Potter, John D.

Abstract Background: High body mass index (BMI) is consistently linked to increased risk of colorectal cancer for men, whereas the association is less clear for women. As risk estimates from observational studies may be biased and/or confounded, we conducted a Mendelian randomization study to estimate the causal association between BMI and colorectal cancer. Methods: We used data from 10,226 colorectal cancer cases and 10,286 controls of European ancestry. The Mendelian randomization analysis used a weighted genetic risk score, derived from 77 genome-wide association study–identified variants associated with higher BMI, as an instrumental variable (IV). We compared the IV odds ratio (IV-OR) with the OR obtained using a conventional covariate-adjusted analysis. Results: Individuals carrying greater numbers of BMI-increasing alleles had higher colorectal cancer risk [per weighted allele OR, 1.31; 95% confidence interval (CI), 1.10–1.57]. Our IV estimation results support the hypothesis that genetically influenced BMI is directly associated with risk for colorectal cancer (IV-OR per 5 kg/m2, 1.50; 95% CI, 1.13–2.01). In the sex-specific IV analyses higher BMI was associated with higher risk of colorectal cancer among women (IV-OR per 5 kg/m2, 1.82; 95% CI, 1.26–2.61). For men, genetically influenced BMI was not associated with colorectal cancer (IV-OR per 5 kg/m2, 1.18; 95% CI, 0.73–1.92). Conclusions: High BMI was associated with increased colorectal cancer risk for women. Whether abdominal obesity, rather than overall obesity, is a more important risk factor for men requires further investigation. Impact: Overall, conventional epidemiologic and Mendelian randomization studies suggest a strong association between obesity and the risk of colorectal cancer. Cancer Epidemiol Biomarkers Prev; 24(7); 1024–31. ©2015 AACR.

Abstract Background: Tumor features such as stage, grade, and microsatellite instability (MSI) status have relevance for colorectal cancer survival whereas the prognostic role of body mass index (BMI) is unclear. We assessed the association of BMI and adult weight gain on colorectal cancer survival, overall and by strata of sex and MSI. Methods: Participants were identified from a multi-center cohort that includes 6,763 colorectal cancer patients with invasive colorectal cancer who were enrolled into the Colon Cancer Family Registry from 1997 to 2008. Vital status was updated through direct contact with patients/next-of-kin and/or linkage with mortality records. BMI 2 years before diagnosis, BMI at age 20 years, and adult weight gain were derived from self-reports of height and weight (i.e., weight 2 years before diagnosis and weight at age 20 years). Tumor MSI status was available for 4,987 patients. Multivariable, two-sided hazard ratios (HR) and 95% confidence intervals (CIs) were estimated from delayed-entry Cox proportional hazards models, controlling for age at diagnosis, TNM summary stage (i.e., I, II, III, IV or missing), smoking (current, former, never), and study center. Results: After a maximum of 13.7 years of follow-up from enrollment to end-of-study (median: 5.3 years), 2,335 patients had died. Higher BMI 2 years before cancer diagnosis (per 5-kg/m2) was associated with higher risk of all-cause mortality overall (HR, 1.10; 95% CI, 1.05 to 1.14), with similar associations when stratified by sex (men: HR, 1.07; 95% CI, 1.02 to 1.14; women: HR, 1.11: 95% CI, 1.05 to 1.17; p-interaction: 0.18) and MSI status (MS-stable/MSI-low: HR, 1.08; 95% CI, 1.03 to 1.14; MSI-high, HR: 1.19; 95% CI, 1.02 to 1.40; p-interaction: 0.88). In joint models, with MS-stable/MSI-low and normal BMI as the referent group, risk of death was lower for those with MSI-high and normal BMI (HR: 0.77; 95% CI: 0.59 to 1.00), higher for MS-stable/MSI-low and high (≥30) BMI (HR: 1.23; 95% CI: 1.07 to 1.42), and essentially the same for MSI-high and high BMI (HR: 0.97; 95% CI: 0.72 to 1.30). Similar patterns of association were observed for BMI at age 20 years and when the outcome was colorectal cancer-specific mortality, although not all associations remained statistically significant for some of the smaller sub-group analyses. After controlling for BMI at age 20 years, adult weight gain was only modestly associated with all-cause mortality (per 5 kg HR: 1.01; 95% CI: 1.00 to 1.02; p-trend: 0.07). Conclusion: High pre-diagnosis BMI was associated with increased mortality after colorectal cancer diagnosis; this association was consistent for men and women and by tumor MSI status. These results also suggest that obesity attenuates the survival advantage observed with MSI-high tumors. Citation Format: Peter T. Campbell, Christina Newton, Polly A. Newcomb, Dennis Ahnen, John Baron, Sean Cleary, Michelle Cotterchio, A. Brad Farris, Jane Figueiredo, Roger C. Green, Loic Le Marchand, John McLaughlin, Amanda Phipps, John D. Potter, Andrew Renehan, Aung Ko Win, Noralane Lindor, Paul Limburg. Prospective study of body mass index and adult weight change with colorectal cancer survival, overall and by tumor microsatellite instability status. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr LB-276. doi:10.1158/1538-7445.AM2014-LB-276

Abstract Background: Cigarette smoking (smoking), hormone therapy (MHT), and folate intake (folate) are each thought to influence colorectal cancer risk, but the underlying molecular mechanisms remain incompletely defined. Expression of estrogen receptor β (ESR2) has been associated with colorectal cancer stage and survival. Methods: In this prospective cohort study, we examined smoking, MHT, and folate-associated colorectal cancer risks by ESR2 protein expression level among participants in the Iowa Women's Health Study (IWHS). Self-reported exposure variables were assessed at baseline. Archived, paraffin-embedded colorectal cancer tissue specimens were collected and evaluated for ESR2 protein expression by IHC. Multivariate Cox regression models were fit to estimate relative risks (RR) and 95% confidence intervals (CI) for associations between smoking, MHT, or folate and ESR2-defined colorectal cancer subtypes. Results: Informative environmental exposure and protein expression data were available for 491 incident colorectal cancer cases. Positive associations between ESR2-low and -high tumors and several smoking-related variables were noted, most prominently with average number of cigarettes per day (RR, 4.24; 95% CI, 1.81–9.91 for ESR2-low and RR, 2.15; 95% CI, 1.05–4.41 for ESR2-high for ≥40 cigarettes compared with nonsmokers). For MHT, a statistically significant association with ESR2-low tumors was observed with longer duration of exposure (RR, 0.54; 95% CI, 0.26–1.13 for >5 years compared with never use). No associations were found for folate. Conclusions: In this study, smoking and MHT were associated with ESR2 expression patterns. Impact: These data support possible heterogeneous effects from smoking and MHT on ERβ-related pathways of colorectal carcinogenesis in older women. Cancer Epidemiol Biomarkers Prev; 24(4); 713–9. ©2015 AACR.

Abstract Background: Genetic and environmental factors influence colorectal cancer (CRC) risk. Previous studies have provided additional etiologic insight by examining multiplicative gene-environment interactions for individual susceptibility loci. However, individual loci confer only modest risks, which may limit statistical power and clinical significance. A genetic risk score comprising known CRC loci may provide a more comprehensive assessment of risk. Further, there is a paucity of data on the role of additive gene-environment interactions, which may have greater public health implications than multiplicative interactions. We thus evaluated both additive and multiplicative interactions between a genetic risk score and 15 key environmental factors on risk of CRC. Methods: We conducted an analysis of 10,491 CRC cases and 10,725 controls of European ancestry in 7 cohort and 6 case-control studies participating in the Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO) or Colon Cancer Family Registry (CCFR). To provide a global measure of genetic predisposition, information across multiple risk variants was combined by calculating a genetic risk score based on 24 polymorphisms near 21 genetic loci identified in previous genome-wide association studies. For the genetic score and environmental factors, the reference category corresponded to that associated with lower CRC risk. We tested for additive interactions by estimating relative excess risk due to interactions (RERIs) using logistic regression; for multiplicative interactions we used an empirical-Bayes approach. Nominal P values ≤ 0.05 were considered statistically significant. Results: After adjustment for age, sex, center, study, principal components, and total energy, as appropriate, we observed super-additive gene-environment interactions for CRC risk between the genetic risk score and body mass index (RERI=0.15; 95% CI: 0.00-0.31), ever smoking (RERI=0.14; 95% CI: 0.00-0.28), pack-years of smoking (RERI=0.23; 95% CI: 0.05-0.41), postmenopausal hormone therapy use (RERI=0.38; 95% CI: 0.17-0.59), and intake of processed meat (RERI=0.23; 95% CI: 0.06-0.40). Of the 15 environmental risk factors, 12 showed RERIs > 0 – suggesting an overall trend toward super-additive interactions for these factors. In addition, we observed evidence of sub-multiplicative interactions for use of aspirin and non-steroidal anti-inflammatory drugs. There were no other statistically significant multiplicative interactions. Conclusions: We observed evidence for super-additive effects of genetic predisposition and environmental risk factors on risk of CRC. Our findings suggest that certain environmental risk factors have stronger effects on absolute risk among individuals with higher genetic risk of CRC. If confirmed in future studies, these results may have implications for targeted prevention strategies. Citation Format: Mengmeng Du, Sonja I. Berndt, Hermann Brenner, Bette J. Caan, Peter T. Campbell, Graham Casey, Andrew Chan, Jenny Chang-Claude, Stephen J. Chanock, Nilanjan Chatterjee, David V. Conti, David Duggan, Jane C. Figueiredo, Steven Gallinger, Jian Gong, Robert W. Haile, Tabitha A. Harrison, Richard B. Hayes, Michael Hoffmeister, John L. Hopper, Li Hsu, Thomas J. Hudson, Carolyn M. Hutter, Eric J. Jacobs, Mark A. Jenkins, Shuo Jiao, Laurence N. Kolonel, Peter Kraft, Loic Le Marchand, Mathieu Lemire, Yi Lin, Noralane M. Lindor, Polly A. Newcomb, John D. Potter, Robert E. Schoen, Fredrick R. Schumacher, Daniela Seminara, Martha L. Slattery, Stephen N. Thibodeau, Cornelia M. Ulrich, Aung Ko Win, Emily White, Brent W. Zanke, Ulrike Peters. Additive and multiplicative gene-environment interactions for colorectal cancer risk. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4831. doi:10.1158/1538-7445.AM2013-4831

Abstract Background: Hormone therapy (HT) may prevent colorectal cancer (CRC), but the molecular mechanisms have not been fully defined. Expression of estrogen receptor beta (ERβ) has been associated with CRC stage and survival and may influence HT-related prevention pathways. In this prospective cohort study, we examined HT-associated CRC risk by ERβ protein expression levels in the population-based Iowa Women's Health Study (IWHS). Methods: The IWHS recruited 41,836 randomly selected Iowa women, ages 55-69 years, with a valid driver's license at study entry (1986). Exposure variables were assessed at baseline, by self-report. Incident CRC cases were ascertained by annual linkage with the Iowa Cancer Registry. Archived, paraffin-embedded tissue specimens were collected from 732 CRC cases, diagnosed through 12/31/2002, for molecular epidemiology studies. HT was categorized as never, former, and current and with durations of <=5 years or >5 years. ERβ protein expression was determined by immunohistochemistry on tissue microarrays. For each tumor core, both staining intensity (0-3) and proportion of cells stained (0-5) were scored. These two scores were added to determine negative (0), low (1-5) or high (6-8) ERβ protein expression levels. Multivariate Cox regression models were fit to estimate relative risks (RRs) and 95% confidence intervals (CIs). Results: Complete HT and ERβ data were available from 484 CRC cases. Associations between HT and CRC subtypes defined by ERβ expression level are shown in the Table. Conclusions: Data from the IWHS cohort suggest that current and longer-duration HT exposure favorably influence the risk for ERβ-neg or ERβ-low CRC cases to a greater degree than ERβ-high CRC cases. Further investigation is required to confirm these observations, and to clarify the underlying mechanisms. ERβ Neg or Low ERβ High Hormone Therapy Person years Events, N RR* (95% CI) Events, N RR* (95% CI) Never 341377 128 1.00 (Ref.) 196 1.00 (Ref.) Ever 212696 62 0.74 (0.54,1.03) 98 0.82 (0.63,1.07) Former 151535 48 0.79 (0.56,1.12) 67 0.73 (0.54,0.99) Current 61161 14 0.60 (0.33,1.09) 31 1.11 (0.74,1.64) P-trend 0.05 0.52 Duration < = 5yrs 148704 47 0.84 (0.59,1.19) 68 0.80 (0.60,1.08) >5yrs 60064 14 0.5 (0.26,0.93) 28 0.89 (0.59,1.34) P-trend 0.03 0.25 Citation Format: Lori S. Tillmans, Robert A. Vierkant, Alice H. Wang, Charles F. Lynch, Kristin E. Anderson, Amy J. French, Robert W. Haile, John D. Potter, Susan L. Slager, Thomas C. Smyrk, Stephen N. Thibodeau, James R. Cerhan, Paul J. Limburg. Hormone therapy and incident colorectal cancer risk by ER beta protein expression level in a population-based cohort of older women. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3652. doi:10.1158/1538-7445.AM2013-3652

Abstract Background: Individuals with a family history of colorectal cancer in first-degree relatives have an elevated risk of developing colorectal cancer themselves, particularly colorectal cancer exhibiting high microsatellite instability (MSI-high). Given that MSI-high colorectal cancer is associated with a favorable prognosis, it is plausible that having a family history of colorectal cancer could, in turn, be favorably associated with colorectal cancer survival. Methods: This study comprised N = 4,284 incident colorectal cancer cases enrolled in the Colon Cancer Family Registry via population-based cancer registries. Using Cox proportional hazards regression, we evaluated the association between family history and both overall and disease-specific survival, accounting for MSI status and tumor site via stratified analyses and statistical adjustment. Results: There was no evidence of association between family history and overall [hazard ratio (HR), 0.92; 95% confidence interval (CI), 0.79–1.08] or disease-specific survival (HR, 1.03; 95% CI, 0.85–1.24) for all cases combined, after adjustment for MSI status or tumor site. Only for rectal cancer cases was colorectal cancer family history modestly associated with more favorable overall survival (HR, 0.75; 95% CI, 0.56–0.99). Conclusions: Although individuals with a family history of colorectal cancer were more likely to have MSI-high tumors than those with nonfamilial disease, this did not translate to a survival benefit. Impact: Overall, there is no evidence that family history of colorectal cancer is associated with colorectal cancer survival; however, specific mechanisms underlying family history may have prognostic impact and merit further study. Cancer Epidemiol Biomarkers Prev; 23(8); 1700–4. ©2014 AACR.

Objective To estimate risk of colorectal cancer (CRC) for first-degree relatives of CRC cases based on CRC molecular subtypes and tumour pathology features. Design We studied a cohort of 33 496 first-degree relatives of 4853 incident invasive CRC cases (probands) who were recruited to the Colon Cancer Family Registry through population cancer registries in the USA, Canada and Australia. We categorised the first-degree relatives into four groups: 28 156 of 4095 mismatch repair (MMR)-proficient probands, 2302 of 301 MMR-deficient non-Lynch syndrome probands, 1799 of 271 suspected Lynch syndrome probands and 1239 of 186 Lynch syndrome probands. We compared CRC risk for first-degree relatives stratified by the absence or presence of specific tumour molecular pathology features in probands across each of these four groups and for all groups combined. Results Compared with first-degree relatives of MMR-proficient CRC cases, a higher risk of CRC was estimated for first-degree relatives of CRC cases with suspected Lynch syndrome (HR 2.06, 95% CI 1.59 to 2.67) and with Lynch syndrome (HR 5.37, 95% CI 4.16 to 6.94), but not with MMR-deficient non-Lynch syndrome (HR 1.04, 95% CI 0.82 to 1.31). A greater risk of CRC was estimated for first-degree relatives if CRC cases were diagnosed before age 50 years, had proximal colon cancer or if their tumours had any of the following: expanding tumour margin, peritumoral lymphocytes, tumour-infiltrating lymphocytes or synchronous CRC. Conclusions Molecular pathology features are potentially useful to refine screening recommendations for first-degree relatives of CRC cases and to identify which cases are more likely to be caused by genetic or other familial factors.

Abstract Cigarette smoking (CS), hormone therapy (HT), and folate intake (FI) are each thought to influence colorectal cancer risk, but the underlying molecular mechanisms remain incompletely defined. The TP53 (p53) protein, encoded by the TP53 tumor-suppressor gene that is commonly mutated in colorectal cancer, can be readily assessed to differentiate biologically distinct colorectal cancer subtypes. In this prospective cohort study, we examined CS-, HT-, and FI-associated colorectal cancer risks by TP53 protein expression level among Iowa Women's Health Study (IWHS) participants. The IWHS recruited 41,836 randomly selected Iowa women, ages 55 to 69 years, with a valid driver's license at study entry in 1986. Self-reported exposure variables were assessed at baseline. Incident colorectal cancer cases were ascertained by annual linkage with the Iowa Cancer Registry. Archived, paraffin-embedded tissue specimens were collected and evaluated for TP53 protein expression by immunohistochemistry. Multivariate Cox regression models were fit to estimate relative risks (RR) and 95% confidence intervals (CI) for associations between CS, HT, or FI and TP53-defined colorectal cancer subtypes. Informative environmental exposure and protein expression data were available for 492 incident colorectal cancer cases: 222 (45.1%) TP53 negative, 72 (14.6%) TP53 low, and 198 (40.2%) TP53 high. Longer duration (>5 years) of HT was inversely associated with TP53 high colorectal cancers (RR, 0.50; 95% CI, 0.27–0.94). No other statistically significant associations were observed. These data support possible heterogeneous effects from HT on TP53-related pathways of colorectal carcinogenesis in older women. Cancer Epidemiol Biomarkers Prev; 23(2); 350–5. ©2013 AACR.