Family History of Colorectal Cancer Is Not Associated with Colorectal Cancer Survival Regardless of Microsatellite Instability Status

Individuals whose first-degree relatives have a history of colorectal cancer have an elevated risk of developing colorectal cancer themselves. However, several studies have suggested that having a colorectal cancer family history is favorably associated with prognosis after colorectal cancer diagnosis (1–4). In a recent retrospective study of 10,782 colorectal cancer cases, Morris and colleagues reported 11% lower all-cause mortality for those with familial versus nonfamilial colorectal cancer, despite a lack of difference in the distribution of age or stage at diagnosis by family history (1). This observed survival difference may reflect, at least partly, the fact that familial colorectal tumors are more likely than nonfamilial tumors to exhibit high microsatellite instability (MSI-high), as MSI-high status has been consistently associated with more favorable prognosis (5). Most studies, however, have not considered the possible impact of MSI on the relationship between family history and colorectal cancer survival.

Using data from the Colon Cancer Family Registry (CCFR), we evaluated the association between colorectal cancer family history and survival after colorectal cancer diagnosis, accounting for MSI status.

Details of the CCFR are provided elsewhere (6, 7). For the present analysis, we included cases with incident invasive colorectal cancer, diagnosed between 1998 and 2007, who were enrolled into four CCFR sites following population-based case-ascertainment. Information on family history and other risk factors was collected via telephone-administered or self-administered questionnaires. We excluded cases for whom MSI status was unknown (N = 1,091). Because Lynch Syndrome is particularly associated with MSI-high status, and has a better prognosis than sporadic disease (8), we excluded 116 cases with a germline mutation in one of four DNA mismatch repair genes (MLH1, MSH2, MSH6, and PMS2). In total, 4,284 population-based colorectal cancer cases were included.

We used Cox regression to evaluate associations between colorectal cancer family history in first-degree relatives and survival after colorectal cancer diagnosis, where the time-axis was defined as days since diagnosis. Staggered entry was used to account for time between diagnosis and study enrollment, and sampling weights were used to account for differences in sampling strategies across CCFR sites. We fit models separately for associations with the presence (yes/no) and extent of family history (0/1/≥2 affected relatives), and for associations with overall and disease-specific survival. All analyses were adjusted for age and year at diagnosis, study site, sex, history of endoscopic screening in the 2 years before diagnosis, smoking, and body mass index. Additional analyses were further adjusted for tumor site and MSI. We also performed analyses stratified by tumor site (proximal colon, distal colon, and rectal cancer) and MSI [microsatellite stable (MSS)/MSI-low and MSI-high]. All analyses were conducted in STATA v13.0.

Relative to nonfamilial cases, familial cases had a later age at diagnosis and were more likely to have MSI-high and proximal colon tumors (Table 1). The distribution of family history differed across CCFR study sites, largely reflecting differences in sampling strategies.

For all cases combined, there was no evidence of association between colorectal cancer family history and survival [HR_overall, 0.93; 95% confidence interval (CI), 0.80–1.08; HR_{disease-specific}, 1.02; 95% CI, 0.84–1.23; Table 2]. Adjustment for tumor site and MSI had a negligible impact on results. The majority (88%) of familial cases reported only one affected relative; thus, statistical power was limited for associations comparing those with 0 versus ≥2 affected relatives.

There was also no evidence of associations between family history and survival in stratified analyses, with the exception that family history was marginally significantly associated with more favorable overall survival in rectal cancer cases [hazard ratio (HR), 0.75; 95% CI, 0.56–0.99]. Conversely, family history was suggestively, but nonsignificantly, associated with poorer overall and disease-specific survival among distal colon cancer cases.

In this large cohort of colorectal cancer cases, we found no evidence of an association between colorectal cancer family history in first-degree relatives and prognosis. Consistent with previous studies, we found those with familial colorectal cancer were more likely to have proximal colon cancer (1) and MSI-high tumors (2, 8); however, these differences in tumor attributes did not translate to differences in overall and colorectal cancer–specific survival. Adjustment for, and stratification by, tumor site and MSI had minimal impact on associations.

Several studies have suggested a modest inverse association between family history and colorectal cancer survival. Only one previous study has considered the contribution of MSI status to this association (2). In an analysis of 1,087 stage III colorectal cancer cases, Chan and colleagues reported no change in the association between family history and disease-free survival after MSI-adjustment, and no interaction in associations with family history by MSI.

The absence of an association between family history and survival observed here, and the modest associations noted by some previous studies, do not rule out the possibility that prognosis differs for those with family history attributable to specific predisposing genetic or environmental factors. Rather, the observed null association may reflect heterogeneity in the mechanisms resulting in colorectal cancer family history. Additional research into the relationship between specific germline genetic factors and colorectal cancer survival is necessary.

No potential conflicts of interest were disclosed.

The content of this article does not necessarily reflect the views or policies of the National Cancer Institute or any of the collaborating centers in the Cancer Family Registry (CCFR), nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government or the CFR.

Conception and design: A.I. Phipps, D.J. Ahnen, A.K. Win, R. Gryfe, P.A. Newcomb

Development of methodology: A.I. Phipps, P.T. Campbell, P.A. Newcomb

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): D.J. Ahnen, P.T. Campbell, A.K. Win, M.A. Jenkins, N.M. Lindor, J.D. Potter, P.A. Newcomb

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): A.I. Phipps, P.T. Campbell, A.K. Win, P.A. Newcomb

Writing, review, and/or revision of the manuscript: A.I. Phipps, D.J. Ahnen, P.T. Campbell, A.K. Win, M.A. Jenkins, N.M. Lindor, R. Gryfe, J.D. Potter, P.A. Newcomb

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): P.T. Campbell, P.A. Newcomb

Study supervision: R. Gryfe, J.D. Potter

This work was supported by grant UM1 CA167551 from the National Cancer Institute, NIH, and through cooperative agreements with members of the CCFR and Principal Investigators. Collaborating centers include Australasian Colorectal Cancer Family Registry, Mayo Clinic Cooperative Family Registry for Colon Cancer Studies, Ontario Registry for Studies of Familial Colorectal Cancer, and Seattle Colorectal Cancer Family Registry. This work was also supported by National Cancer Institute grants K07CA172298 and K05CA152715.