Refining and simplifying the classical oral tyramine challenge test of MAO inhibitors

Evaluation of tyramine Cmax to replace the highly variable blood pressure response

Tyramine and monoamino oxidases

Tyramine is a constituent of in particular fermented foods and drinks like old cheese, beer, wine and herring. It’s a bioactive amine; when reaching the systemic circulation it may cause potentially harmful increases in blood pressure, presumably by displacing noradrenaline. Fortunately, under physiological conditions we are protected by the very high oral first pass effect of tyramine, mediated by monoamino oxidases (MAOs) in intestine and liver (Ostadkarampour & Putnins, 2021; Finberg & Rabey, 2016). As a consequence, the systemic uptake of tyramine is almost negligible, i.e. presumably less than 1% (Rafehi et al. 2019; VanDenBerg et al. 2003).

MAO is a therapeutic target, though. Inhibiting MAO is therapeutically pursued in the treatment of depression and Parkinson’s disease, and other indications for MAO inhibitors are contemplated (Duarte at al. 2021; Iacovino et al. 2020). In the clinical use of MAO inhibitors, the potential risk is that the desired clinical effect comes at the cost of tyramine-mediated side effects, when the MAO inhibitor increases the uptake of tyramine from food.

The oral tyramine challenge

To address that concern, the blood pressure effect of any MAO inhibitor in combination with orally administered tyramine is evaluated in a ‘tyramine challenge’ test, as performed for e.g. rasagiline, tedizold, and safinamide (Goren et al. 2010; Flanagan et al. 2013; Marquet et al. 2012).

The typical oral tyramine challenge comprises a baseline period in which every subject receives escalating single oral doses of tyramine until a threshold systolic blood pressure (SBP) change versus baseline of 30 mm Hg or more is hit; the tyramine dose at that threshold is called the TYR30. After this baseline assessment, test, positive control, or placebo treatment is initiated with randomization, and the TYR30 assessment is repeated when the test and reference treatments are expected to have reached pharmacokinetic steady-state. If test or reference treatment increases tyramine bioavailability compared to baseline in the same subject, the TYR30 on treatment is a lower dose than the TYR30 at baseline. The ratio of TYR30 at baseline over TYR30 at treatment is called the tyramine sensitivity factor (TSF); a TSF of 2 implies that tyramine bioavailability in that subject has doubled on treatment compared to baseline.

The issues with the classical tyramine challenge

One downside of the current traditional ‘classical’ oral tyramine challenge is the poor predictability of the extent of SBP change, since minor variations in tyramine first-pass clearance can cause major changes in systemic tyramine levels and in downstream pressor response. As a consequence, SBP excursions very much above 30 mm Hg may occur in an uncontrolled manner, exposing study participants to potential risk. A further downside of the classical oral tyramine challenge is that it addresses impact on tyramine bioavailability not by measuring tyramine bioavailability itself, but by evaluating a downstream effect of tyramine, i.e. change in SBP. By using this surrogate for tyramine bioavailability, the current oral tyramine challenge adds variation in pharmacodynamic response on top of pharmacokinetic variability. Also, any pharmacodynamic effect of the test treatment itself may—in theory— counteract or enhance the effect of tyramine on blood pressure, thereby obscuring the interpretation of the TSF.

The picture below illustrates the variation issues with the classical tyramine challenge (Figure 1). The graph on the left shows the ratio of tyramine peak concentrations (Cmax) observed within the same subject during placebo treatment over baseline, for all individual subjects in a previously completed tyramine challenge up to 800 mg. Since everyone in the graph received placebo, the ideal outcome—in absence of any intra-subject variation—would be that the ratio of Cmax values is 1.00 for all. In practice that is not the case; Cmax values can easily change up to 10-fold, which is fully consistent with the variation in first pass metabolism.

The graph at the right pushes the point further; it displays the difference in maximum change in systol blood pressures (SBP) within individuals subjects between study periods. Again, in case of negligible intra-subject variation, all subjects would have a difference of 0 mm Hg between placebo treatment and baseline. In practice, differences may easily reach 40 mm Hg if not more, which exceeds that critical number of a 30 mm Hg change that defines the tyramine sensitivity factor.

Figure 1: Within-subject variation in oral tyramine exposure and response; ratio of Cmax on placebo treatment (Period 3) over Cmax during baseline (Period 1) (left panel), and difference in maximum change from baseline (CFB) for SBP between both periods (right panel), in individual subjects receiving oral tyramine doses escalating from 100 mg to 800 mg. Samples sizes vary from n=1 (800 mg) to n=45 (200 mg) (ICON; data on file).

The simplified and refined tyramine challenge

We designed and tested a reduced and refined oral tyramine challenge design that focusses on the point of primary concern, i.e. the increase of tyramine plasma concentrations due to MAO inhibition, as per schematic below (figure 2).

The new design very much resembles the classical design, with only a few, essential differences:

– It escalates to 400 mg of tyramine only, to reduce safety risk.

– It tests all escalating tyramine doses in all subjects, unless an individual subject reaches a reproducible SBP increase. In that case, escalation is stopped in that subject to minimize risk, but the subject will remain in the study if period 3 is still to be done.

– Tyramine plasma concentrations are analysed off-line, and for each subject the tyramine C is assessed.

– The key outcome of the escalation for each subject is the lowest tyramine dose (TYRC10) during the escalation that results in a tyramine Cmax of 10 ng/mL or higher in that subject.

– The TSF for the subject is defined as the ratio of TSFC10 at baseline over treatment in that subject.

Figure 2: Schematic of refined and simplified tyramine challenge design with tyramine escalation to a dose with Cmax ≥10 ng/mL or to 400 mg of tyramine.

*Also in the new design, escalation is stopped when a subject meets the SBP ≥30 mm Hg change from baseline criterion, to minimize risk. A sample size of 16 subjects per treatment is proposed for the new design.

Testing the refined tyramine design

We tested the refined design by retrospective analysis of a completed double-blind, placebo-and active-controlled, oral tyramine challenge study in healthy participants (7 groups of n=16-24 each), receiving escalating tyramine doses at baseline, and after repeated dosing of reference or test treatments or placebo. TYRC10 detected effects on tyramine sensitivity similar to TYR30; TYRC10 was reached sooner in the escalation, typically 2-4 fold below TYR30 (Table 1).

Phenelzine proved an exception, with stronger SBP effect than reflected by tyramine Cmax, possibly due to a direct max effect of phenelzine on SBP (data not shown).

Further to the retrospective analysis above, we evaluated the refined tyramine challenge by trial simulation, in comparison with the classical challenge with TYR30. When simulating 10,000 trials in 12 subjects per treatment, TSF30 and TSFC10 performed similarly in estimating true TSFs of 1, 2, and 4 with the note that TSFC10 approached the true TSF closer than TSF30 (green values in Table 2). Variability in TSF estimates was less in terms of more narrow ranges of simulated values for TSFC10 compared to TSF30 (data not shown).

Importantly, based on confidence intervals (CIs) of TSF estimates, TSFC10 performed much better than TSF30 in terms of at least 25-fold reduced likelihood of making the wrong decision on TSF being 1, 2, or at least 2, whilst the true TSF is another value (green ovals in Table 2).

Table 1: Doses to Cmax ≥10 ng/mL (TYRC10) at baseline (Period 1) and at active test and reference treatment (Period 3), and ratios of these doses (TSFC10), and TYR30 ratios for the same treatments and Periods (TSF30); all values are median values in a sample on 10-14 subjects per treatment, in a completed classical tyramine challenge with oral tyramine doses up to 800 mg (ICON, data on file)

Table 2. Simulation results (10000 simulations, with n=12 per treatment) for TSF30 and for TSFC10 in the classical tyramine challenge escalating to 800 mg of tyramine, applied to placebo treatment and different active treatments with their indicated ‘true’ TSF. For the estimated TSFs, geometric means (gmeans) are presented, as well as proportions of 95% confidence intervals (CIs) including 1, including 2, wholly excluding 2, and including or greater than 2, respectively.

Conclusion

The simplified and refined tyramine challenge study design evaluates the tyramine dose to Cmax of ≥10 ng/mL as primary endpoint. In terms of subject safety, study duration, costs, and interpretability for trial objectives, it outperforms the classical tyramine challenge design that focuses SBP changes and escalates to higher tyramine doses. In trial simulations, the new design has an improved performance in terms of identifying the correct TSF, with more than 25-fold reduced false positive and false-negative rates in estimating TSFs. By escalating to tyramine doses not exceeding 400 mg, the new design reduces risk to subjects associated with tyramine induced blood pressure spikes. Also, it spends less time on escalating, and is thereby less demanding to study participants, and more efficient with shorter overall duration of the trial.

Results warrant the further consideration and application of this new design for the clinical evaluation of novel MAO inhibitors for which there would be a concern on their safety when dosed in combination with tyramine containing foods or liquids.

Read the article here.