TY - JOUR
AU - Markowitz, Sanford D.
AB - Abstract Background: Siblings and other first-degree relatives of patients with “sporadic” (i.e., apparently nonfamilial) colorectal cancer or precursor adenomatous colon polyps have an increased risk of developing colon neoplasia. This observation suggests the presence of inherited genetic determinants for sporadic colon neoplasia. Mice homozygous for a null cyclooxygenase 2 (COX2) (also called PTGS2) allele have a dramatically reduced susceptibility to the development of intestinal adenomas. In humans, use of pharmacologic inhibitors of COX2 enzyme activity are associated with reduced risk of colon neoplasia. This study examined whether the human COX2 locus may be linked to colon neoplasia in humans. Methods: We used the affected sibling-pair method to test for linkage of the human COX2 locus to colon neoplasia. Results: We examined 74 concordantly affected sibling pairs from 46 sibships with colon neoplasia. One hundred five siblings from these sibships were diagnosed with either colorectal cancer or colon adenomatous polyps before age 65 years. No linkage between COX2 and colon neoplasia was found by use of a multipoint model-free linkage analysis (estimate of allele sharing was 0.44; standard error = ±0.04; 95% confidence interval = 0.36 to 0.52). Moreover, even allowing for heterogeneity, the potential that a COX2 colon neoplasia susceptibility variant was present within a substantial subset of these sibships was strongly excluded under either a recessive or a dominant inheritance model (95% confidence to exclude a model in which 2.7% or more of the sibling pairs harbor a dominant susceptibility allele). Conclusions: This study of concordantly affected sibling pairs thus demonstrates that variations in the COX2 gene are unlikely to be a source of individual susceptibility to colon neoplasia in humans. An inherited predisposition to colorectal cancer has been associated with several highly penetrant single gene disorders that lead to colon cancer in young individuals. Relevant genes include APC, whose germline mutations account for the well-recognized familial adenomatous polyposis (FAP) syndrome, and DNA mismatch repair genes, whose germline mutations cause hereditary nonpolyposis colorectal cancer (HNPCC) (1). However, it is estimated that these syndromes account for no more than 5%–10% of all cases of colorectal cancer in humans [reviewed in (2)]. Several lines of evidence suggest that the general population harbors an unidentified susceptibility gene or genes that increase the risk of “sporadic” (i.e., apparently nonfamilial) colorectal cancer arising in typical middle-aged adults. For example, siblings and other first-degree relatives of individuals with either adenomatous polyps or colorectal cancer have a twofold to threefold elevation in risk for developing colon adenomas and cancers of the colon or rectum compared with the general population (3,4). Segregation analysis of colorectal cancer kindreds also suggests the presence of susceptibility alleles for adenomatous polyps (5,6). Cyclooxygenase 2 (COX2) (also called PTGS2) encodes an inducible cyclooxygenase that catalyzes the first committed step in prostaglandin synthesis [reviewed in (7,8)]. COX2 has an important role in colorectal tumorigenesis, since both the COX2 message and the protein product are increased in colon tumors and in premalignant adenomatous colon polyps (9,10). Furthermore, a reduction in COX2 activity powerfully retards the development of colon neoplasia in mice and in humans. For example, mice predisposed to develop intestinal adenomas by virtue of a heterozygous germline mutation in the APC gene show a remarkable 86% reduction in the number of tumors if they are also homozygous for a null COX2 allele compared with littermates who carry wild-type COX2 alleles (11). In human populations, patients who are treated with pharmacologic inhibitors of COX2 enzymatic activity, such as aspirin and nonsteroidal anti-inflammatory drugs (NSAIDs), demonstrate a one-third to two-thirds reduction in colorectal cancer risk (12,13). Moreover, one such NSAID, sulindac, has been shown to induce directly the regression of premalignant human colon adenomas in individuals with FAP (14). Genetic variation at the COX2 locus would thus be expected in humans, as in mice, to alter strongly an individual's risk of colon neoplasia development. The affected sibling-pair method is a powerful linkage technique for detecting the presence of susceptibility alleles at a given genetic locus (15,16). This method is based on the principle that two affected siblings will tend to share identical marker alleles if the marker locus is linked to the disease in question. Thus, affected siblings are statistically more likely to display increased identity by descent (IBD) at a linked locus as compared with an unlinked locus. Stated alternatively, allele sharing among concordantly affected sibling pairs should be greater than 0.5 at a marker locus linked to a disease susceptibility locus; in contrast, at any random locus, allele sharing will equal 0.5. We have employed this method to determine whether the COX2 locus itself harbors allelic variants that alter the risk of colon neoplasia susceptibility in humans. Patients and Methods Colon Neoplasia Kindred Ascertainment The sibships used for this study were recruited to join the Case Western Reserve University (CWRU) Colon Neoplasia Sibling Study (CNSS), an affected sibling-pair linkage study designed to identify novel colon cancer susceptibility genes. The Institutional Review Board of University Hospitals of Cleveland approved the study and all informed consent documents. All participants reviewed and signed the informed consent document before entering the study. Selection criteria for this study included the following: 1) an index case patient and a full sibling who were both diagnosed with colorectal cancer or colon adenomatous polyps by age 65 years, 2) histologic verification of colorectal cancer or colon polyps, 3) no histologic evidence of inflammatory bowel disease, and 4) donation of a blood sample for genetic analysis. In addition, blood samples were requested from all living first-degree family members for genotyping purposes. Four-generation pedigrees were constructed for each kindred and were reviewed for known autosomal dominant forms of colon cancer, such as FAP and HNPCC (17–20). FAP kindreds were excluded by review of the clinical records for evidence of polyposis. HNPCC kindreds were excluded by microsatellite instability analysis of tumor samples from relevant cases in kindreds that met the Bethesda criteria for HNPCC (17–20). In addition, DNA samples from all Ashkenazi Jewish patients with colorectal cancer or colon adenomatous polyps were evaluated to exclude all individuals carrying the APC I1307K allele that is found in 6% of this population and has been shown to increase the risk for colorectal cancer (17). Of the 263 probands and their kindreds reviewed for this study, 46 were eligible for genotyping and linkage analysis. Of the excluded kindreds, 206 were not eligible because of the lack of an affected full sibling of the proband or to incomplete family data, five were not eligible because of FAP, and six were not eligible because of HNPCC. No Ashkenazi sibling harbored the APC I1307K-variant allele. One hundred five affected siblings with histopathologically verified colon neoplasia and 48 additional siblings for whom no clinical colon screening information was available donated a blood sample for genotyping and statistical analysis. Twenty-five (24%) of the affected siblings were diagnosed with colorectal cancer (23 colon cancers and two rectal cancers) and 80 (76%) were diagnosed with adenomatous polyps, at an average age of 52.8 and 49.6 years, respectively. In addition, 28 (61%) of the probands were female and 54 (51%) of the siblings were female. Forty-four (96%) of the kindreds were of Northern European (Caucasian) descent. Genetic Analysis Between 15 and 20 mL of whole blood was obtained from each consenting participant, and DNA was extracted by use of standard procedures (Puregene; Gentra Systems, Inc., Minneapolis, MN). Genomic DNA (100 ng) was used in each of the 10-μL volume polymerase chain reactions (PCRs) along with 2 mM deoxyadenosine triphosphate, deoxyguanosine triphosphate, deoxycytidine triphosphate, and deoxythymidine triphosphate, 5 μM of reverse and forward primer (5′ end-labeled with T4 polynucleotide kinase and 33γ-adenosine triphosphate), 50 mM KCl, 20 mM Tris–HCl (pH 8.0), 15 mM MgCl2, and 0.75 U Taq polymerase (Life Technologies, Inc. [GIBCO BRL], Rockville, MD). The PCR reaction for D1S191 and D1S2848 was processed on an MJ PTC-100 thermocycler (MJ Research, Inc., Waltham, MA) with a 5-minute denaturation at 96 °C, followed by a 30-second denaturation at 95 °C, a 30-second annealing at 60 °C, and a 30-second extension at 72 °C, for a total of 30 cycles. After completion, 4 μL of reaction product was added to 4 μL of formamide-containing buffer and was loaded in sequential lanes on a 6% polyacrylamide/8 M urea-sequencing gel. Reaction products along with Centre d'Etude du Polymorphisme Humain (CEPH) control DNA (Coriell Cell Repository (Camden, NJ), CEPH/UTAH Pedigree 1347, #NA10859) (21) and a sequencing ladder were visualized by overnight autoradiography by use of Kodak BioMax film (Eastman Kodak Co., Rochester, NY). Each autoradiogrph was inspected and scored independently by two researchers before making the final allele assignment. The genotypes were then used to score each kindred for Mendelian inheritance and to evaluate for potential biologic incompatibility. Sibships were removed from the statistical analysis in the event of suspected misidentified parentage. A convenience sample of 100 control chromosomes was also genotyped for both D1S191 and D1S2848. Statistical Analysis Siblings with pathologically verified colon adenomatous polyps or colorectal cancer were scored as affected. Other siblings without colon screening were scored as unknown affected state. The genotypes from parents and these other siblings were used to better define the parental haplotypes of markers flanking the COX2 locus and to improve the determination of parental allele transmission. All possible distinct-affected sibling pairs were used from the triplet and quadruplet sibships. Marker alleles for each affected sibling can be shared or not shared. π, or the proportion of alleles shared IBD by an affected-sibling pair, was estimated by use of the GENIBD and analyzed by use of the SIBPAL2 multipoint linkage programs in the program package Statistical Analysis for Genetic Epidemiology (S.A.G.E.) that is based on the methods of Idury, Elston, and Haseman (22–24). π would equal 0.5 under no linkage and would be greater than 0.5 if COX2 were linked to colon neoplasia. Under the null hypothesis of no linkage, the allele sharing from different pairs in a sibship is pairwise independent and thus can be treated as independent events for the mean test (16,25). Map distances between the two flanking polymorphic dinucleotide repeat markers D1S191 and D1S2848 were estimated to be between 0.6 and 1.4 cM (26). The polymorphism information contents for markers D1S191 and D1S2848 were 0.74 and 0.84, respectively. The transmission of a preferential parental allele was analyzed by use of the global exact transmission–disequilibrium (TDTEX) program in S.A.G.E. and by the method of Spielman and Ewens (27,28) and McGinnis et al. (29). The TDTEX program analyzes all marker alleles simultaneously to discern preferential transmission of any allele to the affected siblings, so there is no need for additional correction because of multiple tests. The transmission–disequilibrium test (TDT) and the sibling–TDT (S-TDT) are methods that examine each allele separately, allowing for the development of a statistic that tests whether specific alleles are associated with the disease (30). Confidence for excluding a COX2 susceptibility allele present in a fraction (λ) of the population was based on the normal distribution of the estimate of IBD sharing (π̂) and a one-sided test. For example, in a reasonably large sample, π can be estimated as π̂, with a standard error of s, so that P(π<π̂ + 1.65s) = 0.95 for a one-sided interval. Results The degree of allele sharing at the COX2 locus was studied in siblings from 46 sibships that were identified from the CWRU CNSS. Each of the 46 sibships included at least two concordantly affected siblings in whom colon neoplasia had been diagnosed at or before the age of 65 years. In total, the affected siblings included 105 affected individuals and defined 74 concordantly affected sibling pairs. Eighty individuals (76%) were affected with adenomatous colonic polyps and 25 (24%) were affected with colorectal cancer. Thirty-five sibships had two affected siblings, nine had three, and two had four. Fig. 1 summarizes the structure of the affected sibships studied. We specifically excluded from this analysis any sibling pairs from kindreds demonstrating HNPCC or FAP and any sibship demonstrating the presence of the APC I1307K variant observed among some Ashkenazi Jews (17). COX2 Locus and Colon Neoplasia Susceptibility Genotypes were determined for two highly polymorphic microsatellite markers, D1S191 and D1S2848, that flank the COX2 locus at a distance of 0.6–1.4 cM (26). With the use of the haplotypes formed by these two loci, more than 90% of the individuals studied were fully informative for two distinguishable COX2 alleles. Genotypes of the COX2 locus markers were determined for each of the 105 affected siblings within the 46 sibships. In addition, DNA samples from 48 siblings of unknown affection status and 21 parents were genotyped. The degree of allele sharing because of IBD among the concordantly affected sibling pairs was determined by use of a model-free multipoint linkage analysis carried out in the GENIBD program within S.A.G.E. (24). For these 74 sibling pairs, the multipoint estimates of IBD allele sharing (π̂) were 0.44 for both the centromeric marker D1S191 and the telomeric marker D1S2848, with a calculated standard error of 0.04 for both estimates (Table 1). In a model-free analysis, linkage to a candidate susceptibility locus is recognized by an IBD estimate greater than 0.5. Our finding a π̂ less than 0.5 (0.44; standard error = ±0.04; 95% confidence interval = 0.36 to 0.52) excludes with 93% confidence that any future study of the COX2 locus would yield an IBD greater than 0.5 (Table 1). This estimate effectively excludes the possibility of COX2 colon neoplasia susceptibility variants segregating in the human population. A more intuitive understanding of the strength of our results can be achieved by assuming that a recessively acting COX2 colon neoplasia susceptibility variant does account for the colon neoplasia in each of our affected sibships. In this case, the expected allele sharing (π̂) would be 1.0. Our measured π of 0.44 completely excludes this model. Similarly, if we assumed that a dominant COX2 neoplasia susceptibility variant accounts for all colon neoplasia in each of our affected sibships, the expected allele sharing in the affected sibling pairs would be 0.75. In this case, our measured π̂ of 0.44 ± 0.04 excludes this model with a one-sided P value of less than 5 × 10–10. The frequencies of the different alleles of markers D1S191 and D1S2848 that we observed in these sibships were similar to both published values and to frequencies that we observed in a set of 100 control chromosomes (data not shown). However, our conclusion that the COX2 locus is not linked to colon neoplasia susceptibility is not dependent on our use of published population frequencies for the different microsatellite alleles. Indeed, π̂ determined for the COX2 locus among our sibling pairs remained less than 0.5 when the analysis was repeated, assuming that all of the alleles at D1S191 and D1S248 were equally frequent, or, alternatively, when the relative frequency of each allele was made proportional to its complement, i.e., 1 – pi, where pi is the published frequency. Evaluation of COX2 Susceptibility Alleles in Subpopulations of Sibships With Colon Neoplasia Our measured π̂ of 0.44 ± 0.04 also effectively excludes models in which colon neoplasia susceptibility would be linked to the COX2 locus among any substantial subpopulation of sibling pairs from which our sibships were sampled. For example, assume that these sibling pairs are composed of two subgroups, one accounting for a proportion (1 – λ), in which COX2 susceptibility variants are not present (i.e., in which π for COX2 = 0.5), and a second accounting for a proportion of (λ), in which a dominant or recessive COX2 susceptibility variant is present. For a dominantly acting allele in the subset of size λ, the π̂ IBD sharing of the total sibling-pair population would be 0.5(1– λ) + 0.75(λ), or 0.5 + 0.25(λ), reflecting a π of 0.5 in the unlinked subset and 0.75 in the subset with linkage. If a recessive COX2 susceptibility variant were present in a subset of size λ, the IBD for the total sibling-pair population would be 0.5 + 0.5(λ), reflecting a π of 1.0 and 0.5 for the linked and unlinked subsets, respectively. As shown in Table 2, the observed data exclude with 95% confidence a model in which 2.7% or greater of sibling pairs harbor a dominant COX2 susceptibility allele. Our data similarly exclude with 99% confidence any model in which 13% or more of our sibling-pair population harbors a dominant susceptibility allele. For a recessive susceptibility allele, we similarly exclude with 99% confidence any model in which 7% or more harbor a susceptibility allele. Thus, we can exclude with high confidence the hypothesis that a COX2 susceptibility allele is a determinant of colon neoplasia susceptibility in any substantial subgroup of the population studied. This conclusion is supported independently by our analysis of these sibling pairs by use of the TDT and the S-TDT. These tests search for linkage of colon neoplasia susceptibility to the COX2 locus by examining the number of transmissions of a given parental allele of the COX2 locus markers, D1S191 and D1S2848, to affected offspring (27,31). These tests have statistical power to discern linkage only when underlying linkage disequilibrium or allelic association exists in the population under study. Thus, these tests would be particularly powerful if linkage disequilibrium does exist between a specific COX2 susceptibility variant and COX2 marker allele, as might be the case if a susceptibility variant were segregating among affected siblings from a distinct subpopulation. The TDTEX program in S.A.G.E. was first employed to examine linkage of colon neoplasia susceptibility to the COX2 locus (24,32). Thirty-eight and 41 informative-affected siblings were analyzed at D1S191 and D1S2848, respectively. No association was detected between the transmission of marker alleles and colon neoplasia (multisided P = .93 for D1S191 and P = .88 for D1S2848; Table 3). We also found no evidence for linkage with the use of the combined TDT/S-TDT, although two alleles for marker D1S2848 resulted in a z-statistic of 1.649 for the 123-base-pair allele and a z-statistic of 1.603 for the 119-base-pair allele (corresponding to a one-sided P value = .05 and .054, respectively). These values do not reach statistical significance when allowance was made for the multiple tests performed (seven alleles for D1S191 and 11 for D1S2848). Discussion We find that the COX2 locus is highly unlikely to harbor colon neoplasia susceptibility variants in any substantial fraction of the general human population. This conclusion is based first on a model-free analysis demonstrating an absence of linkage of colon neoplasia susceptibility to the COX2 locus in 74 sibling pairs concordantly affected with colon cancer or adenomatous polyps. Moreover, we have developed a simple test statistic, λ, calculated from our estimate of IBD sharing (π̂) and from the standard error of that estimate, that allows us to exclude—with a high degree of confidence—linkage of colon neoplasia to COX2 in any substantial subpopulation of the sibships participating in our study. The results of our study are supported independently by the lack of detecting disease susceptibility alleles by use of a TDT, which is a powerful test of linkage to a candidate gene. Thus, while activity of the COX2 gene product is clearly required for the development of colon neoplasms, inherited COX2 alleles that increase the risk for intestinal neoplasia are unlikely to be a feature of the human population. We note that a common COX2 resistance variant that decreased colon neoplasia risk would also likely be detected in a study of affected sibling pairs, but that greater power for observing such a resistance variant would be obtained in an alternate design that employed concordant-unaffected sibling pairs. However, studies of attenuated phenotypes in some familial polyposis kindreds previously have not found evidence for such COX2 resistance variants acting in humans (33). In addition to excluding COX2 as an important human colon neoplasia susceptibility locus, this study illustrates the utility of the affected sibling-pair method for approaching the general problem of evaluating putative colon neoplasia susceptibility genes. One advantage of the affected sibling-pair linkage study design is that the flanking markers used to mark a given locus act as surrogates to test for disease association with all potential polymorphisms contained within the given genetic locus, including the coding region, promoters, and noncoding regions of the gene. Additionally, the affected sibling-pair study design yields conclusions more applicable to the general population of individuals with colon neoplasia and not just to potentially atypical highly selected kindreds. Moreover, experimental study designs that compare affected case patients versus control populations may be influenced by selection differences between case patients and control subjects with regard to ethnic background or other parameters. The affected sibling-pair method used here avoids this source of error by use of a disease-affected population only. Last, the affected sibling-pair method allows testing genetic linkage to diseases, such as sporadic colon neoplasia, for which disease onset is later in life and for which multigenerational kindreds may be rare. The population of affected sibling pairs employed in this study will thus be useful for the evaluation of other candidate loci as well as for supporting a future whole genome scan for detecting novel colon neoplasia susceptibility loci. Table 1. Model-free multipoint linkage analysis* Marker  π̂  SE  95% CI  P  *Model-free multipoint linkage analysis of 74 affected sibling pairs with colon adenomatous polyps or colorectal cancer. One-sided P values were nonsignificant for linkage for each flanking the cyclooxygenase 2 marker. P values were calculated from the estimate of allele sharing due to identity by descent (π̂) and its associated standard error (SE) and 95% confidence interval (CI).  D1S191  .4406  .04  .08  .93  D1S2848  .4431  .04  .08  .92  Marker  π̂  SE  95% CI  P  *Model-free multipoint linkage analysis of 74 affected sibling pairs with colon adenomatous polyps or colorectal cancer. One-sided P values were nonsignificant for linkage for each flanking the cyclooxygenase 2 marker. P values were calculated from the estimate of allele sharing due to identity by descent (π̂) and its associated standard error (SE) and 95% confidence interval (CI).  D1S191  .4406  .04  .08  .93  D1S2848  .4431  .04  .08  .92  View Large Table 2. Confidence levels for excluding the presence of a cyclooxygenase 2 susceptibility variant in a subset greater than or equal to a proportion λ of the sibling pairs concordantly affected by colon neoplasia Autosomal recessive   Autosomal dominant  λ  π̂  Confidence excluded at —  λ  π̂  Confidence excluded at —  0.0133  0.5054  95%  0.0266  0.5054  95%  0.0670  0.5323  99%  0.1340  0.5323  99%  Autosomal recessive   Autosomal dominant  λ  π̂  Confidence excluded at —  λ  π̂  Confidence excluded at —  0.0133  0.5054  95%  0.0266  0.5054  95%  0.0670  0.5323  99%  0.1340  0.5323  99%  View Large Table 3. Significance of linkage of colon neoplasia susceptibility to any cyclooxygenase 2 locus allele by use of the global exact transmission–disequilibrium test in S.A.G.E. (24) Marker  No. of siblings informative for marker  P*  *P values reflect simultaneous testing of all alleles (32).  D1S191  38  .93  D1S2848  41  .88  Marker  No. of siblings informative for marker  P*  *P values reflect simultaneous testing of all alleles (32).  D1S191  38  .93  D1S2848  41  .88  View Large Fig. 1. View largeDownload slide Shown is the family structure of each of 46 full sibships genotyped with the cyclooxygenase 2 (COX2)-flanking markers, D1S191 and D1S2848. The majority (n = 35) of the sibships consisted of two concordantly affected full siblings with colorectal cancer or adenomatous polyps (n = 19) or both siblings with adenomatous polyps (n = 16). Nine sibships had three concordantly affected siblings, and two had four concordantly affected siblings. Solid diamond = male or female with colon cancer; shaded diamond = male or female with adenomatous polyps; number in parentheses = number of sibships in each category. Fig. 1. View largeDownload slide Shown is the family structure of each of 46 full sibships genotyped with the cyclooxygenase 2 (COX2)-flanking markers, D1S191 and D1S2848. The majority (n = 35) of the sibships consisted of two concordantly affected full siblings with colorectal cancer or adenomatous polyps (n = 19) or both siblings with adenomatous polyps (n = 16). Nine sibships had three concordantly affected siblings, and two had four concordantly affected siblings. Solid diamond = male or female with colon cancer; shaded diamond = male or female with adenomatous polyps; number in parentheses = number of sibships in each category. G. L. Wiesner and P. Platzer contributed equally to this study. Supported by Public Health Service grants P30CA43703, U01CA82901 (to S. D. Markowitz), and 1K23CA81308 (to G. L. Wiesner) (National Cancer Institute), P41RR03655 (to R. C. Elston) and M01RR0008 (National Center for Research Resources), and GM28356 (National Institute of General Medical Sciences) from the National Institutes of Health, Department of Health and Human Services. S. D. Markowitz is an associate investigator of the Howard Hughes Medical Institute. We thank the individuals and families who generously participated in this study. References 1 Rustgi AK. Hereditary gastrointestinal polyposis and nonpolyposis syndromes. N Engl J Med  1994; 331: 1694–702. Google Scholar 2 Kinzler KW, Vogelstein B. Colorectal tumors. In: Vogelstein B, Kinzler KW, editors. The genetic basis of human cancer. New York (NY): McGraw-Hill; 1997. p. 566–8. Google Scholar 3 Gaglia P, Atkin WS, Whitelaw S, Talbot IC, Williams CB, Northover JM, et al. Variables associated with the risk of colorectal adenomas in asymptomatic patients with a family history of colorectal cancer. Gut  1995; 36: 385–90. Google Scholar 4 Winawer SJ, Zauber AG, Gerdes H, O'Brien MJ, Gottleig LS, Sternberg SS, et al. Risk of colorectal cancer in the families of patients with adenomatous polyps. National Polyp Study Workgroup. N Engl J Med  1996; 334: 82–7. Google Scholar 5 Burt RW, Bishop DT, Cannon LA, Dowdle MA, Lee RG, Skolnick MH. Dominant inheritance of adenomatous colonic polyps and colorectal cancer. N Engl J Med  1985; 312: 1540–4. Google Scholar 6 Cannon-Albright LA, Skolnick MH, Bishop DT, Lee RG, Burt RW. Common inheritance of susceptibility to colonic adenomatous polyps and associated colorectal cancers. N Engl J Med  1988; 319: 533–7. Google Scholar 7 Taketo MM. Cyclooxygenase-2 inhibitors in tumorigenesis (part I). J Natl Cancer Inst  1998; 90: 1529–36. Google Scholar 8 Taketo MM. Cyclooxygenase-2 inhibitors in tumorigenesis (part II). J Natl Cancer Inst  1998; 90: 1609–20. Google Scholar 9 Kargman SL, O'Neill GP, Vickers PJ, Evans JF, Mancini JA, Jothy S. Expression of prostaglandin G/H synthase-1 and -2 protein in human colon cancer. Cancer Res  1995; 55: 2556–9. Google Scholar 10 Eberhart CE, Coffey RJ, Radhika A, Giardiello FM, Ferrenbach S, DuBois RN. Up-regulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology  1994; 107: 1183–8. Google Scholar 11 Oshima M, Dinchuk JE, Kargman SL, Oshima H, Hancock B, Kwong E, et al. Suppression of intestinal polyposis in Apc delta716 knockout mice by inhibition of cyclooxygenase 2 (COX-2). Cell  1996; 87: 803–9. Google Scholar 12 Reddy BS, Tokumo K, Kulkarni M, Aligia C, Kelloff G. Inhibition of colon carcinogenesis by prostaglandin synthesis inhibitors and related compounds. Carcinogenesis  1992; 13: 1019–23. Google Scholar 13 Thun MJ, Nambandiri MM, Health CW. Aspirin use and reduced risk of fatal colon cancer. N Engl J Med  1991; 325: 1593–6. Google Scholar 14 Giardiello FM, Hamilton SR, Krush AJ, Piantadosi S, Hylind LM, Celano P, et al. Treatment of colonic and rectal adenomas with sulindac in familial adenomatous polyposis. N Engl J Med  1993; 328: 1313–6. Google Scholar 15 Lander ES, Schork NJ. Genetic dissection of complex traits. Science  1994; 265: 2037–48. Google Scholar 16 Blackwelder WC, Elston RC. A comparison of sib-pair linkage tests for disease susceptibility loci. Genet Epidemiol  1985; 2: 85–97. Google Scholar 17 Laken SJ, Petersen GM, Gruber SB, Oddoux C, Ostrer H, Giardiello FM, et al. Familial colorectal cancer in Ashkenazim due to a hypermutable tract in APC. Nat Genet  1997; 17: 79–83. Google Scholar 18 Myeroff LL, Parsons R, Kim SJ, Hedrick L, Cho KR, Orth K, et al. A transforming growth factor beta receptor type II gene mutation common in colon and gastric but rare in endometrial cancers with microsatellite instability. Cancer Res  1995; 55: 5545–7. Google Scholar 19 Parsons R, Myeroff LL, Liu B, Willson JK, Markowitz SD, Kinzler KW, et al. Microsatellite instability and mutations of the transforming growth factor beta type II receptor gene in colorectal cancer. Cancer Res  1995; 55: 5548–50. Google Scholar 20 Rodriguez-Bigas MA, Boland CR, Hamilton SR, Henson DE, Jass JR, Kahn PM, et al. A National Cancer Institute Workshop on Hereditary Nonpolyposis Colorectal Cancer Syndrome: meeting highlights and Bethesda guidelines. J Natl Cancer Inst  1997; 89: 1758–62. Google Scholar 21 http://locus.umdnj.edu/cor/. Google Scholar 22 Idury RM, Elston RC. A faster and more general hidden Markov model algorithm for multipoint likelihood calculations. Hum Hered  1997; 47: 197–202. Google Scholar 23 Haseman JK, Elston RC. The investigation of linkage between a quantitative trait and a marker locus. Behav Genet  1972; 2: 3–19. Google Scholar 24 S.A.G.E. Statistical analysis for genetic epidemiology. Release 4.0. Computer program package available from the Department of Epidemiology and Biostatistics, Case Western Reserve University. Cleveland (OH): Case Western Reserve University; 2000. Google Scholar 25 Hodge SE. The information contained in multiple sibling pairs. Genet Epidemiol  1984; 1: 109–22. Google Scholar 26 http://compgen.rutgers.edu/chr1. Google Scholar 27 Spielman RS, Ewens WJ. The TDT and other family-based tests for linkage disequilibrium and association. Am J Hum Genet  1996; 59: 983–9. Google Scholar 28 Spielman RS, Ewens WJ. A sibship test for linkage in the presence of association: the sib transmission/disequilibrium test. Am J Hum Genet  1998; 62: 450–8. Google Scholar 29 McGinnis RE, Ewens WJ, Spielman RS. The TDT reveals linkage and linkage disequilibrium in a rare disease. Genet Epidemiol  1995; 12: 637–40. Google Scholar 30 Spielman RS. TDT/S-TDT; version 1.1. http://spielman07.med.upenn.edu/TDT.htm.; 1999. Google Scholar 31 Spielman RS, McGinnis RE, Ewens WJ. Transmission test for linkage disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus (IDDM). Am J Hum Genet  1993; 52: 506–16. Google Scholar 32 Cleves MA, Olson JM, Jacobs KB. Exact transmission-disequilibrium tests with multiallelic markers. Genet Epidemiol  1997; 14: 337–47. Google Scholar 33 Spirio LN, Dixon DA, Robertson J, Robertson M, Barrows J, Traer E, et al. The inducible prostaglandin biosynthetic enzyme, cyclooxygenase 2, is not mutated in patients with attenuated adenomatous polyposis coli. Cancer Res  1998; 58: 4909–12. Google Scholar © Oxford University Press
TI - Testing for Colon Neoplasia Susceptibility Variants at the Human COX2 Locus
JF - JNCI: Journal of the National Cancer Institute
DO - 10.1093/jnci/93.8.635
DA - 2001-04-18
UR - https://www.deepdyve.com/lp/oxford-university-press/testing-for-colon-neoplasia-susceptibility-variants-at-the-human-cox2-VODi0VD1Lt
SP - 635
EP - 639
VL - 93
IS - 8
DP - DeepDyve
ER -