Patrick S. Parfrey

Background: Interactions among genetic variants are rarely studied but may explain a part of the variability in patient outcomes.Objectives: In this study, we aimed to identify 1 to 3 way interactions among SNPs from five Wnt protein interaction networks that predict the 5-year recurrence risk in a cohort of stage I-III colorectal cancer patients.Methods: 423 patients recruited to the Newfoundland Familial Colorectal Cancer Registry were included. Five Wnt family member proteins (Wnt1, Wnt2, Wnt5a, Wnt5b, and Wnt11) were selected. The BioGRID database was used to identify the proteins interacting with each of these proteins. Genotypes of the SNPs located in the interaction network genes were retrieved from a genome-wide SNP genotype data previously obtained in the patient cohort. The GMDR 0.9 program was utilized to examine 1-, 2-, and 3-SNP interactions using a 5-fold cross validation step. Top GMDR 0.9 models were assessed by permutation testing and, if significant, prognostic associations were verified by multivariable logistic regression models.Results: GMDR 0.9 has identified novel 1, 2, and 3-way SNP interactions associated with 5-year recurrence risk in colorectal cancer. Nine of these interactions were multi loci interactions (2-way or 3-way). Identified interaction models were able to distinguish patients based on their 5-year recurrence-free status in multivariable regression models. The significance of interactions was the highest in the 3-SNP models. Several of the identified SNPs were eQTLs, indicating potential biological roles of the genes they were associated with in colorectal cancer recurrence.Conclusions: We identified novel interacting genetic variants that associate with 5-year recurrence risk in colorectal cancer. A significant portion of the genes identified were previously linked to colorectal cancer pathogenesis or progression. These variants and genes are of interest for future functional and prognostic studies. Our results provide further evidence for the utility of GMDR models in identifying novel prognostic biomarkers and the biological importance of the Wnt pathways in colorectal cancer.

Background: Paradoxically epidemiological data illustrate a negative relationship between dietary folate intake and colorectal cancer (CRC) risk. The occurrence and progression of CRC may be influenced by variants in some key enzyme coding genes in the folate metabolic pathway. We investigated the correlation between genetic variants in methionine synthase reductase (MTRR) and methylenetetrahydrofolate reductase (MTHFR) and CRC survival. Methods: This study used data collected from the Newfoundland Familial Colorectal Cancer Study. A total of 532 patients diagnosed with CRC for the first time from 1999 to 2003 were enrolled, and their mortality were tracked until April 2010. DNA samples were genotyped by Illumina’s integrated quantum 1 million chip. Cox models were established to assess 33 tag single-nucleotide polymorphisms in MTRR and MTHFR in relation to overall survival (OS), disease-free survival (DFS) and CRC-specific survival. Results: The MTRR and MTHFR genes were associated with DFS and CRC-specific survival in CRC patients at the gene level. After multiple comparison adjustment, MTRR rs1801394 A (vs. G) allele was associated with increased DFS (p = 0.024), while MTHRT rs3737966 (G vs. A), rs4846049 (T vs. G), rs1476413 (A vs. G), rs1801131 (C vs. A), rs12121543 (A vs. C), rs1801133 (C vs. T), rs4846052 (T vs. C), rs2066471 (A vs. G) and rs7533315 (T vs. C) were related to worse CRC-specific survival. Additionally, significant interactions were seen among pre-diagnostic alcohol consumption with MTRR rs1801394, rs3776467, rs326124, rs162040, and rs3776455, with superior OS associated with those protective variant alleles limited to patients with alcohol consumption under the median. The MTHFR rs3737966 (G vs. A) allele seemed to be detrimental to CRC survival only among subjects with fruit intake below the median. Conclusions: Polymorphic variants in MTRR and MTHFR genes that code for key enzymes for folate metabolism may be associated with survival in patients with CRC. The gene-CRC outcome association seems modulated by alcohol drinking and fruit intake.

We examined dietary fiber intake for its relevance to Colorectal cancer (CRC) survival in a cohort of CRC patients and a meta-analysis including results from four prospective cohort studies. We analyzed 504 CRC patients enrolled in the Newfoundland Familial Colorectal Cancer Study (NFCCS) who were newly diagnosed with CRC between 1999 and 2003. Follow-up for deaths was through April 2010. All participants completed a self-administered food frequency questionnaire to evaluate their dietary intakes one year before diagnosis. Multivariable Cox proportional hazard models were used to explore the associations of dietary fiber intake with all-cause mortality and CRC-specific mortality. In the meta-analysis, we identified prospective cohort studies published between January 1991 and December 2021 by searching PubMed, EMBASE, and Cochrane Library. Fixed-effects or random-effects models were used to combine the study-specific hazard ratio (HR) from our original analysis and three other cohorts. In the NFCCS, we found that CRC patients with the second quartile of dietary fiber intake had a 42% lower risk of all-cause mortality (HR: 0.58, 95% CI: 0.35–0.98) and 58% lower risk of CRC-specific mortality (HR: 0.42, 95% CI: 0.21–0.87) compared with those with the lowest quartile. In the meta-analysis, a similar inverse association between dietary fiber and total mortality was detected among CRC patients; each 10 g/day increase in dietary fiber intake was associated with a 16% decreased risk of total mortality. The dose–response meta-analysis showed a linear relationship between dietary fiber intake and all-cause mortality, with no sign of a plateau. For CRC-specific mortality, intriguingly, the benefit associated with increasing dietary fiber intake achieved its maximum at approximately 22 g/day, and no further reduction in CRC-specific mortality was observed beyond this intake level. Our results suggest that high dietary fiber intake may be associated with prolonged survival among CRC patients. Our findings add to the sparse literature on the role of dietary fiber in CRC survival.

Background: SNP interactions may explain the variable outcome risk among colorectal cancer patients. Examining SNP interactions is challenging, especially with large datasets. Multifactor Dimensionality Reduction (MDR)-based programs may address this problem.Objectives: 1) To compare two MDR-based programs for their utility; and 2) to apply these programs to sets of MMP and VEGF-family gene SNPs in order to examine their interactions in relation to colorectal cancer survival outcomes.Methods: This study applied two data reduction methods, Cox-MDR and GMDR 0.9, to study one to three way SNP interactions. Both programs were run using a 5-fold cross validation step and the top models were verified by permutation testing. Prognostic associations of the SNP interactions were verified using multivariable regression methods. Eight datasets, including SNPs from MMP family genes (n = 201) and seven sets of VEGF-family interaction networks (n = 1,517 SNPs) were examined.Results: ∼90 million potential interactions were examined. Analyses in the MMP and VEGF gene family datasets found several novel 1- to 3-way SNP interactions. These interactions were able to distinguish between the patients with different outcome risks (regression p-values 0.03–2.2E-09). The strongest association was detected for a 3-way interaction including CHRM3.rs665159_EPN1.rs6509955_PTGER3.rs1327460 variants.Conclusion: Our work demonstrates the utility of data reduction methods while identifying potential prognostic markers in colorectal cancer.

We aimed to examine the associations of a genome‐wide set of single nucleotide polymorphisms (SNPs) and 254 copy number variations (CNVs) and/or insertion/deletions (INDELs) with clinical outcomes in colorectal cancer patients (n = 505). We also aimed to investigate whether their associations changed (e.g., appeared, diminished) over time. Multivariable Cox proportional hazards and piece‐wise Cox regression models were used to examine the associations. The Cancer Genome Atlas (TCGA) datasets were used for replication purposes and to examine the gene expression differences between tumor and nontumor tissue samples. A common SNP (WBP11‐rs7314075) was associated with disease‐specific survival with P‐value of 3.2 × 10⁻⁸. Association of this region with disease‐specific survival was also detected in the TCGA patient cohort. Two expression quantitative trait loci (eQTLs) were identified in this locus that were implicated in the regulation of ERP27 expression. Interestingly, expression levels of ERP27 and WBP11 were significantly different between colorectal tumors and nontumor tissues. Three SNPs predicted the risk of recurrent disease only after 5 years postdiagnosis. Overall, our study identified novel variants, one of which also showed an association in the TCGA dataset, but no CNVs/INDELs, that associated with outcomes in colorectal cancer. Three SNPs were candidate predictors of long‐term recurrence/metastasis risk.

The intention-to-treat analysis is the gold standard for evaluating the efficacy in a randomized controlled trial. However, when non-adherence to randomized treatments is high the actual treatment effect may be underestimated. The impact of drop-out from the intervention group or drop-in to the control group may be controlled by trial design, increasing the sample size, effective study execution, and a pre-specified analytical plan to take contamination into account. These analyses may include censoring at the time of co-interventions associated with stopping treatment, lag censoring which allows an additional period after discontinuation of study treatment to account for residual treatment effects, inverse probability of censoring weights (IPCW), accelerated failure time models, and contamination adjusted intent-to-treat analysis. These methods are particularly useful in assessing the “prescribed efficacy” of the study treatment, which can aid clinical decision-making.

The quantity of health spending does not equate directly to the quality of healthcare. Social spending and nonmedical determinants of health impact on health outcomes. The value of health spending in the United States is limited by the high cost of goods and labor, and administrative costs, high use of low-level care, and lack of basic healthcare coverage in a substantial minority of the population. There are additional reasons for differences in the value of health spending as revealed by a comparison of Canada and Australia, where similar spending per capita results in better health system performance in Australia than in Canada. Within Canada, there are large differences between provinces in the value of health spending. Overutilization of health resources occurs when an intervention is not indicated, or of low value, or is provided at the wrong time. The Choosing Wisely Initiative is a solution for overutilization of low-value care. Recognition of the reality that effective treatments are often applied inconsistently and inappropriately has spurred the development of a science concerned with improvement of quality of care.

Quality-of-life (QoL) outcomes are important elements of randomized controlled trials. The instruments for measurement of QoL vary but usually multiple comparisons are possible, a concern that can be offset by prespecifying the outcomes of interest. Missing data may threaten the validity of QoL assessments in trials. Therefore, familiarity with the strategies used to account for missing data is necessary. Measures that incorporate both survival and QoL are helpful for treatment decisions. The definition of minimal clinically important differences in QoL scores is important and often derived using inadequate methods.

When analyzing the results of a trial, the primary outcome variable must be kept in clear focus. In the analysis plan, consideration must be given to comparing the characteristics of the subjects, taking account of differences in these characteristics, intention-to-treat analysis, interim analyses and stopping rules, mortality comparisons, composite outcomes, research design including run-in periods, factorial, stratified and crossover designs, number needed to treat, power issues, multivariate modeling, subgroup analysis, competing risks, and hypothesis-generating analyses.