sample obtained within 24 hours before Infliximab infusion. Clinical examination was performed 2 months after that infusion. Results: Fifty-three patients were included (52.8% female). Thirtythree patients (62.3%) had Crohn’s disease and 20 (37.7%) had ulcerative colitis. All patients were in remission at inclusion. After 2 months, 41 patients (77.4%) remained in clinical remission and 12 (22.6%) presented a relapse. FC (mean ± SD) in relapsing and notrelapsing disease was 332 ± 168 and 110 ± 163 mg/g, respectively (P < 0.005). A FC concentration > 160 mg/g had a sensitivity of 91.7%, and specificity of 82.9% to predict relapse. Conclusions: In IBD patients under Infliximab maintenance therapy, high FC levels allow predicting relapse within the following 2 months. Long-term remission is associated with low calprotectin levels. Further studies are required to confirm these results. Key Words: calprotectin, inflammatory bowel disease, Crohn’s disease, ulcerative colitis, Infliximab
(J Clin Gastroenterol 2016;50:147–151)
C
hronic Inflammatory Bowel Diseases (IBD) are characterized by flares and periods of remission. The inflammatory activity in IBD determines the onset of Received for publication September 7, 2014; accepted February 13, 2015. From the Departments of *Gastroenterology and Hepatology; and wLaboratory Medicine, University Hospital of Santiago de Compostela, Santiago de Compostela, Spain. All authors have made substantial contributions to all of the following: R.F.I., M.B.A.: the conception and design of the study; R.F.I., M.O.S.: acquisition of data; R.F.I.; drafting the article; J.E.D.M., M.B.A., M.O.S., A.L.G., A.J.B.E., C.A.P., J.E.D.M.: revising it critically for important intellectual content; and M.B.A., J.E.D.M.: final approval of the version to be submitted. The authors declare that they have nothing to disclose. Reprints: Rocio Ferreiro-Iglesias, MD, Department of Gastroenterology and Hepatology, University Hospital of Santiago de Compostela, c/Choupana, s/n, Santiago de Compostela 15706, Spain (e-mail: [email protected]). Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 50, Number 2, February 2016
symptoms and the development of complications, which usually require modification of treatment. Different clinical and endoscopic indexes, as well as inflammatory biomarkers are used to evaluate the activity of the disease.1–6 The risk of relapse is however unpredictable. The correlation between clinical indexes and the presence of inflammatory activity is far to be optimal, and the sensitivity and specificity of clinical indexes for diagnosing IBD relapse is limited, especially for mild flares.6 Endoscopic indexes could be considered as the gold standard to predict IBD relapse due to their high sensitivity and specificity, but colonoscopy is an invasive, expensive, and not always well-tolerated technique.7,8 Some biomarkers have been proposed to predict IBD relapse, which are more specific than clinical indexes and better tolerated and cheaper than endoscopic indexes. Blood markers are the most commonly used, mainly C-reactive protein (CRP). However, CRP is unspecific and it can be elevated in other inflammatory processes different from IBD.9,10 Fecal markers such as calprotectin are more specific than blood markers, and the main advantage is that of being unaffected by extraintestinal processes. Calprotectin binds to calcium and zinc and constitutes about 30% to 60% of the total cytosolic proteins of neutrophils. It is found in plasma, in some body fluids, and in activated macrophages. In the presence of inflammation of the intestinal mucosa, exudation of these cells into the lumen is produced. The presence of this protein in feces is thus directly related to the migration of neutrophils into the gastrointestinal lumen. Calprotectin is highly stable in feces and only a single small stool sample is required for its quantification.7,8,11–13 Fecal calprotectin (FC) levels allow differentiating accurately organic intestinal diseases from functional disorders such as irritable bowel syndrome. However, it is not specific for IBD and it can also be elevated in other intestinal diseases such as neoplasms, celiac disease, gastrointestinal infections, and use of nonsteroidal anti-inflammatory drugs.14–18 According to studies recruiting patients with IBD in remission, FC could be considered as an appropriate tool for predicting relapse of the disease and thus allowing for early changes in treatment.6,19–25 However, data are scarce regarding the usefulness of close monitoring of FC levels to predict relapse in IBD patients receiving biological therapies.26 The prognostic value of FC varies most probably with different treatments and different severity of the disease. In addition, previous reported data are not solid enough due to the wide heterogeneity of patients included www.jcge.com |
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
147
Ferreiro-Iglesias et al
and the low number of studies including patients receiving biological therapy.6,19–26 As a result, well-defined predictive factors being of help to optimize biological therapy to prevent IBD relapses are needed. The aim of the study was to evaluate the accuracy of a rapid FC test to predict flares in IBD patients in remission under maintenance treatment with Infliximab.
MATERIALS AND METHODS A cross-sectional, observational, prospective cohort study with consecutive patient recruitment was designed and conducted at the IBD Unit of the Department of Gastroenterology of the University Hospital of Santiago de Compostela, Spain. Patients aged 18 years or above with a previous diagnosis of IBD based on clinical, endoscopic, radiologic, and histologic criteria were considered for inclusion in the study. Patients should be in clinical remission for at least 6 months under maintenance Infliximab therapy (5 mg/kg every 8 weeks). Patients receiving different Infliximab regimens such as 10 mg/kg or those who received the drug at 6-week intervals were excluded. Any antibiotic or nonsteroidal anti-inflammatory drug therapy within the 6 months before inclusion, as well as the presence of any relevant cardiorespiratory, liver, hematological, neurological, renal, or serious psychiatric disorders were also considered as exclusion criteria. Changes in concomitant therapies were allowed according to clinical practice. All included patients collected a small sample of feces in a disposable plastic container at inclusion, within the 24 hours before Infliximab infusion. Clinical and demographic characteristics, as well as clinical disease activity indexes were recorded at that time point and at week 8, just before Infliximab infusion. Relapse was defined in Crohn’s disease as a Harvey and Bradshaw index higher than 4, and in ulcerative colitis (UC) as a Partial Mayo index higher than 2. The Harvey and Brashaw index consists of clinical data such as general well-being (0 = very well, 1 = slightly below average, 2 = poor, 3 = very poor, 4 = terrible), abdominal pain (0 = none, 1 = mild, 2 = moderate, 3 = severe), number of liquid stools per day, abdominal mass (0 = none, 1 = dubious, 2 = definite, 3 = tender), and complications (arthralgia, uveitis, erythema nodosum, aphthous ulcers, pyoderma gangrenosum, anal fissure, new fistula, and abscess). The Partial Mayo index consists of clinical data such as rectal bleeding (0 = no blood seen, 1 = streaks of blood with stool less than half of the time, 2 = obvious blood with stool most of the time, 3 = blood alone passed), stool frequency (0 = normal number of stools for this patient, 1 = 1 to 2 stools more than normal, 2 = 3 to 4 stools more than normal, 3 = 5 or more stools more than normal), and physician’s global assessment (0 = normal, 1 = mild disease, 2 = moderate disease, 3 = severe disease). These 3 categories rated from 0 to 3 that are summed to give a total score ranging from 0 to 9. All patients gave their informed consent before inclusion. The study was approved by the local Ethics Committee and it was performed according to the declaration of Helsinki and following the rules for Good Clinical Practice.
FC Quantification Stool samples were processed freshly within the same day they were received by a trained technician who was blinded to any clinical data. Calprotectin extracts were prepared by using the BU¨HLMANN Smart-Prep fecal
J Clin Gastroenterol
Volume 50, Number 2, February 2016
TABLE 1. Clinical and Sociodemographic Data
All Patients (n = 53) Age (median, range) 46 (y) Smoking habit [n (%)] Current 8 Never 26 Former 19 Sex [n (% females)] 28 Disease duration 9 (median, range) (y) Previous IBD 20 surgery [n (%)] Extraintestinal 23 manifestations [n (%)] Extent of disease [n (%)] Left-sided colitis 2 Extensive colitis 18 Ileitis 10 Ileocolitis 18 Colitis 5
Crohn’s Disease (n = 33)
Ulcerative Colitis (n = 20)
(18-68)
41 (18-43)
51 (19-68)
(15.1) (49.1) (35.8) (52.8) (2-24)
7 14 12 20 9
1 12 7 8 9.5
(37.7)
18 (54.5)
2 (10.0)
(43.4)
12 (36.4)
11 (55.0)
(3.7) (34.0) (18.9) (34.0) (9.4)
— — 10 (30.3) 18 (54.5) 5 (15.2)
2 (10.0) 18 (90.0) — —
(21.2) (42.4) (36.4) (60.6) (2-21)
(5.0) (60.0) (35.0) (40.0) (2-24)
IBD indicates inflammatory bowel disease.
sample preparation kit (Bu¨hlmann Laboratories AG, Scho¨nenbuch, Switzerland).27 Fecal extracts were pressed into the bottom cap of a specific device (Roche Diagnostics, West Sussex, United Kingdom). The device was closed and 4 mL extraction buffer was added. The extract was vortexed for 1 minute, diluted 1:16 with extraction buffer, vortexed again, and then centrifugated for 5 minutes at 3000g. Diluted extract (60 mL) was transferred into a test cartridge coated with a monoclonal anticalprotectin antibody. A second monoclonal anticalprotectin conjugated to gold
FIGURE 1. Box graph representation of calprotectin measurements in inflammatory bowel disease patients at inclusion according to the development of relapse during follow-up. The lower and higher limits of the box represent the first and third quartiles, respectively. The black crossbar line in the box represents the median. The top and the bottom black crossbar lines represent the highest and lowest values, respectively. The symbol “J” indicates outliers; “*” extreme values.
148 | www.jcge.com Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved. Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 50, Number 2, February 2016
Fecal Calprotectin as Predictor of Relapse
TABLE 2. Accuracy of Fecal Calprotectin as a Test for Predicting Relapse in Patients With Inflammatory Bowel Disease (IBD Patients, n = 53), Crohn’s Disease (n = 33), and Ulcerative Colitis (n = 20)
Diagnostic Accuracy for Relapse Type of IBD
Calprotectin Cutoff Value (lg/g)
S (%)
Sp (%)
PPV (%)
NPV (%)
Area Under the ROC Curve (95% CI)
P
> 160 > 160 > 198
91.7 87.5 100
82.9 84.0 81.3
68.7 66.9 48.5
96.1 94.8 100
0.888 (0.771-0.958) 0.890 (0.732-0.972) 0.859 (0.632-0.972)
0.005 0.005 0.005
IBD patients Crohn’s disease Ulcerative colitis
CI indicates confidence interval; IBD, inflammatory bowel disease; NPV, negative-predictive value; PPV, positive-predictive value; ROC, receiver-operating characteristic; S, sensitivity; Sp, specificity.
colloid was released into the test cartridge by addition of the diluted extract. After incubation for 12 minutes at room temperature, the result was read using BU¨HLMANN Quantum Blue Reader as indicated by the manufacturer. The lower range Quantum Blue calprotectin has been developed to cover a calprotectin level range of 30 to 300 mg/g and the higher range has been developed to cover a calprotectin level range of 30 to 1800 mg/g. The intra-assay coefficient of variation of FC test is
