Efficacy and Safety of 3 Doses of BI1356 (Linagliptin) in Type 2 Diabetes Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of 3 Doses of BI1356 (Linagliptin) in Type 2 Diabetes Patients (NCT00328172), a randomized phase 2 study. The primary outcome is Change From Baseline in HbA1c (Glycosylated Haemoglobin) at Week 12.

2. Clinical Question

The study was designed to investigate efficacy, safety, and tolerability across several doses of BI 1356 over 12 weeks. The primary endpoint was a continuous measure, so its central comparison concerns the difference in mean HbA1c change between treatment groups. Negative changes indicate a decrease from baseline; the placebo-adjusted estimates describe how much the estimated change differed between each comparison group and placebo.

The metformin group was open-label and had a distinct sensitivity-measurement role. It should not be treated as another blinded placebo-controlled dose comparison. The registry record does not report enough detail to establish the precise intended interpretation of that arm beyond its stated role. The principal placebo comparisons are therefore the three linagliptin doses versus placebo.

3. Trial Design

This was a Phase 2, randomized, double-blind, placebo-controlled, parallel-group study with five treatment groups. The registry lists 302 participants enrolled. The planned groups were placebo, BI 1356 at 0.5 mg, 2.5 mg, or 5.0 mg once daily, and an open-label metformin group. Treatment continued for 12 weeks. The study is marked completed in the registry record.

The study description identifies participants with type 2 diabetes and insufficient glycemic control and includes both drug-naive and treated patients. The registry record does not report the detailed eligibility criteria, the randomization ratio, allocation concealment procedures, or the planned group sizes. It also does not report the original sample-size justification or the statistical assumptions used to design the trial.

Double masking applied to the placebo-controlled BI 1356 comparisons. The metformin arm was explicitly open-label, so masking should not be assumed for that group. The primary outcome analysis included participants with both baseline and on-treatment HbA1c measurements and used last observation carried forward for imputation. That analysis population is not identical to the enrollment count: the primary outcome results list group-specific analysis sizes, while the safety results give separate at-risk counts.

The study also assessed fasting plasma glucose and the percentage of patients with an absolute efficacy response defined as HbA1c <= 7.0% at 12 weeks. The registry lists population pharmacokinetics as an objective in the study summary. A population pharmacokinetic result is not reported in the registry record discussed in these results.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set includes all randomized patients with baseline and on-treatment value of HbA1c. Last observation carried forward (LOCF) was used as the imputation rule.

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 Change From Baseline in HbA1c (Glycosylated Haemoglobin) at Week 12, measured over Baseline, week 12.

The primary endpoint is change from baseline in HbA1c at Week 12. HbA1c is a continuous measure of glycosylated haemoglobin; the registry expresses this endpoint in percent. The stated change is Week 12 HbA1c minus Week 0 HbA1c, so a negative value represents a decrease relative to baseline and a positive value represents an increase.

The reported group results are least-squares means adjusted for baseline HbA1c. In general, adjustment for a baseline value accounts for the fact that participants can begin treatment with different measurements, and the adjusted mean estimates the group result after accounting for that baseline measure. The registry describes the primary analysis population as randomized patients with both a baseline and an on-treatment HbA1c value. Last observation carried forward was used as the imputation rule.

HbA1c is used as an endpoint because it provides a continuous measure that can be compared across randomized groups. A continuous outcome retains information about the size and direction of change, rather than reducing every participant to a response category. The registry also lists a responder outcome based on an HbA1c threshold, but the continuous change is the designated primary endpoint. The registry record does not report the clinical rationale for selecting the endpoint or the threshold used for the responder outcome.

7. Results Posted in the Registry

• Linagliptin (BI 1356) 0.5 mg: mean 0.04 versus 0.18 on control; difference -0.14 (95% CI -0.42 to 0.14), standardized effect d = -0.18; z-test p = 0.322.
• Linagliptin (BI 1356) 2.5 mg: mean -0.24 versus 0.18 on control; difference -0.42 (95% CI -0.70 to -0.14), standardized effect d = -0.55; z-test p = 0.003.
• Linagliptin (BI 1356) 5.0 mg: mean -0.28 versus 0.18 on control; difference -0.46 (95% CI -0.74 to -0.18), standardized effect d = -0.60; z-test p = 0.001.
• Metformin: mean -0.68 versus 0.18 on control; difference -0.86 (95% CI -1.12 to -0.60), standardized effect d = -1.13; z-test p = <0.001.

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.

Change From Baseline in Fasting Plasma Glucose (FPG) at Week 12

• 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:
  • Linagliptin (BI 1356) 0.5 mg: mean 6.69 versus 4.24 on control; difference 2.45 (95% CI -10.03 to 14.93), standardized effect d = 0.07; z-test p = 0.700.
  • Linagliptin (BI 1356) 2.5 mg: mean -15.20 versus 4.24 on control; difference -19.44 (95% CI -32.06 to -6.82), standardized effect d = -0.56; z-test p = 0.003.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Linagliptin (BI 1356) 0.5 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin (BI 1356) 2.5 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry identifies fasting plasma glucose change at Week 12 and the percentage of patients achieving HbA1c <= 7.0% at 12 weeks as secondary outcomes. The posted secondary results include arm-level fasting plasma glucose changes and sponsor-reported placebo comparisons for some, but not all, groups. A result for the responder outcome is not reported in the registry record. The registry also indicates that two secondary outcomes have posted results, but not every listed endpoint has a complete set of reported analyses across groups.

At an overall level, the fasting plasma glucose results show different directions and magnitudes across the treatment groups rather than a uniform pattern across all doses. The registry’s primary HbA1c findings and fasting plasma glucose findings concern related measures of glycemic change, but they are not interchangeable endpoints. Their patterns can be considered for coherence, while keeping each outcome’s measurement scale, analysis, and available comparisons distinct. The responder endpoint would add a threshold-based view, but its results are not reported in the registry record.

Secondary outcomes generally carry less inferential weight than a prespecified primary endpoint because they may be among several questions examined. If multiple tests are performed without an appropriate adjustment, the chance of at least one apparently notable result can rise. Whether any multiplicity adjustment was planned or used for the secondary outcomes is not reported in the registry record. The posted secondary analyses should therefore be treated as supportive and exploratory rather than as definitive evidence on their own.

The registry does not provide full secondary endpoint analysis details, complete confidence intervals for every comparison, or a documented hierarchy of hypotheses. Those omissions limit how strongly the secondary patterns can be combined with the primary endpoint to make broader claims.

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

• Change From Baseline in Fasting Plasma Glucose (FPG) at Week 12
• Percentage of Patients With Absolute Efficacy Response (HbA1c <= 7.0%) at 12 Weeks

9. Statistical Interpretation

10. Statistical Methods Explained

Several statistical ideas help distinguish the reported treatment estimates from the way they were calculated and the uncertainty around them.

What does a least-squares mean represent?

A least-squares mean is an adjusted group mean from a statistical model. In general, model adjustment estimates group results while accounting for specified covariates. For the primary endpoint, the registry states that means were adjusted for baseline HbA1c. This matters because a comparison of adjusted means is not necessarily identical to subtracting two unadjusted group averages. The posted primary values are labelled least-squares means, and the sponsor’s reported comparisons are described as ANCOVA results.

ANCOVA, or analysis of covariance, compares group outcomes while adjusting for covariates such as a baseline measurement. In general, including a baseline value can improve the precision of a treatment comparison and account for baseline differences. The registry reports ANCOVA for the sponsor’s primary comparisons. The full model specification, additional covariates, assumptions checks, and details of the estimation procedure are not reported in the registry record.

How should a mean difference be read?

A mean difference expresses the estimated contrast between two group means, using the order and definition set by the analysis. Here, the primary outcome is change from baseline in HbA1c, and the posted sponsor comparisons are against placebo. A negative estimate means the comparison group’s estimated change was lower than placebo’s estimated change. Since the endpoint is a change score, the sign must be read together with the outcome definition; it is not a percentage-point response rate or a relative effect.

The metformin comparison is also expressed against placebo in the reported results, but its open-label status and stated sensitivity role should be kept in view. It does not share the same blinded treatment context as the linagliptin dose groups.

What do the recalculations represent?

The recalculated comparisons are simple, unadjusted estimates based on the posted summary statistics. Their pooled standard deviations are approximated from the group-level standard errors, as indicated in the registry data. General statistical reasoning explains why this is an approximation: a standard error describes uncertainty in a mean, whereas a standard deviation describes variation among individual observations. Converting between them requires information about the sample size and assumes that the reported quantities correspond to the relevant groups and outcome. The recalculation does not reproduce the sponsor’s adjusted ANCOVA model.

The recalculated p-values are unadjusted, as are the recalculated comparison estimates. An unadjusted analysis evaluates each contrast without accounting for the number of comparisons being examined. General statistical practice recognizes that multiple testing can increase the chance of finding at least one apparently unusual result when many hypotheses are tested. Whether the sponsor applied a multiplicity procedure is not reported in the registry record.

What role do p-values play?

A p-value is a measure, calculated under a statistical model and a null hypothesis, of how unusual the observed data or more extreme data would be if that null hypothesis held. It is not a measure of effect size, and it does not indicate the probability that the null hypothesis is true. A smaller p-value can be evidence against the null model, but interpretation should also consider the estimate, its confidence interval, the analysis plan, and the study context.

The sponsor-reported p-values relate to its ANCOVA comparisons. The recalculated unadjusted p-values are based on a different, simpler calculation. The two sets of values should not be blended or treated as if they came from the same model. Given the discrepancy recorded for this trial, the sponsor’s analysis is the primary source for the trial’s reported inference.

Why does imputation matter?

The registry states that last observation carried forward was used for the primary outcome. In general, this approach substitutes a participant’s last recorded measurement for a later missing measurement. It can retain participants with incomplete follow-up, but it relies on assumptions about what the unobserved outcome would have been. Those assumptions may be unrealistic if the outcome would have changed after the last measurement, and the method does not itself remove bias from missing data. The amount and pattern of missingness, reasons for missing measurements, and any sensitivity analyses are not reported in the registry record.

The primary population also has a specific definition: randomized participants with a baseline and an on-treatment HbA1c value. This differs from simply including everyone enrolled. The registry does not report the number excluded from that population or the reasons for exclusion.

11. Confidence Intervals and Precision

A confidence interval gives a range of values that are reasonably compatible with the data and the statistical model at the stated confidence level. In general, a narrower interval indicates a more precise estimate, while a wider interval indicates greater uncertainty. A confidence interval is not a guarantee that the true value lies in its particular range, nor does it describe the range of individual participants’ outcomes.

The sponsor reported 95% confidence intervals for its ANCOVA comparisons. For the 0.5 mg dose, the interval extended from -0.41 to 0.14 and included zero. For the 2.5 mg dose, it extended from -0.69 to -0.14. For the 5.0 mg dose, it extended from -0.74 to -0.18. The metformin comparison interval was -1.1 to -0.59. The intervals communicate uncertainty around the estimated mean differences, and their relationship to zero is relevant to the corresponding null comparison.

Precision is also affected by the analysis population and the amount of outcome variability. The primary results list different numbers of analyzed participants across groups, and the registry gives standard errors for the least-squares means. The registry record does not report a formal precision target for the trial or a clinical threshold against which these intervals could be judged.

The recalculated intervals are based on an unadjusted approximation from summary statistics and are not identical to the sponsor’s ANCOVA intervals. In view of the recorded discrepancy, the sponsor-reported intervals should be used to describe the trial’s primary analysis. Neither interval width nor statistical significance alone resolves questions about clinical importance or applicability beyond the population and follow-up studied.

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 figures are planning illustrations based on observed effects, not the trial’s original design assumptions. For these continuous outcomes, the figures refer to participants per arm, not total events. The planning calculation assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the stated fraction of the observed effect as the effect to detect. The trial’s own original sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the estimated number of participants per arm for a design with an 80% chance of detecting the specified effect under the planning assumptions, if that effect and the model assumptions hold. The 90% power column describes the corresponding planning size for a 90% chance. These are prospective design quantities: they show how a future study might be planned under the stated assumptions, not how likely the completed trial was to find a result after the data were collected. This is not observed power.

Each planning scenario scales the effect to be detected relative to the observed effect. Planning for 100% of the observed effect uses the full observed estimate; planning for 75% or 50% uses smaller assumed effects. Smaller effects are harder to distinguish from random variation, so they require more participants for the same target power. The planning figures also depend on the variability estimate and the approximation used, so they should not be read as a guarantee of the size needed in a future study.

For each contrast, interpret the participant counts alongside the outcome and the comparison being planned. The figures do not include an adjustment for multiple comparisons, attrition, or other design considerations unless those are explicitly included in the stated assumptions. Such additions are not reported in the registry record.

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 = -0.140; pooled standard deviation σ = 0.776.
2. n = 2 × (1.96 + 0.842)² × 0.776² / -0.140² = 481.8, rounded up to 482 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 <- -0.14                      # observed difference in means
sigma <- 0.775580700977541                     # 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(328172)
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 = 482 (80% power), 645 (90% power)
 75% of observed effect: n per arm = 857 (80% power), 1,147 (90% power)
 50% of observed effect: n per arm = 1,928 (80% power), 2,580 (90% power)
power.t.test: 482.7 
Simulated power at n = 482 per arm: 0.798 (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 sponsor reported ANCOVA comparisons of each group with placebo for the primary endpoint. Those analyses are the trial’s reported adjusted analyses. The recalculated contrasts are simple unadjusted calculations from the posted summary statistics, with pooled variability approximated from arm-level standard errors. They therefore answer a related but not identical statistical question.

The posted sponsor estimates for the three linagliptin doses were -0.14, -0.42, and -0.46, with 95% confidence intervals of -0.41 to 0.14, -0.69 to -0.14, and -0.74 to -0.18, respectively. The sponsor-reported p-values were 0.327, 0.003, and 0.001. For metformin, the sponsor reported an estimate of -0.85 and a 95% confidence interval from -1.1 to -0.59. The recalculated values are not a substitute for these analyses.

The registry marks a discrepancy between the recalculations and the sponsor’s reported results. The sponsor’s own analysis takes precedence. Differences can arise because the recalculation is unadjusted and uses approximated variability, while ANCOVA uses a model that adjusts for baseline HbA1c. The registry record does not report enough methodological detail to reconcile every numerical difference. The sponsor’s results should therefore anchor interpretation of the primary comparisons, with recalculations treated only as a supplementary check.

15. Multiplicity, Missing Data and Analysis Populations

The study contains three placebo comparisons for the primary endpoint, additional treatment comparisons for metformin, and more than one secondary outcome. When several hypotheses are tested, general statistical theory shows that the chance of at least one false-positive result can increase if each test is interpreted independently at the same significance threshold. Methods such as a prespecified hierarchy or multiplicity adjustment can address that issue, but the registry record does not report whether such procedures were part of the sponsor’s analysis plan.

The posted sponsor analyses identify ANCOVA and give estimates, 95% confidence intervals, and p-values for the reported comparisons. The registry record does not report whether those p-values or confidence intervals were adjusted for multiple testing. The recalculated p-values are explicitly unadjusted. Consequently, the reported values should not be assumed to control an overall error rate across all treatment arms and outcomes.

For missing data, the primary analysis used last observation carried forward in the Full Analysis Set, defined as randomized participants with a baseline and on-treatment HbA1c result. General statistical explanation: carrying the last measurement forward is one way to fill a missing endpoint, but it assumes that the carried-forward value is an appropriate stand-in for the unobserved value. If outcomes would have changed after the last measurement, this can affect the estimated group comparison. The amount of missing data, reasons for missingness, handling of missing secondary outcomes, and sensitivity analyses are not reported in the registry record.

The registry does not report whether the primary endpoint, population, imputation method, multiplicity procedures, or planned contrasts were specified in a full statistical analysis plan before the data were examined. It also does not report details of interim analyses. These omissions limit assessment of how closely the reported analyses follow the original plan and how robust the results are to alternative assumptions.

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: pancreatitis acute (1.8%), injection site abscess (1.8%), renal colic (1.8%), myocardial ischaemia (1.7%), schizoaffective disorder (1.5%).

The registry reports serious adverse events and other adverse events using separate at-risk counts for each treatment group. For serious adverse events, the listed affected counts were 1 of 67 for placebo, 1 of 58 for linagliptin 0.5 mg, 2 of 57 for linagliptin 2.5 mg, 0 of 55 for linagliptin 5.0 mg, and 1 of 65 for metformin. These are counts as posted; no new percentages or comparative rates are calculated.

The registry also reports other adverse events, with affected counts of 18 of 67, 15 of 58, 9 of 57, 8 of 55, and 13 of 65 for those groups, respectively. The at-risk counts differ from the group sizes in the primary outcome analysis, because the safety and efficacy results use distinct analysis populations or denominators. The registry record does not explain the participant-level reasons for those differences.

Deaths are not reported in the registry record. Serious adverse event counts are descriptive and do not establish that an event was caused by treatment. The duration of treatment was 12 weeks, and these counts alone do not characterize the full safety profile or provide a formal comparison of risk between groups.

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

The results describe changes in HbA1c over 12 weeks in a randomized, double-blind, placebo-controlled study, together with an open-label metformin group. For the primary outcome, the sponsor’s reported adjusted comparisons indicate that the 2.5 mg and 5.0 mg groups had negative estimated differences relative to placebo, while the 0.5 mg comparison interval included zero. Those are findings from the registry-reported statistical analyses, not a complete account of the clinical value of the treatments.

A statistically detectable difference is not automatically a clinically important difference. The registry record does not report a threshold for a meaningful HbA1c change, nor does it provide patient-level outcomes that would establish how participants experienced the differences. A p-value does not give the probability that a result is true, and a confidence interval does not describe the range of individual responses. Each answers a narrower statistical question.

The open-label metformin group should be interpreted in light of its stated sensitivity-measurement role and its different masking status. It is not the same kind of comparison as a blinded linagliptin-versus-placebo contrast. The secondary results provide additional outcome information, but their completeness and multiplicity procedures are not reported in the registry record.

The trial’s 12-week follow-up and Phase 2 designation define the scope of the evidence. The registry record does not establish persistence of the observed changes beyond the study period, effects on outcomes not measured in the posted results, or results in populations not represented by the trial. These results should be read as statistical findings from this particular study, not as individualized guidance or a general treatment recommendation.

18. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation requires more than reading a point estimate or a p-value. The primary results combine baseline-adjusted means, ANCOVA comparisons, confidence intervals, and a defined analysis population. Each contributes a different piece of information: the estimated direction and size of a group difference, the uncertainty around that estimate, the model used to make the comparison, and the participants whose data entered the analysis.

The record also makes an important distinction between a sponsor-reported adjusted analysis and a simple recalculation from summary statistics. The recalculated contrasts use an approximation to the pooled standard deviation and do not reproduce the baseline adjustment in the sponsor’s ANCOVA. When the registry flags a discrepancy, the sponsor’s own analysis takes precedence; the two approaches should not be presented as equivalent independent confirmations.

The use of last observation carried forward provides a concrete example of how missing-data choices shape the analysis. Understanding the method is not enough to assess its impact: missingness patterns and sensitivity analyses are also important, and those details are not reported in the registry record. Similarly, several treatment comparisons and secondary outcomes raise questions about multiplicity, but the applicable adjustment procedures are not reported.

Finally, the study shows why safety counts and efficacy results must be read with their own denominators and analysis populations. The registry reports different at-risk counts for safety and different group sizes for the primary endpoint. Descriptive adverse-event counts can inform what was recorded, but they cannot alone establish causation or comparative risk.

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 registry record leaves important design and analysis details unspecified. It does not report the original sample-size justification, the randomization ratio, the detailed eligibility criteria, or the full statistical analysis plan. It also does not report the covariates and model diagnostics for ANCOVA, the treatment of multiplicity, or details of interim analyses. These gaps limit independent evaluation of whether the reported analyses match the original plan.

The primary analysis used last observation carried forward, but the registry record does not give the amount or pattern of missing HbA1c data, reasons for missingness, or sensitivity analyses under alternative missing-data assumptions. The listed primary analysis population is restricted to randomized participants with baseline and on-treatment HbA1c values, and the record does not report the reasons participants were absent from that population.

The study lasted 12 weeks and was Phase 2. Its follow-up therefore describes outcomes over that period; longer-term results are not reported in the registry record. The metformin comparison was open-label and had a sensitivity-measurement role, unlike the double-blind placebo comparisons. That distinction limits direct interpretation across all five groups as if they were identical treatment contrasts.

There is also incomplete reporting across the secondary outcomes: the registry lists a threshold-based HbA1c response outcome, but its result is not reported in the registry record. Safety deaths are not reported. The registry does not report a clinical-importance threshold for the primary outcome, which limits the interpretation of whether statistically estimated differences are large enough to matter clinically.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety of 3 Doses of BI1356 (Linagliptin) in Type 2 Diabetes Patients — NCT00328172.
• Registry record last updated 2014-03-14; status last verified 2014-02.
• Results first posted 2011-06-07.
• 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.

23. Record Summary

The registry describes a completed Phase 2, randomized, double-blind, parallel-group study of BI 1356, identified in the results as linagliptin, in people with type 2 diabetes and insufficient glycemic control. It lists 302 participants enrolled and five groups: placebo, three once-daily BI 1356 doses, and an open-label metformin arm. The study ran for 12 weeks.

The primary outcome was baseline-adjusted change in HbA1c at Week 12. The primary population included randomized participants with baseline and on-treatment HbA1c measurements, with last observation carried forward used for imputation. The sponsor reported ANCOVA comparisons against placebo. The registry flags a discrepancy with simple unadjusted recalculations, so the sponsor’s analyses take precedence.

Secondary outcomes included fasting plasma glucose change and a threshold-based HbA1c response. Safety results report serious and other adverse-event counts with separate at-risk group sizes; deaths are not reported in the registry record. The original sample-size justification, multiplicity procedures, detailed missing-data patterns, and several other analysis details are not reported.