Purpose (diagnosis/evaluation)	Grade I recommendations	Grade II recommendations	Grade III recommendations
Definitive diagnosis	Gastroscopy + biopsy (Evidence 1A)	Cytological examination a (Evidence 2A)
Location evaluation	Gastroscopy (Evidence 1A) Abdominal enhanced CT (Evidence 1A)	Abdominal MRI (Evidence 2A)	X‐ray barium double contrast radiography (Evidence 2B)
Staging evaluation	Abdominal and pelvic enhanced CT b (Evidence 1B) Chest CT c (Evidence 1B) EUS d (Evidence 1A)	Abdominal MRI e (Evidence 2A) PET/CT (Evidence 2A) Diagnostic laparoscopy and examination of intraperitoneal washings f (Evidence 1B)
Treatment efficacy evaluation	Abdominal and pelvic enhanced CT g (Evidence 1A)	Gastroscopy (Evidence 2A) PET/CT (Evidence 1B) Abdominal MRI (Evidence 2A)	Functional imaging examination h (Evidence 3)

Abbreviations: CT, computed tomography; MRI, Magnetic resonance imaging; PET, positron emission tomography.

Notes

^a If it is not possible to obtain a pathological diagnosis of gastric cancer despite repeated gastroscopic biopsies, cytological examination of ascites/pleural effusion or pathological examination of metastatic lesions can be used as the basis for qualitative diagnosis.

^b Ensure that the gastric cavity is fully dilated through the gas or water under effective hypotonic procedures before examination [1, 2]. Multiphase contrast‐enhanced scans combined with multi‐planar reconstruction are recommended for diagnosis [3]. Plain abdominal CT scans are not recommended. If patients have contraindications to the contrast agent for enhanced CT, MRI or EUS is recommended. The staging accuracy of the radiologist might be potentially increased with the assistance of radiomics [4, 5].

^c Chest CT can detect lung metastasis more effectively than X‐ray plain film [3]. For EGJ carcinoma, enhanced CT scan of the chest is recommended to assess the extent of esophageal infiltration and status of mediastinal lymph nodes.

^d EUS should be performed in qualified centers. In the 8th edition of the AJCC/UICC staging system for gastric cancer and EGJ cancer, EUS is recommended as the preferred modality for the clinical evaluation of tumor invasion depth (cT) [2]. EUS cT staging not only enables direct observation of the lesions but can also provide visual descriptions regarding the different anatomical layers of the gastric wall and non‐homogeneous hypoechoic regions of the cancers, which could suggest the destruction of corresponding layers of the gastric wall. Simultaneously, EUS can detect enlarged perigastric lymph nodes, EGJ area and nearby metastatic lesions in the liver and peritoneal cavity. EUS is helpful for the diagnosis and clinical staging of gastric cancer and assessment of response to neoadjuvant therapy. A systematic meta‐analysis reported that the overall sensitivity and specificity of EUS for distinguishing T1 to T2 (superficial) versus T3 to T4 (advanced) gastric cancer was 0.86 and 0.90, respectively [6]. Further, the diagnostic ability of EUS to distinguish T1 (early gastric cancer) versus T2 (muscle‐infiltrating) tumor was 0.85 and 0.90, and T1a (mucosal) versus T1b (submucosal) cancer was 0.87 and 0.75, respectively [6].

^e Liver contrast MRI is recommended for further confirmation of CT undetermined liver metastasis. Hepatocyte‐specific contrast agent can be used based on clinical conditions [7].

^f Diagnostic laparoscopic exploration and examination of intraperitoneal washings are recommended for detecting occult metastasis when peritoneal metastasis is suspected [2]. For intraperitoneal lavage, 200 mL of normal saline can be infused into the different quadrants of the abdominal cavity and collect ≥50 mL of the lavage fluid for cytological examinations.

^g According to the response evaluation criteria in solid tumors (Response Evaluation Criteria in Solid Tumors [RECIST]) criteria (version 1.1) [8], metastasis nodules of the liver, lung, or peritoneum with a long diameter ≥1 cm or lymph nodes with a short diameter ≥1.5 cm could be enrolled as target lesions for treatment evaluation. The thickness of primary lesions in the stomach can be used as a reference for therapeutic assessment but could not be considered as a target lesion. In regard to immunotherapy, treatment efficacy can be evaluated by referring to the immune RECIST (iRECIST) criteria [9].

^h Small sample‐sized studies have shown that volume measurement [10] and functional imaging parameters such as the apparent diffusion coefficient value of diffusion‐weighted MRI (DW‐MRI) [11] and iodine concentration of spectral CT examinations [12] can assist in the evaluation of treatment efficacy of gastric cancer and can be used as a reference for evaluating treatment of atypical cases. Further, deep learning technology for CT has also shown potential in assisting the response evaluation of gastric cancer efficacy [13].