Efficacy and Safety of Gemcabene in Patients With Homozygous Familial Hypercholesterolemia on Stable, Lipid-Lowering Therapy (COBALT-1): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Gemcabene in Patients With Homozygous Familial Hypercholesterolemia on Stable, Lipid-Lowering Therapy (COBALT-1) (NCT02722408), a na phase 2 study. The primary outcome is Percent Change From Baseline in LDL-C at Day 28.

2. Clinical Question

3. Trial Design

COBALT-1 was a phase 2, open-label, dose-finding study with a single-group model. The registry lists enrollment as 8 and reports that the study is completed. The recorded study dates are 2016-06 for start and 2017-04 for primary completion. The sponsor is NeuroBo Pharmaceuticals Inc.

Participants with HoFH on stable lipid-lowering therapy received gemcabene in a sequential dose regimen. The arm description states that participants received 300 mg orally once daily from day 1 to 28, then 600 mg orally once daily from day 29 to 56, followed by 900 mg orally once daily. The description stops after “from” and does not report the remainder of that schedule. The record therefore does not support a fuller account of the duration or timing of the 900 mg dose.

Open-label means that masking was not used. This can matter when outcomes involve judgment or participant reporting, although the primary endpoint is a laboratory measure. A single-group study can describe changes observed during treatment, but without a randomized control group it cannot separate those changes from other influences in the same way a controlled comparison can. The registry itself cautions that the dose groups are not a like-for-like control comparison.

The primary endpoint was percent change from baseline in LDL-C at day 28. Secondary outcomes included absolute and percent changes in fasting LDL-C, non-HDL-C, total cholesterol, triglycerides, and HDL-C. The registry lists 49 outcome measures and 234 analyses. Those counts indicate a broad set of recorded outcomes and analyses, but the record does not provide enough information to establish how those analyses were prioritized or adjusted for multiple testing.

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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS): All enrolled participants who received at least 1 dose of study drug and had a valid post-baseline efficacy assessment, for a given dose level.

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: Full Analysis Set (FAS): All enrolled participants who received at least 1 dose of study drug and had a valid post-baseline efficacy assessment, for a given dose level.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline in LDL-C at Day 28, measured over Baseline, day 28.

The primary endpoint is percent change from baseline in LDL-C at day 28. LDL-C is a measure of cholesterol carried in low-density lipoproteins. Expressing change as a percentage relates the change to the participant’s baseline value, rather than reporting only an absolute difference in the measurement scale. This can be useful when baseline levels vary, because it describes change relative to each participant’s starting point.

The registry identifies the endpoint as a least-squares mean and lists standard error as its dispersion measure. A least-squares mean is a model-based estimate of an average, usually adjusted for terms in the specified statistical model. The statistical model used for this estimate is not reported in the registry record, so the adjustments and assumptions behind the estimate cannot be described. Standard error characterizes the precision of an estimated mean; it is not the same as the spread of individual participant measurements.

The timeframe is baseline to day 28. This is an early assessment within the recorded dose-escalation regimen, not an endpoint covering the full schedule described in the record. The listed endpoint does not itself indicate whether LDL-C rose or fell, or by how much. The estimate and its standard error are not reported in the registry record, so the size, direction, and precision of the primary result cannot be assessed from the information recorded.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods determine what an estimate represents, how uncertainty is quantified, and which observations contribute. For COBALT-1, the record identifies the primary endpoint’s summary as a least-squares mean with standard error, and defines the Full Analysis Set for a given dose level. It does not report the full analysis plan or the model used to produce the estimate. The interpretation must therefore stay close to those recorded details.

What does a least-squares mean describe?

A least-squares mean is a model-based average. Unlike a simple arithmetic mean, it can account for specified factors in a statistical model. Its meaning depends on which factors were included, how the model was formulated, and what assumptions were made. Those details are not reported in the registry record for the primary endpoint. The term identifies the type of summary, but does not reveal whether the estimate was adjusted for baseline LDL-C or other variables.

The primary endpoint is percent change from baseline in LDL-C at day 28. In general, a percentage change expresses a measurement’s change relative to its starting value. The endpoint is therefore a summary of change relative to baseline, not a direct comparison between gemcabene and a control group. The registry does not give the estimated value, so no interpretation of its direction or magnitude is possible from the recorded details.

What does the standard error tell the reader?

Standard error describes the precision of an estimated mean under the statistical model. A smaller standard error generally indicates less sampling uncertainty around the estimate, while a larger one indicates more uncertainty. It does not describe the range of individual participants’ responses. A standard error can contribute to constructing a confidence interval or conducting a test, but the registry record does not report the numerical standard error, confidence interval, or test result for the primary endpoint.

A confidence interval is a range of values compatible with the data under a specified statistical procedure. It helps readers judge both the estimated effect and the uncertainty around it. The record does not report a confidence interval for the primary result. It would not be appropriate to reconstruct one without the estimate, its standard error, and the relevant analysis details.

How does the analysis population affect the result?

The primary endpoint population is the Full Analysis Set: participants who received at least 1 dose of study drug and had a valid post-baseline efficacy assessment, for a given dose level. This definition is tied to the dose-specific endpoint analysis. It differs conceptually from simply counting everyone enrolled, because eligibility for the stated analysis requires both drug exposure and a post-baseline assessment. Enrollment is listed as 8, but the number contributing to each endpoint analysis is not reported in the registry record.

Analysis populations can affect interpretation when participants lack measurements or do not meet analysis criteria. The registry does not state how many measurements were missing, why any were missing, or how missing values were handled. It also does not report whether analyses were repeated under alternative assumptions. Without those details, the possible influence of missing data cannot be evaluated.

What can a single-group design estimate?

A single-group study can describe outcomes observed among participants who received the study intervention. A before-and-after endpoint can characterize change over time, but there is no randomized comparator to estimate the change that might have occurred without the intervention. This limits causal interpretation. In addition, an open-label design means masking was not used. The primary measure is a laboratory endpoint, but the design still lacks the protection against confounding and selection differences that random allocation can offer.

The registry describes dose levels and flags that the study is not a randomized comparison. The dose-specific safety counts should therefore be read as descriptive records, not as a formal comparison of dose effects. The dose schedule is sequential in the arm description, but its recorded description is incomplete after the 900 mg dose is introduced. The timing and exposure details not stated in the registry record should not be inferred.

What do the planning assumptions mean?

The registry includes generic assumptions for a sample-size planning illustration: a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect defined as a stated fraction of the observed effect. These are planning assumptions, not a description of the trial’s single-group design. No calculated planning values are reported in the record. Such illustrations are hypothetical design calculations, not observed power and not evidence about the precision or significance of this trial’s result.

More generally, planning calculations depend on the effect chosen for detection, the variability assumed, the statistical test, and the design. An illustration based on an observed effect can be useful for considering future study designs, but does not establish that the original study was adequately sized. The registry does not report the trial’s own sample-size justification, and it would be inappropriate to infer one from the generic illustration assumptions.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Gemcabene")
n   <- c(8)
m   <- c(-25.54)
sp  <- c(8.301)
sd  <- sp * sqrt(n)      # standard errors converted to SDs
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Gemcabene  8  -25.540  23.479  -45.169  -5.911

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 49 outcome measures and 234 analyses. The secondary outcomes include fasting lipid measures reported as both absolute change and percent change, including LDL-C, non-HDL-C, total cholesterol, and triglycerides, as well as percent change in HDL-C. Multiple outcomes can provide a broader description of the data, but repeated statistical testing raises a multiplicity issue: as the number of tests grows, some findings may appear notable by chance even when there is no corresponding underlying effect.

Methods for addressing multiplicity can include prespecifying a primary endpoint, adjusting significance thresholds, or treating secondary analyses as exploratory. The record designates percent change from baseline in LDL-C at day 28 as the primary endpoint, but does not report whether any multiplicity adjustment was used for secondary outcomes or other analyses. The number of analyses alone does not reveal which were inferential, how they were prioritized, or whether adjustment was applied.

The primary endpoint population is the Full Analysis Set, defined as participants who received at least 1 dose and had a valid post-baseline efficacy assessment, for a given dose level. The registry does not report the number contributing to each endpoint estimate. It also does not describe the amount or reasons for missing outcome data, whether missing values were imputed, or whether sensitivity analyses were conducted. These details matter because analyses restricted to participants with valid follow-up measurements may differ from analyses using other approaches to missing data.

The statistical plan, model specifications, and analysis-population details beyond the stated FAS definition are not reported in the registry record. Consequently, neither the multiplicity strategy nor the robustness of the efficacy results to missing-data assumptions can be assessed from the recorded information.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry records serious adverse events for three dose labels: Gemcabene 300 mg, Gemcabene 600 mg, and Gemcabene 900 mg. Each lists 8 participants at risk and 0 participants affected by serious adverse events. Deaths are recorded as 0 for each dose label. The other-adverse-event counts are 7, 4, and 6, respectively, with 8 at risk in each listed group.

These figures are descriptive counts. The registry does not list serious adverse-event terms, so the types of serious events cannot be characterized from this record. The recorded count of 0 means that no serious adverse events are listed for the corresponding dose entry; it does not establish that an event is impossible or that risk is absent. Small study experience can leave uncommon events poorly characterized.

The dose entries should not be interpreted as randomized groups. The trial is listed as single-group, open-label, and not a randomized comparison. The registry’s arm description presents sequential dosing, and the safety entries alone do not establish whether or how exposure periods overlapped for individual participants. Comparisons of the other-event counts across dose labels therefore cannot establish a dose-related difference. The record does not provide event timing, duration, severity details for other adverse events, or an analysis comparing safety by dose.

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

The registry identifies a phase 2, open-label, single-group study of gemcabene in patients with HoFH on stable lipid-lowering therapy. It specifies percent change from baseline in LDL-C at day 28 as the primary endpoint and describes a Full Analysis Set for dose-level analyses. It also records safety counts for three gemcabene dose labels. These features describe the study’s question and some of its reporting, but they do not supply the numerical primary efficacy result.

The primary endpoint’s estimated least-squares mean, standard error, confidence interval, and statistical test result are not reported in the registry record. As a result, the posted information does not show how LDL-C changed by day 28, how variable the estimated change was, or how compatible the result was with a range of possible values. No conclusion about the direction or statistical precision of the primary endpoint can be drawn from the endpoint title alone.

The single-group design also limits what a before-and-after result could establish, even if a numerical estimate were available. Without randomized control participants, change over time cannot be attributed solely to the intervention. Open-label conduct means masking was not used, and the registry specifically cautions that the dose groups are not like-for-like controls. The study can describe observed experience under its design; it cannot provide a randomized comparison of gemcabene against a control treatment.

For safety, the record lists no serious adverse events and no deaths in the dose-specific entries, alongside other adverse events recorded for participants in each entry. Those observations describe what was recorded in this study; they do not rule out uncommon harms or establish differences between doses. The registry does not list serious adverse-event terms.

Accordingly, the recorded information is useful for identifying the study design, endpoint, analysis-population definition, and reported safety counts. It is insufficient to judge the magnitude or precision of the primary efficacy result, the handling of missing data, or the statistical significance of outcomes.

14. Why This Trial Matters Statistically

This study illustrates why a clinical result cannot be understood from an endpoint label alone. The registry specifies a clear primary outcome—percent change from baseline in LDL-C at day 28—and identifies the summary as a least-squares mean with standard error. But the numerical estimate and its uncertainty are not reported in the registry record. Without those values, readers cannot assess the endpoint’s direction, magnitude, or precision.

The design also shapes the type of statistical conclusion that is available. COBALT-1 is listed as an open-label, single-group study, not a randomized comparison. A change from baseline can describe the experience of participants receiving gemcabene, but it does not supply the counterfactual comparison needed to isolate what would have happened without the intervention. The registry’s warning that the dose groups are not like-for-like controls reinforces the need to avoid interpreting dose-specific entries as randomized treatment contrasts.

The record lists 49 outcome measures and 234 analyses. This makes the distinction between a designated primary endpoint and the wider collection of analyses especially important. Multiplicity is a general statistical concern when many outcomes or tests are examined; the record does not report the strategy, if any, used to address it. Likewise, a defined analysis population is important, but without participant counts for each endpoint, missing-data details, and the model specification, readers cannot fully assess how the efficacy summaries were generated.

The safety entries are also a useful reminder that zero recorded serious events in a small study is a descriptive observation, not proof that a risk does not exist. The most informative statistical reading combines study design, endpoint definition, uncertainty, analysis population, missingness, multiplicity, and safety reporting rather than relying on any single label or count.

15. 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 follow directly from the registry record. The study is open-label and single-group, with no randomized comparator. This means the recorded design cannot distinguish a treatment-associated change from changes that might occur over time or under continued background therapy. The registry explicitly cautions that the dose groups are not like-for-like controls, so dose-specific results should not be treated as controlled comparisons.

The primary endpoint is defined, but its numerical estimate and standard error are not reported in the registry record. The record also does not report a confidence interval or test result for that endpoint. This prevents assessment of effect direction, magnitude, and precision. Although the endpoint is described as a least-squares mean, the model specification and any adjustment variables are not reported.

The Full Analysis Set definition is given for a given dose level, but the number of participants contributing to each endpoint analysis is not reported. The record does not describe missing observations, reasons for missingness, methods for handling missing data, or sensitivity analyses. The 49 outcome measures and 234 analyses raise questions about prioritization and multiplicity, but any adjustment strategy is not reported in the registry record.

The arm description documents the 300 mg and 600 mg schedule and introduces a 900 mg dose, but the recorded text ends before completing the 900 mg schedule. The safety data list zero serious adverse events and zero deaths in each dose entry, but serious-event terms are not listed. Other-event counts are reported without additional event details. Finally, the registry’s generic planning assumptions do not establish the study’s own sample-size justification, which is not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Efficacy and Safety of Gemcabene in Patients With Homozygous Familial Hypercholesterolemia on Stable, Lipid-Lowering Therapy (COBALT-1) — NCT02722408.
• Registry record last updated 2020-06-25; status last verified 2020-06.
• Results first posted 2020-06-25.
• 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.

19. Record Summary

COBALT-1 was a completed phase 2, open-label, single-group dose-finding study of gemcabene in patients with HoFH on stable lipid-lowering therapy. The registry lists enrollment as 8, with a start date of 2016-06 and primary completion in 2017-04. The primary endpoint was percent change from baseline in LDL-C at day 28, summarized as a least-squares mean with standard error. The numerical primary result is not reported in the registry record.

The record describes sequential gemcabene dosing and lists 49 outcome measures and 234 analyses. It records 0 serious adverse events and 0 deaths for each of the three dose entries, alongside other adverse-event counts. The design has no randomized comparator, and the registry cautions against treating the dose groups as like-for-like controls. The model details, multiplicity approach, missing-data methods, and trial-specific sample-size justification are not reported in the registry record.