LIPIDS-P Trial Phase I/II Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for LIPIDS-P Trial Phase I/II Trial (NCT03405870), a randomized phase 1, phase 2 study. The primary outcome is Phase II - Primary Outcome - Change in Total Cholesterol (48 Hours - Enrollment).

2. Clinical Question

3. Trial Design

The registry identifies LIPIDS-P as a randomized, single-masked, parallel Phase I/II trial sponsored by the University of Florida. It is listed as completed, with enrollment of 59. The listed start date is 2018-08-17 and the primary completion date is 2023-04-26. The registry record does not say which participants were masked.

The design includes a dose-escalation component and a Phase II comparison. Phase I has active groups at 1.0, 1.2, 1.4 and 1.6 g/kg of Smoflipid. Its primary outcome was maximum tolerated dose. Phase II has active groups at 1.2, 1.4 and 1.6 g/kg, together with a control group that received no drug and was followed as an active control. Phase II’s stated primary outcome was change in total cholesterol at 48 hours.

The arm descriptions say Smoflipid was infused over a 10-16.5 hour period once per day for the first two days of study enrollment. The registry record does not report how randomization was generated, how allocation was concealed, or how the single masking arrangement was implemented. It also does not report the planned allocation ratio, the trial’s original sample-size justification, or whether any interim analyses were planned.

Enrollment in the registry is 59, while the primary-outcome description identifies 47 Phase II patients for the cholesterol analysis across the three dose groups. The posted arm-level summaries also include Phase I participants, for whom cholesterol change was not the primary outcome. Those Phase I summaries should therefore be read as descriptive results for that measure, not as the Phase I primary analysis.

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 8 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Phase II participants were analyzed for the primary outcome of change in total cholesterol at 48 hours (47 patients) at doses of 1.2, 1.4 and 1.6 g/kg. Phase I participants were also analyzed for this outcome, although it was not the primary outcome for this phase (which was maximum tolerated dose).

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: 10 patients were enrolled in the Phase I trial. One patient was withdrawn leaving nine patients for analysis. 49 patients were enrolled in the Phase II trial. Two patients were withdrawn leaving 47 patients who completed the study and were included in the analysis.

Only the first 6 of 9 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Phase II - Primary Outcome - Change in Total Cholesterol (48 Hours - Enrollment), measured over 48 hours.

The primary Phase II endpoint is change in total cholesterol at 48 hours, expressed in mg/dL as the 48-hour value minus the enrollment value. The registry describes the outcome as change in total cholesterol of 0 to +5 mg/dL. The posted group summaries report mean changes and standard deviations, which describe the average change and the spread of individual changes within each group.

Cholesterol change was selected as a direct measure of whether the study drug elevated cholesterol, the stated primary purpose in the trial summary. A change score uses each participant’s enrollment value as a reference point, rather than treating the 48-hour value alone as the outcome. The registry record does not report whether the endpoint was analyzed with adjustment for enrollment cholesterol or other baseline characteristics.

The primary outcome for Phase I was maximum tolerated dose, not cholesterol change. Although Phase I participants were also analyzed for the cholesterol measure, the primary posted comparison is a Phase II comparison: 1.2 g/kg Smoflipid versus control. The record lists organ dysfunction as a secondary outcome and says that measures of organ dysfunction, mortality and biological activity were also among the secondary outcomes described in the study summary. The registry record does not provide further detail on the other outcome measures beyond the single posted secondary outcome.

7. Results Posted in the Registry

• Phase II - 1.2 g/kg Smoflipid: mean 9.00 versus 2.00 on control; difference 7.00 (95% CI -5.76 to 19.76), standardized effect d = 0.39; z-test p = 0.282.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous measure: change in total cholesterol in mg/dL. A mean summarizes the average observed change in a group. A standard deviation describes how dispersed individual changes are around that group mean; it is not the uncertainty in the mean itself. The posted results give the mean and standard deviation for each arm, which allow readers to see both the central group result and the variability among participants.

What does the mean difference represent?

The posted comparison is a mean difference between the Phase II 1.2 g/kg group and control. The group means are 9.0 mg/dL and 2.0 mg/dL, and the reported difference is 7.0 mg/dL. This contrast expresses the difference in average change between the two groups on the endpoint’s original scale. It is not a difference in individual responses, and it does not indicate that every participant’s change was close to either group mean.

The posted standard deviations are 17.0 mg/dL for the 1.2 g/kg group and 18.0 mg/dL for control. Both are larger than the corresponding group means. In general, wide variation within groups makes it harder to estimate a between-group mean difference precisely, especially when group sizes are limited. The endpoint’s observed spread and the size of the group samples are therefore important when interpreting the difference, rather than considering the two means in isolation.

How should the confidence interval be read?

The 95% confidence interval for the mean difference runs from -5.762 to 19.76 mg/dL. In general, a confidence interval is constructed by a procedure that, under its assumptions, would cover the true effect at its stated rate over repeated samples. For an individual trial, it conveys uncertainty about the estimated effect, not a guarantee that the true value lies within the interval and not a range containing a stated percentage of participants.

Here, the interval covers values on either side of zero. Zero represents no mean difference between groups on this scale. Since the interval includes zero as well as effects in both directions, it does not tightly distinguish among no difference, a negative difference, or a positive difference. The confidence interval is more informative about precision and plausible effect values than a binary label based only on a p-value.

What does the p-value add?

The registry posts an unadjusted p-value of 0.282 for the comparison. In general, a p-value describes how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions held. It is not the probability that the observed result occurred by chance, not the probability that the null is true, and not a measure of clinical importance. The registry record does not state the exact inferential test or model used for this comparison, so the p-value should be read as the posted unadjusted result rather than as evidence about an unreported analysis specification.

The p-value and confidence interval are two ways to express inferential uncertainty around the same comparison. They do not establish whether the endpoint difference matters clinically. The record does not report a threshold for a clinically important cholesterol change, nor does it report a decision rule linking this biomarker endpoint to a clinical outcome.

Why can over-dispersion matter?

As a general statistical point, over-dispersion means that observed variability is greater than a model assumes. When analysis relies on a variance model that is too small, standard errors can be understated and confidence intervals can be too narrow; tests may then appear more decisive than the data warrant. For a continuous outcome, the analogous concern is that a model’s assumed spread may not adequately represent the observed variation. The posted standard deviations make variability visible, but the registry record does not report an assessment of model fit or over-dispersion, or identify a model that would permit such an assessment.

The record labels the primary outcome as a mean with standard deviation. It does not report whether the analysis adjusted for baseline cholesterol, used a repeated-measures model, accounted for other covariates, or handled departures from distributional assumptions. Those details should not be inferred from the group summaries or the posted comparison alone.

How does the planning illustration differ from trial power?

The planning values are illustrations based on the observed effect, not a report of the trial’s original design calculations. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect defined as a stated fraction of the observed effect. These assumptions describe the planning illustration only. The registry record does not report the original sample-size assumptions or a prespecified target effect for the trial.

More generally, statistical power is the chance that a planned analysis will detect an effect of a specified size under specified assumptions. A planning calculation can help show how sample size changes when the target effect changes, but it cannot make a small trial more precise after the data are collected. The illustration should not be confused with an estimate of the probability that this completed trial detected an effect.

10. Confidence Intervals and Precision

The principal precision information for the posted comparison is the 95% confidence interval, from -5.762 to 19.76 mg/dL, around the mean difference of 7.0 mg/dL. This interval communicates appreciable uncertainty in the estimated contrast. It includes a negative value, zero, and positive values, so the observed data are compatible with more than one direction of difference under the interval’s assumptions.

In general, precision depends on the amount of information in the data and the variability of the outcome. The Phase II groups in this comparison have 10 participants in the 1.2 g/kg group and 24 in control. Their standard deviations are 17.0 mg/dL and 18.0 mg/dL. The confidence interval is therefore important context for interpreting the mean difference; the point estimate alone does not convey how uncertain that estimate is.

The interval does not say that 95% of participants experienced changes within its limits. It concerns the estimated group difference, not individual cholesterol changes. Nor does the interval measure clinical importance: the registry record does not report a threshold for a meaningful between-group difference in this outcome. The description of a change of 0 to +5 mg/dL is part of the registry’s endpoint description; the record does not explain how that range was intended to guide interpretation.

The posted p-value is 0.282 and is unadjusted. It should be interpreted alongside the interval, not as a substitute for it. The confidence interval gives a range of effect estimates compatible with the data under its calculation, while the p-value summarizes compatibility with a null model. The registry record does not identify the calculation method for the confidence interval or specify whether any covariate adjustment was used. Accordingly, the posted uncertainty should be read with those analysis details in mind.

11. 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 illustration gives participants per arm for 80% power and participants per arm for 90% power. It is not a total number of events: the outcome is continuous, and the stated assumptions specify equal allocation between two arms. The calculations use a two-sided significance level of 0.05 and a normal approximation. The effect to detect is set as a stated fraction of the observed effect, with planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect.

For planning for 100% of the observed effect, the illustration gives 101 participants per arm for 80% power and 135 participants per arm for 90% power. For planning for 75% of the observed effect, it gives 179 participants per arm for 80% power and 240 participants per arm for 90% power. For planning for 50% of the observed effect, it gives 403 participants per arm for 80% power and 539 participants per arm for 90% power.

The 80% power column represents a planning target under which the calculation is designed to detect the specified effect in 80% of repeated studies, if the assumptions hold. The 90% column uses a higher probability target and therefore calls for more participants under the same effect assumptions. Smaller effects need more participants to detect, which is why planning for a smaller fraction of the observed effect produces larger sample-size figures.

These figures are planning illustrations based on the observed effect, not observed power. They do not describe the original trial’s design calculation. The registry record does not report the original sample-size justification, or whether the trial had a different planned allocation, target effect, significance level, or power target.

12. 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 = 7.000; pooled standard deviation σ = 17.724.
2. n = 2 × (1.96 + 0.842)² × 17.724² / 7.000² = 100.6, rounded up to 101 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 <- 7                      # observed difference in means
sigma <- 17.7244534471447                     # 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(3405870)
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 = 101 (80% power), 135 (90% power)
 75% of observed effect: n per arm = 179 (80% power), 240 (90% power)
 50% of observed effect: n per arm = 403 (80% power), 539 (90% power)
power.t.test: 101.6 
Simulated power at n = 101 per arm: 0.803 (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.

13. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several hypotheses or outcomes are tested: as the number of tests increases, the chance of at least one apparently notable result can increase. The registry lists a primary cholesterol outcome for Phase II, maximum tolerated dose as the Phase I primary outcome, and organ dysfunction as a secondary outcome. The study summary also mentions mortality and biological activity among the secondary areas of interest. The record posts one inferential comparison for the Phase II cholesterol outcome, and it does not describe a multiplicity adjustment or a prespecified hierarchy for interpreting multiple outcomes or dose comparisons.

The posted p-value of 0.282 is explicitly unadjusted. It should not be interpreted as if the record had reported a familywise or other multiplicity-controlled analysis. Conversely, the existence of multiple listed outcomes does not establish that every possible comparison was tested. The full analysis plan, the planned number of comparisons, and any adjustment strategy are not reported in the registry record.

The record does not provide a complete account of missing-data handling. It does not state how many participants had missing 48-hour cholesterol measurements, whether analyses used complete cases or imputation, or whether sensitivity analyses were performed. The endpoint description identifies the Phase II analysis population as 47 patients for the primary outcome, but does not give details about eligibility for analysis or reasons for any unavailable measurements.

Analysis populations are also not fully specified. The registry record does not report whether the primary analysis followed intention-to-treat, per-protocol, or another population definition. It identifies the Phase II groups and says Phase I participants were also analyzed for cholesterol change, but does not explain how participants were assigned to the posted analysis set or how protocol deviations were handled. Randomization is listed, but that alone does not establish the population used for each analysis.

14. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: intubation (100.0%), death (75.0%), hospital readmission (11.1%), vasopressor support (4.2%), cardiac arrest (4.2%).

The registry posts serious adverse event counts by arm, alongside counts for other adverse events and deaths. For example, the Phase II 1.2 g/kg group has 2 participants affected by serious adverse events among 10 at risk; the Phase II control group has 8 among 24 at risk. The Phase II 1.4 g/kg group has 3 among 4 at risk, and the 1.6 g/kg group has 1 among 9 at risk. Phase I serious-event counts are also reported by dose group. These are descriptive counts and denominators, not adjusted comparisons.

The record separately reports deaths by arm. It does not state whether deaths overlap with serious adverse event counts, so those figures should not be added together or treated as separate sets of people. The registry’s leading serious-event terms include Intubation at 100.0%, Death at 75.0%, Hospital Readmission at 11.1%, Vasopressor Support at 4.2%, and Cardiac Arrest at 4.2%. The record does not provide further detail on the denominator or coding method for these term percentages in the registry record presented.

Adverse-event counts require context: group sizes vary, and the registry record does not report event timing, severity grading beyond the serious classification, relatedness assessment, or a statistical comparison for safety. These summaries do not establish that events were caused by the study drug or that event rates differed between groups.

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

The posted primary comparison estimates a mean difference in cholesterol change of 7.0 mg/dL for Phase II 1.2 g/kg Smoflipid versus control. The 95% confidence interval, from -5.762 to 19.76 mg/dL, indicates uncertainty that includes differences in either direction as well as zero. The unadjusted p-value is 0.282. Taken together, these statistics describe an imprecise estimate rather than a settled direction or size of difference.

This result is about a biomarker outcome measured at 48 hours. It does not by itself establish effects on organ dysfunction, mortality, or other clinical outcomes. Although the registry summary names those areas as secondary interests, detailed results for them are not reported in the registry record among the registry entries available for this analysis. The posted cholesterol result also does not show whether any observed biomarker change led to a clinical benefit or harm.

The results do not support treating the dose-group means as a dose-response pattern. The registry posts only one inferential comparison, and the other dose means are group summaries. Phase I cholesterol values are especially descriptive because Phase I’s primary purpose was maximum tolerated dose and its group sizes were small.

Safety counts are also descriptive and cannot establish causation or comparative risk. The record does not give a formal safety analysis, timing and relatedness details, or a complete explanation of how the adverse-event term percentages were constructed. As with the efficacy result, the figures are most useful when read with their group denominators and the limited analysis detail.

Overall, the registry record documents a randomized, single-masked pilot Phase I/II study and posts a primary cholesterol comparison, but does not provide the full statistical analysis plan or a broad set of outcome analyses. The findings should be understood as estimates with uncertainty and as a limited account of the trial’s reported results, not as a definitive evaluation of clinical outcomes.

16. Why This Trial Matters Statistically

This trial raises statistical issues common to early-phase studies that combine dose exploration with a controlled comparison. Phase I focused on maximum tolerated dose, while Phase II specified cholesterol change as the primary outcome. Those aims answer different questions: dose exploration concerns tolerability and dose selection, whereas a controlled endpoint comparison estimates a difference between groups. Keeping the phases and their purposes distinct prevents descriptive Phase I results from being mistaken for the primary Phase II analysis.

The primary endpoint is a continuous biomarker, and the posted comparison includes a mean difference, variability estimates, a confidence interval and an unadjusted p-value. This gives readers more context than a point estimate alone. In particular, the interval spans values on both sides of zero, highlighting that a group difference estimate can be uncertain even when numerical group means differ. Statistical interpretation must account for that uncertainty, not simply compare the means.

The small arm-level samples, alongside the observed variability, also illustrate why early studies often yield imprecise estimates. The planning illustration shows how required sample size changes with the target effect and desired power. Such calculations can inform the design of later research, but they rely on assumptions and an observed effect that may itself be unstable in a small study. They do not substitute for the original trial rationale, which is not reported in the registry record.

Finally, this record shows the importance of reporting analysis populations, missing-data methods, multiplicity plans and model details. Without those details, readers can describe the posted result but cannot fully assess the analysis choices that produced it or the range of analyses that may have been considered. This is particularly relevant in a trial with several dose groups and multiple stated outcome areas.

17. 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 specific to this trial follow from the available registry record. The Phase II primary comparison posted is limited to the 1.2 g/kg dose and control. The record does not provide inferential comparisons for the other Phase II doses, so their means should not be used to infer statistically supported differences or a dose-response relationship.

The Phase I dose groups have very small arm-level samples, and cholesterol change was not the Phase I primary outcome. The registry describes maximum tolerated dose as the Phase I primary outcome, but detailed results for that outcome are not included in the registry record. The Phase I cholesterol summaries therefore do not provide a strong basis for comparing dose effects.

The endpoint’s stated analysis population and the overall enrollment are different quantities: the registry lists enrollment of 59 and says 47 Phase II patients were analyzed for the primary outcome. The record does not explain the difference, identify missing 48-hour measurements, or describe exclusions and analysis-set rules. It also does not report baseline adjustment, the specific inferential model, missing-data sensitivity analyses, or the original sample-size rationale.

The safety record provides event counts by arm but lacks detail on event timing, relatedness, and the construction of the serious-event term percentages. It does not report formal between-group safety analyses or clarify overlap between serious adverse events and deaths. These gaps limit interpretation of safety comparisons. More broadly, the registry record does not provide detailed results for organ dysfunction, mortality, or biological activity despite identifying these as study interests.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: LIPIDS-P Trial Phase I/II Trial — NCT03405870.
• Registry record last updated 2025-08-11; status last verified 2025-02.
• Results first posted 2025-08-11.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

LIPIDS-P is listed as a completed randomized, single-masked, parallel Phase I/II trial sponsored by the University of Florida, with enrollment of 59. The study evaluated Smoflipid doses in patients with sepsis and moderate organ dysfunction, with a no-drug active control in Phase II.

Phase II’s primary outcome was change in total cholesterol at 48 hours. The posted comparison of 1.2 g/kg Smoflipid with control reports a mean difference of 7.0 mg/dL, a 95% confidence interval from -5.762 to 19.76 mg/dL, and an unadjusted p-value of 0.282. The interval indicates substantial uncertainty around the estimated difference.

Phase I’s primary outcome was maximum tolerated dose, while Phase II also listed organ dysfunction as a secondary outcome. The registry record does not report detailed results for most other outcomes or the full statistical analysis and missing-data plans. Safety counts are descriptive and do not establish causation or comparative risk.