Study of Inhaled Carbon Monoxide to Treat Idiopathic Pulmonary Fibrosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Inhaled Carbon Monoxide to Treat Idiopathic Pulmonary Fibrosis (NCT01214187), a randomized phase 2 study. The primary outcome is Serum MMP7 Level.

2. Clinical Question

3. Trial Design

The registry describes a randomized, double-masked, parallel-group Phase II study sponsored by Brigham and Women's Hospital. The study is listed as completed, with enrollment of 58. Its purpose was to determine whether low-concentration inhaled carbon monoxide was effective in treating idiopathic pulmonary fibrosis. The registry lists inhaled carbon monoxide and Oxygen 21% as the interventions.

The carbon monoxide arm description specifies 100-200 ppm, two times weekly for two hours per dose, for 12 weeks. The Oxygen 21% arm has no descriptive text in the registry record. The allocation method beyond being randomized, the precise masking procedures, and the details of how the comparator was administered are not reported in the registry record.

The primary endpoint was serum MMP7 change from baseline to Week 12. Secondary measures listed in the registry include total lung capacity percent predicted, diffusing capacity for carbon monoxide percent predicted, six-minute walk distance, and the St George's Respiratory Questionnaire. Results are posted for the first three of these secondary measures, but a result for the questionnaire is not reported in the registry record.

For MMP7, the posted analysis includes 29 participants in the carbon monoxide group and 28 in the oxygen group. The registry says one person randomized to placebo withdrew consent before the MMP7 blood draw at visit 2, accounting for the oxygen-group count of 28. The secondary endpoint summaries list 29 participants per group. The record does not give a fuller account of analysis populations, participant flow, or reasons for any other missing endpoint data.

The registry record does not report the study's original sample-size justification, a detailed statistical analysis plan, or whether interim analyses were conducted. Those details should not be inferred from the sample-size illustrations or the posted outcome summaries.

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: Sixty-five subjects were screened, of which 7 subjects were screen failures. Reasons for screen failure included subjects who completed screening changed their mind about participation before visit 2, FEV1/FVC less than 70% predicted, and infection within 1 month before screening. The screening period up to 28 days from date of consent to Visit 2.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: One subject randomized to placebo withdrew consent prior to the MMP7 blood draw at visit 2. This is missing data and the reason why we have n=28 for placebo group.

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 Serum MMP7 Level, measured over Baseline to Week 12.

The primary endpoint was change in serum MMP7 concentration, measured in ng/ml, from baseline to Week 12. The registry reports least-squares means with standard errors for the two groups. The displayed changes are -0.15 for carbon monoxide inhalation and 0.88 for Oxygen 21%. These are group-level summaries of change over the stated interval.

The registry describes serum MMP7 as easily measurable in peripheral blood and as reflecting changes in the alveolar microenvironment. It says the measure was chosen as a surrogate biomarker for the effect of inhaled carbon monoxide on disease progression after three months of treatment. A surrogate biomarker can be useful for studying biological changes, but a difference in a biomarker does not by itself establish a difference in symptoms, function, or longer-term disease course. Those broader interpretations require evidence from relevant clinical outcomes and a justified relationship between the surrogate and those outcomes.

The posted primary comparison estimates a mean difference of 1.03 for Oxygen 21% relative to carbon monoxide, with a recalculated confidence interval from -0.564 to 2.624 and an unadjusted p-value of 0.205. The registry's reported ANOVA result gives a net mean difference of -1.04 and p = 0.207; the sign reflects the reverse contrast direction. The result is imprecise and the interval includes zero, so the posted summary does not identify a clear between-group difference in the mean MMP7 change.

7. Results Posted in the Registry

• Oxygen 21%: mean 0.88 versus -0.15 on control; difference 1.03 (95% CI -0.56 to 2.62), standardized effect d = 0.34; z-test p = 0.205.

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.

Total Lung Capacity % Predicted Values (TLC)

• 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:
  • Oxygen 21%: mean -1.44 versus -2.29 on control; difference 0.85 (95% CI -2.23 to 3.93), standardized effect d = 0.14; z-test p = 0.588.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Oxygen 21% and the control. The interval gives the range of effects that remain compatible with the data.

Diffusing Capacity for Carbon Monoxide (DLCO) % Predicted Values

• 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:
  • Oxygen 21%: mean 0.46 versus 1.10 on control; difference -0.64 (95% CI -4.43 to 3.15), standardized effect d = -0.09; z-test p = 0.740.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Oxygen 21% and the control. The interval gives the range of effects that remain compatible with the data.

Six Minute Walk Distance

• 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:
  • Oxygen 21%: mean 12.92 versus -35.54 on control; difference 48.46 (95% CI 12.90 to 84.02), standardized effect d = 0.70; z-test p = 0.008.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Oxygen 21% than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary endpoints provide a broader set of functional and physiological measurements alongside the primary serum biomarker. The posted pattern is not uniform: the lung-volume and gas-transfer comparisons are uncertain in the recalculations, while the walk-distance comparison is more separated. The St George's Respiratory Questionnaire is listed as a secondary outcome, but its result is not reported in the registry record.

These measures relate to the primary result by addressing different aspects of the study question. MMP7 is described as a surrogate biomarker, whereas the secondary measures include lung function, walking distance, and a respiratory health questionnaire. Agreement across different kinds of measures could provide complementary evidence, while differences in their patterns may reflect that they measure distinct things. The registry summaries alone do not establish why the outcome patterns differ.

Secondary results generally carry less weight than a primary result when many outcomes are examined, particularly if the analysis plan and multiplicity handling are not available. Each additional comparison creates more opportunity for a nominally unusual result. The registry record does not report a multiplicity adjustment, a hierarchy among secondary outcomes, or the full prespecified statistical plan. The secondary p-values therefore should be treated as exploratory evidence rather than as independent confirmation of the primary finding. The posted results also do not state whether the listed secondary endpoints were all analyzed in the same population or with identical modeling choices.

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

• Total Lung Capacity % Predicted Values (TLC)
• Diffusing Capacity for Carbon Monoxide (DLCO) % Predicted Values
• Six Minute Walk Distance
• St George's Respiratory Questionnaire

9. Statistical Interpretation

10. Statistical Methods Explained

The posted primary summaries are least-squares means for a continuous outcome: change in serum MMP7 from baseline to Week 12. The accompanying dispersion is a standard error. A mean describes the central value of the measured changes across participants, while a standard error describes the uncertainty in the estimated mean. It is not the same as the spread of individual participant measurements. The registry reports group means of -0.15 and 0.88 ng/ml, with standard errors of 0.58 and 0.57.

The recalculated comparison is a simple, unadjusted difference between the posted group means. It is presented as Oxygen 21% minus carbon monoxide, giving 1.03 ng/ml. The confidence interval, from -0.564 to 2.624, quantifies uncertainty around that estimated contrast under the recalculation's assumptions. A confidence interval is a range of values compatible with the data and method; it is not a guarantee that the true value lies in that particular interval. An interval spanning zero indicates that the data do not distinguish the estimated contrast from no difference at the chosen confidence level.

The recalculated p-value is 0.205. A p-value measures how unusual data at least as discordant with a specified no-difference model would be, if that model and the analysis assumptions were correct. It is not the probability that the no-difference model is true, and it does not measure the size or clinical importance of a difference. The registry separately reports an ANOVA analysis, with p = 0.207. The ANOVA is the sponsor's reported analysis; the unadjusted recalculation is not a replacement for it.

The sign of a mean difference depends on which group is subtracted from which. The recalculated estimate uses Oxygen 21% minus carbon monoxide. The registry's reported net mean difference is -1.04, consistent in direction with carbon monoxide minus Oxygen 21%. Comparisons should therefore be read with their stated contrast direction rather than by the sign alone. The record does not provide the full ANOVA specification, covariates, or detailed analysis plan, so the model's implementation cannot be reconstructed from the registry summary.

For the recalculated confidence intervals and p-values, the posted summary statistics are used without adjustment. The standard deviations used in that recalculation are approximated from the reported standard errors and group sizes. That approximation is explicitly indicated in the analysis data. General statistical principle: interval estimates and tests based on summary-statistic approximations may differ slightly from the trial's own analysis, especially when the sponsor used a model or variance estimate not captured by the posted summaries. In this trial, the primary recalculated and reported p-values are close, but the reported ANOVA remains the reference for the sponsor's analysis.

The secondary endpoints are also continuous measures, summarized as least-squares mean changes with standard errors. The same basic ideas apply: a group mean describes average change, a mean difference describes the contrast, and a confidence interval communicates uncertainty. The total lung capacity and diffusing capacity recalculations have intervals that include zero. The six-minute walk recalculation has an interval from 12.9 to 84.02 meters, while its posted group means are -35.54 meters and 12.92 meters. Its sponsor-reported ANOVA p-value is 0.010. These are separate outcome comparisons, and their p-values do not account for the number of outcomes considered.

General statistical principle: testing several outcomes increases the chance that at least one will appear unusual under a no-effect model, even when no real effects are present. Multiplicity procedures can address that issue, but the registry record does not report whether an adjustment was planned or used. The count of five outcome measures and five analyses is not enough to infer a multiplicity strategy or identify all prespecified hypotheses. The St George's Respiratory Questionnaire is listed as a secondary outcome, but a result for it is not reported in the registry record.

Missing data matter because the primary endpoint was not available for one participant assigned to the group described as placebo in the registry's missing-data note. The arm title elsewhere is Oxygen 21%. The record states that this person withdrew consent before the MMP7 blood draw at visit 2. It reports 28 in the oxygen group for the primary endpoint and 29 in the carbon monoxide group. It does not report a detailed missing-data method, whether any imputation was used, or how analysis populations were defined. The secondary endpoint summaries list 29 per group, but that does not establish that every endpoint had complete data for every participant.

General statistical principle: over-dispersion arises when observed variability exceeds what a chosen statistical model assumes. It can make standard errors too small and tests too optimistic if ignored. The posted outcomes are continuous, and no over-dispersion assessment is described in the registry record; it would be inappropriate to infer that such an issue was assessed or present. Likewise, randomization and double masking are listed design features, but detailed procedures for allocation concealment, masking, model diagnostics, and interim monitoring are not reported in the registry record.

11. Confidence Intervals and Precision

The primary recalculated mean difference is 1.03 ng/ml, with a confidence interval from -0.564 to 2.624. The interval is wide enough to include zero and values on either side of it. That means the observed summary does not pin down the direction or size of the population contrast with high precision. The p-value of 0.205 is consistent with this uncertainty, but it should not be read as a measure of how likely the result is true.

The confidence interval is recalculated from posted summary statistics, using an approximate standard deviation, and is not the interval from the sponsor's ANOVA. The registry record reports no confidence limits for that ANOVA. Its p-value is 0.207, and the net mean difference is -1.04 with the contrast expressed in the reverse direction. The close p-values do not make the methods identical, and the sponsor's analysis remains the reference for the trial's own result.

For the secondary outcomes, the recalculated confidence intervals are also useful for judging precision. The total lung capacity interval runs from -2.227 to 3.927 percentage points; the diffusing capacity interval runs from -4.426 to 3.146 percentage points. Each spans zero and includes differences in either direction. The six-minute walk interval runs from 12.9 to 84.02 meters and does not span zero under the recalculation, but it is still an estimate with a range of uncertainty. Its sponsor-reported ANOVA p-value is 0.010.

General statistical principle: a confidence interval combines information about the estimate and its uncertainty, whereas a p-value answers a different question about compatibility with a specified null model. Neither alone establishes clinical importance. No minimally important differences are reported in the registry record, so these estimates cannot be classified against such thresholds from the posted information. Precision also depends on the population and analysis assumptions; the recalculated intervals should not be mistaken for intervals reported by the trial investigators.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size illustrations describe hypothetical future planning based on the observed effect, not the trial's original design. For continuous outcomes such as these, the figures are participants per arm, not total events. The planning assumptions listed 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 80% power column gives the illustrative number of participants per arm that would yield an 80% chance of detecting an effect of the specified size under those assumptions, if that effect and the planning model were correct. The 90% power column describes the corresponding illustration for a 90% chance. Power is a property of a planned design under assumed conditions; it is not the probability that the observed result is correct.

The rows use the full observed effect and smaller fractions of it. As the effect to detect becomes smaller, the required number of participants per arm increases. This pattern reflects the greater information needed to distinguish a smaller difference from random variation. The estimates are illustrations generated from the observed effects, not the study's own sample-size justification. The original design assumptions and justification are not reported in the registry record. These figures are not observed power.

The illustrated requirements can vary substantially by outcome because the observed contrasts and variability differ. They should be read as conditional planning aids, not as recommendations or as evidence that a particular future study would achieve the stated power. They also do not address issues not specified in the planning assumptions, such as multiplicity, missing outcomes, or details of a model-based analysis.

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 = 1.030; pooled standard deviation σ = 3.071.
2. n = 2 × (1.96 + 0.842)² × 3.071² / 1.030² = 139.6, rounded up to 140 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 <- 1.03                      # observed difference in means
sigma <- 3.0712188874363                     # 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(1214187)
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 = 140 (80% power), 187 (90% power)
 75% of observed effect: n per arm = 249 (80% power), 333 (90% power)
 50% of observed effect: n per arm = 559 (80% power), 748 (90% power)
power.t.test: 140.5 
Simulated power at n = 140 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. Analyses Reported by the Sponsor

The registry reports an ANOVA for the primary outcome with a net mean difference of -1.04 and p = 0.207. The unadjusted recalculation from posted summary statistics gives a mean difference of 1.03 and p = 0.205, with Oxygen 21% minus carbon monoxide as the contrast. The sign changes because the contrast direction is reversed. The registry's analysis and the recalculation therefore express the group comparison in opposite directions; their p-values are close.

The sponsor's own analysis takes precedence over a simple recalculation from summary statistics. The recalculation is useful for a transparent cross-check, but it may not reproduce the ANOVA's full model or variance calculation. The registry record does not give enough detail about the ANOVA specification to reconstruct it.

For the posted secondary analyses, the same direction issue applies. The sponsor reports -0.85 for total lung capacity, 0.64 for diffusing capacity, and -48.46 for six-minute walk distance, with p-values of 0.588, 0.740, and 0.010, respectively. Recalculated contrasts are expressed using Oxygen 21% minus carbon monoxide, so their signs need not match the sponsor's net differences. No confidence intervals for the sponsor-reported analyses are included in the registry record. Comparisons between the reported and recalculated results should focus on the contrast definition and the stated analysis, rather than treating a sign difference as a contradiction.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and five analyses, including the primary endpoint and secondary measures. It posts results for the primary MMP7 measure and three secondary measures. A result for the St George's Respiratory Questionnaire is not reported in the registry record. The record does not state whether the analyses were organized in a prespecified hierarchy or whether p-values were adjusted for multiple comparisons.

General statistical principle: when multiple outcomes or hypotheses are tested, the chance of at least one apparently positive result can increase. A multiplicity adjustment or a prespecified testing hierarchy can help control that risk, but neither should be assumed without documentation. The unadjusted recalculated p-values are not multiplicity-adjusted results. The registry's reported p-values are given as posted, but the record does not say whether they reflect any adjustment. This uncertainty is especially relevant when interpreting an isolated secondary result.

Missingness is noted for the primary endpoint: one participant randomized to placebo withdrew consent before the MMP7 blood draw at visit 2, leaving 28 in the group called placebo in that note. The arm is otherwise titled Oxygen 21%. The carbon monoxide primary group has 29 participants. The registry does not describe the missing-data assumptions, any imputation, or sensitivity analyses. It also does not provide a full analysis-population definition or participant flow for all outcomes. Secondary summaries list 29 participants in each arm, but that information alone does not clarify how missing observations were handled across measures.

The original sample-size justification, the trial's detailed statistical analysis plan, and any interim analyses are not reported in the registry record. The sample-size illustrations based on observed effects do not answer those design questions. Without the plan, it is not possible to establish which analyses were prespecified, how multiplicity was addressed, or whether the posted analyses correspond to the intended primary analysis in all details.

16. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (6.9%), ipf progression (6.9%), acute pancreatitis (3.4%), acute coronary syndrome (3.4%), worsening cirrhosis (3.4%).

The registry lists 29 participants at risk for serious adverse events in each arm. Serious adverse events affected 3 participants in the Carbon Monoxide Inhalation group and 7 in the Oxygen 21% group. The registry also lists other adverse events affecting 28 participants in the carbon monoxide group and 19 in the oxygen group, with 29 at risk in each. These are counts in the posted safety summary and should be read as such.

The serious-event terms listed include pneumonia and IPF progression, each at 6.9%, as well as acute pancreatitis, Acute Coronary Syndrome, worsening cirrhosis, and colonic obstruction, each at 3.4%. The registry does not specify the arm allocation for these listed terms in the summary, so they should not be attributed to a particular group based on this information. Death counts are not reported in the registry record.

General statistical principle: adverse-event counts describe events observed during the reported study period; they do not by themselves establish that an intervention caused an event or that one group has a different underlying risk. The registry summary does not provide event timing, severity detail beyond the serious-event classification, exposure-adjusted rates, or a causal assessment. These counts should therefore be interpreted as descriptive safety reporting, not as a complete comparative safety analysis.

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

The posted data describe a randomized comparison of inhaled carbon monoxide and Oxygen 21% in a Phase II study of idiopathic pulmonary fibrosis. On the primary biomarker, the group means changed by -0.15 and 0.88 ng/ml, and the unadjusted contrast is uncertain: its confidence interval includes zero. The sponsor's ANOVA reports p = 0.207. These results do not establish a clear between-group difference in the mean MMP7 change, but they also do not prove that the groups are identical.

The study's primary measure is described as a surrogate biomarker. A biomarker result does not by itself show whether participants experienced better or worse symptoms, function, quality of life, or disease course. Secondary outcomes give additional information, but the results vary by measure and are subject to the limitations of multiple comparisons and incomplete reporting. The six-minute walk result is notable in the posted analyses, but one secondary comparison cannot establish a general treatment effect, especially without the full statistical plan and multiplicity information.

General statistical principle: absence of a statistically clear difference is not the same as evidence of equivalence. Demonstrating equivalence or non-inferiority requires an appropriate design, prespecified margins, and analyses; those details are not reported in the registry record. Likewise, a nominally unusual result does not alone establish clinical importance. The registry does not report thresholds for clinically meaningful differences.

The safety summary reports participants affected by serious and other adverse events, but it does not provide enough detail to establish causation or fully compare safety. Taken together, the posted results are a record of estimated group changes and their uncertainty in this trial. They do not provide a basis for medical advice or a definitive conclusion about treatment effects beyond the measurements and analyses recorded.

18. Why This Trial Matters Statistically

This trial is statistically instructive because it brings together a randomized design, a continuous biomarker endpoint, and several distinct secondary measures. The primary result illustrates why the estimated contrast and its uncertainty matter together: the means differ numerically, but the recalculated confidence interval spans zero and covers a wide range of possible group contrasts. A point estimate alone would hide that uncertainty.

It also illustrates the importance of specifying contrast direction. The recalculation gives Oxygen 21% minus carbon monoxide, while the sponsor's reported net difference is in the reverse direction. The signs differ, but the comparisons are not necessarily substantively inconsistent. Reading the group labels and subtraction order prevents an apparent sign reversal from being mistaken for a disagreement in findings.

The trial also highlights distinctions among statistical significance, clinical meaning, and surrogate measurement. A biomarker may be informative about biology without directly demonstrating a change in outcomes important to participants. Secondary measures can add context, but several comparisons raise multiplicity concerns when adjustment and prespecification are not documented. The registry's incomplete detail about analysis populations, missing data, and the original statistical plan limits how fully those issues can be assessed.

Finally, the sample-size illustrations show how planning depends on an assumed effect size: smaller effects require larger samples under the stated assumptions. Because these illustrations use observed effects, they cannot be read as the trial's original design rationale or as evidence about the probability that the observed result is correct. Separating observed results from planning scenarios is essential when interpreting registry data.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The primary endpoint is a surrogate biomarker, and the registry's explanation does not establish how changes in MMP7 translate into clinical outcomes. The endpoint covers baseline to Week 12, so the posted summaries do not describe longer-term changes. Longer-term follow-up is not reported in the registry record.

The primary group sizes are 29 and 28, with one missing MMP7 measurement attributed to withdrawal of consent before the visit 2 blood draw. The missing-data note calls the group placebo, while the arm title is Oxygen 21%. The record does not explain this wording difference. It does not report a detailed missing-data method, sensitivity analyses, or complete participant flow. The secondary outcome summaries list 29 participants per arm, but no fuller explanation of endpoint-specific analysis populations is available.

The statistical detail is limited. The registry reports ANOVA results but does not provide the full model specification or confidence intervals for those analyses. Recalculated estimates use summary statistics and an approximate standard deviation, so they are a simple cross-check rather than a reconstruction of the sponsor's complete analysis. Multiplicity handling, the original sample-size justification, interim monitoring, and the detailed analysis plan are not reported in the registry record.

The comparator arm's descriptive text is blank, and the registry does not provide further details about oxygen administration or masking procedures. Safety reporting is also limited: event counts and selected serious-event terms are available, but event timing and causal assessment are not. Death counts are not reported in the registry record. These limitations constrain interpretation of the posted efficacy and safety summaries.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Inhaled Carbon Monoxide to Treat Idiopathic Pulmonary Fibrosis — NCT01214187.
• Registry record last updated 2017-06-14; status last verified 2017-05.
• Results first posted 2017-06-14.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

The registry describes a completed, randomized, double-masked, parallel-group Phase II study with enrollment of 58 people with idiopathic pulmonary fibrosis. The comparison was inhaled carbon monoxide versus Oxygen 21%. The primary endpoint was serum MMP7 change from baseline to Week 12.

The primary recalculated contrast is 1.03 ng/ml for Oxygen 21% minus carbon monoxide, with a confidence interval from -0.564 to 2.624 and p = 0.205. The sponsor's reported ANOVA gives a net mean difference of -1.04 in the opposite contrast direction and p = 0.207. The registry also posts three secondary outcome analyses, with a result for the listed respiratory questionnaire not reported in the registry record.

The record notes one missing primary MMP7 measurement and provides adverse-event counts, but important details about the statistical plan, multiplicity, analysis populations, and safety-event attribution are not reported in the registry record. The estimates should be read with their uncertainty and the limits of the posted information.