Conversion surgery for positive peritoneal washing cytology in pancreatic cancer

Abstract

Positive peritoneal washing cytology (PPC) of pancreatic carcinoma is defined as distant metastasis in the American Joint Committee on Cancer or Union for International Cancer Control’s tumour, node, metastases classification. However, surgical resection was believed to be the only method that prolong survival; thus, many institutions perform pancreatectomy for PPC, despite the unfavourable prognosis. Therefore, a more preferable alternative treatment for PPC is required. A 64-year-old man with resectable pancreatic tail cancer presented to our hospital. PPC was detected at first laparotomy; thus, pancreatectomy was avoided and gemcitabine with nabpaclitaxel (GnP) was administered. After four courses of GnP treatment, PPC converted to negative, as evaluated by abdominal port cytology. Thus, distal pancreatectomy was performed, and R0 resection was achieved. He has been healthy for more than 24 months since the first laparotomy. Initial chemotherapy with the intention of converting the cytological status followed by surgical treatment might become a useful treatment strategy for PPC.

Background

Pancreatic carcinoma is one of the worst types of carcinoma,¹ and postoperative recurrence is very frequent²; thus, surgical indication is strictly defined by several guidelines.^3–6 Although resection of metastatic carcinoma is contraindicated, its indication for positive peritoneal washing cytology (PPC) is controversial. In the American Joint Committee on Cancer (AJCC) Staging Manual (eighth edition)⁴ and the Union for International Cancer Control’s (UICC)-tumour, node, metastases classification (TNM) classification (eighth edition),⁵ PPC is defined as distant metastasis; therefore, the National Comprehensive Cancer Network (NCCN) guideline does not recommend surgical resection.³ Meanwhile, in the Classification of Pancreatic Carcinoma (fourth English edition) by the Japan Pancreatic Society, PPC alone, without disseminated nodule is not defined as distant metastasis.⁶ Hence, surgical indication for PPC also remains controversial, and many institutions now perform resection. However, several reports revealed worse prognosis of PPC compared with negative peritoneal cytological status^7–12 and suggested the need for a more preferable treatment, such as preoperative treatment or strong adjuvant chemotherapy.

Conversion surgery for unresectable (UR) PC has a more favourable prognosis, suggesting a better chance of achieving cure.¹³ This strategy suggested that surgical treatment for PC that responds favourably to non-surgical treatment could be expected to achieve long-term survival. Therefore, initial treatment intending conversion surgery might become a good strategy even for PPC. We herein present a case report of conversion surgery of PPC using gemcitabine with nab-paclitaxel (GnP) therapy; we also introduce the use of an abdominal port to evaluate the response of peritoneal washing cytology by chemotherapy.

Case presentation

A 64-year-old man presented to his local hospital with a 6-month history of abdominal colic pain and weight loss. Abdominal CT scan revealed a pancreatic tail tumour, and he was referred to our hospital. He was being treated for hypertension and had a history of surgical treatment for right inguinal hernia and femoral fracture. There was no family history of carcinoma. Laboratory data revealed high levels of Duke pancreatic monoclonal antigen type 2 (Dupan-2; 3732 U/mL) and S-pancreas-1 antigen (Span-1; 76.9 U/mL), but the carcinoembryonic antigen (CEA; 2.1 ng/mL) and cancer antigen 19–9 (CA19-9; 15.4 U/mL) levels were within normal limits. CT revealed a 35 mm mass in the pancreatic tail that had invaded the splenic artery, splenic vein and left kidney. However, the main tumour had not invaded other major vessels such as the celiac artery or portal vein (figure 1A,B). Endoscopic ultrasonography-guided fine-needle aspiration cytological evaluation confirmed the diagnosis as pancreatic ductal adenocarcinoma. Positron emission tomography (PET) findings did not identify any distant metastasis, and the maximum standard uptake value (SUV_max) of the main tumour was 6.2 (figure 2A). Based on these findings, we diagnosed the patient with resectable PC, and surgical resection was planned. Intraoperative findings showed neither peritoneal nor liver metastasis. However, PPC revealed aggregation of cancer cells by Papanicolaou stain and was classified as positive findings (class V). Therefore, we categorised it as M1, according to the AJCC or UICC-TNM classification.

Figure 1.

CT images. (A, B) Before chemotherapy. (C, D) After gemcitabine with nab-paclitaxel (GnP) therapy. Tumour volume decreased after GnP therapy.

Figure 2.

Positron emission tomography images. (A) Before chemotherapy. (B) Standard uptake value (SUV_max) after gemcitabine with nab-paclitaxel (GnP) therapy. The maximum SUV_max value of the main tumour after GnP therapy decreased after treatment compared with that before treatment.

Treatment

At the first laparotomy, we did not perform pancreatectomy and instead implanted an abdominal port (BARDPORT; Bard Access Systems, Salt Lake City, Utah, USA) to evaluate the response to chemotherapy through PPC. The catheter of the abdominal port was indwelled in the pelvic area, and the port was implanted into the subcutaneous area of the right lower quadrant (figure 3A,B). The operation time was 137 min, and blood loss was 10 mL.

Figure 3.

Abdominal port. (A) Patient with abdominal port. (B) CT image of the abdominal port. The port was implanted into the subcutaneous area of the right lower quadrant of the abdomen. The catheter was indwelled in the pelvic area.

Fifteen days after the primary operation, he received GnP according to the standard regimen (ie, gemcitabine (1000 mg/m²) and nab-paclitaxel (125 mg/m²) injection on days 1, 8 and 15, repeated every 4 weeks). During the course of chemotherapy, no major adverse events occurred. After two courses of GnP treatment, PPC was performed using the implanted abdominal port. We injected 100 mL of warm physiological saline through the abdominal port and collected it from the abdominal cavity. Cytological findings were negative (class II); thus, we continued the same chemotherapy regimen for two additional courses. After a total of four courses of GnP treatment, washing cytology of the abdominal port again revealed no apparent malignant cells (class III). Furthermore, the tumour size decreased to 20 mm, and the response evaluation criteria in solid tumors (RECIST) evaluation was defined as partial response (figure 1C,D). PET SUV_max also decreased to 2.8 (figure 2B). There was no evident distant metastasis, and the elevated serum tumour markers gradually decreased (Span-1, 34.9 U/mL; DUPAN-2, 1617 U/mL). Based on these findings, we planned surgical treatment 5 months after the first laparotomy.

Intraoperative PPC again revealed negative findings (class II), and curative resection was planned. Distal pancreatectomy was performed as an en bloc resection, and because the tumour invaded the left kidney, we performed additional partial resection of the left kidney (figure 4A,B). Operation time was 348 min, and blood loss was 350 mL.

Figure 4.

(A, B) Intraoperative findings and the resected tumour. Distal pancreatectomy was performed as an en bloc resection with additional partial resection of the left kidney. The white arrow shows the renal resection margin. (C, D) Histopathological findings. (C) Stromal hyperplasia with myxoma-formed change was conspicuous, suggesting the therapeutic effect of the treatment. (D) Invasive findings of cancer in a renal capsule (black arrows).

Outcome and follow-up

Histopathological findings showed a 17×15 mm, moderately differentiated tubular adenocarcinoma at the pancreatic tail. The main tumour invaded the anterior and retroperitoneal serosa, splenic vein and artery, and renal capsule. Six metastatic lymph nodes were found among 37 resected lymph nodes. The effect of chemotherapy was defined as grade IIb in the Evans classification¹⁴ (figure 4C). According to the UICC-TNM Classification of Malignant Tumours (eighth edition),⁵ the tumour was classified as ypT1N1M0, ypStage IIB. Although direct invasion of the renal capsule was detected, there were no residual carcinoma cells either at the renal resection margin or the pancreatic resection margin (figure 4D); thus, R0 resection could be achieved.

His postoperative course was uneventful, and he was discharged from the hospital 13 days after surgery. The patient was subsequently treated with S-1 as adjuvant chemotherapy for 6 months. Postoperative follow-up, which includes laboratory tests and CT scans, was performed at least once every 2–3 months. He has been doing well for more than 24 months after the first laparotomy and 19 months after surgical resection.

Discussion

Some studies suggested the benefit of surgical resection for PPC.^{11 15–17} However, postoperative survival of PPC was unfavourable. Furthermore, PPC itself could be a predictive risk factor of malignant prognosis,^8–11 and recurrence, especially at the peritoneum, was frequently found even after curative resection.^{8 10 11} Therefore, local resection for PPC was not enough to prolong survival. Nevertheless, many institutions aggressively perform surgical resection for PPC because no alternative strategy that prolongs survival has been established aside from upfront surgery combined with adjuvant chemotherapy.

Reports on conversion surgery for UR PC after non-surgical anticancer treatment have been increasing because it could prolong survival and increase the chances to achieve cure.^{13 18} Considering that PPC is equivalent to M1 disease,^{4 5} we believe that intending conversion surgery for PPC might be a useful strategy for PPC. Satoi et al demonstrated that intravenous and intraperitoneal (IP) paclitaxel with S-1 treatment for peritoneal dissemination in PPC patients can achieve a cytological conversion rate of 55%, and conversion surgery was successfully performed for 24% of the patients. Furthermore, the median survival time after conversion surgery achieved long-term survival (27.8 months).¹⁹ These results suggested that chemotherapy with the intention of conversion surgery could improve the prognosis even for peritoneal dissemination.

When considering intending conversion surgery for PPC, we should discuss the adequate method for initial treatment. IP and intravenous therapies are a good drug delivery system, with expected direct exposure to peritoneal nodule. However, if only floating cancer cells were evident without mass metastasis in the peritoneum, IP chemotherapy targeting peritoneal nodules does not necessarily have an important role. Furthermore, previous reports on clinicopathological characteristics showed that PPC correlated with local status, such as tumour size, anterior serosa invasion or portal vein invasion,^{7 8} which might supply cancer cells to the peritoneal cavity. Thus, treatment intended to shrink the main tumour might also be important. Taken together, we hypothesised that if only PPC was evident, strong systemic chemotherapy might be effective in diminishing the floating cancer cells and keeping the negative status by responding to the main tumour. Recent new chemotherapeutic treatments, such as GnP or FOLFIRINOX, showed high response rates for PC^{20 21}; thus, these strong systemic chemotherapy might be preferable to achieve conversion surgery for PPC. However, whether these systemic chemotherapies are enough to diminish cancer cells in the peritoneal cavity remains unclear. Our case is valuable because it is the first report on successful conversion of PPC by GnP.

To determine the optimal timing of conversion surgery for PPC, the cytological status should be examined by evaluating the response to chemotherapy. However, to evaluate the cytological status, staging laparotomy or laparoscopy should be performed. These methods are invasive, require general anaesthesia and need a drug holiday period. Thus, they cannot be performed frequently. For easier evaluation of the cytological status, an abdominal port was placed during the initial surgery in our current case. This cytological examination was feasible and less invasive compared with re-laparotomy or laparoscopic examination. In addition, this port was effective in evaluating the response to chemotherapy and to determine the timing of surgical resection. The effectiveness of the abdominal port was previously reported for ovarian cancer, which revealed high accuracy of cytology (sensitivity: 0.82, specificity: 1.00, positive predictive value: 1.00, negative predictive value: 0.88) in 40 patients.²² Although further studies are needed to confirm the accuracy of this examination in PC, serial examination using abdominal port cytology might aid in assessing the response to chemotherapy with the aim of conversion surgery.

This case report revealed the negative aspect of not performing staging laparoscopy. As noted in previous studies, staging laparoscopy should be performed in cases such as ours.^{23 24} According to the Cochrane Database Systemic Review,²⁵ with diagnostic laparoscopy, 21% of unnecessary laparotomies with curative intent for resection could be avoided by staging laparoscopy. In this case, we diagnosed the tumour as resectable before the first laparotomy, based on the CT findings. However, PPC was evident in intraoperative findings, and the diagnosis was changed to UR PC. Fortunately, this patient’s recovery from the initial laparotomy was quick and he was able to start with chemotherapy immediately, but for the assessment of PPC, examination with a laparoscope is less invasive than laparotomy and is suitable. Considering this case, we now try to perform staging laparoscopy for relatively advanced PC cases such as those with huge tumour or those with suspected invasion of the anterior serosa of the pancreas before surgical treatment.

In conclusion, this is the first case report of the use of conversion surgery for PPC. Although whether conversion surgery for PPC could improve long-term survival remains unclear, the observation period of our case has reached 24 months without distant metastasis, and thus, the patient might be expected to achieve long-term survival. This report suggests that conversion surgery for PPC could be successfully obtained by strong systemic chemotherapy. A previous study revealed that conversion surgery for PC can possibly prolong survival¹⁰; therefore, this treatment might become a useful treatment strategy for PPC in the near future.

Learning points.

• The prognosis after surgical treatment for positive peritoneal washing cytology (PPC) in pancreatic carcinoma is poor and a better treatment strategy is desired.

• Intending conversion surgery for PPC might be a useful treatment because strong chemotherapy could convert the cytological status to negative.

• An abdominal port is useful for evaluating the response to chemotherapy through PPC.