Study of Efficacy and Safety of Twice Daily Oral LNP023 in Adult PNH Patients With Residual Anemia Despite Anti-C5 Antibody Treatment: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Efficacy and Safety of Twice Daily Oral LNP023 in Adult PNH Patients With Residual Anemia Despite Anti-C5 Antibody Treatment (NCT04558918), a randomized phase 3 study. The primary outcome is Marginal Proportion (Expressed as Percentages) of Participants With Sustained Increase in Hemoglobin Levels From Baseline of ≥ 2 g/dL in the Absence of Red Blood Cell Transfusions.

2. Clinical Question

The clinical question concerns residual anemia while participants were already receiving anti-C5 therapy, rather than treatment-naive PNH. The primary endpoint combines an increase in hemoglobin with the absence of transfusions over specified periods. That makes the endpoint a response criterion with more than one component: a hemoglobin change must be sustained across scheduled measurements, and a participant must not receive a transfusion or meet the protocol’s stated transfusion criteria during the defined interval.

The trial was designed to compare the two randomized groups, not to compare different anti-C5 drugs against one another. It was open-label, so participants and study personnel were not masked to assignment. Randomization supports a between-group comparison, while the open-label design and the endpoint’s reliance on measurements and transfusion decisions are relevant context when interpreting the result. The registry states that the primary estimates included adjustment for baseline covariates and accounted for missing data, but the detailed adjustment and missing-data procedures are not reported in the registry record.

3. Trial Design

This was a Phase 3, randomized, multicenter, active-comparator-controlled, open-label, parallel-group trial. The registry lists an enrollment of 97 participants. It records a start date of 2021-01-25, a primary completion date of 2022-09-26, and a completed status. Participants were assigned to LNP023 200mg twice daily or to continued anti-C5 antibody treatment.

In the randomized treatment period, the LNP023 arm received oral iptacopan in a hard gelatin capsule. After 24 weeks, participants in that arm had the option to enter an extension treatment period and receive an additional 24 weeks of LNP023. Participants assigned to anti-C5 antibody continued the same stable regimen they had received before randomization. The registry describes eculizumab as an intravenous infusion every 2 weeks, but the comparator-arm description is incomplete; further regimen details are not reported in the registry record.

The primary analysis used the Full Analysis Set, defined in the registry as patients to whom study treatment had been assigned by randomization. The posted arm sizes for the primary endpoint were 62 for LNP023 and 35 for anti-C5 antibody. The unequal arm sizes are the reported allocation for this outcome; the allocation ratio or reason for those numbers is not reported in the registry record.

The active-comparator design addresses the question of adding a different treatment approach for participants who remained anemic despite anti-C5 therapy. Because the study was open-label, treatment assignment was known during the trial. Randomization and the stated baseline adjustment are important features of the comparison, while the lack of masking is part of the context for interpreting outcomes that may involve clinical decisions. The registry record does not report the original sample-size justification, details of the randomization procedure, or a fuller account of how the open-label design was managed.

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: Vaccination against Neisseria meningitidis, Streptococcus pneumoniae and Haemophilus influenzae infections was required prior to the start of treatment, if the patient had not been previously vaccinated, or if a booster was required. The vaccines were given according to local regulations, at least 2 weeks prior to first dosing. If iptacopan treatment had to start earlier than 2 weeks post vaccination, prophylactic antibiotic treatment was initiated.

Randomized Treatment Period

Extension Treatment Period

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS): patients to whom study treatment had been assigned by randomization.

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 Marginal Proportion (Expressed as Percentages) of Participants With Sustained Increase in Hemoglobin Levels From Baseline of ≥ 2 g/dL in the Absence of Red Blood Cell Transfusions, measured over Baseline, hemoglobin between Day 126 and Day 168 and absence of transfusions between Day 14 and Day 168.

The primary endpoint was the marginal proportion of participants with a sustained increase in hemoglobin from baseline of at least 2 g/dL, in the absence of red blood cell transfusions. The increase had to be present on three out of four measurements at visits in the last six weeks of the randomized treatment period, between Day 126 and Day 168. The transfusion-free requirement covered Day 14 through Day 168.

For this definition, requiring a transfusion included receiving one or meeting protocol-defined criteria: hemoglobin at or below 9 g/dL with signs or symptoms sufficiently severe to warrant a transfusion, or hemoglobin at or below 7 g/dL regardless of signs or symptoms. The endpoint therefore measures a sustained hemoglobin response together with avoidance of transfusion or meeting the stated criteria. It is not simply a single hemoglobin measurement at one visit.

The registry describes a marginal proportion as the population-average probability of being a responder in each treatment group. The posted values account for baseline covariates, and missing data were also taken into account. The precise statistical procedures used for those adjustments are not reported in the registry record.

Using a sustained response can make an endpoint less dependent on a single measurement, while including transfusion status gives the response definition clinical context. Those are general features of this type of endpoint, not evidence about how a particular participant experienced the trial. The registry’s specified threshold and time windows determine who counts as a responder; participants who did not meet every part of the definition were not counted as responders for this outcome.

7. Results Posted in the Registry

• Anti-C5 Antibody: 2.9% (1/35) versus 82.3% (51/62) on control; risk difference -79.4 percentage points (95% CI -90.4 to -68.4), risk ratio 0.03 (0.01–0.24); unadjusted two-proportion z-test p = <0.001.

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.

Marginal Proportion (Expressed as Percentages) of Participants Who Remain Free From Transfusions

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Anti-C5 Antibody: 25.7% (9/35) versus 95.2% (59/62) on control; risk difference -69.4 percentage points (95% CI -84.9 to -54.0), risk ratio 0.27 (0.15–0.48); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Anti-C5 Antibody than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry identifies secondary outcomes that extend beyond the primary sustained hemoglobin response. They include remaining free from transfusions, hemoglobin change, fatigue, absolute reticulocyte count, log-transformed lactate dehydrogenase, clinical breakthrough hemolysis, and major adverse vascular events. These outcomes address related but distinct aspects of the disease and treatment period; they do not all measure the same construct as the primary responder definition.

The posted secondary result for remaining free from transfusions is conceptually close to the primary endpoint because both involve transfusion status. It is not identical: the secondary outcome focuses on remaining transfusion-free during its stated time window, while the primary outcome additionally requires a sustained hemoglobin increase and applies its own transfusion criteria. That relationship makes the secondary measure relevant context, but not a substitute for the primary outcome.

The secondary outcomes cover several domains, so they can help describe a broader pattern than a single endpoint. However, the registry record does not support a general claim that all secondary outcomes moved in the same direction or had equally strong evidence. Their estimates, uncertainty, and analytic details need to be considered outcome by outcome. The outcomes also differ in measurement scale and interpretation, so a single summary across them would obscure important distinctions.

Secondary findings generally carry less confirmatory weight when multiple outcomes are examined and multiplicity handling is not documented. Testing many outcomes increases the chance of at least one apparently notable result arising by chance if each is interpreted in isolation. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for the secondary outcomes. Accordingly, secondary results are best treated as supporting or exploratory context rather than as independent confirmation of the primary result. The registry lists 17 outcome measures and 17 analyses, while the posted secondary outcome set includes 7; the record does not explain the relationship between those counts in greater detail.

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

• Marginal Proportion (Expressed as Percentages) of Participants Who Remain Free From Transfusions
• Change From Baseline in Hemoglobin Between Day 126 and 168
• Change From Baseline in FACIT-Fatigue Questionnaire in the Randomized Treatment Period
• Change From Baseline in Absolute Reticulocyte Count in the Randomized Treatment Period
• Ratio to Baseline in Log-transformed LDH in the Randomized Treatment Period
• Adjusted Annualized Clinical BTH Rate in the Randomized Treatment Period
• Adjusted Annualized Major Adverse Vascular Events Rate in the Randomized Treatment Period

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: a participant either met the complete responder definition or did not. The registry then presents each treatment group’s marginal proportion, which it defines as the population-average probability of response. This is different from reporting only the raw fraction of participants who responded. The registry states that its marginal proportions include adjustment for baseline covariates and take missing data into account, but the covariates and the precise missing-data approach are not reported in the registry record.

What does a marginal proportion represent?

A marginal proportion puts the treatment-group result on a probability scale. It answers, in the registry’s stated terms, what average probability of meeting the response definition is estimated for each group after the analysis adjustments. The reported values are 82.3% for LNP023 and 2.0% for anti-C5 antibody. These estimates are not interchangeable with the unadjusted observed proportions, because adjustment and missing-data handling can change the estimate. General statistical principle: a probability-scale result is often easier to interpret clinically than an odds ratio, but it remains specific to the outcome definition, population, and analysis model.

What is the difference in marginal proportions?

A difference in marginal proportions compares the two groups on the same probability scale. The sponsor reports a difference of 80.2 percentage points, with a 95% confidence interval from 71.2 to 87.6. Its sign and verbal interpretation depend on the order in which groups are contrasted, so the arm-specific proportions should be read alongside the reported difference. The registry classifies the comparison as a superiority analysis.

Why does the report also include an odds ratio?

The sponsor reports a logistic-regression odds ratio of 338.25, with a 95% confidence interval from 25.07 to 4564.14. Logistic regression models the log odds of a binary outcome, and its exponentiated contrast is an odds ratio. An odds ratio is not a risk ratio or a percentage-point difference. When response probabilities are far apart, an odds ratio can be numerically large and can be less intuitive than group proportions. The wide interval also shows that the odds-ratio magnitude is imprecisely estimated, despite the reported superiority classification. The registry identifies logistic regression as the method for this odds-ratio analysis; further model details are not reported in the registry record.

What does the unadjusted recalculation add?

The unadjusted recalculation uses the posted counts of responders and participants in each arm: 51 of 62 for LNP023 and 1 of 35 for anti-C5 antibody. It is a direct comparison of these summary counts and has a reported p-value of <0.001. General statistical principle: an unadjusted analysis does not account for covariates or missing-data procedures in the same way as an adjusted model. It can serve as a transparent descriptive check, but it answers a more limited question than the sponsor’s adjusted analysis. Since the calculations differ in method, numerical estimates need not be identical.

How should a p-value be understood?

A p-value describes how compatible data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The registry reports a p-value for the unadjusted recalculation and a smaller reported threshold for the sponsor’s odds-ratio analysis. These results should be considered with the effect estimates and confidence intervals rather than treated as stand-alone measures.

Why does over-dispersion matter in general?

Over-dispersion is a general concern when observed variability exceeds what a statistical model assumes. If a model understates variability, standard errors can be too small and confidence intervals or tests can look more certain than warranted. The primary endpoint is binary, and the registry record does not report an over-dispersion assessment or a model adjustment for it. That absence does not establish that over-dispersion occurred; it means the record does not document an assessment. More generally, model assumptions should be considered alongside the reported estimates, especially when event counts are sparse or group sizes differ.

Other statistical details are not reported in the registry record, including the full covariate set, model diagnostics, any planned sensitivity analyses, interim analyses, and the original statistical analysis plan. Those omissions limit independent evaluation of the exact modeling choices. The registry does identify the primary outcome, analysis population, endpoint timing, main reported analyses, and treatment groups, which permit a substantive interpretation of the posted primary result while leaving those details unresolved.

11. Confidence Intervals and Precision

A confidence interval describes a range of values compatible with the data and the analysis model at the stated confidence level. In general, narrower intervals indicate greater precision for the parameter on the scale being estimated; they do not guarantee that an estimate is unbiased or that the model is correct. A 95% confidence interval is not a guarantee that the true value lies in that particular interval.

For the primary adjusted marginal proportions, the 95% confidence interval is 73.4% to 90.2% for LNP023 and 1.1% to 4.0% for anti-C5 antibody. The sponsor-reported 95% interval for the difference in marginal proportions is 71.2 to 87.6 percentage points. These intervals give uncertainty around the respective estimates and retain the endpoint’s specified responder definition.

The odds-ratio interval, from 25.07 to 4564.14, spans a broad range of possible magnitudes on the odds scale. This indicates much less precision about the size of the odds ratio than a single point estimate might suggest. Its scale is different from the percentage-point difference, so the two intervals should not be compared as if they measured the same quantity. The large odds-ratio estimate does not mean that the response probability is hundreds of times greater.

Precision also depends on the amount of information in each group and the endpoint’s event pattern. The primary analysis includes 62 participants in the LNP023 arm and 35 in the anti-C5 antibody arm, with 51 and 1 responders, respectively, in the posted counts used for the unadjusted recalculation. Sparse outcome counts can make estimates on some scales unstable, particularly odds ratios. General statistical principle: confidence intervals quantify sampling uncertainty under the analysis assumptions, not uncertainty from every source, such as measurement choices, missing-data assumptions, or limitations in generalizability. The registry does not provide enough methodological detail to assess all of those sources independently.

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

For the primary endpoint, planning for 100% of the observed effect gives 5 participants per arm for 80% power and 6 participants per arm for 90% power. Planning for 75% of the observed effect gives 10 participants per arm for 80% power and 13 participants per arm for 90% power. Planning for 50% of the observed effect gives 23 participants per arm for 80% power and 30 participants per arm for 90% power.

In plain terms, the 80% power column describes the number of participants per arm estimated to provide an 80% chance of detecting the specified effect under the stated assumptions, if that effect and model assumptions hold. The 90% column uses a higher target chance of detection. Smaller effects require more participants because they are harder to distinguish from random variation. The illustrated numbers are sensitive to the unusually large observed separation in this trial and should not be treated as a validated design recommendation for another study. This is not observed power; it is a planning calculation based on the observed effect.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.823; treatment proportion p₁ = 0.029; average p̄ = 0.426.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6992 = 1.3705.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4168 = 0.3508.
4. n = (1.3705 + 0.3508)² / (-0.7940)² = 4.7, rounded up to 5 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
p0 <- 0.82258064516129                          # control proportion
p1 <- 0.0285714285714286                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(4558918)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 5 (80% power), 6 (90% power)
 75% of observed effect: n per arm = 10 (80% power), 13 (90% power)
 50% of observed effect: n per arm = 23 (80% power), 30 (90% power)
power.prop.test: 4.7 
Simulated power at n = 5 per arm: 0.879 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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 sponsor’s primary reported analysis is the adjusted comparison of marginal proportions. It reports an 80.2 percentage-point difference, with a 95% confidence interval from 71.2 to 87.6, and also reports a logistic-regression odds ratio of 338.25, with a 95% confidence interval from 25.07 to 4564.14. The registry classifies the sponsor analyses as superiority analyses. The two sponsor estimates use different scales: the marginal difference is a percentage-point contrast, while the odds ratio compares odds.

The recalculation is a simple unadjusted calculation from the posted responder counts: 51 of 62 participants in the LNP023 group and 1 of 35 in the anti-C5 antibody group. Its reported p-value is <0.001. It is useful as a transparent check on the observed counts, but it is not the same estimand or procedure as the sponsor’s adjusted analysis. It does not reproduce the registry’s baseline adjustment or missing-data handling.

The registry record does not flag a discrepancy. The sponsor analysis should therefore be read as the main reported analysis, while the recalculation is a separate descriptive comparison. Differences in estimates across these approaches do not alone indicate an error; the methods and target quantities differ. The very wide odds-ratio interval is especially important context: the reported contrast is large, but the precise odds-ratio magnitude is uncertain. The marginal proportions and their difference provide a more direct probability-scale account of the endpoint.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes, time points, comparisons, or analyses. General statistical principle: each additional statistical test creates another opportunity to observe a result that appears notable by chance, even when no true effect exists for that test. A prespecified testing hierarchy or an adjustment to the testing procedure can help control that issue, but the registry record does not report whether either was used for the primary and secondary analyses.

The registry identifies the primary endpoint and reports a superiority analysis, which helps distinguish the main question from the other outcomes. It also lists multiple secondary outcome domains and reports 17 outcome measures and 17 analyses. The number of secondary outcomes posted is 7. The record does not describe a multiplicity strategy, whether secondary tests were intended to be confirmatory, or how any hierarchy among outcomes was defined. Consequently, nominal evidence from separate secondary analyses should not be interpreted as if each were the sole prespecified test.

Missing data are acknowledged in the primary endpoint description: the marginal proportions take missing data into account. The exact approach is not reported in the registry record. It is therefore not possible to determine from the record how missing measurements were handled, what assumptions were made about why data were missing, or whether sensitivity analyses tested alternative assumptions. General statistical principle: missing-data methods can affect estimates when the likelihood of missingness is related to participant characteristics or outcomes, so details of the method and sensitivity analyses matter.

The primary analysis population is the Full Analysis Set, defined as patients assigned treatment by randomization. The record does not provide additional detail on exclusions after randomization, adherence, or analysis populations used for every secondary or safety outcome. It also does not report the original statistical analysis plan, interim analyses, or a full account of model diagnostics. These gaps do not invalidate the posted estimates, but they limit independent assessment of whether every analysis followed a prespecified plan and how robust the results are to analytic choices.

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: covid-19 (5.7%), breakthrough haemolysis (2.9%), extravascular haemolysis (2.9%), jaundice (2.9%), arthritis bacterial (2.9%).

During the randomized treatment period, the registry reports serious adverse events for 6 of 62 participants in the LNP023 group and 5 of 35 in the anti-C5 antibody group. Other adverse events were reported for 34 of 62 and 21 of 35, respectively. No deaths are reported in either group for this period. These are counts with the group sizes at risk, not a complete account of event severity, timing, duration, or attribution.

The registry also reports serious adverse event counts over the randomized period plus the extension period for LNP023, and for any LNP023 exposure. Those groupings have different exposure definitions and denominators, so their counts should not be compared as if they represented the same follow-up period or risk. The registry lists COVID-19 at 5.7% among the top serious-event terms, and breakthrough haemolysis, extravascular haemolysis, jaundice, arthritis bacterial, and intervertebral discitis at 2.9% each.

Safety counts alone do not establish that an event was caused by treatment, and counts across arms cannot be interpreted without considering exposure and follow-up. The registry record does not provide the event-level detail needed for a fuller comparison of timing, duration, severity, or treatment attribution. These figures describe reported events in the specified groups and periods, rather than serving as a complete safety assessment.

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

The primary result describes a substantial separation between the randomized groups for the registry-defined responder outcome. The adjusted marginal proportions were 82.3% for LNP023 and 2.0% for anti-C5 antibody, and the sponsor-reported difference was 80.2 percentage points. The endpoint required both a sustained hemoglobin increase and no transfusion or protocol-defined transfusion criterion during the stated period. The estimate therefore pertains to that specific combined response definition.

The result does not establish that every participant experienced the same change, that all outcomes improved to the same extent, or that the estimated contrast applies to people outside the trial population. The study enrolled adults with PNH and residual anemia despite anti-C5 antibody treatment. The comparator was continued anti-C5 therapy, and the trial was open-label. Extrapolation to different patient groups, treatment histories, or comparators is not established by the registry record.

The result also does not resolve how durable the response is beyond the randomized treatment period or the extension period described in the record. The primary endpoint’s assessment window was between Day 126 and Day 168, with transfusion status considered between Day 14 and Day 168. No inference about outcomes beyond those stated windows follows from the primary result alone.

Confidence intervals communicate uncertainty under the analysis assumptions, but they do not incorporate every limitation. The odds-ratio interval is very wide, while the probability-scale estimates make the group contrast easier to describe. The primary analysis accounts for baseline covariates and missing data according to the registry, but the precise methods are not reported. The unadjusted recalculation is a separate check based on posted counts and does not supersede the sponsor’s adjusted analysis.

These statistical results are evidence about outcomes observed and analyzed in this trial; they are not medical advice, a guarantee of an individual response, or a complete judgment of benefits and harms. The registry’s safety counts provide context, but they do not by themselves establish causation or fully characterize safety.

18. Why This Trial Matters Statistically

Statistically, this trial illustrates why the choice of outcome scale matters. The primary endpoint is presented as adjusted marginal proportions and a difference in those proportions, which communicate response probability and an absolute percentage-point contrast. The sponsor also reports a logistic-regression odds ratio. Because odds ratios and probability differences answer different quantitative questions, reading both alongside the arm-specific proportions prevents a large odds ratio from being mistaken for a similarly sized probability change.

The trial also shows why uncertainty needs to accompany a point estimate. The sponsor’s odds-ratio interval extends from 25.07 to 4564.14, leaving considerable uncertainty about the odds-ratio magnitude. The confidence intervals for the marginal proportions and their difference give a probability-scale view of precision. A statistically classified superiority result does not eliminate uncertainty about the size or broader applicability of the observed contrast.

The endpoint combines a sustained laboratory threshold with transfusion status over specified time periods. That design makes the definition more informative than a single hemoglobin value, but it also means the result depends on several criteria being met. The registry’s explicit responder definition is essential to understanding what the analysis counts.

Finally, the trial’s open-label design, missing-data adjustment, and multiple secondary outcomes illustrate common interpretive considerations in clinical research. The registry documents some key features, including randomization, the Full Analysis Set, and baseline adjustment for the primary marginal proportions. It does not report the original sample-size justification, multiplicity strategy, detailed missing-data method, or full statistical plan. This distinction between what is documented and what remains unknown is central to a careful statistical reading.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several trial-specific limitations affect how broadly the result can be interpreted. The study was open-label, so treatment assignment was known. Randomization supports comparison between assigned groups, but lack of masking remains relevant when outcomes involve clinical decisions, including transfusion criteria. The registry describes thresholds for qualifying as requiring transfusion, but does not report further safeguards against differences in assessment between groups.

The primary analysis uses adjusted marginal proportions and accounts for missing data, but the registry does not report the covariates, the missing-data method, or sensitivity analyses. These details would help readers assess how dependent the estimates are on modeling and missingness assumptions. The posted unadjusted recalculation is informative as a simple comparison of counts, but it does not address those same considerations.

The arm sizes are unequal, with 62 participants in the LNP023 group and 35 in the anti-C5 group for the primary outcome. The posted counts include only 1 responder in the anti-C5 group, which contributes to uncertainty in estimates on scales such as the odds ratio. The odds-ratio confidence interval is especially wide. The reported response contrast is clear on the marginal-proportion scale, but the exact odds-ratio magnitude is less precise.

The primary endpoint is tied to a specific treatment period and assessment window. Participants in the LNP023 group could enter an extension, which creates additional safety groupings but does not extend the stated primary endpoint window. The registry record does not report the original sample-size justification or the details of any interim analyses. It also does not describe a multiplicity strategy for the numerous outcome measures and analyses. These limitations counsel against treating secondary results as independent confirmation or treating the planning illustrations as the trial’s original design basis.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Efficacy and Safety of Twice Daily Oral LNP023 in Adult PNH Patients With Residual Anemia Despite Anti-C5 Antibody Treatment — NCT04558918.
• Registry record last updated 2024-10-09; status last verified 2024-10.
• Results first posted 2023-11-28.
• 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

The registry record describes a completed Phase 3, randomized, multicenter, open-label, parallel-group trial of oral LNP023 200mg twice daily against continued anti-C5 antibody treatment in adults with PNH and residual anemia. Enrollment was 97, and the primary analysis used the Full Analysis Set.

The primary outcome was a sustained hemoglobin increase of at least 2 g/dL without red blood cell transfusions, assessed using the stated measurement and time-window criteria. The adjusted marginal response proportions were 82.3% for LNP023 and 2.0% for anti-C5 antibody. The sponsor reported a difference in marginal proportions of 80.2 percentage points and a logistic-regression odds ratio of 338.25; the odds-ratio confidence interval was wide.

The registry also lists secondary outcomes and reports adverse-event counts for the randomized and extension periods. The open-label design, unreported details of missing-data and multiplicity methods, and limited information on the original design assumptions constrain interpretation. The posted results describe the defined outcomes in this trial and do not settle questions beyond its population, periods, and analyses.