Elsevier

Human Pathology

Volume 41, Issue 10, October 2010, Pages 1359-1368
Human Pathology

Original contribution
Clinicopathologic study of 85 colorectal serrated adenocarcinomas: further insights into the full recognition of a new subset of colorectal carcinoma,☆☆

https://doi.org/10.1016/j.humpath.2010.04.002Get rights and content

Summary

Colorectal serrated adenocarcinoma represents a subtype of colorectal carcinoma that originates from serrated adenomas. Previous studies have suggested a more aggressive course, but this has not been verified. The aim of this work was to test the diagnostic reproducibility of previously proposed histologic criteria for serrated adenocarcinoma and to analyze the clinicopathologic features and outcome that would warrant its recognition as a new subtype of colorectal cancer. Nine hundred twenty-seven consecutive colorectal cancer specimens were used to search for cases fulfilling the criteria of serrated adenocarcinoma and matched controls. Clinicopathologic findings of 85 serrated adenocarcinomas were compared with a matched control group of conventional cancers. Serrated adenocarcinomas were encountered in 9.1% (n = 85) of cases. Residual serrated adenoma was present in 44 (51.7%). Absence of residual adenoma did not have any influence on the parameters studied. Interobserver variation between 2 Spanish and a Finnish pathologist showed moderate agreement (κ = .5873). Compared with their matched controls, serrated adenocarcinomas were more often accompanied by synchronous residual serrated adenomas (P < .0001), remote serrated adenomas (P = .0035), and serrated adenocarcinomas or cancers representing partial features of these tumors (P = .002). They had a less favorable 5-year survival than conventional cancers (P = .048 Breslow, Kaplan-Meier), and left-sided ones had the worst prognosis (P = .001). Serrated adenocarcinoma is an identifiable subset of colorectal cancer; and the histopathologic differences, in addition to its less favorable prognosis, may justify its recognition as a distinct subset of colorectal cancer warranting the search for specific clinical management strategies.

Introduction

The first observation of a colorectal carcinoma with architectural and histochemical similarities to hyperplastic polyps was made by Jass and Smith in 1992 [1]. Thereafter, the first systematic documentation that serrated adenomas frequently progress to cancers with a serrated morphology, that is, “serrated adenocarcinoma” (SAC), was reported by Mäkinen et al [2]. Serrated adenocarcinoma is currently considered an end point carcinoma of the so-called serrated polyp pathway that can be reliably identified by morphologic features or by the presence of a residual serrated polyp. This route has been proposed as an alternative pathway to the time-honored Vogelstein adenoma-carcinoma sequence [3].

According to the review of Mäkinen [4], more than 80 SACs arising from serrated adenoma or large hyperplastic polyp have been reported by different groups. Earlier descriptions of SAC were based on the recognition of a serrated adenoma immediately adjacent to the carcinoma [4]. Tuppurainen et al [5] reported that in 17% of SACs an adjacent lesion presented characteristics of a sessile serrated adenoma with variable degrees of atypia. This is probably an underestimate because sessile serrated adenomas may undergo high-grade transformation [6], [7] and cancer tissue frequently destroys a preexisting adenoma [8].

Diagnostic criteria for distinguishing SAC from conventional colorectal cancer (CC) are essential for cases where precursor lesions are no longer evident. The histologic criteria for the diagnosis of SAC include epithelial serrations, clear or eosinophilic cytoplasm, abundant cytoplasm, vesicular nuclei, absence of or less than 10% necrosis of the total surface area, mucin production, and cell balls and papillary rods in mucinous areas of a tumor [4]. These histologic features of SAC have recently been validated by messenger RNA expression profiling that showed that SACs differ from CCs at both the morphologic and molecular level and that SACs do indeed form a distinct subclass of colorectal cancer (CRC) [9]. Furthermore, there are conflicting reports as to whether SACs may behave differently from CCs in terms of survival and speed of tumor progression [2], [9].

Previous studies have given varying estimates about the frequency of cancers originating from serrated polyps, making it difficult to know the actual incidence of SAC. Morphologic studies [5], [10] have yielded lower numbers (7.5%-12%) than estimates of a 10% to 20% frequency drawn from the molecular characteristics of serrated polyps and sporadic CRC with concurrent DNA hypermethylation and BRAF mutations [11].

The purpose of this study was 4-fold, as follow:

  • To assess the validity of the histologic criteria for the diagnosis of SAC in a different geographical setting.

  • To determine the prevalence of SAC in our region by reviewing a demographic series of 927 cases.

  • To provide a comparative description of the main clinicopathologic findings of SACs versus control CCs.

  • To investigate whether SACs have a different prognosis from that of CCs.

Section snippets

Materials and methods

In advance, our study was approved by the Institutional Review Board of Santa María del Rosell University Hospital. A total of 927 consecutive surgical specimens with CRC (Department of Pathology at the Santa María del Rosell University Hospital) collected from January 1995 to December 2008 were included in the study. Clinical and pathologic records available at the time of diagnosis were retrieved for reassessment. Surgical specimens had been routinely fixed in 10% formaldehyde for at least 24

Results

Specific search for SAC and dSAC conducted by Spanish pathologists yielded a group of 141 patients. This group with 152 tumors with CRC (either SAC or dSAC) was later reviewed by another gastrointestinal pathologist (M. J. M.) who found 85 SACs from 81 patients, 29 dSACs (including one “overlap” polyp with a carcinoma component and possible invasion, and 4 “overlap” CRCs), and 38 CCs.

Eleven patients in the 81 SAC group had, in addition, a synchronous carcinoma: 4 SACs, 6 dSACs, and 1 CC. This

Discussion

To date, 45 cases of SACs have been published by Finnish authors [9]; and another series of 11 SACs has been reported by O'Brien et al [10]. Here we report a series of 85 SACs (8.7% referred to patients or 9.2% referred to all CRCs) collected from a review of 927 consecutive surgical specimens over a 14-year period. It is the largest reported series, doubling the total number of reported SACs to more than 160 cases; and the cancer incidence is quite similar to the 7.5% reported by the Finnish

Acknowledgments

We are grateful to I. Molina-Martínez, A. García-Utiel, M. C. Marín-Serrano, D. Torres, and S. Navarro-Mateos for their technical assistance and to Ms Anna Vuolteenaho, MA; Diego Arcas, PhD; and Dr M. V. Tobin for proofreading.

References (18)

There are more references available in the full text version of this article.

Cited by (0)

All authors declare that there are no conflicts of interest.

☆☆

This work was supported by grants from the Instituto de Salud Carlos III, Ministerio de Sanidad, Madrid, Spain (ref P1081210) and from the Academy of Finland, Helsinki, Finland (ref 127990). This work was also supported by Fundación para la Formación e Investigación Sanitarias de la Consejería de Sanidad de la Región de Murcia.

View full text