Endoscopic ultrasonography in esophageal and gastric cancer

Background

The global burden of esophageal cancer is high ranked as the fifth and eighth most common cause of cancer-related deaths worldwide in men and women, respectively. In 2021, an estimated 604,127 new cases were diagnosed, and the disease resulted in 544,076 deaths, accounting for 5.5% of all cancer-related deaths. At the time of presentation, almost half of patients have locally advanced or metastatic disease. The 5-year survival is poor, measuring 19.9% overall in the United States. The incidence and mortality of esophageal cancer has wide variations on a global basis, with the highest incidence-to-mortality ratio in Northern and Southern Africa among males, while in Western Europe and Southern America among females. Higher socioeconomic development is associated with a higher proportion of adenocarcinoma versus squamous cell carcinoma (SCC). Potential targets for disease prevention include alcohol use, smoking, poor oral hygiene, caustic injury, diet low in fruit and vegetables, and obesity.

Globally, gastric cancer is the fifth most common cancer and the fourth most common cause of cancer-related death. In 2021, an estimated 1,089,103 new cases were diagnosed, and the disease resulted in 768,793 deaths, accounting for 7.7% of all cancer-related deaths. There has been a significant improvement in survival rates in the United States ; however, the 5-year relative survival rate remains low at 32.4%. Potential targets for disease prevention include Helicobacter pylori infection, diet, smoking and alcohol use, and early neoplasia detection through population screening in high-incidence countries.

AJCC staging of esophageal and gastric cancer

Esophageal and gastric cancer staging is defined by the American Joint Committee on Cancer (AJCC) Staging System that establishes tumor-node-metastasis (TNM) subclassifications based on the depth of invasion of the primary tumor (T), lymph node involvement (N), and extent of metastatic disease (M) ( Tables 7.1 and 7.2 ). Tumors involving the esophagogastric junction (EGJ) with the tumor epicenter no more than 2 cm into the proximal stomach are staged as esophageal cancer. EGJ tumors with the epicenter located more than 2 cm into the proximal stomach are staged as gastric cancers, as are all cardiac cancers not involving the EGJ, even if they are within 2 cm of the EGJ. The TNM components for staging esophageal adenocarcinoma and SCC are the same; however, the AJCC anatomic stage groups differ depending on histologic type because of differing mortality rates between tumor type. The 8th edition subdivides N3 disease according to the number of involved nodes, reclassifies T4aN2 and T4bN0 tumors as stage IIIA disease, and has separate prognostic stage groups for clinical and pathologic staging, including pathologic staging after neoadjuvant therapy.

Management is dependent on tumor stage, and accurate preoperative determination of disease stage is therefore essential to select the appropriate treatment, including endoscopic resection, surgery, chemotherapy, radiation, or palliative care.

Workup of newly diagnosed esophageal and gastric cancer

An upper endoscopy is first performed to assess likely tumoral stage, location, length, configuration, degree of luminal stenosis, presence of synchronous dysplastic lesions, and to remove retained food or liquid to reduce the risk of aspiration. This information also informs on the endoscopic or surgical approach, is used in planning radiation fields, and can indicate the likelihood for needing adjunctive procedures to maintain enteral nutrition, such as enteric feeding tubes, luminal stenting, or surgical bypass.

Following the diagnostic gastroscopy and biopsy, computed tomography (CT) of the chest, abdomen, and pelvis with oral and intravenous contrast is recommended. ^, The prime strength of CT scanning in preoperative staging is the detection of distant metastases that identify patients who are not surgical candidates. These patients require histologic confirmation of metastatic disease, after which further investigations to evaluate the T and N status are generally not required. If CT is negative or equivocal for metastatic disease, positron emission tomography (PET) with integrated CT (FDG-PET/CT) is performed if available, as it detects previously unsuspected metastatic disease in 15% to 20% of patients ^, and assesses malignant involvement of lymphadenopathy detected on CT. Once distant metastases have been ruled out, NCCN guidelines ^, recommend endoscopic ultrasonography (EUS) staging to establish the extent of locoregional disease prior to initiation of treatment. The risk of performing an EUS also needs to be considered, particularly in patients with esophageal cancer.

EUS has an overall accuracy of 71% to 92% for assessing the T stage for esophageal cancer. ^, Lymph node staging is usually performed with the combination of EUS, CT, and FDG-PET/CT. The strengths of EUS are in evaluating local lymph node involvement and its ability to sample suspicious nodes for tissue confirmation; however, it may miss lymph nodes beyond the depth of ultrasound penetration or those located past a stricturing tumor or involving distal lymph node groups. With advancements in FDG-PET/CT staging, the role of EUS in determining treatment plan in esophageal cancer may be decreasing.

In gastric cancer, EUS is most accurate in distinguishing between T1-2 disease versus T3-T4 disease (pooled accuracy of 88%), and the pooled accuracy of EUS in determining nodal stage (positive vs negative) is 75%. A meta-analysis on EUS in gastric cancer showed that EUS was most accurate for more advanced tumors—T3 cancers (85%), followed by T4 (79%), T1 (77%), and T2 (65%) cancers. The pooled accuracy, sensitivity, and specificity of EUS for N staging (N0 vs N+) were 64%, 74%, and 80%, respectively. The results are similar in esophageal cancer, with a meta-analysis showing EUS had greater accuracy for advanced compared to early stage disease (pooled sensitivity of 92% for cT4 vs 82% for cT1 tumors).

For superficial esophageal and gastric cancers, ER is the best method for determining depth of invasion; and CT, FDG-PET, and EUS have a limited role in the initial workup. ER of superficial lesions provides a larger specimen for histology assessment of the degree of differentiation, presence of lymphovascular invasion and the depth of infiltration, and is potentially curative. EUS may assist in staging early esophageal and gastric cancer when the endoscopist cannot exclude locally advanced disease based on the endoscopic appearance of nodular or ulcerated lesions, and in patients with T1b (sm1) disease on endoscopic resection, to evaluate for lymph node involvement. However, there is a risk of overstaging and selecting incorrect treatment based on EUS evaluation in early disease, therefore EUS is not routinely performed in this group. ^,

Diagnostic laparoscopy with peritoneal cytology is usually performed to detect radiologically occult metastatic disease in patients with T3 or node-positive gastric cancer being considered for surgical resection without neoadjuvant therapy, and for patients planned for neoadjuvant therapy. ^, There is some evidence for peritoneal staging in patients with early-stage gastric cancer too. In a study of 56 patients with clinical stage T1-2 N0 disease, 17.9% had peritoneal disease on diagnostic laparoscopy and peritoneal cytology. Cost-effectiveness analysis suggests that a selective approach for higher risk patients should be considered.

The role of EUS-FNA or FNB in staging

Tissue confirmation of nodal involvement or metastatic disease is important for selecting appropriate treatment. The presence of nodal involvement in T1-T2 disease determines whether patients receive neoadjuvant chemoradiotherapy or proceed directly to surgery, and confirmation of metastatic disease by EUS-FNA obviates the need for staging laparoscopy and establishes nonoperative disease. ^,

Fine-needle aspiration or biopsy (FNA or FNB, respectively) can be performed for cytological confirmation if it will impact the treatment plan and can be performed without passing the needle through the primary tumor. For both esophageal and gastric cancer, the addition of FNA/FNB to EUS improves identification of malignant lymph nodes compared to EUS morphology assessment alone. In a prospective blinded comparison study in 125 patients with esophageal cancer, EUS, EUS-FNA, and helical CT results were compared to the pathological evaluation of resected lymph nodes. EUS-FNA was more accurate than CT (87% vs 51%; P < 0.001) or EUS (87% vs 74%; P = 0.012) for nodal staging. Furthermore, EUS-FNA significantly modified tumor stage determined by helical CT in 38% of patients, usually toward a higher stage. EUS-guided FNA may also aid with diagnosis in patients with ascites, peritoneal deposits, or other metastatic disease identified on staging EUS.

EUS examination technique and staging

T stage

The esophageal and gastric wall has four main layers—the mucosa, submucosa, muscularis propria, and adventitia or serosa. T1a cancers are confined within the mucosa, T1b tumors involve the submucosa, T2 tumors invade the muscularis propria, T3 the adventitia/serosa, and T4 denotes invasion of adjacent structures.

T staging can be assisted by techniques to improve acoustic coupling and identification of the wall layers. These include using a balloon on the tip of the echoendoscope or positioning the scope over the lesion in the dependent position and then instilling a low volume of water and removing air. Aspiration risk is reduced by using small water volumes, raising the head of the bed, and regularly suctioning the esophagus and stomach. The scope is maintained perpendicular to the tumor without contact to avoid compressing the lesion and tangential imaging.

Five distinct layers of the esophageal and gastric wall are seen on EUS, visible as alternating hyperechoic and hypoechoic layers. The first two layers correspond to the mucosa (hyperechoic and hypoechoic), the third hyperechoic layer corresponds to the submucosa, the fourth hypoechoic layer to the muscularis propria, and the fifth hyperechoic layer to the adventitia or serosa. Tumors appear as a hypoechoic expansion, and the degree of infiltration of the tumor through the esophageal wall layers determines the tumor stage. Expansion of layers 1 to 3 corresponds with infiltration of the superficial and deep mucosa and submucosal layers (T1 disease; Fig. 7.1 ), and layers 1 to 4 correlates with penetration into the muscularis propria (T2 disease, Fig. 7.2 , Video 7.1 ). Expansion beyond the smooth outer border of the muscularis propria correlates with invasion of the adventitia (esophagus) or serosa (stomach), corresponding to T3 disease ( Fig. 7.3 , Video 7.2 ). Loss of a bright tissue plane between the tumor and surrounding structures shows extension beyond the esophagus and stomach, correlating with T4 disease ( Fig. 7.4 , Video 7.3 ).

N stage

The regional lymph node stations are surveyed. For esophageal cancer, these include a mediastinal examination and evaluation of the left gastric, common hepatic, splenic, and celiac nodes. The regional lymph node groups in gastric cancer are the left and right paracardial, perigastric along the lesser and greater curvatures, suprapyloric and infrapyloric, celiac, left gastric, common hepatic, hepatoduodenal, splenic artery, and splenic hilum. Involvement of lymph node groups beyond these are classified as distant metastasis.

Several nodal characteristics are associated with malignant involvement: size >5 to 10 mm, round shape, sharp boarders, and hypoechoic pattern ( Fig. 7.5 ). The likelihood of a lymph node being malignant increases as more EUS criteria are met with the presence of all four features having a sensitivity of 89% and specificity of 92% for lymph node metastasis. Modified lymph node criteria on EUS were proposed by Vazques-Sequeiros et al. On multivariable logistic regression analysis, the presence of lymph nodes ≥5 mm in size, the lymph nodes having a rounded shape, five or more lymph nodes identified, and tumor stage T3-T4 best predicted malignant nodal involvement.