Advertisement
Original article Clinical endoscopy: Editorial| Volume 93, ISSUE 5, P1074-1076, May 2021

Download started.

Ok

Larger esophageal Lugol’s unstained lesions need more follow-up: size as an important predictor for risk of premalignant lesions

      Abbreviations:

      ESCC (esophageal squamous cell carcinoma), LCE (Lugol’s chromoendoscopy), LUL (Lugol’s unstained lesion), OR (odds ratio), SDA (severe dysplasia or above)
      Esophageal squamous cell carcinoma (ESCC) is the most common histologic subtype of esophageal cancer globally.
      • Bray F.
      • Ferlay J.
      • Soerjomataram I.
      • et al.
      Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.
      Mortality from this cancer remains high, mainly because of late presentation, and early detection and screening have continued to be challenging.
      • Codipilly D.C.
      • Qin Y.
      • Dawsey S.M.
      • et al.
      Screening for esophageal squamous cell carcinoma: recent advances.
      The current criterion standard for screening is Lugol’s chromoendoscopy (LCE), which identifies dysplastic areas that do not take up the stain.
      • Dawsey S.M.
      • Fleischer D.
      • Wang G.
      • et al.
      Mucosal iodine staining improves endoscopic visualization of squamous dysplasia and squamous cell carcinoma of the esophagus in Linxian, China.
      LCE is affordable and easy to perform, but it still requires endoscopic equipment and expertise with the technique. It has been shown, however, to be effective in identifying the dysplastic lesions and reducing mortality.
      • Dawsey S.M.
      • Fleischer D.
      • Wang G.
      • et al.
      Mucosal iodine staining improves endoscopic visualization of squamous dysplasia and squamous cell carcinoma of the esophagus in Linxian, China.
      • Wang G.
      • Abnet C.C.
      • Shen Q.
      • et al.
      Histological precursors of oesophageal squamous cell carcinoma: results from a 13 year prospective follow up study in a high risk population.
      • Wei W.Q.
      • Chen Z.F.
      • He Y.T.
      • et al.
      Long-term follow-up of a community assignment, one-time endoscopic screening study of esophageal cancer in China.

      Chen R, Liu Y, Song G, et al. Effectiveness of one-time endoscopic screening programme in prevention of upper gastrointestinal cancer in China: a multicentre population-based cohort study. Gut. Epub 2020 Apr 2.

      In this issue of Gastrointestinal Endoscopy, Liu et al
      • Liu M.
      • Zhou R.
      • Guo C.
      • et al.
      Size of Lugol-unstained lesions as a predictor for risk of progression in premalignant lesions of the esophagus.
      investigate the size of Lugol’s unstained lesions as an additional predictor for risk of premalignant lesions of the esophagus. They evaluated subjects from the screening arm of the Endoscopic Screening for Esophageal Cancer in China randomized screening trial, which block-randomized 668 villages based on population size to either a screening arm or a control (unscreened) arm. Each arm had approximately 16,000 subjects. In the screening arm, LCE was performed during a standard endoscopy, with biopsy specimens taken from all of Lugol’s unstained lesions (LULs) and standard biopsies performed at 28 cm and 33 cm if no LULs were seen.
      The 1058 subjects enrolled in the current analysis needed to have at least 2 LULs at the baseline endoscopy and no biopsy diagnosis of severe dysplasia, carcinoma in situ, or invasive ESCC, collectively called “severe dysplasia or above” (SDA). The endoscopic features of the baseline LULs that were evaluated included size (the smaller of the diameter or length of the lesion, categorized as ≤5 mm, 6-10 mm, or >10 mm), uniformity of staining (uniformly unstained vs mosaic staining), irregularity of the lesion border (>50% of the border irregular), sharpness of the lesion border (>50% of the border sharply demarcated), and dark staining of the lesion border (accentuated dark staining of the Lugol’s stained mucosa around >50% of the LUL
      • Liu M.
      • Liu Z.
      • Liu F.
      • et al.
      Absence of iodine staining associates with progression of esophageal lesions in a prospective endoscopic surveillance study in China.
      ). These endoscopic features were reviewed and scored retrospectively by 2 experienced researchers who were not endoscopists. Of the 1058 subjects, 705 (66%) underwent follow-up endoscopy with biopsies at the same location as that of the index endoscopy. The other 353 (33%) of the subjects were followed up either by annual door-to-door active surveillance or by electronic linkage to a nearly universal health insurance database that was previously shown to be highly accurate at identifying ESCC cases.
      Three kinds of per-subject analyses were performed: calculation of cumulative incidence of SDA, overall and by various strata; univariate and multivariate logistic regression models to calculate odds ratios (OR) and 95% confidence intervals of various strata to predict progression to SDA; and “adjusted cumulative incidence” of SDA, calculated by combining cumulative incidence data and the multivariate adjusted ORs.
      Liu et al
      • Liu M.
      • Zhou R.
      • Guo C.
      • et al.
      Size of Lugol-unstained lesions as a predictor for risk of progression in premalignant lesions of the esophagus.
      found that in 46 (4.3%) of the 1058 study participants, an SDA lesion developed over a median follow-up period of 5.8 years. Stratified by initial pathologic diagnosis, 2.7% of the subjects with no dysplasia, 7.3% of those with mild dysplasia, and 22.0% of those with moderate dysplasia experienced SDA. In univariate regression analyses, they found that body mass index <22, number of biopsies >2, a dysplastic pathologic diagnosis, and all evaluated endoscopic features except sharp border were associated with a higher risk of progression to SDA. In multivariate analyses, adjusted for age, body mass index, number of biopsies, and initial pathologic diagnosis, only lesion size predicted progression to SDA (odds ratio [OR] 6-10 mm = 4.4 [95% confidence interval [CI], 1.9-10.4]; OR >10 mm = 9.4 [95% CI, 3.8-23.5]). In addition, in multivariate analyses stratified by initial pathologic diagnosis, only patients with no dysplasia at baseline had an increasing risk of progression to SDA with increasing lesion size (OR 6-10 mm = 6.7 [95% CI, 1.7-25.7]; OR >10 mm = 27.9 [95% CI, 7.3-105.7]). Comparisons of the adjusted cumulative incidence rates in patients with no dysplasia but different lesion sizes and patients with mild or moderate dysplasia led to the authors’ recommendations for referring patients for endoscopic surveillance on the basis of lesion size along with pathologic features on biopsy specimens.
      There were several limitations to this study. As noted by the authors, image review and scoring were performed retrospectively, rather than during endoscopy, so the endoscopists were not trying to get comparable images, and there was no standardization of variables that affect the apparent size of lesions such as angulation of the endoscope or distance of the endoscope from the lesion.
      In addition, several details that are important for understanding the protocol and the results of the study were not given. In the initial baseline endoscopic examinations, when a patient had >1 lesion, Liu et al
      • Liu M.
      • Zhou R.
      • Guo C.
      • et al.
      Size of Lugol-unstained lesions as a predictor for risk of progression in premalignant lesions of the esophagus.
      did not mention how it was decided which lesion was used to characterize the patient’s endoscopic features and histologic diagnosis, and when a lesion had >1 biopsy, which biopsy diagnosis was used to characterize the lesion. In addition, when >1 biopsy specimen was taken from an individual, it was unclear whether or not there was a standard protocol (eg, 1 biopsy for lesions ≤5 mm, 2 biopsies for lesions 6 to 10 mm). This raises the question that when the patient was stratified by the number of biopsy specimens taken during the initial examination, and had, for example, 6 biopsies, did that mean that the patient had 6 different LULs and each was sampled by 1 biopsy, or 1 large LUL with 6 biopsies taken from that LUL, or other combinations of LUL numbers and biopsies/LUL, or any of the above?
      Follow-up examinations for patients with worrisome LULs are key in LCE screening. In these examinations, was a follow-up biopsy specimen taken from each original lesion site or just the site of the original lesion that was used to characterize the patient at baseline? In addition, at a median of 5.8 years after baseline, how was the original biopsy site identified endoscopically? Were there scars or LULs to target, or were the follow-up biopsies directed only by endoscopic measurements (eg, 26 cm at 4:00)? We were also curious whether when additional new LULs were found during the follow-up endoscopies, did these new LULs also undergo biopsy, and if so, how often were the new SDA diagnoses from the original biopsy sites and how often were they from these new biopsy sites? It would be interesting to know how many of the 46 SDA endpoint diagnoses were made in the follow-up endoscopy examinations (where lesion localization was possible) and how many were made in the patients who were followed up only by active surveillance or by electronic linkage (where lesion localization was not possible).
      Another important limitation of this article was the absence of endoscopic images illustrating LULs in the size range described, especially baseline lesions of ≤5 mm, 6 to 10 mm, and >10 mm that were histologically diagnosed as no dysplasia in patients who experienced progression to SDA within the follow-up period because this is who the authors say should be surveilled at the same intervals as patients with mild or moderate dysplasia. The only recent endoscopic images we could find from the authors’ group were in their earlier article from the same cohort,
      • Liu M.
      • Liu Z.
      • Liu F.
      • et al.
      Absence of iodine staining associates with progression of esophageal lesions in a prospective endoscopic surveillance study in China.
      and nearly all of the illustrated lesions in that article were much larger than the size range discussed in the current article.
      In the current study, most of the increased risk of progression associated with LUL size was found in patients with an initial biopsy diagnosis of no dysplasia. There was also some increased risk associated with LUL size in patients with mild dysplasia. Possible reasons for these findings include (1) the initial biopsy(ies) missed the target lesion, (2) the initial biopsy(ies) missed the most dysplastic part of the target lesion, (3) the pathologist misread the initial biopsy(ies), and (4) the SDA diagnosed during follow-up developed de novo in a different location from the lesion that underwent biopsy at baseline. It is always possible for biopsies to miss small lesions, especially during rapid screening endoscopies. It is also possible to miss the histologically worst part of a larger lesion if not enough biopsy specimens are taken to reasonably sample the histologic heterogeneity of the lesion.
      Many endoscopically resected lesions show variations in the severity of dysplasia (± early invasion), and it is not uncommon for a patient’s diagnosis to be upgraded after the full lesion is resected and examined. Thus, insufficient biopsy sampling of larger lesions can be a real problem, and endoscopic protocols should include guidelines for recommended numbers of biopsies in lesions of various sizes to minimize this biopsy sampling error. Finally, esophageal squamous dysplasia often develops under the influence of a carcinogenic exposure (such as cigarette smoke) that affects large areas of the esophageal mucosa (a carcinogenic field), which is why it is not uncommon to see multiple disconnected or barely connected dysplastic lesions in the esophagus (such as Fig. 5B in the authors’ earlier article
      • Liu M.
      • Liu Z.
      • Liu F.
      • et al.
      Absence of iodine staining associates with progression of esophageal lesions in a prospective endoscopic surveillance study in China.
      ), and new lesions can develop in this exposure field in different locations over time (certainly within 5.8 years of follow-up). We think it is probable that most of the increased risk of progression with LUL size found in patients with initial biopsy diagnoses of no dysplasia or mild dysplasia in the current study by Liu et al
      • Liu M.
      • Zhou R.
      • Guo C.
      • et al.
      Size of Lugol-unstained lesions as a predictor for risk of progression in premalignant lesions of the esophagus.
      resulted from a combination of the initial biopsies missing the targeted lesion, biopsy sampling error in larger lesions, and the later diagnoses of SDAs developing in a different site than the initial lesion. But all 3 of these situations occur, especially in rapid screening endoscopies, so it is important to be aware of the potential importance of lesion attributes (such as size) other than biopsy diagnoses.
      It has been previously shown that there is a positive relation between LUL size
      • Dawsey S.M.
      • Fleischer D.
      • Wang G.
      • et al.
      Mucosal iodine staining improves endoscopic visualization of squamous dysplasia and squamous cell carcinoma of the esophagus in Linxian, China.
      or multiplicity
      • Katada C.
      • Yokoyama T.
      • Yano T.
      • et al.
      Alcohol consumption and multiple dysplastic lesions increase risk of squamous cell carcinoma in the esophagus, head, and neck.
      and the presence of SDA, and this has led to the recommendation that endoscopic therapy be performed on LULs with these “worrisome” endoscopic features and a biopsy diagnosis of moderate dysplasia, and also in all LULs with biopsy specimens showing SDA.
      • Wei W.Q.
      • Hao C.Q.
      • Guan C.T.
      • et al.
      Esophageal histological precursor lesions and subsequent 8.5-year cancer risk in a population-based prospective study in China.
      The current study by Liu et al
      • Liu M.
      • Zhou R.
      • Guo C.
      • et al.
      Size of Lugol-unstained lesions as a predictor for risk of progression in premalignant lesions of the esophagus.
      confirms the importance of special consideration for larger lesions and the importance of trusting clinical judgement when biopsy diagnoses (eg, no dysplasia) do not match the endoscopic impression (a large LUL). As noted in the current article, additional studies in general populations are needed to confirm the findings of this study, to highlight potential screening pitfalls (eg, large lesions with biopsy diagnoses of no dysplasia), to improve our triage of patients to surveillance, and to optimize surveillance intervals based on both LCE findings and pathologic diagnoses.

      Disclosure

      Both authors disclosed no financial relationships.

      References

        • Bray F.
        • Ferlay J.
        • Soerjomataram I.
        • et al.
        Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.
        CA Cancer J Clin. 2018; 68: 394-424
        • Codipilly D.C.
        • Qin Y.
        • Dawsey S.M.
        • et al.
        Screening for esophageal squamous cell carcinoma: recent advances.
        Gastrointest Endosc. 2018; 88: 413-426
        • Dawsey S.M.
        • Fleischer D.
        • Wang G.
        • et al.
        Mucosal iodine staining improves endoscopic visualization of squamous dysplasia and squamous cell carcinoma of the esophagus in Linxian, China.
        Cancer. 1998; 83: 220-231
        • Wang G.
        • Abnet C.C.
        • Shen Q.
        • et al.
        Histological precursors of oesophageal squamous cell carcinoma: results from a 13 year prospective follow up study in a high risk population.
        Gut. 2005; 54: 187-192
        • Wei W.Q.
        • Chen Z.F.
        • He Y.T.
        • et al.
        Long-term follow-up of a community assignment, one-time endoscopic screening study of esophageal cancer in China.
        J Clin Oncol. 2015; 33: 1951-1957
      1. Chen R, Liu Y, Song G, et al. Effectiveness of one-time endoscopic screening programme in prevention of upper gastrointestinal cancer in China: a multicentre population-based cohort study. Gut. Epub 2020 Apr 2.

        • Liu M.
        • Zhou R.
        • Guo C.
        • et al.
        Size of Lugol-unstained lesions as a predictor for risk of progression in premalignant lesions of the esophagus.
        Gastrointest Endosc. 2021; 93: 1065-1073.e3
        • Liu M.
        • Liu Z.
        • Liu F.
        • et al.
        Absence of iodine staining associates with progression of esophageal lesions in a prospective endoscopic surveillance study in China.
        Clin Gastroenterol Hepatol. 2020; 18: 1626-1635
        • Katada C.
        • Yokoyama T.
        • Yano T.
        • et al.
        Alcohol consumption and multiple dysplastic lesions increase risk of squamous cell carcinoma in the esophagus, head, and neck.
        Gastroenterology. 2016; 151: 860-869
        • Wei W.Q.
        • Hao C.Q.
        • Guan C.T.
        • et al.
        Esophageal histological precursor lesions and subsequent 8.5-year cancer risk in a population-based prospective study in China.
        Am J Gastroenterol. 2020; 115: 1036-1044

      Linked Article