Steady-State Constraints on Synthesis and Degradation

Why a Baseline Concentration Is Not a Pair of Rates

A biomarker measured repeatedly at a stable baseline may appear to provide a direct summary of its biology. But the same baseline concentration can result from slow synthesis paired with slow degradation, or from rapid synthesis paired with rapid degradation. Separating these possibilities matters when interpreting turnover, designing a longitudinal study, or deciding which parameters the data can support.

The mass-balance model for a single biomarker pool provides a precise way to state this issue. At steady state, the amount entering the pool per unit time must equal the amount leaving it. That balance constrains synthesis and degradation, but a baseline measurement alone does not tell us how large either rate is.

Turnover Model and Steady-State Constraint

This model assumes a well-mixed pool, constant synthesis over the period being considered, and loss proportional to the current concentration. It also assumes that there are no unmodeled inputs or outputs that materially affect the measured pool. The steady-state relationship below applies only if the observed baseline is genuinely stable under those conditions; a single measurement does not establish that it is.

At steady state, substitute \(B(t)=B_0\) and set the rate of change to zero. This gives

The first equality says that the synthesis rate equals the degradation rate at baseline. The final expression shows that the baseline concentration is a ratio of the two rates, not a separate measurement of each one.

Why Baseline Alone Does Not Identify Both Rates

Suppose \(B_0\) is known exactly and is positive. The steady-state equation gives one constraint involving two unknown positive parameters. For every chosen value of \(k_{\mathrm{deg}}>0\), there is a matching synthesis rate

All such pairs reproduce the same baseline. More formally, if \((k_{\mathrm{syn}},k_{\mathrm{deg}})\) satisfies the equation, then for any constant \(c>0\), the pair \((c k_{\mathrm{syn}},c k_{\mathrm{deg}})\) also satisfies it, because

This is structural non-identifiability from baseline alone: even ideal, error-free knowledge of \(B_0\) does not determine the two rates separately. Better precision for the same baseline measurement narrows uncertainty in the ratio, but does not resolve this ambiguity. The conclusion is about the information in the measurement, not about whether a numerical fitting algorithm can return two parameter values.

How Longitudinal Information Adds a Rate Constraint

Away from steady state, the turnover equation can be rearranged as

If concentration and its local rate of change are estimated at multiple times, each time point provides information about how the rate of change depends on concentration. Under the model, that relationship is a straight line: its intercept is \(k_{\mathrm{syn}}\), and its slope is \(-k_{\mathrm{deg}}\). In practice, derivatives are not observed directly; they must be estimated from longitudinal measurements, and that estimation can be noisy. The example below treats local slopes as estimates obtained from a sufficiently informative hypothetical profile.

Assumptions, Diagnostics, and Alternatives

The steady-state relation is simple, but its interpretation depends on assumptions that should be checked against the data and study design. A baseline concentration should be supported by repeated observations or other evidence of stability. If the biomarker is trending before a perturbation, setting \(dB/dt=0\) can produce a false rate constraint. Likewise, unexplained changes in synthesis, additional inputs or outputs, changing distribution volume, or non-first-order loss can invalidate the single-pool equation.

With longitudinal data, inspect whether estimated changes in concentration are compatible with a single linear relationship between \(dB/dt\) and \(B\). Systematic curvature, different relationships during rising and falling phases, or residual patterns that depend on time can indicate that constant synthesis or first-order degradation is inadequate. Sparse measurements can make local slopes unstable; noise in both concentration and estimated slope can also distort a fitted line. These are reasons to qualify rate estimates, not reasons to treat a baseline ratio as two separately identified parameters.

There are several ways to obtain information beyond baseline, each relying on different evidence:

• Longitudinal concentration measurements can inform the dynamic rates when the time course spans enough change and the model is appropriate.
• Independent information on one rate, such as a separately justified estimate of degradation, allows the other rate to be calculated from \(k_{\mathrm{syn}}=k_{\mathrm{deg}}B_0\).
• Additional baseline measurements under distinct, characterized conditions may add constraints, but only if the relevant model changes are specified and supported; simply repeating the same stable baseline improves measurement precision rather than separating the rates.

No approach removes the need to state its assumptions. An independent estimate may be uncertain or may not apply to the population being modeled. A longitudinal profile may be too short, too sparse, or too noisy to determine a reliable slope. If the actual loss process is nonlinear, forcing a constant first-order \(k_{\mathrm{deg}}\) can yield parameters that summarize the observed range without representing a universal degradation rate.

Common Mistake

A common error is to calculate \(k_{\mathrm{syn}}=k_{\mathrm{deg}}B_0\) after choosing an assumed value for \(k_{\mathrm{deg}}\), then report both rates as if the baseline data had estimated them. The arithmetic is valid, but the inference is not: the chosen degradation rate supplied the missing information. Report the dependence explicitly, or justify the rate using data or evidence beyond the baseline measurement.