Toward more accurate endoscopic diagnosis of colorectal neoplasm: Balancing between innovation and applicability

Image enhanced endoscopic (IEE) technology was developed to enhance images of mucosal changes of the gastrointestinal tract [1] . IEE is widely used currently for the detection and diagnosis of esophageal squamous cell carcinoma, Barrett’s esophagus, colorectal polyps and gastric intestinal metaplasia [2] ; [3] ; [4] ; [5]  ;  [6] . Colonoscopy has proven effective in reducing colorectal cancer (CRC) incidence and mortality [7] . It is currently one of the most widely applied endoscopic procedures, not only in western countries, but also in Taiwan and other Asian countries. Its increased use corresponds to the increasing incidence of CRC and the related increased popularity of organized CRC screening programs, or opportunistic screening using primary screening colonoscopy [8] . Detection, accurate diagnosis and proper management of detected lesions testifies to the high effectiveness of the CRC screening program and minimizes the related cost, making any new modality that may contribute to these aspects worthy of investigation [9] .

In this issue of AIDM, Cheng and colleagues have demonstrated that the net change of index of hemoglobin (IHb) values between normal mucosa and colon polyps can be used to differentiate non-neoplastic polyps (hyperplastic polyps) from neoplastic lesions, including adenoma, advanced adenoma and cancer. However, even though this study has demonstrated significant differences in IHb between non-neoplastic lesions, benign neoplastic lesions, and cancerous lesions, several issues must be addressed. First, in optical diagnosis of colorectal polyps using narrow band imaging (NBI), one of the most popular IEE modalities, high-confidence predictions can provide higher than >90% sensitivity and NPV [10] . In this context, any newly developed modality must be compared with this high standard, not only the test accuracy but the convenience of use as well. The current NBI platform is handy and convenient, requiring only one-touch operation with minimal time spent on using it in ordinary colonoscopy. However, inter-observer discrepancy has been noted for differentiating neoplastic from non-neoplastic lesions by NBI, and therefore, it would be helpful if any new modality, including IHb, could offer quantitative measurement and objective evaluation of a lesion to reduce chances of inaccuracy by human interpretation [11] . Second, the Cheng study used still images for IHb analysis, which is contrary to use in the real setting. If this technology could further incorporate artificial intelligence technology and automatic mechanical function and interpretation function in order to provide real-time probability or likelihood of a certain diagnosis, its application in the real clinical setting could be anticipated [12]  ;  [13] . Third, it is not clear from this study whether this function could be used to guide treatment, especially for suspicious cancerous lesions. Although the ability to differentiate neoplastic from non-neoplastic lesions can help to avoid over-treatment (treating hyperplastic polyps) and under-treatment (leaving neoplastic lesions untreated), it is not clear whether it can be used to predict invasion depth in cancerous lesions. Moreover, only three cancerous lesions were included in this study, including two T1 cancers and one T2, thus its validity is unclear. As for T2 cancer, for which surgery rather than endoluminal therapy is the optimal treatment, it can be diagnosed easily even just using conventional white-light endoscopy, so the added diagnostic value of applying additional imaging technology is limited. Currently, either NBI or chromoendoscopy with concurrent use of magnifying observation is used to predict invasion depth of cancerous lesions with accuracy as high as 90%. This is of utmost importance, as curative resection can be achieved with endoluminal therapy (EMR or ESD) for mucosal cancer (carcinoma in situ) or shallow invasive cancer, which may significantly reduce the cost and improve the quality of life compared with surgery. Further study on this issue using IHb is mandatory. Fourth, it is not clear whether lesion size may have affected the diagnostic accuracy of IHb. In screening detected neoplasms, almost 90% are sub-centimetric lesions, thus the accuracy of diagnosing small or diminutive lesions becomes important. Take NBI for instance—its diagnostic accuracy for diagnosing small lesions is lower than that for large lesions because some diminutive polyps have only a paucity of capillary network in their mucosal layer. Therefore, its endoscopic appearance is sometimes rather ambiguous. Thus, further study on small lesions may be helpful. Finally, establishing a platform for clinical use is important. Time and ease of use have a large impact on cost-effectiveness, which determines the popularity of a clinical tool, regardless of its proven diagnostic accuracy. This issue is of utmost importance, especially in the era of constrained medical resources and financial difficulty.

References

1. [1] T. Kaltenbach, Y. Sano, S. Friedland, et al.; American Gastroenterological Association (AGA) Institute technology assessment on image-enhanced endoscopy; Gastroenterology, 134 (2008), pp. 327–340
2. [2] M. Muto, K. Minashi, T. Yano, et al.; Early detection of superficial squamous cell carcinoma in the head and neck region and esophagus by narrow band imaging: a multicenter randomized controlled trial; J Clin Oncol, 28 (2010), pp. 1566–1572
3. [3] P. Sharma, J.J. Bergman, K. Goda, et al.; Development and Validation of a Classification System to Identify High-Grade Dysplasia and Esophageal Adenocarcinoma in Barretts Esophagus Using Narrow-Band Imaging; Gastroenterology, 150 (2016), pp. 591–598
4. [4] N. Uedo, R. Ishihara, H. Iishi, et al.; A new method of diagnosing gastric intestinal metaplasia: narrow-band imaging with magnifying endoscopy; Endoscopy, 38 (2006), pp. 819–824
5. [5] H.M. Chiu, C.Y. Chang, C.C. Chen, et al.; A prospective comparative study of narrow-band imaging, chromoendoscopy, and conventional colonoscopy in the diagnosis of colorectal neoplasia; Gut, 56 (2007), pp. 373–379
6. [6] K.I. Fu, Y. Sano, S. Kato, et al.; Chromoendoscopy using indigo carmine dye spraying with magnifying observation is the most reliable method for differential diagnosis between non-neoplastic and neoplastic colorectal lesions: a prospective study; Endoscopy, 36 (2004), pp. 1089–1093
7. [7] R. Nishihara, K. Wu, P. Lochhead, et al.; Long-term colorectal-cancer incidence and mortality after lower endoscopy; N Engl J Med, 369 (2013), pp. 1095–1105
8. [8] H.M. Chiu, S.L. Chen, A.M. Yen, et al.; Effectiveness of fecal immunochemical testing in reducing colorectal cancer mortality from the One Million Taiwanese Screening Program; Cancer (2015)
9. [9] C. Hassan, D.K. Rex, A. Zullo, et al.; Loss of efficacy and cost-effectiveness when screening colonoscopy is performed by nongastroenterologists; Cancer, 118 (2012), pp. 4404–4411
10. [10] S.K. McGill, E. Evangelou, J.P. Ioannidis, et al.; Narrow band imaging to differentiate neoplastic and non-neoplastic colorectal polyps in real time: a meta-analysis of diagnostic operating characteristics; Gut, 62 (2013), pp. 1704–1713
11. [11] U. Ladabaum, A. Fioritto, A. Mitani, et al.; Real-time optical biopsy of colon polyps with narrow band imaging in community practice does not yet meet key thresholds for clinical decisions; Gastroenterology, 144 (2013), pp. 81–91
12. [12] Y. Takemura, S. Yoshida, S. Tanaka, et al.; Computer-aided system for predicting the histology of colorectal tumors by using narrow-band imaging magnifying colonoscopy (with video); Gastrointest Endosc, 75 (2012), pp. 179–185
13. [13] Y. Mori, S.E. Kudo, K. Wakamura, et al.; Novel computer-aided diagnostic system for colorectal lesions by using endocytoscopy (with videos); Gastrointest Endosc, 81 (2015), pp. 621–629