ABSTRACT
Introduction
In patients with advanced gastric cancer invading adjacent organs, extended multivisceral resection is required to achieve R0 resection. However, the survival benefit of combined gastrectomy and pancreatic resection remains controversial. This study aimed to investigate the safety and efficacy of combined gastrectomy and pancreatectomy for gastric cancer.
Materials and Methods
This study retrospectively included 64 patients who underwent pancreaticoduodenectomy (PD) or distal pancreatectomy with splenectomy (DP) for primary gastric cancer with invasion of the pancreas by the primary tumor, from lymph node metastasis or via duodenal invasion.
Results
PD was performed in 18 patients (28%) and DP in 46 (72%). Macroscopic invasion of the pancreas from the primary tumor was observed in 47 patients (73%), from lymph nodes in 13 (20%) and via duodenal invasion in four (6%). Pathological pancreatic invasion was observed in 27 patients (57%). Morbidity due to postoperative intra‐abdominal infectious complications (PIICs) of Clavien–Dindo Grade III or higher was observed in 33 patients (52%). Multivariate analysis showed that duodenal invasion was an independent risk factor for PIICs (p = 0.039). Surgery resulted in R1 resection in 12 patients (18%). Multivariate analysis identified R1 resection (hazard ratio (HR): 5.315, p < 0.001) and PIICs (HR: 2.067, p = 0.027) as independent prognostic factors for overall survival. Multivariate analysis of relapse‐free survival identified PIICs as an independent prognostic factor (HR: 2.345, p = 0.036).
Conclusion
Pancreatic resection is considered safe and effective in patients with pancreatic or duodenal invasion from gastric cancer when R0 resection is possible.
Keywords: combined resection, gastric cancer, pancreatic invasion, pancreaticoduodenectomy, pancreatosplenectomy
Short‐term and long‐term results of patients underwent conbined pancreatectomy was investigated in 64 patients with gastric cancer. R1 resection and postoperative intraabdominal infectious complications were identified as independent prognostic factors.
1. Introduction
Worldwide, gastric cancer is the fifth most common cancer and the third leading cause of cancer‐related mortality [1]. Radical gastrectomy with lymph node dissection is the standard treatment for gastric cancer, and complete resection without residual disease (R0 resection) is the only way to achieve a cure. In patients with invasion of adjacent organs, extended multivisceral resection is required to achieve complete tumor clearance [2]. Pancreaticoduodenectomy (PD) or distal pancreatectomy (DP) is needed for R0 resection when the tumor invades the pancreas. Previous reports have suggested that R0 resection with gastrectomy and pancreatectomy might prolong survival in gastric cancer patients [3]. We also previously reported that PD would be beneficial for gastric cancer with invasion of the pancreas if R0 resection is possible [4]. However, several studies have reported that survival in patients who undergo PD might be worse than that in patients who undergo DP or segmental resection [5], and that PD for advanced gastric cancer with invasion of the duodenum does not prolong survival [6]. Furthermore, there is reportedly no difference in survival between cases with invasion of the head of the pancreas via direct tumor invasion versus lymph node metastasis [7]. In addition, Tran et al. [8] recently reported that combined resection of the pancreas for locally advanced gastric cancer might increase the incidence of complications, leading to poor survival. Therefore, whether combined resection of the pancreas improves long‐term survival in gastric cancers invading the pancreas is still unclear. We investigated short‐term and long‐term outcomes and prognostic indicators to elucidate the safety and efficacy of combined pancreatectomy and gastric resection in gastric cancer patients.
2. Materials and Methods
2.1. Patients
The flow diagram of patient is shown in Figure 1. Of the 2366 patients with locally advanced gastric cancer at Shizuoka Cancer Center from January 2009 to December 2022, 51 (2.2%) were preoperatively diagnosed with cT4b (pancreas), 3 of whom received neoadjuvant chemotherapy (NAC) and performed DP. Of the remaining 48 patients, 7 with intraoperative diagnosis as T4b (pancreas) (sT4b) and 1 with non‐sT4b underwent combined pancreatic resection due to invasion from lymph node metastasis (PD in 5 patients, DP in 3). Of the 2315 with non‐cT4b (pancreas), 146 received NAC. 5 Of them with sT4b and 3 with non‐sT4b (pancreas) but with pancreatic invasion from lymph node metastasis underwent pancreatic resection (PD in 1, DP in 7). Of the remaining 2169 patients, 32 patients with sT4b (pancreas) and 13 patients with non‐sT4b (pancreas) (4 with duodenal invasion and 9 with invasion from lymph node metastasis) underwent combined pancreatic resection (PD in 12, DP in 33). Among the patients received pancreatic resection, splenic artery invasion was also observed in 4 patients (2 from primary tumor and 2 from metastatic lymph node). Ultimately, 64 patients who underwent combined pancreatic resection were included in the present analysis. Patient's data were collected from the prospectively registered database and medical records. The histological depth of the tumor and degree of lymph node metastasis were determined by the UICC TNM classification [9]. Histological classification and pathological response rate were classified according to the 3rd edition of the Japanese classification of gastric carcinoma [10]. Gastrectomy and lymph node dissection were performed in accordance with the Japanese Guidelines for the Treatment of Gastric Cancer 6th Edition [2]. This study was approved by the ethics committee of Shizuoka Cancer Center (approval number: J2023‐64‐2023‐1).
FIGURE 1.
Flow diagram of inclusion in this study.
2.2. Surgery and Perioperative Treatment
Pancreatectomy was performed as PD or DP. Either distal gastrectomy (DG) or total gastrectomy (TG) was performed in PD, whereas TG was performed in all patients who underwent DP. PD was performed in patients with invasion of the pancreatic head or body by the primary tumor or metastatic lymph nodes, or in those in whom a negative distal surgical margin could not be obtained due to invasion of the duodenum. In contrast, DP was performed when the primary tumor or metastatic lymph nodes infiltrated the pancreatic body or tail. If invasion of the pancreatic body was confined to the left side of the left edge of the portal vein, DP was selected, whereas PD was performed if it extended to the right side. In all patients, cytology specimens were collected from the pouch of Douglas and the left subphrenic area immediately after laparotomy.
NAC was indicated for patients with other organ invasion, extensive lymph node metastasis, or severe duodenal invasion. The regimen was determined at the discretion of the attending physician. Adjuvant chemotherapy was performed for pStage II and III cancers.
2.3. Evaluation
Short‐term and long‐term outcomes were evaluated by examining the clinicopathological characteristics and survival of the patients. Postoperative complications occurring within 30 days after surgery were classified according to the Clavien–Dindo classification (CD) [11]. Pancreatic fistula, anastomotic leakage, and intra‐abdominal abscess were defined as postoperative intra‐abdominal infectious complications (PIICs). Postoperative follow‐up was conducted in accordance with the Japanese gastric cancer treatment guidelines [2]. The median observation period was 27.0 (interquartile range [IQR]: 10.7–66.4) months.
2.4. Statistical Analyses
Continuous variables are presented as the median and range. Nominal variables were compared using Fisher's exact test. Risk factors for PIICs were evaluated using logistic regression analysis. Overall survival (OS) and recurrence‐free survival (RFS) were estimated using the Kaplan–Meier method to create survival curves, and differences were assessed using the log‐rank test. p values of 0.05 (two‐tailed) were considered significant. Independent prognostic factors were identified by multivariate analyses using the Cox proportional hazards model. All statistical analyses were conducted using SPSS software (version 29.0, IBM Corporation, NY, USA).
3. Results
3.1. Patients' Characteristics
The background characteristics of the eligible patients are shown in Table 1. Median patient age was 70 years, and the male‐to‐female ratio was 2:1. In terms of histological type, undifferentiated‐type was observed in about two‐thirds of the patients. Clinical T4b (cT4b (pancreas)) was observed in about 20% of the patients, and a clinical diagnosis of pancreatic invasion was observed in only 14% of the patients. Clinical lymph node involvement was seen in approximately 80% of the patients. The most common macroscopic type was type 3, followed by types 2 and 4. NAC was performed in 11 patients, and the median treatment duration was 56 days. Surgical factors are summarized in Table S1. As for the surgical approach, open surgery was performed in all patients. Approximately 30% of the patients underwent PD, and the remaining patients underwent DP. Regarding the extent of gastrectomy, 75% of the patients underwent TG, and 25% underwent DG. Only two patients underwent TG with PD. The main reason for combined resection of the pancreas was direct invasion from the primary tumor, followed by lymph node metastasis and invasion via the duodenum. The patients' pathological characteristics are shown in Table S2. Invasion of the pancreas was histologically confirmed in 27 of 47 patients (57%) who were diagnosed intraoperatively as positive for invasion. Pathological stage III was the most common, accounting for about two‐thirds of the patients. Approximately 20% of the patients underwent R1 resection. The reason for R1 was intraperitoneal cytology positivity, except in one patient with positive proximal margins. In the 11 patients who underwent NAC, the pathological response rate was 27.3% (3 patients). Adjuvant chemotherapy was initiated in 37 patients (57.8%), and treatment completion was achieved in 29 patients (45.3%). Reasons for not receiving postoperative adjuvant chemotherapy included patient preference in 13 patients (20.3%), deterioration of general condition due to postoperative complications in 9 patients (14.1%), ineligibility for adjuvant therapy based on postoperative pathological findings in 3 patients (4.7%), and early recurrence in 2 patients (3.1%). Of the 13 patients who refused adjuvant chemotherapy, 4 were found to have grade III or higher PIICs.
TABLE 1.
Background characteristics of eligible patients.
| N = 64 [%] | |
|---|---|
| Age (year) | |
| Median | 70 |
| Range | 23–82 |
| Sex | |
| Male | 43 [67.2] |
| Female | 21 [32.8] |
| Macroscopic type | |
| 0 | 2 [3.1] |
| 1 | 3 [4.7] |
| 2 | 20 [31.3] |
| 3 | 29 [45.3] |
| 4 | 6 [9.4] |
| Histological type (dominant) | |
| Differentiated | 19 [29.7] |
| Undifferentiated | 45 [70.3] |
| Clinical T | |
| 2 | 2 [3.1] |
| 3 | 3 [4.7] |
| 4a | 47 [73.4] |
| 4b | 12 [18.8] |
| Pancreas | 9 [14.1] |
| Spleen | 2 [3.1] |
| Liver | 1 [1.6] |
| Clinical N | |
| 0 | 13 [20.3] |
| 1 | 14 [25.0] |
| 2 | 20 [31.3] |
| 3 | 17 [26.6] |
| Preoperative chemotherapy | |
| Not performed | 53 [82.8] |
| Performed | 11 [17.2] |
| SP | 7 [10.9] |
| SOX | 3 [4.7] |
| FOLFOX | 1 [1.6] |
| Postoperative chemotherapy | |
| Not performed | 27 [42.2] |
| Performed | 37 [57.8] |
| S‐1 | 31 [48.4] |
| DS | 4 [6.2] |
| CapeOX | 2 [3.1] |
Abbreviations: CapeOX, capecitabine + oxaliplatin; DS, docetaxel + S‐1; FOLFOX, 5‐FU + leucovorin + oxaliplatin; SOX, S‐1 + oxaliplatin; SP, S‐1 + cisplatin.
3.2. Postoperative Complications
Details of the postoperative complications are summarized in Table S3. Postoperative complications of CD grade II or higher were observed in 70% of the patients.
There was no postoperative mortality. Surgery‐related complications with CD grade II or higher were noted in 66% of patients. Pancreatic fistula, intra‐abdominal abscess, and suture failure were the most common complications, in that order. PIICs occurred in all patients who developed surgery‐related complications. Non‐surgery‐related complications of grade II or higher were observed in approximately 20% of patients. Of the non‐surgical complications, postoperative pneumonia, pleural effusion, and thromboembolic events were the most frequently observed, and one patient had both pulmonary thromboembolism and portal vein embolism. The univariate and multivariate analyses for PIICs are presented in Table 2. Univariate analysis demonstrated that patients who underwent pancreatic resection due to invasion to the duodenum had a significantly higher incidence of grade III or higher PIICs. Multivariate analysis identified invasion to the duodenum as the sole independent risk factor for PIICs.
TABLE 2.
Multivariate analysis to identify independent risk factor for PIICs.
| Univariate | Multivariate | |||
|---|---|---|---|---|
| OR (95% CI) | p | OR (95% CI) | p | |
| Age (year) | ||||
| ≥ 70 vs. < 70 | 1.275 (0.429–3.828) | 0.803 | — | — |
| Sex | ||||
| Male vs. female | 0.953 (0.294–3.069) | 1.000 | — | — |
| BMI (kg/m2) | ||||
| ≥ 25 vs. < 25 | 0.721 (0.128–3.752) | 0.729 | — | — |
| Age‐adjusted Charlson comorbidity index (pt) | ||||
| ≥ 4 vs. ≤ 3 | 0.778 (0.259–2.312) | 0.627 | — | — |
| Tumor depth | ||||
| cT4b vs. others | 0.478 (0.091–2.151) | 0.331 | 0.482 (0.120–1.940) | 0.305 |
| Lymph node metastasis | ||||
| cN+ vs. cN− | 0.893 (0.215–3.597) | 1.000 | — | — |
| Gastrectomy | ||||
| TG vs. DG | 0.557 (0.142–2.015) | 0.392 | — | — |
| Pancreatectomy | ||||
| PD vs. DP | 2.349 (0.674–9.036) | 0.169 | 2.050 (0.589–7.100) | 0.260 |
| Histological type | ||||
| Undifferentiated vs. differentiated | 1.265 (0.379–4.281) | 0.786 | — | — |
| Preoperative chemotherapy | ||||
| Performed vs. not performed | 0.748 (0.159–3.350) | 0.747 | — | — |
| Reason for combined resection | ||||
| Pancreatic invasion vs. duodenal invasion | NA (0.000–1.211) | 0.038 | 0.254 (0.069–0.935) | 0.039 |
Note: Multivariate analysis identified invasion to the duodenum as the sole independent risk factor for PIICs.
Abbreviations: BMI, body mass index; DG, distal gastrectomy; DP, distal pancreatectomy; PD, pancreaticoduodenectomy; TG, total gastrectomy.
3.3. Long‐Term Outcomes
Thirty‐six patients (56%) died over the follow‐up period, 32 (50%) due to the gastric cancer, and four (6%) because of other diseases. The OS curve is shown in Figure 2A, showing a 3‐year OS of 46% and a 5‐year OS of 42%, with a median survival time of 34.0 months. In multivariate analysis of OS, R1 resection and grade III or higher PIICs were identified as independent prognostic factors (Table 3). The 5‐year OS rate was 54% in patients who achieved R0 resection, whereas all patients with R1 resection died within 3 years after surgery (Figure 3A). Patients with grade III or higher PIICs had significantly worse overall survival (Figure 3B). Postoperative recurrence occurred in 31 patients (48%), with the most common site being peritoneal dissemination in 17 patients (27%), followed by lymph node metastasis in 13 (20%) and liver metastasis in nine patients (14%). As shown in Figure 2B, the RFS for the 52 patients with R0 resection showed a 3‐year RFS rate of 55%, 5‐year RFS rate of 45%, and median RFS rate of 55.1 months. Multivariate analysis identified grade III or higher PIICs as an independent prognostic factor for RFS (Table 4). Survival curves related to PIICs (Figure 4) indicated that patients with grade III or higher PIICs had significantly worse survival outcomes.
FIGURE 2.
Kaplan–Meier estimates of survival. (A) Kaplan–Meier estimates of overall survival in the entire cohort. (B) Kaplan–Meier estimates of relapse‐free survival in patients who underwent R0 resection.
TABLE 3.
Multivariate analysis for overall survival.
| Univariate | Multivariate | |||
|---|---|---|---|---|
| HR (95% CI) | p | HR (95% CI) | p | |
| Age (year) | ||||
| ≥ 70 vs. < 70 | 1.362 (0.705–2.631) | 0.358 | — | — |
| Sex | ||||
| Male vs. female | 0.911 (0.451–1.827) | 0.793 | — | — |
| Macroscopic type | ||||
| Types 3, 4 vs. types 0, 1, 2 | 1.858 (0.924–3.735) | 0.078 | — | — |
| Preoperative chemotherapy | ||||
| Performed vs. not performed | 0.360 (0.110–1.178) | 0.091 | — | — |
| Gastrectomy | ||||
| TG vs. DG | 1.408 (0.678–2.924) | 0.359 | — | — |
| Pancreatectomy | ||||
| PD vs. DP | 0.925 (0.446–1.919) | 0.834 | — | — |
| Histological type | ||||
| Undifferentiated vs. differentiated | 2.332 (1.019–5.339) | 0.045 | 2.286 (0.990–5.274) | 0.053 |
| Residual tumor | ||||
| R1 vs. R0 | 5.432 (2.536–11.63) | < 0.001 | 4.858 (2.255–10.46) | < 0.001 |
| Tumor depth | ||||
| pT4b vs. others | 0.934 (0.484–1.801) | 0.838 | — | — |
| Lymph node metastasis | ||||
| pN+ vs. pN− | 1.759 (0.621–4.985) | 0.288 | — | — |
| Postoperative complications (CD) | ||||
| III, IV vs. 0, I, II | 1.852 (0.952–3.600) | 0.069 | — | — |
| PIICs (CD) | ||||
| III, IV vs. 0, I, II | 2.242 (1.133–4.438) | 0.017 | 2.067 (1.034–4.134) | 0.027 |
| Postoperative chemotherapy | ||||
| Performed vs. not performed | 0.562 (0.291–1.083) | 0.081 | — | — |
Note: In multivariate analysis for OS, R1 resection and grade III or higher PIICs were identified as independent prognostic factors.
Abbreviations: CD, Clavien‐Dindo classification; DG, distal gastrectomy; DP, distal pancreatectomy; PD, pancreaticoduodenectomy; PIICs, postoperative intra‐abdominal infectious complications; TG, total gastrectomy.
FIGURE 3.
(A) Kaplan–Meier estimates of overall survival according to residual tumor status. (B) Kaplan–Meier estimates of overall survival according to postoperative intra‐abdominal infectious complications.
TABLE 4.
Multivariate analysis for independent prognostic factors of RFS.
| Univariate | Multivariate | |||
|---|---|---|---|---|
| HR (95% CI) | p | HR (95% CI) | p | |
| Age (year) | ||||
| ≥ 70 vs. < 70 | 1.134 (0.531–2.425) | 0.745 | — | — |
| Sex | ||||
| Male vs. female | 1.094 (0.474–2.523) | 0.636 | — | — |
| Macroscopic type | ||||
| Types 3, 4 vs. types 0, 1, 2 | 1.640 (0.760–3.541) | 0.058 | — | — |
| Preoperative chemotherapy | ||||
| Performed vs. not performed | 0.260 (0.061–0.907) | 0.049 | 0.340 (0.077–1.500) | 0.154 |
| Gastrectomy | ||||
| TG vs. DG | 0.342 (0.157–0.747) | 0.005 | 0.478 (0.213–1.077) | 0.075 |
| Pancreatectomy | ||||
| PD vs. DP | 1.838 (0.839–4.027) | 0.122 | — | — |
| Histological type | ||||
| Undifferentiated vs. differentiated | 2.063 (0.829–5.130) | 0.111 | — | — |
| Tumor depth | ||||
| pT4b vs. others | 0.790 (0.368–1.694) | 0.544 | — | — |
| Lymph node metastasis | ||||
| pN+ vs. pN− | 0.879 (0.354–2.182) | 0.781 | — | — |
| Postoperative complications (CD) | ||||
| III, IV vs. 0, I, II | 2.055 (0.921–4.588) | 0.072 | — | — |
| PIICs (CD) | ||||
| III, IV vs. 0, I, II | 2.622 (1.195–5.755) | 0.013 | 2.354 (1.059–5.233) | 0.036 |
| Postoperative chemotherapy | ||||
| Performed vs. not performed | 0.686 (0.320–1.469) | 0.329 | — | — |
Note: Multivariate analysis identified grade III or higher PIICs as an independent prognostic factor for RFS.
Abbreviations: CD, Clavien–Dindo classification; DG, distal gastrectomy; DP, distal pancreatectomy; PD, pancreaticoduodenectomy; PIICs, postoperative intra‐abdominal infectious complications; TG, total gastrectomy.
FIGURE 4.
Kaplan–Meier estimates of relapse‐free survival according to postoperative intra‐abdominal infectious complications.
4. Discussion
This study aimed to elucidate the efficacy of combined resection of the pancreas along with gastrectomy for advanced gastric cancer by investigating short‐term and long‐term outcomes, as well as prognostic factors. In the short‐term analysis, invasion to the duodenum was selected as an independent risk factor for PIICs. Furthermore, R0 resection and grade III or higher PIICs were shown to be independent prognostic factors for OS. Furthermore, in patients who underwent R0 resection, multivariate analysis identified PIICs as an independent prognostic factor for RFS.
The sensitivity, specificity, and accuracy of preoperative imaging for the intraoperative diagnosis of pancreatic invasion have been previously reported as 60.0%, 93.3%, and 91.6%, respectively, for abdominal CT, and 20.0%, 100.0%, and 95.8%, respectively, for abdominal ultrasonography (USG) [12]. In the present study, of the 51 patients diagnosed with pancreatic invasion by preoperative imaging, 30 patients (59%) were intraoperatively confirmed as positive for pancreatic invasion; the sensitivity, specificity, and accuracy of preoperative imaging diagnosis and intraoperative diagnosis was 33.7%, 99.1%, and 96.6%, indicating a low sensitivity, as in previous reports. This suggests that the accurate diagnosis of pancreatic invasion from gastric cancer using preoperative imaging seems to be difficult. In the present study, of the 89 patients intraoperatively diagnosed with invasion of the pancreas, 39 did not undergo radical resection, of whom 24 (62%) showed positive cytology by intraoperative rapid cytological diagnosis. In addition, of the 47 patients diagnosed as positive for pancreatic invasion during surgery, invasion was histologically confirmed in 27 patents (57%), which is similar to the accuracy of 39% reported previously [13, 14]. Inflammation and associated adhesions have been reported as the causes of intraoperative overdiagnosis [15]. In fact, it is often difficult to distinguish adhesion from invasion during surgery. Improving the accuracy of preoperative diagnosis is considered important to avoid unnecessary combined resections; however, previous reports suggest that accurately assessing pancreatic invasion with current modalities remains challenging. Recently, there have been reports indicating that the diagnostic accuracy for T4a disease has improved through the use of artificial intelligence (AI) [16], and it is expected that future applications of AI will also enhance the diagnostic accuracy for pancreatic invasion.
The incidence of postoperative complications is reportedly significantly higher with combined resection of the pancreas for gastric cancer than with gastric resection alone or resection of other organs such as the colon and spleen [17]. In PD, the incidence of postoperative complications has been reported as approximately 22%–74%, with pancreatic fistula, intra‐abdominal abscess, and postoperative enteritis being the most common, in that order [13, 18]. In contrast, postoperative complications in DP reportedly occur in 21%–75% of patients, with pancreatic fistula, intra‐abdominal abscess, and anastomotic leakage occurring at significantly higher rates [19, 20]. However, factors related to the occurrence of complications have not been analyzed in either surgical procedure. In gastric cancer surgery, PIICs have been reported to not only prolong postoperative hospital stay, but also worsen survival [21]. In the present study, the incidence of complications was comparable to that of previous reports. All patients with surgery‐related complications of CD grade III or higher also experienced PIICs. In addition, the occurrence of PIICs was identified as an independent prognostic factor suggesting that PIICs associates with a poor survival in patients who underwent pancreatectomy. In a previous report comparing the incidence of postoperative complications according to surgical procedure, no difference was found among PD, DP, and segmental resection [5]. There have been no other reports comparing the outcomes of PD and DP. In the present study, no significant difference in the incidence of PIICs was observed between PD and DP. Invasion to the duodenum was the independent risk factor of PIICs. Gastric cancer with invasion to the duodenum reportedly has a higher incidence of lymph node metastasis, such as to the posterior pancreatic lymph nodes and the lymph nodes along the SMV [22]. However, in the present study, no significant association was found between the presence of pathological lymph node metastasis and the occurrence of postoperative complications. Kakeji et al. [23] reported that, in primary gastric cancer with histological invasion of the duodenum, postoperative complications, such as intra‐abdominal abscess, duodenal stump leakage, pancreatitis, and intra‐abdominal hemorrhage, were more frequently observed. They also reported that gastric cancer with invasion of the duodenum was associated with a significantly higher incidence of serosal invasion, vascular factors, and lymph node metastasis, and was recognized as an independent prognostic factor for OS in tumors located in the gastric antrum [24]. This suggested that gastric cancer extending to the duodenum exhibits a stronger tendency for infiltrating surrounding tissues, requiring more extensive dissection and lymph node dissection, which might lead to an increase in complications.
In previous reports evaluating the results of multivisceral resection, R0 resection was shown to be an independent prognostic factor for OS [13, 17]. In the present study, R0 resection was also identified as independent prognostic factor as well as PIICs. Cytology was positive in most patients who underwent R1 resection. Thus, it is recommended to perform intraoperative rapid cytological diagnosis and to avoid combined resection in patients with positive cytology results. Regarding the association between the surgical procedure of pancreatectomy and survival, patients undergoing PD reportedly had significantly worse OS than those undergoing DP or segmental resection [5], inconsistent with the present results. However, in that study, the number of patients who underwent PD was small (n = 9), the breakdown of the 58 patients undergoing DP or wedge resection was not specified, and no information was provided regarding residual tumor status. Therefore, caution is warranted when interpreting these results. In contrast, Saka et al. [18] reported no difference in survival rates between PD and wedge resection in patients who achieved R0 resection. Thus, the difference in survival between PD and DP or wedge resection is unclear. Combined resection of the pancreas, including PD, might be considered if R0 resection is deemed feasible.
There have been no previous reports examining the prognostic factors for RFS in gastric cancer patients who underwent pancreatectomy. In the present study, the occurrence of PIICs was identified as the only independent prognostic factor for RFS. The impact of postoperative complications as a poor prognostic factor has already been reported in both gastric [21] and pancreatic cancers [25]. In addition, postoperative complications can induce inflammation, leading to immune suppression, which in turn promotes tumor progression, leading to poor survival [26]. Therefore, care should be taken to avoid postoperative complications following gastrectomy with pancreatectomy for advanced gastric cancer.
Adjuvant chemotherapy was initiated in less than two‐thirds of patients, and the completion rate was less than half, both of which were low. These findings may be attributed to the invasiveness of combined pancreatectomy and the occurrence of postoperative complications. Previous studies have reported that postoperative complications reduce the completion rate of adjuvant chemotherapy and further worsen survival [27]. In highly invasive procedures such as combined pancreatic resection for gastric cancer, the decrease in postoperative general condition may make it difficult to initiate or complete adjuvant chemotherapy, suggesting that NAC could be an important opportunity for treatment. In the present study, although NAC was performed in 11 patients (17%), it did not help to avoid combined resection or to prolong survival. However, Chan et al. [28] reported that NAC significantly prolonged OS in patients with gastric cancer invading the head of the pancreas or duodenum. Based on these findings, NAC may also represent a potentially effective treatment option for gastric cancer with pancreatic invasion.
In the present study, all patients underwent open surgery. Although minimally invasive surgery (MIS) for locally advanced gastric cancer has been rapidly adopted worldwide, evidence regarding its safety and efficacy in patients requiring combined resection of adjacent organs remains limited. In contrast, several meta‐analyses evaluating MIS for pancreatic cancer have demonstrated that minimally invasive PD offers short‐term and long‐term outcomes equivalent to those of open PD [29]. Furthermore, meta‐analyses of DP have reported that minimally invasive DP achieves superior outcomes to open DP [30]. Based on these findings, the introduction of MIS in gastric cancer requiring pancreatic resection may reduce the incidence of PIICs and contribute to improved survival. Therefore, MIS is expected to be increasingly adopted for gastric cancer patients requiring PD or DP in the future.
This study has several limitations. First, the number of cases was relatively small. Second, this was a retrospective study conducted at a single institution. Treatment strategies for surgery, NAC, and postoperative adjuvant chemotherapy were not standardized, and, consequently, the efficacy of NAC could not be sufficiently evaluated. Third, since all patients underwent open surgery, the role of MIS could not be assessed.
5. Conclusion
Pancreatic resection is considered safe and effective in patients with pancreatic or duodenal invasion from gastric cancer when R0 resection is possible.
Author Contributions
Taku Hattori: conceptualization, methodology, data curation, formal analysis, writing – original draft. Masanori Terashima: conceptualization, methodology, supervision, writing – review and editing, writing – original draft. Yusuke Koseki: methodology, data curation, writing – review and editing. Kenichiro Furukawa: methodology, data curation, writing – review and editing. Keiichi Fujiya: methodology, data curation, writing – review and editing. Yutaka Tanizawa: methodology, data curation, writing – review and editing. Katsuhisa Ohgi: methodology, writing – review and editing, supervision. Teiichi Sugiura: writing – review and editing, supervision, methodology. Etsuro Bando: supervision, writing – review and editing.
Disclosure
The authors have nothing to report.
Ethics Statement
All procedures were conducted in accordance with the ethical standards of the corresponding committees on human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and later versions. All aspects of this study were approved by the institutional review board at Shizuoka Cancer Center (J2023‐64‐2023‐1).
Consent
The need for written, informed consent was waived because of the retrospective nature of the study. The opt‐out method was used for obtaining patient consent.
Conflicts of Interest
M.T. reports receiving personal fees from Taiho Pharmaceutical Co. Ltd., Chugai Pharmaceutical Co. Ltd., Ono Pharmaceutical Co. Ltd., Bristol Myers Squib Japan K.K., Yakult Honsha Co. Ltd., Takeda Pharmaceutical Co. Ltd., Eli Lilly Japan K.K., Pfizer Japan Inc., Daiichi Sankyo Ltd., Johnson and Johnson K.K., Medtronic Japan Co. Ltd., Intuitive Japan Inc., and Olympus Co. Ltd. E.B. reports receiving personal fees from Daiichi Sankyo Ltd., Johnson and Johnson K.K., Medtronic Japan Co. Ltd., Intuitive Japan Inc., Kaken Pharmaceutical, Eizo Co., Terumo Co., and Eisai Co. The other authors declare no conflicts of interest.
Supporting information
Data S1: ags370107‐sup‐0001‐DataS1.xlsx.
Table S1: Surgical factors. The surgical factors are summarized.
Table S2: Pathological factors. The patients' pathological factors are shown.
Table S3: Postoperative complications. Details of the postoperative complications are summarized.
Data S2: ags370107‐sup‐0002‐DataS2.docx.
Figure S1: Flow diagram of patient inclusion in this retrospective, observational study from the patient database and medical record is shown.
Figure S2: (A) The OS curve shows 3‐year OS of 46% and 5‐year OS of 42%, with a median survival time of 34.0 months. (B) The RFS curve for the 52 patients with R0 resection shows a 3‐year RFS rate of 55%, 5‐year RFS rate of 45%, and median RFS of 55.1 months.
Figure S3: (A) The 5‐year OS rate is 54% in patients who achieved R0 resection, whereas all patients with R1 resection die within 3 years after surgery. (B) The 5‐year OS rate is 62.0% in patients with grade II or lower PIICs, compared with 24.8% in those with grade III or higher (p = 0.017).
Figure S4: Survival curves related to PIICs show that patients with grade III or higher postoperative complications have significantly poorer survival outcomes.
Acknowledgments
A part of the content of this manuscript (Tables 2, 3, 4) has been incorporated into a Japanese‐language review article in the Japanese journal Geka, published by Nankodo on June 27, 2025. We have obtained formal approval from the editorial office of Geka for secondary publication of this work in English. We confirm that the English version faithfully reflects the data and interpretations of the primary version. The above review article was published on June 27, 2025. The priority of the primary publication will be respected by maintaining an interval of at least one week between the two publications.
Funding: The authors received no specific funding for this work.
References
- 1. Bray F., Ferlay J., Soerjomataram I., Siegel R. L., Torre L. A., and Jemal A., “Global Cancer Statistics 2018: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” CA: A Cancer Journal for Clinicians 68 (2018): 394–424. [DOI] [PubMed] [Google Scholar]
- 2. Japanese Gastric Cancer Association , “Japanese Gastric Cancer Treatment Guidelines 2021 (6th Edition),” Gastric Cancer 26, no. 1 (2023): 1–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Bra S. S., Seevaratnam R., Cardoso R., et al., “Multivisceral Resection for Gastric Cancer: A Systematic Review,” Gastric Cancer 15 (2012): S100–S107. [DOI] [PubMed] [Google Scholar]
- 4. Makuuchi R., Irino T., Tanizawa Y., Bando E., Kawamura T., and Terashima M., “Pancreaticoduodenectomy for Gastric Cancer,” Journal of Cancer Metastasis and Treatment 4 (2018): 26. [Google Scholar]
- 5. Min J. S., Jin S. H., Park S., Kim S. B., Bang H. Y., and Lee J. I., “Prognosis of Curatively Resected pT4b Gastric Cancer With Respect to Invaded Organ Type,” Annals of Surgical Oncology 19 (2012): 494–501. [DOI] [PubMed] [Google Scholar]
- 6. Chang S. C., Tang C. M., Le P. H., et al., “Impact of Pancreatic Resection on Survival in Locally Advanced Resectable Gastric Cancer,” Cancers (Basel) 13 (2021): 1289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Nunobe S., Hiki N., Ohyama S., Fukunaga T., Seto Y., and Yamaguchi T., “Survival Benefits of Pancreatoduodenectomy for Gastric Cancer: Relationship to the Number of Lymph Node Metastases,” Langenbeck's Archives of Surgery 393 (2008): 157–162. [DOI] [PubMed] [Google Scholar]
- 8. Tran T. B., Worhunsky D. J., Norton J. A., et al., “Multivisceral Resection for Gastric Cancer: Results From the US Gastric Cancer Collaborative,” Annals of Surgical Oncology 22 (2015): S840–S847. [DOI] [PubMed] [Google Scholar]
- 9. Edge A. B. and Compton C. C., “The American Joint Committee on Cancer: The 7th Edition of the AJCC Cancer Staging Manual and the Future of TNM,” Annals of Surgical Oncology 17, no. 6 (2010): 1471–1474. [DOI] [PubMed] [Google Scholar]
- 10. Japanese Gastric Cancer A , “Japanese Classification of Gastric Carcinoma: 3rd English Edition,” Gastric Cancer 14, no. 2 (2011): 101–112. [DOI] [PubMed] [Google Scholar]
- 11. Clavien P. A., Barkun J., de Oliveira M. L., et al., “The Clavien–Dindo Classification of Surgical Complications: Five–Year Experience,” Annals of Surgery 250 (2009): 187–196. [DOI] [PubMed] [Google Scholar]
- 12. Kim J. J., Jung H. C., Song I. S., et al., “Preoperative Evaluation of the Curative Resectability of Gastric Cancer by Abdominal Computed Tomography and Ultrasonography: A Prospective Comparison Study,” Korean Journal of Internal Medicine 12 (1997): 1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Piso P., Bellin T., Aselmann H., Bektas H., Schlitt H. J., and Klempnauer J., “Results of Combined Gastrectomy and Pancreatic Resection in Patients With Advanced Primary Gastric Carcinoma,” Digestive Surgery 19 (2002): 281–285. [DOI] [PubMed] [Google Scholar]
- 14. Kitamura K., Tani N., Koike H., et al., “Combined Resection of the Involved Organs in T4 Gastric Cancer,” Hepato‐Gastroenterology 47 (2000): 1769–1772. [PubMed] [Google Scholar]
- 15. Carboni F., Lepiane P., Santoro R., et al., “Treatment for Isolated Loco–Regional Recurrence of Gastric Adenocarcinoma: Does Surgery Play a Role?,” World Journal of Gastroenterology 11 (2005): 7014–7017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Shang H., Feng T., Han D., et al., “Deep Learning and Radiomics for Gastric Cancer Serosal Invasion: Automated Segmentation and Multi‐Machine Learning From Two Centers,” Journal of Cancer Research and Clinical Oncology 151 (2025): 60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Mita K., Ito H., Katsube T., et al., “Prognostic Factors Affecting Survival After Multivisceral Resection in Patients With Clinical T4b Gastric Cancer,” Journal of Gastrointestinal Surgery 21 (2017): 1993–1999. [DOI] [PubMed] [Google Scholar]
- 18. Saka M., Mudan S. S., Katai H., Sano T., Sasako M., and Maruyama K., “Pancreaticoduodenectomy for Advanced Gastric Cancer,” Gastric Cancer 8 (2005): 1–5. [DOI] [PubMed] [Google Scholar]
- 19. Otsuji E., Yamaguchi T., Sawai K., Okamoto K., and Takahashi T., “Total Gastrectomy With Simultaneous Pancreaticosplenectomy or Splenectomy in Patients With Advanced Gastric Carcinoma,” British Journal of Cancer 79 (1999): 1789–1793. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Kasakura Y., Fujii M., Mochizuki F., Kochi M., and Kaiga T., “Is There a Benefit of Pancreaticosplenectomy With Gastrectomy for Advanced Gastric Cancer?,” American Journal of Surgery 179 (2000): 237–242. [DOI] [PubMed] [Google Scholar]
- 21. Tokunaga M., Tanizawa Y., Bando E., Kawamura T., and Terashima M., “Poor Survival Rate in Patients With Postoperative Intra–Abdominal Infectious Complications Following Curative Gastrectomy for Gastric Cancer,” Annals of Surgical Oncology 20 (2013): 1575–1583. [DOI] [PubMed] [Google Scholar]
- 22. Tokunaga M., Ohyama S., Hiki N., Fukunaga T., Inoue H., and Yamada K., “Therapeutic Value of Lymph Node Dissection in Advanced Gastric Cancer With Macroscopic Duodenum Invasion: Is the Posterior Pancreatic Head Lymph Node Dissection Beneficial?,” Annals of Surgical Oncology 16 (2009): 1241–1246. [DOI] [PubMed] [Google Scholar]
- 23. Kakeji Y., Korenaga D., Baba H., et al., “Surgical Treatment of Patients With Gastric Carcinoma and Duodenal Invasion,” Journal of Surgical Oncology 59 (1995): 215–219. [DOI] [PubMed] [Google Scholar]
- 24. Kakeji Y., Tsujitani S., Baba H., et al., “Clinicopathologic Features and Prognostic Significance of Duodenal Invasion in Patients With Distal Gastric Carcinoma,” Cancer 15 (1991): 380–384. [DOI] [PubMed] [Google Scholar]
- 25. Watanabe Y., Nishihara K., Matsumoto S., Okayama T., Abe Y., and Nakano T., “Effect of Postoperative Major Complications on Prognosis After Pancreatectomy for Pancreatic Cancer: A Retrospective Review,” Surgery Today 47 (2017): 555–567. [DOI] [PubMed] [Google Scholar]
- 26. Goldfarb Y., Sorski L., Benish M., Levi B., Melamed R., and Ben–Eliyahu S., “Improving Postoperative Immune Status and Resistance to Cancer Metastasis: A Combined Perioperative Approach of Immunostimulation and Prevention of Excessive Surgical Stress Responses,” Annals of Surgery 253 (2011): 798–810. [DOI] [PubMed] [Google Scholar]
- 27. Tsujimoto H., Kouzu K., Sugasawa H., et al., “Impact of Postoperative Infectious Complications on Adjuvant Chemotherapy Administration After Gastrectomy for Advanced Gastric Cancer,” Japanese Journal of Clinical Oncology 3 (2021): 379–386. [DOI] [PubMed] [Google Scholar]
- 28. Chan W. H., Cheow P. C., Chung A. Y., Ong H. S., Koong H. N., and Wong W. K., “Pancreaticoduodenectomy for Locally Advanced Stomach Cancer: Preliminary Results,” ANZ Journal of Surgery 78 (2008): 767–770. [DOI] [PubMed] [Google Scholar]
- 29. Ricci C., D'Ambra Y., Alberici L., et al., “Minimally Invasive Pancreatoduodenectomy: A Comprehensive Systematic Review and Metanalysis of Randomized Controlled Clinical Trials,” Annals of Surgical Oncology 32 (2025): 3614–3622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Ramos J. V. O., Fernandes J. V. A., Ferreira G. H. O., et al., “Minimally Invasive Versus Open Distal Pancreatectomy for Pancreatic Tumors: Updated Meta‐Analysis and Meta‐Regression,” Pancreatology 25 (2025): 569–579. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1: ags370107‐sup‐0001‐DataS1.xlsx.
Table S1: Surgical factors. The surgical factors are summarized.
Table S2: Pathological factors. The patients' pathological factors are shown.
Table S3: Postoperative complications. Details of the postoperative complications are summarized.
Data S2: ags370107‐sup‐0002‐DataS2.docx.
Figure S1: Flow diagram of patient inclusion in this retrospective, observational study from the patient database and medical record is shown.
Figure S2: (A) The OS curve shows 3‐year OS of 46% and 5‐year OS of 42%, with a median survival time of 34.0 months. (B) The RFS curve for the 52 patients with R0 resection shows a 3‐year RFS rate of 55%, 5‐year RFS rate of 45%, and median RFS of 55.1 months.
Figure S3: (A) The 5‐year OS rate is 54% in patients who achieved R0 resection, whereas all patients with R1 resection die within 3 years after surgery. (B) The 5‐year OS rate is 62.0% in patients with grade II or lower PIICs, compared with 24.8% in those with grade III or higher (p = 0.017).
Figure S4: Survival curves related to PIICs show that patients with grade III or higher postoperative complications have significantly poorer survival outcomes.