Keywords

Introduction

In 2017, the Canadian Cancer Society estimated gastric adenocarcinoma to be the 14th most commonly diagnosed malignancy, with 3500 new cases and 2100 deaths. The age-standardized incidence and mortality rate for gastric cancer have decreased from 19.0/100,000 cases and 15.5/100,000 deaths in 1980 to 8.6/100,000 and 5.1/100,000 deaths, respectively, in 2017 [1]. Enormous geographic variation in the incidence of gastric cancer exists with the highest incidence being observed in East Asia. Similarly, wide geographic variation in treatment outcomes is observed with overall 5-year survival rates of 40–60% reported in Asia and Europe, compared to 25–29% in Canada and the USA [1,2,3].

Risk Factors

Established risk factors for gastric cancer (GC) include Helicobacter pylori infection, smoking, alcohol, and dietary factors (such as processed meats and salt-preserved foods). Hereditary gastric cancers represent <5% of all gastric cancers. Main gastric cancer familial predispositions are hereditary diffuse gastric cancer (HDFC), gastric adenocarcinoma and proximal polyposis of stomach (GAPPS), familial intestinal gastric cancer (FIGC). Other hereditary cancer syndromes associated with increased risk of gastric cancer include Lynch syndrome, Peutz-Jeghers syndrome (PJS), familial adenomatous polyposis (FAP), Li-Fraumeni syndrome, and BRCA syndromes (BReast CAncer) (see Table 11.1 below).

Table 11.1 Selected familial predispositions to gastric cancer
Full size table

Special Notes

  • Hereditary Diffuse Gastric Cancer: criteria for diagnosis of HDGC and genetic testing for CDH1 mutation are as mentioned below [8]. Of note, in countries with low incidence of sporadic gastric cancer (such as Canada), approximately 10–18% are identified with CDH-1 mutation.

    1. 1.

      Two gastric cancer cases in first- or second-degree relatives regardless of age, at least one confirmed to be diffuse gastric cancer (DGC) or

    2. 2.

      One case of DGC diagnosed below the age of 40 years in a first- or second-degree relative or

    3. 3.

      Personal or family history of DGC and LBC, one diagnosed below the age of 50 years.

  • Gastric Adenocarcinoma and Proximal Polyposis of Stomach Syndrome: Proposed criteria for diagnosis are [11] as follows:

    1. 1.

      Gastric polyps restricted to the body and fungus with no evidence of colorectal or duodenal polyposis

    2. 2.

      More than 100 polyps in the proximal stomach of the affected patient (or more than 30 polyps in the first-degree relative)

    3. 3.

      Some FGPs having regions of dysplasia (or a family member with FGP and adenocarcinoma)

    4. 4.

      Autosomal dominant inheritance

  • Peutz-Jeghers Syndrome: Clinical diagnosis of PJS is made when any of the following criteria are present [18]:

    1. 1.

      Two or more histological confirmed Peutz-Jeghers (PJ) polyps

    2. 2.

      Any number of PJ polyps detected in someone with family history of PJS

    3. 3.

      Any number of PJ polyps in someone with characteristic mucocutaneous pigmentation

  • Lynch Syndrome [19,20,21]:

    1. 1.

      Patients are at risk of extra-colonic malignancies of endometrium, stomach, ovaries, hepatobiliary, renal pelvis/ureteric, brain and skin

    2. 2.

      Consensus guidelines for gastric cancer surveillance are variable. In general, a baseline upper GI scope at age of 30–35 years and subsequent scopes every 1–5 years are recommended, especially in patients with risk factors such as intestinal metaplasia, gastric atrophy, family history of gastric cancer, and immigration from countries with high incidence of gastric cancer. In addition, H. pylori testing and eradication are recommended.

Classification and Staging

Histopathology

Gastric adenocarcinomas are classified histologically according to the Lauren classification as (1) intestinal (well-differentiated) or (2) diffuse (undifferentiated) histologic subtypes [22]. Such classification can have clinical implications with respect to prognosis and management decision-making.

Intestinal-type adenocarcinoma of stomach is more sporadic and believed to be causally related to H. pylori and environmental risk factors of gastric cancer (GC) [23]. Whereas intestinal-type GC is believed to follow sequential progression of dysplasia to invasive carcinoma, the development of diffuse-type GC is not believed to follow defined preneoplastic stages [23]. Diffuse-type GC has defective intercellular adhesions and tend to spread within deeper layers of gastric wall and in a less coherent fashion [24, 25]. Fashion, which can lead to underestimation of its extent by visual assessment of gastric mucosa. As such, wider gross surgical resection margins may be needed in patients with diffuse-type GC, and if intra-operative frozen sections are being done, full gastric wall thickness assessment of the resection margin should be considered by the pathology team.

In addition, multiple retrospective studies suggest prognostic and predictive value in Lauren classification of gastric adenocarcinoma. Diffuse-type GC has been shown to be associated with worse disease-free survival (DFS) and overall survival (OS) rates [24,25,26]. Furthermore, in phase 2/3 of the prospective randomized FLOT4 study, comparing pathological response to two different perioperative chemotherapy regimens (ECF/ECX vs. FLOT), analysis of the pooled population of both groups showed that patients with intestinal-type GC had 16% complete pathological response vs. only 3% in patients with diffuse-type GC (p = 0.004) [27]. In this study, it was also shown that oxaliplatin-based chemotherapy (FLOT) resulted in more frequent partial tumor response in patients with intestinal-type GC compared to diffuse type (42% vs. 23%, P = 0.04), but tumor response between the two Lauren classification subtypes was similar in the non-oxaliplatin group (ECF/ECX).

Molecular Classification

Recently, as part of the Cancer Genome Atlas (TCGA), a molecular classification for gastric cancer has been developed, dividing gastric cancer into four subtypes: Epstein-Barr virus (EBV) positive, microsatellite unstable (MSI), genomically stable tumors (GS), and those with chromosomal instability (CIN) [28]. These molecular subtypes have been shown to have distinct salient genomic features which may provide guidance in using targeted agents in the future. In Fig. 11.1 below, salient features associated with each subtype, and their distribution in the stomach, are summarized.

Fig. 11.1
figure 1

Molecular subtypes of gastric cancer and their distribution within the stomach [28]. (Permission for use of this figure was obtained from Macmillan Publishers Limited)

Full size image

Molecular classification is emerging, as potential biomarkers to explore personalized treatment strategies in gastric cancer are in the experimental stages at this time. For instance, studies have shown that EBV-positive and MSI high, gastric cancers have higher PD-L1 expression, making them potential candidates for immunotherapy [28, 29]. In addition, there is data to support prognostic value in this molecular classification, suggesting EBV-positive tumors have the best prognosis and GS subtype is associated with the worst outcomes [29]. There is also preliminary evidence to suggest that MSI high status may be a negative prognostic marker in patients treated with perioperative chemotherapy [30]. Ongoing research is needed to better define the role of molecular classification in clinical practice.

Staging

Staging of gastric adenocarcinoma is according to the American Joint Committee on Cancer (AJCC), eighth edition. Gastroesophageal junction tumors with their epicenter located less than 2 cm into proximal stomach are classified, staged, and treated as esophageal cancers [31]. This most recent edition of AJCC has separated clinical from pathological staging and has incorporated post-neoadjuvant staging for gastric cancer (see Table 11.2 below).

Table 11.2 Gastric cancer patient outcomes according to the eighth edition of AJCC [32, 33]
Full size table

Staging Workup

The initial treatment plans are made based on the clinical stage of the patient. There are multiple tools that can be considered to improve the accuracy of the clinical stage and guide clinical decisions. CT scan, MRI, PET scan, endoscopic ultrasound (EUS), and staging laparoscopy ± peritoneal washings are some of these tools.

To evaluate the extent of locoregional disease, diagnostic tools have different accuracies, as summarized in Table 11.3.

Table 11.3 Diagnostic tool accuracy when used for assessment of gastric cancer [34, 35]
Full size table

According to a meta-analysis [35], EUS was most accurate for T3 disease (85%), followed by T4 and T1 (79% and 77%, respectively). Pooled accuracy of EUS in staging T2 lesions was only 65% in this meta-analysis. CT scan accuracy in assessment of T stage was suggested to be lowest in T1 lesions, being only 63% [34]. When comparing CT against MRI for assessment of T stage [34], MRI’s accuracy is statistically significantly higher overall (83% vs. 72%) and when identifying T1 lesions (86% vs. 63%). When assessing for N-Stage, meta-analysis results suggest that both CT and MRI are statistically significantly more sensitive than PET scan, but PET was shown to be more specific than both other techniques [34].

Early vs. Advanced Gastric Cancer

One clinically useful way of classifying gastric cancer is early vs. advanced. This classification can help guide the management strategy:

Early Gastric Cancer (EGC)

tumors confined to the mucosa (Tis or T1a) or submucosa (T1b), independent of the presence of lymph node involvement [36]. EGC is predominately identified by subtle changes in color, vascularity, or texture and is rarely diagnosed outside areas where population-based screening is offered, such as in Japan and Korea.

Advanced Gastric Cancer (AGC)

T2 to T4 (invading muscularis propria, subserosa, perforating serosa, or invading adjacent structures), without distant metastasis.

Management

In this section, we discuss the management of gastric cancer classified into early gastric cancer and advanced gastric cancer (see above for definitions of this classification). Below are definitions of some of the terminologies that are used in the chapter.

Endoscopic Mucosal Resection (EMR)

employs endoscopic techniques to elevate (e.g., injection and suction) and resect (e.g., cautery and banding) mucosal lesions en bloc.

Endoscopic Submucosal Dissection (ESD)

a variation of EMR that employs submucosal injection and a specialized needle-knife to permit en bloc resection of mucosal and submucosal lesions.

Subtotal Gastrectomy (SG)

removal of one-half to three-fourths of the gastric tissue, including omentum and all associated lymph nodes appropriate for a D1 or D2 lymphadenectomy. For distal gastric cancers, SG has been shown to have an equivalent oncological outcome and lesser morbidity when compared to total gastrectomy. SG is also associated with a better nutritional status and quality of life [37].

Total Gastrectomy (TG)

removal of all of the gastric tissue and distal esophagus, including omentum and all associated lymph nodes appropriate for a D1 or D2 lymphadenectomy. TG is preferred for tumors confined to the proximal one-third of the stomach.

Palliative Gastrectomy (PG)

gastrectomy performed with the intent to alleviate symptoms from the primary gastric cancer in the context of metastatic disease. A gastrectomy performed otherwise in a patient with metastatic disease is considered a non-curative gastrectomy [38].

D1 Lymph Node Dissection

includes removal of the omentum with perigastric lymph nodes (stations 1–6) and lymph nodes along the left gastric artery (station 7). It is important to note that station 1 (right paracardial) is part of a D1 LND, but station 2 (left paracardial) is not removed for SG [39]. See Fig. 11.2 for schematic of the lymph node stations.

Fig. 11.2
figure 2

Lymph node stations according to the Japanese gastric cancer treatment guidelines. (Figure adopted from 2014 Japanese gastric cancer treatment guidelines, Springer publications [40]). Numbers in blue color: D1 lymphadenectomy stations; numbers in orange color: D1+ lymphadenectomy stations; numbers in red color: D2 lymphadenectomy stations

Full size image

D2 Lymph Node Dissection

D1 nodes and lymph nodes along the common hepatic artery (station 8a), celiac axis (station 9), splenic artery (stations 10 and 11), and hepatic artery proper (station 12a) [39]. Clearance of station 10 and 11 nodes may require splenectomy (See Special Notes – Extent of Lymphadenectomy) (Fig. 11.2) [39].

Bursectomy

Removal of anterior leaflet of the transverse mesocolon and the pancreatic capsule, along with total omentectomy.

Early Gastric Cancer (EGC)

Workup

Surgery

Adjuvant therapy

Follow-up (f/u)

Recommended:

 History and physical exam

 Upper endoscopy

 Imaging:

  CT abdomen/pelvis

  EUS

  Staging laparoscopya

Optional:

 CT chest

 PET is not indicated for EGC

Gastrectomy with D1 lymph node dissectiona

OR

Endoscopic resection can be considered for lesions fulfilling all of the following [41]a:

 Intestinal type

 Confined to mucosa (Tis or T1a) and cN0

 Elevated lesions <20 mm or flat lesions <10 mm in diameter

 Absence of high-risk features (ulceration, poorly differentiated, lymphovascular invasion)

 Clear lateral and deep margins after excision

Indicated for all node-positive disease, and those who are found to be T2 or higher after resection (please see section on “Advanced Gastric Cancer”)

Recommended:

 Iron, B12, calcium supplements

 Every 3–6 months for 1–2 years, then every 6–12 months for 3–5 years, and yearly thereafter with:

  History and physical exam

  B12, Fe, bone

density if TG was performed

Optional:

 CT abdomen/pelvisa

 EGDa

  1. EGC early gastric cancer, EUS endoscopic ultrasound, EMR endoscopic mucosal resection, ESD endoscopic submucosal dissection, SG subtotal gastrectomy, TG total gastrectomy, RCT randomized controlled trials, EGD esophago-gastro duodenoscopy
  2. aSee Special Notes

Special Notes: Early Gastric Cancer

Endoscopic resection

EMR/ESD may be used in appropriately selected lesions amenable to en bloc resection that have minimal or no risk of nodal metastasis by experienced providers. Expanded criteria for ESD outside of the criteria listed above are considered investigational. ESD expertise and regional outcomes should be considered when choosing ESD as the treatment strategies, as recent meta-analysis has suggested worse endoscopic outcomes in Western countries compared to Eastern countries [42]. If after endoscopic resection it is revealed that the lesion is outside of the above-mentioned criteria (i.e. non-curative endoscopic resection), further treatment with gastrectomy and lymphadenectomy should be considered [43]. In the case of T upstage after endoscopic resection, management as per recommendations in the section “Advanced Gastric Cancer” should be considered.

Staging Laparoscopy (SL)

Limited use in EGC. In cases where the tumor is reliably felt to be clinically T1 and N0, then SL can be omitted.

Extent of Lymphadenectomy

Considerable controversy surrounds the role of extended lymphadenectomy (D1 vs. D2 resection) in early gastric cancer. Adequate staging requires 15 or more lymph nodes to be harvested. For cT1N0 tumors, D1 with splenopancreatic preservation is generally recommended. Worse outcomes have been associated with D2 lymphadenectomy in patients with EGC [44]. If clinically node positive, the staging should be reassessed to ensure not AGC.

Resection Margin (Early and Advanced Gastric Cancer)

Positive microscopic margins after gastrectomy are associated with inferior outcomes when compared to those in whom R0 status was achieved [45,46,47,48]. When subtotal gastrectomy is performed, in general a gross proximal margin of at least 4 cm is recommended to ensure R0 resection [48,49,50]; however, guidelines differ in their recommendation (see table below). Likely, smaller gross margins can be used in resection of EGC (T1), advocated by Japanese gastric cancer treatment guidelines, suggesting a 2 cm gross proximal margin in such cases [36]. Of note, diffuse or signet ring cell subtypes are at higher risk of positive margin, and in these cases a greater gross resection margin can be considered. Recommendations from three different guidelines for surgical resection margins in gastric cancer are outlined in Table 11.4 below.

Table 11.4 Recommended gastric cancer macroscopic proximal resection margin based on guidelines
Full size table

Intra-operative frozen sections of resection margin can be considered selectively, and in retrospective studies, they have been shown to be associated with low (1.7%) false-negative rates [52]. However, patients with signet ring cell or diffuse-type histology are at higher risk of false-negative intra-operative frozen section assessment [52].

To address a microscopically positive margin (R1 resection), consideration for re-resection or post-op CRT is recommended by clinical guidelines [50, 51] in selected cases.

  • Decision to re-resect in this scenario is complex and requires careful consideration of anatomical feasibility, patient factors, and disease factors. Microscopically positive margins in gastric cancer may not be an independent predictor of outcomes in patients with more advanced disease [45, 48, 53]; therefore, re-resection after a microscopically positive margin, when technically feasible, may only be considered in patients who have favorable stage of disease.

  • Demonstrated recurrence and survival benefits of post-operative CRT after R1 resection are based on retrospective studies only [54,55,56,57], and its potential risks/benefits should be carefully discussed in multidisciplinary cancer conferences on a case by case basis.

Laparoscopic Gastrectomy (LG)

LG is appropriate for EGC in experienced, high-volume centers, where results are monitored and assessed against international benchmarks [58]. It is safe and improved short-term outcomes have been demonstrated, but oncologic outcomes are currently being evaluated with ongoing RCTs [59].

Follow-Up Surveillance

Evidence to support the benefit of early detection of recurrence is lacking. Most providers perform surveillance with serial CT scans. Surveillance EGD should be offered to patients at risk of local recurrence (e.g., following endoscopic resection) when complete gastrectomy would be considered.

Advanced Gastric Cancer

Workup

Surgery

Perioperative/adjuvant therapy

Follow-up (f/u)

Recommended tests:

 History and physical exam

 Upper endoscopy

 Imaging:

  CT abdomen/pelvis

 Staging laparoscopya

Optional tests:

 CT chest

 EUSa

 PET is not indicated

Gastrectomy

 D2 LND

 SG or TG depending on location of tumora

 Consider intraoperative margin assessmenta

 Multi-visceral

resection should be performed if the patient is considered a candidate for curative resection

Options are:

 Perioperative FLOT chemo (preferred) [56, 60]

  ORa

 Adjuvant 5-FU-based CRT (if D1 LND or less) [61]

  ORa

 If no pre-op therapy, consider adjuvant chemo after D2 LND (If N+ may consider addition of CRT to the post-op regimen)

Each of the options above has been shown to be superior to resection alone in RCTs [62]. For guidance on choice of multimodality therapy, see Special Notes below

Every 3–6 months for 1–2 years, then every 6–12 months for 3–5 years, and yearly thereafter with:

 History and physical exam

 B12, Fe, bone density if TG was performed

Optional tests:

 CT abdomen/pelvisa

 EGDa

  1. EUS endoscopic ultrasound, SG subtotal gastrectomy, TG total gastrectomy, RCT randomized controlled trial, ECF epirubicin, cisplatin and fluorouracil (5-FU), FLOT docetaxel, oxaliplatin, fluorouracil, and leucovorin, CRT chemoradiotherapy, EGD esophago-gastro duodenoscopy
  2. aSee Special Notes

Special Notes: Advanced Gastric Cancer

Staging Laparoscopy (SL)

Radiologically occult peritoneal metastases are found in 20–30% of patients with T2 or higher disease [63]. SL is indicated in patients with clinical T2 or higher, or node positive on clinical staging to rule out radiologically occult peritoneal metastasis or positive peritoneal cytology [50]. Patients with positive peritoneal washings experience outcomes comparable to those with overt metastatic disease and should be considered palliative [64]. In patients who are being considered for preoperative therapy, SL with peritoneal washings should be obtained prior to preoperative therapy. Even though there are some data to suggest that patients who are converted from cytology positive to negative with systemic therapy have better outcomes [50, 65], role of surgery (gastric resection ± intra-peritoneal chemotherapy) is considered experimental and not the standard of care. Further studies are ongoing to better define the role of surgery in patients with peritoneal disease [66, 67].

Endoscopic Ultrasound (EUS)

EUS is valuable in the distinction between EGC and AGC and is critical if considering EMR/ESD. In patients with an established diagnosis of AGC, EUS is unlikely to change management and is not routinely required.

Resection Margin:

Please refer to the “Resection Margin” section under Early Gastric Cancer management.

Extent of Lymphadenectomy

Evidence suggests improved cancer-specific outcomes with D2 resection, particularly in higher staged tumors (T2–4) [44, 68]. Splenopancreatectomy is clearly associated with higher operative morbidity and is avoided unless required to achieve R0 resection margins [39, 69]. Involvement of nodes beyond a D2 resection (i.e., mesenteric, para-aortic, retroperitoneal) is classified as distant metastases [31]. The role of “D3” resections is not supported in the management of gastric cancer [70].

Bursectomy

Bursectomy was routinely considered/performed for serosa-positive gastric cancers according to Japanese guidelines, but has been studied further in JCOG 1001 randomized control trial in patients with resectable cT3/T4a GC [60]. The results of the study were published early, after second interim analysis in 2017, on the basis of futility. Based on these results, bursectomy did not provide survival advantage over non-bursectomy and was significantly associated with more pancreatic fistula [60]. A recent meta-analysis also was consistent with the results of JCOG 1001, but did not demonstrate superior overall or recurrence-free survival in patients with resectable cT3/T4 GC who received bursectomy [71].

Laparoscopic Gastrectomy (LG)

LG is not recommended for AGC due to limited available evidence on oncologic outcome [40, 50]. A Korean RCT is investigating oncologic outcomes of LG in AGC [72].

Combined Modality Treatment:

Strong evidence exists to support adding systemic therapy or chemoradiotherapy (CRT) to surgical resection as part of the treatment for patients with advanced GC improves outcomes [61, 73,74,75]. Below, we discuss peri-operative vs. postoperative therapy treatment strategies, as well as roles of chemotherapy and CRT in treatment of advanced GC. This is followed by Fig. 11.3, which summaries this discussion.

Fig. 11.3
figure 3

Peri-operative/postoperative therapy decision tree for advanced gastric cancer (AGC). Please see section above on “Combined Modality Treatment” for further details. ∗For treatment of AGC, we advocate for peri-operative chemotherapy approach over adjuvant therapy only, given low compliance rate with adjuvant therapy post-gastric surgery. ∗∗The benefit of post-op CRT in this scenario is only demonstrated in retrospective studies. Its risk/benefit or indication should be discussed on a case-by-case basis in multidisciplinary rounds. R1 resection microscopically positive margin, CRT chemoradiotherapy, LND lymphadenectomy, AGC advanced gastric cancer, FLOT docetaxel, oxaliplatin, fluorouracil, and leucovorin, CapeOx capecitabine and oxaliplatin, CapeCis capecitabine and cisplatin, FOLFOX FOLinic acid, fluorouracil, oxaliplatin

Full size image

  • Peri-operative Chemotherapy

    • Currently, in North America, peri-operative chemotherapy is the favored multi-modality approach for treatment of AGC. Peri-operative FLOT has been recently adopted as the standard of care in North America for management of patients with cT2 or greater and/or cN-positive patients [75].

    • Three preoperative and three postoperative cycles of ECF/ECX were compared against four pre- and 4 postoperative cycles of FLOT in a phase 3 randomized trial. The results showed that FLOT was associated with improved OS and PFS, with no increased complications rates [75]. More patients in the FLOT arm were able to complete all allocated treatment cycles compared to ECF/ECX. Peri-operative FLOT also resulted in improved R0 resection rates.

    • In the phase 2 of the same trial, four cycles of preoperative FLOT was associated with significantly higher rates of pathological complete regression (16%) compared to three cycles of preoperative ECF/ECX (6%) [27].

    • The role of replacing postoperative chemotherapy with CRT after preoperative chemo and adequate surgery (at least D1+) in patients with stage 1B-4a was investigated in the CRITICS trial [76]. There was no improvement in outcomes with incorporating CRT postoperatively in the treatment of these patients.

    • The CRITICS trial [76] once again highlighted the poor compliance with post-operative therapies after gastric resection (59% and 62% in the two groups) regardless of whether chemo or CRT was used postoperatively. In the FLOT study [75], only 52% of patients in the ECF/ECX arm and 60% of patients in the FLOT arm started the allocated postoperative chemotherapy. Low compliance has been seen in other gastric cancer adjuvant therapy trials as well, and should be a factor considered when deciding between peri-operative or adjuvant therapy approach in treatment of patients with GC. CRITICS-2 trial will be looking at the value of incorporating CRT in the neoadjuvant setting, in an attempt to find the most effective therapy that can be administered preoperatively, when patients have higher chance of tolerating the therapy [77].

  • Postoperative Chemoradiotherapy

    • The landmark INT-0116 trial showed long-term, improved, relapse-free survival and overall survival in patients with resectable stage 1B-4 disease who received postoperative CRT compared to those with surgery alone [61]. However, in this study, only 10% of patients had D2 lymph node dissections, and 54% did not even have complete D1 lymphadenectomy [19].

    • A phase 3 randomized trial in Korea investigated the role of postoperative CRT after curative resection of advanced gastric cancer with D2 lymphadenectomy, and did not demonstrate benefit with addition of adjuvant CRT in this group of patients compared to adjuvant chemotherapy alone [78, 79]. Unplanned subgroup analysis [79] suggested improved disease-free survival in node-positive patients who received concurrent adjuvant CRT compared to adjuvant chemotherapy alone. Benefits of adjuvant CRT in this subset of patients will be explored further in ARTIST-2 trial.

    • There may be a role for considering postoperative radiation in the case of microscopically positive resection margins. Please see the section “Resection Margin” above for more details.

    • To summarize the role of radiotherapy, patients with resected advanced GC who had curative resection but only D1 lymphadenectomy and no neoadjuvant chemo should be considered for adjuvant CRT [61]. In patients who receive neoadjuvant chemo followed by curative resection and at least D1 lymphadenectomy, no clear benefit has been demonstrated in post-op CRT compared to post-op chemo [76]. Lastly, in node-positive patients with completely resected gastric cancer and D2 lymphadenectomy who did not receive preoperative chemotherapy, there may be benefit in incorporating CRT in their adjuvant regimen [79].

  • Postoperative Chemotherapy:

    • Following curative resection (R0) and D2 lymphadenectomy of advanced GC, in patients who did not receive preoperative chemo, the results of phase 3 randomized trials as well as meta-analysis support use of adjuvant chemotherapy over surgery alone, when possible [74, 78, 80,81,82,83]. However, the role of adjuvant chemotherapy in this patient population who have received D1 lymphadenectomy (or less) is not well defined, and adjuvant CRT tends to be the treatment of choice [50, 61].

Unresectable or Metastatic Gastric Cancer

Workup

Management

Follow-up (F/U)

Recommended tests:

 History and physical exam

 Upper endoscopy

 HER-2 status

 Imaging:

  CT abdomen/pelvis

Optional tests:

 Staging laparoscopya

 CT chest

Consider chemotherapy, radiotherapy, and nonoperative management for symptomatic patients

Palliative gastrectomy should be avoided and only performed for symptomatic patients, for whom all nonsurgical and less morbid options have been considered

Stenting is associated with less morbidity than resection or bypass for palliation of obstruction and is typically preferred

Radiation or angioembolization can be effective for transfusion-dependent bleeding

As symptoms warrant

  1. aSee Special Notes

Special Notes: Unresectable or Metastatic Gastric Cancer

Staging Laparoscopy

may have utility in confirming metastatic disease, especially carcinomatosis, if suspected on imaging.

Criteria for Nonoperative Management

  • Unresectable

    • Level 3 or 4 suspicious nodes on imaging or confirmed by biopsy. Level 3 nodes include the posterior surface of the pancreas (nodal station 13), superior mesenteric artery, and vein (station 14). Level 4 nodes are middle colic vessels (station 15) and the para-aortic nodes (station 16).

    • Invasion or encasement of major vascular structures, such as celiac axis and its branches, is considered unresectable. Isolated left gastric artery involvement can be treated with curative intent if an R0 margin is obtainable.

  • Metastatic spread or peritoneal seeding (including positive peritoneal cytology) identified at surgical resection is considered incurable. Unless symptoms exist, systemic therapy should be considered rather than resection.

  • Non-curative gastrectomy has been demonstrated to impart no benefit in the setting of metastatic disease and exposes patients to unnecessary surgical procedures and risks of complications. In a phase 3 trial, survival of gastrectomy (followed by postoperative chemotherapy) in patients with advanced gastric cancer and one non-curative factor was compared against modern chemotherapy only, showing no survival benefit from gastrectomy and higher serious adverse events [84].

Landmark Surgical Publications (D1 vs. D2 Lymphadenectomy)

Study

Methods

Results

Dutch Trial

Bonenkamp et al. [69]

RCT

N = 711

D1 vs. D2 resection

D2 resection included distal pancreatectomy (30%) and splenectomy (38%)

Morbidity: 43% D2 vs. 25% D1 (p < 0.001)

Mortality: 10% D2 vs. 4.0% D1 (p = 0.004)

Median postoperative stays: D2 25 days vs. D1 18 days; p < 0.001

5-year update [39]:

 No difference in 5-year OS rates: 35% D1 vs. 33% D2

15-year update [68]:

 Overall 15-year survival: 22% D1 vs. 28% D2; p = 0.34

 Deaths from gastric cancer: 48% D1 vs. 37% D2; p = 0.01

Medical Research Council (MRC) ST01

Cuschieri et al. [85]

RCT

N = 400

D1 vs. D2 resection

D2 resection includes distal pancreatectomy and splenectomy (56%), or only splenectomy (66%)

Morbidity: 46% D2 vs. 28% D1; p < 0.001

Mortality: 13% D2 vs. 6.5% D1; p = 0.04

5-year update [86]:

 No difference in 5-year OS rates: 35% D1 vs. 33% D2

Italian Gastric Cancer Surgical Group (IGCSG)

Degiuli et al. [87]

RCT

N = 267

D1 vs. D2 resection

In the D2 arm, spleen and pancreas were preserved unless direct tumor extension. Splenectomy was performed for T1 or higher tumors on the greater curvature of the proximal or middle one-third of the stomach

No difference in 5-year OS: 66.5% D1 vs. 64.2% D2

Morbidity: 10.5% D1 vs. 16.3% D2; p < 0.29

In-hospital mortality: 0% D2 vs. 1.3% D1; not statistically significant

5-year update [44]:

 Trend toward improved 5-year OS for advanced disease (T2-4; N+): 59% D2 vs. 38% D1; p = 0.055

 5-year DSS for pT1 cancers were worse in the D2 arm compared to the D1 group (83% vs. 98%; p = 0.015)

  1. CRT chemoradiotherapy, OS overall survival, RCT randomized control trial

Landmark Chemotherapy and Chemoradiation Publications

Study

Methods

Results

FLOT Trial

Al-Batran et al. [75]

RCT

N = 716

Stage ≥cT2 and/or cN+, M0 resectable gastric and GEJ adenocarcinoma

3 preoperative and 3 postoperative 3-week cycles of ECF/ECX or 4 preoperative and 4 postoperative 2-week cycles of FLOT

Peri-op FLOT improved overall survival and progression-free survival compared to peri-op ECF/ECX

Median OS 50 months vs. 35 months (HR 0.77 [0.63–0.94]; p = 0.012)

PFS 30 vs. 18 months (HR 0.75 [0.62–0.91]; p = 0.004)

More grade 3 and 4 nausea/vomiting within ECF/ECX group compared to FLOT

CRITICS Trial

Cats et al. [76]

RCT

N = 788

Stage 1B-4. Induction. 3 cycles of pre-op ECX, then curative gastrectomy and at least D1 LND, then randomized to post-op chemo (3 cycles of ECX) or CRT (45 Gy + weekly and daily capecitabine)

Post-op only 59% of chemo group and 62% of CRT group started post-op therapy

Post-op CRT did not improve overall survival vs. post-op chemo

Median OS 43 months (95% CI 31–57) in chemo group and 37 months (30–48) in CRT group (HR 1.01 m, 95% CI 0.84–1.22; p = 0.90). Median follow-up 61.4 months

No mortality in post-op period. Grade 3 and 4 complications during post-op were 48% and 9% in chemo group vs. 41% and 4% in CRT group

INT-0116 Trial

MacDonald et al. [61]

RCT

N = 556

Surgery plus adjuvant CRT vs. surgery alone

Adjuvant treatment was 5-FU + leucovorin followed by 4500 cGy

All patients received curative-intent surgery:

 Only 10% received D2 resection

 54% received D0 resection

Improved overall and relapse-free survival with adjuvant CRT

Median OS: 36-month CRT vs. 27-month surgery alone; p = 0.005

Median RFS: 30-month CRT vs. 19-month surgery alone; p < 0.001

3-year OS: 50% CRT vs. 41% surgery alone; p = 0.005

MAGIC Trial

Cunningham et al. [73]

RCT

N = 503, T2 or higher

Surgery with perioperative ECF vs. surgery alone

ECF was administered for 3 cycles preoperatively and 3 cycles postoperatively

Improved PFS and OS with perioperative ECF

5-year OS: 36% ECF vs. 23% surgery alone; HR 0.75 (95% CI 0.60–0.93), p = 0.009

PFS: HR 0.66 (95% CI 0.53–0.81), p < 0.001

GASTRIC Study

Paoletti et al. [74]

Patient-level meta-analysis of 17 RCTs

N = 3838

Chemotherapy after complete resection vs. surgery alone

Improved OS and DFS with adjuvant chemotherapy in resectable gastric cancer

OS: HR = 0.82 (95% CI 0.76–0.90; P < 0.001)

DFS: HR = 0.82 (95% CI 0.75–0.90; P < 0.001)

CLASSIC Trial

Noh et al. [80]

 Multicenter RCT

 n = 1035 patients, stage II–IIIB

 Surgery plus adjuvant capecitabine and oxaliplatin vs. surgery alone

All patients underwent D2 resection

Improved DFS and OS with chemo

5-year DFS: 68% vs. 53%; HR 0.58 (95% CI 0.47–0.72)

5-year OS: 78% vs. 69%; HR 0.66 (95% CI 0.51–0.85)

ARTIST-I Trial

Park et al. [78, 79]

RCT

n = 458

All patients underwent D2 gastrectomy

Chemotherapy alone (6 cycles capecitabine + cisplatin) vs. CRT (4 cycles chemo; 45 Gy with concurrent capecitabine)

No difference in DFS and OS at 7years of median follow-up

5-year DFS: HR 0.74 (95% CI 0.52–1.05; p = 0.092)

5-year OS: 73% vs. 75%, HR 1.13 (95% CI 0.78–1.65; p = 0.53)

Subgroup analysis suggests benefit of CRT for node-positive disease and intestinal subtype (awaiting results of ARTIST-II trial)

  1. CRT chemoradiotherapy, OS overall survival, RFS relapse-free survival, PFS progression-free survival, DFS disease-free survival, HR hazard ratio, RCT randomized control trial, ECF epirubicin/cisplatin/5-fluorouracil, FLOT docetaxel, oxaliplatin, fluorouracil, and leucovorin

Landmark Palliative Publications

Study

Methods

Results

Chemotherapy vs. best supportive care in non-curable gastric cancer

Glimelius et al. [88]

RCT

N = 61, unresectable

Chemotherapy + best supportive care vs. best supportive care alone

Chemotherapy was ELF-regimen consisting of 5-fluorouracil, leucovorin, and etoposide

Improved or prolonged high-quality life at 4 months: 45% chemotherapy group vs. 20% best supportive care group; p < 0.05

TOGA Trial

Bang et al. [89]

RCT

N = 584, inoperable or metastatic, HER-2+ gastric cancer

Chemotherapy alone (capecitabine or 5-FU + cisplatin) vs. chemotherapy + trastuzumab

Improved median OS in HER2+ patients treated with trastuzumab: median OS 13.8-month trastuzumab vs. 11.1-month chemotherapy alone (p = 0.0046)

22% of patients assessed were HER2+

REGATTA Trial

Fujitani et al. 2016 [84]

RCT

N = 175 (planned N = 330)

Eligibility: gastric cancer (cT1-3), single non-curable site of disease confined to liver, peritoneum or para-aortic lymph node, PS 0-1

Gastrectomy (D1 without resection of metastases) followed by chemotherapy (S-1 plus cisplatin) vs. chemotherapy alone

Terminated early by DSMC based on futility: 2-year OS 25.1% for gastrectomy followed by chemotherapy vs. 31.7% for chemotherapy alone (p = 0.68)

  1. OS overall survival, RCT randomized control trial, 5-FU fluorouracil, PS performance status, DSMC data safety monitoring committee

Referral to Medical Oncology and Radiation Oncology

  • As the decision regarding adjuvant treatments should be made preoperatively, all patients should be referred to medical oncology and radiation oncology prior to resection and discussed at a multidisciplinary care conference.

  • Relative contraindications to chemotherapy [62]

    • Impaired cardiac function such as congestive heart failure, baseline left ventricular ejection fraction less than 50%, transmural myocardial infarction, valvular heart disease, high-risk arrhythmias

    • Impaired renal function (Cr clearance of <60 ml/min)

    • Disorders of the nervous system and diabetes are relative contraindications for chemotherapy with neuropathic agents (e.g., platinums)

  • Relative contraindications to radiation

    • Prohibitive toxicities anticipated due to volume or adjacent structures

    • Connective tissue disease

    • Previous irradiation to area

Referral to Multidisciplinary Cancer Conference

  • All cases of advanced gastric cancer should be discussed at a Multidisciplinary Cancer Conference (MCC), before surgical intervention to devise an individual plan for each patient.

  • Gastric cancer cases that were not discussed at MCC preoperatively should be discussed if the final pathology is >T1N0.