This PSM cohort study revealed that LRAMPS did not offer oncological advantages over LDP for left-sided PDAC within the specified selection criteria. The LRAMPS procedure did not yield a higher number of positive lymph nodes or a more extensive margin-negative rate. On the contrary, it resulted in longer operation time, increased intraoperative bleeding and postoperative abdominal drainage volume. It also led to a prolonged recovery period for gastrointestinal function and an increase in postoperative bowel movement frequency. These factors can impact the regimen selection and completion of postoperative adjuvant chemotherapy, consequently compromising the potential benefits of LRAMPS in achieving better local control. The study suggests that LDP is sufficient to meet the oncologic requirements for selected left-sided PDAC.

Due to the resection plane of the RAMPS procedure being located behind Gerota’s fascia, as opposed to the traditional distal pancreatosplecnectomy resection plane which is located behind the pancreatic capsule, it significantly increases the number of N1 lymph nodes harvested and improves the R0 rate of the posterior tangential margin of the specime [5,6,7]. The study conducted by Grossman et al. included 78 cases of RAMPS, with 93.6% of patients achieving negative tangential margins. The average number of lymph nodes retrieved was 20, with an average of 1.4 positive lymph nodes [26]. Meta-analyses further support the superiority of RAMPS over standard procedures, demonstrating improved R0 resection rates and lymph node retrieval rates [27, 28]. Recent studies have shown that LRAMPS is technically feasible and provides comparable long-term oncological outcomes to open RMAPS [10, 29]. Additionally, LRAMPS demonstrates similar rates of postoperative complications to open RMAPS but also exhibits lower intraoperative blood loss and faster postoperative recovery of eating time [9]. Given the oncological equivalence of LRAMPS and its advantages in short-term outcomes, several studies have examined the feasibility of LRAMPS as the standard treatment for resectable left-sided PDAC [13,14,15].

However, in previous studies focusing on the oncology and safety of RAMPS, most of the tumors analyzed were at T2-stage or above [8, 12, 13]. For tumors within T2 stage and confined to the pancreatic parenchyma, the necessity of expanding the resection extent and dissecting more N1 lymph nodes requires further exploration. Referring to the Korean LRAMPS Yonsei criteria [12], we proposed our selection criteria and conducted this PSM cohort analysis. Although the LRAMPS group retrieved more lymph nodes compared to the LDP group (16.0 vs. 13.0, P = 0.008), this did not result in a higher detection rate of positive lymph nodes (35.2% vs. 33.3%, P = 0.839). Recent studies on anti-tumor immunity of solid tumors have found that uninvolved lymph nodes are enriched with progenitor exhausted CD8 + T cells, while the anti-tumor immune hallmarks were impaired in metastatic lymph nodes and exhibited immunosuppressive cellular niches [30]. Therefore, removing too many negative lymph nodes, particularly for immunologically “cold” tumors like PDAC, may potentially worsen the immune microenvironment. Moreover, LRAMPS did not increase the detection rate of positive retroperitoneal margins in PDAC cases meeting the selection criteria (16.7% vs. 13.0%, P = 0.588). But, it did result in longer operation time (240.5 vs. 219.0 min, P = 0.020) and increased intraoperative blood loss (200 vs. 150 mL, P = 0.001). However, given that most left-sided PDACs are discovered at a stage greater than T2 and invade the fascial layer behind the pancreas in clinical practice, RAMPS remains the primary approach for achieving oncological resection in left-sided PDAC cases beyond our proposed selection criteria.

A study conducted in the United States analyzed data from the National Surgical Quality Improvement Project database, comparing 236 distal pancreatosplecnectomy patients with 117 RAMPS patients. The study found that RAMPS had comparable surgical safety to distal pancreatosplecnectomy, as it did not increase postoperative complications such as overall POPF, clinically relevant POPF, and Clavien–Dindo grade 3 or higher complications [14]. However, it is important to note that in this study, the RAMPS group had a low proportion of minimally invasive surgery, accounting for only 6.8% of cases. Currently, there is limited research comparing LRAMPS and LDP [31]. In our study, we found no significant difference in the length of postoperative hospital stay between the two groups, with a median hospital stay of 10 days after surgery. It is understandable that LRAMPS, with its larger resection extent, would result in a higher postoperative TAD compared to LDP (85.0 vs. 40.0 mL, P = 0.001), which in turn leads to dilution of the DFA (2802.0 vs. 5129.5 U/L, P = 0.025). Moreover, although not reaching statistical significance, the incidence of chyle leak in the LRAMPS group appears to be higher than that in the LDP group (14.8% vs. 5.6%). We believe that with an increased sample size, statistical significance may be achieved. However, these differences did not result in a significant difference in the occurrence of POPF between the two groups. While one patient in the LRAMPS group experienced early mild hemorrhage that required reoperation for hemostasis and was classified as having a grade C POPF and a Clavien–Dindo grade 3 complication, there were no statistically significant differences between the two groups in terms of delayed gastric emptying, and no deaths occurred within 90 days of surgery in either group.

However, it should be noted that LRAMPS did prolong the time required for patients to recover to a semi-liquid diet after surgery. Additionally, due to the removal of more nerve plexus during LRAMPS [16], there was an increase in daily bowel movement frequency after surgery, which could potentially impact the patients' quality of life. The impact of LRAMPS surgery on patients' physical recovery can also be observed in the timing of initiation, regimen selection, and completion of adjuvant chemotherapy. It appears that patients in the LRAMPS group started adjuvant chemotherapy later, although this difference did not reach statistical significance. However, in the LDP group, a higher proportion of patients received intravenous chemotherapy (98.0% vs. 80.7%) and had a higher completion rate of 6-month intravenous chemotherapy (75.9% vs. 51.9%). Therefore, when evaluating the potential oncological benefits of LRAMPS surgery, it is crucial to fully consider the impact of this technique on overall treatment outcomes.

To date, there is no evidence to suggest that RAMPS surgery improves postoperative RFS or overall survival. However, it may be effective in controlling local tumor recurrence. A study conducted in Japan analyzed the clinical data of 174 patients who underwent open distal pancreatosplecnectomy and open RAMPS between 2009 and 2016. Following PSM analysis, the study found no significant difference in 3-year RFS and overall survival between the two groups. However, the open RAMPS group did demonstrate a lower 3-year local recurrence rate compared to the open distal pancreatosplecnectomy group (10% vs. 34%, hazard ratio = 0.275, P = 0.024) [8]. Our own study on LRAMPS compared to LDP found no RFS benefit. Among the PSM cohort, a total of 39 patients experienced tumor recurrence, with 21 patients in the LRAMPS group and 18 patients in the LDP group. We further examined the recurrence patterns between the two groups. Consistent with the Japanese study, LRAMPS, with its wider resection range, showed potential for improved local control. This was evident in the lower rates of early recurrence (61.9% vs. 66.7%) and local recurrence (23.8% vs. 38.9%) in the LRAMPS group compared to the LDP group. However, these differences were not statistically significant. Nevertheless, LRAMPS had an impact on postoperative physical recovery, leading to a lower completion rate of adjuvant intravenous chemotherapy. This significantly compromised the potential benefits of better local control. Overall, our study found that, for selected early-stage left-sided PDAC, LRAMPS did not provide oncological advantage over LDP.

Our study is the first to discuss the necessity of LRAMPS surgery for selected early-stage left-sided PDAC, but it still has some limitations. Firstly, although we utilized PSM to minimize intergroup bias, we were unable to achieve the effects of a randomized controlled clinical trial. Secondly, although all surgeons involved in this study had surpassed their learning curve for LRAMPS, it is important to note that this study was conducted at a high-volume center in China, which may impact the generalizability of the results to other centers. Finally, the follow-up duration in this study was relatively short, and longer follow-up is necessary to observe the overall survival impact of LRAMPS on left-sided PDAC within the specified selection criteria. However, we believe that our study offers valuable insights into the application of LRAMPS in early-stage resectable left-sided PDAC.