Leucine-rich alpha-2 glycoprotein for detecting small bowel lesions in Crohn’s disease: A critical review and the path forward

Abstract

The study by Ohno et al provides valuable insights into the role of leucine-rich alpha-2-glycoprotein (LRG) as a potential biomarker for identifying small bowel lesions in Crohn's disease (CD). However, several methodological challenges hinder its immediate use in clinical practice. Notably, the current research was retrospective, lacks comparative studies with fecal calprotectin, and did not provide long-term predictive data. Further prospective studies are needed to improve the applicability of LRG. Moreover, integrating LRG with additional biomarkers and employing artificial intelligence techniques may improve its effectiveness in disease monitoring. Future research should address interobserver variability, assess LRG's cost-effectiveness, and standardize endoscopic healing definitions to ensure broader applicability. Advancing these areas is vital for establishing LRG's role in precision medicine strategies for the management of CD.

Key Words: Crohn's disease; Leucine-rich alpha-2 glycoprotein; Biomarkers; Small bowel lesions; Inflammatory bowel disease; Disease monitoring; Precision medicine

Core Tip: A recent study by Ohno et al indicated that leucine-rich alpha-2 glycoprotein (LRG) has potential as a biomarker for identifying small bowel lesions in Crohn's disease. However, several methodological challenges are required to be addressed. The retrospective nature, the lack of direct comparisons with fecal calprotectin, and the absence of long-term data highlight the need for further validation. Moreover, there is potential for improving the utility of LRG by integrating it with other biomarkers and artificial intelligence to improve its effectiveness in disease monitoring. Future research should address interobserver variability, assess the cost-effectiveness, and standardize definitions for endoscopic healing.

TO THE EDITOR

Crohn's disease (CD) is a chronic inflammatory bowel disease (IBD) that can cause transmural inflammation throughout any part of the gastrointestinal tract[1]. One of the key treatment objectives is achieving endoscopic healing (EH), which is linked to improved long-term outcomes, such as decreased hospitalization rates, fewer surgeries, and improved quality of life for patients[2]. However, accurately assessing EH, particularly in the small intestine, presents challenges due to the limitations of existing biomarkers like C-reactive protein (CRP) and fecal calprotectin (FC)[3]. Recently, leucine-rich alpha-2 glycoprotein (LRG), a 50-kilodalton protein produced in response to inflammatory conditions, has garnered attention as a potential biomarker for monitoring disease activity in CD[4]. Emerging research indicates a strong correlation between LRG levels and endoscopic and clinical activity in CD, particularly concerning small bowel lesions[1,4]. However, its effectiveness in predicting EH compared to CRP has not been thoroughly investigated.

LRG shows promise for application in several clinical contexts, particularly in the early detection and monitoring of small bowel lesions associated with CD. It can complement endoscopic evaluations, offering a more comprehensive assessment of disease activity[2]. For patients with suspected CD who have inconclusive imaging or endoscopic results, LRG levels can serve as an additional diagnostic tool, aiding in the direction of further investigations. Furthermore, LRG has the potential to be incorporated into routine follow-up protocols to monitor treatment responses and predict the likelihood of disease relapse. By delivering real-time insights into disease activity, LRG can assist clinicians in tailoring treatment strategies to meet the specific needs of individual patients, thereby enhancing overall patient outcomes.

In observance of this, we read with great interest the manuscript by Ohno et al[5] entitled "Leucine-rich alpha-2 glycoprotein as a superior biomarker to CRP for detecting small bowel lesions in Crohn's disease". The study studies an important clinical question regarding the utility of LRG as a biomarker for detecting small bowel lesions in CD, particularly in comparison to the widely used CRP. The study presented important preliminary data suggesting that LRG may be more effective than CRP in identifying small bowel lesions in CD. These findings show a basis for future research and emphasize the necessity for additional validation of LRG as a biomarker.

While the study provides valuable insights into the potential superiority of LRG over CRP in assessing EH in CD, several methodological, statistical, and interpretative limitations warrant discussion that need to be addressed to understand the clinical applicability of LRG in practice. By addressing the methodological and analytical limitations pointed out in this critique, subsequent studies can enhance the work of Ohno et al[5] and work towards establishing LRG as a dependable tool for managing CD.

Methodological limitations

The study was retrospective in nature, which could introduce the risk of selection bias. Despite the analysis of consecutive patients, the established exclusion criteria, including active perianal disease and varying disease conditions or treatments, may have resulted in the omission of patients with more severe or complex disease phenotypes. This limitation could affect the generalizability of the findings to the broader CD population, particularly those with comorbidities or more aggressive forms of the disease. We recommend that the authors consider adopting a prospective study design with clearly defined inclusion and exclusion criteria, thereby minimizing selection bias to improve upon this. Additionally, it would be beneficial to include a more diverse patient population, particularly those with perianal disease or other comorbidities, to enhance the external validity of the findings.

The current exclusion criteria did not consider systemic inflammatory conditions that could elevate LRG levels. Furthermore, this study only included one patient using steroids, restricting the ability to generalize the results to those on corticosteroids or other immunosuppressants. Future studies should consider stratifying patients based on their disease-modifying therapies and any coexisting inflammatory conditions to help mitigate this source of bias.

Another point of consideration is the median period of 18 days between LRG measurement and BAE, with a range extending up to 60 days. This delay may introduce variability in the correlation between LRG levels and endoscopic findings, as disease activity can fluctuate over time. Although the authors assert that this timeframe is similar to prior studies, it still represents a potential confounding factor. Future studies should minimize the interval between biomarker measurements and endoscopic evaluations to reflect disease activity during assessment better and increase accuracy. This study primarily concentrated on the ileum and colon, offering a limited assessment of the jejunum due to the retrograde approach of BAE. Considering that CD can affect any area of the gastrointestinal tract, excluding the jejunum could lead to an underestimation of the true extent of small bowel involvement and the efficacy of LRG as a biomarker[6]. Future research should utilize antegrade BAE or alternative imaging techniques, such as capsule endoscopy, to comprehensively evaluate the entire small bowel, including the jejunum[7].

Analytical concerns

The study's sample size of 133 participants was relatively small for conducting subgroup analyses, which may increase the risk of type. It is recommended that a power analysis be conducted and the sample size II errors be increased to ensure that future studies yield statistically meaningful comparisons.

Spearman's rank correlation coefficient was employed to evaluate the relationship between LRG and endoscopic activity, which is suitable for non-parametric data[8,9]. However, the analysis did not account for potential confounding factors such as medication usage (e.g., biologics, immunomodulators) or disease behavior (e.g., stricturing vs penetrating disease). These variables may affect LRG levels and endoscopic results, potentially introducing bias into the findings. Future research should utilize multivariate regression analysis to adjust for these confounding variables, mainly focusing on medication use and disease phenotype to more accurately determine the independent predictive value of LRG.

The receiver operating characteristic analysis indicated that LRG exhibited a higher area under the curve than CRP for predicting EH in the ileum. However, the study did not investigate the possible added value of combining LRG with other biomarkers, such as FC, which has been previously shown to correlate with small bowel disease activity. Moreover, the LRG cutoff value of 12.4 μg/mL was derived from a single cohort, raising questions about its generalizability to other populations. Exploring the combined use of LRG and FC or other biomarkers is important to enhance EH's prediction accuracy. Furthermore, external validation of the LRG cutoff value in independent cohorts is crucial to confirm its clinical utility[10].

While patients were categorized based on EH, the study did not fully account for disease severity at baseline, treatment history, or previous surgical interventions. These factors could significantly influence biomarker levels. Future research should integrate disease severity scoring and perform stratified analyses based on prior treatments to enhance the interpretation of results. Lastly, the study did not include FC, a well-established biomarker for CD. Without comparing LRG to FC, the clinical significance of LRG as a superior biomarker remains uncertain. Future studies should aim to compare LRG with FC and other emerging biomarkers to establish its relative efficacy.

Bias and result interpretation

The study involved three expert endoscopists who conducted the procedures; however, it is important to note that interobserver variability was not evaluated. This lack of assessment introduces a degree of subjectivity in the endoscopic evaluations, which may have influenced the findings[11]. Future research should employ standardized scoring methods and interobserver agreement analyses to reduce potential bias.

Additionally, the study presents cross-sectional data, which limits the ability to determine whether LRG levels can predict long-term clinical outcomes or treatment responses. Longitudinal studies are needed to understand better LRG’s role in predicting disease progression and treatment efficacy[12]. The patient population under study demonstrated significant heterogeneity concerning disease behavior, medication usage, and prior intestinal resections. For instance, the use of biologics differed notably between the EH group, with 83.6% receiving treatment, and the ileal group, at just 42.6%. This variability could potentially confound results, as biologics are known to affect both biomarker levels and EH. Future studies should stratify patients according to disease behavior and medication use to evaluate LRG's performance in more homogeneous subgroups.

The definition of EH as the absence of ulcerative lesions (mSES-CD < 3 points) is used in this study. However, it is worth noting that definitions of EH can differ across studies, with some employing stricter criteria (e.g., mSES-CD = 0). This variability in the definition may impact both the interpretation of the results and the generalizability of the findings. Standardizing definitions of EH, as suggested by international consensus guidelines, could facilitate comparison across different studies. Moreover, the study did not control for disease severity, duration, or previous surgical interventions, which can affect biomarker levels[13]. Variations in treatment regimens among different patient subgroups may also influence the results. Therefore, it is recommended that future studies include stratified analyses based on disease duration, phenotype, and treatment history.

Furthermore, the analysis relied on a single measurement of LRG and CRP, which might not adequately reflect the dynamic nature of disease activity in CD. Although three experts carried out the endoscopic scoring, the lack of reported interobserver variability raises concerns about potential measurement bias[14]. Future research should incorporate repeated biomarker measurements to account for fluctuations in disease activity and ensure that interobserver agreement for endoscopic scoring is reported to confirm the findings' reliability.

Finally, the study did not address the possibility of publication bias, especially given its positive findings on LRG's superiority over CRP. Negative or inconclusive results from similar studies may remain unpublished, potentially leading to an overestimation of LRG's effect size. Therefore, further analysis should be conducted to comprehensively assess the overall evidence for LRG as a biomarker in CD, including consideration of unpublished data to provide a more objective perspective on its clinical utility.

Future directions

Future studies need to be conducted across multiple centers, employing a prospective design that includes larger and more diverse patient populations to overcome the limitations identified in the current research. Such an approach would significantly improve the generalizability of the findings, enabling researchers to perform subgroup analyses that consider various factors, including disease phenotype, types of medication used, and other pertinent characteristics that could influence outcomes.

Moreover, future investigations should focus on integrating LRG with additional biomarkers such as FC and advanced imaging techniques like magnetic resonance enterography[15]. This multifaceted methodology could lead to a more holistic understanding of disease activity in CD, facilitating improved diagnostic precision and management strategies.

Longitudinal studies are also crucial to thoroughly assess LRG's potential as a biomarker for monitoring the progression of CD and evaluating the response to various treatment regimens over an extended period. Gaining insights from such studies would enhance our understanding of LRG's role within the treat-to-target strategy, ultimately improving patient therapeutic outcomes.

Additionally, the application of artificial intelligence (AI) holds promise for gastroenterology[16]. By utilizing AI-driven pattern recognition algorithms, researchers can refine predictive models for assessing disease activity, allowing for more accurate forecasts of disease course and better-informed clinical decisions[17,18]. Establish consistent definitions of EH to enhance the ability to compare results across different studies and streamline clinical decision-making processes. It is important to use an advanced statistical methodology, including sophisticated machine learning algorithms, to delve into extensive datasets and uncover innovative combinations of biomarkers that could significantly elevate the efficacy of disease monitoring[19].

Similarly, the cost-effectiveness of LRG, in contrast to traditional biomarkers, should be assessed to assess its viability for integration into routine clinical practice. Evaluating the cost-effectiveness of LRG as a biomarker for managing CD is another crucial area for further investigation for its potential implementation in clinical settings. Although LRG demonstrates promise in identifying small bowel lesions, it is important to assess its economic viability compared to established biomarkers like CRP and FC. This includes considering the costs associated with LRG testing, such as laboratory processing and necessary equipment, its diagnostic accuracy, and its potential to lower long-term healthcare expenses through enhanced disease monitoring and improved treatment outcomes. Future research should consider incorporating cost-benefit analyses to determine if LRG presents a cost-effective or complementary alternative to current biomarkers. Furthermore, it would be beneficial to examine how integrating LRG with other diagnostic modalities, including imaging techniques, may influence overall healthcare costs. Finally, explore the promise of LRG in steering personalized treatment approaches for CD, aligning with a treat-to-target strategy aimed at maximizing patient outcomes and ensuring tailored interventions.

CONCLUSION

The present study offered encouraging insights into the potential of LRG as a biomarker for detecting small bowel lesions in CD; it is essential to recognize several methodological and interpretative challenges that need to be addressed. Tackling these limitations and adopting the recommendations outlined previously can pave the way for future research to elucidate the true significance of LRG in CD management, ultimately enhancing patient care and outcomes. Furthermore, integrating cutting-edge AI-driven analytics and comprehensive multimodal biomarker evaluations is crucial for solidifying LRG's pivotal role in the intricate landscape of managing CD.