A Multicenter Study Comparing the Safety and Efficacy of ABT-335 and Rosuvastatin Calcium Combination Therapy to Monotherapy in Subjects With Dyslipidemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Multicenter Study Comparing the Safety and Efficacy of ABT-335 and Rosuvastatin Calcium Combination Therapy to Monotherapy in Subjects With Dyslipidemia (NCT00463606), a randomized phase 3 study. The primary outcome is Mean Percent Change From Baseline to the Final Visit in High-density Lipoprotein Cholesterol (HDL-C) (Full Analysis Set).

2. Clinical Question

The trial addressed whether adding ABT-335 to rosuvastatin was associated with a different short-term change in HDL-C than rosuvastatin alone. HDL-C was the primary endpoint, while the listed secondary outcomes included other lipid measures and high-sensitivity C-reactive protein. The posted result is an average percent change in each group, not a description of every participant’s response. It also does not answer questions about outcomes beyond the 12-week timeframe. The record identifies the study as a Phase 3 randomized trial and reports enrollment of 760 participants, but the primary outcome groups include 236 and 241 participants in the Full Analysis Set. The clinical question is therefore framed around these observed analysis groups and the endpoint definition in the registry record, rather than assuming that every enrolled participant contributed a primary endpoint value.

3. Trial Design

This was a Phase 3, randomized, double-blind, parallel-group study with a planned treatment period of 12 weeks. The registry record describes three drug arms: ABT-335 135mg with rosuvastatin calcium 5mg, ABT-335 135mg alone, and rosuvastatin calcium 5mg alone. Each regimen was administered orally once daily. The primary comparison used the combination and rosuvastatin monotherapy arms; the ABT-335 monotherapy arm appears in the listed secondary comparisons.

The registry record reports enrollment of 760 and classifies the trial as completed. The recorded start is 2007-04 and primary completion is 2008-02. The conditions listed are hypercholesterolemia and dyslipidemia. AstraZeneca is identified as the sponsor. These details describe the registered design and status; they do not establish how many participants completed treatment or the reasons any participants may have discontinued.

For the primary HDL-C outcome, the Full Analysis Set contained 236 participants in the combination arm and 241 in the rosuvastatin arm. The endpoint definition required randomized participants to have a baseline and at least one post-baseline HDL-C value. Last observation carried forward (LOCF) was used to impute values for participants missing a post-baseline visit value. The registry record does not report the original sample-size justification, the planned allocation ratio, or details of any interim analyses. The record identifies randomization, double masking, and a parallel model, but does not provide further operational details about sequence generation or how masking was maintained.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: A total of 168 study sites screened participants, wtih 144 of these sites randomizing participants.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set was used and was defined as all randomized participants who had both a baseline and at least 1 post-baseline value for high-density lipoprotein cholesterol. Last observation carried forward (LOCF) was used to impute values for participants missing a post-baseline visit value. Only post-baseline values were carried forward.

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Mean Percent Change From Baseline to the Final Visit in High-density Lipoprotein Cholesterol (HDL-C) (Full Analysis Set), measured over Baseline to 12 Weeks.

The primary endpoint was mean percent change in high-density lipoprotein cholesterol (HDL-C) from baseline to the final visit, with the timeframe given as baseline to 12 weeks. Percent change expresses the outcome relative to each participant’s baseline level, allowing the endpoint to summarize change rather than just a follow-up concentration. The analysis compared the combination of ABT-335 135 mg and rosuvastatin 5 mg with rosuvastatin 5 mg monotherapy.

HDL-C was one of several lipid measures listed in the registry record. As the primary endpoint, it represents the central prespecified question described in the record, while the additional lipid and inflammatory measures are secondary. The posted means were 23.0% for the combination arm and 12.4% for rosuvastatin alone. These are group averages; they do not show how responses varied among individuals. The record gives standard errors for the arm means and bounds for the reported comparison. It does not report an endpoint value in the concentration unit of HDL-C, so the result should be read as a percent change, not as an absolute difference in HDL-C concentration.

The Full Analysis Set included randomized participants with a baseline and at least one post-baseline HDL-C value, and LOCF was used for missing post-baseline visit values. This definition matters because the endpoint estimate is tied to that analysis population and imputation approach; it is not necessarily an estimate based only on participants with a measured value at the final visit.

7. Results Posted in the Registry

• Rosuvastatin Calcium: mean 12.40 versus 23.00 on control; difference -10.60 (95% CI -14.27 to -6.93), standardized effect d = -0.52; 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.

Mean Percent Change From Baseline to the Final Visit in Non-high-density Lipoprotein Cholesterol (Non-HDL-C), With ABT-335 135 mg in Combination With Rosuvastatin 5 mg Versus ABT-335 135 mg Monotherapy (Full Analysis Set)

• 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:
  • ABT-335: mean -16.00 versus -37.40 on control; difference 21.40 (95% CI 18.48 to 24.32), standardized effect d = 1.31; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for ABT-335 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Mean Percent Change From Baseline to the Final Visit in Non-high-density Lipoprotein Cholesterol (Non-HDL-C), With ABT-335 135 mg in Combination With Rosuvastatin 5 mg Versus Rosuvastatin 5 mg Monotherapy (Full Analysis Set)

• 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:
  • Rosuvastatin Calcium: mean -31.80 versus -37.40 on control; difference 5.60 (95% CI 2.68 to 8.52), standardized effect d = 0.34; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Rosuvastatin Calcium than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Mean Percent Change From Baseline to the Final Visit in Very-low-density Lipoprotein Cholesterol (VLDL-C) (Full Analysis Set)

• 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:
  • Rosuvastatin Calcium: mean -22.20 versus -41.30 on control; difference 19.10 (95% CI 9.27 to 28.93), standardized effect d = 0.35; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Rosuvastatin Calcium than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry record lists six secondary outcomes, including non-HDL-C, VLDL-C, ApoB, hsCRP, and total cholesterol. Numeric results are posted for three secondary comparisons: two comparisons involving non-HDL-C and one involving VLDL-C. The listed ApoB, hsCRP, and total cholesterol outcomes do not have numeric results in the registry record. Across the posted non-HDL-C and VLDL-C mean changes, the combination arm’s changes are more negative than the corresponding comparator arm’s values. This is directionally consistent with a greater reduction in these measures in the combination arm, while the primary HDL-C result describes a greater average increase in the combination group.

The secondary outcomes add information about other lipid measures, but they are not interchangeable with the primary HDL-C endpoint. Each measures a distinct biomarker, and results for one do not establish what occurred for the others. The record’s secondary comparisons include the combination against ABT-335 monotherapy and the combination against rosuvastatin monotherapy. They should be understood in relation to their particular comparator, rather than treated as a single overall treatment contrast.

Unadjusted secondary results carry less weight because multiple outcomes and comparisons create more opportunities for apparently notable results to arise by chance. The posted p-values for these recalculated contrasts are unadjusted, and the registry record does not report a multiplicity procedure. The record also lists no reported analyses, so it does not establish which inferential procedures were prespecified or whether an adjustment was part of the sponsor’s analysis. Secondary findings are best regarded as supporting or exploratory context rather than as a replacement for the primary comparison. No numeric interpretation can be made for the secondary endpoints whose results are not reported in the registry record.

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

• Mean Percent Change From Baseline to the Final Visit in Non-high-density Lipoprotein Cholesterol (Non-HDL-C), With ABT-335 135 mg in Combination With Rosuvastatin 5 mg Versus ABT-335 135 mg Monotherapy (Full Analysis Set)
• Mean Percent Change From Baseline to the Final Visit in Non-high-density Lipoprotein Cholesterol (Non-HDL-C), With ABT-335 135 mg in Combination With Rosuvastatin 5 mg Versus Rosuvastatin 5 mg Monotherapy (Full Analysis Set)
• Mean Percent Change From Baseline to the Final Visit in Very-low-density Lipoprotein Cholesterol (VLDL-C) (Full Analysis Set)
• Mean Percent Change From Baseline to the Final Visit in Apolipoprotein B (ApoB) (Full Analysis Set)
• Median Percent Change From Baseline to the Final Visit in High Sensitivity C-reactive Protein (hsCRP) (Full Analysis Set)
• Mean Percent Change From Baseline to the Final Visit in Total Cholesterol (Full Analysis Set)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is a continuous outcome: mean percent change in HDL-C. For a continuous outcome, the analysis summarizes the measured quantity with a group mean and compares average values between groups. This differs from a binary outcome, which would classify each participant into categories, or a time-to-event outcome, which would incorporate the timing of an event. In this trial, the unit is percent change from baseline to the final visit, and the posted arm values are means.

What does the mean describe?

The mean is the arithmetic average of the participant-level values included in the analysis. It offers a concise estimate of the group’s average change, but it does not imply that every participant changed by the mean amount. In a clinical study, the distribution of individual responses may be wider or narrower than the mean alone suggests. The registry record reports standard errors for the primary arm means, not the individual-level distribution or the standard deviation of participant changes.

The arm means are 23.0% in the combination group and 12.4% in the rosuvastatin group. The comparison is reported as a mean difference of -10.6. The sign depends on the order used to express the contrast, so it should be interpreted together with the arm means and the named groups. The negative sign accompanies a lower mean in the rosuvastatin arm relative to the combination arm. It is not a negative change in the combination arm, whose posted mean change is positive.

What does the standard error indicate?

A standard error describes the uncertainty in an estimated mean due to sampling variability, under the assumptions of the statistical method. It is not the same as the standard deviation of individual participant values. The primary arm standard errors are 1.33 and 1.32. A relatively small standard error compared with the mean indicates that the group mean is estimated with greater precision than if the standard error were large, although precision also depends on the scale and the method. Standard errors alone do not show the full distribution of participant outcomes.

How should the interval and p-value be read?

The posted comparison bounds run from -14.27 to -6.927. An interval estimate represents a range of values compatible with the observed data and statistical assumptions; it communicates uncertainty around a point estimate rather than a single exact value. The record does not state the confidence level for these bounds. The interval is entirely on the negative side of zero under the reported contrast convention. The unadjusted p-value is < 0.001. A p-value measures how unusual data at least as incompatible with a no-difference reference would be under a specified statistical model, if that reference and model were correct. It is not the probability that the no-difference reference is true, and it does not measure clinical importance.

Here, the comparison and p-value are recalculated unadjusted results from the posted arm summary statistics. The registry record lists no reported analyses. A recalculated result is therefore not a claim about the exact model, covariate adjustment, or inferential procedure used by the sponsor. The record does not report those details, nor does it identify a multiplicity adjustment for the primary or secondary outcomes.

Why does over-dispersion matter in general?

In general, over-dispersion means that observed variability is greater than a statistical model expects. When a model understates variability, standard errors may be too small and intervals too narrow, which can make evidence appear more precise than it is. This concept is especially relevant to outcomes analyzed with count or binary models. The primary endpoint in this study is continuous, and the registry record does not report an over-dispersion assessment; this general point should not be mistaken for a finding about this trial.

How do missing values affect this endpoint?

The Full Analysis Set required randomization, a baseline HDL-C value, and at least one post-baseline HDL-C value. LOCF was used for participants missing a post-baseline visit value. In general, LOCF carries the participant’s last observed value forward to represent a later missing value. This can preserve participants in the analysis, but it relies on assumptions about how the outcome would have evolved after the last measurement. The approach may not reflect the uncertainty of unobserved values, and its suitability depends on the pattern and reasons for missingness. The record does not provide the number or pattern of missing values, reasons for missingness, or sensitivity analyses under alternative assumptions.

Randomization supports comparisons between assigned groups by reducing the likelihood that baseline factors systematically differ, but it does not eliminate uncertainty or problems arising after randomization. The registered design is randomized, double-blind, and parallel-group. The registry record does not provide enough detail to assess sequence generation, concealment, or the conduct of masking. These limits matter when interpreting the statistical result: the posted estimates describe the endpoint and analysis definition that are recorded, while unreported analytical details remain unknown.

11. Confidence Intervals and Precision

The primary comparison is a mean difference of -10.6, with reported interval bounds from -14.27 to -6.927. In combination with the arm means, 23.0% for the combination and 12.4% for rosuvastatin monotherapy, the interval supports a contrast in the direction of a higher average percent change in the combination arm. Its width conveys uncertainty around the estimated difference: a narrower interval would indicate greater precision than a wider one, all else equal. The interval does not capture every possible source of uncertainty, particularly uncertainty related to unreported design or analysis choices.

The registry record does not identify the confidence level associated with these bounds. It is therefore appropriate to describe the reported interval endpoints without assigning a confidence level that is not stated. The bounds exclude zero, and the recalculated unadjusted p-value is < 0.001. These features indicate statistical separation under the comparison represented by the posted summary statistics, but they do not determine the clinical importance of the difference.

The standard errors for the primary arm means are 1.33 and 1.32. Standard errors quantify sampling uncertainty in the means, whereas the interval pertains to the between-group contrast. Neither should be confused with the spread of individual participant changes. The record provides no standard deviation for participant-level HDL-C percent changes, and no individual-level distribution can be reconstructed from the displayed summary information.

Precision also depends on the analysis population and handling of missing values. The primary means are based on the Full Analysis Set, and LOCF was used to impute missing post-baseline visit values. Without counts and patterns of missingness, the impact of this choice on the interval cannot be assessed from the registry record. The interval should thus be read as precision conditional on the posted analysis definition, not as a complete accounting of uncertainty in the study.

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, not total events. They are illustrative calculations based on the observed effect, using a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is a stated fraction of the observed effect. The registry record does not report the trial’s original sample-size assumptions or justification, so these figures do not reconstruct the study’s actual design planning.

For the primary HDL-C comparison, planning for 100% of the observed effect gives 59 participants per arm for 80% power and 79 participants per arm for 90% power. Planning for 75% of the observed effect gives 104 participants per arm for 80% power and 140 participants per arm for 90% power. Planning for 50% of the observed effect gives 234 participants per arm for 80% power and 314 participants per arm for 90% power.

In plain terms, the 80% power column gives the illustrative number per arm for a design with an 80% chance of detecting the specified effect under the stated assumptions, if that effect is present. The 90% column gives the corresponding number for a higher chance of detection. These are design probabilities, not guarantees for any particular study. Smaller effects need more participants to detect than larger effects, which is why the participant requirements rise as the assumed fraction of the observed effect falls.

These calculations are planning illustrations, not observed power. They use an effect estimated from the trial’s posted results, so they should not be taken as an independent measure of how successful the completed trial was or as evidence that its original sample size was adequate. The estimates also rely on the listed simplifying assumptions and may not match a design using different analysis methods or allocation decisions.

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 = -10.600; pooled standard deviation σ = 20.462.
2. n = 2 × (1.96 + 0.842)² × 20.462² / -10.600² = 58.5, rounded up to 59 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 <- -10.6                      # observed difference in means
sigma <- 20.4622173002367                     # 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(463606)
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 = 59 (80% power), 79 (90% power)
 75% of observed effect: n per arm = 104 (80% power), 140 (90% power)
 50% of observed effect: n per arm = 234 (80% power), 314 (90% power)
power.t.test: 59.5 
Simulated power at n = 59 per arm: 0.799 (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. Multiplicity, Missing Data and Analysis Populations

The registry record describes one primary endpoint and multiple secondary outcomes. Examining several outcomes raises a multiplicity issue: as the number of statistical comparisons increases, the chance of at least one result appearing unusual by chance can also increase. In general, a prespecified strategy may address this by controlling an overall error rate or by organizing the analyses into a hierarchy. The registry record does not report whether such a strategy was planned or used. The posted comparison p-values are unadjusted recalculations from summary statistics, so they do not account for the full set of outcomes.

The record lists no reported analyses. It therefore does not provide a complete statistical analysis plan, details of covariate adjustment, or a sponsor-reported inferential model for the primary result. The recalculated estimates are straightforward unadjusted comparisons of posted summaries. No discrepancy is identified in the record, but that does not establish that every detail of a sponsor analysis has been reproduced: the underlying participant-level data and full analysis specification are not included.

The primary outcome population is the Full Analysis Set, defined as randomized participants with a baseline and at least one post-baseline HDL-C value. LOCF was used for participants missing a post-baseline visit value. The registry record does not report the number of missing values, the timing or reasons for missingness, or the number excluded from the Full Analysis Set. It also does not report sensitivity analyses to evaluate whether conclusions depend on the imputation method. In general, missing-data methods depend on assumptions about why observations are missing and what unobserved values might have been; those assumptions cannot be evaluated fully from the posted record.

The record also does not report an interim analysis plan or original sample-size justification. These items should not be inferred from the completed study’s enrollment or from planning illustrations based on the observed effect. The limits are material: they leave uncertainty about multiplicity control, missing-data robustness, and the precise analysis procedures, while the endpoint definition and posted group summaries remain interpretable as recorded.

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: coronary artery disease (0.8%), acute myocardial infarction (0.4%), cardiogenic shock (0.4%), pancreatitis acute (0.4%), renal cell carcinoma stage unspecified (0.4%).

For serious adverse events, the registry record reports 7 affected participants among 253 at risk in the combination arm, 6 among 254 in the ABT-335 arm, and 4 among 251 in the rosuvastatin arm. For other adverse events, it reports 52 among 253, 41 among 254, and 45 among 251, respectively. These are group-specific counts with their corresponding at-risk denominators. They should be read as descriptive safety summaries, not as adjusted comparisons or evidence that differences between groups are statistically meaningful.

The registry record lists serious adverse-event terms including coronary artery disease at 0.8%, acute myocardial infarction at 0.4%, cardiogenic shock at 0.4%, pancreatitis acute at 0.4%, renal cell carcinoma stage unspecified at 0.4%, and anxiety at 0.4%. The record does not report deaths; a missing death value should not be interpreted as zero deaths. The safety observations cover a 12-week trial, and the posted counts alone do not describe event timing, severity beyond the serious classification, causality assessment, or longer-term safety.

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

The primary result describes average HDL-C percent change over 12 weeks in the analyzed groups. The combination arm mean was 23.0%, and the rosuvastatin monotherapy arm mean was 12.4%; the reported contrast was -10.6 with bounds from -14.27 to -6.927 and unadjusted p < 0.001. Taken together, these statistics describe a difference in group averages under the endpoint definition and analysis population recorded for the trial.

The result does not mean that every participant experienced the average change, nor does it identify the outcome for a particular individual. Group means hide variation among participants, and the record does not provide the individual-level distribution. It also does not establish a difference in longer-term outcomes: the study’s stated timeframe was baseline to 12 weeks. Nor does a biomarker result by itself report whether participants experienced a clinical event or how they felt.

The statistical result is not a statement about the probability that a treatment works or fails. An unadjusted p-value below 0.001 is evidence against a no-difference reference under the assumptions of the comparison; it is not the probability that the reference is true. The interval bounds describe uncertainty around the estimated contrast, but the confidence level is not reported in the registry record. Statistical separation also does not, by itself, establish clinical importance.

The result is conditional on the Full Analysis Set and LOCF handling of missing post-baseline values. The record does not report missing-value counts or sensitivity analyses, so the robustness of the result to alternative missing-data assumptions cannot be judged. Multiple secondary outcomes are listed, and the record does not report multiplicity control. Secondary findings therefore provide context but do not carry the same interpretive priority as the primary endpoint. Finally, the registry record describes the trial design and posted results, but does not provide every detail needed to evaluate the sponsor’s complete analysis procedures.

17. Why This Trial Matters Statistically

This trial illustrates how a registry record can support a transparent reading of a continuous clinical endpoint while leaving important analytical details unresolved. The primary outcome is a mean percent change, so interpretation depends on distinguishing the arm means, the between-group contrast, and the uncertainty around that contrast. The sign of the reported difference could be confusing in isolation; reading it alongside the arm means clarifies that the combination group had the higher average percent change.

The record also illustrates why endpoint populations and missing-data rules are part of the result, not merely technical footnotes. The primary analysis used a Full Analysis Set defined by availability of baseline and post-baseline values, with LOCF for missing post-baseline visits. Those choices determine which participants contribute and how incomplete follow-up is represented. Without missingness details or sensitivity analyses, readers cannot assess how much the result depends on those assumptions.

Another statistical lesson is the distinction between a primary outcome and a collection of secondary measures. Additional endpoints can show whether observed patterns extend across related biomarkers, but unadjusted comparisons across multiple outcomes need cautious interpretation. Here, numeric registry results are available for some secondary comparisons, while other listed secondary outcomes have no numeric results in the registry record. The record also lists no reported analyses, limiting insight into the sponsor’s intended inferential strategy.

Finally, planning illustrations based on observed results are not substitutes for the original design rationale. They show how participant requirements vary with the effect size targeted under explicit assumptions, but do not establish what assumptions guided the completed trial. Separating the observed result, its uncertainty, and the unreported design details helps prevent a statistical summary from being read as a broader clinical conclusion.

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.

Several limitations are specific to what the registry record reports for this trial. The study lasted 12 weeks, so the posted endpoint results describe short-term changes and do not provide longer-term follow-up. The primary outcome is a biomarker measure, and no clinical event outcome is reported as the primary result. The registered enrollment is 760, while the primary HDL-C analysis includes 236 participants in the combination group and 241 in the rosuvastatin group. The record does not explain the difference between overall enrollment and the primary analysis group counts.

The Full Analysis Set required a baseline and at least one post-baseline HDL-C measure, and LOCF was used for missing post-baseline values. The record does not report the amount or pattern of missing data, reasons for missingness, or sensitivity analyses. This limits assessment of how robust the primary estimate is to alternative assumptions about missing measurements. The record also does not report the original sample-size justification, analysis plan details, interim analyses, or a multiplicity strategy.

Statistical comparisons presented from the posted summary values are unadjusted recalculations, and the registry record lists no reported analyses. The confidence level for the primary interval bounds is not stated. These limitations do not negate the descriptive estimates, but they constrain how fully their inferential context can be evaluated. Secondary results are incomplete in the registry record: although six outcomes are listed, numeric results are posted for only selected comparisons. Finally, the adverse-event summaries are counts over the study period; the record does not provide enough detail to assess timing, causal attribution, or longer-term safety. Death values are not reported in the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Multicenter Study Comparing the Safety and Efficacy of ABT-335 and Rosuvastatin Calcium Combination Therapy to Monotherapy in Subjects With Dyslipidemia — NCT00463606.
• Registry record last updated 2012-10-03; status last verified 2012-09.
• Results first posted 2012-08-01.
• 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

NCT00463606 is a completed Phase 3 randomized, double-blind, parallel-group study sponsored by AstraZeneca. The registry record reports enrollment of 760 and a 12-week comparison of ABT-335 135mg plus rosuvastatin calcium 5mg against ABT-335 monotherapy and rosuvastatin calcium 5mg monotherapy in participants with dyslipidemia.

The primary endpoint was mean percent change in HDL-C from baseline to the final visit. In the Full Analysis Set, the combination and rosuvastatin groups had posted means of 23.0% and 12.4%. Their reported mean difference was -10.6, with interval bounds from -14.27 to -6.927 and unadjusted p < 0.001. LOCF was used for missing post-baseline HDL-C values. The record lists secondary lipid and inflammatory outcomes, but numeric results are not available for every listed outcome. It does not report the original sample-size justification, a complete analysis plan, or multiplicity handling.