El asma y las alergias son enfermedades muy frecuentes en la población que se inician durante la infancia. Padecer asma se ha asociado con un riesgo reducido de algunos cánceres, entre ellos el cáncer de páncreas. Éste es uno de los tumores más agresivos, que está aumentando su frecuencia en nuestro entorno y para el que no existe ningún tratamiento eficaz ya que los pacientes se diagnostican demasiado tarde. Con este proyecto queremos estudiar el papel de otras enfermedades, como por ejemplo, la obesidad, la diabetes o las exposiciones como el tabaco, en la relación entre el asma/alergia y el cáncer de páncreas. En este proyecto, también queremos investigar si el tratamiento del asma/alergia afecta a esta asociación.
Resultados previos obtenidos en el marco del estudio PanGenEU, del que la Dra. Malats es investigadora principal indicaron que los sujetos con asma y alergia tuvieron un riesgo 40% menor de desarrollar cáncer de páncreas que individuos sin estas dos condiciones. No obstante, el papel de algunas exposiciones/factores tales como el tabaco, la obesidad, la diabetes, y, sobre todo, el tratamiento con antiinflamatorios/antihistamínicos, debe ser descartado.
Proponemos aplicar un enfoque multidisciplinario innovador para el estudio de A/A y PC, incluyendo métodos epidemiológicos, proporcionando también un marco para evaluar la asociación/interacción de las comorbilidades y los tratamientos con respecto al riesgo de CP. El proyecto producirá, por primera vez, un mapa completo que represente las complejas relaciones entre los endotipos A/A y el CP, así como una indicación sobre su relación temporal.

Aim: to assess the Predictive Power of Bidirectional Encoder Representations from Transformers (BERT) Models in Type 2 Diabetes mellitus (T2DM): Specific Cause Mortality associated with Adherence to Glucose Lowering Treatment
Methodology: Using data collected from the Information from primary care database (SIDIAP from Spain and CPRD from the UK) with linkage to specific cause mortality from death certificates, we will investigate how the BERT encoder performs in the analysis of sequences of non-fatal and fatal clinical events among individuals with T2DM in the period 2006 – 2018 in Relation to Adherence to Glucose Lowering Treatment (GLT). The cohort will include incident cases with T2DM identified within SIDIAP (exposed individuals). The model input will contain an initial health status vector composed of information on the patient at the beginning of the sequence of clinical events, e.g., sex, age, previous diagnosis and adherence to the GLT. The BERT-based encoder will be pre-trained to analyze the sequences of clinical records starting from an initial time point and fine-tuned to predict the desired outcome. The outcomes will be defined as occurrence of diabetes-related complications and specific cause mortality. Specific cause mortality rates will be estimated by age, sex, and adherence to GLT (proper or improper adherence). The model will be validated using CPRD.
Statistical analysis: Participant characteristics will be reported using mean and SD for the clinical and analytical variables, and percentages for categorical variables. Fine-tuned models will be trained using the negative log-likelihood of the joint distribution of the different outcomes hitting time considering the right censoring, and evaluated using the Area Under the receiver operating characteristic Curve (AUC). Welch Two Sample t-test, (a=0.05) will be used to test cross-validated model performances.
Relevance to diabetes: Our model can be easily fine-tuned for various clinical prediction tasks such as prediction of diabetes complications and specific cause mortality, which could help in the design of new prevention strategies and improvement of adherence to GLT.

Each year, an estimated 14,000 children and adolescents in Europe are diagnosed with cancer. Cancer treatments have improved so that most childhood cancer patients in Europe will become long-term survivors. However, around 20% of them cannot be cured and will die, many in hospital. Paediatric palliative care (PPC) encompasses care regardless of diagnosis or stage of disease, and focuses on the prevention and relief of suffering of paediatric patients and their informal caregivers. The value of PPC for children with life-threatening or life-limiting conditions is well-established in general, and for cancer specifically. Currently, care for children with cancer focuses primarily on disease treatment and symptom management, and specialised PPC is often not received. As a result, many children with cancer and their informal caregivers experience gaps in care, such as poor management of psychosocial and even physical symptoms, failures in communication, inadequate support for informal caregivers, and inconsistent bereavement follow-up.

In view of the known differences between European countries concerning the availability of PPC services, knowledge and experience in PPC and available resources, a Joint Action is the best way to facilitate the development of high quality PPC in all European countries, as countries with less knowledge or resources can take advantage of the experiences, knowledge and lessons learned in countries with more advanced PPC programmes. The overall aim of HOPE4Kids is to bring together relevant actors from across Member States and countries associated to the EU4Health programme to advance PPC in paediatric oncology. Activities across the action cover a Landscape Analysis of PPC in Europe, guideline development, pilots and education and training, as well as sustainability, focusing on six cross-cutting themes: Symptoms Treatment, Advance Care Planning, Shared-Decision Making, Psychosocial care, Pre-Loss and Bereavement Care and Models of Care.

Background
Climate change leading to extreme events represent pressing challenges for humanity. Short- term exposure
to ambient air pollution and temperature has been associated with exacerbations of chronic respiratory
diseases, yet significant knowledge gaps remain regarding specific diseases, such as interstitial lung diseases
and bronchiectasis, and potential adaptation strategies to decrease the risk of exacerbations.
Objectives
To investigate the short-term effects of air pollution and temperature, including their co- occurrence, on
chronic respiratory diseases exacerbations and sick leaves in Catalonia; to evaluate potential adaptation
strategies, including the role of inhaled corticosteroids, statins, SGLT-2 inhibitors, and vaccines in modifying
these effects; to incorporate patient perspectives to understand coping mechanisms against climate change
stressors and disease burden.
Methods
Using a well-characterised cohort with detailed environmental exposure data, we will analyze associations
between air pollution and temperature, and chronic respiratory diseases exacerbations, accounting for
hospitalisations, mortality, and sick leaves. We will apply advanced statistical modelling to disentangle
exposure-outcome relationships, assess effect modification, and integrate a qualitative research approach
to capture patient experiences.
Expected results
Identification of environmental triggers of chronic respiratory disease exacerbations, evaluation of
protective effects of pharmacological and behavioural adaptation measures, and generation of evidence to
inform public health policies and clinical guidelines.