Preclinical development brings together pharmacology, toxicology, formulation and manufacturing science to confirm that a candidate is safe, stable and ready to progress. It's also one of the stages where programmes most often lose time, through late discovery errors, weak study design or gaps in regulatory documentation that trigger costly delays or full resets.
With the experts at Abzena, the leading biologics CDMO and CRO, we look at where programmes most commonly go wrong and how getting the fundamentals right early can protect both timelines and data quality.
Translation is where assumptions get tested
Many programmes move too quickly from a promising result in discovery to a full preclinical model, without confirming that the result will hold up elsewhere. What performs well in one system does not always behave the same way across species, dose levels or exposure conditions, and without that confirmation, a study can end up built on assumptions that do not translate.
"The transition from discovery to preclinical models is where assumptions get tested, and not always favourably," says Gary Watts, Head of Formulation Development at Abzena. "What looks promising in one system can behave very differently across species or dose ranges. Taking time to ensure translational relevance before committing to a full study design is not a delay; it is risk management."
Analytical characterisation needs to be right from the start
Once a molecule enters regulated studies, every attribute carries weight. Purity, potency, stability and structure all require supporting evidence, and inconsistent or missing analytical data tends to trigger regulatory questions or repeat testing later in the programme.
"Phase-appropriate, qualified analytical methods need to be in place before studies begin, not introduced once questions start coming in," explains Watts. "Getting the analytical foundation right from the outset is one of the simplest ways to avoid losing time further down the line."
The need for manufacturing readiness
Biologics manufacturing decisions made early in a programme have consequences that run the full length of preclinical work and beyond. If the material used in preclinical studies is unstable or the process behind it can’t scale, the data generated may not reflect the product that eventually reaches the clinic, and that gap tends to surface at the worst possible time.
"In biologics, the process is the product," says Watts. "If you are running preclinical studies with material that does not reflect your intended commercial process, the data you generate may not hold up later. Getting manufacturing involved at the preclinical stage is not early; it is necessary."
Documentation gaps undermine good science
Even strong scientific results can run into regulatory difficulty if the records behind them are incomplete. Regulators expect clear traceability from material source through to test result, and informal or fragmented documentation slows submissions and often forces teams to reconstruct work that should have been captured the first time.
"Regulators do not just evaluate your science; they evaluate your evidence that the science was conducted properly," adds Watts. "Incomplete records or informal reporting can undermine a strong data package. Traceability needs to be built into the workflow from day one, not reconstructed at submission."
Coordination across partners determines pace
Preclinical programmes routinely involve multiple vendors and CROs, from labs running studies to partners generating supporting data. When communication between these groups breaks down, data gets lost, timelines drift, and conclusions become harder to defend by the time they reach a regulatory submission. A single point of coordination and shared visibility over data as it is generated tends to prevent most of these issues before they start.
A connected workflow moves faster
"When analytical development, manufacturing and preclinical testing operate as a connected workflow rather than separate workstreams, programmes move more predictably," says Watts. "You catch issues earlier, you avoid duplication, and the data package you take to regulators reflects a coherent body of work."
Preclinical development will always carry pressure and uncertainty, but the patterns behind most failures are well understood and avoidable. Programmes that plan for translational relevance, get their analytics and manufacturing right early, maintain disciplined documentation, and keep partners coordinated tend to reach the clinic faster and with considerably more confidence in their data.
