Surgical treatment of pancreatic metastases: a clinicopathologic study of 43 patients

doi:10.21203/rs.3.rs-1989969/v1

Download PDF

Research Article

Surgical treatment of pancreatic metastases: a clinicopathologic study of 43 patients

https://doi.org/10.21203/rs.3.rs-1989969/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Purpose It is rare to find surgically-resectable metastatic lesions that are limited to the pancreas, by preoperative imaging. In this retrospective study, we reviewed the histopathologic findings in pancreatic surgical resection specimens to determine the factors that contribute to the prognostic improvement of surgical treatment for pancreatic metastases.

Methods From 1979 to 2018, 43 patients underwent pancreatic resection for metastatic tumors in the Department of Hepato-Biliary-Pancreatic Surgery of the Cancer Institute of the Japanese Foundation for Cancer Research. Survival curves were created using the Kaplan–Meier method, and comparisons of survival were performed using the log-rank test.

Results Primary tumors comprised renal cell carcinoma (RCC), colorectal carcinoma (CRC), and miscellaneous malignancies (MM) in 23, 9, and 11 cases, respectively. Histopathologically, the patterns of metastases were classified as hematogenous (32), lymphatic (8), and peritoneal (3). Regarding the prognosis after metastasectomy of the pancreas, there were significant differences for disease-free interval (< 5 years versus ≥5 years) and primary tumor (RCC versus CRC/MM). The rate of secondary or accompanying lymph node metastasis around the main metastasis was 5/9 (56%) for CRC, 5/11 (45%) for MM, and 1/23 (4%) for RCC.

Conclusion Survival outcomes are more favorable for disease-free intervals ≥ 5 years, pancreatic metastasis from RCC, and with lymphatic metastasis. Lymph node dissection is not strictly necessary for the surgical removal of pancreatic metastasis from RCC, whereas this is highly recommended for patients with metastasis from CRC or MM.

Pancreatic metastasis

Surgical treatment

Clinicopathologic study

Survival curve

Disease-free interval

Lymph node dissection

Pancreatic metastasis detected at autopsy is occasionally found as the “tip of the iceberg” in the terminal stages in patients with advanced cancers, and it is rare to find surgically-resectable metastatic lesions that are limited to the pancreas on preoperative imaging. Until recently, surgical treatment of pancreatic metastases has been unsuccessful, with recurrence in another organ soon after resection because few modalities accurately diagnose the systemic spread of recurrent tumors. However, advances in clinical imaging, especially magnetic resonance imaging and positron emission tomography, have made the discovery of recurrent tumors of the pancreas relatively simple.

In this retrospective study, we reviewed the histopathologic findings in pancreatic surgical resection specimens to determine the factors that contribute to the prognostic improvement of surgical treatment for pancreatic metastases.

From 1979 to 2018, 43 patients underwent pancreatic resection for metastatic tumors at our institute, which corresponded to 2.3% of 1896 pancreatic resections for pancreatic neoplasms. Patients with involvement of the pancreas by direct extension of carcinomas of the gastrointestinal tract, namely carcinomas of the stomach, duodenum, extrahepatic bile duct, and ampulla of Vater, were excluded from our study. The patients’ ages ranged from 46 to 83 years (average, 63 years), and the male-to-female ratio was 1.4:1.

Forty patients had metastatic pancreatic tumors, excluding three synchronous tumors, detected less than 6 months after the completion of therapy for primary neoplasms or at routine yearly ultrasonography or computed tomography (CT). Although pancreatic metastases from renal cell carcinoma (RCC) were relatively easy to differentiate from primary pancreatic tumors because of the rich vascularity and sharp demarcation on enhanced CT of RCCs, it was difficult to differentiate between metastatic and primary pancreatic tumors in other cases. Therefore, the final diagnoses in the 43 cases were established by histologic similarities between pancreatic and primary tumors. Three patients with metastatic pancreatic tumors from colorectal carcinoma (CRC) and two with RCC metastasis underwent concurrent pancreatic and partial hepatic resections.

All resected pancreatic sections were fixed in 10% formalin and sliced serially at 5–8-mm intervals along the plane perpendicular to the main pancreatic duct (MPD), generally. All slices were processed following standard procedures for hematoxylin and eosin-stained histologic preparations, and all slides were examined microscopically.

Survival was calculated from the dates of initial laparotomy and the last known follow-up or death. Survival curves were prepared using the Kaplan–Meier method, and comparisons of survival were performed using the log-rank test. A p-value of <0.05 was considered statistically significant. Disease-free intervals (DFIs), defined as the interval between primary and metastatic tumors, were calculated as the period from the date of initial treatment for the primary tumor to the date of the first diagnosis of pancreatic metastasis.

No adjuvant therapy was performed after pancreatic resection for metastases from RCC, whereas some patients with metastases from CRC and miscellaneous malignancies (MM) received supplementary therapy after resection.

The clinicopathologic characteristics of the pancreatic metastases of the 43 patients are shown in Table 1. Most pancreatic metastases were solitary, and the number of metastases from RCC was unable to be determined in only one case. The DFI in 22 cases, including 3 with RCC (who simultaneously underwent nephrectomy for primary RCC), was less than 5 years. The DFI in 21 cases, including 14 with RCC, was more than 5 years. The primary organs in 22 cases with DFI < 5 years varied, as follows: 9 RCCs, 7 MMs, and 6 CRCs. The median DFI of all 43 cases was 4 years and 2 months; specifically for RCC, CRC, and MM, the DFIs were 7 years and 9 months, 4 years and 2 months, and 4 years and 4 months, respectively. On the basis of histopathologic findings and anatomical relationships between the pancreatic parenchyma and metastatic tumors, the patterns of metastasis were classified as hematogenous (32 cases), lymphatic (8 cases), and peritoneal (3 cases). Regarding the three patterns of pancreatic metastasis, a typical gross view of a cut section and histopathologic findings are shown in Figure (Fig.) 1, in which lymphatic or peritoneal metastases frequently appeared relatively poorly demarcated.

The histopathologic characteristics of the pancreatic metastases are shown in Table 2. Lymphatic invasion [1] was observed in 10 (23%), and venous invasion [1] was detected in 11 (26%) of the 43 samples. Additionally, the primary organs varied, namely, 4 MMs, 3 RCCs, and 3 CRCs in cases with lymphatic invasion and 5 CRCs, 3 RCCs, and 3 MMs in cases with venous invasion. Tumor extension into the MPD [1] was observed in 9 (21%) of the 43 cases, comprising 4 RCC, 3 MM, and 2 CRC cases. Figure 2 shows the macroscopic and microscopic appearances of the resected specimen from a typical patient with tumor extension into the MPD. The patient underwent resection of the metastases to the body of the pancreas from non-small-cell pulmonary carcinoma 4 years and 9 months previously. Secondary or accompanying lymph node metastasis surrounding the metastatic tumor was observed in 11 cases (5 CRCs, 5 MMs, and 1 RCC), with a metastatic rate to lymph nodes of 5/9 (56%), 5/11 (45%), and 1/23 (4%) for CRCs, MMs, and RCCs, respectively.

Figures 3 and 4 show the details of pancreatic infiltration from metastatic tumors to the peripancreatic lymphatic space or connective tissue in eight and three patients, respectively. Eight cases of lymphatic metastasis are shown in Fig. 3, in which five were characterized as CRC. As shown in the schematic diagrams of the cut sections for the eight cases, metastatic tumors were present in the space between the pancreatic tissue and the duodenum/stomach wall. Therefore, it can be easily speculated that pancreatic infiltration of the metastases occurs from outside the pancreas. Secondary or accompanying lymph node metastasis was detected in four of the eight cases. As shown in Fig. 4, three cases had peritoneal metastases. It is highly characteristic that these cases included two cases of primary ovarian carcinoma.

Fig. 5a shows the 5-year survival for all cases, which was 55.1%. Comparisons of survival by the DFI, primary tumor, and metastatic pattern are shown in Fig. 5b–d, respectively. There were significant differences in the survival curves between the different DFIs (Fig. 5b) and primary tumors (Fig. 5c), and there was a convincing trend (p=0.049) toward improved survival by the pattern of metastasis; i.e., 66.0% 5-year survival (Fig. 5d) in the hematogenous group versus 37.5% in the lymphatic group. As noted, the survival curves between the different DFIs (Fig. 5b) and primary tumors (Fig. 5c) were significantly different (p=0.007 and p<0.001, respectively), with 78.7% 5-year survival in the group with DFI ≥ 5 years and 85.3% 5-year survival in the RCC group.

The widely accepted traditional criteria for surgical resection of metastatic tumors provided by Thomford et al. [2] are the following: (1) the patient is a good candidate for surgical intervention, (2) the primary malignancy is controlled, and (3) there is no evidence of metastatic disease elsewhere in the body. Reddy and Wolfgang [3] suggested the following criteria for the selection of patients for pancreatic metastasectomy: primary cancer type associated with successful outcomes, control of the primary cancer site, isolation of metastases, resectability of metastases, and patient fitness to tolerate pancreatectomy. We typically follow these criteria but have recently expanded them. For metastatic tumors to the pancreas from RCC or CRC, even if several metastases to the liver exist simultaneously, we aim to resect the pancreatic as well as the hepatic metastases if there are fewer than two or three liver metastases. We use this approach because it is widely known that patients with a few metastases to the liver from RCC or CRC have a better prognosis than that of patients with metastases from other malignancies.

The 5-year actuarial survival for all patients after metastasectomy of the pancreas is commonly reported to be approximately 60%; e.g., Bahra et al.: 63.8% (n = 20) [4], Strobel et al.: 56.8% (n = 44) [5], and Konstantinidis et al.: 61% (n = 40) [6]. Likewise, in our study, the 5-year survival rate was 55.1% (n = 43). Although there was a significant difference between the < 5-year and ≥ 5-year DFI groups in our study, Choi et al. [7] reported that DFI was not associated with prolonged survival.

Survival after the resection of pancreatic metastatic lesions varies greatly among different primary tumor types. RCC has the best prognosis [7, 8], with a 5-year survival of 57%–75% [5, 9, 10]. In our study, the survival rate in patients with RCC was also the most favorable at 85.3% among the different primary surgical sites, namely the colon/rectum, lung, stomach, breast, ovary, skin, uterus, and sarcoma [3, 6, 11, 12], which generally have less favorable outcomes after metastasectomy than that of RCC [6]. Survival after pancreatic resection for metastatic lesions from CRC is somewhat favorable, with a median survival of 17 months according to Sperti et al. [13] and a 5-year survival rate of 33.3% in our study. Thus, the surgical removal of secondary tumors to the pancreas, excluding metastases from RCC, is insufficient for prognostic improvement. We found no previous reports evaluating the prognosis by metastatic patterns (hematogenous, lymphatic, and peritoneal); however, there was a convincing trend toward improved survival in the hematogenous group, in our study, with a 66.0% 5-year survival rate.

Few studies [14]have reported the microscopic invasion of vessels by metastatic lesions to the pancreas. Reddy et al. [14] reported that 13 of 37 patients (35%) had perineural or vascular invasion. In our specimens, lymphatic invasion was observed in 10 of 43 patients (23%) and vascular invasion in 11 of 43 patients (26%). Tumor extension into the MPD [15, 16] is a characteristic of metastatic tumors of the pancreas. Koide et al. [15]reported that tumor extension into the MPD was detected in 3 of 5 cases of pancreatic metastases from RCC, whereas this finding was observed in 9 of 43 cases (21%) in our study. Importantly, it is noteworthy that the MPD may extend far from the main tumor. On the basis of our histopathologic findings, wedge or partial resection of the pancreas is not usually performed for pancreatic metastasis. In contrast to pulmonary or hepatic metastasectomy, pancreatic metastases are associated with a rather high risk of incomplete metastasectomy; i.e., an incompletely excised tumor or post-operative leakage of pancreatic juice from the cut surface of the pancreatic parenchyma.

There have also been few reports on secondary or accompanying lymph node metastases adjacent to metastatic lesions to the pancreas [5, 14]. Strobel et al. [5] described lymph node involvement in 10 of 44 patients (26%); Reddy et al. [14] reported this finding in 13 of 39 patients (35%). In our study, lymph node involvement was detected in 11 of 43 patients (26%). Notably, the rate of secondary or accompanying lymph node metastasis adjacent to the pancreas was profoundly different from that of the primary tumor (4/9 (44%) in CRCs, 3/11 (27%) in MMs, and 1/23 (4%) in RCCs) in our study. These results indicate that standard pancreatectomy, including pancreatoduodenectomy or distal pancreatectomy, with lymph node dissection is not strictly necessary for the surgical treatment of metastatic tumors to the pancreas from RCC, whereas this approach is highly recommended for pancreatic metastases from CRC or MM.

It would be of interest to determine whether lymph node involvement arises from the main metastasis (i.e., tertiary metastasis) or whether this occurs simultaneously. Generally, lymphatic drainage for the right colon (cecum, ascending colon, and transverse colon) runs along the ileocolic or middle colic vessels toward the root of the superior mesenteric artery and vein, adjacent to the head and body of the pancreas [17, 18]. Charnsangavej et al. [19] described nodal metastases that occurred from the right colon to the nodes around the head and body of the pancreas, and stated that further progression of these principal nodal metastases is likely to involve the pancreatic parenchyma. For this reason, pancreatic metastasis from the right colon may simultaneously accompany lymph node involvement. However, we consider that quite a few lymph node metastases around the metastatic lesion to the pancreas are likely secondary because the cases whose main metastasis was histopathologically classified as hematogenous frequently showed simultaneous nodal metastases, in our study. Specifically, four of our nine cases with pancreatic involvement from CRC were hematogenous metastases, of which three showed nodal metastasis (3/4, 75%); whereas five of nine cases were lymphatic metastases, of which two showed nodal metastasis (2/5, 40%).

We performed retrospective chart reviews of 43 patients who underwent surgical resection of the pancreas for metastatic tumors. The primary malignancies comprised 23 RCCs, 9 CRCs, and 11 MMs. The histological patterns of metastasis were classified as hematogenous (32 cases), lymphatic (8 cases), and peritoneal (3 cases).

Regarding the prognosis after metastasectomy of the pancreas, there were significant differences for the different DFIs (< 5 year versus ≥ 5 years) and primary tumors (RCC versus CRC/MM), as well as a convincing trend and nearly significant difference in the survival rates between the hematogenous group and the lymphatic group.

The rates of secondary or accompanying lymph node metastasis adjacent to the pancreas were 5/9 (56%) for CRC, 5/11 (45%) for MM, and 1/23 (4%) for RCC, respectively, whereas these rates were 4/8 (50%) for the lymphatic group, 1/3 (33%) for the peritoneal group, and 6/32 (19%) for the hematogenous group, respectively. Consequently, lymph node dissection is not strictly necessary for the surgical removal of pancreatic metastases from RCC, whereas this is highly recommended for pancreatic metastases from CRC or MM.

Authors’ contributions

・Study conception and design: M. Seki, M. Katori, Y. Kato

・Acquisition of data: A. Saiura, Y. Takahashi, Y. Inoue, M. Takamatsu

・Analysis and interpretation of data: M, Seki, N. Yamamoto

・Drafting of manuscript: M. Seki, A. Saiura

・Critical revision of manuscript: Y. Takahashi, M. Takamatsu

Statements and Declarations: The protocol for this research project was approved by our institutional ethics review committee (registration number: 2021-GB-105), and the study confirms to the provisions of the Declaration of Helsinki.

Informed consent: All patients consented to the use of their clinical data for research purposes when the operations were performed.

Competing Interests and Funding: The authors have no conflict of interest or financial support for the study to declare.

Acknowledgments: We thank Jane Charbonneau, DVM, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.

Kato Y, Mukai K (2003) Histological findings of pancreatic tumors. In: Classification of pancreatic carcinoma, 2nd English ed., Kawarada (ed.). Kanehara, Tokyo. http://www.so-net.ne.jp/medipro/kanehara/
Thomford NR, Woolner LB, Clagett OT (1965) The surgical treatment of metastatic tumors in the lungs. J Thorac Cardiovasc Surg 49:357-363.
Reddy S, Wolfgang CL (2009) The role of surgery in the management of isolated metastases to the pancreas. Lancet Oncol 10:287-293
Bahra M, Jacob D, Langrehr JM, Glanemann M, Schumacher G, Lopez-Haenninen E, Neuhaus P (2008) Metastatic lesions to the pancreas. When is resection reasonable? Chirurg 79:241-248
Strobel O, Hackert T, Hartwig W, Bergmann F, Hinz U, Wente MN, Fritz S, Schneider L, Buechler MW, Werner J (2009) Survival data justifies resection for pancreatic metastases. Ann Surg Oncol 16:3340-3349
Konstantinidis IT, Dursun A, Zheng H, Wargo JA, Thayer SP, Castillo CF, Warshaw AL, Ferrone CR (2010) Metastatic tumors in the pancreas in the modern era. J Am Coll Surg 211:749-753
Choi TW, Kim SH, Shin CI, Han JK, Choi BI (2015) MDCT findings of pancreatic metastases according to primary tumors. Abdom Imaging 40:1595-1607
Thompson LDR, Heffess CS (2000) Renal cell carcinoma to the pancreas in surgical pathology material. Cancer 89:1076-1088
Sweeney AD, Fisher WE, Wu MF, et al (2010) Value of pancreatic resection for cancer metastatic to the pancreas. J Surg Res 160:268-276
Law CHL, Wei AC, Hanna SS, Al-Zahrani M, Taylor BR, Greig PD, Langer B, Gallinger S (2003) Pancreatic resection for metastatic renal cell carcinoma: presentation, treatment, and outcome. Ann Surg Oncol 10:922-926
Roland CF, Heerden JA (1989) Nonpancreatic primary tumors with metastasis to the pancreas. Surg Gynecol Obstet 168:345-347
Adsay NV, Andea A, Basturk O, Kilinc N, Nassar H, Cheng JD (2004) Secondary tumors of the pancreas: an analysis of a surgical and autopsy database and review of the literature. Virchows Arch 444:527-535
Sperti C, Pasquali C, Berselli M, Frison L, Vicario G, Pedrazzoli S (2009) Metastasis to the pancreas from colorectal cancer: is there a place for pancreatic resection? Dis Colon Rectum 52:1154-1159
Reddy S, Edil BH, Cameron JL, Pawlik TM, Herman JM, Gilson MM, Campbell KA, Schulick RD, Ahuja N, Wolfgang CL (2008) Pancreatic resection of isolated metastases from nonpancreatic primary cancers. Ann Surg Oncol 15:3199-3206
Koide N, Yokoyama Y, Oda K, Nishio H, Ebata T, Abe T, Igami T, Nimura Y, Nagino M (2008) Pancreatic metastasis from renal cell carcinoma: results of the surgical management and pathologic findings. Pancreas 37:104-107
Seki M, Tsuchiya E, Hori M, Nakagawa K, Ohta H, Ueno M, Takahashi T, Ohashi K, Ishikawa Y, Yanagisawa A (1998) Pancreatic metastasis from lung cancer. Int J Pancreatol 24:55-59
Gray H (1985) Lymphatic vessels and nodes of the abdomen and pelvis. In: Gray’s anatomy 30^th ed., Clemente (ed.). Lea & Febiger, Philadelphia.
Goligher J (1984) Surgery of the anus, rectum and colon. Bailliere Tindall, London.
Charnsangavej C, Whitley NO (1993) Metastases to the pancreas and peripancreatic lymph nodes from carcinoma of the right side of the colon: CT findings in 12 patients. AJR Am J Roentgenol 160:49-52

Tables 1 and 2 are available in the Supplementary Files section

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Surgical treatment of pancreatic metastases: a clinicopathologic study of 43 patients

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Results

Discussion

Conclusion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1