Introduction

Pancreatic ductal adenocarcinoma (PDAC) is one of the most fatal malignancies [1, 2]. The number of PDAC patients is expected to increase to 88,000 in 2030 from 43,000 in 2010, with 63,000 deaths in 2030 compared to 36,888 in 2010 in the USA [1]. The distal PDAC usually only presents symptoms during terminal stages [3]. The current only chance of cure for a localized distal PDAC is a surgical resection with negative surgical site margins [4]. Radical antegrade modular pancreatosplenectomy (RAMPS), first reported by Strasberg et al. in 2003 [5], has been an attractive radical procedure for distal PDAC. The concept of this procedure is designed to increase the rate of negative posterior retroperitoneal margin, characterized by en bloc retroperitoneal fat resection (including the left adrenal gland, if necessary) compared with conventional distal pancreatectomy (DP) [6, 7]. With the development of the minimally invasive surgery, laparoscopic RAMPS (Lap-RAMPS) has been performed in high-volume centers recently, and several technical reports on Lap-RAMPS that ensure safety and curability have been published [8,9,10]. While the usefulness of Lap-RAMPS compared to open RAMPS (Open-RAMPS) has been advocated by several retrospective studies, the clinical applicability of the procedure still needs to be further discussed [11,12,13], particularly by evaluating the clinical outcomes of Lap-RAMPS compared to the current gold standard treatment for resectable PDAC. Knowing the evidence of adjuvant chemotherapy in improving recurrence-free survival (RFS) and overall survival (OS) after resection [14,15,16], the current best treatment is local control with surgery followed by systemic treatment with adjuvant chemotherapy. Therefore, the benefit of aggressive RAMPS can be proven by high R0 resection rate, a low local recurrence rate, and a high induction rate of adjuvant chemotherapy [17].

The aim of this study is to evaluate the efficacy of Lap-RAMPS compared to Open-RAMPS, while comparing the results to the modern gold standards in treating resectable PDAC.

Methods

Patients

This is a retrospective cohort study performed on consecutive patients who underwent RAMPS with curative intent for the distal anatomically resectable PDAC at Cancer Institutional Hospital of the Japanese Foundation for Cancer Research between January 2016 and March 2021. The indication for Lap-RAMPS was resectable distal PDAC. The exclusion criteria were as follows: (1) patients with distant metastasis, (2) concomitant resection of other organ(s) including celiac axis, portal vein, and colon (except for the left adrenal gland), and (3) initial resectability with borderline resectable or locally advanced PDAC [18], which were basically indication of Open-RAMPS or DP with celiac axis resection. The choice of Lap- or Open-RAMPS was based on the surgeon’s skill and preference, but in recent years, Lap-RAMPS has been increasingly selected for the eligible cases. Tumors were evaluated by multidetector computed tomography (CT), contrast-enhanced magnetic resonance imaging (MRI), endoscopic ultrasonography, and/or positron emission tomography CT. The study was approved by the Ethics Committee of Cancer Institute Hospital, Japanese Foundation for Cancer Research, prior to collecting identifiable patient information and analysis (2021-GB 013).

Operation technique

The technical details of Lap-RAMPS and Open-RAMPS have been described previously [7, 10]. In brief, in both Lap- and Open-RAMPS, the root of the splenic artery is dissected and ligated under the anterocranial view (we named “anterocranial splenic artery-first approach”) early in the operation to prevent venous congestion in the distal pancreas and spleen [7, 10]. Lap-RAMPS was done by several surgeons who have experienced Open-RAMPS and laparoscopic surgery including conventional distal pancreatectomy, cholecystectomy, hepatectomy, and gastrectomy. According to the temporal distribution between Lap- and Open-RAMPS, there was no clear learning curve formed in the current series (Supplement figure 1). Either anterior or posterior RAMPS was planned at the multidisciplinary conference based on the location of the tumor, so the left adrenal gland is resected en bloc (which case is defined as posterior RAMPS), if necessary, in order to achieve posterior retroperitoneal tumor-free margins (Fig. 1). After resection, drain placement and conservative drain management were done as previously described [10].

Fig. 1
figure 1

The view after laparoscopic RAMPS. A Posterior RAMPS. B Anterior RAMPS. RAMPS: radical antegrade modular pancreatosplenectomy; CHA: common hepatic artery; SPA: splenic artery; SMA: superior mesenteric artery; SMV: superior mesenteric vein

Full size image

Patient demographics and operative and postoperative outcomes

We reviewed the electronic medical records. Patient demographics including age, gender, body mass index (BMI), American Society of Anesthesiologists (ASA) score, and presence of diabetes; tumor characteristics including location (body/tail), size (mm), and the serum level of CEA and CA19-9 level at diagnosis and before surgery; and the operative outcomes including the procedure method (Lap- or Open-RAMPS and anterior or posterior RAMPS), operative time, intraoperative blood loss, and hospital stay were recorded. Intraoperative blood loss was calculated in accordance with the conventional method, i.e., as the sum of intraoperative suction fluids (after subtracting the amount of irrigation fluids) and increase in operative gauze weight [19]. Most of the patients remained hospitalized until all drains were removed and were rarely discharged home with drains. The severity of complications was defined according to the Clavien–Dindo classification system [20]. POPF, delayed gastric emptying (DGE), surgical site infection (SSI), and intra-abdominal hemorrhage were defined according to the International Study Group of Pancreatic Surgery [21,22,23]. Mortality and morbidity were defined as a complication occurring during the hospital stay. The patients were followed up every 3 months after surgery and performed CT and/or MRI every 6 months routinely.

Neoadjuvant and adjuvant chemotherapy

Before December 2018, resectable PDAC was treated by upfront surgery. Thereafter, all patients except for those with clinical stage I received two courses of neoadjuvant gemcitabine (GEM) and tegafur/gimeracil/oteracil (S-1) chemotherapy. After resection, S-1 was administrated as the first choice of adjuvant chemotherapy based on the criteria of the randomized controlled trial (adequate oral intake, Eastern Cooperative Oncology Group performance status of 0 or 1, adequate bone marrow/liver/kidney function) [15, 24]. Four cycles of S-1 were administered, with each cycle having a daily dose of 80 mg/m2 orally for 28 days, followed by 14 days of rest. GEM was used for patients who could not tolerate S-1 (for example, allergy and nausea). The administration of GEM was repeated every 4 weeks for up to six cycles [15].

Pathologic evaluation

The surgical specimens were assessed according to the Union for International Cancer Control (UICC) 8th edition TNM system [25]. It has been our standard to slice the specimen transversely and evaluate/report microscopic superior, inferior, and posterior margin addition to the pancreas stump margin status. We inked the posterior retroperitoneal margin to ensure accurate histologic evaluation. The posterior retroperitoneal margin was inked before fixation, the fixed specimen was sliced transversely with a 5 mm thickness, and its tangential margin was evaluated. R0 resection was defined by the absence of microscopic cancer cells in all margins, regardless of the distance between the tumor and the closest margin, and also intraoperative abdominal cytopathology negative.

Statistical analysis

Continuous data are presented as a median (range), and categorical data are expressed as a number (percentage). All analyses were performed using StatFlex 6.0 software (Artech Co., Osaka, Japan). The Mann-Whitney U test and the chi-square test were used for comparisons, as appropriate. RFS and OS rates were determined from the time of resection to the time of detection of first recurrence and death or last follow-up, respectively. Survival curves were constructed using the Kaplan–Meier method and compared with the log-rank test. A Cox proportional hazards model was used to assess the prognostic significance of Lap-RAMPS. For Cox model, significant covariates included induction of neoadjuvant chemotherapy, the level of CA19-9, pathological lymph node positive, and induction of adjuvant chemotherapy. Results were considered significant when p < 0.05.

Results

Patients’ characteristics

178 patients underwent RAMPS for PDAC during this period (January 2016 to March 2021). 60 patients were excluded in the analysis: 44 underwent RAMPS with another organ(s) resection and 13 and 3 with initially borderline resectable/unresectable PDAC. Finally, 118 patients were included in the analysis. 43 patients underwent Lap-RAMPS and 75 patients underwent Open-RAMPS for the localized distal PDAC. The patients’ characteristics are shown in Table 1. The Lap-RAMPS group received neoadjuvant chemotherapy more frequently (Lap-RAMPS vs. Open-RAMPS, 53.5% vs. 17.3%, p < 0.001).

Table 1 Patients’ characteristics data
Full size table

Operative and postoperative outcomes

The details of operative and postoperative outcomes are shown in Table 2. The median operative time was longer (364 min [250–547] vs. 303 min [182–579] p < 0.001), and the median blood loss was lower (125 mL [5–500] vs. 390 mL [40–1340], p < 0.001) in the Lap-RAMPS group. The anterior RAMPS procedure was performed more frequent in the Lap-RAMPS (anterior/posterior 26/17 vs. 24/51, p = 0.003). The postoperative stay was shorter (17 days vs. 21 days, p = 0.018) in the Lap-RAMPS group. The details of postoperative complications are shown in Table 2. The incidence of POPF, DGE, SSI, and hemorrhage was not different between the two groups. There was no significant difference in complication of Clavien–Dindo grade 3a or higher (p = 0.862). There was no postoperative mortality in both groups, and there was no open conversion in the Lap-RAMPS group.

Table 2 Operative and postoperative outcome
Full size table

Histopathological outcomes

Histopathological outcomes are shown in Table 3. Pathological tumor size in the Lap-RAMPS group was smaller than that in the Open-RAMPS group (24 mm [4–78] vs. 30mm [2–70], p = 0.013). Lymph node metastases were observed in 18 patients (41.9%) of the Lap-RAMPS group and 40 patients (53.3%) of the Open-RAMPS group (p = 0.230). There was no significant difference in UICC stage between two groups. The achievement rate of R0 resection was 100.0% in the Lap-RAMPS and 90.7% in the Open-RAMPS (p = 0.039). In the Open-RAMPS group, R1 resection was observed in seven patients: positive posterior retroperitoneal margin in three patients, positive proximal pancreatic stump in two patients, positive nerve plexus stump in one patient, and intraoperative abdominal cytopathology positive in one patient.

Table 3 Histopathological outcome and adjuvant chemotherapy data
Full size table

Adjuvant chemotherapy

Patients not eligible for adjuvant chemotherapy were as follows: 4 elderly patients (Lap: 2, Open: 2), 3 pathological stage 0 (Lap: 1, Open: 2), 4 patients with early recurrence (Lap: 1, Open: 3), and 1 patient treated for another malignancy (Lap: 0, Open: 1). Of the candidates for adjuvant chemotherapy excluding those, 39 patients (100.0%) received adjuvant chemotherapy in the Lap-RAMPS group and 60 out of 67 patients (89.6%) received adjuvant chemotherapy in the Open-RAMPS group (p = 0.037) (Fig. 2). The reasons for not receiving adjuvant chemotherapy in the Open-RAMPS group were as follows. In the Open-RAMPS group: an impaired performance status after surgery in 5 patients and refusal of adjuvant chemotherapy in 2 patients. The intervals between initiation of adjuvant chemotherapy and resection were shorter in the Lap-RAMPS group compared to the Open-RAMPS group (54.0 days vs. 59.5 days, p = 0.026).

Fig. 2
figure 2

The induction rate of postoperative adjuvant chemotherapy between the Lap-RAMPS and the Open-RAMPS group

Full size image

Overall survival outcome

The median follow-up month of all patients was 23.1 months (8.6–42.2) in the LAP-RAMPS group and 31.5 months (7.1–71.6) in the Open-RAMPS group (p = 0.0002). Figure 3 shows Kaplan–Meier curves for OS, RFS, local recurrence, and distant metastasis among all patients. Three-year OS, RFS, local recurrence rates, and distant metastasis rate after resection in the Lap-RAMPS group were 69.8%, 58.1%, 8.7%, and 33.6%, respectively, compared to 71.1%, 58.3%, 10.0%, and 35.8% in the Open-RAMPS group, respectively (RFS: p = 0.433, OS: p = 0.996, local recurrence rates: p = 0.976, and distant metastasis rates: p = 0.391). Among the patients with distant metastases, 8 patients (18.6%) had liver metastases, 1 patient (2.3%) had peritoneal metastases, and 3 patients (7.0%) had para-aortic lymph node metastases in the Lap-RAMPS group; and 7 patients (9.3%) had liver metastases, 7 patients (9.3%) had peritoneal metastases, and 5 patients (6.7%) had para-aortic lymph node metastases in the Open-RAMPS group. Lap-RAMPS was not associated with poor prognosis compared to Open-RAMPS even after adjusting for induction of neoadjuvant chemotherapy, the level of CA19-9, pathological lymph node positive, and induction of adjuvant chemotherapy (the hazard ratio 1.185, 95% CI = 0.480–2.927; p = 0.713) (Table 4).

Fig. 3
figure 3

Kaplan–Meier analysis of overall survival (A), recurrence-free survival (B), local recurrence (C), and distant metastasis (D) between the Lap-RAMPS group and the Open-RAMPS group

Full size image
Table 4 Results of Cox multivariate analyses of open/laparoscopic and other prognostic factors after surgical treatment
Full size table

Discussion

Although the RAMPS continues to be the attractive and reasonable procedure for adequate local control of resectable distal PDAC with sufficient posterior retroperitoneal margin, its usefulness needs to be reevaluated since it is more technically demanding than the conventional DP and its short- or long-term outcomes have not been discussed extensively in literature [5, 6, 26, 27]. PDAC is a systemic disease with a recurrence rate after curative resection of 70% to 80%, with 70% to 80% of which being distant metastases. Since adjuvant chemotherapy plays an important role in controlling recurrence [28, 29], the rate of induction of adjuvant chemotherapy is an important factor in evaluating the outcome of radical PDAC resection, which should be accompanied by pathologic evaluation of the resected specimen and the local recurrence rate after resection. We have previously reported on the technical aspects of Open-RAMPS [7] and later on Lap-RAMPS (“anterocranial splenic artery-first approach”) as the minimally invasive surgery progressed [10]. The present study demonstrated that these efforts have resulted in a high induction rate of adjuvant chemotherapy (Lap-RAMPS: 100.0%, Open-RAMPS: 89.6%), a high R0 rate (Lap-RAMPS: 100.0%, Open-RAMPS: 90.7%), and a low 3-year local recurrence rate (Lap-RAMPS: 8.7%, Open-RAMPS: 10.0%) and shows the certain usefulness of RAMPS in short- and long-term outcomes. Lap-RAMPS was not associated with poor prognosis compared to Open-RAMPS even after adjusting for other prognostic factors (hazard ratio 1.185; p = 0.713). On the other hand, Lap-RAMPS was able to introduce adjuvant chemotherapy earlier postoperatively at a higher induction rate than Open-RAMPS (100.0% vs. 89.6%, p = 0.037), suggesting that Lap-RAMPS may be a useful procedure that combines the benefits of minimally invasive surgery with the radicality of RAMPS. It is practically difficult to plan a randomized control trial comparing conventional DP and RAMPS since the decision to perform conventional DP or RAMPS is dependent on location and institution, just as the reason why conventional DP for PDAC is rarely performed in our institution [7]. The results of the LEOPARD-2 study, which compared the time to functional recovery between laparoscopic and open pancreatoduodenectomy and failed to demonstrate the safety of laparoscopic pancreatoduodenectomy, suggest that the quality control of RCTs of minimally invasive surgery can be challenging [30]. Considering this fact, the result of this study should be thoroughly taken into account when implementing evidence-based practice for resectable PDAC.

The current results showed that the rate of R0 in the Lap-RAMPS group was higher than that of the Open-RAMPS group (Lap vs. Open: 100% vs. 90.9%, p = 0.039). A major reason could be that the patients having tumor near the splenic artery or the superior mesenteric vein are often considered ineligible for Lap-RAMPS. Neoadjuvant chemo(radio)therapy has been shown to reduce preoperative tumor size and the level of CA19-9 and to increase the rate of R0 achievement in postoperative outcomes [31, 32] and to improve OS compared with upfront surgery in the patients with borderline resectable pancreatic cancer [33]. The difference in the rate of neoadjuvant chemotherapy could also play in the difference in R0 rate due to the overlapping period during which Lap-RAMPS and neoadjuvant treatment were introduced. The preoperative tumor size was smaller in the Lap-RAMPS group, and as a result, the anterior RAMPS was selected more often (Lap vs. Open: 60.4% vs. 32.0%, p < 0.001).

Caudal approach during Lap-RAMPS that includes opening of the mesotransverse colon to secure a surgical margin in advanced cases with transverse mesocolon invasion or tumor adjacency to the superior mesenteric artery has been advocated as well. However, there are few literatures that show short- and long-term outcomes of this approach [34,35,36]. We have recently reported the anterocranial splenic artery-first approach contributed to reduction of blood loss during dissection following our open-RAMPS concept. The current results demonstrated that the anterocranial splenic artery-first approach can achieve high rates of negative posterior retroperitoneal margin, low rate of local recurrence, and low volume of intraoperative blood loss [10]. It can be explained that these good results were achieved by initially ligating the splenic artery blood inflow to prevent congestion, thereby ensuring a dry field of vision, and by unifying the left-sided dissection of the superior mesenteric artery and en bloc dissection of the retroperitoneum procedures.

Induction of postoperative adjuvant chemotherapy is indispensable in maximizing OS in the treatment of PDAC. Currently, S-1 and FOLFIRINOX are the best adjuvant chemotherapy with comparable efficacy in terms of hazard ratio for OS compared to GEM, and oral S-1 continues to be the first choice in clinical practice in Japan due to its convenient mode of administration along with minimal side effects [14, 15, 37]. Regardless of which drug is used, the currently available evidence suggests that adjuvant chemotherapy following radical surgery is the best treatment for OS extension. In this study, 39 patients (100.0%) in the Lap-RAMPS group and 60 patients (89.6%) in the Open-RAMPS group were induced adjuvant chemotherapy. Five patients in the Open-RAMPS group could not initiate adjuvant chemotherapy due to their impaired performance status after surgery, whereas all patients in the Lap-RAMPS underwent adjuvant treatment. Although the present results showed somehow short-term advantage of minimally invasive Lap-RAMPS with less body wall destruction and intestinal dryness, the positive impact on long-term outcomes (RFS, OS) could not be demonstrated.

There are some limitations in this study. First, patient selection bias could not be excluded due to the retrospective nature of the study. Second, the patient backgrounds between two groups are different since the induction of neoadjuvant chemotherapy and Lap-RAMPS at our hospital started later in the study period. However, the current results also showed that Lap-RAMPS can be safely performed after neoadjuvant treatment, and moreover, Lap-RAMPS was not associated with poor prognosis compared to Open-RAMPS even after adjusting for significant covariates including neoadjuvant chemotherapy (the hazard ratio 1.185; p = 0.713). These are considered important results that could justify the continued adoption of this relatively new technique. Further prospective multicenter trials are essential to demonstrate the true efficacy and non-inferiority of Lap-RAMPS to Open-RAMPS. Third, our discharge criteria were not strictly defined. This might affect the results of postoperative hospital stay.

In conclusion, the safety and efficacy of Lap-RAMPS were comparable to those of Open-RAMPS in terms of achieving local control and adjuvant chemotherapy induction, which are deemed as the current treatment gold standards for resectable distal PDAC. Higher induction rate of adjuvant chemotherapy in Lap-RAMPS might suggest that minimally invasive approach for a radical resection may be promising in improving the care and outcomes of patients.