BRAINS FAQs

It will be federal reviewers - there will never be non-federal reviewers on these evaluations. 

ARPA-H will not seek ownership of the IP developed.    

Data rights are determined on a program/project basis. There is not an overarching policy. 

The NIH definition of an Early-Stage Investigator (ESI) is a Program Director/Principal Investigator who has completed their terminal research degree or end of post-graduate clinical training, whichever is later, within the past 10 years. For the purposes of BRAINS we are not strictly defining or requiring that applicants be ESIs, rather we are encouraging them to apply. Any institution that can put forward a competitive proposal will be considered. 

There is no limit on the number of non-overlapping applications an organization may submit. However, under the BRAINS program, ARPA-H anticipates making no more than one award per proposing organization. 

Absolutely. This would absolutely be considered early-stage entity. We are certainly not requiring a proposer to even be near ready for toxicology studies.  In fact, we are not currently aware of any mature or near-mature products that would already meet the BRAINS program metrics or be ready for toxicology studies. If such products do exist, we would be very interested to hear about them. 

We anticipate that most, if not all, applicants will be at a very early stage in developing the kinds of products we are seeking. That said, this does not mean the program is limited to early-stage companies or startups; rather, it means the product itself is likely to be at an early stage of development.  

Important to note - much of what is currently available is cell-based rather than tissue-based, and some approaches are described loosely as ‘organoids’, which, without further innovation, or at current state-of-the-art, would not meet the program’s metrics. 

 Assays can include in vivo work, just no work with human fetal tissue.  

We view these TAs as sufficiently challenging and substantial to stand alone. As a result, we will not give preference to teams that propose both TA1 and TA2 over teams that propose only one or the other. This approach also applies to Phase 2. That said, in Phase 2, for successful delivery of the graph, it may be easier and a more robust proposal, if the team includes - whether within the same institution or through collaboration - a partner (if not themselves) who can provide source tissue. 

Species are not specified; however, the goal is for translation of the results to humans in the not-so-distant future. The closer you are to human the more similar you’ll be to mimicking normal human neurodevelopment, and therefore results are more likely and readily translatable. 

Non-invasive approaches of that kind would be outside the scope of BRAINS. BRAINS is specifically focused on tissue-level repair, which, based on current understanding, is not achievable with non-invasive approach(s). Accordingly, neuromodulation or other non-invasive strategies would not be considered in scope unless they can demonstrably achieve tissue-level repair. 

We are open to a range of methods for demonstrating connectivity, including approaches for tracing axons and dendrites, as well as newer high-resolution methods such as viral library-based techniques. The key expectation is that connectivity should be assessed against an appropriate stage-matched benchmark for the corresponding tissue in a normal brain. At a minimum, the evaluation should include a normal (age-matched) control, as that comparison is the most important requirement. Beyond that, we are open to different available methods that can make a strong and well-justified comparison. 

 We do not have a preference for either acute or chronic injury models. Teams should choose the time frame that is best justified by their overall approach. If you are designing your work with a specific clinical indication in mind, it may make sense to emphasize a timeline that is more clinically relevant. On the other hand, if your primary goal is to establish proof of concept, you could reasonably argue that an earlier post-injury time point is more practical for demonstrating initial feasibility, with later work extending to more chronic injury states. In either case, the selected approach should be well justified in the broader translational context. 

We are open to either approach. Validation against relevant existing human fetal reference data may be appropriate, but teams should not propose collecting human fetal tissue for this effort as it is unallowed. At the same time, engraftment can also be a useful assay within TA1 to help demonstrate that the generated cells, cell types, or tissue construct are appropriate. If engraftment is used in that way, it does not mean the full TA2 surgical implementation requirements or milestones must also be met within TA1.  

Molecular profiling approaches such as genomics or spatial transcriptomics may be valuable components of characterization, but they would not, by themselves, necessarily be sufficient. Teams should make a strong case that their validation strategy meaningfully demonstrates the developmental appropriateness and readiness of the precursor tissue. 

We are primarily focused on teams achieving the program milestones as quickly and effectively as possible. We are aware that substantial existing human developmental datasets are available, and we strongly encourage teams to leverage those resources to the greatest extent possible. These datasets can provide an important foundation for early natural-tissue characterization and precursor-tissue design. 

At the same time, existing resources are not fully exhaustive - for example, they may not capture every developmental stage or specific time point with the level of resolution needed for a given approach. Accordingly, while the use of these datasets is absolutely encouraged, teams should also anticipate the need for complementary validation using other experimental approaches. In particular, proposers should be cautious about relying on computational analysis alone to satisfy a milestone.  

In short - existing datasets can strongly inform and support the work, but milestone completion should be backed by appropriate biological validation. 

We are certainly open to innovative methods that can guide cells down the appropriate lineage and give rise to the right complexity of tissue. In the call, we use the term “mimetics,” a somewhat novel way of describing how iPSCs, as they differentiate, can be patterned into more structured tissues. That is one possible approach, complementary to strategies in which iPSCs are differentiated into the relevant precursor cell types and then assembled at that stage. There could also be intermediate approaches between those two ends of the spectrum. We are open to all of those possibilities. 

Viral methods and genetic modifications to cells can certainly be useful for facilitating assays. However, if they are used not simply as reporters but as mechanisms of control, I would caution that there is a risk that the resulting process may not recapitulate normal development and therefore may be at high risk of not producing normal tissue. There is also a translational risk: you are taking a product that is already complex and making it even more complex through additional modifications and viral delivery methods. So while we are open to these approaches, they do come with caveats that we will be paying close attention to, and proposers should keep that in mind. 

Looking at connectivity, projection patterns, or the targets of those neuronal projections are all important ways of assessing how well you are recapitulating normal development, in order to produce a normal tissue. All mentioned approaches are encouraged. 

Yes, potentially - but not on their own. What is being pursued here is highly challenging, and it may ultimately be necessary to complement the core tissue-repair approach with additional technologies to achieve optimal outcomes in graft maturation, connectivity, and function. However, it is not yet clear at the outset whether such complementary approaches will be needed, or to what extent. 

Because a tissue-repair approach that closely mimics normal development has not yet been (widely) established (to our knowledge), proposals should think carefully about whether and why these kinds of enhancements are being included. If such an approach is proposed, it should be supported by a strong rationale and make clear why it is justified at this stage, rather than being added speculatively.

Not per se, no, unless there's a way you can get mesenchymal stem cells to differentiate into the brain tissue type that you're focused on with all its complexity in terms of cell types and architecture. As far as we know that is not possible.  

Not in and of itself. Unless you could argue that the scaffold will facilitate generation of a new structured part of the brain in situ with all its cell types and cytoarchitecture. If you have a scaffold that does that, please submit. From our understanding that is unlikely to exist. 

We have seen a lot of pre-clinical studies with positive effects on behavior, recovery of behavior after damage, but that has not translated to humans. Many types of cell-based therapies have been tried clinically and fail based on these pre-clinical positive results. Likely the positive results are due to the bystander effects or paracrine effects where mouse brains are small and the paracrine effect can be potent.  

The spirit or goal of BRAINS ET is to come up with approaches to generate properly structured tissue with the right cell types.  

Organoids are generally incomplete in their cell types and abnormally structured… we are not excluding them, but they have significant drawbacks that make them less attractive unless you can propose improvements of the state of the art.

 We are open to any tech that can be used to generate the cells needed for making the precursor tissue. So if that tech is part of your strategy to differentiate IPSc, we would be open to it, but it wouldn’t be sufficient on its own to meet the goals / metrics of the program. 

Within the 18-month exploratory topic we do not expect to have functional readouts relevant for moving to the clinic, even though ultimately, we are headed in that direction. For the 18 months - we would be looking more at measures such as survival and maturation of the precursor tissue, along with integration – for example, whether the graft sends projections to the appropriate targets and receives inputs from the appropriate parts of the brain. 

We are open to both - whatever you think will best recapitulate normal development. The closer you can recapitulate normal development the higher the likelihood that as the tissue matures it will function normally. We are open to either engineering approaches or approaches that allow development to follow its normal course from iPSCs. 

That depends how you define organoids… state-of-the-art organoids as published don’t meet the standards for precursor tissue that we are looking for here. It's not out-of-scope, but for example, organoids that are spherical and can be delivered through needle easily would lack pattern and orientation. Delivery approaches for tissues with natural structure and orientation may be different. It matters how you put them in. With an organoid-based approach you have to make a strong argument about why it would be feasible to meet the metrics and milestones. 

In deciding the stage of precursor tissue, there are several factors. If the tissue is too mature we know mature neurons do not integrate as well or project as well in the adult environment as less mature neurons do. So, maturity can work against you when developing a precursor tissue. But maturity can work for you in ease of working with or grafting the tissue. If you are looking at very early brain pre-cursor tissues, they are exceedingly soft and hard to manipulate. Ideally you argue why your selected maturity state is best… we are not prescribing that.  

One that is indistinguishable from its natural counterpart. 

Part of integration milestone is the cytoarchitecture; it is hard to imagine how a mix of dissociated cells would form that in the absence of normal developmental processes. If you can argue that will happen then it can be considered, but at the outset, it seems unlikely. 

Meeting the milestones as described in the BRAINS ISO. They are not there by accident. They are the steps we believe are needed to get to the clinic with a product that can repair tissue-level damage by regrowing new tissue for that part of the brain. 

The latter- the intention is to transplant intact pieces of tissue. Modifications to support engraftment may be acceptable, if they are unlikely to interfere with the natural and normal development of the tissue.