Short answer: In a 10-person Phase I study, taking the investigational glucocorticoid-receptor antagonist PT150 did not produce a statistically detectable change in the study’s breath-derived estimated blood alcohol concentration after an ethanol challenge. That finding is narrower than saying PT150 cannot interact with alcohol. All ten study completers were men, the protocol used an ethanol-containing “placebo” beverage, compared pretreatment day 1 with PT150-exposed day 5, and did not analyze venous whole-blood ethanol or test driving performance. It can inform how a small drug–alcohol safety experiment should be read; it cannot establish an individual’s forensic BAC, impairment, or event causation.
The primary source is Morice and colleagues, Scientific Reports (2021), DOI 10.1038/s41598-021-88609-6. Its actual comparisons, sample size, exposure measurements, and statistical uncertainty are essential to interpreting the phrase “no significant difference.”
What question did this study actually answer?
The investigators asked whether short-term PT150 exposure, given alongside a controlled ethanol beverage, was associated with detectable changes in selected safety and pharmacodynamic measures. PT150 is a selective antagonist of the glucocorticoid receptor, a component of stress-hormone signaling. The trial was designed as an early safety evaluation, not as a test of whether a person taking PT150 can operate a vehicle or whether ethanol is causally responsible for an observed behavior.
Its main measurements were breathalyzer-derived ethanol estimates, heart rate, blood pressure, electrocardiograms, adverse-event reports, and several subjective rating scales. Blood samples were used for selected laboratory tests, including plasma cortisol, but the researchers expressly acknowledged that the ethanol concentrations came from a breath device rather than from direct analysis of blood ethanol. That distinction prevents a casual reader from turning “estimated BAC” in a graph into an independently measured forensic whole-blood concentration.
The most defensible conclusion is therefore conditional: in the ten studied participants and under the specified challenge conditions, the measured safety variables and breath-estimated alcohol summaries did not reveal the particular large differences the investigation sought. The absence of statistical significance does not prove the underlying effect is zero or define a reliable upper limit on an interaction in other populations.
How were the ten participants exposed to PT150 and ethanol?
The research took place at one U.S. clinical research site in 2018–2019. Eleven people were randomized, one withdrew before the first alcohol challenge, and ten completed the procedures. All ten completers were men; median age was 44.5 years, with a reported range of 28–63 years. The enrollment requirements favored otherwise healthy, alcohol-experienced adults. People with certain psychiatric disorders, substance-use disorders, abnormal screening laboratory results, and current medication use were excluded. Results from this selected group cannot be carried uncritically to women, patients with complex comorbidities, people taking several medications, or people undergoing alcohol withdrawal.
Each participant completed two beverage sessions on day 1, before PT150, and two on day 5, after repeated PT150 dosing. On both challenge days, they consumed an ethanol beverage and a lower-ethanol comparator. Beverage order was randomized on day 1 and reversed on day 5. The sessions were separated by at least four hours, and the participants received standardized meals. This design gives each person an internal comparison, but the PT150 exposure condition was inseparable from study day: everybody was unexposed first and exposed later. It was not a fully counterbalanced PT150-versus-no-PT150 drug treatment sequence.
The alcohol beverage contained 0.8 g ethanol per kilogram body weight in a 16%-by-volume mixture. The comparator contained approximately 0.05 g/kg ethanol in a 1% mixture to help mask taste. A low-ethanol comparator is not a chemically ethanol-free control. PT150 was administered orally at 900 mg once daily, beginning after the day-1 challenges and continuing through day 5. The authors described the later exposure as steady-state PT150 treatment; this report should not be taken as an individual measurement of PT150 plasma concentration.
What was directly measured, and what was calculated?
An Alco-Sensor IV breath instrument supplied the ethanol measurements from which the investigators reported estimated blood alcohol concentration, or eBAC. Breath tests were taken before the beverage and at scheduled points over the ensuing two hours. The investigators calculated conventional concentration–time summaries: a maximum observed eBAC (Cmax), a time of the maximum (Tmax), and a two-hour area under the curve (AUC0–2). Those summaries are functions of the recorded breath-based estimates and the sampling schedule.
This sequence has multiple inferential steps. The device observes exhaled ethanol under its operating rules. The paper then reports a blood-concentration-style estimate. The investigator identifies the highest sampled value and integrates the observed time series. None of those operations retroactively creates a gas-chromatographic measurement of venous blood from the same participant. Nor does the estimated peak or AUC specify how the individual performed on attention, reaction-time, divided-attention, field sobriety, or driving tasks—because those were not measured as validated functional endpoints.
| Reported quantity | How it arose | What it does not establish |
|---|---|---|
| eBAC series, Cmax, Tmax, AUC0–2 | Alco-Sensor IV exhaled-breath results, expressed or summarized as blood-alcohol equivalents | A laboratory-assayed venous whole-blood BAC, an earlier driving-time BAC, or task impairment |
| Heart rate and blood pressure | Serial physiological observations before and after beverages | Presence or absence of all drug effects, cognition, driving capacity, or arrhythmia risk for another person |
| ECG findings and adverse-event reports | Safety monitoring during a short trial in ten selected men | Population-level assurance of no rare adverse events or chronic toxicity |
| AUQ, BAES, PANAS and ARCI scores | Participants’ answers to subjective questionnaires | Measured ethanol concentration, objective psychomotor performance, or individual credibility |
| Plasma cortisol and other clinical chemistry | Blood laboratory assays collected on scheduled study days | Plasma ethanol analysis, measured receptor occupancy, or a forensic impairment threshold |
Did PT150 change the observed alcohol concentration–time curve?
For the ethanol challenge, the study reported a mean peak eBAC of 0.104 in the paper’s eBAC reporting convention both before PT150 exposure and after repeated dosing. The modeled between-condition difference in the peak was reported as 0.00, with a 95% confidence interval from −0.03 to 0.03 and p = .99. For the measured time to peak, the estimated difference was −0.60 with a wide 95% interval from −25.08 to 23.89 and p = .96. The two-hour AUC comparison was −0.16, 95% interval −3.40 to 3.08, p = .91. These figures are the investigators’ breath-derived summary comparisons, not paired gas-chromatographic blood samples.
It is tempting to say that PT150 “does not affect alcohol pharmacokinetics.” That is too strong. A concentration curve measured at a limited set of times may fail to detect small, delayed, or otherwise different effects. The design did not independently measure drug concentrations in plasma and did not identify the full absorption, distribution, metabolic, and excretory mechanisms for PT150–ethanol coexposure. The authors explicitly described this Phase I study as groundwork for a future pharmacokinetic interaction investigation.
Equally important, a test of statistical significance is not a formal equivalence test. A high p-value does not by itself establish that two exposure conditions have equivalent effects within a prespecified clinically acceptable margin. The estimate, its uncertainty interval, the small sample, and what was actually sampled are necessary for a defensible statement.
What did the cardiovascular and adverse-event results show?
The investigators did not find statistically significant evidence of a difference in the selected heart-rate or blood-pressure responses between the day-1 ethanol challenge without PT150 and the day-5 challenge with PT150. For example, the estimated difference in mean heart-rate change was 0.57, with a 95% confidence interval from −1.01 to 2.16 and p = .48. The corresponding systolic blood-pressure difference was 1.94, interval −0.61 to 4.49, p = .14; the diastolic estimate was 0.50, interval −1.41 to 2.40, p = .61.
During the brief study, five mild adverse events were recorded in one participant; the investigators classified them as unrelated to study drug. No serious adverse events, unexpected serious adverse reactions, or clinically significant abnormal ECGs were reported in this small completed cohort. Those observations matter, but the appropriate wording is “none observed in these ten participants over this protocol,” not “impossible” or “excluded” in an arbitrary patient. A ten-person study cannot reliably quantify rare event rates, clinically important subgroups, or long-term combined exposure.
These outcomes also do not establish the absence of a psychomotor drug–alcohol interaction. The trial gathered safety physiology and self-report measures, not a comprehensive battery of validated fine-motor, attention, visual tracking, or simulated-driving tests. A normal ECG is not a test of divided attention. A stable blood-pressure trace is not a demonstration of safe driving.
Did subjective alcohol effects and craving scores stay the same?
No. The largely negative cardiovascular and eBAC comparisons should not be used to erase the questionnaire findings. The investigators reported a difference of 1.30 AUQ points in the change in alcohol-urge scores between the two ethanol challenges, with a reported 95% interval of 0.66–1.95 and a tabled p < .001. The original authors also observed that the pre-challenge AUQ baseline was higher on day 1 than on day 5, which complicates attribution of the change to an acute ethanol–PT150 interaction.
Some Addiction Research Center Inventory contrasts also varied: for the LSD-labeled subjective-effects subscale, the overall PT150-exposure contrast was 0.58 (95% interval 0.06–1.09; p = .03), while the specific comparison of ethanol challenges across PT150 exposure was 0.57 with interval −0.15 to 1.29 (p = .12). Those are different statistical contrasts. They should not be combined into a single assertion that PT150 clearly potentiated or clearly prevented intoxication.
The AUQ concerns a participant’s current desire to drink. The ARCI subscales ask about subjectively experienced effects. Neither one is an instrument for direct ethanol assay, nor is a questionnaire score a measured decrement in reaction time or legally defined impairment. A person can report a change in desire or feeling without the study proving that their objectively measured functional performance changed in the same direction.
What about cortisol—was it measured in blood?
Yes, but which analyte matters. The investigators collected blood for selected clinical laboratory measures and found higher mean plasma cortisol following PT150 exposure. Mean cortisol was reported as 11.19 µg/dL at baseline, 16.20 µg/dL on day 3, and 15.69 µg/dL on day 6. The reported day-3 change from baseline was 5.01 µg/dL (95% interval 0.95–9.06; p = .02), and the day-6 change was 4.50 µg/dL (95% interval 0.45–8.56; p = .03).
That pattern is pharmacologically plausible for glucocorticoid-receptor antagonism because blocking negative-feedback signaling can be associated with increased circulating cortisol. But a cortisol assay is not a measurement of plasma ethanol. The cortisol findings do not identify a driver’s BAC, demonstrate receptor occupancy in an individual, or prove that a particular behavior was caused by either drug. They support a separate endocrine observation within this controlled dosing schedule.
Why does the 1% “placebo” beverage matter?
The authors deliberately added a small amount of ethanol to the comparator beverage to make the sensory experience more similar to the 16% ethanol drink. As a matter of chemical fact, the comparator was not alcohol-free. That is not automatically a defective study design: masking is an important goal, and the investigators identified the dose in their methods. It does, however, mean an interpretation should not treat the comparison as an absolute zero-ethanol physiological baseline.
The protocol also separated the two beverages on a given study day by at least four hours, with the order reversed on day 5. Without individual-level ethanol time courses, it is not justified to reconstruct an unmeasured residual concentration at a particular later moment, nor to assert that no carryover was possible. The correct approach is to describe the administered beverages, timing, observed results, and design limitations without inventing an individual concentration.
How much can a Phase I trial of ten men exclude?
Small mechanistic and safety studies can be valuable for identifying clear signals under controlled conditions. They are usually poorly suited for ruling out modest differences, heterogeneous effects, low-frequency serious events, and interactions that emerge under other dosing patterns. This sample had ten male completers and no parallel group with randomized PT150 versus placebo-drug assignment. The drug sequence was fixed by day, even though beverage order was randomized. The analyses were exploratory and the authors did not adjust the reported p-values for multiple comparisons.
Selection matters as much as sample size. The trial excluded several populations particularly likely to appear in real clinical or forensic disputes: people with certain active psychiatric or substance-use disorders, medication coexposures, and abnormal screening results. It did not investigate alcohol withdrawal, chronic disease, prolonged PT150 treatment, different PT150 doses, or exposure to additional sedatives. The lack of a statistically detectable effect on a selected physiological variable cannot be generalized to a different substance or to a person taking a different medication merely because both medicines influence the central nervous system.
| Original comparison | Reported result | Narrow interpretation |
|---|---|---|
| Breath-derived eBAC peak, day 5 versus day 1 alcohol challenge | Mean peak 0.104 under both conditions; difference 0.00, 95% CI −0.03 to 0.03; p = .99 | No detected peak difference in this cohort; not direct whole-blood equivalence or proof of no pharmacokinetic interaction |
| eBAC time to sampled peak | Difference −0.60, 95% CI −25.08 to 23.89; p = .96 | No detected difference with this sampling plan; broad uncertainty relative to a single person’s timing |
| eBAC two-hour AUC | Difference −0.16, 95% CI −3.40 to 3.08; p = .91 | No detected two-hour exposure-summary difference; not a measurement of metabolic enzyme activity |
| Heart-rate change during ethanol challenge | Difference 0.57, 95% CI −1.01 to 2.16; p = .48 | No detected average difference in this selected physiological endpoint; not a driving test |
| Alcohol Urge Questionnaire change | Difference 1.30, 95% CI 0.66 to 1.95; p < .001 (source Table 2) | Subjective scale difference, complicated by different baseline values; no direct assay or demonstrated impairment |
| Plasma cortisol after repeated PT150 | Day 3 increase 5.01 µg/dL, 95% CI 0.95–9.06; p = .02 | A measured endocrine response—not a measured plasma alcohol concentration |
What questions would matter in forensic testimony?
Suppose a party cites the 2021 trial to assert that PT150 cannot change an individual’s alcohol concentration or that an unaffected breath estimate rules out a drug–alcohol interaction. A careful scientific evaluation would first identify what is being asserted and what source result is actually offered. Does the opinion concern a laboratory whole-blood BAC, an evidential breath-test result, a later model estimate, a subjective report, objective performance, an adverse medical event, or causation? These are different propositions with different evidentiary requirements.
The underlying case record would need the actual medication name, dose, timing, adherence, coexposures, testing instrument, sample type, collection time, analytical method, calibration or quality-control records, and, where impairment is alleged, observations or performance evidence. The Morice study provides none of those case-specific facts. It also did not validate a conversion formula that takes a participant’s PT150 status and computes an earlier BAC or predicts a driver’s error rate. A research-group p-value is not a person’s diagnostic likelihood.
For a measured concentration, begin with the forensic toxicology research library and identify the assay and matrix. For a disputed earlier concentration, the two-blood-sample elimination research brief illustrates why an observed interval and an extrapolated event-time value are separate questions. The site’s research and source policy explains the distinction between published literature and an adopted case-specific expert opinion.
Frequently asked questions about PT150 and alcohol evidence
Did the trial prove PT150 has no interaction with alcohol?
No. It found no statistically significant differences in selected physiological and breath-estimated concentration summaries under one small, short protocol. It was not designed to exclude all possible pharmacokinetic, behavioral, or rare safety interactions.
Was blood alcohol measured directly?
No. The reported ethanol concentration summaries were derived from an Alco-Sensor IV breath instrument. Blood collected for cortisol and other clinical chemistry does not convert the ethanol results into an independently assayed whole-blood BAC.
Did a placebo group receive absolutely no ethanol?
No. The comparator contained 1% ethanol, approximately 0.05 g/kg. The higher-dose challenge contained 16% ethanol, providing 0.8 g/kg. The lower dose was used to aid masking.
Did everyone receive PT150 in random order?
No. Beverage order was randomized and later reversed, but every participant completed pre-PT150 challenges on day 1 and PT150-exposed challenges on day 5. The drug-treatment sequence itself was not counterbalanced.
Were objective driving or field sobriety tests administered?
No validated driving, crash-risk, standardized field sobriety, or individual impairment classification was a reported trial endpoint. Self-rated sedation or craving and ordinary safety vital signs cannot replace those measurements.
Do the findings establish safety for women, patients with alcohol use disorder, or people taking other drugs?
No. All completers were men, and the screening criteria excluded several comorbidities and medication exposures. Extrapolation beyond the studied group requires additional data.
What is the strongest single conclusion for a legal or scientific reader?
This ten-person experiment found no detectable difference in selected breath-estimated ethanol peaks or brief physiological safety outcomes after short PT150 exposure, while some subjective and endocrine measures differed. It does not provide a validated person-specific inference about blood BAC, sobriety, functional impairment, or causation.
Primary research source
Morice C, Baker DG, Patel MM, Nolen TL, Nowak K, Hirsch S, Kosten TR, Verrico CD. A randomized trial of safety and pharmacodynamic interactions between a selective glucocorticoid receptor antagonist, PT150, and ethanol in healthy volunteers. Scientific Reports. 2021;11:9876. doi:10.1038/s41598-021-88609-6. The discussion above appraises this original small primary experiment; its data are not a clinical prescribing recommendation, a statute-specific legal standard, or a conclusion about an unexamined individual.