Efficacy and Safety Study of Maribavir Treatment Compared to Investigator-assigned Treatment in Transplant Recipients With Cytomegalovirus (CMV) Infections That Are Refractory or Resistant to Treatment With Ganciclovir, Valganciclovir, Foscarnet, or Cidofovir: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety Study of Maribavir Treatment Compared to Investigator-assigned Treatment in Transplant Recipients With Cytomegalovirus (CMV) Infections That Are Refractory or Resistant to Treatment With Ganciclovir, Valganciclovir, Foscarnet, or Cidofovir (NCT02931539), a randomized phase 3 study. The primary outcome is Percentage of Participants Who Achieved Confirmed Clearance of Plasma Cytomegalovirus (CMV) Deoxyribonucleic Acid (DNA) (CMV Viremia Clearance) at End of Week 8.

2. Clinical Question

The central statistical question is whether the observed proportion with confirmed clearance differed between the randomized maribavir and investigator-assigned treatment groups. The posted primary outcome includes participants regardless of whether study-assigned treatment was stopped before the end of the stipulated 8 weeks, provided they had not received alternative anti-CMV treatment. That definition matters: the outcome is not simply a measure of completing treatment, and the clearance endpoint does not by itself capture symptom control, recurrence, safety, or other aspects of a participant’s course.

3. Trial Design

This was a Phase 3, multicenter, randomized, open-label, active-controlled study using a parallel-group design. The registry record lists Shire as sponsor and the study status as completed. The study began on 2016-12-22, and primary completion was 2020-08-17. Enrollment was 352. The registry record does not report the original sample-size justification or the design assumptions used to choose enrollment.

Participants were randomized to maribavir or investigator-assigned anti-CMV treatment. The maribavir regimen was 400 mg orally twice daily for 8 weeks, with doses separated by a minimum of 8 hours. For the comparison group, the investigator selected an anti-CMV agent considered best suited to the participant, using ganciclovir, valganciclovir, foscarnet, or cidofovir, with dosing prescribed by the investigator for 8 weeks. The active comparator is therefore a treatment strategy that could involve different drugs and dosing regimens, rather than a single fixed medication regimen.

The study was open-label, so masking was not used. The registry record does not report whether outcome assessors or laboratory personnel were masked separately. Randomization supports a treatment-group comparison, while open-label conduct can matter for outcomes affected by participant or investigator knowledge of treatment. The primary endpoint was based on plasma CMV DNA measurements, with a specified assay and confirmation rule; its interpretation still depends on the registry’s outcome definition and analysis methods.

The primary outcome was assessed at the end of Week 8. Additional listed outcomes address clearance and symptom control, persistence of treatment effects, and recurrence during treatment or follow-up. Those outcomes broaden the description of response but do not change what the primary endpoint measures. The detailed randomization procedure, allocation concealment, prespecified stratification factors, and original statistical analysis plan are not reported in the registry record.

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: Participants with cytomegalovirus (CMV) infections were enrolled and randomized into two treatment groups: IAT (control), and Maribavir 400 mg. Participants randomized to IAT treatment arm, if met the stringent criteria for lack of improvement/worsening of CMV infection, considered eligible for entry into Maribavir rescue arm at Week 3 based on medical monitor review. As planned, combined data has been reported for IAT control group.

Period 1: Treatment Phase

Reasons for not completing this period:

Period 2: Maribavir Rescue Therapy

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The randomized set consisted of all participants who had signed informed consent and had begun some study procedures and were randomized to the study.

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: The randomized set consisted of all participants who had signed an informed consent and had begun some study procedures, and randomized to the study.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Who Achieved Confirmed Clearance of Plasma Cytomegalovirus (CMV) Deoxyribonucleic Acid (DNA) (CMV Viremia Clearance) at End of Week 8, measured over Week 8.

The primary endpoint is the percentage of participants with confirmed clearance of plasma CMV DNA at the end of Week 8. The registry definition requires plasma CMV DNA concentration below the lower limit of quantification, specified as less than 137 IU/mL using the COBAS AmpliPrep/COBAS TaqMan CMV Test, in 2 consecutive postbaseline samples separated by at least 5 days. Requiring consecutive measurements makes the endpoint a confirmation of clearance rather than a single low measurement.

The endpoint is useful for a treatment comparison because it expresses the response as the proportion meeting a prespecified laboratory criterion at a defined assessment time. It focuses on CMV viremia, not on every clinical consequence of infection. Other listed outcomes separately address symptom control, maintenance of response, and recurrence, so those questions should not be treated as interchangeable with the primary endpoint.

The primary outcome description says that participants were counted regardless of whether either study-assigned treatment was discontinued before the end of the stipulated 8 weeks, and that participants could not have received alternative anti-CMV treatment. This helps specify how treatment discontinuation and alternative treatment relate to the endpoint. The registry record does not report the full missing-data procedure or every rule used for incomplete or unavailable laboratory assessments. The result should therefore be read as a comparison under the posted endpoint definition, not as a complete account of all possible clinical outcomes.

7. Results Posted in the Registry

• Maribavir 400 mg: 55.7% (131/235) versus 23.9% (28/117) on control; risk difference 31.8 percentage points (95% CI 21.8 to 41.8), risk ratio 2.33 (1.65–3.28); 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.

Percentage of Participants Who Achieved Confirmed CMV Viremia Clearance and CMV Infection Symptom Control at End of Week 8, Followed by Maintenance of Treatment Effect at Week 16

• 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:
  • Maribavir 400 mg: 18.7% (44/235) versus 10.3% (12/117) on control; risk difference 8.5 percentage points (95% CI 1.0 to 15.9), risk ratio 1.83 (1.00–3.32); unadjusted two-proportion z-test p = 0.041.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Maribavir 400 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Who Completed 8 Weeks of Study-assigned Treatment With Recurrence of CMV Viremia During the First 8 Weeks of the Treatment

• 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:
  • Maribavir 400 mg: 15.3% (28/183) versus 10.8% (4/37) on control; risk difference 4.5 percentage points (95% CI -6.8 to 15.8), risk ratio 1.42 (0.53–3.79); unadjusted two-proportion z-test p = 0.480.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Maribavir 400 mg and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants Who Completed 8 Weeks of Study-assigned Treatment With Recurrence of CMV Viremia During the 12 Weeks of Follow-up Period

• 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:
  • Maribavir 400 mg: 41.0% (75/183) versus 35.1% (13/37) on control; risk difference 5.8 percentage points (95% CI -11.1 to 22.8), risk ratio 1.17 (0.73–1.87); unadjusted two-proportion z-test p = 0.508.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Maribavir 400 mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes extend the primary Week 8 laboratory result to questions about symptom control, maintenance of response, and recurrence during treatment or follow-up. Across the posted secondary comparisons, the pattern is not uniform: some response-and-maintenance measures favor maribavir descriptively, while the recurrence outcomes do not provide the same clear separation. This broader set of outcomes helps distinguish initial clearance from durability and recurrence, but each outcome has its own definition and analysis population.

The registry lists 25 secondary outcomes. The posted results include a reported Cochran-Mantel-Haenszel analysis for the combined clearance-and-symptom-control outcome maintained through Week 16. The registry also posts recurrence comparisons among participants who completed 8 weeks of assigned treatment. Those completion-based analyses concern a selected subset and should not be treated as equivalent to the primary randomized-set comparison. The endpoint definitions and denominators differ, so direct numerical comparisons across outcomes can be misleading.

Secondary findings generally carry less inferential weight when many outcomes are examined and no multiplicity procedure is reported. Repeated testing can increase the chance that at least one result appears statistically notable by chance. The registry record does not report whether secondary analyses were adjusted for multiplicity, how the outcome hierarchy was prespecified, or whether any other multiplicity strategy was used. The unadjusted secondary p-values should therefore be interpreted as exploratory evidence rather than as a collection of equally definitive confirmatory tests.

The secondary results complement, rather than replace, the primary outcome. They can indicate whether the Week 8 clearance finding aligns with measures of symptoms or later response, but they do not establish a single overall treatment effect across all aspects of CMV infection. The registry record does not report every detailed secondary analysis method or missing-data procedure. Conclusions should remain tied to each outcome’s stated timeframe and analysis population.

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

• Percentage of Participants Who Achieved Confirmed CMV Viremia Clearance and CMV Infection Symptom Control at End of Week 8, Followed by Maintenance of Treatment Effect at Week 16
• Percentage of Participants Who Achieved Confirmed CMV Viremia Clearance After Receiving 8 Weeks of Study-assigned Treatment
• Percentage of Participants Who Achieved Confirmed CMV Viremia Clearance and CMV Infection Symptom Control After Receiving 8 Weeks of Study-assigned Treatment Through Weeks 12, 16 and 20
• Percentage of Participants Who Maintained CMV Viremia Clearance and CMV Infection Symptom Control at the End of Study Week 8 Through Weeks 12 and 20 Regardless of Whether Either Study-assigned Treatment Was Discontinued Before 8 Weeks of Therapy
• Percentage of Participants With Recurrence of CMV Viremia During the First 8 Weeks of Study Regardless of Whether Study-assigned Treatment Was Discontinued Before 8 Weeks of Therapy
• Percentage of Participants With Recurrence of CMV Viremia During the 12 Weeks Follow-up Period Regardless of Whether Study-assigned Treatment Was Discontinued Before 8 Weeks of Therapy
• Percentage of Participants With Recurrence of CMV Viremia at Any Time on Study Regardless of Whether Study-assigned Treatment Was Discontinued Before 8 Weeks of Therapy
• Percentage of Participants Who Completed 8 Weeks of Study-assigned Treatment With Recurrence of CMV Viremia During the First 8 Weeks of the Treatment

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: each participant either met the confirmed-clearance definition by the endpoint time or did not. Reporting the percentage of participants meeting that criterion gives a response proportion for each group. The response percentages are 23.9% for investigator-assigned anti-CMV treatment and 55.7% for maribavir 400 mg. The corresponding posted response counts are 28 of 117 and 131 of 235. These are the endpoint-specific group denominators and counts reported for the primary comparison; they should not be confused with enrollment or with denominators used for other outcomes.

A risk difference expresses the contrast between two response proportions on the proportion scale. The simple recalculation gives 0.318, with a confidence interval from 0.218 to 0.418. A positive risk difference in this comparison corresponds to a higher observed response proportion in the maribavir group. The recalculation is unadjusted and is based on posted summary statistics; it is not a reconstruction of the sponsor’s full analysis. No additional adjustment should be inferred from it.

The registry reports a Cochran-Mantel-Haenszel analysis for the primary endpoint, expressed as a difference in percentage of responders. In general, the Cochran-Mantel-Haenszel method combines information across strata to estimate or test an association while accounting for the stratification structure. The specific strata and implementation details are not reported in the registry record, so those should not be guessed. The registry’s reported result is 32.8, with a 95% confidence interval from 22.80 to 42.74 and p < 0.001. The registry labels the analysis as superiority. That reported analysis is distinct from the simple unadjusted recalculation.

A confidence interval describes a range of estimates compatible with the data under the method’s assumptions. It communicates precision as well as direction: narrower intervals indicate less statistical uncertainty than wider intervals, while an interval that includes the no-difference value leaves more uncertainty about the direction of the effect. The registry’s reported primary interval does not include no difference. For a risk difference, the no-difference value is zero; for other measures, the reference value depends on the measure. Confidence intervals do not account automatically for bias, missing data, multiplicity, or limitations in the study design.

A p-value summarizes how unusual the observed data, or data more extreme, would be under a specified null hypothesis and statistical model. The reported p < 0.001 is evidence against the null hypothesis of no difference under the registry’s analysis. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of the result. The registry also posts an unadjusted recalculation p-value of <0.001. That value belongs to the simple recalculation and should not be substituted for the registry’s reported analysis.

Risk ratios and odds ratios are alternative ways to express a binary-outcome comparison. A risk ratio compares response proportions as a ratio; an odds ratio compares response odds. They answer related but different questions and can have different numerical values. The registry record includes these alternative estimates for the primary comparison, but the main sponsor-reported result is the difference in percentage of responders. The choice of measure affects interpretation, so estimates on different scales should not be treated as if they were the same quantity.

The primary outcome’s measurement rule is also part of the statistical method. The threshold is below the assay’s lower limit of quantification, and clearance must be seen in 2 consecutive postbaseline samples with at least 5 days between them. This confirmation requirement is intended to distinguish a sustained qualifying laboratory response from a single measurement. The result still depends on sample collection and availability and on the registry’s rules for handling observations that do not meet the endpoint definition. The detailed procedures for missing or unevaluable assessments are not reported in the registry record.

In general, over-dispersion is a concern when observed binary or count data vary more than a statistical model assumes, which can make standard errors too small if the model does not account for the extra variation. The primary endpoint is a binary participant-level response, and the registry identifies a Cochran-Mantel-Haenszel analysis rather than a count model. The registry record does not report an over-dispersion assessment or a model-based adjustment for it. No such analysis should be presumed.

Finally, the open-label design and the active-control strategy affect the scope of inference, not the arithmetic of the response proportions. Randomization provides the basis for comparing assigned groups, but it does not remove every possible source of uncertainty. The comparator allowed several agents and investigator-selected regimens. The exact randomization procedures, any stratification variables, and the detailed analysis plan are not reported in the registry record. The posted estimates should be read with those boundaries in mind.

11. Confidence Intervals and Precision

The sponsor-reported primary estimate is a difference in percentage of responders of 32.8. Its 95% confidence interval extends from 22.80 to 42.74, and the reported p-value is <0.001. The interval is entirely on the positive side of no difference, which is consistent with the reported superiority analysis. Its width still represents uncertainty about the magnitude of the group difference; the point estimate alone should not be read as exact.

The simple unadjusted recalculation gives a risk difference of 0.318, with an interval from 0.218 to 0.418. The recalculation’s interval also lies above zero. These estimates are on the risk-difference scale and come from a simple analysis of the posted counts. The registry’s reported analysis uses a Cochran-Mantel-Haenszel method and reports the difference in percentage of responders. The registry record does not provide enough method detail to treat the recalculation interval as a replacement for the sponsor’s interval.

In general, a confidence interval describes statistical uncertainty under the analysis assumptions; it does not capture every source of uncertainty. It does not correct for possible bias, incomplete outcome data, open-label conduct, or multiple comparisons unless the analysis specifically accounts for those issues. The registry record does not report whether the primary interval was adjusted for other outcomes or interim analyses. The width and placement of the interval are therefore best interpreted as precision for the reported primary comparison under the stated method, not as a guarantee that the result generalizes unchanged to other settings or endpoints.

Precision also depends on the number of participants with outcome data for the specific analysis, not enrollment alone. The endpoint-specific denominators are 117 for investigator-assigned treatment and 235 for maribavir. Other outcomes have different denominators, so their precision may differ. The registry record does not report the full flow of missing outcome observations or the procedures used to address them.

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. The registry record does not report the original sample-size justification or its design assumptions. The posted illustration assumes a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It uses the stated fraction of the observed effect as the effect to detect. The figures are participants per arm for the binary outcomes in this comparison, not total events.

The 80% power column describes the planning number of participants per arm under which a study would have an 80% chance of detecting the specified effect if the assumptions and effect were correct. The 90% power column applies the same idea with a higher chance of detecting that effect, and consequently generally requires more participants. These are design-planning concepts, not guarantees about what a future study will observe.

The rows describe planning for 100%, 75%, and 50% of the observed effect. For the primary endpoint, the figures are 36 and 48 participants per arm for 80% and 90% power when planning for 100% of the observed effect; 63 and 83 when planning for 75%; and 134 and 179 when planning for 50%. The smaller the effect a study aims to detect, the more participants it generally needs, all else equal. This is not observed power.

These calculations are illustrations based on the observed result and specified assumptions. They do not establish what the original investigators expected, what effect they considered important, or how they planned the completed trial. The registry record does not report those original design decisions. Differences between the illustration and actual enrollment should not be treated as evidence of a design flaw or as a retrospective evaluation of trial adequacy.

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 = 0.239; treatment proportion p₁ = 0.557; average p̄ = 0.398.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6923 = 1.3570.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6548 = 0.5511.
4. n = (1.3570 + 0.5511)² / (0.3181)² = 36.0, rounded up to 36 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 <- 0.239316239316239                          # control proportion
p1 <- 0.557446808510638                          # 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(2931539)
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 = 36 (80% power), 48 (90% power)
 75% of observed effect: n per arm = 63 (80% power), 83 (90% power)
 50% of observed effect: n per arm = 134 (80% power), 179 (90% power)
power.prop.test: 36 
Simulated power at n = 36 per arm: 0.803 (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 registry’s own primary analysis is the Cochran-Mantel-Haenszel comparison: a difference in percentage of responders of 32.8, with a 95% confidence interval from 22.80 to 42.74 and p < 0.001. This is the main reported inferential result. The registry also permits a simple unadjusted recalculation from the posted counts, giving a risk difference of 0.318, with an interval from 0.218 to 0.418 and p < 0.001.

The recalculation is a check based on summary proportions, not a substitute for the sponsor’s analysis. It does not reproduce the Cochran-Mantel-Haenszel method or any analysis details that are not reported. The registry’s discrepancy indicator is false, and both analyses give the same broad statistical reading: the posted primary comparison favors a higher clearance proportion in the maribavir group. The sponsor’s analysis takes precedence when a discrepancy is flagged; no such discrepancy is flagged in this record.

For interpretation, keep the measures and methods distinct. The sponsor reports a difference in percentage of responders, while the simple recalculation reports a risk difference using the posted summary counts. The confidence intervals and p-values belong to their respective analyses. The registry record does not report enough detail to reconstruct every aspect of the sponsor’s method independently, so the sponsor-reported analysis should remain the reference result.

15. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a multiplicity adjustment for the primary and secondary outcome analyses. It lists 25 secondary outcomes, alongside the primary endpoint, and the posted results include multiple comparisons and timeframes. When many statistical tests are conducted, the chance of at least one small p-value can increase even if there is no real effect for the corresponding outcome. A prespecified hierarchy or adjustment can address this issue, but no such procedure should be assumed from the posted information.

The record also does not report whether interim analyses were planned or conducted, whether any stopping boundaries were used, or whether the primary endpoint was the only confirmatory test. The registry identifies the reported primary analysis as a superiority analysis, but the detailed statistical analysis plan and its version history are not reported. These omissions limit assessment of how the endpoint and analysis were selected and how secondary findings should be weighted.

The primary outcome definition identifies a randomized set and says that participants were considered regardless of discontinuation before 8 weeks, provided they had not received alternative anti-CMV treatment. The registry record does not provide a full account of missing laboratory assessments, exclusions after randomization, imputation, sensitivity analyses, or how alternative treatment was handled in every analysis. It also does not report a complete participant flow for the primary endpoint. Those details matter because the observed response proportion depends on who contributes outcome information and how unobserved outcomes are treated.

The listed recurrence outcomes include analyses restricted to participants who completed 8 weeks of study-assigned treatment. Restriction based on post-randomization treatment completion can change the group being compared, because completion may be related to participant condition, tolerability, or other factors. This is a general statistical concern; the registry record does not report the reasons for noncompletion or methods to address selection in these analyses.

The endpoint-specific analysis population, the missing-data approach, the multiplicity strategy, and interim-analysis details are not reported in the registry record. Their absence does not prove that no such procedures existed, but it means they cannot be evaluated from this record. The primary reported comparison and secondary findings should be interpreted with that uncertainty in view.

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: vomiting (16.7%), cytomegalovirus infection (16.7%), spinal fracture (16.7%), hypokalaemia (16.7%), transplant rejection (14.3%).

The registry reports serious adverse events and deaths by adverse-event group. In the investigator-assigned groups, serious adverse events affected 33 participants among 56 at risk in the ganciclovir/valganciclovir group, 26 among 47 in the foscarnet group, 3 among 6 in the cidofovir group, and 1 among 7 in the foscarnet plus ganciclovir/valganciclovir group. For maribavir 400 mg, the registry reports 131 participants affected among 234 at risk. The maribavir rescue arm is also listed, with 14 affected among 22 at risk. These are the adverse-event group figures, not the primary endpoint denominators.

The record lists deaths of 6, 7, 0, 0, 27, and 0 for those respective groups. These counts alone do not establish that treatment caused an event or death. The registry also lists serious-event terms including vomiting, cytomegalovirus infection, spinal fracture, hypokalaemia, transplant rejection, and acute kidney injury, with the displayed percentages attached to those terms. The record’s adverse-event summary does not by itself provide enough context to compare causality, duration, severity, or background risk across treatment groups.

Safety counts should be read alongside the study’s group definitions and populations. The investigator-assigned arm contains several drug-specific subgroups, while the primary efficacy comparison is between randomized treatment strategies. The registry record does not provide a full causal assessment or a detailed account of exposure-adjusted safety analyses. These figures describe reported events in the listed groups; they are not, on their own, a complete comparative safety conclusion.

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

The posted primary result shows a higher observed percentage of participants meeting the confirmed Week 8 CMV viremia-clearance definition in the maribavir group than in the investigator-assigned treatment group. The reported Cochran-Mantel-Haenszel analysis also indicates a statistically detectable difference under its stated method. The confidence interval gives a range of estimates compatible with the data and method, while the p-value indicates evidence against the no-difference hypothesis.

The result does not mean that every participant cleared CMV, that every person would have the same response, or that the observed group difference applies to all transplant recipients. The enrolled population was restricted to transplant recipients with CMV infection refractory or resistant to specified prior anti-CMV treatments. The comparator was investigator-selected and could involve several different agents and dosing regimens. The record does not establish a separate randomized comparison against each agent.

The primary endpoint is a laboratory measure at Week 8, with a confirmation requirement. It is not equivalent to symptom resolution, continued response after treatment, recurrence prevention, survival, or overall safety. Those are distinct questions, some of which are addressed by listed secondary outcomes. The secondary results are subject to their own definitions, denominators, and uncertainties, and the registry record does not report a multiplicity adjustment.

The open-label design, unreported details of missing-data handling, and incomplete reporting of the statistical plan limit how fully the result can be assessed. The numerical findings describe the posted analysis; they do not resolve every question about bias, generalizability, or treatment effects outside this study. Statistical significance should not be confused with clinical importance, and the registry record does not report a threshold for a clinically meaningful difference.

Accordingly, the most direct interpretation is confined to the randomized comparison, the defined Week 8 clearance endpoint, and the analysis reported in the registry. It should not be extended to unreported populations, outcomes, or follow-up periods, and it does not amount to medical advice.

18. Why This Trial Matters Statistically

This trial raises a useful statistical distinction between a treatment product and a treatment strategy. Maribavir was administered under a fixed regimen, while the active comparator allowed investigators to select among several anti-CMV agents and use prescribed dosing. The randomized comparison therefore estimates a contrast between maribavir and an investigator-assigned strategy as used in the study, rather than a set of separate head-to-head estimates against each comparator drug.

The primary endpoint also illustrates why outcome definitions matter. Confirmed plasma CMV DNA clearance requires a threshold result in consecutive samples separated by a specified interval. That definition is more precise than a general statement that CMV improved, but it answers a narrower question than symptom control, durable response, recurrence, or safety. Statistical interpretation depends on keeping the endpoint’s operational definition aligned with the claim being discussed.

The registry provides both a sponsor-reported Cochran-Mantel-Haenszel result and a simple recalculation from posted summary counts. Their distinction is educational: a transparent recalculation can help readers inspect the direction and scale of a result, but it does not reproduce a sponsor’s method when adjustment details are unavailable. The reported analysis remains the principal inferential result.

The secondary outcomes show why multiplicity and analysis populations matter. Outcomes across different timeframes and subsets can add clinical context, but they do not automatically constitute independent confirmation. A completion-based analysis may answer a different question from a randomized-set analysis, while multiple unadjusted tests require caution. The record’s omissions make those issues visible rather than allowing them to be resolved by assumption.

Finally, the sample-size illustration separates planning assumptions from the completed trial’s original design. Calculations based on observed effects can help readers understand how smaller effects require more participants under stated assumptions, but they cannot recover the original justification or establish observed power. The registry record supports that distinction and provides a focused example of how statistical evidence, endpoint scope, and reporting completeness shape interpretation.

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.

The active comparator was not one uniform drug regimen. Investigator-assigned treatment could involve ganciclovir, valganciclovir, foscarnet, or cidofovir, with the investigator selecting an agent and dosing regimen. Consequently, the primary comparison does not isolate the effect of maribavir relative to each individual drug. The record does not report treatment-specific randomized efficacy comparisons for each agent.

The study was open-label. The registry record does not report separate masking of laboratory personnel or endpoint assessors, so masking beyond the stated open-label design cannot be assumed. The primary endpoint is based on laboratory measurements, but its interpretation still depends on the outcome assessment process and available samples.

The primary endpoint has a restricted interpretation: confirmed plasma DNA clearance at Week 8 under a definition that excludes participants who received alternative anti-CMV treatment. The registry record does not provide the full missing-data procedure, a complete participant flow for outcome assessment, or sensitivity analyses addressing unavailable measurements. It also does not report the detailed handling of treatment changes and other intercurrent events beyond the endpoint wording.

Several listed secondary outcomes concern maintenance, symptoms, or recurrence. Some reported recurrence analyses are limited to participants who completed 8 weeks of treatment, which is a post-randomization subset rather than the full randomized comparison. The reasons for completion or noncompletion and any methods addressing selection are not reported. The registry record also does not report a multiplicity adjustment, the detailed statistical analysis plan, interim-analysis information, or the original sample-size justification.

These limitations do not negate the posted primary result, but they define its scope. The available record supports a comparison of the stated randomized groups on the stated Week 8 endpoint. It gives less basis for detailed conclusions about individual comparator drugs, missing outcomes, confirmatory status of secondary findings, or the original design rationale.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety Study of Maribavir Treatment Compared to Investigator-assigned Treatment in Transplant Recipients With Cytomegalovirus (CMV) Infections That Are Refractory or Resistant to Treatment With Ganciclovir, Valganciclovir, Foscarnet, or Cidofovir — NCT02931539.
• Registry record last updated 2021-11-03; status last verified 2021-09.
• Results first posted 2021-09-23.
• 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

NCT02931539 is a completed, randomized, open-label, parallel-group Phase 3 study of maribavir versus investigator-assigned anti-CMV treatment in transplant recipients with CMV infection refractory or resistant to prior treatment. Enrollment was 352. The primary endpoint was confirmed CMV viremia clearance at Week 8, defined using plasma CMV DNA below the stated assay threshold in consecutive samples.

The registry reports clearance in 55.7% of the maribavir group and 23.9% of the investigator-assigned treatment group. Its Cochran-Mantel-Haenszel analysis reports a difference in percentage of responders of 32.8, a 95% confidence interval from 22.80 to 42.74, and p < 0.001. A simple unadjusted recalculation gives a risk difference of 0.318, with an interval from 0.218 to 0.418 and p < 0.001.

Secondary outcomes address response maintenance, symptoms, and recurrence. The registry record does not report the original sample-size justification, detailed missing-data procedures, or a multiplicity adjustment. The results should be interpreted in light of the open-label design, the flexible active comparator, and the distinction between the primary laboratory endpoint and other clinical outcomes.