Different Molecular Profiles Characterize Well-Differentiated Endocrine Tumors and Poorly Differentiated Endocrine Carcinomas of the Gastroenteropancreatic Tract Daniela Furlan, Roberta Cerutti, Silvia Uccella, Stefano La Rosa, Elena Rigoli, Anna Genasetti, and Carlo Capella Department of Human Morphology, University of Insubria and Ospedale di Circolo, Varese, Italy ABSTRACT Purpose: The molecular pathogenesis of gastroentero- pancreatic endocrine tumors (ETs) is still largely unknown. The purpose of this work was a molecular characterization of 38 gastroenteropancreatic ETs with respect to the pri- mary site and to the morphofunctional profile, pointing out useful diagnostic or prognostic molecular markers. Experimental Design: Twenty-four well-differentiated ETs or carcinomas (WDET/Cs; 11 pancreatic, 3 gastric, and 10 intestinal) and 14 poorly differentiated endocrine carci- nomas (1 pancreatic, 6 gastric, and 7 colorectal) were mi- croallelotyped using 38 polymorphic microsatellite markers covering chromosomes 1, 3, 5q, 6, 11, 17, and 18. Results: Regardless of the primary site, a significantly higher percentage of allelic imbalances (AIs) was observed in poorly differentiated endocrine carcinomas than in WDET/Cs (P  0.012), except for 3 of 8 nonfunctioning pancreatic endocrine tumors and 1 colorectal WDEC, ex- hibiting multiple AIs on chromosomes 1, 3, 6, and 11. A strong positive correlation between AI percentage and Ki-67 proliferation index was detected considering both the whole series of ETs (P  0.004) and the group of WDET/Cs alone (P  0.011). The survival analysis showed a positive corre- lation between low percentage of AI and longer survival (P  0.01). No recurrent AIs at specific chromosomal re- gions were identifiable with respect to the primary site. Conclusions: The malignant progression of endocrine tumors seems to be associated with complex allelotypes and chromosomal instability. Although no specific molecular markers of malignancy can be defined with certainty, the ploidy status and the degree of chromosomal derangements appear to be the most informative genetic factors with prog- nostic significance. INTRODUCTION Endocrine tumors (ETs) of the gut and pancreas are rela- tively rare neoplasms traditionally considered to originate from the cells of the dispersed endocrine system, which are scattered through the gastrointestinal mucosa or are present in the islets within the pancreatic parenchyma. They constitute a heteroge- neous group of tumors, which includes different histopatholog- ical and prognostic classes. Regardless of the tumor site, they are classified into two major categories with significant differ- ences in phenotype and clinical behavior, namely, well-differ- entiated ETs (WDETs) and poorly differentiated endocrine carcinomas (PDECs; Ref. 1). WDETs frequently show bland cytologic atypia, low proliferative rate, and a benign or low- grade malignant behavior. On the contrary, PDECs exhibit a highly atypical morphology with elevated proliferative index and behave in a highly malignant fashion presenting local in- vasion, early metastases, and a very poor prognosis. From both a morphological and a biological standpoint, these latter tumors may be considered as a gastroenteropancreatic (GEP) counter- part of the most common poorly differentiated endocrine carci- nomas (small cell carcinoma; SCC) of the lung, although it is still unclear whether they share a common clinical course and a similar response to therapy. Currently, GEP ETs may represent a difficult task for the diagnosis and/or the clinical management, PDECs, because they may be misdiagnosed as a metastatic SCC of the lung and WDETs, because the histological evaluation alone is often not reliable in predicting the biological behavior and the prognosis of these tumors. The past decade has seen the employment of cytogenetic and molecular approaches to improve our understanding of the biology of GEP ETs, and to define a panel of diagnostic and therapeutic targets for these neoplasms. However, no specific molecular markers of malignancy useful for the prognostic evaluation of these neoplasms have been found, and still very little is known about the pathogenesis of ETs. Molecular data for GEP PDECs are missing overall, with the exception of the high frequency of loss of heterozygosity at APC, DCC, and TP53 loci reported in 7 colorectal PDECs (2), and the evidence of “hot spots” for loss of heterozygosity on 8p, 15q, 17p, 11pq, 12p, and 13q, and Xpq found in a small series of gastric PDECs (3–5). Regarding the GEP WDETs, more information is available, especially for pancreatic ETs (PETs), but only small or hetero- geneous series have been analyzed, without a consistent tumor phenotype analysis among different studies. The application of comprehensive genome-wide approaches such as comparative genomic hybridization and high-resolution allelotyping have led to the identification of a wide spectrum of genetic aberrations in PETs, including chromosomal losses of 1p, 3pq, 6q, 11q, Xq, Received 7/2/03; revised 10/6/03; accepted 10/13/03. Grant Support: University of Insubria, Varese, Italy. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Requests for reprints: Carlo Capella, Department of Pathology, Os- pedale di Circolo, Viale Borri, 57, 21100 Varese, Italy. Phone: 39- 0332264557; Fax: 39-0332262313; E-mail: carlo.capella@ospedale. varese.it. 947 Vol. 10, 947–957, February 1, 2004 Clinical Cancer Research Downloaded from http://aacrjournals.org/clincancerres/article-pdf/10/3/947/1954363/zdf00304000947.pdf by guest on 30 June 2022