Summary
Squamous dysplasia is the histological precursor of esophageal squamous cell carcinoma (ESCC). The optimal management for distinct squamous dysplasia grades remains unclear because the corresponding risk of developing ESCC is unknown. We aimed to assess the ESCC risk in patients with esophageal squamous dysplasia in a Western country. This nationwide cohort study included all patients with esophageal squamous dysplasia, diagnosed between 1991 and 2020 in the Dutch nationwide pathology databank (Palga). Squamous dysplasia was divided in mild-to-moderate dysplasia (mild, low-grade, and moderate dysplasia) and higher-grade dysplasia (high-grade dysplasia, severe dysplasia, carcinoma in situ). ESCC were identified in Palga and the Netherlands Cancer Registry. The primary endpoint was diagnosis of prevalent (≤6 months) and incident (>6 months after squamous dysplasia) ESCC. In total, 873 patients (55% male, aged 68 years SD ± 13.2) were diagnosed with esophageal squamous dysplasia, comprising mild-to-moderate dysplasia (n = 456), higher-grade dysplasia (n = 393), and dysplasia not otherwise specified (n = 24). ESCC was diagnosed in 77 (17%) patients with mild-to-moderate dysplasia (49 prevalent, 28 incident ESCC) and in 162 (41%) patients with higher-grade dysplasia (128 prevalent, 34 incident ESCC). After excluding prevalent ESCC, the annual risk of ESCC was 4.0% (95% CI: 2.7–5.7%) in patients with mild-to-moderate dysplasia and 8.5% (95% CI: 5.9–11.7%) in patients with higher-grade dysplasia. All patients with squamous dysplasia, including those with mild-to-moderate dysplasia, have a substantial risk of developing ESCC. Consequently, endoscopic surveillance of the esophageal mucosa or endoscopic resection of dysplasia should be considered for patients with mild-to-moderate dysplasia in Western countries.
KEY MESSAGES
What is already known on this topic?
Squamous dysplasia is the histological precursor of ESCC and is divided in distinct grades, based on the proportion of the squamous epithelium with histopathological abnormalities.
In Western countries, the optimal management for distinct squamous dysplasia grades remains unclear because the corresponding risk of developing ESCC is unknown.
What this study adds
The ESCC risk of patients with squamous dysplasia was increased for all patients with squamous dysplasia in a Western country; 2.1% for patients with mild dysplasia, 5.1% for low-grade dysplasia, and 5.2% for moderate dysplasia. Increasing grades of squamous dysplasia were associated with an increased ESCC risk.
How this study might affect research, practice, or policy
We recommend that endoscopic follow-up or treatment should be considered in all patients with esophageal squamous dysplasia in Western countries: 1) for patients with mild, low-grade, and moderate dysplasia, endoscopic surveillance with careful inspection with narrow band imaging or dye-based chromoendoscopy of the esophageal mucosa is indicated; and 2) for patients with high-grade dysplasia, severe dysplasia and carcinoma in situ adequate endoscopic staging and in case of suspected neoplasia endoscopic treatment should be performed.
Keywords: endoscopy, esophageal squamous cell carcinoma, squamous dysplasia
INTRODUCTION
Over 85% of the esophageal cancers are esophageal squamous cell carcinoma (ESCC) worldwide.1 In Western countries, the age-standardized incidence rate of ESCC ranged between 1.0 and 2.5 per 100.000 persons in 2018.1 As most ESCC are detected in advanced and incurable stages, the 5-year survival rate of patients with ESCC is merely 22%.2,3 The detection of ESCC at early stages is associated with a considerably better 5-year survival of 85–100%, as early-stage ESCC can potentially be treated curatively with endoscopic resection (ER).2,3
The cornerstone in detecting ESCC at early stages consists of the identification of high-risk patients for ESCC. An important group of high-risk patients are patients with esophageal squamous dysplasia, a histological precursor lesion of ESCC. Squamous dysplasia is defined as neoplastic alterations of the esophageal squamous epithelium, without invasion.4 ESCC is thought to develop via the dysplasia-carcinoma cascade: from normal squamous epithelium via increasing grades of dysplasia to ESCC.5,6 Adequate endoscopic detection and treatment of patients with squamous dysplasia allows for early detection of ESCC or can even prevent ESCC development.7,8
The pathological assessment of squamous dysplasia can be challenging and currently two classification systems are used worldwide: a three-tiered and two-tiered classification.4,9 Both classifications are based on the proportion of the squamous epithelium with histopathological abnormalities.4,9 The three-tier system is predominantly used in Asian countries and classifies squamous dysplasia in mild, moderate, and severe dysplasia.9 In an Asian study with 13.5 years of endoscopic surveillance, the risk of neoplastic progression was up to 24% for mild dysplasia, 50% for moderate dysplasia, and 74% for severe dysplasia.9–14 However, it is unknown whether this risk of ESCC can be generalized to patients with squamous dysplasia in Western countries, as the incidence of ESCC differs strongly between Western and Asian countries.1
In Western countries, the World Health Organization advises to use the two-tiered classification with low- and high-grade dysplasia to increase the level of inter-observer agreement among pathologists.4,15 Current guidelines in Western countries advocate that ER should be performed for high-grade dysplasia and ESCC limited to the mucosa, but it remains controversial whether endoscopic surveillance or treatment is indicated for low-grade dysplasia.16 The optimal management for distinct squamous dysplasia grades remains unclear because the corresponding risk of developing ESCC for each distinct grade of squamous dysplasia is unknown. We, therefore, aimed to assess the ESCC risk in patients with squamous dysplasia in a Western country.
METHODS
Study design and patients
We performed a nationwide, retrospective study including all patients diagnosed with esophageal squamous dysplasia between January 1991 and December 2020 in the Netherlands. Patients were identified via the Dutch nationwide pathology databank (Palga).17 The development of ESCC in included patients was identified from Palga and the Netherlands Cancer Registry (NCR; nationwide registry of all cancers). All patient data were coded and anonymized by a third trust party and, therefore, no informed consent was needed. This study was approved by the Medical Ethical Review Committee of the Erasmus Medical Centre (MEC-2022-0274). The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the institution’s Human Research Committee.
Palga search
The Palga database contains all Dutch pathology reports with nationwide coverage since 1991, linked to an encrypted individual patient identification number and diagnostic code.17 The diagnostic code reflects the location, type, and histopathological diagnosis of the tissue sample (e.g. esophagus × biopsy × low-grade dysplasia). The Palga database was searched for the diagnostic codes for dysplasia and atypia in the esophagus (search details are described in Supplementary Table S1). Inclusion criteria were all diagnostic codes for squamous dysplasia in the esophagus. Exclusion criteria were dysplasia in a Barrett’s esophagus or columnar epithelium, dysplasia located in the stomach and patients with previous or simultaneous (i.e. in the pathology specimen of the same date) esophageal cancer, and patients developing esophageal adenocarcinoma.
Histopathological definitions
Squamous dysplasia is characterized by the presence of both cytological and architectural atypia.4 Characteristics of cytological atypia include cell enlargement, pleomorphism, hyperchromasia, loss of polarity, and overlapping. Architectural atypia is defined as abnormal maturation of the epithelium. The grade of dysplasia is based on the proportion of the squamous epithelium with pathological abnormalities.4,9 Mild, moderate, and severe dysplasia are limited to the lower third, middle third and three thirds of the squamous epithelium (Fig. 1).9 Low-grade dysplasia is defined as mild cytological atypia confined to the lower half of the squamous epithelium.4 High-grade dysplasia is characterized by severe cytological atypia or the presence of mild cytological atypia in more than half of the squamous epithelium.4 Carcinoma in situ (CIS) is defined as the presence of dysplastic cells throughout the full thickness of the squamous epithelium, without invasion.9 If squamous dysplasia could not be graded, because of biopsy size or orientation, this is referred to as dysplasia not otherwise specified (NOS).9
Fig. 1.
Distinct grades of squamous dysplasia in the esophagus. (A) Mild dysplasia with dysplastic cells limited to the lower third of the squamous epithelium. (B) Moderate dysplasia with dysplastic cells limited to the middle third of the squamous epithelium. Both image (A) and (B) are referred to as low-grade dysplasia in the two-tiered classification. (C) Severe dysplasia with dysplastic cells extending to the full thickness of the squamous epithelium (hematoxylin-eosin staining, original magnification ×100 in A and B and ×50 in C).
Data collection
From Palga, anonymized pathology reports with conclusions and microscopic assessment were collected from 5 years before the diagnosis squamous dysplasia and all follow-up reports till July 2022. We extracted the following characteristics for all included patients: sex, age, year of squamous dysplasia, number of endoscopies with biopsies and time intervals, and date of last follow-up or the diagnosis of ESCC. The time interval between squamous dysplasia and ESCC diagnosis was divided in prevalent (within 6 months) and incident (> 6 months). For patients with ESCC, the following characteristics were collected from the NCR; age at ESCC diagnosis, histopathological characteristics, and location of ESCC (cervical; <18 cm from the incisors, upper third; 18–24 cm from the incisors, middle third; 24–32 cm from the incisors, lower third; 32–40 cm from the incisors, and overlap; between two parts of the esophagus). The TNM stage of ESCC and treatment strategy (on 31 January 2021) were also assessed.
Study endpoints
The primary endpoint of this study was the proportion of patients with squamous dysplasia that were subsequently diagnosed with ESCC. Secondary endpoints included 1) the risk of ESCC for distinct grades of squamous dysplasia, 2) the time between first squamous dysplasia diagnosis and the detection of ESCC, and 3) characteristics and outcomes of patients with squamous dysplasia and subsequent ESCC.
Statistics
Descriptive statistics are presented as means with standard deviations (SD), medians with inter-quartile ranges (IQR) and counts with percentages, according to the nature of the data. For sub-group analyses, patients with mild, low-grade, and moderate dysplasia were combined in the group mild-to-moderate dysplasia and patients with high-grade dysplasia, severe dysplasia, and CIS were combined in the group higher-grade dysplasia. Sub-groups were compared using the X2 test. The percentage annual risk of ESCC was calculated with number of events divided by number of patient years at risk, multiplied by 100. Cox proportional hazards analyses were performed to identify and quantify potential risk factors for the detection of ESCC and were presented as hazard ratios (HR) with 95% confidence intervals (CI). The statistical package (survminer) in R was used for the cumulative incidence plot. Two-side P-values <0.05 were considered significant. Analyses were performed in IBM SPSS for Windows version 28 (SPSS Inc) and R version 4.2.2 (The R Foundation Statistical Computing, Vienna, Austria).
RESULTS
Patients
The Palga search identified 9.687 patients with dysplasia in esophageal pathology specimen between January 1991 and December 2020 in the Netherlands (Figure S1). After review of pathology reports, 873 patients with a confirmed first diagnosis of squamous dysplasia in the esophagus were included. The mean age of included patients was 68.0 years (SD ±13.2) and 55.1% was male.
Baseline characteristics of squamous dysplasia
The baseline grade of dysplasia of included patients was mild (n = 179), low-grade (n = 80), moderate (n = 197), high-grade (n = 77), and severe (n = 244) dysplasia, and CIS (n = 72) (Table 1). In 79/197 patients with moderate dysplasia, the grade of dysplasia could be divided into low-grade (69.6%) and high-grade dysplasia (30.4%), based on complete pathology reports. Squamous dysplasia was diagnosed between 2020 to 2010 (38.9%), 2000 to 2010 (33.9%), and 1991 to 2000 (27.2%). Most cases of low-grade (75.0%) and high-grade dysplasia (71.4%) were diagnosed between 2011 and 2020 (Fig. 2) (P < 0.001).
Table 1.
Baseline characteristics of included patients with esophageal squamous dysplasia.
| Characteristic | Total cohort | Mild-to-moderate dysplasia | Higher-grade dysplasia |
|---|---|---|---|
| No. of patients | 873 | 456 | 393 |
| Sex, male | 481 (55.1%) | 255 (55.9%) | 213 (54.2%) |
| Age, years | 68 (SD ± 13.2) | 66 (SD ± 13.6) | 71 (SD ± 12.2) |
|
Year of diagnosis 1991–2000 2000–2010 2010–2020 |
237 (27.2%) 296 (33.9%) 340 (38.9%) |
139 (30.5%) 146 (32.0%) 171 (37.5%) |
92 (23.4%) 142 (36.1%) 159 (40.5%) |
Data presented as n with (%) or mean with standard deviation (SD). Patients with mild (n = 179), low-grade (n = 80), and moderate dysplasia (n = 197) were combined in the group mild-to-moderate dysplasia. Patients with high-grade dysplasia (n = 77), severe dysplasia (n = 244), and CIS (n = 72) were combined in the group higher-grade dysplasia. The total cohort (n = 873) also included patients with baseline squamous dysplasia NOS (n = 24).
Fig. 2.
The proportion of patients with distinct grades of squamous dysplasia during recent decades. For patients with dysplasia NOS, grading of squamous dysplasia was not possible, due to biopsy size or orientation. Most cases of low- (75.0%) and high-grade dysplasia (71.4%) were diagnosed between 2011 and 2020 (P < 0.001).
Treatment strategies for baseline mild, low-grade, and moderate dysplasia
The cohort included 456 patients with mild-to-moderate dysplasia of which 57.0% of the patients underwent endoscopic re-assessment with histopathology or treatment. This was performed after a median of 12 weeks (IQR 6–29). During the first endoscopic-reassessment, ESCC was detected in 5.1%, 1.8%, and 12.8% of patients with baseline mild, low-grade, and moderate dysplasia, respectively (Fig. 3). The median histopathological follow-up time was 10 months (IQR 3–42) and patients received a median of 2 (IQR 1–4) endoscopies. Thirteen (2.8%) patients with mild-to-moderate dysplasia were treated with primary ER (n = 12; 2.6%) or surgery (n = 1; 0.2%). The ER and surgery specimens showed mild-to-moderate dysplasia (n = 6), higher-grade dysplasia (n = 5), and ESCC (n = 2). The remaining patients (43.0%) had no histopathological follow-up.
Fig. 3.
Most advanced lesion detected during first endoscopic re-assessment and complete follow-up in patients with distinct grades of squamous dysplasia at baseline. Data presented in groups according to the grade of confirmed first squamous dysplasia diagnosis. (A) shows the most advanced lesion detected during first endoscopic re-assessment. Patients with dysplasia NOS as first dysplasia diagnosis (n = 24) or detected during first endoscopic re-assessment (n = 11) are not shown. The median time to first endoscopic re-assessment with histopathology was 17, 12, 11, 4, 5, and 6 weeks for patients with mild, low-grade, moderate, high-grade, and severe dysplasia, and CIS, respectively. (B) shows the most advanced lesion detected during complete follow-up. (B) contains seven patients with a clinical diagnosis of ESCC without pathology confirmation, based on data from the Netherlands cancer registry. The median follow-up time was 20 months for mild dysplasia, 10 months for low-grade dysplasia, 7 months for moderate dysplasia, 4 months for high-grade dysplasia, 2 months for severe dysplasia, and 5 months for CIS.
Treatment strategy for baseline high-grade dysplasia, severe dysplasia, and CIS
A total of 71.5% of 393 included patients with higher-grade dysplasia underwent endoscopic re-assessment with histopathology or treatment. The initial endoscopic re-assessment was performed after a median of 5 weeks (IQR 2–10) and revealed mild-to-moderate dysplasia (5.0%), higher-grade dysplasia (39.6%), and ESCC (32.7%). Patients underwent a median of 2 endoscopies (IQR 1–4) during the follow-up till the diagnosis of ESCC or last follow-up. 71 (18.1%) patients with higher-grade dysplasia underwent treatment with ER (9.7%), surgery (4.3%), or chemotherapy and/or radiotherapy (4.1%). Pathological assessment of ER and surgery specimens showed no dysplasia (n = 1), mild-to-moderate dysplasia (n = 1), higher-grade dysplasia (n = 25), and ESCC (n = 28). In the remaining patients (28.5%), no histopathological follow-up was available.
Association between increasing grades of dysplasia and risk of ESCC
ESCC was diagnosed in 28.4% of included patients with baseline squamous dysplasia. Table 2 depicts the proportions of patients diagnosed with ESCC, according to their distinct grades of baseline squamous dysplasia. Increasing grades of dysplasia were associated with a significantly increased risk of ESCC (P < 0.001) (Table 3). Patients with moderate dysplasia had a significantly increased risk of ESCC compared with mild dysplasia (HR 2.40, 95% CI: 1.35–4.29) and a showed a trend towards an increased risk compared with low-grade dysplasia (HR 1.71, 95% CI: 0.93–3.16). Patients with low-grade dysplasia had a tendency towards an increased risk to develop higher-grade dysplasia, compared with patients with mild dysplasia (HR 3.10, 95% CI: 0.98–9.80), but the results were not significant. Baseline high-grade dysplasia was associated with a significantly increased risk of ESCC, compared with mild (HR 3.64, 95% CI: 1.91–6.94) and low-grade dysplasia (HR 2.59, 95% CI: 1.32–5.10). The risk of ESCC between patients with baseline moderate dysplasia and high-grade dysplasia did not differ significantly (HR 0.66, 95% CI: 0.41–1.08). Results were consistent after adjusting for sex, age, year of first dysplasia diagnosis, primary treatment strategy, and time to first endoscopic re-assessment (Table 3, Supplementary Table S2).
Table 2.
The proportion of patients with distinct grades of baseline squamous dysplasia diagnosed with prevalent and incident ESCC.
| Baseline grade of dysplasia | No. of patients | No. of patients diagnosed with ESCC | Prevalent ESCC | Incident ESCC 1 | Annual ESCC risk per PY 1 | PY at risk 1 |
|---|---|---|---|---|---|---|
| Mild dysplasia | 179 | 15 (8.4%) | 9 (5.0%) | 6 (3.5%) | 2.1% | 279.7 |
| Low-grade dysplasia | 80 | 13 (16.3%) | 8 (10.0%) | 5 (6.9%) | 5.1% | 97.3 |
| Moderate dysplasia | 197 | 49 (24.9%) | 32 (16.2%) | 17 (10.3%) | 5.2% | 324.4 |
| High-grade dysplasia | 77 | 24 (31.2%) | 18 (23.4%) | 6 (10.2%) | 8.9% | 67.5 |
| Severe dysplasia | 244 | 110 (45.1%) | 92 (37.7%) | 18 (11.8%) | 7.5% | 239.7 |
| Carcinoma in situ | 72 | 28 (38.9%) | 18 (25.0%) | 10 (18.5%) | 10.6% | 94.6 |
| Dysplasia NOS 2 | 24 | 9 (37.5%) | 4 (16.7%) | 5 (25.0%) | 10.1% | 49.6 |
| Total cohort | 873 | 248 (28.4%) | 181 (20.7%) | 67 (9.7%) | 5.8% | 1152.7 |
Data presented as n with (%). ESCC were divided in prevalent (≤ 6 months) and incident (>6 months) after baseline diagnosis of squamous dysplasia. 1Calculated for patients at risk of ESCC at 6 months after baseline squamous dysplasia (n = 692). 2Grading of squamous dysplasia was not possible, due to biopsy size or orientation. ESCC, esophageal squamous cell carcinoma; NOS, not otherwise specified, PY, patient-years.
Table 3.
Risk factors associated with the detection of ESCC in patients with distinct grades of squamous dysplasia (n = 873).
| No. of ESCC/total no. of patients | Univariate HR | 95% CI | P | Adjusted HR | 95% CI | P | |
|---|---|---|---|---|---|---|---|
|
Sex Male Female |
139/481 109/392 |
Reference 0.96 |
— 0.75–1.24 |
0.765 |
Reference 0.91 |
— 0.71–1.17 |
— 0.460 |
| Age (years) | 248/873 | 1.02 | 1.01–1.03 | <0.001 | 1.02 | 1.00–1.03 | 0.009 |
|
Year of first dysplasia diagnosis 1991–2000 2000–2010 2010–2020 |
50/237 101/296 97/340 |
Reference 1.43 1.22 |
— 1.02–2.00 0.87–1.72 |
— 0.040 0.251 |
Reference 1.37 1.19 |
— 0.98–1.94 0.83–1.70 |
— 0.069 0.340 |
|
Grade of baseline dysplasia Mild dysplasia Low-grade dysplasia Moderate dysplasia High-grade dysplasia Severe dysplasia Carcinoma in situ Dysplasia NOS |
15/179 13/80 49/197 24/77 110/244 28/72 9/24 |
Reference 1.40 2.40 3.64 5.33 4.07 2.43 |
— 0.67–2.95 1.35–4.29 1.91–6.94 3.10–9.15 2.17–7.62 1.06–5.55 |
— 0.373 0.003 <0.001 <0.001 <0.001 0.036 |
Reference 1.29 2.23 2.96 4.70 3.43 2.34 |
— 0.60–2.75 1.25–3.99 1.52–5.77 2.72–8.11 1.82–6.49 1.02–5.34 |
— 0.513 0.007 0.001 <0.001 <0.001 0.045 |
Results were obtained with univariate and multivariate Cox proportional hazards analyses. Two-side P-values <0.05 were considered statistically significant. In multivariate analyses, results were adjusted for sex, age, and grade of baseline dysplasia. For patients with dysplasia NOS, grading of squamous dysplasia was not possible, due to biopsy size or orientation. Data are presented as HR with 95% CI with the detection of ESCC as outcome. CI, confidence interval; ESCC, esophageal squamous cell carcinoma; HR, hazard ratio; NOS, not otherwise specified.
Prevalent and incident ESCC
Prevalent ESCC was diagnosed in 181/873 (20.7%) patients and incident ESCC in 67/692 (9.7%) patients (Table 2, Fig. 4). Incident ESCC was detected after a median of 23 months (IQR 11–49). After excluding patients with prevalent ESCC, the annual ESCC risk was 2.1%, 5.1%, and 5.2% per patient-year for patients with mild, low-grade, and moderate dysplasia (Table 2) with a total of 701.3 patient-years of follow-up. The risk for both prevalent and incident ESCC increased with increasing grades of baseline squamous dysplasia (Table 2, Supplementary Table S3). Multivariable analyses, adjusted for sex and age, showed similar results.
Fig. 4.
Cumulative incidence of esophageal squamous cell carcinoma in 873 patients with baseline squamous dysplasia. Data are shown for patients with mild-to-moderate dysplasia (n = 456), higher-grade dysplasia (n = 393), and dysplasia NOS (n = 24). Logrank test between groups: p < 0.001.
Characteristics of ESCC
Patients diagnosed with ESCC had a mean age at diagnosis of 69.0 years (SD ± 10.7) and 56.0% was male (characteristics of prevalent and incident ESCC are shown in Supplementary Table S4). The tumor stage of ESCC was low (0–II) in 48.4% and high (III–IV) in 27.5%. Distant metastases at time of diagnosis were detected in 8.9% of patients. In total, patients with ESCC were treated with ER (14.9%), surgery (35.1%), and chemo-/radiotherapy (47.6%). The median survival after ESCC diagnosis was 25 months (IQR 10–75).
DISCUSSION
Squamous dysplasia is the histological precursor of ESCC and is divided in distinct grades, based on the proportion of squamous epithelium with histopathological abnormalities. In Western countries, the risk of ESCC for these distinct grades of dysplasia is unknown and, consequently, optimal management remains unclear. We performed a nationwide, retrospective study on the risk of ESCC in patients with distinct grades of esophageal squamous dysplasia in the Netherlands. In our study, all patients with squamous dysplasia, including patients with mild, low-grade, and moderate dysplasia, had a substantially increased risk of developing ESCC. Therefore, endoscopic surveillance or treatment should be considered for all patients with squamous dysplasia in Western countries.
The currently published studies on squamous dysplasia and the associated risk of ESCC originate from Asian countries.9–11,13,14 These studies report a cumulative 5-year incidence of ESCC ranging from 1% to 24% for patients with mild dysplasia, 5% to 50% for moderate dysplasia, and up to 100% for severe dysplasia and CIS.9–11,13,14 In rural areas of China, one-time endoscopic screening and treatment in case of dysplasia resulted in both a decreased incidence and mortality of ESCC in residents aged 40–69 years, compared with controls.18,19 These findings confirm that in Asian countries, endoscopic surveillance or treatment is warranted for patients with all grades of squamous dysplasia. Unfortunately, comparisons between Asian and Western populations are difficult, caused by the large differences in ESCC incidence between these countries and, therefore, also differences in screening, surveillance, and treatment strategies.1
In the current study, we report on prevalent and incident ESCC separately, attempting to distinguish patients with a potentially underlying baseline ESCC, from patients developing ESCC during the follow-up. Prevalent ESCC were diagnosed in one fifth of included patients and up to 16% of patients with baseline mild, low-grade, and moderate dysplasia. A part of the patients with prevalent ESCC potentially had a visible suspicious lesion during endoscopy without histopathological confirmation. In others, the pathology report of dysplasia may have resulted in an additional endoscopy with ER, during which the diagnosis ESCC was established.
The pathological assessment of esophageal squamous dysplasia can be challenging and may be subject to interobserver variability and sampling bias. To decrease interobserver variability in Western countries, the two-tiered classification into low-grade or high-grade dysplasia was introduced in the 5th edition of the WHO classification.4 In line with the introduction of the two-tiered classification, low- and high-grade dysplasia were diagnosed more frequently in this study during recent years. Nevertheless, we found that both the three-tiered (mild, moderate, and severe dysplasia) and two-tiered classification are currently used in the Netherlands. In line with the recommendation of the WHO, we think that one uniform classification for patients with squamous dysplasia, used by all pathologists in Western countries, would be desirable. Standardized advice for distinct grades of squamous dysplasia with indications for endoscopic surveillance and treatment, will promote exchangeability and comparability of scientific data and may help to improve the outcomes of patients with squamous dysplasia.
Sampling bias, when biopsies do not adequately reflect the grade of dysplasia, can be caused by several endoscopic and histopathological factors. Endoscopic factors include for example the experience of the endoscopist and the number, chosen location, and depth of the biopsies. Histopathological factors include a lack of orientation and presence of other histopathological abnormalities such as active inflammation in case of reflux- or candida esophagitis. The occurrence and clinical relevance of sampling bias is confirmed by previous studies, which reported a discordance of the grade of squamous dysplasia of up to 45% between biopsy and corresponding ER specimen.20–22 The study of Chen et al. (2022) reported on 202 patients with low-grade dysplasia in biopsies, of which the corresponding ER specimen showed high-grade dysplasia in 33% of patients.21 These results are in line with the proportion of prevalent ESCC of 20.7% detected in our current study, and emphasize the importance of adequate endoscopic (re-)assessment with representative biopsies and accurate pathological assessment.
This nationwide cohort study is one of the first Western studies reporting on the ESCC risk in patients with squamous dysplasia, but has some inherent limitations. The current study was based on characteristics available in the Palga and NCR databases. The Palga database contains pathology reports from clinical practice. The NCR contains certain characteristics of patients diagnosed with ESCC, but clinical data, including medical history, symptoms of dysphagia and odynophagia, and endoscopy characteristiscs such as the presence, size, and macroscopic appearance of lesions, are not available. Endoscopic assessment and follow-up or treatment strategies were performed upon clinician’s expert opinion and daily clinical practice, and no pathology slides were reassessed. No histopathological follow-up was available in a substantial proportion of included patients (i.e. 42% of patients with mild-to-moderate dysplasia and 31% of patients with higher-grade dysplasia), which may have resulted in an underestimation of the risk of ESCC.
In conclusion, all patients with esophageal squamous dysplasia in Western countries, including those with mild, low-grade and moderate dysplasia, have a substantial risk of developing ESCC. Consequently, endoscopic surveillance of the esophageal mucosa or ER of dysplasia should be considered for patients with mild-to-moderate dysplasia in Western countries. For patients with high-grade dysplasia, severe dysplasia and CIS, adequate endoscopic staging and in case of suspicion for neoplasia, aggressive treatment is required as ESCC is already present in a substantial proportion of patients.
Supplementary Material
ACKNOWLEDGEMENTS
The authors thank the registration team of the NCR and Palga for the collection of data.
Financial support: The authors received no financial support for the research, authorship, and/or publication of this article. This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflicts of Interest statement: None declared.
Contributor Information
Laurelle van Tilburg, Department of Gastroenterology and Hepatology, Erasmus MC Cancer Institute, University Medical Center, 3000 CA Rotterdam, the Netherlands.
Manon C W Spaander, Department of Gastroenterology and Hepatology, Erasmus MC Cancer Institute, University Medical Center, 3000 CA Rotterdam, the Netherlands.
Marco J Bruno, Department of Gastroenterology and Hepatology, Erasmus MC Cancer Institute, University Medical Center, 3000 CA Rotterdam, the Netherlands.
Lindsey Oudijk, Department of Pathology, Erasmus MC Cancer Institute, University Medical Center, 3000 CA Rotterdam, the Netherlands.
Lara R Heij, Department of Pathology, Erasmus MC Cancer Institute, University Medical Center, 3000 CA Rotterdam, the Netherlands.
Michail Doukas, Department of Pathology, Erasmus MC Cancer Institute, University Medical Center, 3000 CA Rotterdam, the Netherlands.
Arjun D Koch, Department of Gastroenterology and Hepatology, Erasmus MC Cancer Institute, University Medical Center, 3000 CA Rotterdam, the Netherlands.
ETHICS APPROVAL
This study was approved by the Medical Ethical Review Committee of the Erasmus Medical Centre (MEC-2022-0274).
References
- 1. Arnold M, Ferlay J, van Berge Henegouwen M I, Soerjomataram I. Global burden of oesophageal and gastric cancer by histology and subsite in 2018. Gut 2020; 69(9): 1564–71. [DOI] [PubMed] [Google Scholar]
- 2. van Putten M, de Vos-Geelen J, Nieuwenhuijzen G A P et al. Long-term survival improvement in oesophageal cancer in the Netherlands. Eur J Cancer 2018; 94: 138–47. [DOI] [PubMed] [Google Scholar]
- 3. Yokoyama A, Ohmori T, Makuuchi H et al. Successful screening for early esophageal cancer in alcoholics using endoscopy and mucosa iodine staining. Cancer 1995; 76(6): 928–34. [DOI] [PubMed] [Google Scholar]
- 4. World Health O . WHO classification of tumours: digestive system tumours: World Health Organization. WHO, 2019. [Google Scholar]
- 5. Savant D, Zhang Q, Yang Z. Squamous neoplasia in the esophagus. Arch Pathol Lab Med 2021; 145(5): 554–61. [DOI] [PubMed] [Google Scholar]
- 6. Liu X, Zhang M, Ying S et al. Genetic alterations in esophageal tissues from squamous dysplasia to carcinoma. Gastroenterology 2017; 153(1): 166–77. [DOI] [PubMed] [Google Scholar]
- 7. van de Ven S E M, de Graaf W, Bugter O et al. Screening for synchronous esophageal second primary tumors in patients with head and neck cancer. Dis Esophagus 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Bugter O, van de Ven S E M, Hardillo J A, Bruno M J, Koch A D, Baatenburg de Jong R J. Early detection of esophageal second primary tumors using Lugol chromoendoscopy in patients with head and neck cancer: a systematic review and meta-analysis. Head Neck 2019; 41(4): 1122–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Wang G Q, 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(2): 187–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Li H, Zhang S, Zhou J et al. Endoscopic surveillance for premalignant esophageal lesions: a community-based multicenter, prospective cohort study. Clin Gastroenterol Hepatol 2022. [DOI] [PubMed] [Google Scholar]
- 11. Dawsey S M, Lewin K J, Wang G Q et al. Squamous esophageal histology and subsequent risk of squamous cell carcinoma of the esophagus. A prospective follow-up study from Linxian, China. Cancer 1994; 74(6): 1686–92. [DOI] [PubMed] [Google Scholar]
- 12. 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(5): 1065–73.e3. [DOI] [PubMed] [Google Scholar]
- 13. 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(7): 1036–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Wen D, Zhang L, Wang X et al. A 5.5-year surveillance of esophageal and gastric cardia precursors after a population-based screening in China. J Gastroenterol Hepatol 2015; 30(12): 1720–5. [DOI] [PubMed] [Google Scholar]
- 15. Schlemper R J, Dawsey S M, Itabashi M et al. Differences in diagnostic criteria for esophageal squamous cell carcinoma between Japanese and Western pathologists. Cancer. Cancer 2000; 88(5): 996–1006. [DOI] [PubMed] [Google Scholar]
- 16. Pimentel-Nunes P, Libânio D, Bastiaansen B A J et al. Endoscopic submucosal dissection for superficial gastrointestinal lesions: European Society of Gastrointestinal Endoscopy (ESGE) guideline–update 2022. Endoscopy 2022; 54: 591–622. [DOI] [PubMed] [Google Scholar]
- 17. Casparie M, Tiebosch A, Burger G et al. Pathology databanking and biobanking in the Netherlands, a central role for PALGA, the nationwide histopathology and cytopathology data network and archive. Anal Cell Pathol 2007; 29(1): 19–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. 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(17): 1951–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. 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 2021; 70(2): 251–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Park Y J, Kim G H, Park D Y et al. Histopathologic discrepancies between endoscopic forceps biopsy and endoscopic resection specimens in superficial esophageal squamous neoplasms. J Gastroenterol Hepatol 2019; 34(6): 1058–65. [DOI] [PubMed] [Google Scholar]
- 21. Chen H, Zhou X Y, Li S et al. Endoscopic detection of esophageal low-grade squamous dysplasia: how to predict pathologic upgrades before treatment? J Dig Dis 2022; 23(4): 209–19. [DOI] [PubMed] [Google Scholar]
- 22. Yang L, Jin H, Xie X L et al. Endoscopic resections for superficial esophageal squamous cell epithelial neoplasia: focus on histological discrepancies between biopsy and resected specimens. BMC Gastroenterol 2021; 21(1): 114. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.