Annals of Surgical Oncology Sign the Guestbook
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
10.1245/ASO.2006.05.002
Annals of Surgical Oncology 13:582-594 (2006)
© 2006 Society of Surgical Oncology
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Raut, C. P.
Right arrow Articles by Evans, D. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Raut, C. P.
Right arrow Articles by Evans, D. B.

Original Article

Intraductal Papillary Mucinous Neoplasms of the Pancreas: Effect of Invasion and Pancreatic Margin Status on Recurrence and Survival

Chandrajit P. Raut, MD1, Karen R. Cleary, MD2, Gregg A. Staerkel, MD2, James L. Abbruzzese, MD3, Robert A. Wolff, MD3, Jeffrey H. Lee, MD4, Jean-Nicolas Vauthey, MD1, Jeffrey E. Lee, MD1, Peter W. T. Pisters, MD1 and Douglas B. Evans, MD1

1 Department of Surgical Oncology, Unit 444, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, Texas 77030
2 Department of Pathology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, Texas 77030
3 Department of Gastrointestinal Medical Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, Texas 77030
4 Department of Gastrointestinal Medicine and Nutrition, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, Texas 77030

Correspondence: Address correspondence and reprint requests to: Douglas B. Evans, MD; E-mail: devans{at}mdanderson.org.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: The natural history and prognosis for patients with intraductal papillary mucinous neoplasms (IPMN) with and without invasion remain poorly defined. This study evaluated the outcome after pancreatectomy for IPMN according to the pancreatic transection margin status and the presence or absence of invasive carcinoma.

Methods: Data from a prospective pancreatic tumor database and medical records were reviewed for all patients who underwent pancreatic resection for IPMN at our institution between July 1990 and July 2003. Surgical specimens were re-reviewed by a single pathologist.

Results: IPMN was diagnosed in 35 (26%) of 137 patients who underwent pancreatic resection for cystic neoplasms. Invasive IPMN was confirmed in 13 (37%) of 35 patients. Noninvasive IPMN was found in 22 (63%) of 35 patients; pathology re-review changed the original diagnosis from invasive to noninvasive IPMN in 6 patients. Noninvasive IPMN was found at the final pancreatic margin in eight patients; none developed recurrent disease at a median follow-up of 34 months. Recurrent disease was identified in 7 (58%) of 13 patients with invasive IPMN and in none with noninvasive IPMN. The median overall survival was 22.9 and 84.9 months in patients with invasive and noninvasive IPMN, respectively (P = .0009).

Conclusions: Distinction between invasive and noninvasive IPMN is essential in estimating prognosis and determining the need for adjuvant therapy and the frequency of follow-up surveillance. Noninvasive IPMN, even if present at the pancreatic margin, was not associated with recurrent disease. In contrast, invasive IPMN was associated with early recurrence and short survival.

Key Words: Intraductal papillary mucinous neoplasm • Invasive adenocarcinoma • Pancreas • Survival • Pancreatic margin


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Intraductal papillary mucinous neoplasms (IPMNs) form an increasingly well-recognized, albeit rare, category of pancreatic neoplasms.1 IPMNs are defined by the World Health Organization as papillary mucin-producing neoplasms arising in the main pancreatic duct or its major branches.2 Although the true incidence of IPMNs remains unknown, they are thought to represent approximately 1% of pancreatic neoplasms and 24% of pancreatic cystic neoplasms.3–6 IPMNs account for 8% to 20% of pancreatic resections for malignancy in modern surgical series.7,8

Relatively little is known about the molecular genetic events surrounding the development of IPMN, although these events are currently under investigation. IPMNs have a lower frequency of mutations in K-ras (40%–60%), p53 (50%), and MUC1 (25%–38%) than do pancreatic ductal adenocarcinomas, which have mutation rates of 70% to >90% in these genes.9 However, expression of the high-molecular-weight glycoproteins MUC2 mucin and MUC5 mucin is higher in IPMNs than in pancreatic ductal adenocarcinomas (>80% vs. 0%–13%).4 Furthermore, whereas 50% of pancreatic ductal adenocarcinomas demonstrate loss of DPC4 gene expression, 84% to 100% of IPMNs express the DPC4 protein, thus suggesting that these neoplasms are genetically distinct.8,10 In addition, 32% of IP-MNs have loss of heterozygosity of STK11/LKB1; the biological significance of this is uncertain.11 Finally, cyclooxygenase 2 is expressed by 70% of IP-MNs, thus potentially rendering these tumors candidates for chemoprevention with cyclooxygenase inhibitors.12

IPMNs are distinguished from mucinous cystic neoplasms by their direct communication with the main or branch pancreatic ducts, their proximal location, male predominance, occurrence in older patients (usually in the seventh decade of life), and absence of ovarian-like stroma. In contrast, mucinous cystic neoplasms occur as circumscribed unilocular or multilocular cysts at least partially encapsulated by fibrous tissue with no communication with the pancreatic ducts, arise most often in the body or tail of the pancreas, have a female predominance, and usually occur in younger individuals (aged 40–50 years).1,2,4,13–15 Similar to mucinous cystic neoplasms, IPMNs may occur as invasive or noninvasive tumors. However, according to standard preoperative studies, it may be difficult, if not impossible, to differentiate noninvasive from invasive IPMN before pancreatic resection. It is important to note that the natural histories of both forms of IPMN remain poorly defined, and little is known about the time course for progression of noninvasive to invasive IPMN, the frequency of recurrence in the pancreatic remnant after partial (right or left) pancreatectomy, and the survival duration of patients with nonmetastatic invasive IPMN.16 Elucidation of the natural history of IPMN is needed to guide surgeons in determining the extent of pancreatic resection necessary, especially when they are faced with a transection margin interpreted as positive for noninvasive IPMN.

Therefore, we reviewed our institutional experience with IPMN over the past 13 years to determine the value of standard preoperative studies, to evaluate the outcome associated with a positive pancreatic transection margin, and to define the natural history of noninvasive and invasive IPMNs.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
After approval by the institutional review board, all patients who underwent pancreatic resection for cystic neoplasms of the pancreas at The University of Texas M. D. Anderson Cancer Center between July 1990 and July 2003 were identified from a prospective pancreatic tumor database. Surgical specimens for all cystic neoplasms were re-reviewed by a single pathologist (K.R.C.) without knowledge of patient outcome. IPMNs were distinguished from other cystic neoplasms by the presence of papillary mucin-producing neoplastic epithelium arising within the main pancreatic duct or its major branches without evidence of ovarian-like stroma, as defined by the World Health Organization;2 the presence of extracellular mucin was not necessary for diagnosis. Invasive IPMN (i.e., IPMN with invasive adenocarcinoma) was characterized by the unequivocal infiltration of the pancreatic parenchyma by neoplastic epithelial cells and ductal structures.17 Although invasive IPMNs are generally accompanied by pools of extracellular mucin, the presence of acellular pools of mucin alone within the stromal connective tissue without evidence of malignant cells was not considered sufficient evidence for a diagnosis of invasive IPMN.17

The medical records of the subset of patients with pathologically confirmed IPMN were retrospectively reviewed. Data on patient demographics, presenting symptoms, CA 19-9 levels, type of pancreatic resection, histopathology, adjuvant therapy, recurrence, and survival were collected from the medical records.

The results of preoperative computed tomography (CT)-guided or endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA) biopsies were obtained from the medical record. For the purposes of this study, aspirates interpreted as positive for malignancy, suspicious for malignancy, or positive for mucinous neoplasms were all considered to indicate a positive FNA result. Aspirates that were negative for malignant cells (in the absence of mucin), atypical, or nondiagnostic were considered to indicate a negative FNA result.

The surgical procedure performed was determined by the operating surgeon. In general, tumors predominantly in the head, neck, or uncinate process of the pancreas were treated with pancreaticoduodenectomy. Tumors predominantly in the body or tail of the pancreas were treated with distal pancreatectomy. Intraoperative frozen-section examination of the pancreatic transection margin was performed, and additional pancreatic tissue was resected if invasive IPMN was present at the margin. The management of noninvasive IPMN at the pancreatic transection margin was individualized and based on patient factors such as age and medical comorbidities. Total pancreatectomy was reserved for patients with radiographic evidence of diffuse involvement of the entire pancreas or with pancreatic transection margins positive for invasive IPMN.

The histopathology results are reported on the basis of pathologic evaluation of surgical specimens at three distinct time periods representing three distinct and separate reports: (1) the frozen-section evaluation of the pancreatic transection margin as reported at the time of operation; (2) the permanent-section evaluation of the surgical specimen as reported on the final pathology report approximately 1 week after surgery, which included review of the frozen sections of the pancreatic margins that were previously interpreted at the time of operation; and (3) re-review of all pathology material, including the frozen-section slides, performed for this study by one of the authors (K.R.C.).

The surgical margins were evaluated for evidence of gross and microscopic disease on frozen and/or permanent sections, as described in the sixth edition of the AJCC Cancer Staging Manual.18 The common bile duct and pancreatic transection margins were evaluated on frozen sections by taking en face or multiple perpendicular sections of the margin and assessing them for the presence of invasive or non-invasive neoplastic epithelium. Re-review of the pancreatic margins for this study was based on review of the original frozen-section slides. For the purposes of pathology re-review, margins positive for invasive disease were defined as those with histologic evidence of invasive neoplastic epithelium present at the transection margin. Margins positive for noninvasive disease were defined as those with any neoplastic epithelium within ducts (no matter how small) present at the transection margin. Negative margins had no histologic evidence of invasive or noninvasive neoplastic epithelium at the transection margin. Margins with acellular mucin alone were not considered positive for IPMN. Data on the status of the retroperitoneal margin (soft tissue adjacent to the proximal superior mesenteric artery), the presence or absence of lymph node metastases, and the presence or absence of perineural, vascular, and lymphatic invasion were assessed on permanent sections, reported in the final pathology report, and not re-reviewed for this analysis.

The patients’ postoperative courses were documented. The length of hospital stay was calculated by considering day 1 as the date of operation and by not counting the day of discharge. The time to first recurrence was calculated from the date of cytological or histologic diagnosis, and the site or sites of first recurrence were defined as follows: local indicated the pancreatic bed; regional indicated the peritoneal cavity; and distant indicated the liver, lung, bone, or other sites of distant organ metastases.

Categorical variables were compared by the {chi}2 test. Continuous variables were evaluated by Student’s t-test. Survival and follow-up were measured from the time of the initial cytological or histologic diagnosis to the date of death or last follow-up. Survival curves were created by using the method of Kaplan and Meier.19 Differences in survival were compared by using the log-rank test. A P value <.05 was considered statistically significant.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Patient and Tumor Characteristics
IPMN was histologically confirmed in 35 (26%) of 137 patients who underwent pancreatic resection for pancreatic cystic neoplasms during the period of this analysis. Re-review of all histology specimens confirmed invasive IPMN in the surgical specimens of 13 (37%) of 35 patients and noninvasive IPMN in 22 (63%; Table 1Go).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Results of initial review versus re-review of permanent pancreatic specimens
 
Patient characteristics and presenting symptoms are listed in Table 2Go. There were no differences in age, sex, or race/ethnicity between patients with invasive and noninvasive disease. Jaundice was an uncommon presenting symptom overall (23% of patients), even though most (26 of 35; 74%) tumors involved the head of the pancreas. However, jaundice was significantly more common in patients with invasive disease (6 of 13; 46%) than in those with noninvasive disease (2 of 22; 9%; P = .0321).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Patient and tumor characteristicsa
 
Tumor characteristics, based on histopathologic re-review, are summarized in Table 2Go. The mean tumor diameter was 3.8 cm. Tumors with an invasive component were larger (mean, 4.9 cm; range, 2.0–8.0 cm) than noninvasive IPMNs (mean, 3.2 cm; range, 1.0–7.0; P = .0148). No patient had a positive common bile duct margin, and only one had a positive retroperitoneal margin. Of 13 patients with invasive IPMN, 5 (38%) had lymph node metastases, 4 (31%) had perineural invasion, 1 (7.5%) had vascular invasion, and 2 (15%) had lymphatic invasion.

Preoperative Evaluation
Levels of the serum tumor marker CA 19-9 were increased in 8 (42%) of 19 patients. However, the CA 19-9 level did not reliably discriminate invasive from noninvasive IPMN (Fig. 1Go).


Figure 1
View larger version (8K):
[in this window]
[in a new window]
  		FIG. 1. Scatter plot of preoperative serum CA 19-9 levels in 19 patients. Levels were increased (above the dotted line) in 8 patients (42%). There was significant overlap between the levels in patients with noninvasive intraductal papillary mucinous neoplasms (IPMNs; black circles) and patients with invasive IPMNs (open circles).

 
Preoperative CT- or EUS-guided FNA biopsies were performed in 27 (77%) of 35 patients (Table 3Go). Positive aspirates (positive for malignancy [adenocarcinoma], suggestive of malignancy, or positive for mucinous neoplasm) were identified in 20 patients. Nine (45%) of 20 patients with positive FNA results had histologic confirmation of invasive IPMN. Among the 20 patients with positive aspirates, the cytological diagnosis was reported as adenocarcinoma in 11. Only 6 of 11 patients had invasive IPMN on final histologic analysis of the resected surgical specimen. Negative aspirates were obtained in seven patients. Two (29%) of seven patients with negative FNA results had invasive IPMN on histologic evaluation of the surgical specimen.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Results of preoperative CT- or EUS-guided FNA (n = 27)
 
Surgery
The surgical procedures performed are summarized in Table 4Go. In six of the eight patients who underwent a total pancreatectomy, this was the planned procedure because of radiographic evidence of diffuse, extensive cystic disease involving the entire pancreas. The other two patients underwent total pancreatectomy because intraoperative frozen-section analysis detected invasive IPMN at the pancreatic transection margin after pancreaticoduodenectomy. In one of these two cases, intraoperative pancreatoscopy demonstrated multiple mural abnormalities along the length of the remaining pancreatic duct.


View this table:
[in this window]
[in a new window]
  		TABLE 4. Surgery and adjuvant therapy
 
The median length of hospital stay after surgery was 10 days; this was identical for patients with invasive and noninvasive disease. There were no inhospital deaths or deaths within 30 days of operation. One patient died of complications of an arterial-enteric fistula approximately 12 months after pancreaticoduodenectomy. One patient with invasive IPMN was lost to follow-up.

Pancreatic Margin Analysis
Partial pancreatectomy was performed in 10 patients with invasive IPMN (all pancreaticoduodenectomies) and 19 patients with noninvasive IPMN (15 pancreaticoduodenectomies and 4 distal pancreatectomies). Intraoperative frozen-section analysis was performed on the pancreatic transection margins in all of these patients. Among the 10 patients who underwent pancreaticoduodenectomy for invasive IPMN, 4 (40%) had invasive IPMN at the pancreatic transection margin on frozen-section analysis (Table 5Go; Fig. 2Go). In two of these patients, additional pancreatic tissue was resected to achieve a negative margin on frozen sections. In the other two patients, the procedure was converted to a total pancreatectomy.


View this table:
[in this window]
[in a new window]
  		TABLE 5. Management of the pancreatic margin
 

Figure 2
View larger version (11K):
[in this window]
[in a new window]
  		FIG. 2. Evaluation of pancreatic transection margins in patients with invasive intraductal papillary mucinous neoplasm (IPMN; the main specimen diagnosis was based on histological re-review; the margin status was based on initial review of frozen sections). Margins were positive for invasive disease in 4 patients who underwent PD. Two underwent additional resection to a negative margin, and 2 underwent a TP. No patient had invasive IPMN at the final pancreatic margin. PD, pancreaticoduodenectomy; DP, distal pancreatectomy; TP, total pancreatectomy.

 
Among the 15 patients who underwent pancreati-coduodenectomy for noninvasive IPMN, 5 patients (33%) had noninvasive IPMN at the pancreatic transection margin on frozen-section analysis (Table 5Go; Fig. 3AGo). Among the four patients with noninvasive IPMN who underwent distal pancreatectomy, one (25%) had noninvasive IPMN at the pancreatic transection margin on frozen-section analysis. Thus, six patients with noninvasive IPMN had noninvasive IPMN at the pancreatic margins. In two of these six patients (both after pancreaticoduodenectomy), additional pancreatic tissue was resected to achieve a negative margin. In the remaining four patients (three after pancreaticoduodenectomy and one after distal pancreatectomy), no further resection was performed; the surgeon accepted a pancreatic transection margin that was positive for noninvasive IPMN on frozen-section examination.


Figure 3
View larger version (14K):
[in this window]
[in a new window]
  		FIG. 3. Evaluation of pancreatic transection margins in patients with noninvasive intraductal papillary mucinous neoplasm (IPMN; the main specimen diagnosis was based on pathologic re-review; the margin status was based on the initial review of frozen sections). (A) Positive margins on frozen sections. Margins were positive for noninvasive disease in 6 patients (5 who underwent PD and 1 who underwent DP). Two underwent additional resection to a negative margin. In the remaining 4 patients (3 after PD and 1 after DP), the surgeon accepted a pancreatic transection margin positive for noninvasive IPMN on frozen sections in the absence of confirmed invasion because of medical comorbidities. (B) Negative margins on frozen sections but positive margins in the final pathology report. Initial frozen-section analysis of the pancreatic margin was negative for IPMN in 1 patient who underwent a PD for noninvasive disease, but re-examination for the final pathology report revealed a pancreatic margin positive for noninvasive IPMN. The original intraoperative frozen-section diagnosis was confirmed in all other patients with noninvasive IPMN and in all patients with invasive IPMN. PD, pancreaticoduodenectomy; DP, distal pancreatectomy; TP, total pancreatectomy.

 
All frozen-section pancreatic transection margins evaluated at the time of operation were also reexamined for the final pathology report. In one patient who underwent a pancreaticoduodenectomy for noninvasive disease, the frozen-section margin was interpreted as negative at the time of operation, but re-examination for the final pathology report revealed the presence of a microscopic focus of noninvasive IPMN (Fig. 3BGo). Therefore, this patient did not undergo immediate re-resection. In all other patients with noninvasive IPMN and in all patients with invasive IPMN, the final diagnosis of the pancreatic margin (as listed in the pathology report in the medical record) confirmed the original intraoperative frozen-section diagnosis.

Histologic re-review of the final pancreatic transection margins was possible in 25 of the 35 patients; 6 underwent total pancreatectomy without pancreatic transection and margin analysis, and in 4 patients frozen-section slides were not available for re-review. Of the 25 patients, a final pancreatic margin positive for noninvasive IPMN was reconfirmed on histologic re-review in all 5 patients in whom a positive margin was stated in the final pathology report. However, the status of the pancreatic transection margins was reinterpreted as positive for noninvasive IPMN (from negative for IPMN) in 3 (15%) of the remaining 20 patients for whom margins were assessable. Two of these three patients had distorted foci of noninvasive IPMN in small ducts at the margin. The other patient had low-grade pancreatic intraepithelial neoplasia (PanIN 1) at the margin. One of the three patients had invasive IPMN in the main pancreatic specimen; the other two had noninvasive IPMN.

Thus, on the basis of histologic re-review, a total of eight patients (seven with noninvasive IPMN and one with invasive IPMN) had noninvasive IPMN at the final transection margin. Five had a positive margin at the time of resection diagnosed by immediate intraoperative frozen-section analysis or on final review of the margin, and three (including the one patient with invasive IPMN in the main specimen) had a margin positive for noninvasive IPMN identified on histologic re-review. There were no cases in which histologic re-review changed the diagnosis of the pancreatic transection margin from positive (for invasive or noninvasive IPMN) to negative or from negative to positive for invasive IPMN.

Neoadjuvant/Adjuvant Therapy
Eight patients received neoadjuvant therapy based on a preoperative diagnosis of invasive carcinoma on CT- or EUS-guided FNA biopsy (Table 4Go). Histologic re-review confirmed invasive IPMN in five of these patients and noninvasive IPMN in three patients. The degree to which preoperative therapy affected the final histologic diagnosis in the last three patients could not be determined.

Six patients received postoperative adjuvant therapy for assumed invasive IPMN. However, subsequent histologic re-review determined that there was no invasive disease in two of these six patients.

Five patients received intraoperative radiotherapy. All five also received either preoperative (n = 1) or postoperative (n = 4) adjuvant therapy. Histologic re-review determined that there was no invasive disease in three of these five patients (one of these three had preoperative therapy; the other two had postoperative therapy).

Recurrence and Survival
Follow-up data were available for 12 of 13 patients with invasive IPMN and for all 22 patients with noninvasive disease (Table 6Go). One patient with invasive IPMN was lost to follow-up and was excluded from the analysis of recurrence and survival. The median follow-up was 61.0 months in patients with invasive IPMN and 30.3 months in patients with noninvasive IPMN (no significant difference).


View this table:
[in this window]
[in a new window]
  		TABLE 6. Recurrence and survival ratesa
 
Seven (59%) of 12 patients with invasive IPMN developed recurrent disease, and the other 5 (41%) were disease free at the time of death or last follow-up. The median time to first recurrence was 8.2 months. The single patient with invasive IPMN whose pancreatic transection margin was reclassified as positive for noninvasive IPMN on histologic re-review remained alive and disease-free after 39.5 months of follow-up. None of the 22 patients with noninvasive IPMN developed recurrent disease, including the 7 with pancreatic margins positive for noninvasive disease and the 6 who were originally believed to have invasive IPMN (before histologic re-review).

Among the seven patients with invasive IPMN who had recurrences, one developed a local recurrence as the first site of failure 10.0 months after resection. This patient had extrapancreatic invasion extending through the transverse mesocolon at the time of total pancreatectomy and experienced recurrence in the pancreatic bed. This patient also developed rapidly progressive distant metastases, including soft tissue metastases to the left axilla and left breast. Three of these seven patients developed regional recurrences (malignant ascites) 4.1, 5.0, and 8.2 months after resection; one of these three also developed distant metastases (liver). The remaining three patients developed isolated distant disease (liver) as the first site of failure 5.6, 8.0, and 25.4 months after diagnosis. Four of five patients with pathologically confirmed lymph node metastases developed local and distant (n = 1), regional (n = 2), or distant-only (n = 1) recurrent disease, whereas one remained disease-free at the time of death from other causes 14.0 months after surgery.

Four of the 10 patients who underwent pancreaticoduodenectomy for invasive IPMN had frozen-section pancreatic margins positive for invasive IPMN; 2 underwent re-resection to negative margins, and 2 underwent total pancreatectomy. Of the two patients who underwent local re-resection, one experienced disease recurrence in the liver and died 6.0 months after diagnosis, and the second died of complications of an arterial-enteric fistula 14.0 months after diagnosis. Of the two patients who underwent total pancreatectomy, one died 15 months after diagnosis, and the other patient was alive and disease free 83.0 months after diagnosis.

Kaplan-Meier survival curves are shown in Figs. 4Go and 5Go. The median disease-free survival (DFS) was 13.8 months in patients with invasive IPMN (Fig. 4Go). The median DFS was not reached in patients with noninvasive disease; this difference was significant (P = .0001). The median overall survival (OS) was 22.9 months in patients with invasive disease and 84.9 months in patients with noninvasive disease (P = .0009; Fig. 5Go).


Figure 4
View larger version (12K):
[in this window]
[in a new window]
  		FIG. 4. Kaplan-Meier estimate of disease-free survival. The median disease-free survival was 13.8 months for patients with invasive intraductal papillary mucinous neoplasm (IPMN) and was not reached for patients with noninvasive IPMN (P = .0001).

 

Figure 5
View larger version (13K):
[in this window]
[in a new window]
  		FIG. 5. Kaplan-Meier estimate of overall survival. The median overall survival was 22.9 months for patients with invasive intraductal papillary mucinous neoplasm (IPMN) and 84.9 months for patients with noninvasive IPMN (P = .0009).

 
Follow-Up of Patients With Noninvasive IPMN at the Pancreatic Transection Margin
At a median follow-up of 34.1 months (range, 21.3–59.2 months), there have been no disease recurrences in the pancreatic remnant in the 8 patients found to have noninvasive IPMN at the pancreatic margin. One of these 8 patients was followed with serial imaging and remained disease-free at the time of her death from other causes nearly 5 years later; the remaining 7 patients are alive at last follow-up.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Originally recognized by Ohhashi et al.,20 IPMNs have been given various names, including intraductal papillary hyperplasia/neoplasm, duct-ectatic mucinous cystadenoma, villous adenoma of the main pancreatic duct, and mucinous ductal ectasia.4 The unifying nomenclature of IPMN was adopted in 1994 and endorsed by the World Health Organization.21 Advances in diagnostic imaging of the pancreas, an increased awareness of IPMN, and standardization of nomenclature have contributed to an increase in the recognition of IPMNs. Many of these tumors were previously misclassified as mucinous cystic neoplasms, ductal adenocarcinomas, or simply villous adenomas. Nevertheless, increased identification and improved classification together may not account for what seems to be a true and as-yet unexplained increase in the relative frequency of IPMNs.8,22

Despite the increased identification and greater frequency of this disease, there is little consensus to guide surgeons in the evaluation and operative management of IPMNs. We reviewed our institutional experience with IPMN to determine the value of preoperative diagnostic studies, to evaluate the outcome associated with a positive pancreatic transection margin, and to define the natural history of noninvasive and invasive IPMNs.

Between July 1990 and July 2003, 35 patients underwent pancreatectomy for IPMNs at our institution; they represented 26% of pancreatectomies for cystic neoplasms, a percentage similar to that recently reported by others.6 The characteristics of our patient cohort, including the male preponderance and age at presentation, were also similar to those found in the literature.13–15,23 Although jaundice was uncommon in our study, six (75%) of eight patients who had jaundice at the time of diagnosis had invasive IPMN. In a single-institution study, Kitagawa et al.23 reported that jaundice, increased serum CA 19-9, abnormal liver function tests, and p53 overexpression were predictive of malignancy (defined as invasive disease and carcinoma in situ), although only abnormal liver function tests and p53 overexpression were predictive of malignancy on multivariate analysis. Rivera et al.24 reported that a serum CA 19-9 level >2000 U/mL was predictive of unresectability. We found that the CA 19-9 level did not distinguish invasive from noninvasive disease (Fig. 1Go).

Proximal tumor location also may be predictive of malignancy. Navarro et al.25 reviewed 116 cases of IPMN and reported that lesions in the head of the pancreas are often malignant, whereas those that arise in the body and tail are often benign. Of the 13 patients with invasive IPMN in this study, 8 were treated with pancreaticoduodenectomy and 5 with total pancreatectomy. All four patients who underwent distal pancreatectomy for distal lesions had noninvasive disease. We also found that invasive IPMNs were larger than their noninvasive counterparts (mean diameter, 4.9 vs. 3.2 cm; P = .0148), and this is consistent with other studies.25–29

In patients with IPMN, the diagnosis of invasion can be made definitively only when tumor cells are seen infiltrating the pancreatic connective tissue. In contrast to our experience with solid neoplasms of the pancreas, for which EUS-guided FNA has a diagnostic accuracy of 92%, we found that FNA cannot reliably distinguish invasive from noninvasive IPMN.30 FNA may be of value in documenting the presence or absence of mucin when the radiographic characteristics of the cystic neoplasm are not typical for IPMN; therefore, the differential diagnosis may include nonmucinous benign tumors in addition to mucinous cystic neoplasm and branch duct IPMN. When main duct IPMN is apparent on CT or endoscopy (endoscopic retrograde cholangiopan-creatography or EUS), FNA biopsy is unlikely to provide additional information necessary to develop an optimal treatment plan. Given the difficulty in differentiating invasive from noninvasive IPMN by FNA biopsy before definitive surgical resection, we currently do not deliver preoperative (neoadjuvant) therapy on the basis of an FNA diagnosis of adenocarcinoma when we are dealing with a cystic neoplasm without an obvious solid component on cross-sectional imaging.

We did not have a single patient with a final pancreatic transection margin positive for invasive IPMN. In this study, all four patients with a frozen-section diagnosis of invasive IPMN at the pancreatic margin underwent either resection of additional pancreatic tissue to achieve a negative margin (two patients) or total pancreatectomy (two patients). The presence of invasive IPMN at the margin was confirmed on histological re-review in all 4 cases. There were no cases in which the histological re-review changed a margin interpreted as negative or positive for noninvasive IPMN to one positive for invasive IPMN. There is no apparent reason to treat invasive IPMN any differently than the more typical ductal adenocarcinoma. In general, we prefer not to leave the operating room with a pancreatic transection margin positive for invasive adenocarcinoma. However, three of the four patients with a frozen-section diagnosis of invasive IPMN at the pancreatic margin (who underwent further resection to achieve a negative margin) died of recurrent disease within 15 months of diagnosis. Therefore, re-resection would be attempted only in good-risk patients—those who could clearly tolerate a more extensive operation.

Management of a pancreatic transection margin microscopically positive for noninvasive IPMN is controversial because the natural history of such a positive margin remains largely unknown. In our study, noninvasive IPMN was present at the pancreatic transection margin in 9 (47%) of 19 patients with noninvasive IPMN. Only two (22%) of these nine patients underwent re-resection to achieve a negative margin. At a median follow-up of almost 3 years, there have been no recurrences in the remaining seven patients left with a positive margin. Our data suggest that it may not be necessary to perform additional resection for a microscopically positive (grossly negative) pancreatic transection margin in the absence of histologic evidence of invasive IPMN. Additional follow-up is needed to allow for a more definitive treatment recommendation with regard to initial surgery and follow-up imaging.

Recurrent tumor in the remaining pancreas has been reported after resection for noninvasive IPMNs, but pathologic pancreatic margin status has not been consistently reported in some of these cases.31 Recurrence of noninvasive tumors has also been reported even after resections with a negative margin.28,32 Uneven distribution of invasive components within a lesion and the multifocal nature of IPMN combined with the associated risk of suboptimal sampling for histologic evaluation may contribute to understaging of the primary tumor.22 Alternatively, a field-defect phenomenon, whereby an intrinsic genetic defect exists throughout the ductal system, or extensive exposure to an exogenous mutagen, may induce neoplastic transformation in numerous foci at different times.22 In this study, we found no recurrences of noninvasive IPMN at a median follow-up of 30 months, even among the patients with positive pancreatic margins. However, our follow-up remains short, especially relative to the assumed biology of this disease. Therefore, we recommend that patients with noninvasive IPMN receive cross-sectional abdominal imaging on a regular basis at 12- to 18-month intervals.

The pancreatic margin status was reclassified in three patients (one with invasive IPMN in the main specimen) from negative for IPMN to positive for noninvasive IPMN. These changes in frozen-section diagnosis reveal the difficulty in evaluating frozen-section material for the presence of noninvasive IPMN. It remains uncertain whether a small focus of atypical or dysplastic epithelium lining a duct is sufficient for the diagnosis of IPMN, given the inherent technical limitations of the frozen-section technique. Nevertheless, as with pancreatic margins in cases of typical ductal carcinoma, limited foci of atypical or neoplastic epithelium within ducts may be of uncertain clinical and biologic significance.

The median DFS was 13.8 months, and the median OS was 22.9 months for patients with invasive IPMN. These rates are similar to those in previously reported patients from our institution with ductal adenocarcinoma33 and are consistent with rates in reports from other investigators.13 Both DFS and OS differed significantly between patients with invasive and noninvasive IPMN. Because invasive and non-invasive IPMNs have such different natural histories, accurate pathologic diagnosis is critical to proper management. Invasive IPMN was originally diagnosed in 19 patients, but the diagnosis was changed to noninvasive IPMN in 6 patients on pathologic re-review; none of these 6 patients had recurrences, thus supporting the diagnosis of noninvasive IPMN. These cases had findings such as disruption of large ducts with IPMN, extensive involvement of small ducts by IPMN, and/or pools of acellular mucin, all in a background of extensive fibrosis, that were misinterpreted as evidence of invasive carcinoma. One of these six patients received preoperative chemotherapy and radiotherapy for assumed invasive IPMN; the extent to which preoperative therapy affected the histologic diagnosis cannot be determined. The diagnosis was not changed in any of the 16 patients with an initial diagnosis of noninvasive disease, and the absence of recurrences supports the diagnosis of noninvasive IPMN.

The importance of accurate pathologic diagnosis cannot be overstated. Pools of extracellular mucin in the absence of malignant cells invading the surrounding stroma do not justify a diagnosis of invasive IPMN. The preoperative, intraoperative, and postoperative management of patients with IPMN is difficult, and interpretation of the histopathologic results is challenging. The care of patients with IPMN is facilitated by an experienced team of physicians that includes a radiologist, gastroenterologist, surgeon, pathologist, and medical oncologist.


   ACKNOWLEDGMENTS
 
The authors thank Melissa G. Burkett, Adrienne Denny, Henry Gomez, MD, and Susan Hogue for assistance in the preparation of this manuscript. Supported by the Lockton Fund for Pancreatic Cancer Research at The University of Texas M. D. Anderson Cancer Center.


   FOOTNOTES
 
Presented at the 57th Annual Cancer Symposium of the Society of Surgical Oncology, New York, New York, March 18–21, 2004.

Chandrajit P. Raut is now at Department of Surgery, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115.

Received for publication May 10, 2005. Accepted for publication October 6, 2005.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Adsay NV. Intraductal papillary mucinous neoplasms of the pancreas: pathology and molecular genetics. J Gastrointest Surg 2002; 6:656–9.[CrossRef][Medline]
  2. Longnecker DS, Adler G, Hruban RH, Kloppel G. Intraductal papillary-mucinous neoplasms of the pancreas. In: Hamilton SR, Aaltonen LA, eds. World Health Organization Classification of Tumours: Tumours of the Digestive System. Lyon, France: IARC Press, 2000; 237–240.
  3. Agostini S, Choux R, Payan MJ, et al. Mucinous pancreatic duct ectasia in the body of the pancreas. Radiology 1989; 170(3 Pt 1):815–6.[Abstract/Free Full Text]
  4. Farrell JJ, Brugge WR. Intraductal papillary mucinous tumor of the pancreas. Gastrointest Endosc 2002; 55:701–14.[CrossRef][Medline]
  5. Nickl NJ, Lawson JM, Cotton PB. Mucinous pancreatic tumors: ERCP findings. Gastrointest Endosc 1991; 37:133–8.[Medline]
  6. Warshaw AL, Brugge WR, Lewandrowski KB, Pitman MB. Case records of the Massachusetts General Hospital. Weekly clinicopathological exercises. Case 35-2003. A 75-year-old man with a cystic lesion of the pancreas. N Engl J Med 2003; 349:1954–61.[Free Full Text]
  7. McDonald JM, Williard W, Mais D, Beitler A. The incidence of intraductal papillary mucinous tumors of the pancreas (1). Curr Surg 2000; 57:610–4.[CrossRef][Medline]
  8. Sohn TA, Yeo CJ, Cameron JL, et al. Intraductal papillary mucinous neoplasms of the pancreas: an increasingly recognized clinicopathologic entity. Ann Surg 2001; 234:313–21, discussion 321–2.[CrossRef][Medline]
  9. Hruban RH, Petersen GM, Ha PK, Kern SE. Genetics of pancreatic cancer. From genes to families. Surg Oncol Clin North Am 1998; 7:1–23.[Medline]
  10. Iacobuzio-Donahue CA, Klimstra DS, Adsay NV, et al. Dpc-4 protein is expressed in virtually all human intraductal papillary mucinous neoplasms of the pancreas: comparison with conventional ductal adenocarcinomas. Am J Pathol 2000; 157:755–61.[Abstract/Free Full Text]
  11. Sato N, Rosty C, Jansen M, et al. STK11/LKB1 Peutz-Jeghers gene inactivation in intraductal papillary-mucinous neoplasms of the pancreas. Am J Pathol 2001; 159:2017–22.[Abstract/Free Full Text]
  12. Kokawa A, Kondo H, Gotoda T, et al. Increased expression of cyclooxygenase-2 in human pancreatic neoplasms and potential for chemoprevention by cyclooxygenase inhibitors. Cancer 2001; 91:333–8.[CrossRef][Medline]
  13. Doi R, Fujimoto K, Wada M, Imamura M. Surgical management of intraductal papillary mucinous tumor of the pancreas. Surgery 2002; 132:80–5.[CrossRef][Medline]
  14. Kimura W, Sasahira N, Yoshikawa T, et al. Duct-ectatic type of mucin producing tumor of the pancreas—new concept of pancreatic neoplasia. Hepatogastroenterology 1996; 43:692–709.[Medline]
  15. Sugiyama M, Atomi Y, Kuroda A. Two types of mucin-producing cystic tumors of the pancreas: diagnosis and treatment. Surgery 1997; 122:617–25.[CrossRef][Medline]
  16. Maire F, Hammel P, Terris B, et al. Prognosis of malignant intraductal papillary mucinous tumours of the pancreas after surgical resection. Comparison with pancreatic ductal adenocarcinoma. Gut 2002; 51:717–22.[Abstract/Free Full Text]
  17. Fukushima N, Mukai K. Pancreatic neoplasms with abundant mucus production: emphasis on intraductal papillary-mucinous tumors and mucinous cystic tumors. Adv Anat Pathol 1999; 6:65–77.[Medline]
  18. Greene FL, Page DL, Fleming ID, et al., eds. Exocrine pancreas. In: AJCC Cancer Staging Manual. Chicago: Springer-Verlag, 2002; 157–64.
  19. Kaplan E, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958; 53:457–81.[CrossRef]
  20. Ohhashi K, Murakami Y, Takekoshi T. Four cases of mucin producing cancer of the pancreas on specific findings of the papilla of Vater. Prog Dig Endosc 1982; 20:348–51.
  21. Kloppel G, Solcia E, Longnecker DS, et al. Histologic Typing of Tumors of the Exocrine Pancreas. World Health Organization International Histologic Classification of Tumors. Geneva: Springer, 1996.
  22. Adsay NV. The "new kid on the block": intraductal papillary mucinous neoplasms of the pancreas—current concepts and controversies. Surgery 2003; 133:459–63.[CrossRef][Medline]
  23. Kitagawa Y, Unger TA, Taylor S, et al. Mucus is a predictor of better prognosis and survival in patients with intraductal papillary mucinous tumor of the pancreas. J Gastrointest Surg 2003; 7:12–8; discussion 18–9.[CrossRef][Medline]
  24. Rivera JA, Fernandez-del Castillo C, Pins M, et al. Pancreatic mucinous ductal ectasia and intraductal papillary neoplasms. A single malignant clinicopathologic entity. Ann Surg 1997; 225:637–44; discussion 644–6.[CrossRef][Medline]
  25. Navarro F, Michel J, Bauret P, et al. Management of intraductal papillary mucinous tumours of the pancreas. Eur J Surg 1999; 165:43–8.[CrossRef][Medline]
  26. Kaserer K, Schmaus J, Bethge U, et al. Staining patterns of p53 immunohistochemistry and their biological significance in colorectal cancer. J Pathol 2000; 190:450–6.[CrossRef][Medline]
  27. Obara T, Maguchi H, Saitoh Y, et al. Mucin-producing tumor of the pancreas: surgery or follow-up? Nippon Shokakibyo Gakkai Zasshi 1994; 91:66–74.[Medline]
  28. Traverso LW, Peralta EA, Ryan JA Jr, Kozarek RA. Intraductal neoplasms of the pancreas. Am J Surg 1998; 175:426–32.[CrossRef][Medline]
  29. Jang J-Y, Kim S-W, Ahn YJ, et al. Multicenter analysis of clinicopathologic features of intraductal papillary mucinous tumor of the pancreas: is it possible to predict the malignancy before surgery? Ann Surg Oncol 2005; 12:124–32.[Abstract/Free Full Text]
  30. Raut CP, Grau AM, Staerkel GA, et al. Diagnostic accuracy of endoscopic ultrasound-guided fine-needle aspiration in patients with presumed pancreatic cancer. J Gastrointest Surg 2003; 7:118–28.[CrossRef][Medline]
  31. Nakagohri T, Asano T, Kenmochi T, et al. Long-term surgical outcome of noninvasive and minimally invasive intraductal papillary mucinous adenocarcinoma of the pancreas. World J Surg 2002; 26:1166–9.[CrossRef][Medline]
  32. Zamora C, Sahel J, Cantu DG, et al. Intraductal papillary or mucinous tumors (IPMT) of the pancreas: report of a case series and review of the literature. Am J Gastroenterol 2001; 96:1441–7.[CrossRef][Medline]
  33. Breslin TM, Hess KR, Harbison DB, et al. Neoadjuvant chemoradiotherapy for adenocarcinoma of the pancreas: treatment variables and survival duration. Ann Surg Oncol 2001; 8:123–32.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Ann. Surg. Oncol.Home page
T. Nakagohri, T. Kinoshita, M. Konishi, S. Takahashi, and N. Gotohda
Surgical Outcome of Intraductal Papillary Mucinous Neoplasms of the Pancreas
Ann. Surg. Oncol., November 1, 2007; 14(11): 3174 - 3180.
[Abstract] [Full Text] [PDF]

Home page
 GutHome page
P. Ghaneh and J. Neoptolemos
A new approach to managing intraductal papillary mucinous pancreatic neoplasms
Gut, August 1, 2007; 56(8): 1041 - 1044.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Raut, C. P.
Right arrow Articles by Evans, D. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Raut, C. P.
Right arrow Articles by Evans, D. B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS