Introduction

Pancreatic ductal adenocarcinoma (PDAC), arising from the pancreatic ductal epithelium, is one of the most aggressive malignant digestive system tumors, with a 5-year survival rate of less than 10%1. Surgical resection is taken as the only curative therapy for PDAC, and the 5-year survival rate after radical resection is approximately 20%2,3. Unfortunately, more than half of the patients present with distant metastases and are considered an absolute contraindication for surgical resection4,5. Additionally, even among PDAC patients evaluated as resectable, there are 14%-30% of patients have abdominal metastasis (such as liver metastasis, mesenteric metastasis, or peritoneal implant metastasis) during open surgery or laparoscopic exploration without any positive signs on preoperative imaging examination, which is also known as occult metastasis6. The most common site of metastasis is the liver, which accounts for approximately 80% of PDAC cases with distant metastases7.

Palliative chemotherapy is the standard therapeutic strategy for the treatment of unresectable metastatic PDAC (UR-M PDAC). Gemcitabine plus nab-paclitaxel (GA) and FOLFIRINOX (fluorouracil, irinotecan, oxaliplatin, and calcium folinate) have become the first-line medications for UR-M PDAC, which could prolong overall survival (OS)8,9. Historically, resection of recurrent or metastatic pancreatic cancer seemed to be inconceivable. However, in recent years, pancreatectomy combined with synchronous or metachronous metastasectomy has been considered for selected patients with UR-M PDAC. Synchronous resection of primary tumors and metastases in PDAC may provide survival benefits without compromising safety and quality of life in a highly selected group of patients10,11,12.

In this context, our aim is to evaluate the clinical value of surgical resection of PDAC with synchronous metastases and to confirm the prognostic factors of PDAC patients with simultaneous liver metastasis.

Materials and methods

The systematic review and meta-analysis were done according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement13. This systematic review and meta-analysis was also conducted following the recommendations of the AMSTAR (Assessing the methodological quality of systematic reviews) guidelines14.

Search strategies

The PubMed, Embase, and Cochrane Library databases were searched for eligible articles up to September 1, 2024. The search was conducted using medical subject headings (MeSH) in combination with free-text words. The following search headings were used: “pancreatic cancer”, “liver”, “metastasis”, “resection”, and “hepatectomy”, and we used “AND”,“OR”,“NOT”for combination of these headings to avoid missing and wrong articles. The search strategy is described in the Supplementary Materials. Our protocol was registered on the International Platform of Registered Systematic Review and Meta-analysis Protocols database (INPLASY2024100064).

Inclusion and exclusion criteria

All studies included in the meta-analysis were selected according to the following inclusion criteria: (1) all patients were diagnosed with PDAC with synchronous liver metastases; (2) primary tumor and metastases were simultaneously resected with or without preoperative chemotherapy; (3) survival data can be collected in the literature; (4) Newcastle–Ottawa Quality Assessment Scale (NOS) score ≥ 6. The exclusion criteria were as follows: (1) neuroendocrine tumor and other pathological types; (2) metachronous liver metastasis; (3) studies with incomplete survival data; (4) abstracts, case reports, editorials, letters, systematic reviews, and comments; (5) studies that enrolled the overlapped or same population; and (6) duplicate studies.

Data extraction

Two investigators (PCZ and ZHW) independently extracted the necessary data from the included studies, and any disagreements were resolved by discussion until a consensus was reached. The following data were extracted from each study: first author, publication year, country, study design, number of patients, follow-up duration, tumor site, number of liver metastases, median survival time, and overall survival.

Quality assessment

The Newcastle–Ottawa Quality Assessment Scale (NOS) was used to evaluate the quality of the included studies. NOS consists of three aspects: selection, comparability, and outcomes. Studies with a score of six or higher were considered high-quality studies15. This work was also performed independently by two investigators (PCZ and ZHW). (Supplementary Table S1).

Statistical analysis

Meta-analysis was conducted using Stata 14.0 software, following the Meta-analysis of Observational Studies in Epidemiology (MOOSE) guidelines. The heterogeneity of the pooled effect was assessed using Cochran’s Q test and the Higgins I2 statistic. Q test p value < 0.1 or I2 > 50% was considered significant heterogeneity, and a random-effect model was applied to estimate the pooled HR. While heterogeneity was not significant (Q test p value > 0.1 or I2 < 50%), a fixed-effect model was used. Sensitivity analysis was applied to reduce and explain heterogeneity among the studies. Furthermore, publication bias was visually checked through a funnel plot and then quantitatively analyzed using Begg’s and Egger’s tests. All statistical tests were two-sided, and a p value less than 0.05 were defined as statistically significant.

Results

Study selection

We searched the PubMed, Embase, and Cochrane Library databases, and 6560 articles were initially retrieved. After removing 363 duplicates, 6197 articles remained. After screening the titles and abstracts, 6127 articles were excluded for being irrelevant topics, reviews or meta-analysis, conference abstracts, letters, case reports, case series, or comments. Finally, 14 articles met our inclusion criteria, and 1107 patients had UR-M PDAC with synchronous hepatic metastasis16,17,18,19,20,21,22,23,24,25,26,27,28,29. Among the 1107 patients, 553 patients who underwent synchronous resection of the primary tumor and metastases were included in this meta-analysis, while 554 patients received palliative chemotherapy and constituted the non-surgical group. The detailed selection process was illustrated in Fig. 1.

Fig. 1
figure 1

PRISMA flow diagram of eligible studies selection.

Full size image

Clinical characteristic of enrolled studies

All included studies were retrospective and were mainly published in the past ten years. Among the 553 patients who underwent surgery, 185 underwent preoperative chemotherapy, while the others underwent upfront surgery. Palliative chemotherapy was administered to the rest 554 patients. The scores of study quality assessed by the NOS ranged from 6 to 8. The main characteristics of the surgical and the non-surgical populations were presented in Tables 1 and 2.

Table 1 The main characteristics of surgical group.
Full size table
Table 2 The main characteristics of non-surgical group.
Full size table

Clinical benefits of curative surgery in PDAC patients with liver metastases

Seven studies investigated patients who underwent synchronous resection of primary tumors and liver metastases16,18,22,23,26,27,29. There were 553 patients in surgical group, whereas the non-surgical group included 554patients. The pooled HR was 0.496 with 95% CI: 0.414–0.594 (I2 = 0, p = 0.531). It revealed that simultaneous resection could prolong the survival of PDAC patients with hepatic metastases. (Fig. 2).

Fig. 2
figure 2

Forest plot of meta-analysis on OS between surgical and non-surgical groups. (OS overall survival, HR hazard ratio, CI confidence interval).

Full size image

Clinical values of preoperative chemotherapy in PDAC patients with liver metastases

Four studies reported the survival data of PDAC patients with liver metastases underwent preoperative chemotherapy16,17,18,19. A total of 107 patients who underwent synchronous resection of the primary tumor and liver metastases with preoperative chemotherapy were enrolled, while 72 patients underwent upfront surgery without chemotherapy. The characteristics of patients who underwent preoperative chemotherapy were reported in Table 3. According to the pooled results (HR = 0.579, 95% CI: 0.473–0.709, I2 = 0, p = 0.418), patients performed preoperative chemotherapy had better OS. Forest plot was illustrated in Fig. 3.

Table 3 The main characteristics of preoperative group.
Full size table
Fig. 3
figure 3

Forest plot of meta-analysis on OS between preoperative chemotherapy and no preoperative chemotherapy groups. (OS overall survival, HR hazard ratio, CI confidence interval).

Full size image

Subgroup analyses and meta regression analyses

To explore and explain the heterogeneity, meta regression analyses were performed for the number of patients in the surgical group, the use of propensity score matching (PSM), and the administration of preoperative chemotherapy, yielding p-values of 0.825, 0.848, and 0.785, respectively. In addition, we performed subgroup analyses according to the aforementioned three factors, setting a cut-off value of 20 for the number of patients in the surgical group; the results were illustrated in Fig. 4A–C.

Fig. 4
figure 4

Forest plot of subgroup analyses based on (A) the number of patients in surgical group; (B) the use of PSM; (C) the administration of preoperative chemotherapy. (OS overall survival, HR hazard ratio, CI confidence interval, PSM propensity score matching).

Full size image

Sensitivity analysis and publication bias

Sensitivity analysis was conducted to assess the effect of individual studies on the pooled HR of OS in studies that included synchronous surgery, and the results suggested that omitting any individual studies had no significant effect on the pooled HR (Fig. 5). Furthermore, publication bias was investigated, and there was no obvious asymmetry in the funnel plot upon visual inspection (Fig. 6), then Begg’s and Egger’s tests yielded p-values of 0.035 and 0.072, respectively, which also indicated that there was some extent of publication bias among the included studies.

Fig. 5
figure 5

Sensitivity analysis of the surgical group. (OS overall survival).

Full size image
Fig. 6
figure 6

Funnel plot illustrating publication bias test result.

Full size image

Prognostic factors in PDAC patients with liver metastasis

Clinicopathological variables were analyzed to identify the risk factors associated with prognosis and survival in PDAC patients with liver metastasis. Finally, preoperative chemotherapy, resection of the primary tumor, R0 resection margin status at pancreatic resection, and postoperative chemotherapy could be considered as independent prognostic factors in PDAC patients with liver metastasis,and the results were presented in Table 4.

Table 4 Meta-analysis of the prognostic factors in PDAC patients with liver metastasis.
Full size table

Discussion

Synchronous resection the primary and metastatic tumors has gained wide acceptance as a curative treatment in colorectal cancer patients with liver metastases30,31. The role of synchronous surgery for hepatic metastases from pancreatic neuroendocrine tumors on long-term outcomes has also been well-documented32,33,34. Nonetheless, radical surgery was used to be heresy in most instances for PDAC patients with liver metastases, and palliative chemotherapy seemed to be the only option. With advances in systemic treatment and surgical techniques, synchronous resection has been introduced in recent years for PDAC patients with liver metastasis. Some studies reported that radical surgery was valuable and could prolong survival35, whereas some articles were discouraged because it did not improve survival36.

To clarify the controversy regarding synchronous resection, we collected relevant studies and conducted a meta-analysis. In our study, 14 articles with 1107 PDAC patients with synchronous hepatic metastases were enrolled16,17,18,19,20,21,22,23,24,25,26,27,28,29. Our pooled analysis showed that simultaneous resection could significantly improve overall survival of PDAC patients with hepatic metastases compared to those who did not undergo surgery (HR = 0.496, 95% CI: 0.414–0.594, I2 = 0, p = 0.531). Furthermore, according to our further study, PDAC patients with hepatic metastases who received preoperative chemotherapy had better survival than those who underwent upfront surgery (HR = 0.579, 95% CI: 0.473–0.709, I2 = 0, p = 0.418). We also confirmed several risk factors associated with the prognosis of liver metastases. These included preoperative chemotherapy, resection of the primary tumor, and resection margin status et al.

Chemotherapy, especially preoperative chemotherapy, played a crucial role in the management of PDAC patients with liver metastases. For metastatic patients, GA, FOLFIRINOX, and modified FOLFIRINOX (m-FOLFIRINOX) were the standard regimens. The results of a phase III clinical trial showed that the median OS of the FOLFIRINOX regimen for the treatment of UR-M PDAC was 11.1 months, which was significantly longer than that of the gemcitabine single-agent group at 6.8 months (p < 0.01). The progression-free survival (PFS) was also significantly prolonged (6.4 months vs. 3.3 months, p < 0.001)37,38,39. A phase II clinical trial also investigated m-FOLFIRINOX, which revealed that m-FOLFIRINOX prolonged survival, reduced the incidence of side effects, and improved chemotherapy tolerance40. Recently, NALIFIROX (liposome irinotecan, 5-fluorouracil, leucovorin, and oxaliplatin) has been introduced as a novel reference regimen for the treatment of metastatic patients. The OS and progression-free survival (PFS) of the NALIFIROX group were 11.1 months and 7.4 months, respectively, which were superior to the 9.2 months and 5.6 months in the GA group (P values were 0.036 and < 0.001, respectively)41. In our study, preoperative chemotherapy ( pooled HR: 0.307, 95% CI: 0.219–0.433) was an independent risk factor associated with survival. According to Crippa et al., the median OS of 11 metastatic patients with partial or complete response after preoperative chemotherapy was obviously longer than that of the remaining 116 patients who did not undergo surgery (median OS: 39 vs 11 months, p < 0.0001)27. Preoperative chemotherapy was helpful in assessing tumor biological behavior, screening potential patients for surgery, and eliminating or controlling liver metastases. Therefore, compared to upfront surgery (surgery-first), conversion surgery (chemotherapy-first) had significantly better median survival in PDAC patients with synchronous hepatic metastases.

The optimal duration of preoperative chemotherapy is still debated. In the current studies, the median duration of preoperative chemotherapy for PDAC patients with liver metastases was 6.2 to 12 months. Ushida et al. believed that a duration of preoperative chemotherapy ≥ 8 months was an independent protective factor and could improve the prognosis of patients42. Conversely, in accordance with Hank’s study, a time interval from initial diagnosis to surgery of < 5 months or ≥ 9 months was an independent risk factor and was associated with poor prognosis43. The criteria for conversion surgery are also debatable. Ushida et al. proposed that the indications for conversion surgery for UR-M PDAC were serum CA19-9 level within the normal range, imaging examination results showing tumor shrinkage, progression-free survival of 8 months or more, and disappearance of metastatic lesions42. In addition, Hank et al. defined a relatively lax indication: the primary lesion was evaluated as stable or improved according to the Response Evaluation Criteria in Solid Tumors (RECIST) standard, the metastatic lesion was partially responded to or better, and the tumor marker level was reduced. More clinical trials were required to determine the best surgical indication43.

The role of postoperative chemotherapy in patients with metastasis was also unclear. Nagal et al. showed that postoperative chemotherapy had little impact on the median OS after surgical resection of UR-M PDAC with liver metastases17. The main reasons why patients did not undergo postoperative chemotherapy were poor physical condition, severe complications, older age, and patient unwillingness44. In our study, we enrolled three articles, and the pooled results (HR = 0.430, 95% CI: 0.283–0.653) indicated that postoperative chemotherapy was associated with better OS19,20,23.

This meta-analysis has several limitations. Firstly, all the included studies were retrospective, so the overall level of clinical evidence was relatively low, and potential selection bias was inevitable. Secondly, the survival data of the studies were estimated from the survival curve and survival time, which might not be very accurate. Thirdly, the sample size was relatively small. Therefore, caution should be exercised in the pooled results of the meta-analysis.

Conclusion

In summary, our meta-analysis revealed that preoperative chemotherapy combined with simultaneous resection of both primary and metastatic tumors was able to prolong the survival of PDAC patients with synchronous liver metastases and that these patients could also benefit from postoperative chemotherapy. Moreover, owing to the existing limitations, large, multicenter RCTs are urgently needed.