Abstract
Background:
Multimodal treatment strategies with surgery as its centerpiece have been accepted as the standard of care in nonmetastatic cardia gastric cancer (CGC). There remains a lack of consensus regarding the optimal multimodal treatment strategy.
Method:
We queried National Cancer Database from 2004 to 2016 to identify patients with resected nonmetastatic CGC who received perioperative chemotherapy (PEC), postoperative chemoradiation therapy (POCR), or postoperative chemotherapy (POC). A subgroup analysis was performed in optimally treated patients defined as initial chemotherapy within 45 days of diagnosis, resection within 45 days of diagnosis, negative margins, adjuvant chemotherapy within 90 days of resection, and standard radiation dose (45 Gy). Kaplan–Meier, Univariate analysis (UVA), and Multivariable analysis (MVA) were performed.
Results:
We identified 2387 patients. Median survival was 38.8 months in the PEC group, 36 months in the POCR group, and 32.3 months in the POC group (p = 0.1025). On UVA, patients treated with PEC had an association with improved survival (HR, 0.83; p = 0.037) when compared with POC. On MVA, no significant difference was noted in overall survival (OS) between PEC, POCR, and POC, similar to subgroup analysis of optimally treated cohort.
Conclusion:
OS rate in nonmetastatic CGC is not significantly different between patients receiving PEC, POCR, or POC.
Keywords: multimodal treatment in cardia gastric cancer, multimodal treatment in gastric cancer, nonmetastatic cardia gastric cancer, nonmetastatic gastric cancer, resectable cardia gastric cancer
1 |. INTRODUCTION
Gastric cancer (GC) was responsible for over 1 million new cases in 2018 and is the 3rd leading cause of cancer-related mortality worldwide.1 In 2019, an estimated 123 920 people were living with GC in the United States.2 GC represents 1.4% of all new cancer cases diagnosed in the United States, with an estimated 26 380 new cases and 11 090 deaths in 2022.2 Incidence rates vary by region with markedly elevated rates in East Asian countries like Japan and Korea compared with North America and Europe, with a higher prevalence in males.1,3 GC is anatomically divided into the cardia and noncardia based on the location. Risk factors such as obesity and gastroesophageal reflux disease are exclusively associated with GC in the cardia.4 Cardia gastric cancers (CGC) have increased sevenfold, especially in the developed world, while the eradication of H. pylori infections, has reduced the incidence of non-CGC.3,5 CGC are known to be more aggressive compared with non-CGC with a greater depth of invasion and a higher rate of lymphovascular invasion.6,7
Surgery has been a vital cog of multimodal treatment strategies including chemotherapy and chemoradiation therapy—the accepted standard of care for nonmetastatic CGC. The ideal sequence and type of therapy remains to be determined. The Intergroup trials adjuvant chemoradiation (postoperative chemoradiation therapy [POCR]) regime was the first to demonstrate the benefits of multimodal treatment strategy in GC over surgery alone,8 subsequently followed by the Magic trial utilizing perioperative chemotherapy (PEC) and the Classic trial employing postoperative chemotherapy (POC),9,10 establishing the benefits of multimodal treatment strategies over surgery alone. While these three treatment strategies have independently proven to improve survival compared with surgery alone, a head-to-head comparison to identify the ideal treatment strategy is lacking.
The purpose of this study is to query the National Cancer Database (NCDB) database to compare PEC, POCR, and POC to reveal the optimal treatment strategy in patients with resected nonmetastatic CGC.
2 |. MATERIALS AND METHODS
2.1 |. Data source
The NCDB was queried to examine patients with resected nonmetastatic CGC from 2004 to 2016. The NCDB is a nationwide oncology outcomes database, a joint initiative of the American College of Surgeons and the American Cancer Society. It contains hospital registry data that are collected in more than 1500 Commission on Cancer—accredited facilities, representing 72% of newly diagnosed cancer cases nationwide and more than 40 million records. Data on all types of cancer are captured and are used to explore trends in cancer care and serve as the basis for quality improvement.11
2.2 |. Study patients
Patients above 18 years of age, diagnosed with GC in the cardia—C16.0—according to the 3rd edition of International Classification of Diseases for Oncology (ICD-O-3), who underwent definitive resection were identified. The variable “days from diagnosis to definitive surgery” was used to identify these patients. Patients treated with a palliative care intent were excluded. Patients who received PEC (Systemic Surgery Sequence—4), POCR (Systemic Surgery Sequence—3 and Radiation Surgery Sequence—3), and POC (Systemic Surgery Sequence—3) were identified. Patients with non-CGC, metastatic disease, patients who received intraoperative systemic therapy, patients with unknown information on timing of chemotherapy or radiation therapy in relation to surgery, patients treated with any other modality of treatment strategy (like neoadjuvant chemotherapy or radiation therapy or surgery alone), and patients previously treated for cancer were excluded.
2.3 |. Methods
We conducted a retrospective analysis in patients with nonmetastatic CGC who underwent definitive resection to compare different treatment strategies including PEC, POCR, and POC. The NCDB defines PEC as receipt of at least two cycles of chemotherapy before and after surgery.
The primary objective was to evaluate and compare the overall survival (OS) benefits of these three treatment strategies. Additional subgroup analysis was performed in the optimally treated cohort, defined as initial chemotherapy within 45 days of diagnosis (PEC), resection within 45 days of diagnosis (POCR and POC), negative margins, adjuvant therapy within 90 days of resection (POCR and POC), and adequate standard (45 Gy) radiation dose (POCR).
2.4 |. Statistical analysis
Statistical analysis was carried out using SAS 9.3. Descriptive statistics were used to summarize patient characteristics. The Kaplan–Meier method was used to estimate OS in the three treatment cohorts and survival curves were compared using log-rank tests. Univariate analysis (UVA) and Multivariable analysis (MVA) utilizing cox proportional hazard model were used to compare OS between different treatment strategies (PEC, POCR, and POC), demographics (age, sex, race, facility type, and location), surgical variables (lymph nodes, margins), and tumor grade. A backward selection method was used for MVA with an α level of removal of 0.05.
3 |. RESULTS
3.1 |. Demographics, tumor characteristics, and treatment cohorts
We identified 202 216 patients with GC in the NCDB from 2004 to 2016 (Figure 1). We included 2387 patients in our final analysis as per our selection criteria. The median age of this cohort was 63 years (Table 1). More than half the patients were below the median age (53%), while the majority were male (82%) and of White ethnicity (89.5%) with a Charlson-Deyo score of 0 (70%). Majority of the patients presented with a clinical T stage 1–2 (48.7%) and T3 (47.4%) and were clinically N0 (59.2%). Poorly differentiated/undifferentiated tumor (63%), positive regional lymph nodes (80%), and negative surgical margins (81.5%) were commonly noted histopathological findings. Majority of the patients were treated with POCR (n = 1427, 60%), followed by POC (n = 591, 25%) and PEC (n = 369, 15%) (Table 2).
FIGURE 1.
Schematic depicting inclusion and exclusion criteria of patients with nonmetastatic resected cardia gastric cancer including stratification by treatment strategy. NCDB, National Cancer Database; PEC, perioperative chemotherapy; POCR, postoperative chemoradiation therapy; POCR, postoperative chemotherapy.
TABLE 1.
Descriptive statistics including demographics and tumor characteristics of patients with nonmetastatic resected cardia gastric cancer for all included patients and patients who met optimal treatment selection criteria.
| Variable | Level | All patients, N (%) = 2387 | Optimally treated cohort, N (%) = 705 |
|---|---|---|---|
| Median age = 63 years | Median age = 61 years | ||
| Age | ≤Median | 1265 (53%) | 377 (53.5%) |
| >Median | 1122 (47%) | 328 (46.5%) | |
| Sex | Male | 1952 (81.8) | 585 (83.0) |
| Female | 435 (18.2) | 120 (17.0) | |
| Race | White | 2137 (89.5) | 632 (89.6) |
| Black | 131 (5.5) | 37 (5.2) | |
| Other | 119 (5.0) | 36 (5.1) | |
| Facility type | Nonacademic/research program | 1287 (55.2) | 368 (54.0) |
| Academic/research program | 1043 (44.8) | 314 (46.0) | |
| Missing | 57 | 23 | |
| Facility location | Northeast | 680 (29.2) | 215 (31.5) |
| Midwest | 566 (24.3) | 177 (26.0) | |
| West | 354 (15.2) | 88 (12.9) | |
| South | 730 (31.3) | 202 (29.6) | |
| Missing | 57 | 23 | |
| Primary payor | Private/not insured | 1289 (54.0) | 423 (60.0) |
| Medicare/Medicaid/other government | 1098 (46.0) | 282 (40.0) | |
| Year of diagnosis | 2006–2010 | 1360 (57.0) | 391 (55.5) |
| 2011–2016 | 1027 (43.0) | 314 (44.5) | |
| Charlson-Deyo score | 0 | 1665 (69.8) | 536 (76.0) |
| 1+ | 722 (30.2) | 169 (24.0) | |
| Clinical T | 1–2 | 640 (48.7) | 163 (39.4) |
| 3 | 622 (47.4) | 234 (56.5) | |
| 4 | 51 (3.9) | 17 (4.1) | |
| Missing | 1074 | 291 | |
| Clinical N | 0 | 997 (59.2) | 258 (51.1) |
| 1 | 496 (29.4) | 172 (34.1) | |
| 2–3 | 192 (11.4) | 75 (14.9) | |
| Missing | 702 | 200 | |
| Grade | Well/moderately differentiated | 845 (36.8) | 259 (37.6) |
| Poorly differentiated/undifferentiated | 1450 (63.2) | 430 (62.4) | |
| Missing | 92 | 16 | |
| Regional lymph nodes positive | Negative | 453 (20.3) | 170 (24.8) |
| Positive | 1779 (79.7) | 516 (75.2) | |
| Missing | 155 | 19 | |
| Surgical margins status | Negative | 1877 (81.5) | 705 (100%) |
| Positive | 427 (18.5) | 0 | |
| Missing | 83 | 0 | |
| Surgery of primary site | Unknown | 578 (24.20) | 156 (22.12) |
| Local tumor destruction, NOS | 1 (0.04) | 0 | |
| Photodynamic therapy | 5 (0.20) | 0 | |
| Electrocautery | 8 (0.33) | 0 | |
| Cryosurgery | 2(0.08) | 0 | |
| Laser | 1 (0.04) | 0 | |
| Local tumor excision, NOS | 25 (1.04) | 3 (0.42) | |
| Laser ablation | 1 (0.04) | 0 | |
| Laser excision | 4 (0.16) | 0 | |
| Polypectomy | 13 (0.54) | 3 (0.42) | |
| Excisional biopsy | 50 (2.09) | 4 (0.56) | |
| Gastrectomy, NOS (partial, subtotal, hemi-) | 125 (5.23) | 39 (5.53) | |
| Upper (Proximal) gastrectomy | 135 (5.65) | 46 (6.52) | |
| Near total or total gastrectomy | 204 (8.54) | 72 (10.21) | |
| Gastrectomy, NOS with removal of a portion of esophagus | 295 (12.35) | 79 (11.20) | |
| Partial or subtotal gastrectomy with removal of a portion of esophagus | 575 (24.08) | 152 (21.56) | |
| Near total or total gastrectomy with removal of a portion of esophagus | 174 (7.28) | 57 (8.08) | |
| Gastrectomy, NOSa | 30 (1.25) | 12 (1.70) | |
| Partial or subtotala | 61 (2.55) | 27 (3.82) | |
| Near total or total gastrectomya | 65 (2.72) | 31 (4.39) | |
| Radical gastrectomya | 27 (1.13) | 4 (0.56) |
Note: Values in bold indicate statistical significance.
Abbreviation: NOS, not otherwise specified.
Gastrectomy resections with organs other than esophagus. Portions of esophagus may or may not be included in the resection.
TABLE 2.
Descriptive statistics including clinicopathological features of all patients with nonmetastatic resected cardia gastric cancer stratified by treatment groups.
| Covariate | Statistics | Level | Treatment group |
Parametric p valuea | ||
|---|---|---|---|---|---|---|
| Perioperative chemotherapy (N = 369) | Postoperative chemoradiation (N = 1427) | Postoperative chemotherapy (N = 591) | ||||
| Age | N (Col %) | Below median | 224 (60.7) | 758 (53.12) | 283 (47.88) | <0.001 |
| N (Col %) | Above median | 145 (39.3) | 669 (46.88) | 308 (52.12) | ||
| Race | N (Col %) | White | 329 (89.16) | 1280 (89.7) | 528 (89.34) | 0.335 |
| N (Col %) | Black | 19 (5.15) | 85 (5.96) | 27 (4.57) | ||
| N (Col %) | Other | 21 (5.69) | 62 (4.34) | 36 (6.09) | ||
| Sex | N (Col %) | Male | 304 (82.38) | 1182 (82.83) | 466 (78.85) | 0.103 |
| N (Col %) | Female | 65 (17.62) | 245 (17.17) | 125 (21.15) | ||
| Median income quartiles 2008–2012 | N (Col %) | <$38 000 | 48 (13.01) | 188 (13.29) | 76 (12.88) | 0.969 |
| N (Col %) | $38 000–$47 999 | 93 (25.2) | 343 (24.24) | 155 (26.27) | ||
| N (Col %) | $48 000–$62 999 | 96 (26.02) | 391 (27.63) | 160 (27.12) | ||
| N (Col %) | ≥$63 000 | 132 (35.77) | 493 (34.84) | 199 (33.73) | ||
| Facility type | N (Col %) | Nonacademic/research program | 141 (39.39) | 856 (61.58) | 290 (49.83) | <0.001 |
| N (Col %) | Academic/research program | 217 (60.61) | 534 (38.42) | 292 (50.17) | ||
| Primary payor | N (Col %) | Private/not insured | 226 (61.25) | 755 (52.91) | 308 (52.12) | 0.009 |
| N (Col %) | Medicare/Medicaid/other government | 143 (38.75) | 672 (47.09) | 283 (47.88) | ||
| Charlson-Deyo score | N (Col %) | 0 | 287 (77.78) | 987 (69.17) | 391 (66.16) | <0.001 |
| N (Col %) | 1+ | 82 (22.22) | 440 (30.83) | 200 (33.84) | ||
| Year of diagnosis | N (Col %) | 2006–2010 | 136 (36.86) | 885 (62.02) | 339 (57.36) | <0.001 |
| N (Col %) | 2011–2016 | 233 (63.14) | 542 (37.98) | 252 (42.64) | ||
| Facility location | N (Col %) | Northeast | 133 (37.15) | 356 (25.61) | 191 (32.82) | <0.001 |
| N (Col %) | Midwest | 81 (22.63) | 379 (27.27) | 106 (18.21) | ||
| N (Col %) | West | 65 (18.16) | 208 (14.96) | 81 (13.92) | ||
| N (Col %) | South | 79 (22.07) | 447 (32.16) | 204 (35.05) | ||
| Grade | N (Col %) | Well/moderately differentiated | 129 (37.07) | 522 (38.05) | 194 (33.74) | 0.198 |
| N (Col %) | Poorly differentiated/undifferentiated | 219 (62.93) | 850 (61.95) | 381 (66.26) | ||
| Regional lymph nodes positive | N (Col %) | Negative | 130 (36.83) | 230 (17.46) | 93 (16.55) | <0.001 |
| N (Col %) | Positive | 223 (63.17) | 1087 (82.54) | 469 (83.45) | ||
| Clinical T | N (Col %) | 1–2 | 65 (21.31) | 405 (58.19) | 170 (54.49) | <0.001 |
| N (Col %) | 3 | 216 (70.82) | 272 (39.08) | 134 (42.95) | ||
| N (Col %) | 4 | 24 (7.87) | 19 (2.73) | 8 (2.56) | ||
| Clinical N | N (Col %) | 0 | 106 (32.02) | 627 (66.35) | 264 (64.55) | <0.001 |
| N (Col %) | 1 | 169 (51.06) | 223 (23.6) | 104 (25.43) | ||
| N (Col %) | 2–3 | 56 (16.92) | 95 (10.05) | 41 (10.02) | ||
| Pathologic T | N (Col %) | 0–2 | 90 (32.14) | 255 (28.27) | 93 (25.55) | 0.116 |
| N (Col %) | 3 | 169 (60.36) | 583 (64.63) | 232 (63.74) | ||
| N (Col %) | 4 | 21 (7.5) | 64 (7.1) | 39 (10.71) | ||
| Pathologic N | N (Col %) | 0 | 122 (38.98) | 215 (20.44) | 78 (17.93) | <0.001 |
| N (Col %) | 1 | 90 (28.75) | 479 (45.53) | 178 (40.92) | ||
| N (Col %) | 2 | 51 (16.29) | 228 (21.67) | 95 (21.84) | ||
| N (Col %) | 3 | 50 (15.97) | 130 (12.36) | 84 (19.31) | ||
| Surgical margins status | N (Col %) | Negative | 327 (90.58) | 1066 (77.58) | 484 (85.06) | <0.001 |
| N (Col %) | Positive | 34 (9.42) | 308 (22.42) | 85 (14.94) | ||
| Chemotherapy type | N (Col %) | Chemotherapy administered, type and number of agents not documented | 13 (3.52) | 113 (7.92) | 75 (12.86) | - |
| N (Col %) | Single-agent chemotherapy | 4 (1.08) | 586 (41.07) | 105 (17.77) | ||
| N (Col %) | Multiagent chemotherapy | 352 (95.39) | 728 (51.02) | 410 (69.37) | ||
| Surgery of primary site | N (Col %) | Unknown | 79 (21.41) | 406 (28.45) | 93 (15.74) | - |
| N (Col %) | Local tumor destruction, NOS | 0(0) | 1 (0.07) | 0(0) | ||
| N (Col %) | Photodynamic therapy | 0 (0) | 3 (0.21) | 2 (0.34) | ||
| N (Col %) | Electrocautery | 0 (0) | 6 (0.42) | 2 (0.34) | ||
| N (Col %) | Cryosurgery | 0 (0) | 1 (0.07) | 1 (0.17) | ||
| N (Col %) | Laser | 0 (0) | 1 (0.07) | 0(0) | ||
| N (Col %) | Local tumor excision, NOS | 1 (0.27) | 21 (1.47) | 3 (0.51) | ||
| N (Col %) | Laser ablation | 0 (0) | 0(0) | 1 (0.17) | ||
| N (Col %) | Laser excision | 0 (0) | 3 (0.21) | 1 (0.17) | ||
| N (Col %) | Polypectomy | 0 (0) | 11 (0.77) | 2 (0.34) | ||
| N (Col %) | Excisional biopsy | 1 (0.27) | 37 (2.59) | 12 (2.03) | ||
| N (Col %) | Gastrectomy, NOS (partial, subtotal, hemi-) | 15 (4.06) | 79 (5.54) | 39 (6.60) | ||
| N (Col %) | Upper (Proximal) gastrectomy | 20 (5.42) | 72 (5.05) | 43 (7.28) | ||
| N (Col %) | Near total or total gastrectomy | 43 (11.65) | 101 (7.08) | 60 (10.15) | ||
| N (Col %) | Gastrectomy, NOS with removal of a portion of esophagus | 48 (13.01) | 167 (11.7) | 80 (13.54) | ||
| N (Col %) | Partial or Subtotal gastrectomy with removal of a portion of esophagus | 75 (20.33) | 336 (23.55) | 164 (27.75) | ||
| N (Col %) | Near total or total gastrectomy with removal of a portion of esophagus | 40 (10.84) | 95 (6.66) | 39 (6.6) | ||
| N (Col %) | Gastrectomy, NOSb | 7 (1.9) | 15 (1.05) | 8 (1.35) | ||
| N (Col %) | Partial or subtotalb | 11 (2.98) | 35 (2.45) | 15 (2.54) | ||
| N (Col %) | Near total or total gastrectomyb | 26 (7.05) | 22 (1.54) | 17 (2.88) | ||
| N (Col %) | Radical gastrectomyb | 3 (0.81) | 15 (1.05) | 9 (1.52) | ||
Abbreviation: NOS, not otherwise specified.
The parametric p value is calculated by x2 test.
Gastrectomy resections with organs other than esophagus. Portions of esophagus may or may not be included in the resection.
3.2 |. UVA and MVA
Variables associated with OS on UVA (Supporting Information: File 1) were also associated with OS on MVA except facility location and primary payer (Table 3). On MVA, age above median (HR, 1.46; p < 0.001) compared with below median, a Charlson-Deyo score of 1 and 1+ (HR, 0.028; p = 0.028) compared with a score of 0, poorly differentiated/undifferentiated tumor (HR, 1.21; p = 0.002) compared with well-differentiated tumor, positive regional lymph nodes (HR, 1.90; p < 0.001) and positive surgical margins (HR, 2.02; p < 0.001) compared with negative were associated with worse OS.
TABLE 3.
Multivariable analysis of all patients with nonmetastatic resected cardia gastric cancer treated with perioperative chemotherapy, postoperative chemoradiation therapy, and postoperative chemotherapy including subgroup analysis of optimally treated cohort.
| All patient analysis |
Optimally treated cohort |
||||||
|---|---|---|---|---|---|---|---|
| Covariate | Level | Hazard ratio (95% CI) | HR p value | Overall p value | Hazard ratio (95% CI) | HR p value | Overall p value |
| Treatment group: weighted against POC | Perioperative chemotherapy | 1.04 (0.86–1.26) | 0.701 | 0.680 | 1.01 (0.77–1.33) | 0.919 | 0.993 |
| Postoperative chemoradiation | 0.97 (0.85–1.10) | 0.607 | 1.00 (0.78–1.29) | 0.992 | |||
| Postoperative chemotherapy | - | - | - | - | |||
| Treatment group: weighted against PEC | Postoperative chemoradiation | 0.93 (0.78–1.11) | 0.416 | 0.680 | 0.99 (0.77–1.26) | 0.918 | 0.993 |
| Postoperative chemotherapy | 0.96 (0.80–1.17) | 0.701 | 0.99 (0.75–1.30) | 0.919 | |||
| Perioperative chemotherapy | - | - | - | - | |||
| Age | Above median | 1.46 (1.31–1.63) | <0.001 | <0.001 | - | - | - |
| Below median | - | - | |||||
| Median income quartiles 2008–2012 | $38 000–$47 999 $48 000–$62 999 | 0.91 (0.76–1.09) 0.84 (0.70–1.00) | 0.331 0.052 | 0.038 | - | - | - |
| ≥$63000 | 0.79 (0.67–0.94) | 0.009 | |||||
| <$38 000 | - | - | |||||
| Primary payor | Private/not insured | - | - | - | 0.73 (0.60–0.90) | 0.003 | 0.003 |
| Medicare/Medicaid/other government | - | - | - | - | |||
| Charlson-Deyo score | 1+ | 1.14 (1.01–1.29) | 0.028 | 0.028 | - | - | - |
| 0 | - | - | |||||
| Year of diagnosis | 2011–2016 | 0.87 (0.77–0.98) | 0.024 | 0.024 | - | - | - |
| 2006–2010 | - | - | |||||
| Grade | Poorly differentiated/undifferentiated | 1.21 (1.07–1.36) | 0.002 | 0.002 | - | - | - |
| Well/moderately differentiated | - | - | |||||
| Regional lymph nodes positive | Positive | 1.90 (1.62–2.24) | <0.001 | <0.001 | 2.22 (1.67–2.95) | <0.001 | <0.001 |
| Negative | - | - | - | - | |||
| Surgical margins status | Positive | 2.02 (1.76–2.32) | <0.001 | <0.001 | - | - | - |
| Negative | - | - | |||||
Note: Values in bold indicate statistical significance.
Median OS was 38.8 months in the PEC group, 36 months in the POCR group, and 32.3 months in the POC group, showcasing no significant difference in OS between the three treatment groups (p = 0.1025) (Figure 2). On UVA, patients treated with PEC had an association with improved survival (HR, 0.83; p = 0.037) when compared with POC. This association was not maintained on MVA, when PEC (HR, 1.04; p = 0.701) and POCR (HR, 0.97; p = 0.607) were compared with POC. Similarly, no significant difference in OS was seen on MVA between the three treatment strategies, when POCR (HR, 0.93; p = 0.416) and POC (HR, 0.96; p = 0.701) were weighted against PEC.
FIGURE 2.
Kaplan–Meier survival curves in all patients with nonmetastatic resected cardia gastric cancer treated with perioperative chemotherapy, postoperative chemoradiation therapy, and postoperative chemotherapy. Median survival, 5-year and 10-year survival rates included. CI, confidence interval; OS, overall survival.
3.3 |. Subgroup analysis of optimally treated cohort
In the subgroup analysis of our optimally treated cohort, 705 patients were identified of which 242 received PEC, 278 POCR and 185 POC (Supporting Information: File 2). Descriptive statistics were similar to our all-patient analysis (Table 1). Median survival was 48.1 months in the PEC group, 39.9 months for POCR, and 44.4 months for POC, revealing no statistically significant difference between the three treatment groups (p = 0.4095) (Figure 3). UVA showed no significant difference in OS when PEC (HR, 0.84; p = 0.194) and POCR (HR, 0.94; p = 0.631) were weighted against POC (Supporting Information: File 3). On MVA, no significant difference in OS was seen between PEC, POCR, and POC (Table 3).
FIGURE 3.
Kaplan–Meier survival curves in patients with nonmetastatic resected cardia gastric cancer treated with perioperative chemotherapy, postoperative chemoradiation therapy, and postoperative chemotherapy in patients who met optimal treatment selection criteria. Median survival, 5-year and 10-year survival rates included. CI, confidence interval; OS, overall survival.
4 |. DISCUSSION
The purpose of this study was to examine the NCDB to reveal the optimal treatment strategy in patients with nonmetastatic CGC who underwent definitive resection. Our study revealed no significant difference in independent association with OS between PEC, POCR, and POC in this population. This result was maintained in our optimally treated cohort.
Surgical resection with regional lymphadenectomy has been the mainstay of treatment, and multimodal treatment strategies have been developed to counter the risk of incomplete resections and poor survival achieved with surgery alone. Our study revealed no significant difference in OS when we analyzed and compared these multimodal treatment strategies including PEC, POCR, and POC. The ARTIST trial compared POC with POCR in D2 resected GC patients where POCR did not significantly prolong disease-free survival (DFS) (78.2% vs. 74.2%, p = 0.0862). In a subgroup analysis of pathologically node-positive patients, POCR was associated with significantly improved DFS compared with POC (3-year DFS: 76% vs. 72%; p = 0.04).12 Our study is in accord with the results of the ARTIST trials all patient analysis. No separate subgroup analysis on node-positive patients was done in our study. The ARTIST-II trial compared single-agent POC, multiagent POC, and POCR in a 1:1:1 ratio in node-positive GC patients after D2 resection.13 While no difference between multiagent POC and POCR (HR, 0.971; p = 0.879) was noted, both multiagent POC (HR, 0.69; p = 0.04) and POCR (HR, 0.72; p = 0.07) were associated with improved DFS compared with single agent POC. Our study is different from the ARTIST 2 trial as we did not categorize POC into single-agent or multiagent chemotherapy in our treatment cohorts, and our inclusion criteria was not limited to pathological node-positive patients.
The focus in recent clinical trials has gradually been shifting toward optimizing the neoadjuvant approach in resectable GC, primarily because only 60% of the patients are able to start the allocated postoperative treatment.14 The PRODIGY trial compared two groups—neoadjuvant chemotherapy followed by surgery and chemotherapy (CSC) versus upfront resection followed by chemotherapy (SC). The addition of neoadjuvant therapy led to significant tumor downstaging and improved 3-year DFS (66.3% for CSC vs. 60.2% for SC; HR 0.70, log-rank p = 0.023).15 Our study compared OS between PEC and POC and showed no difference between the two treatment strategies. We were not able to analyze DFS as it is not available in the NCDB.
Limitations of this study include the retrospective nature of its design leading to a selection bias. The NCDB does not contain longitudinal data such as DFS. It does not capture the details of treatment strategies such as drugs used, dosage, frequency, and duration of treatment or data on dose reduction and its causes. Additionally, there is no data on the toxicity of a regimen, complications during treatment or hospital stay, and cause of death. The exact number of patients in the intent to treat cohort is unknown as only resected patients were analyzed. Interestingly, studies have reported that only 5%–10% of patients do not receive surgery following preoperative treatment.9,14 Exclusion of patients with missing information including sequence of therapy in relation to surgery is another limitation of this study. Despite these limitations, our study examines a large population database and compares PEC, POCR, and POC in resectable CGC patients and includes a subgroup analysis of optimally treated cohort.
5 |. CONCLUSION
OS rates in resectable CGC patients receiving PEC, POCR, or POC is not significantly different. This data may allow oncologists to offer any of these three treatment strategies based on patient factors such as comorbidities, social support, and functional status in resectable CGC patients. Large randomized controlled trials are needed to further investigate the optimal treatment strategy in resectable CGC patients as survival remains far from satisfactory and efforts to improve outcomes are warranted.
Supplementary Material
ACKNOWLEDGMENTS
This study was supported in part by the Contardi Research Fellowship and the Adriaan Weststrate Memorial Fund. Research reported in this publication was supported in part by the Biostatistics Shared Resource of Winship Cancer Institute of Emory University and NIH/NCI under award number P30CA138292. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The acknowledged parties had no role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.
Abbreviations:
- CGC
cardia gastric cancer
- PEC
perioperative chemotherapy
- POC
postoperative chemotherapy
- POCR
postoperative chemoradiation therapy.
Footnotes
SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the end of this article.
DATA AVAILABILITY STATEMENT
The data that supports the findings of this study are available in the supplementary material of this article
REFERENCES
- 1.Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394–424. [DOI] [PubMed] [Google Scholar]
- 2.National Cancer Institute. Surveillance, epidemiology and end results program. Cancer stat facts: Stomach cancer. 2022. https://seer.cancer.gov/statfacts/html/stomach.html [Google Scholar]
- 3.Balakrishnan M, George R, Sharma A, Graham DY. Changing trends in stomach cancer throughout the world. Curr Gastroenterol Rep. 2017;19(8):36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Karimi P, Islami F, Anandasabapathy S, Freedman ND, Kamangar F. Gastric cancer: descriptive epidemiology, risk factors, screening, and prevention. Cancer Epidemiol Biomarkers Prevent. 2014;23(5):700–713. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Clinton SK, Giovannucci EL, Hursting SD. The World Cancer Research Fund/American Institute for Cancer Research third expert report on diet, nutrition, physical activity, and cancer: impact and future directions. J Nutr. 2020;150(4):663–671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tajima Y, Yamazaki K, Makino R, et al. Differences in the histological findings, phenotypic marker expressions and genetic alterations between adenocarcinoma of the gastric cardia and distal stomach. Br J Cancer. 2007;96(4):631–638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tajima Y, Nakanishi Y, Yoshino T, Kokawa A, Kusano M, Shimoda T. Clinicopathological study of early adenocarcinoma of the gastric cardia: comparison with early adenocarcinoma of the distal stomach and esophagus. Oncology. 2001;61(1):1–9. [DOI] [PubMed] [Google Scholar]
- 8.Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med. 2001;345(10):725–730. [DOI] [PubMed] [Google Scholar]
- 9.Cunningham D, Allum WH, Stenning SP, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med. 2006;355(1):11–20. [DOI] [PubMed] [Google Scholar]
- 10.Bang YJ, Kim YW, Yang HK, et al. Adjuvant capecitabine and oxaliplatin for gastric cancer after D2 gastrectomy (CLASSIC): a phase 3 open-label, randomised controlled trial. Lancet. 2012; 379(9813):315–321. [DOI] [PubMed] [Google Scholar]
- 11.American College of Surgeons, National Cancer Database. https://www.facs.org/quality-programs/cancer/ncdb
- 12.Park SH, Sohn TS, Lee J, et al. 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. J Clin Oncol. 2015;33(28):3130–3136. [DOI] [PubMed] [Google Scholar]
- 13.Park SH, Lim DH, Sohn TS, et al. A randomized phase III trial comparing adjuvant single-agent S1, S-1 with oxaliplatin, and postoperative chemoradiation with S-1 and oxaliplatin in patients with node-positive gastric cancer after D2 resection: the ARTIST 2 trial(☆). Ann Oncol. 2021;32(3):368–374. [DOI] [PubMed] [Google Scholar]
- 14.Cats A, Jansen EPM, van Grieken NCT, et al. Chemotherapy versus chemoradiotherapy after surgery and preoperative chemotherapy for resectable gastric cancer (CRITICS): an international, open-label, randomised phase 3 trial. Lancet Oncol. 2018;19(5):616–628. [DOI] [PubMed] [Google Scholar]
- 15.Kang YK, Yook JH, Park YK, et al. Phase III randomized study of neoadjuvant chemotherapy (CT) with docetaxel(D), oxaliplatin(O) and S-1(S) (DOS) followed by surgery and adjuvant S-1, vs surgery and adjuvant S-1, for resectable advanced gastric cancer (GC) (PRODIGY). Ann Oncol. 2019;30:v876–v877. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that supports the findings of this study are available in the supplementary material of this article