Telmisartan to Reduce AIDS-Related Fibrotic and Inflammatory Contributors (TRAFIC Study): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Telmisartan to Reduce AIDS-Related Fibrotic and Inflammatory Contributors (TRAFIC Study) (NCT01928927), a randomized phase 2 study. The primary outcome is Change in Percent Collagen I Deposition on Lymph Node Pathology From Baseline to Week 48.

2. Clinical Question

3. Trial Design

TRAFIC was a phase 2, randomized, parallel-group study with no masking. The registry record lists 58 participants enrolled and gives the study status as completed. The study started in 2014-01 and its primary completion date was 2016-03. The named intervention was telmisartan, and the comparison arm received no study drug while following the week 0–48 evaluation schedule.

The primary outcome was change in percent collagen I deposition on lymph-node pathology from baseline to week 48. The analysis population was defined by protocol completion, treatment completion for participants assigned telmisartan, absence of confirmed virologic failure, paired biopsies at baseline and week 48, and non-missing lymphoid-tissue collagen I deposition. Thus, the primary endpoint summaries refer to participants meeting these criteria, not automatically to everyone enrolled. The registry record lists 17 participants in the telmisartan primary-outcome group and 12 in the no-study-drug primary-outcome group.

The registry describes additional outcomes involving fibronectin deposition in lymph-node and subcutaneous abdominal adipose tissue, collagen VI deposition in adipose tissue, adverse events, and inflammatory markers including IL-6 and IL-7. It lists 66 outcome measures and 64 posted secondary outcomes. These other outcomes broaden the study's scope, but the primary comparison remains the week 48 lymph-node collagen I measure. Details such as the randomization procedure, allocation concealment, telmisartan dose, and any interim analyses 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: All participants enrolled to Step 1. Participants with successful Step 1 biopsies and eligible for Step 2 were randomized to the two study arms using a 2:1 allocation ratio with permuted blocks of size 3 and without institutional balancing. There was no stratification.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Per-protocol population: Participants who 1) completed the protocol, 2) completed treatment (if on telmisartan arm), 3) did not have confirmed virologic failure, 4) had paired biopsies at baseline and week 48. Additionally, have non-missing lymphoid tissue collagen I deposition.

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: All Step 2 randomized participants with data available for specific measures.

6. Primary Endpoint

The registry lists the primary outcome as Change in Percent Collagen I Deposition on Lymph Node Pathology From Baseline to Week 48, measured over baseline and week 48.

The primary endpoint measures the absolute change in the percentage of lymph-node tissue staining positive for collagen I between baseline and week 48. The registry defines the underlying measure as the average percentage of collagen stained across multiple uniform-sized, high-magnification images in each sample. Change is calculated as the week 48 value minus the baseline value, so a negative value represents a lower measured percentage at week 48 than at baseline for that participant.

Collagen I deposition is a tissue measure of fibrosis, or scarring. Biopsy-based measures can capture structural features in the sampled tissue rather than relying only on circulating markers. The registry's stated study goal was to assess whether telmisartan changed fibrosis and inflammation, and this endpoint operationalizes part of that goal in lymph-node tissue. It does not capture every possible aspect of fibrosis, inflammation, or health.

The endpoint is summarized with medians and inter-quartile ranges. A median describes the middle of the observed distribution, while the inter-quartile range describes its central spread. The per-protocol definition means that the summary applies to participants meeting the stated completion, biopsy, virologic, and measurement criteria. The registry record does not report a separate analysis of the primary outcome for all randomized participants.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports a theta statistic and identifies the method as “Theta statistic; DeLong & Clarke-Pearson.” It also reports a 95% confidence interval and a p-value for the superiority analysis. The record does not give a fuller account of how theta was defined, the precise handling of ties, or whether any covariate adjustment was used. Those details are not reported in the registry record.

What does the median summarize?

The median is the point that divides an ordered set of observations into a lower and an upper half. It is useful when a distribution may be skewed or when extreme observations could strongly affect an average. Here, the endpoint summaries are medians of individual changes, rather than changes in group means. A group median should not be interpreted as the change experienced by every participant.

The inter-quartile range describes the central half of a distribution, from its lower quartile to its upper quartile. It indicates how dispersed the middle observations are, but it is not a confidence interval for the median. The ranges reported for the two arms overlap substantially. Overlap alone is not a formal test of a between-group difference, and it does not determine the result of the sponsor's analysis.

What does the theta analysis add?

A between-group statistic can compare the distributions of outcomes without reducing them to a difference in means. The registry's theta statistic is paired with a confidence interval and a p-value, making it the reported inferential comparison for the primary outcome. The record does not supply enough explanation to assign a clinical meaning to particular theta values or to identify which direction of theta favors which arm. Interpretation should therefore remain anchored to the reported estimate, interval, p-value, and the endpoint definition rather than assigning an unsupported directional meaning.

In general, a confidence interval describes a range of values compatible with the data and the assumptions of the statistical procedure. It also conveys precision: a broader interval represents greater uncertainty than a narrower one. It is not a range containing a specified proportion of individual participant outcomes, and it is not a guarantee that the true value lies within the interval. The p-value measures how unusual data at least as incompatible with a null hypothesis would be if that null and the model assumptions were correct. It does not measure the probability that the null is true or the probability that a result occurred by chance.

Why does the analysis population matter?

The primary analysis was per protocol, with eligibility requiring completion of specified study and biopsy procedures as well as non-missing endpoint data. This is a defined subset of the enrolled study population. Such restrictions can make the analyzed group different from the full randomized group, particularly if completion or availability of paired biopsies is related to participant characteristics or outcomes. The registry record does not report the reasons for each exclusion or a sensitivity analysis assessing the effect of missing endpoint data.

Randomization is intended to support fair comparisons between assigned groups, but a per-protocol analysis does not necessarily preserve all the protections of an analysis based on original assignment. The registry record does not report an intention-to-treat analysis of this endpoint, so the primary estimate should be understood as applying to the stated per-protocol population. The record also does not report whether the analysis accounted for any other design or clinical variables.

What does the study design imply?

Random allocation supports a comparative design, and the parallel-group structure means participants were assigned to one study arm rather than crossing between treatments. The absence of masking means treatment assignment was not concealed from participants and study personnel as described by the registry. This could matter for outcomes affected by expectations or decisions about assessment; the tissue endpoint is biopsy-based, but the record does not describe masking of pathologists or other outcome assessors.

General statistical considerations also apply to small studies: estimates can be imprecise, and a non-significant test may reflect uncertainty rather than evidence of equivalence. The registry record does not provide a sample-size justification or a formal equivalence or non-inferiority margin. It also does not report an interim-analysis plan. Any planning illustration based on an observed effect is a hypothetical design calculation, not observed power and not proof that the original study was adequately powered.

How should multiple outcomes be considered?

The registry lists 66 outcome measures and 64 posted secondary outcomes. Assessing many outcomes can increase the chance that some results appear notable by chance if each is tested separately. The statistical methods for addressing multiplicity, if any, are not reported in the registry record. The primary outcome is identified in advance in the record, but the listed secondary outcomes should be interpreted with awareness that no adjustment procedure is described.

10. Analyses Reported by the Sponsor

The registry reports the sponsor's primary comparison as a theta statistic of 0.50, with a 95% confidence interval from 0.26 to 0.73 and p = 0.97. It identifies the method as a theta statistic using DeLong & Clarke-Pearson. That is the inferential result reported for the primary endpoint.

The arm summaries are medians and inter-quartile ranges, while the sponsor's comparison is expressed as theta. These are different forms of summary, so the medians cannot be subtracted or otherwise used to reproduce the reported theta result from the information in the registry record. The record does not explain the theta statistic's direction in clinical terms, nor does it give enough methodological detail to independently reconstruct the test.

The sponsor's own analysis is therefore the relevant reported comparison; the descriptive summaries provide context but do not replace it. The registry record does not identify a discrepancy between the reported comparison and a recalculation, and it does not report enough information for a separate numerical recalculation of theta from the arm summaries alone.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 66 outcome measures, including 64 posted secondary outcomes. The secondary outcomes span tissue measures, adverse events, and inflammatory markers at different time points. Considering many outcomes creates a multiplicity issue: when many statistical tests are conducted, the chance of at least one apparently notable result can increase. The statistical plan for controlling multiplicity, or a statement that no adjustment was planned, is not reported in the registry record. The primary outcome is designated, but that designation does not itself establish how secondary comparisons should be interpreted.

The primary analysis population required protocol completion, treatment completion for the telmisartan arm, no confirmed virologic failure, paired biopsies at baseline and week 48, and non-missing lymphoid-tissue collagen I deposition. The registry reports 17 participants in the telmisartan primary-outcome group and 12 in the no-study-drug group. These are the primary endpoint analysis groups; they should not be confused with the overall enrollment of 58 or with the denominators reported for safety outcomes.

The record does not report why participants were absent from the primary analysis, how missing measurements were handled beyond the non-missing-data eligibility criterion, or whether sensitivity analyses assessed missing outcomes. It also does not report an intention-to-treat primary endpoint analysis. These omissions limit assessment of whether the per-protocol results might differ from results among everyone assigned to a group.

The record does not report a sample-size justification, statistical power calculation for the original design, or an interim-analysis plan. A planning illustration based on an observed effect, where available, is not observed power and should not be read as evidence about the design's original assumptions. The registry record also does not report whether a prespecified analysis plan addressed multiplicity, protocol deviations, or missing data in greater detail.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: angina unstable (6.7%), diarrhoea (3.4%), herpes zoster disseminated (3.4%), hepatic enzyme increased (3.4%).

The safety record reports serious adverse events for 3 of 29 participants at risk in the telmisartan group and 1 of 15 in the no-study-drug group. Other adverse events are reported for 24 of 29 and 10 of 15, respectively. The recorded number of deaths was 0 in each group. These counts use safety denominators, which differ from the primary endpoint analysis groups, and should not be combined with the biopsy outcome summaries.

The registry also lists serious adverse-event terms with percentages: Angina unstable, 6.7%; Diarrhoea, 3.4%; Herpes zoster disseminated, 3.4%; and Hepatic enzyme increased, 3.4%. The record does not map these listed terms to a particular treatment arm in the terms list, so they should not be assigned to an arm based on this information alone.

Adverse-event counts describe what was recorded during the study; they do not by themselves establish that an event was caused by study treatment. The record does not provide event timing, severity details for each listed term, or a comparative statistical analysis of these safety outcomes. The numbers should be read as descriptive safety data, with attention to the separate denominators and limited study context.

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

The result shows the observed distribution of change in lymph-node collagen I deposition among participants meeting the primary per-protocol criteria, together with the sponsor's reported between-group test. The group medians were negative, but the inter-quartile ranges extended across zero. The sponsor's comparison reported theta = 0.50, a 95% confidence interval from 0.26 to 0.73, and p = 0.97. Taken together, these data do not provide evidence of a between-group difference in the reported test, while leaving uncertainty about the size and direction of any underlying difference.

This is not evidence that the treatments are equivalent. Equivalence requires a prespecified margin and a suitable design and analysis; such a margin is not reported in the registry record. Nor does a high p-value show that there is no effect. It indicates that the reported test did not distinguish the groups against its null hypothesis, given the observed data and the test's assumptions.

The result is specific to the primary tissue endpoint, its baseline-to-week-48 interval, and the per-protocol analysis population. It does not establish effects on other tissue measures, inflammatory markers, clinical outcomes, or adverse events. The registry describes several additional outcomes, but those require their own interpretation and consideration of multiplicity and uncertainty.

The findings also should not be generalized beyond the population and study conditions described in the record without further evidence. In particular, the primary analysis excludes participants who did not meet the stated completion and biopsy requirements. The record does not provide an all-randomized analysis of this endpoint. The most defensible reading is therefore limited: the reported per-protocol data did not identify a difference in the sponsor's primary statistical comparison, and the available summary does not establish equivalence or resolve effects outside this endpoint and population.

14. Why This Trial Matters Statistically

This trial is statistically informative because it combines a randomized comparison with a biopsy-based primary endpoint and a clearly described per-protocol population. The endpoint focuses on a tissue measure related to fibrosis, rather than relying solely on a circulating marker. At the same time, the analysis illustrates why it is important to distinguish the observed endpoint summaries from the formal between-group test: medians and inter-quartile ranges describe each group's distribution, while theta is the statistic used for the reported comparison.

The registry also highlights the importance of reading analysis populations carefully. Enrollment was 58, while the primary endpoint summaries include 17 participants in the telmisartan group and 12 in the no-study-drug group. The safety summaries use different denominators again. These figures refer to different aspects of the study and cannot be treated as interchangeable. The per-protocol eligibility rules make clear that paired biopsies and other completion conditions shaped who contributed to the primary outcome.

Finally, the study illustrates how a non-significant result should be interpreted alongside its confidence interval, endpoint definition, analysis population, and the reporting limitations. The theta estimate is accompanied by uncertainty, and the registry record does not report a clinically important threshold, multiplicity procedure, sample-size justification, or fuller explanation of theta's direction. These gaps do not invalidate the reported comparison, but they constrain what can be concluded from it. The statistical lesson is to preserve that distinction between what the analysis reports and broader claims that the record cannot support.

15. Limitations

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

The primary endpoint analysis was per protocol rather than reported for all randomized participants. Eligibility depended on completion, treatment completion where relevant, no confirmed virologic failure, paired biopsies, and non-missing tissue measurements. The registry record does not describe the reasons participants were not included in this analysis or assess how missing biopsy data might affect the comparison.

The primary summaries are based on 17 participants in the telmisartan group and 12 in the no-study-drug group. The registry record does not report a sample-size justification or power calculation for the original design, so it is not possible from this record to assess the planned precision or sensitivity of the study. The confidence interval for theta also indicates uncertainty, and the registry does not define a clinically important difference for this tissue outcome.

The reported analysis is not fully reproducible from the record. Theta is named, and the method is identified as DeLong & Clarke-Pearson, but the record does not define the statistic's direction in clinical terms or provide enough detail about its implementation. The arm-level summaries are medians and inter-quartile ranges, not the inputs needed to independently reconstruct the reported comparison.

The study was not masked. The registry record does not state whether outcome assessors or pathologists were masked to treatment assignment. It also does not report the randomization procedure, allocation concealment, multiplicity handling across the many outcomes, or an interim-analysis plan. The list of adverse-event terms is not assigned to arms in the terms list. These reporting limitations restrict assessment of potential bias, selective interpretation, and the full context of the safety findings.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Telmisartan to Reduce AIDS-Related Fibrotic and Inflammatory Contributors (TRAFIC Study) — NCT01928927.
• Registry record last updated 2021-08-03; status last verified 2020-06.
• Results first posted 2017-04-12.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

TRAFIC was a completed, randomized, open-label, parallel-group phase 2 study of telmisartan versus no study drug in people with HIV-1 infection described as well controlled on antiretroviral therapy. The registry reports enrollment of 58, a start date of 2014-01, and primary completion in 2016-03.

The primary endpoint was the change in percent collagen I deposition on lymph-node pathology from baseline to week 48, defined as week 48 minus baseline. The primary analysis was per protocol. The reported medians were -2.44 for telmisartan and -6.08 for no study drug; the sponsor's theta comparison was 0.50, with a 95% confidence interval from 0.26 to 0.73 and p = 0.97.

The result is specific to this tissue endpoint and analysis population. The registry record does not report a clinically important threshold, a sample-size justification, or the statistical approach to multiplicity and missing data.