Better control of immune-mediated diseases, non-invasive imaging of immune cells, machine learning for drug discovery and advances towards better, more patient-friendly clinical trials are the topics of the new IMI calls for proposals. The deadline to present the project is 14 June 2018.
The call will provide a total budget of approx. € 167 million, € 84,92 mln of which will be covered by EFPIA companies and IMI2 Associated Partners and € 82,36 mln by IMI2 Joint Undertaking.
The interested companies and research centres can find all documents needed to apply – including the IMI Manual for evaluation, submission and grant award (version 1.6) – on the Participant Portal and the IMI2 Call documents page. Short proposals must be submitted via the electronic submission system of the Horizon 2020 Participant Portal.

Targeted immune intervention to treat non-response and relapse

The first topic of the Call addresses the needs of patients suffering from immune-mediated diseases who fail to respond to current standard-of-care treatments or undergo a rapid relapse. The goal is to identify new methods able to predict how patients will respond to treatment, in order to identify potential novel patient-centric treatment approaches.
Novel immunological biomarkers are expected to be identified to predict clinical responses; activities might include the analysis and profiling of immune cells obtained from non-blood tissues, the discovery of new, individual disease and cross-disease predictive biomarkers for disease, non-response, relapse and flare-up. These can be validated by mean of early phase clinical trials (e.g. enriched study populations for certain molecular pathways; adaptive and basket trial designs etc.), with a special focus on well-characterised immune-mediated diseases, such as lupus erythematosus, rheumatoid arthritis, ulcerative colitis, Chron’s disease, asthma and chronic obstructive pulmonary disease (COPD).
The translational precision-immunology research platform that should represent the main outcome of the Call should provide a more specific tool to improve patient management and help identify personalised treatment. The more accurate definition of subcategories of auto-immune disorders and their responses to particular therapies on the individual patient level are expected helping in the reduction of failure rates in early clinical trials; they will also support discovery of novel therapeutics and access to the most appropriate patient populations. Results might be also used in future to design platform trials for single indications with multiple mechanisms. Another expected target is the decrease of phase 2 proof of concept (POC) attrition and of the costs of development to be sustained by pharmaceutical companies to achieve regulatory approval and reimbursement.

Non-invasive clinical molecular imaging of immune cells

The second topic will parallel and support the goals of the first one though the establishment of a consortium to develop and validate a quantitative, non-invasive, immune cell imaging platform that should include novel and target-specific molecular imaging agents, (hybrid) imaging modalities, and image processing algorithms. The availability of non-invasive diagnostic methodologies is important to improve the penetration of precision medicine through the identification and stratification of patients and the prediction of therapeutic outcomes. Early diagnosis of the disease and/or its classification based on the immune phenotype should be also facilitated.
The topic aims to overcome the limits typical of current pharmacodynamic (PD) assessments of immune cells based on peripheral blood biomarkers or biopsy samples acquired by invasive procedures. Furthermore, already available imaging technologies provide limited information on time-dependent and disease-specific relevant immune cell subpopulations and compartments types, or measures of direct engagement of immune targets.
Immunotracers able to bind specific immune cells or targets within immune-mediated pathways would represent a definite advantage to address immune cell subtypes and immune markers of disease in a clinical setting, providing in vivo insights and improving knowledge about the pathophysiology of various immune-mediated diseases. The methodologies developed under this topic could provide an early indicator of whether patients are likely to benefit from a given (immuno-)therapeutic intervention (surrogate of response), also at the level of tissue/organ sites that are not biopsy-accessible. Another expected outcome of this action is related to a greater regulatory acceptance of standardised protocols using validated immune-imaging approaches, an important element that might significantly reduce the time and cost of clinical trials.

A platform for federated machine learning

Artificial intelligence and machine learning algorithms are rapidly emerging to play a central role in the development of new therapeutic interventions based on the acquisition and analysis of large amounts of big data. Large quantities of data are also generated during drug discovery activities, and their analysis has to be refined order to meet the complexity typical of biological systems. The third topic of the Call addresses this need, in order to optimise research activities, decrease development costs and improve regulatory acceptance of the generated data.
The new federated and privacy-preserving machine learning platform that is expected to be developed under the topic shall be initially validated on publicly accessible data in order to demonstrate the ability to preserve from illegitimate use of data. The technology is also expected to be scalable enough to be deployed to a significant representation of private data in the actual preclinical data warehouses of the participating major pharmaceutical companies in yearly evaluation runs.
In silico generated predictions obtained from the platform are expected to replace in future years pre-clinical in vitro testing and compound synthesis. A possible extension of this target might involve the application of similar concepts to clinical data, to enable faster recruitment of more targeted patients populations and real-world evidence analysis. The results of this action are also expected to improve access to data by third parties, “providing data owners with the confidence that their data and the corresponding predictive models will remain private”. Furthermore, federated learning represents a line of research and product development beyond that of data federation and specifically addressed to knowledge and Information and communication partners.

A Centre of Excellence for remote decentralised clinical trials

The last topic of the Call is focused on the creation of a new Centre of Excellence for the management of remote decentralised clinical trials. This action addresses the need to facilitate patients’ recruitment and retention in order to improve the adherence to clinical trial protocol. “Geography and the distance to the clinical site” has been identified by the 2017-global CISCRP survey as a burden for the 60% of patients, representing one of the reasons for a negative decision upon the participation to a clinical trial.
The topic aims to improve patients’ experience by disaggregating the current model of running clinical studies, while mapping new technologies (e.g. telemedicine, mobile health, etc.) that might support the new decentralised model. The projects should be focus on the demonstration of the feasibility of such model, reaching an higher speed of recruitment and a better retention. The follow-up phase of the trial should be also considered, to provide increased flexibility to patients and reducing the burden both on patients and hospitals. Among other targets of this topic is also the support to the update of the ICH guidelines all along the process by generating evidence.