Portal de Periódicos da CAPES

Precision Medicine in Inflammatory Bowel Diseases: Challenges and Considerations for the Path Forward

2022; Elsevier BV; Volume: 162; Issue: 7 Linguagem: Inglês

10.1053/j.gastro.2022.02.049

1528-0012

Amanda Ricciuto, Ingrid Rauter, Dermot McGovern, Robert M. Mader, Walter Reinisch,

Pancreatic and Hepatic Oncology Research

Precision medicine strives to optimize health by delivering care tailored to each person's unique biology. The striking heterogeneity of inflammatory bowel disease (IBD) in phenotypic presentation, disease course, and treatment response renders it ideally suited to benefit from precision medicine approaches. Despite recent innovative research, implementations of truly tangible precision medicine applications in IBD are few. Although the past 20 years have ushered in a slew of novel therapies, our ability to use them in a rational, biology-driven fashion remains fairly crude. Although the limitations of a one-size-fits-all approach in IBD are widely acknowledged, clinical practice has struggled to move meaningfully beyond them. In this regard, the IBD community must cover substantial ground to catch up with other fields, such as oncology. Despite the important differences between cancer and IBD, including the clonal nature of malignancy, the IBD community can still glean important lessons from the successes of precision oncology. Here, we review some of the major barriers to progress in precision IBD and propose calls to action while drawing on precision oncology for insights. We have omitted certain generic challenges, such as the need for harmonization, global data sharing, and ethical/privacy considerations, which are reviewed elsewhere,1Denson L.A. Curran M. McGovern D.P.B. et al.Challenges in IBD research: precision medicine.Inflamm Bowel Dis. 2019; 25: S31-S39Crossref PubMed Scopus (39) Google Scholar,2Liu X. Luo X. Jiang C. et al.Difficulties and challenges in the development of precision medicine.Clin Genet. 2019; 95: 569-574Crossref PubMed Scopus (20) Google Scholar focusing on challenges specific to precision IBD (Figure 1). A fundamental first challenge is that current IBD diagnostic and classification schemes, such as the Lennard-Jones criteria and Montreal classification,3Silverberg M.S. Satsangi J. Ahmad T. et al.Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: report of a working party of the 2005 Montreal World Congress of Gastroenterology.Can J Gastroenterol. 2005; 19: 5A-36ACrossref PubMed Scopus (2350) Google Scholar do not capture the significant heterogeneity of IBD. Even the segregation of IBD into Crohn's disease (CD), ulcerative colitis (UC), and the ill-defined entity IBD-unclassified is a vast oversimplification of complex pathophysiologic processes. More than 240 genetic risk variants interact with epigenetic, immunologic, and microbial factors, under the influence of dynamic environmental exposures, to elicit the rich array of clinical manifestations of IBD.4Chang J.T. Pathophysiology of inflammatory bowel diseases.N Engl J Med. 2020; 383: 2652-2664Crossref PubMed Scopus (138) Google Scholar The narrow focus of existing diagnostic/classification systems on phenotypic features, such as disease location and behavior, and their constraint by categorical definitions of CD as discontinuous and transmural and of UC as continuous and mucosal prevent them from encompassing and meaningfully describing the full breadth of IBD. One potential approach is to reorganize the clinical criteria that make up existing classification systems. For example, a novel index including a subset of the Lennard-Jones criteria was shown to better distinguish CD from UC than the original criteria.5Reinisch S. Schweiger K. Pablik E. et al.An index with improved diagnostic accuracy for the diagnosis of Crohn's disease derived from the Lennard-Jones criteria.Aliment Pharmacol Ther. 2016; 44: 601-611Crossref PubMed Scopus (2) Google Scholar However, a solution that leverages molecular characterization will almost certainly be better equipped to capture and delineate the numerous distinct IBD subtypes. Work from the International IBD Genetics Consortium, which examined associations between genetic risk scores and IBD subphenotypes, supported a continuum of disorders better explained by three IBD subgroups (ileal CD, colonic CD, and UC) compared with the traditional 2 subgroups (CD and UC).6Cleynen I. Boucher G. Jostins L. et al.Inherited determinants of Crohn's disease and ulcerative colitis phenotypes: a genetic association study.Lancet. 2016; 387: 156-167Abstract Full Text Full Text PDF PubMed Scopus (432) Google Scholar Here, there are lessons to take from oncology's playbook. At comprehensive cancer centers, systematic molecular tumor characterization, especially for genetic aberrations using next-generation sequencing, is the standard of care. Malignancies are not solely defined by qualitative descriptors but also by their genetic architecture. In many cases, these molecular signatures can be used to risk-stratify patients and to select biology-specific treatment regimens.7Malone E.R. Oliva M. Sabatini P.J.B. et al.Molecular profiling for precision cancer therapies.Genome Med. 2020; 12: 8Crossref PubMed Scopus (187) Google Scholar The success of this approach in oncology has hinged on the availability of large numbers of novel targeted drugs; in 2020 alone, 21 therapeutic compounds were approved by the US Food and Drug Administration.8New oncology drug approvals in 2020.Clinical Oncology News. December 15, 2020; (Available at:) (Accessed December 11, 2021)https://www.clinicaloncology.com/FDA-Watch/Article/12-20/New-Oncology-Drug-Approvals-in-2020/61464Google Scholar It is time to move beyond our current phenotypically driven IBD classification to outline the molecular blueprint of IBD and to reveal all its subtypes, each likely with a distinct disease trajectory and treatment response profile. Thus, molecularly defined subphenotypes can be explored for their ability to predict prognosis and likelihood of treatment response, which are central goals in precision IBD. Although past studies have largely sought to identify biomarkers that align with conventional IBD types/phenotypes (eg, CD vs UC, ileal vs colonic CD), we must instead define novel IBD subtypes that align with molecular biomarkers. Important tools to accomplish this will include unsupervised clustering methods9Shen R. Mo Q. Schultz N. et al.Integrative subtype discovery in glioblastoma using iCluster.PLoS One. 2012; 7e35236Crossref Scopus (162) Google Scholar and systems biology methods that conceptualize IBD pathogenesis as a network rather than linear interactions, thus emphasizing nodes amenable to therapeutic interventions. Broad racial/ethnic representation and longitudinal assessment will be vital to capture the heterogeneity of IBD. Any novel biomarkers or precision medicine tools, whether for classification or other purposes (eg, prognosis, treatment response), will need to be the product of processes that ensure reproducibility. Industry collaborations can facilitate standardized access to samples under monitored conditions. Classic endoscopic, histologic, and radiographic assessment will continue to play an important, but perhaps complementary, role in IBD classification. Emerging approaches, such as artificial intelligence (AI) methods, including computer vision and deep learning/neural networks, can help to automate, standardize, and enhance the reproducibility of these traditional assessments and aid in identifying novel subgroups by detecting patterns not previously discernible.10Chahal D. Byrne M.F. A primer on artificial intelligence and its application to endoscopy.Gastrointest Endosc. 2020; 92: 813-820Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar,11van der Laak J. Litjens G. Ciompi F. Deep learning in histopathology: the path to the clinic.Nat Med. 2021; 27: 775-784Crossref PubMed Scopus (64) Google Scholar The big data multiomics approach advocated here is not without its own inherent challenges, including the small effect sizes of many analytic measures, such as the majority of genetic variants,12McGovern D.P.B. Kugathasan S. Cho J.H. Genetics of inflammatory bowel diseases.Gastroenterology. 2015; 149: 1163-1176Abstract Full Text Full Text PDF PubMed Scopus (221) Google Scholar relative to intersubject heterogeneity. This mandates processes to reconcile statistical and clinical significance (one should not be equated with the other) and to integrate numerous—at times, seemingly disparate—effects into a cohesive model that explains biological phenomena. In addition, although current technologies allow us to probe many multiomics facets of IBD, modalities to quantify the exposome remain limited—a real challenge given the very significant effect of the environment on IBD pathophysiology and likely on treatment response. A second challenge is the lack of criterion standard, precision medicine–appropriate endpoints. Such endpoints must be objective, clearly defined, reproducible, attainable within a sensible timeframe, and deeply rooted in disease biology and, thus, causally associated with clinically meaningful long-term outcomes. Clinical activity indices, despite being highly relevant for patients and regulatory agencies, are not suitable, given their subjective nature and imperfect correlation with inflammatory activity. Even seemingly "hard" outcomes, such as surgery, hospitalization, and treatment escalation, are suboptimal because they lack standardized definitions and may reflect physician practices and/or patient preferences. Regulatory agencies are increasingly embracing objective outcomes, such as endoscopy and histology in trial design, but have simultaneously mandated patient-reported outcomes (PROs) as coprimary endpoints.13Williet N. Sandborn W.J. Peyrin-Biroulet L. Patient-reported outcomes as primary end points in clinical trials of inflammatory bowel disease.Clin Gastroenterol Hepatol. 2014; 12: 1246-1256Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar Although the importance of the patient experience cannot be undermined, PROs capture a construct driven not only by inflammation but by myriad other influences (eg, anatomy, diet, microbiome, functional processes). As such, the juxtaposition of PROs and endpoints like mucosal healing dilutes the objectivity of the latter. The differential biological determinants of clinical activity and endoscopic healing are illustrated by the PROgECT trial, in which a colonic gene expression signature predicted mucosal healing in patients with UC treated with golimumab while showing no association with clinical response or remission.14Telesco S.E. Brodmerkel C. Zhang H. et al.Gene expression signature for prediction of golimumab response in a phase 2a open-label trial of patients with ulcerative colitis.Gastroenterology. 2018; 155: 1008-1011Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar An additional unknown is the optimal timepoint at which to assess outcomes for each therapy, reflecting our incomplete understanding of pharmacodynamics. We must also account for variability in drug exposure because this can obscure the interpretation of changes in molecular and cellular pharmacodynamic markers of treatment response. A fundamental first step is for the IBD community to achieve consensus on a core set of precision medicine–appropriate endpoints to be systematically applied across studies. Although overlap is to be expected with endpoints identified through initiatives such as Selecting Therapeutic Targets in IBD (STRIDE),15Turner D. Ricciuto A. Lewis A. et al.STRIDE-II: an update on the Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE) initiative of the International Organization for the Study of IBD (IOIBD): determining therapeutic goals for treat-to-target strategies in IBD.Gastroenterology. 2021; 160: 1570-1583Abstract Full Text Full Text PDF PubMed Scopus (274) Google Scholar discrepancies are anticipated (regarding PROs, for example), which must be reconciled moving forward. Although endoscopic, histologic, and radiographic endpoints are likely the most objective and appropriate of currently available outcome measures, they are far from perfect. None encapsulates the totality of the concepts of inflammatory burden, disease severity, and prognosis, and all have operational challenges, such as optimal indices/cutoffs and training standards.16Gottlieb K. Daperno M. Usiskin K. et al.Endoscopy and central reading in inflammatory bowel disease clinical trials: achievements, challenges and future developments.Gut. 2021; 70: 418-426PubMed Google Scholar As previously alluded to, AI may facilitate the reproducible interpretation of image-based outcomes and assist with the development of novel instruments. Ultimately, the IBD community must develop novel endpoints more tailored to the needs of precision medicine. Molecular biomarkers (eg, a molecular inflammatory score based on intestinal gene expression)17Argmann C. Tokuyama M. Ungaro R.C. et al.Molecular characterization of limited ulcerative colitis reveals novel biology and predictors of disease extension.Gastroenterology. 2021; 161: 1953-1968Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar warrant exploration because they have a more direct relationship to disease biology. Peripheral biomarkers from blood, stool, urine, etc represent the holy grail because their easy collection allows more frequent monitoring and treatment adjustment. In oncology, bioengineering advances in micro- and nanofluidics have enabled peripheral blood-based biomarkers, such as liquid biopsies, from which circulating tumor cells can be isolated. Biosensors and wearables offer the advantage of continuous monitoring with instantaneous feedback. Such emerging technologies hold great promise to revolutionize the way in which we measure treatment success in IBD.18Ho D. Quake S.R. McCabe E.R.B. et al.Enabling technologies for personalized and precision medicine.Trends Biotechnol. 2020; 38: 497-518Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar A major limitation of studies to date is the lack of longitudinal assessment. Prospective, longitudinal studies are needed to validate prognostic, predictive, and outcome biomarkers. In addition, longitudinal serial assessment is necessary to elucidate the temporal evolution of biomarkers in relation to disease stage and severity, as well as treatment response. A common shortcoming of studies has been a failure to account for patient differences in disease duration/severity and treatment, all of which are likely to affect disease biology and molecular profiles. Prospective longitudinal follow-up of inception cohorts, with initial sampling of treatment-naive patients and subsequent repeated sampling as the disease evolves and is treated, is likely the design best suited to address this. Serial sampling is vital because short-term changes in biomarkers may better predict long-term outcomes than single baseline parameters.19Waljee A.K. Liu B. Sauder K. et al.Predicting corticosteroid-free endoscopic remission with vedolizumab in ulcerative colitis.Aliment Pharmacol Ther. 2018; 47: 763-772Crossref PubMed Scopus (49) Google Scholar,20Waljee A.K. Liu B. Sauder K. et al.Predicting corticosteroid-free biologic remission with vedolizumab in Crohn's disease.Inflamm Bowel Dis. 2018; 24: 1185-1192Crossref PubMed Scopus (33) Google Scholar Although inception cohort studies are the criterion standard, they are hugely resource intensive. Accordingly, we must strive to develop infrastructure and protocols to accomplish similar objectives by leveraging clinical care processes. The widespread implementation of electronic health records (EHRs) can greatly facilitate the use of routinely collected health data to ascertain longitudinal phenotypes, exposures, and outcomes, variables for which we must establish standardized definitions to optimize efficiency and accuracy. Coupling this approach to standardized biospecimen collection at key timepoints during routine follow-up presents an opportunity to democratize precision medicine studies to include more community-based centers. This recruitment strategy increases the potential "audience," enhancing study power, and, importantly, helps include a more representative spectrum of IBD severity. Powerful tools result from the union of EHRs and machine learning, particularly deep learning. Deep learning is well suited for medical data because it can identify patterns in sparse, noisy data and requires little input-feature engineering.21Norgeot B. Glicksberg B.S. Butte A.J. A call for deep-learning healthcare.Nat Med. 2019; 25: 14-15Crossref PubMed Scopus (98) Google Scholar Nelson et al22Nelson C.A. Butte A.J. Baranzini S.E. Integrating biomedical research and electronic health records to create knowledge-based biologically meaningful machine-readable embeddings.Nat Commun. 2019; 10: 3045Crossref PubMed Scopus (17) Google Scholar have developed a method to embed clinical features from EHRs into a heterogeneous knowledge network, termed the Scalable Precision Medicine Oriented Knowledge Engine (SPOKE), which integrates data from 29 publicly available databases, effectively creating medically relevant "barcodes" for individual medical variables in EHRs that map onto SPOKE. These high-dimensional, knowledge-guided patient health signatures have been used as features in random forest modeling to recognize prodromal features of disease, such as multiple sclerosis, several years before diagnosis.23Nelson C.A. Bove R. Butte A.J. et al.Embedding electronic health records onto a knowledge network recognizes prodromal features of multiple sclerosis and predicts diagnosis.J Am Med Inform Assoc. 2022; 29: 424-434Crossref PubMed Scopus (3) Google Scholar Norgeot et al24Norgeot B. Glicksberg B.S. Trupin L. et al.Assessment of a deep learning model based on electronic health record data to forecast clinical outcomes in patients with rheumatoid arthritis.JAMA Network Open. 2019; 2e190606Crossref PubMed Scopus (76) Google Scholar have also leveraged machine learning and structured EHR data to predict future outcomes in patients with rheumatoid arthritis. The concept of repeated biological sampling, recognizing spatiotemporal disease evolution, is well established in precision oncology. When a patient's cancer relapses, its molecular underpinnings are assumed to have shifted; the tumor is resampled and reanalyzed for molecular alterations, and therapies are adjusted accordingly. This is possible because drugs have been developed that directly target specific aberrations. Similarly, molecular characterization at a single timepoint for the purpose of treating IBD is potentially insufficient and may need to be repeated at the time of relapse. For example, such repeated sampling can help elucidate the molecular shift in patients previously responsive to an antitumor necrosis factor who have lost the response to this mechanism of action despite adequate drug exposure and the absence of immunogenicity. A significant translational gap separates precision medicine discovery and implementation in IBD. A leading contributing factor is the current model of discovery by siloed academic groups. Although this has led to numerous novel and groundbreaking discoveries, it is not a pathway that typically culminates in biomarkers that are robustly validated, reproducible according to technical standards, approved by regulatory bodies, and widely available and reimbursed, all prerequisites for global acceptance and implementation. Oncology again leads by example, having developed a model that incorporates biomarkers into drug development programs, including companion diagnostics. As an example, pembrolizumab was shown to be efficacious for non–small-cell lung cancer tumors that express the marker programmed death ligand 1 (PD-L1) in ≥50% of cells.25Reck M. Rodríguez-Abreu D. Robinson A.G. et al.Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer.N Engl J Med. 2016; 375: 1823-1833Crossref PubMed Scopus (5718) Google Scholar In this randomized controlled trial, PD-L1 was measured using a commercially available companion diagnostic.25Reck M. Rodríguez-Abreu D. Robinson A.G. et al.Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer.N Engl J Med. 2016; 375: 1823-1833Crossref PubMed Scopus (5718) Google Scholar As a result, both pembrolizumab and its companion diagnostic have received regulatory approval for this indication. To accommodate the inclusion of biomarkers, oncology trials have undergone a major shift in design. Several novel biomarker-focused trial designs have been developed and are reviewed elsewhere.26Hu C. Dignam J.J. Biomarker-driven oncology clinical trials: key design elements, types, features, and practical considerations.JCO Precis Oncol. 2019; 3 (PO.19.00086)Google Scholar By comparison, biomarker-driven trials remain rare in IBD. Although industry-partnered research programs that intertwine drug and biomarker development, with a focus on biomarker-driven trials, represent one arm of the solution, much work remains on the clinical implementation side. Decision support tools are needed to guide clinicians on which tests to perform and how to apply results. One example of this is the decision tool incorporating clinical, genetic, and serologic markers for predicting disease progression in CD.27Siegel C.A. Horton H. Siegel L.S. et al.A validated web-based tool to display individualised Crohn's disease predicted outcomes based on clinical, serologic and genetic variables.Aliment Pharmacol Ther. 2016; 43: 262-271Crossref PubMed Scopus (66) Google Scholar Molecular tumor boards, interdisciplinary meetings during which experts achieve consensus on how to translate complex biological data into concrete clinical recommendations, have been widely adopted in oncology. An analogous approach may be of great utility in IBD, and virtual communication platforms can facilitate access to necessary experts. Mechanisms must also be put in place to migrate these data to EHRs so that they are available for health care providers and patients to make informed decisions on biomarker-aided clinical strategies. Feasibility and cost-effectiveness are not challenges to be diminished. Cost-effectiveness studies are, in fact, entirely lacking in the precision IBD space and represent an important area in need of future research. The costs of bringing precision medicine applications to the clinic can be colossal and prohibitive in some cases. This is especially true because large randomized controlled trials are the criterion standard to establish effectiveness before implementation. Nevertheless, a whole blood gene expression profile that has been associated with more severe CD is being used to stratify patients to a more aggressive treatment approach vs the standard of care in a randomized controlled trial in the first study of its kind28Biasci D. Lee J.C. Noor N.M. et al.A blood-based prognostic biomarker in IBD.Gut. 2019; 68: 1386-1395Crossref PubMed Scopus (78) Google Scholar,29Lee J.C. Lyons P.A. McKinney E.F. et al.Gene expression profiling of CD8+ T cells predicts prognosis in patients with Crohn disease and ulcerative colitis.J Clin Invest. 2011; 121: 4170-4179Crossref PubMed Scopus (201) Google Scholar (discussed later). That said, certain interventions where the balance is highly weighted toward benefit over risk and cost may be suitable for implementation even without costly randomized controlled trials. Examples that may be ready for prime time include screening for NUDT15 variants, which are associated with thiopurine-induced myelosuppression, before medication initiation, particularly in patients of East Asian ancestry as well as the use of HLA-DQA1∗05–based assays to determine people at high risk of developing immunogenicity to anti–tumor necrosis factor (TNF) therapies.30Walker G.J. Harrison J.W. Heap G.A. et al.Association of genetic variants in NUDT15 with thiopurine-induced myelosuppression in patients with inflammatory bowel disease.JAMA. 2019; 321: 773-785Crossref PubMed Scopus (73) Google Scholar,31Sazonovs A. Kennedy N.A. Moutsianas L. et al.HLA-DQA1∗05 carriage associated with development of anti-drug antibodies to infliximab and adalimumab in patients with Crohn's disease.Gastroenterology. 2020; 158: 189-199Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar The success of precision oncology would not have been possible without access to drugs with multiple mechanisms of action. This will likely be true in IBD as well. Fortunately, numerous novel biologics and small molecules are in the IBD pipeline. However, with increasing therapeutic options comes the threat of careless cycling through medications without attempts to optimize drug performance. Even with the handful of currently available agents, it is likely that we can substantially improve outcomes by optimizing their use. To maximize drug optimization, we must deepen our limited understanding of drug mechanisms of action, including counterregulatory processes. In a study published more than 20 years ago, almost all 24 patients with CD treated with a single infliximab infusion achieved clinical remission at 1 week, and TNF secretion was undetectable within a day, suggesting almost universal efficacy of the anti-TNF mechanism.32Nikolaus S. Raedler A. Kühbacker T. et al.Mechanisms in failure of infliximab for Crohn's disease.Lancet. 2000; 356: 1475-1479Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar However, all patients subsequently experienced a recurrence of symptoms at variable timepoints, preceded by a rise in TNF. Greater insight into the processes that prompt a loss of response in specific patients at specific timepoints, potentially even very early after treatment initiation, could enable individualized drug regimens that prevent such loss of response. Between 2008 and 2013, 26% of oncology trials investigated combination therapy to maximize efficacy, compared to <7% of nononcology trials.33Wu M. Sirota M. Butte A.J. Chen B. Characteristics of drug combination therapy in oncology by analyzing clinical trial data on ClinicalTrials.gov.Pac Symp Biocomput. 2015; : 68-79PubMed Google Scholar We have repeatedly witnessed a ceiling effect in IBD clinical trials; therapeutic regimens that combine multiple biologics and/or small molecules may represent one way to break through. Such combination therapy appears intuitive when considering the multifactorial nature of IBD and the intricacies and redundancies of the immune system. Illustrating the concept of redundancy, using single-cell RNA sequencing, Smillie et al34Smillie C.S. Biton M. Ordovas-Montanes J. et al.Intra- and inter-cellular rewiring of the human colon during ulcerative colitis.Cell. 2019; 178: 714-730Abstract Full Text Full Text PDF PubMed Scopus (316) Google Scholar found that oncostatin M phenocopies TNF in UC and that this might be one of the mechanisms of anti-TNF resistance, a finding that might inform subsequent treatment strategies. The rising number of drugs greatly complicates the task of treatment selection, especially when considering combination regimens. AI-based methods could assist. One example is personalized phenotypic medicine, which can be used to predict optimal drug dosing for efficacy and safety using relatively few data points from the treated patient only.18Ho D. Quake S.R. McCabe E.R.B. et al.Enabling technologies for personalized and precision medicine.Trends Biotechnol. 2020; 38: 497-518Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar Throughout this commentary, we have turned to precision oncology as a paradigm to inform steps forward in IBD. Although we acknowledge that cancer may be uniquely suited to precision medicine applications given its clonal nature and the chance to identify and to therapeutically address molecular drivers, we remain convinced that there are important lessons to learn from our oncology colleagues and that it is precisely the differences between cancer and IBD that allow these lessons to stand out. Other chronic immune-mediated diseases may appear to be more intuitive comparators, but they suffer from many of the same challenges as IBD. That said, conditions such as the inflammatory arthritides have witnessed important successes on the precision medicine front, some of which may have direct applications to IBD, such as markers of treatment response or prognostic markers. As an example, PreductSURE IBD,35PredictImmune. PredictSURE IBD.https://www.predictimmune.com/predictsureibd/Google Scholar a commercially available 17-gene quantitative polymerase chain reaction–based blood assay intended to identify IBD patients at high risk of early and frequent relapse, is based on a T-cell exhaustion signature that was first linked to prognosis in autoimmune conditions such as systemic lupus erythematosus and anti-neutrophilic cytoplasmic autoantibody–positive vasculitis.36McKinney E.F. Lee J.C. Jayne D.R.W. et al.T-cell exhaustion, co-stimulation and clinical outcome in autoimmunity and infection.Nature. 2015; 523: 612-616Crossref PubMed Scopus (0) Google Scholar The availability of multiomics data and computational proficiency will continue to grow exponentially over the coming years, extending the boundaries of what is achievable for IBD precision medicine. A concerted effort by the IBD community to overcome the obstacles discussed here is necessary to achieve the full potential of precision medicine for patients with IBD as it concerns disease classification/prediction, endpoints, longitudinal assessment, translational commitment, and therapeutic insights. These efforts must engage all parts of the IBD family, including patients, clinicians, and scientists, and should embrace advances from other disciplines even beyond inflammation/immunity to substantiate and ensure wide reach of the benefits of precision IBD.