Cancer registration in Europe originated at the national level in the first half of the 20th century in Denmark. Since then, the process has undergone considerable expansion both in geographic terms as well as in its overall scope (from the basic initial incidence indicators to survival and clinical outcomes). Cancer registration is a key component of ‘big data’ in oncology and has become an important instrument for aetiological and epidemiological research at the population level. In October 2016 the European Network of Cancer Registries (ENCR) and the European Commission’s Joint Research Centre (JRC) organized the 2016 ENCR Scientific Meeting and General Assembly with the theme: ‘Joining forces for better cancer registration in Europe’. Approximately 170 European cancer-registry representatives joined the event, contributing with 36 oral and 50 poster presentations. To mark the importance of this event, the JRC together with the ENCR decided to request publication of a special issue of the European Journal of Cancer Prevention (EJCP) focused on cancer registration in Europe, selecting some of the contributions of the scientific conference. This special issue presents a snapshot of the wide range of cancer registries’ activities in Europe and includes examples of the descriptive and analytical studies currently being carried out with cancer-registry data. The opening article focuses on the substantial and long-term effort of Germany in integrating its regional population-based cancer registries already in place to a nationwide implementation of clinical cancer registries (Holleczek and Katalinic, 2017). Within this strategy, half of the German states are already setting up comprehensive cancer registries integrating the tasks of both clinical and population-based cancer registries. Several methodological issues still remain to be resolved, including standards on data collection, processing and utilization to reach uniform top-quality standards. The EJCP special issue also includes the first evaluation of case-ascertainment completeness in the whole of Switzerland (Lorez et al., 2017). By applying two innovative approaches, the authors report satisfactory levels of completeness across all Swiss registries. A paper from Estonia (Paapsi et al., 2017) evaluates the impact of case under-reporting in childhood cancer estimates for incidence and survival in the calendar period 2000–2011. The estimated overall completeness of 89.5% is reflected in considerable underestimation for both incidence and survival measures, mostly because of the exclusion of nonfatal childhood cancer cases. Two other papers propose ad hoc statistical methods applied to cancer registration data. One addresses the validation of incidence rates in an Italian region not fully covered by registration (Nannavecchia et al., 2017). The method uses neighbouring incidence data, adjusting for hospitalization and mortality. The second estimates the population-based cancer-specific potential years of life lost in a cohort of cancer patients from Belgium and proposes a method applicable when accurate information on the cause of death is not available (Silversmit et al., 2017). The authors report on how the cancer-specific fraction of the potential years of life lost increases with increasing cancer-specific mortality and decreasing age at diagnosis. A number of papers report on cancer incidence, survival and end-outcome trends in different European geographic areas. Solans et al. (2017) describe the changes in the incidence and survival of Hodgkin lymphoma in Girona (Spain) over three decades, considering stage of diagnosis, histological subtype and presence of B-symptoms, that is, systemic symptoms of fever, nocturnal sweats and weight loss. They find constant Hodgkin lymphoma incidence throughout the period, and lower survival for patients with older age at diagnosis, higher clinical stage and presence of B-symptoms. Antunes et al. (2017) report on cancer survival trends for the first decade of the millennium in the northern region of Portugal, focusing on the 20 cancer sites with the highest incidence. Survival improvements are not consistent for all cancer sites; although a very good prognosis is confirmed for thyroid and prostate cancers, some cancer sites still report poor survival values. Suteu et al. (2017) show increasing cancer incidence trends over 14 years for skin melanoma and squamous cell carcinoma in Cluj County, Romania. They also analyse survival, finding improvements over the period, but low values in older ages, advanced stages and rural areas. Low registration completeness for squamous cell carcinoma may affect survival results, especially in stratified analyses. Jakab et al. (2017) consider trends of incidence and survival of childhood leukaemia over 45 years in Hungary, analysing the prognostic role of socioeconomic level of the different geographic areas. They observe a 1% annual increase in leukaemia incidence, mainly because of an increase in acute lymphoblastic leukaemia; they also report significant improvements in survival and an inverse association with deprivation level of the area of residence at diagnosis, especially for longer-term survival estimates. Zadnik et al. (2017) explore the time trends and spatial variations of mesothelioma incidence in Slovenia over the last 50 years. After an increment since 1970, mesothelioma incidence levels off after 2003. This peak was reached 30 years after the maximum value of the asbestos import curve. Spatial analysis shows mesothelioma clusters around known asbestos sources in the past, whereas more recent geographic distribution is more scattered. Kaceniene et al. (2017) report on the suicide risk among cancer patients in Lithuania over the period 1993–2012. An increment of suicide risk is shown compared with the general population, being three-fold higher for patients with advanced stages, and with the highest risk in the critical period shortly after diagnosis. This study underlines the unsatisfied clinical and psychological needs of patients, and the importance of multidisciplinary preventive interventions. Two papers integrate cancer-registry data with other sources, such as screening programme data and outdoor air pollution. Bastos et al. (2017) evaluate the effectiveness of population-based breast cancer screening in cancer mortality in the central region of Portugal. They estimate a more than 30% reduction in breast cancer mortality following the introduction of an organized screening programme. The association between several outdoor air pollutants and lung-cancer mortality is analysed at regional level in Crete, Greece (Sifaki-Pistolla et al., 2017). The authors find that high levels of analysed pollutants, especially particulate matter, are associated with a higher risk of lung-cancer mortality across geographic regions. Limited data on adjustment for smoking habits, including lack of information on duration and amount of cigarettes smoked, and absence of accounting for the latency period of lung cancer limit the interpretation of these provocative findings. The collection of detailed clinical information on patient and tumour characteristics, including stage, diagnostic exams and treatments, enables cancer registries to evaluate the cancer care patterns at the population level. Guevara et al. (2017) examine the treatment patterns and trends in the last decade for women with nonmetastatic breast cancer according to recent European guidelines in Navarra, Spain. The authors find increasing adherence to European guidelines over the last decade in this Spanish region, but still an inadequate adherence for elderly women. The last two articles address the participation of cancer patients in clinical trials. Of these, a study from the Netherlands evaluates the association between trial participation of long-term Hodgkin’s lymphoma survivors with health-related quality of life (Thong et al., 2017). The authors report nonsignificant differences in quality of life associated with the participation in clinical trials. A cohort study of cancer patients in Northern Ireland including over 50 000 patients (Donnelly et al., 2017) investigates population and disease factors associated with participation in clinical trials. The authors show unsatisfactory trial participation, especially among adult and elderly patients, but a more satisfactory one among children. The establishment of a regional cancer-trial network implied increase patient participation. The linkage between clinical trials and cancer-registry databases provides an effective instrument for monitoring changes in cancer treatment and clinical trial participation at the population level. The wide scope of relevant issues covered in this EJCP supplement, spanning from coverage and methodology of cancer registration to its implications and inferences on cancer causes, management and outcome, illustrates the critical role and importance of cancer registries over seven decades since their first establishment in Europe. The synergistic partnership of the ENCR and the European Commission since 2012 has given new impetus to the work of cancer registration in Europe. It has contributed with harmonization of data quality standards (Martos et al., 2014), the JRC-ENCR quality check software (Giusti et al., 2016) and the ENCR-JRC project on ‘Incidence and Mortality in Europe’ (http://www.encr.eu/index.php/activities/encr-jrc-project, accessed 31 August 2017), in collaboration with all European cancer registries and stakeholders that participated actively.
The use of palm oil by the food industry is increasingly criticized, especially in Italy, for its purported negative effects on human health and environment. This paper summarizes the conclusions of a Symposium on this topic, gathered by the Nutrition Foundation of Italy, among experts representing a number of Italian Medical and Nutritional Scientific Societies. Toxicological and environmental issues were not considered. Participants agreed that: no evidence does exist on the specific health effects of palm oil consumption as compared to other saturated fatty acids-rich fats; the stereospecific distribution of saturated fatty acids in the triacylglycerol molecule of palm oil limits their absorption rate and metabolic effects; in agreement with International guidelines, saturated fatty acids intake should be kept <10% of total energy, within a balanced diet; within these limits, no effect of palm oil consumption on human health (and specifically on CVD or cancer risk) can be foreseen.
Relevant references from the academic literature search. Results from the academic literature search are listed in the form of relevant publications. (DOCX 14 kb)
Read moreWe confirmed the lack of association between exposure to asbestos-free talc, lung cancer, and mesothelioma. Increased mortality from pneumoconiosis among miners is attributable to past exposure to silica.
Read moreLevel of education is a strong indicator of premature mortality. The magnitude of the association between educational level and mortality differs across sexes, marital status and causes of death.
Read moreBased on a large pooled analysis, we identified several occupations and related exposures that are associated with elevated odds of gastric cancer. These findings have potential implications for risk attenuation and could be used to direct investigations evaluating the impact of targeted gastric cancer prevention/early detection programmes based on occupation.
Read moreDiets high in glycemic index (GI) and glycemic load (GL) have been related to an increased risk of selected cancers, but additional quantification is required. We updated a systematic review and meta-analysis published in 2015 to May 2019 to provide quantitative information on GI/GL and cancer risk. Relative risks (RR) and the corresponding 95 % confidence intervals (CI) for the highest versus the lowest categories of GI and GL were extracted from selected studies and pooled using random-effects models. Twenty reports (>22,000 cancer cases) have become available after January 2015, and 15 were added to the meta-analyses by cancer sites, which considered a total of 88 investigations. The five additional reports were reviewed, but not included in the meta-analyses, since data were inadequate to be pooled. For hormone-related cancers, summary RRs for the highest versus lowest GI and GL intakes were moderately increased. They ranged from 1.04 (breast) to 1.12 (endometrium) for GI and from 1.03 (prostate) to 1.22 (ovary) for GL, of borderline significance. High GI was associated with small increased risks of colorectal (summary RR for GI: 1.20, 95% CI, 1.07-1.34-GL: 1.09, 95% CI, 0.97-1.22, 19 studies), bladder (GI: 1.25, 95% CI, 1.11-1.41-GL: 1.10, 95% CI, 0.85-1.42, four studies) and kidney cancers (GI: 1.16, 95% CI, 1.02-1.32-GL: 1.14, 95% CI, 0.81-1.60, five studies). GL was not significantly related to those cancer sites. Stomach, prostate and lung cancers were not associated with GI and GL. The present analysis, based on an updated comprehensive evaluation of the epidemiological literature, indicates moderate unfavorable effects of high versus low GI on colorectal, and possibly bladder and kidney cancers, and a possible moderate positive association between GL and endometrial cancer.
Read moreInfertility represents a very peculiar area of medicine. Contrary to other areas, where signs and symptoms lead to a diagnosis, which in turn leads to a specific treatment, in reproduction the lack of signs and symptoms for more than 12 months suggests the diagnosis of 'unexplained subfertility', and if this condition has lasted for some years, couples qualify for IVF. Diagnosis and treatments can extend over long periods of time (even years) and the accuracy of the diagnostic armamentarium is not optimal. Uncertainty about diagnosis and the need for significant perseverance is demanding on both couples and physicians, and actually constitute a very favourable situation for overdiagnosis ('unexplained subfertility') and overtreatment (IVF) on one hand, and, on the other, it may also affect compliance with treatments. To improve our capacity to properly handle this challenging situation, increased attention should be given to the duration of pregnancy seeking. Initiating treatments earlier in older women is unwise because this population has a lower fecundity and, therefore, duration of pregnancy seeking is even more important to achieve a reliable diagnosis of infertility. Moreover, if the infertility work-up is unremarkable, duration of pregnancy seeking should be extended up to more than 2 years prior to making a diagnosis of unexplained infertility regardless of age. An adequate period of pregnancy seeking is also required for couples who are diagnosed with conditions that can interfere with fertility to avoid overdiagnosis and overtreatment. Indeed, most causes of infertility will reduce but not impair natural conception. Within this sometimes long-term management, physicians should also pay attention to detrimental life habits in order to optimize the chances of both natural and assisted reproduction technology -mediated pregnancy. Even if interventional studies are not conclusive, it is advisable to address the problems of obesity and smoking. Focussing on frequency of sexual intercourse may be also beneficial for natural conception. Finally, there is the need for improving our capacity to handle compliance. Providing information on the importance of persevering at the start of treatment, promoting shared decision-making and tackling patient, clinic and treatment causes of drop-out can all improve the overall chances of parenthood. Thus, we plead for a wiser and more pragmatic approach to infertility, paying more attention to these neglected, but in our opinion essential, aspects of infertility care.
Read moreSpline functions, defined as piecewise polynomials with a fixed degree, whose joint points are called knots, are highly flexible tools to modeling non-linearity between a response and some continuous covariates. In epidemiological studies, the number and position of knots usually have an important meaning. Therefore, special attention should be posed to techniques that allow to choose the number and position of knots. Here, we will follow one of the most recent approaches to variable selection in a Bayesian context. Estimating the positions of the knots is not easy and, for a fixed degree, regression coefficients and locations of knots have to be estimated simultaneously, turning the estimation into a non-linear optimisation problem. \nThe aim of the present work is to: 1. introduce a two-step Bayesian procedure within the semiparametric generalised linear model framework, to be applied in epidemiological studies where the effect of a continuous exposure on risk is under investigation; 2. show how this framework is applied in a bivariate context, where the aim is to modeling the joint effect of intensity and duration of alcohol drinking in cancer of the oral cavity.
Read moreCarbohydrate foods with high glycaemic index (GI) and load (GL) may negatively influence cancer risk. We studied the association of dietary carbohydrates, GI, GL, intake of bread and pasta with risk of bladder cancer using data from an Italian case-control study. The study included 578 men and women with histologically confirmed bladder cancer and 608 controls admitted to the same hospitals as cases for acute, non-neoplastic conditions. OR were estimated by logistic regression models after allowance for relevant confounding factors. OR of bladder cancer for the highest v. the lowest quantile of intake were 1·52 (95 % CI 0·85, 2·69) for available carbohydrates, 1·18 (95 % CI 0·83, 1·67) for GI, 1·96 (95 % CI 1·16, 3·31, P trend<0·01) for GL, 1·58 (95 % CI 1·09, 2·29, P trend=0·03) for pasta and 1·92 (95 % CI 1·28, 2·86, P trend<0·01) for bread. OR for regular consumption of legumes and whole-grain products were 0·78 (95 % CI 0·60, 1·00) and 0·82 (95 % CI 0·63, 1·08), respectively. No heterogeneity in risks emerged across strata of sex. This case-control study showed that bladder cancer risk was directly associated with high dietary GL and with consumption of high quantity of refined carbohydrate foods, particularly bread. These associations were apparently stronger in subjects with low vegetable consumption.
Read moreThe aim of this study was to investigate the relation between bladder cancer risk and the use of selected drugs for cardiovascular disease (CVD) prevention, such as aspirin, statins, and calcium channel blockers (CCBs). We analyzed data from a multicentric case-control study carried out in Italy between 2003 and 2014, including 690 bladder cancer cases and 665 hospital controls. Odds ratios (ORs) of bladder cancer and corresponding 95% confidence intervals (CIs) were estimated using unconditional multiple logistic regression models. The ORs for bladder cancer were 1.21 (95% CI: 0.87-1.68) for regular use of aspirin, 0.72 (95% CI: 0.54-0.97) for use of any CCBs, and 1.32 (95% CI: 0.87-1.99) for use of any statins. A slight inverse association was found with duration of use of CCBs, whereas no consistent association was found with duration of use, age at first use, and frequency for aspirin and statin use, or with indication of use for aspirin (as an analgesic or, for CVD prevention). No significant association was found for various combinations of drugs or for all drugs combined (OR=1.23, 95% CI: 0.31-4.85). Our data indicate the lack of a relevant association between the use of selected drugs for CVD prevention and bladder cancer risk, although suggest a potential favorable role for CCBs.
Read more
