Highlights of the EHA-EMA Joint Symposium on RWE

8237 20240615 blur

The fourth EHA-EMA Joint Symposium at EHA2024 brought together investigators, regulators and patients to discuss the use of real world evidence (RWE) in the evaluation of new drugs. The symposium comprised three presentations addressing different aspects of RWE, followed by a panel discussion.

Results of joint EMA-EHA survey on attitudes to RWD

Tarec El-Galaly discussed the preliminary results of a recent survey run by EMA and EHA looking at attitudes among hematologists and regulators towards real world data (RWD) studies compared to other trial designs. Results from the survey showed that whereas both groups have very favorable attitudes towards randomized control trials (RCTs), when it comes to studies using RWD, healthcare professionals (HCPs) tend to have a more favorable view whereas regulators tend to be more critical.

The EMA-EHA survey also investigated how convinced respondents were by evidence from different trial designs for various outcomes. The results showed, unsurprisingly, that RWD critics are not strongly convinced about overall survival (OS) effect, even with high quality RWD. In fact this group considers poor quality RWD to be less convincing than studies with no formal comparator. For RWD proponents, the quality of the data has to be high to persuade them of the OS effect. All responders, regardless of their attitude to RWD, were less willing to accept toxicity trade-offs in scenarios with a comparator and lower OS gain and/or low quality data.

The differences in attitudes observed in the survey highlight the need for more activities that work towards increasing alignment between regulators and HCPs.

Case study on using data innovatively

Martin Kaiser described a case study in multiple myeloma in which near-RWD was used as the external comparator arm in the OPTIMUM trial. In this case study, high unmet need and the fact that randomization was ethically challenging for this patient group were important considerations in the choice of trial design. The investigators partnered with Myeloma UK and agreed to run an external comparator trial. Trust in the external data, in terms of availability, accessibility, completeness, quality and comparability, was essential in this decision. Martin explained that although the study used data from another large pragmatic trial as the external comparator rather than true RWD, the lessons learned from this study could be extrapolated to studies using RWD.

The first lesson of the OPTIMUM trial, in which the initial large dataset from a trusted source was whittled down significantly when the criteria for data completeness were applied, highlights the need for careful pre-planning in a RWD setting. Secondly, the analysis of the results, in this case study, turned out to be straightforward, as the results were unequivocal in all areas. However, in the situation where the results are more ambiguous, interpretation and deciding how to go to the next step are more difficult. These are important considerations for planning large RWD comparison trials.

This case study shows that external data, including RWD, can be acceptable to answer specific questions. Martin proposed that label extension and dose optimization could be the place to start. However, trust and new expertise is needed.

Building quality sources for RWE

Jan Cornelissen, national seconded expert to EMA, presented the regulatory perspective on using external RWD as control arm in hematological studies. Jan outlined the current framework within the EU to enable use of RWD, describing the different areas of work and ways in which its integration into regulatory decision-making is being facilitated.

Jan then discussed three studies:

  • A methodological review of externally controlled study designs using RWD found that most of the studies identified included RWD that were retrospectively monitored.
  • The second study aimed to answer the question: to what extent can external RWD compare to the actual control arm in a randomized study?
  • Lastly Jan discussed an ongoing pilot study, in which prospective real-time RWD is being compared to an RCT control cohort.

Limitations highlighted in the first two studies show that using retrospective RWD can lead to deficiencies in elements such as matching, follow up, and statistical analysis, and that even prospective RWD may need to be supplemented with additional data. Preliminary findings from the third study suggests that the use of digitally available hospital data may help to address these limitations.

Patient perspective and panel discussion

Patient advocate Natacha Bolaños applauded the undertaking of the EMA-EHA survey. She stressed the importance of robust data with transparent data sources and highlighted the need to integrate patient experience data (PED) and quality of life (QoL) data into RWD. “My aspiration, as a patient representative, is to have trial designs, which are currently very selective, move to designs that are representative of patients from the real world.”

When asked about the patients’ views on data sharing, Natacha emphasized the need for transparency on data collection and use. It is necessary to have a good and timely informed consent process (not just a form). The involvement of patients is vital. She further stressed the need for a robust framework for both data collection and reporting. This is also important to establish harmonization of endpoints.

Pierre Demolis (representing EMA) pointed out that the reasons for the difference in attitudes between HCPs and regulators observed in the EMA-EHA study are the differing roles that these two groups hold: regulators need to look at safety of the drug in real use. They have to answer the question is the benefit/risk ratio positive, using the evidence that is presented to them. Pierre went on to support Martin’s proposal that treatment optimization post marketing approval is the ideal use for RWD, given that physicians have priority access to RWD and could use it to answer these types of questions that are often not answered in industry-led trials.

To a question on the role of a pre-qualification process for RWD sources, Pierre and Jan both emphasized that EMA already has processes in place for investigators to seek advice on trial development, including for qualification of a system to gather RWD, through the Scientific Advice Working Party. When using RWD it is particularly important to make use of this process to ensure deficiencies in elements such as statistical analysis plan, follow up, matching etc, are addressed before starting the trial.

Responding to a question on how EMA engages with HTA bodies, Pierre shared that there is progress on this front. Although the evidence required by the HTAs can be different from that required by the regulator, when an applicant seeks advice from the regulator, this could be done in parallel with HTAs. Furthermore, EMA will now be sharing the results of their assessments throughout the process rather than at the end. So, the dialogue between the regulator and HTA bodies is becoming more open, allowing the HTA bodies to start their assessments earlier.

Key takeaways

The panel concluded that, whilst there is promise in RWE, the data used needs to be prospective and contemporary. The prospective accrual of toxicity data – crucial for regulators and payers – still presents a challenge.

Key determinants for trust are the availability and accessibility of the data, its quality and completeness, and the comparability of the data.

Health professionals and regulators often hold different views on the use of RWE. For this reason further dialogue is necessary to ensure alignment on when the use of RWD is appropriate and how to define and improve data quality.

(Re-)watch the joint symposium

If you were unable to join the symposium, or would like to revisit the discussion, you can do so on the Congress platform until August 15.

Last Updated on Friday 05 July 2024.