Newsletter

The eleventh issue of the IDEA4RC newsletter

Hello,

In this newsletter, we interview Roberta Gazzarata, biomedical engineer at HL7 Europe, R&D Director and Co-founder at Healthropy srl, and member of the IDEA4RC consortium, about the the project’s efforts to develop a European common data model for cancer. Gazzarata shares insights on how this initiative emerged from the IDEA4RC data model and outlines the next steps in advancing cancer data standardization.

If you missed the previous issue, where we covered the fifth plenary meeting of the IDEA4RC consortium, you can find it here.

Subscribe to our newsletter for bi-monthly updates on the project’s advancements. If you’d like to invite your colleagues or friends to subscribe, share this link with them.

IDEA4RC aims at building a rare cancer data space that facilitates re-using data collected in clinical settings to advance research on these diseases.

A core challenge in this endeavour is that of defining a minimal set of clinical concepts to include in the data space and determining how to logically connect these concepts- this is the essence of building a data model.

In developing such model, IDEA4RC researchers considered several requirements and constraints. First, the data are intended for secondary use, which entails a set of legal requirements distinct from those governing primary use cases – those directly related to patient care. Currently these requirements are set by the General Data Protection Regulation (GDPR), but the European Health Data Space (EHDS) will soon come into force, changing the legal basis for processing health data for research purposes. Second, the selected clinical concepts should be sufficient to support the use cases envisaged by the clinicians involved in the project (outlined in deliverable D8.1). Third, the selected concepts should be available or easily retrievable by most of the clinical centres involved in the project.

Within IDEA4RC, the development of the data model followed an iterative process involving clinicians and data engineers with experience in healthcare. Through this work, researchers realised that many of the challenges they encountered apply not only to rare cancers but to all cancers and the solutions they propose could have broader relevance.

This is why, thanks to the support and interest of HL7 Europe, they will present their findings at the HL7 Europe Working Group Meeting and EU-a-thon 2025 in Lisbon from 10 to 14 February to initiate a project on a European common data model for cancer.

The meeting will involve several other European and national projects that are working on cancer data related topics and are tackling similar challenges.

We interviewed Roberta Gazzarata, a bioengineer at HL7 Europe, member of the IDEA4RC consortium and co-leader of this initiative. Gazzarata explained why Europe would benefit from a common data model for cancer and how they plan to build one.

Why would Europe benefit from having a common data model for cancer?

Europe is placing significant emphasis and investing substantial resources in improving cancer care and research. At the same time, this is a pivotal moment for data-driven research and innovation. With the European Health Data Space (EHDS) soon coming into force, the European Union aims to unlock the power of data for the benefit of all European patients.

However, these benefits will not be realised unless health data are easily accessible and interoperable, while also being securely protected from malicious threats and used ethically. To achieve these objectives there is still a lot of work ahead.

Rare cancers would benefit more than common cancer from the possibility to collectively analyse data from different centres in Europe, considering the very low incidence of these tumours. However, as our ability to analyse the genetic features of cancers improves, we continue to uncover greater differences between cancer types and common cancers will be facing the same data scarcity issues of rare cancers.

Enabling comprehensive data analysis will be crucial in identifying risk factors, evaluate treatment effectiveness and quality of care. It will also help bridge disparities in health outcomes—both between academic and non-academic settings and between rare and less rare cancers.

What will be your starting point?

So far, we have identified various starting points as the basis of our discussion.

The first one is mCODE, promoted by the American Association of Clinical Oncology (ASCO), with the objective of facilitating cancer data interoperability and improve overall cancer data quality for patient care and research in US. It does so by establishing a core set of elements that should populate all Electronic Health Records for cancer patients. While mCODE provides an excellent foundation, it was primarily developed to support primary use cases, meaning those directly related to cancer care. Additionally, it has been developed based on US core specifications and the legal landscape governing personal and healthcare data in the United States differs significantly from that of Europe, which presents additional challenges.

The second starting point is MEDOC (Minimal Essential Description of Cancer) data set, developed by Digital Oncology Network for Europe (DigiONE), an initiative promoted by the Digital Institute for Cancer Outcome Research (DIGICORE), public–private partnership founded in 2020 to digitize cancer out- comes research, which is also IDEA4RC partner.

MEDOC includes 40 clinical concepts and was the result of an iterative consensus process conducted with 16 hospitals in 13 countries in Europe with the objective of creating a digital learning system for precision oncology. This process also considered data availability in non-academic settings, where IT skills and resources are often more limited.

Finally, data models developed by European and national projects such as the one developed by IDEA4RC will be inputs. As the DigiONE position paper on MEDOC states, oncology is evolving towards a paradigm where each cancer is characterised as a unique rare cancer with distinct molecular specifications. For this reason, we believe that the IDEA4RC data model will serve as an excellent starting point for our work.

Who will contribute to this project?

HL7 Europe is involved in multiple data-related projects on cancer beyond IDEA4RC, including PanCareSurPass and Flute. Therefore, we will draw on the expertise of our colleagues at HL7 Europe who are involved in these projects, as well as IDEA4RC researchers.

We have also leveraged the workshops organized by the European Commission in 2023 on the re-use of health data resources for cancer research to foster synergies among projects in the field of data-driven cancer research. Several of the projects involved in these workshops have been invited to collaborate on an initial minimal set of clinical concepts, and we will discuss their feedback in Lisbon this week to start to create the European common data model.

Why is HL7 Europe well positioned to lead this activity?

A model is a great starting point, but for it to have a real impact, it should be implemented so that it can be easily adopted. With our expertise in health data standards implementation, we aim to turn this model into a practical tool and facilitate its adoption by hospitals across European member states.

Moreover, HL7 has successfully experimented with accelerators in the past, particularly through its US affiliate, where they have proven to be an effective means of turning projects into reality. Phoenix, the HL7 Europe initiative for cancer, aiming to become the first European accelerator of its kind, will be launched in Lisbon. While it will not be limited to the development of a common data model for cancer, this will be the first project initiated within the initiative.

We aim to involve as many clinical centres as possible, as broad participation is crucial to developing a model with the highest probability of widespread adoption.

What are the first steps you would like to take?

We hope that in Lisbon we will agree upon an initial small set of core concepts on which we will start developing a FHIR implementation guide. We need to start small for the whole process to be sustainable and to be able to show its value. This is a very ambitious project which we hope will keep us busy for the next several years. This first meeting marks the beginning of a journey we are truly excited about.