Clinical Development Strategy
The plan for what evidence a programme must generate — endpoints, populations, phases, regions, comparators, and the decision criteria that determine whether to continue after each stage. Strategy is where the evidence a regulator will require and the evidence a payer will require are reconciled, before either has been asked. A trial that runs perfectly against the wrong question wastes years and exposes participants for nothing.
What an explainer is not
A topic explainer is SPEQ’s synthesis of what a practice involves, cited to the standards that govern it. It does not reproduce their text, and it does not determine which of them apply to your product or process.
[ POSITION IN THE FRAMEWORK ]
7 DIMENSIONS · 21 LINKSA development plan is a sequence of questions, each of which should be able to stop the programme: a plan whose studies can only confirm what is hoped has bought reassurance rather than evidence.
06 · QUALITY MATURITY — CLINICAL DEVELOPMENT STRATEGY, REACTIVE TO ADAPTIVE
Studies are planned one at a time. Each is designed once the previous one has read out.
A development plan exists covering the programme, built around the registration requirements rather than around the uncertainties.
The plan is organised by what is not yet known: each study answers a stated question, and decision criteria are set before the data exists.
The evidence a payer, a prescriber and a regulator each need is planned together, so the programme does not finish approvable and unusable.
Uncertainty is actively retired in the cheapest available order, and a study that would not change a decision is not run.
SPEQ’s shared five-stage progression, labelled synthesis — not the FDA QMM rating scale. Where does your organization sit? Score your quality system →
07 · REGULATORY & EVIDENCE
GOVERNING STANDARDS · 4
Derived from the 4 standards SPEQ maps to this subject, across 1 regulatory body: ICH.
RECORDS & OBJECTIVE EVIDENCE
- The clinical development plan, with the questions each study answers
- Decision criteria set before each readout, and the decisions taken against them
- The evidence requirements of regulators, payers and prescribers, and how the plan meets them
- Risk assessments informing study design and sequencing
- Records where the plan changed because of what a study showed
COMMON INSPECTION FINDINGS
- A programme whose studies cannot fail in any way that would stop it
- Decision criteria defined after the data was seen
- A registration package that meets regulatory requirements and no reimbursement evidence need
- Studies run in a sequence that repeats an uncertainty rather than retiring it
- A plan unchanged by results that contradicted its assumptions
Quality by design, applied to a programme
ICH E8(R1) reframed clinical study design around quality by design: identify the factors critical to the quality of the study, and design the study to protect them, rather than attempting to inspect quality in afterwards. It also introduced proportionality — the design and its oversight should be scaled to what actually matters for the reliability of the results and the safety of participants.
The practical consequence is a shift in where effort goes. Under the older model, effort concentrated on monitoring and data cleaning after the fact; under E8(R1) it concentrates on identifying the small number of things whose failure would invalidate the study, and building the design to protect those specifically.
Two audiences, one programme
A regulator asks whether the product is safe and effective for the proposed indication. A payer asks whether it is better than what is already reimbursed, and by how much, in the population it will actually be used in. These questions overlap and are not the same, and a programme designed only for the first arrives at launch with an approval and no evidence anyone will pay for it.
Reconciling them is a design activity, not a communications one: comparator choice, endpoints that are clinically meaningful rather than only statistically tractable, and populations representative of real use rather than optimised for signal. Each of those has a cost, and each is far cheaper decided at strategy than added as a post-approval study.
Decision criteria set before the data arrive
The most valuable thing a development strategy can contain is what will constitute a reason to stop. Written before the data exist, a stopping criterion is a scientific judgement; written afterwards, it is a negotiation with sunk cost. Programmes very rarely fail on a single unambiguous result — they fail slowly, through a series of results each of which is individually explicable.
This is also where regulatory advice earns its cost. Agreeing the evidence standard with the authority before generating the evidence removes the most expensive category of failure: a programme that ran competently and produced evidence the reviewer will not accept.
SPEQ interpretation — strategy is where participant burden is set
Every element of a development strategy has an ethical dimension that is usually discussed only at protocol level: how many participants are exposed, to what, for how long, and to answer what question. A programme that generates evidence nobody needed exposed people for nothing, and that judgement belongs at strategy where the questions are chosen — not at protocol where they are merely operationalised.
The practical form this takes is asking, of each planned study, what decision its result will change. A study whose result changes no decision is a study that should not be run, and that test is easier to apply honestly before a programme has momentum behind it.
FREQUENTLY ASKED
What did ICH E8(R1) change?
It reframed study design around quality by design and proportionality: identify the factors critical to study quality and design to protect them, rather than inspecting quality in afterwards. Effort shifts from post-hoc monitoring and data cleaning toward protecting the small number of things whose failure would invalidate the study.
Why do regulatory and payer evidence requirements diverge?
A regulator asks whether the product is safe and effective for the proposed indication; a payer asks whether it is better than what is already reimbursed, in the population it will actually be used in. Comparator choice, clinically meaningful endpoints and representative populations reconcile them, and each is far cheaper at strategy than as a post-approval study.
Why write stopping criteria before the data exist?
Because afterwards they become a negotiation with sunk cost. Programmes rarely fail on one unambiguous result; they fail through a series of individually explicable ones. A criterion written in advance is a scientific judgement, and it is the only kind that survives the moment it is needed.
What is the test of whether a planned study should run?
What decision its result will change. A study whose outcome changes no decision should not expose participants, and the question is far easier to answer honestly before a programme has accumulated momentum.