Reduced Exposure Study Using THS 2.2 Menthol With 5 Days in a Confinement Setting Followed by 86 Days in an Ambulatory Setting: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Reduced Exposure Study Using THS 2.2 Menthol With 5 Days in a Confinement Setting Followed by 86 Days in an Ambulatory Setting (NCT01989156), a randomized na study. The primary outcome is Concentration of Monohydroxybutenylmercapturic Acid (MHBMA).

2. Clinical Question

3. Trial Design

This was a randomized, controlled, open-label, three-arm, parallel-group, multicenter study. The registry lists the allocation as randomized and masking as none. The arms were THS 2.2 Menthol, menthol conventional cigarettes, and smoking abstinence. Participants assigned to the product-use arms used their assigned product ad libitum; the abstinence arm involved abstaining from smoking.

The study had a 5-day confinement period followed by an 86-day ambulatory period. The primary endpoint focused on measurements at Day 5, so it concerns the confinement phase rather than the full follow-up period. The registry record lists enrollment as 160. It lists the study start as 2013-12, primary completion as 2014-10, and status as completed. The phase is recorded as NA.

The primary analysis population was the Per Protocol set. It included randomized subjects who had no major protocol deviation impacting evaluability during confinement. This is narrower than a population consisting of every randomized subject, and the registry record does not give further detail about exclusions or the reasons for them. Arm-level numbers for the primary endpoint are listed as 65.0 for THS 2.2 Menthol, 30.0 for menthol conventional cigarettes, and 21.0 for smoking abstinence.

The registry record does not report the randomization sequence method, allocation concealment, site count, planned sample-size justification, or a detailed analysis plan. It also does not report why the Per Protocol set differs in size across arms. Those omissions limit assessment of design implementation and the representativeness of the primary analysis population.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Enrolled and randomized population = 160 subjects * 80 subjects in mTHS 2.2 * 41 subjects in mCC * 39 subjects in SA Number of subjects exposed to the product test = 165 * Number of subjects enrolled but NOT randomized = 4 * Number of subjects failed screening = 1

Confinement Period (Population at Day 5)

Reasons for not completing this period:

End of Study (Population at Day 90)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The analysis was performed on the Per Protocol set (PP) which included all randomized subjects who had no major protocol deviation impacting the evaluability during the confinement period.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Study population = all randomized subjects who had at least one post randomization product use experience and at least one valid biomarker of exposure measurement. 160 randomized subjects: 80 in mTHS 2.2, 41 in mCC and 39 in SA arms

6. Primary Endpoint

The registry lists the primary outcome as Concentration of Monohydroxybutenylmercapturic Acid (MHBMA), measured over 5 days.

The primary endpoint was the concentration of monohydroxybutenylmercapturic acid (MHBMA) in urine at Day 5, adjusted for creatinine. The unit was pg/mg creat. The registry describes MHBMA concentrations as a biomarker of exposure to selected smoke constituents. Creatinine adjustment expresses the urinary concentration relative to creatinine, which is used to account for variation in urine concentration.

The endpoint is a biomarker rather than a clinical outcome such as illness, symptoms, or survival. Biomarkers can help quantify exposure over a defined measurement period, making them useful in studies that compare exposure under different product-use or abstinence conditions. They do not, by themselves, establish how a participant’s health will change over time. That distinction is particularly relevant because the primary endpoint was measured at Day 5, while the study regimen continued into an ambulatory period.

The registry reports geometric means and 95% confidence intervals as descriptive statistics for the endpoint. Geometric summaries are suited to positive measurements that may vary multiplicatively or have a skewed distribution. The registry does not report the laboratory assay details, the handling of measurements below detection limits, or further validation information for this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a positive concentration measure, and the registry reports geometric least-squares means. In general, a geometric mean summarizes multiplicative variation by working on a logarithmic scale and then expressing the result on the original scale. It is often useful when measurements are right-skewed, because a small number of high values can pull an arithmetic mean upward. The registry does not state the distributional rationale used by the investigators, so this is a general explanation rather than a claim about their specific decision process.

What does a confidence interval express?

A confidence interval describes the uncertainty around an estimated quantity under the statistical model and its assumptions. The reported 95% intervals accompany the geometric least-squares means and the ANCOVA ratio. A narrower interval indicates greater precision for that estimate, while a wider interval indicates more uncertainty. The interval does not mean that nearly all individual measurements fall within its bounds, nor does it account automatically for design or measurement limitations.

For THS 2.2 Menthol, the reported interval for the Day 5 geometric least-squares mean is 93.49 to 131.7 pg/mg creat. For menthol conventional cigarettes, it is 687.2 to 1133.0 pg/mg creat. For smoking abstinence, it is 70.67 to 127.6 pg/mg creat. The ANCOVA ratio has a 95% confidence interval from 9.27 to 17.05. These intervals describe different quantities: the first set concerns individual arm summaries, and the latter concerns the modeled comparison.

What does ANCOVA add?

ANCOVA, or analysis of covariance, compares groups while accounting for covariates included in the model. In general, adjustment can improve precision or address baseline differences when the model and covariates are appropriate. The registry reports an ANCOVA geometric least-squares mean ratio for the THS 2.2 Menthol and menthol conventional-cigarette arms. It does not report the covariates, their definitions, or the model specification. The adjustment therefore cannot be independently assessed from the registry information summarized in this analysis.

The ANCOVA result is not necessarily identical to a ratio calculated directly from the descriptive means. Adjusted estimates can depend on covariates, model form, and the analysis population. The registry’s reported estimate should be understood as a model-based result. No fresh contrast should be inferred by combining the arm summaries, and no recalculated value is reported in the registry record.

Why use a ratio?

For positive, potentially skewed outcomes, a model may analyze log-transformed values and report a ratio on the original scale. In general, a ratio compares the modeled level in one group with that in another. Its interpretation depends on which group is the numerator and which is the denominator. Although the registry identifies the two groups in the analysis, the available record does not give the contrast coding or covariate specification. The ratio of 12.58 should therefore be quoted as the reported ANCOVA result without assigning an unreported direction.

How should the p-value be read?

The reported p-value is p < 0.001. In general, a p-value describes how compatible the observed data are with a specified null hypothesis, under the model and assumptions used. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. The registry does not provide the full test specification, so the reported value should be interpreted as attached to the sponsor’s ANCOVA comparison.

Statistical significance does not remove uncertainty about bias, measurement, or generalizability. Nor does a small p-value convert a biomarker result into evidence about long-term clinical outcomes. The estimated ratio and its confidence interval provide information about the magnitude and precision of the modeled contrast; the p-value addresses compatibility with the null under the analysis.

What matters about the analysis population?

The primary analysis used the Per Protocol set, defined in the registry by randomization and absence of a major protocol deviation impacting evaluability during confinement. In general, a Per Protocol analysis can describe results among participants who followed specified requirements, but exclusions after randomization may affect comparability if the reasons for exclusion relate to outcomes or treatment. The registry record does not report the detailed deviations, exclusion counts by reason, or a sensitivity analysis using another population.

Sample-size planning illustrations, when shown alongside trial results, are not observed power. The registry’s planning assumptions describe an illustration based on an observed effect, two-sided significance level 0.05, equal allocation, and a normal approximation. Such planning calculations do not change the observed analysis or establish that the trial was designed with those assumptions. The registry record does not report the trial’s own sample-size justification in the information.

10. Analyses Reported by the Sponsor

The sponsor-reported analysis is an ANCOVA comparison of THS 2.2 Menthol and menthol conventional cigarettes. It gives a geometric least-squares mean ratio of 12.58, a 95% confidence interval from 9.27 to 17.05, and p < 0.001. This is the result to use when describing the sponsor’s own inferential analysis.

The registry also gives descriptive geometric least-squares means and confidence intervals for all three arms. Those descriptive summaries are not a substitute for the ANCOVA estimate: the adjusted contrast can depend on the model and covariates, which are not reported in the registry record. The ratio should not be reconstructed from the displayed group summaries, and its direction should not be inferred without the contrast definition.

A separate recalculated estimate for this primary comparison is not reported in the registry record. The record does not flag a discrepancy between a sponsor analysis and a recalculation. Accordingly, there is no conflicting recalculated result to discuss; the sponsor’s reported ANCOVA analysis is the available inferential comparison.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and reports one inferential analysis for the comparison between THS 2.2 Menthol and menthol conventional cigarettes. The endpoint is MHBMA concentration at Day 5. The record also indicates five outcome measures and five analyses, but does not provide additional secondary endpoint results in the information available for this analysis.

The registry record does not report whether the investigators planned a multiplicity adjustment, how the primary comparison was selected among possible comparisons, or whether any interim analyses were conducted. Without that information, it is not possible to determine how the reported p-value fits into a broader family of planned tests. In general, when several endpoints or comparisons are tested, the chance of obtaining at least one small p-value can increase; adjustment or a prespecified testing hierarchy may be used to address that issue. This general principle does not establish what this trial did.

The record identifies the Per Protocol set for the primary endpoint, but detailed missing-data methods are not reported. The registry does not state how missing urine measurements were handled, whether any imputation was used, or whether sensitivity analyses assessed the effect of missing values. It also does not report the number of participants with missing endpoint data or reasons for missingness by arm. Those details matter because missingness can affect both the analyzed population and the reliability of comparisons.

The primary population excluded randomized subjects with a major protocol deviation impacting evaluability during confinement. The registry does not give a full account of deviations or explain whether an intention-to-treat analysis was conducted for this endpoint. These omissions limit assessment of the effect of post-randomization exclusions. The registry record also does not report a prespecified analysis plan or a sample-size justification, so the intended treatment of multiplicity, missingness, and analysis populations cannot be verified from the available information.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists serious adverse events for each arm as 0 affected participants. The corresponding numbers at risk are 80 for THS 2.2 Menthol, 41 for menthol conventional cigarettes, and 39 for smoking abstinence. These are the posted serious-adverse-event figures; they do not establish that no event could occur outside the recorded observation or reporting process.

The registry also lists other adverse events, with 31 affected participants in the THS 2.2 Menthol arm, 12 in the menthol conventional-cigarette arm, and 16 in the smoking-abstinence arm. The numbers at risk shown for those figures are 80, 41, and 39, respectively. No totals or rates are calculated. The registry record does not give the event terms, severity breakdown, attribution, or timing in the information available for this analysis. It also does not report death counts as numeric values.

Safety figures should be read alongside the study design, follow-up, and completeness of reporting. The primary result concerns an exposure biomarker, not safety. The adverse-event counts alone do not establish a comparative safety conclusion, particularly without event-level detail and a stated analysis method.

13. What the Result Does, and Does Not, Mean

The posted results describe urinary MHBMA concentrations at Day 5 among participants in the Per Protocol set. The descriptive geometric least-squares mean for menthol conventional cigarettes is higher than the reported means for THS 2.2 Menthol and smoking abstinence. The sponsor’s ANCOVA analysis reports a geometric least-squares mean ratio of 12.58 for the THS 2.2 Menthol and menthol conventional-cigarette comparison, with a 95% confidence interval from 9.27 to 17.05 and p < 0.001. These are findings about the stated biomarker and analysis.

The endpoint is a measure of exposure to a selected smoke constituent, adjusted for creatinine. It is not a direct measure of disease, symptoms, or survival. The study’s description of THS 2.2 Menthol as a candidate Modified Risk Tobacco Product does not make the biomarker result proof of reduced health risk. A change in one exposure marker cannot, by itself, establish the overall health effects of a product or predict long-term outcomes.

The primary measurement was at Day 5, during the confinement period. Although the assigned regimens continued for 86 days in an ambulatory setting, the primary endpoint result does not characterize the full course of exposure or later clinical consequences. The registry record does not report further primary endpoint measurements beyond the stated timeframe.

Interpretation is also bounded by the analysis population. The result comes from the Per Protocol set, and the available registry information does not detail exclusions, missing measurements, model covariates, or sensitivity analyses. The reported comparison is therefore best understood as the sponsor’s model-based estimate for the specified endpoint and population, rather than a comprehensive assessment of product effects.

These results should not be generalized automatically to populations outside the study description or to outcomes that were not reported. The registry record does not provide evidence about long-term health, other biomarkers, or comparative clinical benefit. The findings support a description of the Day 5 biomarker data and the reported ANCOVA contrast, with the limitations stated above.

14. Why This Trial Matters Statistically

This study illustrates the difference between measuring exposure and measuring health. A urinary biomarker can quantify exposure to a selected constituent over a defined period, and the MHBMA endpoint offers a measurable outcome for comparing the study arms at Day 5. Such a result is informative about the biomarker, but it does not directly answer whether a product changes long-term clinical outcomes.

It also illustrates why the estimand and analysis population matter. The descriptive values apply to the Per Protocol set, not necessarily to every randomized participant. The sponsor’s ANCOVA comparison is a model-based result, while the arm summaries are descriptive. Keeping those quantities distinct prevents an adjusted inference from being treated as a simple calculation based on group means.

The reported confidence intervals help communicate precision, and the p-value conveys evidence against a specified null under the analysis assumptions. Neither resolves unreported details such as model covariates, missing-data handling, or multiplicity procedures. A statistically strong result can still have a narrow scope if it concerns one endpoint, one time point, and a selected analysis population.

For clinical researchers, the record highlights the value of transparent reporting: defining the analysis contrast, specifying model covariates, documenting exclusions, and explaining missing-data and multiplicity strategies. Those details help readers judge how robust an estimate may be and what population it describes. The registry results are most useful when read within the endpoint’s measurement scope and the limitations of the available record.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The primary result is limited to MHBMA concentration in urine, adjusted for creatinine, at Day 5. It does not establish findings for other biomarkers, later time points, or clinical outcomes. The trial continued into an 86-day ambulatory period, but the primary endpoint described in the registry is measured during the confinement phase.

The primary endpoint analysis used the Per Protocol set. The registry does not report the detailed protocol deviations, the reasons participants were excluded from this set, or an accompanying intention-to-treat analysis. The primary arm counts are 65.0, 30.0, and 21.0, while the adverse-event entries list different numbers at risk; the registry information does not explain these differences. These populations should not be assumed to be interchangeable.

The sponsor reports an ANCOVA ratio, but the registry record does not describe the covariates, model specification, contrast direction, or assumptions used. As a result, the adjusted estimate cannot be independently reconstructed or fully evaluated from the available record. No separate recalculated estimate is reported for the primary comparison.

The record does not report detailed missing-data methods, endpoint missingness, sensitivity analyses, multiplicity procedures, interim analyses, or the trial’s sample-size justification. It also does not provide assay details or information about measurements below detection limits. These gaps constrain judgments about robustness, precision, and the potential impact of analytic choices.

The study was open-label, with masking listed as none. The registry does not describe whether knowledge of assignment could have affected measurement or participant behavior. The record’s description of the enrolled group is limited to apparently healthy smokers, which also limits how broadly the result can be applied. These considerations do not negate the reported biomarker result, but they define the scope of the evidence available.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Reduced Exposure Study Using THS 2.2 Menthol With 5 Days in a Confinement Setting Followed by 86 Days in an Ambulatory Setting — NCT01989156.
• Registry record last updated 2020-03-24; status last verified 2020-03.
• Results first posted 2017-04-05.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry describes a randomized, controlled, open-label, three-arm, parallel-group, multicenter study of THS 2.2 Menthol, menthol conventional cigarettes, and smoking abstinence. The study included 5 days in confinement followed by 86 days in an ambulatory setting. The primary endpoint was urinary MHBMA concentration, adjusted for creatinine, at Day 5.

For the Per Protocol analysis, the reported geometric least-squares means were 111.0 pg/mg creat for THS 2.2 Menthol, 882.2 pg/mg creat for menthol conventional cigarettes, and 94.97 pg/mg creat for smoking abstinence. The sponsor-reported ANCOVA comparison between the first two arms was a geometric least-squares mean ratio of 12.58, with a 95% confidence interval from 9.27 to 17.05 and p < 0.001.

The findings describe a short-term exposure biomarker, not a clinical outcome. The registry does not report several details needed to fully assess the analysis, including model covariates, missing-data methods, and multiplicity procedures.