Skip to content
For clinical investigators

A defensible comparator for the study you can’t randomize.

When patients cannot wait in a placebo arm, or the budget cannot pay for one, an external control can carry the comparison — if it is designed in from the start. Variacle helps you build it so that funders, ethics committees and industry partners take it seriously — and, when your hospital is the sponsor, takes care of the regulatory submission.

Design comparisonIllustrative
1:1 randomized3:1 + external control
Patients on the new treatment50%75%
Information on the control outcomeConcurrent arm onlyConcurrent arm and external cohort
What limits precisionSize of the control armQuality of the external data and its assumptions
The small concurrent arm anchors the comparison; the external cohort supplies most of the control information, under assumptions that are written down and tested.
Situations we see

When a full control arm is not an option.

No standard of care to compare against

Last-line and advanced therapies, where the ethics committee will not allow the innovative treatment to be withheld.

Small populations

Rare and paediatric diseases, where a full control arm would take years to recruit.

A new therapy with small, uncontrolled studies

Promising early results, an ethical objection to a large placebo arm, and no budget for a large trial.

Reviewers said “the data are unbalanced”

And suggested matching that throws most of your patients away.

What we do

Methods you can defend, written for the people who decide.

We work alongside your team and your hospital's research unit, from the first sketch of the design to the reviewers' last question.

Design and sample size

Single arm with an external control, or unbalanced randomization (for example 3:1) that keeps a small concurrent control to anchor the comparison. Power is computed with the external information included.

Protocol, SAP and ethics-committee annex

Estimands per ICH E9(R1), the identifying assumptions, the estimation method and a sensitivity analysis — written so that a committee can follow them.

Funding applications

Methodology sections for competitive public calls and for industry programmes that fund investigator-initiated trials.

Re-analysis for resubmission

Efficient estimation on the full sample instead of discarding data, external data to strengthen sparse groups, and a sensitivity analysis reviewers can check.

Why it matters

A stronger proposal with much less risk.

Industry partners and funders back risk they can see. “Promising single-arm results” is a hard sell; a pre-specified comparator with stated assumptions and a sensitivity analysis is a value proposition they can evaluate.

The same work protects you later. When the comparison is fixed before outcomes are seen, a reviewer cannot dismiss the result as an artefact of the analysis — and if the treatment does not work, you learn that honestly.

Your therapy may well work. What you are missing is a comparator the committee will accept.

Questions

What investigators ask us.

Is a study with an external control publishable?
Yes, when the comparison is pre-specified and its assumptions are reported with a sensitivity analysis. We write the methods so that journal reviewers can check them, in line with the estimand framework they increasingly expect.
Where would the external data come from?
From completed trials, registries, hospital records or published cohorts in your indication. Part of the design work is checking which sources measure your endpoint in a comparable way, and how to request access.
Can we add an external control to a study that is already running?
Not as confirmatory evidence: an external control has to be pre-specified. For a running or closed study we can run a retrospective or exploratory analysis. It shows what your data could support next time, without touching the current submission.
When should we contact you?
As early as possible — ideally before the protocol or the funding application is written, because the external control has to be designed in. If your study is already running or published, a re-analysis may still help.
Can you do the submission in CTIS for our hospital?
Yes. When the hospital or its research foundation is the sponsor, we prepare Part I and Part II, submit through CTIS with a role the sponsor delegates to us, and answer the regulators' requests for information within the 12-day deadline. The hospital remains the legal sponsor; the paperwork is ours.
Who is responsible when the hospital sponsors and a company owns the drug?
It depends on what the company wants from the data. If it wants to use them for registration or hold rights over them, the trial is no longer non-commercial research and the company should be a co-sponsor, with responsibilities split in writing. If not, the hospital is the sponsor and the company supplies the drug and the investigator's brochure. We put that split on one page before anything is submitted.
Can we talk before sharing anything confidential?
Yes. The first call needs no data. We are happy to sign your NDA before we see protocols or datasets.

Tell us about the study you want to run.

A 30-minute conversation about your design. We will tell you honestly whether an external control helps — and if it does not, what would.

Talk through your study