The epidemiology of rare types of hepatobiliary and pancreatic cancer from national cancer registry

To the best of our knowledge, this was the first study to comprehensively elucidate the incidence and clinical features of uncommon subtypes in hepatobiliary and pancreatic cancers. Except for the common histological subtypes, such as hepatocellular carcinoma, cholangiocarcinoma, and adenocarcinoma, all of uncommon subtypes were included in this study, using NCR data. There is no internationally agreed upon definition of rare cancers. In the USA, a recent analysis held at the National Cancer Institute in 2007 employed the definition of < 15 incident cases per 100,000 per year [11]. Conversely, the Surveillance of Rare Cancers in Europe (RARECARE) project proposed a list of rare cancers and developed a new incidence-based definition of < 6 incident cases per 100,000 per year [12]. Our study revealed that the incidence of uncommon subtypes was extremely rare and less than 0.3 per 100,000 population for hepatobiliary cancer and at most 1.1 per 100,000 population for pancreatic cancer. Moreover, these incidences easily met any suggested definition of rare cancers. Although there have been many studies about rare tumors, many of them were retrospective observational studies targeted at single rare tumor types on a limited scale [13, 14]. These studies could analyze practice treatment, but were inappropriate to obtain epidemiologic information. Recently, PBCRs data were established in many countries, and were used for the study of rare tumors [12, 15,16,17]. A European report used the RARECARE definition to hierarchically structure the list of rare tumors into three layers: families of tumors (tumors with the same referral pattern), clinically meaningful tumors (perceived by clinicians as single diseases), and WHO tumor entities. In this list, the first layer is marked with the number 1 (Tier 1), the second layer with 2 (Tier 2), and the WHO entities are marked with 3. A total of 190 types of tumors are classified as rare cancers, including nine types of Tier 1 tumors and 181 types of Tier 2 tumors [12]. Regarding rare hepatobiliary and pancreatic tumors, many of the uncommon subtypes covered in our study were included in the RARECARE lists. However, neuroendocrine tumors and sarcoma are not divided by each primary site, and the details of adenocarcinoma with variants are currently unclassified. We listed details of adenocarcinoma with variants, neuroendocrine neoplasms, and sarcoma. Furthermore, the primary sites of the intrahepatic bile duct and the ampulla of Vater were not included for “liver” or “biliary tract,” but divided individually. Thus, we believe that this study reveals a more comprehensive and detailed burden of rare tumors than previous reports.

When evaluating individual tumors, the incidence of several tumors differed from some previous reports. Fibrolamellar hepatocellular carcinoma is a rare variant of hepatocellular carcinoma, and usually found in the young (< 40 years) and in Caucasians without underlying hepatitis or cirrhosis. Fibrolamellar hepatocellular carcinoma accounts for 0.85% of all cases of primary liver cancer in the United States, and its incidence is 0.02 per 100,000 population in Europe [15, 18]. In Japan or in Asia, fibrolamellar hepatocellular carcinoma is rare [17]. This difference of incidence between Caucasians and Asians were similarly suggested in our study. Intraductal papillary mucinous neoplasm (IPMN) is a common lesion, with an estimated prevalence of 3–6% in the general population. When IPMN progresses to an invasive pancreatic ductal adenocarcinoma, it is referred to as “IPMN with an invasive carcinoma” or “intraductal papillary mucinous carcinoma” (IPMC). IPMC accounts for approximately 10% of resected pancreatic cancers of ductal origin; however, the incidence of IPMC remains ambiguous [19, 20]. From the previous reports, although the incidence of IPMC was 0.01 per 100,000 population in Europe, it was 0.14 per 100,000 population in Korea and 0.50 per 100,000 population in Japan [15, 17]. In our study, the crude incidence further increased to 0.58 per 100,000 population (Supplementary Table S6). Similarly, the incidence of undifferentiated carcinoma with osteoclast-like giant cells was higher than previous reports (< 0.01 per 100,000 population in Europe and Japan) [15, 17]. These differences indicate that the Japan NCR project could more comprehensively cover all rare tumors, whereas many differences may exist in the pathological diagnoses or real incidence between countries.

Furthermore, our study elucidated not only the burden, but also the clinical features of rare tumors, which were not fully covered in previous studies [12, 15,16,17]. Regarding the sex ratio, there were no major difference between common morphology types and rare tumors as a whole. As is well-known, liver cancer, extrahepatic bile duct cancer, and ampulla of Vater cancer present more frequently in men, whereas gallbladder cancers present more frequently in women [1, 21]. This trend was common in rare tumors. However, rare tumors of the intrahepatic bile duct or gallbladder were seen more frequently in men than common morphology type tumors. On the whole, based on our study, rare hepatobiliary and pancreatic tumors occurred more frequently in men. Individually considering each rare tumor, acinar cell carcinoma and combined hepatocellular carcinoma and cholangiocarcinoma were common in men, whereas mucinous cystic neoplasm with associated invasive carcinoma and solid pseudopapillary neoplasm were common in women, as previously reported [22, 23].

Many rare tumors of the extrahepatic bile duct were diagnosed via symptomatic reports, which indicated rare tumors may have similar characteristics to common morphology type because of tumor localization. However, only 17% of rare tumors of the extrahepatic bile duct were diagnosed with distant metastasis, which was less than reported in common morphology types [21]. Similarly, the tumors with distant metastasis were less frequent in intrahepatic bile duct and pancreas. These features indicate that some rare tumors have less aggressive characteristics than those of common morphology types.

In Japan, cancer treatment is provided through multiple facilities, rather than being centralized. According to an analysis using a hospital-based cancer registry in Japan, even rare cancers with an estimated annual incidence of one case per 100,000 individuals were treated at nearly 300 DCCHs [24]. Conversely, in some European countries, rare cancer treatment is centralized through a small number of large specialized institutions. For example, the number of hospitals providing 75% of treatments for rare liver cancers was 22 in Belgium (population, 10.5 million), 12 in Bulgaria (population, 7.7 million), and 36 in the Netherlands (population, 16.3 million) [15]. Our study revealed the real-world status of diagnosis and treatment of rare hepatobiliary and pancreatic tumors in Japan. There were no large differences in diagnosis between each primary site, which means roughly half of rare tumors were diagnosed in non-DCCHs (not specialized facilities). This is not only because of the distribution characteristics of cancer treatment facilities, but also because of difficulties of pre-treatment biopsies and diagnosis of rare tumors. Although we expected patients diagnosed with rare cancer to be referred to a DCCH for specialized treatment, only about 10% more patients received treatment at DCCHs in comparison with those who were diagnosed at DCCHs. This difference may be explained by the lack of system to centralize rare tumors treatment. Notably, patients with rare tumors of the gallbladder received surgery less frequently at DCCHs because of difficulties of pre-operative biopsy and diagnosis of rare tumors in gallbladder, or because the cancer was found incidentally after resection due to suspected cholecystitis. Since information on rare cancers is lacking, it is important to improve access for patients and facilitate communication between DCCHs and non-DCCHs. Low incidence is a major obstacle to conducting clinical trials to develop effective new treatments for rare tumors. One way to overcome this obstacle would be to centralize patients with rare tumors to DCCHs that excel in management and treatment of that specific cancer subtype.

This study had several limitations that require consideration when interpreting the results. First, because the registry started in 2016, we could only analyze data between 2016 and 2017. The number of uncommon cancers is small, leading to the unstability of the incidence data. Moreover, because the follow-up duration was still short, we do not yet have sufficient data on survival. We plan to solve these issues after we accumulate sufficiently mature data. Regarding the classification of rare tumors, we used both the worldwide WHO classification of tumors and domestic classification, as described in the latest versions as of 2016 [6,7,8,9]. However domestic classification (general rules) of LC, BTC, and PC have been revised, and there are some differences in pathological classification of rare tumors, such as neuroendocrine neoplasm. Finally, several important epidemiological factors that are recorded in HBCRs are not captured by the NCR. While HBCRs did not cover all the cancer cases in Japan, they prioritized the clinical information, including the classification of the Union for International Cancer Control (UICC) Tumor Node Metastasis of malignant tumor stage and treatment dates of each patient, which are not obtained by the NCR. Thus, an HBCR is superior to PBCR in facility comparison. However, the size of the present study compensates for this weakness to a great extent and provides a comprehensive epidemiological understanding of all rare hepatobiliary and pancreatic tumors.

In conclusion, we have provided comprehensive information on the epidemiological status on all rare hepatobiliary and pancreatic tumors in Japan by utilizing population-based NCR data for the first time. We plan to expand this study and analyze survival data and trends in mortality. Moreover, detailed information will be analyzed for each individual rare cancer in the future. The current study serves as an important and useful basis for understanding and developing treatment for rare tumors.

留言 (0)

沒有登入
gif