The data were collected from 187 participating hospitals across the country. The geographical distribution of the registered patients among Japan’s 47 prefectures is illustrated in Fig. 1. More than 1000 patients per year were registered in the prefectures of Tokyo and Osaka; on the other hand, the number of registered patients was less than 100 in 15 prefectures. The hospital volumes in the participating hospitals are indicated in Fig. 2. The median hospital volume was 66 patients per year.
Data of 13067 patients who had undergone surgery in 2001 for primary gastric tumors were eventually accumulated. Of these, 88 patients with benign tumor or non-epithelial tumor were excluded from the analysis. Ninety-four patients who received endoscopic mucosal resection were also excluded. Data of 881 patients lacked essential items. Consequently, data of the remaining 12004 patients were used for the final analysis.
The results are shown in Tables 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, and 28; data in these Tables are for the total number of patients, survival rates by year, standard error of 5YSR, direct death within 30 postoperative days, numbers lost to follow-up within 5 years, 5-year survivors, and main causes of death (such as local and/or lymph node metastasis, peritoneal metastasis, liver metastasis, distant metastasis, recurrence at unknown site, other cancer and other disease). Figures 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, and 14 show cumulative survival curves of patients stratified by essential categories.
The 5YSR in the 12004 patients with primary gastric cancer was 69.1% (Table 3; Fig. 3). Within 5 postoperative years, 1976 patients were lost to follow-up; the follow-up rate was 83.5%. Of the 12004 patients, 11261 underwent gastric resection; 350 were unresected; and in 393 the type of surgery was not specified. Accordingly, the resection rate was 97.0% (11261/11611). Sixty-three of the 11261 patients who had undergone gastrectomy died within 30 postoperative days; the direct death rate was 0.6% (Table 4; Fig. 4).
The most frequent cause of death in patients who had received gastrectomy was peritoneal metastasis (n = 1040), followed, in descending order, by other diseases (n = 501), liver metastasis (n = 357), recurrence at an unknown site (n = 298), and local recurrence including node metastasis (n = 267).
The proportion of male patients was 69.6% and their 5YSR was lower than that of female patients (P < 0.01; Table 5; Fig. 5). The proportion of patients who were more than 80 years old was 7.0%, and their 5YSR was 48.7% (Table 6; Fig. 6). Upper-third gastric cancer accounted for 21.4% of the cases, and the 5YSR (65.3%) of patients with cancer at this site was lower than that for the middle- and lower-third cancers (P < 0.001; Table 7; Fig. 7). The proportion of patients with type 4 cancer was 7.0%, and their 5YSR was markedly low, at 20.4% (P < 0.001; Table 8; Fig. 8). In regard to the histological type, the 5YSR of patients with undifferentiated type, including poorly differentiated adenocarcinoma, signet-ring cell carcinoma, and mucinous adenocarcinoma, was 64.6%. The undifferentiated type showed a poorer prognosis than the differentiated type (P < 0.001; Tables 9, 10). The grade of venous invasion (v0–v3) and that of lymphatic invasion (ly0–ly3) showed significant correlations with prognosis (P < 0.001; Tables 11, 12).
There was a high incidence of early-stage cancer, as indicated in Tables 13 and 14 and Figs. 9 and 10. The proportion of pathological T1 (pT1; mucosal or submucosal) cancer was 51.2%. The 5YSR of this population was 90.8%, and the primary cause of death was not cancer recurrence (n = 96), but other diseases (n = 207).
Peritoneal washing cytology (CY) was carried out for 3481 of 5857 patients with T2, T3, and T4 cancer (59.4%). The 5YSR of cytology-positive patients (CY1) was 12.3%, which corresponded with that of the patients with peritoneal metastasis (P1) (Tables 17, 18).
The 5YSRs of the patients stratified by the JGCA staging system were 91.9% for stage IA, 85.1% for stage IB, 73.1% for stage II, 51.0% for stage IIIA, 33.4% for stage IIIB, and 15.8% for stage IV. These JGCA 5YSRs seemed to correlate well with the TNM 5YSRs (Tables 20, 21, 22, 23; Figs. 12, 13).
In regard to the operative procedure, the proportion of patients who underwent laparoscopic gastrectomy was 3.6%, and their 5YSR was 97.4%. Laparoscopic surgery was carried out mainly in patients with early gastric cancer. Only 1.0% of the patients were treated by thoraco-laparotomy, and their 5YSR was 40.7%. Thoraco-laparotomy was carried out in patients with gastric cardia cancer invading the esophagus (Table 24). Thirty percent of the patients underwent total gastrectomy, and their 5YSR was 53.7%. The proportion of patients treated by modified surgery such as proximal gastrectomy, pylorus-preserving gastrectomy, segmental gastrectomy, and local resection was 9.4% (Table 25). D0, D1, D1+α, and D1+β dissections were carried out in 7.4, 21.7, 12.5, and 5.5% of the patients, respectively. According to the JGCA gastric cancer treatment guidelines [7, 8], D1+α dissection with modified gastrectomy was indicated for T1(M)N0 tumors and T1(SM)N0 differentiated tumors <1.5 cm in diameter, while D1+β dissection with modified gastrectomy was indicated for T1(SM)N0 undifferentiated tumors, T1(SM)N0 differentiated tumors larger than 1.6 cm, T1(M)N1 tumors, and T1(SM)N1 tumors <2.0 cm. D0 and D1 dissections were carried out mainly in patients with non-curative factors or poor surgical risks. D2 lymph node dissection was carried out in 49.3% of the patients and the risk of direct death in those with D2 gastrectomy was 0.5% (28/5403; Table 26).
The curative potential of gastric resection was an important prognostic factor. The proportion of patients with a high probability of cure (resection A) was 63.7%, and their 5YSR was 88.7%. On the other hand, the proportion of patients with definite residual tumor (resection C) was 12.8%, and their 5YSR was 13.4% (Table 28; Fig. 14).