Introduction

Pancreatic cancer is associated with a dismal prognosis and a 5-year survival rate of only 9% [1]. At the rate its incidence is increasing, it is anticipated that by 2030, it will be the second leading cause of cancer-related death [2]. Despite advances in chemotherapy and chemoradiotherapy, pancreatic resection is essential to cure pancreatic cancer. However, the multi-organ resection that pancreatic resection requires, together with glucose tolerance insufficiency and severe complications such as postoperative pancreatic fistula, impose a heavy burden on patients. In the 1970s, the mortality rate after pancreaticoduodenectomy (PD) was approximately 20%, but progress in surgical procedures and devices, combined with better postoperative management, have decreased the rate to around 3% [3,4,5].

During the late twentieth century, minimally invasive surgery (MIS); namely, laparoscopic and robotic surgery, encompassed digestive surgery, with laparoscopic surgery in the 1980s and robotic surgery in the 1990s [6]. Although the initial indication for MIS was benign conditions such as appendicitis or cholelithiasis, it expanded gradually to include gastrointestinal cancers [7, 8]. The several advantages of MIS, such as reduced postoperative pain, fewer wound complications, and early postoperative recovery, made MIS an attractive treatment option for cancer. According to a recent national survey by the Japan Society for Endoscopic Surgery, more than 60% of colorectal cancers are now treated by MIS [9]. Even open pancreatic surgery is a challenging procedure for surgeons because of the retroperitoneal location, anatomical complexity, and proximity to major vessels, but now minimally invasive pancreatic resection (MIPR) is being performed in clinical practice. MIPR for benign or low-grade malignant tumors has several advantages over open pancreatic resection in perioperative outcomes [10]. According to a worldwide survey on MIPR, 90% of participating surgeons thought that MIPR had overall advantages for patients [11]. Thus, the application of MIPR to pancreatic cancer treatment may benefit patients with this refractory disease. To promote a better understanding of MIPR for pancreatic cancer, we review its history and current status, and discuss its future perspectives.

History of minimally invasive pancreatic resection for pancreatic cancer

Following the successful application of laparoscopy to several hepato-pancreato-biliary procedures such as cholecystectomy, choledochotomy, and liver resection, laparoscopic pancreatic resection was introduced [6]. In 1994, Gagner et al. reported their first laparoscopic PD (LPD) for chronic pancreatitis, performed in 1992 [12]. Then, in 1996, Cuschieri et al. described the first laparoscopic distal pancreatectomy (LDP), also performed for chronic pancreatitis [13]. The first description of MIPR for pancreatic cancer was in a case series of laparoscopic pancreatic resections (LPRs) reported by Gagner et al. in 1997 [14]. They reported 23 cases of LPR, four of which were LPD for pancreatic adenocarcinoma and one of which was LDP for pancreatic cystadenocarcinoma. The first description of robotic pancreatic resection for pancreatic cancer was in a case series reported by Giulianotti et al., who described three cases of robotic PD and three cases of robotic DP for pancreatic cancer [15]. The first case series of minimally invasive PD (MIPD) for pancreatic cancer, including oncological outcomes such as prognosis, was reported by Palanivelu et al. [16] in 2007, with 40 cases of LPD for periampullary malignancy, including nine for pancreatic adenocarcinoma and four for pancreatic cystadenocarcinoma. In the same year, Fernández-Cruz et al. [17] reported the first case series of minimally invasive DP (MIDP) for pancreatic cancer with oncological outcomes, including 13 cases of LDP performed for pancreatic cancer. They described the application of radical antegrade modular pancreatosplenectomy (RAMPS), which was proposed by Strasberg et al., to laparoscopic procedures for en bloc resection of left-sided pancreatic cancer in open surgery [18].

Current status of minimally invasive pancreatic resection for pancreatic cancer

The favorable perioperative outcomes of MIPR such as less blood loss and shorter hospital stay have resulted in MIPR becoming an accepted treatment option for pancreatic cancer in clinical practice. According to studies in the National Cancer Database (NCDB) of the United States, 27.9% (506/1807) of patients who underwent DP and 14.9% (1191/7967) of those who underwent PD for pancreatic cancer, were treated by minimally invasive approaches between 2010 and 2012 [19, 20]. In the National Comprehensive Cancer Network (NCCN) guidelines for pancreatic adenocarcinoma, MIS is described as equal to open surgery as a treatment option for resectable disease [21]. The Japanese Clinical Practice Guidelines for Pancreatic Cancer recommend LDP for pancreatic cancer except if there is multiple organ invasion or if combined vascular resection is required [22]. However, the Japanese guidelines state that LPD for pancreatic cancer is not recommended in clinical practice and should be performed in clinical studies. This is because LPD for cancer was not permitted by Japanese health insurance at the time of publication of the guidelines.

Several studies comparing MIPR and open pancreatic resection (OPR) for pancreatic cancer have been published. Tables 1, 2, 3 and 4 summarize these comparative studies between MIPR and OPR for pancreatic cancer (adenocarcinoma) [19, 20, 23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45]. All the studies were retrospective and ten of them collected data from nationwide databases including the NCDB [19, 20, 24, 28, 29, 34, 35, 37, 38, 42].

Table 1 Perioperative outcomes of minimally invasive vs. open distal pancreatectomy for pancreatic cancer (adenocarcinoma)
Full size table
Table 2 Oncological outcomes of minimally invasive vs. open distal pancreatectomy for pancreatic cancer (adenocarcinoma)
Full size table
Table 3 Perioperative outcomes of minimally invasive vs. open pancreaticoduodenectomy for pancreatic cancer (adenocarcinoma)
Full size table
Table 4 Oncological outcomes of minimally invasive vs. open pancreaticoduodenectomy for pancreatic cancer (adenocarcinoma)
Full size table

Table 1 compares the perioperative outcomes between MIDP and open distal pancreatectomy (ODP) [19, 23, 27, 28, 30, 32,33,34,35,36, 38,39,40, 42,43,44]. In most of these studies, operation time, postoperative complications, and mortality were comparable between MIDP and ODP, but MIDP was associated with less blood loss and a shorter hospital stay. Meta-analyses of comparative studies comparing MIDP and ODP for benign and malignant conditions also revealed less blood loss and a shorter hospital stay [10, 46, 47]. Although most studies showed comparable postoperative complication rates, two studies using nationwide databases revealed fewer postoperative complications after MIDP. Sulpice et al. [42] analyzed data from healthcare databases in France and found a significantly lower incidence of major abdominal complications after LDP. The study using data from the American College of Surgeons-National Surgical Quality Improvement Program identified a lower incidence of overall postoperative complications as well as pneumonia, surgical site infection, and sepsis [34]. The meta-analyses also showed reduced postoperative complications after MIDP [10, 46,47,48]. Thus, MIDP for pancreatic cancer may be associated with a lower incidence of postoperative complications. Table 2 compares the oncological outcomes of MIDP and ODP [19, 23, 27, 28, 30, 32, 33, 35, 36, 38,39,40, 42,43,44]. R0 resection rates, number of harvested lymph nodes, adjuvant chemotherapy, and overall survival were mostly comparable. Some large cohort studies revealed significantly higher R0 resection rates with MIDP [19, 28, 38, 43]. However, as bulky tumors or tumors close to major vessels tended to require open surgery rather than minimally invasive surgery in these studies, selection bias may have influenced the outcome. Meta-analyses of comparative studies between MIDP and ODP for pancreatic cancer also revealed comparable R0 resection rates, numbers of harvested lymph nodes, adjuvant chemotherapy, and overall survival except for one meta-analysis that showed a smaller number of harvested lymph nodes with MIDP [49,50,51].

Table 3 compares the perioperative outcomes of MIPD and open PD (OPD) [20, 24,25,26, 29, 31, 37, 41, 45]. Most studies showed similar postoperative complications and mortality after MIPD and OPD, but MIPD was associated with longer operation time, less blood loss, and a shorter hospital stay. MIPD was also associated with a longer operation time, less blood loss, and a shorter hospital stay in meta-analyses of studies comparing MIPD and OPD for benign and malignant periampullary disease [52, 53]. Although mortality was comparable for MIPD and OPD, a low hospital volume was associated with increased mortality in MIPD [37]. International Evidence-based Guidelines on MIPR recommend that MIPD should be performed at high-volume centers [54].

Table 4 compares the oncological outcomes of MIPD and OPD [20, 24,25,26, 29, 31, 37, 41, 45]. In most studies, MIPD and OPD showed comparable R0 resection rates, adjuvant chemotherapy, and overall survival, but MIPD achieved larger numbers of harvested lymph nodes. A meta-analysis of randomized controlled trials and high-quality nonrandomized studies comparing MIPD and OPD also showed a higher number of harvested lymph nodes in MIPD [53]. Magnified high-resolution images and meticulous manipulation of minimally invasive surgery may facilitate lymph node dissection.

RAMPS is often used in MIDP for pancreatic cancer [17, 32, 55, 56]. Medial-to-lateral dissection of the retroperitoneum in RAMPS may allow for a better laparoscopic view than the lateral-to-medial approach of conventional pancreatosplenectomy. Some surgeons use the ligament of Treitz approach to expose a dissection plane anterior to the left renal vein [57, 58]. Pancreatic cancer often requires combined vascular resection. Although some investigators have described MIPR with major vessel resection (portal vein resection or celiac axis resection) [59, 60], evidence of its safety and efficacy is limited. Therefore, it should be performed in high volume centers by experienced surgeons for the purpose of prospective investigations.

Future perspectives

Although MIPR for pancreatic cancer appears to be oncologically comparable to OPR and may have some better perioperative outcomes, the current evidence is based on retrospective studies. Further analyses according to prospective investigations including randomized controlled trials are necessary. Evidence of the usefulness of neoadjuvant therapy for resectable or borderline resectable pancreatic cancer is accumulating and the number of cases of conversion surgery for primary unresectable pancreatic cancer are increasing. However, the feasibility of MIPR after neoadjuvant therapy or as conversion surgery has not been established and requires further investigation.

MIPR for pancreatic cancer is still in development. Standardization of surgical procedures and widespread educational programs for MIPR may improve outcomes, as demonstrated by a nationwide training program in MIDP in the Netherlands, which reduced blood loss, conversion, margin-positive resection, and the length of hospital stay [61]. Further advances in imaging technology and surgical devices will also improve the precision of surgical procedures. For example, the application of augmented reality during MIPR may allow surgeons to locate tumors or vessels accurately despite the lack of tactile sensation [62]. Postoperative pancreatic fistula (POPF) is the most concerning complication of pancreatic surgery. A randomized controlled study suggested that stapler reinforcement may inhibit the development of POPF in distal pancreatectomy [63]. Thus, we await the development of methods or devices to overcome POPF.

The improvements in prognosis after pancreatic resection for pancreatic cancer resulting from better multidisciplinary treatments are unfortunately accompanied by an increasing number of cases of metachronous cancer in the remnant pancreas [64]. Several authors suggest that resection may improve the prognosis of patients with remnant pancreatic cancer [65,66,67]. If the initial pancreatic resection is performed by MIS, less adhesion is expected. One of the predictors of difficulty in laparoscopic repeat liver resection for recurrent hepatocellular carcinoma is the history of an open approach for the initial liver resection [68]. Hence, initial pancreatic resection according to MIS may facilitate secondary surgery for remnant pancreatic cancer.

Conclusion

MIPR for pancreatic cancer is being adopted in clinical practice more slowly than MIS for other abdominal malignancies. Current evidence suggests that it has some perioperative outcome advantages, with further advantages evolving through progress in techniques and devices. Whether MIPR benefits patient survival needs to be verified prospectively. Centralization, standardization, and education are future issues of MIPR for pancreatic cancer.