Safety Study of Tezepelumab (AMG 157) in Healthy Adults and Adults With Atopic Dermatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety Study of Tezepelumab (AMG 157) in Healthy Adults and Adults With Atopic Dermatitis (NCT00757042), a randomized phase 1 study. The primary outcome is Number of Participants With Adverse Events.

2. Clinical Question

3. Trial Design

This was a Phase 1, randomized, double-blind, placebo-controlled, ascending single-dose study. Its official title describes evaluation of safety, tolerability, pharmacokinetics, and pharmacodynamics in healthy subjects and subjects with moderate to severe atopic dermatitis. The registry classifies the study model as sequential. It lists enrollment of 78 and a completed status.

The study had two parts. Part A included healthy adults and tested multiple subcutaneous and intravenous dose groups, alongside placebo groups using both routes. Part B included adults with moderate to severe atopic dermatitis and compared tezepelumab 700 mg intravenously with intravenous placebo. Participants were assigned to receive one dose; the registry description does not indicate repeated dosing.

The safety observation period varied across the listed groups. For Part A tezepelumab and placebo at 2.1 mg, 7 mg, 21 mg, 70 mg, and 210 mg subcutaneously, the adverse-event timeframe was 85 days. For Part A at 420 mg subcutaneously, 210 mg intravenously, and 700 mg intravenously, and for Part B, it was 113 days. This matters when interpreting counts: the outcome is tied to the stated observation windows, which were not identical for every group.

The registry identifies Amgen as sponsor and records a study start of 2008-09-18 and primary completion of 2011-01-05. Further details about the sequence of dose cohorts, randomization procedures, sample-size justification, and statistical analysis plan are not reported in the registry record. The posted results therefore support a descriptive reading of the safety counts, but not a reconstruction of every design or analysis decision.

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: In Part A the first 2 participants enrolled in Cohort 1 were randomized in a 1:1 ratio and subsequent participants in this cohort were randomized in a 5:1 ratio to receive tezepelumab or placebo. For Cohorts 2 to 8 in Part A, participants were randomized in a 6:2 ratio to receive tezepelumab or placebo. In Part B, the first 2 participants were randomized in a 1:1 ratio, and subsequent participants in Cohort 9 were randomized in an 8:2 ratio to receive tezepelumab or placebo.

Only the first 6 of 11 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All participants who received at least 1 dose of study drug

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: Participants who received at least 1 dose of Investigational product.

Only the first 6 of 12 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Adverse Events, measured over For Part A Tezepelumab/Placebo 2.1 mg, 7 mg, 21 mg, 70 mg, and 210 mg SC: 85 days. For Part A Tezepelumab/Placebo 420 mg SC, 210 mg IV, and 700 mg IV and Part B: 113 days.

The primary endpoint is the number of participants with adverse events. It counts participants rather than the total number of event episodes, so a participant with more than one event is still represented as a participant with an event for this measure. The endpoint includes any untoward medical occurrence in a participant administered study drug and does not require a causal relationship with treatment. Worsening of a pre-existing condition and laboratory-value changes requiring therapy or adjustment of prior therapy are included in the registry definition.

Adverse events were assessed for severity using CTCAE Version 3: Grade 1 is mild, Grade 2 moderate, Grade 3 severe, Grade 4 life-threatening, and Grade 5 death due to an adverse event. Investigators determined the relationship between an event and study treatment. Serious adverse events are separately defined by criteria such as fatality, life threat, hospitalization, persistent or significant disability, congenital anomaly, or another significant medical hazard.

Counting participants with events is useful in an early safety study because it provides a direct descriptive account of how many people in each group experienced at least one event over the specified follow-up. It does not, by itself, establish whether an event was caused by treatment, capture the number of episodes, or provide a comparative estimate adjusted for other factors. The registry record does not report a separate primary endpoint estimate with an inferential comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation begins with the form of the endpoint. The primary outcome is a count of participants who experienced an adverse event, reported separately across treatment groups. Because the outcome is whether a participant had an event, it is a binary participant-level outcome even though the registry expresses the result as a count. The posted information gives the number at risk and the number with events for each group. It does not provide an adjusted treatment effect or a formal test of group differences.

What does a participant count represent?

A participant count records whether each person met the endpoint definition during the follow-up period. It is distinct from an event count: repeated events in one person do not necessarily increase the number of participants with an event. It is also distinct from an event rate per unit of time, which would account for differing amounts of observation time. The registry reports group counts and a stated timeframe, not person-time rates. As a result, the counts should be read within their respective follow-up windows, rather than as if every group had identical observation time.

In general, a proportion is formed by relating the number of participants with an event to the number at risk. Comparisons of proportions can be informative when groups and observation conditions are sufficiently comparable, but the registry’s listed results are spread across two study parts, several dose levels, two administration routes, and different follow-up periods. No additional comparison has been reported. The interpretation should therefore remain descriptive and should not convert the collection of groups into a single contrast that the record does not specify.

Why do small groups require caution?

In general, estimates based on small groups are imprecise: a change in the outcome of one participant can noticeably alter the group summary. This is particularly important for uncommon outcomes such as serious adverse events. A zero count in a small group is compatible with uncertainty about the event experience beyond that group and observation period. It does not demonstrate that the underlying risk is zero. Confidence intervals, when reported, help show the range of values compatible with the data under a statistical model. Such intervals are not reported in the registry record for these results.

Inference also requires a clearly defined comparison and a method suited to the endpoint and design. The registry record identifies randomization and double masking, but does not report the test, model, adjustment strategy, or analysis plan used for these safety counts. It lists zero analyses. No p-value or model-based treatment effect should be inferred from the counts alone. In a randomized study, random allocation can support a causal comparison when the comparison, analysis population, and assumptions are prespecified and appropriately analyzed. The registry information available for these results does not supply those elements in sufficient detail to make such a comparison.

What do the planning illustrations mean?

In general, a sample-size calculation estimates how much information a future study might need to detect an assumed effect at a chosen error rate and power target. The registry’s planning assumptions describe a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a stated fraction of the observed effect. These are planning assumptions, not evidence about the trial’s original sample-size rationale. The record explicitly describes the illustration as based on the observed effect rather than the trial’s own design.

Such planning values are conditional on the assumed effect and method. If the effect used as a reference is uncertain, the resulting planning estimate inherits that uncertainty. A smaller target effect generally requires more information to detect than a larger one, all else being equal; this is a general statistical principle, not a new result from this trial. These illustrations are not observed power. Observed power is not a substitute for describing the precision of the actual results, and the registry does not report a trial-specific power analysis.

How should multiple groups and outcomes be considered?

The registry lists 16 outcome measures, including pharmacokinetic measures for both study parts, as well as the primary safety outcome. Looking across many doses, routes, and outcomes creates multiple opportunities to notice patterns. In general, multiplicity refers to the increased chance of false-positive findings when many statistical questions are tested without an appropriate plan. The record does not report a multiplicity strategy or a family of confirmatory hypotheses. Because no inferential analyses are listed, the safety counts should not be treated as statistically tested findings, and apparent differences across groups should not be assigned significance.

General methods for binary outcomes include comparisons of proportions, regression models, and exact methods for sparse data. The appropriate choice depends on the planned contrast, group sizes, design, and analysis assumptions. No method can be attributed to this trial from the registry record. The safest statistical reading is to retain the stated group-level counts and denominators, distinguish serious from other adverse events, respect the follow-up periods, and avoid treating descriptive variation as proof of a treatment effect.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcomes and multiple treatment groups. The secondary outcomes include pharmacokinetic measures such as maximum observed concentration, time of maximum concentration, and area under the concentration-time curve in Part A and Part B. The primary safety endpoint is also represented across several dose levels, routes, and study parts. In general, testing many outcomes or comparisons can increase the chance of finding an apparently notable result by chance. A prespecified multiplicity approach can define which comparisons are confirmatory and how error rates are controlled. Such an approach is not reported in the registry record.

The registry lists no analyses. It does not report a statistical analysis plan, hypothesis-testing hierarchy, confidence intervals, or adjusted comparisons for the adverse-event counts. It is therefore not possible to tell from the record whether there were planned interim analyses, how any unreported analyses were handled, or whether a particular multiplicity adjustment was used. These details are not reported in the registry record and should not be inferred from the trial’s randomized or double-blind description.

The primary outcome population is described as all participants who received at least 1 dose of study drug. This defines a treated population for the safety endpoint, rather than a full randomized population. The registry does not report the numbers randomized, the numbers excluded from this endpoint population, missing outcome status, reasons for missing data, or an imputation method. It also does not describe how participants with incomplete follow-up were handled. Missing-data procedures and a detailed analysis-population description are not reported in the registry record.

For safety interpretation, the posted counts should be read as the recorded outcomes in the listed groups, with the registry’s stated at-risk counts and follow-up periods. The record does not establish whether all events were captured for every participant throughout those periods. No assumptions about completeness or missingness can be verified from the available information.

11. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: infection (33.3%), procedural pain (33.3%).

The serious adverse-event figures are participant counts, not counts of every serious event episode. The registry lists no participants with serious adverse events in the tezepelumab groups or in the Part A placebo groups. In Part B, 1 participant in the intravenous placebo group had a serious adverse event, with 3 participants at risk. The event count should be interpreted in the context of the small group and the stated 113-day follow-up for Part B; it is not a comparative estimate of risk.

The registry identifies INFECTION and PROCEDURAL PAIN as top serious terms, each reported at 33.3%. The listed information does not show which group or groups those terms correspond to, whether the terms refer to participants or events, or how they relate to the Part B serious event. That context is not reported in the registry record. The definitions indicate that serious adverse events meet at least one specified seriousness criterion, while severity grades describe intensity; these are related but distinct concepts.

Deaths are not reported in the registry record. A missing death count should not be read as a reported zero. Likewise, zero serious-event counts in some groups do not establish absence of risk outside the observed participants and periods.

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

The posted results provide a descriptive account of adverse events recorded across a single-dose study with healthy adults and adults with moderate to severe atopic dermatitis. They show that non-serious adverse events were recorded in every listed group, with the count varying across dose, route, and study part. Serious-event counts were zero in the listed tezepelumab groups and Part A placebo groups, while Part B intravenous placebo listed 1 participant with a serious event. The registry’s reported counts and denominators allow those observations to be read group by group.

These data do not by themselves establish that an event was caused by tezepelumab. The registry definition includes events without requiring a causal relationship, and treatment relationship was determined by investigators. The counts also do not show the number of events per participant, the timing of each event, or a formally adjusted comparison. Different follow-up windows and the division into study parts further limit direct comparison across every group.

The results do not establish equivalence between tezepelumab and placebo, nor do zero counts prove that serious events cannot occur. Small group sizes make each participant’s outcome influential, and no confidence intervals or inferential analyses are reported. A lack of a reported comparison is not evidence that the groups were the same.

The study’s broader aims included tolerability, immunogenicity, pharmacokinetics, and pharmacodynamics. The registry lists pharmacokinetic outcomes, but the result details do not provide a basis for conclusions about those measures or about clinical benefit. The record does not report a primary efficacy question. The appropriate conclusion is limited to the descriptive safety observations recorded for the specified groups and follow-up periods, with substantial uncertainty about how well those observations generalize beyond this study.

13. Why This Trial Matters Statistically

This trial illustrates a common statistical challenge in early clinical research: safety outcomes are examined across multiple doses, administration routes, and populations, while each group may contain relatively few participants. Descriptive counts are essential in this setting because they preserve what was observed without implying a precision or a causal contrast that the registry does not provide. The group-by-group structure matters; combining results across doses or study parts could obscure differences in population, route, or follow-up.

It also shows why adverse-event counts need careful labels. A count of participants with an event is not the same as a count of event episodes, a severity score, a rate over time, or a treatment-attributable event count. Seriousness is determined by specified criteria, while severity grades describe intensity. Keeping those concepts separate prevents a reader from treating every adverse event as serious or assuming that every recorded event was caused by study treatment.

Statistically, a zero count is not a guarantee of zero underlying risk, especially when group sizes are small. The absence of confidence intervals and inferential analyses means the posted results should remain descriptive. This is not a defect that can be repaired by deriving an unplanned comparison from raw counts: a comparison requires a defined estimand, a suitable method, and knowledge of the analysis population and missing-data handling.

Finally, planning illustrations based on an observed effect can help explain how assumptions influence future sample-size planning, but they are not a retrospective verdict on the design or an account of observed power. The registry’s description explicitly distinguishes the illustration from the trial’s own design. This distinction is important for readers using early safety data to understand uncertainty without overstating what the study establishes.

14. 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 registry record presents the safety outcome across many separate groups, but it does not report a formal comparison between tezepelumab and placebo. The listed groups differ by study part, dose, administration route, and, for some groups, follow-up period. Raw counts across those groups cannot be read as a controlled dose-response analysis. A formal dose-response result is not reported in the registry record.

The group sizes are small, and no confidence intervals or other precision measures are reported for the event counts. This limits how precisely the underlying event experience can be assessed. The absence of a serious event in a listed group is only an observation in that group during its specified period, not proof that the event cannot occur.

The registry does not report detailed randomization methods, allocation concealment, the operational procedures for double masking, a statistical analysis plan, multiplicity handling, interim analyses, or a sample-size justification. It also does not report how missing safety observations were handled, whether follow-up was complete, or the number randomized and excluded from the safety population. These omissions prevent a fuller assessment of potential bias and analysis choices.

The registry’s serious-event terms are not linked in the available information to particular groups or detailed event narratives. Death counts are not reported in the registry record. The safety endpoint includes events whether or not they were caused by treatment, and investigator attribution does not replace a controlled comparison. These limitations constrain interpretation to the recorded counts and their stated group and follow-up context.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Safety Study of Tezepelumab (AMG 157) in Healthy Adults and Adults With Atopic Dermatitis — NCT00757042.
• Registry record last updated 2022-09-21; status last verified 2022-09.
• Results first posted 2022-07-18.
• 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.

18. Record Summary

The registry record identifies NCT00757042 as a completed Phase 1 randomized, double-blind, placebo-controlled, ascending single-dose study of tezepelumab in healthy adults and adults with moderate to severe atopic dermatitis. It lists enrollment of 78, Amgen as sponsor, and two study parts. Part A included multiple subcutaneous and intravenous doses and placebo groups; Part B compared intravenous tezepelumab 700 mg with intravenous placebo.

The primary outcome was the number of participants with adverse events among those who received at least 1 dose. Follow-up was 85 days for the specified lower Part A subcutaneous dose groups and 113 days for the other listed groups and Part B. The registry also lists pharmacokinetic secondary outcomes. It reports group-level adverse-event counts but no analyses, confidence intervals, or formal treatment comparisons. Details of the statistical plan, missing-data handling, and multiplicity procedures are not reported in the registry record.