ORIGINAL ARTICLE High Risk of Rectal Cancer and of Metachronous Colorectal Cancer in Probands of Families Fulfilling the Amsterdam Criteria Laura Cirillo, MD, ∗ Emanuele DL. Urso, MD,† Giovanni Parrinello, MD,‡ Salvatore Pucciarelli, MD,† Dario Moneghini, MD, ∗ Marco Agostini, MD, PhD,† Donato Nitti, MD,† and Riccardo Nascimbeni, MD ∗ Objective: To investigate the risk of metachronous colorectal cancer (CRC), its impact on survival, and the risk of rectal cancer in a cohort of probands meeting the Amsterdam criteria. Background: Several determinants of decision-making for the management of CRC in patients with a putative diagnosis of Lynch syndrome are scarcely defined, and many of them undergo segmental bowel resection instead of the advised total colectomy. Methods: A retrospective cohort study was conducted on 65 probands of the Amsterdam-positive families who had surgery for primary CRC and at least 5-year surveillance thereafter. The rates of metachronous CRC and of rectal cancer were evaluated, together with their association with preoperatively available clinical predictors. Differences in overall survival between patients with and without metachronous CRC were evaluated using a time-dependent Cox model. Results: Seventeen patients (26.2%) had metachronous CRC. No clinical fea- ture was associated with an increased risk of its development. The risk of death in patients with metachronous CRC was 6-fold increased. Neither a 2-year interval endoscopic surveillance after surgery, nor total colectomy was asso- ciated with a significant reduction in metachronous CRC. Eighteen patients (23.7%) had rectal cancer at first presentation, 5 patients of the remainder (10.6%) developed rectal cancer after primary colon resection. Two patients undergoing total colectomy developed a metachronous rectal cancer (18.2%). A first-degree family history of rectal cancer was associated with an increased risk of rectal cancer. Conclusions: Probands of families fulfilling the Amsterdam criteria carry a high risk of rectal cancer and of metachronous CRC. Total proctocolectomy, or total colectomy and a 1-year interval of proctoscopic surveillance should be advised when a high risk of rectal cancer can be predicted. Keywords: Amsterdam criteria, colorectal surgery, familial colorectal cancer type X, Lynch syndrome, proband, rectal cancer (Ann Surg 2012;00: 1–5) T he diagnosis of Lynch syndrome relies on the detection of germline mutations in mismatch repair genes. 1,2 Families quali- fying for mismatch repair gene testing are generally selected starting from a single subject presenting with colorectal cancer (CRC), or another Lynch-associated cancer, whose personal and/or family char- acteristics fulfil specific criteria, such as the Amsterdam criteria or From the ∗ Department of Medical & Surgical Sciences, University of Brescia, Italy; †Department of Oncological & Surgical Sciences, University of Padua, Italy; and ‡Medical Statistics Unit, University of Brescia, Italy. Disclosure: The authors declare no conflict of interest.Sources of funding for research and publication: University of Brescia & University of Padua. No external sources of financial or material support to disclose. Reprints: Riccardo Nascimbeni, MD, Department of Medical & Surgical Sci- ences, University of Brescia, Viale Europa 11, 25100 Brescia, Italy. E-mail: nascimbr@med.unibs.it. Copyright C  2012 by Lippincott Williams & Wilkins ISSN: 0003-4932/12/00000-0001 DOI: 10.1097/SLA.0b013e31826bff79 the revised Bethesda guidelines 3 : this first subject is defined as the proband or index patient. In patients with a mismatch repair-mutation-proved Lynch syn- drome, total colectomy with ileorectal anastomosis is advised when colon cancer develops. 4,5 Probands presenting with CRC, however, often undergo primary intestinal surgery without a defined mismatch repair gene status, because of the time required to perform molec- ular/gene testing. Accordingly, even when selection is carried out utilizing the more specific Amsterdam criteria, 6 we may expect that both mismatch repair–mutation-positive (Lynch syndrome) and mis- match repair–mutation-negative (familial CRC type X) subjects 7,8 are included among probands at the time of primary surgery. Specific evidence on the risk of metachronous cancer, its im- pact on survival, and on the prophylactic value of total colectomy is scarce for this heterogeneous group of subjects. Consequently, surgi- cal decision-making is guided prevalently by considerations similar to that adopted in sporadic cancer, such as tumor site/multiplicity, co- morbidity, and life expectancy of patients. 9 This approach, together with a frequently overlooked family history, 10 may explain the high proportion of patients with putative Lynch syndrome and primary CRC undergoing a segmental resection of the colorectum. 11 To improve the surgical decision-making in probands, manda- tory information should encompass the risk of metachronous cancer in the colon and in the rectum, and the impact of metachronous CRC on survival. To these aims, we reviewed the data at the time of first diagnosis and the long-term outcomes of a cohort of probands with Amsterdam-positive criteria who underwent primary CRC surgery. PATIENTS AND METHODS This retrospective observational study was conducted on a co- hort of unrelated consecutive patients with putative Lynch syndrome as identified by the Amsterdam criteria from the Brescia and Padua Hereditary CRC Registries. The protocol was approved by the Institutional Review Boards at both institutions. All patients had consented to the anonymous use of their data for scientific purposes when they were originally included in the registries. All data were collected retrospectively from computerized reg- istry databases, and further verified directly with patients on outpa- tient visits during 2010, or with other family members in the case of previous death of the proband. Only probands of families fulfilling the Amsterdam I or II criteria 12 were eligible for the study, a proband (also known as index case) being defined as the first member of a family being suspected of Lynch syndrome by the Amsterdam criteria, and undergoing formal registration at 1 of the 2 Institutions. Further inclusion criteria were the following: adherence of family history to the Amsterdam I or II criteria with documented histology of cancers running in the family, personal history of surgery for CRC, and at least 5 years of docu- mented surveillance and follow-up after primary surgery. Probands undergoing proctocolectomy as a primary procedure were excluded from the study. Copyright © 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Annals of Surgery  Volume 00, Number 00, 2012 www.annalsofsurgery.com | 1