Tumors of the stomach are diverse in presentation, symptoms, and prognosis. In this chapter, the authors will first describe the epidemiology, presentation, and management of gastric adenocarcinoma. Subsequently, gastric polyps, mesenchymal tumors (eg, gastrointestinal [GI] stromal tumors), and the rare gastric sarcoma and lymphoma will be discussed.

Historic Background

The first description of stomach cancer documented in Western literature is generally thought to be that of Avicenna (980-1037). Many years later, in 1761, Morgani published a manuscript on malignancies of the stomach. In 1879, Pean was believed to perform the first gastric resection for cancer, followed by Billroth performing the first described pyloric resection in 1881, and Schlatter successfully performing the first total gastrectomy (TG) in 1897. In 1951, McNeer et al recommended a more extensive resection for cancer, including TG with distal pancreatectomy and splenectomy.¹

Incidence and Pathology

While gastric cancer (GC) is the third leading cause of cancer-related death worldwide, significant differences in its incidence exist across the continents.² Specifically, a higher incidence is found in Japan and Eastern Asia (approximately 18-25 cases/100,000) than in Europe and North America (approximately 8-10 cases/100,000).³ The incidence of GC in the United States is low as it is currently the 15th most prevalent cancer. In 2015, 24,500 patients were diagnosed with GC, and nearly 10,000 persons are projected to die from GC in 2016. The estimated overall 5-year survival approaches 30%.⁴

Gastric cancer is a malignant solid organ tumor of older adults (>65 years). The median age of diagnosis is 69 years of age. Similar to other solid organ cancers, older adults are primarily affected.⁵ In recent years, the incidence of GC has been rising in younger adults (age <50 years). Initially, their outcomes were mistakenly perceived to be worse than older adults. However, a recent large population-based study showed that younger patients were more likely to present with advanced or metastatic disease; however, they have a more favorable stage for stage prognosis than their older counterparts.⁶

In addition to age, race and ethnicity also impact the presentation, treatment, and prognosis of GC. In the United States, Caucasians typically present with proximal GC, often involving the gastroesophageal junction (GEJ), whereas Asians tend to present with early stage disease, distal tumors, and have a more favorable prognosis. In contrast, African Americans and Hispanics are more likely to present with advanced stage disease and harbor worse outcomes, likely due to a combination of issues related to access to care and multiple morbidities.^7,8

Gastric Cancer Risk Factors

The development of GC has been attributed to several risk factors. The most significant appears to be an infection with Helicobacter pylori. This is particularly an issue in developing countries and is more often observed in GC outside the cardia, as supported by a robust meta-analysis of 42 observational studies.⁹ Diets that contain salt, smoked or poorly preserved foods, nitrates, nitrites, and secondary amines have been shown to contribute to development. In contrast, diets that are rich in raw vegetables, fresh fruits, vitamin C, vitamin A, calcium, and antioxidants have been found to be protective.¹⁰ Cigarette smoking is another major environmental risk factor with 2- to 3-fold increased risk for GC. Excessive alcohol consumption may also pose an increase.¹¹ Finally, GC is associated with a wide host of hereditary GC syndromes outlined below.

HEREDITARY GASTRIC CANCER SYNDROMES

Ten percent of GCs concentrate in families, and less than 3% are due to an inherited cancer syndromes.^12–14 In evaluating these patients, close attention should be paid to the family history along with the pathology to uncover those with hereditary GC. Features suggestive of hereditary risk include GC in two or more first-degree relatives and/or second-degree relatives, cancers in multiple generations, signet ring cell histology, and early age of onset (<45 years).¹⁵

Hereditary diffuse gastric cancer (HDGC) was the first hereditary GC to be identified. HDGC is caused by germline mutations in CDH1, the gene that encodes the E-cadherin protein.¹⁶ The average age of onset for HDGC is 38 years (range 14-69 years); premenopausal lobular breast cancer can occur as well^16,17 By the age of 80 years, 70% of men and 56% of women with HDGC developed diffuse gastric cancer (DGC); lobular breast cancer develops in 42% of women.¹⁸

To better manage these risks, the International Gastric Cancer Linkage Consortium has proposed criteria for HDGC. Germline CDH1 testing is recommended in three groups: (1) families with two or more patients with GC at any age and one confirmed DGC, (2) individuals with DGC before the age of 40, and (3) families with diagnoses of both DGC and lobular breast cancer (one diagnosis before the age of 50). CDH1 testing should also be considered in the following three groups: (1) patients with bilateral lobular breast cancer or a family history of two or more cases of lobular breast cancer occurring before the age of 50, (2) patients with a personal or family history of cleft lip/palate and DGC, and (3) tumors with in situ signet ring cells and/or the pagetoid spread of signet ring cells.

In addition to the early age of onset, one of the key challenges when families with HDGC is the relative ineffectiveness of endoscopic surveillance in detecting DGC alone. In this regard, TG for DGC patients and prophylactic gastrectomy (for at-risk family members) is appropriate for CDH1 carriers. These recommendations are less clear in managing families who meet established criteria but in whom no mutation is present.

Other different hereditary cancer syndromes with increased risk of GC include Lynch syndrome, Li-Fraumeni syndrome, familial adenomatous polyposis (FAP), Peutz-Jeghers syndrome (PJS), juvenile polyposis syndrome, MUTYH-associated polyposis, hereditary breast, and ovarian cancer syndrome.¹⁹ As such, it is imperative to refer at-risk patients and their families to cancer genetics professional to discuss options for management.

Prognostic Factors and Surgical Outcomes

OVERALL SURVIVAL OUTCOMES

In western populations, patients predominantly present with advanced stage disease and an overall 5-year survival rate of less than 30%.⁴ Nearly 65% of GC patients, collected in the US National Cancer Database, are found to have advanced disease (T3/T4). Indeed, up to 85% of these patients harbor nodal metastases at the time of diagnosis.¹¹ The median survival in persons who undergo curative gastrectomy is 24 months (5-year survival 20%-30%). However, when palliative or no GC therapy is performed, these median survival rates drop to 8 and 5.4 months, respectively.¹¹

In this current era of personalized cancer medicine, the prognosis after R0-gastric cancer resection has been examined across several externally validated and electronically available patient-centered nomograms. The nomograms go beyond elements of the current American Joint Committee on Cancer (AJCC) staging system to also account for the patients’ age, sex, tumor location, tumor size, negative/positive lymph node status, and pathologic features/classification of the tumor(s).^20,21

GASTRIC CANCER SURGERY OUTCOMES

Due to regionalizing complex surgical care, including major cancer surgery, mortality rates after gastrectomy (namely TG) have become relatively very low (<2%). A high hospital volume with at least 11 gastrectomies per year is predicted to have a 3% to 6% lower in-hospital mortality than lower volume centers.^22,23

While postgastrectomy mortality rates are low, the postoperative morbidity rate remains high, approaching 40%.²⁴ These complications include systemic (pulmonary embolism, pneumonia, myocardial infarction, deep vein thrombosis) and technically-related issues (anastomotic leak, anastomotic stricture).

Postgastrectomy readmission rates are estimated to range from 7% to 20%. These readmissions are largely driven by operative GI complications. It was noted that within this subset of patients, those with a higher preoperative nutritional risk and postoperative infections were found to be at the highest risk of complications requiring readmission.^25,26

PATHOLOGIC STAGING

The pathologic staging is a T-, N-, M-based AJCC staging system (Table 31-1).

CLINICAL MANIFESTATIONS

Gastric cancers can present with specific or nonspecific GI symptoms. Specific symptoms include early satiety, dysphagia, or weight loss, often prompting the treating caregiver to request an upper endoscopy for diagnosis and further workup.¹⁰ However, when patients present with nonspecific symptoms that are easily mistaken for benign conditions (eg, gastroesophageal reflux, peptic ulcer disease), the diagnostic process can be considerably delayed. Indeed, the relatively low incidence rates and lack of cost effective screening program for GC in the United States only add to the difficulty of diagnosing these patients early. Patients with advanced disease can present with a palpable abdominal mass, cachexia, ascites, or bowel obstruction.

The cardinal physical examination findings of metastatic disease, such as an enlarged supraclavicular node (Virchow node) or a drop metastasis in the pouch of Douglas (Blumer shelf), are typically rare.¹¹ Overall, most physical examination findings are unfortunately found to be nonspecific and unreliable in making a definitive diagnosis.

Pretherapy Workup

HISTORY AND PHYSICAL EXAMINATION

The workup should start with a thorough history and physical examination. Specific questions in the history should include whether the patient has experienced unintentional weight loss, anorexia, early satiety, vomiting, bleeding, epigastric burning, pain, or discomfort. The surgeon should also ask about social factors, such as tobacco and alcohol use, and the consumption of large amounts of nitrate-rich or smoked/preserved foods. A previous history of H. pylori infection or a family history of GC are also important considerations.¹⁰ Finally, assessing the patients’ performance status and frailty are also critical to predict their ability to tolerance GC therapies.

LABORATORY WORKUP

In addition to a thorough history and physical examination, the pretherapy evaluation in patients with GC should include:

• 
  

  Complete blood count (CBC): To evaluate and treat GC- or treatment-related anemia

  
  
• 
  

  Basic metabolic panel (BMP): To detect electrolyte abnormalities, especially in gastric outlet obstruction, and to detect renal functional abnormalities prior to receiving contrast-enhanced imaging and/or preoperative systemic therapy

  
  
• 
  

  Liver function panel: To check prior to the induction of preoperative systemic therapy

  
  
• 
  

  Albumin and prealbumin: To uncover malnutrition especially since approximately 30% to 80% of patients diagnosed with GC are malnourished.²⁵

Genetic Testing for Hereditary Gastric Cancer

As previously described, the operating surgeon should be aware of potential features in the history or pathology report that are suggestive of hereditary GC. As part of the multidisciplinary approach to GC, a referral to a professional cancer genetic counselor is imperative to better manage this unique cohort of patients and their families with potential hereditary GC.

In family carriers of the CDH1 mutation, a shared decision is made around prophylactic TG. Furthermore, support from a social worker and a psychologist should be available to address and manage the burden of dealing with such major life-changing treatment decisions.

Diagnostic and Staging Modalities

SCREENING FOR GASTRIC CANCER (EAST VS WEST)

Unlike the United States and Europe, screening esophagogastroduodenoscopy (EGD) is typically performed in countries with a high prevalence of GC, such as Japan and South Korea. The overall goal is to improve overall outcomes. However, in regions outside of Asia with a lower prevalence, endoscopic screening has not been shown to be beneficial for primary prevention in the general population.²⁷

As previously discussed in other parts of this chapter, screening should be considered in certain cases of hereditary GC syndromes.²⁸ For example, in persons with adenomatous polyposis syndromes (eg, FAP), screening EGD is recommended starting at age 25 to 30 years and repeated every 0.5 to 4 years based on the Spigelman stage of duodenal polyposis.²⁹ In cases of HDGC, prophylactic TG is the recommended approach at a young age, obviating the need for screening EGD and potentially preventing the development of GC.

Endoscopy and Endoscopic Ultrasound for Gastric Cancer

Upper GI endoscopy plays a key role in establishing the diagnosis and in treatment planning of GC. However, upper endoscopy lacks standardized quality measures for adequate endoscopy similar to colonoscopy for colonic polyps or cancer (eg, caecal intubation, time for completions). As such, surgeons and their GI counterparts need to work collaboratively to identify key GC-relevant elements including: (1) the exact location of the cancer within the stomach to determine the extent of gastrectomy, (2) the relation of the cancer to the esophagus to ascertain the need for distal esophageal resection, and (3) the presence of linitis plastica as a marker of systemic disease pointing toward the need for preoperative systemic therapy.

Experienced endoscopists can leverage several other techniques to assist with identifying high-risk lesions using magnifying endoscopy, chromoendoscopy, narrow band imaging, flexible spectral color enhancement endoscopy (FICE), and confocal laser endomicroscopy (CLE). These techniques have shown promise in improving the endoscopic detection of worrisome lesions and in guiding biopsy site selection.³⁰

Endoscopy also represents an opportunity for tissue collection and molecular profiling (eg, Her2neu) for advanced and progressive GC for possible enrollment in emerging GC immune therapy trials (eg, tissue for PD1 status). When preoperative systemic therapy is employed, this diagnostic tool can also provide useful treatment and prognostic information, such as the status of the response to preoperative therapy or distal esophageal involvement.

Nearly 10% of gastric patients diagnosed before the age of 50 years also have synchronous colon cancer, and a preoperative screening colonoscopy should be performed on all these patients.³¹

Endoscopic ultrasound (EUS) has also emerged as an accurate staging tool in GC to compliment high-resolution cross-sectional imaging and staging laparoscopy.^32,33 By placing the EUS probe directly over the primary tumor, EUS can distinguish T1 (early GC) from T2 (invasion of muscularis propria) lesions with a sensitivity of 85% and a specificity of 90%. Additionally, the distinction between T1 and T2 (superficial) versus T3 and T4 (advanced) lesions can be achieved with a sensitivity of 86% and a specificity of 90%.³⁴

N-staging is also important. EUS is used to confirm the presence of nodal involvement with a sensitivity of 83% and a specificity of 67%. Nodal areas of particular interest include the paracardial, superior gastric, inferior gastric, and pancreaticolienal regions.³⁴ EUS is highly operator-dependent and provides accurate loco-regional staging of gastric adenocarcinoma that can inform decisions regarding preoperative systemic therapy versus surgery-first followed by adjuvant therapy.

Contrast-Enhanced Cross-Sectional Imaging

In addition to EUS, high-resolution multidetector computed tomography (CT) is the preferred imaging modality for the staging and post-treatment surveillance of gastric adenocarcinoma. Because of its thin slices and multiplanar capabilities, this modality provides accurate visualization of the primary tumor and surveys the chest, abdomen, and pelvis for metastatic disease.

At our institution (MedStar Georgetown University Hospital), we developed a gastric tumor protocol CT for patients with suspected or known gastric masse; the stomach is distended with oral contrast material or water prior to performing the CT. Distention of the stomach helps differentiate a collapsed gastric wall from tumor. It has been shown that the combination of focal or eccentric wall thickening greater than 1 cm and intravenous contrast enhancement is highly specific.³⁵ This gastric tumor protocol CT is extremely useful when considering the laparoscopic resection of other gastric neoplasms, such as gastric GI stromal tumors or neuroendocrine tumors. Further details are found in other parts of this chapter. We also recommend taking advantage of the coronal and sagittal reconstructions, which display helpful gastric anatomy and, in some cases, the optimal planes for visualizing tumors with adjacent organ involvement.

Overall, the accuracy of CT for staging is estimated to range from 66% to 93%.³⁶ The limitations of CT are its inability to detect subtle serosal invasion, metastatic disease in normal-sized lymph nodes, and small peritoneal deposits that may be below the resolution of CT. Hence, our strong preference is to also add staging laparoscopy to overcome these known limitations.

Positron emission tomography (PET) with 2-deoxy-w-(¹⁸F) fluoro-D-glucose (FDG PET) combined with CT (FDG PET CT) is valuable in some patients for the detection of occult disease. However, it lacks accuracy in mucinous or diffuse disease, such as linitis plastica.

DIAGNOSTIC LAPAROSCOPY AND PERITONEAL CYTOLOGY

Diagnostic laparoscopy (DL) is highly recommended as an additional staging tool. DL was found to upstage 20% to 25% of patients and thus can prevent nontherapeutic laparotomies in patients with subradiographic or occult hepatic or peritoneal metastasis.³⁷ The risk of peritoneal disease is much higher in those with linitis plastica and AJCC stage T3+, N+ disease; DL can inform the treatment strategy of a patient with suspected stage IV disease. DL can also be utilized to re-evaluate disease response to systemic therapy, to obtain peritoneal cytology, or to place a preoperative feeding tube.^38,39

The use of peritoneal cytology continues to be an area of controversy. Peritoneal cytology that is positive for tumor cells has been shown to be a poor prognostic marker in the absence of visible tumor spread (C1 disease).^40,41 The estimated median survival in patients with C1 disease was only 20 months.^39,40 At our institution, we employ a selective approach for peritoneal cytology in the following situations: (1) linitis plastica, (2) borderline performance status, or (3) evidence of AJCC T4 disease on imaging.

In summary, our multidisciplinary staging approach is to employ the following three tools: (1) EUS performed by an experienced gastroenterologist, (2) high-resolution CT of the chest, abdomen, and pelvis, and (3) staging laparoscopy with or without cytology.

Multidisciplinary Treatment Strategy for Operable Gastric Cancer

A multidisciplinary treatment strategy is crucial in patients newly diagnosed with GC. We recommend a stage-dependent treatment approach informed by the three patient-centered aspects:

1. 
   

   The patient’s suitability to undergo curative gastrectomy

   
   
2. 
   

   Accurate three-tool pretherapy staging

   
   
3. 
   

   Sequence of GC therapy (surgery-first vs perioperative therapy)

SUITABILITY TO UNDERGO CURATIVE GASTRECTOMY

The operating surgeon should consider several important preoperative variables prior to undertaking surgical therapy. Evaluating the underlying comorbidities and performance status are key preoperative considerations. Managing the burden of existing comorbidities is also important to enhance operative recovery after gastrectomy. As previously mentioned, reversing preoperative malnutrition and electrolyte abnormalities will lead to better operative outcomes. In some regards, improving some factors, such as frailty or performance status, may not be an achievable goal. However, a preoperative rehabilitation stay prior to surgical treatment may reduce the impact of frailty on operative outcomes.^42,43

ACCURATE THREE-TOOL PRETHERAPY STAGING

Up to 30% of newly diagnosed GC patients harbor occult radiographic metastases. As such, we recommend accurate stage-dependent treatment using a combination of EUS, high-resolution contrast-enhanced cross-sectional imaging, and DL. Together, patients can avoid the pitfalls of nontherapeutic laparotomy for GC.

SEQUENCE OF GASTRIC CANCER THERAPY (SURGERY-FIRST VERSUS PERIOPERATIVE THERAPY)

Margin-negative (R0) gastrectomy and adequate lymphadenectomy together represent the pillars of surgical therapy for operable GC. Level I evidence continues to strongly support a multimodal approach in persons with GC to enhance their overall and disease-free survival. In this regard, most patients with AJCC T2+ or N+ operable GC are offered one of the following two treatment sequences: (1) surgery-first, followed by adjuvant chemotherapy and/or chemoradiotherapy, or (2) perioperative systemic therapy. The latter is gaining more traction in Europe in light of emerging evidence that shifts the operable GC treatment paradigm to that of a perioperative approach. Additional details will be provided in parts of the current chapter.

At our institution, we take patient- and tumor-related factors into consideration to guide the sequence of therapy. Factors such as the presence of linitis plastica and the tumor location are considered during the treatment strategy planning. For example, a proximal GC location, the presence of linitis plastica, or borderline performance status all favor perioperative therapy. In those with proximal GC, prolonged operative recovery and higher operative complication rates are to be expected after TG; hence the rationale for perioperative therapy. In contrast, factors including an early stage GC or distal GCs (and their subsequent tumor-related complications, including bleeding or obstruction) favor a surgery-first approach.

Surgical Approaches for Operable Gastric Cancer

TOTAL GASTRECTOMY

Total gastrectomy is typically performed in patients with proximal GCs (cancers of the cardia or fundus). This operation entails the extirpation of the entire stomach, GEJ, and omentum, with subsequent restoration of intestinal continuity using a Roux-en-Y reconstruction (Figs. 31-1 A and B). The stomach is carefully dissected and mobilized free of all attachments. The arterial supply of the stomach is then ligated at its origin, followed by the removal of the stomach. TG is considered complete when normal esophageal and duodenal mucosa is included in the margins.^11,44

SUBTOTAL GASTRECTOMY

A subtotal gastrectomy (STG) is recommended for patients with midbody or distal GCs (Fig. 31-2). Unlike TG, STG entails the removal of 70% to 80% of the stomach. Adequate negative resection margins (ie, 4-6 cm proximal and 2 cm distal margins) are necessary to ensure an appropriate oncologic resection. In line with TG, it is imperative to ligate the gastric arteries at their origins, with the exception of the short gastric vessels; these should be maintained to prevent remnant ischemia. Gastrointestinal continuity is restored either via a Roux-en-Y gastrojejunal reconstruction (our preference) or a loop gastrojejunostomy. The rationale behind the use of a Roux-en-Y gastrojejunal reconstruction instead of a loop gastrojejunostomy is to avoid bile reflux into the gastric remnant.¹¹ Hand-sewn or a stapled anastomosis is considered safe and appropriate.

Equivalent overall and disease-free survival have been noted after TG versus STG for distal GC (overall 5-year survival rate of 62.4% vs 65.3% for TG vs STG, respectively).²² When compared to TG, STG has been shown to provide more favorable nutritional outcomes and quality-of-life.⁴⁵

Extent of Lymphadenectomy for Gastric Cancer

While adequate lymphadenectomy with histopathological nodal evaluation are important components of GC staging and therapy, the extent of lymphadenectomy has been an area of significant debate and controversy.⁴⁶

The classification of lymphadenectomy for operable GC falls into two categories: (1) the topographic location of the lymph node stations and (2) the extent of nodal removal, extending away from the stomach. The Japanese Research Society for Gastric Cancer has described the topographic classification of histopathological and nodal evaluations. This classification is based on nodal stations within various parts of the stomach and its arterial supply, and extends to the para-aortic nodal region.^47,48 The second classification is based on the extent of nodal removal and is also known using the “D” nomenclature. As such, four tiers of lymphadenectomy exist: (1) D0 denotes incomplete removal and therefore is considered an inadequate nodal dissection, unless palliative gastric resection is considered; (2) D1 entails the removal of the perigastric lymph nodes; (3) D2 is D1 combined with the removal of the nodal stations around the celiac trunk, along with a distal pancreatectomy and splenectomy; and (4) D3 includes D2 + resection of the nodes from the celiac axis to the inferior mesenteric artery.⁴⁸

The evaluation of lymphadenectomy for GC has progressed in Europe. Initially, two large European trials from the United Kingdom and Netherlands demonstrated no survival differences between D1 versus D2 lymphadenectomy (Fig. 31-3). In both trials, enrollees suffered worse operative outcomes after D2 lymphadenectomy.^49–51 However, the long-term results from the Dutch Gastric Cancer Group trial demonstrated a more favorable survival benefit for D2 nodal dissection. Specifically, the 15-year overall survival (OS) rates were 21% and 29%, respectively, for the D1 and D2 groups (P = 0.34). Lower rates of local (12% vs 22%) and regional recurrence (13% vs19%) were also associated with D2 lymph node dissection.^50,52,53 Contemporary European studies are currently evaluating survival benefits with D2 compared to D1 lymphadenectomy in the setting of improved D2 operative outcomes.^54–56

To better answer this question of survival benefit of D2 lymphadenectomy in the Asian population where GC is more prevalent, JCOG9501 was a Japanese randomized controlled trial conducted to compare D2 dissection alone versus D2 with para-aortic nodal dissection (PAND) for operable T2b–T4 GC (T2b, T3, or T4). D2 nodal dissection with PAND did not improve the overall and relapse-free survival rates compared to D2 dissection alone (5-year OS rates were 70.3% and 69.2%, respectively).^57,58 Recent meta-analyses of D1 versus D2 trials have demonstrated that D2 dissection is associated with a significantly higher postoperative risk, but with equivalent long-term survival rates between D1 versus D2 lymphadenectomy.^59,60

A Cochrane review meta-analysis of over 2500 patients enrolled in eight Asian and European lymphadenectomy (D1, D2, or D3) GC trials showed no difference in survival between D2 and D3 even in Asian lymphadenectomy trials. Furthermore, no significant differences were found in the overall and disease-free survival in trials of D1 versus D2 lymphadenectomy. However, D2 lymphadenectomy was associated with a significantly improved disease-specific survival rate compared to D1, albeit with two higher operative mortality rates.^61,62

The differences in Asian versus western results are perhaps attributable to differences in disease biology, surgical expertise, variations in where GC surgery is performed (especially in the United States), and differences in BMI in Eastern versus Western GC patients.^51,63

In light of this mixed level of evidence, most current western guidelines recommend at least a D1 lymphadenectomy with a total nodal yield of 15 or more lymph nodes. A modified D2 (also known as pancreas and spleen-preserving D2 lymphadenectomy) remains an approach in expert centers.

Minimally Invasive Gastrectomy for Gastric Cancer

In addition to open GC surgery, minimally invasive gastrectomy (MIG) has emerged as an investigational surgical therapy. Nonrandomized and observation studies, using propensity score case-matching, have shown that MIG is associated with reductions in surgical site pain, the length of hospital stay, the use of narcotics, and postgastrectomy complication rates.^64,65 In a small prospective randomized investigation of laparoscopic versus open STG, operative mortality rates (3.3% vs 6.7%, respectively), 5-year OS (58.9% vs 55.7%, respectively), and disease-free survival rates (57.3% vs 54.8%, respectively) were more favorable for MIG, although the difference was not statistically significant.^66,67

While emerging investigations point toward improved outcomes with MIG in operable GC, several studies have included subjects with smaller tumor sizes or early-stage GC. This observation should be factored into comparisons of MIG with open GC surgery in western populations that typically have larger tumor sizes and more advanced disease.

There are two ongoing large prospective randomized MIG versus open gastrectomy trials in Asia. The first is a multihospital phase III Japanese study to assess the OS of laparoscope-assisted distal gastrectomy compared to open distal gastrectomy in patients with early-stage GC.⁶⁸ Klass 01 is another ongoing large Korean prospective randomized trial of laparoscopic versus open gastrectomy for distal GC.⁶⁹

To date, MIG for GC in the United States remains in its infancy and will require additional larger randomized clinical trials for more adoption among the surgical community in the United States.

SYSTEMIC CHEMOTHERAPY

Multimodality therapy for GC is designed to provide patients and their family with better care by prolonging the survival outcomes, reducing the risk of recurrence along with minimizing the burden of the disease. In this regard, the benefit of integrating surgical therapy (gastrectomy and lymphadenectomy) with systemic therapy has been demonstrated to reduce the recurrence risk of GC following surgical resection. In one meta-analysis, the use of any form of chemotherapy as adjuvant therapy to surgical resection for GC produced an 18% overall reduction in the risk of cancer recurrence.⁷⁰ Currently postoperative chemotherapy and radiation therapy is being compared to the perioperative use of chemotherapy alone.

Adjuvant Chemotherapy and Radiation Therapy

In 2001, the US Intergroup trial (INT-0116) established the combination of chemotherapy with radiation therapy as one possible adjuvant care standard for operable GC. In this randomized, phase III, open-label trial, patients were eligible for enrollment if their tumor was stage Ib to stage IVM0; patients were assigned to the treatment arm (5-fluorouracil and leucovorin, concurrent with radiation) versus surgery alone.⁷¹ In the adjuvant chemotherapy and radiotherapy arm, OS was statistically better (36 months), in comparison to the control arm (27 months, P = 0.005).

In the Intergroup trial, only 64% of patients assigned to the treatment arm were able to complete therapy, underscoring the toxicity of adjuvant therapy especially in the context of the morbidity and operative recovery associated with the necessary surgical therapy. In addition, 77% of enrolled patients had a distal tumor location within the stomach. While specific histologic subtypes were not reported, a distal location is typically associated with the less aggressive intestinal histology of gastric adenocarcinoma.⁷² Thus, the tumor location and biology could explain some of the benefit of this adjuvant approach. Finally, this trial was also notable for the lack of adequate nodal evaluation in as many as 50% of trial participants, thus raising questions about whether adjuvant therapy compensated for inadequate GC surgery or unrecognized node-positive disease.

Perioperative Chemotherapy

In 2006, the Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial in the United Kingdom established an alternative approach to adjuvant therapy for resectable GC through the use of perioperative chemotherapy alone. This phase III, open-label study randomized patients deemed to have resectable GC to either six cycles of chemotherapy (three each, presurgery and postsurgery) or to surgery alone.⁷³ Patients with nonmetastatic gastric adenocarcinoma of stage II or higher were eligible. Unlike the Intergroup trial, chemotherapy in the MAGIC trial consisted of a triple regimen of epirubicin, cisplatin, and 5-fluorouracil. Modifications to this regimen, using oxaliplatin instead of cisplatin, and capecitabine instead of 5-fluorouracil, have been shown by the Randomized ECF for Advanced and Locally Advanced Esophagogastric Cancer 2 (REAL-2) study to be acceptable, affording similar outcomes but with reduced toxicity.⁷⁴

In the MAGIC trial, the use of perioperative chemotherapy was associated with a significant survival benefit versus surgery alone (5-year survival rates of 36% and 23%, respectively; P = 0.009). As in the case with the Intergroup trial, several characteristics of the MAGIC trial need to be highlighted. First, while 90% of patients assigned to the treatment arm were able to complete the preoperative cycles of chemotherapy, only 57% began the postoperative chemotherapy cycles and only 43% completed them. While again highlighting the challenge of adjuvant therapy postoperatively, the relatively consistent ability of patients to tolerate neoadjuvant therapy is also noted. Second, the majority of tumors were located proximally, including 15% that were at the GEJ and 11% that were in the distal esophagus. While the histologic subtypes were not reported in the MAGIC trial either, the more aggressive diffuse subtype tends to predominate in this proximal tumor location.⁷² Together, these data suggest that the perioperative approach to adjuvant therapy could be optimal for a proximal tumor location and histology.