The PI has decided to close the study due to the outdated study design.

This is an open-label, stratified, two arm design. All patients receive same initial standard preoperative chemotherapy and surgical resection. Patients will then be assigned to either standard postoperative chemotherapy if node negative at surgery or standard postoperative chemoradiation if node positive at surgery.

The primary objective of this study is to determine the feasibility of patients enrolling and receiving either postoperative chemoradiation or chemotherapy alone, based upon nodal status at surgery, following preoperative chemotherapy.

The secondary Objectives is to evaluate the rate of cancer recurrence in patients assigned to treatment based upon node status. To explore the potential correlation between changes in expression of a pre-specified panel of genes identified as relevant to gastrointestinal cancers in response to preoperative chemotherapy, using presence of nodal involvement at time of surgery as an indicator of response.

Gastric adenocarcinoma, also known as stomach cancer, is a global health problem. While surgical resection remains the only curative option for patients who develop this disease, because of its aggressive nature and high recurrence rates, all but the lowest stages of gastric cancer require additional, adjuvant therapy in addition to surgery. Currently, two competing standards exist for the adjuvant management of gastric cancer: peri-operative chemotherapy (based upon the MAGIC study), and postoperative combination chemoradiation therapy (based upon the MacDonald Intergroup study). These two standards were arrived at based upon two separate trials that did not directly compare the two approaches, and thus either approach is considered acceptable. Several modifications to the chemotherapy used in the peri-operative approach have been studied and are now accepted as standard as well. Selection oftentimes derives from physician preference and tumor characteristics such as symptoms or nodal stage, and a clear guideline for when to select one adjuvant approach over the other is lacking.

Currently, analysis of patient outcomes from the large, randomized phase III CRITICS trial is ongoing, and will attempt to answer if one approach is clearly favored. Patients enrolled in the CRITICS trial all received preoperative chemotherapy and standardized surgical resection. Patients were then randomized to receive either chemotherapy or chemoradiation therapies postoperatively. Enrollment is closed for the CRITICS trial; preliminary data suggest that neither arm of postoperative treatment is superior. These early data support what practitioners have long suspected, which is that no one adjuvant standard is universally superior, but that tumors from different patients and with different characteristics will respond better to different adjuvant approaches.

Unfortunately, knowing which tumors will respond better to which therapy remains unpredictable. Certain clinical features can help clinicians to make a recommendation for their adjuvant approach. For example, patients who have tumors that are symptomatic, such as with bleeding or gastric obstruction, are less likely to tolerate several months of surgical delay while chemotherapy is administered preoperatively, and thus will typically be offered upfront surgery, followed by adjuvant chemoradiation. For most patients, however, the administration of systemic therapy prior to surgery as a part of a perioperative approach is recommended. This approach is favored because of the local control and reduction in tumor that can result prior to surgery, which is thought to lead to better outcomes in the long term.

However, not all patients will experience a response to preoperative chemotherapy. In those patients whose tumors are found at surgery to remain advanced, with nodal involvement, despite preoperative chemotherapy, long-term outcomes have been shown to be poorer than those whose tumors responded well, with the absence of nodal involvement upon dissection. These data are based upon the retrospective analysis of those patients who were enrolled in the original MAGIC study. Approximately 75% of tumors were found to be in the category of having nodes involved with cancer at resection despite preoperative chemotherapy. Thus, for this majority of patients with gastric cancer who have an inadequate response to preoperative chemotherapy (i.e., who remain node-positive at surgery), an alternative approach to their postoperative adjuvant treatment rather than mere completion of postoperative chemotherapy is apparently necessary.

One reasonable alternative for the management of patients who remain node positive at surgery could be to incorporate radiation into their postoperative adjuvant treatment. There is some evidence for this approach, coming from the South Korean ARTIST trial. In the ARTIST trial, patients who underwent gastrectomy for stomach cancer randomly received either post-operative chemotherapy or chemoradiation therapy. Although no overall survival benefit was seen overall in the ARTIST study, in subset analysis, node positive patients had a significant improvement in disease-free survival with adjuvant chemoradiotherapy versus adjuvant chemotherapy alone.

The disparity in tumor responses to preoperative chemotherapy supports biologic heterogeneity of these tumors. Understanding these biologic differences would be invaluable to the future therapeutic direction of their management. These differences are likely complex, existing as baseline genetic characteristics at diagnosis but also as dynamic changes in these characteristics in response to chemotherapy. Identifying these characteristics could allow clinicians to determine which patients' tumors are unlikely to respond to therapy upfront or how to modify ongoing therapy in response to tumor response and evolution as therapy is administered. Knowing that tumor nodal response to preoperative chemotherapy is a surrogate for outcomes in these patients can provide an important comparator for groups of gene expression and evolution.

Recent work into the characterization of gastric adenocarcinoma suggests that four distinct molecular subtypes exist. These subtypes are characterized as Epstein-Barr virus positive, microsatellite unstable, genomically stable, and chromosome instable. Determination of the genetic features giving rise to each category is based upon analysis of virgin tumor samples, and thus does not consider variations that arise in response to treatment. Nevertheless, each subtype is characterized by the expression or mutation of specific genes or gene categories that can be profiled through routine analysis. Correlating these gene expressions to clinical nodal responses to chemotherapy could potentially offer insights into the biomarker driven determination of adjuvant treatment selection in gastric cancer. Additionally, monitoring these gene expression levels in response to therapy and making similar correlates could provide valuable information into the adaption of adjuvant therapy to the tumor's response to treatment in real time.

The genes associated with these four proposed subtypes of gastric cancer are by no means exhaustive, however, and the expression or alteration of other genes may yield further predictive and therapeutic information for the adjuvant treatment of these cancers. Casting a wider net is increasingly easy to accomplish, with the advent of genomic profiling of tumors. Particularly in the case of profiles that assess for expression of targetable genes, the ability to assess or monitor for their expression as they relate to chemotherapy response could have important implications for being able to incorporate the therapies into the adjuvant treatment process at critical junctures in the treatment sequence.

Inclusion Criteria:

1. Must have pathologically-proven adenocarcinoma of the stomach or gastroesophageal (GE)-junction, stage M0, as established by both imaging and surgical pathologic staging.

   Imaging: Clinical stage of M0 will be established by either CT (chest with contrast and abdomen/pelvis with and without contrast), or CT/PET (positron emission tomography) (skull base to mid-thigh). This is standard post-surgery imaging.

   Surgery: Surgical pathologic staging must be M0.

2. Must have completed 3 cycles of neo-adjuvant chemotherapy. Either CAPEOX or FOLFOX is allowed. Dose modifications are allowed, but all 3 cycles must have been completed.

3. Must have undergone a surgical resection with definitive intent, either by open or laparoscopic resection of the primary gastric or GE junction cancer. Patients must have undergone a total gastrectomy, subtotal gastrectomy, or distal gastrectomy (depending on the location of primary gastric lesion) with at least a modified D2 lymphadenectomy.

4. Must be deemed as a good candidate for adjuvant chemotherapy or chemoradiation (to start within 3 months of surgery), in the opinion of the treating investigator. Plan must be to start adjuvant therapy within 90 days of surgery; adjuvant treatment cannot begin more than 90 days after surgery.

5. Must have diagnostic biopsy tissue (pre-neoadjuvant chemo) available for genetic testing.

6. Must have surgical tissue (post-neoadjuvant chemo) available for genetic testing.

7. Must be > 18 years of age.

8. Must be able to provide informed consent.

9. Must have adequate kidney, liver, and bone marrow function, within 28 days prior to registration, as follows:

   i. Hemoglobin ≥ 8.0 gm/dL

   ii. Absolute neutrophil count (ANC) ≥ 1500 cells/mm3

   iii. Platelet count ≥ 75,000 /mm3

   iv. Calculated creatinine clearance of > 60 mL/min/m2, calculated as follows:

   For males = ((140 - age [years]) x (body weight [kg])) / ((72) x (serum creatinine [mg/dL])

   For females = 0.85 x male value

   v.Total bilirubin ≤ 1.5 times upper limit of normal (ULN)

   vi.AST (aspartate aminotransferase) (SGOT) and ALT (alanine transaminase) (SGPT) ≤ 3.0 times the ULN

10. Must have life expectancy of greater than 3 months.

11. Must have an ECOG (Eastern Cooperative Oncology Group) performance status 0-2.

12. Male or female patients of childbearing potential must be willing to use contraceptive precautions throughout the trial and 3 months following discontinuation of study treatment. Post-menopausal women must be amenorrheic for at least 12 months to be considered of non-childbearing potential.

Exclusion Criteria:

1. Other than the 3 cycles of neoadjuvant chemotherapy and surgery (mentioned above), must not have received other treatment for their gastric cancer.
2. Female patients who are pregnant, breast feeding, or of childbearing potential without a negative pregnancy test prior to baseline. Women of childbearing potential must have a negative serum pregnancy test as a part of eligibility, within 28 days of registration.
3. Patients unwilling or unable to comply with the protocol, or provide informed consent.
4. Patients with clinical evidence of metastatic disease.
5. Any medical condition that, in the opinion of the investigator, would exclude the patient from participating in this study and treatment plan.

From the end of completion of assigned therapy, subjects undergo follow-up every 3 months for a total of 36 months(planned) after the date of surgery or until the study closure