| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Massachusetts General Hospital, Harvard Medical School, 15 Parkman Street, ACC/336, Boston, Massachusetts 02114
Correspondence: Address correspondence and reprint requests to: Carlos Fernández-del Castillo, MD; E-mail: cfernandez{at}partners.org.
Just over two decades ago, Ohashi et al.1 described four cases of pancreatic cancer that had unusual endoscopic features characterized by a bulging papilla and a distended pancreatic duct with overproduction of mucus. The cases were also unique in that they had very good survival compared with the usual pancreatic ductal adenocarcinoma. Their observation was soon followed by reports of similar cases, initially from Japan, but later on from all over the world. As it often happens with newly described entities, this tumor received many different names. In Europe and the United States, it was commonly referred to as mucinous ductal ectasia, but in the mid 1990s, it was recognized as a separate pancreatic entity by the World Health Organization and given the name of intraductal papillary mucinous neoplasm (IPMN).2
In this issue of Annals of Surgical Oncology, Jang et al.3 report on 208 patients with IPMN. This is the largest reported series on IPMN, and 208 is an impressive number considering that 10 years ago clinicians were still reporting case reports and small series with this condition. The cohort was compiled from the experiences of 28 medical institutions in Korea, and the goal of the authors, in addition to describing the clinicopathologic features and surgical treatment of IPMN in their country, was to identify factors that predict the presence of malignancy.
Of the clinicopathologic characteristics of the cohort, two merit contrast to other experiences. One is the preponderance of men, with a male-female ratio of 2.4:1. Although this has also been reported in other series from Japan, it is not the case in the four largest series from the United States, in which the distribution is nearly equal.4–7 Another important difference is the frequency of malignancy—in particular of lymph node metastases—in cases with invasive cancer. Overall, 38.5% of IPMNs were malignant (14% in situ and 24.5% invasive), and only 12% of invasive IPMNs had positive lymph nodes. By comparison, the recently published series from Johns Hopkins with 136 patients describes 34% with in situ carcinoma and 38% with invasive carcinoma; 54% of the latter had lymph node metastases.6 It is unclear whether these represent real geographic variations in the presentation of IPMN or whether they reflect differences in the way the diagnosis is made.
It is now known that IPMNs can arise in the main duct (IPMN-M) or in the side branch ducts (IPMN-Br) of the pancreas. Several studies have suggested that IPMN-Br is a more indolent disease with a lesser incidence of malignancy and slower growth when compared with IPMN-M.8–11 Inasmuch as an increasing number of IPMNs and other cystic lesions of the pancreas are being incidentally discovered12 and because the proportion of asymptomatic IPMN-Br can be as high as 53%,11 the necessity for resection in all patients has been questioned, and conservative management with observation alone has been described.10,13
In the series of Jang et al.,3 110 (53%) of the 208 cases were IPMN-M, 79 (38%) were IPMN-Br, and the remainder were a combined type. Although the incidence of malignancy in the main duct or combined variants was significantly higher than in the branchduct type (46% vs. 28%), there was no difference whatsoever in survival, and after multivariate analysis this distinction had no predictive value for the likelihood of malignancy. However, we need to keep in mind that this is not only a retrospective study dating as far back as 1993, but also one that involves 48 surgeons from 28 different institutions. Because the classification of IPMN into IPMN-Br and IPMN-M is relatively recent (and the awareness that they could have different biological behavior even more so), clinicians and pathologists have not paid much attention to making this distinction. In my experience and that of others, it is difficult (and outright impossible in many cases) to determine accurately in a retrospective fashion whether an IPMN affects predominantly the main duct, the side branches, or both. Unless one has the actual radiological studies (usually computed tomography or magnetic resonance cholangiopan-creatography) and the pathologist has made a very detailed description of the macroscopic specimen, this information cannot be retrieved. In addition, we do not know how many patients (if any) with IPMN-Br were seen by the authors in that same time period and not operated on.
Pancreatic resection has certainly become progressively safer, but this does not mean that we need to adopt a blanket policy of resection for anything that appears abnormal in the pancreas. The concern for actual or potential malignancy in IPMNs is real, and the recommendation to proceed with resection may be justified in most suitable surgical candidates. However, we also have the obligation to learn from our experience. If IPMN-Br is indeed an indolent disease, we will find out only by scrupulously classifying tumors with adequate preoperative imaging and by collaborating closely with radiologists and pathologists.
Received for publication September 8, 2004. Accepted for publication October 4, 2004.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
