Introduction

Scar excision with endoscopic full-thickness resection (eFTR) offers a minimally invasive alternative to completion surgery after incomplete resection of low-risk T1 colorectal cancer (CRC). While this strategy has demonstrated promising technical and short-term results, long-term data on oncological safety are lacking.

Aims & Methods

Consecutive patients undergoing eFTR scar excision after incomplete resection of low-risk T1CRC were selected from the Dutch prospective eFTR registry. Incomplete resection was defined as indeterminate (Rx), tumour-positive (R1, <0.1mm) or <1mm tumour-free (R0) resection margins. Lesions were considered low-risk in absence of lymphovascular invasion, poor tumour differentiation and tumour budding grade 2-3 (if assessed). A retrospective analysis was conducted to evaluate the procedural outcomes, subsequent treatment after eFTR, and surveillance data, including endoscopy, CEA and imaging studies, as outlined by local protocols. Study outcomes included the 3-year locoregional recurrence rate (LRR), distant recurrence rate (DRR), salvageable recurrence rate (SRR), overall survival (OS) and organ preservation rate. The 3-year estimates were calculated using the Kaplan-Meier method. Locoregional recurrence was defined as intramural cancer recurrence and/or locoregional lymph node metastasis. Other cancer recurrence was defined as distant recurrence. Salvageable recurrence was defined as cancer recurrence that could be treated with curative intent.

Results

Out of the 1341 procedures registered in the Dutch prospective eFTR registry between 2015 and 2023, 314 patients met the inclusion criteria (mean age 69 years, 62% male). Initial histology showed Rx/R1 margins in 260/314 (83%) and R0 <1mm in 54/314 (17%). Mean time from initial resection to eFTR was 2 months (SD 2). Macroscopic complete scar excision was achieved in 259/314 (83%) patients. At histology, 36/314 (12%) eFTR specimens had residual cancer (15 low-risk T1CRC, 21 high-risk T1CRC/≥T2CRC). Residual cancer was found in 26/143 (18%) patients after initial piecemeal resection versus 10/171 (6%) after en bloc resection. No residual cancer was found after R0 <1mm (0/54). The R0-resection rate for residual cancer was 25/36 (69%). After eFTR, 30/314 (10%) patients underwent completion surgery, of whom 9/30 (30%) had residual cancer. One patient received adjuvant chemoradiotherapy. Surveillance was initiated in the remaining 283/314 (90%) patients. In the surveillance-only group, 246/283 (87%) had at least one recorded follow-up visit, with a median follow-up time of 36 months (IQR 16-52). Of these, 17/246 (7%) had residual cancer in the eFTR specimen. The 3-year LRR, DRR, SRR and OS for this subgroup (n=17) were 13%, 0%, 100% and 94%, respectively. Of the 229/246 (93%) patients without residual cancer in the scar, 187/229 (82%) had no risk factors at initial histology, while the presence of risk factors could not be reliably assessed in 42/229 (18%). In the low-risk group (n=187), the 3-year LRR, DRR, SRR and OS were 1%, 1%, 99% and 98%, respectively. In the group with indeterminate risk factors (n=42), these were 7%, 3%, 97% and 92%. The overall 3-year organ preservation rate for all 314 eFTR procedures initiated was 87%.

Conclusion

In this analysis of a large Dutch prospective eFTR registry, the use of completion eFTR after incomplete resection of low-risk T1CRC demonstrated excellent 3-year oncological outcomes. However, caution should be exercised if the presence of histological risk factors cannot be adequately assessed. Further prospective studies with strict surveillance are warranted.

Introduction

The Lyon 2.0 Consensus accepts that wireless pH monitoring provides equivalent information to catheter-based studies and recommends 96-hr wireless pH studies for gastroesophageal reflux disease (GERD) diagnosis.¹ However, the pH sensor may detach early, and 24-hr and 48-hr studies are routine (and cost-efficient) in many centers. Published studies have used different analyses to summarize prolonged pH data, including “average acid exposure time (AET),” “worst-day AET,” and “dominant pattern analysis."

Aims & Methods

This study aimed to determine (i) when a shorter recording time is sufficient for conclusive diagnosis of GERD; (ii) the impact of prolonged pH studies on patients with an inconclusive diagnosis of GERD based on 24- and 48-hr measurements; (iii) the replicability and diagnostic yield of 24-hr, 48-hr, and 72-hr studies compared with the 96-hr reference standard; (iv) the performance of different analyses of prolonged pH data.
We analyzed endoscopy and pH data from 944/2045 (46%) consecutive patients (age 46.5 ± 14.6 years, 65% F) referred for investigation of typical reflux symptoms off medication that had at least 92-hr measurement. Patients were classified at 24, 48, and 72-hr against the 96-hr reference standard. AET <4% was considered conclusively negative, and AET >6% conclusively positive for GERD. Los Angeles (LA) esophagitis was an independent marker of disease. The effect of prolonging pH measurement on GERD diagnosis was evaluated using average day, worst-day, and dominant pattern analysis (≥2 negative or positive days). Receiver operating curve (ROC) analysis was performed on the 24- and 48-hr pH data in patients with normal endoscopy, LA A and B esophagitis. Groups defined by 24-hr AET thresholds from 1% to 7% were followed up to 96 hr to assess when 24-hr data were sufficient for conclusive diagnosis.

Results

The proportion of patients with inconclusive results (average AET 4-6%) reduced from 113 at 24 hours to 40 at 96 hours (35% of the subgroup; p=0.02), with similar results for dominant pattern analysis, but not for worst-day analysis (p=0.2). When 24- and 48-hr test periods were assessed, prolonging average day studies to 72 and 96 hr increased conclusive GERD diagnosis to 71% and 65%, respectively. On dominant pattern analysis, prolonging pH studies up to 96-hr in the subgroup of patients with inconclusive diagnoses at 48-hr showed a decrease in inconclusive diagnoses from 100% to 59% (p < 0.05). A stepwise improvement in diagnostic accuracy and yield was observed from 24, 48, and 72-hr measurements compared to the 96-hr reference standard (p<0.00001 for trend).
On ROC analysis, 24-hr AET > 6.4% and 48-hr AET > 6.1% predicted GERD based on 96-hr measurements in all patients with normal endoscopy (table). 93% of patients with LA grade B esophagitis had AET >6%. Only 2.6% of patients with a 24-hr AET ≤ 2.0% had GERD at 96 hr. Over 93% of these had physiologic acid exposure (AET <4%) on reference standard 96-hr studies. When the analysis was repeated for 48-hr data, there was a <1% probability of GERD at 96 hr.

Conclusion

This study confirms that prolonging studies from 24-hr catheter-based or 48-hr wireless pH measurements to ≥72-hr is sufficient for conclusive GERD diagnosis; however, clearly negative (< 2% AET) or positive results (> 6% AET) from 24-hr studies are sufficient to exclude or confirm GERD. Prolonged pH studies of up to 96-hr duration improve the diagnostic yield for conclusive GERD. Analysis methods that utilize average day AET or dominant pattern are preferred.

References

1. Gyawali CP, Yadlapati R, Fass R, et al. Updates to the modern diagnosis of GERD: Lyon consensus
2.0. Gut 2024;73(2):361-371.