Why the Traditional Approach to Evaluating Gene Editing is Not Your Only Option

Most organizations evaluating gene editing are still operating with outdated development and collaboration models. The traditional approach has long been framed as a binary choice: build capabilities internally or commit to large-scale R&D partnerships centered around a single technology or editing modality.  ElevateBio offers a third path: one that doesn’t require organizations to choose between the two options before the science is ready to support that decision.

Gene editing shouldn’t be based on a single platform decision or early-stage commitment. What’s needed is a framework that lets organizations evaluate gene editing opportunities with clear go/no-go decisions: generating evidence at each stage before deciding whether to invest further.

This shift reflects a broader reality in biotechnology. There is no single editor, delivery system, or development strategy capable of addressing every therapeutic challenge. Yet, traditional licensing structures and large-scale R&D collaborations still assume a level of upfront certainty that doesn’t match the complexity of gene editing development. What is needed instead are more iterative, decision-driven frameworks that reduce uncertainty step-by-step. This model is what ElevateBio replaces.

ElevateBio’s partnership approach is structured around three sequential questions:

Each stage addresses a distinct scientific decision point, without requiring large-scale upfront commitment, allowing organizations to follow the science.

---

01 Proof of Concept

Can this target actually be edited, and with which approach?

Not all disease targets are created equal. The underlying biology – the type of mutation, the tissue involved, the gene's expression pattern, and how precisely the edit needs to be made – determines not just which modality might work, but whether the target is editable at all. A target accessible to one approach may be out of reach for another, and the margin between a viable edit and an ineffective or unsafe one can be narrow.

When looking for the right modality to make an edit, organizations are faced with a wide set of technologies to choose from: more traditional approaches like nucleases and base editors or newer modalities like epigenetic editing and targeted gene insertion. Each approach carries different strengths, constraints, and biological tradeoffs, and the best choice isn’t always obvious. And historically, this has meant figuring out whether a target is even feasible has required immense investment into building internal capabilities or entering a large-scale R&D collaboration – before there is sufficient evidence that the target is viable at all.

Instead, what is needed at this stage is a clear, data-driven answer to whether a viable editing strategy exists. And being able to determine that without long-term structural commitments.

A proof-of-concept approach is designed to address this gap. Rather than committing to a single modality upfront, multiple editing strategies can be evaluated in parallel against the same target to determine whether effective editing is biologically achievable. The goal is not to advance a program into development, but to generate a clear go/no-go decision supported by experimental data.

At ElevateBio, this approach is operationalized through Proof of Concept partnerships that deliver 1–3 optimized gene editors for a given target for in-house evaluation, along with supporting characterization data, in under six months. The emphasis is on speed and clarity: providing a defined experimental output that allows organizations to determine whether gene editing is a viable path forward before committing to larger investment decisions.

What this stage answers

1. Can the target be edited?
2. Which modality is most promising?
3. Is there data to support continued advancement?

---

02 Clinical Candidate

If the answer is yes, what does the path to the clinic require?

Once a target has been shown to be editable, the question is no longer whether gene editing can produce a biological effect, but rather, can that effect be translated into a safe, clinically viable therapeutic candidate. This transition introduces a different set of challenges, including potency, specificity, delivery, manufacturability, and regulatory expectations.

This is also where many traditional partnership models can fall short. Development activities such as editor optimization, delivery engineering, and regulatory planning are often distributed across multiple providers. Decisions made in one domain, can significantly influence outcomes in another, yet these decisions are often made in isolation.

ElevateBio’s Clinical Candidate model replaces this fragmented development structure with a fully integrated, decision-driven workflow. It brings together all the components required to develop a clinical candidate – editor engineering, delivery, translational testing, development planning – within a single coordinated system. The result is an editor ready for development candidate, optimized for safety and potency with IND-supporting data. Our integrated model helps you transition from discovery into the clinic, while maintaining the discipline of a data-driven staged approach to development.

What this stage answers

1. Can the edit be optimized for therapeutic development?
2. What delivery, specificity, potency, and translational requirements must be addressed?
3. What would it take to advance toward a development candidate?

---

03 Pipeline Model

What does this mean for a broader pipeline strategy?

For organizations managing multiple programs, the challenge eventually shifts from optimizing a single target to building repeatable success across an entire pipeline.
Historically, this has been difficult to achieve. Gene editing programs have relied on a network of specialized partners across discovery, delivery, development, and manufacturing. While this approach can support individual assets, it often creates fragmented execution, where learnings from one program are not easily applied to the next.

Once an organization has validated its editing approach at one target and advanced a candidate toward the clinic, a natural question follows: can this be repeated, faster and in parallel across more targets? Insights generated in earlier programs – like editor design, delivery engineering, and translational insights – become transferable across programs. A pipeline approach is built for this stage, where after learnings from earlier programs compound into the next, made possible by a single organization with end-to-end capabilities under one roof.

At ElevateBio, this is reflected in our Pipeline Model partnership option that evolves from program-by-program development with a portfolio-scale, continuously learning system. These partnerships provide access to the full gene editing toolbox, including AI-enabled discovery, editor engineering, and delivery optimization, applied across multiple programs in parallel. Rather than treating each program as an isolated effort, the model is designed to enable continuous iteration of gene editing solutions across a portfolio, where insights from one program directly inform the design and performance of others.

This approach differentiates ElevateBio by combining integrated access to a full gene editing platform with a long-term pipeline commitment and portfolio-scale collaboration that emphasizes continuous discovery and iteration, rather than one-time delivery of individual assets.

What this stage answers

1. Can learnings from one program accelerate others?
2. How should gene editing be integrated across a broader pipeline?
3. What partnership structure supports portfolio-scale execution?

---

The Real Shift

For years, organizations have been asked to make early, high-stakes decisions between building internal capabilities or committing to large external collaborations, often before there is sufficient evidence that a given target is tractable, or that gene editing is the right approach at all.

That model is no longer sustainable. With increased pressures to invest R&D dollars wisely, development paths with clear go/no-go decisions are nonnegotiable. Separating feasibility, therapeutic strategy, and investment decisions to be addressed sequentially, rather than collapsed into a single upfront choice, gives teams the best chance to maximize the chance of therapeutic success. This is why we created flexible partnership models with three distinct paths to therapeutic progress: Proof of Concept, Clinical Candidate and Pipeline Model

Downstream decisions about program direction, development strategy, and partnership structure become clearer, faster, and more grounded in data.

The future of gene editing will not be defined solely by improvements in editors, delivery systems, or manufacturing capabilities. It will also be defined by whether organizations adopt frameworks that allow them to evaluate opportunity before they commit to it.