SEARCH

SEARCH BY CITATION

Keywords:

  • gastric cancer;
  • preoperative radiation;
  • lymph node dissection;
  • gastrectomy;
  • Surveillance, Epidemiology, and End Results

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

BACKGROUND:

Several trials have been conducted to determine the feasibility of preoperative radiotherapy (RT) for gastric cancer. However, the absolute benefit from radiotherapy remains to be defined. In this study, the authors examined the use of preoperative RT (Pre-RT) and postoperative RT (PORT) in patients with gastric cancer from the Surveillance, Epidemiology, and End Results (SEER) database.

METHODS:

The overall survival of patients who had nonmetastatic, resected gastric cancer between 2000 and 2006 was analyzed from the SEER database. Kaplan-Meier survival curves comparing Pre-RT, PORT, and no RT (No-RT) were analyzed using the log-rank test. A multivariate analysis (MVA) was conducted using Cox proportional hazards regression.

RESULTS:

The authors identified 10,251 patients. There was no survival benefit for patients who received Pre-RT or PORT compared with No-RT patients for the entire cohort. Conversely, among lymph node-positive patients, there was a significant survival benefit from both Pre-RT and PORT compared with No-RT (log-rank test: PORT, P < .0001; Pre-RT, P = .0261). The median survival and 5-year overall survival among lymph node-positive patients were 22 months and 24%, respectively, for Pre-RT;29 months and 34%, respectively, for PORT; and 19 months and 20%, respectively, for No-RT. MVA demonstrated that Pre-RT, PORT, and removing ≥15 lymph nodes were independent predictors of improved survival, whereas tumor classification, lymph node status, tumor size, and tumor location were independent predictors of death.

CONCLUSIONS:

The current results supported the use of Pre-RT in select patients with gastric cancer. However, additional trials will be needed to confirm these findings. Cancer 2011;. © 2011 American Cancer Society.

Gastric cancer is a major cause of cancer mortality and represents the second most common cause of cancer death in the world.1, 2 Surgical resection is the only curative option, but outcomes are poor.3 Adjuvant therapies have been investigated with various results. The Intergroup 0116 (INT0116) trial demonstrated a significant improvement in 3-year overall survival with the addition of adjuvant chemoradiotherapy for patients with gastric cancer who underwent surgical resection.4 Although consensus guidelines have adopted this trial as a standard of care for gastric cancer, the trial was criticized for its low rates of D2 lymphadenectomies (10%) and the inclusion of patients with stage IB disease.5-7 Perioperative chemotherapy was investigated in the Medical Research Council Adjuvant Gastric Cancer Infusional Chemotherapy (MAGIC) trial and also resulted in a statistically significant improvement in overall survival; however, <50% of patients in that study completed the full course of therapy.8 Several randomized studies from Europe and Asia as well as a meta-analysis demonstrated a survival benefit for preoperative radiotherapy (Pre-RT) in patients with gastric cancer compared with surgery alone.9-13

Pre-RT theoretically offers multiple advantages. Because the vasculature is intact, there may be enhanced delivery of radiosensitizing chemotherapy and increased response because of the presence of aerated cells. Pre-RT delivery also facilitates improved delineation and coverage of target volumes with potential avoidance of normal structures, which would minimize treatment-related toxicity. In addition, Pre-RT may yield tumor downstaging and reduce the burden of residual microscopic disease at surgery, thus increasing the potential of margin-negative resections. Finally, the delivery of Pre-RT before surgery helps to ensure that patients receive all components of their multimodality treatment.

Studies from the Surveillance, Epidemiology, and End Results (SEER) database have investigated the survival impact of postoperative radiotherapy (PORT) since publication of the INT0116 results,5, 14, 15 but none have examined the use of Pre-RT. Thus, the objective of the current study was to examine the outcomes of patients in the with gastric cancer in the SEER database who received Pre-RT.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Patients

Data were obtained from the SEER limited-access database. In the SEER-17 Registries, the 2008 release of the public use dataset from 2000 and 2006 was queried to identify patients with nonmetastatic gastric cancer who underwent resection. Initially, 12,995 patients were identified. Patients who were included in the analysis were aged ≥20 years, had American Joint Committee on Cancer (AJCC) (sixth edition) stage I through M0 IV disease (according to the sixth edition of the AJCC Cancer Staging Manual), received either PORT, Pre-RT, or no RT (N0-RT), and had the histologic subtypes adenocarcinoma (not otherwise specified [NOS], intestinal type, mucinous or mucin-producing, mixed cell or with mixed subtypes, tubular, and papillary) or carcinoma (NOS, diffuse type, and signet ring cell). All tumor characteristics were derived from pathologic analysis. Patients were excluded from the analysis if they survived for <4 months after surgery, if they had metastatic disease, if they received intraoperative radiation (IORT), if they had received both Pre-RT and PORT, or if the sequence of RT and surgery was unknown. Patients also were excluded if they had the histologic subtypes squamous cell carcinoma, carcinoid, neuroendocrine tumor, small cell carcinoma, leiomyosarcoma, sarcoma, gastrointestinal stromal sarcoma, linitis plastica, adenocarcinoma in villous or tubulovillous adenoma, superficial spreading adenocarcinoma, clear cell adenocarcinoma, adenosquamous carcinoma, melanoma, carcinosarcoma, rhabdoid tumors, or malignant peripheral nerve sheet tumors. All patients underwent some form of gastrectomy and lymph node dissection. Our initial query identified 12,995 patients. Patients were excluded because they were aged <20 years (5 patients), because they had received IORT (2 patients), because they had received both Pre-RT and PORT (18 patients), because the sequence of surgery and RT were unknown (25 patients), because they did not meet histologic criteria (1031 patients), because their survived was <4 months after surgery (1610 patients), and because the type of surgery was unknown (53 patients). Patient characteristics are presented in Table 1.

Table 1. Patient Characteristics
 No RTPostop RTPreop RT
CharacteristicNo.%No.%No.%
  1. RT indicates radiotherapy; Postop, postoperative; Preop, preoperative; AJCC, American Joint Committee on Cancer; NOS, not otherwise specified.

Patients67671003060100424100
Age, y      
 20-59129319.1125340.918944.6
 60-69144421.387228.515436.3
 ≥70403059.693530.68119.1
Sex      
 Men408060.3198264.836786.6
 Women268739.7107835.25713.4
AJCC tumor classification      
 T1199529.52177.18219.3
 T2329448.718065922753.5
 T399314.775324.68419.8
 T447272759317.3
 Tx130.280.300
AJCC lymph node classification      
 N0342750.657718.923154.5
 N1223333157251.416138
 N282712.269422.7276.4
 N32804.12177.151.2
Histology      
 Adenocarcinoma523477.320216636485.8
  NOS366954.214684831875
  Intestinal106315.732510.6194.5
  Mucinous2123.11193.9215
  Tubular1131.7341.120.5
  Mixed1422.1662.240.9
  Papillary350.590.300
 Signet ring124418.488028.85513
 Carcinoma2894.31595.251.2
Location      
 Fundus2103.1963.120.5
 Cardia157523.373223.939292.5
 Greater curvature3294.91876.140.9
 Lesser curvature91713.639612.920.5
 Body5558.22267.451.2
 Antrum186827.67962671.7
 Pylorus3445.11635.330.7
 Overlapping40662417.951.2
 NOS5638.32237.340.9
Surgery      
 Partial gastrectomy362353.5125140.96515.3
  Antrectomy27041043.430.7
  Upper4116.11555.1255.9
  NOS132219.553517.5194.5
  Lower162023.963920.981.9
 Total gastrectomy79611.844314.5133.1
 Gastrectomy and organs234834.7118438.735684

Data that are not included in the SEER database include patient comorbidities, nutritional status, performance status, surgical margin status, postoperative complications, type of lymphadenectomy, chemotherapy, and RT dose and field design. The surgical procedure labeled “gastrectomy with other organs removed” did not specify which organs were removed (ie, spleen, pancreas, esophagus, or liver).

Statistical Analysis

Survival curves were calculated according to the Kaplan-Meier method. Survival data in the current study were for overall survival. Differences between survival curves were examined with the log-rank test. Multivariate analysis (MVA) of prognostic factors related to survival (RT, AJCC tumor [T] classification, number of lymph nodes removed, AJCC lymph node classification, tumor size, surgery type, tumor location, histologic subtype, and year of diagnosis) was performed using a the Cox proportional hazards model. Data were analyzed using SAS statistical software (SAS Institute Inc., Cary, NC).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

We identified 10,251 patients from the initial 12,995 patients who met the inclusion criteria. The survival effect of Pre-RT, PORT, and No-RT is illustrated in Figure 1. For all patients, neither Pre-RT nor PORT had any effect on overall survival (Fig. 1A). The median survival and 5-year overall survival were 33 months and 36%, respectively, for Pre-RT; 34 months and 38%, respectively, for PORT; and 38 months and 41%, respectively, for No-RT. In lymph node-negative patients, there was no benefit from PORT (P = .1181); however, Pre-RT had a negative impact on survival compared with No-RT and PORT (P = .0002) (Fig. 1B). The median survival and 5-year overall survival for lymph node-negative patients were 45 months and 46%, respectively, for Pre-RT; >84 months (median survival not reached) and 56%, respectively, for PORT, and 82 months and 61%, respectively, for No-RT. In lymph node-positive patients, a survival benefit was observed for both Pre-RT (P = .0261) and PORT (P < .0001) compared with No-RT (Fig. 1C). The median survival and 5-year overall survival in lymph node-positive patients were 22 months and 24%, respectively, for Pre-RT; 29 months and 34%, respectively, for PORT; and 19 months and 20%, respectively, for No-RT. Of the 12,995 patients who initially were identified, 1610 patients were not included in the analysis because of survival <4 months status postresection. Of those 1610 patients, 13 received Pre-RT (0.8%); whereas, among patients who met inclusion criteria, 424 received Pre-RT (4.1%; data not shown). This suggests that Pre-RT is not associated with increased postoperative mortality defined as death within 3 months after surgery.

thumbnail image

Figure 1. (A) These Kaplan-Meier curves illustrate the survival of all patients who did not receive radiotherapy (RT) (No RT), who received postoperative RT (Postop RT), and who received preoperative RT (Preop RT). (B) These Kaplan-Meier curves illustrate the survival of patients with negative lymph node status who did not receive RT, who received postoperative RT, and who received preoperative RT. (C) These Kaplan-Meier curves illustrate the survival of patients with positive lymph node status who did not receive RT, who received postoperative RT, and who received preoperative RT. Survival curves were analyzed with the log-rank test.

Download figure to PowerPoint

The pathologic differences in lymph node and primary tumor status between patients in the Pre-RT group and patients in the PORT and No-RT groups combined are presented in Table 2. There was no difference in tumor size (mean, 4.7 ± 3.1 cm vs 4.8 ± 3.0 cm; P = .7498) or distribution according to AJCC T-classification (19.3% vs 22.5% for T1 [P = .1191], 53.5% vs 51.9% for T2 [P = .4747], 19.8% vs 17.8% for T3 [P = .2943], and 7.3% vs 8.6% for T4 [P = .8086]). However, there were significantly fewer positive lymph nodes (mean, 1.7 ± 3.0 vs 3.8 ± 5.9 [P < .0001]) and fewer lymph nodes removed (mean. 11.0 ± 8.5 vs 14.1 ± 11.1 [P < .0001]) in the Pre-RT group compared with the PORT and No-RT groups combined. This suggests that the treatment effect of Pre-RT was on microscopic lymph node disease rather than on the primary tumor.

Table 2. Comparison of Pathologic Characteristics in Patients Who Received Preoperative Radiotherapy
CharacteristicPreop RTNo RT and Postop RTP
  • Preop indicates preoperative; RT, radiotherapy; Postop, postoperative; LN, lymph nodes; SD, standard deviation.

  • a

    Analysis of variance.

  • b

    Chi-square test.

No. of positive LN: Mean±SD1.7±3.03.8±5.9<.0001a
No. of LN removed: Mean±SD11.0±8.514.1±11.1<.0001a
Tumor size: Mean±SD, cm4.7±3.14.8±3.0.7498a
Tumor classification, %   
 T119.322.5.1191b
 T253.551.9.4747b
 T319.817.8.2943b
 T47.38.6.8086b

MVA revealed that Pre-RT, PORT, and removing ≥15 lymph nodes were strong independent predictors of survival (Table 3). AJCC T-classification and N-classification, tumor size, and tumor location were strong independent predictors of death on MVA. A subset analysis of lymph node-negative patients suggested that only PORT and removing ≥15 lymph nodes were strong independent predictors of survival on MVA. AJCC T-classification, tumor size, and tumor location were independent predictors of death in lymph node-negative patients (Table 4). Finally, Pre-RT, PORT, and removing ≥15 lymph nodes were strong independent predictors of survival in lymph node-positive patients, whereas AJCC T-classification and N-classification, tumor size, and tumor location were independent predictors of death on MVA (Table 5). Extent of gastrectomy and histologic subtype were not independent prognostic factors on MVA.

Table 3. Multivariate Analysis of All Patients
VariableHR95% CIP
  • HR indicates hazard ratio; CI, confidence interval; RT, radiotherapy; Postop, postoperative; Preop, preoperative; AJCC, American Joint Committee on Cancer; NOS, not otherwise specified; AC, adenocarcinoma.

  • a

    Continuous variable.

RT (vs no RT)   
 Postop RT0.6110.570-0.655<.0001
 Preop RT0.7070.580-0.861.0006
AJCC tumor classification (vs T1)   
 T21.8141.627-2.023<.0001
 T32.3422.066 - 2.655<.0001
 T43.0072.603-3.474<.0001
AJCC lymph node classification (vs N0)   
 N12.1431.975-2.325<.0001
 N23.5373.195-3.917<.0001
 N35.2764.555-6.110<.0001
No. of lymph nodes dissected (vs <15)0.6520.607-0.700<.0001
Tumor sizea1.0031.002-1.004<.0001
Surgery (vs partial lower)   
 Antrectomy1.0530.876 1.265.5836
 Partial upper1.1250.974-1.300.1095
 Partial NOS1.0810.978-1.195.1275
 Total1.0960.975-1.232.1255
 Gastrectomy with organs1.0970.996-1.209.0608
Location (vs lesser curve)   
 Cardia1.4241.271-1.597<.0001
 Fundus1.2341.017-1.497.0327
 Body1.0020.867-1.159.9742
 Greater curve1.1460.976-1.346.0965
 Antrum1.0590.949-1.181.3097
 Pylorus1.2241.042-1.438.0138
 Overlapping lesion1.1010.947-1.279.2100
 NOS1.2061.042-1.397.0122
Histology (vs papillary AC)   
 Tubular AC1.0620.583-1.933.8447
 Intestinal AC1.1530.664-2.003.6131
 AC NOS1.2990.752-2.245.3486
 Mucinous AC1.2670.719-2.233.4123
 Mixed AC1.4040.783-2.516.2548
 Carcinoma1.0930.620-1.929.7583
 Signet ring1.2550.724-2.176.4182
Table 4. Multivariate Analysis of Lymph Node-Negative Patients
VariableHR95% CIP
  • HR indicates hazard ratio; CI, confidence interval; RT, radiotherapy; Postop, postoperative; Preop, preoperative; AJCC, American Joint Committee on Cancer; NOS, not otherwise specified; AC, adenocarcinoma.

  • a

    Continuous variable.

RT (vs no RT)   
 Postop RT0.7570.631-0.908.0028
 Preop RT0.8910.654-1.214.4637
AJCC tumor classification (vs T1)   
 T21.7251.478-2.015<.0001
 T32.6992.159-3.374<.0001
 T43.9882.997-5.307<.0001
No. of lymph nodes dissected (vs <15)0.6440.553-0.749<.0001
Tumor sizea1.0061.003-1.008<.0001
Surgery (vs partial lower)   
 Antrectomy1.1600.835-1.612.3770
 Partial upper1.1390.869-1.493.3451
 Partial NOS1.1340.937-1.373.1973
 Total1.0570.825-1.355.6596
 Gastrectomy with organs1.1340.927-1.387.2219
Location (vs lesser curve)   
 Cardia1.4771.167-1.868.0012
 Fundus1.1780.807-1.718.3955
 Body1.0850.823-1.431.5635
 Greater curve1.2010.884-1.633.2415
 Antrum1.1410.918-1.418.2355
 Pylorus1.3651.006-1.852.0454
 Overlapping lesion1.3010.914-1.853.1439
 NOS1.3501.011-1.804.0418
Histology (vs papillary AC)   
 Tubular AC1.2520.529-2.962.6093
 Intestinal AC0.7610.355-1.631.4829
 AC NOS0.8780.414-1.860.7336
 Mucinous AC1.1960.531-2.696.6654
 Mixed AC0.7580.297-1.939.5634
 Carcinoma0.6810.296-1.567.3663
 Signet ring0.8740.408-1.875.7300
Table 5. Multivariate Analysis of Lymph Node-Positive Patients
VariableHR95% CIP
  • HR indicates hazard ratio; CI, confidence interval; RT, radiotherapy; Postop, postoperative; Preop, preoperative; AJCC, American Joint Committee on Cancer; NOS, not otherwise specified; AC, adenocarcinoma.

  • a

    Continuous variable.

RT (vs no RT)   
 Postop RT0.5840.542-0.629<.0001
 Preop RT0.5930.455-0.772.0001
AJCC tumor classification (vs T1)   
 T21.5891.351-1.869<.0001
 T31.9701.657-2.342<.0001
 T42.4872.057-3.007<.0001
AJCC lymph node classification (vs N1)   
 N21.6791.542-1.828<.0001
 N32.5392.213-2.912<.0001
No. of lymph nodes dissected (vs <15)0.6490.599-0.704<.0001
Tumor sizea1.0021.001-1.003<.0001
Surgery (vs partial lower)   
 Antrectomy1.0100.810-1.261.9270
 Partial upper1.1200.944-1.329.1941
 Partial NOS1.0550.938-1.187.3740
 Total1.1020.964-1.260.1531
 Gastrectomy with organs1.0890.974-1.217.1362
Location (vs lesser curve)   
 Cardia1.4151.241-1.614<.0001
 Fundus1.2591.005-1.576.0452
 Body0.9870.832-1.171.8784
 Greater curve1.1320.937-1.386.1994
 Antrum1.0440.919-1.185.5117
 Pylorus1.1800.976-1.427.0873
 Overlapping lesion1.0790.913-1.276.3710
 NOS1.1720.988-1.391.0678
Histology (vs papillary AC)   
 Tubular AC1.2140.515-2.861.6576
 Intestinal AC1.6040.715-3.602.2519
 AC NOS1.7740.795-3.962.1617
 Mucinous AC1.6240.715-3.689.2467
 Mixed AC1.9780.861-4.548.1081
 Carcinoma1.5130.666-3.439.3227
 Signet ring1.7120.765-3.834.1909

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Gastric cancer remains a major cause of cancer mortality, and neoadjuvant treatments have been used in an attempt to improve outcomes. Our analysis of the SEER database suggested that both Pre-RT and PORT are associated with a survival benefit for patients who have lymph node-positive gastric cancer on both univariate analysis and MVA. It is noteworthy that there was an association between Pre-RT and poorer survival outcomes in lymph node-negative patients in our univariate Kaplan-Meier analysis (Fig. 1B), yet this association was not demonstrated on MVA (Table 4).

We noted that patients in the Pre-RT group had significantly decreased numbers of positive lymph nodes and lymph nodes dissected compared with the combined PORT and No-RT groups (Table 2). However, there appeared to be no effect of Pre-RT on the primary tumor, because there was no difference in tumor size or T-classification distribution compared with the combined PORT and No-RT groups. Pre-RT, PORT, and removing ≥15 lymph nodes were strong independent predictors of survival on MVA, whereas AJCC T-classification and N-classification, tumor size, and tumor location were independent predictors of death on MVA. Neither extent of surgery nor histology was predictive on MVA.

The main difference in pathologic characteristics observed in the Pre-RT group was draining lymph nodes and not tumor stage or size. It was reported previously that preoperative chemoradiotherapy for gastric cancer achieved a pathologic complete response rate of 26%.16 One explanation for the lack of pathologic tumor differences in the groups of patients analyzed in the studies discussed below, although published data reported pathologic complete responses, is that patients who were treated on some trials may not have been covered by a SEER registry and, thus, are not reflected in the SEER database. The SEER database also lacks any information on either concurrent or sequential chemotherapy, which also can affect response rates. Another issue that arises is how many lymph node-negative patients in the Pre-RT group were clinically lymph node-negative before the initiation of neoadjuvant treatment. Table 1 indicates that 54% of patients in the Pre-RT group were lymph node-negative, whereas 41% of patients in the combined No-RT and PORT groups were lymph node-negative. If patients were converted from clinically positive lymph node status to pathologically negative lymph node status with Pre-RT, then mortality would have been overestimated in the lymph node-negative Pre-RT group, because that group would include clinically lymph node-positive patients. In addition, there would be an underestimation of the survival benefit in the lymph node-positive Pre-RT group, because this potentially may exclude clinically lymph node-positive patients who had a complete response to therapy in the draining lymph nodes. Unfortunately, 1 of the limitations of SEER is that there are no clinical staging data available for gastric cancer.11

Multiple randomized trials have evaluated the benefit of Pre-RT in patients with gastric cancer. Shchepotin et al compared the outcomes of 293 patients who were randomized to surgery alone, Pre-RT, or Pre-RT with local hyperthermia.13 RT was delivered to a total dose of 20 Gray (Gy) in 5-Gy fractions. Pre-RT alone did not increase 3-year or 5-year survival. However, local hyperthermia in combination with RT followed by surgery produced a significant improvement in 3-year survival of 22.1% (from 35.5% ± 4.9% to 57.6% ± 6.3%; P < .05) and an improvement ion 5-year survival of 21.3% (from 30.1% ± 4.7% to 51.4% ± 6.6%; P < .05). Zhang et al conducted a randomized trial of Pre-RT versus surgery alone in 370 patients.9 RT was delivered by anterior-posterior, opposing, parallel fields to deliver 40 Gy. At 5 years and 10 years, significant improvements were noted in overall survival among the patients who received Pre-RT; at 10 years, the overall survival rate was 20% in the Pre-RT group versus 13% in the surgery-alone group (P = .009). Skoropad et al randomized 78 patients to surgery alone versus Pre-RT and intraoperative RT.12 Those authors noted a marked survival advantage in lymph node-positive patients (P = .04) and patients with extragastric extension (P = .042). In a follow-up study by the same group, Pre-RT alone failed to improve survival versus surgery alone in 102 patients with gastric cancer.10 A trend toward improved survival was noted in the RT arm (10-year survival rate, 32% vs 18%; P = .555). Finally, a meta-analysis of randomized trials for gastric cancer revealed that Pre-RT compared with surgery alone significantly reduced the 3-year mortality rate (odds ratio [OR], 0.57; 95% confidence interval [CI], 0.43-0.76; P = .0001) and the 5-year mortality rate (OR, 0.62; 95% CI, 0.46-0.84; P = .002) in 832 patients.11

In an effort to improve outcomes associated with Pre-RT, several prospective, single-arm phase 2 trials have been completed that analyzed the effect of preoperative chemoradiotherapy. Balandraud et al treated 38 patients with gastroesophageal adenocarcinoma with concurrent, infusional 5-fluorouracil (5-FU) and cisplatin using 45 Gy for Pre-RT,17 and 34 patients achieved complete (R0) resection. A pathologic complete response was noted in 6 patients (17.6%), the median survival was 23 months (range, 15-31 months), and the median disease-free survival was 19 months (range, 15-23 months). At 1 year and 2 years, 70.7% and 45.6% of those patients remained alive, respectively.

Klautke et al treated 21 patients using preoperative, concurrent chemoradiotherapy (45 Gy) and infusional 5-FU, cisplatin, and paclitaxel.18 Twelve patients in that study underwent surgery, and 11 achieved R0 resection. Three patients had a pathologic complete response, 13 patients had a partial response, and 14 patients had no response. In that study, the median and 2-year survival rates were 18 months and 42%, respectively, for patients who underwent R0 resection compared with 10 months and 0%, respectively, for the remaining patients (P = .035). The local control at 4 years for patients who achieved R0 resection was 89%.

Ajani et al treated 49 patients with localized gastric adenocarcinoma from 20 institutions using 2 cycles of induction of 5-FU, leucovorin, and cisplatin followed by concurrent chemoradiotherapy with infusional 5-FU, weekly paclitaxel, and 45 Gy. The pathologic complete response and R0 resection rates were 26% and 77%, respectively. Fifty percent of those patients underwent D2 dissection. The median survival was 33.7 months, and the authors noted that patients who achieved a partial or complete pathologic response had a significantly longer median survival than patients who had less than a partial response (63.9 months vs 12.6 months; P = .03).

Rivera et al treated 23 patients with gastric cancer and gastroesophageal cancer using 2 cycles of induction cisplatin and irinotecan followed by concurrent chemoradiotherapy with 45 Gy, cisplatin, and irinotecan.19 Two patients (9%) in that trial achieved a pathologic complete response, and R0 resection was obtained in 15 patients (65%). The median survival was 14.5 months, and the 2-year actuarial survival rate was 35%.

The current study does suffer from several limitations associated with the SEER database. Although SEER is the largest epidemiologic cancer registry in the United States, essential information can be missing: 1) no information is included on resection margins, whether resection was complete, whether additional organs were resected, or the extent of lymphadenectomy; 2) there is no information on comorbidities, nutritional status, or performance status of the patients; 3) the receipt of either concurrent or sequential chemotherapy is unknown; 4) no data are provided on local control; 5) no information is provided on the use of salvage therapy in patients with local recurrences; and 6) no information is provided on which target volumes or prescribed doses for RT were used. In addition, it is impossible to fully understand from SEER why some patients may have received Pre-RT versus PORT versus No-RT; indeed, it is possible that healthier patients, who would have a better prognosis, received Pre-RT rather than patients in the No-RT group. However, despite these inherent biases, SEER documents the real-world outcomes of these patients. Although these limitations may confound the results, our findings do elucidate a possible association between Pre-RT and improved survival for patients who have lymph node-positive gastric cancer.

In conclusion, our analysis of the SEER database demonstrates an association between the receipt of Pre-RT and survival in patients with lymph node-positive gastric cancer. Pre-RT was an independent predictor of survival for patients with lymph node-positive disease. Although Pre-RT was associated with poorer overall survival for patients with lymph node-negative disease on univariate analysis, this association was not demonstrated on MVA. However, stage migration may play a role in overestimating the mortality in lymph node-negative patients. The current results also demonstrated an association between PORT and improved overall survival. Additional trials will be needed to clarify the benefit of Pre-RT in patients with gastric cancer.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES