1 INTRODUCCIÓNEl cáncer colorrectal (CCR) es la tercera causa de diagnóstico de cáncer en todo el mundo. El 11-27% de pacientes se diagnostican en estadio avanzado o metastático. De los inicialmente diagnosticados de CCR localizado, un 25-50% acabarán presentando metástasis. En los últimos años hemos asistido a un leve decremento de la incidencia y de la mortalidad por CCR, probablemente debido a la mejora en los programas de ¨screening¨ y a las mejoras en las estrategias terapéuticas. El CCRM tiene un pronóstico pésimo. La supervivencia global se sitúa entre los 5-6 meses, en pacientes que no reciben tratamiento. En los últimos años, la implementación de las nuevas terapias ha elevado esta media hasta los 15-18 meses. En la actualidad, ¿existe alguna manera para reconocer con que tipo de CCRM estamos tratando? ¿Cómo podemos determinar de antemano como se comportará el CCRM, en términos de pronóstico y respuesta a los tratamientos? Hasta la fecha se han realizado multitud de estudios con la intención de determinar factores pronósticos y predictivos de respuesta en CCRM. En el texto se recogen detalladamente los más importantes. Sin embargo, existen diferentes factores que explican que estos factores no se hayan implementado en la práctica clínica habitual. Entre ellos está la falta de estudios clínicos prospectivos confirmativos y, por otra parte está la heterogeneidad que impera a la hora de realizar los estudios sobre marcadores. Para solventar este último punto, el Nacional Cancer Institute ha creado unas guías a seguir (REMARK). Actualmente hace falta diseñar convenientemente estudios prospectivos confirmativos. En el Hospital Clínic de Barcelona hemos desarrollado (2001) un algoritmo terapéutico basado en el patrón de metástasis al debut de la enfermedad y los algunos de los factores más utilizados como pronósticos. 2 HIPÓTESIS Y OBJETIVOS: En nuestra hipótesis, hemos seleccionado dos determinantes moleculares (MMP-7 y sFAS/sFASL) como potenciales marcadores pro'osticos y predictivos de respuesta a terapia en CCRM. Matrilisina o metaloproteasa-7 (MMP-7) es una enzima proteolítica perteneciente a la familia de las metaloproteasas. Es sintetizada y secretada al medio por células tumorales. Detalles sobre la regulación de su expresión y función pueden hallarse en el texto original. Entre sus actividades destacamos su capacidad de cortar la proteína de membrana FASL. El receptor FAS y su ligando FASL son receptores transmebrana. Su interacción induce la activación de la vía extrínseca de apoptosis. Existen fracciones solubles de estas proteínas: sFAS y sFASL. sFAS es resultante de fenómenos de "splicing" alternativo. Su función es predominantemente proapoptótica. sFASL resulta de cortar la proteína de membrana, a cargo de proteasas como MMP-7. La función de sFASL es básicamente antiapoptótica. Las fracciones solubles pueden anularse entre sí. Su función final depende de los balances entre ellas, por lo que el ratio pude ser un estimador de la función apoptótica final. FAS y FASL son proteínas que se han relacionado con la respuesta apoptótica al tratamiento quimioterápico. El CCRM muestra un patrón de resistencia a la inducción de apoptosis vía FAS-FASL. El papel de las formas solubles sFAS y sFASL en quimioresistencia nunca ha sido evaluado. MMP-7 se ha relacionado con un fenotipo agresivo de CCRM, aunque también con quimioresistencia, debido a su capacidad de corte de FASL. Los niveles séricos de sFAS han demostrado estar aumentados en pacientes con CCRM. La medición de los niveles séricos de sFASL y MMP-7 nunca se ha llevado a cabo en pacientes con CCRM.Nuestra hipótesis es que los niveles séricos de MMP-7, sFAS y sFASL en pacientes con CCRM puden ser marcadores biológicos que estimen la agresividad y la quimioresistencia. A modo de concepto nuevo y, teniendo en cuenta que el tumor varía en el tiempo, está el hecho que los marcadores biológicos deberían poderse obtener no solo en el momento del diagnóstico sino en cualquier momento durante la evolución de la enfermedad. Si esto fuera así, el método de obtención debería ser fácil y no invasivo. Los niveles séricos de MMP-7, sFAS y sFASL y sus variaciones durante el tiempo deberían ser predictores de quimioresistencia en CCRM en cualquier momento durante la enfermedad.Los objetivos se pueden resumir en:A) Determinar si los niveles séricos basales de MMP-7 en pacientes con CCRM y establecer su valor pronóstico. B) Determinar los niveles séricos de sFAS y sFASL en pacientes con CCRM, antes y durante el tratamiento quimioterápico, y establecer una correlación con la respuesta tumoral, de manera que podamos valorar su papel como marcador predictivo de respuesta. C) En función de los resultados obtenidos, diseñar estudios clínicos prospectivos para validar el valor de los niveles séricos de MMP-7 y sFAS/sFASL como nuevos marcadores solubles de pronóstico y respuesta al tratamiento en CCRM. 3. RESULTADOS: Los resultados han sido hechos públicos en dos artículos, uno de ellos aparecido en 2005 y otro de ellos en prensa. Ambos se recogen en la tesis. 4. CONCLUSIONES: En resumen, concluimos que:-MMP-7 puede medirse en el suero y sus niveles basales son un factor pronóstico independiente en pacientes con CCRM.-Las variaciones de los niveles séricos de sFAS/FASL en pacientes con CCRM en tratamiento quimioterápico correlacionan con la respuesta tumoral.-La detección de un descenso del los valores séricos del ratio sFAS/sFASL, habitualmente debido al incremento de sFASL, se relaciona con quimioresistencia.-Los valores séricos del ratio sFAS/sFASL podrían utilizarse como un predictor dinámico de respuesta al tratamiento quimioterápico en pacientes con CCRM y su valor debería validarse en estudios clínicos prospectivos. -Nuestras observaciones, en los campos básico y clínico, indican que MMP-7 estaría implicada en quimioresistencia primaria o intrínseca y adquirida, en CCRM.-Una hipótesis generada "de novo" es que un patrón sérico consistente en niveles altos de MMP-7 y de FASL, tanto si se detecta antes de iniciar el tratamiento como durante éste, implicaría qumioresistencia y, en consecuencia, mal pronóstico. -MMP-7 y sFAS/sFASL, los nuevos marcadores solubles que proponemos, se pueden detector fácilmente a través de una técnica no invasiva.-MMP-7 y sFAS/sFASL, los nuevos marcadores solubles propuestos, se pueden detectar en cualquier momento a lo largo de la enfermedad y refleja la biología tumoral cambiante, de una manera dinámica. -De acuerdo a los resultados presentados y las hipótesis generadas a partir de éstos, hemos diseñado estudios clínicos prospectivos para determinar la relevancia clínica de los niveles séricos de MMP-7 y sFAS/sFASL, como nuevos marcadores solubles de quimioresistencia en CCRM.

The primary aim of this thesis was to conduct a longitudinal study, to investigate the quality of life (QOL) in a large sample of people diagnosed with colorectal cancer (CRC), immediately post-diagnosis and one-year later. Various measures were utilised to capture the multifaceted concept of QOL, including psychological distress, satisfaction with life, positive affect and cancer-related quality of life. The investigation was conducted within Lazarus and Folkman’s stress and coping framework. Specifically, this research aimed to test the utility of the stress and coping framework in the context of CRC, and to explore the role of benefit-finding within the context of the stress and coping model components. One thousand, eight hundred participants (1078 men and 722 women) with a CRC diagnosis duration of 1 to 12 months, completed a telephone interview and written questionnaire, assessing demographics, disease/treatment characteristics, threat appraisal, social support, optimism, coping, benefit-finding and quality of life domains, at approximately five- and 12-months post-diagnosis. To identify CRC-specific coping strategies, the Coping with Colorectal Cancer measure was developed in the initial study. The eight factor structure was confirmed, and the subscales (Positive Perceptual Change; Religion/Spirituality; Rumination; Acceptance; Humour; Palliative; Seeking Social Support; and Lifestyle Reorganisation) found to have reliability and preliminary criterion-related validity within the context of the stress and coping framework. As predicted, regression analyses showed that, after controlling for demographics, disease/treatment characteristics and stress/coping variables, the coping subscales uniquely predicted Time 1 QOL outcomes, with Seeking Social Support continuing to predict Time 2 Positive Affect. In the second study, the benefit-finding domains specific to those with CRC were identified, and relationships with quality of life outcomes assessed, to determine the inclusion of benefit-finding in the path model. Based on completed data from 1757 of the initial 1800 participants, confirmatory factor analyses revealed three domains of benefit-finding: Personal Growth; Interpersonal Growth and Acceptance. As hypothesised, regressions analyses found that benefit-finding domains at Time 1 was associated with Time 1 QOL outcomes, specifically, Positive Affect and Cancer-related Quality of Life (both the aggregate score and its Social/Family, Functional and Colorectal Cancer-specific Well-being subscales). Time 1 Personal Growth also predicted Time 1 Psychological Distress. After controlling for Time 1 Positive Affect, Personal Growth continued to predict Time 2 Positive Affect. The final study drew on the results of the first two studies, which informed the coping strategies and benefit-finding domains to be included in regression analyses initially, and then, structural equation modelling. The final study included 1276 complete data sets of the initial 1800 participants. Parameters of the initial hypothesised model of the stress and coping framework, including relationships with benefit-finding (based on empirical findings) failed to fit the model to the data. After several revisions, the analysis revealed that the final model fit the data, where stress, coping and benefit-finding accounted for 63% of the variance in Time 1 QOL. The model showed that threat appraisal, coping resources, avoidant coping and benefit-finding directly impacted on Time 1 QOL, while threat appraisal, social support and approach coping directly impacted on benefit-finding. In this study, the approach coping strategies included in the path model could also be conceptualised as meaning-based coping strategies, as they appeared to facilitate a meaning-making process. However, benefit-finding, which some researchers have suggested is also a meaning-based coping strategy, had differential relationships with stress, coping and outcome variables, compared with the approach coping strategies. These results indicate that benefit-finding is an empirically distinct construct in the context of CRC. Finally, in this study, the impact of stress, coping and benefit-finding on Time 2 QOL, was indirect, being mediated by Time 1 QOL outcomes. The accumulated findings of these three studies have extended the cancer coping and benefit-finding research by revealing new relations between stress, coping and benefit-finding and QOL in a mixed-gender, older population with CRC. There are implications for measurement of, and theory building around benefit-finding. Finally, these studies inform the development of clinical interventions to enhance the quality of life in the short- and longer-term for individuals diagnosed with CRC.

The primary aim of this thesis was to conduct a longitudinal study, to investigate the quality of life (QOL) in a large sample of people diagnosed with colorectal cancer (CRC), immediately post-diagnosis and one-year later. Various measures were utilised to capture the multifaceted concept of QOL, including psychological distress, satisfaction with life, positive affect and cancer-related quality of life. The investigation was conducted within Lazarus and Folkman’s stress and coping framework. Specifically, this research aimed to test the utility of the stress and coping framework in the context of CRC, and to explore the role of benefit-finding within the context of the stress and coping model components. One thousand, eight hundred participants (1078 men and 722 women) with a CRC diagnosis duration of 1 to 12 months, completed a telephone interview and written questionnaire, assessing demographics, disease/treatment characteristics, threat appraisal, social support, optimism, coping, benefit-finding and quality of life domains, at approximately five- and 12-months post-diagnosis. To identify CRC-specific coping strategies, the Coping with Colorectal Cancer measure was developed in the initial study. The eight factor structure was confirmed, and the subscales (Positive Perceptual Change; Religion/Spirituality; Rumination; Acceptance; Humour; Palliative; Seeking Social Support; and Lifestyle Reorganisation) found to have reliability and preliminary criterion-related validity within the context of the stress and coping framework. As predicted, regression analyses showed that, after controlling for demographics, disease/treatment characteristics and stress/coping variables, the coping subscales uniquely predicted Time 1 QOL outcomes, with Seeking Social Support continuing to predict Time 2 Positive Affect. In the second study, the benefit-finding domains specific to those with CRC were identified, and relationships with quality of life outcomes assessed, to determine the inclusion of benefit-finding in the path model. Based on completed data from 1757 of the initial 1800 participants, confirmatory factor analyses revealed three domains of benefit-finding: Personal Growth; Interpersonal Growth and Acceptance. As hypothesised, regressions analyses found that benefit-finding domains at Time 1 was associated with Time 1 QOL outcomes, specifically, Positive Affect and Cancer-related Quality of Life (both the aggregate score and its Social/Family, Functional and Colorectal Cancer-specific Well-being subscales). Time 1 Personal Growth also predicted Time 1 Psychological Distress. After controlling for Time 1 Positive Affect, Personal Growth continued to predict Time 2 Positive Affect. The final study drew on the results of the first two studies, which informed the coping strategies and benefit-finding domains to be included in regression analyses initially, and then, structural equation modelling. The final study included 1276 complete data sets of the initial 1800 participants. Parameters of the initial hypothesised model of the stress and coping framework, including relationships with benefit-finding (based on empirical findings) failed to fit the model to the data. After several revisions, the analysis revealed that the final model fit the data, where stress, coping and benefit-finding accounted for 63% of the variance in Time 1 QOL. The model showed that threat appraisal, coping resources, avoidant coping and benefit-finding directly impacted on Time 1 QOL, while threat appraisal, social support and approach coping directly impacted on benefit-finding. In this study, the approach coping strategies included in the path model could also be conceptualised as meaning-based coping strategies, as they appeared to facilitate a meaning-making process. However, benefit-finding, which some researchers have suggested is also a meaning-based coping strategy, had differential relationships with stress, coping and outcome variables, compared with the approach coping strategies. These results indicate that benefit-finding is an empirically distinct construct in the context of CRC. Finally, in this study, the impact of stress, coping and benefit-finding on Time 2 QOL, was indirect, being mediated by Time 1 QOL outcomes. The accumulated findings of these three studies have extended the cancer coping and benefit-finding research by revealing new relations between stress, coping and benefit-finding and QOL in a mixed-gender, older population with CRC. There are implications for measurement of, and theory building around benefit-finding. Finally, these studies inform the development of clinical interventions to enhance the quality of life in the short- and longer-term for individuals diagnosed with CRC.

The aim of this research was to assemble sufficient experimental evidence about candidate gene transcript expression changes between non-neoplastic and neo- plastic colorectal tissues to justify future assay development involving promis- ing leads. To achieve this aim, this thesis explores the hypothesis that gene expression-based biomarkers can be used to accurately discriminate colorectal neoplastic tissues from non-neoplastic controls. This hypothesis was tested by first analysing multiple, large, quality controlled data sets comprising gene expression measurements across colorectal phenotypes to discover potential biomarkers. Candidate biomarkers were then subjected to validation testing using a custom-design oligonucleotide microarray applied to independently derived clinical specimens. A number of novel conclusions are reached based on these data. The most important conclusion is that a defined subset of genes expressed in the colorectal mucosa are reliably differentially ex- pressed in neoplastic tissues. In particular, the apparently high prediction accu- racy achieved for single gene transcripts to discriminate hundreds of neoplastic and non-neoplastic tissues provides compelling evidence that the resulting can- didate genes are worthy of further biomarker research. In addition to addressing the central hypothesis, additional contributions are made to the field of colorectal neoplasia gene expression profiling. These contributions include: The first systematic analysis of gene expression in non-diseased tissues along the colorectum To better understand the range of gene expression in non-diseased tissues, RNA extracts taken from along the longitudinal axis of the large intestine were studied. The development of quality control methodologies for high dimen- sional gene expression data Complex data collection platforms such as oligonucleotide microarrays introduce the potential for unrecognized confound- ing variables. The exploration of quality control parameters across five hundred microarray experiments provided insights about quality control techniques. The design of a custom microrray comprised of oligonucleotide probe- sets hybridising to RNA transcripts differentially expressed in neo- plastic colorectal specimens A custom design oligonucleotide microarray was designed and tested combining the results of multiple biomarker discovery projects. Introduction of a method to filter differentially expressed genes dur- ing discovery that may improve validation efficiencies of biomarker discovery based on gene expression measurements Differential expression discovery research is typically focused only on quantitative changes in transcript concentration between phenotype contrasts. This work introduces a method for generating hypotheses related to transcripts which may be quali- tatively “switched-on” between phenotypes. Identification of mRNA transcripts which are differentially expressed between colorectal adenomas and colorectal cancer tissues Transcripts differentially expressed between adenomatous and cancerous RNA extracts were discovered and then tested in independent tissues. In conclusion, these results confirm the hypothesis that gene expression profiling can discriminate colorectal neoplasia (including adenomas) from non-neoplastic controls. These results also establish a foundation for an ongoing biomarker development program.

Together, colorectal (CRC) and bladder cancer (BlCa) are responsible for a large percentage of cancer related morbidity and mortality in Western society. A dramatic reduction in patient survival occurs as these cancers progress towards invasive and metastatic disease, from a five year survival rate of about 90% for localised disease to approximately 5-10% for advanced disease involving distant metastasis. A greater understanding of disease progression will lead to enhanced screening, diagnostic and treatment strategies, in turn providing an improved prognosis for the patient. The purpose of this study was to expand the current molecular knowledge of CRC and BlCa by elucidating the role of Mxi1 mutations and MTSS1 expression in CRC and BlCa respectively, and to examine the diagnostic potential of these genes. The Mxi1 coding region for 41 tumours, collected by the South Western Sydney Colorectal Cancer Tumour Bank from 2000-2001, was screened for mutations using Dideoxy Fingerprinting (ddF) and sequencing. Sequence alterations were detected in 34% of tumours. Three different polymorphisms and three mutations were detected. One mutation could possibly affect the tumour suppressor function of Mxi1. The presence of a gene mutation did not correlate to any clinical characteristics and is therefore not a suitable diagnostic marker. Microsatellite instability (MSI) status however, significantly correlated with tumour grade. Expression levels of MTSS1 and an associated gene, MTSS2, were examined in 16 BlCa cell lines, 9 clonally-derived BlCa sublines, and 30 transitional cell carcinomas (TCCs) collected by the Heinrich-Heine University from 1993-2000. Variable gene expression was observed in BlCa cell lines and tumour samples. No significant correlation of MTSS expression and invasive ability was observed for the cell lines or tumour samples. Further studies eliminated promoter methylation and p53 functional status as mechanisms involved in MTSS1 and MTSS2 down-regulation. Functional studies performed on stable MTSS1-expressing BlCa lines found that although migration was increased, cells displayed reduced anchorage-independent growth. The invasive ability of these cells was unchanged confirming that expression does not correlate with invasive ability. These data support the role of MTSS1 as a tumour suppressor and not as a metastasis suppressor gene. Although MTSS1 may not be useful in predicting more invasive disease, its role as a tumour suppressor in cancer may be useful.

Together, colorectal (CRC) and bladder cancer (BlCa) are responsible for a large percentage of cancer related morbidity and mortality in Western society. A dramatic reduction in patient survival occurs as these cancers progress towards invasive and metastatic disease, from a five year survival rate of about 90% for localised disease to approximately 5-10% for advanced disease involving distant metastasis. A greater understanding of disease progression will lead to enhanced screening, diagnostic and treatment strategies, in turn providing an improved prognosis for the patient. The purpose of this study was to expand the current molecular knowledge of CRC and BlCa by elucidating the role of Mxi1 mutations and MTSS1 expression in CRC and BlCa respectively, and to examine the diagnostic potential of these genes. The Mxi1 coding region for 41 tumours, collected by the South Western Sydney Colorectal Cancer Tumour Bank from 2000-2001, was screened for mutations using Dideoxy Fingerprinting (ddF) and sequencing. Sequence alterations were detected in 34% of tumours. Three different polymorphisms and three mutations were detected. One mutation could possibly affect the tumour suppressor function of Mxi1. The presence of a gene mutation did not correlate to any clinical characteristics and is therefore not a suitable diagnostic marker. Microsatellite instability (MSI) status however, significantly correlated with tumour grade. Expression levels of MTSS1 and an associated gene, MTSS2, were examined in 16 BlCa cell lines, 9 clonally-derived BlCa sublines, and 30 transitional cell carcinomas (TCCs) collected by the Heinrich-Heine University from 1993-2000. Variable gene expression was observed in BlCa cell lines and tumour samples. No significant correlation of MTSS expression and invasive ability was observed for the cell lines or tumour samples. Further studies eliminated promoter methylation and p53 functional status as mechanisms involved in MTSS1 and MTSS2 down-regulation. Functional studies performed on stable MTSS1-expressing BlCa lines found that although migration was increased, cells displayed reduced anchorage-independent growth. The invasive ability of these cells was unchanged confirming that expression does not correlate with invasive ability. These data support the role of MTSS1 as a tumour suppressor and not as a metastasis suppressor gene. Although MTSS1 may not be useful in predicting more invasive disease, its role as a tumour suppressor in cancer may be useful.

Together, colorectal (CRC) and bladder cancer (BlCa) are responsible for a large percentage of cancer related morbidity and mortality in Western society. A dramatic reduction in patient survival occurs as these cancers progress towards invasive and metastatic disease, from a five year survival rate of about 90% for localised disease to approximately 5-10% for advanced disease involving distant metastasis. A greater understanding of disease progression will lead to enhanced screening, diagnostic and treatment strategies, in turn providing an improved prognosis for the patient. The purpose of this study was to expand the current molecular knowledge of CRC and BlCa by elucidating the role of Mxi1 mutations and MTSS1 expression in CRC and BlCa respectively, and to examine the diagnostic potential of these genes. The Mxi1 coding region for 41 tumours, collected by the South Western Sydney Colorectal Cancer Tumour Bank from 2000-2001, was screened for mutations using Dideoxy Fingerprinting (ddF) and sequencing. Sequence alterations were detected in 34% of tumours. Three different polymorphisms and three mutations were detected. One mutation could possibly affect the tumour suppressor function of Mxi1. The presence of a gene mutation did not correlate to any clinical characteristics and is therefore not a suitable diagnostic marker. Microsatellite instability (MSI) status however, significantly correlated with tumour grade. Expression levels of MTSS1 and an associated gene, MTSS2, were examined in 16 BlCa cell lines, 9 clonally-derived BlCa sublines, and 30 transitional cell carcinomas (TCCs) collected by the Heinrich-Heine University from 1993-2000. Variable gene expression was observed in BlCa cell lines and tumour samples. No significant correlation of MTSS expression and invasive ability was observed for the cell lines or tumour samples. Further studies eliminated promoter methylation and p53 functional status as mechanisms involved in MTSS1 and MTSS2 down-regulation. Functional studies performed on stable MTSS1-expressing BlCa lines found that although migration was increased, cells displayed reduced anchorage-independent growth. The invasive ability of these cells was unchanged confirming that expression does not correlate with invasive ability. These data support the role of MTSS1 as a tumour suppressor and not as a metastasis suppressor gene. Although MTSS1 may not be useful in predicting more invasive disease, its role as a tumour suppressor in cancer may be useful.

There are two chief molecular pathways to sporadic colorectal cancer (CRC), the chromosomal instability (CIN) and the CpG island methylator phenotype (CIMP) pathways. A third pathway, the pure microsatellite instability pathway, is important in inherited CRC specifically hereditary non-polyposis colorectal cancer. The CIN pathway is characterized by an adenomatous pathological precursor, aneuploidy and microsatellite stability. CIMP pathway cancers, however, are frequently proximal, develop from serrated rather than adenomatous polyps and are strongly associated with BRAF mutation. The CIMP pathway is driven primarily by epigenetic rather than genetic instability. These pathway-specific molecular traits are evident within the pathological precursors to these cancers and thus pathway divergence must occur at the beginning of carcinogenesis or even before. Although DNA methylation is recognized as a key mechanism in colorectal carcinogenesis, relatively little is known about its pattern, regulation and relevance in normal colorectal mucosa. This PhD thesis characterized the profile of DNA methylation in the normal human colorectum and explored its associations with luminal, environmental, dietary and pathological factors. The genes methylated in CRC are characterized as “type A” (Age-related) genes and “type C” (Cancer-specific) genes. Generally, “type A” genes are methylated in both normal and neoplastic tissue with the degree of methylation proportional to the age of the tissue. The methylation of “type C” genes, however, is more specific for neoplastic tissue. The primary study recruited 166 patients undergoing colonoscopy. At colonoscopy, mucosal biopsies were taken from the caecum, transverse colon, sigmoid colon and rectum. DNA methylation was analysed by MethyLight at “type A” (ESR1, GATA5, HIC1, HPP1, SFRP1) and “type C” methylation markers (MGMT, MLH1, CDKN2A, MINT2, MINT31, IGF2, CACNA1G, NEUROG1, SOCS1, RUNX3). LINE-1 methylation was quantified by pyrosequencing. The last 5 “type C” markers comprise a CIMP panel used to identify CIMP cancers. Mean “type A” and CIMP panel methylation Z-scores were calculated. The PMR for each of these CpG island loci was compared to patient age, gender, previous colorectal polyps, smoking history and the presence of concomitant pathology. Most “type A” genes demonstrated strong and direct correlations between methylation and patient age (e.g. ESR1, ρ=0.66, p<0.0001) and had greater methylation within the distal compared to the proximal colorectum (e.g. ESR1, p<0.0001). On multivariate analysis, the mucosal “type A” methylation Z-score had a strong, independent, inverse association with the diagnosis of colorectal adenomas (OR=0.23, p<0.001), the precursor to CIN cancers. The mean CIMP methylation Z-score in normal mucosa, however, was significantly and independently associated with advanced proximal serrated polyps (OR=5.1, p=0.009), the precursor to CIMP cancers. The luminal and epithelial associations with colorectal methylation were explored by a randomized, double-blind, placebo-controlled trial. This experiment was undertaken to determine whether dietary supplementation could modulate epithelial DNA methylation. In addition, the study was designed to evaluate intra-individual reproducibility of the MethyLight technique. The study consisted of a 4 week cross-over trial of resistant starch and Bifidobacterium lactis either alone or as a combined synbiotic preparation, in 20 human volunteers. Rectal biopsies, faeces and serum were collected. Rectal mucosal endpoints included DNA methylation at the CpG island loci and LINE-1, epithelial proliferation (Ki67 immunohistochemistry) and crypt cellularity. Faecal short-chain fatty acid concentrations, pH, ammonia and microbiological profiles (by DGGE and sequencing) were examined. The synbiotic intervention fostered a significantly different faecal stream bacterial community than either the prebiotic or the probiotic interventions alone, but did not show any significant associations with the epithelial or luminal parameters. To explore possible associations between luminal and epithelial parameters and mucosal DNA methylation, the baseline indices were further analysed. There was a strong positive correlation between baseline epithelial proliferation and “type A” marker methylation (ρ = 0.7, p = 0.0001). Thus, “type A” methylation may reflect the cellular age or mitotic burden of a tissue, which is a function of both time and cell turnover. There were consistent inverse trends evident between faecal short-chain fatty acid levels and rectal mucosal DNA methylation. This PhD project found that DNA methylation within the normal colorectal mucosa varied with patient age and region and was strongly associated with the development of pathway-specific pathology, suggesting that the background colorectal field may predict both the at-risk patients and at-risk pathways. Diet and the luminal environment more broadly may influence levels of DNA methylation in the colorectal mucosa and could help to explain regional patterns of colorectal DNA methylation.

No description available.

Diss. (sammanfattning) Stockholm : Karol. Inst., 2002. / Härtill 4 uppsatser.