Anamorelin HCl in the Treatment of Non-Small Cell Lung Cancer-Cachexia (NSCLC-C): An Extension Study (ROMANA 3): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Anamorelin HCl in the Treatment of Non-Small Cell Lung Cancer-Cachexia (NSCLC-C): An Extension Study (ROMANA 3) (NCT01395914), a randomized phase 3 study. The primary outcome is Percentage of Participants With Treatment-emergent Adverse Events.

2. Clinical Question

This is an extension-study question, rather than a description of the full treatment history of the original trial. The secondary outcome timeframes refer to change from baseline of the original trial through Week 12 of the extension. The record does not give further detail about how participants entered the extension, their treatment exposure before it, or the criteria used to define treatment-emergent events; those details are not reported in the registry record.

3. Trial Design

ROMANA 3 is registered as a phase 3, randomized, parallel-group extension study. The masking designation is quadruple. Its listed interventions were anamorelin HCl and placebo, each given orally once daily. The active-arm description specifies a 100 mg tablet; the placebo tablets were described as identical in appearance to the active tablets. The study status is completed.

The registry lists enrollment of 513. That figure describes the registered study enrollment and should not be substituted for the endpoint-specific denominators. For the primary safety result, the record gives 343 participants at risk in the anamorelin HCl arm and 167 in the placebo arm. The secondary outcome summaries use 345 and 168, respectively. These differences in denominators are visible in the posted results; the registry record does not explain them.

The primary completion date is 2014-06 and the study start is 2011-07. The record identifies Helsinn Therapeutics (U.S.), Inc as the sponsor. Its stated primary endpoint covers the 12-week treatment period. The secondary outcomes measure change from baseline of the original trial through Week 12 of the extension, so they refer to a different measurement frame from the adverse-event endpoint.

The record does not report the trial's original sample-size justification, the allocation ratio used in the design, the detailed randomization procedure, or the full analysis plan. The planning illustrations associated with the posted comparisons use equal allocation as a stated assumption, but that is not evidence of the allocation ratio used in the trial. Details about eligibility criteria, participant flow, treatment adherence, and discontinuations 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: The primary purpose of this extension study was to permit patients who completed dosing in the original 12-week trials to have the option of continuing to receive randomized study drug for an additional 12 weeks, to further evaluate the safety and tolerability of anamorelin HCl.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety Population, defined as patients who received any extension trial study drug.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: The intent-to-treat (ITT) Population included all randomized patients.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Treatment-emergent Adverse Events, measured over Over the 12-week treatment period.

The primary endpoint is the percentage of participants with treatment-emergent adverse events over the 12-week treatment period. It is a binary outcome at the participant level: a participant is counted as having an event or not having one for the purpose of this percentage. The registry defines the primary safety population as patients who received any extension trial study drug. The endpoint description says it was used to evaluate safety and tolerability.

Summarizing the outcome as a percentage allows the event occurrence to be compared across arms despite different numbers of participants at risk. The accompanying risk difference expresses the contrast in the arm-level event proportions. It does not describe the number, duration, severity, or clinical consequences of events for each participant. Nor does the single overall percentage distinguish among types of events. Separate serious adverse-event information and listed serious terms provide additional context, but do not replace the primary endpoint.

The registry does not report the detailed operational definition of treatment-emergent, how repeated events in one participant were handled, or how events were attributed or coded. Those definitions matter when interpreting a safety percentage and are not reported in the registry record. The endpoint therefore describes the posted event proportion within the stated population and timeframe, rather than a complete account of every aspect of safety.

7. Results Posted in the Registry

• Anamorelin HCl: 51.9% (178/343) versus 56.3% (94/167) on control; risk difference -4.4 percentage points (95% CI -13.6 to 4.8), risk ratio 0.92 (0.78–1.09); unadjusted two-proportion z-test p = 0.351.

8. Secondary Efficacy Endpoints

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

Change in Body Weight

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Anamorelin HCl: mean 3.06 versus 0.92 on control; difference 2.14 (95% CI 0.30 to 3.98), standardized effect d = 0.20; z-test p = 0.023.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Anamorelin HCl than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change in Handgrip Strength of the Non-Dominant Hand

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Anamorelin HCl: mean -0.83 versus -0.55 on control; difference -0.28 (95% CI -3.01 to 2.45), standardized effect d = -0.02; z-test p = 0.841.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Anamorelin HCl and the control. The interval gives the range of effects that remain compatible with the data.

Change in A/CS Domain Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Anamorelin HCl: mean 4.46 versus 3.24 on control; difference 1.22 (95% CI -1.46 to 3.90), standardized effect d = 0.08; z-test p = 0.372.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Anamorelin HCl and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes broaden the picture beyond treatment-emergent adverse events by describing change in body weight, handgrip strength of the non-dominant hand, and A/CS domain score. They differ in their measurement scales and in what they represent. Their estimates therefore complement the safety endpoint rather than serving as interchangeable measures of one common outcome.

Across these outcomes, the posted pattern is not uniform. The body-weight comparison has a positive estimated mean difference with an interval that excludes zero, while the handgrip-strength and A/CS comparisons have intervals that include zero. This pattern does not determine the primary safety result, because the outcomes address different questions. Nor does a result on one secondary measure establish a general effect across the other measures.

The secondary comparisons are unadjusted, and the registry lists no reported analyses for them. When several outcomes are examined, the chance of at least one apparently notable result can increase, even if the outcomes are not independent. A multiplicity adjustment or a prespecified hierarchy can help control how evidence is interpreted across multiple tests, but no such procedure is reported in the registry record. Accordingly, an unadjusted secondary p-value carries less confirmatory weight than a result supported by a documented analysis plan that addresses the family of outcomes.

The secondary timeframes refer to change from baseline of the original trial through Week 12 of this extension, not simply change measured from an extension-only baseline as far as the registry description states. The record does not provide further detail about the modeling, missing observations, or how the secondary outcomes were selected for analysis. Those omissions limit interpretation of the apparent pattern, so the secondary results are best treated as descriptive evidence requiring context rather than as definitive conclusions.

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

• Change in Body Weight
• Change in Handgrip Strength of the Non-Dominant Hand
• Change in A/CS Domain Score

9. Statistical Interpretation

10. Statistical Methods Explained

For the primary outcome, each arm's result is a percentage of participants with at least one treatment-emergent adverse event as represented in the registry outcome. The posted arm summaries give 52.0% for anamorelin HCl and 56.0% for placebo. A percentage communicates the proportion affected within the relevant arm; it is not a count of every event, because one participant may have more than one event. The record does not report whether the endpoint's event definition uses any event, a first event, or another detailed counting rule.

What does the risk difference express?

A risk difference subtracts one arm's event proportion from the other arm's proportion. The posted risk difference is -0.044 for anamorelin HCl relative to placebo. Its negative direction indicates a lower observed proportion in the anamorelin HCl arm under that comparison's ordering. The estimate is not a relative reduction, and it should not be read as a count of participants. Unlike a ratio measure, a risk difference remains on the scale of an absolute difference in proportions.

The registry also posts relative measures for this comparison: the risk ratio is 0.922, with an interval from 0.779 to 1.091, and the odds ratio is 0.838, with an interval from 0.578 to 1.215. A risk ratio compares proportions as a ratio; an odds ratio compares the odds of the event. These measures answer different mathematical questions and may look different from an absolute difference. Their intervals include the no-association value for a ratio, while the risk-difference interval includes zero. These measures do not remove the uncertainty evident in the posted comparison.

How should a p-value be understood?

The unadjusted p-value for the primary comparison is 0.351. In general, a p-value measures how unusual data at least as discrepant as those observed would be under a specified null model and the assumptions of the analysis. It is not the probability that the null model is true, nor the probability that the result arose by chance. The registry does not report a detailed statistical analysis plan or identify any adjustment for covariates. The p-value should therefore be interpreted as an unadjusted comparison, not as a substitute for a complete account of the analysis.

For the continuous secondary outcomes, the registry reports least-squares means and standard errors, as well as a mean difference for each comparison. A least-squares mean is an estimated mean from a statistical model; it is not necessarily the same as a raw arithmetic average. The record does not describe the model or its adjustment variables. A standard error describes the estimated precision of a mean, rather than the spread of individual participant measurements. The comparisons list approximate pooled standard deviations for planning illustrations, but these are not direct measures of the standard error.

Why does over-dispersion matter?

As a general statistical point, over-dispersion occurs when observed variability is greater than a model's assumptions allow. For count outcomes, for example, a model that assumes a particular relationship between the mean and variance may understate uncertainty when the data are more variable than expected. That can make intervals too narrow and tests too optimistic. This trial's primary endpoint is presented as a participant percentage, not as an event count model, and the registry does not report a method addressing over-dispersion. It would be inappropriate to assume that such a method was used.

For the continuous outcomes, the posted standard errors are useful for understanding uncertainty in the arm means. The registry's comparison intervals give uncertainty for the estimated mean differences. Since the record does not report the model specifications, covariance structure, or missing-data approach, readers cannot fully reconstruct how those intervals were obtained. General statistical explanations help identify what the measures mean, but they do not establish which unreported procedures were used in this study.

What do the planning illustrations estimate?

The sample-size illustrations are based on the observed effects, not on the trial's original design. For continuous outcomes, the stated assumptions include a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation; the effect to detect is the stated fraction of the observed effect. The values describe participants per arm. They are planning illustrations and are not observed power. The original trial's sample-size justification and design assumptions are not reported in the registry record.

Planning calculations based on an observed effect inherit uncertainty from that observed estimate. They do not verify that an effect is reproducible, establish an appropriate target effect, or account for analysis features absent from the registry record. The calculations should be used as illustrations of how assumptions affect planning, not as retrospective evidence about the success or failure of the completed study.

11. Confidence Intervals and Precision

The primary risk-difference interval runs from -0.136 to 0.048 around the estimate of -0.044. It includes zero, the value corresponding to no difference in proportions. Its span indicates that the posted estimate is compatible with a range of possible contrasts, including directions opposite to the point estimate. The result is therefore not precise enough to settle the direction of the difference on its own. A confidence interval describes uncertainty under the statistical method and its assumptions; it is not a guarantee that the true value lies in the interval.

The risk-ratio interval, from 0.779 to 1.091, and odds-ratio interval, from 0.578 to 1.215, likewise include their respective no-association values. Ratios and differences use distinct scales, so their numerical endpoints should not be directly compared as though they represented the same quantity. The registry does not state the confidence level in the outcome record, so no level is assigned in this explanation.

The secondary mean-difference intervals also show different degrees and patterns of uncertainty. The body-weight interval runs from 0.297 to 3.983, whereas the handgrip-strength interval runs from -3.007 to 2.447 and the A/CS domain score interval from -1.458 to 3.898. The latter intervals include zero; the body-weight interval does not. These intervals apply to different outcome scales and should be interpreted separately. The record does not report the complete model details needed to assess every source of uncertainty, including missing-data handling and any adjustment strategy.

12. Planning a Similar Trial: Sample Size

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

The planning figures are participants per arm for the continuous-outcome illustrations; the assumptions specifically state that non-time-to-event tables give participants per arm. Time-to-event tables, if present, would give total events, but the posted outcomes in this record are not time-to-event outcomes. The figures are based on a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect equal to the stated fraction of the observed effect. These are planning assumptions, not the trial's original design assumptions, which are not reported in the registry record.

The 80% power column represents the planned sample size per arm under which the calculation would have an 80% probability of detecting the specified effect under its assumptions. The 90% power column represents the corresponding planning target for a 90% probability. Power is conditional on the assumed effect and analysis conditions; it is not a guarantee about the result of a future study. The original study's design justification is not reported in the registry record.

Rows for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect illustrate how the target effect changes the required sample size. A smaller effect requires more participants under these assumptions, because a subtler signal is harder to distinguish from variability. The curve should be read as a visual expression of this planning relationship, not as a prediction of what a future trial will observe. These values are not observed power.

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 = 0.563; treatment proportion p₁ = 0.519; average p̄ = 0.541.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7047 = 1.3813.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.7041 = 0.5925.
4. n = (1.3813 + 0.5925)² / (-0.0439)² = 2,019.3, rounded up to 2,020 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 <- 0.562874251497006                          # control proportion
p1 <- 0.518950437317784                          # 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(1395914)
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 = 2,020 (80% power), 2,703 (90% power)
 75% of observed effect: n per arm = 3,584 (80% power), 4,798 (90% power)
 50% of observed effect: n per arm = 8,048 (80% power), 10,774 (90% power)
power.prop.test: 2019.3 
Simulated power at n = 2,020 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. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes, comparisons, or analyses. Even when each individual test uses a conventional threshold, examining a collection of tests increases the possibility that at least one result will look unusual by chance. This is a general statistical concern, not a claim that any particular posted result is false. The registry lists three secondary outcomes in addition to the primary safety outcome, and no multiplicity adjustment or testing hierarchy is reported in the registry record.

The secondary comparisons are reported with unadjusted p-values. Without a documented plan for how those outcomes were prioritized and interpreted together, the p-values should not be treated as if each were the sole prespecified test. The record reports no analyses under the secondary outcomes, and no formal analysis details are available for evaluating the relationship between the posted comparisons and an original statistical plan.

Missing-data procedures are also not reported in the registry record. This matters because outcome availability can affect who contributes to an analysis and can influence estimates when missingness is related to outcomes or treatment. The primary safety population is defined as patients who received any extension trial study drug. For the secondary outcomes, the record gives arm-level sample sizes, but it does not provide a full analysis-population definition, participant flow, reasons for missing measurements, or an imputation method. No assumptions about those procedures should be inferred.

The primary comparison's p-value is described as unadjusted. The record does not report covariate adjustment, interim analyses, or a complete statistical plan. It also does not report the original sample-size justification. These gaps do not prove that such procedures were absent in the study itself; they mean the registry record does not document them. Readers should distinguish the posted numerical summaries from the additional methodological information needed to assess confirmatory evidence.

15. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: neoplasm progression (4.7%), death (1.8%), thrombocytopenia (1.2%), pneumonia (1.2%), anaemia (0.9%).

The serious adverse-event summary reports 44 of 343 participants at risk in the anamorelin HCl arm and 21 of 167 at risk in the placebo arm. The other-adverse-event summary reports 94 of 343 and 57 of 167, respectively. These figures describe participants affected in the listed categories, not necessarily the total number of individual event episodes. The registry record does not explain whether participants could contribute to both categories or provide the detailed event-counting rules.

The registry also lists serious adverse-event terms with percentages: neoplasm progression at 4.7%, death at 1.8%, thrombocytopenia at 1.2%, pneumonia at 1.2%, anaemia at 0.9%, and cardiopulmonary failure at 0.6%. The record does not clarify the denominator or arm distribution for these term percentages. They should therefore be read as listed in the registry, not assigned to a particular arm or interpreted as rates of unique participants without further information.

Arm-specific death counts are not reported in the registry record. The serious-event term list includes death, but that does not supply the missing arm-specific counts. These summaries provide a descriptive account of reported events; they do not by themselves establish comparative safety or explain the clinical course of participants.

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

The posted results describe event proportions and other measured outcomes in a randomized, masked extension study. For the primary safety endpoint, the observed event percentage was lower in the anamorelin HCl arm than in the placebo arm, but the risk-difference interval includes zero and the unadjusted p-value is 0.351. This is evidence with uncertainty, not proof that the arms are equivalent, and not a basis for saying that either treatment is safe or unsafe.

The secondary outcomes add information about measured changes in body weight, handgrip strength, and A/CS domain score. Their pattern differs across measures, and each is subject to the limitations of unadjusted comparisons and missing analysis details. The positive body-weight estimate should not be used to infer a broad effect across all outcomes; similarly, intervals including zero do not establish that there is no effect. Estimates and intervals describe the data under the methods used, with their limitations.

A registry result is not a full report of the study conduct or analysis. The record does not report detailed event definitions, missing-data methods, a multiplicity strategy, the original sample-size justification, or full model specifications. It also does not provide arm-specific death counts. These absences restrict what can be concluded from the posted summaries and make it important to distinguish measured results from unreported aspects of the trial.

The study's randomized and quadruple-masked design are relevant features, but they do not eliminate uncertainty or substitute for complete reporting. Statistical results alone do not provide medical advice, and this explanation does not determine an individual treatment choice. It describes the registry's posted outcomes and the precision and limitations of the associated comparisons.

17. Why This Trial Matters Statistically

This study illustrates why safety outcomes and other clinical measures need to be interpreted on their own scales and with their own denominators. The primary endpoint is a percentage of participants with treatment-emergent adverse events, while the secondary outcomes are model-based estimates of change in continuous measures. A single narrative of benefit or harm would obscure these distinctions.

It also shows the importance of reading an estimate together with its interval. A point estimate gives a central summary, but the interval conveys how uncertain that summary is under the analysis. In the primary comparison, the interval includes zero, so the observed direction does not by itself resolve the possible direction of the underlying difference. The same principle applies to the secondary measures, whose intervals differ in relation to zero.

The record further highlights how much interpretation depends on analysis reporting. Definitions of outcomes, handling of missing observations, multiplicity plans, and model specifications help readers judge whether comparisons are exploratory or confirmatory and how robust they may be. When those details are not reported in the registry record, the numerical summaries remain useful, but they cannot answer every methodological question.

Finally, planning illustrations based on observed effects can show the relationship between target effect and required sample size, but they do not recover the original design rationale or establish post hoc power. Keeping planning calculations separate from observed results is a central statistical distinction. Together, these points make the trial useful for examining how estimates, uncertainty, safety summaries, and reporting gaps fit into a careful reading of clinical evidence.

18. 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 primary safety comparison is unadjusted in the posted results, and the registry record does not report a detailed analysis plan or covariate-adjustment strategy. The comparison's interval and p-value therefore cannot be assessed against a complete description of the intended analysis. The record also does not report how treatment-emergent events were defined, classified, or counted when participants had repeated events.

Endpoint-specific denominators differ between the primary safety results and secondary outcomes, while the registry does not explain the differences. The secondary outcomes are least-squares means, but model specifications and adjustment variables are not reported. Missing-data handling, reasons for unavailable measurements, and participant flow are also not reported in the registry record. These omissions limit assessment of potential selection and analysis effects.

The record lists several outcomes but does not report a multiplicity procedure or testing hierarchy. The secondary p-values are unadjusted, which reduces their value as confirmatory evidence when considered across multiple outcomes. The registry also does not report the original sample-size justification, interim analyses, or the trial's full statistical plan. Planning calculations based on the observed effects should not be mistaken for the study's own design assumptions.

For serious adverse-event terms, percentages are listed without a stated denominator or arm distribution. Arm-specific death counts are not reported in the registry record, although death appears among the listed serious terms. These details prevent a more complete comparison of individual serious events. The limitations reflect information not documented in the registry record; they do not establish that unreported methods or analyses were not conducted.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Anamorelin HCl in the Treatment of Non-Small Cell Lung Cancer-Cachexia (NSCLC-C): An Extension Study (ROMANA 3) — NCT01395914.
• Registry record last updated 2017-09-14; status last verified 2017-08.
• Results first posted 2017-07-11.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

ROMANA 3 is a completed phase 3 randomized, quadruple-masked, parallel-group extension study of anamorelin HCl versus placebo in non-small cell lung cancer-cachexia. The primary endpoint was the percentage of participants with treatment-emergent adverse events over the 12-week treatment period. The registry reports arm-level percentages, a risk difference with an interval, and an unadjusted p-value.

Posted secondary outcomes cover change in body weight, non-dominant handgrip strength, and A/CS domain score through Week 12 of the extension, measured from baseline of the original trial. The results show differing patterns across these outcomes. The record does not report a complete statistical plan, multiplicity procedure, missing-data methods, original sample-size justification, or arm-specific death counts. The comparisons should therefore be read as posted statistical summaries with material uncertainty and reporting limitations.