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
DESTINATION was a randomized, parallel, triple-masked, phase 3 trial of tezepelumab in adults and adolescents with severe, uncontrolled asthma. The registry reports 951 participants, two intervention groups, two registered primary endpoints, three posted outcome measures, and two posted statistical analyses.
| Feature | DESTINATION |
|---|---|
| Trial name | DESTINATION |
| ClinicalTrials.gov identifier | NCT03706079 |
| Phase | Phase 3 |
| Status | Completed |
| Therapeutic area | Pulmonology |
| Condition | Asthma |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Triple |
| Primary purpose | Treatment |
| Enrollment | 951 |
| Interventions | Tezepelumab (biological); Placebo (other) |
| Lead sponsor | AstraZeneca |
| Sponsor type | Industry |
2. Clinical Question
The registry describes DESTINATION as an extension study evaluating the safety and tolerability of tezepelumab in adults and adolescents with severe, uncontrolled asthma. The statistical ClinicalTrials.gov record also include a comparison of annualized asthma exacerbation rate between randomized groups from predecessor studies.
Population
Adults and adolescents with severe, uncontrolled asthma enrolled in the DESTINATION extension study.
Intervention
Tezepelumab, identified in the registry data as a biological intervention.
Comparator
Placebo, identified in the registry data as an other intervention.
Primary question
How do the registered exposure-adjusted incidence rates of adverse events and serious adverse events, together with total time at risk, characterize the safety and tolerability of the study treatment?
3. Trial Design
Tezepelumab
- Tezepelumab.
- Registry classification: biological intervention.
Placebo
- Placebo.
- Registry classification: other intervention.
The registry record identifies the study as an extension study and gives a follow-up framework extending to Week 104. The registry-reported endpoint text also refers to subjects switching from placebo in a predecessor study. The available trial data therefore need to be interpreted in the context of both the DESTINATION extension and the predecessor-study treatment histories.
4. Endpoints
DESTINATION has two registered primary endpoints. Both are safety/time-at-risk measures rather than conventional time-to-event efficacy endpoints.
| Registered primary endpoint | Time frame | Endpoint type | Registry definition reported |
|---|---|---|---|
| Exposure Adjusted Incidence Rates of AEs/SAEs | Baseline (Week 0 in predecessor study) to Week 104. For subjects switching treatments from placebo in the predecessor to tezepelumab in DESTINATION, all data collected after first dose of tezepelumab are excluded. | Count / rate | Includes adverse events with an onset date between the date of first dose of IP in the predecessor and minimum (date of last dose of IP + 33 days, date of death, date of study withdrawal, day prior to start of another biologic). The analysis is based on the Safety Analysis Set. |
| Total Time at Risk | Baseline (Week 0 in predecessor study) to Week 104. For subjects switching treatments from placebo in the predecessor to tezepelumab in DESTINATION, all data collected after first dose of tezepelumab are excluded. | Other / unclear | Includes time between the date of first dose of IP in the predecessor and minimum (date of last dose of IP + 33 days, date of death, date of study withdrawal, day prior to start of another biologic). The analysis is based on the Safety Analysis Set. |
Secondary endpoint represented in the posted analyses
The statistical analyses posted on ClinicalTrials.gov for DESTINATION concern Annualized Asthma Exacerbation Rate (AAER), measured in events per year. The posted effect measure is a rate ratio, with superiority as the hypothesis type.
| Outcome measure | Unit | Endpoint type | Effect measure | Hypothesis |
|---|---|---|---|---|
| Annualized Asthma Exacerbation Rate (AAER) | Events per year | Count / rate | Rate ratio | Superiority |
5. Statistical Methodology
The ClinicalTrials.gov record identifies rate ratio as the effect measure for the posted statistical analyses. However, the statistical method and method category are both explicitly reported as Not reported. This distinction matters: the effect measure can be described without attributing an unreported regression model or distributional assumption to the trial.
Exposure-adjusted incidence rates
The primary safety endpoint is expressed as an exposure-adjusted incidence rate. Conceptually, this type of measure accounts for differing amounts of time at risk by relating the number of subjects experiencing adverse events to the aggregate time at risk contributed by the safety analysis population.
The DESTINATION registry definition specifies the safety analysis set and a time-at-risk framework based on treatment exposure and specified stopping points. The registry-reported extract does not report a named statistical model for the primary endpoint.
Total time at risk
Total time at risk is important because a simple count of adverse events does not account for how long participants were observed. The registered endpoint explicitly measures the amount of time contributed under the study's predefined exposure framework.
Rate ratio
A rate ratio compares event rates between two groups. A value below 1 indicates a lower estimated event rate in the numerator group relative to the denominator group, whereas a value above 1 indicates a higher estimated rate.
For the posted AAER analyses, the reported rate ratios compare randomized tezepelumab and placebo groups from the specified predecessor studies.
Superiority framework
The posted AAER analyses are classified in the registry data as superiority analyses. That means the statistical question is framed as whether the event rates differ in favor of one randomized group, rather than whether one treatment meets a non-inferiority margin.
6. Results
The ClinicalTrials.gov record contains two posted statistical analyses. Both are secondary analyses of Annualized Asthma Exacerbation Rate rather than formal comparative analyses of either registered primary endpoint.
Annualized Asthma Exacerbation Rate: NAVIGATOR randomized groups
Rate ratio
95% CI: 0.35–0.51 · Two-sided 95% CI · Superiority hypothesis
NAVIGATOR randomized tezepelumab vs NAVIGATOR randomized placebo
The reported rate ratio of 0.42 means that the estimated annualized exacerbation rate in the NAVIGATOR randomized tezepelumab group was 42% of the corresponding rate in the NAVIGATOR randomized placebo group, using the reported rate-ratio measure.
The estimate describes a relative rate comparison, not an absolute number of exacerbations prevented and not the proportion of individual patients who benefited. A rate ratio of 0.42 corresponds to a 58% lower estimated rate relative to the comparison group when expressed as 1 − 0.42.
The 95% confidence interval of 0.35–0.51 describes statistical uncertainty around the reported rate-ratio estimate. It does not mean that individual patients have rate ratios somewhere inside that interval.
The confidence interval is also not a measure of effect size. The estimate itself describes the relative rate; the interval describes its precision under the underlying statistical framework.
The ClinicalTrials.gov record does not identify the statistical model used to obtain this rate ratio. Therefore, this page does not attribute a Poisson model, negative-binomial model, offset specification, or other particular regression method to the analysis. The rate ratio should also not be interpreted as a hazard ratio or risk ratio.
Annualized Asthma Exacerbation Rate: SOURCE randomized groups
Rate ratio
95% CI: 0.38–0.96 · Two-sided 95% CI · Superiority hypothesis
SOURCE randomized tezepelumab vs SOURCE randomized placebo
The reported rate ratio of 0.61 means that the estimated annualized exacerbation rate in the SOURCE randomized tezepelumab group was 61% of the corresponding rate in the SOURCE randomized placebo group. Expressed as a relative rate reduction, this corresponds to 39% from 1 − 0.61.
The 95% confidence interval of 0.38–0.96 quantifies uncertainty around the reported rate-ratio estimate. Because the interval is relatively broad compared with the point estimate, the exact magnitude of the relative rate difference is less precisely determined than the point estimate alone might suggest.
The interval also remains below 1.00, which is consistent with a lower estimated exacerbation rate for the tezepelumab group under the reported superiority analysis. The p-value is not reported in the ClinicalTrials.gov record, so no p-value should be inferred from the confidence interval.
As with the NAVIGATOR analysis, the registry extract does not identify the statistical model used to generate the rate ratio. No specific count-regression model should therefore be assumed from the rate-ratio result alone.
What is and is not reported for the primary endpoints
The registry data state that both registered primary endpoints have results posted, but the posted statistical analyses do not provide a numerical estimate, confidence interval, or p-value for either primary endpoint. The two formal statistical-analysis records instead concern the secondary AAER endpoint.
| Primary endpoint | Results posted | Comparative estimate reported |
|---|---|---|
| Exposure Adjusted Incidence Rates of AEs/SAEs | Yes | No formal comparative estimate is included in the ClinicalTrials.gov record. |
| Total Time at Risk | Yes | No formal comparative estimate is included in the ClinicalTrials.gov record. |
For an exposure-adjusted rate endpoint, a formal comparison would ordinarily require an appropriate rate-based statistical framework that accounts for differing observation time. For total time at risk, the measure is principally a denominator/exposure quantity used to characterize how much observation time contributes to the rate calculation. The ClinicalTrials.gov record does not identify the specific inferential procedure used for either primary endpoint.
7. Safety Results
The ClinicalTrials.gov record reports serious adverse events by randomized predecessor-study treatment history and by treatment-switching group. These figures are presented exactly as affected participants over the corresponding number at risk.
| Group | Serious adverse events affected / at risk |
|---|---|
| NAVIGATOR Rand Teze | 82/528 |
| NAVIGATOR Rand Pbo | 94/531 |
| NAVIGATOR Pbo to Teze | 17/206 |
| SOURCE Rand Teze | 18/74 |
| SOURCE Rand Pbo | 19/76 |
| SOURCE Pbo to Teze | 4/32 |
The presence of predecessor-study treatment-switching groups is important for interpretation. The DESTINATION record is an extension study, and the registry-reported safety groups distinguish participants randomized to tezepelumab or placebo as well as participants who switched from placebo to tezepelumab.
8. Statistical Methods Explained
Why is a rate ratio used for AAER?
AAER is a count/rate endpoint expressed as events per year. A rate ratio provides a relative comparison of event rates while retaining the time dimension of the outcome. This is different from simply comparing the total number of exacerbations between groups.
What does a rate ratio of 0.42 mean?
A rate ratio of 0.42 means that the estimated annualized exacerbation rate in the numerator group is 42% of the rate in the comparison group. In relative terms, that is a 58% lower estimated rate. It does not mean that 42% of patients experienced exacerbations, nor does it mean that each patient's personal risk was reduced by exactly 58%.
Why does total time at risk matter?
Participants may contribute different amounts of observation time. A rate based only on event counts can therefore be misleading when exposure differs. Total time at risk provides the denominator needed to express events relative to the amount of observation contributed by the population.
What does a 95% confidence interval tell us?
The confidence interval describes uncertainty around the estimated treatment effect under the statistical framework used to generate it. For NAVIGATOR, the reported interval is 0.35–0.51; for SOURCE, it is 0.38–0.96. Neither interval describes the range of individual patient responses.
Why should the rate ratio not be called a hazard ratio?
A rate ratio compares event rates, while a hazard ratio is a time-to-event model parameter describing relative instantaneous event rates under a survival model. Although both are relative measures and can be numerically below 1, they are not interchangeable statistical quantities.
Why does the missing method matter?
The ClinicalTrials.gov record identifies the effect measure and hypothesis type but list the method and method category as not reported. Different count/rate models can make different assumptions about event distributions, exposure time, and variability. Without the named method, it would be inappropriate to attribute a particular regression model to DESTINATION.
What does the superiority designation tell us?
The posted analyses are classified as superiority analyses. This indicates that the comparison was framed around demonstrating a treatment difference rather than proving that two treatments were sufficiently similar within a prespecified non-inferiority margin. The ClinicalTrials.gov record does not report a non-inferiority margin because this is not identified as a non-inferiority analysis.
9. Confidence Intervals and Precision
The two AAER estimates illustrate why a point estimate should not be interpreted in isolation.
| Comparison | Rate ratio | 95% CI | Statistical interpretation |
|---|---|---|---|
| NAVIGATOR Rand Teze vs Rand Pbo | 0.42 | 0.35–0.51 | Estimated annualized exacerbation rate was lower in the tezepelumab group, with the registry-reported interval below 1. |
| SOURCE Rand Teze vs Rand Pbo | 0.61 | 0.38–0.96 | Estimated annualized exacerbation rate was lower in the tezepelumab group, with a wider registry-reported interval that remains below 1. |
The intervals provide information about precision that the point estimates alone cannot. The NAVIGATOR estimate is accompanied by a narrower interval relative to its point estimate, while the SOURCE estimate has greater visible uncertainty around its magnitude.
A confidence interval is not a probability statement that the true treatment effect has a particular probability of lying inside the interval. It is an interval produced by a specified statistical procedure under repeated-sampling assumptions. It is also not a substitute for reporting the absolute event rates or event counts.
10. Randomization and Triple Masking
DESTINATION is identified as randomized, with a parallel design and triple masking. Randomization is the central design feature that supports a comparative treatment analysis by assigning participants to study groups according to a predefined random mechanism rather than investigator selection.
Randomization
Creates the treatment-group comparison specified by the trial design and reduces systematic allocation differences that could otherwise confound the comparison.
Parallel design
Participants are evaluated within their assigned study group rather than all participants receiving each intervention sequentially as in a crossover design.
Triple masking
The registry identifies the study as triple-masked. The ClinicalTrials.gov record does not specify which three parties were masked, so no additional attribution is made here.
Safety analysis set
The registered primary exposure-adjusted incidence-rate endpoint explicitly states that its analysis is based on the Safety Analysis Set.
11. Analysis Populations and Treatment Switching
The registered primary safety endpoint specifies the Safety Analysis Set. The endpoint definition reported in the ClinicalTrials.gov record also establishes exposure-related stopping rules involving the last dose, death, study withdrawal, and the start of another biologic.
The predecessor-study history is especially relevant because the registry's statistical comparisons are labeled according to randomized treatment in NAVIGATOR and SOURCE, while separate groups identify participants who switched from placebo to tezepelumab.
| Group label | Role in the ClinicalTrials.gov record |
|---|---|
| NAVIGATOR Rand Teze | Randomized tezepelumab group from NAVIGATOR |
| NAVIGATOR Rand Pbo | Randomized placebo group from NAVIGATOR |
| NAVIGATOR Pbo to Teze | Participants switching from placebo to tezepelumab |
| SOURCE Rand Teze | Randomized tezepelumab group from SOURCE |
| SOURCE Rand Pbo | Randomized placebo group from SOURCE |
| SOURCE Pbo to Teze | Participants switching from placebo to tezepelumab |
12. Exposure-Adjusted Analysis: Why It Is Useful
Safety follow-up in an extension study can involve different amounts of time under observation. An exposure-adjusted rate addresses this by using time at risk as part of the denominator rather than treating every participant as though they contributed identical observation time.
This is why the registered endpoint includes both an exposure-adjusted incidence rate and total time at risk. The latter supplies information about the amount of observation contributing to the former.
The registry definition also establishes specific endpoints for stopping the risk period, including the minimum of specified treatment-exposure and follow-up events. These rules are important because the rate depends directly on which person-time is counted in the denominator.
13. What the AAER Results Do — and Do Not — Establish
What the estimates show
The statistical analyses posted on ClinicalTrials.gov report rate ratios below 1 for both NAVIGATOR and SOURCE randomized tezepelumab-versus-placebo comparisons.
What they do not show
They do not provide an individual-patient probability of exacerbation, an absolute rate difference, or a hazard ratio.
What the confidence intervals add
The intervals quantify uncertainty around the two reported rate-ratio estimates and show that the precision differs between the two comparisons.
What remains unspecified
The ClinicalTrials.gov record does not identify the statistical model used to produce the rate ratios or provide p-values for these two analyses.
14. Limitations
- Primary statistical method not identified: the ClinicalTrials.gov record lists the method and method category as not reported.
- Secondary rather than primary comparative analyses: the two numerical statistical analyses registry-reported concern AAER, while the two registered primary endpoints are exposure-adjusted AE/SAE incidence rates and total time at risk.
- No p-values reported: the statistical-analysis records provide rate ratios and 95% confidence intervals but do not provide p-values.
- Predecessor-study context: the outcome analyses are labeled according to randomized treatment in NAVIGATOR and SOURCE, while the extension includes participants with prior placebo exposure who switched to tezepelumab.
- Safety interpretation: serious adverse-event counts are provided for multiple predecessor-study treatment histories and should not be collapsed into a single randomized comparison without the underlying analysis specification.
- No model-specific assumptions should be inferred: because the analysis method is not reported in the ClinicalTrials.gov record, assumptions associated with a particular Poisson, negative-binomial, generalized linear, or other model cannot be attributed to the trial.
- Registry-level evidence: this analysis is constrained to the ClinicalTrials.gov record and the linked publication identifiers. It does not add unpublished statistical-analysis details.
15. Why This Trial Matters Statistically
DESTINATION is a useful teaching example because it illustrates a statistical problem that differs from the more familiar survival-analysis structure of many phase 3 trials. Its registered primary endpoints focus on exposure-adjusted safety rates and total time at risk, while the registry-reported comparative statistical analyses focus on annualized asthma exacerbation rate.
| Concept | How it appears in DESTINATION |
|---|---|
| Randomization | The study is identified as randomized. |
| Parallel design | The registered design model is parallel. |
| Triple masking | The registry identifies the trial as triple-masked. |
| Exposure-adjusted incidence rate | A registered primary endpoint measures exposure-adjusted incidence rates of AEs/SAEs. |
| Time at risk | Total time at risk is a second registered primary endpoint. |
| Rate ratio | The posted AAER analyses use rate ratio as the reported effect measure. |
| Confidence interval | Both posted AAER comparisons include two-sided 95% confidence intervals. |
| Superiority testing | Both posted AAER analyses are classified as superiority hypotheses. |
| Safety analysis set | The primary exposure-adjusted incidence-rate definition states that the analysis is based on the Safety Analysis Set. |
| Treatment switching | The ClinicalTrials.gov record separately identify participants switching from placebo to tezepelumab. |
16. A Practical Guide to Reading the Two Rate Ratios
The two estimates should be read as relative rates, not as percentages of patients.
| Step | NAVIGATOR | SOURCE |
|---|---|---|
| Reported rate ratio | 0.42 | 0.61 |
| Relative-rate interpretation | Estimated rate was 42% of the comparison rate. | Estimated rate was 61% of the comparison rate. |
| Relative-rate reduction from 1 | 58% | 39% |
| 95% confidence interval | 0.35–0.51 | 0.38–0.96 |
| Hypothesis type | Superiority | Superiority |
The reductions in the table are simple interpretations of the reported rate ratios, using 1 − rate ratio. They are not independently estimated effects and do not replace the reported rate ratios or their confidence intervals.
The ClinicalTrials.gov record does not report p-values for these two statistical analyses. More importantly, a p-value would answer a question about compatibility with a null hypothesis under a specified testing framework; it would not measure the magnitude of the rate difference. The rate ratio and confidence interval together provide the central information available here about relative effect and statistical precision.
17. Trial Timeline
Trial start
The DESTINATION study began on January 7, 2019, according to the ClinicalTrials.gov record.
Randomized extension study
The trial is classified as phase 3, randomized, parallel, and triple-masked, with tezepelumab and placebo as the two interventions.
Registered endpoint horizon
The registry-reported primary-endpoint time frame runs from baseline, defined as Week 0 in the predecessor study, to Week 104, with additional registry wording concerning subjects switching from placebo in the predecessor study.
Primary completion
The ClinicalTrials.gov record gives October 26, 2021 as the primary completion date.
18. Sources and Linked Publications
- ClinicalTrials.gov: NCT03706079 — DESTINATION.
- Linked PubMed record: PMID 39326921.
- Linked PubMed record: PMID 38697286.
- Linked PubMed record: PMID 36702146.
- Linked PubMed record: PMID 33087119.
19. Related Tutorials
Learn more about the methods used in this trial:
20. Related Calculators
Continue through the Clinical Biostats statistical tutorials
Explore the statistical concepts behind randomized trials, confidence intervals, rate comparisons, and clinical-trial design.
21. Record Summary
DESTINATION provides a useful statistical case study in the analysis of clinical-trial event rates over differing amounts of exposure. The registered primary endpoints are exposure-adjusted incidence rates of adverse events and serious adverse events and total time at risk. The registry-reported formal statistical analyses instead provide two secondary AAER comparisons, reporting rate ratios of 0.42 for NAVIGATOR and 0.61 for SOURCE, with corresponding two-sided 95% confidence intervals of 0.35–0.51 and 0.38–0.96.
The key statistical lesson is that event counts, exposure time, rates, and relative rates answer different questions. A rate ratio summarizes the relative frequency of events over time, while the confidence interval describes uncertainty around that estimate. The safety results additionally demonstrate why treatment history and treatment switching must be preserved when interpreting an extension study rather than collapsing heterogeneous groups into a single comparison.