Safety and oncologic efficacy of robotic surgery compared to open surgery after neoadjuvant therapy for pancreatic cancer

Abstract

Background

Neoadjuvant therapy (NAT) has become the recommended treatment approach for patients with borderline resectable or locally advanced pancreatic cancer. However, pancreatic peritumoral fibrosis and proximity of the tumor to major vessels make surgery after NAT challenging. The feasibility and safety of robotic approaches remain unclear.

Methods

A retrospective analysis was conducted on consecutive pancreatic cancer patients who underwent either open pancreatic surgeries (OPS) or robotic pancreatic surgeries (RPS) following NAT between February 2017 and April 2025. Outcomes, including R0 resection rates, postoperative complication rates, disease-free survival (DFS), and overall survival (OS), were compared between the two groups.

Results

The rate of combined vascular resection and reconstruction was significantly higher in OPS group compared to RPS group following NAT. A total of 43 patients underwent surgical resection which did not require venous segmental resection and reconstruction, including 23 RPS and 20 OPS, were finally included in the analysis. The two groups were well balanced in baseline characteristics, except for BMI. In the RPS group, one case (4.3%) was converted into laparotomy. The RPS group was associated with longer operative time (340.0 min vs. 254.0 min, p = 0.015). The number of harvested lymph nodes was significantly higher in the RPS group compared to the OPS group(25 vs. 18, p = 0.049). There were no significant differences in R0 resection rates between the groups. The overall complication rate was significantly lower in the RPS group (21.7% vs. 55.0%, p = 0.024), while the rates of severe complications, clinically relevant pancreatic fistula, and postoperative hemorrhage did not differ between groups. Postoperative hospital stay was slightly shorter in the RPS group, though not statistically significant. There were no significant differences in DFS or OS.

Conclusion

In carefully selected patients with pancreatic cancer following neoadjuvant therapy who did not require vascular segmental resection and reconstruction, a robotic approach appeared technically feasible and was associated with favorable short-term postoperative outcomes and a greater lymph node yield. These findings should be considered exploratory, and further prospective studies are warranted to better define the role of robotic surgery in this setting.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12893-026-03942-9.

Introduction

In recent years, neoadjuvant therapy has been an important treatment modality for borderline resectable or locally advanced pancreatic cancer, as it can help explore cancer’s biological behavior, alleviate the tumor-vessel relationship, and improve the R0 resection rate [1]. Moreover, neoadjuvant therapy has gradually increased in patients with resectable pancreatic cancer [2, 3]. However, neoadjuvant therapy can lead to fibrosis of the tissues surrounding the pancreas and blur the tissue planes, and the tumor itself often being closely associated with adjacent vessels, thus increasing the complexity of the surgery.

Pancreatic cancer resections after neoadjuvant therapy are predominantly performed as open surgeries. However, with rapid advancements in laparoscopic and robotic surgical systems in recent years, minimally invasive surgery has undergone substantial transformation. Vascular resection under vascular clamping has become safe and feasible, and precise dissection and multi-dimensional cutting and suturing techniques are now well-established for minimally invasive surgery [4, 5]. Robotic surgery system, compared to traditional laparoscopic system, offers unique advantages—such as a three-dimensional 10-fold magnified view, seven degrees of freedom for flexible wrist movements, tremor filtering, and superior ergonomic design—significantly enhancing surgical precision and quality, earning them the reputation of being a minimally invasive approach that more closely resembles open surgery [6]. Surgeons can fully leverage the platform’s strengths to perform complex procedures that are challenging with conventional laparoscopic systems.

Currently, most studies focus on upfront resectable pancreatic cancer, benign or low-grade malignant tumors, demonstrating that robotic surgery can shorten postoperative hospital stays, reduce intraoperative blood loss, and achieve oncological outcomes comparable to open surgery [7–9]. The safety and feasibility of minimally invasive surgery in cases after neoadjuvant therapy were unclear. Both the Miami international evidence-based guidelines on minimally invasive pancreas resection and the Brescia international validated European guidelines on minimally invasive pancreatic surgery indicate that there is insufficient evidence to support the use of minimally invasive procedures for pancreatic cancer resection after neoadjuvant therapy and further studies are warranted [10, 11]. Therefore, it is worth exploring the safety of robotic resection for pancreatic cancer following neoadjuvant therapy.

This study aims to compare the efficacy of robotic surgery and open surgery in patients with pancreatic cancer following neoadjuvant therapy, exploring the feasibility and safety of robotic procedures in complex cases of pancreatic cancer.

Methods

Study design and data collection

From August 2016 to April 2025, patients with pancreatic cancer who underwent either open surgery or robotic-assisted surgery after receiving neoadjuvant therapy were included. All surgeries were performed by a single surgical team. The surgical team began performing robotic distal pancreatectomy (RDP) in 2019 and robotic pancreatoduodenectomy (RPD) in 2020. In 2021, the surgical team performed the first robotic pancreatic cancer resection after neoadjuvant chemotherapy, having already completed the learning curve for robotic pancreatic surgery, while also possessing extensive experience with open pancreatic surgeries [12, 13]. This retrospective study was approved by the institutional review board (IRB) of Peking Union Medical College Hospital (Approval Number: K7514), and was conducted according to the principles of the Declaration of Helsinki (75th Helsinki Finland, October 2024), and followed the guidelines of the Strengthening The Reporting Of Cohort Studies in Surgery (STROCSS) [14].

For patients classified as borderline or locally advanced resectability on imaging, or with CA19-9 levels ≥ 500 U/mL, neoadjuvant therapy was recommended in our institution. Patients were recommended to receive the following regimens: AS, AG, GS, mFOLFIRINOX, or NALIRIFOX. The number of cycles was decided based on the tumor response and changes in CA19-9 levels. After neoadjuvant therapy, if there was no tumor progression in relation to the vasculature and CA19-9 levels had decreased by more than 50%, and the involved vasculature was deemed amenable to resection or reconstruction. The decision to proceed with surgery was made after multidisciplinary team discussion, and surgical exploration was performed 4–6 weeks after completion of neoadjuvant therapy. The choice of surgical procedure was determined by the tumor location, including pancreaticoduodenectomy, distal pancreatectomy, or total pancreatectomy.

For patients who did not require vascular reconstruction, which was judged based on preoperative imaging showing that the tumor invaded less than 180° of the PV/SMV circumference, with shorter-segment abutment (< 2 cm), allowing for complete tumor removal through wedge resection, the choice between robotic pancreatic surgery (RPS) and open pancreatic surgery (OPS) was made by the patients after detailed informed consent, which included an explanation of the benefits, risks, and medical costs. For cases where imaging showed tumor invasion exceeding 180° or an encasement length greater than 2 cm, requiring segmental resection and vascular reconstruction, OPS was preferred.

Patients were included in the study if they met the inclusion and exclusion criteria as follows. Inclusion criteria were: I) Age over 18 years. II) Received neoadjuvant chemotherapy. III) Postoperative pathology confirmed pancreatic cancer. IV) Underwent either open or robotic surgery. Exclusion criteria were: I) Intraoperative discovery of unresectable tumor or presence of distant metastasis. II) ASA score ≥ 4. All data were retrospectively collected from prospectively maintained databases from the hospital’s electronic patient records system, including demographic data, operation data, oncological data, and postoperative outcomes. R0 resection was defined as microscopic radical resection of at least 1 mm between the tumor and the resection margin. Pathological findings were independently confirmed by two pathologists. Venous resection types were defined by the International Study Group on Pancreatic Surgery (ISGPS) [15]. The Clavien-Dindo classification was used to assess the severity of complications [16]. Clinically relevant postoperative pancreatic fistula (CR-POPF), delayed gastric emptying (DGE), and postoperative pancreatectomy hemorrhage (PPH) were defined by ISGPS [17–19]. Complications graded as Clavien-Dindo grade 3 or higher were considered as severe complication. Prognostic data included disease-free survival (DFS) and overall survival (OS). DFS was defined as the time from surgery to the first recurrence or metastasis. OS was defined as the time from surgery to death or the last follow-up. Patients were followed up every six months through outpatient visits or telephone calls until death or loss to follow-up. The follow-up period for this study was censored on April 1, 2026.

Operation procedure

Depending on the extent of vascular involvement, venous wedge resection, partial venous excision using a patch, segmental resection with veno-venous anastomosis or arterial resection was performed. In all robotic cases, the da Vinci^® Si or Xi Surgical System (Intuitive Surgical Inc., Sunnyvale, CA, USA) was used with a 5-port trocar setup. The detailed steps for robotic pancreaticoduodenectomy are described in a previously published article [12], while the steps for robotic distal pancreatectomy are outlined in another publication [20]. For venous resection with or without construction, laparoscopic bulldog clamps were used for venous clamping before dissecting the tumor from the PV/SMV confluence, followed by continuous suturing with 5 − 0 PROLENE sutures. For arterial resections, such as those involving the celiac axis, an Endo-GIA stapler was used for transection.

Statistical analysis

This study employed an intention-to-treat analysis, where patients who underwent conversion from robotic surgery to laparotomy were still classified in the robotic surgery group. Since RPS was only performed in cases that did not require venous segmental resection and reconstruction, to ensure comparability between OPS and RPS, we compared the outcomes of them in well-selected cases where vascular reconstruction was not required. Statistical analyses were performed using IBM SPSS Statistics software (version 26.0; SPSS Inc.). Quantitative data are presented as mean ± standard deviation (SD) or median (interquartile range, IQR), as appropriate. Continuous variables were compared between groups using the independent samples t-test for normally distributed data and the Mann–Whitney U test for non-normally distributed data. Categorical data are presented as frequencies and percentages, with comparisons made using Pearson’s chi-squared test, or Fisher’s exact test. Survival analysis was conducted using the Kaplan-Meier method, and comparisons between groups were made using the log-rank test and Cox regression model. A p-value < 0.05 was considered statistically significant.

Results

Baseline characteristics of patients

A total of 56 pancreatic cancer patients after neoadjuvant chemotherapy received surgery during the period from August 2016 to April 2025. Of them, 33 (58.9%) underwent open pancreatic surgery (ORS) and 23 (41.1%) underwent robotic pancreatic surgery (RPS) (Supplementary Table 1). In the OPS group, 12 patients (36.4%) underwent venous reconstruction, and 1 patient (3.0%) underwent arterial reconstruction, while no vascular reconstructions were performed in the RPS group due to selection criteria (p < 0.001) (Supplementary Table 2). All OPS patients who underwent vascular reconstruction experienced postoperative complications, with 5 cases (15.2%) developing severe complications and 1 patient (3.0%) succumbing to PPH on the first day after surgery.

To ensure comparability between OPS and RPS, we compared the outcomes in cases where vascular segmental resection and reconstruction was not required (Fig. 1). A total of 43 patients were finally included in the study. Of these, 23 (53.5 %) had tumors located in the head of the pancreas, 19 (44.2%) had tumors located in the body or tail of the pancreas, and 1 patient (2.3%) had lesions in both the head and the body/tail. Twenty-seven patients (62.8%) received the AG (albumin-bound paclitaxel and gemcitabine) or AS (albumin-bound paclitaxel and S-1) regimen, while 8 (18.6%) received the mFOLFIRINOX (modified treatment with fluorouracil, leucovorin, oxaliplatin, and irinotecan) regimen, and 5 (11.6%) received the GS (gemcitabine and S-1) regimen. After neoadjuvant therapy, 21 patients (48.8%) experienced weight gain, while 17 patients (39.5%) experienced weight loss. Twenty patients (46.5%) underwent open surgery, and 23 patients (53.5%) underwent robotic surgery. Baseline characteristics between the open and robotic surgery groups were well balanced, except for BMI (Table 1).

Fig. 1.

Flowchart of the study

Table 1.

Baseline characteristics of patients not requiring vascular reconstruction

	OPS (N = 20)	RPS (N = 23)	P value
Age, median (IQR), years	56.0 (49.5, 65.5)	63.0 (56.0, 68.0)	0.093
Male, n (%)	11 (55.0)	10 (43.5)	0.451
ASA, n (%)			0.747
I	11 (55.0)	10 (43.5)
II	6 (30.0)	9 (39.1)
III	3 (15.0)	4 (17.4)
BMI, median (IQR), kg/m2	22.37 (20.22, 23.93)	24.02 (22.49, 26.29)	0.030
Hypertension, n (%)	6 (30.0)	5 (21.7)	0.536
Cardiovascular disease, n (%)	2 (10.0)	1 (4.3)	0.590†
Diabetes, n (%)	4 (20.0)	9 (39.1)	0.173
History of abdominal surgery, n (%)	5 (25.0)	3 (14.3)	0.440†
History of smoking, n (%)	6 (30.0)	5 (21.7)	0.536
History of drinking, n (%)	6 (30.0)	5 (21.7)	0.536
Neoadjuvant regime, n (%)			0.287
AG	6 (30.0)	13 (56.5)
AS	3 (15.0)	5 (21.7)
GS	4 (20.0)	1 (4.3)
mFOLFIRINOX	5 (25.0)	3 (13.0)
Nalirifox	1 (5.0)	1 (4.3)
G	1 (5.0)	0 (0.0)
Tumor location, n (%)			0.538
Head	10 (50.0)	13 (56.5)
Body and tail	10 (50.0)	9 (39.1)
Both Head and body/tail	0 (0.0)	1 (4.3)
Resectability before NAT, n (%)			0.597
Resectable	3 (15.0)	6 (26.1)
Borderline resectable	15 (75.0)	14 (60.9)
Locally advanced	2 (10.0)	3 (13.0)
Weight change during NAT, n (%)			0.278
Increase	9 (45.0)	12 (52.2)
Decrease	7 (35.0)	10 (43.5)
Remained	4 (20.0)	1 (4.3)
RECIST 1.1 from imaging, n (%)			0.362
CR	0 (0.0)	0 (0.0)
PR	6 (30.0)	10 (43.5)
SD	14 (70.0)	13 (56.5)

OPS Open pancreatic surgery, RPS Robotic pancreatic surgery, ASA The American Society of Anesthesiologists, BMI Body mass index, AG Albumin-bound paclitaxel and gemcitabine, AS Albumin-bound paclitaxel and S-1, GS, Gemcitabine and S-1, mFOLFIRINOX Modified treatment with fluorouracil, leucovorin, oxaliplatin, and irinotecan. Nalirifox, Nanoliposomal Irinotecan, fluorouracil, leucovorin and  oxaliplatin, NAT Neoadjuvant therapy, CR Complete response, PR Partial response, SD Stable disease

A p-value <0.05 was considered statistically significant

^†Fisher’s exact test

Intraoperative outcomes

Among the 20 patients who underwent open pancreatic surgery (OPS), 8 had pancreaticoduodenectomy (PD), 11 had radical antegrade modular pancreatosplenectomy (RAMPS), and 1 patient was initially scheduled for pancreaticoduodenectomy but intraoperative frozen pathology from the pancreatic stump revealed malignancy. Due to the small remaining pancreatic volume, total pancreatectomy was performed instead. Among the 23 patients who underwent robotic pancreatic surgery (RPS), 11 underwent pancreaticoduodenectomy, 11 underwent RAMPS, and 1 underwent total pancreatectomy due to cancer involvement in both the head and body/tail of the pancreas. One case required conversion to open surgery due to a close relationship between the tumor and the sidewall of the SMV. After laparoscopic suturing of the SMV sidewall, the vessel narrowed, necessitating conversion to open surgery for re-clamping and transverse suturing of the SMV sidewall. The operation time was significantly longer in the RPS group compared to the OPS surgery group (340.0 min vs. 254.0 min, p = 0.015). There was no statistically significant difference in EBL between the two groups (350 ml vs. 400 ml, p = 0.961). In terms of short-term oncological outcomes, the R0 resection rate was slightly higher in the RPS group (91.3%) compared to the OPS group (85.0%), though the difference was not statistically significant. The number of lymph node resected was higher in RPS than in OPS ( (25 vs. 18, p = 0.049). Additionally, there were no statistical significance in terms of postoperative receiving adjuvant therapy rate and the time interval between surgery and postoperative adjuvant therapy (Table 2).

Table 2.

Perioperative and pathological outcomes of patients not requiring vascular reconstruction

	OPS (N = 20)	RPS (N = 23)	p value
Operation type, n (%)			0.876
Pancreatoduodenectomy	8 (40.0%)	11 (47.8)
RAMPS	11 (55.0)	11 (47.8)
Total pancreatectomy	1 (5.0)	1 (4.3)
Conversion into laparotomy, n (%)	——	1 (4.3)	——
Operation time, median (IQR), min	254.0 (213.5, 325.5)	340.0 (262.0, 463.0)	0.015
Estimated blood loss, median (IQR), ml	400 (300, 525)	350 (150, 800)	0.961
Blood transfusion, n (%)	5 (25.0)	6 (26.1)	0.935
Resection status, n (%)			0.650†
R0	17 (85.0)	21 (91.3)
R1	3 (15.0)	2 (8.7)
Vascular resection, n (%)	5 (20.0)	4 (17.4)	0.711†
Artery	1 (5.0)	1 (4.3)
Vein	3 (15.0)	3 (13.0)
Artery+Vein	1(5.0)	0 (0.0)
Tumor size, median (IQR), cm	2.45 (1.80, 3.15)	2.50 (1.70, 3.70)	0.951
Perineural invasion, n (%)	15 (75.0)	16 (69.6)	0.692
Peripancreatic adipose tissue invasion, n (%)	17 (85.0)	17 (73.9)	0.467†
Lymph node resected, n (%)	18 (13, 23)	25 (20, 30)	0.049
Positive lymph node rate, median (IQR)	0.00 (0.00, 0.07)	0.00 (0.00, 0.06)	0.715
Pathology, n (%)			0.590†
PDAC	18 (90.0)	22 (95.7)
Others	2 (10.0)	1 (4.3)
TNM stage, n (%)			0.648
ypT0N0M0	1 (5.0)	1 (4.3)
I	11 (55.0)	13 (56.5)
II	6 (30.0)	4 (17.4)
III	2 (10.0)	5 (21.7)
CAP grade, n (%)			0.481
0	1 (5.0)	1 (4.3)
1	3 (15.0)	1 (4.3)
2	5 (20.0)	10 (43.5)
3	11 (55.0)	11 (47.8)

OPS Open pancreatic surgery, RPS Robotic pancreatic surgery, RAMPS Radical antegrade modular pancreatosplenectomy, PDAC Pancreatic ductal adenocarcinoma, CAP score College of American Pathologists score

A p-value <0.05 was considered statistically significant

^†Fisher’s exact test

Postoperative outcomes

Regarding postoperative complications, the RPS group had a significantly lower overall complication rate compared to the OPS group (21.7% vs. 55.0, p = 0.024). However, there were no significant differences between the groups in terms of severe complications, CR-POPF. No significant differences were observed between the groups regarding PPH, pulmonary complications, or abdominal infections. The median postoperative hospital stay was 8 days for the RPS group, which was shorter than the 10-day stay for the OPS group, though this difference was not statistically significant (p = 0.368) (Table 3).

Table 3.

Postoperative outcomes of patients not requiring vascular reconstruction

	OPS (N = 20)	RPS (N = 23)	p value
Postoperative Adjuvant Therapy, n (%)	17 (85.0)	20 (87.0)	1.000†
Time to first adjuvant therapy, median (IQR), day*	55.0 (45.0, 64.0)	48.0 (41.5, 61.0)	0.218
Overall complication, n (%)	11 (55.0)	5 (21.7)	0.024
Severe complication, n (%)	0 (0.0)	1 (4.3)	1.000†
CR-POPF, n (%)	6 (30.0)	3 (17.4)	0.263†
CR-POPF for PD, n (%)	0 (0.0)	1 (9.1)	1.000†
CR-POPF for RAMPS, n (%)	6 (54.5)	2 (18.2)	0.118†
PPH, n (%)	2 (10.0)	1 (4.3)	0.590†
Delayed gastric emptying, n (%)	0 (0.0)	1 (4.3)	1.000†
Pulmonary complication, n (%)	1 (5.0)	0 (0.0)	0.465†
Abdominal infection, n (%)	2 (10.0)	1 (4.3)	0.590†
Other complication, n (%)	2 (10.0)	4 (17.4)	0.669†
Reoperation, n (%)	0 (0.0)	0 (0.0)	——
Reintervention, n (%)	0 (0.0)	1 (4.3)	1.000†
90-day mortality associated with surgery, n (%)	0 (0.0)	0 (0.0)	——
Postoperative hospital stay, median (IQR), days	10.0 (8.5, 12.0)	8.0 (8.0, 13.0)	0.368

OPS Open pancreatic surgery, RPS Robotic pancreatic surgery, PD Pancreatoduodenectomy, RAMPS Radical antegrade modular pancreatosplenectomy, CR-POPF Clinically relevant postoperative pancreatic fistula, PPH Postpancreatectomy hemorrhage. A p-value <0.05 was considered statistically significant

^*Especifically for PDAC (n = 40)

^†Fisher’s exact test

Long-term oncologic outcomes

In terms of prognosis, we performed a subgroup analysis of patients with pathologically confirmed PDAC (n = 40). The median follow-up period was 30.6 months. The median DFS and OS were 18.7 months and 26.4 months in the OPS group, respectively, compared with 12.3 months and 29.9 months in the RPS group (Fig. 2). No statistically significant differences were observed between the two groups in terms of DFS or OS (p = 0.153 and p = 0.711, respectively). Furthermore, univariate Cox regression analysis was conducted to identify factors associated with DFS and OS; however, no significant prognostic factors were identified (Supplementary Tables 3 and Supplementary Table 4). In addition, after adjusting for age, sex, BMI, chemotherapy regimen, resection margin status, TNM stage, CAP grade, and receipt of adjuvant chemotherapy, DFS and OS remained comparable between the two groups (p = 0.084 and p = 0.603, respectively).

Fig. 2.

Disease-free and overall survival after RRS and ORS in patients with pancreatic ductal adenocarcinoma after NAT. Survival analyses were performed using Kaplan–Meier estimates and log-rank tests. A Kaplan–Meier curves for disease-free survival: median DFS was 12.3 months for RPS and 18.7 months for OPS (p = 0.153). B Kaplan–Meier curves for overall survival: median OS was 28.5 months for RPS and 26.4 months for OPS (p = 0.711)

Discussion

Our study is exploratory in nature and provides preliminary evidence supporting the feasibility and safety of robotic surgery for radical resection in patients with pancreatic cancer following neoadjuvant therapy. In our cohort, no patients experienced postoperative complications of Clavien-Dindo grade ≥ 3b, further supporting the safety and feasibility of the robotic approach. Moreover, RPS was associated with a higher number of lymph node harvested and a lower overall complication rate, albeit with a longer operation time. Postoperative hospital stay was also slightly shorter in the RPS group. Importantly, the oncologic outcomes of robotic and open surgery were comparable, as reflected by similar R0 resection rates, DFS, and OS.

It should be emphasized that patients included in this study were well-selected. Only cases not requiring vascular segmental resection and reconstruction were considered for robotic surgery. Patients with preoperative imaging indicating PV/SMV involvement necessitating segmental resection were preferentially assigned to open surgery. These cases are inherently more complex, with a higher conversion rate from minimally invasive to open surgery. While some studies have reported cases of laparoscopic venous patch repair and segmental resection with reconstruction, these cases remain rare, constituting only around 14% of the total, and the study did not recommend the use of synthetic venous grafts or arterial resection with reconstruction [21].

We observed a significantly lower overall postoperative complication rate in the RPS group compared with the OPS group. Given that standardized criteria were applied for complication assessment and postoperative management, the comparability between groups was ensured. Notably, the incidences of CR-POPF and pulmonary complications were lower in the RPS group. These findings may be attributable to the intrinsic advantages of the robotic platform. The high-definition, magnified three-dimensional visualization and articulated instruments with seven degrees of freedom facilitate precise identification of the main pancreatic duct and enable meticulous suturing and reinforcement of the pancreatic stump, thereby potentially reducing the risk of CR-POPF. In addition, the absence of pulmonary infections in the RPS group may reflect the minimally invasive nature of robotic surgery, which is associated with reduced surgical trauma, less postoperative pain, and consequently better preservation of respiratory function.

Currently, evidence regarding robotic resection following neoadjuvant therapy remains limited. To date, only two studies have compared robotic versus open pancreatic cancer resections following neoadjuvant treatment, with results aligning with ours. Nassour et al. used the U.S. National Cancer Database to compare the outcomes of robotic pancreaticoduodenectomy (RPD) and open pancreaticoduodenectomy (OPD) after neoadjuvant therapy. The results indicated that RPD was associated with an increased number of lymph node harvested, reduced postoperative hospital stays, and a higher rate of adjuvant chemotherapy. Both groups had similar R0 resection rates, 90-day mortality, and long-term outcomes [22]. However, this study did not assess vascular involvement comparability or postoperative complications [22]. Another study conducted by Ocuin et al. focused on cases of distal pancreatectomy with celiac axis resection (DP-CAR), including 11 robotic and 19 open cases, with 27 patients having received neoadjuvant therapy. The results showed that robotic DP-CAR reduced operative time and intraoperative blood loss, and there were no significant differences in postoperative complications, pancreatic fistula rates, or overall survival between the two groups [23]. AIMasri et al. also reported a case of robotic resection following neoadjuvant therapy for pancreatic head cancer. In this case, the tumor invaded 90–180 degrees of the involved vessels, and a wedge resection of the vein was performed using a side-bite technique. The patient was safely discharged on the 8th postoperative day [24]. Based on these findings and our study, robotic surgery after neoadjuvant therapy appears to be safe and feasible, showing efficacy noninferior to open surgery in well-selected cases.

In addition to safety, another major concern regarding robotic-assisted surgery for pancreatic cancer is its short- and long-term oncologic efficacy. Previous studies have shown that, in patients who did not receive neoadjuvant therapy, robotic surgery was associated with a significantly greater number of lymph nodes harvested [25, 26]. Similarly, our study found that patients undergoing robotic pancreatectomy after neoadjuvant chemotherapy also had a higher number of lymph nodes retrieved, consistent with the findings of Nassour et al. [22]. Furthermore, RPS was associated with comparable rates of negative resection margins, receipt of adjuvant therapy, and time to initiation of postoperative adjuvant treatment. No significant differences in OS were observed between the RPS and OPS groups. These findings support the use of robotic surgery as an oncologically effective alternative to open surgery for patients undergoing pancreatectomy following neoadjuvant therapy.

Several limitations should be acknowledged. First, this was a retrospective study and is therefore subject to potential selection bias. Second, data on postoperative adjuvant chemotherapy—such as completion rates and relative dose intensity—were incomplete, limiting the evaluation of how surgical approach may influence systemic treatment delivery. Third, only patients with tumors involving vessels less than 180° and not requiring vascular reconstruction were included. More complex cases requiring venous patch repair or segmental resection were managed with open surgery. With increasing experience, robotic surgery combined with vascular reconstruction is being explored, and future studies should evaluate its safety and feasibility. Fourth, the follow-up duration in the RPS group was relatively short, necessitating longer follow-up to better assess survival outcomes. Finally, the sample size was limited, and larger, multicenter prospective studies are warranted to validate our findings.

Conclusion

In carefully selected patients with pancreatic cancer following neoadjuvant therapy who did not require vascular segmental resection and reconstruction, a robotic approach appeared technically feasible and was associated with favorable short-term postoperative outcomes and a greater lymph node yield. Given the retrospective design and limited sample size, these findings should be interpreted with caution and viewed as exploratory. Further prospective studies are required before conclusions regarding safety, oncologic adequacy, or clinical indications can be established.

Authors’ contributions

Menghua Dai, Pengyu Li: Conceptualization, Design. Pengyu Li, Hui Zhang, Juan Li: Methodology. Pengyu Li, Hui Zhang, Juan Li, Peng Zhang, Qiang Xu, Wenjing Liu, Haomin Chen, Weijie Chen, Yaqi Wang: Data collection. Pengyu Li, Mingfei Wang, Yuqian Kan, Xinwei Cui: Formal analysis and investigation. Pengyu Li: Writing - original draft preparation. Menghua Dai, Huadan Xue, Zhiyong Liang: Writing - review and editing; Huadan Xue, Zhiyong Liang: Supervision.

Funding

This work was supported by the National High Level Hospital Clinical Research Funding (2022-PUMCH-B-004), CAMS Innovation Fund for Medical Sciences (CIFMS) (2024-I2M-ZD-001), and the project of National Natural Science Foundation of China (82573271).

Data availability

Given privacy and ethical restrictions, the data supporting this study’s findings are available on request from the corresponding author.

Declarations

Ethics approval and consent to participate

This study was performed in line with the principles of the Declaration of Helsinki. The Institutional Review Board (IRB) of Peking Union Medical College Hospital approved the study (Approval Number: K7514).Informed consent was obtained from all individual participants included in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.