Personalized Medicine in IBD: Where Are We in 2013 David A. Schwartz, MD Director, Inflammatory Bowel Disease Center Associate Professor of Medicine Vanderbilt University Medical Center What is Personalized Medicine? Personalized medicine: Use of patient s own information (genetic or other molecular biomarker information) to improve the safety, effectiveness and health outcomes via more efficiently targeted risk stratification, prevention and tailored treatment management approaches Examples: HER2/neu: Herceptin - Herceptin, Tamoxifen, hormonal therapies PGx Predict: warfarin dosing Examples in IBD: TPMT : azathioprine dosing 1
Examples of Uses for Personalized Medicine 1) Risk stratification Patient MORE or LESS likely to develop disease/condition? 2) Inform treatment selection Is it safe? Is it effective? 3) Inform dosage Rapid / Slow metabolizers 4) Prognostic testing How likely is the patient to respond to standard treatments? 5) Treatment monitoring Is it working? Should we switch therapies or treatment strategies? 6) Improve or optimize clinical treatment pathways WhydoweneedtopersonalizeIBD therapy? 2
Heterogeneity in IBD Patient examples: A) 25 yo male with new diagnosis and 10 stools per day. He has wt loss, and deep ileal ulcers on scope B) 42 yo male with new diagnosis, 4 stools per day and mild patchy colitis on scope Differences in Patient presentation Disease behavior Disease genetics and immune response Development of complications Response to therapy Complications of therapy Effectiveness of Step-up Treatment 3
Personalizing therapy Natural history Response to therapy Therapeutic Choice Adverse effects Can We Predict Disease Course? 4
Differing Rates of Progression Predictors of Disabling Crohn s Disease Factors significantly associated with disabling Crohn s disease within 5 years of diagnosis (n=1123) Factor OR 95% CI Steroid use 3.1 2.2 4.4 <40 years 2.1 1.3 3.6 Perianal disease 1.8 1.23 2.8 Positive Predictive Value # Factors % 0 61 1 67 2 91 3 93 CI = confidence interval Beaugerie et al. Gastroenterology. 2006; 130:650-6 OR = odds ratio 5
Predictors of Rapid Progression to Surgery Factor Odds Ratio (95% CI) Current smoker 3.1 (1.5 6.5) Abdominal pain 1.8 (1.1 3.2) Nausea/vomiting 2.1 (1.0 4.1) Ileal localization only 2.2 (1.3 3.8) Oral steroid use in 1 st 6months 3.8 (1.9 7.6) Sands et al., Am J Gastroenterol. 2003 Prognosis of CD Patients with Severe Ulcerations Retrospective study, 102 patients with active CD Severe endoscopic lesions (SEL) defined as deep ulcerations >10% of mucosal area with at least one colonic segment Risk of colectomy associated with SELs, high CDAI, absence of immunosuppression % Colectomy 100 80 60 40 20 0 6% 31% No SEL 8% 42% Years SEL 18% 62% 1 3 5 Allez M, et al. Am J Gastroenterol. 2002;97(4):947-53. 6
Ulcerative colitis - predictors Factors significantly associated colectomy within 10 years of diagnosis for pts with UC (n=519): Age > 50 years 0.28 (95% CI 0.12 0.65) Extensive colitis 2.98 (95% CI 1.25 7.08) ESR > 30 mm 2.94 (95% CI 1.58 5.46) Solberg et al.,scand J Gastro 2009 Antibodies and NOD2 Genotype Associated with Specific Crohn s Disease Phenotypes Small bowel disease Fibrostenosi s Internal Perforatin g Small Bowel Surgery UC-like disease Anti-I2 - P<0.027 - P=0.01 - Anti- - - P<0.006 - - OmpC ASCA P=0.023 P<0.001 P<0.001 P<0.001 P<0.001 a panca P<0.0130 013 a P<0.0020 002 a - P<0.0010 001 P<0.0010 001 NOD2 P<0.003 P<0.002 - P<0.001 P<0.008 a Anti-CBir1 P<0.018 P=0.05 P=0.008 - - 1. Mow WS, et al. Gastroenterology. 2004; 126:414-423. 2. Targan SR, et al. Curr Opin Gastroenterol. 2007;23:390-394. 3. Targan SR, et al. Gastroenterology. 2005;128:2020-2028. a 7
David A. Schwartz, MD, FACG Complications Increase with Number of Positive Serologies Dubinsky et al., Am J Gastro 2006 Dubinsky et al., Clin Gastro Hep, 2008 NOD2/CARD15 Gene Status and Need for Surgery Presence of NOD2/CARD15 variants was associated with: Cum mulative Survival 1.00 P=0.0089 0.8 More frequent surgery for stricture Earlier requirement of 0.6 surgery 0.4 More frequent surgical recurrence 0.2 No variant NOD2/CARD15 gene variants Earlier requirement for reoperation 0.0 0 20 40 60 80 100 120 Months Alvarez-Lobos M, et al. Ann Surg. 2005;242:693-700. 8
David A. Schwartz, MD, FACG Serology and Genetics Predictors of Disease Severity Lichtenstein et al., Inflamm Bowel Dis. 2011 Serology and Genetics Predictors of Disease Severity 9
ANCA Predicts Chance of Relapse in UC N=432 Hoie et al. J Crohn s Colitis,2008 Precolectomy panca Reactivity and Postoperative Risk of Pouchitis 42% 20% Fleshner PR, et al. Gut. 2001;49:671-677. 10
David A. Schwartz, MD, FACG Weighing The Value of Top-Down Therapy Pros Maintenance of i i remission Improved function and QOL Early promotion of mucosal healing to prevent complications Cons Side effects Cost Majority of patients may not require more potent treatments initially Under-treatment of most severe patients with ith episodic i di strategy? t t? Lichtenstein et al. Inflamm Bowel Dis. 2004; 10:S2-S10 Caprilli et al. Dig Liver Dis. 2005; 37:973-9 Can We Predict Response to Therapy? 11
Infliximab in Children Study (REACH) (ie-shorter Disease Duration) Median disease duration 2 years Response Remission 100 90 88 p=0.002 Patients (%) 80 70 60 50 40 30 20 10 59 64 n=99 n=66 n=33 n=29 n=17 56 33 P<0.001 24 n=12 0 Week 10 Week 54 q8 Week 54 q12 Overall number of subjects n=112 Hyams J, et al. Gastroenterology. 2007;132(3):863-873. Higher Remission Rates with ADA and CZP with Shorter Disease Duration Post-hoc Analyses % of Patients 100 90 80 70 60 50 40 30 20 10 0 Placebo All ADA 59 40 41 25 17 14 <2 years 2 to <5 years 5 years ** % IN CDA1 Response or Remiss sion100 90 80 70 60 50 40 30 20 10 0 Placebo All CZP 68 55 47 44 37.1 36.4 29.1 23.5 < 1 year 1 - < 2 years 2 - < 5 years 5 years Week 26 *p=0.002; **p<0.001; p=0.014; p=0.001; all vs placebo Schreiber S, et al. DDW 2007. #985. Sandborn, et al ACG 2007. 12
TPMT Testing 6-TG Levels and Clinical Response Dubinsky et al, Gastroenterology 2000 13
Clinical Characteristics as Predictors of Response to Biologic Therapy Positive Younger age Shorter disease duration Inflammatory phenotype Isolated colonic disease Negative Smoking Isolated ileitis Presence of intestinal stricture No effect Gender and race Albumin levels and platelet counts Serology Parsi et al. Gastroenterology. 2002; 132:707-13 Arnott et al, Aliment Pharmacol Ther. 2003; 17:1451-7 Vermeire et al, Am J Gastroenterol. 2002; 97:2357-63 Schnitzler et al, Gut. 2009; 58:492-500 Biomarkers and Likelihood of Respone Data is mixed. Pts can still respond to Biologic Treatment if not elevated but likelihood is reduced. P=0.012 Reinisch et al. APT 2012 14
David A. Schwartz, MD, FACG High Infliximab Levels are Associated with Mucosal Healing in Crohn s Disease Serum samples in 210 CD i d i treatment patients undergoing with infliximab were collected 6 Trou ugh level (mcg/ml) Infliximab trough levels were correlated with endoscopic healing (complete, partial or none) 7 5.77 5 4 3.89 3 2 1 0.95 0 Complete Partial None Van Moerkercke W. et al. DDW 2010. Abs #405b Van Moerkercke et al. SONIC: Clinical Remission Without Corticosteroids By Trough IFX Concentration at Week 26 Primary Endpoint 100 Patients (%) 80 60 59 57 19/32 13/23 73 74 72 43/59 36/49 31/43 40 20 0 0 >0-1 >1-3 >3-6 IFX Concentration (mg/ml) at Week 30 >6 Colombel JF, et al. N Engl J Med. 2010;362(15):1383-95. 15
Clinical Utility of Measuring Infliximab and HACA Levels in Patients with IBD Clinical Outcomes in Patients with Detectable HACA (N=35)* Clinical Outcomes in Patients with Subtherapeutic Concentrations (N=69)* P<0.004 P<0.016 (11/12) (1/6) (2/6) (25/29) Afif W, et al. Am J Gastroenterol. 2010;105(5):1133. ADA Levels and Anti-Adalimumab Antibodies (ATA) and Outcomes 66 patients: 27% with detectable ATA 59 (89%) CD, 7 (11%) UC ADA level 5 µg/ml is associated with lower CRP Conclusions: ADA level 5 μg/ml correlates with lower CRP and mucosal healing; Yarur, AJ, et al. Presented at DDW; May 21, 2013. Abstract Tu1147. 16
Anti-Drug Antibodies Associated With Treatment Failure N=76 N=196 The development of anti-drug antibodies are associated with poor treatment prognosis Bartelds G M, et al. JAMA. 2011;305(14):1460-1468. Assessing Response to Treatment 17
Efficacy Endpoints Clinical response CDAI decrease 70-100 points Clinical remission CDA <150 Decrease in CRP, stool calprotectin Mucosal healing Absence of mucosal ulceration on colonoscopy Deep remission Clinical remission plus mucosal healing Pineton de Chambrun G, et al. Nat Rev Gastroenterol Hepatol. 2010;7(1):15-29. Using Biomarkers to Assess Response to Therapy (Calprotectin and CRP) N=64 Bjorkesten et al. Scand J Gastro 18
David A. Schwartz, MD, FACG Symptomatic Recurrence Stratified According to Endoscopic Lesions Rutgeerts P, et al. Gastroenterology. 1990;99:956-963 Endoscopic Healing and Reduced Hospitalizations and Surgeries: Infliximab maintenance for Crohn s disease Rate of Hospitalizations and Surgeries (%) a 50 Patients with no healing (n=74) 46% Patients with healing at 1 visit (10 or 54 wk) (n=16) 40 30 Patients with healing at both 10 and 54 wks 25% 20 8% 10 (34*) (4*) 0 0% Hospitalization *Number per 100 patients (6*) 0% 0% Surgery Rutgeerts P et al. Gastroenterology. 2002;123(suppl):43.M2138. 19
Mucosal Healing Predicts Sustained Clinical Remission in Early CD % of Patients Baert FJ, et al. Gastroenterology, 2010;138(2):463 468. 3 Identifying Patients at Increased Risk Of Adverse Events 20
Long-Term Safety: Association Between Use of Specific Medications and Odds for Opportunistic Infection in IBD Medication OR (95% CI) 1 p value 2 No medication 1.0 (reference) Mesalamine 1.0 (0.6-1.6) NS Corticosteroids 3.3 (1.8-6.1) <.001 AZA/6MP 3.8 (2.0-7.0) <.001 Anti-TNF agent 4.4 (1.1-17).03 # of Immunosuppressants None 10( 1.0 (reference) 1 2.9 (1.5-5.3) <.001 2 or 3 14.5 (4.9-43) <.001 1 Odds ratio and 95% confidence intervals assessed by multiple variable conditional logistic regression. 2 P value using conditional logistic regression. Toruner M, et al. Gastroenterology. 2008; 134:929-936. Elderly IBD At Increased Risk For Severe Infection/Mortality with Anti-TNF Therapy 95 patients (37 U.C., 58 CD) over 65 of an IBD database of 2475 pts. given IFX. S.I. % Neo % Mort. % Elderly 11 3 10 Younger 0.5 2 2 Cottone et al., Clin Gastro Hep 2011 21
Risk of lymphoma 6 5.41 5 4 3 2 258 2.58 1.88 1.68 Current Past Never 1 0.66 0.37 0.4 0 0 0 < 50 years 50-65 years > 65 years Siegel et al., Gastro Hep 2009 HSTCL Risk in men younger than 35 years 1 in 7,404 AZA 1 in 45,000 1 in 3,534 AZA/IFX Thai et al., J Crohn s colitis 2010 Kotylar DS. Clin Gastroenterol Hepatol. 2011 Jan;9(1):36-41.e1. 22
Current Stratification of Natalizumab-Associated PML Risk Anti-JCV Antibody Status Negative Positive Prior IS Use 0.1/1,000 /, 95% CI 0 0.38 Natalizumab Exposure NO YES 1 24 months 0.6/1,000 patients (95% CI 0.4 0.9) 1.8/1,000 patients (95% CI 1.1 2.8) >24 months 52/1000 5.2/1,000 patients t 10.6/1,000 000 patients t 95% CI 4.3 6.2) (95% CI 8.1 13.8) Note: Low body mass has recently been identified as an additional potential risk factor for the development of PML IS = immunosuppressant; CI = confidence interval. Dong-Si T, et al. Presented at AAN; March 16 23, 2013; San Diego, CA. Abstract P04.271; Foley J. Presented at AAN; March 16 23, 2013; San Diego, CA. Abstract S30.002. Conclusion 23
Where We are Now with Personalized Medicine in IBD 1) Risk stratification yes = Clinical, serologic and genetic factors 2) Inform treatment selection Just Beginning: clinical factors but not much for genetics, etc 3) Inform dosage Evolving (TPMT for immunomods, Nothing yet for initial TNF dosage) 4) Prognostic testing Evolving: Clinical factors, biomarkers, genetics (too early to use) 5) Treatment monitoring Evolving: Endoscopy (yes), TNF levels (+/-), Biomarkers (+/-) 6) Improve or optimize clinical treatment pathways Evolving: Drug and metabolite levels 24