A Phase 3 Study of VX-121 Combination Therapy in Participants With Cystic Fibrosis (CF) Heterozygous for F508del and a Minimal Function Mutation (F/MF): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 3 Study of VX-121 Combination Therapy in Participants With Cystic Fibrosis (CF) Heterozygous for F508del and a Minimal Function Mutation (F/MF) (NCT05033080), a randomized phase 3 study. The primary outcome is Absolute Change in Percent Predicted Forced Expiratory Volume in 1 Second (ppFEV1).

2. Clinical Question

3. Trial Design

This was a randomized, quadruple-masked, parallel-group phase 3 study of VX-121 combination therapy in participants with cystic fibrosis who were F/MF. The registry record lists enrollment of 435 and a completed study. The start date was 2021-09-14, and the primary completion date was 2023-05-12. The record names Vertex Pharmaceuticals Incorporated as sponsor.

The active comparison was VX-121/TEZ/D-IVA versus ELX/TEZ/IVA. Both arms had an ELX/TEZ/IVA run-in period of 4 weeks, followed by a treatment period of 52 weeks. This means the primary endpoint timeframe, baseline through Week 24, covers part of the treatment period rather than the full stated treatment duration. The posted endpoint is a between-arm comparison in a population that had already completed the run-in.

The primary outcome analysis included participants meeting the registry’s Full Analysis Set definition: randomized participants with the intended CFTR mutation or mutations who received at least 1 dose of study drug during the treatment period. For this endpoint, the analyzed number is described as those evaluable for the outcome. The registry record does not provide an account of how participants were assigned to the treatment arms beyond stating that allocation was randomized, nor does it report the detailed masking procedures.

The posted results include a continuous primary outcome and secondary sweat chloride outcomes. Some secondary results pool data from study VX20-121-103, so those results do not represent only the enrolled groups described for the primary comparison. The record also lists IVA and matched placebos among the interventions, but the posted primary arm descriptions concern the active VX-121/TEZ/D-IVA and ELX/TEZ/IVA comparison.

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 total of 435 participants were enrolled in this study, of which 37 were included in the run-in period but were not dosed in treatment period. Therefore, results are presented for only 398 participants dosed in the treatment period.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Full Analysis Set (FAS) included all randomized participants who carried the intended CFTR mutation(s) and received at least 1 dose of study drug during Treatment Period. Here, "Overall Number of Participants Analyzed" signifies participants who were evaluable for this outcome measure.

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: All randomized participants who carry the intended CFTR genotype and received at least 1 dose of study drug in a study.

6. Primary Endpoint

The registry lists the primary outcome as Absolute Change in Percent Predicted Forced Expiratory Volume in 1 Second (ppFEV1), measured over From Baseline Through Week 24.

The primary endpoint was absolute change in percent predicted forced expiratory volume in 1 second (ppFEV1), from baseline through Week 24. FEV1 is the volume of air that can forcibly be blown out in one second after full inspiration. The reported unit is percentage points, which expresses a change on the ppFEV1 scale rather than a percentage change in the participant’s own baseline value.

FEV1 is used as a measure of lung function. In this study, the endpoint allowed the trial to compare average change between the two randomized treatment groups over a specified period. A continuous outcome retains information about the measured values rather than reducing each participant to a binary category. The sponsor’s primary analysis reports least-squares means for each arm and a least-squares mean difference between arms.

Least-squares means are model-based estimates, not necessarily simple arithmetic averages. The registry identifies mixed models for repeated measures as the comparison method, but does not report the model’s full specification, including the covariates or assumptions used. The primary endpoint’s confidence interval describes uncertainty around the estimated arm difference under that analysis. The noninferiority label makes the prespecified margin central to interpretation; that margin is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

General explanation: What does a least-squares mean represent?

A least-squares mean is an estimate from a statistical model. It can differ from a raw average because the model may account for repeated observations and other prespecified factors. The registry reports least-squares means for the primary continuous outcome and a mixed model for repeated measures for the arm comparison. The covariates, covariance structure, and other detailed model choices are not reported in the registry record. Consequently, the posted estimates can be described, but the full model cannot be reproduced from the record alone.

General explanation: How should the confidence interval be read?

A confidence interval communicates the precision of an estimate under the stated analysis. A narrower interval indicates greater precision than a wider interval, while an interval spanning zero for a difference is compatible with no difference as well as with effects in either direction. It is not a guarantee about the true value and does not replace the study’s decision rule. For the primary comparison, the posted 95% confidence interval for the least-squares mean difference extends from -0.7 to 1.1 percentage points. Its interpretation is conditional on the model and analysis population used.

General explanation: Why does noninferiority require a margin?

In a noninferiority design, the question is whether the outcome for the intervention is not worse than the comparator by more than a prespecified amount considered acceptable for the study. That amount is the noninferiority margin. The confidence interval is assessed in relation to the margin and the direction of the comparison. A test result cannot be fully evaluated without knowing the margin and the direction in which the contrast was defined. The registry record labels the primary analysis as noninferiority but does not report the margin, its rationale, or the detailed decision rule. The reported p-value therefore should not be interpreted as an ordinary test for superiority or as evidence of equality.

General explanation: What does a repeated-measures model do?

A mixed model for repeated measures can use observations collected at multiple visits while accounting for correlation among measurements from the same participant. This approach can estimate changes over time and compare arms while using the observed repeated measurements under its modeling assumptions. Those assumptions and the handling of missing visits matter. The registry identifies the method but does not report the visit-level model specification, missing-data assumptions, or sensitivity analyses. The result is therefore best treated as the sponsor’s model-based estimate rather than a directly reproducible calculation from the displayed arm summaries.

How do the secondary binary outcomes differ?

The secondary sweat chloride threshold outcomes classify participants according to whether their value was below a stated threshold. The sponsor’s reported method is a generalized estimated equation model, with odds ratios and confidence intervals. The unadjusted recalculations instead express the comparison as a risk difference, with a confidence interval and an unadjusted p-value. A risk difference describes the absolute difference between proportions; an odds ratio compares the odds of meeting the threshold. These are different effect measures and need not have the same numerical scale or interpretation.

The recalculated secondary comparisons are simple, unadjusted summaries from posted counts and denominators. They do not incorporate the sponsor’s generalized estimated equation model or adjust for other variables. The secondary outcomes also pool data from study VX20-121-103, so their denominators are not the same as those of the primary endpoint. The registry’s listed continuous sweat chloride outcome does not have values available for a two-arm recalculation. No inference should be made from an unavailable estimate.

General explanation: Why can over-dispersion matter?

For binary outcomes, a simple model may assume that variation follows the expected binomial pattern. When the observed variation is greater than that assumption allows, standard errors can be too small and confidence intervals too narrow. Methods that account for clustering or repeated measurements can address some forms of dependence or extra variation. The registry names a generalized estimated equation model for the sponsor’s binary analyses, but does not give enough detail to establish exactly which correlation structure or variance adjustments were used. This general consideration is not evidence that over-dispersion occurred in these data.

What do the planning calculations mean?

The planning illustrations use the observed effect as a starting point and consider detecting a stated fraction of that effect. Their stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For non-time-to-event outcomes, the reported planning quantity is participants per arm for the stated power. These are hypothetical planning illustrations, not the trial’s original sample-size justification, and they are not observed power. The trial’s own sample-size assumptions and justification are not reported in the registry record. Planning estimates based on an observed effect can be sensitive to the estimate chosen and should not be treated as confirmation of the original design or of the effect’s reproducibility.

10. Analyses Reported by the Sponsor

The sponsor’s primary analysis reports a mixed model for repeated measures, a least-squares mean difference of 0.2 percentage points, a 95% confidence interval from -0.7 to 1.1, and p < 0.001 for a noninferiority analysis. The registry record does not flag a discrepancy between the posted primary analysis and a recalculation. The sponsor’s modeled result is the relevant posted primary result; the displayed arm-level summaries alone are not a replacement for that repeated-measures analysis.

For the secondary binary outcomes, the sponsor reports generalized estimated equation model odds ratios. The recalculated comparisons are unadjusted risk differences derived from posted summary counts and denominators. They are not a replication of the sponsor’s model: they use a different effect scale and do not apply the reported modeling approach. Accordingly, they serve as descriptive checks on the posted summaries, not as corrections to or substitutes for the sponsor’s analyses.

The noninferiority p-value does not by itself establish how the conclusion relates to a clinically or scientifically acceptable margin. Since that margin and its rationale are not reported in the registry record, the registry’s noninferiority claim cannot be independently reconstructed from the posted information. The sponsor’s own analysis remains the source for the reported model-based result, while its full interpretation is limited by the absent design and analysis details.

11. Multiplicity, Missing Data and Analysis Populations

The registry record identifies a primary ppFEV1 endpoint and lists secondary outcomes, but a multiplicity strategy is not reported in the registry record. It is therefore unknown whether the secondary hypotheses were ordered hierarchically, whether testing was adjusted across outcomes, or whether the reported p-values are intended to support confirmatory claims. The unadjusted p-values from the secondary recalculations should be interpreted with this absence in mind; they do not account for the possibility of false-positive findings when multiple outcomes are examined.

The record labels the primary analysis as noninferiority but does not report the noninferiority margin, its clinical or statistical rationale, or the detailed rule for declaring noninferiority. The sample-size justification and any planned interim analyses are also not reported in the registry record. These omissions prevent an independent assessment of how the design was powered for its primary objective and how the reported test was intended to guide the conclusion. The planning illustrations based on observed effects are not a substitute for those protocol details.

The primary analysis population is described as the Full Analysis Set, including randomized participants who carried the intended CFTR mutation or mutations and received at least 1 dose during the treatment period. For this endpoint, the record says the analyzed number refers to participants evaluable for the outcome. The registry does not report the missing-data methods, the number or pattern of missing endpoint measurements, or sensitivity analyses addressing missingness. A repeated-measures model can use observed data under assumptions, but those assumptions cannot be verified from the information posted.

The secondary sweat chloride threshold results pool data from study VX20-121-103. The registry record does not report enough detail to evaluate the pooling procedure or how study-level differences were handled. The listed continuous sweat chloride outcome lacks values for a two-arm recalculation. These gaps, together with the absent multiplicity plan, mean that secondary findings should not be treated as a complete, independently verified family of confirmatory tests.

12. 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: infective pulmonary exacerbation of cystic fibrosis (11.4%), influenza (1.5%), covid-19 (1.5%), pneumonia (1.0%), syncope (1.0%).

The safety figures use denominators that differ from the primary endpoint’s analyzed numbers. For serious adverse events, the registry reports 41 affected participants among 202 at risk in the ELX/TEZ/IVA group and 28 among 196 at risk in the VX-121/TEZ/D-IVA group. For other adverse events, it reports 184 among 202 and 177 among 196, respectively. No deaths are reported in either group. These counts are descriptive; they do not by themselves establish a causal difference between treatments.

The most frequently listed serious adverse event term is infective pulmonary exacerbation of cystic fibrosis, reported as 11.4%. Other listed terms are influenza and COVID-19, each 1.5%, and pneumonia, syncope, and suicidal ideation, each 1.0%. The registry excerpt does not specify in this list whether the percentages are treatment-specific or pooled, so they should not be assigned to an individual arm. Nor does the list establish that the events were caused by study treatment.

Serious adverse events and other adverse events are different categories, and these summaries do not describe event duration, severity beyond the serious classification, timing, or treatment attribution. Safety comparisons require attention to exposure and follow-up as well as the number of participants affected. Those details are not reported in the registry record in the information summarized for these results.

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

The primary result is a model-based estimate of the difference in ppFEV1 change between VX-121/TEZ/D-IVA and an active ELX/TEZ/IVA comparator. Its confidence interval includes zero and spans differences in both directions. This describes uncertainty in the estimated average difference; it does not prove that the treatments are identical, that there is no meaningful difference, or that one is superior.

The trial’s primary analysis is labeled noninferiority, a conclusion that depends on comparison with a prespecified margin. Since the margin and its justification are not reported in the registry record, the posted estimate and p-value cannot alone establish whether the noninferiority criterion was met. A p-value associated with a noninferiority analysis should not be read as a test of superiority without support from an appropriate superiority analysis.

The secondary sweat chloride threshold results have a different endpoint, a pooled data basis, and a different analytic approach from the primary ppFEV1 result. Their direction may be of interest, but they do not answer the primary lung-function question. They also carry less inferential weight when no multiplicity adjustment or testing hierarchy is reported. The absence of a recalculable continuous sweat chloride result further limits the scope of interpretation.

These registry results apply to the specified F/MF cystic fibrosis population and the regimens described, including the ELX/TEZ/IVA run-in and subsequent treatment periods. They do not establish effects in other populations, under other treatment conditions, or beyond the reported outcome timeframes. The record does not provide detailed information on missing data, model assumptions, or the full protocol analysis plan, so conclusions should remain tied to the estimates and methods actually posted.

Finally, the safety counts describe participants with reported events and should not be treated as proof that either regimen caused or prevented a particular event. Taken together, the results provide a registry-reported comparison with uncertainty and important gaps in design and analysis details. They support statistical description, not individual treatment advice or a broader causal claim beyond the trial’s stated comparison.

14. Why This Trial Matters Statistically

This trial is statistically informative because it compares an investigational combination with an active regimen rather than with placebo for the primary endpoint. An active comparator changes the interpretation: the principal question is framed as noninferiority, not simply whether the intervention differs from no treatment. The registry’s designation makes the noninferiority margin essential, yet that margin is not reported in the registry record. This illustrates how an effect estimate and p-value can be insufficient without the decision rule they are meant to address.

The primary outcome also highlights the distinction between point estimates and precision. The estimated least-squares mean difference is close to zero on the reported scale, but its confidence interval permits differences in either direction. Readers need the interval and the model context, rather than the p-value alone, to understand what range of effects remains compatible with the data.

The secondary outcomes show why analysis scale and multiplicity matter. Odds ratios from a modeled analysis and unadjusted risk differences are related descriptions, but are not interchangeable. Pooled secondary data, multiple outcomes, and the absence of a reported multiplicity plan all affect how strongly those findings can be interpreted. Statistical significance in an unadjusted secondary comparison is not equivalent to confirmatory evidence.

Finally, the posted record separates efficacy and safety populations and does not fully document missing-data handling, the model specification, or the sample-size rationale. Those are not peripheral details: they affect reproducibility and the strength of inferences. This study is a useful example of why careful registry interpretation distinguishes the reported result, its uncertainty, the prespecified objective, and the information needed to independently assess that objective.

15. Limitations

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

The registry record leaves several trial-specific uncertainties. Most importantly for the primary result, it labels the analysis noninferiority without reporting the margin, the rationale for the margin, or the complete decision rule. The primary p-value and confidence interval therefore cannot be independently evaluated against the intended noninferiority criterion. The original sample-size assumptions and justification are also not reported; the planning illustrations based on observed effects do not fill that gap.

The record names a mixed model for repeated measures but does not give the full model specification, including included covariates, covariance structure, or sensitivity analyses. It describes the Full Analysis Set and says analyzed participants were evaluable for the primary outcome, but does not provide detailed missing-data patterns or methods. The completeness of follow-up and robustness to missing measurements therefore cannot be assessed from the posted information.

The primary endpoint timeframe ends at Week 24, while the described treatment period lasts 52 weeks. The posted primary estimate does not by itself characterize later outcomes. The study population is specifically F/MF cystic fibrosis participants, which limits generalization beyond that group. Both active arms also followed an ELX/TEZ/IVA run-in, a feature that forms part of the trial context and should be retained when interpreting the comparison.

The secondary sweat chloride threshold results pool data with study VX20-121-103, but the pooling methods and study-level comparability are not reported. No unadjusted two-arm recalculation is available for the listed continuous sweat chloride outcome. The registry also does not report how multiplicity was handled, limiting the inferential weight of secondary p-values.

Finally, safety denominators differ from the primary endpoint analyzed numbers, and the listed serious event percentages are not identified by treatment group in the available record. The figures are useful as descriptive reports, but do not support a detailed comparative safety analysis without information on exposure, timing, attribution, and event-level context.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Phase 3 Study of VX-121 Combination Therapy in Participants With Cystic Fibrosis (CF) Heterozygous for F508del and a Minimal Function Mutation (F/MF) — NCT05033080.
• Registry record last updated 2024-10-03; status last verified 2024-09.
• Results first posted 2024-06-10.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT05033080 is a completed randomized, quadruple-masked, parallel phase 3 study sponsored by Vertex Pharmaceuticals Incorporated. It enrolled 435 participants with cystic fibrosis who were heterozygous for F508del and a minimal function mutation. The primary comparison was VX-121/TEZ/D-IVA versus ELX/TEZ/IVA after an ELX/TEZ/IVA run-in period of 4 weeks, with a stated treatment period of 52 weeks.

The primary endpoint was absolute change in ppFEV1 from baseline through Week 24. The sponsor reported a mixed-model repeated-measures noninferiority analysis, with a least-squares mean difference of 0.2 percentage points and a 95% confidence interval from -0.7 to 1.1. The noninferiority margin is not reported in the registry record.

Posted secondary sweat chloride threshold results are pooled with data from another study and should be interpreted separately from the primary endpoint. The record includes descriptive adverse event counts and reports no deaths. Multiplicity handling, detailed missing-data methods, the full model specification, and the original sample-size justification are not reported in the registry record.