Esophageal resection (esophagectomy) has been a standard surgical treatment option for any resectable esophageal cancer. Esophagectomy is one of the most complex, invasive procedures in the upper gastrointestinal tract, and is associated with high mortality and morbidity even with the recent refinement of surgical techniques and perioperative care.^1,2 Accumulating data have suggested that the probability of lymph node involvement in patients with early esophageal cancer (T1a intramucosal cancer) is unlikely (<2%), and esophagectomy may be unnecessarily invasive for patients with early esophageal cancer.^3–5 With the introduction of endoscopic surveillance program and advancement in optic technology, patients with early esophageal cancer have been increasingly encountered, and interest in esophageal-preserving endoscopic treatments has grown. Esophageal-preserving treatments include any endoluminal procedure that is performed in an attempt to completely eradicate disease while preserving the anatomical structure of esophagus. There are primarily two esophageal-preserving treatments: endoscopic ablation (radiofrequency ablation and cryotherapy) and endoscopic resection (endoscopic mucosal resection and endoscopic submucosal dissection). However, the concept of esophageal-preserving treatments has caused confusion in the decision-making among health care providers. Patients with early esophageal cancer have a chance for cure, and it is therefore extremely important to determine if esophageal-preserving treatments are adequate to achieve complete cure, that is, minimizing residual or recurrent disease. In this chapter, we focus on esophageal-preserving treatments for patients with early esophageal cancer.

The most critical step for esophageal-preserving treatments is appropriate patient selection, and patients with high risk of lymph node involvement and/or metastatic disease need to be excluded as candidates for esophageal-preserving treatments. Esophageal-preserving treatments start with meticulous endoscopic examination of esophageal epithelium with extensive biopsies for tissue diagnosis. To provide high-quality endoscopic images, several new endoscopic technologies (e.g., optical coherent tomography, autofluorescent imaging, confocal laser endomicroscopy) combined with enhancement techniques (e.g., narrow band imaging, chromoendoscopy) have been introduced and investigated; however, none of them has been routinely used in general practice. For early esophageal cancer, the depth of tumor invasion and assessment of certain pathological features such as ulceration or grade are extremely important to estimate potential lymph node involvement. Therefore, endoscopic mucosal resection for staging purposes is essential for accurate clinical staging and risk stratification based on pathological assessment. Endoscopic ultrasound (EUS) has been utilized to exclude lymph node metastasis and to determine the depth of tumor invasion, and EUS can accurately differentiate T1 and T2 tumors;⁶ although EUS⁶ has a limited accuracy, it still has a limited accuracy⁷ to discriminate between T1a and T1b tumors even with high-frequency miniprobe (20 or 30 MHz).

T1a (Intramucosal) Esophageal Cancer

Esophageal-preserving treatments can be indicated for patients with T1a adenocarcinoma with low risk or no risk of lymph node involvement or metastatic disease. High-risk factors for lymph node involvement include submucosal invasion (T1b), lymphovascular invasion (L+ and/or V+), poor differentiation, and a nodule >3 cm in diameter.^8–11 By contrast, low-risk factors include type I, IIa < 2 cm, IIb, IIc < 1 cm, well or moderately differentiated adenocarcinoma, and no lymphovascular invasion (L- and V-).^9,10 Risk factors for T1a adenocarcinoma are summarized in Table 85-1.

Since esophageal squamous cell cancer appears to be biologically more aggressive than adenocarcinoma,¹² the probability of lymph node involvement is higher in patients with squamous cell cancer compared to those with adenocarcinoma. Previous studies have demonstrated that the risk of lymph node involvement in patient with intraepithelial cancers (m1) and cancers invading the lamina propria (m2) is highly unlikely,^13–15 whereas the risk of lymph node involvement in cancers invading the muscularis mucosae (m3) and the submucosa (sm) ranges from 0% to 10%¹⁵ and from 50% to 55%,¹⁴ respectively. Based on this risk stratification, patients with superficial squamous cell cancers (m1 and m2) with well-to-moderate differentiation and no lymphovascular invasion can be a candidate for esophageal-preserving treatments. Patients with m3 squamous cell cancers could be a candidate for esophageal-preserving treatments if there are no further risk factors for lymph node involvement. Patients with submucosal squamous cell cancers (T1b) require esophageal resection. Risk factors for early squamous cell carcinoma (SCC) are summarized in Table 85-2.

T1b Esophageal Adenocarcinoma

Due to the abundant lymphatic network in the submucosal layer, the probability of lymph node involvement is exponentially increased once tumors invade into the submucosa.^16,17 Based on this, esophagectomy has been recommended as a standard of care for patients with T1b esophageal cancers. A recent review analyzing the pooled data of 7645 patients with T1b submucosal esophageal cancer has demonstrated that the overall rate of lymph node involvement in patients with T1b cancers was 37%. However, there was a substantial difference between T1sm1 and T1sm2/3 adenocarcinomas (6% vs. 23% or 58%, respectively). These data suggest that highly selected patients with T1sm1 adenocarcinoma could be treated with esophageal-preserving approaches.¹² The most recent study involving 66 patients with low-risk T1sm1 cancer (polypoid or flat lesion, well-to-moderate differentiation and no lymphovascular invasion) has demonstrated that complete remission was achieved in 97% of patients with nodules ≤2 cm, and long-term remission without any metachronous disease was achieved in 90%. Furthermore, there were no tumor-related deaths and the estimated 5-year survival was 84%, although one patient (1.5%) developed lymph node metastasis.¹⁸ Although further studies with long-term follow-up data are required, these data suggest that the risk of developing lymph node metastasis in patients with T1sm1 adenocarcinoma after esophageal-preserving treatments may be lower than the postoperative mortality related to esophagectomy, which still can exceed 3%. Patients with low-risk T1sm1 adenocarcinoma could be treated with esophageal preserving treatments, especially when poor functional status and comorbid conditions make esophagectomy too risky. T1sm2 and T1sm3 adenocarcinoma and all T1b SCCs are associated with a substantially higher risk of lymph node involvement, and esophagectomy should be considered.^19,20 Since these data were achieved within high-volume, experienced centers, these recommendations may not be transferable to patients at all centers.

Radiofrequency ablation (RFA) and cryotherapy have been primarily performed as endoscopic ablation therapy. Since the purpose of ablation therapy is to eradicate disease by ablating the abnormal esophageal epithelium, no specimen is available for histological assessment.

Radiofrequency Ablation

Radiofrequency ablation (RFA) using the Barrx™ Ablation System (Covidien, Sunnyvale, CA) has been most commonly performed as endoscopic ablation therapy, especially for Barrett’s esophagus (BE). Either an ablation balloon catheter (Barrx™ 360 RFA Balloon Catheter) for circumferential ablation or an endoscopic mounted device (Barrx™ 90, 60, Ultra Long RFA Focal Catheter) for focal ablation can be selected based on the length, extension, and location of disease. RFA delivers a high-power, ultra-short burst of ablative energy to the abnormal esophageal epithelium, and the energy delivered provides uniform treatment to a depth of approximately 500 µm. Therefore, the depth of treatment is limited to the mucosal layer and the risk of stricture formation is significantly reduced. The multicenter, randomized, sham-controlled trial involving 127 patients with BE demonstrated that 81% of patients with high-grade dysplasia (HGD) and 90% of those with low-grade dysplasia had complete eradication of dysplasia compared to 19% in the control group (no RFA) (p < 0.001), and patients who underwent RFA had significantly less disease progression (3.6% vs. 16.3%, p = 0.03) and reduced progression to invasive cancer (1.2% vs. 9.3%, p = 0.045) during the follow-up of 12 months. The rate of stricture formation was 6%.²¹ However, RFA is currently not recommended for cancer due to the limited depth of treatment. RFA has been commonly used to ablate the remaining flat BE following endoscopic resection of nodular BE.

Cryotherapy

Cryotherapy is a noncontact ablative technique that involves the topical application of spraying aerosolized liquid nitrogen or carbon dioxide onto the abnormal esophageal epithelium, providing intracellular disruption and ischemia while preserving the extracellular matrix and thereby minimizing fibrosis. The depth of treatment can be adjustable, depending on the dose of applications, and therefore cryotherapy can be used for early esophageal cancer. Current devices for cryotherapy require a venting system (e.g., nasogastric tube) to remove excessive nitrogen gas from the esophagus and stomach, thus preventing perforation of the gastrointestinal tract. Since cryotherapy is a relatively new ablation technique, the clinical data for cryotherapy in the treatment of esophageal cancer is limited. In the multicenter, retrospective cohort study to evaluate the efficacy and safety of cryotherapy for esophageal cancer, complete eradication of T1a tumors was achieved in 75% (18/24) of patients. For T1b tumors, complete eradication was achieved in 60% (4/6) of patients with a mean follow-up of 11.8 months.²² Current device versions have several technical issues such as the nonuniform application using a handheld catheter, the fogging of the scope lens, and the prolonged duration of treatment. A novel through-the-scope cryoballoon device, which does not require a venting system and potentially delivers a uniform and reproducible ablation, has been under investigation.