A Study of LY2484595 in Patients With High LDL-C or Low HDL-C: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of LY2484595 in Patients With High LDL-C or Low HDL-C (NCT01105975), a randomized phase 2 study. The primary outcome is Percent Change From Baseline to 12 Weeks Endpoint in High Density Lipoprotein Cholesterol (HDL-C) With LY2484595 in Combination With Atorvastatin and Atorvastatin Monotherapy.

2. Clinical Question

The trial also addressed broader questions about LY2484595 alone, its use with other statins, statin pharmacokinetics, CETP measures, and adverse events. Those objectives provide context, but they do not replace the prespecified primary comparison. The record does not report a clinical outcome such as cardiovascular events as the primary endpoint, so the result is specifically about a short-term change in a blood lipid measure.

3. Trial Design

This was a randomized, quadruple-masked, parallel-group Phase 2 study sponsored by Eli Lilly and Company. The registry lists an enrollment of 398 and a completed status. The study started in 2010-04, and primary completion was recorded as 2011-06. The listed condition is dyslipidemia, and the title describes patients with high LDL-C or low HDL-C.

The treatment groups covered several study questions: LY2484595 monotherapy at 30 mg, 100 mg, or 500 mg; placebo; statin monotherapy; and LY2484595 combined with atorvastatin, simvastatin, or rosuvastatin. The primary endpoint specifically contrasts 100 mg LY2484595 plus 20 mg atorvastatin with 20 mg atorvastatin monotherapy. The primary outcome analysis includes participants with both baseline and Week 12 HDL-C measurements; its group sizes are 31 and 34, respectively.

Random allocation can help balance measured and unmeasured participant characteristics across groups on average, while masking can reduce the influence of treatment knowledge on care or outcome assessment. These design features do not, by themselves, establish how allocation concealment was implemented or how masking was maintained; those details are not reported in the registry record. The record also does not report the randomization method, the original sample-size justification, planned interim analyses, or the trial’s complete statistical analysis plan.

The parallel-group structure means the primary comparison is between separate treatment groups, rather than a within-participant crossover. The endpoint is reported at Week 12 relative to baseline. The registry describes a mixed model repeated measures analysis for the sponsor’s primary comparison, while the displayed group summaries are adjusted least-squares means. Details of follow-up beyond the stated outcome time frame are not reported for this primary efficacy result.

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.

Only the first 6 of 10 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants who had both baseline and Week 12 HDL-C measurements.

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.

Only the first 6 of 11 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline to 12 Weeks Endpoint in High Density Lipoprotein Cholesterol (HDL-C) With LY2484595 in Combination With Atorvastatin and Atorvastatin Monotherapy, measured over Baseline, Week 12.

The primary endpoint is percent change from baseline to the Week 12 endpoint in high-density lipoprotein cholesterol (HDL-C), comparing LY2484595 combined with atorvastatin against atorvastatin monotherapy. HDL-C is a continuous laboratory measurement, and expressing change as a percentage places the endpoint in terms of relative change from each participant’s baseline value rather than an absolute change in mg/dL. The registry labels its unit as percent change of mg/dL.

For the primary analysis, the registry reports least-squares mean values controlled for region, baseline measurement, treatment, visit, and treatment-by-visit interaction. Adjusted means can account for specified model terms, making the treatment comparison conditional on the analysis model rather than simply a difference between raw group averages. The analysis population consisted of participants who had both baseline and Week 12 HDL-C measurements.

HDL-C was used as a lipid-profile measure, consistent with the study’s stated question about whether adding LY2484595 to a statin changes blood fats. It is a surrogate laboratory endpoint, not a direct measure of symptoms, cardiovascular events, or long-term health outcomes. The registry record does not report whether the observed HDL-C change was linked to clinical outcomes, and the primary endpoint alone cannot answer that question.

7. Results Posted in the Registry

• 100 mg LY2484595 + 20 mg Atorvastatin: mean 79.90 versus 1.40 on control; difference 78.50 (95% CI 62.28 to 94.72), standardized effect d = 2.36; 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.

Percent Change From Baseline to 12 Weeks Endpoint in High Density Lipoprotein Cholesterol (HDL-C) With LY2484595 and Placebo

• 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:
  • 30 mg LY2484595 Monotherapy: mean 53.60 versus -3.00 on control; difference 56.60 (95% CI 41.08 to 72.12), standardized effect d = 1.70; z-test p = <0.001.
  • 100 mg LY2484595 Monotherapy: mean 94.60 versus -3.00 on control; difference 97.60 (95% CI 81.94 to 113.26), standardized effect d = 2.90; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for 30 mg LY2484595 Monotherapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for 100 mg LY2484595 Monotherapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent Change From Baseline to 12 Weeks Endpoint in Low Density Lipoprotein Cholesterol (LDL-C) With LY2484595 and Placebo

• 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:
  • 30 mg LY2484595 Monotherapy: mean -13.60 versus 3.90 on control; difference -17.50 (95% CI -25.82 to -9.18), standardized effect d = -0.98; z-test p = <0.001.
  • 100 mg LY2484595 Monotherapy: mean -22.30 versus 3.90 on control; difference -26.20 (95% CI -34.52 to -17.88), standardized effect d = -1.47; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for 30 mg LY2484595 Monotherapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for 100 mg LY2484595 Monotherapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent Change From Baseline to 12 Weeks Endpoint in High Density Lipoprotein Cholesterol (HDL-C) With LY2484595 in Combination With Simvastatin or Rosuvastatin and Simvastatin/Rosuvastatin Monotherapy

• 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:
  • 100 mg LY2484595 + 40 mg Simvastatin: mean 86.60 versus 7.30 on control; difference 79.30 (95% CI 63.78 to 94.82), standardized effect d = 2.41; z-test p = <0.001.
  • 10 mg Rosuvastatin Monotherapy: mean 5.50 versus 7.30 on control; difference -1.80 (95% CI -17.32 to 13.72), standardized effect d = -0.05; z-test p = 0.820.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for 100 mg LY2484595 + 40 mg Simvastatin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 10 mg Rosuvastatin Monotherapy and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes broaden the study’s scope beyond the primary combination-versus-atorvastatin comparison. They include HDL-C and LDL-C changes with LY2484595 monotherapy versus placebo, HDL-C and LDL-C comparisons involving LY2484595 with simvastatin or rosuvastatin, pharmacokinetic assessment of LY2484595, CETP activity and mass, and rash episodes. The posted secondary results cover only some of these outcome families, so their pattern should not be treated as a complete account of every registered secondary measure.

Across the posted lipid results, LY2484595 monotherapy comparisons with placebo show modeled changes in HDL-C and LDL-C, and the combination comparisons with simvastatin or rosuvastatin show differences in HDL-C relative to their respective statin-only groups. These outcome families are related to the primary question because they assess whether the study drug’s lipid-profile patterns extend beyond the atorvastatin comparison. They do not replace the primary endpoint or establish that findings across different drugs and comparisons are equivalent.

The secondary findings carry less inferential weight when interpreted individually because the registry record does not report a complete plan for controlling multiplicity across the many outcomes and comparisons. When numerous hypotheses are tested, the chance of obtaining apparently notable results by chance can increase. An unadjusted p-value addresses a particular comparison under its test assumptions; it does not account for the full set of analyses unless an adjustment is reported. The record does not report such a full adjustment strategy.

Accordingly, secondary results are best read as supportive or exploratory context rather than as independent confirmation of the primary result. The result for the primary HDL-C comparison is not evidence that all secondary outcomes changed, and a secondary result does not establish effects on clinical outcomes. The registry does not report clinical endpoint findings that would connect these laboratory measures to patient-important outcomes.

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

• Percent Change From Baseline to 12 Weeks Endpoint in High Density Lipoprotein Cholesterol (HDL-C) With LY2484595 and Placebo
• Percent Change From Baseline to 12 Weeks Endpoint in Low Density Lipoprotein Cholesterol (LDL-C) With LY2484595 and Placebo
• Percent Change From Baseline to 12 Weeks Endpoint in High Density Lipoprotein Cholesterol (HDL-C) With LY2484595 in Combination With Simvastatin or Rosuvastatin and Simvastatin/Rosuvastatin Monotherapy
• Percent Change From Baseline to 12 Weeks Endpoint in Low Density Lipoprotein Cholesterol (LDL-C) With LY2484595 in Combination With Simvastatin or Rosuvastatin and Simvastatin/Rosuvastatin Monotherapy
• Pharmacokinetics - LY2484595 Area Under the Concentration-Time Curve (AUC) at Steady-State
• Percent Change From Baseline to 12 Weeks Endpoint in Plasma Cholesteryl Ester Transfer Protein (CETP) Activity
• Percent Change From Baseline to 12 Weeks Endpoint in Plasma Cholesteryl Ester Transfer Protein (CETP) Mass
• The Number of Episodes of Rashes at Any Time From Baseline Through Week 12

9. Statistical Interpretation

10. Statistical Methods Explained

The registry reports a mixed model repeated measures (MMRM) analysis for the primary comparison. This model uses repeated outcome measurements over visits and can account for the correlation among measurements from the same participant. In general, repeated measurements from one person are not independent: a participant’s readings tend to be related over time. Modeling that structure can make the estimated treatment contrast and its uncertainty more appropriate than treating every reading as unrelated.

What does the adjusted mean represent?

The primary endpoint is reported as a least-squares mean, with adjustment for region, baseline measurement, treatment, visit, and treatment-by-visit interaction. In general, a least-squares mean is a model-based mean adjusted for the factors included in the analysis. The treatment-by-visit interaction allows the modeled treatment difference to vary across visits. The registry gives baseline and Week 12 as the endpoint timeframe, but does not report the full model specification, covariance structure, estimation method, or handling of all possible departures from model assumptions.

The primary analysis population is restricted to participants who had both baseline and Week 12 HDL-C measurements. This is the population described for the endpoint, not necessarily every randomized participant. The registry record does not report the number of participants missing the primary endpoint, reasons for missingness, or whether any imputation or sensitivity analyses were used. Missing-data assumptions matter because a model’s reliability depends partly on how the absence of measurements relates to observed and unobserved outcomes.

How should the mean difference be read?

The sponsor-reported mean difference in final values is 78.5, comparing combination therapy with atorvastatin monotherapy. It is a contrast on the reported percent-change scale. A positive value indicates that the modeled percent-change estimate was higher for the combination group in this comparison. The result is not a risk ratio or an odds ratio: it is a difference between continuous outcome estimates.

The unadjusted recalculation uses the posted group means and standard errors to produce a simple comparison. It gives a mean difference of 78.5, an interval from 62.28 to 94.72, and p < 0.001. Because the group standard errors are used to approximate the variability for that calculation, it is not the same analysis as the sponsor’s MMRM. In general, a recalculation from summary statistics cannot reproduce adjustments, repeated-measure handling, participant-level covariance, or other details that require the original analysis data.

What does a p-value tell the reader?

A p-value describes how compatible the observed data are with a specified null hypothesis under the assumptions of the test. Smaller p-values indicate greater statistical incompatibility with that null model, but they do not state the probability that the null is true. Nor do they describe the magnitude, precision, or clinical importance of an effect. The primary sponsor-reported p-value is < 0.001. The registry does not report whether the primary test was part of a broader multiplicity adjustment across outcomes or treatment comparisons.

Statistical significance should be considered alongside the estimated difference and its confidence interval. A small p-value can coexist with uncertainty about the magnitude that would be seen in another study. The interval communicates that uncertainty more directly, subject to the modeling assumptions. The posted secondary results include multiple lipid, pharmacokinetic, CETP, and safety outcomes, and the registry record does not report a complete multiplicity strategy for interpreting them together.

Why does over-dispersion matter?

Over-dispersion is a general concern when the variability in observed data is larger than a statistical model expects. For continuous outcomes, the related issue is whether the model adequately represents the outcome’s variability and correlation. If uncertainty is understated, intervals may be too narrow and p-values too small; if the model is unnecessarily conservative, the reverse can occur. The registry does not provide enough detail to assess these diagnostics for the primary MMRM. The simple recalculation’s approximate standard deviations are derived from standard errors and group sizes, and should be understood as an approximation rather than participant-level variance estimates.

Randomization supports a causal comparison in principle, but the analysis population’s requirement for observed baseline and Week 12 values means that interpretation also depends on missing-data processes and analysis choices. The registry does not report a full intention-to-treat analysis description or sensitivity analyses to evaluate robustness to missing outcomes. Those omissions limit independent evaluation of the modeling and analysis decisions; they do not change the values reported for the registered result.

11. Confidence Intervals and Precision

The sponsor-reported primary mean difference is 78.5, with a 90% confidence interval from 64.9 to 92.1. The interval is centered around the estimated contrast and conveys sampling uncertainty under the reported MMRM and its assumptions. Its bounds are both above zero, consistent with the reported p < 0.001 for a difference in the modeled endpoint. The interval does not guarantee that the true effect lies within those bounds, nor does it account for uncertainty from model choice, missing data, or unreported design decisions.

A confidence interval is a range generated by a procedure that, under its assumptions and repeated use, would cover the underlying parameter at the stated rate. In plain language, it communicates which effect sizes are reasonably compatible with the data and model, rather than giving only a single estimate. A narrower interval indicates greater statistical precision than a wider one, but precision is not the same as clinical importance or freedom from bias.

The simple recalculation gives a mean difference of 78.5 with an interval from 62.28 to 94.72. Its endpoints differ from the sponsor-reported 90% interval because the calculations use different approaches and uncertainty specifications. The recalculation is based on posted summary statistics and approximate variability; it is not a replication of the participant-level MMRM. The sponsor-reported model result is the relevant registered analysis for the primary endpoint.

Precision is also affected by the endpoint population: the primary outcome analysis included participants with both baseline and Week 12 HDL-C measurements. The registry does not report missing-outcome counts or sensitivity analyses, so the reported interval does not resolve uncertainty about the consequences of missing data. Nor does it indicate whether a difference of this size is clinically meaningful; no clinical-importance threshold is reported in the registry record.

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

The 80% power column gives the estimated participants per arm for a future comparison to have an 80% chance of detecting the specified effect under those assumptions, if that effect and the planning model are correct. The 90% power column gives the corresponding planning figure for a 90% chance. Power is a property of a study design under stated assumptions, not a guarantee that a future study will detect an effect.

Each planning row considers a different fraction of the observed effect: 100%, 75%, or 50%. Smaller target effects generally require more participants to distinguish them from random variation with the same planned power and significance level. This pattern is especially important when the observed estimate may be larger than the effect a future study should reasonably expect.

These figures are not observed power. They do not describe the probability that the completed trial detected its true effect, and they should not be read as evidence that a very small planning number would be adequate for a real study. Planning would also need to account for choices and uncertainties not captured by this illustration, including the original analysis model, missing outcome data, multiplicity, and the effect size considered scientifically important. The registry record does not report the trial’s own design assumptions.

13. How the Sample Size Is Derived

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

Formula

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

Worked calculation for this trial

1. Observed difference Δ = 78.500; pooled standard deviation σ = 33.318.
2. n = 2 × (1.96 + 0.842)² × 33.318² / 78.500² = 2.8, rounded up to 3 participants per arm for 80% power.

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

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 78.5                      # observed difference in means
sigma <- 33.3175629360852                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

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

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1105975)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 3 (80% power), 4 (90% power)
 75% of observed effect: n per arm = 6 (80% power), 7 (90% power)
 50% of observed effect: n per arm = 12 (80% power), 16 (90% power)
power.t.test: 4.1 
Simulated power at n = 3 per arm: 0.588 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The sponsor’s registered primary analysis used a mixed model repeated measures method and reported a mean difference in final values of 78.5, with a 90% confidence interval from 64.9 to 92.1 and p < 0.001. The registry also reports adjusted least-squares means of 1.4 for atorvastatin monotherapy and 79.9 for LY2484595 plus atorvastatin. These are the sponsor’s model-based results for the primary endpoint.

The simple recalculation from the posted summaries gives the same displayed mean difference, 78.5, and an interval from 62.28 to 94.72, with an unadjusted p < 0.001. Similarity in the central estimate is reassuring as a basic arithmetic check, but the recalculated uncertainty differs because it uses summary statistics and approximate variability rather than the sponsor’s repeated-measures model. It does not reproduce the model’s adjustments or its treatment of repeated measurements.

The registry marks no discrepancy between the posted sponsor result and the recalculation. Even so, these are not interchangeable analyses. The sponsor’s reported MMRM is the trial’s own analysis and should be used as the primary inferential result; the simpler calculation is supplementary context. The registry record does not provide sufficient model details or participant-level data to independently reproduce the sponsor’s result.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcome measures and reports analyses across several treatment comparisons. Testing several hypotheses creates a multiplicity problem: even if no real effects existed, some results could appear statistically notable by chance. Statistical approaches to address this can include prespecified hierarchical testing or adjustment of significance thresholds, but the appropriate method depends on the study’s analysis plan and objectives.

The registry record does not report a complete multiplicity strategy for the primary and secondary endpoints, nor does it say whether the displayed secondary p-values were adjusted for the number of analyses. The primary result should therefore be distinguished from the larger set of secondary comparisons. Unadjusted secondary p-values are not equivalent to family-wise evidence across all outcomes.

The primary analysis population is described as participants with baseline and Week 12 HDL-C measurements. The registry record does not report how many randomized participants lacked either measurement, the reasons for missing values, or the methods used to address missingness. It also does not report sensitivity analyses under alternative missing-data assumptions. These details matter because the observed-outcome population may differ from the full randomized population if missingness is related to treatment or outcome.

Other analysis-population details are not reported in the registry record, including whether the primary analysis followed an intention-to-treat principle, how protocol deviations were handled, or whether additional per-protocol or safety populations were used for efficacy analyses. The sponsor’s MMRM is identified, but the full specification and diagnostic information are not reported. These gaps prevent complete assessment of robustness and leave uncertainty about how analysis choices affect inference. They do not justify assuming that particular procedures were or were not used.

16. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: coronary artery disease (2.9%), influenza (2.6%), bursitis (2.5%), anal cancer (2.5%), metastases to lymph nodes (2.5%).

The registry reports serious adverse event counts by listed treatment group and separately includes follow-up groups. During the treatment groups, serious adverse events affected 0 participants in the 20 mg atorvastatin monotherapy group and 1 participant in the 100 mg LY2484595 plus 20 mg atorvastatin group. In the monotherapy and other statin groups, the listed serious-event counts range from 0 to 1 per group. These are descriptive counts, not a formal comparison of safety risks.

The registry also lists serious adverse event terms and percentages, including coronary artery disease at 2.9%, influenza at 2.6%, bursitis at 2.5%, anal cancer at 2.5%, metastases to lymph nodes at 2.5%, and fall at 2.4%. The record does not provide enough context in these entries to attribute these events to a particular treatment or to establish causation. Serious events may occur during a trial without being caused by the intervention.

Death values are null in the adverse-event data, so deaths are not reported in the registry record. Non-serious adverse-event counts are also listed by group, but they do not establish comparative safety without a specified analysis. The trial’s adverse-event data should be interpreted descriptively; the registry does not report a formal inferential safety comparison for the primary treatment contrast.

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

The primary result describes a difference in a short-term laboratory endpoint: the modeled percent change in HDL-C from baseline to Week 12 was higher in the combination group than in the atorvastatin-only group. The sponsor-reported mean difference is 78.5, with a 90% confidence interval from 64.9 to 92.1 and p < 0.001. This is evidence about the measured endpoint in the analyzed participants, under the reported model.

It does not establish that LY2484595 improves health or prevents disease. HDL-C change is not itself a clinical outcome, and the registry does not report a validated relationship between this trial’s endpoint difference and cardiovascular events or other patient-important outcomes. The record also does not report a clinical threshold that would define a meaningful change for participants.

The result is specific to 100 mg LY2484595 combined with 20 mg atorvastatin versus 20 mg atorvastatin monotherapy. It should not be generalized automatically to LY2484595 alone, other doses, combinations with simvastatin or rosuvastatin, or other populations. Each comparison has its own outcome estimates and uncertainty. The study is listed as Phase 2 and its primary endpoint timeframe is Week 12, so it does not establish longer-term effects.

Statistical significance does not settle questions about clinical significance, safety, durability, or generalizability. Nor does the primary finding answer the study’s pharmacokinetic or CETP questions. The broader evidence would need to consider the other endpoints, adverse events, missing data, and future research. Those questions are not resolved by the primary HDL-C comparison alone.

18. Why This Trial Matters Statistically

This trial illustrates several important statistical distinctions. The primary endpoint is a continuous laboratory measure, the treatment contrast is a difference in adjusted means, and the sponsor’s reported inference comes from a mixed model repeated measures analysis. Reading the effect estimate together with its confidence interval is more informative than relying on a p-value alone.

The study also offers a useful example of why a summary-statistic recalculation and a sponsor’s participant-level model are not identical. The recalculated and sponsor-reported mean differences are both 78.5, while their intervals differ. A simple check can help assess the scale and direction of a result, but it cannot reconstruct model adjustments or repeated-measure structure without the underlying data and full specification.

The broad set of treatment arms and registered outcomes raises questions about multiplicity, analysis populations, and missing data. The primary comparison remains identifiable, but the registry does not report enough detail to evaluate a complete family of tests or all robustness analyses. This is a reminder that a statistically significant primary endpoint does not confer the same evidential status on every secondary result.

Finally, the endpoint is a surrogate laboratory measure observed over the stated Week 12 timeframe. Statistical evidence for a difference in that measure is distinct from evidence about clinical outcomes or long-term effects. Careful interpretation keeps the estimand, model, uncertainty, and scope of the data in view.

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.

Several trial-specific limitations affect interpretation. The primary analysis includes only participants with baseline and Week 12 HDL-C measurements, while the registry record does not report missing-outcome counts, reasons for missingness, or sensitivity analyses. This makes it difficult to assess how the observed-data analysis relates to all randomized participants.

The registry identifies an MMRM and adjustment factors but does not report the complete model specification, covariance structure, diagnostics, or detailed handling of missing measurements. The trial’s original sample-size assumptions and justification are also not reported. The planning illustrations are based on the observed effect and should not be mistaken for the trial’s original design calculations.

The record lists multiple secondary outcomes and analyses but does not report a complete multiplicity strategy. This limits the strength of conclusions drawn from secondary p-values. In addition, the primary endpoint is a Week 12 lipid measure rather than a clinical outcome, and no clinical-importance threshold is reported. The duration of the primary endpoint limits inference about persistence or longer-term consequences.

Although the trial was randomized and quadruple-masked, implementation details for allocation and masking are not reported in the registry record. The record also does not provide complete eligibility criteria or detailed reasons for discontinuation. The primary comparison concerns one LY2484595 dose with one atorvastatin regimen; results should not be assumed to apply to other regimens or populations. These are limits on interpretation of the registry record, not evidence that the reported analysis is invalid.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of LY2484595 in Patients With High LDL-C or Low HDL-C — NCT01105975.
• Registry record last updated 2018-03-22; status last verified 2018-02.
• Results first posted 2018-03-22.
• 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

NCT01105975 is a completed randomized, quadruple-masked, parallel-group Phase 2 study sponsored by Eli Lilly and Company. The registry lists an enrollment of 398, a start in 2010-04, and primary completion in 2011-06. The trial studied LY2484595 alone and in combination with atorvastatin, simvastatin, or rosuvastatin in patients with high LDL-C or low HDL-C.

The primary endpoint was percent change from baseline to Week 12 in HDL-C for LY2484595 plus atorvastatin versus atorvastatin alone. The sponsor-reported MMRM mean difference was 78.5, with a 90% confidence interval from 64.9 to 92.1 and p < 0.001. The result is specific to this laboratory endpoint, treatment comparison, and analysis population; it does not establish a clinical outcome effect.