The ROI Case for External Control Arms: Why Not Investing in One Is the Riskier Decision

If your biotech is developing a rare disease therapy or an advanced modality and you haven't committed budget to an external control arm, you are not saving money, you're stacking up risk.

Regulatory risk, commercial risk, the kind that compounds quietly until it shows up at exactly the wrong moment. A major interaction with the FDA, an HTA submission that falls apart or at a board meeting where someone asks a question your data can't answer.

The conversation around external control arms (ECAs) has been stuck in a narrow frame for years. An external control arm is a comparator group drawn from real-world data or historical studies rather than from concurrent randomisation within the trial itself. Despite their growing importance, ECAs still get treated as a regulatory workaround, something you fall back on when a randomised controlled trial isn't feasible. That framing sells them short. Badly.

A properly conducted ECA is the single highest ROI evidence activity a biotech can fund. And the case for that claim keeps getting stronger.

You were never getting an RCT anyway

Let's be honest about the starting point. If you are working in ultra-rare diseases where patient populations number in the hundreds, or developing gene therapies and CAR-T products where single-arm designs dominate, the traditional RCT was never really on the table. Randomising a handful of children with a devastating genetic condition to placebo is ethically fraught and practically impossible. Everyone knows this, regulators included.

The FDA's 2023 draft guidance on externally controlled trials says as much, explicitly recognising that certain disease settings make concurrent controls infeasible [1]. The EMA's reflection paper on single-arm trials, which we unpacked in detail previously, makes a similar concession beneath its firm preference for randomisation.

So here is the problem. Too many biotechs run their single-arm trial, collect their data, and then scramble for comparative context after the fact. We have written before about what happens when regulators start asking pointed questions about a single-arm submission and it is rarely a comfortable experience. The reactive approach works sometimes, but it is slow, it is expensive, and it leaves value on the table at every other stage of the programme.

This post is about the proactive case. Plan your external control arm before your pivotal trial reads out, fund it properly, and it will pay for itself several times over.

FDA and EMA are not just tolerating ECAs, they are building frameworks around them.

Something has shifted in the last three years. I'd argue it deserves more attention than it has received.

The FDA's 2023 draft guidance didn't simply acknowledge that ECAs exist. It went further. Specific expectations for design and conduct: pre-specified protocols, feasibility assessments, comparability across multiple domains, blinded outcomes adjudication [1]. We broke down the full checklist in a previous post. The key signal is that the agency wants sponsors to treat ECAs with the same methodological seriousness as a randomised comparator arm. That is a high bar, but it is also an invitation.

The EMA is moving in the same direction. Concept paper published in 2025, dedicated workshop in November, CHMP discussion expected Q4 2026 [2]. The MHRA has started engaging as well. What caught my attention was the Clinical Research Data Sharing Alliance (CRDSA) analysis from October 2025 [3], which found all three agencies converging on the same core principles: pre-specification, transparency in methods, early scientific engagement. As of early 2026, three major regulators across three separate processes have reached essentially the same conclusions on external control arms. That does not happen by accident.

What does this mean for the governance conversation at your company? It's pretty simple. Investing in an ECA is not a speculative bet on regulatory acceptance. The agencies are telling you, in writing, what they expect. The uncertainty is lower than it has ever been.

SCHOLAR-1, SCHOLAR-3, and Zolgensma: how ECAs have already changed outcomes

Theory is useful. Precedent is better.

SCHOLAR-1 and YESCARTA. Before axicabtagene ciloleucel could be evaluated as a CAR-T therapy for relapsed/refractory diffuse large B-cell lymphoma, someone had to quantify just how grim the outcomes were for these patients on standard of care. SCHOLAR-1 did that. Crump and colleagues pooled retrospective data from two phase 3 trials and two observational cohorts, establishing a 26% objective response rate and a median overall survival of 6.3 months for patients with refractory disease [4]. That single dataset changed what the ZUMA-1 single-arm results actually meant. They went from "high response rates in a single-arm trial" to "a 73% reduction in mortality risk compared to the best available alternatives." Five-year follow-up data from ZUMA-1 later confirmed durable responses [6]. Without SCHOLAR-1, YESCARTA's story would have been told very differently to regulators, to NICE, and to the physicians making treatment decisions.

SCHOLAR-3 and TECARTUS. Kite took the same playbook into B-cell acute lymphoblastic leukaemia. SCHOLAR-3 used propensity score matching to compare ZUMA-3 patients treated with brexucabtagene autoleucel against matched historical standard-of-care cohorts [5]. Median overall survival exceeded 25 months for the ZUMA-3 arm versus less than 6 months for the external control [10]. More than four times the survival. That comparison played a key part in the strategy of both approval and launch of TECARTUS in this indication.

Zolgensma. The gene therapy that set the template. Novartis secured approval for onasemnogene abeparvovec in infantile-onset spinal muscular atrophy using natural history data from the Pediatric Neuromuscular Clinical Research (PNCR) cohort of 23 patients and a NeuroNext cohort of 16 patients as external comparators. The FDA clinical reviewer noted that the natural history of infantile-onset SMA follows "a relatively predictable course that can be objectively measured and verified" and that the treatment effect was "large" and "readily ascertained" [7]. A broader analysis of FDA decisions found that 80% of approvals supported by external control data between 2000 and 2019 involved rare diseases [8].

Same pattern across all three. The ECA was deliberate, pre-planned, methodologically serious. And it served the programme at every stage, not just the regulatory filing.

What ROI does an external control arm deliver beyond the regulatory submission?

Here is where the ROI argument gets interesting, because the regulatory submission is actually the most obvious use case for an ECA. Possibly not even the most valuable one.

1. Regulatory interactions. This is the one everyone thinks of first, and fair enough. An ECA gives you comparative context. Walk into a Type B meeting or scientific advice session with a pre-specified ECA protocol and the conversation changes. You are not asking the agency to take your single-arm results on faith. The FDA guidance explicitly encourages early engagement on ECA design [1].

2. De-risking the development programme. Building an ECA forces you to do something that too many development teams skip. You have to deeply characterise the disease you are targeting before your pivotal data arrives. What does the standard of care actually look like in practice? What are the real-world progression rates, what prognostic variables drive outcomes? In my experience, this intelligence surfaces issues with trial design, flags potential safety signals, and stress-tests your assumptions about treatment effect size before it is too late to act on them. If the ECA data suggests your expected effect size is marginal against the comparator, you learn that before your phase 3 read-out. Not after. This is why an external control arm should be part of any serious integrated evidence plan.

3. Investor and governance communications. A single-arm trial generates data that is hard for non-specialists to evaluate. "Our therapy achieved a 71% complete remission rate" is a number without context. But "our therapy achieved a 71% complete remission rate versus 26% for the best available standard of care, with median survival exceeding 25 months compared to less than 6" is a story that a board member, an investor, or a potential acquirer can understand and act on. We have written before about how RWE transforms investor narratives. An ECA is the sharpest version of that capability.

4. HTA and market access. Payers need comparative effectiveness data. Full stop. A review of HTA submissions across four European countries found that ECAs are increasingly submitted but acceptance remains limited [11] which is precisely the argument for getting the methodology right from day one, not an argument against doing one. Putnam's analysis makes a similar point about the growing strategic role of ECAs in HTA [12]. Here is the practical upside: the data you generate for a regulatory ECA can be repurposed for your NICE, G-BA, or HAS submission. Wait until post-approval and you lose months. You also negotiate from a weaker position, which tends to show up in the price.

5. Medical affairs and commercial strategy. The data that sits behind an ECA; disease burden quantification, treatment patterns, outcomes benchmarking, patient journey mapping  does not vanish after the regulatory filing. It feeds your publication plan, informs KOL engagement, provides ammunition for formulary discussions. Done well, an ECA becomes an evidence asset that the medical affairs team will draw on for years. I have seen teams treat ECAs as a single-use regulatory deliverable, file the submission, and never look at the dataset again. Genuinely painful to watch when you know what that data could have done for them.

Add those five together. The maths is straightforward. A properly conducted external control arm pays for itself through the regulatory submission alone, and everything after that is upside. Not investing in one when you are running a single-arm programme in a rare disease or an advanced modality is the genuinely expensive decision.

What separates a regulatory-grade external control arm from a desk exercise?

A quick word on execution, because none of the above applies if the ECA is done badly.

When should you start building your external control arm? Before your pivotal trial completes. Pre-specification is non-negotiable. Both the FDA and EMA have made this explicit: an ECA protocol and statistical analysis plan should exist before your main trial completes, ideally before it begins [1][2]. Selecting your comparator data after you have seen your results is exactly the kind of post-hoc rationalisation that regulators will reject. And they should.

Data source selection matters enormously. A registry exists for your disease? Great. That does not mean it is fit for purpose as an ECA data source. Prognostic variable capture, endpoint comparability, temporality, missingness all of these need rigorous assessment against regulatory-grade standards. The FDA's guidance calls out multiple specific domains of comparability that must be evaluated [1].

Then there is the team. You need both methodological depth and disease-area expertise. Propensity score matching and sensitivity analyses require real statistical sophistication. But understanding whether a particular prognostic variable actually matters in your disease requires clinical knowledge. Getting an ECA to the standard where it holds up under regulatory scrutiny, and then pulling value from it across HTA, commercial, and medical affairs, requires both. This is where working with a partner who brings scientific rigour and strategic perspective together makes a real difference. At Inovia Bio, this is core to what our team does: designing ECAs as multi-purpose evidence assets, built to regulatory grade and delivered at speed.

For the full technical checklist on FDA expectations, our previous breakdown covers the detail.

The question is not "should we do an ECA?" It is "can we justify not doing one?"

For any biotech developing a rare disease or advanced therapy asset on the back of single-arm data, the strategic case for an external control arm is about as close to obvious as drug development gets. Regulators are formalising frameworks for them. Precedent-setting approvals in CAR-T and gene therapy have shown their value. HTA agencies increasingly expect them. And the evidence they generate serves every function in your organisation, from clinical development through to commercial launch.

So the next time someone in a governance meeting asks why the programme should fund an external control arm, flip the question. What is the cost of not having one? What regulatory questions will you be unable to answer? What comparative data will your HTA submission lack?

When you do the maths honestly, the ECA is often the highest-ROI evidence investment you will make.

 References

1. FDA. Considerations for the Design and Conduct of Externally Controlled Trials for Drug and Biological Products: Draft Guidance for Industry. February 2023. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/considerations-design-and-conduct-externally-controlled-trials-drug-and-biological-products
2. EMA. Development of a Reflection Paper on the Use of External Controls for Evidence Generation in Regulatory Decision-Making: Concept Paper. 2025. https://www.ema.europa.eu/en/development-reflection-paper-use-external-controls-evidence-generation-regulatory-decision-making-scientific-guideline
3. CRDSA. External Controls: Evolving Regulatory Approaches — MHRA, EMA, and FDA. October 2025. https://crdsalliance.org/wp-content/uploads/CRDSA_External-Controls_Evolving-Regulatory-Approaches_October-2025-1.pdf
4. Crump M, Neelapu SS, Farooq U, et al. Outcomes in refractory diffuse large B-cell lymphoma: results from the international SCHOLAR-1 study. Blood. 2017;130(16):1800-1808. https://ashpublications.org/blood/article/130/16/1800/36474/Outcomes-in-refractory-diffuse-large-B-cell
5. Shah BD, Ghobadi A, Oluwole OO, et al. Two-year follow-up of KTE-X19 in patients with relapsed or refractory adult B-cell acute lymphoblastic leukemia in ZUMA-3 and its contextualization with SCHOLAR-3, an external historical control study. J Hematol Oncol. 2022;15:170. https://jhoonline.biomedcentral.com/articles/10.1186/s13045-022-01379-0
6. Neelapu SS, Jacobson CA, Ghobadi A, et al. Five-year follow-up of ZUMA-1 supports the curative potential of axicabtagene ciloleucel in refractory large B-cell lymphoma. Blood. 2023;141(19):2307-2315. https://ashpublications.org/blood/article/141/19/2307/494672/Five-year-follow-up-of-ZUMA-1-supports-the
7. FDA. Summary Basis for Regulatory Action: Zolgensma (onasemnogene abeparvovec-xioi). May 24, 2019. https://www.fda.gov/media/127961/download
8. Jahanshahi M, Gregg K, Davis G, et al. The Use of External Controls in FDA Regulatory Decision Making. Ther Innov Regul Sci. 2021;55(5):1019-1035. https://pmc.ncbi.nlm.nih.gov/articles/PMC8332598/
9. Zou KH, Vigna C, Talwai A, et al. The Next Horizon of Drug Development: External Control Arms and Innovative Tools to Enrich Clinical Trial Data. Ther Innov Regul Sci. 2024. https://link.springer.com/article/10.1007/s43441-024-00627-4
10. Gokbuget N, Stelljes M, Topp MS, et al. Updated Outcomes from the Historical Control Study SCHOLAR-3 Contextualizing ZUMA-3 Results of Brexucabtagene Autoleucel (KTE-X19) in Adult Patients with Relapsed or Refractory B-Cell Acute Lymphoblastic Leukemia. Blood. 2022;140(Supplement 1):3158. https://ashpublications.org/blood/article/140/Supplement%201/3158/487262/Updated-Outcomes-from-the-Historical-Control-Study
11. Monnereau M, Delord JP, Michiels S, et al. Acceptance of external control arms by HTA agencies: a review of oncology submissions in France, England, Germany and Norway from 2021 to 2023. Br J Cancer. 2025. https://www.nature.com/articles/s41416-025-03155-6
12. Putnam Associates / Inizio. Why Real-World External Control Arms Are Gaining Ground in Health Technology Assessment. 2025. https://www.putassoc.com/insights/why-real-world-external-control-arms-are-gaining-ground-in-health-technology-assessment/
13. Parexel. EMA's Evolving Stance on External Controls: Key Takeaways and Preparation Strategies for Sponsors. 2025. https://www.parexel.com/insights/blog/emas-evolving-stance-on-external-controls-key-takeaways-and-preparation-strategies-for-sponsors
14. Khachatryan A, Read SH, Madison T. External control arms for rare diseases: building a body of supporting evidence. J Pharmacokinet Pharmacodyn. 2023;50(6):501-506. https://pmc.ncbi.nlm.nih.gov/articles/PMC10673956/