Study of the Nasal Decolonisation of Staphylococcus Aureus (SA) and the Safety and Tolerability of XF-73 Nasal Gel in Healthy Subjects: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of the Nasal Decolonisation of Staphylococcus Aureus (SA) and the Safety and Tolerability of XF-73 Nasal Gel in Healthy Subjects (NCT02282605), a randomized phase 1, phase 2 study. The primary outcome is Apparent Eradication of Nasal SA After the Last Dose of XF-73 Based on Semi-quantitative SA Scores From a Broth Enrichment Method..

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, placebo-controlled, parallel-group, single-centre Phase I/II study. The registry identifies Destiny Pharma Plc as the sponsor and lists the study as completed. Its recorded start is 2014-09 and primary completion is 2014-11. The registry enrollment is 60. Details of the randomization sequence, allocation concealment, blinding procedures, and the number of centres beyond the single-centre description are not reported in the registry record.

The three recorded arms were XF-73 2.0 mg/g nasal gel, XF-73 0.5 mg/g nasal gel, and placebo nasal gel. Each arm description included twice-daily nasal gel application for two days and chlorhexidine gluconate 2% body and face cloths before each morning dose. The active arms differed in the concentration of XF-73. The registry describes the study aim as assessing efficacy, safety, and tolerability in eradicating nasal carriage of Staphylococcus aureus.

The primary endpoint was apparent eradication assessed 48 hours after the last dose (Day 4, 84 hours). The registry also lists secondary assessments of apparent eradication after the last dose, the time-point at which clearance was first observed, the area under the curve of semi-quantitative scores, and changes in vital signs, ECG, and routine laboratory tests. The record lists 5 outcome measures and 6 analyses. Details of the planned sample-size justification, the full statistical analysis plan, interim analyses, and analysis populations are not reported in the registry record.

4. Participant Flow and Accounting

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

Before assignment: Persistent nasal SA carriage was defined by 3 separate, SA positive cultures from nasal swabs. The protocol allowed for dosing to commence before the results of the final swab on day -1 were available. Dosing was thus commenced in 3 subjects in the XF-73 2.0mg/g group but was withdrawn as the day -1 SA results were negative.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Exploratory comparisons between active groups versus placebo of the number of Responder subjects were performed using a one-sided Fisher's exact test at 5% significance 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.

6. Primary Endpoint

The registry lists the primary outcome as Apparent Eradication of Nasal SA After the Last Dose of XF-73 Based on Semi-quantitative SA Scores From a Broth Enrichment Method., measured over The primary endpoint was 48 hours after the last dose (Day 4, 84 hours)..

The primary endpoint counted participants classified as having apparent eradication of nasal Staphylococcus aureus after treatment. Assessment used a broth-enriched semi-quantitative culture method with a 0-6 point scale. Under the registry definition, negative and 0 scores indicated absence of SA and counted as a responder; scores of 1 or greater indicated its presence and counted as a non-responder.

The primary time point was 48 hours after the last dose, specified as Day 4, 84 hours. This endpoint turns a culture score into a binary classification, then compares the number of responders between each active group and placebo. It is a measure of apparent absence at the specified assessment time, not a measure of every aspect of colonisation over time. The record lists other assessments of clearance timing and semi-quantitative scores, but they are separate from the primary responder count.

Binary endpoints are commonly used when the clinical or laboratory question is defined by whether a participant meets a criterion. General statistical point: this simplifies the outcome into two categories, but it does not preserve all the gradations in the underlying score. The registry’s endpoint definition sets the classification rule; additional details about validation of the scale or the handling of indeterminate results are not reported in the registry record.

7. Results Posted in the Registry

• XF-73 2.0 mg/g Nasal Gel: 11.1% (3/27) versus 16.7% (4/24) on control; risk difference -5.6 percentage points (95% CI -24.6 to 13.5), risk ratio 0.67 (0.17–2.68); unadjusted two-proportion z-test p = 0.565.
• XF-73 0.5 mg/g Nasal Gel: 8.3% (1/12) versus 16.7% (4/24) on control; risk difference -8.3 percentage points (95% CI -29.9 to 13.3), risk ratio 0.50 (0.06–4.00); unadjusted two-proportion z-test p = 0.496.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary: each participant is classified as a responder or non-responder according to the culture score. The registry defines a responder as a participant with a negative or 0 score and a non-responder as a participant with a score of 1 or greater. For each comparison, the posted counts are the number of responders and the number assessed in the arm. A risk is the proportion of participants in an arm who meet the responder definition. The risk difference contrasts the active-arm risk with the placebo risk; a negative estimate means the observed responder risk was lower in the active group under that subtraction direction.

Why use Fisher’s exact test?

The registry describes exploratory active-versus-placebo comparisons of responder counts using a one-sided Fisher’s exact test at a 5% significance level. Fisher’s exact test is a method for comparing categorical outcomes in a two-group table. It calculates a probability for the observed arrangement of counts and other arrangements at least as extreme under its null model, conditional on the margins of the table. It is often used when counts are small, as is relevant to the posted responder counts. General statistical explanation: unlike a large-sample approximation, the exact procedure uses the discrete possible arrangements in the observed table.

The posted re-calculations present risk differences with confidence intervals and unadjusted p-values. These summaries address related but distinct questions. The risk difference estimates the magnitude and direction of the contrast; the confidence interval expresses uncertainty around that estimate under the calculation’s assumptions. The p-value describes how unusual the data would be under a specified null hypothesis and test procedure. It is not an effect-size estimate and does not indicate the probability that a null hypothesis is true.

What does a confidence interval convey?

General statistical explanation: a confidence interval is a range generated by a procedure designed to capture the underlying parameter at a stated rate over repeated use of that procedure, if its assumptions hold. It should be read alongside the point estimate and the endpoint definition. An interval that includes zero for a risk difference is compatible with no difference under the interval procedure, while also allowing differences in either direction. It does not prove equivalence or show that every value in the interval is equally plausible.

The re-calculated intervals are unadjusted comparisons based on the posted summary counts. They are not described as the sponsor’s model-based or protocol analysis, and no adjusted estimate is reported in the registry record. The sponsor’s Fisher Exact p-values are separately reported and should be distinguished from the simple re-calculated p-values. The registry does not report confidence intervals for the sponsor analyses.

Why are there more than one comparison?

Each active concentration is compared with the placebo group. This produces more than one primary-outcome contrast, alongside the additional secondary assessments listed in the record. General statistical explanation: when multiple hypotheses are tested, the chance of at least one nominally notable result can increase if each test is interpreted separately. A multiplicity strategy can address this, but a correction or hierarchy is not reported in the registry record. The record also does not state whether the exploratory comparisons were intended to support confirmatory claims.

The primary outcome is a count, not a continuous score in the posted primary analysis. The underlying culture method is semi-quantitative, but applying the responder threshold discards distinctions between the positive score categories. The secondary outcomes include score-based measures and time to first observed clearance, which would require their own methods and interpretation. Those methods and their results are not reported in the registry record described for this primary comparison.

Finally, the result applies to the recorded endpoint, assessment time, arms, and study population. General statistical explanation: randomization can support causal comparison when the design and analysis are properly carried out, but interpretation also depends on allocation procedures, follow-up, outcome ascertainment, and the handling of missing observations. Details of those procedures and the analysis population are not reported in the registry record, so the posted counts should be read as the reported group results rather than as a complete account of every analytic choice.

10. Confidence Intervals and Precision

The re-calculated risk difference for XF-73 2.0 mg/g versus placebo is -0.056, with a 95% confidence interval from -0.246 to 0.135. For XF-73 0.5 mg/g versus placebo, the risk difference is -0.083, with a 95% confidence interval from -0.299 to 0.133. Both intervals include zero and extend into positive and negative values. The estimates therefore have limited precision: the data do not pin down the direction or size of the underlying difference closely.

General statistical explanation: a confidence interval reflects sampling uncertainty under the method used to calculate it; it is not a guarantee that the true value lies in that particular interval. The width of an interval is influenced by the amount of information in the data and by the observed outcome pattern. With few responder events in the posted counts, the estimates are especially sensitive to the pattern of a small number of outcomes. This is one reason to read the interval rather than relying only on the point estimate.

These intervals belong to the simple unadjusted re-calculations. The sponsor’s own Fisher Exact analyses are reported with p-values but without confidence intervals in the registry record. The intervals do not account for any multiplicity adjustment, and no such adjustment is reported. They also do not address uncertainty from outcome measurement, missing data, or generalizability beyond the study population. Those issues require information beyond the interval calculation itself.

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 sample-size figures are planning illustrations based on the observed effect, not a reconstruction of the trial’s original design. The listed 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 set to the stated fraction of the observed effect. The table gives participants per arm for the 80% and 90% power targets; these are not total event counts. The record’s note that time-to-event tables give total events does not apply to these binary-outcome figures.

The 80% power column gives the illustrative participants per arm for a design with an 80% chance of detecting the specified effect under the assumptions, if that effect and the other planning conditions hold. The 90% power column gives the corresponding illustrative participants per arm for a higher target chance of detection. For the 2.0 mg/g comparison, planning for 100% of the observed effect gives 608 participants per arm for 80% power and 813 for 90% power; planning for 75% gives 1,126 and 1,506; planning for 50% gives 2,633 and 3,524. For the 0.5 mg/g comparison, the corresponding figures are 247 and 329; 470 and 628; and 1,126 and 1,506.

Smaller effects require more participants to detect under the same assumptions, which is why the figures increase when planning for a smaller fraction of the observed effect. These are hypothetical planning calculations, not observed power. The trial’s own original sample-size assumptions and justification are not reported in the registry record. The curve should be read as a visual representation of the planning relationship, not as an additional observed trial result.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.167; treatment proportion p₁ = 0.111; average p̄ = 0.139.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4891 = 0.9586.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4875 = 0.4103.
4. n = (0.9586 + 0.4103)² / (-0.0556)² = 607.1, rounded up to 608 participants per arm for 80% power.

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

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.166666666666667                          # control proportion
p1 <- 0.111111111111111                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2282605)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 608 (80% power), 813 (90% power)
 75% of observed effect: n per arm = 1,126 (80% power), 1,506 (90% power)
 50% of observed effect: n per arm = 2,633 (80% power), 3,524 (90% power)
power.prop.test: 607.1 
Simulated power at n = 608 per arm: 0.803 (20000 replicates)

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

13. Analyses Reported by the Sponsor

The registry reports the sponsor’s Fisher Exact p-value as 0.792 for XF-73 2.0 mg/g versus placebo and 0.887 for XF-73 0.5 mg/g versus placebo. The simple unadjusted re-calculations from the posted summary counts give p-values of 0.565 and 0.496, respectively, with risk differences and confidence intervals also reported. These are distinct analyses in the record; the posted sponsor p-values should not be replaced by the re-calculated values.

The registry describes the planned exploratory comparisons as one-sided Fisher’s exact tests at a 5% significance level. It does not provide the sponsor’s confidence intervals or the treatment-effect estimates associated with those p-values. The re-calculations are unadjusted, summary-statistic calculations and do not establish that they reproduce the full sponsor analysis. The registry does not explain why the reported p-values differ from the re-calculated values. The record’s discrepancy indicator is false, so the posted results are not flagged as discrepant.

For interpretation, use the sponsor’s reported analyses as the trial’s posted analysis results, while treating the re-calculations as a separate aid to understanding the counts and approximate contrast. Neither set of p-values alone conveys the full size and precision of the treatment contrast.

14. Multiplicity, Missing Data and Analysis Populations

The primary endpoint involved two exploratory active-versus-placebo comparisons: one for each XF-73 concentration. The record also lists secondary assessments of apparent eradication, first observed clearance, area under the curve of semi-quantitative scores, and changes in vital signs, ECG, and routine laboratory tests. It lists 5 outcome measures and 6 analyses. This means the record describes multiple statistical questions, not a single isolated test.

General statistical explanation: when multiple comparisons are made, interpreting every result against the same nominal threshold can increase the chance of a false-positive finding across the set. Researchers may prespecify a testing order, adjust significance thresholds, or identify analyses as exploratory. The registry says the primary comparisons were exploratory, but a multiplicity adjustment, testing hierarchy, or other family-wise error strategy is not reported in the registry record. It is therefore not possible to determine from this record how multiplicity was handled across all outcomes and analyses.

The missing-data approach is not reported in the registry record. The record does not state whether all randomized participants were included in the primary analysis, how participants without an endpoint culture were handled, or whether missing observations were imputed or excluded. The primary population is described as exploratory comparisons of the number of responder subjects, but a named analysis population and its inclusion rules are not reported. Details on protocol deviations and withdrawals relevant to the endpoint are also not reported in the registry record.

These gaps matter because analysis choices can affect both the denominator and the interpretation of a responder count. The posted arm counts and results remain informative as recorded, but the registry does not provide enough detail to assess all prespecified analyses, missing-data sensitivity, or the relationship between the primary and secondary testing procedures.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The serious adverse event counts listed are 0 for XF-73 2.0 mg/g, 0 for XF-73 0.5 mg/g, and 0 for placebo. The corresponding numbers at risk are 27, 12, and 24. These figures describe serious adverse events recorded in the registry for the listed groups; they do not establish that no serious events could occur outside the observed study experience. Death counts are not reported in the registry record.

The record also lists participants with other, non-serious adverse events: 9 in the 2.0 mg/g group, 2 in the 0.5 mg/g group, and 5 in the placebo group. These are separate from the serious adverse event counts. The registry does not provide event terms in the listed serious-event summary, and no serious-event terms are listed. Further detail about types, timing, attribution, severity grading, or withdrawals due to adverse events is not reported in the registry record.

These safety counts should be read descriptively. They do not establish comparative safety, particularly without additional event-level detail and a prespecified comparative analysis.

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

The posted primary outcome shows responder counts of 3.0 among 27.0 participants in the XF-73 2.0 mg/g group, 1.0 among 12.0 in the XF-73 0.5 mg/g group, and 4.0 among 24.0 in the placebo group. Under the registry’s definition, response meant a negative or 0 broth-enriched semi-quantitative culture score at 48 hours after the last dose (Day 4, 84 hours). The results describe apparent eradication at that assessment time.

The re-calculated risk differences are negative for both active-versus-placebo comparisons, but their confidence intervals include zero and extend in both directions. The sponsor’s reported Fisher Exact p-values are 0.792 and 0.887. Together, these results do not offer a precise estimate of the difference between groups. They should not be translated into a claim that XF-73 does or does not have an effect.

The finding does not establish sustained clearance, the effect on clinical infection, or outcomes beyond the recorded culture threshold and time point. The intervention included a shared chlorhexidine gluconate 2% washing regimen in all arms, so the comparisons concern the assigned nasal gel alongside that regimen. The record does not provide results for all listed secondary outcomes in the registry record for this interpretation.

The study was conducted in healthy subjects and is recorded as a single-centre Phase I/II trial. Its findings therefore describe this study context, not necessarily other populations, treatment schedules, or settings. Safety counts are also descriptive and do not by themselves establish comparative safety. Further details on missing data, analysis populations, and multiplicity are not reported in the registry record.

The appropriate statistical reading is limited: the posted counts and analyses provide a snapshot of the responder outcome in the recorded groups, with substantial uncertainty around the re-calculated contrasts. They do not settle a broader clinical question.

17. Why This Trial Matters Statistically

This study illustrates why a binary endpoint should be interpreted alongside the rule used to create it. The primary result depends on classifying negative and 0 scores as responders and scores of 1 or greater as non-responders. The underlying method is semi-quantitative, but the primary comparison reduces its scale to a yes-or-no outcome. That makes the endpoint straightforward to count while setting aside variation among score levels.

It also illustrates the distinction between an effect estimate, its precision, and a test result. The re-calculated risk differences describe the observed contrasts, their confidence intervals show broad uncertainty, and the sponsor’s Fisher Exact p-values summarize evidence under a separate test procedure. These quantities answer related but different statistical questions. A p-value alone does not communicate the magnitude or precision of a difference.

Small responder counts make interpretation especially dependent on a limited number of observations. The confidence intervals extend across zero, so the direction and size of the underlying difference are not tightly determined by the posted summaries. The planning illustrations further show how much larger future studies might need to be under specified assumptions, while remaining distinct from the original trial design and from observed power.

Finally, the record brings together an exploratory primary comparison, several listed secondary outcomes, and limited detail about multiplicity and missing-data handling. This is a reminder that statistical interpretation depends not only on the headline endpoint but also on the analysis plan, the population analyzed, and how multiple questions are handled. Where those details are not reported, conclusions should stay close to the observable counts and estimates.

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.

The primary endpoint is a thresholded measure of apparent eradication at one stated time point. It does not represent every possible interpretation of the semi-quantitative culture score, and the result cannot be taken as evidence of sustained clearance. The registry does not report validation details for the scale, handling of indeterminate results, or whether outcome assessors remained blinded during culture assessment.

The trial record describes a single-centre study in healthy subjects. Baseline characteristics, detailed eligibility criteria, and information needed to assess how representative the participants were are not reported in the registry record. The record also does not report the allocation-concealment process, randomization details, or operational details of masking.

For the primary endpoint, sponsor confidence intervals and treatment-effect estimates are not reported. The simple recalculations are unadjusted and should not be confused with the sponsor’s Fisher Exact analyses. The registry does not explain the difference between the posted sponsor p-values and the re-calculated p-values. Multiplicity adjustment, missing-data handling, the named analysis population, and the original sample-size justification are also not reported in the registry record.

The primary comparisons are identified as exploratory. Secondary outcomes are listed, but their results are not included in the registry record for this interpretation. Safety information is limited to group counts for serious and other adverse events; event descriptions, attribution, timing, and detailed follow-up are not reported in the registry record. These limitations restrict how far the results can be generalized or used to make claims beyond the recorded endpoint and study context.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study of the Nasal Decolonisation of Staphylococcus Aureus (SA) and the Safety and Tolerability of XF-73 Nasal Gel in Healthy Subjects — NCT02282605.
• Registry record last updated 2016-10-07; status last verified 2016-06.
• Results first posted 2016-10-07.
• 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

NCT02282605 is a completed randomized, double-blind, placebo-controlled, parallel-group Phase I/II study sponsored by Destiny Pharma Plc. It studied XF-73 nasal gel at 2.0 mg/g and 0.5 mg/g against placebo in healthy subjects, with chlorhexidine gluconate 2% body and face cloths used before each morning dose in all arms. The registry lists an enrollment of 60 and describes a single-centre study.

The primary outcome was the number of responders at 48 hours after the last dose (Day 4, 84 hours), using a semi-quantitative broth-enriched culture definition. Posted counts were 3.0, 1.0, and 4.0 responders in the 2.0 mg/g, 0.5 mg/g, and placebo groups, respectively. The record reports Fisher Exact p-values of 0.792 and 0.887 for the active-versus-placebo comparisons. The re-calculated risk-difference intervals are broad and include zero. These data describe the recorded endpoint and do not by themselves establish a broader clinical conclusion.