Multimodal treatment combining neoadjuvant therapy, laparoscopic subtotal distal pancreatectomy and adjuvant therapy for pancreatic neck-body cancer: Case series

Abstract

BACKGROUND

Pancreatic cancer involving the pancreas neck and body often invades the retroperitoneal vessels, making its radical resection challenging. Multimodal treatment strategies, including neoadjuvant therapy, surgery, and postoperative adjuvant therapy, are contributing to a paradigm shift in the treatment of pancreatic cancer. This strategy is also promising in the treatment of pancreatic neck-body cancer.

AIM

To evaluate the feasibility and effectiveness of a multimodal strategy for the treatment of borderline/locally advanced pancreatic neck-body cancer.

METHODS

From January 2019 to December 2021, we reviewed the demographic characteristics, neoadjuvant and adjuvant treatment data, intraoperative and postoperative variables, and follow-up outcomes of patients who underwent multimodal treatment for pancreatic neck-body cancer in a prospectively collected database of our hospital. This investigation was reported in line with the Preferred Reporting of Case Series in Surgery criteria.

RESULTS

A total of 11 patients with pancreatic neck-body cancer were included in this study, of whom 6 patients were borderline resectable and 5 were locally advanced. Through multidisciplinary team discussion, all patients received neoadjuvant therapy, of whom 8 (73%) patients achieved a partial response and 3 patients maintained stable disease. After multidisciplinary team reassessment, all patients underwent laparoscopic subtotal distal pancreatectomy and portal vein reconstruction and achieved R0 resection. Postoperatively, two patients (18%) developed ascites, and two patients (18%) developed pancreatic fistulae. The median length of stay of the patients was 11 days (range: 10-15 days). All patients received postoperative adjuvant therapy. During the follow-up, three patients experienced tumor recurrence, with a median disease-free survival time of 13.3 months and a median overall survival time of 20.5 months.

CONCLUSION

A multimodal treatment strategy combining neoadjuvant therapy, laparoscopic subtotal distal pancreatectomy, and adjuvant therapy is safe and feasible in patients with pancreatic neck-body cancer.

Key Words: Pancreatic neck-body cancer; Multimodal treatment; Neoadjuvant therapy; Laparoscopic subtotal distal pancreatectomy; Adjuvant therapy

Core Tip: Radical resection of the tumor and associated vascular reconstruction for pancreatic neck-body cancer is mostly accomplished at our center using laparoscopic subtotal distal pancreatectomy. The application of this minimally invasive technique not only ensures oncologic outcomes but also preserves the intact gastrointestinal tract and part of the pancreas, allowing a quicker recovery after surgery and the accomplishment of subsequent adjuvant therapy. Our current study shows that the multimodal treatment strategy combining neoadjuvant therapy, laparoscopic subtotal distal pancreatectomy, and adjuvant therapy is safe and feasible in patients with pancreatic neck-body cancer.

INTRODUCTION

The incidence of pancreatic cancer is increasing, but the prognosis is poor, with an overall five-year survival rate of less than 9%[1]. Surgery is currently the only potential cure for pancreatic cancer. Owing to the insidious onset of this condition, only 10%-20% of newly diagnosed patients can undergo surgical resection[2]. The introduction of neoadjuvant therapy has allowed some patients with borderline/locally advanced tumors to be down staged and thus gain access to surgical cures[3]. Adjuvant therapy has also further improved the long-term survival of patients undergoing radical resection[4]. The therapeutic paradigm for borderline/locally advanced pancreatic cancer is entering the era of multimodal treatment. The pancreatic neck is the transition zone between the head and the body and lies anterior to the superior mesenteric vein (SMV) and portal vein (PV). Pancreatic cancers occurring at different sites present different lymphatic metastasis, peripheral invasion and prognostic features[5]. Pancreatic neck cancers are more challenging to achieve R0 resection because they often involve posterior vascular structures, which also results in lower resection rates than those of tumors located at the head or body/tail[6]. When the tumor involves both the neck and body, the selection of the surgical approach is more complicated. There is no optimal surgical plan at present, and extended pancreatoduodenectomy, subtotal pancreatectomy or even total pancreatectomy are potentially available options[7].

Compared with open pancreatic resection, laparoscopic pancreatectomy may allow for faster postoperative recovery and provide comparable oncologic safety[8,9]. Laparoscopic subtotal pancreatectomy preserves both part of the pancreatic tissue and the intact digestive tract, which may result in faster recovery after surgery and improve patient tolerance to adjuvant therapy. We envision that this surgical procedure combined with neoadjuvant therapy can improve long-term survival in patients with pancreatic neck-body cancer by both achieving R0 resection and ensuring that the patient completes adjuvant therapy as planned (Figure 1). In this study, we elucidated the feasibility and effectiveness of integrating preoperative neoadjuvant chemotherapy, laparoscopic subtotal pancreatectomy and postoperative adjuvant therapy as a multimodal strategy for the treatment of borderline/locally advanced pancreatic neck-body cancer.

Figure 1 Flowchart for multimodal treatment of borderline/locally advanced pancreatic neck-body cancer. MDT: Multidisciplinary team; SMV: Superior mesenteric vein; PV: Portal vein.

MATERIALS AND METHODS

Definitions

On the basis of the American Joint Committee on Cancer 8^th edition tumor-node-metastasis system and previous studies, we defined the pancreatic subdivision as follows (Figure 2A): (1) The pancreas head, the part to the right of the PV/SMV; (2) The pancreas neck, the part in front of the PV/SMV; (3) The pancreas body, the part from the left of the PV/SMV to the left edge of the abdominal aorta (AA); and (4) The pancreas tail, the part to the left of the AA[10]. In this study, subtotal pancreatectomies were defined as pancreatic operations with transection planes along the left edge of the common bile duct (CBD). The margin status of the surgical specimen was classified according to the criteria of the Royal College of Pathologists, with R0 referring to a margin ≥ 1 mm from the tumor, R1 referring to a margin < 1 mm from the tumor, and R2 referring to a macroscopically positive margin[11]. The status of tumor invasion of the PV was classified into four types: (1) Type A (no involvement); (2) Type B (involvement of one side of the PV); (3) Type C (involvement of both sides of the PV); and (4) Type D (portal stenosis or obstruction with collateral circulation)[12]. Response Evaluation Criteria In Solid Tumors 1.1 criteria was used to evaluate the efficacy of neoadjuvant therapy for pancreatic tumors[13]. Complications associated with pancreatic surgery, such as postoperative pancreatic fistula, delayed gastric emptying, and post-pancreatectomy hemorrhage, were evaluated and documented according to the International Study Group of Pancreatic Surgery definitions[14-16].

Figure 2 The definition of pancreatic subdivisions. A: The pancreatic subdivisions were defined as follows: Pancreatic head, right to the portal vein (PV)/superior mesenteric vein (SMV); pancreatic neck, in front of the PV/SMV; pancreatic body, left of the PV/SMV to left of the abdominal aorta; pancreatic tail, left of the abdominal aorta; B: Pancreatic cancer involving the neck and body of the pancreas; pancreatic cancer (yellow arrow), portal vein (blue arrow). PV: Portal vein; GDA: Global domain adaptation; CBD: Common bile duct; SMV: Superior mesenteric vein; Ph: Pancreatic head; Pn: Pancreatic neck; Pb: Pancreatic body; Pt: Pancreatic tail; CHA: Common hepatic artery; CA: Celiac artery; LGA: Left gastric artery; SA: Spleen artery; AA: Abdominal aorta; SMA: Superior mesenteric artery.

Patients

On the basis of a prospectively collected database, the clinical features of patients with pancreatic ductal adenocarcinoma who underwent radical surgery after neoadjuvant chemotherapy at our hospital from January 2019 to December 2021 were analyzed retrospectively. The inclusion criteria were as follows: (1) The initial diagnosis of borderline or locally advanced pancreatic cancer was determined by a multidisciplinary team (MDT) with reference to imaging within 1 week and the 2021 National Comprehensive Cancer Network definition[17]; (2) The tumor involved both the neck and the body (Figure 2B); (3) At least 2 cycles of neoadjuvant chemotherapy were performed before surgery; (4) Tumors were evaluated by the MDT as resectable or borderline after neoadjuvant chemotherapy (Figure 3); and (5) Laparoscopic subtotal pancreatectomy was performed. Patients who underwent arterial resection and patients with missing data were excluded. This study was approved by the Ethics Committee of Hunan Provincial People’s Hospital (The First Affiliated Hospital of Hunan Normal University), No. 2020-03. Written consent was obtained from all patients for the publication of all images, clinical data, and other data.

Figure 3 Changes in pancreatic lesions before and after neoadjuvant therapy. A and B: Case 1; C and D: Case 2. Yellow arrows indicate the lesions.

Preoperative evaluation and management

Nutritional risk screening was performed according to nutrition risk screening (2002) before surgery, and parenteral nutrition support was provided if the body mass index was < 18.5. The preoperative tests included routine blood and liver biochemistry, blood coagulation, abdominal ultrasound, contrast-enhanced thin-slice computed tomography, vascular imaging of the upper abdomen, three-dimensional reconstruction of the pancreas, and whole-body positron emission tomography-computed tomography. All patients received a pathological diagnosis of pancreatic cancer via endoscopic ultrasound-guided and fine-needle aspiration before initiating neoadjuvant therapy. According to the relationship between the tumor and the surrounding tissue, the location and length of vascular invasion, and the status of collateral circulation formation, we determined the method of resection and reconstruction of the PV/SMV and whether artificial vessels would need to be prepared. Surgery was performed 4 weeks after neoadjuvant chemotherapy. The indications for surgical resection were as follows: The tumor touched or encircled (< 180 degrees) the superior mesenteric artery (SMA) and/or the celiac artery (CA) without irregular stenosis, the PV/SMV could be reconstructed, preoperative imaging indicated that lymph node metastasis did not exceed the scope of resection, the involvement of second-station lymph nodes (specifically No. 13 and 16) was a contraindication for surgery[7,18], and there was no invasion of the AA and no distant metastasis.

Laparoscopic subtotal distal pancreatectomy

Patients were routinely placed in a horizontal position with separated legs, and a V-shaped five-hole trocar layout was used. The whole abdominal cavity was examined to exclude metastasis, and an artery-first approach was applied to explore whether the SMA and the CA had been invaded (Figure 4A). If the SMA had been invaded, surgery was terminated. If not, the SMA and CA were skeletonized, and then the posterior plane behind the neck of the pancreas was dissected. The No. 14 and No. 9 lymph nodes and the nerve fibers and connective tissue in Heidelberg’s triangle were cleaned to expose the left edge of the surface of the uncinate mesentery. The No. 8, No. 12a, and No. 12p lymph nodes and the fibrous connective tissue on the left side of the hepatoduodenal ligament were cleaned at the upper edge of the pancreas. The global domain adaptation (GDA) was then dissected, and the blood supply to the duodenum was confirmed. The left edge of the CBD was used as the transection plane of the pancreas to ensure that the resection margin was at least 1 cm away from the tumor (Figure 4B and C). The extrathecal tissues of the SMA, common hepatic artery (CHA), and CA, as well as the PV, SMV, and pancreatic neck stumps, were sent for quick-frozen pathology to ensure negative resection margins. If a poor duodenal blood supply is found during the operation, total pancreatectomy is performed instead.

Figure 4 Laparoscopic subtotal pancreatectomy with portal vein/superior mesenteric vein reconstruction. A: The artery-first approach was applied in our laparoscopic subtotal pancreatectomy; B: Schematic diagram of the pancreas transection plane for laparoscopic subtotal pancreatectomy; C: We transected the pancreas along the left edge of the common bile duct during surgery; D: The involved portal vein was resected, rapid frozen section pathology confirmed no residual tumor cells at the portal vein transection margin; E: End-to-end anastomosis of the portal vein was performed, the soft tissues adjacent to the tumor in the Heidelberg triangle were removed; F: The posterior plane of resection was behind the left adrenal gland and Gerota’s fascia (similar to the posterior radical antegrade modular pancreatosplenectomy). SMA: Superior mesenteric artery; CBD: Common bile duct; CA: Celiac artery.

End-to-end anastomosis was performed when PV and/or SMV invasion was ≤ 3 cm (Figure 4D and E). If the PV and/or SMV was invaded by > 3 cm, an artificial vessel was used. The vascular anastomosis is usually sutured continuously with 5-0 prolene, and the knot should not be too tight to avoid excessive tension of the anastomosis after PV reflow. Posterior radical antegrade modular pancreatosplenectomy was used for en bloc resection of the sample (Figure 4F). The extrathecal tissue of the SMA, CHA, CA, and the end of the PV, SMV, and pancreatic neck were sent for intraoperative pathology to ensure negative margins. All of the operations were performed by the same group of surgeons.

Postoperative management

Antibiotics, low-molecular-weight heparin and somatostatin were routinely administered. The diet was gradually increased beginning on the first postoperative day. Pancreatic fistula and postoperative bleeding were evaluated by the International Study Group of Pancreatic Surgery definition and symptomatic management, and other adverse events were judged by the Clavien-Dindo operation complication classification[19]. Four weeks later, the MDT decided the adjuvant treatment plan according to the intraoperative findings, the pathology results, and the patient’s physical condition and individual wishes.

Follow-up

Follow-up visits were accomplished through outpatient clinics, WeChat, and telephone. The first follow-up visit was conducted 1 month after surgery. During postoperative adjuvant therapy, individualized follow-up was performed according to the treatment plan. After adjuvant therapy, follow-up was performed every 3 months, with routine blood, liver and renal function, blood glucose, and carbohydrate antigen 19-9 (CA19-9) tests and abdominal computed tomography enhancement examination.

Statistical analysis

All of the analyses were conducted via SPSS software, version 26.0 (IBM Inc., Chicago, IL, United States). The survival time and disease-free survival time of the patients were calculated from the first day after chemotherapy and the operation, respectively. Continuous variables with a normal distribution are represented by the mean and range, and those that did not conform to a normal distribution are represented by the median and range. Categorical variables are expressed in terms of numbers and proportions. This investigation has been reported in line with the Preferred Reporting of Case Series in Surgery criteria[20].

RESULTS

Baseline characteristics

This study enrolled 11 patients, comprising 5 males and 6 females, with a median age of 55 years (range: 41-68 years) (Table 1). The initial diagnosis was borderline pancreatic cancer in six patients and locally advanced pancreatic cancer in five patients. Four patients had diabetes mellitus, and one patient had a history of biliary surgery. The pathological results of the preoperative biopsies revealed that all the patients had pancreatic ductal adenocarcinoma. Before chemotherapy, the median CA19-9 level was 871 U/mL (range: 112-2154 U/mL), and the median tumor diameter was 4.4 cm (range: 2.8-5.9 cm). Six borderline patients were treated with albumin-bound paclitaxel and gemcitabine (AG) for 2-6 cycles, one locally advanced patient was treated with AG, S1 and tirelizumab for 2 cycles, one locally advanced patient was treated with AG and tirelizumab for 6 cycles, and 3 locally advanced patients were treated with the modified FOLFIRINOX regimen for 6 cycles. During neoadjuvant chemotherapy, 3 patients experienced grade 2 myelosuppression, 4 developed diarrhea, and 6 experienced alopecia. Myelosuppression and diarrhea resolved with symptomatic treatment, and hair regrowth occurred following the completion of chemotherapy. The chemotherapy effect was a partial response in 8 patients and stable disease in 3 patients. Two borderline and five locally advanced patients were successfully down staged after chemotherapy. After neoadjuvant therapy, the median CA19-9 level was 24 U/mL (range: 6-157 U/mL), and the average tumor diameter was 2.2 cm (range: 1.3-4.1 cm).

Table 1 Patients baseline information and results of neoadjuvant therapy, n (%).

Characteristic	N = 11
Sex
Female	6 (55)
Male	5 (45)
Age (years), median (range)	55 (41-68)
Blood glucose level
Normal	6 (55)
Diabetes	4 (36)
Impaired glucose tolerance	1 (9)
Before neoadjuvant therapy
Borderline	6 (55)
Locally advanced	5 (45)
Arterial involvement
CA	1 (9)
CA/SMA	5 (45)
CHA/CA	3 (27.3)
CHA/CA/SMA	1 (9.1)
SMA	1 (9.1)
Vein involvement
PV/SMV/SV	8 (73)
PV/SV	3 (27)
Biopsy
Ductal adenocarcinoma	11 (100)
Neoadjuvant therapy regimen
AG	6 (55)
AG + tirelizumab	1 (9)
AG + S-1 + tirelizumab	1 (9)
The mFOLFIRINOX	3 (27)
Neoadjuvant therapy cycle, median (range)	5 (2-6)
Initial CA19-9 (U/mL), median (range)	871 (112-2154)
Preoperative CA19-9 (U/mL), median (range)	24 (6-157)
Initial diameter (cm), median (range)	4.4 (2.8-5.9)
Preoperative diameter (cm), median (range)	2.2 (1.3-4.1)
Neoadjuvant therapy effect
PR	8 (73)
SD	3 (27)
After neoadjuvant therapy
Borderline	8 (73)
Resectable	3 (27)

CA: Celiac artery; SMA: Superior mesenteric artery; CHA: Common hepatic artery; PV: Portal vein; SMV: Superior mesenteric vein; SV: Spleen vein; AG: Albumin-bound paclitaxel and gemcitabine; S-1: Tegafur, gimeracil, and oteracil potassium; mFOLFIRINOX: Modified FOLFIRINOX regimen, including 5-fluorouracil, leucovorin, irinotecan and oxaliplatin; CA19-9: Carbohydrate antigen 19-9; PR: Partial response; SD: Stable disease.

Perioperative outcomes

All 11 patients underwent artery-first laparoscopic subtotal distal pancreatectomy (DP) combined with PV/SMV resection and reconstruction. The average operation time and amount of intraoperative blood loss were 400 minutes (range: 330-510 minutes) and 300 mL (range: 100-700 mL), respectively. There were no cases of blood transfusion. Obvious collateral circulation formed around the pancreas in 7 patients. End-to-end anastomosis was performed in 8 patients after segmental resection of the PV/SMV. Artificial vascular reconstruction was performed in 3 patients after segmental resection of the PV/SMV. Two patients had complications of grade B pancreatic fistula and recovered after puncture and drainage. Intractable ascites occurred in 2 patients, causing partial destruction of the collateral circulation during the operation; in these two patients, the drainage tube was removed on the 21^st and 28^th days after the operation. There were no patients with gastric emptying disturbances, bleeding, anastomotic thromboses or reoperations. The median postoperative hospital stay was 11 days (range: 10-15 days) (Table 2).

Table 2 Operation details and patient recovery information, n (%).

Characteristic	N = 11
Operation time (minute), median (range)	400 (330-510)
Blood loss (mL), median (range)	300 (100-700)
Collateral circulation	7 (64)
SMA/CA invasion
Yes	0 (0)
No	11 (100)
Pancreatic margin
Positive	0 (0)
Negative	11 (100)
PV/SMV invasion classification
A	0 (0)
B	3 (27)
C	3 (27)
D	5 (45)
Vein reconstruction
Artificial vascular	3 (27)
End-to-end	8 (73)
RAMPS excision plane
Posterior	11 (100)
Complication
No	7 (64)
Intractable ascites	2 (18)
Pancreatic fistula	2 (18)
Postoperative hospital stays (days), median (range)	11 (10-15)

SMA: Superior mesenteric artery; CA: Celiac artery; PV: Portal vein; SMV: Superior mesenteric vein; RAMPS: Radical antegrade modular pancreatosplenectomy.

Postoperative pathology and adjuvant therapy

Postoperative pathology revealed that all 11 patients underwent R0 resection, including 2 patients with high differentiation, 4 patients with medium differentiation, 3 patients with medium-low differentiation and 2 patients with low differentiation. The average number of lymph node dissections was 16 (range: 11-25). Three patients had one positive lymph node, and one patient had three positive lymph nodes. Nerve invasion was found in seven patients. Seven patients had grade 2 tumor regression grading (TRG), and 3 patients had grade 3 TRG. According to the tumor-node-metastasis staging, 3 patients were in stage IA, 3 patients were in stage IB, 1 patient was in stage IIA, and 4 patients were in stage IIB (Table 3). Four patients received radiotherapy (50.4 Gy/28 f) combined with capecitabine (1 g orally, twice per day) after 2 cycles of AG. One patient received 2 cycles of modified FOLFIRINOX chemotherapy after surgery. Four patients received oral S-1 (120 mg/day on days 1-28 of each cycle, repeated every 6 weeks for up to 6 months). Among these patients, 3 completed 6 courses of treatment, and one had abnormal liver function and discontinued the drug after 2 cycles of oral administration. One patient received intravenous gemcitabine (1000 mg/m² on days 1, 8, and 15 of each cycle, which was repeated every 4 weeks for up to 6 months). One patient received 4 cycles of AG, and the other received 6 cycles of tirelizumab. All 11 patients were treated with Oryz-Aspergillus enzyme and pancreatin tablets (1 tablet, oral, three times per day) after the operation (Table 3).

Table 3 Postoperative pathology data and follow-up results, n (%).

Characteristic	N = 11
Degree of differentiation
High	2 (18)
Medium	4 (36)
Medium-low	3 (27)
Low	2 (18)
Pathological staging
IA	3 (27)
IB	3 (27)
IIA	1 (9)
IIB	4 (36)
R0 resection
Yes	11 (100)
Lymph node harvest, median (range)	16 (11-25)
Positive lymph node
0	7 (64)
11d	1 (9)
8a	2 (18)
8a/9/14p	1 (9.1)
Nerve invasion	7 (64)
Tumor regression grading
0	1 (9)
1	0 (0)
2	7 (64)
3	3 (27)
Postoperative adjuvant therapy
AG + PD-1	1 (9)
Chemoradiotherapy	4 (36)
Gemcitabine	1 (9)
The mFOLFIRINOX	1 (9)
S-1	4 (36)
Blood glucose level
Normal	4 (36)
Diabetes	5 (46)
Impaired glucose tolerance	2 (18)
Follow-up (months), median (range)	15 (8-33)
Recurrence and metastasis
No	8 (73)
Liver	1 (9)
Local and liver	1 (9)
Local and radial bone	1 (9)
Survival state
Death	1 (9)
Survived with tumor	2 (18)
Tumor-free	8 (73)
Median overall survival (month)	20.5
Median disease-free survival (month)	13.3

AG: Albumin-bound paclitaxel and gemcitabine; PD: Pancreaticoduodenectomy; S-1: Tegafur, gimeracil, and oteracil potassium; mFOLFIRINOX: Modified FOLFIRINOX regimen, including 5-fluorouracil, leucovorin, irinotecan and oxaliplatin.

Follow-up outcomes

All 11 patients were followed up, and data collection ended on March 31, 2023. The follow-up time ranged from 8-33 months. The median overall survival and median disease-free survival of patients were 20.5 months and 13.3 months, respectively. During the follow-up period, three patients experienced tumor recurrence. One patient developed liver metastasis in the 3^rd month after surgery and then underwent laparoscopic microwave ablation and died at the 11^th month after surgery. One patient developed local recurrence and radial metastases at the 21^st month after surgery and survived with the tumor. The last patient developed local recurrence 11 months after surgery and liver metastases 14 months later. Impaired glucose tolerance and weight loss occurred in one patient and were controlled by adjusting the diet and prescribing metformin. One patient was diagnosed with new-onset diabetes after the operation and was managed with insulin and dietary adjustments to control blood glucose.

DISCUSSION

Pancreatic cancer is a malignant tumor of the digestive tract with a poor prognosis[1]. R0 resection is currently a potential cure for pancreatic cancer. However, most patients with pancreatic cancer miss the optimal window for surgery at the time of diagnosis, and only 20% of patients are eligible for resection[21]. Owing to the dismal long-term prognosis of pancreatic cancer patients and the successful application of neoadjuvant and postoperative adjuvant therapy in other tumors, the therapeutic paradigm for pancreatic cancer has begun to shift from a single surgical resection to a multimodal strategy[22]. The feasibility and effectiveness of the latter is supported by a growing number of clinical trials[23,24]. However, due to postoperative complications or clinical deterioration, a significant proportion of patients fail to complete adjuvant therapy[25]. In our center, patients with borderline/locally advanced pancreatic neck-body cancer are treated with neoadjuvant therapy combined with laparoscopic subtotal pancreatectomy, which not only helps achieve R0 resection but also avoids postoperative complications caused by gastrointestinal reconstruction and total pancreatectomy. This allows patients to recover quickly after surgery, improves their tolerance to adjuvant chemoradiotherapy, and ultimately completes multimodal treatment. Our results suggest that this strategy is feasible and effective. Adjuvant chemoradiotherapy combined with surgical resection is currently the only possible cure for pancreatic cancer. Most patients with pancreatic cancer receive selective pancreaticoduodenectomy (PD) or DP according to the location of the tumor. However, the neck of the pancreas is located in a small area surrounded by the PV, CHA and GDA, and the body of the pancreas is located from the left side of the SMV to the left margin of the AA[7]. Anatomically, both are located in the dorsal pancreas. Owing to the specific anatomic location of pancreatic neck-body cancer, it often extends beyond the pancreas and results in the involvement of critical peripancreatic vessels, making complete surgical resection with negative margins challenging[26].

Previous studies have shown that total pancreatectomy can achieve a higher R0 resection rate than can partial pancreatectomy, but there was no significant difference in median survival between the two groups[27,28]. Complete loss of the internal and external endocrine functions of the pancreas in patients after total pancreatectomy leads to poor postoperative quality of life[18,29,30], and the proportion of patients receiving postoperative adjuvant therapy is significantly reduced[28]. Therefore, total pancreatectomy is not a preferred option for pancreatic cancer patients. In addition to total pancreatectomy for pancreatic neck-body cancer, there are various modified procedures for partial pancreatectomy, such as PD with CHA[19] or splenic artery resection[31], proximal subtotal pancreatectomy[32], Whipple at the splenic artery[31], DP with CA resection[33], and extended DP[34]. However, pancreatic neck cancer often involves lymph node metastasis along the CHA (57%), pancreatic head (19%) and SMA (11%)[7]. Carcinoma of the pancreatic body easily invades the nerve plexus around the root of the splenic artery[35,36], the splenic vessels are embedded into the pancreatic body, and lymph node metastasis around the splenic artery is common[37,38]. Therefore, in addition to the resection margin, lymph node metastasis is another important factor that is controversial in surgical selection.

Borderline or locally advanced pancreatic cancer located above the PV, SMA and CA usually involves both the neck and body of the organ; thus, it is difficult to accurately define one region or the other as the tumor’s primary origin. Therefore, some scholars have chosen proximal pancreatectomy or DP according to the distance between the tumor and the GDA[7,39]. Previous studies have shown that the R0 resection rate and survival time of pancreatic neck-body cancer patients after DP with CA resection or extended DP without invasion of the GDA are similar to those of patients after PD or PD with splenic artery resection. However, combined arterial resection and reconstruction increase the difficulty of surgery and the incidence of complications, and the oncological benefits are controversial[40-42]. Moreover, extended pancreatoduodenectomy without reconstruction of the splenic vein could lead to left portal hypertension, gastrointestinal bleeding and hypersplenism and might also lead to acute splenic congestion and forced splenectomy[43]. Owing to tumor retraction after neoadjuvant therapy, the resection margins of the arteries (CHA, CA, and SMA) that previously touched or were encircled by the tumor became clear. It has been verified that a radical cure may be achieved by periarterial divestment[44]. Regrettably, few studies on pancreatic neck-body cancer after neoadjuvant chemotherapy exist, and the number of cases is limited. Considering the lymph node metastasis pathway of pancreatic neck-body cancer and the low positive lymph node rate after neoadjuvant chemotherapy[45,46], we selectively performed subtotal DP in these patients.

Our results revealed that all 11 patients achieved R0 resection. Four of the patients in groups 8a or 11d had lymph node metastasis, which is consistent with previous studies and confirms the necessity of complete resection of the tail of the pancreas for tumors involving the body of the pancreas[47]. Eight patients survived without tumor recurrence. Recurrence and metastasis occurred in 3 patients during follow-up. Among these three patients, the patient with local recurrence and radial metastasis 21 months after the operation received only 2 cycles of neoadjuvant chemotherapy “AG”. The diameter of the tumor reached 3.8 cm before the operation; lymph node metastasis was detected, and the TRG grade was 3 after the operation. An insufficient number of cycles of preoperative chemotherapy might have been the main cause of this recurrence. The CA19-9 level of the patients with liver metastases was 2154.5 U/mL at the initial diagnosis. After 5 cycles of AG chemotherapy, the CA19-9 level decreased to 157.1 U/mL. Although concurrent chemoradiotherapy was performed postoperatively, liver metastases still occurred within the third month after surgery. Therefore, the decision of surgical timing during neoadjuvant chemotherapy for borderline and locally advanced pancreatic cancer patients is directly related to prognosis, and it is necessary to strike a balance between chemotherapy and surgery. Another patient who presented with local recurrence followed by liver metastasis had a preoperative CA199 of 1398.4 U/mL. Despite six cycles of AG + tirelizumab preoperatively and tumor shrinkage from 5.9 cm to 1.5 cm, three positive lymph nodes (8a, 9, and 14p) were observed postoperatively. Owing to the positive lymph nodes, we used aggressive postoperative concurrent radiotherapy, but local recurrence with secondary liver metastases developed 10 months after surgery. This patient illustrates the current dilemma of pancreatic cancer treatment; therefore, we need to explore pancreatic cancer more deeply in basic research.

Neoadjuvant chemotherapy provides a time window for the establishment of adequate collateral circulation, especially for invasion of the C and D types of the PV/SMV[12]. In this group, there were 7 cases of C-type or D-type PV/SMV invasion with obvious collateral circulation around the pancreas. Fibrotic changes and tissue adhesions after neoadjuvant chemotherapy led to the disappearance of the original, clear anatomical layers. All of the above factors increase the difficulty of performing laparoscopic operations. Accurate preoperative assessment and delicate intraoperative surgery are essential. Owing to the low adiposity of patients after chemotherapy, the artery-first approach usually easily exposes the SMA and SMV, reducing intraoperative bleeding and allowing the resectability of the tumor to be reevaluated in the early stage of surgery[48]. When the branches of the PV/SMV and pancreatic parenchyma of pancreatic cancer patients are dissected after chemotherapy, bleeding easily occurs. Excellent psychological quality and laparoscopic suture hemostasis techniques are necessary. When incising the pancreatic parenchyma with an ultrasonic knife, surgeons should adopt a strategy of progressing from shallow to deep, cutting a small amount at a time, and advancing slowly to avoid massive bleeding and bile duct injury; this strategy is also conducive to confirming the presence of the pancreatic duct. Of course, intraoperatively, we could have used indocyanine green fluoroscopy to visualize the CBD[49], but we did not use it because the frequent switching between 3D laparoscopy and fluoroscopy lenses affected the surgical process. In the future, with the integration of fluorescence and 3D, surgery will be safer. Owing to the thickening of the PV/SMV wall after chemotherapy, laparoscopic portal reconstruction becomes relatively simple in chemotherapy recipients compared with nonrecipients.

The ideal extent of lymph node dissection for pancreatic neck-body cancer has not been determined[18]. Although the lymph node positivity rate significantly decreased after neoadjuvant chemotherapy and preoperative imaging revealed that the degree of lymph node metastasis did not exceed the scope of resection[50], there was still a risk of lymph node metastasis in group 13. Therefore, the precise indications for subtotal DP need to be rigorously established. Moreover, this study has several limitations: (1) A larger sample size and prospective studies are needed to further confirm the effectiveness of subtotal DP; (2) The study had strict inclusion criteria and required good patient compliance; and (3) Our follow-up is not yet complete, and not all patients’ recurrence and metastasis data were obtained in full. The short period of follow-up may reveal the high incidence of recurrence. Longer follow-up periods are needed to determine the recurrence site and to further verify the oncological effects of subtotal DP.

CONCLUSION

In conclusion, laparoscopic subtotal DP for pancreatic neck-body cancer with neoadjuvant and adjuvant chemotherapy is safe and feasible in elective cases.