Audio Transcript from January 13, 2026 R3 Seminar

This transcript is intended to provide an accurate, verbatim representation of the language used by the speakers during the R3 Seminar Using RECOVER Biosamples to Advance Your Research held on January 13, 2026. You can view a timestamped version of the transcript with the video on YouTube.

Watch the recording or read the summary on the R3 Recap page.

Disclaimer: This transcript has been generated using AI technology and lightly edited. Please note that some errors or omissions may have occurred due to the limitations of automated transcription.

Patrick Ahearn:

Good afternoon and good morning, everyone. Welcome to today’s RECOVER webinar. This is Part 2, “Using RECOVER Biosamples to Advance Your Research.” My name is Patrick Ahearn with RTI, and I’ll be helping out with the virtual room today.

So just a few quick housekeeping notes before we get started. When you log in, your microphone and web camera will be automatically muted and turned off, but if you have any questions for our presenters today, please submit those in the Q&A window at any time. If you run into any technical issues with Zoom, please let me know in the Q&A window as well. We do have closed captions available during today’s webinar, just click on the “Show Captions” button on your main Zoom toolbar to turn those on. At this time, I’ll go ahead and turn things over to my colleague, Christine, to kick off today’s session.

Thank you.

Dr. Christine Bevc:

Thank you, Patrick. Welcome everyone to today’s R3 seminar. My name is Christine Bevc, and I’ll be your moderator. If you joined us last month, welcome back. Today’s session is the second part in our series focused today on available RECOVER data and biospecimens. Over the next 90 minutes, we’re going to be providing a high-level overview of the biospecimens and samples collected through the autopsy tissue study and the observational adult and pediatric cohort studies. We’ll also provide a quick refresh of the tools available to help you navigate and estimate these samples. If you missed our session last month, the recording is now available on recoverCOVID.org.

Next slide. All right. The data, biospecimens, and tissue samples discussed during today’s session represent the contribution of over 17,000 individuals since RECOVER’s observational studies began in 2021. Respect for and protection of the interests of research participants is fundamental to NIH’s stewardship of people’s personal health data.

For questions about access and availability, we encourage you to review the short guide distributed to all registrants, and we might get a link to that in the chat below. In compliance with NIH’s data management and sharing policy, RECOVER allows those interested to view the data and for researchers to analyze it, learn from it, and incorporate it into future studies.

All right. Next slide.

For interested researchers, this series also directly complements the current Research Opportunity Announcement and notice for ancillary studies from NIH. These are some of the ways that researchers both within and outside of RECOVER can apply to use participant data and access the biospecimens in their proposed research studies. If you check the chat window, you’ll see that link to the opportunity posted. Thank you, Kate. And it’s also available online where you can learn more and apply. You can click on the link or go to recoverCOVID.org/funding.

All right. So, without further ado, let’s quickly introduce our panel of experts for today’s session. It is my pleasure to welcome back Jim Stone. Jim is the site principal investigator for the RECOVER autopsy cohort at Massachusetts General Hospital and also serves as the chair of the Autopsy Coordinating Committee and is a member of RECOVER’s Observational Cohort Steering Committee.

We’re also pleased to welcome Madeleine Thorn and Trisha Balan from RECOVER’s Data Resource Core, also located at Massachusetts General Hospital. Madeleine serves as the assistant program manager, supporting RECOVER investigator requests for sample access. She’s also joined by Trisha, coming back to us for Part 2 as one of RECOVER’s leading data analysts supporting the preparation and harmonization of data from across the study sites. Rounding out our panel today, we’re joined once again by Emily Hughes and Cera Fisher. Emily is a product manager for BioData Catalyst Powered by PIC-SURE and will be demonstrating how you can use PIC-SURE to quickly query the data and identify samples using this publicly available data exploration tool. Cera is our program manager for NHLBI programs at Velsera and supports BioData Catalyst Powered by Seven Bridges, the secure cloud-based ecosystem that hosts RECOVER’s observational study datasets.

As we jump in, as Patrick mentioned, if you have questions throughout the session, you can use that Q&A feature to ask your question. It’s located again in your Zoom menu bar. Thank you to those who submitted questions in advance. After we go through today’s panel, our speakers will answer as many questions as possible, and, as usual, we cannot answer questions about individual cases, analyses, or proposals.

So with that, I’m going to turn things over to our first speaker, Dr. Stone.

Dr. Jim Stone:

Hello, I’m Jim Stone, and I direct the autopsy service here at Massachusetts General Hospital. I’m the chair of the RECOVER Autopsy Cohort Coordinating Committee, and I’m really happy to be able to talk to you about all of the biospecimens we’ve been collecting as part of the RECOVER autopsy cohort. I want to make it clear that we’re very excited about as many studies as possible getting launched and using these biospecimens to help understand the mechanisms of Long COVID. We want your proposals, we want the science to be completed, and the specimens are ... Essentially most of them are now in the biobank and waiting for the appropriate studies to be recommended and suggested.

Okay. Let’s go to the next slide. So why autopsy? I know some of you may have never performed a study using autopsy tissue. There’s really 4 key topics in this category. One is that, compared with serum analyses, which are really indirect assessments of tissue changes, with autopsy, you’re actually able to directly assess all of those tissues without needing to rely on indirect measurements. Compared with imaging, it’s far more sensitive than imaging to detect changes, and obviously you can detect the molecular changes as well and get real insight into mechanisms.

Compared with biopsy, you’re able to get larger quantities of tissue, particularly the difficult-to-obtain tissues like heart and brain. So, you do have a lot more tissue from many sites throughout the body—frequently tissues that are very hard to obtain. Also, you can simultaneously analyze multiple organs throughout the body, which is very difficult to do with biopsies, as you would have to then biopsy the same patient multiple times.

I want to stress that the tissue that we’re obtaining is not routine autopsy tissue. These are what I would call expedited autopsies, and the tissue is of high quality, much higher than that of a typical routine autopsy. It’s suitable for use with modern advanced experimental -omics approaches. As you’ll see, we do this in a way that the RNA is intact and that you’re able to do many types of analyses that would otherwise, in a routine autopsy, not really be potentially an option.

Okay. Let’s go to the next slide, please. In the autopsy enrolling cohort, we have 9 sites that have so far contributed material to the cohort. You can see this is not a single institution study; it is a multi-institution study. We have good coverage on the East Coast and also coverage of the Southeast, the Southwest, and the Midwest. We’re very happy that we have coverage across most of the United States, and this is not just a single-institution study.

Okay. Next slide, please. I do want to point out a particular issue with autopsy study permissions. We do obtain permission from the family, either the next of kin or the healthcare proxy specifically, to do the study and to obtain the tissue. We are not getting study from the decedent prior to death. So these are obtained retrospectively after death. Just be aware of that when you’re writing your proposals either for funding or your IRB [institutional review board], that we are getting permission, but it is from the next of kin, typically.

Okay. Next slide, please. I need to go over our group definitions because our cohort operates somewhat independently of the other RECOVER cohorts, and we do have a slightly different definition for our main patient groups. You can see the 3 patient groups here. We have the post-acute with Long COVID, and we define Long COVID as the presence of any new symptom or significant worsening of a preexisting symptom that’s not due to a cause that is independent of a PASC [post-acute sequelae of SARS-CoV-2]-related process. That new symptom also has to last for 60 days following the initial infection.

The reason we put in that clause, “Not due to a cause that is independent of a PASC-related process,” we want to be able to exclude symptoms that are due, for example, to motor vehicle accidents or other forms of trauma. Clearly, you can have a motor vehicle accident, have trauma, and have a pain that lasts more than 60 days. That would not count as PASC. If it’s a potential PASC process, we include it as Long COVID if the symptom lasts long enough. Now, granted, this definition is very broad. It’s actually very similar to the National Academy definition. So, there will be some patients in the Long COVID group who don’t have Long COVID because we’re using such a broad definition, but we’re also not excluding patients who have Long COVID by using too-narrow definitions. It is a broad definition, and we are sure that it has very high sensitivity but not necessarily complete specificity. That’s what you need to keep in mind.

The other group is the post-acute patients without Long COVID. They are 60 days, at least, following the initial infection, and they do not have any new symptoms or significant worsening of a preexisting symptom that can’t be attributed to some process unrelated to PASC. Then our third group is the acute group, and these are patients who are 15 to 59 days from initial infection. Notice that we do not enroll acute patients who are less than 15 days, so they have to survive at least 15 days to be in the acute group.

I also want to stress that we do not have any non-infected patients in this cohort, and there’s many reasons for that, but we do not have pre-pandemic controls or non-infected patients as part of the RECOVER cohort. If your study requires that type of tissue, you’ll have to obtain that on site, and it can’t be part of the tissue you obtain from RECOVER. Now, for the autopsy cohort, we essentially use the terms “Long COVID” and “PASC” interchangeably. PASC, meaning the post-acute sequelae of SARS-CoV-2 infection. The other cohorts don’t necessarily use those 2 terms interchangeably, so you have to be cognizant of the definitions that each cohort is using. For our cohort with autopsy, we essentially use those interchangeably. If you hear me say the word “PASC,” I mean Long COVID. I’m not trying to specify some different definition for this presentation.

Okay. Next slide, please. This is where we are to date with enrolling. Enrolling is actually still ongoing in this cohort. We have enrolled a total of 298 decedents, actually 299 as of noon today. There were 112 post-acutes with Long COVID enrolled, 155 without Long COVID, and 31 who are acute. We will continue enrolling through the end of January, so we have 3 more weeks of enrolling. We expect to be in the low 300s somewhere by the time we finish enrolling for total numbers, which is a fairly reasonable size of autopsy cases that have been acquired in this expedited manner, as we’ll talk about.

Okay. Next slide, please. I’m very happy that we have been able to enroll a diverse population that does essentially mirror the US adult population. We are somewhat much more diverse, I should say, than the average typical autopsy population. We’re about 65% White, non-Hispanic. At the moment, we’re 9% Black, non-Hispanic ;18% Hispanic; 4% for other underrepresented groups. I’m happy to say 47% female sex. As those who do autopsy studies know, the typical autopsy study in the US tends to be very male predominant and tends to be very Caucasian, White predominant, much more than 65%. So, we’re happy that we’ve been able to enroll a diverse population with relatively equal numbers of both sexes.

Okay. Next slide, please.

The big question for today is what are we actually collecting so you can start planning your studies? The key thing is for the cases; we get 52 anatomic sites throughout the body, and they’re shown on the right. We’ll go over these in more detail, so you don’t have to try to memorize them now looking at the diagram, but they’re tissue sites throughout the body. Actually, the autopsy protocol has been loaded into the chat, so you can download that. It’s referenced at the bottom, so please, it’s free to download off PLOS One, and I’m going to make reference to it many times. So, if you’re interested at all in submitting a RECOVER proposal, you should really download and read this protocol.

Now, there may be slight differences in definitions over what I’ve told you today. The current definitions are the ones we use, but the protocol is up to date in terms of the tissue sites we collect.

The key points: the postmortem interval is less than 24 hours. These are expedited. The standard autopsy typically has postmortem intervals of 2 to 3 days or more. These are expedited; they’re within 24 hours. And also, more importantly, the formalin fixation time is only 16 to 32 hours. In a lot of the formalin-fixed brain tissue that’s available, through your local autopsy services, the brains are actually fixed for 3 weeks. We’re only fixing the tissue for 16 to 32 hours, which allows us to preserve both the RNA and a lot of the antigens that are lost when you do those extended fixation times that are routine at most autopsy centers. The tissue is of very high quality, and it’s being collected from throughout the body, which is also very important.

Okay. Next slide, please.

I want to go through the individual tissues, and they were on the diagram, but it’s helpful just to talk through them. For the neural and muscular tissues, starting with the nerve, ganglia, and muscle, we have 7 tissue sites just for those. Those include nerves from the leg, the sural nerve, and peroneal nerve. Then we have 2 skeletal muscle sites, the calf muscle and the psoas muscle. We have the spinal cord, and we have 2 ganglia sites, the dorsal root ganglia and sympathetic chain ganglia.

For the brain itself, there are 12 distinct tissue sites: olfactory bulb, frontal cortex, basal ganglia, thalamus, hippocampus, occipital cortex, amygdala, pons, choroid plexus, medulla, cerebellum, and then the dura, the outer covering of the brain.

Okay. Next slide, please.

For the cardiovascular and pulmonary tissues, these are our next 2 largest groups of tissues. For the cardiovascular, we have 9 tissue sites. From the heart, we have the lateral left ventricle, the basal left ventricle, the right ventricle, both right and left atria. If there’s any lesional tissue grossly obvious at the time of the autopsy, that’ll be taken as a separate block. We also sample both left and right coronary arteries and the aorta. Now, for the pulmonary system, we have 1 block from each lobe: right upper lobe, right middle lobe, right lower lobe, left upper, and left lower. Again, if there is grossly obvious lesional tissue, we’ll put that in a separate tissue block, and then we also sample the trachea.

So go ahead to the next slide. For the remaining organs, this is what we’re sampling. For gastrointestinal, there’s 7 tissue sites: duodenum, ileum, ascending colon, descending colon, pancreas, liver, and mesenteric fat. For the renal tissue, we take both regular kidney and then also kidney if there’s any obvious lesional tissue. In the hematopoietic and lymphatic systems, there are 3 sites. We have bone marrow, which is typically expressed from the rib, so it’s what we call a rib squeeze. We also have spleen and mediastinal lymph node. For the reproductive organs, we have ovary, fallopian tube, and testis. For the skin, we have both calf skin and the attached subcutaneous tissue. And then for endocrine tissue, we have 1 tissue site, the adrenal gland.

Now, if I’ve mentioned the tissue site or you saw it on the screen, then we have it. If it’s not listed here, we do not collect it. So that’s very important to keep in mind when you’re submitting your proposals.

Okay. Let’s go to the next slide. Okay. This is what we’re collecting from the tissue sites, particularly. The left 2 items on the top and the bottom, for each of those 52 tissue sites, we have 2 pieces of frozen tissue that are snap frozen, typically on dry ice. And we also have 1 formalin-fixed paraffin-embedded piece of tissue. This is where most are getting their tissues from, either this frozen tissue or this formalin-fixed paraffin-embedded tissue. For the autopsy cases, in addition to those 52 tissue sites, we also collect cerebral spinal fluid, serum, and then we have whole blood spot cards as well. We collect stool, which can be used for microbiome analyses, for example, and we collect bronchial swabs.

Now, we don’t do this on all the cases, but for a very limited number of the cases, we do have some intact fixed hemibrains. The purpose of this was so groups could do postmortem ex vivo MRI [magnetic resonance imaging] on these brains if they wanted to do so. We actually have one study ongoing, but it’s not utilizing all of the hemibrains that we have. If there is a group out there wanting to do more hemibrain MRI, please reach out to us. In that particular case, if you just want to contact me, I can talk to you about what the resources are, but that’s certainly the option for more ex vivo hemibrain MRI exists for those groups that have the funding and wish to pursue that. So overall, we have collected over 24,000 biospecimens on the autopsy with the autopsy cohort. So, it’s an immense amount of biospecimens that are waiting to be analyzed.

Now, obviously many of them are being analyzed, but there’s clearly room for a lot more studies, which is why we’re having these webinars.

Okay. Let’s go to the next slide. I want to talk a little bit more about the snap frozen tissue because this is certainly a tissue that a lot of people are interested in for their studies. So, we have 2 pieces that are frozen in dry ice in these little cryo cassettes. The tissue is frozen without any media or reagents, which is good for most studies. You just have tissue that is snap frozen. You need to be trying to request milligram amounts. I realize, because it’s autopsy, if you are doing it prospectively on a single case, you could get gram or even kilogram amounts of tissue, but we can’t reasonably, in an economic way, bank grams and kilograms of tissue from all of these tissue sites that we’re collecting from, so you need to be requesting milligram amounts of tissue. Do not try to request an entire frozen block because we want more than 2 studies to be able to access the tissue. We’re trying to get as many groups to be able to be involved as possible.

Basically, you will receive a core from the frozen tissue blocks. We’ll just core into it. The core, depending on your needs, will be typically around 1 to 6 millimeters in diameter, with the tissue having a depth of about 1 to 4 millimeters in the core, depending on which tissue it is. Now, some tissues are much more limited than others. The bone marrow that’s expressed from the rib; that’s obviously not going to be nearly as abundant in the frozen cryo cassettes as the liver or the spleen. There’s a lot less of the bone marrow, but we try to give each group what they need in order to complete their studies.

A key point, we are not freezing tissue in OCT, the optimal cutting temperature compound. We cannot provide you frozen section slides. So do not write your proposal based on receiving those from us. We can give you the core of frozen tissue, and you can then prepare frozen section slides from that core, but do not request frozen section slides from RECOVER.

Okay. Next slide. I need to stress what is not available, and this is because groups have, unfortunately, spent time preparing proposals for things that were not available. I think it’s helpful upfront to know this. We actually do not have specimens from patients enrolled in other cohorts, to any significant extent. These are non-overlapping cohorts, so you can’t identify a key subset of patients in the adult cohort and then try to request the autopsy tissue from them. That’s not how it works. So, these are separate cohorts.

Unfortunately for the studies, but fortunately for the children, we do not have any specimens from children, because children very rarely die at all, and certainly very rarely from Long COVID. We do not have specimens from children, and we also, as I said, do not have specimens from non-infected control. Likewise, we do not have specimens from anyone who’s acutely infected for less than 15 days. Now, a key point: we do not have peripheral blood mononuclear cells [PBMCs] as part of the autopsy cohort. Those are available in other cohorts but not the autopsy cohort. So don’t request PBMCs through the autopsy cohort; we only have serum in the biobank. Once again, tissue sites not in the protocol are not available. When I went through all the ones that are available. And again, this Troxel et al., PLOS One 2024 protocol paper, is available in the chat. Please download that and please refer to that before submitting any proposals.

I will stress, just because we’ve had requests, we do not collect tonsil, vagus nerve, thyroid, and stomach, for example, are some key organs that we do not collect on, so we cannot help you for studies on those particular organs.

Okay. Next slide, please.

We also have associated clinical data. Now remember, these patients are completely separate, for the most part, from the adult cohort. You can’t go into the adult cohort part of BioData Catalyst and obtain clinical information on the autopsy patients. They’re 2 different cohorts. We are collecting pre-mortem clinical data. We do collect the demographics, the past medical history. We have extensive documentation of medications, which is something that we extensively document. We do document their pre-infection laboratory tests. Unfortunately, we have very limited imaging results, which are limited only to selected pre-infection imaging. So, we do not have much imaging, and we do not have the post-infection laboratory results. We do have their COVID symptoms, tests, treatments, vaccinations, also dates of infection, so you can sort out the most likely dominant variant at the time. Of course, we try to provide you with the specific PASC symptoms so that you’re able to identify the specific types of PASC cases that you want to study. Again, I’m using PASC synonymous with Long COVID, so the specific subtypes of Long COVID that you want to study.

Some of the pathologic findings are already in BioData Catalyst, including some of the gross pathology findings and some microscopic findings. Now, all of this data initially flows through our REDCap database, but it’s available to you upon application and requests through BioData Catalyst.

Okay. Let’s go to the next slide.

So, a few key points about our pre-mortem clinical data. This is obtained retrospectively after death from the electronic health records, typically Epic with Care Everywhere, but sometimes it’s other electronic health record formats. Occasionally, we have to rely on charts that are faxed to our enrolling sites from outside entities. In some cases, there may be very limited clinical data through these other 2 mechanisms, and we actually directly interview the family to get what information we have.

We do not have standardized enrolling type questionnaires that are asked of the patient themselves the way that the other cohorts have because we’re dealing with deceased patients, so we cannot have them actually directly tell us exactly on certain time points what their symptoms are. So, this is all done retrospectively. Now, for some decedents, only very limited pre-mortem clinical data is available, so that’s something to keep in mind. We don’t have extensive data on every patient. Again, we do not have post-infection imaging or laboratory results in our REDCap database.

Okay. Next slide, please.

We do have a big ongoing study called Pathology of PASC, and there will eventually be a histologic assessment of every block from all of those 300 patients, all 52 tissue sites. There’ll be a histologic read from a subspecialty-level pathologist, meaning for every piece of brain tissue, a neuropathologist will have given their opinion on that. For every lung block, there’ll be a pulmonary pathologist, et cetera. We don’t expect that to make its way through to BioData Catalyst until about 2027 because that’s still ongoing, but keep that in mind because it will very much help everyone who’s trying to write for studies in the future, because you can actually directly go in and find, not only the specific patient, but the specific pathologic change that you’re interested in and narrow that down to actually the specific tissue block. That will be coming, but we’re about a year or so away from that.

Okay. Next slide, please.

Key considerations for your tissue request: be reasonable with the number of patients for your study. Tissue analyses are expensive, and there’s very few studies that actually try to analyze more than 50 patients when they’re doing tissue studies, so be reasonable. We understand what things cost, and be sure that you have the right amount of funding for all the studies you’re suggesting.

Be as specific as possible as to what types of Long COVID patients you are most interested in. Are you really interested in patients with neurologic problems, patients with pulmonary problems? That’s very helpful for the success of the study. Request the minimum amount of tissue necessary to do the planned study. This is what holds up some studies: they try to request way more tissue than is really necessary to do it. So, at the beginning, just request the minimum amount. You need to justify the need for the tissue in your protocol in your request. This goes for also the different types of tissues. It doesn’t look very good if you write a protocol focusing on questions about lung changes, but then you request tissue from other parts of the body and then don’t even talk about that. So please justify why you’re asking for each piece of tissue that you’re asking for. The reason, again, is we’re trying to make this tissue as available as we can for all the groups that want access to it.

Okay. Next slide, please. I get this question a lot, and it was even a pre-conference question. How do you go about starting this process of having an autopsy-based RECOVER study? So, first, you either apply for a ROA [Research Opportunity Announcement] or an ancillary study. It depends on whether you’re also needing money from RECOVER or you have external funding, meaning funding external to RECOVER. If you need to get both funding and the tissue, then you should try a ROA application. If you do have a source of external funding and you just need permission for the tissue, you would submit for an ancillary study through ... And again, this is all on the RECOVER website.

Once those are approved, they are subject to review by the RECOVER Biospecimens Access Committee. So even if a ROA has been approved, all of the details are carefully gone over by the Biospecimens Access Committee, which also refers to, and consults with, the RECOVER Autopsy Cohort Coordinating Committee because we have to make sure we can actually provide what is being requested and that it’s reasonable and will not impact other studies that are ongoing. Basically, it starts with either a ROA or an ancillary study application, and that’s how you get the ball rolling.

Okay. Next slide, please. So, there’s multiple ongoing studies now involving RECOVER cohort biospecimens. I can tell you the first autopsy ROA data that’s going to be presented outside of RECOVER will be presented at USCAP [United States and Canadian Academy of Pathology] in the spring. So, if you’re attending that meeting, please look for that. You will see as the year goes on and the years follow, a lot of Long COVID presentations coming out using RECOVER data and specimens. We’re very hopeful, of course, we can have even more applications for projects.

All right. Next slide. My final slide. So, the RECOVER autopsy cohort has enrolled a diverse population of decedents, and the tissue is an invaluable resource that can offer unique opportunities to uncover the mysteries of Long COVID. It’s of high quality, and it will be utilized in multiple ongoing future studies.

Okay. Thank you all.

Dr. Christine Bevc: 

Thank you, Jim. Next, Maddie and Trisha are going to share a little more detail about the schedule of biospecimen collection and labs collected as part of the adult and pediatric observational studies, and that’ll help us better understand what’s available.

Maddie, go ahead and get us started.

Madeleine Thorn:

Great. Thank you, Christine. Thank you for the introduction.

So, as Christine mentioned, we’ll be starting with the adult cohort. Next slide, please. Great, thank you.

This is a snapshot of an overview of an individual’s entire participation within the RECOVER adult cohort. This just goes to show that biospecimens are one aspect of participation, and you can see below, in the little red box, that biospecimens are collected at multiple time points, but they also occur concurrently with different surveys and different study activities throughout the course of their participation.

Next slide, please.

We can start with the baseline biospecimens. So, the baseline visit occurs when a participant first enrolls in their RECOVER study. And so, with this, we want to mention that someone’s baseline visit may occur at any point regarding their first SARS-CoV-2 infections. This can happen either during kind of an active infection, maybe a year or longer after their first infection, or even before some participants’ first infections.

Next slide, please.

With that, regardless of a participant’s infection status, they still have the same samples that are collected during their baseline study. We start with the nasal swab or nasopharyngeal swab. One thing I want to note is that nasal swabs were the first type of sample that was collected. And then in June 2022, we switched over from nasopharyngeal, sorry, swabs into nasal swabs just to kind of reduce participant burden and discomfort. With that, we also have blood collections that occur during the baseline visit. We collect plasma, serum, PBMCs, DNA from white blood cells, and then also RNA as part of those blood samples. There is also a urine kit that is collected during baseline, and then a stool kit that is a take-home collection kit. And then with that, the baseline visit concludes with DNA that comes from saliva.

We want to highlight here that the Oragene 600 saliva collection kit is what is used for collecting saliva. We just want people to know that saliva can only be used for DNA purposes, rather than any other assays. Next slide, please.

So, this is more details on the baseline biospecimens that are collected, just so people can use the sample volumes to inform themselves on what they might be able to ask for. What I do want to note with these is you are seeing the maximum amount of volume that could be collected, because these are just the transport vials and the different collection tubes that are used, but this slide is just for everyone’s reference to know the gist of how much and the maximum amount of sample can be collected per a baseline visit.

Next slide. Thank you.

Now we can go into follow-up biospecimens. This is the same exact schedule as we just saw, but now in the highlighted yellow box are the follow-up visits that may occur for a participant. As I mentioned before, somebody may enroll either within their first infection or a year out from their first infection. What changes with that is that their collections happen on an annual basis. However, participants may have the opportunity to collect samples for the 3-month and 6-month visits in between those. So, this varies on a participant basis, but in general, this is what’s expected to be collected from participants. The one thing I want to note here is that we collect the nasal swabs/nasopharyngeal swabs with every visit, and then also the blood collections. However, urine and stool samples are only collected at baseline and the 24-month visit, just to also try to reduce participant burden, but these are the samples that we can expect to be collected.

And then I also want to point out the crossover baseline visit as well as the first on study reinfection as they kind of fit within a participant’s follow-up schedule because they can occur at any point after the enrollment or baseline visit. The crossover baseline visit happens when a participant is enrolled as uninfected, and then we capture their first on-study infection over the course of their participation in RECOVER. With that, as you can see, we collect samples similar to the baseline visit. And then the first on-study reinfection is something to the tune of that as well where we collect somebody’s first re-infection just to catch an acute infection if we’re able to. Next slide, please.

These are more details about the types of samples collected. These are the exact same parameters that we use for the baseline vials, but we just also wanted to make everyone aware of the different specifications and parameters of what we collect for each of the samples during the follow-up visits as well.

Next, please. Then we also wanted to highlight tier 3 biospecimens, as some of them are available for distribution. However, we wanted to really highlight the caveat that only a very small amount of participants will have these samples collected from themselves because these come from tier 3 visits and tier 3 tests, and so this happens with the participants that are eligible for these collections. With that, we have cerebral spinal fluid and then also residual tissue that may result from a skin biopsy. However, we really want to highlight that this is a small amount of sample available, but it still is available for distribution if you want to use that for your study.

I’ll pass this on to Trisha for peds. Thank you.

Trisha Balan:

Hello, everyone. Moving on to the pediatric cohort. Next slide, please.

Baseline study categories are determined by the timing of infection relative to enrollment in peds. These include post-acute infected, acute infected, and uninfected participants who get randomized to either the post-acute or acute arms. We also have an approximate number of participants for all those groups displayed on the slide. In addition to this, it is important to note that only a subset of pediatric participants are selected for long-term follow-up, with even a smaller group further selected for tier 3 participation or more invasive tests. In addition to this, another note that we would like to highlight is that the ABCD [Adolescent Brain Cognitive Development Study] cohort also is entirely post-acute and completes baseline assessments only. We would like to reiterate that pediatric participants’ study schedules are based on the time from enrollment, not infection, like in the adult study.

Next slide, please.

All right, moving on to baseline biospecimens. Next slide. Presented here is the biospecimen collection schedule for pediatric participants. Since pediatric participants have varying ages, there are also some slight differences in the kit types that are collected. You’ll notice a legend in the corner for what’s collected and what age range it applies to. For baseline biospecimens, only blood and saliva are collected. For those enrolled in the post-acute arm, they do remote collections for their blood and saliva samples. For remote blood collections, a blood sponge is used specifically, it’s called the Tasso device to collect the sample. For those in the acute arm, week 8 is what serves as their baseline visit, and saliva and blood are collected in person. The blood collection is done with a standard blood draw, and there are different types of blood collection tubes that are used, depending on the age of the participant, namely; SST [serum separator tube], PBMC, plasma, white blood cells, and red blood cells. Saliva is only collected once for the study, mostly at baseline, but if baseline collection is missed, it can happen later on, but only once.

Next slide. All right. We have some details here on the collections and volumes for post-acute baseline. Next slide, please.

You can find further details of what I just explained for acute baseline as well on volume types and ages of collection. Next slide, please.

All right, moving on. As mentioned, we’re going to move on to tier 2 biospecimens. Next slide. So, as mentioned before, out of those enrolled in the study, only some are promoted for long-term follow-up. So, this is good to keep in mind when you are thinking of the number of participants start to dwindle down. There are no differences in collection between post-acute and acute for long-term follow-up. They have a similar study schedule and biospecimen collection. Only a blood collection is done for long-term follow-up moving forward. At 12 months from enrollment for long-term follow-up, in addition to the blood collection, there’s also a PAXgene blood RNA tube that is used for participants who are 6 and over. Next slide, please.

More details on the 6-month collections are listed here. Next slide. And more details as well. As shown here, as mentioned before, RNA PAXgene is collected for those 6 and over for the 12-month visit onwards.

All right. So, moving on to tier 3 biospecimens, there’s a limited number of pediatric participants that are selected for tier 3, like we mentioned at the start. The biospecimens are collected as part of the microbiome laboratory study that is a pediatric tier 3 test. All pediatric participants are expected, if they are of age, to complete all the tier 3 tests. For the microbiome laboratory, these include samples such as needle swabs, stool, urine, sputum, skin, and tongue swabs. Next slide, please.

So yes, you can see more details about that, and that only a limited number of participants are selected for tier 3. Next slide, please. We’ll move on to caregiver biospecimens. So, in addition to pediatric participants, caregiver biospecimens are collected from the consented caregivers at baseline. These include dried blood spots for antibody testing and saliva samples for DNA extraction.

Additional biological parents may also contribute saliva samples. Participants under 18 may have what’s called a non-consenting caregiver, meaning they don’t give any biospecimens. Also, participants over 18 don’t require a caregiver to enroll, so there are no expected biospecimens for that as well. Next slide, please.

Moving on to the congenital cohort. So finally, the congenital cohort includes participants that are linked in the adult RECOVER study. So, these participants, based on the study design, kind of age into their study visits, with biospecimen collection occurring only at the 24-month visit, when the child is approximately 2 years old. So, blood is collected either using a home-based Tasso device or the standard venipuncture, so a classic blood draw, with samples processed for serum and EDTA [ethylenediaminetetraacetic acid].

In summary, RECOVER collects a variety of longitudinal biospecimens across adult, pediatric, caregiver, and congenital populations, which can be very useful for investigators looking at and utilizing biospecimens for their studies.

Thank you.

Dr. Christine Bevc: 

Thanks, Trisha and Maddie. As a reminder, if you have any questions, please use the Q&A option in the Zoom menu. Our panelists are being awesome in answering those questions in the written tab or writing those answers in the answer tab there. Next, we’re going to bring in Emily and Cera to help us better understand how you can find all this information, data, and be able to search for biospecimens using the tools available. Handing it over to you. Yep. There you are, Emily.

Emily Hughes:

Appreciate it.

Dr. Christine Bevc:

You’re welcome.

Emily Hughes: 

All right. Thanks for the introduction. Hello, everyone. My name is Emily Hughes, and I’m joined by my colleague Cera Fisher today, and we’re going to tell you about NHLBI BioData Catalyst and how you can explore RECOVER datasets, particularly related to biospecimens. So next slide. Thanks.

So, for those of you who may have missed the first session of the RECOVER webinar series, I’ll briefly introduce NHLBI BioData Catalyst, which I’ll refer to as BDC. BDC is a cloud-based ecosystem of tools and resources to support research. There are various platforms in the ecosystem that support search, analysis, tools, and workflows to democratize data and computational access, all aimed at ultimately advancing science to improve the lives of patients.

Next slide. So today we’ll be covering 2 main parts of BDC. The first, which I will be discussing, is BioData Catalyst Powered by PIC-SURE. This is a search and data exploration platform that researchers can use to find data that supports their research questions. In other words, they can assess the feasibility of their hypotheses. BDC PIC-SURE also supports cohort building and preparing data in an analysis-ready format. And then the second part of the session will be covering BDC Powered by Seven Bridges, which Cera will discuss later.

So, let’s talk about how RECOVER data can be explored on BDC PIC-SURE. BDC PIC-SURE enables cohort building and data exploration through 3 main actions. There’s searching, filtering, and retrieving. So, first, researchers can search for key terms of interest across all data hosted on BDC. So not just the RECOVER datasets, but that is including the RECOVER datasets. So, for example, I can search for specific symptoms related to my research idea, such as head pain or for demographics like age. I can also do searches related to biosamples or biospecimens of interest such as saliva. In BDC, we refer to these participant characteristics or attributes as variables. When you hear me say “variable,” this is what I’m referring to.

Once I find the symptoms, clinical outcomes, or terms of interest, I can apply filters on these characteristics. For example, I could filter for participants who had head pain or for participants over 30 years of age. Then, after applying the filters, I can retrieve counts of participants that meet my filters or meet my query criteria. If I have access, I can actually get that participant-level data to do my research.

Next slide. Our team has created a publicly available version of this tool that anyone can use, and this is because data discovery can be difficult. Complex data can be hard to understand, and oftentimes researchers have a research question in mind, but they aren’t sure how to find the data that supports their idea. BDC PIC-SURE is a publicly available cohort-building tool that anyone can use, so there’s no login and no registration required to use this. This promotes equity to data exploration and allows everyone to explore data available on BDC, including the RECOVER datasets. Today, a little bit later, I’ll be demonstrating what is possible with this tool, which you can do yourself.

Currently in BDC, there are 3 RECOVER cohorts available. There’s the adult, the pediatric, and the autopsy cohorts. I’d like to note that each of these cohorts were described in detail in previous sessions—so either in the previous webinar or in this webinar, so I’ll just briefly cover them here. One thing I strongly encourage all researchers to do, regardless of the data, is that when you’re starting a project and have a dataset in mind is to really dig in and understand how that data is organized, because this will really help you to understand what data is available and plan your research project and research analyses a bit more in advance.

All right. Getting into some of the nitty-gritty here—for the adult and pediatric cohorts specifically, the majority of the variables or participant characteristics are separated by infection status, so whether a participant was infected or non-infected, and months post-index or visit type. It’s months post-index in that adult cohort and visit type in that pediatrics cohort. I’ve included a few examples of this on the slide where saliva variables for those infected at 6 months post-index, and there’s a saliva variable for those non-infected at 9 months post-index.

Now, focusing on the biospecimen data a little bit more specifically, I want to highlight that the biospecimen information on BDC PIC-SURE refers to the information at the sample collection for the participant. So, this means that the variables are not necessarily representative of the current state or the number of samples that are available, but rather what was collected at what time point. For example, the saliva variables I’ve included on this slide, these refer to whether a saliva sample was collected from a participant at these time points.

Let’s take a look at how to use the public version of BDC PIC-SURE to perform a feasibility assessment. For this demo, or for this example, let’s say that we have the following research question in mind: is the virus detectable in samples collected from past participants 6 months post-index? In other words, can we detect the virus in oral and nasal biosamples from participants with Long COVID 6 months post-index? To explore this, we’re interested in the nasopharyngeal swabs and the saliva samples that were collected, which we intend to analyze as part of our research project. So to determine the feasibility of this hypothesis, we want to see how many participants, first, were part of the infected cohort in the RECOVER Adult Study; second, had PASC at 6 months post-index; and then, third, either had saliva samples or nasopharyngeal swabs collected at 6 months post-index.

The next slide, I’ve recorded a demo of the publicly available version of BDC PIC-SURE, which you can follow along and do this as well because like I said, no authorization or login required here. Great. The link was dropped in the chat for you.

Here we can see the page when I first navigate to the BioData Catalyst Powered by PIC-SURE platform. We can see that I am not logged in, but there are still some options I can take to explore data. Let’s use the Discover page to conduct the feasibility assessment. Here at the top, we can see the search bar. This can be used to search for clinical outcomes, phenotypes, and other terms of interest. On the left-hand panel, you can see some faceted search options that can help narrow down the search results. Let’s try to search for PASC.

Behind the scenes in PIC-SURE, this is searching across all datasets in BDC for search results related to PASC. I want to note that this is searching across all parts of the variable for relevant results; the variable name, description, values, and even dataset- and study-level metadata. Here, we see results that look promising, some from the RECOVER adult cohort, some from the autopsy, and some from the pediatric. Let’s narrow down the search results using the faceted search, clicking the “RECOVER_Adult” option under “Dataset.”

Now we can see that the results were updated to only show search results from RECOVER_Adult. These results show several different PASC-related variables. We can see that some are based on different definitions of PASC, including from 2023 and 2024. If we want to learn more about the variable, we can click on the row or the “i” icon to view additional information. If you recall, we are interested in those that had PASC at 6 months post-index, so let’s use the PASC PG2024 at 6 months post-index variable. We can apply a filter by clicking on the “Filter” icon in the “Actions” column. Here, we see the options for filtering that are available. I will select PASC and click the plus button. Here, we can see that there are about 862 participants who had PASC at 6 months post-index.

Next, we are interested in the biosamples. Let’s search for saliva first. With this search, we can see that there are many variables related to saliva collection over time. We are interested in the 6 months post-index variables, so let’s narrow down the search results even further. I can use the “Consortium Curated Facets” section, which was created in collaboration with the RECOVER DRC [Data Resource Center]. Here we can see different months post-index times. I’m going to select the 6 months post-index.

Now we can see that this narrowed down the search results to 6 months post-index variable. I’m going to scroll down to see if there’s something related to my research question. Here, I can see that the saliva infected at zero months post-index variable fits what I’m looking for. Let’s apply a filter on this variable. Here, I’m going to select “Collected” to indicate participants that had this sample collected. Now, we can see that the count went down to about 182 participants. This means that about 182 participants had PASC at 6 months post-index and had saliva samples collected at 6 months post-index.

Now, I can search for the second biosample I’m interested in, the nasopharyngeal swab. Similarly, I can add the filter for the infected at 6 months post-index time. I’m again going to select “Collected.” Now we can see the count went down to 127 participants. This is because this indicates the number of participants that had PASC, had a saliva sample collected, and had a nasopharyngeal swab collected at 6 months post-index, but if you recall the original feasibility assessment question, I’m interested in those that had either the saliva sample collected or the swab collected. We can change our query by turning on the Advanced Filtering feature with this toggle here.

I want to note that Advanced Filtering is currently under development, and so if you have feedback, please let us know using this link here.

We can now see that this display changed a little bit, with some dropdown menu options between the filters. Between the 2 biosample filters, I will go ahead and change that “and” to an “or” instead. Now that count went up to about 261 participants.

This was a brief demonstration of the BioData Catalyst Powered by PIC-SURE public tool. You can now go try it yourself and retrieve aggregate counts based on your feasibility assessments and research questions in mind. These counts can also help support your grant applications if you’re interested in requesting access to the dataset.

Great. I hope that demonstration was useful, and I really encourage you to go and try it yourself. So go check out the BioData Catalyst Powered by PIC-SURE site using the publicly available tool, and you can search for terms of interest that you have in mind, related to your specific research question. Now, if while you’re exploring, you realize that you need some assistance using the platform because this was a very high-level overview of the platform, you can reach out to us using the link that’s on this slide. Using this link will actually submit a help desk ticket, but I wanted to mention that I personally respond to the questions about PIC-SURE, so this is a really great way to contact me directly with your questions and get some more support on using the platform. You can also submit general questions using this link that doesn’t have to be PIC-SURE specific, but my team will direct your question to the right person within the BioData Catalyst Consortium.

Thanks so much for your time and attention, and I’ll hand it over to Cera.

Dr. Cera Fisher:

Thank you so much, Emily. I really appreciate that wonderful demonstration of how researchers can use PIC-SURE for feasibility assessments. My name is Cera Fisher. I’m the program manager for BioData Catalyst Powered by Seven Bridges. I’m going to spend the next few minutes giving a quick overview of this analytical platform and describe how researchers who have gotten access to RECOVER biospecimens and perform lab tests on them can use BDC Seven Bridges to analyze their results alongside the RECOVER study data.

Next slide. Yes. BDC Seven Bridges is an analytical platform that’s part of the BioData Catalyst ecosystem. I want to apologize in advance because I’m going to be using a little bit more technical research jargon than Emily did in her discussion, and that’s because while you can use PIC-SURE without registration and without having an authorized research plan already in place, Seven Bridges is a cloud computing analytical space for researchers who have completed registered research plans.

Seven Bridges does require registration. Access to the data requires that registered research plan. This is because on BDC Seven Bridges, you can access and analyze the participant-level data, and that requires secure workspaces to protect individuals’ privacy. On Seven Bridges, researchers can set up these secure workspaces that allow them to collaborate with researchers across different institutions, while staying inside a regulatory compliant environment. Researchers are able to analyze the data where it lives, rather than downloading it locally and making a copy of it. You’re able to set up your analyses using a graphical user interface, or if you’re more code savvy, you can use an API [application programming interface]. There’s also the opportunity to do interactive data analysis using Jupyterlab Notebooks for Python, RStudio, or SAS Studio, all of which provide high-level statistical analysis options. The platform allows researchers to access top-of-the-line computer hardware through AWS, which is Amazon cloud computing, or through Google cloud computing. There are associated costs for using cloud computing, which are directly passed through, but I will talk a little bit about support to defray some of those costs at the end of this.

Next slide.

I want to take a step back for a moment and think together about the flow of scientific research. BDC Seven Bridges comes into this flow much closer to the end of the project. While the work that Emily demoed for us, a feasibility assessment, is determining how many participants had biospecimen collected that are of interest for your research question before you’re actually even putting together your whole research plan. So after you determine, using that feasibility assessment, that the biospecimens will support your research question, then you will work with the RECOVER program to get access to the biospecimens, and then you do the truly difficult work of actually doing the lab benchwork to get the data out of the biospecimens. Once you have your data, the results of whatever bioassays you’re doing with the biospecimens, then you need to analyze that data and connect it to the participant-level data (i.e., the clinical and demographic data that’s generated by the RECOVER program). Data analysis is where BDC Seven Bridges comes in.

Next slide. For researchers, here’s what this computing environment is going to give you. As I mentioned, you analyze the data where it lives. The data stays within that secure environment protecting participant privacy. You do not have to worry about making sure that your own local computing environment, or your university’s computing environment, meets all of the regulatory requirements for working with sensitive personal data; we have done that work. BDC Seven Bridges also democratizes access to high-powered computing. This is particularly important for researchers who are at small colleges and universities. At many of the big R1 research universities, researchers may have access to a university-owned data center that is in regulatory compliance, but smaller schools don’t necessarily have that.

However, with BDC Seven Bridges, if you are a researcher with an approved research plan, all you need is an internet connection. You can get started analyzing the RECOVER data as long as you can connect to the internet. There is no computer system that you have to manage. You don’t have to install software. You don’t have to download files. You don’t have to get everything in place. You can just log in and start analyzing.

I mentioned that there are costs, but you also only pay for what you use. The most common computing resource is 34 cents an hour. You are renting time on advanced hardware, rather than having to purchase the whole computer. We also have 24/7 technical support, scientific research support, and robust documentation to help you with your work.

Next slide. For researchers who are using RECOVER biospecimens, in order for you to analyze your resulting data alongside the relevant RECOVER data, you’ll need to upload the results of your biospecimen assays to BDC Seven Bridges. An important thing to remember when you’re getting ready to do this is that your data results have to include the masked ID that will be the key that you need to connect each biospecimen’s data to the study participant’s data.

If you’re unsure about how to do that, make sure you reach out for help. There are 2 ways to upload data to the Seven Bridges platform. You can either use the web interface, or for large files, you can use the Seven Bridges command line interface. I’m going to explain very briefly about these 2 options.

To upload your file using the web interface, you would first open up your project on BDC Seven Bridges. From there, step 1 would be to navigate to the files area. Step 2 is to select the “Add Files” button in the files area and select your computer as the place you want to add files from. On step 3, the website will bring you to a page where you can select the files that are on your computer and upload them, either by dragging and dropping them directly onto the page or by clicking a button that will let you browse your computer’s files. I strongly recommend using this method for the smaller-size files that are less than a gigabyte in size, because this is a typical and, hopefully, familiar workflow for any website that lets you upload files from your own computer.

Next slide. Thank you. If the results of your bioassays come in much larger file sizes, we recommend that you use the Seven Bridges command line interface. This is a bit more technical than the web interface. It requires installing the Seven Bridges command line interface software to your computer. Once you’ve done that, step 1 is to open your terminal program where you can perform command line functions. Step 2 is to run a command that will start an upload, and step 3 is to return to the web interface and see that your uploaded files are present in your project’s file area.

Next slide. At that point, if you have also created a cohort with study variables at PIC-SURE and you have that registered research plan in place, you can then import the data from PIC-SURE to your BDC Seven Bridges project space with a provided script. Navigate to Public Resources and then Projects. You would open the project, data export from the PIC-SURE UI [user interface] and navigate to the Data Studio tab. You can copy the data studio analysis to your own project workspace, and you can follow along with this beautiful script that Emily wrote that will teach you how to import the data.

At that point, you can actually unite your bioassay results, from the biospecimens, with the clinical and demographic data collected by the PIC-SURE program in an easy-to-work-with data frame from BDC PIC-SURE. Next slide.

That was a very high-level overview. It was intended to show you what’s possible. For many researchers, moving from analyzing data locally on your own computer to analyzing it in the cloud is a bit of an adjustment. We are here to help. As I mentioned, there are costs associated with using cloud computing, and NHLBI [National Heart, Lung, and Blood Institute] offers new users $500 in pilot fund credits to get started. $500 goes a long way. I know some researchers who have completed their projects within that budget, but most importantly, that means you’re not going to have to use your grant dollars just to figure out how to use this system. BDC also has a YouTube channel with demos and instructional videos. We also have troubleshooting and technical support available 24/7. You can contact our support team by sending an email to support@sevenbridges.com. For live support, Seven Bridges has drop-in-style office hours twice a week where you can chat with our scientists and get not only technical support, but also research support in designing your workflows, accessing data, and using the tools.

Please don’t hesitate to reach out for help and for more information. Thank you so much for being here today.

Dr. Christine Bevc:

Thank you, Emily and Cera. Now we’re going to invite all of our panelists to turn their videos back on and help us answer some of the questions that we’ve been getting from our audience today. As those videos are coming on, if you’ve submitted a question, you might actually see that question answered in the answer tab there. One of the most common ones, just because this is a really dense set of information here, is whether these slides are going to be available from the webinar. As a reminder, the recording and recap of the seminar is going to be posted on the recovercovid.org R3 Seminar series webpage within a few weeks. Took us about 2 weeks, 3 weeks to get the last session on, but you can also request a copy of the slides with an email to recover_acc@rti.org, and that’s in the answered questions as well.

All right. To kick us off, we’ve got another really common question that, Jim, you did a great job of answering this around the process for requesting biospecimens and samples for the autopsy cohort. Now, I want to pass it over to Maddie and Trisha to answer that same question for the adults and the pediatric cohorts for that. And just as a lead-in there for that, as a reminder, there’s a PDF that was shared, and I think Kate might be able to drop it in there, about the 2 pathways for requesting those biospecimens through the ROA and the ancillary study. So, there’s more information that Jim mentioned. Once you have your research plan, what are some of the next steps there? Maddie, Trisha, can you help shed some light there?

Madeleine Thorn: 

Yes, of course. Thank you, Christine. So yes, as Christine mentioned, you’ll have to first go through an application process to get your proposal, and once that is approved, that’s when you come to the DRC to speak with us to figure out how to set up samples and what samples you are looking for, for your specific study. With that, as Jim nicely laid out for us, is we go through a whole phenotyping process. Once you go through everything, we will work on phenotyping to define your symptom profiles that you’re looking for. As Christine mentioned, you’ll already have a proposal nicely ready to go, and so our main focus is just getting us on the same page of how your goals will align with the RECOVER samples and what we have available for you, and what we can do for you to help you get the samples to do the study that you’re looking to do.

With that, we will do a phenotyping process to make sure everyone’s on the same page, and that’ll definitely include information on figuring out the cohorts and understanding the study design, and that will be a large part of the application process as well. From then on, we will go through a step-by-step process of looking at sample availability, and that’s where the Biospecimen Access Committee that has been mentioned comes into play, as one of their main goals is making sure that we have an equitable distribution of samples as we definitely want samples to go out and we really, really want them to be used for studies. We also want to make sure that we’re not erroneously using them or just distributing more sample than is deemed necessary for specific studies. We definitely want you to meet your minimums and be able to conduct the research that you want to do, but we also don’t want to make sure we’re giving excess amounts in any sort of scenario.

And then from then on, once we have a good guideline set up and once we figure out the right amount of samples going through and we have stamps of approval that go along the way, then that’s when we shepherd off your request to the PBC [PASC Biorepository Core], who will then begin the process of doing sample pull and prep for your sample so that can then be shipped onto you. That’s the process of just making sure that things are aligning with the RECOVER study and making sure that we are answering your study questions in the way that our samples and data are collected. That applies for both adult and pediatric cohorts, and similar to autopsy as well, adult and peds just resemble each other a little more, so that’s why we lump them in together.

Dr. Christine Bevc:

All right.

Madeleine Thorn: 

Thank you, Christine.

Dr. Christine Bevc:

Yeah, my next question is a bit of a follow-up to that one because it’s critical for planning for investigators, and that sounds like a lot of steps there. What’s the timeline for accessing these samples? Just setting a reasonable expectation for investigators. As they’re planning, as they’re looking ahead, how long does all of this take? How much time should they plan for?

Madeleine Thorn: 

Right, definitely. Thank you. This is a very important question to go on, and we understand everybody has timelines, and so this is a great question.

I will say, as far as defining the pick list, this part can take up to 12 to 16 weeks. Defining the pick list, what I mean is that’s the phenotyping and figuring out your symptom profiles and getting everything together. This is a very highly iterative process where you’re going back and forth between your team and the DRC, and so we are just trying to go back and forth and making sure, again, we’re all aligned at the end of the day because we don’t want to be giving you samples that you’re not looking for. This definitely depends on the complexity of studies and also just response times, honestly, and making sure everybody works back and forth together well. We realize that there are so many obstacles and hoops that we have to jump through when it comes to timelines, so that’s why we want to give a rough overall up to 12 to 16 weeks just to set expectations.

And then from there, once we have things defined, that can happen a little bit more quickly. That process of the BAC approvals that we were talking about, the Biospecimen Access Committee, and then also PBC fulfillment of samples and sample shipments takes around 8 to 10 weeks. And so that’s more of a set time, and it’s a little bit less iterative, but that one’s a more defined timeline. Overall, you’re expecting, I would say, sorry, I’m trying to do math, 20 to 39-ish weeks as a very high-level view of the maximum amount of time that you could expect. I see Jim unmuted as well, if he wants to touch on that.

Dr. Jim Stone: 

Yeah. I was just going to add, after all that’s done, a key issue is that your institution will have to execute a material transfer agreement of some type, and that can happen very quickly at some institutions, or it might take months at other institutions. So, you need to also understand how long it will take your institution to execute that material transfer agreement.

Dr. Christine Bevc:

They’re all really great points. And so, for investigators that are coming online, I think we want to reiterate also, and I’m going to switch hats real quick here, is that once your project is approved, that you will receive support and guidance along the way. So as tempting as it is to reach out to Jim, to reach out to Maddie, to reach out to Trisha to say, “Hey, I want to be able to find out how many samples are available or how long it’s going to take,” or “Hey, can I get ahold of these biospecimens?” We really want to encourage you to go through those pathways. Taking that step back, going over to PIC-SURE and also registering with BDC.

Emily, Cera, what do you suggest for investigators who are just starting and going, “All right, I have an idea,” what’s the next step that they should do? What’s the next action item for our attendees today?

Emily Hughes: 

Sure. Yeah. If you have a research question in mind and aren’t exactly sure what data is available or how many participants might be available for your analysis, you’re trying to gauge sample sizes to plan your research a little bit, I encourage you to go over to BDC Powered by PIC-SURE, and using that publicly available tool, you can search for terms that you’re interested in. I really encourage folks to try a couple of different search terms. I know maybe during the last seminar, I think it was, somebody had asked about olfaction tests, but perhaps searching for “smell,” or “nasal,” or “nose,” or other related or similar terms can help you to find maybe some of the variables or the specific data points that you’re looking for. I think that would be one step.

I’ll hand it over to Cera to see if she has anything else.

Dr. Cera Fisher:

Yeah. I would also say that as you are working on preparing your application materials for your grants or your biospecimens, we ... Sorry, my mind went blank.

We have available cloud use statement language on the BioData Catalyst website. We have good, approved language that you can use to include in your grants to say that you’re going to be using BioData Catalyst. We also provide support for estimating cloud costs. Estimating cloud costs is really hard the first time you do it, but it’s not the first time we’ve done it, and so we have assistance in doing that. I am also going to pop in the chat here real quick that we’re having a minimizing cloud cost presentation on January 29. Working in cloud computing is, as I said, kind of downstream, but none of us have any desire for you to spend more of your research dollars on cloud costs than is necessary. So please let us help you with planning that.

Dr. Christine Bevc:

Okay. Great. Yeah. Feel free. Yeah. Cera’s dropped that link in there so you can register for that. Our next question that we have actually comes from one that was submitted, and we’ve got a lot of ongoing studies currently. The question is: is the data generated by different researchers available as new variables in the datasets? If someone checks on the presence of SARS-CoV-2 and saliva, as Emily’s demonstration showed, are those results going to be public and linked to the samples that were used? Let me start with Jim and then let others join in.

Dr. Jim Stone: 

Our goal is to try to make that happen as much as possible, meaning we would love for every time someone does an assessment on a particular biosample, for that to become eventually integrated into the database for multiple reasons. For one, so you don’t waste your time repeating it and your money, but also so we can synergize and maximize the impact and compare it with all the other studies that are being done on biosamples, possibly on the same patient. It’s a little difficult with some of the ancillary studies because you aren’t necessarily obligated to give patient-level data back to RECOVER. We haven’t made that an obligation, but we are encouraging that. I will say it looks very good in your application if you say you will do that, and you’re open to that, but it is not currently a requirement, but it looks very good to say that.

Madeleine Thorn: 

Yes. Thank you, Jim. I can also talk about adult and peds. It’s a little bit different where we have a wider database, a lot more participants. One caveat that we first want to mention with sharing with the different samples is if you, for example, if there’s a study that does the nasal swabs, I think was the example that they use where, for example, there are about 14,000 to 20,000 adult participants. One given study might be doing research on, at maximum, maybe around 300 to 500 participants, and so you’ll have a very small percentage of that data available. There’s a lot of nuances with sharing data. However, we do have a workflow that we’re trying to work on where investigators can share samples, like how Jim was promoting, where it would be great if we can share data with each other, and we want to keep this as collaborative as possible.

The variables will not be made public and will not be added to the dataset. However, through BDC, we have a workflow ready to go if people want to share with other investigators within that platform and see the assays, and the results that other people were getting, just in case there might be participant overlap. Thank you. Sorry about the technical difficulties.

Dr. Christine Bevc:

No, no. We’ve just got a few more minutes left for today, so I’m going to do one quick question. For those that are preparing their study design, what’s the process to determine if a proposal sample size or power considerations are sound and warrant the use of essentially these finite resources that RECOVER has to offer?

Madeleine Thorn: 

Yes. For sample size, it’s mostly a scientific question, as James was just saying off camera, but when people submit their application, there are different review committees who will be going over that. So, there’s not necessarily a very standardized way. We don’t require a threshold for sample size or power, and so that’s just really application dependent. What I’ve noticed is when people are able to justify that and explain their study, there haven’t necessarily been limitations when it comes to somebody’s specific sample sizes. We’ve had sample sizes from 3 to 1,000. So, if there’s the justification available, and if you’re able to explain that and how your power analysis will still work with any given request, then I think that’ll be okay. But again, I’m not part of the review process, but there are review committees who go over the scientific merit and also the statistical significance of the requested amount of samples.

Thank you.

Dr. Jim Stone: 

Yeah. That’s all spot on. I would say we’re open to both pilot analyses and very thorough statistically significant studies. We do not reject a study that’s underpowered, that’s labeled as a pilot analysis that’s trying to get data for a grant application, for example; we are open to that. But the maximum, at the higher end, it really depends on a combination of the potential impact of the work, how it impacts other studies, what’s the feasibility of actually being able to complete the work, being able to afford the studies that are being proposed. So those all come into play.

Dr. Christine Bevc:

All right. Well, thank you all. As expected, we have run out of time to answer all of the questions, but the expertise and the knowledge that is brought to us by our panel today is going to be used to answer any of the remaining questions that were not answered today.

I want to thank our audience for joining and engaging in today’s session. As I mentioned, the recording is going to be available on recoverCOVID.org in about a week or so. We also post the summary of the Q&A documents and transcript, and both of those will include responses to the questions that we received today. You’ll see the short survey that’s popped up on your screen that asks for feedback about today’s seminar as well as future seminar topics you’d like to see. One of the questions that we got was if you were interested in seeing a similar session about the clinical trial biospecimens, or other topics, more about BDC, or any other interests that you might want. This is a great way for our planning committee to be able to learn how this series can best help you.

We thank you for joining us today and hope that you’ll join us again next month. Be sure to check the R3 Seminar webpage for the updates, the information, and registration. All of those for next month’s session are going to be opening shortly, so please stay tuned there. If you have any other further questions, you can drop those into the poll there and reply to any of the emails that you’ve received for our registrations.

Thank you so much for taking the time to complete that survey. We hope that you will all have a great rest of your day and a great rest of your week. Thank you.