Growth patterns of liver metastases as compared to resected tumors of the same colorectal cancer patients.

Abstract

519 Background: Colorectal cancer (CRC) affects one million people a year. 25% of the patients have metastatic disease at the time of diagnosis. Metastatic disease is one of the major challenges in the treatment of cancer. Even though the TNM staging system is a good prognostic marker, tumours have a biological diversity that cannot be evaluated by the TNM system. Recently three different growth patterns of CRC liver metastases were identified. The aim of this study was to determine the growth patterns of CRC liver metastases from patients with more than one metastasis. Methods: A pilot study was conducted including 34 patients resected for CRC liver metastases, 28 patients selected from a database of 200 patients, operated in Copenhagen between 2007-2010 and 6 patients, operated in Montreal. From these 34 patients, 15 patients were resected for two or more liver metastases. All paraffin embedded tissue sections were stained for reticulin and haematoxylin and eosin at the Finsen Laboratory. Evaluation of the growth pattern was done by three different observers from Antwerp and Copenhagen. Reproducibility was >90%. Results: In the fifteen patients, 9 from Copenhagen and 6 from Montreal, that had multiple hepatic metastases, there was uniformity in the growth pattern of the individual metastases. 40% had a desmoplastic, 33% a pushing and 20% a replacement growth pattern and 7% had a mix pattern. Of the 9 patients from Copenhagen 33% were synchronous and 67% metachronous metastases. Within patients, all patterns were identical. Conclusions: The uniformity of growth patterns seen in each patient suggests that these patterns are not random. The identical growth pattern may be determined by specific interactions between the tumor and the host microenvironment, as postulated by Paget’s seed and soil theory. This could represent three different stromal and cytokine responses or oncogenic pathway responses and could represent metastatic tumour growth in different microenvironments.