AGENDA

Clinical Data Sharing and Consent
16th June 2022 at 3pm BST/10am EDT

(dial in numbers below)

Meeting Recording

Zoom Details:

Join Zoom Meeting

https://us02web.zoom.us/j/84801200657?pwd=QU9zZmR2UFlhSUUxN0lCbVZMWVBGdz09

Meeting ID: 848 0120 0657

Passcode: 652103

Attendees:

1. Maili Raven-Adams
2. Lindsay Smith
3. Beatrice Kaiser
4. Emma Tudini
5. Danya Vears
6. Yann Joly
7. Stacey Donnelly
8. Juliann Savatt
9. Alison Hall
10. Danielle Azzariti
11. Ariana Gonzalez
12. Miranda Mourby

Minutes:

Overview of Community

JS: Outline the importance of clinical data, and also what ClinGen is thinking about. From our perspective, there are often requests to clinical labs to share patient information for interpretation. Such data sharing often includes sex at birth, age, family history or inheritance, phenotypic details and sometimes other testing or other variants identified via testing. Ultimately sharing such data is useful to help interpretation and can benefit patient care. The classifications can be changed based on this information. If data are deidentified and only for interpretation, it can be shared without explicit consent. The problem however is the uncertainty over what types of data and what level of detail can be shared, and for what reason (just variant classification? Does it expand to delineating phenotypes?).

ClinGen is an NIH funded resource aimint to better understand the relationship between genes and health. Uses publicly available data. Have a data access protection and confidentiality working group (DAPC) to manage data policy frameworks. Multi-disciplinary team tasked with a number of issues, thinking about what types of data can be shared, where should data be stored. Hope is that we can harmonize with guidance from this group, broader guidance would be useful.

DV: With that group, what kind of role public engagement was playing in terms of using patient or public views to inform those recommendations?

JS: Having our first conversation about this later today! Haven’t gotten to place of collecting stakeholder feedback. First step is to assess current practices.

SD: Since Geisinger is such a big player, has there been any thought to just systematically getting informed consent without having to decide?

JS: We want to think more broadly in not making it geisinger specific, but for any clinical lab or group that wants to share data. We thought about a broad consent and make materials but the implementation has been limited. Generally acceptance that we might not need that consent if its deidentified.

JS: [in chat] - Here are the broad data sharing consent resources Danielle is speaking about and that she helped lead the development of - https://www.clinicalgenome.org/tools/consent-resources/

DA: Even if the clinicians were on board, you need the lab on board. They’re the ones that hold the information and are responsible for sharing it in the responsible way. How do we track that internally who we can share and who we cannot.

YJ: Is the form you speak of somewhat harmonized with BMK’s clinical consent form?

DA: Yes, Bartha was a coauthor on that effort.

AH: Record of discussion form experience across labs. Found that there was a lot of variability between labs, some wanted to amend or add details and hard to get consensus on the exact form that should be used. The genomic medicine service in the UK has adopted a version of that form and has required it to be applied unilaterally for WGS but not for other forms of testing. One of the problems is the time it takes to go through each of those elements and explain them.

MM [in chat]: Here is the Genomic Medicine Service Record of Discussion form that Alison mentioned: https://www.england.nhs.uk/wp-content/uploads/2021/09/nhs-genomic-medicine-service-record-of-discussion-form.pdf

YJ: Are we going to talk also about the research use context, or are we focused on clinical use.

SD: I vote for research use

DV: I think my response depends on what you’re planning on doing in order to develop the recommendations. I saw something about a questionnaire, if you do that I think it makes sense to get the scope of both.

YJ: Maybe try to map out where consent forms might be needed.

DV: I think the answer to a lot of these questions are based on what people expect to happen with their data. Need to map the literature of what already exists with these questions. Maybe we we already have enough data.

YJ: Start with a review. Would be interesting, European colleagues, if there is something that you think would make this data sharing more complex in Europe, please let us know. If we attempt to do something that will have international traction, want to recognize that there are limitations out there.

AH: There are complexities certainly in the UK around other policy drivers. The data strategy published a few days alludes to reforms to the patient opt out, which might change the landscape and take some emphasis away from the professional interaction.

YJ: Draft a small questionnaire that we send to everyone in this group as a pilot, get a sense of where different countries are in tackling this challenge.

BK: Agree, send out a members form with current practices from their institution.

YJ: I like the idea of reaching out broader than our group. Two starting point activities.

Moving on to the end product we’re expecting. Envision a points to consider where we try to achieve some consensus over key things, like what kind of information can be shared. If we want to share for research, suspect there will be more trouble on agreement with this. Consensus document, would that help?

JS: I think a points to consider would go a long way in level setting between labs. And won’t be a small undertaking to reach that consensus!

DV: Q about who the audience is for this, is it labs, all clinical staff? Will make a difference in how we frame it.

YJ: Clinical staff is definitely one part of the audience. Agree we’ll need to consider this. Also whether patients are a potential audience.

SD: For the US, if we do tackle research it would also be the IRBs and things like that. They do love these points to consider documents because it gives them something to hang their hat on. Tend to be risk averse groups.

DV: If we do clinical and research, maybe should do different documents because they have very different considerations.

YJ: Stage 2 could be potential research use of this information?

DV: I think that makes sense, but literature review could be done together because some papers will address both.

DA: Wonder among those tiers if the top most is just broad data sharing. A broad clause at the top.

AG: I see a lot of variants in ClinVar that say classified by internal data, and don’t know what kind of study created that data. Need to decide which path, clinical or research.

Next Steps:

Literature review.

Share thoughts with Yann about how to conduct this review, guidance for Beatrice as she gets it started. Reach out to rews-coordinator@ga4gh.org

Julian and Yann will continue offline on the 5 questions, but if anyone has suggestions, will make sure to take it into account.

DV: Did a massive review of return of results, would be happy to share my experience.

MM [in chat]: Happy to help with EU/UK work.