Efficacy, Safety, and Tolerability of ATM-AVI in the Treatment of Serious Infection Due to MBL-producing Gram-negative Bacteria: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy, Safety, and Tolerability of ATM-AVI in the Treatment of Serious Infection Due to MBL-producing Gram-negative Bacteria (NCT03580044), a randomized phase 3 study. The primary outcome is Percentage of Participants With Clinical Cure at the Test of Cure (TOC) Visit -Microbiological Intent to Treat (Micro-ITT) Analysis Set.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, open-label, comparative study with a parallel-group design. It compared aztreonam-avibactam with best available therapy in serious infections due to multidrug-resistant Gram-negative bacteria producing metallo-beta-lactamase. The registry lists Pfizer as the sponsor. The study began on 2020-12-25, had a primary completion date of 2023-01-23, and its status is terminated. Enrollment was 15.

Randomization means treatment assignment was intended to be determined by a chance process rather than solely by clinical choice. In general, randomization helps limit systematic differences between treatment groups, although small groups can still differ by chance. Open-label means masking was listed as none: treatment assignment was not masked. The registry also describes a parallel model, in which participants were assigned to one treatment group rather than switching between groups as part of the design.

The registry record does not report the reason for termination, the allocation ratio, the randomization procedure, or detailed eligibility criteria. It does not report the planned or actual treatment regimens, the rules for selecting BAT, or a protocol-level sample-size justification. The posted primary outcome includes 12 participants in the ATM-AVI arm and 3 in the BAT arm. These are the analysis counts shown for that outcome, rather than a description of why participants were or were not included in the Micro-ITT set.

The study included 20 outcome measures, with 19 secondary outcomes posted. The secondary outcomes cover clinical cure at different visits or in different analysis sets, as well as microbiological responses. The registry lists no analyses for those outcomes. The termination status and small outcome groups are important context when interpreting the posted estimates; neither establishes why the study ended or what the investigators intended to conclude.

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: A total of 15 participants signed the informed consent form and were randomized in the study. The study was terminated as recruitment of participants with serious infections caused by gram-negative bacteria producing MBL was challenging.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Micro-ITT analysis set was a subset of the ITT analysis set and included all participants who had at least 1 MBL-positive Gram-negative baseline pathogen from an adequate specimen at the start of study treatment.

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: Intent to Treat (ITT) analysis set included all randomized participants regardless of receipt of study drug. Participants were analyzed according to the treatment to which they were randomized.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Clinical Cure at the Test of Cure (TOC) Visit -Microbiological Intent to Treat (Micro-ITT) Analysis Set, measured over Day 28.

The primary endpoint was the percentage of participants with clinical cure at the Test of Cure visit on Day 28, assessed in the Micro-ITT analysis set. It is a binary outcome: each eligible participant is classified as having met or not met the stated cure definition. The reported percentage describes how often the outcome was recorded in that analysis group.

The definition focuses on improvement in the infection’s baseline signs and symptoms and whether further antimicrobial treatment was required after study treatment. For participants with complicated intra-abdominal infection, it also accounts for whether unplanned drainage or surgery was needed after the initial procedure. This makes the endpoint a clinical assessment rather than a measure of microbiological response alone. A blinded independent adjudication committee determined clinical response, even though the overall study was open-label.

A Test of Cure assessment asks whether the clinical criteria are met at a specified follow-up point; it does not by itself describe the whole course of illness or establish a treatment effect. The Micro-ITT restriction focuses the primary analysis on participants with a documented MBL-positive Gram-negative baseline pathogen from an adequate specimen. This connects the outcome to the infection type the study was intended to examine. The registry record does not report further details about how missing Day 28 assessments were handled or how the primary comparison was planned statistically.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a participant-level binary outcome, expressed as a percentage. For each participant in the Micro-ITT analysis set, the clinical response assessment classified the person according to whether the cure definition was met at the Day 28 Test of Cure visit. In general, a percentage for a binary endpoint is an estimate of the proportion of participants meeting the specified outcome in the analysis group. It does not, on its own, show whether a treatment caused the outcome or whether groups differ beyond chance.

What does the confidence interval convey?

A confidence interval conveys statistical uncertainty around an estimate. In repeated use of a method under its assumptions, intervals constructed at a stated confidence level are intended to cover the underlying quantity at that level over the long run. It is not a statement that a particular participant has a corresponding probability of being cured, nor is it a guaranteed range for the true value. For the ATM-AVI group, the registry reports a 95% confidence interval of 18.0% to 68.8% around the 41.7% estimate. Its breadth signals substantial uncertainty in the estimated proportion.

The registry says that Jeffrey's method was used for the primary endpoint confidence interval. In general, this is a method for estimating an interval for a binomial proportion. Methods designed for proportions can behave differently from simple normal-approximation intervals, especially when sample sizes are small or estimates are near the boundaries of the possible range. The registry does not provide the calculation details beyond naming the method. No confidence interval is reported for the BAT percentage, so a like-for-like comparison of interval precision is not available.

Why does the analysis set matter?

The primary population was the Micro-ITT analysis set, defined as a subset of the ITT set containing participants with at least 1 MBL-positive Gram-negative baseline pathogen from an adequate specimen at the start of study treatment. This definition narrows the analysis to participants with the specified microbiological finding. It is therefore important to distinguish the primary result from an outcome calculated in all randomized participants or in a different secondary analysis set. The registry lists other analysis sets for secondary outcomes, including ME and Micro-ITT, but does not give further definitions for each set in the posted primary description.

What does open-label randomization imply?

Randomization is a design method intended to distribute known and unknown prognostic factors without systematic treatment-selection decisions. In general, that protection is less reassuring when groups are very small because chance imbalance may remain. Open-label design means treatment assignment was not masked. Knowledge of assignment can potentially influence care or outcome assessment in general, but the registry states that a blinded independent adjudication committee determined clinical response. That feature concerns endpoint assessment; it does not change the trial’s open-label classification or establish that all potential bias was prevented.

What statistical analysis is reported?

The registry provides arm-level percentages and, for ATM-AVI, a confidence interval. It lists zero analyses and does not report a formal between-group test, adjusted model, or treatment-effect estimate for the primary outcome. The statistical analysis plan, any interim analyses, and the handling of missing primary outcomes are not reported in the registry record. These omissions mean readers cannot assess the intended inferential framework or determine whether the posted descriptive results are the complete analysis planned by the investigators.

The record describes a sample-size planning illustration based on an observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such planning numbers, when available, are hypothetical design calculations rather than observed power. They do not show that a study had a particular probability of detecting an effect after results were observed. No calculated participant requirements for those illustrations are reported in the registry record, and these assumptions should not be mistaken for the original trial design.

Why is caution needed with sparse outcomes?

With small groups, a single participant can materially affect a percentage, and interval estimates may be broad or unavailable. General methods for binary data rely on assumptions about the outcome and the analysis population. Sparse data can make ordinary approximations unreliable; specialized interval methods can address some of those mathematical limitations but cannot create information that was not observed. In this study, the small BAT group and its unreported interval are central limitations on precision. The available descriptive result should be read as an account of observed outcomes in the defined analysis groups, not as a robust estimate of a comparative treatment effect.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Aztreonam- Avibactam (ATM- AVI)", "Best Available Therapy (BAT)")
n      <- c(12, 3)
events <- c(5, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Aztreonam- Avibactam (ATM- AVI)  12  5  41.7  15.2  72.3  19.3  68.0  19.2  68.4
ARM  Best Available Therapy (BAT)  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 20 outcome measures and 19 posted secondary outcomes. These include clinical cure at the Test of Cure and End of Treatment visits in different analysis sets, as well as favorable microbiological responses at those visits and at participant or pathogen level. Assessing many outcomes can increase the chance that at least one result appears notable by chance if each is tested separately. Statistical plans often address this issue by prioritizing outcomes, adjusting inference, or treating secondary findings as exploratory, but the appropriate approach depends on the prespecified plan.

The registry record does not report a multiplicity-control strategy, a hierarchy among secondary outcomes, or which analyses were intended to support confirmatory conclusions. It lists zero analyses. As a result, the secondary outcome list alone does not establish which comparisons were performed or how any multiple testing was handled. The primary outcome is identified, but a formal comparative analysis is not posted.

The Micro-ITT definition requires an adequate baseline specimen and at least 1 MBL-positive Gram-negative pathogen. This criterion is relevant to who contributes to the primary estimate and differs from a simple count of all randomized participants. The registry record does not report a detailed flow of participants into each analysis set, the amount or pattern of missing outcome data, or the methods used to handle missing Day 28 assessments. It also does not report whether missing observations were excluded, imputed, or treated under another rule.

Those omissions limit the ability to assess selection effects and the robustness of results to alternative assumptions about unavailable outcomes. They also make it difficult to reconcile the posted group counts with the full randomized population. Neither a missing-data method nor a multiplicity procedure should be inferred from the endpoint labels or from the presence of several listed outcomes; both are not reported in the registry record.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cardiac arrest (50.0%), multiple organ dysfunction syndrome (50.0%), urinary tract infections (50.0%), acute kidney injury (50.0%), pyelonephritis (8.3%).

Safety figures are reported separately from the primary efficacy outcome and use their own at-risk counts. In the ATM-AVI group, 5 participants were recorded as affected by serious adverse events among 12 at risk, and 9 were recorded as affected by other adverse events among 12 at risk. There were 2 deaths among 12 at risk. In the BAT group, 2 participants were recorded as affected by serious adverse events among 2 at risk, and 1 was recorded as affected by other adverse events among 2 at risk. There was 1 death among 2 at risk.

These are counts, not evidence that study treatment caused the events. The registry record does not give enough detail to determine event timing, relatedness, overlap between categories, or the reason for each death. The safety at-risk count for BAT is 2, while the primary outcome count for BAT is 3; the registry does not explain this difference in the posted information. These denominators should not be treated as interchangeable.

The listed serious-event terms include Cardiac arrest, Multiple organ dysfunction syndrome, Urinary tract infections, and Acute kidney injury, each shown as 50.0%; Pyelonephritis and Sepsis are each shown as 8.3%. The registry does not specify the denominator or arm attribution for these listed percentages in the information available. With small safety groups, counts and percentages should be interpreted descriptively, not as stable comparative rates.

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

The posted primary result describes clinical cure at Day 28 among participants in the Micro-ITT analysis set. The ATM-AVI group had a reported cure percentage of 41.7%, with a 95% confidence interval from 18.0% to 68.8%. The BAT group had a reported percentage of 0.0% among 3 participants, with no confidence interval reported. These figures document outcomes in the registry’s defined groups; they do not by themselves establish a treatment effect.

The result does not show whether any apparent difference between groups is statistically reliable. The registry provides no formal comparative analysis, p-value, or comparative estimate with a confidence interval. It also does not report the statistical plan, missing-outcome handling, or the reason for termination. Without these details, the posted percentages cannot be interpreted as a complete inferential assessment of the randomized comparison.

The study’s small enrollment and very small BAT outcome group limit precision. General statistical reasoning indicates that estimates from small groups can be strongly affected by individual outcomes. A confidence interval helps communicate uncertainty, but the absence of a reported BAT interval leaves that arm’s precision less directly visible. The ATM-AVI interval itself is broad. These limitations do not erase the observed data, but they constrain the conclusions that can be drawn from them.

The endpoint is also specific: it concerns a clinical-cure definition at a particular visit in a microbiologically defined analysis population. It is not interchangeable with microbiological response, cure at End of Treatment, or outcomes in all randomized participants. The registry lists several secondary outcomes, but no analyses for them are posted. Nor do the safety counts establish that adverse events were caused by either treatment.

Accordingly, the record supports a descriptive reading of the observed clinical and safety outcomes, alongside substantial uncertainty and incomplete reporting of the comparative analysis. It does not establish that ATM-AVI or BAT is better or worse, and it cannot be generalized beyond the studied setting without further evidence.

14. Why This Trial Matters Statistically

This trial raises statistical issues that are important in studies of serious infections with narrowly defined microbiological eligibility. The primary population required an MBL-positive Gram-negative baseline pathogen from an adequate specimen. The distinction between that group and broader populations matters: an estimate in the Micro-ITT set answers a more specific question than an estimate among all randomized participants.

The randomized, parallel design supports a comparative question, but the study enrolled 15 participants and the primary outcome groups were small, particularly for BAT. Under general statistical principles, small groups limit the precision of percentages and make chance variation more influential. The reported ATM-AVI interval, 18.0% to 68.8%, illustrates that uncertainty. The BAT result has no reported confidence interval, leaving its precision unquantified in the registry record.

The trial also shows why descriptive outcomes and comparative inference should be kept distinct. Two arm-level percentages do not replace a prespecified statistical comparison. The registry lists zero analyses and does not report a p-value or comparative effect estimate. The absence of these elements means the posted results cannot answer whether the observed arm outcomes differ beyond what could plausibly arise from chance under a specified model.

There are further interpretive questions around termination, missing outcomes, and multiplicity. The reason for termination, missing-data strategy, and approach to the 19 secondary outcomes are not reported in the registry record. These are not minor technical details: they affect how readers assess completeness, potential bias, and the strength of any claims. A careful statistical reading treats the reported outcomes as informative descriptions while recognizing the limits of their precision and inferential 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 record identifies the study as terminated but does not report why it ended or whether termination followed a prespecified rule. It also does not state whether enrollment reached the planned target or provide the trial’s original sample-size justification. A planning illustration based on an observed effect is not a substitute for that information and does not establish the study’s initial design assumptions.

The primary outcome groups were small, with 12 participants in the ATM-AVI group and 3 in the BAT group. The BAT estimate has no reported confidence interval. The broad interval for ATM-AVI indicates limited precision, and the very small comparator group makes the arm-level percentage especially sensitive to individual outcomes. The record provides no formal between-group analysis, adjusted estimate, or p-value.

Several operational details are also unavailable. The registry does not report the BAT regimens used, dosing or duration details, the randomization procedure or allocation ratio, participant flow by analysis set, or the reason the safety and primary outcome counts differ for BAT. It does not explain how missing Day 28 assessments were handled or provide a detailed missing-data account.

The study was open-label, although clinical response was determined by a blinded independent adjudication committee. That adjudication is a relevant safeguard for endpoint assessment, but it does not remove all possible effects of unmasked treatment assignment. Finally, the multiplicity plan for the many listed outcomes and the complete statistical analysis plan are not reported in the registry record. These limitations restrict the strength and scope of conclusions that can be drawn from the posted results.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Efficacy, Safety, and Tolerability of ATM-AVI in the Treatment of Serious Infection Due to MBL-producing Gram-negative Bacteria — NCT03580044.
• Registry record last updated 2024-02-02; status last verified 2024-01.
• Results first posted 2024-02-02.
• 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

NCT03580044 is a Phase 3 randomized, open-label, parallel-group study comparing aztreonam-avibactam with best available therapy for serious infections due to MBL-producing Gram-negative bacteria. The registry lists Pfizer as sponsor, enrollment of 15, study start on 2020-12-25, primary completion on 2023-01-23, and status as terminated.

The primary outcome was clinical cure at the Day 28 Test of Cure visit in the Micro-ITT analysis set. The ATM-AVI result was 41.7% (95% confidence interval, 18.0% to 68.8%); the BAT result was 0.0% among 3 participants, with no confidence interval reported. The registry lists 19 secondary outcomes but no analyses. It does not report a formal primary between-group analysis, the reason for termination, or the missing-data and multiplicity methods.