Germline genetic regulation of the colorectal tumor immune microenvironment.

OBJECTIVE: To evaluate the contribution of germline genetics to regulating the briskness and diversity of T cell responses in CRC, we conducted a genome-wide association study to examine the associations between germline genetic variation and quantitative measures of T cell landscapes in 2,876 colorectal tumors from participants in the Molecular Epidemiology of Colorectal Cancer Study (MECC). METHODS: Germline DNA samples were genotyped and imputed using genome-wide arrays. Tumor DNA samples were extracted from paraffin blocks, and T cell receptor clonality and abundance were quantified by immunoSEQ (Adaptive Biotechnologies, Seattle, WA). Tumor infiltrating lymphocytes per high powered field (TILs/hpf) were scored by a gastrointestinal pathologist. Regression models were used to evaluate the associations between each variant and the three T-cell features, adjusting for sex, age, genotyping platform, and global ancestry. Three independent datasets were used for replication. RESULTS: We identified a SNP (rs4918567) near RBM20 associated with clonality at a genome-wide significant threshold of 5 × 10- 8, with a consistent direction of association in both discovery and replication datasets. Expression quantitative trait (eQTL) analyses and in silico functional annotation for these loci provided insights into potential functional roles, including a statistically significant eQTL between the T allele at rs4918567 and higher expression of ADRA2A (P = 0.012) in healthy colon mucosa. CONCLUSIONS: Our study suggests that germline genetic variation is associated with the quantity and diversity of adaptive immune responses in CRC. Further studies are warranted to replicate these findings in additional samples and to investigate functional genomic mechanisms.

Heterozygote advantage at HLA class I and II loci and reduced risk of colorectal cancer.

Objective: Reduced diversity at Human Leukocyte Antigen (HLA) loci may adversely affect the host's ability to recognize tumor neoantigens and subsequently increase disease burden. We hypothesized that increased heterozygosity at HLA loci is associated with a reduced risk of developing colorectal cancer (CRC). Methods: We imputed HLA class I and II four-digit alleles using genotype data from a population-based study of 5,406 cases and 4,635 controls from the Molecular Epidemiology of Colorectal Cancer Study (MECC). Heterozygosity at each HLA locus and the number of heterozygous genotypes at HLA class -I (A, B, and C) and HLA class -II loci (DQB1, DRB1, and DPB1) were quantified. Logistic regression analysis was used to estimate the risk of CRC associated with HLA heterozygosity. Individuals with homozygous genotypes for all loci served as the reference category, and the analyses were adjusted for sex, age, genotyping platform, and ancestry. Further, we investigated associations between HLA diversity and tumor-associated T cell repertoire features, as measured by tumor infiltrating lymphocytes (TILs; N=2,839) and immunosequencing (N=2,357). Results: Individuals with all heterozygous genotypes at all three class I genes had a reduced odds of CRC (OR: 0.74; 95% CI: 0.56-0.97, p= 0.031). A similar association was observed for class II loci, with an OR of 0.75 (95% CI: 0.60-0.95, p= 0.016). For class-I and class-II combined, individuals with all heterozygous genotypes had significantly lower odds of developing CRC (OR: 0.66, 95% CI: 0.49-0.87, p= 0.004) than those with 0 or one heterozygous genotype. HLA class I and/or II diversity was associated with higher T cell receptor (TCR) abundance and lower TCR clonality, but results were not statistically significant. Conclusion: Our findings support a heterozygote advantage for the HLA class-I and -II loci, indicating an important role for HLA genetic variability in the etiology of CRC.

Transformer-based biomarker prediction from colorectal cancer histology: A large-scale multicentric study.

Deep learning (DL) can accelerate the prediction of prognostic biomarkers from routine pathology slides in colorectal cancer (CRC). However, current approaches rely on convolutional neural networks (CNNs) and have mostly been validated on small patient cohorts. Here, we develop a new transformer-based pipeline for end-to-end biomarker prediction from pathology slides by combining a pre-trained transformer encoder with a transformer network for patch aggregation. Our transformer-based approach substantially improves the performance, generalizability, data efficiency, and interpretability as compared with current state-of-the-art algorithms. After training and evaluating on a large multicenter cohort of over 13,000 patients from 16 colorectal cancer cohorts, we achieve a sensitivity of 0.99 with a negative predictive value of over 0.99 for prediction of microsatellite instability (MSI) on surgical resection specimens. We demonstrate that resection specimen-only training reaches clinical-grade performance on endoscopic biopsy tissue, solving a long-standing diagnostic problem.

Tumor immune infiltration estimated from gene expression profiles predicts colorectal cancer relapse.

A substantial fraction of patients with stage I-III colorectal adenocarcinoma (CRC) experience disease relapse after surgery with curative intent. However, biomarkers for predicting the likelihood of CRC relapse have not been fully explored. Therefore, we assessed the association between tumor infiltration by a broad array of innate and adaptive immune cell types and CRC relapse risk. We implemented a discovery-validation design including a discovery dataset from Moffitt Cancer Center (MCC; Tampa, FL) and three independent validation datasets: (1) GSE41258 (2) the Molecular Epidemiology of Colorectal Cancer (MECC) study, and (3) GSE39582. Infiltration by 22 immune cell types was inferred from tumor gene expression data, and the association between immune infiltration by each cell type and relapse-free survival was assessed using Cox proportional hazards regression. Within each of the four independent cohorts, CD4+ memory activated T cell (HR: 0.93, 95% CI: 0.90-0.96; FDR = 0.0001) infiltration was associated with longer time to disease relapse, independent of stage, microsatellite instability, and adjuvant therapy. Based on our meta-analysis across the four datasets, 10 innate and adaptive immune cell types associated with disease relapse of which 2 were internally validated using multiplex immunofluorescence. Moreover, immune cell type infiltration was a better predictors of disease relapse than Consensus Molecular Subtype (CMS) and other expression-based biomarkers (Immune-AICMCC:238.1-238.9; CMS-AICMCC: 241.0). These data suggest that transcriptome-derived immune profiles are prognostic indicators of CRC relapse and quantification of both innate and adaptive immune cell types may serve as candidate biomarkers for predicting prognosis and guiding frequency and modality of disease surveillance.

Bridging the Gap: Patient Navigation Increases Colonoscopy Follow-up After Abnormal FIT.

INTRODUCTION: Recent studies indicate low rates of follow-up colonoscopy after abnormal fecal immunochemical testing (FIT) within safety net health systems. A patient navigation (PN) program is an evidence-based strategy that has been shown to improve colonoscopy completion in private and public healthcare settings. The aim of this study was to evaluate the effectiveness of a PN program to encourage follow-up colonoscopy after abnormal FIT within a large safety net hospital system. METHODS: We established an enterprisewide PN program at 5 tertiary care hospitals within the Los Angeles County Department of Health Services system in 2018. The PN assisted adult patients aged 50-75 years with an abnormal FIT to a follow-up colonoscopy within 6 months. PN activities included initiating referral for and scheduling of colonoscopy, performing reminder phone calls to patient for their upcoming colonoscopy, and following up with patients who did not attend their colonoscopy. We assess the effectiveness of the PN intervention by comparing follow-up colonoscopy rates with a period before the intervention. RESULTS: There were 2,531 patients with abnormal FIT results (n = 1,214 in 2017 and n = 1,317 in 2018). A majority were women (55% in 2017 vs 52% in 2018) with a mean age of 60 ± 6.2 years. From a previous mean of 163 days without PN in 2017, the mean time from abnormal FIT to colonoscopy with PN improved to 113 days in 2018. The frequency of colonoscopy completion with PN increased from 40.6% (n = 493) in 2017 to 46% (n = 600) in 2018. DISCUSSION: After the introduction of the PN program, there was a significant increase in patients undergoing follow-up colonoscopy after abnormal FIT and patients were more likely to undergo colonoscopy within the recommended 6 months.

Incidence of hospital-acquired pressure ulcers - a population-based cohort study.

Our study sought to estimate the association between race, gender, comorbidity and body mass index (BMI) on the incidence of hospital-acquired pressure ulcer (PU) from a population-based retrospective cohort comprising 242 745 unique patient hospital discharges in two fiscal years from July 2009 to June 2010 from 15 general and tertiary care hospitals. Cases were patients with a single inpatient encounter that led to an incident PU. Controls were patients without a PU at any encounter during the two fiscal years with the earliest admission retained for analysis. Logistic regression models quantified the association of potential risk factors for PU incidence. Spline functions captured the non-linear effects of age and comorbidity. Overall 2·68% of patients experienced an incident PU during their inpatient stay. Unadjusted analyses revealed statistically significant associations by age, gender, race, comorbidity, BMI, admitted for a surgical procedure, source of admission and fiscal year, but differences by gender and race did not persist in adjusted analyses. Interactions between age, comorbidity and BMI contributed significantly to the likelihood of PU incidence. Patients who were older, with multiple comorbidities and admitted for a surgical diagnosis-related groups (DRG) were at greater risk of experiencing a PU during their stay.

Pedigree structure and kinship measurements of a mid-Michigan community: a new North American population isolate identified.

Previous studies identified a cluster of individuals with an autosomal recessive form of deafness that reside in a small region of mid-Michigan. We hypothesized that affected members from this community descend from a defined founder population. Using public records and personal interviews, we constructed a genealogical database that includes the affected individuals and their extended families as descendants of 461 settlers who emigrated from the Eifel region of Germany between 1836 and 1875. The genealogical database represents a 13-generation pedigree that includes 27,747 descendants of these settlers. Among these descendants, 13,784 are presumed living. Many of the extant descendants reside in a 90-square-mile area, and 52% were born to parents who share at least one common ancestor. Among those born to related parents, the median kinship coefficient is 3.7 × 10(-3). While the pedigree contains 2,510 founders, 344 of the 461 settlers accounted for 67% of the genome in the extant population. These data suggest that we identified a new population isolate in North America and that, as demonstrated for congenital hearing loss, this rural mid-Michigan community is a new resource to discover heritable factors that contribute to common health-related conditions.

The Study to Explore Early Development (SEED): a multisite epidemiologic study of autism by the Centers for Autism and Developmental Disabilities Research and Epidemiology (CADDRE) network.

The Study to Explore Early Development (SEED), a multisite investigation addressing knowledge gaps in autism phenotype and etiology, aims to: (1) characterize the autism behavioral phenotype and associated developmental, medical, and behavioral conditions and (2) investigate genetic and environmental risks with emphasis on immunologic, hormonal, gastrointestinal, and sociodemographic characteristics. SEED uses a case-control design with population-based ascertainment of children aged 2-5 years with an autism spectrum disorder (ASD) and children in two control groups-one from the general population and one with non-ASD developmental problems. Data from parent-completed questionnaires, interviews, clinical evaluations, biospecimen sampling, and medical record abstraction focus on the prenatal and early postnatal periods. SEED is a valuable resource for testing hypotheses regarding ASD characteristics and causes.

Nonsteroidal anti-inflammatory drugs and risk of melanoma.

Because nonsteroidal anti-inflammatory drugs (NSAIDs) inhibit tumor growth in vitro, we investigated the association between NSAIDs and melanoma to determine if there was epidemiologic evidence of a chemopreventive effect from these medications. Three hundred twenty-seven subjects with incident melanoma and 119 melanoma-free controls completed a structured interview assessing melanoma risk factors. The unadjusted odds ratio (OR) for use of nonaspirin NSAIDs was 0.58 (95% CI 0.31-1.11), in a comparison of subjects with melanoma to controls. After adjustment for melanoma risk factors, the OR was 0.71 (95% CI 0.23-2.02). Aspirin users had an unadjusted OR of 0.85 (95% CI 0.45-1.69) and an adjusted OR of 1.45 (95% CI 0.44-4.74). In this pilot study, we found no evidence of a significant association between analgesic use and melanoma risk when potential confounders are assessed. Based on conflicting reports in the literature, meta-analysis may be appropriate.

Pathologic predictors of microsatellite instability in colorectal cancer.

Identification of microsatellite unstable (MSI-H) colorectal cancers (CRCs) is important not only for the identification of hereditary nonpolyposis colorectal cancer syndrome but also because MSI-H CRCs have a better prognosis and may respond differently to 5-fluorouracil-based chemotherapy. We present 2 nearly equivalent logistic regression models for clinical use that predict microsatellite instability based on the review of 1649 CRCs from patients of all ages collected in a population-based case control study in northern Israel. One hundred ninety-eight of these 1649 tumors demonstrated a high degree of microsatellite instability (12%). Multivariate analysis found that >2 tumor-infiltrating lymphocyte (TIL) cells per high-powered field, the lack of dirty necrosis, the presence of a Crohn-like reaction, right-sided location, any mucinous differentiation (mucinous or focally mucinous) and well or poor differentiation, and age less than 50 were all independent predictors of MSI-H. We developed 2 logistic regression models that differ only by the statistical approach used to analyze the number of TIL cells per high-powered field, where the slightly more accurate (and complex) model uses the log of the total number of TIL cells. The simpler clinical model uses a cut-off of 2>TIL cells per high-powered field. The accuracy of both models is high, with an 85.4% versus 85.0% probability of correctly classifying tumors as MSI-H. By employing the simpler model, pathologists can predict the likelihood of microsatellite instability by compiling the MSI probability score (Table 4 and Fig. 1) from simple histologic and clinical data available during sign-out. Our model shows that approximately 43% of CRCs have a MSI probability score of 1 or less and hence have little likelihood (<3%) of being MSI-H. Although this model is not perfect in predicting microsatellite instability, its use could improve the efficiency of expensive diagnostic testing.

Genetic variation in 8q24 associated with risk of colorectal cancer.

Chromosome 8q24 harbors oncogenes known to be involved in pathogenesis of colorectal cancer (CRC) as well as uncharacterized genetic variants that have recently been shown to influence inherited risk of prostate cancer. In a population-based case-control study of colorectal cancer in northern Israel, we investigated the association between variation in 8q24 and risk of CRC. Among 1,861 incident cases and 1,937 population-based controls matched on age, gender, ethnicity, and clinic, rs10505477 was associated with risk of CRC in a dominant model, with an odds ratio = 1.23, 95% confidence interval = 1.05-1.43, (p = 0.008). This association was independently validated in an analysis of cancer among relatives of carriers of the risk allele, with a hazard ratio of 3.2 (95% bootstrap CI = 1.16-17.8). Genetic variation at rs10505477 on 8q24 potentially accounts for 14% of CRC in this population and should be replicated in other studies.

APC I1307K and the risk of prostate cancer.

The kin-cohort design has been proposed as an alternative to traditional case-control and cohort measures to evaluate inherited susceptibility to cancer in population-based studies. Here, we used this design to evaluate inherited susceptibility to prostate cancer associated with APC I1307K using data from the Molecular Epidemiology of Colorectal Cancer study. Two techniques were used to compare the incidence of prostate cancer in APC I1307K carriers. First, we compared the incidence of prostate cancer in relatives of mutation carriers and noncarriers using standard techniques for survival analysis. Second, we used the marginal maximum likelihood method for kin-cohort analysis to infer the genotypes in the relatives. We also evaluated APC I1307K in 75 Ashkenazi Jewish individuals with prostate cancer from 27 families enrolled in the University of Michigan Prostate Cancer Genetic Study. We observed a slightly increased risk of prostate cancer in relatives of APC I1307K carriers, however, this difference was not statistically significant (hazard ratio, 1.6; 95% confidence intervals, 0.7-3.4). Similar conclusions were drawn using both techniques for kin-cohort analysis. APC I1307K was found in 7.4% of families genotyped, which is slightly higher than the allele prevalence reported in Ashkenazi Jews in the general population. In addition, we did not observe loss of heterozygosity at APC or a somatic mutation near APC I1307K using microdissected tumor DNA from mutation carriers enrolled in the Prostate Cancer Genetic Study. Overall, the evidence for an association between APC I1307K and prostate cancer is not compelling. APC I1307K is unlikely to play a clinically meaningful role in susceptibility to prostate cancer.

Colorectal polyps in carriers of the APC I1307K polymorphism.

PURPOSE: The probability of colorectal cancer is moderately increased among carriers of the APC I1307K polymorphism. However, it is not known if endoscopic surveillance of this high-risk group is warranted. The prevalence of polyps and adenomas in specimens of colorectal cancer who are carriers and noncarriers of the APC I1307K polymorphism is compared. METHOD: Prevalence of adenomatous polyps in the pathology specimens of the study participants, stratified by their APC I1307K polymorphism status, was studied in 900 consecutive cases of colorectal cancer diagnosed in northern Israel between 1998 and 2002, within the framework of a population-based, case-controlled study (MECC Study). RESULTS: The APC I1307K mutation was detected in 78 colorectal cancer cases (8.7 percent) of the study population. Prevalence was higher among Ashkenazi Jews (11.2 percent) than among non-Ashkenazi Jews (2.7 percent) or Arabs (3.1 percent). After adjustment for age, APC I1307K carriers were significantly more likely than noncarriers to have polyps in their surgical specimen (51.3 percent vs. 32.6 percent, P = 0.002). Adenomas with a tubular component (either tubular adenomas or tubulovillous adenomas), but not villous adenomas, were significantly more frequent among carriers (37.2 percent vs. 23.6 percent, P = 0.005). CONCLUSION: Together with former evidence of I1307K being a risk factor for colorectal cancer, these data suggest that colonoscopic surveillance for colorectal adenomas and cancer may be warranted in I1307K carriers, even in the absence of other identifiable risk factors.

Statins and the risk of colorectal cancer.

BACKGROUND: Statins are inhibitors of 3-hydroxy-3-methylglutaryl coenzyme A reductase and effective lipid-lowering agents. Statins inhibit the growth of colon-cancer cell lines, and secondary analyses of some, but not all, clinical trials suggest that they reduce the risk of colorectal cancer. METHODS: The Molecular Epidemiology of Colorectal Cancer study is a population-based case-control study of patients who received a diagnosis of colorectal cancer in northern Israel between 1998 and 2004 and controls matched according to age, sex, clinic, and ethnic group. We used a structured interview to determine the use of statins in the two groups and verified self-reported statin use by examining prescription records in a subgroup of patients for whom prescription records were available. RESULTS: In analyses including 1953 patients with colorectal cancer and 2015 controls, the use of statins for at least five years (vs. the nonuse of statins) was associated with a significantly reduced relative risk of colorectal cancer (odds ratio, 0.50; 95 percent confidence interval, 0.40 to 0.63). This association remained significant after adjustment for the use or nonuse of aspirin or other nonsteroidal antiinflammatory drugs; the presence or absence of physical activity, hypercholesterolemia, and a family history of colorectal cancer; ethnic group; and level of vegetable consumption (odds ratio, 0.53; 95 percent confidence interval, 0.38 to 0.74). The use of fibric-acid derivatives was not associated with a significantly reduced risk of colorectal cancer (odds ratio, 1.08; 95 percent confidence interval, 0.59 to 2.01). Self-reported statin use was confirmed for 276 of the 286 participants (96.5 percent) who reported using statins and whose records were available. CONCLUSIONS: The use of statins was associated with a 47 percent relative reduction in the risk of colorectal cancer after adjustment for other known risk factors. Because the absolute risk reduction is likely low, further investigation of the overall benefits of statins in preventing colorectal cancer is warranted.

EGF gene polymorphism and the risk of incident primary melanoma.

Overexpression of the epidermal growth factor (EGF) pathway has been implicated in melanoma pathogenesis, and a recent case-control study identified a single nucleotide polymorphism (G to A) in the EGF gene where the G allele was associated with increased EGF expression and an increased risk of melanoma. To further evaluate this association, we conducted a case-control analysis from the Genes, Environment, and Melanoma study at the University of Michigan site using two different study designs. Incident cases of histopathologically confirmed first primary melanoma that were diagnosed between January 1, 2000 and December 31, 2000 from the University of Michigan Melanoma Clinic (n = 330) were compared with the following two different sources of nonmelanoma controls: spouse/friend controls (n = 84) and healthy volunteer controls from a case-control study of psoriasis (n = 148). Using a second analytic design, comparisons between multiple primary melanoma cases (n = 62) and single primary melanoma cases (n = 330) were also evaluated to estimate odds ratios (ORs). Genotyping for the single nucleotide substitution (G to A) at position 61 in the 5' untranslated region of the EGF gene was performed from genomic DNA, and epidemiological risk factors were assessed through a telephone interview. When EGF genotypes were compared between incident primary melanoma cases and the nonmelanoma controls, the risk associated with the homozygous G/G genotype was not statistically significantly associated with an increased risk for incident primary melanoma compared with the homozygous A/A genotype [OR, 1.09; 95% confidence interval (CI); 0.65-1.85]. No strong associations with EGF G/G genotype were observed in comparisons of multiple primary and single primary melanoma cases (OR, 0.66; 95% CI; 0.25-1.73). Case subjects with tumors >/=3.5 mm compared with those <3.5 mm were not significantly associated with the G/G genotype (OR, 0.54; 95% CI; 0.12-2.35). Our data do not support a significant association between melanoma and the EGF 61*G allele or the homozygous G/G genotype. The EGF polymorphism is not a reproducible risk factor for melanoma or thick melanoma in our data. The two analytic approaches used in the study provide evidence against a strong association between EGF 61*G and melanoma and demonstrate the potential utility of case-case designs for evaluating the role of single nucleotide polymorphisms and cancer. Additional independent studies will be required to elucidate relationships between genetic variation in the EGF gene and risk of melanoma.

Use of complementary and alternative medicine in men with family history of prostate cancer: a pilot study.

OBJECTIVES: To describe the use of complementary and alternative medicines (CAMs) among men with a family history of prostate cancer and to evaluate the relationship between selected sociodemographic and behavioral characteristics and the use of CAMs. METHODS: Unaffected brothers of men diagnosed with prostate cancer were asked to participate in a short computer-assisted telephone interview. The survey focused primarily on the use of different vitamins, herbal supplements, and medications, some of which are marketed for prostate health or prostate cancer prevention. RESULTS: A total of 111 men completed the survey, representing 66% of eligible study subjects. Of the 111 men, 61 (55%) reported currently taking some form of CAM, with 30% taking a vitamin or supplement purported to have prostate-specific benefits. The prevalence of CAM use generally increased with increasing age; however, men who were younger than their affected brother at the time of the diagnosis of prostate cancer were more likely to use CAMs than were older brothers. CONCLUSIONS: Most men with a family history of prostate cancer take vitamins and supplements, some of which are believed to prevent future cancer occurrence. The results of this study and others provide some insight into the determinants of potentially beneficial health behaviors in high-risk individuals.

BRCA1 and BRCA2 founder mutations and the risk of colorectal cancer.

BACKGROUND: Mutations in BRCA1 and/or BRCA2 (BRCA1/2) profoundly increase the risks of breast and ovarian cancers, but it is unclear whether mutations in these genes increase the risk of colorectal cancer. We investigated BRCA1/2 founder mutations and a family history of breast cancer as potential risk factors for colorectal cancer. METHODS: In the population-based Molecular Epidemiology of Colorectal Cancer study in northern Israel, 1422 case patients with incident colorectal cancer, diagnosed between March 31, 1998, and December 31, 2002, and 1566 control subjects without colorectal cancer were genotyped for the BRCA1 187delAG, BRCA1 5385insC, and BRCA2 6174delT founder mutations. Genotypes and interview data from all case patients and control subjects and from only those of Ashkenazi Jewish descent (1002 case patients and 1038 control subjects) were used to calculate odds ratios [ORs] from logistic regression. RESULTS: Twenty-four (2.4%) case patients and 20 (1.9%) control subjects carried one of the three mutations (OR = 1.24, 95% confidence interval [CI] = 0.68 to 2.26). A family history of breast cancer in a female relative was not associated with an increased risk of colorectal cancer, even after adjustment for the presence of a BRCA founder mutation (OR = 1.03, 95% CI = 0.75 to 1.41). CONCLUSIONS: Although weak associations cannot be excluded, Ashkenazi BRCA founder mutations do not confer a strongly elevated risk of colorectal cancer. Similarly, a family history of breast cancer does not appear to be a strong risk factor for colorectal cancer in this population.

Phenotype of microsatellite unstable colorectal carcinomas: Well-differentiated and focally mucinous tumors and the absence of dirty necrosis correlate with microsatellite instability.

The phenotypic markers of colorectal carcinomas with microsatellite instability have been widely studied and include mucinous or poor differentiation, prominent host response, a circumscribed growth pattern, histologic heterogeneity, and right-sided location. As part of a population-based case-control study of colorectal cancer in northern Israel, we reviewed the pathology and microsatellite status of 528 consecutively diagnosed colorectal cancers. Phenotypic analysis was performed by one pathologist (J.K.G.) and included assessment of grade, mucinous histology (>50%, or focal), histologic heterogeneity, growth pattern, necrosis, and host response. Microsatellite status was determined on microdissected portions of formalin-fixed, paraffin-embedded tissue using a panel of 5 NCI consensus primers. Fifty-two of 528 colorectal carcinomas were microsatellite unstable (9.85%). Multivariate analysis found that >2 tumor infiltrating lymphocytes per high power field (p <0.0001), the lack of dirty necrosis (p = 0.0054), a Crohn's-like host response (p = 0.0064), right-sided location (p = 0.032), well or poor differentiation (p = 0.037), and any mucinous differentiation (p = 0.039) were independent predictors of microsatellite instability. Tumor infiltrating lymphocytes were the single best histologic predictor of microsatellite instability. The absence of dirty necrosis and the presence of well-differentiated tumors and tumors with only focal mucinous differentiation were also important markers for microsatellite instability that have not been emphasized previously. The combination of >2 tumor infiltrating lymphocytes per high power field and/or any mucinous differentiation and/or the absence of dirty necrosis identified all MSI-H tumors in this study.