Search

Introduction The effectiveness of colorectal cancer (CRC) screening programs is directly related to participation and the number of interval CRCs. The objective was to analyse specific-mortality in a cohort of individuals invited to a CRC screening program according to type of CRC diagnosis (screen-detected cancers, interval cancers, and cancers among the non-uptake group). Material and methods Retrospective cohort that included invitees aged 50–69 years of a CRC screening program (target population of 85,000 people) in Catalonia (Spain) from 2000–2015 with mortality follow-up until 2020. A screen-detected CRC was a cancer diagnosed after a positive faecal occult blood test (guaiac or immunochemical); an interval cancer was a cancer diagnosed after a negative test result and before the next invitation to the program (≤24 months); a non-uptake cancer was a cancer in subjects who declined screening. Results A total of 624 people were diagnosed with CRC (n = 265 screen-detected, n = 103 interval cancers, n = 256 non-uptake). In the multivariate analysis, we observed a 74% increase in mortality rate in the group with interval CRC compared to screen-detected CRC adjusted for age, sex, location and stage (HR: 1.74%, 95% CI:1.08–2.82, P = 0.02). These differences were found even when we restricted for advanced-cancers participants. In the stratified analysis for type of faecal occult blood test, a lower mortality rate was only observed among FIT screen-detected CRCs. Conclusion CRC screening with the FIT was associated with a significant reduction in CRC mortality.

Background: Bilirubin, a byproduct of hemoglobin breakdown and purported anti-oxidant, is thought to be cancer preventive. We conducted complementary serological and Mendelian randomization (MR) analyses to investigate whether alterations in circulating levels of bilirubin are associated with risk of colorectal cancer (CRC). We decided a priori to perform analyses separately in men and women based on suggestive evidence that associations may differ by sex. Methods: In a case-control study nested in the European Prospective Investigation into Cancer and Nutrition (EPIC), pre-diagnostic unconjugated bilirubin (UCB, the main component of total bilirubin) concentrations were measured by high-performance liquid chromatography in plasma samples of 1386 CRC cases and their individually matched controls. Additionally, 115 single-nucleotide polymorphisms (SNPs) robustly associated (P < 5 × 10−8) with circulating total bilirubin were instrumented in a 2-sample MR to test for a potential causal effect of bilirubin on CRC risk in 52,775 CRC cases and 45,940 matched controls in the Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO), the Colon Cancer Family Registry (CCFR), and the Colorectal Transdisciplinary (CORECT) study. Results: The associations between circulating UCB levels and CRC risk differed by sex (Pheterogeneity = 0.008). Among men, higher levels of UCB were positively associated with CRC risk (odds ratio [OR] = 1.19, 95% confidence interval [CI] = 1.04–1.36; per 1-SD increment of log-UCB). In women, an inverse association was observed (OR = 0.86 (0.76–0.97)). In the MR analysis of the main UGT1A1 SNP (rs6431625), genetically predicted higher levels of total bilirubin were associated with a 7% increase in CRC risk in men (OR = 1.07 (1.02–1.12); P = 0.006; per 1-SD increment of total bilirubin), while there was no association in women (OR = 1.01 (0.96–1.06); P = 0.73). Raised bilirubin levels, predicted by instrumental variables excluding rs6431625, were suggestive of an inverse association with CRC in men, but not in women. These differences by sex did not reach formal statistical significance (Pheterogeneity ≥ 0.2). Conclusions: Additional insight into the relationship between circulating bilirubin and CRC is needed in order to conclude on a potential causal role of bilirubin in CRC development.

OBJECTIVES: The REAC-TAVI (Assessment of platelet REACtivity after Transcatheter Aortic Valve Implantation) trial enrolled patients with aortic stenosis (AS) undergoing transcatheter aortic valve replacement (TAVR) pre-treated with aspirin + clopidogrel, aimed to compare the efficacy of clopidogrel and ticagrelor in suppressing high platelet reactivity (HPR) after TAVI. BACKGROUND: Current recommendations support short-term use of aspirin + clopidogrel for patients with severe AS undergoing TAVR despite the lack of compelling evidence. METHODS: This was a prospective, randomized, multicenter investigation. Platelet reactivity was measured at 6 different time points with the VerifyNow assay (Accriva Diagnostics, San Diego, California). HPR was defined as (P2Y12 reaction units (PRU) ≥208. Patients with HPR before TAVR were randomized to either aspirin + ticagrelor or aspirin + clopidogrel for 3 months. Patients without HPR continued with aspirin + clopidogrel (registry cohort). The primary endpoint was non-HPR status (PRU

Modern life involves mistimed sleeping and eating patterns that in experimental studies are associated with adverse health effects. We assessed whether timing of meals is associated with breast and prostate cancer risk taking into account lifestyle and chronotype, a characteristic correlating with preference for morning or evening activity. We conducted a population-based case-control study in Spain, 2008-2013. In this analysis we included 621 cases of prostate and 1,205 of breast cancer and 872 male and 1,321 female population controls who had never worked night shift. Subjects were interviewed on timing of meals, sleep and chronotype and completed a Food Frequency Questionaire. Adherence to the World Cancer Research Fund/American Institute of Cancer Research recommendations for cancer prevention was examined. Compared with subjects sleeping immediately after supper, those sleeping two or more hours after supper had a 20% reduction in cancer risk for breast and prostate cancer combined (adjusted Odds Ratio [OR] = 0.80, 95%CI 0.67-0.96) and in each cancer individually (prostate cancer OR = 0.74, 0.55-0.99; breast cancer OR = 0.84, 0.67-1.06). A similar protection was observed in subjects having supper before 9 pm compared with supper after 10 pm. The effect of longer supper-sleep interval was more pronounced among subjects adhering to cancer prevention recommendations (OR both cancers = 0.65, 0.44-0.97) and in morning types (OR both cancers = 0.66, 0.49-0.90). Adherence to diurnal eating patterns and specifically a long interval between last meal and sleep are associated with a lower cancer risk, stressing the importance of evaluating timing in studies on diet and cancer. What's new? Evidence shows that long-term disruption of endogenous circadian rhythms may be associated with cancer. The effects of mistimed sleeping and eating patterns that come with modern life are however less clear. This large Spanish population-based study examined whether meal timing and sleep patterns are associated with the two most common nightshift-related cancers. Adherence to a more diurnal eating pattern, and specifically an early supper and a long interval between last meal and sleep were associated with a lower breast and prostate cancer risk, stressing the importance of evaluating circadian rhythms in diet and cancer studies and revisiting recommendations for prevention., Grant sponsor: Instituto de Salud Carlos III-FEDER; Grant number: PI11/01889; Grant sponsor: Accion Transversal del Cancer, approved on the Spanish Ministry Council on the 11th October 2007; Grant sponsor: Instituto de Salud Carlos III-FEDER; Grant numbers: PI08/1770, PI08/0533, PI08/1359, PI09/00773-Cantabria, PI09/01286-Leon, PI09/01903-Valencia, PI09/02078-Huelva, PI09/01662-Granada, PI11/01889-FEDER, PI11/02213, PI12/00488, PI12/00265, PI12/01270, PI12/00715, PI14/0613, PI15/00069, PI15/00914, PI15/01032; Grant sponsor: Regional Government of the Basque Country; Grant sponsor: Consejeria de Sanidad de la Region de Murcia; Grant sponsor: European Commission grants FOOD-CT-2006-036224-HIWATE; Grant sponsor: Spanish Association Against Cancer (AECC) Scientific Foundation; Grant sponsor: Catalan Government DURSI grant; Grant number: 2014SGR647; Grant sponsor: Fundacion Caja de Ahorros de Asturias and by the University of Oviedo

Little is known about the dynamics of human papillomavirus (HPV) infection and subsequent development of high-grade cervical intraepithelial neoplasia (CIN2/3), particularly in women >30 years of age. This information is needed to assess the impact of HPV vaccines and consider new screening strategies. A cohort of 1728 women 15-85 years old with normal cytology at baseline was followed every 6 months for an average of 9 years. Women with squamous intraepithelial lesions were referred for biopsy and treatment. The Kaplan-Meier method was used to estimate the median duration of infection and Cox regression analysis was undertaken to assess determinants of clearance and risk of CIN2/3 associated with HPV persistence. No difference in the likelihood of clearance was observed by HPV type or woman's age, with the exception of lower clearance for HPV16 infection in women under 30 years of age. Viral load was inversely associated with clearance. In conclusion, viral load is the main determinant of persistence, and persistence of HPV16 infections carry a higher risk of CIN2/3.

7 pages, 6 tables.-- PMID: 15531428 [PubMed].-- PMCID: PMC1247607.-- Printed version published Nov 2004., Organochlorine compounds have been linked to increased risk of several cancers. Despite reductions in their use and fugitive release, they remain one of the most important groups of persistent pollutants to which humans are exposed, primarily through dietary intake. We designed a case-control study to assess the risk of colorectal cancer with exposure to these chemicals, and their potential interactions with genetic alterations in the tumors. A subsample of cases (n = 132) and hospital controls (n = 76) was selected from a larger case-control study in Barcelona, Catalonia, Spain. We measured concentrations in serum of several organochlorines by gas chromatography. We assessed point mutations in K-ras and p53 genes in tissue samples by polymerase chain reaction/single-strand conformation polymorphism and assessed expression of p53 protein by immunohistochemical methods. An elevated risk of colorectal cancer was associated with higher serum concentrations of mono-ortho polychlorinated biphenyl (PCB) congeners 28 and 118. The odds ratio for these mono-ortho PCBs for middle and higher tertile were, respectively, 1.82 [95% confidence interval (CI) , 0.90-3.70] and 2.94 (95% CI, 1.39-6.20) . alpha-Hexachlorocyclohexane, hexachlorobenzene, and p,p´-DDE (4,4´-dichlorodiphenyltrichloroethene) showed nonsignificant increases in risk. Risk associated with mono-ortho PCBs was slightly higher for tumors with mutations in the p53 gene but was not modified by mutations in K-ras. Mono-ortho PCBs were further associated with transversion-type mutations in both genes. These results generate the hypothesis that exposure to mono-ortho PCBs contributes to human colorectal cancer development. The trend and magnitude of the association, as well as the observation of a molecular fingerprint in tumors, raise the possibility that this finding may be causal., The study was supported by grants from the Fundació La Marató de TV3 (48/95), the Spanish Fondo de Investigaciones Sanitarias (FIS 96/0797, FIS 00/0027, FIS 01/1264, FIS 03/0114, ISCIII, Red de Centros RCESP C03/09, and Red de Centros de Cancer C03/10) and Comisión Interministerial de Ciencia y Tecnología (SAF 00/81), and a European Union Marie Curie Fellowship for M.H. (HPMFCT 2000-00888).

Fundamento: El objetivo del estudio fue evaluar el cambio de estrategia de cribado (test inmunológico cuantitativo) en un programa poblacional de detección precoz de cáncer colorrectal (CCR) en Cataluña. Métodos: La cuarta ronda del programa de cribado de CCR en Hospitalet de Llobregat se implementó en 2008-2010. Se ofreció un test bioquímico a 50.227 individuos y uno inmunológico cuantitativo a 12.707 individuos. Se analizaron diferencias en las dos estrategias de cribado respecto a variables de aceptabilidad (entre participación, abandonos y adherencia a la colonoscopia), de precisión diagnóstica (valor predictivo positivo y tasas de detección), de resultados (tamaño y localización de lesiones, estadio de los cánceres detectados) y de recursos (número necesario de colonoscopias e intervalo de tiempo entre el resultado positivo del test y la colonoscopia). Resultados: La participación en el cribado fue superior entre los individuos que utilizaron el test inmunológico (OR: 1,35; IC95%:1,27-1,42). Las tasas de detección fueron superiores para el test inmunológico destacando la de adenomas de alto riesgo (26,7‰ vs 3,0‰). El valor predictivo positivo para adenomas de alto riesgo fue del 45,0% y del 46,9% en el inmunológico y el guayaco, respectivamente. El número de colonoscopias necesarias para detectar un cáncer fue de casi el doble que en el guayaco (13,6 vs 7,4). Conclusiones: El test inmunológico es una buena estrategia de cribado especialmente sensible para la detección de adenomas de alto riesgo. Sin embargo, requiere realizar un gran número de colonoscopias y por ello se debe disponer de los recursos y medios necesarios. Background: The aim of this study was to evaluate the screening strategy (quantitative immunological test vs biochemical test) in a population-based screening program for colorectal cancer (CRC) in Catalonia. Methods: The fourth round of a screening program for CRC with a fecal occult blood test was implemented in Hospitalet de Llobregat during 2008-2010. A biochemical test was offered to 50,227 individuals and a quantitative immunological test was offered to 12,707 individuals. We analysed differences according to the screening strategy in the following variables: acceptability of the target population (participation, dropouts, and adherence to colonoscopy), diagnostic accuracy (positive predictive value and detection rates), results (size and location of lesions, staging of CRC) and resources (number of colonoscopies needed and time interval between the positive test and colonoscopy). Results: Participation was higher among individuals who used the immunological test (OR: 1.35; CI95%:1.27-1.42). Detection rates for adenomas and cancer were also higher for the immunological test, hightlighting the detection rate for high-risk adenomas (26.7‰ vs. 3.0‰). The positive predictive value for high-risk adenomas was 45.0% and 46.9% in the immunological test and guaiac test, respectively. The number of colonoscopies needed to detect cancer with the immunological test was almost two-fold than those needed with the guaiac test (13.6 vs 7.4). Conclusions: The immunological test is a good screening strategy particularly sensitive for detecting high-risk adenomas. However, it is paramount to have enough resources to assure the quality of the CRC screening due to the large number of colonoscopies that would be required.

Background and objective: altered adhesive interaction between bone marrow (BM) stroma and progenitors in chronic myeloid leukemia (CML) may be in part caused by abnormal expression of cell adhesion molecules (CAMs) on malignant progenitor cells. Treatment of CML with interferon-a (IFN-a) re-establishes normal hemopoiesis in some patients in part by restoring normal adhesive interactions between CML progenitors and BM microenvironment, which may in turn be mediated by correcting CAM expression on progenitors. Design and methods: we investigated the expression of CAMs (L-selectin, b((2))-integrin, LFA-3, ICAM-1, ICAM-3, NCAM) on purified BM CD34(+) cells from CML patients (n= 34) and healthy adults (n= 15) by flow cytometry. Modulation of L-selectin expression on CD34(+) cells from CML after in vitro treatment with IFN-a was also investigated. RESULTS: The mean percentage of CD34(+ )cells expressing L-selectin was significantly lower in CML patients (25.4+/-12.8%) than in normal controls (68.7+/-8.3%, n=15). CD34(+)/HLA-DR(-/low) and CD34(+)/ CD38(-/low) co-expressing L-selectin were also significantly lower in untreated CML (27.4+/-21.5% and 39.8+/-26.7%, respectively, n=8) than in controls (61+/-17% and 83.7+/-10%, respectively, n=7). In vitro treatment with IFN-a of purified CD34(+) BM cells from untreated CML patients (n=8) induced a significant, dose and time-dependent increase in the L-selectin expression as indicated by FACS analysis. Interpretation and conclusions: we hypothesize that this L-selectin deficiency reflects a cell surface adhesion defect of progenitors from CML that is partially restored by in vitro IFN-a treatment. These data may help to explain the adhesive abnormalities of CML progenitors to the BM microenvironment and the in vitro restoration of adhesion capacity after IFN-a treatment.

Fundamentos: la incidencia y factores de riesgo de cáncer tienen grandes variaciones en diferentes áreas geográficas incluso dentro de un mismo país. Es de máximo interés conocer y analizar los datos obtenidos en registros de cáncer como sistema de información sanitaria. Material y métodos: a partir de los datos obtenidos por el registro de cáncer de Girona se analiza la incidencia de cáncer ginecológico en el área santitaria de Girona para el período 1985-1989. Resultados: se registraron 469 casos de cancer ginecológico en sus distintas localizaciones. Las tasas brutas y ajustadas para los tumores más frecuentes fueron, respectivamente, para el cáncer de cuerpo de útero 16,2 y 10,6; para el cáncer de ovario 11,7 y 8,0, y para el cáncer de cérvix 8,9 y 6,9. Al comparar la incidencia por comarcas se encontró una menor incidencia de cáncer de ovario en el Gironès comparada con la incidencia del área sanitaria. A pesar de que las tendencias cronológicas de las tasas de incidencia para el período 1985-1989 para cada uno de los tumores estudiados fueron positivas ninguna de éstas fue estadísticamente significativa. Tampoco se encontraron diferencias en cuanto a incidencia de estos tumores al comparar entre zona rural y urbana. Conclusiones: con los datos obtenidos, observamos que nuestra área sigue la tendencia de los países industrializados, siendo los tumores ginecológicos más frecuentes el cáncer de útero y el de ovario, seguidos de los de cérvix, otros genitales y útero mal especificado.

Inflammation plays a key role in the development of colorectal cancers. We have investigated the relationship between PTGS2 (COX2) polymorphisms and colorectal cancer risk in a hospital based case-control study. We recruited 292 patients with colorectal cancer and 274 controls from new patients admitted to Bellvitge Hospital, Barcelona, Spain, from 1996 to 1998. Subjects responded to a questionnaire on risk factors. Genotypes of the eight more frequent polymorphisms of PTGS2 were determined. Two polymorphisms are located in the promoter sequence, one in the untranslated region of exon 1, one in exon 3, one in intron 5, two in the untranslated region of exon 10, and one downstream of the last polyadenylation (poly-A) signal. Associations were analysed with logistic regression models assuming a dominant effect for rare variants to increase statistical power. An association was detected between colorectal cancer and a polymorphism in the untranslated region of exon 10 of PTGS2, with an odds ratio (OR) of 2.49, 95% confidence interval (CI) of 1.17-5.32, P=0.01. A nearby polymorphism downstream of the last poly-A signal also showed a nonsignificant increase in risk (OR 2.17, 95% CI 0.99-4.78, P=0.05). Analysis of haplotypes confirmed that individuals with these variants were at increased risk of colorectal cancer (OR compared to the most frequent haplotype: 2.17, 95% CI 0.97-4.84, P=0.06) Interactions between PTGS2 genotype and use of nonsteroidal anti-inflammatory drugs and risk of colorectal cancer were also explored.

Background: It is uncertain whether male circumcision reduces the risks of penile human papillomavirus (HPV) infection in the man and of cervical cancer in his female partner. Methods: We pooled data on 1913 couples enrolled in one of seven case-control studies of cervical carcinoma in situ and cervical cancer in five countries. Circumcision status was self-reported, and the accuracy of the data was confirmed by physical examination at three study sites. The presence or absence of penile HPV DNA was assessed by a polymerase-chain-reaction assay in 1520 men and yielded a valid result in the case of 1139 men (74.9 percent). Results: Penile HPV was detected in 166 of the 847 uncircumcised men (19.6 percent) and in 16 of the 292 circumcised men (5.5 percent). After adjustment for age at first intercourse, lifetime number of sexual partners, and other potential confounders, circumcised men were less likely than uncircumcised men to have HPV infection (odds ratio, 0.37; 95 percent confidence interval, 0.16 to 0.85). Monogamous women whose male partners had six or more sexual partners and were circumcised had a lower risk of cervical cancer than women whose partners were uncircumcised (adjusted odds ratio, 0.42; 95 percent confidence interval, 0.23 to 0.79). Results were similar in the subgroup of men in whom circumcision was confirmed by medical examination. Conclusions: Male circumcision is associated with a reduced risk of penile HPV infection and, in the case of men with a history of multiple sexual partners, a reduced risk of cervical cancer in their current female partners.

El document inclou una pàgina final amb una correcció (corrigendum). Aquesta, per si sola, té el següent DOI: 10.1038/ncomms9739 i es va publicar al mateix vol. 6. Genetic susceptibility to colorectal cancer is caused by rare pathogenic mutations and common genetic variants that contribute to familial risk. Here we report the results of a two-stage association study with 18,299 cases of colorectal cancer and 19,656 controls, with follow-up of the most statistically significant genetic loci in 4,725 cases and 9,969 controls from two Asian consortia. We describe six new susceptibility loci reaching a genome-wide threshold of P

El document inclou una pàgina final amb una correcció (corrigendum). Aquesta, per si sola, té el següent DOI: 10.1038/ncomms9739 i es va publicar al mateix vol. 6. Genetic susceptibility to colorectal cancer is caused by rare pathogenic mutations and common genetic variants that contribute to familial risk. Here we report the results of a two-stage association study with 18,299 cases of colorectal cancer and 19,656 controls, with follow-up of the most statistically significant genetic loci in 4,725 cases and 9,969 controls from two Asian consortia. We describe six new susceptibility loci reaching a genome-wide threshold of P

[cat] El càncer colorectal (CCR) representa una de les neoplàsies amb major incidència i mortalitat a escala global. Per aquest motiu els sistemes nacionals de salut europeus han posat en marxa programes de cribratge poblacionals amb l’objectiu d’augmentar la detecció precoç de lesions precursores o del càncer, i així, poder evitar l’aparició de CCR o disminuir-ne la mortalitat. Tanmateix, les modalitats dels programes de cribratge amb prova intermèdia, com la prova FIT (prova immunohistoquímica fecal), tendeixen a tenir un elevat nombre de falsos positius a causa de l’especificitat subòptima d’aquesta prova, el que es tradueix en colonoscòpies innecessàries. Per aquest motiu, és fonamental establir noves estratègies que millorin el rendiment de les actuals. La síndrome de poliposi serrada (SPS) és una entitat clínica caracteritzada per la presència de grans i/o múltiples lesions/pòlips serrats en el còlon i un alt risc a CCR. La causa de la SPS és desconeguda. No obstant una petita proporció de casos podrien ser resultat de factors genètics heretats, sobretot per la multiplicitat de les lesions, el ràpid desenvolupament d’un CCR i les edats joves de diagnòstic. Tanmateix, gran part de la predisposició germinal a la SPS està per determinar. En aquest sentit, pel CCR es coneix que variants genètiques comunes i de baixa penetrància contribueixen a la predisposició germinal a CCR, no obstant, encara no s’ha explorat el paper d’aquest tipus de variants en la predisposició germinal a la SPS. En aquest context, el primer objectiu proposat de la present tesi doctoral va ser identificar quines variants genètiques comunes i de baixa penetrància associades a la predisposició germinal a CCR estan també implicades en la predisposició germinal a SPS. Per contra, el segon objectiu, va ser avaluar si les variants genètiques comunes i de baixa penetrància associades a la predisposició germinal a CCR es poden aplicar en el programa de cribratge actual per optimitzar-lo. Per una banda, per tal d’avaluar si les variants genètiques comunes i de baixa penetrància associades a la predisposició germinal a CCR es poden aplicar en el programa de cribratge actual per optimitzar-lo, es van genotipar 65 variants genètiques germinals associades consistentment a la predisposició germinal del CCR en una cohort de 2.893 participants del programa de detecció precoç del CCR de Barcelona, amb una prova de sang oculta en femta positiva. Posteriorment, es va calcular el PRS per a cada individu. Els resultats obtinguts al comparar la mitjana dels valors del PRS entre casos i controls van assenyalar que els casos tenien valors de PRS més elevats en comparació amb els controls (OR 1,4; IC 95% 1,02-1,06; p-valor < 0,0001). Al comparar el PRS dels individus de l’últim decil amb els del primer decil, s’observava que els de l’últim decil tenien gairebé 2,5 vegades més risc de tenir un CCR o una lesió associada a risc de CCR (OR 2,22; IC 95% 1,59-3,12; p-valor, [eng] Colorectal cancer (CRC) has high incidence and associated mortality worldwide. Screening programs are recommended for men and women over 50. Intermediate screens such as fecal immunochemical testing (FIT) select patients for colonoscopy with suboptimal sensitivity. Additional biomarkers could improve the current scenario. In the other hand serrated polyposis syndrome (SPS) is a clinical entity characterised by large and/or multiple serrated polyps throughout the colon and increased risk for colorectal cancer (CRC). The basis for SPS genetic predisposition is largely unknown. Common, low-penetrance genetic variants have been consistently associated with CRC susceptibility, however, their role in SPS genetic predisposition has not been yet explored. Accordingly, this thesis has two main purposes. The first one was to evaluate the potential of polygenic risk score to improve colorectal cancer screening while the second one was to evaluate if common, low-penetrance genetic variants for CRC risk are also implicated in SPS genetic susceptibility. To assess whether common genetic variants and low penetrance associated with germinal predisposition in CCR can be applied in the current screening program to optimize it, 65 germinal genetic variants associated consistent with the germinal predisposition of the CCR were genotyped in a cohort of 2,893 participants of the Barcelona CCR early detection program with a positive FIT. Subsequently, the PRS was calculated for each individual. To achieve the first objective, all the individuals included in the study were classified as cases when a high-risk lesion for colorectal cancer was detected after colonoscopy, whereas the control group comprised individuals with low-risk or no lesions. 65 colorectal cancer risk genetic variants were genotyped. Polygenic risk score (PRS) and additive models for risk prediction incorporating sex, age, FIT value, and PRS were generated. To achieve the second objective a case-control study was performed in 219 SPS patients and 548 asymptomatic controls analysing 65 CRC susceptibility variants. A risk prediction model for SPS predisposition was developed. The results obtained by comparing the average PRS values between cases and controls noted that risk score was higher in cases compared with controls [per allele OR = 1.04; 95% confidence interval (CI), 1.02–1.06; P < 0.0001]. Moreover, a 2-fold increase in colorectal cancer risk was observed for subjects in the highest decile of risk alleles (≥65), compared with those in the first decile (≤54; OR = 2.22; 95% CI, 1.59–3.12; P < 0.0001). The results obtained by comparing the different risk models showed that the model combining sex, age, FIT value, and PRS reached the highest accuracy for identifying patients with a high-risk lesion [cross-validated area under the ROC curve (AUROC): 0.64; 95% CI, 0.62–0.66]. On the other hand, to identify which common genetic variants and low penetrance associated with germinal predisposition in CCR are also involved in germinal predisposition in SPS a case-control association study was conducted. Therefore, 65 germinal genetic variants associated consistent with germinal CCR predisposition in 219 patients diagnosed with SPS and 548 controls were genotyped. Subsequently, a PRS was calculated to assess whether people with SPS generally had a higher number of risk alleles than those in the control group. The results showed that seven genetic variants were significantly associated to SPS (rs4779584-GREM1, rs16892766-EIF3H, rs3217810-CCND2, rs992157-PNKD1/TMBIM1, rs704017-ZMIZ1, rs11196172-TCF7L2, rs6061231-LAMA5). Moreover, a 4-fold increase in SPS risk was observed when comparing subjects within the highest decile of variants (≥65) with those in the first decile (≤50). Consequently, we could conclude that PRS has capacity for risk stratification of colorectal cancer suggesting its potential for optimizing screening strategies alongside with other biomarkers and that genetic variants for CRC risk are also involved in SPS susceptibility.

[eng] The objective of this thesis is to estimate prospectively the association between environmental cofactors and the risk of developing cervical cancer and pre-cancer, including tobacco smoking, hormonal and reproductive factors, and serological markers of Human Papillomavirus (HPV), Chlamydia trachomatis (CT) and Human Herpesvirus type 2 (HHV-2) infections. We followed a cohort of 308,036 women recruited in the European Prospective Investigation into Cancer and Nutrition (EPIC) study. During a mean follow-up of 9 years, 261 invasive cervical cancer (ICC) cases and 804 cervical intraepithelial neoplasia grade 3 (CIN3) or carcinoma in situ (CIS) cases were reported. A nested case-control study within the EPIC cohort was also performed, including the sera from 184 invasive cases, 425 cases of CIN3/CIS, and 1,218 matched control women. They were tested for L1, E6 and E7 antibodies against several HPV types, CT and HHV-2 to allow the adjustment for these variables in the statistical analyses and thus confirm associations obtained in the cohort study. L1 seropositivity to any mucosal HPV type was significantly associated with both CIN3/CIS and ICC. By HPV type, HPVs 11, 16, 18, and 45 L1 were associated with CIN3/CIS whereas only HPVs 11 and 16 L1 were significantly related to invasive cervical cancer risk. Associations with HPV types 16 and 18 E6 and E7 seropositivity were only significant for invasive cancer, being the strongest association with HPV 16 E6 seropositivity (OR=10.2). Previous exposure to CT was strongly associated with ICC and weakly associated with CIN3/CIS, and HHV-2 seropositivity was marginally associated with both CIN3/CIS and ICC risk. Increasing number of sexually transmitted infections (HPV L1, CT and HHV-2) was associated with increasing both CIN3/CIS and ICC risk. Current smokers showed a two-fold increased risk of CIN3/CIS and ICC compared to never smokers. Smoking duration and intensity increased the risk of cervical cancer and pre-cancer, with a clear dose-response among ever smokers. Overall, smoking cessation was associated with a reduced by a half of the risk of both CIN3/CIS and ICC. Consistent associations were observed after adjustment for HPV L1, CT and HHV-2 serostatus in the nested case-control study, confirming the results obtained in the cohort. Being a parous woman was positively associated with CIN3/CIS risk, with magnitudes of two-fold, but not with ICC. The risk of pre-invasive cancer also increased with increasing number of full-term pregnancies. Duration of oral contraceptives use was associated with a significantly increased risk of CIN3/CIS and ICC, with relative risks of 1.6 and 1.8 respectively for more than 15 years of use compared to never use. Quitting the use for more than 5 years reduced the risk for CIN3/CIS to almost a half. Ever use of menopausal hormone therapy was associated with a significantly reduced risk of ICC (HR=0.5). No association was found between cervical cancer risk and ever use of intrauterine device (IUD). Analyses restricted to all cases and HPV seropositive controls were also conducted, yielding similar results for parity, OC use and hormonal therapy use, and emerging a significant inverse association with IUD for combined CIN3/CIS and ICC (OR=0.7). This large prospective study confirms the role of tobacco smoking as an important cofactor for both CIN3/CIS and ICC. The strong beneficial effect of quitting smoking is an important finding that will further support public health policies for smoking cessation. Our results also suggest that several hormonal and reproductive factors, as well as infections of CT and HHV-2, are involved in cervical carcinogenesis. Adherence to current cervical cancer screening guidelines should minimize the increased risk of cervical cancer associated with hormonal risk factors. Our study further identifies HPV 16 E6 seropositivity as a potential marker to predict invasive cervical cancer before the disease development., [cat] L’objectiu d’aquesta tesi és estimar prospectivament les associacions entre els cofactors ambientals i el risc de desenvolupar un càncer cervical invasor (ICC) o pre-invasor (CIN3/CIS) utilitzant les dades de l’estudi EPIC (European Prospective Investigation into Cancer and Nutrition). Hem seguit una cohort de 308.036 dones durant un període mitjà de 9 anys, identificant 261 casos de ICC i 804 casos de CIN3/CIS, i hem realitzat un estudi de casos i controls aniuat dins la cohort, incloent sèrums de 184 casos invasors, 425 CIN3/CIS i 1.218 controls aparellats, testant anticossos del Virus del Papil·loma Humà (VPH), la Chlamydia trachomatis (CT) i el Herpesvirus Humà tipus 2 (HHV-2). Els principals resultats han determinat que la seropositivitat per la proteïna L1 a qualsevol tipus del VPH mucós està associat a CIN3/CIS i ICC. En canvi les associacions per les oncoproteïnes E6 i E7 dels tipus del VPH 16 i 18 només han estat significatives pel càncer invasor, destacant la del VPH 16 E6 (OR=10.2). L'exposició passada a CT, i en menor mesura a HHV-2, s’ha vist associada a un major risc de desenvolupar un càncer invasor o pre-invasor. Les dones fumadores tenen el doble de risc de desenvolupar un CIN3/CIS o ICC en comparació amb les dones que no han fumat mai, risc que augmenta amb la duració i intensitat d’ús; en canvi el fet de deixar de fumar s’ha associat a una reducció del risc a la meitat. Un major nombre d’embarassos s’ha associat a un augment en el risc de tenir un CIN3/CIS. La durada en l'ús d’anticonceptius orals s’ha associat a un risc significativament major de desenvolupar un CIN3/CIS o ICC, mentre que deixar d’utilitzar-los durant com a mínim 5 anys redueix el risc de CIN3/CIS a gairebé la meitat. L'ús de teràpia hormonal substitutiva s’ha associat a un menor risc de patir un càncer cervical invasor, en canvi no s’ha trobat cap associació significativa entre l'ús de dispositius intrauterins i el càncer cervical. Aquesta cohort prospectiva confirma el rol del tabac, i en menor mesura dels factors hormonals i reproductius i de les infeccions per CT i HHV-2, com a cofactors del càncer de coll uterí.

[eng] Low risk Human Papillomaviruses (LR-HPVs) 6 and 11 are the main causative agents of benign proliferative lesions such genital warts (GWs) and recurrent respiratory papillomatosis (RRP) (Aubin et al. 2008; Ball et al. 2011; Garland et al. 2009; Prétet et al. 2008). High Risk HPV (HR-HPV) 16 is the most oncogenic type and is responsible for invasive cancers (IC) of cervix (ICC), vulva (IVuC), vagina (IVaC), penis (IPeC) and anus (IAnC) (Alemany et al. 2014, 2015, 2016; Larsson et al. 2013; de Sanjose et al. 2010). The well-established connection between HPV16 infection and IC is observed for the most prevalent ICC histological presentations namely squamous cell carcinomas (SCC), adenocarcinomas (ADC) and adenosquamous cell carcinomas (ADSC) (de Sanjose et al. 2010). All these three HPVs (HPV6, HPV11 and HPV16) are the most prevalent in their associated pathological outcomes (Alemany et al. 2014, 2015, 2016; Aubin et al. 2008; Ball et al. 2011; Garland et al. 2009; Larsson et al. 2013; Prétet et al. 2008; de Sanjose et al. 2010). At level of HPV variants, previous studies have addressed differential HPV6 and HPV11 lineage distributions in GWs and RRP, but mainly at a national or regional level (Kocjan et al., 2009). Although without large sample size, some studies describe differential prevalence of HPV variants among distinct pathologies (Danielewski et al. 2013; Jelen et al. 2014). For HPV6, it has been observed a higher presence of HPV6_B1 variants in GWs compared to RRPs (Flores-Díaz et al. 2017b). HPV16 variant distribution has been mainly focused on the uterus cervix and less on other anatomical sites (Cornet et al. 2012; Yamada et al. 1997), what emphasizes the still wanting research of HPV16 viral lineages in other anogenital cancer location (non-cervical cancers). Some studies show that HPV16 variants in anogenital cancers are largely the same regardless of cancer anatomical locations (ICC, IVuC, IVaC, IAnC and IPeC) , showing increased prevalence of HPV16 A1-3 for all IC of squamous nature (de Koning, Quint, and Pirog 2008; Larsson et al. 2013; Ouhoummane et al. 2013; Tornesello et al. 2008; Zuna et al. 2011). HPV16 variants have been widely studied in SCC (Cornet et al. 2012; Zuna et al. 2011), as this histological type remains the most prevalent ICC (Vinh-Hung et al. 2007; Vizcaino et al. 2000). Nonetheless, data available is poorer for other cervical cancer histological presentations such ADC and ADSC. Although few studies had addressed other glandular histologies, they had been restricted to local geographic origin or small populations (Burk et al. 2003; Lizano 2006; Qmichou et al. 2013; Tornesello et al. 2011). Previously described background, shows an increased prevalence of HPV16_A1-3 variants in SCCs (Zuna et al. 2011) while an enhanced presence of HPV16_D in ADCs (Burk et al. 2003; Mirabello et al. 2016; Quint et al. 2010). Regarding HPV16 intratype variability, it has grown the interest of researchers on the T350G E6 polymorphism as data shows an increased oncogenic potential of those variants containing the 350G allele (Grodzki et al. 2006; Jackson et al. 2016; Zehbe et al. 1998, 2001). Finally, most of the data published do not show results regarding age at tumour diagnosis or use different age definitions such age at enrollment (Mirabello et al. 2016). According to the HPV variant context, the works presented in this thesis try to provide all together information about LR-HPV6 and 11 and HR-HPV16 variant diversity among different lesions, cancers and among different geographical regions with a worldwide spectrum, analyzing large number of HPV monoinfected samples, strategy that prevents possible added bias introduced by co-infections or multi-infections. Through these manuscripts we present for the first time the relative contributions of variant differential abundance, geographical origin, cancer anatomical locations and IC histological presentations to the observation of differential prevalence distribution of HPV16 variants. Through our cases data, we show concordant results with previous published works, observing an increased prevalence of HPV6 B1 variants in GWs, of HPV16 A1-3 variants in anogenital cancers of squamous nature and of HPV16 D variants in ADC. We further show determinate geographical structure of HPV16 variants largely based on the dominance of HPV16 A1-3 variants in Europe, the virtually exclusive presence of HPV16 B and C variants in Africa, the increased prevalence of HPV16 A4 variants in Asia and the enrichment of HPV16 D variants in the Americas. For the most oncogenic-related polymorphism, the E6- T350G, we further show different allelic frequencies according to geographical location independently of the anogenital cancer analyzed, revealing an enhanced 350G allele frequency in isolates from Central-South America compared with Europe. Additionally, we confirme the worldwide trend of cervical cancers to be diagnosed significantly earlier than other anogenital cancers and at histological level, we further present that ADC are diagnosed earlier (mid-forties) than SCC (mid-fifties). According to the previous context, our results complement and may expand those communicated. The current data suggests that the outcome of the virus-host interaction depends on the combination of phylogeny (i.e. the individual genetic background of both virus and patient) and ontogeny (i.e. the differential susceptibility of different tissues) and for HPV16, provide integrating knowledge of variant-specific differential risk that may impact the future screening algorithms, helping to ensure proper early detection of, for example, elusive ADCs. Furthermore, our data emphasizes the necessity of developing deep analyses in HPV variant field. Additionally, the monitoring of initial steps in viral colonization of anogenital mucosas and the follow-up of differential viral persistences according to patient genetic background, may be of remarkable importance. All these research might ultimately provide answers about the extent of the differential fitness of HPV viral lineages and will help to understand the virus-host interplay for the most oncogenic HPVs. Additionally, in vaccinated women follow-up, tracing HPV variants may need to keep more attention as, although not described any evidence, they may be indeed, a possible causative agent of not expected pathological outsiders., [spa] Descripción de la distribución diferencial de variantes de los virus del Papiloma humano 6 y 11 (VPH6 y VPH11) en sus patologías asociadas: verrugas genitales y papilomatosis respiratória recurrente. Descripción de la distribución diferencial de variantes del virus del Papiloma humano 16 (VPH16) en cánceres anogenitales humanos y en los distintos tipos hisológicos de cancer de cérvix (escamoso, adenoescamoso y adenocarcinoma). Para las variantes de VPH16, se decribe su distribución geográfica a nivel mundial y se cuantifica la contribución relativa de la distinta abundancia de las variantes de VPH16, localización anatómica del cáncer,histología y geografía respecto a la observación diferencial de las variantes de VPH16.

[cat] INTRODUCCIÓ: La prova de detecció de sang oculta en femta és la prova de cribratge de càncer colorectal (CCR) més implementada a Europa. Tot i que aquesta ha demostrat reduir la mortalitat per CCR, encara dista de ser una prova perfecta donat el seu risc de resultats falsos positius i falsos negatius. Els efectes adversos del cribratge podrien reduir-se si coneguéssim els determinants associats a resultats falsos positius i negatius. A més, la identificació de polimorfismes genètics, en combinació amb altres biomarcadors moleculars i factors ambientals podria permetre personalitzar el cribratge poblacional. OBJECTIUS: 1) ldentificar els factors associats a un resultat fals positiu de la prova de detecció de sang oculta en femta immunològica (PDSOFi). 2) Determinar la proporció de resultats falsos negatius de la PDSOFi i explorar els seus determinants. 3) Elaborar un model de predicció de risc de CCR basat en població espanyola que combini la predisposició genètica juntament amb factors ambientals i antecedents familiars per estratificar la població en grups de risc de CCR. METODOLOGIA: Per respondre als primers dos objectius es van utilitzar dades del programa poblacional de cribratge de CCR que coordina l'lnstitut Català d'Oncologia. Es realitzaren dos estudis transversals amb participants en el cribratge que van obtenir un resultat positiu (n = 427) o negatiu (n = 218) en la PDSOFi i als que posteriorment se'ls va realitzar una colonoscopia i un qüestionari. Finalment, per respondre al tercer objectiu s'aprofitaren les dades de l'estudi MCC-Spain, que es un estudi de casos i controls multicèntric que va recollir 1.336 casos de CCR i 2.744 controls. De tots ells disposàvem d'un qüestionari epidemiologic i un array d'exoma que contenia 21 polimorfismes de nucleotid simple associats a risc de CCR. Amb aquestes dades varem desenvolupar un model de predicció de risc per estratificar la població en grups de risc de CCR. RESULTATS: Estudi 1: Mitjançant l'estudi que avalua el risc de resultats falsos positius en la PDSOFi, s'observa que les variables que van resultar ser determinants de fals positiu foren: ser dona, haver participat préviament en el cribratge, la presència d'hemorroides i/o fissura anal, així com prendre fàrmacs inhibidors de la bomba de protons. Estudi 2: A l'avaluar el risc de resultats falsos negatius en la PDSOFi, la proporció de falsos negatius observada va ser del 15,6%. L'hemoglobina fecal va ser indetectable en un 45,5% dels resultats i va ser inferior a 4 microgr. Hb/g en un 94,0% d'individus amb un resultat fals negatiu. Estudi 3: Vam observar que les variables (ajustades pel model basal) que es van associar de forma independent a CCR van ser: el nombre d'al·lels de risc, antecedents familiars de CCR, l'obesitat, l'activitat física, la ingesta de carn vermella i vegetals, i el no consum d'antiinflamatoris no esteroïdals. Els individus amb més de 25 al·lels de risc tenien un 82% més de risc que els individus amb menys de 19 al·lels de risc. CONCLUSIONS: 1. L'ús d'inhibidors de la bomba de protons durant la realització de la PDSOFi podria incrementar la probabilitat d'un resultat fals positiu. 2. Disminuïr el llindar de positivitat de la PDSOFi no detectaria més neoplàsia avançada i augmentaria els costos i els riscos innecessaris per als pacients. 3. El model de predicció del risc de CCR basat en població espanyola que combina la predisposició genètica amb factors ambientals i antecedents familiars podria ser útil per estratificar individus de risc., [spa] INTRODUCCIÓN: La prueba de detección de sangre oculta en heces es la prueba de cribado de cáncer colorrectal (CCR) más implementada en Europa. Aunque ésta ha demostrado reducir la mortalidad por CCR, dista de ser una prueba perfecta dado su riesgo de resultados falsos positivos y falsos negativos. Los efectos adversos del cribado podrían reducirse si conociéramos los determinantes asociados a resultados falsos positivos y negativos. Además, la identificación de polimorfismos genéticos, en combinación con otros biomarcadores moleculares y factores ambientales podría permitir personalizar el cribado poblacional. OBJETIVOS: 1) ldentificar los factores asociados a un resultado falso positivo de la prueba de detección de sangre oculta en heces inmunológica (PDSOHi). 2) Determinar la proporción de resultados falsos negativos de la PDSOHi y explorar sus determinantes. 3) Elaborar un modelo de predicción de riesgo de CCR basado en población española que combine la predisposición genética junto con factores ambientales y antecedentes familiares para estratificar la población en grupos de riesgo de CCR. METODOLOGIA: Para responder a los primeros dos objetivos se utilizaron datos del programa poblacional de cribado de CCR que coordina el lnstituto Catalán de Oncología. Se realizaron dos estudios transversales con participantes en el cribado que obtuvieron un resultado positivo (n = 427) o negativo (n = 218) en la PDSOHi y a quienes posteriormente se les realizó una colonoscopia y un cuestionario. Finalmente, para responder al tercer objetivo se aprovecharon los datos del estudio MCC-Spain, que es un estudio de casos y controles multicentrico que recogió 1.336 casos de CCR y 2.744 controles. De todos ellos disponíamos de un cuestionario epidemiológico y un array de exoma que contenía 21 polimorfismos de nucleótido simple asociados a riesgo de CCR. Con estos datos pudimos desarrollar un modelo de predicción de riesgo para estratificar la población en grupos de riesgo de CCR. RESULTADOS: Estudio 1: Mediante el estudio que evaluó el riesgo de resultados falsos positivos en la PDSOHi, se observó que las variables que resultaron ser determinantes de falso positivo fueron: ser mujer, haber participado previamente en el cribado, la presencia de hemorroides y/o fisura anal, así como tomar fármacos inhibidores de la bomba de protones. Estudio 2: Al evaluar el riesgo de resultados falsos negativos en la PDSOHi, la proporción de falsos negativos observada fue del 15,6%. La hemoglobina fecal fue indetectable en un 45,5% de los resultados y fue inferior a 4 microgr. Hb/g en un 94,0% de individuos con un resultado falso negativo. Estudio 3: Se observó que las variables (ajustadas por el modelo basal) que se asociaron de forma independiente a CCR fueron: el numero de alelos de riesgo, antecedentes familiares de CCR, la obesidad, la actividad física, la ingesta de carne roja y vegetales, y el no consumo de antiinflamatorios no esteroideos. Los individuos con más de 25 alelos de riesgo tenían un 82% más de riesgo que los individuos con menos de 19 alelos de riesgo. CONCLUSIONES: 1. El uso de inhibidores de la bomba de protones durante la realización de la PDSOHi podría incrementar la probabilidad de un resultado falso positivo. 2. Disminuir el umbral de positividad de la PDSOHi no detectaría mas neoplasia avanzada y aumentaría los costes y los riesgos innecesarios para los pacientes. 3. El modelo de predicción del riesgo de CCR basado en población española que combina la predisposición genetica con factores ambientales y antecedentes familiares podría ser útil para estratificar individuos de riesgo., [eng] INTRODUCTION: ln Europe, the most commonly implemented test to screen colorectal cancer (CCR) is the faecal occult blood test. Although it has shown to reduce CRC mortality, it is far from perfect because of its false positive and false negative results. The adverse effects of screening could be reduced if we knew the determinants associated with false positive and negative results. ln addition, the identification of genetic polymorphisms in combination with other molecular biomarkers and environmental factors could allow personalizing population screening. OBJECTIVES: 1) To identify the factors associated with a false positive result of the faecal immunochemical test (FlT). 2) To determine the false negative proportion in and to explore its determinants. 3) To develop a risk stratification model that combined environmental factors with family history and genetic susceptibility to stratify the population in risk groups of CRC. METHODOLOGY: ln order to respond to the first two objectives, the population of the CRC screening program coordinated by the Catalan lnstitute of Oncology was used. Two cross-sectional studies were performed with screening participants who had a positive (n = 427) or negative (n = 218) FlT result and who subsequently underwent a colonoscopy and a questionnaire. Finally, in order to respond to the third objective, data from the MCC-Spain study was used. This multicentre case-control study recollected an epidemiological questionnaire and an exoma array containing 21 simple nucleotide polymorphisms associated with CRC risk of 1,336 cases of CRC and 2,744 controls. With these data we developed a risk stratification model to stratify the population in groups at risk of CRC. RESULTS: Study 1: While evaluating factors associated with false-positive results in a CRC screening, the variables associated with a false-positive result were: women, successive screening, haemorrhoids and/or anal fissure and the use of proton pump inhibitors. Study 2: When analysing the risk of false negative results in the FlT, the false negative rate observed was 15.6%. Faecal haemoglobin was undetected in 45.5% and it was below 4 micrograms Hb/g in 94.0% of the individuals with a FN result. Study 3: The variables (adjusted by the baseline model) that were independently associated with CRC were: the number of risk alleles, family history of CRC, alcohol consumption, obesity, physical activity, red meat and vegetable intake, and nonsteroidal anti- inflammatory drug use. The risk of subjects with more than 25 risk alleles was 82% higher than subjects with less than 19 alleles. CONCLUSIONS: 1. Concurrent use of proton pump inhibitors at the time of FlT might increase the likelihood of a false-positive result. 2. Decreasing the positivity threshold of FlT does not increase the detection rate of advanced neoplasia, but may increase the costs and potential adverse effects. 3. A Spanish population-based CRC risk prediction model that combines genetic predisposition with environmental factors and family history may be useful for stratifying at-risk individuals.

[eng] Papillomaviruses (PVs) are a numerous family of small dsDNA viruses infecting virtually all mammals. PVs cause infections without triggering a strong immune response, and natural infection provides only limited protection against reinfection. Most PVs are part and parcel of the skin microbiota. In some cases, infections by certain PVs take diverse clinical presentations, from highly productive self-limited warts to invasive cancers. The main aim of this thesis is explore the link between HPV genotypic diversity and the phenotypic, clinical diversity of the diseases associated with HPV infection by means of evolutionary, clinical and ecologically approaches. From and evolutionary perspective, we studied the codon usage preferences of HPVs. By applying phylogenetic inference and dimensionality reduction methods, we found that phylogenetic relationships between HPVs explained only a small proportion of codon usage preferences (CUPrefs) variation. Instead, the most important explanatory factor for viral CUPrefs was infection phenotype, as orthologous genes in viruses with similar clinical presentation displayed similar CUPrefs. Moreover, viral genes with similar spatiotemporal expression patterns also showed similar CUPrefs. Moreover, we also found that HPV genes with similar spatiotemporal expression patterns displayed similar CUPrefs. Hence, Our results suggest that CUPrefs in HPVs reflect either variations in the mutation bias or differential selection pressures depending on the clinical presentation and expression timing. From a clinical point of view, first, we studied the distribution patterns of oncogenic and non-oncogenic HPVs in anal and perianal Squamous Intraepithelial Lesions (SIL) in non-vaccinated heterosexual men, women, and Men who have sex with men (MSM) with known HIV status. We found that there is an increased prevalence of low-grade perianal lesions driven only by oncogenic HPVs. We also found a high prevalence of anal SIL containing foci of high-grade SIL exclusively driven by non-oncogenic HPVs. Our results suggest that there is a disagreement in high-grade/low-grade intraepithelial neoplasia and oncogenic/non-oncogenic HPV infection. Second, we analyzed the presence of HPVs not explicitly targeted by standard molecular epidemiologic methods of detection in squamous carcinoma samples of the vulva, penis and head and neck. These three anatomical locations display a low fraction of cancer cases attributable to HPVs, in sharp contrast with the higher rates of viral DNA prevalence in anal and cervical carcinomas. The standard HPV detection methods target only a subset of clinically important HPVs, namely oncogenic AlphaPVs, and may thus overlook the presence of other HPVs. We tested 2365 samples and found 6 samples containing cutaneous HPVs, suggesting that certain cutaneous HPVs, typically classified as “non- oncogenic” HPVs, may be linked to small number of cancer cases and call for further studies to elucidate the pathogenic role and malignisation mechanism of these HPVs. Finally, from an ecological perspective, we studied the interaction among HPVs inside its host in different stages of the cervical infection and different anatomical regions. By applying established ecological methods, we found that HPVs interact within the host, and that the presence of one given HPV is not neutral for the rest of the HPVs infecting the host. We also study how the interaction among HPVs could be affected by the introduction of ecological pressures linked to vaccination. By applying the same ecological methods, we find that in the initial descriptions of the post-vaccination era, HPVs still form non-neutral communities suggesting that the vaccine is not changing the underlying processes that govern HPV distributions and relative abundances., [cat] Los virus del Papiloma (VPs) son una numerosa familia de virus pequeños de DNA de doble cadena que infectan a todos los mamíferos. La mayoría de infecciones por VPs son subclínicas, sin embargo, en algunos casos, las infecciones por ciertos VPs toman diversas presentaciones clínicas, desde verrugas, lesiones altamente productivas, hasta cánceres invasivos. El objetivo principal de esta tesis es explorar el vínculo entre la diversidad genotípica de los virus del Papiloma Humano (VPH) y la diversidad fenotípica y clínica de las enfermedades asociadas con la infección por VPH mediante 3 enfoques: evolutivo, clínico y ecológico. Desde una perspectiva evolutiva, se estudiaron las preferencias de uso de codones (CUPrefs) de los VPHs. Aplicando métodos de inferencia filogenética y métodos de reducción de la dimensionalidad, se observó que que los genes ortólogos en los virus con presentaciones clínicas similares mostraron CUPrefs similares. Además, los genes virales con patrones de expresión espacio-temporales similares también mostraron CUPrefs similares. Nuestros resultados sugieren que las CUPrefs de los genes de los VPHs reflejan variaciones en los sesgos mutacionales así como presiones de selección diferencial, dependiendo de la presentación clínica de los VPHs y el momento de expresión génica. Desde el punto de vista clínico, se estudiaron los patrones de distribución de VPHs oncogénicos y no oncogénicos en lesiones intraepiteliales anales y perianales. Los resultados muestran que en hombres homosexuales (HSH) VIH positivos hay un aumento de la prevalencia de lesiones perianales de bajo grado debidas a infecciones causadas por VPHs oncogénicos en comparación con HSH que son VIH negativos. Nuestros resultados también muestran que en HSH, independientemente del estatus de VIH, existe una alta prevalencia (>65%) de lesiones anales de bajo grado que contienen focos de lesiones de alto grado que son asociadas a la infección con VPHs no oncogénicos. También se analizó, en muestras de carcinoma escamoso de vulva, pene y cabeza y cuello, la presencia de VPHs cutáneos. Se testaron 2365 muestras y se detectó que 6 muestras contenían AND de VPHs cutáneos. Estos resultados sugieren que ciertos tipos de VPH cutáneos podrían tener potencial oncogénico. Se requieren de estudios adicionales para dilucidar el papel patogénico y el mecanismo de malignización de estos VPHs. Por último, desde una perspectiva ecológica, se estudió la interacción entre los VPHs en diferentes estadios de la infección cervical y diferentes regiones anatómicas. Los resultados mostraron que los VPH interactúan dentro del huésped. También se estudió cómo la interacción entre los VPH podría verse afectada por la introducción de presiones ecológicas vinculadas a la vacunación. Los resultados mostraron que en las descripciones iniciales de la era posterior a la vacunación, los VPH también forman comunidades no neutrales, sugiriendo que la vacuna no está cambiando los procesos subyacentes que gobiernan las distribuciones de VPH y las abundancias relativas.

En esta tesis se han desarrollado e implementado distintas herramientas bioinformáticas que permiten el estudio de las interacciones miRNA-mRNA en contextos celulares específicos. oncretamente se ha creado un paquete de R (miRComb) que calcula las interacciones miRNA-mRNA partiendo de expresión de miRNAs y mRNAs, y predicciones bloinformáticas de bases de datos preexistentes. Las interacciones miRNA-mRNA finales son aquellas que muestran una correlación negativa y han estado predichas por al meno una base de datos. Como valor añadido, el paquete miRComb realiza un resumen en pdf con los resultados básicos del análisis (número de interacciones, número de mRNAs target por miRNA, análisis funcional, etc.), que permite comparar los datos de distintos estudios. Hemos aplicado esta metodología en el contexto de cánceres digestivos. En un primer estudio hemos utilizado datos públicos de 5 cánceres digestivos (colon, recto, esófago, stómago e hígado) y hemos determinado las interacciones miRNA-mRNA comunes entre ellos y específicas de cada uno. En un segundo estudio, hemos utilizado la misma metodología para analizar datos de IRNA-mRNA en biopsias de pacientes del Hospital Clínic de Barcelona con cáncer de páncreas. En este estudio hemos descrito interacciones miRNA-mRNA en el contexto de cáncer pancreático y hemos podido validar dos de ellas a nivel experimental. En resumen, podemos concluir que el paquete miRComb es una herramienta útil para el estudio del interactoma de miRNA-mRNA, y que ha servido para establecer hipótesis biológicas que luego se han podido comprobar en el laboratorio.

Acrylamide is an organic chemical that has several important uses in industry, such as a flocculant in public water supplies, and in biomedical research. In 1994, acrylamide was classified by the International Agency for Research on Cancer as ‘probably carcinogenic’ to humans. In 2002, acrylamide was discovered in some starchy foods that are cooked at high temperature (e.g., potato crisps and chips). Acrylamide also is a common component of cigarette smoke. From 2006-2010, epidemiologic studies began to examine and report on the possible association between dietary acrylamide intake and risk of human cancer, including ovarian and endometrial cancers (two of which suggested possible associations). Endometrial and ovarian are the fourth and fifth most common cancers diagnosed in European women. The general aim of this thesis was to assess the risk of endometrial cancer (EC) and epithelial ovarian (EOC) cancer in relation to dietary acrylamide exposure based on questionnaire information and biomarkers of internal exposure. The study population for this project was 368,010 women from the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. Dietary acrylamide intake in the full EPIC cohort was estimated using the most recent European acrylamide-content food databases, country-specific food intake questionnaires, and 24-hour dietary recalls at baseline. Multivariate Cox Proportional hazard models were used to estimate hazard ratios (HR) and 95% confidence intervals (95% CI) for questionnaire-based acrylamide intake and EC or EOC risk. Cohort analyses were conducted in all women and in subgroups. Two nested case-control studies of acrylamide biomarkers and EC or EOC risk were conducted in non-smoking postmenopausal women. Pre-diagnostic biomarkers of acrylamide and its epoxide metabolite, glycidamide, were measured as hemoglobin adducts (HbAA and HbGA, respectively) using HPLC/MS/MS in red blood cells collected at baseline 5-10 years before diagnosis. Unconditional logistic regression models were used to estimate odds ratios (OR) and 95% CI for the relation between acrylamide biomarkers (HbAA, HbGA, HbAA+HbGA, and HbGA/HbAA) and EC or EOC risk. The relation between dietary and demographic determinants of biomarker levels of acrylamide and glycidamide were also examined using simple and multiple linear regression analyses in 801 control women. After a mean follow-up of 11 years, there were 1,382 cases of EC and 1,191 cases of EOC in the EPIC cohort. In general, the results of the dietary acrylamide intake analyses showed no associations with EOC and EC, except for a possible association between higher acrylamide intake and type-I EC in women who were non-smokers and non-users of oral contraceptives (HRQ5vsQ1: 1.97, 95% CI: 1.08-3.62). The main food group determinants of acrylamide and glycidamide biomarker levels were biscuits, crackers and cakes; and alcohol intake and body mass index were important determinants of the ratio of glycidamide to acrylamide. The nested case-control study of acrylamide and glycidamide biomarker levels and EC (383 cases, 171 type-I, and 385 controls) did not find any support for associations between acrylamide or glycidamide and EC risk. The nested case-control study of acrylamide and glycidamide biomarkers and EOC risk (334 cases and 417 controls) found some evidence for associations with the third quintile of HbGA, and the second and third quintiles of HbAA+HbGA. Participants classified in the fifth quintile of HbGA or HbAA+HbGA showed statistically non-significant higher risks of EOC (ORHbGA.Q5vsQ1: 1.63, 95% CI: 0.92-2.86; P-trend: 0.04, and ORHbAA+HbGGA.Q5vsQ1: 1.60, 95% CI: 0.89-2.87; P-trend: 0.14). In conclusion, results in women from the EPIC cohort did not provide convincing support for the hypothesis that higher acrylamide and glycidamide exposures are associated with higher risk for EC. For EOC, inconsistent results were observed in our study. Additional nested case- control studies with larger sample size in different populations, and pooled analysis of existing studies of the relation between acrylamide and glycidamide biomarkers and EOC risk should be performed.

[cat] Durant aquests últims anys hem viscut el gran impacte de l'era post-genòmica, en la qual la seqüència del genoma humà ja és coneguda. En aquesta etapa, per exemple, s'ha pogut demostrar que el DNA no codificant, inicialment considerat com a no funcional, té un rol essencial en el funcionament de les cèl.lules humanes. Com a conseqüència d'aquest nou escenari, així com de la reducció del cost dels experiments a gran escala i la incorporació de la bioinformàtica com a eina essencial, la recerca en biologia molecular i salut pública ha modificat el seu enfocament clàssic per incorporar aquest canvi de paradigma. El càncer, una malaltia ja tractada per civilitzacions presents segles enrere, per primera vegada podrà ser estudiat des d'un punt de vista holístic a nivell cel.lular. Més específicament, en el cas del càncer colorectal, una malaltia altament heterogènia i molecularment molt complexa, seran de gran ajuda tots els avenços possibles en quant a tècniques de laboratori, tecnologia i noves disciplines. Durant els propers anys molt probablement s'incorporaran a la rutina clínica noves classificacions moleculars, que permetran prescriure un tractament més precís en comptes dels agents quimioteràpics utilitzats actualment amb la majoria dels pacients. També s'identificaran biomarcadors més acurats per al diagnòstic precoç i per predir millor el pronòstic de la malaltia. L'aprofundiment en les bases moleculars del càncer colorectal ens durà fins i tot a desenvolupar eines diagnòstiques basades en criteris moleculars, en substitució dels criteris clínics i patològics emprats majoritàriament a dia d'avui. Amb aquesta tesi el lector podrà endinsar-se en treballs concrets que intenten avançar en aquestes direccions. Els resultats es presenten com a compendi de tres articles científics publicats a revistes científiques internacionals. Tots tres tracten de càncer colorectal, i són el resultat de part del treball realitzat durant els últims anys a la Unitat de Biomarcadors i Susceptibilitat de l'Institut Català d'Oncologia., [eng] Over the last years we have lived the great impact of post-genomic era, in which the human genome was already known. In this stage, for example, it has been demonstrated that noncoding DNA, initially considered as non-functional DNA, has an essential role in human cells. As a result of this new scenario, as well as reducing the cost of large-scale experiments and incorporating the bioinformatics as an essential tool, the research in molecular biology and public health has changed its traditional approach to incorporate this paradigm shift. Cancer, a disease already treated by civilizations many centuries ago, for the first time can be studied from a holistic point of view at the cellular level. In particular, to study the colorectal cancer, a heterogeneous and complex disease, will be helpful all new developments in terms of laboratory techniques, technology and new disciplines. Over the next few years, likely will be incorporated new molecular classifications into the clinical routine, that will able to recommend a more precise medicine instead of the chemotherapeutic agents, currently used with the vast majority of patients. More accurate biomarkers for early diagnosis and to better predict the prognosis of the disease also will be identified. Deepening at the molecular basis of colorectal cancer will lead to even develop diagnostic tools based on the molecular information, replacing the clinical and pathological criteria currently used. In this thesis the reader will be introduced in specific works that try to advance in these directions. The results are based on a collection of three scientific articles published in international journals. These three articles speak about colorectal cancer and represent the work carried out during the last years at the Biomarkers and Susceptibility Unit of the Catalan Institute of Oncology.

Discurs de presentació del professor Victor Moreno Aguado, Discurs de la solemne investidura acadèmica de conferiment del grau de doctor honoris causa al senyor Francesc Xavier Bosch i José. El professor Víctor Moreno Aguado repassa la trajectòria del prestigiós oncòleg, posant èmfasi en les seves importants aportacions en la prevenció del càncer.

Discurs de presentació del professor Victor Moreno Aguado Discurs de la solemne investidura acadèmica de conferiment del grau de doctor honoris causa al senyor Francesc Xavier Bosch i José. El professor Víctor Moreno Aguado repassa la trajectòria del prestigiós oncòleg, posant èmfasi en les seves importants aportacions en la prevenció del càncer.

[cat] En aquesta tesi, s’analitzen les dades d’un GWAS de càncer de mama més enllà de l’anàlisi estadístic. Sota la perspectiva de la biologia de sistemes, s’apliquen unes estratègies basades en conjunts de gens/proteïnes funcionalment relacionats per identificar nous gens de baixa penetrància, els processos biològics en els que participen i els mecanismes moleculars mitjançant els quals confereixen el risc així com les seves interaccions. El càncer de mama és un exemple de malaltia complexa on participen tant factors genètics com ambientals en la seva etiologia. Entendre la base genètica del càncer de mama en profunditat podria permetre entendre millor la malaltia i desenvolupar millors eines de diagnòstic, pronòstic i/o tractament. D’aquesta manera, s’ha determinat que podrien influir en el risc de càncer de mama variants comunes en els gens dels processos biològics Cell Communication i Cell Adhesion; així com en determinats loci de les driver kinases, particularment en els gens que codifiquen per receptors d’EPHs. A més, variants comunes en el locus EPHB1 podrien estar associades amb el risc a càncer de mama en portadors de mutacions en BRCA1 i BRCA2. D’altra banda, s’ha determinat que les interaccions gen-gen podrien contribuir a explicar part de la missing heritability i inclourien freqüentment gens relacionats amb processos fonamentals en diferents tipus cel·lulars, com són el metabolisme i la biosíntesi de molècules. A més, les interaccions gen-gen associades al risc de càncer podrien ser predites en part mitjançant els patrons complexes de coexpressió gènica específics del tipus de càncer. Recentment, a partir dels resultats del metanàlisi de càncer de mama més gran realitzat fins el moment, s’ha suggerit que existeixen molts més loci (potser alguns milers), la majoria amb 0,95 < ORs < 1,05 que contribueixen a la susceptibilitat. Aquests resultats donen suport a la hipòtesi del model infinitesimal per explicar la missing heritability en càncer de mama. Segons aquesta hipòtesi, centenars o milers d’al·lels comuns, amb efectes relativament molt petits, contribuirien a la susceptibilitat a la malaltia. Així, els GWAS portats a terme fins ara, haurien detectat els al·lels amb un efecte major, la resta dels al·lels de susceptibilitat no poden ser detectats degut als estrictes llindars de significació utilitzats en aquests estudis. En aquest context, les interaccions genètiques tampoc poden ser detectades amb robustesa degut als estrictes llindars de significació requerits. Aquest fet, justifica la nostra estratègia basada en la biologia de sistemes per l’anàlisi dels resultats del GWAS en lloc de centrar-nos en els SNPs/gens individualment., [eng] In GWAS, genes are interpreted individually, and a gene ranked below the significance threshold will not be measured or experimentally characterized in relation to the disease or to genes that passed the threshold. Given these statistical constraints of GWAS, a Systems Biology approach can facilitate interpretation of GWAS results. In this thesis, we applied an integrative approach for identifying candidate low- penetrance breast cancer susceptibility genes, their characteristics and molecular networks through the analysis of diverse sources of biological evidence. Breast cancer is a complex disease; both genetic and environmental factors contribute to its etiology. A deep knowledge of genetic basis of breast cancer contributes to disease understanding and better diagnostic, prognostic and treatment tools. As a result of this work we propose that breast cancer risk may be influenced by common variants in Cell Communication and Cell Adhesion Biological Processes; also in driver kinases loci, specifically in genes encoding EPHs receptors. Moreover common variants in EPHB1 locus may be associated to breast cancer in BRCA1 and BRCA2 mutation carriers. In the other hand, gene-gene interactions might contribute to missing heritability. In addition, gene-gene interactions associated to breast cancer may be predicted using complex co-expression patterns in the specific cancer type.

Unitat de Biomarcadors i Susceptibilitat – Institut Català d’Oncologia S'han publicat diversos perfils d'expressió amb intenció predictiva del pronòstic en malalts amb càncer colorectal, però cap s'ha implantat a la pràctica clinica. El projecte consistirà en avaluar la capacitat predictiva dels perfils en una sèrie de datasets publics i mesurar la potencial utilitat.. S'han publicat diversos perfils d'expressió amb intenció predictiva del pronòstic en malalts amb càncer colorectal, però cap s'ha implantat a la pràctica clinica. El projecte consistirà en avaluar la capacitat predictiva dels perfils en una sèrie de datasets publics i mesurar la potencial utilitat

[cat] Al llarg de les dues darreres dècades hem tingut el privilegi d’assistir a un dels esdeveniments més importants en la història de la biomedicina: el desenvolupament del Projecte Genoma Humà. Conjuntament amb aquest fet, les tècniques de laboratori a gran escala, cada dia més potents i fiables, s’utilitzen actualment de manera rutinària en el camp de la recerca biomèdica. Tot i que encara queda un gran camí per recórrer, aquests fets indubtablement han sembrat la llavor per desenvolupar un nou paradigma de recerca en càncer, així com per millorar els futurs tractaments dels pacients. Ens estem movent progressivament d’un escenari on els diagnòstics i els tractaments es basen principalment en criteris patològics a un altre completament personalitzat, on cada pacient serà diagnosticat i tractat d’una manera especialitzada en funció de criteris moleculars. Tot i així, per assolir una medicina del càncer totalment eficient i personalitzada és essencial obtenir una imatge acurada de tots els processos moleculars involucrats en el desenvolupament d’aquesta malaltia complexa. Per obtenir aquesta imatge necessitarem, doncs, obtenir informació a gran escala de la cèl•lula tumoral a nivell de genoma, epigenoma, transcriptoma i proteoma. Un cop tota la información està disponible, s’hauran d’aplicar les tècniques analítiques integratives per detectar les alteracions moleculars que tenen un paper primordial el procés tumoral. Els estudis presentats en aquesta tesi pretenen ser una mostra d’aquests tipus d’anàlisis integratives. Aquesta tesi es basa en una col•lecció de tres articles que són el resultat de la feina duta a terme a la Unitat de Biomarcadors i Susceptibilitat de l’Institut Català d’Oncologia., [eng] Over the last two decades we have been given the privilege of witnessing one of the most relevant breakthroughs in the history of biomedicine: the development and completion of the Human Genome Project. Along with it, large-scale laboratory techniques, every day more powerful and reliable, are now routinely applied in biomedical research. Although there is still a long way to go, this fact has undoubtedly set the seed for a new paradigm in cancer research and the future treatment of patients. We are progressively shifting from a scenario where diagnoses and treatments are mainly based on pathological criteria to a completely personalized one, where every single patient will be diagnosed and treated in a specialized manner according to molecular criteria. Nonetheless, to achieve a fully efficient and personalized cancer medicine, it is essential to obtain an accurate picture of all the molecular processes involved in the development of such a complex pathology as cancer. This picture can only be obtained if we can have a detailed view of a tumor cell’s status at a whole genome, epigenome, transcriptome and proteome levels. Once all the information is available, suitable analytical integrative techniques must be applied to detect the molecular alterations that play a driver role in the tumorigenic process. The studies presented in this thesis aim to be examples of such integrative analyses. This thesis is based on a collection of three articles, which are the result of the work done at the Unit of Biomarkers and Susceptibility at the Catalan Institute of Oncology.

Els avenços que a les darreres dècades han protagonitzat les tècniques de genotipatge i de seqüenciació, unit al desenvolupament de tècniques estadístiques especialitzades i sofisticades, han permès elaborar noves vies de recerca per comprendre l’etiologia de malalties complexes l’origen de les quals, en molts casos, és multifactorial. Així com s’han establert factors ambientals que poden modular el risc de patir certes malalties, també s’han detectat variants genètiques que hi poden estar involucrades. Patologies com la diabetis, el càncer, l’esquizofrènia o l’asma es veuen influenciades per factors genètics en interacció amb factors ambientals. Al capdavant d’aquestes investigacions es troben els mapes de polimorfismes. El polimorfisme més comú al genoma humà és la variació en una sola base de la seqüència genòmica, l’anomenat ”Single Nucleotide Polimorphismï conegut per les seves inicials ”SNP”. Degut a la seva abundància, els SNPs són molt adients per generar mapes genètics i han esdevingut els marcadors més utilitzats en estudis d’associació genètica. Si bé des de fa dècades l’estudi del genoma humà s’ha centrat principalment en analitzar les variacions en la seqüència genòmica, des d’inicis de l’any 2000 sabem per diversos estudis que aquestes variacions tendeixen a donar-se en bloc. D’altra banda, també s’ha demostrat que les recombinacions genètiques que es donen al llarg del genoma no es produeixen de manera uniforme. Per aquest motiu, el genoma presenta zones que es transmeten en bloc, de progenitors a descendents, i que poden incloure blocs de variacions. Aquestes zones de baixa recombinació que es segreguen en bloc són els anomenats haplotips. Els haplotips poden facilitar el descobriment de gens relacionats amb malalties que pateixen els éssers humans. L’inter`Les en l’assignació d’haplotips i l’anàlisi de l’associació entre haplotips i malaltia en mostres d’individus no relacionats ha crescut incommensurablement als darrers anys degut a l’èmfasi que projectes com HapMap han situat sobre l’anàlisi d’haplotips. Ara bé, la deter-minació dels haplotips donada una mostra de genotips per un conjunt d’individus no sempre és immediata, havent de recórrer a tècniques específiques per tal de separar els cromosomes. Les tècniques de tipus molecular són les que aporten menys error però desafortunadament són cares i això dificulta el seu ús, sobretot en estudis poblacionals que tracten amb mostres grans. Per superar aquesta limitació, les investigacions han tendit a utilitzar la inferència estadística com a via més usual a l’hora de determinar els haplotips. La inferència sobre les freqüències haplotípiques és una bona solució per reconstruir la mostra haplotípica, però cal tenir present els efectes que el fet de treballar amb estimacions comportarà sobre tots els càlculs que es realitzin amb la mostra. En aquest sentit, resulta interessant dedicar esfor¸cos per tal d’intentar minimitzar la propagació d’aquests errors en les anàlisis d’associació genètica amb haplotips. Tot i que existeix diversitat de programes per fer anàlisis haplotípiques aplicables a mostres d’individus no relacionats, molts d’ells presenten limitacions que esdevenen una bona mo-tivació per intentar cercar d’altres alternatives teòriques i computacionals per tractar més eficientment la problemàtica dels haplotips. En aquesta tesi doctoral es presenta el desenvolupament i la implementació informàtica d’un mètode per estimar haplotips i els efectes associats a diversos tipus de fenotips. El marc teòric amb que s’ha treballat és la inferència Bayesiana combinada amb tècniques de Markov Chain Monte Carlo que optimitzin les qües-tions computacionals. L’eina resultat és el paquet BayHap, publicada com a paquet a l’entorn estadístic R. El programa s’ha validat sobre escenaris de dades simulats i sobre dades reals. L’aplicació mostra millores en l’estimació d’efectes associats a haplotips amb baixa freqüència i alhora ofereix la possibilitat de dur a terme l’anàlisi d’haplotips sota un punt de vista bayesià amb les avantatges que aquest fet ofereix., Nowadays, haplotypic information has become vitally important to clarify the genetic basis of the etiology of some common diseases. Comparing DNA of healthy and diseased individuals let us to describe changes in the genomic sequence that could modify the risk of su ering from the disease. Association studies are the framework where this class of analysis are carried out. The DNA variations more often analyzed due to its high frequency along the genome are the Single Nucleotide Polimorphisms. One \SNP" is the change in only one nucleotide between individuals at the same position of their genomes. Is well known that there are zones in the genomic sequence with a low rate of recombinations, that are inherited as a block by the o spring. These zones are called haplotypes, and everyone carries two of them. On the other hand, in the last decade researchers have stated that mutations as SNPs are also transmitted in blocks, situated in haplotypic zones. For all of this, the knowledge of haplo- types corresponding to a sample of genotypes observed for some SNPs of a set of unrelated individuals could be very helpful to better understand the genetic association with a phe- notype of interest. Initiatives as the international HapMap project have strongly motivated the scienti c community to use haplotypes in association analysis. Unfortunately, in the absence of family data, obtaining haplotypic information is not straightforward. Since every cell of the human organism contains 22 pairs of homologous chromosomes, plus the sexual chromosome, for each chromosomical location at the autosomal chromosomes there are two bases, one for each homologous chromosome at the same position of the DNA sequence. Given that current lab techniques usually only report genotypic data and do not provide the chromosome for each base, individuals with two or more heterozygous sites have uncertain haplotypes because there is more than one possible haplotype pair compatible with their genotype. To overcome this limitation, research has tend to use statistical inference as the most common way to determine haplotype information. Although statistical inference of haplotypes frequencies is a good solution to reconstruct the haplotypes sample researchers have to take into account the error propagation in posterior analysis. This is an encouraging situation to apply e orts in order to minimize unwanted e ects. Even though there are some programs to do haplotypic analysis over unrelated individuals, most of them have unresolved questions as could be the estimation of association between disease and low frequency haplotypes. In this work we develop and implement a method to infer haplotypes and association with many types of phenotypes. The theoretical frame- work in which we based our algorithm is the bayesian inference and Markov Chain Monte Carlo techniques. The resulting application is an R package called BayHap available from de repository of packages of this free statistical environment. The program was validated through computational simulations and also applied over real data. The application shows better performance than other algorithms in case of low frequencies haplotypes. In addition, the package o ers the chance of make analysis through a bayesian point of view, with all the strength this fact can suppose.

Studies of the role of tuberous sclerosis complex (TSC) proteins (TSC1/TSC2) in pathology have focused mainly on their capacity to regulate translation and cell growth, but their relation with alterations of cellular structures and the cell cycle is not yet fully understood. The transforming acidic coiled-coil (TACC) domain-containing proteins are central players in structures and processes involving the microtubule network. Here, TACC3 interactome mapping identified TSC2 and 15 other proteins, including TACC homo and heterodimers, and two evolutionary conserved interactors (ch-TOG/CKAP5 and FAM161B). TACC3 and TSC2 co-localize and co-purify with components of the nuclear envelope, and their deficiency causes morphological alterations of this structure. During cell division, TACC3 is necessary for the proper localization of phospho-Ser939 TSC2 at the spindle poles and cytokinetic bridges. Consistently, abscission alterations and increased frequency of binucleated cells were observed in Tacc3- and Tsc2-deficent cells relative to controls. In regulating cell division, TSC2 acts epistatic to TACC3 and, in addition to canonical TSC/mTOR signaling and cytokinetic associations, converges to the early mitotic checkpoint mediated by CHFR. Our findings link TACC3 to novel structural and cell division functions of TSC2, which may provide additional explanations of the clinical and pathological manifestations of lymphangioleiomyomatosis disease and TSC syndrome, including the greater clinical severity associated with TSC2 germline mutations relative to TSC1., "ESTUDI DE LA CONNEXIÓ ENTRE EL CENTROSOMA, L'ESCLEROSI TUBEROSA I LA DIFERENCIACIÓ EPITELIAL"TEXT:La major part dels estudis realitzats per tal d'esbrinar les funcions de les proteïnes TSC1 i TSC2 causants de l'Esclerosi Tuberosa (TSC) s'han centrat en la seva capacitat de regular la transcripció i el creixement cel·lular, però la seva relació amb alteracions d'estructures cel·lulars i el cicle cel·lular no estan ben determinades. Les proteïnes TACC tenen un paper primordial en estructures i processos relacionats amb les xarxes de microtúbuls (MTs). En aquest estudi hem realitzat un mapatge d'interaccions centrat en TACC3 i hem identificat TSC2 i 15 proteïnes més, com són els homo- i heterodímers de TACC, i dues proteïnes (ch-TOG/CKAP5 i FAM161B), la interacció de les quals està conservada entre espècies. TACC3 i TSC2 co-localitzen i co-purifiquen amb components de l'embolcall nuclear i la seva deficiència dóna lloc a alteracions d'aquesta estructura. Al llarg de la divisió cel·lular, TACC3 és necessària per a la localització de phospho-Ser939-TSC2 als pols del fus mitòtic i al punt de divisió de citocinesi. En concordança, cèl·lules deficients en Tacc3 i Tsc2 presenten un retràs en la citocinesi i una elevada freqüència de cèl·lules binucleades. Pel que fa a la regulació del cicle cel·lular, TSC2 actua de forma epistàtica sobre TACC3 i, a més de la seva funció de senyalització en la via TSC/mTOR i les associacions amb la citocinesi, convergeix amb el punt de regulació temprà de mitosi mediat per CHFR. Els nostres resultats relacionen TACC3 amb noves funcions estructurals i de divisió cel·lular per part de TSC2, la qual cosa podria proporcionar noves explicacions a les manifestacions clíniques i patològiques de la limfangioleiomiomatosi (LAM) i TSC.

Durant la dècada dels 80 la incidència de i la mortalitat per càncer a Tarragona eren inferiors a la de la majoria d'altres àrees europees i presentaven una tendència ascendent en ambdós casos. La supervivència dels malalts amb càncer era desconeguda. La implantació del Registre de Càncer de Tarragona va permetre conèixer aquestes dades inicials i plantejar com a objectiu de recerca el verificar si la incidència i la mortalitat per càncer als voltants de l'any 2000 seria superior a la de finals dels anys 80, si el paper dels factors de risc en l'increment dels casos incidents seria més important que el dels factors demogràfics i si la supervivència seria similar a la de la mitjana europea.La metodologia emprada consisteix en analitzar l'estudi de la cohort de residents de la província de Tarragona respecte a la incidència, la mortalitat i la seva evolució temporal. També s'analitza, per la cohort de persones amb càncer, la supervivència i la seva evolució temporal. Finalment es realitza un estudi de la prevalença a 31 de desembre de 2001. En el període 1998-2001 la taxa d'incidència ajustada a la població mundial estàndard fou de 310 en els homes i 203 en les dones. La probabilitat de desenvolupar un càncer fins als 85 anys d'edat és aproximadament d'un de cada dos en els homes i d'una de cada tres en les dones. Els tipus tumorals més freqüents són els de pròstata, pulmó i bufeta urinària en els homes i mama, còlon i cos uterí en les dones. En el mateix període la taxa ajustada de mortalitat per càncer fou de 150 en els homes i 75 en les dones. El risc de morir per càncer abans dels 85 anys era de 28% en els homes i 15% en les dones. A nivell europeu, la incidència de Tarragona se situava a la zona mitjana-alta en els homes i mitjana-baixa en les dones. Els percentatges anuals de canvi de la taxa ajustada d'incidència entre 1982 i 2001 foren de 2,4% en els homes i 1,7% en les dones. Els factors de risc foren els responsables del 49% de l'increment en el nombre de casos anuals en els homes i del 36% en les dones. La taxa de mortalitat va augmentar fins 1988 i es va mantenir estable fins 2001. La supervivència relativa als cinc anys dels pacients diagnosticats en el període 1995-1999 fou de 47,5% en els homes i de 57,3% en les dones, i era pràcticament igual a la de la mitjana europea en els homes i lleugerament superior en les dones.Els resultats inclouen múltiples paràmetres dels grans indicadors estudiats (incidència, mortalitat, anys potencials de vida perduts, supervivència i prevalença) i de la seva evolució temporal. Per a cada tipus tumoral es presenten múltiples taules i figures sobre aquests indicadors així com una comparació internacional de la incidència i la supervivència. Es presenten també dades sobre la incidència, la mortalitat i la supervivència del càncer infantil i del càncer en l'ancià. Finalment es quantifiquen els càncers múltiples del període 1980-2001 i el tipus dels segons càncers en relació al dels primers. En els annexos s'inclouen 29 taules amb informació detallada sobre la incidència y la mortalitat del període 1998-2001 per tipus tumoral, grup d'edat i sexe, sobre l'evolució de la incidència i la mortalitat per biennis i sobre la incidència i la mortalitat per regió sanitària i comarca.El Registre de Càncer de Tarragona és l'eina ideal pel monitoratge de la magnitud del càncer i de l'impacte de múltiples activitats de control d'aquesta malaltia en la població de la província de Tarragona., During the eighties cancer incidence and mortality in Tarragona were lower than in most of European areas but both showed an increasing trend. At that time survival of cancer patients was unknown. The implantation of the Tarragona Cancer Registry allowed to knowledge these initial data and to define as objective of research to verify if cancer incidence and mortality would be higher around 2000 than at the end of the eighties, if the role of risk factors in the increase of the number of incidents cases would be more important than the role of demographic factors and, finally, if survival rates would be similar to the European average.Methodology used consisted in analysing the Tarragona residents cohort study, in relation to the incidence, the mortality and their temporal evolution. Analysis is also performed on the cancer patients cohort for survival and its temporal evolution. Finally the study of the 31st December 2001 prevalence is performed.In the period 1998-2001, the age world-standardized incidence rate was 310 in males and 203 in females. The probability of developing a cancer before 85 years of age is of one out of two in males and one out of three in females. The most frequent tumoral types are prostate, lung and urinary bladder in males and breast, colon and corpus uteri in females. During the same period, the age-standardized mortality rate of cancer was 150 in males and 75 in females. The risk of death by cancer before 85 years old was of 28% in males and 15% in females. At European level, the incidence in Tarragona was in the medium-high level in males and in medium-low level in females. The annual percentages of change of the standardized incidence rate between 1982 and 2001 were 2.4% in males and 1.7% in females. Risk factors were responsible for 49% of the increase in the number of annual cases in men and 36% in females. Mortality rate increased until 1988 and maintained stable from this year until 2001. Five-year relative survival of the patients diagnosed in the period 1995-1999 were 47,5% in men and 57,3% in females, and were practically equal to the European average in men and lightly higher in women.The results include multiple parameters of the major indicators studied (incidence, mortality, potential lost years of life, survival, and prevalence, and their temporal evolution). A variety of tables and figures show these indicators as well as an international comparison of incidence and survival for each type or group of cancers. Data on incidence, mortality and survival for childhood cancer (0 to 14 years old) and elderly cancer (75 and more years old) are presented. Finally, multiple primary cancers (second, third and fourth ones) of the period 1980-2001 and the type of the second cancer in relation to the first one are quantified. The annexes include 29 tables with detailed information, classified by type of tumour, age group and sex, concerning incidence and mortality of the period 1998-2001, the biennial evolution of incidence and mortality and incidence and mortality classified by Health Regions and County.The Tarragona Cancer Registry is the ideal tool to monitor the magnitude of cancer and the impact of multiple cancer control activities in the Tarragona province population.

En los últimos 20-30 años se ha descrito una aumento significativo de la incidencia de las neoplasias linfoides, si bien las causas de dicho incremento no son del todo conocidas, parece que la mejora diagnóstica de dichas enfermedades y la epidemia del SIDA han contribuido en parte a la creciente incidencia reportada recientemente, pero en la gran mayoría de los casos los factores de riesgo son aún desconocidos. Los objetivos de este trabajo fueron estimar la incidencia poblacional de las neoplasias linfoides según el sexo y la edad de las diferentes entidades histopatológicas teniendo en cuenta la última clasificación de la OMS para las enfermedades hematológicas; analizar su distribución geográfica de la incidencia; estimar la supervivencia observada y relativa y calcular la incidencia atribuible poblacional del SIDA en la incidencia esta patología.Desde Enero de 1994 a Diciembre de 2001, se registraron 1.288 pacientes con el diagnóstico de neoplasia linfoide en la población cubierta por el registro de cáncer poblacional de Girona. Se revisaron los diagnósticos anatomopatológicos y hematológicos para la reclasificación retrospectiva de los diagnósticos según la última clasificación de la OMS para las enfermedades hematológicas.La distribución de las neoplasias linfoides fue: Neoplasias de células B( 77,3%), neoplasias de células T/NK (5,9%), Linfomas de Hodgkin (8,7%) y el 8,2% de los casos no se pudieron reclasificar. La tasa bruta de incidencia de las neoplasias linfoides fue de 35,8 nuevos casos por 100.000 hombres y año y de 25,7 nuevos casos por 100.000 mujeres y año. En los niños (< 15 años) fueron la leucemia/linfoma linfoblástico de células B precursoras (65%) y linfomas de Hodgkin (20%) las entidades más frecuentes mientras que mielomas (17,8%), linfomas B difuso de células grandes (13,5%), leucemia linfática crónica / linfoma linfocítico de células pequeñas (13,3%) y los linfomas foliculares las más frecuente en los adultos (9,7%). La distribución geográfica de la incidencia de los linfomas en la Región Sanitaria Girona (RSGi), es diferente según la variante histológica. En la actualidad se estima que en la RSGi existe una incidencia de las neoplasias linfoides en el sexo masculino superior al resto del estado español, pero hallándose cercana a la media europea y mundial. La supervivencia relativa a los 5 años de todas las neoplasias linfoides fue del 52,5% (IC95% 49,4-55,8). Los subtipos histológicos con mejor supervivencia relativa a los 5 años son los linfomas de la zona marginal extranodal (MALT) [SR5%: 77,6], Linfomas de Hodgkin [SR5%: 75,4]; Leucemia linfática crónica/linfoma linfocítico de células pequeñas [SR5%: 73,5], neoplasias de células T y NK [SR5%: 64,4]. Las de peor pronóstico son los Leucemia/Linfoma de Burkitt [SR5%: 23,2], mielomas/plasmocitoma [SR5%: 29,6], los linfomas B difuso de células grandes y Leucemia/Linfoma linfoblástico de células B precursoras [SR5%: 40,0].Según el sexo, la supervivencia relativa a los 5 años de las neoplasias linfoides es superior en las mujeres que los hombres, estas diferencias son significativas en el caso de los linfomas de Hodgkin, linfomas de células T/NK y linfomas B difuso de células grandes. La única variante en que la supervivencia es superior significativamente en los hombres, es en la variante leucemias-linfomas linfoblásticos de células B presursoras. Existe un riesgo aumentado de padecer linfomas en los pacientes con SIDA, este riesgo es diferente según el tipo de linfoma, sexo, vía de exposición al VIH y periodo de tiempo en relación al diagnóstico del SIDA. El meta-análisis realizado en este estudio pone evidencia la magnitud del riesgo y la homogeneidad en diferentes territorios. La variante histológica más frecuente en estos pacientes fue el linfoma B difuso de células grandes seguido del linfoma de Burkitt. La fracción atribuible del SIDA en la incidencia poblacional de los linfomas no Hodgkin fue del 8,7% y del 2,7% para los linfomas de Hodgkin. Esta fracción fue superior en los hombres. Se constata un descenso de la fracción atribuible poblacional del SIDA en la incidencia de los LNH a partir de la introducción de la Terapia antiretroviral de gran actividad (TARGA)., In the last 20-30 years has been described a significant incidence increased of the lymphoid neoplasms, the causes of this increment are not known, seems that the diagnosis improvement and the AIDS epidemic has contributed in part in the increased incidence reported lately, but in the majority of cases the risk factors are still unknown.The objectives were to assess the distribution of the lymphoid neoplasms and their histological subtypes in accordance with the World Health Organization (WHO) classification by calculating their incidence rates in our area; to estimate the observed and relative survival and calculate the populational attributable incidence to AIDS.From January 1994 to December 2001, 1.288 patients diagnosed with lymphoid neoplasm were recruited in the population- based Cancer Registry of Girona. Both former pathological and hematological diagnoses were reviewed and some prospectively reclassified following the latest WHO classification.Following criteria established by WHO classification the distribution of lymphoid neoplasms was as follows: 77.3% B-cell neoplasm, 5.9% T-cell neoplasm, 8.7% Hodgkin lymphoma and 8,2% was unclassifiable. From 1994 to 2001 the lymphoid neoplasm crude incidence rates was 35.8 per 100.000 men-year, while it was 25.7 new cases per 100.000 women-year. In children (In Girona Helath Region the geographical incidence pattern of lymphomas was different according to histological subtypes.Nowadays, a higher incidence rate of lymphoid neoplasm was found in men in our area compared with other geographical areas in Spain, which could suggest a faster approximation to the pattern observed in industrialized societies. The cause of this geographical distribution is unknown.5 years relative survival rates of lymphoid neoplasms was 52.5% (CI95% 49.4-55.8). The histological subtypes with a good prognosis were marginal zone-cell lymphoma (MALT) [SR5%: 77.6], Hodgkin lymphoma [SR5%: 75.4]; B-cell chronic lymphocytic leukemia/small lymphocytic lymphoma [SR5%: 73.5]; T and NK-cell lymphoid neoplasms [SR5%: 64.4]. The lymphoid neoplasms with the worst prognosis were Burkitt lymphoma/ leukemia [SR5%: 23.2]; plasma cell myeloma [SR5%: 29.6], diffuse large B-cell lymphoma and precursor B-lymphoblastic leukemia/lymphoma [SR5%: 40.0].By sex, the 5 years relative survival of the lymphoid neoplasms is higher in women than in men, these differences are statistically significant in case of Hodgkin lymphoma, T/NK-cell lymphomas and diffuse large B-cell lymphoma. The variant which has a significant best survival in men, was the precursor B-lymphoblastic leukemia/lymphoma.There is an high risk to devolop lymphomas in AIDS patients, this risk is different according to the type of lymphoma, sex, HIV exposure category and period in relation to AIDS diagnosis. The meta-analysis done in thi study show the risk magnitude and the homogeneity in differents territories. The most frequent histological subtype in these patients were the diffuse large B-cell lymphoma followed by the Burkitt lymphoma. The fraction attributable to AIDS in the populational incidence of non-Hodgkin lymphoma was 8.7% and 2.7% for a Hodgkin lymphoma. This fraction was higher in men. A decreased in the fraction attributable to AIDS in non-Hodgkin lymphoma incidence is confirmed after the introduction of HAART.

At the beginning of the 21st century, primary liver cancer (PLC) remains the fifth most common malignancy in men worldwide, and the eighth in women. Central Africa and South East of Asia are high risk geographic areas for PLC, whereas developed countries appear to be generally low risk. Infections with hepatitis B (HBV) and C (HCV) viruses are the main risk factors for PLC, accounting for well over 80% of PLC cases detected worldwide. The recently detected increase in both incidence and mortality by PLC in developed countries is strongly related to these viral infections. The evaluation of PLC time trends needs to take into consideration the geographic distribution and effect of these viruses. This thesis presents three studies which the aim to describe PLC incidence and mortality issues in different geographic areas, each addressing several epidemiological and methodological issues. For each study, different statistical methods on the basis of the Bayesian inference have been proposed, evaluated and discussed in order to cope with extra-Poisson variability. The first study, entitled "Meta-analysis of cohort studies of risk of liver cancer death among HBV carriers", evaluates the variability in PLC mortality reported in 11 cohort studies of male HBV carriers, taking into consideration the effects of geographic area and the choice of the general population versus a more comparable group such as HBV-negative workers or blood donors as the comparison group. The statistical methods of this study focuses on mixtures of Poisson distributions. The "stickbreaking" method has been used to estimate the number of components of the mixture of Poisson distributions, and, thus to obtain a pooled relative risk (RR) of death for PLC among male HBV carriers. The pooled RR of death by PLC related to HBV infection was 23.5 (95% Credibility Interval (CRI): 14.9 - 44.5). Studies carried out in high risk areas for PLC (China and Taiwan) showed RRs 2 to 5-fold higher than those of studies carried out in Europe, Japan and the U.S.. In low risk areas for PLC, studies which used workers or blood donors as comparison groups had RRs 1.9-fold higher (95% CRI: 1.2 - 3.1) than studies which used the general population. However, in high risk areas, the ratio of RRs was 5.3-fold (95% CRI: 3.4 - 7.9). This is the first time that a "healthy donor effect" has been quantified in longitudinal studies. The second study, entitled "Geographic distribution of primary liver cancer in Europe in 2002" evaluates the effect of HBV and HCV seroprevalence in 38 European countries on PLC incidence and mortality. Mixed Poisson models based on Bayesian inference have been used to smooth Standardized Incidence (SIR) and Mortality (SMR) ratios for PLC accounting for the effect of HBV and HCV prevalences. This approach enabled us to both examine the effect of different levels of HBV and HCV, and to identify remaining variability in PLC after accounting for infection rates. Bayesian inference allowed the determination of posterior probabilities for the somoothed SIRs and SMRs (hereafter RRs). The Deviance Information Criterion (DIC) and the "effective number of parameters" (pD) have been used as tools for model choice. The highest mortality and incidence PLC RRs were found in Southern European countries (RR range 0.9-2.4), whereas Northern European countries showed the lowest RRs (RR range: 0.3-0.9). The effect of HBV infection was not found to be statistically significant in the model which accounted for both HBV and HCV prevalence. Countries with a prevalence of HCV higher than 2% (e.g.: Italy and Spain) had a higher risk of incidence and mortality (RR range: 1.28 - 1.78) than countries with HCV prevalence below 1%. Thus, the high risk of PLC detected in Southern Europe appears to be explained, in part, by HCV infection. The high HCV seroprevalence in this area could be associated with exposure 30-50 years ago. There may be an underestimation of PLC incidence and mortality rates in Eastern European countries given the low PLC RRs reported, despite high HBV and HCV seroprevalences observed. The implementation of population-based cancer registries in Eastern European countries is warranted, as well as HCV prevalence studies across Europe, to better determine the distribution of PLC in Europe and its relationship with that virus. The last study, entitled "Time trends in liver disease in Spain during the period 198397", describes incidence and mortality trends in hepatocellular carcinoma and cholangiocarcinoma as well as mortality trends in liver cirrhosis in Spain. Autoregressive age-period-cohort (APC) models have been used to evaluate the time trends. We found that APC models performed well for those liver diseases with large number of cases, whereas the age-period models did for those liver diseases with low number of cases. We found an increase in incidence and mortality of hepatocellular carcinoma in Spain (annual percent change (APCH) in men's incidence: 6.6%, 95% CRI: 5.8, 8.1: APCH in women's incidence: 4.5%, 95% CRI: 1.4%, 7.3%; APCH in men's mortality: 6.8%, 95% CRI: 5.8%, 8.1%; APCH in women's mortality: 5.1%, 95% CRI: 3.5%, 6.3%), that appear to be related to HCV exposure 30 years ago, as described in other studies of PLC. We also found an increasing trend in cholangiocarcinoma mortality (APCH in men: 17.1%, 95% CRI: 13.5%, 21.2%; APCH in women: 15.0%, 95% CRI: 11.5%, 19.5%) similar to that found in some developed countries, that could be attributed to improvement in diagnosis resulting from better imaging and diagnostic techniques. However, we did not detect a significant increasing trend in cholangiocarcinoma incidence, perhaps due to the low number of cases reported by the Spanish cancer registries. We have observed a decreasing trend in cirrhosis mortality in both sexes during the study period (APCH in men: -3.1%, 95% CRI: -5.1, -1.9%; APCH in women: -2.9%; 95% CRI: -6.2%, -1.3%), although younger cohorts did not show this pattern. This cohort effect suggests the possibility that younger cohorts could be exposed to some additional risk factors besides alcohol consumption. HIV and HCV or HBV co-infection and intravenous drug addiction could explain the increase in liver cirrhosis mortality among younger cohorts. The flexibility of the Bayesian approach allowed us to cope with extra-Poisson variability in three statistical analyses, applying different models, and addressing relevant methodological aspects specific to each problem. Challenging statistical issues in the framework of Bayesian applied modelling are: i) the selection of prior distributions for model parameters, which is related to convergence of the model; and ii) model selection procedures, and these remain important considerations for future research.

Se realizó un estudio prospectivo de casos y controles de base hospitalaria para estimar el riesgo de padecer cáncer colorrectal en función de diversos factores ambientales, entre los que se realizó un énfasis especial en la exposición a medicamentos, y para explorar la posible interacción de los factores ambientales entre sí y con determinadas alteraciones genéticas. De un total de 436 casos de cáncer colorrectal y 416 controles estudiados, 345 (79.13%) casos y 346 (83.17%) controles refirieron haber utilizado medicamentos con anterioridad a su ingreso y de forma continua durante 6 o más meses. De estos 691 sujetos se recogieron datos sobre el uso de un total de 2547 medicamentos (1376 entre los casos y 1171 entre los controles). Se codificaron los medicamentos para describir la frecuencia de uso de los diversos tipos de fármacos y grupos terapéuticos. A partir de los datos se exploró la asociación entre la presencia de cáncer y el uso de los distintos tipos de fármacos, agrupados según los niveles descritos, considerando la duración y la secuencia temporal de la exposición respecto al cáncer, las características clínicas y genéticas de los tumores y ajustando los riesgos por edad y sexo. Se observaron asociaciones significativas entre el uso crónico de laxantes, antianémicos e hipouricemiantes y un incremento de riesgo de cáncer colorrectal, así como asociaciones entre un menor riesgo de cáncer colorrectal y el uso crónico de antiinflamatorios no esteroides, fármacos para el tratamiento de enfermedades cardiovasculares e insulina. Varias de las observaciones estudiadas sugieren mecanismos carcinogénicos comunes en los que de una forma u otra se implica la desregulación de la insulina y/o la alteración de factores de crecimiento relacionados con la insulina. La confluencia de varias observaciones en un mismo mecanismo probablemente está reflejando que las principales causas de morbimortalidad en la cultura occidental son de origen ambiental y por lo tanto modificables en una gran proporción., A prospective case-control hospital-based study was conducted to estimate the risk of colorectal cancer as a function of several enviromental factors, including drug exposure. An special attention was paid to drug use, and also to explore any interaction between drugs and enviromental factors and/or some genetical abnormalities. Overall, 436 colorectal cancer cases and 416 controls were studied, of which 345 cases (79.13%) and 346 controls (83.17%) referred the use of at least one medication continuously for at least 6 months prior to admission. Data on a total of 2547 exposures (1376 from case subjects and 1171 from control subjects) was collected from these 691 subjects. Medications were coded in order to describe frequencies of exposure to different drug types and therapeutic groups. The association between exposures and disease was explored to assess colorectal cancer risk, using drug exposure data, adjusting by age and sex, considering the lenght of the exposure and the temporal sequence with the cancer, as well as some clinical anf genetical features of the disease. Significant associations were observed suggesting an increased risk amongst chronic users of laxatives, of iron supplements and of treatments for hyperuricaemia, as well as reduced risk estimates for patients with chronic exposure to non-steroidal anti-inflammatory drugs, cardiovascular drugs and insulin.Several of the observed associations suggested common carcinogenic mechanisms involving to different degrees features of insulin disregulation and/or affecting insulin-related growth factors. Confluence of several observations into a common mechanism were probably reflecting that the main causes for morbimortality in occidental cultures can be attributed to enviromental factors, and thus could be potentially modifiable to a wide extent.

Justificació: encara que la pneumònia adquirida a la comunitat (PAC) és una malaltia clàssica i àmpliament estudiada, avui dia continua essent una important causa de morbimortalitat i es continua tenint un coneixement limitat dels factors que l'afavoreixen. La PAC es pot considerar una malaltia d'alta incidència, té un maneig clínic controvertit, una mortalitat important (especialment en els vells) i un cost elevat. Si a això s'hi afegeix l'aparició de resistències bacterianes als antibiòtics sembla lògic que en l'abordatge d'aquesta malaltia cada cop prengui més força la perspectiva de la prevenció.Objectius: conèixer els factors de risc d'adquirir una PAC, a la població major de 14 anys, relacionats amb els hàbits tòxics o estils de vida modificables, les condicions ambientals, laborals i de l'ahitatge, així com conèixer l'efecte de la vacuna antipneumocòccica en la prevenció de la PAC en la població major de 14 anys.Metodologia: per donar resposta a aquests objectius es va dissenyar un estudi observacional de tipus cas-control, de base poblacional, multicèntric i coordinat amb una base poblacional de 860.000 persones majors de 14 anys. Aquesta era la població necessària per, en un any, detectar 1.500 casos de PAC. Van participar 64 centres d'atenció primària de Catalunya, Andorra, València i les Illes Balears. Tots els casos de PAC diagnosticats a la població d'estudi durant el període novembre 1999 - novembre 2000 segons uns estrictes criteris clínics, radiològics i evolutius van ser inclosos. Per cada cas es va seleccionar un control aparellat per edat, sexe i centre d'atenció primària segons un procés de mostreig aleatori a partir del padró municipal. A tots ells, se'ls va administrar un extens qüestionari de factors de risc de PAC que feia referència al consum de tabac, alcohol, exposicions laborals, condicions de vida i de l'habitatge, vacunes, així com antecedents mèdics i comorbiditats.Resultats: Durant el període d'estudi es van identificar i reclutar 1.336 casos de PAC (participació 97 %) i 1.326 controls (participació 63 %), amb una edat mitjana de 57 anys i un lleuger predomini del sexe masculí (53 %). La incidència acumulada observada va ser de 1,6 casos/1.000 habitants any amb una clara relació estacional, de manera que la incidència el mes de gener era casi sis vegades superior a la dels mesos de juliol i agost. Pel que fa a la comorbiditat es va observar una associació estadísticament significativa entre la PAC i la infecció respiratòria de vies altes durant l'últim mes, la insuficiència cardíaca, la bronquitis crònica i el càncer, però no amb la diabetis, l'accident vascular cerebral ni la insuficiència renal crònica. Es va observar una associació estadísticament significativa entre el consum de tabac i la PAC independent de les esmentades comorbiditats, així com una relació dosi resposta entre la PAC i la intensitat de l'hàbit, la seva durada i la quantitat total de paquets fumats. Pel que fa al consum passiu de tabac es va observar una important interacció entre aquesta exposició i l'edat, de manera que en les persones de menys de 60 anys el consum passiu de tabac no tenia cap efecte (OR=1,0), mentre que en les de 60 anys o més s'observa un efecte significatiu (OR=1,52) amb una fracció atribuïble poblacional del 10%. En relació al consum d'alcohol es va observar un efecte significatiu només en els homes i quan el consum era superior als 80 g/dia (OR=2,34) sense que s'observi un increment del risc en consums inferiors encara que es facin diàriament. Altres variables associades a la PAC van ser l'exposició laboral a pols (OR=1,65), animals, vísceres o excrements (OR=1,78) i canvis sobtats de temperatura (OR=3,27), conviure més de 10 persones a la llar (OR=2,20), conviure o treballar amb nens de menys de 15 anys (OR=1,48) o tenir animals domèstics (OR=1,37). Les anàlisis multivariades en les que s'ajusta l'efecte de les vacunes per la comorbiditat mostren un efecte protector i estadísticament significatiu tant de la vacuna de la grip (OR=0,76) com de la vacuna antipneumocòccica (OR=0,55).Conclusions: Aquest estudi identifica diversos factors de risc modificables de PAC, el que ha de permetre dissenyar mesures preventives encaminades a disminuir la seva incidència. A més, aporta nova evidència científica sobre l'efectivitat de la vacuna de la grip i la antipneumocòccica en la prevenció de la PAC., Context: Although community-acquired pneumonia (CAP) is a well-known and widely studied disease, morbimortality is high and knowledge of the factors that favor CAP limited. CAP is a disease with high incidence, cost and mortality (especially among the elderly) and disputed clinical management. If to all this one adds antibiotic-resistant bacteria it is hardly surprising that the prevention perspective of the disease is gaining importance.Objectives: to determine the risk factors of acquiring CAP that are related to toxic habits or modifiable lifestyles, environmental, working and living conditions, as well as the effect of the antipneumococcal vaccine in preventing CAP, in a population of over 14 years of age.Methods: A population-based, multicenter, coordinated, observational, case-control study was designed with a population base of 860,000 people of over 14 years of age. This was the population needed to detect 1,500 cases of CAP in one year. Sixty-four primary health care clinics in Catalonia, Andorra, Valencia and the Balearic Islands participated in the study. All cases of CAP, diagnosed according to strict clinical, radiological and evolutional criteria between November 1999 and November 2000, were included in the study. A control matched by age, sex and primary health care clinic was selected for each case, following a random sampling process from the town census. All participants were given a detailed questionnaire on the risk factors of CAP which covered tobacco and alcohol use, occupational exposures, living and housing conditions, vaccines, and medical history and comorbidities. Results: During the study period, 1,336 cases of CAP were identified and enrolled (97% participation) and 1,326 controls were selected (63% participation) with an average age of 57 years and a slight predominance of men (53%). The accumulated incidence observed was of 1.6 cases/1,000 inhabitants/year with a marked seasonal relation reflected by the incidence in January being nearly 6 times greater than the incidence in July or August. With regard to comorbidity, a significant association was found between CAP and upper respiratory infections within the previous month, cardiac insufficiency, chronic bronchitis, and cancer, but not with diabetes, stroke, or chronic renal insufficiency. A significant association was found between smoking and CAP, independent of the above-mentioned comorbidities, and a dose-response relation between CAP and the intensity and duration of the smoking habit and the total number of packets smoked. With regard to passive smoking, a clear interaction between exposure and age was observed, there being no effect on people under 60 years of age (OR=1.0) but a significant effect was observed in people over 60 years (OR=1.52) with a population attributable fraction of 10%. Regarding alcohol consumption, a significant effect was observed for men only and when consumption was greater than 80 g/day (OR=2.34), no increased risk being observed for lower consumption even when daily. Other variables associated with CAP were occupational exposure to dust (OR=1.65), animals, offal, or excrements (OR=3.27) and sharp changes in temperature (OR=3.27), living with more than 10 people in the home (OR=2.20), living or working with children under 15 years of age (OR=1.48), or having pets (OR=1.37). Multivariate analysis where the effect of vaccines was adjusted for comorbidities showed a significant protector effect for both the influenza vaccine (OR=.76) and the antipneumococcal vaccine (OR=.55).Conclusions: This study identified several modifiable risk factors of CAP which should allow preventative measures to be designed to reduce incidence. New scientific evidence on the effectiveness of the influenza and antipneumococcal vaccines in the prevention of CAP has also been presented.

La paraula amiant s'aplica a una família de silicats fibrosos i inclou tres varietats principals: el crisotil·lè, la crocidolita i l'amosita. Totes les varietats d'amiant són carcinogèniques i poden produir càncer del pulmó i mesotelioma. El mesotelioma maligne és un tumor poc freqüent que es localitza sobre tot a la pleura. Té un pronòstic molt dolent, i el període de latència pot ser molt llarg, de 20 a 50 anys. L'exposició ocupacional a l'amiant és una causa reconeguda de mesotelioma; però un cop s'ha controlat l'exposició a concentracions molt elevades que es donaven a l'ambient laboral, la principal preocupació s'ha desplaçat cap els possibles efectes de l'exposició a dosis baixes. S'ha estudiat l'efecte de l'exposició domèstica que pateixen els familiars de treballadores exposats que porten la roba de feina a casa per la seva neteja. També s'ha estudiat l'efecte de l'exposició ambiental de les persones que viuen prop d'una font puntual d'emissió d'amiant. Malgrat tot, no s'havia analitzat, fins ara, si la contaminació ambiental o domèstica en la població general de zones industrials és una causa de mesotelioma.Es va dur a terme un estudi multicèntric de casos i controls de base poblacional en sis àrees d'Itàlia, Espanya i Suïssa. Es van incloure 215 casos nous de mesotelioma pleural amb confirmació histològica i 448 controls. Un grup d'higienistes industrials varen classificar els participants segons la seva probabilitat i intensitat d'exposició a l'amiant. La informació de la història laboral va ser la base de l'exposició ocupacional. Per la classificació de l'exposició ambiental i domèstica es van tenir en compte la utilització o presència d'amiant en cada domicili de la història de residències, així com les activitats industrials de l'entorn.En els 53 casos i 232 controls sense exposició ocupacional es va observar un increment de risc de mesotelioma associat a la utilització domèstica d'amiant o la presència de productes que poden desprendre fibres, així com el fet de viure a menys de 2 km d'un establiment industrial que utilitzava amiant. Aquestes circumstàncies impliquen exposicions a concentracions de 0.1 a 5 f/l, unes mil vegades menys que les mesurades en moltes àrees amb exposició ocupacional. En l'anàlisi dels 232 casos i 257 controls a Espanya es van identificar com ocupacions amb més risc de mesotelioma el treball en la fabricació de fibrociment, la neteja i planxa de roba, el treball com a mecànic electricista, i els conductors de vehicles i maquinària de la construcció. Gairebé el 88% dels casos de mesotelioma són atribuïbles a l'exposició a l'amiant, 62% d'origen ocupacional i 26% no ocupacional.L'amiant està prohibit actualment en la majoria de països industrialitzats, però encara s'utilitza en alguns països en desenvolupament. D'altra banda, atès el llarg període de latència, el mesotelioma pot ser un problema de salut durant les properes dècades. Cal doncs parar molta atenció a certs aspectes, com ara la problemàtica de l'amiant en edificis i instal·lacions, els efectes sobre la salut dels materials utilitzats en substitució de l'amiant i el seguiment i possible compensació dels treballadors que han estat exposats a l'amiant., El término amianto designa de forma genérica un grupo de silicatos fibrosos con tres variedades principales: crisotilo, crocidolita y amosita. Todas ellas son carcinogénicas, capaces de producir mesoteliomas y cáncer de pulmón. El mesotelioma maligno es un tumor poco frecuente que se localiza preferentemente en la pleura. Tiene muy mal pronóstico y un largo período de latencia, de entre 20 y 50 años. Existe una clara relación causal entre el mesotelioma y la exposición ocupacional al amianto. Sin embargo, tras el control de la exposición a concentraciones elevadas propias de algunas ocupaciones, la preocupación por los efectos del amianto se ha centrado en las exposiciones a dosis bajas. Se ha estudiado el efecto de la exposición doméstica de los familiares que lavan la ropa de trabajadores expuestos al amianto; también se ha estudiado el efecto de la exposición ambiental debida a la residencia cerca de una fuente puntual de emisión de amianto. Sin embargo, hasta ahora no se había analizado si la contaminación ambiental o doméstica de la población general en áreas industriales puede ser una causa de mesotelioma pleural.Se realizó un estudio multicéntrico de casos y controles de base poblacional en seis áreas de Italia, España y Suiza. Incluyó 215 casos nuevos de mesotelioma pleural con confirmación histológica y 448 controles. Un grupo de higienistas industriales evaluó la probabilidad e intensidad de la exposición al amianto de origen ocupacional y no ocupacional, a partir de la información de la historia laboral y de residencias. La exposición doméstica y ambiental se basó en la posible utilización o presencia de amianto en el domicilio, las características de la vivienda y las actividades industriales del entorno.En los 53 casos y 232 controles sin evidencia de exposición ocupacional se observó un incremento del riesgo de mesotelioma pleural asociado a la utilización de amianto en el domicilio o a la presencia de productos que pudiesen desprender fibras, así como en los residentes en un radio de 2 km respecto a alguna industria que utilizase amianto. Ello supone una exposición a concentraciones de 0.1 a 5 f/l, unas mil veces menores que las habituales en el ambiente laboral. Las principales ocupaciones de riesgo identificadas en el análisis de los 232 casos y 257 controles de los centros españoles fueron: fabricación de fibrocemento, trabajadores del lavado, limpieza y planchado de ropa, ajustadores y mecánicos electricistas, y conductores de maquinaria de la construcción. El 88% de los casos de mesotelioma son atribuibles al amianto, de los cuales 62% a la exposición ocupacional y 26% a la exposición doméstica o ambiental.El uso de amianto está prohibido casi en la totalidad de países industrializados, pero se utiliza aún en diversos países en vías de desarrollo. Incluso los países donde se ha prohibido el amianto deben prestar especial atención a otros aspectos de interés, tales como la problemática del amianto instalado, los posibles efectos de los materiales utilizados como sustitutos del amianto y el seguimiento de los trabajadores expuestos con vistas a la posible compensación a la que tienen derecho., Asbestos is a generic term applied to a group of fibrous silicates. There are three commercially important forms: chrysotile, crocidolite, and amosite. All of them are carcinogenic, capable of causing mesothelioma and lung cancer. Malignant mesothelioma is a rare tumour mainly located in the pleura. Most mesotheliomas are fatal within 12-24 months of diagnosis, and have a latency period of 20-50 years. There is strong evidence supporting the causal association between mesothelioma and occupational exposure to asbestos. Nevertheless, with the virtual cessation of high-dose occupational exposure to asbestos, public health attention has turned to the risks of exposure at lower doses arising from non-occupational sources. Domestic exposure results from asbestos fibres brought home by workers exposed in the workplace. Environmental exposure may result from residence in the vicinity of a single well-identified source of asbestos pollution. However, further studies are needed to investigate whether the industrial and domestic use of asbestos may produce sufficient environmental pollution to cause mesothelioma.A population-based case control study was carried out in six areas from Italy, Spain and Switzerland, including 215 new histologically confirmed cases of pleural mesothelioma and 448 controls. A panel of industrial hygienists assessed the probability and intensity of asbestos exposure separately for occupational, domestic and environmental sources. Each occupational period was classified based on the information collected in the occupational history. Classification of domestic and environmental exposure of each residence was based on the presence and use of asbestos at home, asbestos industrial activities in the surrounding area and their distance from the dwelling.Among 53 cases and 232 controls without evidence of occupational exposure to asbestos, handling asbestos material or presence of asbestos material at home susceptible to damage, as well as living within 2000 m of asbestos industries was associated with increased risk of mesothelioma. Typically these circumstances entail exposure to concentrations ranging from 0.1 to 5 f/l, about 1000 times lower than those measured in occupational settings. Among the 132 cases and 257 controls in the Spanish centres the main occupations with increased risk of mesothelioma were manufacture of asbestos cement, launderers, cleaners and pressers, electrical fitters, plumbers, and drivers of material-handling and related equipment. Almost 88% of mesotheliomas were attributable to asbestos exposure, 62% due to occupation and 26% from non-occupational origin.Utilisation of asbestos has almost completely ended in most developed countries as the result of government bans. Nevertheless, asbestos manufacture continues in parts of the developing world. Even countries that have banned the material still have to devise strategies to cope with the asbestos that remains in place, monitoring potential effects of fibres used as asbestos substitutes, and surveillance of former asbestos workers and exposed populations in order to provide them with access to compensation.

Introducción: Se propuso la elaboración del diseño estadístico para un ensayo clínico en el que se comparaba AAS con triflusal en la profilaxis secundaria en los 35 días post-infarto agudo de miocardio (IAM) -Triflusal en el Infarto de Miocardio (TIM)-. Con un diseño clásico se requería un tamaño muestral (N) de 4000 pacientes para detectar una diferencia de 2,5% con protección frente a los errores estadísticos de tipo I y II de 0,05 y 0,2 respectivamente. Se realiza una revisión de distintos diseños aplicables al estudio TIM. Objetivos: Principales: Búsqueda de un diseño teórico alternativo al diseño clásico. Comprobación del diseño mediante simulaciones. Realización del diseño, monitorización y análisis del estudio. Secundarios: Revisión de métodos alternativos a los diseños clásicos. Estudio del ahorro en tamaño muestral de los métodos secuenciales. Material y Método: Se realizó una primera selección de alternativas al diseño clásico en base a la posibilidad de finalización prematura bajo la hipótesis nula con los métodos desarrollados en el momento del diseño. Se simularon 10000 estudios para cada diseño y para distintas diferencias entre tratamientos con fortran 90. La precisión fue de 0,98% y de 0,78% para la estimación de los errores I (alfa=0,05) y II (beta=0,2) respectivamente. El ahorro en tamaño muestral fue estudiado con los percentiles 50 y 90 -P50(t) y P90(t)- de N y con el número de estudios que finalizaron con N respecto a un diseño clásico. Finalmente se implementó en el estudio TIM el modelo razón de probabilidades secuencial (RPST) truncando la región de continuación a 5305 pacientes. La primera inspección se programó tras 1000 inclusiones y las sucesivas cada 250, supeditadas a calendario de reuniones. Para el análisis de la variable principal se usó el estimador por máxima verosimilitud no sesgado con su intervalo de confianza (IC) al 95%, y las variables secundarias se ajustaron dada su correlación con la variable principal. Resultados: Algunos de los métodos secuenciales basados en límites continuos (MLC) se mostraron como las mejores alternativas al diseño clásico, (RPST, RPST truncado y triangular), con un ahorro -P50(t)- teórico de N de alrededor del 40%. Durante la fase de monitorización y de revisión tras finalización del estudio TIM, se realizaron cambios en la población de pacientes incluidos en la población de eficacia que no afectaron a la finalización del estudio. Los modelos basados en límites continuos estudiados, así como los métodos basados en la función de gasto de alfa, mostraron un idéntico comportamiento en cuanto al criterio de finalización. La corrección en la reducción del riesgo mediante el estimador no sesgado evitó el sesgo entre un 2,5% y 11% para las variables principal y secundarias. La revisión después de la finalización y previa al análisis estadístico no produjo cambios sustanciales en los resultados (-0,3% en reducción de riesgo). Conclusiones: Los MLC fueron la mejor alternativa al diseño clásico y presentaron un comportamiento de acuerdo a las especificaciones teóricas. La implementación del modelo RPST truncando a 5305 inclusiones permitió finalizar el estudio con un ahorro del 43%.

La información mutua (MI) es una medida general de asociación entre variables que detecta tanto relaciones lineales como no lineales. Esta medida puede ser muy útil en análisis genéticos, donde los genes pueden interactuar entre ellos de una manera no lineal. En este trabajo se lleva a cabo una estimación de la información mutua basada en métodos de kernel, empleándose un algoritmo que permite reducir el coste computacional. La significación estadística de los estimadores se evalúa a partir de un test de permutaciones, utilizando una aproximación en el caso de p-valores pequeños. Además, su error estándar es estimado mediante bootstrap. Se ha creado un paquete en R con esta metodología, aplicándose a datos reales de expresión génica, y se han comparado los resultados con los obtenidos utilizando el coeficiente de correlación de Spearman. Por último, se han explorado otras aplicaciones de la información mutua como la construcción de redes de asociación génica o su uso como distancia en la técnica de MultiDimensional Scaling.. La información mutua (IM) puede ser una medida de asociación interesante para evaluar asociación. Ha sido descrita inicialmente para variables categóricas pero puede extenderse a variables numéricas. En este caso es necesario calcular una doble integral que supone un reto si se desea analizar esta medida para un número importante de combinaciones de variables. El proyecto consistirá en evaluar las propiedades estadísticas de la IM, elaborar un software en R para su cálculo rápido, posiblemente con paralelización que permita el análisis de su error estándar por bootstrap y de la significación por un test de permutaciones. Finalmente se aplicará la IM al análisis de datos genéticos de expresión.