| Stratification* | Grade I recommendations | Grade II recommendations | Grade III recommendations |
|---|---|---|---|
| Non‐EGJ gastric cancer: cT3‐4aN+M0, stage cIII | Neoadjuvant chemotherapy: SOX regimen (Evidence 1) |
|
|
| Gastric cancer invading the EGJ[Link], [Link]: cT3‐4aN+M0, stage cIII | Neoadjuvant chemoradiotherapy: DT 45‐50.4gy (concurrent fluorouracil, platinum or taxanes) (Evidence 1B) |
|
Neoadjuvant radiotherapy (patients unsuitable for chemotherapy) (Evidence 2B) |
| cT4bNanyM0, stage cIVA (without non‐resectable factors) | MDT discussion for optimal personalized treatment |
|
Encourage participation in clinical trials |
| R1/R2 resection after neoadjuvant therapy | MDT discussion for optimal personalized treatment | Encourage participation in clinical trials | |
| Localized disease progression after neoadjuvant therapy | MDT discussion for optimal personalized treatment | Encourage participation in clinical trials |
*According to the 8th AJCC/UICC clinical staging system (cTNM) for gastric cancer
Abbreviations: EGJ, esophagogastric junction; MSI‐H, high microsatellite instability status; cTNM, clinical tumor‐node‐metastasis classification; SOX, S‐1 + oxaliplatin; DOS, docetaxel + oxaliplatin + S‐1; 5‐FU, 5‐fluorouracil; FLOT, 5‐FU+ leucovorin + oxaliplatin + docetaxel; XELOX, capecitabine + oxaliplatin; MDT, multidisciplinary team;
Notes
dAdjuvant treatment for resectable gastric cancer
There are several large phase III clinical trials supporting the use of postoperative adjuvant chemotherapy for patients who had undergone D2 radical gastrectomy [70, 71, 72, 73]. The indications of postoperative adjuvant chemotherapy for resectable gastric cancer are D2 radical gastrectomy and no prior neoadjuvant therapy for stage pII and pIII patients. For stage II patients, the recommended regimen is S‐1 (oral; till 1 year after operation) or capecitabine combined with oxaliplatin or cisplatin [70, 71]. In the JACCRO GC‐07 study [72, 74], the investigators found that S‐1/docetaxel (oral S‐1 on days 1‐14 with 7 days of rest followed by 6 cycles of S‐1 combined with docetaxel on day 1 of each cycle, then 4 further cycles of S‐1 on days 1‐28 every 42 days) was associated with improved survival of patients with stage III advanced gastric cancer compared to S‐1 monotherapy (3‐year recurrence‐free survival [RFS] rate, S‐1/docetaxel arm, 65.9%, vs S‐1 arm, 49.6%; P = 0.0007) and suppressed all types of recurrences, including hematogenous, lymphatic, and peritoneal recurrences. The RESOLVE trial [75] showed that for locally advanced cT4a/N+M0 or cT4b/NxM0 gastric cancer, adjuvant S‐1 plus oxaliplatin (SOX) was not inferior to adjuvant capecitabine plus oxaliplatin (XELOX) (3‐year disease‐free survival [DFS] rate: 60.3% vs. 54.8%, P = 0.162). The ARTIST‐II trial [76] enrolled 900 stage II‐III gastric cancer patients with positive lymph nodes who underwent D2 radical gastrectomy and investigated the curative effects of 1‐year S‐1 monotherapy versus 6 months of SOX versus SOX plus radiotherapy (SOXRT). The results showed that, compared to S‐1 monotherapy, adjuvant SOX or SOXRT could significantly prolong DFS, but compared to adjuvant SOX regimen, adjuvant SOXRT had no additional survival benefit. In recent years, there have been studies investigating the applicability of survival prediction models, such as nomograms, based on tumor and patient characteristics to evaluate the survival benefits of individualized adjuvant chemotherapy for stage II/III gastric cancer. Wang et al. [77] reviewed the data of 1464 pT3‐4 or N+ gastric cancer patients who received adjuvant fluoropyrimidine plus oxaliplatin (F‐OX) after D2 gastrectomy from three major centers across China. The results showed that, compared to the 7th AJCC gastric cancer classification, the nomogram was superior in stratifying patients for predicting benefit from F‐OX. Using the nomogram, patients in the low‐risk group had no improvement in survival with F‐OX, while for those classified in the intermediate‐ and high‐risk groups, F‐OX could reduce the risk of death by over 20%; thereby, the nomogram could more accurately guide the selection of gastric cancer patients who would benefit from F‐OX adjuvant chemotherapy.
At present, it is not clear whether patients with stage pI gastric cancer would benefit from adjuvant chemotherapy. It is suggested for stage pI patients with high‐risk factors, such as younger age (<40 years old), high or low histological grade differentiation, and nervous plexus, vascular or lymphatic invasion, investigational treatment can be offered.
For resectable gastric cancer, the results of phase III clinical studies investigating the efficacy of chemoradiotherapy after radical surgery were different in the East and the West. The INT0116 study [78], from the U.S, confirmed that concurrent radiotherapy and 5‐fluorouracil (5‐FU) chemotherapy after surgery improved OS compared to surgery alone, but the surgery performed were mainly D0/D1 gastrectomy, whilst in countries like China, Korea, and Japan, mostly D2 gastrectomy are performed. The ARTIST study from South Korea [71], which compared 6 cycles of adjuvant capecitabine plus cisplatin (XP) versus 2 cycles of XP followed by concurrent capecitabine combined with RT (XP/XRT/XP) plus 2 additional cycles of XP in gastric cancer patients after D2 R0 gastrectomy. They found no significant reduction in recurrence between the two therapies in the overall population (3‐year DFS rates, XP/XRT/XP arm: 78.2% vs XP arm: 74.2%; P = 0.0862), but in subgroup analysis of patients with positive pathologic lymph nodes, patients from the XP/XRT/XP arm had superior DFS than the XP arm (3‐year DFS rate: 77.5% vs 72.3%; P = 0.0365). Subsequently, the ARTIST‐II study [76] was performed to compare the efficacy of different chemotherapy regimens and chemoradiotherapy (adjuvant S‐1 monotherapy versus SOX versus SOXRT) in patients with D2‐resected, stage II/III, node‐positive gastric cancer. The results were published by the American Society of Clinical Oncology (ASCO) in 2019 and no difference in DFS between SOX and SOXRT (P = 0.667) were observed but both adjuvant SOX and SOXRT were effective in prolonging DFS when compared to S‐1 monotherapy; thereby demonstrating that the addition of radiotherapy could not provide addition survival benefits [76]. Thus, for resectable patients who can undergo R0 and D2 resection, adjuvant chemoradiotherapy is not recommended unless they are diagnosed with advanced pathological stage and associated with high‐risk factors including insufficient dissection distance from tumor margin (<2 cm), vascular tumor thrombus, perineural invasion, N3 or metastatic lymph node ratio >25%, then, after systemic therapy adjuvant radiotherapy could be considered. For those who did not achieve R0 resection (without distant metastasis), postoperative chemoradiotherapy [79] or MDT discussion is recommended.
At present, adjuvant chemotherapy for gastric cancer invading the EGJ is mostly based on the findings of studies from Asia. Among four large‐scale phase III clinical studies, the rate of EGJ‐gastric cancer was 23.4% in the JACCOR GC‐07 study [72, 74], 4.8% in the ARTIST study [71], 2.3% in the CLASSIC study [70], and 1.4% in the ACTS‐GC study [80]. However, there is still a lack of randomized controlled trials investigating the significance of adjuvant chemotherapy or chemoradiotherapy for EGJ carcinoma.
ePreoperative and perioperative chemotherapy for advanced gastric cancer
Perioperative therapy (neoadjuvant chemoradiotherapy + surgery + adjuvant chemotherapy/radiotherapy) for gastric cancer has been proven to be superior to surgery alone in Western countries as they could downstage the tumor, increase the rate of radical resection, and improve survival whilst not increasing the risks of postoperative complications and deaths [81, 82]. Also, neoadjuvant chemotherapy prior to radical gastrectomy in Asian studies has also been associated with significantly improved tumor remission rates, R0 resection rates, and treatment safety [83, 84]. The survival benefits of perioperative chemo‐/radiotherapy, as compared with postoperative chemotherapy after radical D2 gastrectomy, remain to be determined in large phase III clinical trials. The RESOLVE study [75], a large cohort randomized controlled phase III clinical study led by Chinese investigators aiming at comparing the efficacy and safety of adjuvant XELOX (arm A) or adjuvant SOX (arm B) against perioperative SOX (neoadjuvant SOX followed 5 cycles of adjuvant SOX; arm C) in locally advanced gastric cancer patients after D2 radical gastrectomy. They found that perioperative SOX was superior to adjuvant XELOX (3‐year DFS rate: 62.0% vs. 54.8%; P = 0.045) while adjuvant SOX was non‐inferior to adjuvant XELOX (3‐year DFS rate: 60.3% vs. 54.8%; P = 0.162). Therefore, 3 cycles of neoadjuvant SOX chemotherapy and 5 cycles of adjuvant SOX followed with 3 cycles of S‐1 monotherapy is recommended as the perioperative treatment for locally advanced gastric cancer. In addition, during the same period, the PRODIGY study [85] reported that for locally advanced gastric cancer staged as cT2/3N+M0 or cT4/NxM0, 3 cycles of neoadjuvant docetaxel plus oxaliplatin plus S‐1 (DOS) chemotherapy plus 8 cycles of postoperative S‐1 monotherapy, compared to surgery followed by 8 cycles of S‐1 monotherapy, was associated with tumor downstaging and significant improvement in 3‐year DFS. Thus, the DOS regimen can also be recommended as neoadjuvant chemotherapy for locally advanced gastric cancer patients.
Currently, the recommended neoadjuvant chemotherapy regimens for gastric cancer include XELOX [86], FOLFOX [87], cisplatin combined with S‐1 (SP) [88], and SOX [89]. Results of the large prospective phase III FLOT4‐AIO study [90] showed that compared with epirubicin plus cisplatin (ECF)/epirubicin plus cisplatin and capecitabine (ECX) regimen, the docetaxel combined with oxaliplatin, leucovorin and 5‐FU (FLOT) regimen was associated with improved 3‐year OS and DFS and had higher pathological response rate and R0 resection rate. Therefore, the FLOT regimen can also be used as the recommended regimen for preoperative chemotherapy of gastric cancer. In regard to HER2‐positive gastric cancer, there have been several investigations on the perioperative treatment of anti‐HER2 drugs, including double‐ or triple‐drug chemotherapy, combination with a monoclonal antibody (mAb) or double anti‐HER2 treatment. However, as the final results of these trials are to be published, currently no standard treatment strategy has been proposed.
Results of the international multicenter CRITICS study [91] showed that for stage IB‐IVA resectable gastric cancer or EGJ cancer patients who received neoadjuvant epirubicin, cisplatin or oxaliplatin, and capecitabine (ECC/EOC) chemotherapy followed by curative intent gastrectomy with adequate lymph node dissection (D1+ accounted for 86% of the study population) and same chemotherapy regimen as adjuvant treatment administrated alone or in combination with adjuvant radiotherapy. The investigators found no improvement in survival in those who underwent adjuvant chemoradiotherapy compared to patients with resectable gastric cancer treated with neoadjuvant chemotherapy, adequate surgery, and adjuvant chemotherapy. However, the completion rate of the study was only 50%, and 60% of the investigated cohort were stage I‐II patients. As such, the local control rate of radiotherapy could not be fully determined and decreased its clinical referential value.
For gastric cancer patients with T4b disease and without unresectable factors, based on current research evidence [92, 93, 94], the following points could be considered as treatment options: ① R0 resection is an independent prognostic factor for survival; ② the rate of complications after combined organ resection is very high, close to 40%, among which pancreatoduodenectomy is the highest risk procedure; ③ surgery for peripheral organ involvement is very complex, and it is difficult to formulate a standard treatment principle. Therefore, it is suggested that such cases should undergo MDT discussion for an individualized treatment plan. Further, neoadjuvant therapy could improve the R0 resection rate and can be used as a treatment option. For patients who can achieve R0 resection, combined organ resection is acceptable, but combined pancreatoduodenectomy should be carefully assessed for risk and benefits.
A multinational individual‐patient‐data meta‐analysis [95] which explored the associations of MSI status with postoperative prognosis and perioperative chemotherapy efficacy in patients with resectable gastric cancer enrolled in the CLASSIC [70], ARTIST [71], MAGIC [81], and ITACA‐S trials [96] showed that for resectable dMMR/MSI‐H gastric cancer patients, the prognosis of patients who underwent only surgery was better than those who underwent surgery plus adjuvant chemotherapy. Currently, small sample‐sized retrospective studies have shown that the prognosis of dMMR and MSI‐H patients was good but had conflicting results regarding the benefits of adjuvant chemotherapy. Overall, considering the low number of corresponding patients in these studies, conflicting evidence in current literature, adverse reactions related to chemotherapy and patients’ financial implications, it is suggested that for dMMR/MSI‐H patients, (neo) adjuvant treatments such as immunotherapy in clinical trial settings could be first considered, unless unwillingness from the patient's side, after detailed discussion with the patient and families about the risk and benefits of different treatment strategies, post‐operative observation or chemotherapy can be considered.
For EGJ adenocarcinoma, clinical studies [97, 98] have shown that neoadjuvant chemoradiotherapy plus surgery plus adjuvant chemotherapy was effective in tumor downstaging, could increase R0 resection rate and improve survival, and was not associated with increased risk of complications or postoperative deaths. The long‐term follow‐up results of the POET study [97], which investigated the addition of radiotherapy to preoperative chemotherapy in lower esophageal and gastric cardia adenocarcinoma, demonstrated that, although the trial was terminated early as it did not meet its accrual goals and could not provide statistical significance, preoperative chemoradiotherapy appeared to reduce the risk of local recurrence and tended to prolong survival, compared to preoperative chemotherapy, without an increase in adverse events and surgery‐related complications in patients with localized EGJ adenocarcinoma. Also, the RTOG‐9904 multicenter phase II clinical trial [98] demonstrated satisfactory results for locally advanced gastric cancer patients undergoing preoperative chemoradiotherapy. Therefore, the current recommendation for stage III EGJ adenocarcinoma is neoadjuvant chemoradiotherapy followed by radical D2 gastrectomy. For locally advanced gastric cancer, encouragement of patient participation in clinical trials for preoperative chemoradiotherapy is recommended for confirmation evidence. Recommended regimens for concurrent chemotherapy include paclitaxel combined with 5‐FU, paclitaxel combined with platinum, or 5‐FU combined with platinum. At present, the TOPGEAR (NCT01924819) [99], CRITICS‐II (NCT02931890) [100] and the phase III prospective Neo‐CRAG clinical trial (NCT01815853) [101] launched by the Sun Yat‐sen University Cancer Center are actively investigating the effects of preoperative chemoradiotherapy in this category of patients.
Studies investigating neoadjuvant chemotherapy and perioperative chemotherapy in patients with EGJ cancer are gradually increasing, including studies from Asia such as the RESOLVE [75], PRODIGY [85], and RESONANCE studies [102]. In the PRODIGY study [85], EGJ cancer accounted for 5.6% of the study population, and the results showed that neoadjuvant chemotherapy with DOS regimen could achieve tumor downstaging, improve R0 resection rate, and prolong PFS. In the RESOLVE study [75], EGJ cancer accounted for 36.5% of the study population, and the results showed that compared with the adjuvant XELOX chemotherapy, perioperative SOX chemotherapy was associated with improved R0 resection rate and prolonged DFS. A propensity score‐matching study from Zhongshan Hospital of Fudan University [103], comprising of 32% of patients with EGJ cancer, showed that neoadjuvant DOS regimen was more effective than neoadjuvant XELOX regimen in improving both PFS and OS, without any added adverse effects. According to these findings, the DOS and SOX regimens can be considered for neoadjuvant chemotherapy of EGJ cancer patients.
The efficacy of neoadjuvant therapy should be timely evaluated using EUS, CT, or PET/CT imaging modalities. Compared with CT and other non‐invasive imaging examinations, laparoscopic laparotomy can improve the diagnostic rates of occult metastasis within the abdominal cavity, including radiologically undetected small liver metastases. It can be carried out alongside a cytological examination of intraperitoneal washings [104]. As such, prior to neoadjuvant therapy (for T3‐4 or N+ cases), explorative laparoscopic staging and cytological examination of intraperitoneal washings are recommended.
For surgically resected specimens diagnosed as pathological complete response (pCR) after neoadjuvant therapy, it is recommended that the same neoadjuvant regimen to be continued postoperatively. Till present, there is no sufficient evidence attributing to the survival differences between those who undergo different adjuvant regimens as to their initial neoadjuvant regimens or abstain from adjuvant therapies.
For patients who underwent neoadjuvant therapy and achieved R0 resection, if the preoperative imaging or pathological assessments showed improvement in shrinking the cancerous lesion, it is recommended that the same neoadjuvant regimen to be continued postoperatively.
In case of disease progression following neoadjuvant therapy, surgery should be considered if R0 resection can be achieved. If not, the treatment protocol should be discussed via an MDT panel.
For patients who could not achieve R0 gastrectomy despite the absence of distant metastasis after neoadjuvant chemotherapy, either postoperative chemoradiotherapy or MDT discussion is recommended. If neoadjuvant chemoradiotherapy was performed, the subsequent treatment should be discussed via an MDT panel, else palliative treatment is recommended.
fPerioperative treatment for EGJ cancer
The choice of perioperative treatment for EGJ cancer has some particularity because of the differences in clinical research design and results between Eastern and Western countries. The proportion of patients with EGJ cancer was very low in a number of clinical trials which had positive results for postoperative adjuvant chemotherapy after D2 gastrectomy in Asian countries. Although the overall population can benefit from postoperative adjuvant chemotherapy, it is uncertain whether patients with EGJ cancer in Asian countries would benefit from such treatment. Comparatively, in European clinical trials which investigated perioperative treatment for gastric cancer, the proportion of patients with EGJ cancer was higher, i.e., in the FFCD (n = 60%) [105] and FLOT4‐AIO study (n = 56%) [90], suggesting that perioperative chemotherapy was indeed an effective treatment for patients with EGJ cancer in Western countries. In the RESOLVE study [75], EGJ cancer patients comprised of 36.5% of the study population, suggesting that perioperative chemotherapy could also be an effective treatment in the Asian population.
In addition, due to inconsistent findings on neoadjuvant therapy studies, it is difficult to identify the optimal neoadjuvant therapy (neoadjuvant chemoradiotherapy versus neoadjuvant chemotherapy) for EGJ cancer. The POET study [97] only demonstrated potential advantages of preoperative chemoradiotherapy for EGJ cancer. However, a meta‐analysis by Petrelli et al. [106] showed that neoadjuvant chemoradiotherapy, compared with neoadjuvant chemotherapy, was associated with increased pCR rate and reduced local recurrence in EGJ adenocarcinoma, but did not prolong OS, which was different from the conclusion of the POET study. As such, based on the current research evidence, for EGJ cancer, perioperative chemoradiotherapy could be more suitable than adjuvant chemotherapy.