Radiation Oncology/Stomach/Resectable

Resectable Gastric Cancer


  • Possible in 50-60% of patients.
  • Radical subtotal gastrectomy - (for body or antrum tumors) removes 80% of stomach and node-bearing tissue; gastrohepatic and gastrocolic omenta; first portion of duodenum. Margins should be 5 cm or total gastrectomy should be done.
  • Total gastrectomy - (for larger or proximal lesions)

Laparoscopic gastrectomy


Most studies are retrospective studies and meta-analysis of retrospective studies. Only one randomised trial (KLASS Trial)

  • KLASS Trial (2006 - Ongoing)(A Korean Trial) -- laparoscopic distal gastrectomy vs open gastrectomy
    • Randomized. 342 patients, adenocarcinoma, preop Stage I. Arm 1) laparoscopic-assisted distal gastrectomy vs open gastrectomy
    • 2010 PMID 20160637 -- "Morbidity and mortality of laparoscopic gastrectomy versus open gastrectomy for gastric cancer: an interim report--a phase III multicenter, prospective, randomized Trial (KLASS Trial)." (Kim HH, Ann Surg. 2010 Mar;251(3):417-20.)
      • Outcome: Complication rate LADG 10% vs open 15% (NS), postop mortality 1% vs 0% (NS)
      • Conclusion: Laparoscopic gastrectomy safe, trial ongoing

Extent of gastrectomy (Distal Cancer)

  • Prince of Wales Hospital (Hong Kong)(1987-1991) -- subtotal vs. total gastrectomy
    • Randomized. 55 patients with adeno CA of the antrum. Arm 1) R1 subtotal gastrectomy (6 cm proximal resection margin, greater/lesser omentectomy, no further LN dissection) vs. Arm 2) R3 total gastrectomy (omentectomy, splenectomy, distal pancreatectomy, celiac axis LN clearance, skeletonization of porta hepatis vessels)
    • 1994 PMID 8053740 -- "A prospective randomized trial comparing R1 subtotal gastrectomy with R3 total gastrectomy for antral cancer." (Robertson CS, Ann Surg. 1994 Aug;220(2):176-82.)
      • Outcome: median OS R1 4.2 years vs. R3 2.5 years (SS)
      • Conclusion: R3 total gastrectomy should not be routinely used for antral cancer
  • Italy (1982-1993) -- subtotal vs. total gastrectomy
    • Randomized. 622 patients, distal half of the stomach. At surgery, had to have proximal edge of tumor >=6 cm from cardia, no hepatic/intraperitoneal spread, no level 3 LN+. Arm 1) subtotal gastrectomy vs. Arm 2) total gastrectomy. D2 gastrectomy recommended
    • 1997PMID 9389395 --"Total versus subtotal gastrectomy: surgical morbidity and mortality rates in a multicenter Italian randomized trial. The Italian Gastrointestinal Tumor Study Group" (Bozzetti F, Ann Surg. 1997 Nov;226(5):613-20.)
      • Outcome: SG vs TG; Nonfatal complications(9% vs 13%) and death(1% vs 2%) were similar in both groups. Multivariate analysis of postoperative events showed that splenectomy or resection of adjacent organs was associated with a twofold risk of postoperative complications. The mean length of stay, adjusted for extension of surgery, was 13.8 days for subtotal gastrectomy and 15.4 days for total gastrectomy
    • 1999 PMID 10450730 -- "Subtotal versus total gastrectomy for gastric cancer: five-year survival rates in a multicenter randomized Italian trial. Italian Gastrointestinal Tumor Study Group." (Bozzetti F, Ann Surg. 1999 Aug;230(2):170-8.) Median F/U 6 years
      • Outcome: 5-year OS subtotal 65% vs. total 62% (NS). The authors concluded SG is associated with better nutritional status and quality of life.
      • Conclusion: Both procedures similar survival; SG should be procedure of choice, provided proximal margin can be obtained

Extent of lymph node dissection

  • D0 dissection - incomplete removal of D1 level of nodes.
  • D1 dissection - removes only perigastric nodal areas
  • D2 dissection - also removes celiac axis, splenic artery, splenic hilar nodes. Originally, pancreatico-splenectomy was also integral part of Japanese D2 resection, but now no longer practiced
  • There was no survival advantage for D2 dissection over D1 dissection in 2 European trials, but their operative mortality was high (~10%) and the trials were criticized for lack of experience with the surgical procedure and postopoerative care
  • Japanese trial of D2 LND +/- PA-LND had post-op morality of only 0.8%, and D2 dissection continues to be favored in high-volume institutions
  • Maruyama Index (MI) has been developed as a means of evaluating the completeness of nodal dissection. It is a predicted score based on Japanese data, which estimates likelihood of residual disease in lymph nodes unresected by the surgeon. Blinded analysis of INT0116 and of Dutch D1-D2 trials revealed that low MI index (<5) is a strong predictor of survival and can be used to quantify adequacy of lymphadenectomy

  • Japan COG (1995-2001) -- D2 resection vs. D2 + para-aortic LND
    • 5-years; 2008 PMID 18669424 -- "D2 lymphadenectomy alone or with para-aortic nodal dissection for gastric cancer." (Sasako M, N Engl J Med. 2008 Jul 31;359(5):453-62.)
    • Randomized. 523 patients, curable T2b-T4. Arm 1) D2 lymphadenectomy vs. Arm 2) D2 lymphadenectomy + para-aortic LND. No adjuvant therapy
      • Outcome: PA LN+ in 8%. 5-year OS D2 69% vs. D2+PALND 70% (NS)
      • Toxicity: surgery complications D2 21% vs D2+PALND 28% (p=0.07); death rate 0.8% in both groups
      • Conclusion: Para-aortic LND doesn't improve survival in patients undergoing D2 dissection
  • INT 0116 Surgical; 2002 PMID 11923135 -- "Surgical treatment variation in a prospective, randomized trial of chemoradiotherapy in gastric cancer: the effect of undertreatment." (Hundahl SA, Ann Surg Oncol. 2002 Apr;9(3):278-86.)
    • Prospective evaluation of surgical data. 553/556 patients, coded using Japanese node stations. Maruyama program used to estimate likelihood of disease in undissected regional LN stations (Maruyama Index)
    • Outcome: D0 resection in 54%. Median MI score 70. MI independent predictors for survival.
    • Conclusion: Maruyama index, as a measure of unresected nodal disease, as in independent predictor of survival. Surgical undertreatment was frequent and undermined survival
  • Dutch Gastric Cancer Group (1989-1993) -- D1 vs D2 resection
    • Randomized. 711/996 patients (285 found to have incurable disease), adenoCA of stomach. Arm 1) D1 gastrectomy vs. Arm 2) D2 gastrectomy
    • 5-years; 1999 PMID 10089184 -- "Extended lymph-node dissection for gastric cancer." (Bonenkamp JJ, N Engl J Med. 1999 Mar 25;340(12):908-14.)
      • Outcome: 5-year OS D1 45% vs. D2 47% (NS). Similar recurrence patterns, 5-year RFS 63% vs. 62% (NS)
      • Toxicity: surgical complications D1 25% vs. D2 43% (SS), post-op deaths 4% vs. 10% (SS).
      • Conclusion: Results do not support routine use of D2 lymph node dissection
    • Maruyama Index; 2005 PMID 16317484 -- "Low Maruyama index surgery for gastric cancer: blinded reanalysis of the Dutch D1-D2 trial." (Peeters KC, World J Surg. 2005 Dec;29(12):1576-84.)
      • Blinded retrospective. 648/711 patients. Minimum F/U 11 years
      • Outcome: Median MI index 26 (compared with INT0116 = 70). MI <5 strong predictor of survival and relapse risk. Strong dose-response correlation
      • Conclusion: Low MI surgery is associated with enhanced survival; surgeons have more impact on patient survival by low-MI operation than by a particular D-level dissection
    • 15-years; 2010 PMID 20409751 -- "Surgical treatment of gastric cancer: 15-year follow-up results of the randomised nationwide Dutch D1D2 trial." (Songun I, Lancet Oncol. 2010 Apr 19. [Epub ahead of print])
      • Outcome: 15-year local recurrence D1 22% vs D2 12%; regional recurrence 19% vs 13%; 15-year CSS 52% vs 63% (SS); 15-year OS 21% vs 29% (NS)
      • Toxicity: operative mortality D1 10% vs D2 4% (SS)
      • Conclusion: D2 dissection associated with lower locoregional recurrence and improved cancer-related death rates. Spleen-preserving D2 now available, and should be the recommended surgical approach for gastric cancer patients
  • MRC ST01 (1986-1993) -- D1 vs. D2 resection
    • Randomized. 400 patients, staging laparotomy, Stage I-III. Arm 1) D1 resection (removal of LN within 3.0 cm of tumor en bloc with greater omentum and stomach) vs. Arm 2) D2 resection (additional removal of omental bursa, and en bloc distal hemipancreatico-splenectomy for middle/upper lesions, or hepatoduodenal and retroduodenal LN for antral lesions to include Level 1-2 regional LNs)
    • 5-years; 1999 PMID 10188901 -- "Patient survival after D1 and D2 resections for gastric cancer: long-term results of the MRC randomized surgical trial. Surgical Co-operative Group." (Cuschieri A, Br J Cancer. 1999 Mar;79(9-10):1522-30.) Median F/U 6.5 years
      • Outcome: 5-year OS D1 35% vs. D2 33% (NS); death from gastric cancer NS
      • Conclusion: Classic Japanese D2 resection offers no survival benefit over D1

Prognostic factors

  • T-stage: With negative nodes, 5-year survival for T1 (85%-100%), T2 (50-60%), T3 (45%)
  • N-stage: N+ 5-year survival 15%
  • Borrmann type: Types I-II most favorable
  • Minimal node involvement adjacent to the primary only affects prognosis slightly
  • MSKCC nomogram using 7 variables has been validated to predict 5-year DSS, and is superior to AJCC Stage

  • Memorial Sloan Kettering Nomogram
    • MSKCC; 2003 PMID 14512396 -- "Postoperative nomogram for disease-specific survival after an R0 resection for gastric carcinoma." (Kattan MW, J Clin Oncol. 2003 Oct 1;21(19):3647-50.)
      • Nomogram developed based on 1039 patients, R0 resection. Variables: age, sex, primary site, Lauren histotype, number of positive LN, number of negative LN, depth of invasion. Endpoint DSS
      • Conclusion: Nomogram for 5-year DSS constructed
    • Dutch Gastric Cancer Validation; 2005 PMID 15641033 -- "Validation of a nomogram for predicting disease-specific survival after an R0 resection for gastric carcinoma." (Peeters KC, Cancer. 2005 Feb 15;103(4):702-7.)
      • Retrospective. 459 patients. Nomogram performed well when applied to patients treated in a large number of institutions
      • Conclusion: Nomogram better predictions than AJCC staging, regardless of LN dissection

Patterns of failure


Local failure in tumor bed and regional LN in about 40-65% of resected patients. Distant failure by hematogenous or peritoneal spread (in up to 43%, autopsy series).

  • Mass General Hospital; 1990 - clinical failure after resection
    • PMID 2262358 — "Patterns of failure following curative resection of gastric carcinoma." Landry J et al. Int J Radiat Oncol Biol Phys. 1990 Dec;19(6):1357-62.
    • 130 pts. Recurrence in 88 pts.
    • LRF as a component of failure in 38% (49 of 130), sole failure in 16% (21 of 130). By stage, LRR > 35% for T3 or T4, either N0 or N+. Relapse in gastric bed (21%), anastomosis (25%), LN (8%).
    • DM in 52%, high (>50%) in T3 or T4, either N0 or N+.
    • Note: actual incidence is likely to be higher since this was not a reoperative series
  • University of Minnesota; 1982 - PMID 7061243 — "Adenocarcinoma of the stomach: areas of failure in a re-operation series (second or symptomatic look) clinicopathologic correlation and implications for adjuvant therapy." Gunderson LL et al. Int J Radiat Oncol Biol Phys. 1982 Jan;8(1):1-11.
    • 107 pts had planned reoperation; recurrence in 86 pts at reoperation or follow-up.
    • DM alone was uncommon, but present in 25%. LRF as only site of failure in 53% or as a component of failure in 88%.
    • This paper led to the classic illustration depicting sites of local failure, superimposed on a standard radiation field.

Autopsy series:

Outcomes after surgical resection:

Adjuvant RT

  • 1994 British Stomach Cancer Group PMID 7910321 - The second British Stomach Cancer Group trial of adjuvant radiotherapy or chemotherapy in resectable gastric cancer: five-year follow-up. (Hallissey MT, Lancet. 1994 May 28;343(8909):1309-12.) - No benefit for either RT or CT.

Adjuvant Chemo-RT

  • CALGB 80101 / Intergroup (2003-9) link -- Phase III. 5-FU vs ECF
    • 546 pts. Resected gastric or GE junction. ECF = epirubicin, cisplatin, 5-FU (continuous infusion). Randomized after surgery to:
      • Arm 1: 5-FU/LV (bolus) x 1 -> 45 Gy + 5-FU -> 5-FU/LV x 2
      • Arm 2: ECF x 1 -> 45 Gy + 5-FU -> ECF x 2
    • 2011 ASCO Abstract 4003 -- "Postoperative adjuvant chemoradiation for gastric or gastroesophageal junction (GEJ) adenocarcinoma using epirubicin, cisplatin, and infusional (CI) 5-FU (ECF) before and after CI 5-FU and radiotherapy (CRT) compared with bolus 5-FU/LV before and after CRT: Intergroup trial CALGB 80101." (Fuchs CS, J Clin Oncol 29: 2011 (suppl; abstr 4003))
      • MS 37 mo (Arm A) vs 38 mo, NS; 3-yr OS 50% vs 52%. Median DFS 30 mo vs 28 mo (NS); 3-yr DFS 46% vs 47%.
      • Tox: grade 5 (death) in 8 pts (3%) vs 1 (<1%). Grade 4: 40% vs 26% (SS).
      • Conclusion: Compared to bolus 5-FU/LV, ECF does not improve survival.
  • RTOG 0114; 2006 (2001-2004)
    • Phase II, randomized. 73/78 patients. Arm 1) "PCF" - Induction 5-FU, cisplatin, paclitaxel then concurrent 5-FU, taxol vs. Arm 2) "PC" - Induction cisplatin, paclitaxel then concurrent cisplatin, paclitaxel. RT 45/25 in both arms.
      • Induction: 2 cycles. PCF - 5-FU(by continuous infusion, 24/hr x days 1-5,29-33), cisplatin(days 1-5,29-33), taxol(24-hr infusion, days 1,29). PC - cisplatin(days 1,29), taxol(days 1,29)
      • Concurrent: PCF - 5-FU(continuous infusion x 5 days,weekly), taxol(weekly). PC - cisplatin(weekly), taxol(continuous infusion x 5 days,weekly).
    • ASTRO Abstract 2006 -- "A Randomized Phase II Trial Comparing Two Paclitaxel (P))-Cisplatin (C) Containing Chemoradiation (CRT) Regimens As Adjuvant Therapy In Resected Gastric Cancer (RTOG 0114)" (Schwartz G, ASTRO 2006, Abstract 1069)
      • Toxicity (Interim analysis): 5-FU arm significantly higher toxicity (G3 97%) and arm closed. Non 5-FU arm comparable toxicity to INT0116
      • Outcome: 2-year DFS 5-FU/P/C 1.2 years vs. P/C 2.9 years
      • Conclusion: Non-FU regimen (paclitaxel/cisplatin) appears safe and well tolerated. However, lower bound estimate for DFS failed to exceed hypothesis
    • 2009: PMID 19273696 -- "Randomized Phase II Trial Evaluating Two Paclitaxel and Cisplatin–Containing Chemoradiation Regimens As Adjuvant Therapy in Resected Gastric Cancer (RTOG-0114)" (Schwartz GK, J Clin Oncol. 2009 Apr 20;27(12):1956-62.)
      • Closed at interim analysis (with 22 pts entered on PCF arm) due to increased toxicity. Accrual continued on PC arm.
      • Grade 3+ GI toxicity 59% in PCF arm (significantly worse than in INT0116). Median DFS 14.6 mo (PCF), Median DFS not reached in PC arm; 2-yr DFS 52% (PC).
      • Conclusion: Although PC appears safe, the DFS failed to exceed the target goal (set by INT0116) and cannot be recommended.
  • Intergroup INT-0116 (1991-1998) -- Observation vs. Concurrent Chemo-RT + Adjuvant Chemo
    • Randomized. 556 patients. Completely resected (R0) adenocarcinoma of the stomach or GE junction. Stage IB to IV(M0) [1988 staging; IB=T1N1 or T2N0]. Arm 1) Observation vs. Arm 2) Bolus 5-FU (425 mg/m2/d) + LV (20 mg/m2/d) x 1 cycle, followed by concurent chemo-RT one month later. Chemotherapy given on first 4 and last 3 days of RT (5-FU 400 mg/m2 + LV 20 mg/m2). Adjuvant chemo one month following RT with two 5-day cycles of 5-FU/LV given one month apart. A D2 lymph node dissection was recommended, but most (54%) had a less than D1 dissection or had a D1 dissection (31%). 64% completed protocol
    • RT technique: 45 Gy to tumor bed, regional nodes, 2 cm beyond proximal and distal margins of resection. Defined tumor bed by pre-op CT. Lymph nodes included were: perigastric, celiac, local para-aortic, splenic, hepatoduodenal or hepatic-portal, and pancreaticoduodenal. Exclusion of the splenic nodes was allowed in patients with antral lesions if it was necessary to spare the left kidney. For tumors of GE junction, included paracardial and paraesophageal lymph nodes.
    • 2001 PMID 11547741 — "Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction." (MacDonald JS, N Engl J Med. 2001 Sep 6;345(10):725-30.) Median F/U 5 years
      • Outcome: Median survival observation 2.2 years vs chemo-RT 3.0 year (SS, HR for death 1.35). 3-year OS 41% vs 50%. 3-year RFS 31% vs 48%, median 19 months vs 30 months (SS, HR for relapse 1.52). LR 29% vs 19%, regional relapse 72% vs 65% (largely abdominal carcinomatosis), DM higher 18% vs 33%. Regional failure included peritoneal spread or liver mets.
      • Toxicity: Grade 3+ hematologic 54%, GI 33%. 17% stopped treatment due to toxic effects. 32% of pts in chemo/RT group experienced grade 4 toxic effects; 1% had treatment-related deaths.
      • Conclusion: Postop chemo-RT should be considered for patients at high risk for recurrence after curative resection
    • 2012; 10-year PMID 22585691 -- "Updated Analysis of SWOG-Directed Intergroup Study 0116: A Phase III Trial of Adjuvant Radiochemotherapy Versus Observation After Curative Gastric Cancer Resection." (Smalley SR, J Clin Oncol -- May 14, 2012 online before print)
      • 10-year median F/U. Continue to show improved OS (HR=1.32) and RFS (HR=1.51). Second malignancies: 21 pts (RT) vs 8 (Obs); NS. Most subsets showed a benefit, with the exception of pts with diffuse histology (minimal non-significant treatment effect).
      • Conclusion: "Intergroup 0116 (INT-0116) demonstrates strong persistent benefit from adjuvant radiochemotherapy. Toxicities, including second malignancies, appear acceptable, given the magnitude of RFS and OS improvement. LRF reduction may account for the majority of overall relapse reduction. Adjuvant radiochemotherapy remains a rational standard therapy for curatively resected gastric cancer with primaries T3 or greater and/or positive nodes."

Adjuvant chemo vs Adjuvant chemo-RT

  • ARTIST (Adjuvant Chemoradiation Therapy in Stomach Cancer) -- Xeloda/CDDP vs Xeloda/CDDP + concurrent Xeloda/RT
    • Randomized. 458 pts. D2 LN dissection. Randomized to: 1) XP: 6 cycles of capecitabine (2000 mg/m2/day on days 1-14) and cisplatin (60 mg/m2 Day 1), every 3 weeks; or 2) XP/XRT/XP: 2 cycles of XP, then RT 45 Gy with concurrent capecitabine (1650 mg/m2/day), then 2 cycles of XP.
    • 2011 PMID 22184384 -- "Phase III Trial Comparing Capecitabine Plus Cisplatin Versus Capecitabine Plus Cisplatin With Concurrent Capecitabine Radiotherapy in Completely Resected Gastric Cancer With D2 Lymph Node Dissection: The ARTIST Trial." (Lee J, J Clin Oncol. 2011 Dec 19. [Epub ahead of print])
      • Treatment completed as planned in 75.4% (XP) and 81.7% (RT). Addition of XRT to XP did not significantly prolong DFS. However, in the subgroup of pN+ pts, XRT resulted in superior DFS (HR 0.68).
      • Conclusion: "The addition of XRT to XP chemotherapy did not significantly reduce recurrence after curative resection and D2 lymph node dissection in gastric cancer. A subsequent trial (ARTIST-II) in patients with lymph node–positive gastric cancer is planned."
    • 2015 update [1] -- "Phase III Trial to Compare Adjuvant Chemotherapy With Capecitabine and Cisplatin Versus Concurrent Chemoradiotherapy in Gastric Cancer: Final Report of the Adjuvant Chemoradiotherapy in Stomach Tumors Trial, Including Survival and Subset Analyses." (Park S, J Clin Oncol. 2015)
      • With 7 years of follow-up, DFS remained similar between treatment arms (hazard ratio [HR], 0.740; 95% CI, 0.520 to 1.050; P = .0922). OS also was similar (HR, 1.130; 95% CI, 0.775 to 1.647; P = .5272).
      • Subgroup analysis showed DFS improvements for N+ and intestinal type gastric carcinoma
        • N+ -- In 396 patients with node-positive disease, 3-year DFS was significantly different (72% in XP arm v 76% in XPRT arm; P = .04).
        • Intestinal type GC -- 163 patients with intestinal-type GC, 3-year DFS rates were 83% and 94% in the XP and XPRT arms, respectively (P = .01).
  • HeCOG, Greece (2002-2005) -- Taxol/Carbo +/- RT
    • Randomized. Trial stopped early due to slow accrual. 147 out of 200 patients. Gastric CA, stage T3-T4 or N+. Surgery, SM-. LND D0 54%, median LN examined 14. Arm 1) Docetaxel 75 mg + cisplatin 75 mg (replaced with carboplatin AUC 5 due to excessive vomiting) Q3W x 6 cycles vs Arm 2) same chemo + RT 45/25. RT given after 3rd cycle of chemo (delivered in 85%). Immunohistochemistry done on 67 patients
    • 2010 PMID 20130877 -- "A randomized phase III study of adjuvant platinum/docetaxel chemotherapy with or without radiation therapy in patients with gastric cancer." (Bamias A, Cancer Chemother Pharmacol. 2010 Feb 4. [Epub ahead of print]) Median F/U 4.5 years
      • Outcome: Local recurrence chemo 10% vs. chemo-RT 5% (NS). 3-year OS 61% vs 57% (NS); 3-years DFS 51% vs. 48% (NS). Patients with ERCC1+ significantly better OS (median 5.2 years vs 1.6), no difference for Her2 and AMP-Tau
      • Toxicity: Neutropenia 20% vs 24%; diarrhea 7% vs 4%
      • Conclusion: Addition of RT to chemo did not improve outcomes

Neoadjuvant chemotherapy

  • FNCLCC/FFCD (1995-2003) -- surgery alone vs perioperative chemotherapy (cisplatin, 5-FU).
    • Randomized. Trial stopped early after 224 pts (due to slow accrual). Included pts with resectable adenocarcinoma of stomach (25%), GEJ (64%), lower esophagus (11%).
    • 2011 PMID 21444866 -- "Perioperative chemotherapy compared with surgery alone for resectable gastroesophageal adenocarcinoma: an FNCLCC and FFCD multicenter phase III trial." -- Median f/u 25 mos.
      • Results: R0 resection rate 84% (chemo) vs 74% (SS). 5-year overall survival improved chemo/surg 38% vs. surg 24% (SS). 5-year DFS also improved, chemo/surg 34% vs. surg 19% (SS).
      • Conclusion: Preoperative chemotherapy resulted in higher R0 resection rate, OS, and DFS.
      • Note that though the distribution of primary site is fairly different than on MAGIC (with more weight to GEJ), the absolute 5 y OS numbers are pretty close.
  • EORTC 40954 - surgery alone vs preoperative chemotherapy (cisplatin, 5-FU/leucovorin).
    • Randomized. Trial stopped early after 144 pts (due to poor accrual). Included pts with stomach or GE junction primaries (52.8% had tumors of the proximal stomach).
    • 2010 PMID 21060024 -- "Neoadjuvant Chemotherapy Compared With Surgery Alone for Locally Advanced Cancer of the Stomach and Cardia: European Organisation for Research and Treatment of Cancer Randomized Trial 40954." -- Median f/u 4.4 yrs
      • R0 resection rate 81.9% (chemo) vs 66.7%, SS. Positive lymph nodes 61.4% vs 76.5% (SS). Postoperative complications 27.1% vs 16.2% (p=0.09). No differences in overall survival (HR 0.84, NS).
      • Preoperative chemotherapy resulted in higher R0 resection rate but no improved survival
  • MRC MAGIC Trial (1994-2002) -- surgery alone vs peri-operative ECF
    • Randomized. 503 patients. Resectable Stage II+ adenocarcinoma of stomach (75%), GE junction (11%), or lower esophagus (14%). Enrolled at time of diagnosis (vs. after resection in INT0116). Arm 1) Perioperative chemo given as preop ECF (epirubicin, cisplatin, 5-FU) x3 cycles -> surgery -> postop ECF x3 cycles vs. Arm 2) surgery alone. Surgeon decided extent of LND. 42% completed entire chemo-surgery-chemo course
    • 2006 PMID 16822992 -- "Perioperative Chemotherapy versus Surgery Alone for Resectable Gastroesophageal Cancer." (Cunningham D, NEJM 2006 July 6;355(1):11-20.) Median F/U 4 years
      • Outcome: 5-year OS ECF 36% vs surgery 23% (SS); PFS also better. Chemotherapy decreased tumor size and stage. Curative resection ECF 69% vs surgery 66%. LR 14% vs. 21% (significance not stated), DM 24% vs. 37%
      • Conclusion: Survival benefit for chemotherapy in the neoadjuvant/adjuvant setting
  • Dutch FAMTX -- surgery alone vs perioperative FAMTX
    • Randomized. Closed early due to poor outcome in FAMTX arm. 59 patients with adenocarcinoma of the stomach. Arm 1) pre-operative FAMTX (5-FU, doxorubicin, MTX) vs. Arm 2) surgery alone
    • 2004 PMID 15256239 -- Neo-adjuvant chemotherapy for operable gastric cancer: long term results of the Dutch randomised FAMTX trial. (Hartgrink HH, Eur J Surg Oncol. 2004 Aug;30(6):643-9)
      • Outcome: Same resectability rate. Progressive disease in 44%. Median OS FAMTX 1.5 years vs. surgery alone 2.5 years (NS)
      • Conclusion: Early closure due to inadequate rates of curative resection in CT group. No benefit.
    • Comment (in PMID 16822992): Result may reflect inferiority of FAMTX to ECF seen in advanced disease

Neoadjuvant Chemo-RT

  • RTOG 99-04 (1999-2004)
    • Phase II. 49 pts. Preoperative chemotherapy (5-FU + cisplatin) x 2 cycles, followed by chemoradiotherapy (45 Gy, 5-FU + weekly taxol) then resection.
    • 2006 PMID 16921048 -- "Phase II trial of preoperative chemoradiation in patients with localized gastric adenocarcinoma (RTOG 9904): quality of combined modality therapy and pathologic response." (Ajani JA, J Clin Oncol. 2006 Aug 20;24(24):3953-8.)
      • pCR in 26%, R0 resection in 77%. Median f/u 21 months. MS 23 months, 1yr OS 72%. (1yr OS 82% for pts with pCR vs 69% without.)

Neoadjuvant Chemo versus neoadjuvant Chemo-RT

  • German Oesophageal Cancer Study Group (2000-2005) -- Neoadjuvant Chemo vs. neoadjuvant Chemo-RT
    • Randomized. Trial terminated early due to poor accrual. 119 of expected 354 patients with locally advanced T3-4 adenocarcinoma of the lower oesophagus and gastric cardia. Arm 1) 2.5 courses PLF (cisplatin, fluorouracil, leucovorin) vs. Arm 2) 2.0 courses PLF followed by Chemo-RT (30/2) and 1 course of EP (etoposid, cisplatin). Surgery 3-4 weeks after completion of neoadjuvant treatment.
    • 2009 PMID 19139439 -- "Phase III comparison of preoperative chemotherapy compared with chemoradiotherapy in patients with locally advanced adenocarcinoma of the esophagogastric junction." (Stahl M, J Clin Oncol. 2009 Feb 20;27(6):851-6) Median F/U 3.8 years
      • Outcome: 3-y OS: 47% vs. 28% (p=0.07). Higher rates of pCR and ypN0 with RCT than with CT
      • Toxicity: Postoperative mortality 10% vs. 4% (NS).
      • Conclusion: Although study underpowered (due to early closure) to demonstrate a survival benefit through RCT, strong trend towards OS-benefit through RCT.

Adjuvant chemotherapy

  • Historically, adjuvant chemotherapy has not been convincingly shown to provide survival benefit
  • A Japanese trial in patients with D2 dissection was stopped early, after adjuvant chemo with S1 showed an overall survival benefit

  • ACTS-GC, Japan (2001-2005) -- S1 chemo vs observation
    • Randomized. Trial stopped prematurely after significant survival benefit for adjuvant chemo group. 1034 patients, surgery with D2 LND, SM-, Stage II-IIIB. Arm 1) chemo S1 (tegafur + gimeracil + oteracil) x1 years vs Arm 2) surgery only. Primary endpoint OS
    • 2007 PMID 17978289 -- "Adjuvant chemotherapy for gastric cancer with S-1, an oral fluoropyrimidine." (Sakuramoto S, N Engl J Med. 2007 Nov 1;357(18):1810-20.)
      • Outcome: 3-year OS S1 80% vs surgery 70% (HR 0.68, SS); 3-year DFS 72% vs 60%. Local relapse 1% vs 3%, LN relapse 5% vs 9%, peritoneal relapse 11% vs 16%
      • Toxicity: Grade 3-4 anorexia 6%, nausea 4%, diarrhea 3%
      • Conclusion: S1 is effective adjuvant treatment for patients undergoing D2 dissection
  • GOIM 9602 (Italy)(1996-2001) -- observation vs. ELFE (epirubicin/leucovorin/5-FU/etoposide)
    • Randomized. 228 patients, radically resected with total/subtotal gastrectomy and D1 LND, Stage IB-IIIB. Arm 1) observation vs. Arm 2) adjuvant ELFE x6 cycles
    • 2007 PMID 17525087 -- "Adjuvant chemotherapy with epirubicin, leucovorin, 5-fluorouracil and etoposide regimen in resected gastric cancer patients: a randomized phase III trial by the Gruppo Oncologico Italia Meridionale (GOIM 9602 Study)." (De Vita F, Ann Oncol. 2007 Aug;18(8):1354-8. Epub 2007 May 24.) Median F/U 5 years
      • Outcome: 5-year OS control 43% vs. chemo 48% (NS); DFS 39% vs. 44% (NS)
      • Conclusion: No survival benefit for adjuvant ELFE
  • GOIRC (Italy)(1995-2000) -- observation vs. PELF (cisplatin/epirubicin/Leucovorin/5-FU)
    • Randomized. 258 patients, stage IB-IV (T4N2) treated with potentially curative surgery. Arm 1) observation vs. Arm 2) adjuvant PELF x4 cycles. Median time-to-chemo 46 days, 10% never started therapy
    • 2008 PMID 18334706 -- "Adjuvant Chemotherapy in Completely Resected Gastric Cancer: A Randomized Phase III Trial Conducted by GOIRC." (Di Costanzo F, J Natl Cancer Inst. 2008 Mar 11 [Epub ahead of print]) Median F/U 6.1 years
      • Outcome: 5-year OS observation 49% vs. chemo 48% (NS); DFS 42% vs. 42% (NS)
      • Recurrence: no difference between arms; liver 28%, peritoneum 27%, LN 16%
      • Toxicity: Grade 3-4 vomiting 21%, leucopenia 20%; one toxicity-related death
      • Conclusion: No benefit for adjuvant PELF
  • ITMO (Italy)(1992-1997) -- observation vs. EAP (etoposide/adriamycin/cisplatin)
    • Randomized. 274 patients, poor prognisis gastric cancer, s/p subtotal or total gastrectomy with D2 dissection, Stage T3-4 or N+ (90%). Arm 1) observation vs. Arm 2) adjuvant EAP + 5-FU/Leucovorin
    • 2002 PMID 11886009 -- "Adjuvant chemotherapy in gastric cancer: 5-year results of a randomised study by the Italian Trials in Medical Oncology (ITMO) Group." (Bajetta E, Ann Oncol. 2002 Feb;13(2):299-307.) Median F/U 5.5 years
      • Outcome: 5-year OS control 48% vs. chemo 52% (NS); DFS 44% vs. 49% (NS)
      • Toxicity: Grade 3-4 leukopenia 21%, N/V 14%; 2 deaths due to sepsis
      • Conclusion: No benefit for adjuvant EAP
  • EORTC/ICCG (1990-1998) -- 2 trials: observation vs. FAMTX or FEMTX
    • Randomized. 2 trials reported together. Both closed prematurely due to poor accrual. 397 patients, Adeno CA of stomach or GE junction, curative resection (R0-R1), D2 LND, Stage IB-IVM0.
      • EORTC (n=206): observation vs. FAMTX
      • ICCG (n=191): observation vs. FEMTX
    • 2006 PMID 16293676 -- "Randomized phase III trials of adjuvant FAMTX or FEMTX compared with surgery alone in resected gastric cancer. A combined analysis of the EORTC GI Group and the ICCG." (Nitti D, Ann Oncol. 2006 Feb;17(2):262-9. Epub 2005 Nov 17.)
      • Outcome: Combined analysis 5-year OS control 44% vs. chemo 43% (NS); DFS 42% vs. 41% (NS)
      • Conclusion: Neither FAMTX nor FEMTX should be used as adjuvant treatment
  • FFCD 8801 (France)(1989-1997) -- observation vs. 5-FU/cisplatin
    • Randomized. Closed early due to poor accrual. 260 patients, curative resection, Stage II-IVM0 (N+ 80%). Arm 1) observation vs. Arm 2) adjuvant 5-FU/cisplatin x4 cycles
    • 2005 PMID 15939717 -- "Adjuvant chemotherapy with 5-fluorouracil and cisplatin compared with surgery alone for gastric cancer: 7-year results of the FFCD randomized phase III trial (8801)." (Bouche O, Ann Oncol. 2005 Sep;16(9):1488-97. Epub 2005 Jun 6.) Median F/U 8.1 years
      • Outcome: 5-year OS control 42% vs. chemo 47% (NS)
      • Toxicity: Only 49% received >80% of planned chemo dose
      • Conclusion: No benefit for adjuvant 5-FU/cisplatin
  • 1995 SWOG PMID 8591078 - Adjuvant chemotherapy with 5-FU, adriamycin, and mitomycin-C (FAM) versus surgery alone for patients with locally advanced gastric adenocarcinoma (MacDonald JS, Ann Surg Oncol. 1995 Nov;2(6):488-94.) - No benefit.
  • 1994 British Stomach Cancer Group PMID 7910321 - The second British Stomach Cancer Group trial of adjuvant radiotherapy or chemotherapy in resectable gastric cancer: five-year follow-up. (Hallissey MT, Lancet. 1994 May 28;343(8909):1309-12.) - No benefit.
  • 1990 International Collaborative Cancer Group PMID 2199622 - A randomized trial comparing adjuvant fluorouracil, doxorubicin, and mitomycin with no treatment in operable gastric cancer. (Coombes RC, J Clin Oncol. 1990 Aug;8(8):1362-9.) - No benefit.
  • Mitomycin C - Spanish trial (PMID 1900148); Spanish trial (PMID 9508188)
    • Improved OS (small study)
  • FAMTx chemotherapy (replaces mitomycin C with MTX)
  • EAP (etoposide, adriamycin, cisplatin) - trial stopped due to high toxic death rate
  • ELF (etoposide, leucovorin, 5-FU)


  • 2004 PMID 15309728 - "Meta-analysis of intraperitoneal chemotherapy for gastric cancer." (Xu DZ, World J Gastroenterol. 2004 Sep 15;10(18):2727-30.)
    • 11 studies, 3 high quality. Intraperitoneal CT may be beneficial. Continued trials needed.
  • 2002 PMID 12439918 - "Intravenous chemotherapy for resected gastric cancer: meta-analysis of randomized controlled trials." (Hu JK, World J Gastroenterol. 2002 Dec;8(6):1023-8.)
    • 14 trials. IV CT may have positive treatment effect. However, evidence is not strong. Further rigorous trials are required.
  • 2002 PMID 12004845 - "Adjuvant chemotherapy in gastric cancer: a meta-analysis of randomized trials and a comparison with previous meta-analyses." (Panzini I, Tumori. 2002 Jan-Feb;88(1):21-7.)
    • Adjuvant CT results in a significant survival advantage, however, this observation undoubtedly requires confirmation in large randomized controlled trials including cisplatin before adjuvant chemotherapy after curative resection for GC can be proposed for use in clinical practice.
  • 2001 PMID 10997811 - "Efficacy of adjuvant chemotherapy after curative resection for gastric cancer: a meta-analysis of published randomised trials." (Mari E, Ann Oncol. 2000 Jul;11(7):837-43.)
    • Small survival benefit, but adjuvant chemotherapy is still to be considered as an investigational approach
  • 1999 PMID 10533448 -- "Adjuvant chemotherapy after curative resection for gastric cancer in non-Asian patients: revisiting a meta-analysis of randomised trials." (Earle CC, Eur J Cancer. 1999 Jul;35(7):1059-64)
    • Small survival benefit of borderline statistical significance. Continued trials warranted.
  • 1993 PMID 8336183 -- "Adjuvant therapy after curative resection for gastric cancer: meta-analysis of randomized trials." (Hermans J, J Clin Oncol. 1993 Aug;11(8):1441-7.)
    • No benefit

Radiation technique




Volume to irradiate (based on PMID 11979420 - Seminars in Radiation Oncology, Tepper, 2002)
Note: The original table used 2002 AJCC staging. Updated to give 2009 staging equivalents

Stage (2002 AJCC) Remaining stomach Tumor bed Nodes
T1-2 N0 (not into subserosa)
2009 staging: T1-T2
T2 N0 (into subserosa)
2009 staging: T3
Variable Y N
T3 N0
2009 staging: T4a
Variable Y N
T1-2 N+
2009 staging: as above
Y T1=N
T3-4 N+
2009 staging: as above

In general:
Can omit treating remaining stomach if wide (5 cm) surgical margins around the primary tumor and node-negative, but should treat if it won't cause increased morbidity

Specific recommendations:

  • GE junction - include 3-5 cm margin of distal esophagus. Include major portion of left hemidiaphragm if lesion extends through gastric wall.
  • Proximal / cardia -
  • Body - Always treat the remaining stomach. Treat body of pancreas.
  • Antrum - Treat head of pancreas and C-loop of duodenum

RTOG 0114:

  • CTV: Clinical target volume (CTV) is equal to the gastric remnant and the adjacent remaining perigastric nodal tissue, anastamosis(es), lymph node regions of the celiac axis (celiac, splenic, pancreatoduodenal, suprapancreatic nodal beds), porta hepatis lymph node bed, and upper para-aortics (to the level of approximately L3). For patients who had proximal gastric lesions, CTV should also include the lower paraesophageal nodal regions (to the level of approximately T9).
  • PTV: Planning target volume (PTV) is equal to the CTV and an appropriate margin for organ and setup variation as per the discretion of the physician, but a minimum of 0.5 cm. In some cases, this may have to be considerably more, such as when respiratory variation causes significant superior-inferior movement of the gastric remnant.
  • Extension Through Wall: For proximal T3 and T4 lesions the medial 2/3-3/4 of the left hemidiaphragm should be included as target volume with 1.5 cm margins. If the lesion is confined to the gastric wall or is distal, left hemidiaphragm treatment is not necessary
  • Proximal lesions involving the cardia or gastroesophageal junction: The paraesophageal nodes are at risk and should be included in the target volume. A 5-cm margin of esophagus should be included in the cephalad field margin.
  • Distal lesion at or near gastroduodenal junction: A ≥ 5 cm margin of duodenal stump should be included if the gross lesion extended to the gastroduodenal junction as defined pathologically.

  • Consensus report; 2002 PMID 11872272 — "Gastric surgical adjuvant radiotherapy consensus report: rationale and treatment implementation." Smalley SR et al. Int J Radiat Oncol Biol Phys. 2002 Feb 1;52(2):283-93.
    • RT as delivered in INT0116 was difficult to implement, and 35% of plans contained major or minor deviations from protocol. This consensus report was published to clarify RT implementation
    • Review of data supporting RT, clinical and anatomical issues related to RT, and details of practical application of RT to common clinical presentations


  • Tel Aviv; 2007 PMID 17482375 -- "Three-dimensional non-coplanar conformal radiotherapy yields better results than traditional beam arrangements for adjuvant treatment of gastric cancer." (Soyfer V, Int J Radiat Oncol Biol Phys. 2007 Oct 1;69(2):364-9. Epub 2007 May 7.)
    • Treatment planning. 19 patients, adjuvant RT. AP/PA vs. 4F box vs. 4F non-coplanar plans compared
    • Outcome: Adequate CTV coverage in all 3 methods. Kidney doses better with 3D planning
    • Conclusion: Non-coplanar 3D CRT best results for kidney and spinal cord exposure, with adequate CTV coverage
  • Melbourne; 2005 PMID 15763311 -- "3D conformal radiotherapy for gastric cancer--results of a comparative planning study." (Leong T, Radiother Oncol. 2005 Mar;74(3):301-6.)
    • 3D-CRT compared with AP-PA (used in INT0116). 95% PTV dose: 99% 3D-CRT vs. 93% AP-PA
    • Toxicity: RT kidney, LT kidney, SC lower, liver higher (but below tolerance)
    • Conclusion: 3D conformal radiotherapy produces superior dose distributions and reduced radiation doses to the kidneys and spinal cord compared to AP-PA techniques, with the potential to reduce treatment toxicity.


  • Appears to result in superior conformality of PTV and avoidance of critical structures
  • However, there is increased dose inhomogeneity inside the PTV. Given that PTV contains sensitive organs in the post-op setting (e.g small bowel), caution is warranted

  • Tel Aviv; 2008 ASTRO Abstract -- "Does IMRT Offer Clinically Meaningful Advantages in the Adjuvant Management of Gastric Cancer?" (Alani S, International Journal of Radiation Oncology*Biology*Physics, Volume 72, Issue 1, Supplement 1, 1 September 2008, Page S545)
    • Treatment planning. 10 patients. Comparison of IMRT to 3D-CRT (4F non-coplanar approach)
    • Outcome: Satisfactory CTV coverage. OAR liver, spine, and right kidney marginal benefit. Left kidney, mean dose benefit for IMRT 13 Gy vs 3D-CRT 33 Gy
    • Conclusion: IMRT offers marginal benefit, and should only be offered to patients with risk factors for kidney disease
  • Singapore/UCSF; 2008 PMID 18234440 -- "Can all centers plan intensity-modulated radiotherapy (IMRT) effectively? An external audit of dosimetric comparisons between three-dimensional conformal radiotherapy and IMRT for adjuvant chemoradiation for gastric cancer." (Chung HT, Int J Radiat Oncol Biol Phys. 2008 Jul 15;71(4):1167-74. Epub 2008 Jan 30.)
    • Treatment planning. 10 plans, post-op gastric cancer. Planned using 3D-CRT and IMRT at National University Hospital, Singapore (limited IMRT experience) and at UCSF (extensive IMRT experience). Prescribed 45 Gy to 95%
    • Singapore 3D-CRT vs Singapore IMRT: IMRT higher PTV V45 than 3D-CRT, improved liver dose, no difference in kidney dose
    • Singapore IMRT vs UCSF IMRT: UCSF better left and right kidney doses
    • Conclusion: IMRT improves PTV coverage and liver doses; experienced IMRT center can also improve both kidney doses
  • University of Chicago; 2006 PMID 16714752 -- "Intensity-modulated radiation therapy in the treatment of gastric cancer: early clinical outcome and dosimetric comparison with conventional techniques." (Milano MT, Br J Radiol. 2006 Jun;79(942):497-503.)
    • Treatment planning. 7 patients. IMRT compared with AP/PA and 3D-CRT. Then treated to 50.4 Gy
    • Outcome: IMRT better PTV coverage, but greater volume receiving >110% dose. Significant reduction in liver and kidney dose
    • Conclusion: IMRT reduces doses to critical normal tissues. In initial cohort, well tolerated
  • Princess Margaret; 2005 PMID 15978742 -- "IMRT for adjuvant radiation in gastric cancer: A preferred plan?" (Ringash J, Int J Radiat Oncol Biol Phys. 2005 Nov 1;63(3):732-8. Epub 2005 Jun 22.)
    • Treatment planning. 20 patients planned with 3D-CRT (5-fields) replanned using IMRT (7-9 fields).
    • Outcome: Blinded GI radiation oncologists chose IMRT plans 19/20 (95%); better target volume coverage (86%), spinal cord sparing (74%), kidney sparing (69%), liver sparing (71%), heart sparing (69%). However, in 2 plans 3D-CRT deemed comparable/superior due to "hot spots" in small bowel, and greater area of spinal cord irradiated
    • Conclusion: Oncologists frequently preferred IMRT plans based on DVH data for improved PTV coverage and improved OAR sparring
  • Heidelberg; 2004 PMID 15234061 -- "IMRT for postoperative treatment of gastric cancer: covering large target volumes in the upper abdomen: a comparison of a step-and-shoot and an arc therapy approach." (Wieland P, Int J Radiat Oncol Biol Phys. 2004 Jul 15;59(4):1236-44.)
    • Treatment comparison. 15 patients. AP/PA vs 3D-CRT vs step-and-shoot IMRT vs tomotherapy. Prescribed 45 Gy
    • Conclusion: IMRT can deliver efficient dose to PTV. Kidney dose is different from conventional 3D-CRT; advantageous in small number of patients, and probably advantageous but not unequivocal for majority. Choice of treatment depends on department