News
The move from preclinical work into first-in-human (FIH) studies changes what a bioanalytical method needs to do. While the chemistry behind a preclinical assay may be sound, the realities encountered in clinical studies often shift substantially from the controlled parameters of animal models. The need for bioanalytical data to support dose escalation, safety monitoring, and regulatory review requires not only a newly tailored method, but one that can stand up to greater regulatory scrutiny.
Many sponsors encounter this abrupt shift at the point of transition to FIH. Suddenly, it seems, an assay that performed well through toxicology studies is challenged when it meets human matrix, lower analyte concentrations, larger sample volumes, and stricter documentation requirements. Where preparation for FIH has been thin, the outcome is usually extensive rework: additional validation, repeat analysis, or a delayed study start while the method is brought in line with clinical realities.
However, as FIH is the goal of any IND filing, developing a method that anticipates changes in sample type and volume can save extensive time and resources while getting players ready for greater regulatory scrutiny. Bioanalytical readiness is the practice of anticipating those requirements before the time pressures of a clinical trial decrease the risk tolerance for delays.
Greater regulatory expectations are the most obvious change. Bioanalytical methods supporting clinical studies require full validation in the matrices of interest. Further, methods and documentation that were adequate for an internal audience now have to withstand inspection. For example, fit-for-purpose biomarker tests that informed development choices can no longer stand alone. Clinical programs introduce new obligations, such as the ICH M10 requirement of incurred sample reanalysis (ISR).
In addition to regulatory requirements for clinical work, analytical demands also change. Human samples carry a different endogenous background than animal matrices, subject populations vary in ways that controlled animal cohorts might not, and concomitant medications present in patient participants might introduce interferences that never appeared during preclinical development. Further, collection typically happens via multisite trials, the number of samples moving through the laboratory increases sharply, and the samples themselves often have different volumes compared to animal models.
A highly consequential analytical change is that analyte concentrations can be quite different. FIH starting doses are set conservatively, often well below the exposures used in toxicology studies. A method built to quantify toxicology-range concentrations may not reach down to the lower limit of quantitation (LLOQ) that early clinical samples require. In extreme cases, uncharacterized LLOQs could mean that a preclinical method requires substantial modifications, which can take considerable time and resources.
A method validated in animal matrix typically does not transfer to human samples without any modification. As the human assay requires a new validation, often the need for method changes arise. Detection conditions, extraction chemistry, and calibration ranges that were optimized against, say, rat or dog plasma, often need revisiting once human matrix is introduced. Therefore, a new validation rarely requires a similar amount of time and resources compared to a preclinical validation process.
Sensitivity is a common factor for preclinical method insufficiency. When the LLOQ for human samples drops relative to animal models by an order of magnitude or more, incremental optimization rarely closes the gap. Extraction approach, chromatographic column conditions, and instrument parameters may all need to be adjusted; thus validation for human samples typically sets off a new method development exercise, taking correspondingly longer.
Selectivity is a second preclinical method insufficiency factor. Human matrix sample lots often exhibit greater variability compared to animal matrices, and patient populations differ more still. A method that showed clean selectivity across 6 animal lots can produce noisy data sets when measuring samples from human subjects; especially if those human subjects are taking other medications or presenting with the disease state under study.
Validation gaps tend to appear where teams assume continuity that the method itself does not provide. Method validation is not a single protocol that carries forward indefinitely, and several changes routinely encountered at the transition into clinical trials can require additional work:
Where data within a study come from more than one validated method or more than one laboratory, cross-validation assesses the results’ comparability. Cross-validation is best completed before study samples are analyzed, such as once it is clear that differently validated methods will be used, such as in a site transfer. If cross-validation only occurs once a discrepancy surfaces, considerable time can be lost because retrospective reconciliation of 2 datasets is considerably harder to defend than a planned comparison. Further, not conducting a cross-validation before it is clearly warranted could risk considerable delay upon regulatory review, or even a failed trial.
Sponsors changing bioanalytical laboratories between preclinical and clinical phases have the additional burden of demonstrating continuity across that change. Sample traceability, raw data availability, and documented method transfer all become part of the regulatory record; weaknesses for any component can put the integrity of an otherwise sound dataset in question.
Early clinical datasets are typically noisier than preclinical ones, with some of that noise due to biological variability across a human population.
If matrix effects, selectivity, recovery, and precision near the LLOQ have not been characterized thoroughly for human matrix, the differences between sample population and method uncertainty become difficult to distinguish. Exposure relationships that look inconsistent may reflect real subject variability, or they may reflect assay behavior that was never fully mapped. That ambiguity can compromise decision certainty during dose escalation, in turn impacting subject safety and raising regulatory scrutiny.
Thus, careful bioanalytical method development and full validation is not just an important step, it is vital to determining the safety and efficacy of the drug candidate and mitigating the risk faced by clinical participants.
Given the risks of downstream trial delays, regulatory pushback, or patient safety, characterizing assay performance before samples arrive is critical. Unexpected results should be anticipated in the transition to a clinical dataset; upfront bioanalytical work can increase the confidence in how to interpret and attribute variability in the encountered data.
Scalability is a design consideration. Clinical studies bring larger batch sizes, more players, including more analytical scientists, longer sample collection campaigns, and, in many cases, multiple collection sites operating on different schedules. The study complexity alone means that retrofitting the method is not an option.
Methods with narrow operating tolerances tend to show their limits under these expanded conditions, and consistent performance becomes a vital metric. Long run times that were acceptable for a preclinical study of modest size become a throughput constraint at clinical scale. Extraction procedures requiring significant manual handling introduce variability between analytical scientists. Stability windows that were never tested beyond a short interval become a logistical constraint when samples ship from several sites.
Evaluating assay behavior across all realistic parameters of scale, including number of analytical scientists, instruments, and realistic batch sizes during development is what sets up a method to deliver consistent performance as the program grows.
The most effective point to prepare a clinical assay is before the preclinical work is completed. Decisions made during the investigational new drug (IND)-enabling phase, including matrix selection, calibration range, and internal standard choices, shape how much work will be required in the clinical transition.
Appropriate preparation means anticipating several clinical study parameters to assess method scalability. For example:
Considering and testing such parameters in advance vastly lowers the risk of encountering these complications once the clinical phase starts.
Alturas Analytics supports sponsors moving bioanalytical programs from preclinical studies into FIH and early clinical development. This includes best-in-class method development and validation ready to consider human samples, assessment of matrix effects and selectivity across representative lots, cross-validation where methods or laboratories change, and generally conducting sample analysis under GxP (good practice) quality that govern preclinical and clinical bioanalysis, including GLP and GCP. We have extensive method development experience using LC-MS/MS and, where samples call for additional resolution, LC-HRMS/MS.
Our scientific and sample management teams collaborate with sponsors to ready method performance for the requirements of the coming study phase(s) rather than the one just completed. For example, ensuring that pharmacokinetic and toxicokinetic (PK/TK) data generated across preclinical and clinical programs remain comparable and defensible from IND through to NDA filings. We run PK/TK analysis in-house, rather than subcontracting it, to more seamlessly ensure the highest quality of bioanalytical deliverables at each stage of the preclinical to clinical transition. When programs transfer work from another laboratory, we support every step of the documentation, validation, and cross-validation processes needed to demonstrate continuity.
Readiness at any stage should be less about any single validation exercise and more about developing and readying methods both for the present stage and the study it is to support. Sponsors who evaluate their clinical bioanalytical needs while preclinical work is still in progress, and who collaboratively coordinate with their bioanalytical CRO before the protocol is locked, generally find the transition to be a matter of planned execution. Those who wait until the clinical protocol is final find themselves continually solving for similar problems while the pressure continues to ramp.
LC-MS/MS is widely used in bioanalysis because of its high sensitivity, selectivity, and ability to quantitatively detect low concentrations of compounds in complex biological matrices. It is considered the gold standard for quantitative bioanalytical testing in drug development.
Yes. Bioanalytical methods supporting clinical studies require full validation for each matrix being analyzed. For example, a method validated in rat or dog plasma does not carry forward to human samples on its own. The scope of the human sample validation depends on how much of the method, including extraction chemistry, chromatographic conditions, and calibration range, remains unchanged.
ICH M10 expects ISR on the first clinical encounter with human subjects, with additional evaluations for other key studies such as the first trial in patients (i.e., those with a disease state, as opposed to safety assessment alone), and trials for special populations. Where most validation tests are performed on spiked matrix samples, ISR confirms concentration reproducibility and reliability for study samples where the drug has been biologically incorporated.
Disclaimer: This article is intended for educational and informational purposes and reflects the perspectives and expertise of the Alturas Analytics team. It is not a peer-reviewed scientific publication. Readers seeking additional scientific context are encouraged to review the supporting resources that may be referenced within this posting.
