Study of Efficacy and Safety of Erenumab in Adult Chronic Migraine Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Efficacy and Safety of Erenumab in Adult Chronic Migraine Patients (NCT03867201), a randomized phase 3 study. The primary outcome is Change From Baseline in Monthly Migraine Days During the Last 4 Weeks of the 12-week Treatment Period.

2. Clinical Question

The clinical question is whether monthly erenumab 70 mg is associated with a different change in monthly migraine days than placebo among adults with chronic migraine. The registry describes a Phase 3 study in an Asian population. The primary outcome compares change from baseline, defined as the 4 weeks before study drug, to the last 4 weeks of the 12-week treatment period.

The registry’s definition of a migraine day requires a qualified migraine headache on a calendar day. A qualifying headache may involve onset, continuation, or recurrence, must last at least 4 continuous hours, and must meet specified pain-feature or associated-symptom criteria. This makes the endpoint a defined measure of headache-day frequency rather than a general count of any headache or a measure of severity alone.

The study also recorded disability and productivity, acute headache medication days, a threshold response based on reduction in migraine days, and safety outcomes. These outcomes add context, but the registry identifies change in monthly migraine days as the primary endpoint.

3. Trial Design

This was a Phase 3, randomized, triple-masked, parallel-group study comparing erenumab 70 mg with placebo. The official title describes a 12-week, once-monthly subcutaneous treatment study. Both study products were administered by pre-filled syringe. The registry lists 557 participants enrolled and identifies the study as completed.

The record describes the target population as adult chronic migraine patients and its brief summary specifies an Asian population. The primary outcome population is the Full Analysis Set, defined in the registry as all randomized participants with evaluable measurements at the respective timepoint. This definition means that the number contributing to an outcome can depend on whether an evaluable measurement was available at that timepoint. The registry does not report a fuller account of how missing measurements were handled.

For the primary outcome, the registry lists 270 participants in the erenumab group and 274 in the placebo group. The outcome is a least-squares mean change, with standard error as the reported measure of dispersion. These outcome-specific analysis counts should not be treated as interchangeable with the overall enrollment figure or with counts for other endpoints, whose analysis populations differ.

The study start date is 2019-08-26, and the primary completion date is 2021-08-11. The sponsor is Novartis Pharmaceuticals. The registry identifies one reported primary analysis, using a mixed models analysis and reporting a mean difference (net) with a 95% confidence interval and p-value. Details beyond that description, including the model specification and its assumptions, are not reported in the registry record.

Randomization and masking are design features intended to limit systematic differences and influence in treatment assignment, assessment, or conduct. Their presence does not by itself establish that all potential biases were eliminated. The registry record does not report the allocation ratio or the original design assumptions for sample-size planning.

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: A screening period of 2 weeks was used to assess initial eligibility, followed by a 4-week baseline period to assess diary compliance and headache frequency.

Double-Blind Period Until Week 12

Reasons for not completing this period:

Open-Label Period From Week 12

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set. All randomized participants with evaluable measurements at the respective timepoint.

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.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Monthly Migraine Days During the Last 4 Weeks of the 12-week Treatment Period, measured over baseline (4 weeks period prior to start of study drug), week 9 to 12.

The primary endpoint is change from baseline in monthly migraine days during the last 4 weeks of the 12-week treatment period. Baseline is defined as the 4 weeks before study drug began. A migraine day is a calendar day on which a participant experiences a qualified migraine headache, including onset, continuation, or recurrence. The headache must last at least 4 continuous hours and satisfy the registry’s specified pain-feature or associated-symptom criteria.

This endpoint measures frequency over a defined period. A change score expresses how the measured monthly migraine-day count at the treatment-period assessment differs from the baseline measurement. A negative change indicates a reduction relative to baseline; the group estimates are therefore interpreted as reductions in migraine days. The between-group estimate compares those changes rather than comparing the groups’ raw endpoint values alone.

Frequency is useful as an outcome because it directly represents how often qualifying migraine days occur during the assessed period. It is not, by itself, a complete account of migraine burden: it does not capture every aspect of severity, disability, medication use, or participant experience. The registry includes separate secondary outcomes addressing some of these related dimensions.

The primary outcome is reported as a least-squares mean with standard error. Least-squares means are model-based estimates; they are not necessarily identical to simple arithmetic averages. The registry does not provide the full model specification or enough detail to reconstruct the adjustment structure. The primary comparison is the key statistical result for the study, while other outcomes are supportive and should be interpreted with their own uncertainty and the context of multiple testing.

7. Results Posted in the Registry

• Erenumab 70 mg: mean -8.19 versus -6.62 on control; difference -1.57 (95% CI -2.83 to -0.31), standardized effect d = -0.21; z-test p = 0.015.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in Migraine-related Disability and Productivity as Measured by the mMIDAS During the Last 4 Weeks of the 12-week Treatment Period

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Erenumab 70 mg: mean -14.67 versus -12.93 on control; difference -1.74 (95% CI -5.05 to 1.57), standardized effect d = -0.09; z-test p = 0.303.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Erenumab 70 mg and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With at Least 50% Reduction From Baseline in Monthly Migraine Days During the Last 4 Weeks of the 12-week Treatment Period

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Erenumab 70 mg: 47.0% (131/279) versus 36.7% (102/278) on control; risk difference 10.3 percentage points (95% CI 2.1 to 18.4), risk ratio 1.28 (1.05–1.56); unadjusted two-proportion z-test p = 0.014.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Erenumab 70 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Monthly Acute Headache Medication Days During the Last 4 Weeks of the 12-week Treatment Period

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Erenumab 70 mg: mean -5.34 versus -4.66 on control; difference -0.68 (95% CI -1.76 to 0.40), standardized effect d = -0.11; z-test p = 0.218.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Erenumab 70 mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes extend the primary frequency measure into related areas: migraine-related disability and productivity, a threshold reduction in monthly migraine days, acute headache medication days, and safety. Taken together, they provide a broader description of the study’s outcome pattern than the primary endpoint alone. They are related to the primary question but do not measure the same construct, and their results need not move in lockstep.

The posted estimates show a mixed pattern of uncertainty across the secondary outcomes. The threshold-response comparison has an interval that excludes no difference, while the reported intervals for the continuous disability and acute-medication comparisons include no difference. These findings offer context, but the record lists no sponsor-reported analysis for these secondary efficacy outcomes. The calculations should therefore not be presented as equivalent to the sponsor’s reported primary mixed models result.

Secondary outcomes generally carry less confirmatory weight than a designated primary endpoint, particularly when several outcomes are examined. Each additional test creates more opportunity for a result to appear notable by chance. The registry record does not report a multiplicity adjustment or a hierarchical testing procedure for these outcomes. Consequently, unadjusted p-values should be treated as exploratory evidence rather than as a set of independent confirmations.

The secondary results can be considered for coherence and for questions that the primary endpoint cannot answer, but they do not establish whether the primary result is replicated across different measures. Disability, medication days, and a threshold responder definition capture different aspects of outcomes. Their meaning also depends on their measurement definitions, available observations, and uncertainty. A cautious reading keeps these results in their supporting role and avoids selecting one favorable secondary result in isolation.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change From Baseline in Migraine-related Disability and Productivity as Measured by the mMIDAS During the Last 4 Weeks of the 12-week Treatment Period
• Number of Participants With at Least 50% Reduction From Baseline in Monthly Migraine Days During the Last 4 Weeks of the 12-week Treatment Period
• Change From Baseline in Monthly Acute Headache Medication Days During the Last 4 Weeks of the 12-week Treatment Period
• Number of Subjects With Adverse Events as a Measure of Safety
• Number of Subjects With Anti-AMG 334 Antibodies

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous: it is a change in monthly migraine days, measured in days. The registry reports least-squares mean changes and standard errors for each group, followed by a between-group mean difference. In general, a continuous outcome retains information about the magnitude of a measurement, rather than reducing every participant to a yes-or-no category. The interpretation depends on how the outcome is defined, the analysis population, and the statistical model.

What does a least-squares mean represent?

In general, a least-squares mean is a model-based estimate of a group mean. It may account for variables or repeated measurements included in the fitted model, whereas a simple mean is calculated directly from observed values. The registry labels the primary group estimates as least-squares means, but the full model specification, included covariates, and handling of repeated observations are not reported in the registry record. For that reason, the estimates should be read as the posted model-based results rather than as independently reconstructed adjusted means.

How should the standard error be read?

In general, a standard error describes the uncertainty in an estimated quantity, such as a mean. It is not the same as the spread of individual participant outcomes. A smaller standard error generally indicates a more precisely estimated mean under the model and data conditions. The registry reports standard errors for the group-level least-squares means; it does not report standard deviations for the primary endpoint. A standard error alone does not describe the full distribution of participant responses.

What does the between-group estimate compare?

The mean difference (net) compares the change estimate in the erenumab group with the change estimate in the placebo group, in the direction reported by the registry. Because both group changes are negative, the estimated contrast describes how much the reductions differ. In general, a between-group contrast is more informative for a randomized comparison than judging one group’s change in isolation. The estimate does not mean that every participant experienced a change of that size.

What is the role of the confidence interval and p-value?

In general, a confidence interval conveys uncertainty around an estimate and identifies values reasonably compatible with the data and analysis assumptions at the stated confidence level. A p-value summarizes how unusual data at least as incompatible with a null hypothesis would be if that null hypothesis and the model assumptions held. Neither statistic establishes the probability that a treatment works, the probability that a result will repeat, or the importance of an effect to participants. The registry reports the sponsor’s 95% confidence interval and p-value for the primary comparison.

How does the mixed models analysis fit in?

The registry names the sponsor’s primary method as Mixed Models Analysis and reports a mean difference (net). In general, mixed models can represent relationships among repeated measurements and account for within-participant correlation, depending on the model specification. Their validity depends on the chosen structure and assumptions, including assumptions about the outcome and missing data. The registry record does not report enough detail to assess those choices. The posted analysis should therefore be described as the sponsor’s stated model, without inferring its exact implementation.

What does the recalculation add?

The recalculated estimate is a simple, unadjusted calculation from posted summary statistics. It is useful as a check on the direction and approximate scale of the reported contrast, but it does not reproduce the mixed models analysis. For continuous outcomes, a pooled standard deviation may be estimated from summary information when direct individual-level data are unavailable. Such a reconstruction is approximate and does not recover model adjustments, correlations, or other features not represented by the summaries. The primary recalculated contrast agrees with the sponsor’s reported estimate and p-value, and the sponsor’s analysis remains the primary reference.

For binary outcomes, general analysis methods compare the proportions meeting a defined response criterion. A risk difference expresses the difference in proportions; relative measures express a ratio. These measures answer different questions and can look different even when based on the same underlying observations. The threshold-response endpoint reports a risk difference and confidence interval, while the registry’s sponsor-reported analysis list contains no analysis for that secondary outcome.

Over-dispersion is a general concern when count outcomes vary more than a model’s assumed distribution allows. It can make standard errors too small if the extra variability is ignored. The primary endpoint is reported as a continuous least-squares mean change, and the registry does not identify an over-dispersion analysis or provide information sufficient to assess this issue for the underlying migraine-day measurements.

All statistical interpretations are conditional on the endpoint definition, analysis population, model, and handling of incomplete data. The registry’s stated primary population includes randomized participants with evaluable measurements at the respective timepoint. Further missing-data methods, model diagnostics, and sensitivity analyses are not reported in the registry record.

11. Confidence Intervals and Precision

The sponsor’s primary estimate is a mean difference of -1.57 monthly migraine days, with a 95% confidence interval from -2.83 to -0.30. The interval describes the uncertainty around the estimated between-group contrast under the sponsor’s analysis. Its endpoints are both negative, so the range reported by the sponsor does not include zero difference. This is consistent with the reported p-value of 0.015, while neither statistic indicates the probability that the result is true.

The interval is important because the point estimate alone does not show how much uncertainty surrounds the comparison. Values toward either end of the interval represent different possible magnitudes of the between-group reduction compatible with the analysis. The width also reflects that the estimate is not exact. A confidence interval does not guarantee that the true value lies within it, and its interpretation relies on the statistical model and assumptions.

The recalculated interval is based on unadjusted summary-statistic calculations, while the sponsor reports a mixed models analysis. The recalculated primary estimate and p-value agree with the sponsor’s reported values, but the sponsor’s analysis remains the principal result. The sponsor’s confidence interval is therefore the appropriate interval to emphasize when describing the registry’s primary analysis.

Precision is not the same as clinical importance. The registry does not report a threshold defining a clinically important difference for this endpoint, and statistical uncertainty cannot by itself determine whether a difference matters to participants. Nor does a confidence interval describe how much individual participants vary in their responses; it concerns uncertainty in the estimated comparison.

Secondary outcomes have their own intervals and uncertainty. Their interpretation should account for their status as secondary outcomes and the lack of reported multiplicity adjustment. The interval around any estimate is more informative than a binary reading of whether a p-value falls on one side of a conventional cutoff.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are illustrations based on the observed effect, not the trial’s original design calculation. For these outcomes, the reported figures are participants per arm, not total events. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect equal to the stated fraction of the observed effect. The original trial’s sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that the planning calculation indicates would be needed for an 80% chance of detecting the specified effect under those assumptions, if the assumptions and effect were correct. The 90% power column describes the corresponding planning target for a 90% chance. These are prospective design illustrations; they do not describe the probability that the completed trial found a true effect.

The rows for planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect illustrate how the assumed effect size changes the required sample size. Smaller effects need more participants to distinguish them from random variation at the same stated power and significance level. For example, the posted primary-outcome planning figures are 359 participants per arm for 80% power when planning for 100% of the observed effect, and 1,435 participants per arm when planning for 50% of the observed effect.

These figures are not observed power. Observed power is not a useful substitute for interpreting the estimate and its confidence interval after a study has been completed. The planning values also should not be mistaken for a claim that the original study was designed around the observed effect: the registry record does not report its original sample-size rationale.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -1.570; pooled standard deviation σ = 7.503.
2. n = 2 × (1.96 + 0.842)² × 7.503² / -1.570² = 358.6, rounded up to 359 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -1.57                      # observed difference in means
sigma <- 7.50349402006992                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3867201)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 359 (80% power), 481 (90% power)
 75% of observed effect: n per arm = 638 (80% power), 854 (90% power)
 50% of observed effect: n per arm = 1,435 (80% power), 1,921 (90% power)
power.t.test: 359.5 
Simulated power at n = 359 per arm: 0.797 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry identifies the sponsor’s primary method as Mixed Models Analysis and reports a mean difference (net) of -1.57, a 95% confidence interval from -2.83 to -0.30, and p = 0.015. The registry also permits an unadjusted recalculation from the posted summary statistics. That recalculation gives a mean difference of -1.57 and p = 0.015, consistent with the sponsor’s reported point estimate and p-value.

The recalculation is a check based on summary statistics, not a reproduction of the sponsor’s model. It does not incorporate model features or adjustments that are not represented in the posted summaries, and its reconstructed uncertainty is approximate. Agreement in the estimate and p-value supports consistency of the basic contrast, but it does not establish that both procedures are analytically identical.

Because the record reports the sponsor’s mixed models analysis as the primary analysis, that analysis takes precedence. The sponsor’s confidence interval should be used when describing the reported primary result. The registry does not report the full model specification, so more detailed comparisons of adjustment choices or assumptions are not possible from this record.

No sponsor-reported analyses are listed for the posted secondary endpoints. Recalculated secondary comparisons therefore should be understood as unadjusted descriptive analyses, not as sponsor-confirmed analyses.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when an analysis evaluates more than one outcome or comparison. Even if each test is conducted using a conventional significance threshold, examining multiple tests increases the chance that at least one result appears statistically notable by chance. A prespecified primary endpoint can help focus interpretation, while secondary outcomes can provide supporting or exploratory information.

The registry identifies change in monthly migraine days as the primary outcome and lists several secondary outcomes. It reports one primary analysis. A multiplicity adjustment, hierarchical testing sequence, or other procedure for controlling the overall error rate is not reported in the registry record. The unadjusted secondary p-values should therefore be read cautiously and should not be treated as if they were multiplicity-controlled confirmatory findings.

The primary analysis population is the Full Analysis Set, defined as all randomized participants with evaluable measurements at the respective timepoint. The registry does not report a detailed missing-data method, the amount or pattern of missing outcome data, or sensitivity analyses assessing the effect of missingness. The endpoint-specific analysis counts differ, so results should be understood in relation to the available observations for each outcome rather than assumed to use one identical set of participants.

In general, missing-data methods can affect estimates if the probability of having an observed measurement is related to outcomes or other participant characteristics. Different assumptions about unobserved measurements can lead to different results. Without a reported method and sensitivity analyses, the robustness of the posted estimates to missing data cannot be evaluated from the registry record.

The registry names a mixed models analysis for the primary outcome, but does not report its full model specification, covariates, covariance structure, diagnostics, or missing-data assumptions. These details matter for assessing how the analysis uses repeated measurements and for reproducing the sponsor’s result. Their absence limits independent evaluation; it does not itself show that the analysis was inappropriate.

16. Safety Results

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

Deaths

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: intervertebral disc protrusion (1.3%), atrial septal defect (0.9%), abortion (0.9%), breast cancer (0.9%), headache (0.9%).

The registry reports serious adverse events and other adverse events for the double-blind treatment period groups. In the erenumab 70mg DBTP group, 7 participants were reported with serious adverse events among 279 at risk; the placebo DBTP group also had 7 among 278 at risk. The registry reports 58 other adverse events among 279 at risk in the erenumab group and 59 among 278 at risk in the placebo group. Deaths are reported as 0 in each group.

These are counts for the listed groups, not estimates of causality. The registry record does not state in these figures whether events were judged related to study treatment, nor does the aggregate count show event duration, severity beyond the serious classification, or the number of participants with particular combinations of events. Serious and other adverse-event categories should not be conflated.

The record also lists open-label treatment-period groups, with serious and other adverse-event counts and no deaths reported. Those groups represent different treatment-period histories, so their counts should not be combined with the double-blind period groups as though they were one randomized comparison. The listed serious adverse-event terms and percentages describe reported events, but do not by themselves establish treatment attribution or comparative risk.

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

The primary result is evidence of a between-group difference in change in monthly migraine days in this randomized study. The sponsor reports a mean difference of -1.57, with a 95% confidence interval from -2.83 to -0.30 and p = 0.015. The direction indicates a greater reduction in the erenumab group than in the placebo group for the defined endpoint and treatment period.

This result does not mean that every participant had fewer migraine days, or that every participant’s response matched the group estimate. Group averages conceal individual variation. The result also does not show that erenumab eliminates migraine or addresses every aspect of chronic migraine. The primary endpoint measures qualifying migraine-day frequency, while disability, productivity, medication use, and safety are separate dimensions.

A statistically notable result is not automatically a clinically important result. The registry does not report a threshold for a clinically important difference in monthly migraine days, so the practical importance of the estimated contrast cannot be determined from the record alone. The confidence interval describes statistical uncertainty under the sponsor’s analysis, not the range of responses expected for individuals.

The findings apply to the study population and its eligibility, treatment, follow-up, outcome definitions, and analysis procedures. The record describes an Asian population, but does not establish how directly the findings apply to populations not represented in this study. Nor does the 12-week treatment period establish longer-term outcomes. The record includes open-label treatment-period safety groups, but those observations do not provide the same randomized comparison as the double-blind period.

The primary result should be distinguished from the secondary findings. Secondary outcomes can add context, but their unadjusted analyses and the absence of a reported multiplicity procedure limit their confirmatory interpretation. The registry record is a record of the reported study results, not a substitute for the full protocol, complete statistical analysis plan, or detailed clinical assessment.

18. Why This Trial Matters Statistically

This trial illustrates why a randomized comparison is interpreted through the between-group contrast rather than through change in the intervention group alone. The placebo group also showed a reduction from baseline in monthly migraine days. Comparing the changes accounts for that observed control-group pattern and focuses attention on the estimated difference between groups.

It also shows the distinction between an estimate and its uncertainty. The primary difference is reported alongside a confidence interval and p-value. Those statistics answer related but different questions: the estimate gives the direction and scale of the contrast, the interval conveys uncertainty, and the p-value summarizes compatibility with a specified null hypothesis under the analysis assumptions. None alone establishes clinical importance.

The record provides a useful example of an outcome defined in operational terms. A migraine day requires a qualifying headache with a specified duration and symptom criteria. Careful definitions matter because the meaning of a measured endpoint depends on what counts as an event and the period over which it is assessed.

Finally, the trial highlights the limits of secondary analyses and post hoc planning illustrations. Secondary comparisons can broaden interpretation but require attention to multiplicity, analysis populations, and missing data. Sample-size illustrations based on observed effects are not the study’s original design assumptions and do not represent observed power. Keeping these statistical roles separate helps prevent a precise-looking result from being given more certainty than the registry record supports.

19. 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 does not report the original sample-size justification, allocation ratio, or detailed protocol assumptions. The planning illustrations are based on observed effects and should not be used to infer what effect the study was designed to detect. The full statistical analysis plan is not reported in the registry record.

The sponsor’s primary analysis is identified as a mixed models analysis, but its model specification is not reported. The record does not provide details of covariate adjustment, covariance structure, model diagnostics, or the assumptions used for missing observations. The primary population includes randomized participants with evaluable measurements at the respective timepoint, but a detailed account of missingness and sensitivity analyses is not reported.

The registry does not report multiplicity control for the secondary outcomes. Their unadjusted comparisons are consequently less persuasive as confirmatory evidence than a prespecified primary analysis. The record also lists no sponsor-reported analyses for the posted secondary efficacy outcomes, which limits comparison between those recalculations and an official secondary analysis.

The primary endpoint is measured over the last 4 weeks of a 12-week treatment period. It does not establish longer-term efficacy or safety. The brief summary identifies an Asian population, and the record does not establish the generalizability of the findings to populations not represented. Safety counts are reported by treatment-period group, but the aggregate figures do not resolve causality or provide detailed event-level context.

These are limitations of what can be assessed from the registry record. They do not demonstrate that the trial methods were deficient; they identify where the available reporting does not permit a fuller statistical evaluation.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Efficacy and Safety of Erenumab in Adult Chronic Migraine Patients — NCT03867201.
• Registry record last updated 2025-10-16; status last verified 2025-10.
• Results first posted 2025-05-11.
• 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.

23. Record Summary

NCT03867201 is a completed Phase 3 randomized, triple-masked, parallel-group study of once-monthly subcutaneous erenumab 70 mg versus placebo in adults with chronic migraine. The official title describes a 12-week treatment study, and the brief summary identifies an Asian population. The record lists 557 enrolled participants.

The primary endpoint is change from baseline in monthly migraine days during the last 4 weeks of treatment. The sponsor’s mixed models analysis reports a mean difference (net) of -1.57, a 95% confidence interval from -2.83 to -0.30, and p = 0.015. The posted secondary outcomes address disability and productivity, migraine-day response, acute headache medication days, and safety.

Interpretation is limited by unreported details of the original sample-size rationale, full model specification, missing-data handling, and multiplicity procedures. The primary result is the principal reported comparison; secondary results and planning illustrations require more cautious interpretation.