Predose Sampling Across Replicate Periods

Why Predose Samples Matter

In a replicate crossover study, the same participant receives treatments in more than one period. A sample collected immediately before dosing in a period can show whether measurable analyte is present before that period's dose. This is useful because a concentration carried over from an earlier period may complicate interpretation of the new period's pharmacokinetic profile. The sample is a piece of evidence about the participant's predose state; it is not, on its own, a verdict on whether the washout was adequate.

The distinction matters. A detectable predose concentration is evidence that analyte was present at the sampling time, but it does not by itself identify the source or establish a clinically meaningful effect on the next period. Conversely, a result below the assay's quantification limit does not establish that the concentration was zero, or that there could have been no residual pharmacologic effect. Interpretation depends on the treatment history, the elapsed time since the previous dose, the analyte and assay, and the concentration's relevance to the study's measurements.

Placement and Documentation in a Replicate Schedule

Collect the predose sample in each period, including the first, according to the protocol's specified timing relative to dosing. The first-period sample provides a baseline measurement before any study treatment has been administered; later-period samples can be interpreted in light of prior study treatment. A sample taken after the current dose is not a predose sample, even if it was collected early in the dosing visit.

For every result, preserve the information needed to interpret it. At a minimum, the record should identify the participant and period; the scheduled and actual sample times; the current treatment; the treatment in the preceding period; the preceding dose time; and whether any relevant non-study drug use or sample-handling issue was reported. The actual elapsed time matters: a planned washout interval and the interval that occurred for a particular participant need not be identical.

In a replicate design, treatment labels alone are not enough. Two predose results for the same participant might follow different treatments, or the same treatment after different elapsed intervals. Examining results against the preceding treatment and time since that dose helps distinguish a pattern compatible with residual study drug from an isolated value that may have another explanation. This use of treatment history does not require the predose result to be treated as a new endpoint or as an automatic reason to discard data.

What a Concentration Can and Cannot Establish

A result above the assay's lower limit of quantification (LLOQ) establishes that the analyte was quantifiable in that sample under the assay conditions. If the analyte is a study drug or a relevant metabolite, and no alternative source is plausible, that finding supports the presence of residual analyte at the predose time. It does not, by itself, show that the residual concentration altered the period's pharmacokinetic response or that the study's washout plan was inadequate for every participant.

A result reported as below the LLOQ means the assay did not quantify the analyte at or above that limit. It is not a measured concentration of zero. A lower assay limit improves the ability to detect small concentrations, but even a nondetect cannot show that concentrations were zero throughout the interval or rule out every possible residual pharmacologic effect. The LLOQ should therefore be interpreted alongside the expected concentration scale and the purpose of measuring the analyte.

Predose results also need source evaluation. A quantifiable value could reflect residual study treatment, an endogenous or background source, an unreported concomitant product, contamination, or a labeling or analytical issue. A concentration that is unexpected in context should prompt review of timing, sample identity, assay records, and relevant medication history before it is assigned an explanation. The result is not a substitute for those checks.

Using Elimination Calculations Carefully

A simple elimination calculation can help judge whether a measured predose concentration is plausible after a given interval. Under a one-compartment, first-order elimination approximation after absorption and distribution are complete, concentration declines exponentially. If \(C_0\) is the concentration at the start of the interval, \(t\) is elapsed time, and \(t_{1/2}\) is the elimination half-life, the predicted concentration is:

This expression assumes the relevant decline follows a single exponential over the interval and that the half-life is appropriate for the participant and analyte. Real concentration-time profiles may not meet those assumptions, particularly during absorption, distribution, or multiphasic elimination. The calculation is therefore an expectation under a simplified model, not proof of an actual concentration or a universal washout rule.

The familiar five-half-life calculation follows directly from the equation. If \(t=5t_{1/2}\), then the remaining fraction is:

Thus, under the stated model, five half-lives leave 3.125% of the starting concentration, not zero. Whether that amount is negligible depends on the analyte, the starting concentration, assay sensitivity, and potential pharmacologic relevance. Adequacy is a design question about whether the planned interval is sufficiently protective for the study's objectives, supported by appropriate evidence; it cannot be declared solely by counting half-lives.

Interpreting Results Against the Assay and Study Context

The same reported category can have different implications depending on the assay limit and the expected concentration scale. A result of 6 ng/mL is quantifiable if the LLOQ is 5 ng/mL but below quantification if the LLOQ is 10 ng/mL. Report the actual result category and the relevant LLOQ rather than translating a below-LLOQ result into zero. If several analytes are measured, interpret each separately; a parent-drug result does not automatically describe a metabolite's presence or persistence.

A useful interpretation therefore has three parts: what the assay reported; what the participant received and when; and what the concentration means for the study's measurements. A result can be analytically detectable but too small to be relevant to the intended interpretation. Alternatively, even a low or unquantifiable result may not resolve concerns if the assay is insufficiently sensitive for the question or if a persistent pharmacologic effect is plausible.

Diagnostics, Alternatives, and Limitations

When a predose result is unexpected, first verify the basics: the sample was collected before dosing; the participant, period, and treatment labels are correct; actual times are available; and the laboratory result and LLOQ are correctly recorded. Then examine the preceding treatment, elapsed interval, relevant analyte, and possible alternative sources. A pattern of quantifiable results associated with particular prior treatments may support a carryover explanation more strongly than a single isolated value, but a pattern still needs biological and analytical interpretation.

Predose sampling is a diagnostic aid, not a direct test of all forms of carryover. Concentration measurement may not capture pharmacodynamic persistence, and one sample cannot describe the full concentration trajectory during washout. A predose result may also be affected by sampling error, assay performance, or timing deviations. Conclusions should be limited to the analyte measured, the sample collected, and the study question addressed.

Alternative evidence includes pharmacokinetic knowledge about the analyte and its metabolites, the planned washout interval, observed concentration-time data, and relevant pharmacologic information. These sources answer related but distinct questions: a model predicts concentrations under assumptions, while a predose assay measures a sample at a particular time. They are most informative when considered together, rather than substituted for one another.

• Common mistake: “The predose result was below the LLOQ, so the concentration was zero and washout was proven adequate.” Correction: the result was not quantified at or above the LLOQ in that sample. It does not establish zero concentration or rule out every relevant residual effect.
• Common mistake: “A detectable result proves that the previous treatment caused meaningful carryover.” Correction: it establishes a quantifiable analyte in the sample; source and relevance require further evaluation.
• Common mistake: “Five half-lives always mean adequate washout.” Correction: the simplified model predicts 3.125% remaining after five half-lives. Adequacy depends on context and cannot be inferred from that percentage alone.

Predose results should not trigger automatic participant exclusion or an improvised adjustment to pharmacokinetic analyses. Any handling of such findings should follow prespecified study procedures and be interpreted in the context of the complete evidence. The practical goal is to make residual analyte assessable and to avoid confusing the measured predose state with proof that the washout was either adequate or inadequate.