MB&P Tools & Diagnostics

Trial Autopsy: Is the Asset Failing, or Did the Trial Fail the Asset?

A negative or ambiguous result does not automatically mean the biology failed. Before a repeat study, financing decision, licensing discussion or programme stop, the first task is to determine whether the study actually delivered a controlled and interpretable test of the asset.

The decision problem. Teams often move too quickly from “the primary endpoint was not met” to a biological conclusion. That can destroy value in two directions: a viable asset may be abandoned because the experiment was weak, or a weak asset may absorb more capital because operational noise is mistaken for hidden efficacy.

When this diagnostic is useful

Before an irreversible commitment

Use it before a protocol lock, pivotal-design decision, regulatory interaction, major vendor commitment, financing gate, partnering discussion or repeat-study decision.

When the team is not converging

Use it when different functions are giving individually reasonable answers that do not yet add up to one defensible development decision.

What to pressure-test

Biological test integrity – exposure, target engagement, dose and duration
Population and phenotype – whether the enrolled patients matched the mechanism
Endpoint and estimand – whether the effect was measured in the right way and time window
Execution and data – recruitment, eligibility control, raters, vendors and missingness
Statistical interpretability – variance, event-rate and model assumptions
Regulatory and capital relevance – whether the evidence can support the next irreversible decision

Red flags

The team cannot state the dominant failure mode in one sentence.
A repeat study would substantially reuse the same design assumptions.
Subgroup or regional explanations appear only after the headline result disappointed.
The conclusion changes materially under plausible missing-data or sensitivity assumptions.

Why this matters to capital

The decision is not simply whether to “run another trial”. It is whether the next tranche of capital buys new information or repeats the same uncertainty.

What changes after a useful review

A useful diagnostic does not end with a score. It identifies the dominant uncertainty, distinguishes what is known from what is assumed, shows which decision is currently exposed and defines the smallest next action capable of changing that decision.

Illustrative outcome from selected work. Selected rescue work in a multi-country CNS programme combined feasibility re-forecasting, country/site redesign, vendor reset and endpoint-focused RBQM. The result was stabilised enrolment, better forecast accuracy and restored interpretability at database lock.

Download the one-page diagnostic

The downloadable version is intentionally concise. It is designed to structure an internal discussion and reveal where a programme-specific review may be warranted; the full professional working method is not published.

Download the Trial Autopsy PDF

Related expertise

Trial Rescue & Execution Governance CRO underperformance: replace, repair or redesign?Related trial-design and execution insights Results & case snapshots

Start with the decision, not the deliverable

Bring the protocol, topline result and the next board or financing decision. A focused diagnostic can identify which uncertainty must be resolved before more capital is committed.

Start with a 20-minute Clinical Decision Diagnostic

Bring one live decision, the next hard milestone and the time available. The first conversation is designed to determine whether focused senior intervention can create material value.

Educational decision-support tool. It does not replace programme-specific medical, statistical, regulatory, operational, financial or legal review.