Abstract
Background and aims
Radiofrequency ablation (RFA) is accepted treatment for flat Barrett’s neoplasia. Less is known about RFA for esophageal squamous-cell neoplasia (ESCN). Our group has reported several prospective studies of RFA for ESCN in China with promising results through 12 months of follow-up. In the current study, we aimed to evaluate longer-term outcomes after RFA for ESCN.
Methods
Patients with flat unstained lesions (USLs) on Lugol’s endoscopy containing moderate/high grade intra-epithelial neoplasia (MGIN/HGIN) or mucosal cancer were treated with RFA every 3mo until complete remission (CR; no MGIN+). Patients with CR at 12mo (CR12) were included for follow-up and underwent annual Lugol’s endoscopy with biopsies and re-RFA for flat USLs. The clinical course of patients with persistent ESCN at 12mo (‘treatment failures’) is also reported.
Results
Among the 78 CR12 patients, 67 (86%) had sustained CR during a median 48mo (IQR 48–48) of follow-up and 5 endoscopies (IQR 4–6). Recurrence occurred in 7/78 patients (9%; MGIN(6), HGIN(1)); all were managed with RFA. Four other patients (5%) had progression (to HGIN(1); ESCC-sm(3)). During follow-up, protocol violations occurred in 46/78 patients (59%). Of the 12 treatment failures, progression occurred in 6. Overall, two patients developed subepithelial disease that was not visible after Lugol’s. Based on post-hoc analysis, the ‘pink-color sign’ at baseline (a pink color change after Lugol’s) significantly predicted failure after RFA.
Conclusion
RFA is relatively easy to apply and can efficiently treat large areas with ESCN. Despite protocol violations that may have interfered with the efficacy of RFA in 59% of patients, the great majority with CR12 had sustained CR during FU. However, some patients progressed to advanced disease and two developed subepithelial disease, not visible after Lugol’s. Based on currently available data, we advise to restrict the use of RFA for flat MGIN and HGIN without the pink-color sign on Lugol’s chromoendoscopy.
Registration
ClinicalTrials.gov identifier NCT02047305
Keywords: Esophageal squamous cell neoplasia, RFA, durability, recurrence
Introduction
Esophageal cancer is the 8th most common cancer worldwide, with a poor 5-year overall survival of 10–15% (1). Globally over 80% of esophageal cancers occur in developing countries, where nearly all cases are esophageal squamous cell carcinoma (ESCC). China has an especially high burden of disease; almost half of all ESCC cases in the world occur in China, where ESCC is the 4th leading cause of cancer-related death (1, 2).
When ESCC is diagnosed at a symptomatic stage, patients have a poor prognosis, because the majority of cases are already locally advanced and/or have metastasized. The prognosis is excellent, however, when ESCC is diagnosed at an early stage when the neoplasia is confined to the mucosal layer. In these cases, curative endoscopic treatment can be performed with preservation of the esophagus. This mucosal neoplasia is generally asymptomatic, but can be detected during screening endoscopy with the use of Lugol’s chromoendoscopy. Endoscopic screening programs are widely implemented in high-risk areas in China, with more than 200,000 screening endoscopies being performed each year. ESCC and its precursor lesions are detected in about 3% of these screening endoscopies(3).
The development of ESCC is a gradual process, starting with intraepithelial esophageal squamous cell neoplasia (ESCN). In China, precursors lesions are classified in 3 progressive stages according to the proportion of the epithelial layer containing neoplasia. Low grade (LGIN), moderate grade (MGIN) and high grade intra-epithelial neoplasia (HGIN) can be distinguished histologically, with 1/3, 2/3 and 3/3 respectively of the epithelial layer showing nuclear atypia, loss of normal cellular polarity and abnormal tissue maturation (4). MGIN and HGIN are considered indications for treatment given their progression rate to cancer (50% and 74% respectively over 13.5 years), whereas surveillance is indicated for LGIN (5).
Endoscopic treatment options include endoscopic resection (ER), either endoscopic mucosal resection (EMR) or endoscopic submucosal dissection (ESD), or ablation techniques such as radiofrequency ablation (RFA). ER enables en-bloc resection, thereby allowing adequate pathological assessment to evaluate the prognosis and the potential need for additional treatment. However, ER is technically demanding, and has a risk for complications (6). Most high-risk areas for ESCC have only limited endoscopic resources and expertise available and therefore, additional safe and effective treatment modalities of lower complexity are required. In addition, widespread ER may be associated with esophageal stenosis (6) and/or residual ESCN next to the ER scar (7,8). Ablation techniques may offer theoretical advantages in selected patients, such as more widespread and complete eradication of oncogenic abnormalities, which should result in lower local recurrence rates, and lower rates of esophageal stenosis (9).
In 2008 we initiated a prospective trial to assess the safety and efficacy of RFA for eradicating ESCN in patients with flat type MGIN, HGIN and early ESCC (10, 11). A complete remission (CR; absence of MGIN+ in biopsies) was established in 84% of patients at 12 months, with strictures occurring in 21%. After evaluation of several different circumferential RFA regimens, a single application of 12 J/cm2 after application of Lugol’s emerged as the favored regimen, resulting in a CR at 12 months of 82% and a stricture rate of 6%. This study however reported only a short-term follow-up duration of 12 months. The long-term durability of RFA for ESCN is therefore still unknown. The current study is a continuation of the aforementioned trial, and is the first to evaluate long-term durability of endoscopic RFA for eradicating MGIN, HGIN and early flat ESCC for up to 5 years after baseline RFA treatment.
Methods
The original 12-month study (ClinicalTrials.gov identifier NCT02047305) was conducted between October 2008 and October 2011 at the Cancer Institute and Hospital, Chinese Academy of Medical Sciences (CICAMS) in Beijing, China (10, 11). The present study was an extension of this study and was conducted between October 2009 and October 2016. The protocol was approved by the CICAMS Institutional Review Board and a written informed consent was signed by each patient. All authors had access to the study data and reviewed and approved the final manuscript.
Patients
Patients included in the original study were required to have at least one unstained lesion (USL) on Lugol’s endoscopy which measured ≥3cm in length, covered ≥25% of the esophageal circumference, and contained flat MGIN, HGIN or ESCC-m. Multiple USLs were allowed as long as the total USL-bearing esophagus was ≤12cm in length. EUS and CT were performed for patients with HGIN or ESCC-m to exclude submucosal invasion and/or lymphadenopathy. Other exclusion criteria were the presence of any non-flat lesions (lesions other than Paris type 0-IIb), esophageal strictures, and previous ablation therapy or ER. Further details of patient selection have been described previously (11).
All patients with a complete remission, defined as the absence of MGIN, HGIN or ESCC-m in biopsies at 12 months (CR12), were included in this study extension. Patients with persistent ESCN at 12 months (treatment failures) were excluded. Other exclusion criteria were non-circumferential radiofrequency ablation at baseline and discontinuance of FU before 12 months.
Protocol treatment phase (the first 12 months)
The treatment protocol for the original study has previously been described in detail (11). In summary, the USL-bearing esophagus was treated with one of four circumferential RFA regimens (BARXX360 system, Medtronic) at baseline (Supplementary Table 1) and sustainable tattoos were placed at the most proximal and distal edges to identify the treatment area (TA). Subsequently patients underwent endoscopies at 3-month intervals with biopsies and focal RFA (BARRX90, 3×12J/cm2, Medtronic) for persisting USLs, until CR was achieved. All patients underwent a 12 month endoscopy with Lugol’s staining and biopsies to assess the primary endpoint of the original study (i.e. the proportion of patients with a complete remission at 12 months).
Protocol follow-up phase (from 12 to 60 months)
The follow-up protocol consisted of annual follow-up endoscopies from 12 to 60 months after baseline. The TA was carefully evaluated by means of high resolution endoscopy, narrow band imaging, Lugol’s staining and histological analysis of 2 biopsies per 2 cm of the TA. Flat type USLs in the TA were also biopsied and treated with focal RFA. If subsequent biopsies showed MGIN or worse, reiterative RFA (rRFA) treatment was repeated every 3 months until complete remission was re-established. Patients with non-flat lesions (i.e. lesions other than Paris type 0-IIb) or lesions with invasive cancer in biopsies were treated with escape therapy according to institution’s standards of care. If CR was re-established after escape treatment, the regular FU protocol with endoscopies on an annual basis was resumed.
Outside of this formal FU study, we also followed the treatment failures at 12 months, which were treated and followed-up per investigator’s discretion, realizing that both treatments and FU protocols varied in these patients and were not the same as those who achieved CR12.
Based on progressive insight on the significance of the ‘pink-color sign’ (PCS) after Lugol’s staining (i.e. a color change after Lugol’s staining from an initial whitish-yellow color to a pink color 2–3 minutes later), we performed a post-hoc assessment of all baseline images to score the presence of a pink-color sign (Figure 1) (12).
Figure 1. A patient with a pink-color sign after Lugol’s staining.
(A) White light endoscopy image without visible abnormalities. (B) Directly after Lugol’s a large unstained lesion appeared from 3 to 7 o’clock and more proximally from 10 to 12 o’clock. Several minutes after application of Lugol’s, the entire USL turned pink (C).
Missing data for this study were retrospectively completed by review of medical records and all data were monitored by a study coordinator and research fellow from Amsterdam Medical Centre, the Netherlands.
Histopathology examination
The histological analysis of specimens during the original 12-month study has previously been described in detail (10, 11). All biopsy specimens obtained during the current follow-up study were routinely processed and reviewed by an expert gastrointestinal pathologist (L.X. or N.L.), with review of selected specimens by a second expert (S.M.D.). All specimens were scored as no intraepithelial neoplasia, LGIN, MGIN, HGIN, ESCC-m or ESCC>m (13), with the most advanced result determining the histology status of the patient. All ER and surgical specimens in the study were reviewed by one of the expert GI pathologists for depth of ESCC invasion, the presence of ESCC at the deep or lateral resection margins, the grade of ESCC differentiation and the presence of lymphovascular invasion (LVI).
Outcome measures
The primary endpoint of our follow-up study was the proportion of CR12 patients with sustained CR (defined as the absence of MGIN or worse in all TA biopsies) throughout the 48-month follow-up.
Secondary study outcomes were: (i) the proportion of patients with recurrent disease in the TA, defined as flat lesions with MGIN+ and with a lower or equal histological grade than the histological grade at study entry; (ii) the proportion of patients with progressive disease in the TA, defined as any non-flat lesion or detection of a more severe histological grade than the histological grade at study entry; (iii) the occurrence of adverse events, including perforation, infection, bleeding requiring transfusion, stricture or death; and (iv) the proportion of patients with the development of ESCN outside the TA.
Statistical analyses
Data analysis was performed using the IBM SPSS statistical software package (SPSS Inc., Chicago, IL). Means with standard deviations (SD) were computed for normally distributed variables and medians with interquartile ranges (IQR) for variables with a skewed distribution. Categorical variables were presented as frequencies and percentages of the total. Continuous and categorical variables were compared using the student’s t-test and Fisher’s exact test, respectively. All tests were two-sided and P values<0.05 were considered to be statistically significant.
The durability of complete remission was assessed using Kaplan Meier survival curves. All patients with CR12 were included for this analysis and ‘time-zero’ was the 12 months endoscopy. Using COX regression, univariate analysis was conducted to assess patient characteristics possibly associated with recurrent or progressive disease. All variables with p-value <0.3 were subsequently combined in multivariate COX analysis, and the variable with highest p-value was excluded until only variables with p<0.05 in multivariate analysis persisted. Additional analyses were performed to identify predictors for RFA failures in the total study population (i.e. all 90 patients who were initially treated with circumferential RFA and were followed ≥ 12 months), including both patients who failed to achieve a complete remission at 12 months and patients who had a complete remission at 12 months but developed recurrent or progressive disease during the 48-month follow-up (FU) period as RFA failures.
Results
Patients
Ninety-six patients were included in the original study, of which 78 were eligible for inclusion in the current follow-up study. Eighteen patients from the original study were excluded: 12 treatment failures at 12 months, 3 patients with focal RFA at baseline and 3 with discontinued follow-up before 12 months (lost to follow-up (n=2) or unrelated death (n=1)) (Figure 2). The baseline diagnosis of the 78 included patients was MGIN (n=39), HGIN (n=33) or ESCC-m (n=6), and the mean (±SD) USL length at baseline was 6.1 (± 2.9) cm (range 3–15) (Table 1).
Figure 2. Patient flow during the full 5 years of treatment and follow-up.
Abbreviations: CR - Complete response, defined as absence of MGIN or worse; CR12 - CR at 12 months endpoint of the original study; ESCC - esophageal squamous cell carcinoma; ESCC-sm — Submucosal esophageal squamous cell carcinoma; FU - follow-up; HGIN - High Grade Intraepithelial Neoplasia; IQR – Interquartile range; LTFU - Lost to follow-up; MGIN - Moderate Grade Intraepithelial Neoplasia; mo - months; RFA – RadioFrequency ablation therapy.
Table 1.
Characteristics, findings and treatments of the 78 patients in the follow-up study
| Protocol Treatment Phase (the first 12 months) | ||
|---|---|---|
| Male gender, n(%) | 41 (53) | |
| Age (years), mean (±SD) | 59.6 ± 6.6 | |
| Worst pathology grade at baseline, n (%) | MGIN | 39 (50) |
| HGIN | 33 (42) | |
| ESCC-m | 6 (8) | |
| Length of USL (cm), mean (±SD) | 6.1 ± 2.9 | |
| Length of TA (cm), mean (±SD) | 8.1 ± 2.8 | |
| Pink-color sign at baseline, n (%) | 11 (14) | |
| Baseline RFA regimen*, n (%) | A (Lugol-RFA-Clean-RFA) | 31 (40) |
| B (No Lugol-RFA) | 24 (31) | |
| C (Lugol-RFA) | 14 (18) | |
| D (Lugol-RFA-RFA) | 8 (10) | |
| Protocol Follow-up Phase (from 12 to 60 months) | ||
| FU time in months, median (IQR) | 48 (48–48) | |
| Endoscopies per patient, median (IQR) | 5 (4–6) | |
| Patients with rRFA | 25 (37%) | |
| Total rRFA sessions | 59 | |
| rRFA sessions / patient, median (IQR) | 1 (1–1) | |
Abbreviations: Cl – cleaning step; ESCC-m – mucosal esophageal squamous cell carcinoma; cm – centimeter; HGIN – high grade intraepithelial neoplasia; IQR – interquartile range; Lug – Lugol’s staining; MGIN – moderate grade intraepithelial neoplasia ; RFA – radiofrequency ablation therapy; rRFA – reiterative RFA during the follow-up period; SD – standard deviation; USL – unstained lesion
Four different circumferential RFA regimens were used at baseline (table 1), A (Lugol – RFA – cleaning – RFA), B (No Lugol – RFA), C (Lugol – RFA), D (Lugol – RFA - no cleaning – RFA). One patient was treated with a different regimen, consisting of two hits with 12J/cm2, with cleaning in between and without Lugol’s staining.
During a median follow-up of 48 months (IQR 48–48) after the 12-month endpoint of the original study, patients underwent a median of 5 (IQR 4–6) endoscopies. Sixty-seven (86%) patients completed the full 5-year follow-up of the study. The other 11 (14%) patients discontinued follow-up due to unrelated death (n=3) or unrelated comorbidity (n=2), or were lost to follow-up (n=6). The median follow-up of these 11 patients was 24 months (IQR 0–36).
Protocol violations (PV) occurred in a total of 46/78 patients (59%) and were categorized into 3 types: prolonged FU intervals (PV-1), USLs in the TA that were left untreated (PV-2), and inadequate follow-up after treatment of recurrent ESCN (PV-3). Each type of protocol violation (PV-1, PV-2 and/or PV-3) could occur at each FU endoscopy, so each PV could occur more than once in a single patient. PV-1, PV-2 and PV-3 occurred at least once in 24 (31%), 30 (38%) and 4 (4%) patients, respectively (Table 2).
Table 2. Protocol violations.
Three main type of protocol violations occurred during the FU period; 1) patients with prolonged FU intervals (PV-1); 2) patients with USLs that were left untreated (PV-2); and 3) patients with inadequate FU after retreatment for recurrent ESCN (PV-3). Each PV could occur multiple times in a single patient.
| Number of patients (%) | Median frequency of occurrence per patient (range) | |
|---|---|---|
| Prolonged FU intervals (PV-1) | 24 (31%) | 1 (1–3) |
| USLs that were left untreated (PV-2) | 30 (38%) | 1 (1–4) |
| Inadequate FU after new ESCN (PV-3) | 4 (5%) | 1 (1–3) |
| Any violation | 46 (59%) | 1 (1–4) |
Sustained CR, recurrence, and progression
Of the 78 patients with CR12, 67 patients (86%) had sustained CR during follow-up (Figure 2). In 25 of these 67 patients (37%), focal rRFA sessions were performed for USLs that had no ESCN (i.e. a USL was biopsied and ablated in the same session, and subsequent histology showed no ESCN), with a median of 1 treatment session per patient (IQR 1–1).
Of the 78 patients with CR12, 11 patients (14%) developed recurrent (n=7) or progressive (n=4) disease during follow-up (Figure 2). The mean recurrence-free survival time was 45 months (95% confidence interval 43–47) (Figure 3).
Figure 3. Kaplan Meijer Curve for durability of ESCN eradication after RFA.
Black curve (grey zone representing 95% confidence interval) shows the durability of CR12 after RFA for MGIN, HGIN and early ESCN during the extended follow-up. Eleven patients with recurrent or progressive disease were considered to be failures, even if CR was re-established after Radiofrequency Ablation.
Seven patients (9%) had recurrent disease (flat MGIN (n=6) or flat HGIN (n=1)) (Figure 4). Baseline pathology for these patients was MGIN (n=4) or HGIN (n=3). All recurrences were treated with rRFA, with CR re-established in 4 patients and CR results pending in 3 patients who were treated at their last follow-up endoscopy.
Figure 4. A recurrent USL during follow-up was successfully retreated with RFA.
At baseline, the patient had a large USL after Lugol’s, extending from 3 to 11 o’clock (A), with HGIN histology, which was treated with circumferential RFA (B). At 12 months patient was in complete remission with absence of USLs after Lugol’s (C) and negative biopsies. One year later, a recurrent USL was found (D), with MGIN histology. The patient was re-treated with a single session of focal RFA (E), and during subsequent follow-up endoscopies through 60 months, the patient re-achieved and sustained a complete remission (F).
Progressive disease was found in 4 other patients (5%), all with HGIN at baseline (Table 3). All were treated with endoscopic submucosal dissection (ESD); 2 were diagnosed with disease stages that required additional non-endoscopic treatment, but both refused further therapy. Patient 1 presented with a flat USL containing MGIN 2 years after baseline and was treated with focal rRFA. The patient returned 1 year later (PV-3) when a Paris type 0-IIa/c lesion was found. ESD was performed and pathology showed poorly differentiated ESCC-sm2. Patient 2 maintained CR through the 2-year follow-up endoscopy, but then, after a prolonged FU interval (PV-1), a slightly elevated lesion was observed that was normally stained after Lugol’s staining. ESD found an ESCC-sm3 that was buried under non-neoplastic epithelial cells (‘subepithelial disease’) and had a positive deep resection margin. Patient 3 had sustained CR through his 2-year follow-up endoscopy, but then did not return for 2 additional years (PV-1), at which time he was found to have a superficially elevated USL (Paris type 0-IIa). ESD showed a moderately differentiated ESCC-sm1. Patient 4 initially presented with persistent recurrent flat disease at 2, 3 and 4 years (HGIN, MGIN, MGIN) and was repeatedly treated with rRFA without achieving CR (PV-3s). At the 5-year endoscopy a Paris type 0-IIa/c USL containing HGIN was found and resected with ESD.
Table 3. Patients with progressive disease after an initial complete remission at 12 months.
Progressive disease was defined as any non-flat lesion or detection of a more severe histological grade than the grade at study entry.
| Pt | Baseline | Time (mo) | Endoscopic appearance * | Escape treatment | Worst pathology** |
|---|---|---|---|---|---|
| 1 | 4cm USL, HGIN, PCS+ | 40 | 0-IIa, USL, 1cm | ESD | ESCC-sm2, G3, LVI−, R0 |
| 2 | 11cm USL, HGIN, PCS− | 48 | 0-IIa, normally stained, 6cm | ESD | Subepithelial ESCC-sm3, G2, LVI−, R1 |
| 3 | 10cm USL, HGIN, PCS− | 48 | 0-IIa, USL, 1cm | ESD | ESCC-sm1, G2, LVI−, R0 |
| 4 | 11cm USL, HGIN, PCS− | 60 | 0-IIa/c, USL, 3cm | ESD | HGIN |
Abbreviations: CR – complete remission, ESCC-sm – submucosal esophageal squamous cell carcinoma, ESD – endoscopic submucosal dissection, HGIN – high-grade intraepithelial neoplasia, LVI− - no lymphovascular infiltration, MGIN – moderate-grade intraepithelial neoplasia, mo – months; PCS + – pink-color sign present, PCS− - pink-color sign absent, Pt – patient, R0 – radical resection with deep margins free of ESCN/ESCC, R1 – radical resection with deep resection margin involved by ESCN/ESCC, USL – unstained lesion
Appearance according to the Paris Classification of endoscopic lesions(27): 0-IIa = slight elevation of the mucosa, 0-IIb = flat mucosa, 0-IIc = slight mucosal depression, 0-IIa/c = combined type 0-IIa and 0-IIc lesion; the characteristics after Lugol’s staining; and the maximum length in cm.
Differentiation according to American Joint Committee on Cancer(28): G1 = well differentiated, G2 = moderately differentiated, G3 = poorly differentiated, G4 = undifferentiated
Other secondary outcomes
There were no serious adverse events such as perforation, infection, bleeding requiring transfusion or death during the study period. No new strictures were observed during the follow-up period. In the treatment phase a stricture occurred in twenty patients, as described in previous publications (11). Four of these patients required endoscopic dilatation (median 1 (IQR 1–1) session) during this FU phase and all strictures were resolved at the last study endoscopy. Stricturing during the treatment phase was not associated with recurrent or progressive disease during FU. Two patients developed a stricture after circumferential ESD, which was resolved after 7 and 8 dilatations, respectively.
ESCN-containing USLs outside the TA were found in 11 of 78 patients (14%; MGIN, n=7; HGIN, n=3; ESCC-m, n=1), and were treated with RFA (n=6), ESD (n=2) or otherwise (n=3).
Clinical course of treatment failures at 12 months
Of the 96 patients initially included in the original study, the 12 with residual ESCN at 12 months were identified as treatment failures (Figure 2). These patients were treated and followed-up per investigator’s discretion outside this formal FU study. We report on the clinical course after 12 months, realizing that both treatments and FU protocols varied in these patients and were not the same as those who achieved CR12.
Five (42%) of the 12 treatment failures achieved and sustained CR after additional RFA (i.e. RFA sessions after the 12 months of the original study) (n=4) or after EMR for a flat lesion containing HGIN (N=1), and 1 was lost to FU directly after 12 months. The other six patients (50%) showed progressive disease and were treated with surgery (n=4), chemoradiotherapy (n=1) or ESD (n=1) (Table 4). Three of these progressors had first achieved CR upon additional RFA and subsequently developed progressive disease, whereas the other 3 developed progressive disease from persistent ESCN. Of the 6 patients with progressive disease, 4 (67%) developed advanced ESCC that exceeded the limits of endoscopic therapy (1 ESCC-T3N1Mx, 1 ESCC-T1smN1M1, 1 ESCC-T2N0Mx and 1 ESCC-T1smNxMx). One patient developed subepithelial disease which was not clearly visible on Lugol’s chromoendoscopy (Table 4, Patient 2; Figure 5). This patient had a persistent USL with HGIN at 12 months and was treated with RFA twice, after which the USL had disappeared and biopsies were negative at 24 months. However, a small, atypical nodule had developed, with normal staining characteristics after Lugol’s and with negative biopsies. This lesion persisted on subsequent FU endoscopies, with repeatedly normal Lugol’s staining and negative biopsies. Biopsies at the 60 month endoscopy showed ESCC, and the patient was referred for surgery. Pathology assessment showed a poorly differentiated ESCC-sm covered by non-neoplastic squamous epithelium. One month after surgery, an enlarged cervical lymph node and squamous gastric metastasis were detected.
Table 4. Patients with persisting ESCN at 12 months that subsequently developed progressive disease.
Progressive disease was defined as any non-flat lesion or detection of a more severe histological grade than the grade at study entry.
| Pt | Baseline | 12 month PA | Time (mo) | Endoscopic appearance* | Escape treatment | Worst pathology** |
|---|---|---|---|---|---|---|
| 1 | 6cm USL, MGIN, PCS+ | HGIN | 48 | 0-IIa/c, USL, 5cm | Surgery, chemotherapy | ESCC-T3N1Mx, G2, LVI−, R0 |
| 2 | 4cm USL, ESCC, PCS+ | HGIN | 60 | Nodule, normally stained, 0.8cm | Surgery, chemoradiotherapy | ESCC- T1smN1M1, G3, LVI+, R0 |
| 3 | 12cm USL, HGIN, PCS− | HGIN | 42 | 0-I, USL | Surgery, chemotherapy | ESCC-T2N0Mx, G3, LVI−, R0 |
| 4 | 10cm USL, HGIN, PCS+ | MGIN | 55 | 0-IIa/c, USL, 6cm | ESD | ESCC-T1sm2NxMx, G2, LVI−, R0 |
| 5 | 6cm USL, HGIN, PCS− | HGIN | 60 | 0-I, USL | Chemoradiotherapy | HGIN*** |
| 6 | 8cm USL, HGIN, PCS− | HGIN | 36 | 0-IIa/c, USL, 2cm | Surgery | HGIN, R0 |
Abbreviations: APC – Argon Plasma Coagulation, EMR – endoscopic mucosal resection, ESCC – esophageal squamous cell carcinoma, ESD – endoscopic submucosal dissection, HGIN – high-grade intraepithelial neoplasia, LVI− - no lymphovascular infiltration, MGIN – moderate-grade intraepithelial neoplasia, mo – months; PA – pathology; PCS+ - pink-color sign present, PCS− - pink-color sign absent, Pt – patient, R0 – radical resection with vertical margins free of ESCN/ESCC, R1 – irradical resection with positive vertical resection margin, USL – unstained lesion
Appearance according to the Paris Classification(27) of endoscopic lesion: 0-IIa = slight elevation of the mucosa, 0-IIb = flat mucosa, 0-IIc = slight mucosal depression, 0-IIa/c = combined type 0-IIa and 0-IIc lesion; the characteristics after Lugol’s staining; and the maximum length in cm.
Differentiation according to American Joint Committee on Cancer(28): G1 = well differentiated, G2 = moderately differentiated, G3 = poorly differentiated, G4 = undifferentiated
This diagnosis was based on biopsies only.
Figure 5.
Post-RFA development of submucosal ESCC under normal squamous epithelium
Predictors for recurrence or progression
Of the 78 patients with CR at 12 months, 67 sustained CR whereas 11 developed recurrent or progressive disease. On univariate analysis, the likelihood of developing recurrent or progressive disease during follow-up after RFA was associated with a longer baseline USL length (hazard ratio (HR) 1.26, 95% confidence interval (95% CI) 1.07–1.49) (Table 5). Multivariate testing including baseline USL length and PCS showed independent association with recurrent or progressive disease only for baseline USL length.
Table 5.
Predictors of recurrent and progressive disease during 5 years of follow-up.
| Five year follow-up | COX analysis (univariate) | |||
|---|---|---|---|---|
| Sustained CR | Recurrence or progression | Hazard Ratio [95% CI] | P | |
| All patients | 67/78 (86%) | 11/78 (14%) | ||
| Baseline findings and treatments | ||||
| Baseline grade of ESCN | 0.40 | |||
| MGIN, n(%) | 35/67 (52%) | 4/11 (36%) | REF | REF |
| HGIN, n(%) | 26/67 (39%) | 7/11 (64%) | 2.33 [0.68–7.98] | 0.18 |
| ESCC-m, n(%) | 6/67 (9%) | 0/11 (0%) | N.A. | N.A. |
| Length of unstained lesions, cm | 5 (IQR 4–7) | 10 (IQR 4–12) | 1.26 [1.07–1.49] | <0.01 |
| Pink-color sign at baseline | 8/67 (12%) | 3/11 (27%) | 2.64 [0.70–9.96] | 0.23 |
| Baseline RFA regimen* | 0.64 | |||
| A, n(%) | 26/66 (39%) | 5/11 (45%) | REF | REF |
| B, n(%) | 23/66 (35%) | 1/11 (9%) | 0.27 [0.03–2.33] | 0.24 |
| C, n(%) | 11/66 (17%) | 3/11 (27%) | 1.53 [0.37–6.41] | 0.56 |
| D, n(%) | 6/66 (9%) | 2/11 (18%) | 1.52 [0.30–7.85] | 0.62 |
| Findings during the protocol treatment or follow-up phase | ||||
| Post-RFA stricture | 16/67 (24%) | 3/11 (27%) | 0.96 [0.25–3.60] | 0.95 |
| LGIN at 12mo | 11/67 (16%) | 3/11 (27%) | 1.67 [0.44–6.30] | 0.45 |
| Protocol violations | 37/67 (55%) | 9/11 (82%) | 2.53 [0.55–11.70] | 0.24 |
Abbreviations: CI – Confidence Interval; CR – complete remission, defined as absence of MGIN or worse in pathology assessment; ESCC-m – Mucosal esophageal squamous cell carcinoma; ESCN – esophageal squamous cell neoplasia, HGIN – high-grade intraepithelial neoplasia, , MGIN – moderate-grade intraepithelial neoplasia, RFA – Radio Frequency Ablation Therapy; USL – unstained lesion.
Four different circumferential RFA regimens were used at baseline (table 1), A (Lugol – RFA – cleaning – RFA), B (No Lugol – RFA), C (Lugol – RFA), D (Lugol – RFA - no cleaning – RFA)(11). One patient was treated with a different regimen, consisting of two hits with 12J/cm2, with cleaning in between and without Lugol’s staining.
In order to identify predictors for failure to achieve or sustain CR during 5 years after initial RFA, additional analyses were performed that included all 90 patients (78 with CR12 + 12 treatment failures) and considered both treatment failures at 12 months and CR12 patients who developed recurrent or progressive disease during follow-up as failures. On univariate analysis, baseline USL length (HR 1.22, 95% CI 1.08–1.37) and PCS (HR 4.01, 95% CI 1.75–9.23) were significantly associated with failure after RFA (Supplementary Table 2). Multivariate testing demonstrated that both USL length (HR 1.20, 95% CI 1.07–1.35) and PCS (HR 3.66, 95% CI 1.59–8.41) were independent predictors for failure to achieve or sustain CR during 5 years after initial RFA (Supplementary Figure 2).
Discussion
This is the first study to assess the long-term durability of RFA treatment of MGIN, HGIN and mucosal ESCC. In the 78 patients whose RFA resulted in complete eradication of ESCN during the treatment phase (CR12), the great majority (86%) sustained this eradication through 5 years after initial treatment. A few patients (9%) developed recurrent disease, and all could be treated with rRFA. Progressive disease was observed in 5%, and half of these could be curatively treated with ESD. In contrast, the 12 patients whose ESCN eradication was not successful during the 12 month treatment phase had relatively poor outcomes. Although 42% achieved and sustained CR after additional RFA, 50% developed progressive disease and the majority of these required non-endoscopic therapy. Two patients, including one with and one without CR12, developed subepithelial disease that was not clearly visible on Lugol’s chromoendoscopy.
The results of this study are the only information currently available on the long-term outcome after RFA for ESCN. It is important to note that the study was limited by suboptimal follow-up and protocol violations in a substantial number of patients. Nevertheless, three important lessons can be learned from our data. First, cases in whom complete remission of ESCN is not accomplished within a 12 month treatment period are prone to disease progression. Most of the progressors and the most advanced progressors were found in the group of patients with persistent ESCN at 12 months. These patients should therefore undergo escape treatment by EMR or ESD at 12 months to allow full histological review of the lesion and further treatment if clinically indicated. Second, RFA of ESCN requires rigorous endoscopic follow-up with Lugol’s chromoendoscopy. Stringent follow-up is required, given the rate of recurrent lesions in the treatment area and the occurrence of metachronous lesions in the remaining esophagus. All USLs seen after Lugol’s staining should be biopsied for histological diagnosis. Absence of USLs does not exclude the possibility of subepithelial neoplasia, and therefore, all non-flat areas (i.e. all lesions other than Paris type 0-IIb) in the treatment area should be sampled with keyhole biopsies, with a low threshold for a diagnostic EMR or ESD. Third, RFA should be restricted to patients with flat type (i.e. Paris type 0-IIb) USLs which contain MGIN or HGIN and do not have a pink-color sign after Lugol’s staining. ‘Flat lesions’ was already a major selection criterion in the current study, however, the distinction between 0-IIb and 0-IIa or 0-IIc may be difficult, especially in the West where ESCN is a rare disease and where endoscopists are less experienced with ESCN (14). We suggest adding the pink-color sign after Lugol’s chromoendoscopy as an additional endoscopic exclusion criterion for RFA treatment. We believe that RFA should be restricted to patients with lesions limited to the epithelium, i.e. MGIN or HGIN. The ablation effect of RFA covers the epithelium but may be insufficient for eradication of invasive neoplasia (i.e. ESCC) (15–17). The pre-treatment distinction between these two entities is therefore crucial, since RFA ablation therapy lacks the histological staging of endoscopic resection. In our study, the differentiation between ESCN and ESCC was based on endoscopic appearance (i.e. type 0-IIb lesions only) and pre-treatment biopsies (with an inevitable risk of biopsy sampling error). We feel that these two features may not suffice for adequate case selection prior to RFA. Therefore, we suggest adding the pink-color sign as a third selection criterion. This reddish or rose-pink color change, that typically occurs 2 to 3 minutes after Lugol’s staining, appears to be a characteristic of ESCC and more advanced stages of ESCN (12, 18–21). Although the recognition of the PCS seems rather straightforward and does not require advanced endoscopic skills, the inter- and intra-observer agreement has never been studied. Our results showed that the PSC was significantly associated with a failure to achieve or sustain CR after RFA: it independently predicted initial failure at 12 months (data not shown), and it predicted failure to achieve or sustain CR during 5 years after initial RFA. The PCS at baseline identified 58% of the treatment failures at 12 months and 27% of the CR12 patients with recurrent or progressive disease during follow-up. Thus the PCS may well be an extra safety criterion which can partially overcome the current limitations inherent in macroscopic assessment and biopsy sampling error of ESCN.
Data on short-term efficacy and safety of RFA for treatment of ESCN are limited but promising (10, 11, 22, 23). In our original 12-month study, the currently advised RFA regimen (Lugol staining followed by a single 12J/cm2 ablation) was associated with CR12 in 82% and a stenosis rate of 6% (11). This approach emerged as the optimal regimen from a number of small patient groups in which different regimens were used (10, 11). It is important to note that the number of cases treated with this “Lugol-12J” regimen was limited (only 17 cases), and these cases were assigned to this treatment on a temporal basis. To evaluate this regimen further, a new prospective study using this regimen in 100 additional patients is underway which incorporates lessons learned from the current study (Table 6). Inclusion criteria for this new trial require patients to have flat type USLs with MGIN or HGIN and no PCS, and a more rigorous follow-up protocol will include standard follow-up endoscopies every 3 months in the first 12-month treatment phase independent of pathology results. Patients with ESCN persisting at 12 months (i.e. treatment failures) will undergo EMR or ESD, whereas patients with a complete remission at 12 months will enter a four-year follow-up phase consisting of annual Lugol’s chromoendoscopy with dedicated inspection by experienced endoscopists and histologic assessment of biopsies. All lesions that exceed the initial inclusion criteria for RFA (i.e. non-flat lesions, PCS positive lesions or lesions with an ESCC diagnosis) will directly be treated with EMR or ESD to prevent progression to advanced disease, and keyhole biopsies will be performed on normal-staining non-flat lesions to evaluate the possibility of subepithelial ESCC. Based on this, we suggest 12 “do’s and don’ts” to optimize the chance of a successful outcome for RFA treatment of ESCN (Table 6).
Table 6.
Lessons learned from the current study to optimize outcomes of RFA for ESCN These suggestions are based on our experience and the current limited data. We would advise using RFA cautiously for ESCN, only by experienced endoscopists, and according to the following rules.
| 1 | Perform work-up with high-definition white light endoscopy and 1.25% Lugol’s chromoendoscopy, to:
|
| 2 | Perform EUS and CT for HGIN, to exclude ESCC and metastatic disease |
| 3 | Only include cases with type 0-IIB lesions containing MGIN or HGIN and no pink-color sign |
| 4 | Wait at least 2 weeks after the last endoscopy with Lugol’s staining before performing endoscopic treatment (the caustic effect of recent previous Lugol is associated with increased bleeding). |
| 5 | Re-inspect the esophagus with high-definition endoscopy and Lugol’s staining at the RFA session, to confirm that you are still dealing with only type 0-IIB lesions without the pink-color sign |
| 6 | The currently recommended RFA regimen is 1×12J/cm2 for circumferential RFA, and 3×12J/cm2 (no cleaning) for focal RFA |
| 7 | Tattoo the proximal and distal margins of the treatment area with 2 minute ≤0.5cc injections |
| 8 | Repeat follow-up endoscopies with Lugol’s staining and biopsy and RFA of all USLs >5mm every 3 months during the treatment phase (the first 12 months) |
| 9 | Perform EMR/ESD for all non-flat USLs, pink-color sign lesions or USLs with ESCC in biopsies during the treatment phase |
| 10 | After the 12 month endoscopy (the end of the treatment phase):
|
| 11 | Obtain biopsies of all USLs during follow-up and obtain keyhole biopsies of all non-flat lesions in the TA, even if there is normal Lugol’s staining, with a low threshold for EMR/ESD |
| 12 | Perform direct EMR/ESD of (i) all non-flat USLs, (ii) all lesions with a pink-color sign and (iii) all lesions with ESCC in the biopsies |
| 13 | In case of retreatment with RFA, begin follow-up again as if it were the initial treatment (follow steps 8–12, above) |
We found two patients with subepithelial ESCC that was not clearly visible after Lugol’s staining. These patients developed a non-flat lesion while the epithelium appeared normal on WLE, NBI and after Lugol’s. Pathology assessment for these patients showed ESCC-sm covered by non-neoplastic squamous epithelium. These subepithelial lesions are rightly feared for the risk they may pose to silently progress to advanced cancer without being visible on endoscopy. Besides being hard to detect, these lesions can be difficult to eradicate because it is easy to underestimate the true size of the tumor (24). ESCN has the potential to extend down the pre-existing epithelial-lined ducts of submucosal glands. This neoplastic extension may go as deep as the submucosal layer, without the epithelial neoplasia actually being invasive (“submucosally located intra-epithelial neoplasia”). This was shown by Tajima et al in a study of 83 surgically resected specimens with mucosal ESCC: neoplastic extension in these ducts was found in 14% (11/83), of which 45% (5/11) showed submucosally located intra-epithelial neoplasia (25). Jansen et al reported even higher rates of neoplastic extension into the ducts in 65 ESCC ESD specimens: neoplastic ductal extension was found in 60% (39/65), with submucosally located intra-epithelial neoplasia in 33% (13/39) (14).
Extension of ESCN down pre-existing ducts of submucosal glands may be associated with residual ESCN after RFA treatment if the ablation remains too superficial. RFA intentionally aims to ablate the epithelium and muscularis mucosa but preserve the submucosa, reasoning that this will reduce the risk for complications like bleeding, fibrosis and stricturing. Three studies have assessed the depth of ablation after RFA in humans at a histopathological level. Using different circumferential treatment regimens (either 1, 2 or 3 hits with 8 to 14 J/cm2) (15–17), all reported damage limited to the mucosa without any injury in the submucosa. However, the depth of ablation in these studies was only assessed through 2 days post ablation, and ablation may extend deeper over subsequent days. An animal study that assessed ablation depth after RFA in pig esophagi reported more advanced injury at 3 days post-ablation than at 0, 2, 5 or 7 days post-ablation (26). Further study with a larger number of patients and careful evaluation with follow-up endoscopies by experienced endoscopists will be required to further evaluate the risk of subepithelial ESCN after the use of RFA for squamous neoplasia.
This is the first long-term follow-up study after RFA for patients with ESCN. Strengths of our study include the prospective design, the relatively large number of patients, and the fact that the great majority of patients completed 60 months of follow-up. Furthermore, we used a standardized biopsy protocol with a large number of biopsies performed to sample both the normally stained treatment area and recurrent or persistent USLs at each visit, and all biopsies were reviewed by an expert gastrointestinal pathologist with selected review by a second expert. Lastly, we used clear histological endpoints.
Two major limitations need to be addressed. The first was the suboptimal follow-up with protocol violations in a substantial number of patients. Although no association was found between PVs and the occurrence of recurrent or progressive disease (Table 5), this still may have affected our results and the results might have been better if we had strictly adhered to the study protocol. Although this wouldn’t have influenced the incidence of recurrent disease, it might have contributed to disease progression in some of our patients. This emphasizes the importance of strict follow-up regimens with a low threshold for escape treatment during treatment and follow-up phases.
The second major limitation was the use of different RFA regimens at baseline. Given the relatively small numbers of patients per subgroup, the temporal assignment of cases to each subgroup and the variability of regimen among the subgroups, the long-term results by treatment group were of little value in the choice of an optimal treatment regimen.
Another limitation of our study was the post-hoc assessment of the pink-color sign by reviewing pictures of the baseline endoscopy. The pink-color sign was first reported in the literature as predictor of HGIN or ESCC during the course of our study and, therefore, at baseline we did not report on its presence.
Conclusion
The great majority (86%) of patients with ESCN or mucosal ESCC who had a complete remission (no residual disease) 12 months after initial RFA treatment (CR12) experienced sustained eradication of neoplasia during an additional 4 years of follow-up. Among the few patients with a complete remission at 12 months who experienced new ESCN during follow-up, the majority could be treated with RFA and only a few showed progression of disease. Patients with residual ESCN at 12 months, however, were at higher risk for progression to advanced disease. Overall, a significant number of those who progressed developed advanced ESCC that required non-endoscopic therapy, and two patients developed subepithelial ESCC that was not clearly visible on Lugol’s chromoendoscopy.
Based on our study, which was limited by protocol violations and inadequate follow-up in a significant number of patients, we conclude that RFA may be best suited for treatment of non-invasive epithelial neoplasia (i.e. MGIN or HGIN), and cannot be recommended for treatment of ESCC. Given the complexity of differentiating ESCN from ESCC prior to treatment, we advise to use RFA cautiously and by experienced endoscopists only. We have suggested several “lessons learned” to optimize the chance of a successful outcome. Further studies with careful selection of patients and strict treatment and follow-up protocols performed by experienced endoscopists are needed to further clarify the role of RFA in the treatment of patients with ESCN.
Supplementary Material
Acknowledgments
This was a sponsor initiated trial, supported by Medtronic Inc., Medtronic supported study personnel cost and provided ablation devices and access to a central electronic data management system. Radiofrequency ablation is not FDA cleared for use in esophageal squamous cell neoplasia in the United States.
Abbreviations
- CI
Confidence interval
- CICAMS
Cancer Institute and Hospital, Chinese Academy of Medical Sciences
- CR
Complete remission
- CR12
Complete remission at 12 months
- ESCC
Esophageal squamous cell carcinoma
- ESCC-m
Mucosal esophageal squamous cell carcinoma (=ESCC-T1m)
- ESCC-m2
Mucosal esophageal squamous cell carcinoma, invading the lamina propria
- ESCC-m3
Mucosal esophageal squamous cell carcinoma, extending to or invading the muscularis mucosae
- ESCC>m
Esophageal squamous cell carcinoma with invasion deeper than the mucosa
- ESCC-sm
Submucosal esophageal squamous cell carcinoma (=ESCC-T1sm)
- ESCC-sm1
Submucosal esophageal squamous cell carcinoma, invading the upper 1/3 of the submucosa
- ESCC-sm2
Submucosal esophageal squamous cell carcinoma, invading the upper 2/3 of the submucosa
- ESCC-sm3
Submucosal esophageal squamous cell carcinoma, invading >2/3 of the submucosa
- ESCC-T1m
Mucosal esophageal squamous cell carcinoma (=ESCC-m)
- ESCC-T1sm
Submucosal esophageal squamous cell carcinoma (=ESCC-sm)
- ESCC-T2
Esophageal squamous cell carcinoma, invading the muscularis propria
- ESCC-T3
Esophageal squamous cell carcinoma, invading the adventitia
- ER
Endoscopic resection
- ESCN
Esophageal squamous cell neoplasia
- ESD
Endoscopic submucosal dissection
- FU
Follow-up
- HGIN
High grade intraepithelial neoplasia
- HR
Hazard ratio
- IQR
Interquartile range
- LGIN
Low grade intraepithelial neoplasia
- LVI
Lymphovascular invasion
- MGIN
Moderate grade intraepithelial neoplasia
- MGIN+
Moderate grade intraepithelial neoplasia or a worse histologic grade
- PCS
Pink-color sign
- RFA
Radiofrequency ablation therapy
- rRFA
Reiterative radiofrequency ablation therapy (i.e. after 12 months)
- PV
Protocol violations
- PV-1
Protocol violations type 1
- PV-2
Protocol violations type 2
- PV-3
Protocol violations type 3
- SD
Standard deviation
- TA
Treatment area
- USL
Unstained lesion
References
- 1.Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136(5):E359–86. [DOI] [PubMed] [Google Scholar]
- 2.Lin Y, Totsuka Y, He Y, Kikuchi S, Qiao Y, Ueda J, et al. Epidemiology of esophageal cancer in Japan and China. J Epidemiol. 2013;23(4):233–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.China Cancer Foundation. Rural cancer early diagnosis and treatment committee of experts aCfdcapC. Work report on the early diagnosis and treatment for cancer in rural areas of China 2015/2016. People’s medical publisher; 2016. p. 1–2. [Google Scholar]
- 4.Dawsey SM, Lewin KJ, Wang GQ, Liu FS, Nieberg RK, Yu Y, 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]
- 5.Wang GQ, Abnet CC, Shen Q, Lewin KJ, Sun XD, Roth MJ, 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]
- 6.Sgourakis G, Gockel I, Lang H. Endoscopic and surgical resection of T1a/T1b esophageal neoplasms: a systematic review. World J Gastroenterol. 2013;19(9):1424–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Muto M, Takahashi M, Ohtsu A, Ebihara S, Yoshida S, Esumi H. Risk of multiple squamous cell carcinomas both in the esophagus and the head and neck region. Carcinogenesis. 2005;26(5):1008–12. [DOI] [PubMed] [Google Scholar]
- 8.Ohashi S, Miyamoto S, Kikuchi O, Goto T, Amanuma Y, Muto M. Recent Advances From Basic and Clinical Studies of Esophageal Squamous Cell Carcinoma. Gastroenterology. 2015;149(7):1700–15. [DOI] [PubMed] [Google Scholar]
- 9.van Vilsteren FG, Pouw RE, Seewald S, Alvarez Herrero L, Sondermeijer CM, Visser M, et al. Stepwise radical endoscopic resection versus radiofrequency ablation for Barrett’s oesophagus with high-grade dysplasia or early cancer: a multicentre randomised trial. Gut. 2011;60(6):765–73. [DOI] [PubMed] [Google Scholar]
- 10.Bergman JJ, Zhang YM, He S, Weusten B, Xue L, Fleischer DE, et al. Outcomes from a prospective trial of endoscopic radiofrequency ablation of early squamous cell neoplasia of the esophagus. Gastrointest Endosc. 2011;74(6):1181–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.He S, Bergman J, Zhang Y, Weusten B, Xue L, Qin X, et al. Endoscopic radiofrequency ablation for early esophageal squamous cell neoplasia: report of safety and effectiveness from a large prospective trial. Endoscopy. 2015;47(5):398–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Goda K, Dobashi A, Yoshimura N, Kato M, Aihara H, Sumiyama K, et al. Narrow-Band Imaging Magnifying Endoscopy versus Lugol Chromoendoscopy with Pink-Color Sign Assessment in the Diagnosis of Superficial Esophageal Squamous Neoplasms: A Randomised Noninferiority Trial. Gastroenterol Res Pract. 2015;2015:639462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Dawsey SM, Lewin KJ, Liu FS, Wang GQ, Shen Q. Esophageal morphology from Linxian, China. Squamous histologic findings in 754 patients. Cancer. 1994;73(8):2027–37. [DOI] [PubMed] [Google Scholar]
- 14.Jansen M, Scholvinck DW, Kushima R, Sekine S, Weusten BL, Wang GQ, et al. Is it justified to ablate flat-type esophageal squamous cancer? An analysis of endoscopic submucosal dissection specimens of lesions meeting the selection criteria of radiofrequency studies. Gastrointest Endosc. 2014;80(6):995–1002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ganz RA, Utley DS, Stern RA, Jackson J, Batts KP, Termin P. Complete ablation of esophageal epithelium with a balloon-based bipolar electrode: a phased evaluation in the porcine and in the human esophagus. Gastrointest Endosc. 2004;60(6):1002–10. [DOI] [PubMed] [Google Scholar]
- 16.Dunkin BJ, Martinez J, Bejarano PA, Smith CD, Chang K, Livingstone AS, et al. Thin-layer ablation of human esophageal epithelium using a bipolar radiofrequency balloon device. Surg Endosc. 2006;20(1):125–30. [DOI] [PubMed] [Google Scholar]
- 17.Smith CD, Bejarano PA, Melvin WS, Patti MG, Muthusamy R, Dunkin BJ. Endoscopic ablation of intestinal metaplasia containing high-grade dysplasia in esophagectomy patients using a balloon-based ablation system. Surg Endosc. 2007;21(4):560–9. [DOI] [PubMed] [Google Scholar]
- 18.Ishihara R, Kanzaki H, Iishi H, Nagai K, Matsui F, Yamashina T, et al. Pink-color sign in esophageal squamous neoplasia, and speculation regarding the underlying mechanism. World J Gastroenterol. 2013;19(27):4300–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Shimizu Y, Omori T, Yokoyama A, Yoshida T, Hirota J, Ono Y, et al. Endoscopic diagnosis of early squamous neoplasia of the esophagus with iodine staining: high-grade intra-epithelial neoplasia turns pink within a few minutes. J Gastroenterol Hepatol. 2008;23(4):546–50. [DOI] [PubMed] [Google Scholar]
- 20.Ishihara R, Yamada T, Iishi H, Kato M, Yamamoto S, Yamamoto S, et al. Quantitative analysis of the color change after iodine staining for diagnosing esophageal high-grade intraepithelial neoplasia and invasive cancer. Gastrointest Endosc. 2009;69(2):213–8. [DOI] [PubMed] [Google Scholar]
- 21.Takahashi M, Shimizu Y, Ono M, Suzuki M, Omori S, Yoshida T, et al. Endoscopic diagnosis of early neoplasia of the esophagus with narrow band imaging: correlations among background coloration and iodine staining findings. J Gastroenterol Hepatol. 2014;29(4):762–8. [DOI] [PubMed] [Google Scholar]
- 22.van Vilsteren FG, Alvarez Herrero L, Pouw RE, ten Kate FJ, Visser M, Seldenrijk CA, et al. Radiofrequency ablation for the endoscopic eradication of esophageal squamous high grade intraepithelial neoplasia and mucosal squamous cell carcinoma. Endoscopy. 2011;43(4):282–90. [DOI] [PubMed] [Google Scholar]
- 23.Becker V, Bajbouj M, Schmid RM, Meining A. Multimodal endoscopic therapy for multifocal intraepithelial neoplasia and superficial esophageal squamous cell carcinoma - a case series. Endoscopy. 2011;43(4):360–4. [DOI] [PubMed] [Google Scholar]
- 24.Erard-Poinsot D, Pioche M, Chabrun E, Rivory J, Belleannee G, Hervieu V, et al. Buried adenocarcinoma hidden by normal squamous epithelium in Barrett’s esophagus: should we enlarge the margins for endoscopic resections? Endoscopy. 2015;47 Suppl 1 UCTN:E439–41. [DOI] [PubMed] [Google Scholar]
- 25.Tajima Y, Nakanishi Y, Tachimori Y, Kato H, Watanabe H, Yamaguchi H, et al. Significance of involvement by squamous cell carcinoma of the ducts of esophageal submucosal glands. Analysis of 201 surgically resected superficial squamous cell carcinomas. Cancer. 2000;89(2):248–54. [DOI] [PubMed] [Google Scholar]
- 26.Kruger L, Gonzalez LM, Pridgen TA, McCall SJ, von Furstenberg R, Harnden I, et al. Ductular and proliferative response of esophageal submucosal glands in a porcine model of esophageal injury and repair. Am J Physiol Gastrointest Liver Physiol. 2017:ajpgi 00036 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Group ECR. Update on the Paris classification of superficial neoplastic lesions in the digestive tract. Endoscopy. 2005(37):570–8. [DOI] [PubMed] [Google Scholar]
- 28.Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol. 2010;17(6):1471–4. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.