IDOTCOVID

International Database on Organ Donation and Transplantation - COVID19 

Concept and Hypothesis

Establishing treatment protocols in the current epidemic status of a novel disease is incompatible with awaiting for large multicentre retrospective studies that may be published in a later future. In order to provide evidence towards clinical decision making, transplant physicians need to be able to access a large database which enables them to compare outcomes using different strategies. 

HypothesisH1: In the current pandemic with the new SARS-CoV2 coronavirus, disease presentation and severity of the infection are higher in patients receiving SOT.
H2: Immunosuppressive treatment and management in SARS-CoV2 infected patients will influence the evolution and cure of the disease.
H3. A personalized decision support algorithm will enable to optimize outcomes at individual patient data

Objectives

The main objectives of this project are the development of a tool that helps in the management of immunosuppression and the treatment of COVID19 infection in SOT recipients.
To achieve this, the following objectives have been designed to be developed in stages and progressively:
   1. Creation of an international database that includes all SOT recipient patients with COVID19 infection (confirmed or suspected);
   2. Inclusion of different clinical and analytical data with recognized prognostic factor in the general population;
   3. Inclusion of treatment data, including management of immunosuppression and clinical outcomes;
  4. Development of a Decision Support Algorithm (DSA) that can assist the scientific community in updating their treatment management and immunosuppression protocols in this high-risk population;

Descision Support Algorithm (DSA)

​The DSA to be developed aims at aiding clinicians in their decision regarding treatment alternatives at the individual level. In summary, the Algorithm will be developed and trained based on the already available data on the database, with focus on pre-admission (i.e. recipient demographics, transplant type) and at admission (i.e. symptoms and biochemical values) data to explore and predict outcomes.  The DSA will be developed in two phases (see methdology).
Once trained, upon introduction of a new patient on the database, expected outcomes according to the different therapeutic approaches will be provided to clinicians (i.e. expected outcomes according to withdrawal of immunosuppression or administration of COVID-19 therapeutic alternatives). This information will provide clinicians with real-time personalized information, and hopefully will aid in clinical decision making.

Methodology

Study design
Multicenter international prospective observational study

Participating Centers
Centers from all over the world performing follow-up of transplanted patients will be contacted and invited to participate in the study.
Upon agreement to participate, principal investigator (PI) from each collaborating center will be provided a login to the IDOTCOVID database (IDOTCOVID.org). Centers performing more than one type of SOT shall define a PI for each SOT, and a login be provided to each PI It is the PI’s responsibility to introduce the data on the database. 

Patient selection
All solid organ transplanted patients (Kidney, pancreas, liver, heart, lung, intestine/multivisceral, VCA [vascularized composite allograft]) with confirmed COVID19 infection will be included prospectively using electronic medical records.
Patients with negative test results but with clinical symptoms/radiologic findings suggestive of COVID19 infection will also be included. 

Data Inclusion
Data to be included in the study concerns:

• Patient demographics: age, gender, race, country
• Patients’ previous medical history: hypertension, diabetes, dyslipidemia, pulmonary disease, ischemic cardiac disease, or any other disease deemed significant (to be introduced as free-text);
• Transplant history: type of transplant, date of transplant, number of previous transplants, induction and maintenance immunosuppression, previous history of graft rejection, number of rejection episodes, date of last rejection episode, treatment of last rejection episode;
• Medication: Treatment with ACEI/ARB medication; Maintenance Immunosuppression;
  
• COVID19 history: Disease presentation symptoms, date of first symptoms, patient management (outpatient or hospital admission), hypertension at Hospital admission, hematological and biochemical values at hospital admission,gasometric values at hospital admission, pharmacological treatment for COVID19, management of immunosuppression during COVID19 infection, requirement for ICU and Invasive mechanical ventilation (IMV) or non-invasive mechanical ventilation (NIMV), date of ICU/MVI/NIMV, date of ICU/MVI/NIMV withdrawal, requirement for renal replacement therapy (RRT), type of RRT required, date of RRT indication, date of RRT withdrawal, patient and graft outcomes (functioning/alive vs failed/death), date of graft failure and/or date of patient death.

User Guide

Decision Support Algorithm (DSA)

The development of the Machine Learning Algorithm will be performed using all the variables included on the database, including patients’ demographics, transplant characteristics (including previous graft rejections), as well as all variables associated with the COVID-19 infection.
The DSA will be developed in two phases:

1. Phase 1: Based on medical experts’ pre-defined and codified rules. This is expected to be performed on very short notice once the knowledge of the experts in the field has been captured and encoded.
2. Phase 2: Though Machine Learning, based on the different variables captured and the ‘outcomes’ of each of the cases, to identify those variables that have an impact on the result. In this case training of the model will be started once the initial 1000 cases have been reached and as new cases are added, it is retrained

 
The entire decision support system will be based on Ingenuous Pty Ltd software, which will be installed on the servers where the data shall be hosted. Said software called ‘Intuition’ uses as a data layer ‘elasticsearch’ which is a search engine specialized in search analysis in an unstructured data set. Both the web used to collect data through web forms and the software to support decisions will run on the same infrastructure, offering users who provide data a web application that integrates all the functionalities. A data scientist would access the data remotely and for a limited time for the development of the Machine Learning model.

Data management and protection

Access to data

The treatment, communication and transfer of personal data of all participants will be in compliance with EU Regulation 2016/679 of the European Parliament and of the Council of April 27, 2016, being mandatory as of May 25, 2018 and the Organic Law 3/2018, of December 5, on the Protection of Personal Data and guarantee of digital rights. 

Study principal investigators and Technical Committee will have access to all data included on the webpage. Local PIs will have access to data included from their center. When required, and following an agreement, official national transplant organizations will have access to pseudo-anonymized data from their national cohort.
Upon request from any of the study collaborators, and only for research purposes of studies previously approved by local Ethical Investigation Review Boards, pseudo-anonymized individual patient data may be shared in an exportable database.
Data will be retained for a maximum of 10 years, and afterwards eliminated from the host service. 

Ethical approval
Hospital Clinic of Barcelona

clinical_research_ethics_committee_opinion.pdf
File Size:	101 kb
File Type:	pdf

Download File

---