This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
TIMELESS was a randomized, parallel, quadruple-masked phase 3 trial evaluating tenecteplase versus placebo in stroke patients between 4.5 and 24 hours. The registry identifies the ordinal modified Rankin Scale (mRS) score at Day 90 as the single registered primary endpoint.
| Feature | TIMELESS |
|---|---|
| Trial name | TIMELESS |
| ClinicalTrials.gov identifier | NCT03785678 |
| Brief title | Tenecteplase in Stroke Patients Between 4.5 and 24 Hours |
| Therapeutic area | Neurology |
| Condition | Thrombolysis |
| Phase | Phase 3 |
| Status | Completed |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Primary purpose | Treatment |
| Enrollment | 458.0 |
| Arms | 2 |
| Interventions | Tenecteplase (biological); Placebo (other) |
| Lead sponsor | Genentech, Inc. |
| Sponsor type | Industry |
2. Clinical Question
The clinical question can be expressed directly from the registered population, interventions, and primary endpoint: in stroke patients between 4.5 and 24 hours, what is the difference between tenecteplase and placebo in the ordinal modified Rankin Scale score at Day 90?
Population
Stroke patients between 4.5 and 24 hours, according to the trial's brief title.
Intervention
Tenecteplase, classified in the registry as a biological intervention.
Comparator
Placebo, classified in the registry as an other intervention.
Primary question
How do the randomized groups differ in ordinal mRS score at Day 90?
3. Trial Design
Patients were allocated to the study arms using randomization.
The registry identifies a parallel-group design with 2 arms.
The ClinicalTrials.gov record identifies the trial as quadruple-masked.
The registered primary purpose is treatment.
Tenecteplase
- Intervention type: biological
- Randomized comparison arm
Placebo
- Intervention type: other
- Randomized comparison arm
Trial timeline
Trial start
The registry lists March 2, 2019 as the study start date.
Primary completion
The registry lists February 28, 2023 as the primary completion date.
Study status
The ClinicalTrials.gov record identifies the study status as completed.
4. Primary Endpoint
| Endpoint | Time frame | Registry definition |
|---|---|---|
| Ordinal Modified Rankin Scale (mRS) Score at Day 90 | Day 90 | the modified Rankin score (mRS) is a 6-point scale commonly used to assess disability due to stroke, with higher values indicating worse outcomes. 0 = No symptoms 1. = No significant disability 2. = Slight disability 3. = Moderate disability 4. = Moderately severe disability 5. = Severe disability 6. = Death |
The endpoint is ordinal rather than simply binary. The possible mRS values progress from 0 through 6, with higher values indicating worse outcomes. That structure contains more information than a single yes/no endpoint because it preserves the ordering of disability categories.
The numerical values are ordered categories. The difference between adjacent scores should not automatically be treated as an equal quantitative distance.
5. Results
The available trial data identify the registered primary endpoint as ordinal mRS at Day 90, but they do not provide a formal between-group statistical analysis for that endpoint. No hazard ratio, odds ratio, confidence interval, p-value, median, event count, or category-specific mRS result is reported in the ClinicalTrials.gov record used for this page.
What would normally be analysed?
For a randomized trial with an ordinal mRS endpoint, an appropriate primary analysis can use an ordinal regression framework that respects the ordering of the mRS categories. A commonly used approach is a proportional-odds ordinal logistic model. Depending on the prespecified statistical analysis plan, the treatment effect can be summarized as an odds ratio for being at or below a given level of disability across the ordered categories.
Here, k represents an mRS threshold and the model uses the ordered nature of the outcome. The proportional-odds assumption means that the treatment coefficient is shared across the cumulative thresholds.
This is a description of a typical statistical approach for an ordinal endpoint, not a claim about the exact analysis used in TIMELESS. The ClinicalTrials.gov record does not report the formal statistical-analysis method.
6. Safety Results
The ClinicalTrials.gov record reports serious adverse events by randomized arm using affected patients over the corresponding at-risk denominator.
| Safety measure | Tenecteplase | Placebo |
|---|---|---|
| Serious adverse events, affected / at risk | 91/218 | 102/214 |
The denominators posted on ClinicalTrials.gov for the serious-adverse-event measure are 218 and 214, respectively. They are not the same as the overall enrollment of 458.0, so the safety figures should be interpreted using the reported affected/at-risk denominators rather than treating them as percentages of total enrollment.
Tenecteplase
Serious adverse events were reported in 91 of 218 patients at risk.
Placebo
Serious adverse events were reported in 102 of 214 patients at risk.
7. Statistical Methodology
Randomization
Randomization is central to the design because the treatment assignment is determined before the outcome is observed. In a parallel randomized trial, the randomized groups provide the basis for comparing outcomes while reducing systematic allocation differences that could otherwise confound treatment comparisons.
Ordinal outcome analysis
The Day 90 mRS is an ordered categorical endpoint. Its defining feature is that the categories have a meaningful order, but the numerical labels 0 through 6 should not automatically be interpreted as measurements on an interval scale.
The ordering is meaningful, while the numerical spacing between categories does not necessarily represent equal changes in disability.
Proportional-odds model
A proportional-odds model is one standard framework for analysing an ordinal endpoint such as mRS. Rather than collapsing the endpoint into one threshold, it models the cumulative probability of being at or below each possible threshold.
If a treatment odds ratio were reported from such a model, interpretation would depend on the direction used for the cumulative outcome. The precise definition of the reported effect measure therefore matters: an odds ratio cannot be interpreted correctly without knowing which outcome category is being modelled and which group is in the numerator.
Confidence intervals
A confidence interval around an ordinal treatment-effect estimate would quantify statistical uncertainty under the specified model and sampling framework. It would not describe the range of individual patient responses and would not establish that every threshold of the mRS has the same treatment effect.
Missing outcome data
Ordinal outcomes measured at a fixed time point require careful handling of patients whose Day 90 assessment is unavailable. The ClinicalTrials.gov record does not describe the missing-data strategy or any imputation method used in TIMELESS. Consequently, no specific imputation method is attributed to this trial.
Analysis population
The ClinicalTrials.gov record identifies randomized allocation but do not specify the statistical analysis population used for the primary endpoint. An analysis based on randomized assignment would preserve the principal advantages of the randomized design, but the exact analysis population and any exclusions should be taken from the prespecified statistical analysis plan before making that claim about TIMELESS.
8. Statistical Methods Explained
Why is mRS analysed as an ordinal endpoint?
The mRS categories have a natural order from no symptoms through increasingly severe disability and death. Treating the endpoint as ordinal allows the analysis to use that ordering rather than reducing the outcome to a single arbitrary cutoff.
What does an ordinal regression model estimate?
An ordinal regression model evaluates the relationship between treatment assignment and the probability distribution across ordered categories. In a proportional-odds formulation, the same treatment coefficient is used across cumulative mRS thresholds, subject to the proportional-odds assumption.
What is the proportional-odds assumption?
The proportional-odds assumption states that the treatment effect on the cumulative odds is constant across the outcome thresholds represented in the model. If that assumption does not hold, a single odds ratio may not adequately describe the treatment effect across the full mRS distribution.
Why is a binary analysis potentially less informative?
A binary analysis requires selecting a cutoff, such as defining a favorable outcome below a particular mRS value. That can be clinically useful, but it discards information from the other categories. An ordinal analysis can retain more of the observed disability distribution.
Why does randomization matter for the analysis?
Randomization establishes treatment assignment independently of patients' observed and unobserved characteristics in expectation. This makes the between-group outcome comparison the principal basis for estimating the treatment effect, subject to the trial's conduct and analysis assumptions.
Why should the registry result and the statistical interpretation be separated?
The ClinicalTrials.gov record identifies the endpoint and report that results are posted, but they do not provide a formal statistical-analysis record. A statistically appropriate method can therefore be explained educationally without presenting an unreported estimate as though it were an observed TIMELESS result.
9. Blinding and Randomization
TIMELESS is described in the ClinicalTrials.gov record as randomized and quadruple-masked. These design features are particularly relevant when interpreting an outcome such as the mRS, which involves assessment of functional status after stroke.
Randomization
Randomization provides the design framework for comparing the tenecteplase and placebo groups without assigning treatment according to observed prognosis.
Quadruple masking
The registry identifies the study as quadruple-masked. Masking can reduce the influence of treatment knowledge on trial conduct and outcome assessment.
The ClinicalTrials.gov record does not identify the four parties covered by the quadruple-masking designation. This page therefore does not assign specific roles to the masked groups.
10. Endpoint Interpretation
The Day 90 mRS measures functional disability on an ordered scale. Lower values represent better outcomes under the registry definition, while a score of 6 represents death.
The mRS is not a direct measurement of thrombolytic activity, infarct volume, treatment exposure, or mortality alone. It is a functional outcome that combines disability states into an ordered clinical scale.
Two treatment groups can have the same proportion below one chosen mRS threshold while having different distributions across the remaining categories. Ordinal analysis is designed to preserve information across those ordered categories.
11. Important Statistical Interpretation Issues
- Ordinality: mRS categories are ordered, but the numerical distance between adjacent categories should not automatically be treated as equal.
- Proportional odds: if an ordinal logistic model is used, its proportional-odds assumption should be evaluated or otherwise justified.
- Direction of effect: interpretation of an ordinal odds ratio depends on whether the model is defined around higher or lower mRS categories and which treatment group is the reference.
- Fixed-time assessment: the registered endpoint is measured at Day 90, making missing Day 90 assessments relevant to the analysis population and missing-data strategy.
- Randomized comparison: treatment comparisons should remain anchored to randomized assignment when the estimand is intended to reflect treatment assignment rather than treatment received.
- Multiple analyses: the ClinicalTrials.gov record does not describe a multiplicity strategy, so no claims about multiplicity adjustment are made here.
- Formal analysis record: no formal statistical analyses were posted. Educational descriptions of suitable methods should not be mistaken for reported TIMELESS analysis results.
12. What the Available Results Can and Cannot Establish
| Question | Supported by the ClinicalTrials.gov record? | Reason |
|---|---|---|
| Was TIMELESS randomized? | Yes | the ClinicalTrials.gov record specifies randomized allocation. |
| Was the design parallel? | Yes | The registry design model is parallel. |
| Was the study quadruple-masked? | Yes | The registry masking field specifies quadruple masking. |
| What was the enrollment? | Yes | the ClinicalTrials.gov record reports 458.0. |
| What was the registered primary endpoint? | Yes | Ordinal mRS Score at Day 90. |
| What were the primary-endpoint effect estimates? | No | No statistical analyses are reported. |
| What were the primary-endpoint confidence intervals? | No | No statistical analyses are reported. |
| What were the primary-endpoint p-values? | No | No statistical analyses are reported. |
| Were serious adverse events reported? | Yes | 91/218 for tenecteplase and 102/214 for placebo. |
| Was a formal comparative safety analysis reported? | No | No statistical analysis is included in the ClinicalTrials.gov record. |
13. Why This Trial Matters Statistically
TIMELESS is useful as a teaching case because it illustrates a common situation in clinical-trial statistics: the scientific question and endpoint can be clearly defined even when a registry extract does not provide the formal statistical comparison needed to quantify the treatment effect.
| Concept | How it appears in TIMELESS |
|---|---|
| Randomization | The trial uses randomized allocation. |
| Parallel-group design | The registry identifies a parallel design with 2 arms. |
| Blinding | The trial is described as quadruple-masked. |
| Ordinal outcome | The primary endpoint is ordinal mRS at Day 90. |
| Functional outcome | The mRS is used to assess disability after stroke. |
| Ordinal regression | A proportional-odds model is a standard framework that can be considered for an ordered endpoint of this type. |
| Missing data | Fixed-time functional assessment makes incomplete Day 90 measurements analytically important. |
| Safety analysis | Serious adverse events are reported by arm using affected/at-risk counts. |
| Registry interpretation | Results are posted, but the registry-reported statistical-analysis field contains no formal analyses. |
14. Limitations
- No formal statistical analyses reported: no formal statistical analyses were posted, so treatment-effect estimates, confidence intervals, and p-values cannot be reported from the registry record.
- Limited endpoint results: although the registry indicates that results are posted and 11 outcome measures are posted, the ClinicalTrials.gov record provides the primary endpoint definition but not its numerical result.
- No subgroup results reported: no subgroup estimates or interaction analyses are included in the trial data.
- No baseline characteristics reported: the available data do not contain a baseline demographic or clinical characteristics table.
- No analysis population details reported: the ClinicalTrials.gov record does not identify an ITT, modified ITT, per-protocol, or other primary efficacy population.
- No missing-data strategy reported: the registry extract does not specify how missing Day 90 mRS measurements were handled.
- No multiplicity strategy reported: the available data do not describe endpoint hierarchy, alpha allocation, or multiplicity adjustment.
- No interim-analysis information reported: the available trial data do not identify interim looks, stopping boundaries, or alpha-spending procedures.
- No crossover information reported: the available data do not describe treatment crossover or rescue treatment.
- Safety denominators differ from total enrollment: the reported serious-adverse-event counts use denominators of 218 and 214, rather than the overall enrollment of 458.0.
15. Clinical Interpretation vs Statistical Interpretation
Clinical interpretation
The registered clinical outcome is functional status at Day 90, measured using the modified Rankin Scale. Lower mRS values correspond to better functional outcomes under the registry definition.
Statistical interpretation
The endpoint is ordinal and therefore calls for an analysis that respects ordering. The ClinicalTrials.gov record does not identify the formal TIMELESS treatment-effect estimate or its inferential statistics.
Keeping these two levels separate is important. The clinical endpoint tells us what patient outcome the trial was designed to evaluate. The statistical analysis determines how the randomized comparison is quantified and how uncertainty around that comparison is assessed.
16. Serious Adverse Events: Statistical Perspective
The reported serious-adverse-event data are presented as affected patients divided by the number at risk:
These are the exact affected/at-risk counts reported in the ClinicalTrials.gov record.
A comparative safety analysis could express the difference using a risk ratio, risk difference, odds ratio, or another prespecified estimand. The appropriate choice depends on the analysis plan, follow-up definition, and denominator convention. None of those formal comparative estimates is reported here, so they are not calculated on this page.
17. Statistical Concepts in This Trial
Learn more about the methods and design concepts most relevant to understanding TIMELESS:
18. Related Statistical Calculators
19. Sources
- ClinicalTrials.gov: TIMELESS — NCT03785678.
- Linked publication: PubMed record — PMID 38329148.
- Linked publication: PubMed record — PMID 35369714.
- Linked publication: PubMed record — PMID 35262424.
Continue through the Clinical Biostats library
Explore statistical tutorials and calculators related to randomized trials, ordinal outcomes, blinding, randomization, and clinical safety analysis.
20. Record Summary
TIMELESS is a completed randomized, parallel, quadruple-masked phase 3 trial with 458.0 enrolled participants evaluating tenecteplase versus placebo in stroke patients between 4.5 and 24 hours. Its registered primary endpoint is the ordinal modified Rankin Scale score at Day 90. The statistical structure of that endpoint is important: mRS is an ordered categorical outcome, making ordinal methods such as proportional-odds regression a natural framework for consideration when specified by the statistical analysis plan.
The ClinicalTrials.gov record also report serious adverse events as 91/218 for tenecteplase and 102/214 for placebo. However, the ClinicalTrials.gov record is empty. Consequently, this page deliberately separates the documented trial design and reported safety counts from statistical quantities that are not provided, rather than reconstructing an unreported treatment effect, confidence interval, or p-value.