A Drug Interaction Study to Assess the Effect of LY2603618 on the Metabolic Pathway of Desipramine: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Drug Interaction Study to Assess the Effect of LY2603618 on the Metabolic Pathway of Desipramine (NCT01358968), a na phase 1 study. The primary outcome is Plasma Pharmacokinetics of LY2603618: the Maximum Concentration of LY2603618 in the Plasma (Cmax).

2. Clinical Question

These two descriptions identify related but distinct aspects of the study. The clinical pharmacology question concerns a possible effect on the metabolic pathway of desipramine, while the named primary endpoint quantifies peak plasma exposure to LY2603618. That endpoint can describe the concentration profile of LY2603618, but it is not itself a direct numerical measurement of desipramine metabolism. The registry record does not report a result for either question.

3. Trial Design

The registry record describes a completed, phase 1, single-group study in participants with cancer, with enrollment of 20. The study began in 2011-06 and had primary completion in 2012-04. Masking is listed as none, and allocation is listed as NA. These entries indicate that the record does not describe a masked, randomized comparison between independent groups.

The drug-interaction portion is organized into two periods. In Period 1, participants received a single 50 mg oral dose of desipramine alone. In Period 2, the description specifies a single 275 mg intravenous infusion of LY2603618 over 1 hour, followed by a single 50 mg oral dose of desipramine at the end of the infusion. The registry record does not report details such as a washout interval or the scheduling of assessments between periods.

The record also lists LY2603618 with gemcitabine and LY2603618 with pemetrexed as continued-access groups. These entries are separate from the two-period desipramine interaction sequence and should not be taken as additional randomized comparators. The record does not describe random assignment to these groups. The listed intervention names include all four drugs, but detailed dosing information for the continued-access combinations is not reported in the registry record.

The enrollment figure is not necessarily the number contributing to every outcome. The safety listings show 20 participants at risk in the desipramine-alone period and 19 in the LY2603618 plus desipramine period; the continued-access groups list 12 and 7 participants at risk. The registry record does not explain all transitions between these groups. Thus, an overall enrollment number alone cannot establish the number with evaluable pharmacokinetic data or the number contributing to each safety measure.

The study’s stated purpose was to assess the effect of LY2603618 on CYP2D6 in relation to desipramine. The record separately names maximum plasma concentration of LY2603618 as the primary outcome. This difference matters when describing what the design can answer: the design description identifies the drug sequence, but the endpoint title identifies a specific concentration measure. The registry record does not provide an analysis plan clarifying how all outcomes were intended to address the broader interaction question.

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: This was a nonrandomized, open-label, fixed-sequence, 2-period study followed by a continued access phase.

Period 1

Period 2

Reasons for not completing this period:

Continued Access Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All participants who received LY2603618 and had pharmacokinetic data collected to calculate the Cmax of LY2603618 in Period 2.

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 participants who received at least 1 dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Plasma Pharmacokinetics of LY2603618: the Maximum Concentration of LY2603618 in the Plasma (Cmax), measured over Period 2 only: Predose, 1, 2, 4, 6, 8, 12, 24, 48, 72, 96, 120 and 144 hours post dose.

The primary endpoint is the maximum concentration of LY2603618 in plasma, abbreviated Cmax, with units of nanograms/milliliter (ng/mL). It is specified for Period 2, with pharmacokinetic sampling described at predose and 1, 2, 4, 6, 8, 12, 24, 48, 72, 96, 120 and 144 hours post dose. The stated endpoint summary is a geometric mean with geometric coefficient of variation as its dispersion measure. The endpoint population is participants who received LY2603618 and had pharmacokinetic data collected to calculate Cmax in Period 2.

In general, Cmax describes the highest measured concentration in a drug’s observed plasma profile. It is used in pharmacokinetic work because peak exposure provides one summary of how much drug reaches the circulation at the upper point of the sampled concentration-time profile. The sampling schedule is relevant: a maximum inferred from measured samples depends on when measurements are taken, and may not identify a concentration between scheduled samples.

A geometric mean is a multiplicative summary commonly used for positive, right-skewed concentration data; the geometric coefficient of variation describes relative spread on a multiplicative scale. These are general statistical interpretations, not findings from this study. The registry record does not report the Cmax value, the geometric coefficient of variation, participant-level values, or a confidence interval. Consequently, the endpoint’s estimated level and precision are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The record identifies a single-group, unmasked study with no randomized allocation listed. That design description is important for interpreting the treatment sequence. A period in which participants receive desipramine alone and a later period in which they receive LY2603618 followed by desipramine can support a within-study pharmacology comparison if appropriate measurements and analysis are available. It is not equivalent to assigning separate participants at random to two independent arms. Period, sequence, participant characteristics, and changes over time may all matter in a repeated-period design.

What does Cmax summarize?

In general, a concentration-time profile consists of measurements taken at specified times after dosing. Cmax is the highest concentration observed or derived from that profile. It is a concise measure of peak exposure, but it does not describe the entire profile. Measures such as the timing of the peak or total exposure answer different questions. The registry record names Cmax as the primary outcome and gives a schedule extending from predose through 144 hours post dose. It does not report the observed concentrations or the resulting Cmax estimate.

The distinction between a measured maximum and an underlying biological maximum is also relevant in general. Sampling only occurs at scheduled times, so a peak between samples may not be directly observed. The adequacy of sampling depends on the drug and the profile, but the registry record does not provide a rationale for the selected times. No claim about whether the schedule captured the true peak can be made from this record alone.

Why report a geometric mean?

In general, pharmacokinetic measurements are positive and may vary multiplicatively. A geometric mean summarizes such values by working on a logarithmic scale and then expressing the result on the original scale. The geometric coefficient of variation describes relative variability rather than variability in the original unit as an absolute distance. These summaries can be useful when the distribution is skewed or when proportional differences are more meaningful than fixed-unit differences.

A summary statistic does not replace an uncertainty interval. A geometric mean alone would not show how precisely the population value is estimated. A confidence interval, when reported, gives a range of values compatible with the data and analysis under its assumptions. The record gives the planned summary and dispersion type but does not report the geometric mean, geometric coefficient of variation, confidence interval, or model used to calculate an interval.

What comparison would address an interaction?

In general, a drug-interaction analysis may compare pharmacokinetic measurements under a drug alone with measurements under a coadministration condition. The estimand must be explicit: it should specify which drug’s exposure is compared, which summary is used, and how within-participant or between-participant data are handled. When repeated measurements come from the same participant, a paired analysis can account for participant-level differences. A ratio-based comparison on a logarithmic scale is common in pharmacokinetic settings, but the registry record does not report such an analysis for this study, so no particular method should be attributed to it.

The stated primary measure is LY2603618 Cmax in Period 2. The broader study purpose refers to LY2603618’s effect on CYP2D6 and desipramine metabolism. Those statements do not by themselves specify a direct statistical contrast for desipramine exposure. The analysis plan, any model, covariate adjustments, and the estimate addressing the CYP2D6 question are not reported in the registry record.

How should the posted safety counts be read?

In general, adverse-event counts describe the number of participants with events under a specified exposure and observation period. A participant count is not necessarily an event count: one participant may experience multiple events, and a count of participants affected does not describe the number of episodes. The registry record lists participants at risk and participants affected for serious and other adverse events in each group. It does not provide person-time, detailed event timing, or a comparative statistical model. The continued-access groups also have different participant counts from the two desipramine periods, so their raw counts are not directly comparable as if they arose from a common randomized allocation.

The listed serious-event terms include neutropenia, intestinal obstruction, arthritis infective, device related infection, sepsis, and anaemia, with percentages attached to the terms in the registry. Those entries do not identify, in the information, a treatment-attributable rate or a comparative risk estimate. A percentage attached to a listed term should be read as a descriptive registry entry, not as evidence of causation.

What do planning calculations mean?

In general, a sample-size illustration based on an observed effect is a planning exercise, not an estimate of what the trial actually achieved. The stated illustration assumptions use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is expressed as a fraction of the observed effect. Such assumptions do not establish that the original study used that design or that the effect was observed in this registry record. They also do not constitute observed power. The registry record does not report the study’s original sample-size justification or a protocol-based statistical plan.

Power is a design property under specified assumptions, not a retrospective score proving whether a result is true. In general, recalculating a sample size from a point estimate can be unstable because the observed estimate itself is uncertain. This distinction is especially important when no primary endpoint result is reported: there is no posted study estimate in this record on which to base a study-specific interpretation. Any planning illustration should therefore be kept separate from the trial’s reported findings.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 7 outcome measures, but the record available for interpretation gives a detailed description only for the primary outcome and lists one secondary outcome, number of participants with a tumor response. Statistical analysis results are not reported for the primary measure in the registry record, and no analysis results are reported for the other outcomes. The registry record does not state how outcome testing was ordered or whether any adjustment for multiple comparisons was planned.

Multiplicity arises when several outcomes, time points, or comparisons are tested. In general, testing more hypotheses can increase the chance of at least one apparently unusual result, so an analysis plan may identify a primary hypothesis or specify an adjustment. The outcome count alone does not reveal whether tests were performed, which outcomes were prespecified, or how any multiplicity was handled. Those details are not reported in the registry record.

The primary endpoint population is described as participants who received LY2603618 and had pharmacokinetic data collected to calculate Cmax in Period 2. The record does not report how many participants met that definition, how missing samples were handled, whether any values were excluded, or whether the endpoint analysis used an alternative population. The overall enrollment of 20 should not be assumed to be the endpoint denominator.

More generally, missing-data methods can affect an estimate when participants lack outcome measurements. Analyses may use only observed measurements, make assumptions about the missingness process, or apply other methods. No such method is reported in the registry record. The safety listings give group-specific numbers at risk, but they do not explain all reasons for participant flow or missing information.

The record also does not report interim analyses, stopping rules, covariate adjustment, or a prespecified statistical analysis plan. Their absence from the registry entry does not show whether such procedures were used elsewhere; it means they are not reported in the registry record. Without these details and the endpoint estimates, the multiplicity, missing-data, and analysis-population implications cannot be evaluated fully.

11. 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: neutropenia (14.3%), intestinal obstruction (14.3%), arthritis infective (14.3%), device related infection (14.3%), sepsis (14.3%).

The registry reports serious adverse events for four exposure groups. The listed participant counts affected are 0 of 20 during desipramine in Period 1, 1 of 19 during LY2603618 plus desipramine in Period 2, 4 of 12 during LY2603618 plus gemcitabine in continued access, and 2 of 7 during LY2603618 plus pemetrexed in continued access. The counts for other adverse events are 8 of 20, 12 of 19, 12 of 12, and 7 of 7, respectively. These are descriptive counts, not treatment-effect estimates.

The groups differ in treatment context and size, and the design is not described as randomized. It would therefore be inappropriate to interpret the raw counts as a controlled comparison of risk. The record does not report event duration, timing, participant overlap between periods, person-time, or a statistical comparison. It also does not establish whether events were judged related to study drugs. Deaths are not reported in the registry record, which is distinct from a reported zero.

The serious-event terms listed include neutropenia, intestinal obstruction, arthritis infective, device related infection, sepsis, and anaemia. The attached percentages are registry descriptions and do not establish causation or comparative rates. Counts of participants affected also do not necessarily equal counts of individual event episodes.

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

The study record describes a phase 1 drug-interaction investigation in participants with cancer. It documents a sequence in which desipramine was given alone in Period 1 and with LY2603618 in Period 2. This design is relevant to the stated question about CYP2D6 and desipramine metabolism. The record also identifies a primary pharmacokinetic measure, LY2603618 Cmax, and defines the measurement schedule and summary approach.

Those design details do not establish the study’s findings. The numerical primary endpoint result is not reported in the registry record. Nor is a result that directly quantifies the effect on desipramine metabolism reported. Without an estimate and its uncertainty, the direction, magnitude, and precision of the pharmacokinetic findings cannot be assessed. The absence of a posted result is not evidence that the interaction was absent, nor does it establish that an interaction occurred.

The safety counts document participants affected in the listed periods and continued-access groups. They do not establish that any event was caused by a study drug, and the nonrandomized group structure does not support a like-for-like comparison. The record also does not provide enough detail to assess how exposure duration, participant flow, or event timing may have shaped the counts.

In general, a clear interpretation of a clinical pharmacology study requires alignment between the question, the endpoint, the analysis population, and the estimate. Here, the stated purpose and the named primary endpoint address related but distinct concepts. The registry record does not provide the analysis details or results needed to connect them quantitatively. The record can therefore describe what was intended and how treatment was sequenced, but not the size or certainty of a pharmacokinetic effect.

These limits are about what the registry record reports, not a judgment that the study had no scientific value or that no analyses were conducted elsewhere. The result information needed to evaluate the primary statistical question is not reported in the registry record. No clinical recommendation or treatment conclusion can be drawn from the design description alone.

13. Why This Trial Matters Statistically

This trial is statistically instructive because a study’s broad scientific question and its registered primary endpoint are not always identical in wording or scope. The stated purpose concerns whether LY2603618 affects CYP2D6 in the metabolic pathway of desipramine. The primary outcome is LY2603618’s maximum plasma concentration in Period 2. Both concern clinical pharmacology, but the endpoint title alone is not a direct numerical answer to the desipramine metabolism question.

The distinction underscores why interpretation begins by identifying the estimand: the specific quantity the analysis is intended to estimate. A useful estimand description states the treatment conditions, population, outcome, and summary contrast. For a drug-interaction study, it is particularly important to specify which drug’s exposure is being assessed and how the conditions are compared. The registry record identifies a treatment sequence and a primary concentration measure, but does not report the analysis contrast or the result.

The design also illustrates why group labels should not be mistaken for randomized arms. The record describes a single-group study with no allocation listed and no masking. Its periods and continued-access groups represent distinct exposure contexts. Descriptive counts across such groups do not automatically support causal comparisons, even when the counts are numerically different.

Finally, a registered outcome definition is not a substitute for reporting an estimate and its uncertainty. The record provides the unit, sampling times, summary statistic, and endpoint population, which help readers understand what was intended. Without the outcome value and an uncertainty measure, however, readers cannot assess the precision of the primary estimate. Clear statistical reporting requires both a well-defined measure and the results needed to interpret it.

14. 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 main limitation for statistical interpretation is that the registry record does not report the primary pharmacokinetic result. Although Cmax is defined with units, a geometric-mean summary, a dispersion measure, a sampling schedule, and an endpoint population, no estimate or confidence interval is reported. The record also does not report a numerical outcome that directly quantifies the stated CYP2D6 and desipramine metabolism question.

The design is listed as single-group, with allocation NA and no masking. Consequently, the periods should not be interpreted as randomized, independent treatment arms. Period effects and participant-level differences may matter in a repeated-period pharmacokinetic comparison. The registry record does not report a washout interval, the details of period scheduling, or a statistical method to account for repeated measurements.

The endpoint population is described by data availability and receipt of LY2603618, but its size is not stated. The overall enrollment of 20 does not establish how many participants contributed usable Cmax data. The record does not report missing-data handling, exclusions, or reasons for changes in participant counts across the listed safety groups.

Safety data are presented as participant counts affected in different treatment contexts. The continued-access groups have different numbers at risk from the two desipramine periods, and no randomized comparison or statistical adjustment is reported. Event timing, duration, person-time, and attribution are not reported in the registry record. Death counts are also not reported in the registry record.

The record does not report the original sample-size justification, analysis plan, interim analyses, multiplicity procedures, or detailed statistical models. These omissions limit evaluation of precision, bias, and the intended inferential framework. They do not establish that such methods were not used outside the registry entry; they are simply not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Drug Interaction Study to Assess the Effect of LY2603618 on the Metabolic Pathway of Desipramine — NCT01358968.
• Registry record last updated 2018-10-25; status last verified 2018-03.
• Results first posted 2018-10-25.
• 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.

18. Record Summary

This completed phase 1, single-group study enrolled 20 participants with cancer. It began in 2011-06 and reached primary completion in 2012-04. The record lists no masking and no randomized allocation. Its stated purpose was to assess the effect of LY2603618 on CYP2D6 in relation to desipramine metabolism.

The treatment description specifies 50 mg oral desipramine alone in Period 1, followed in Period 2 by a 275 mg intravenous LY2603618 infusion over 1 hour and then 50 mg oral desipramine. The primary endpoint is LY2603618 Cmax in Period 2, summarized as a geometric mean with geometric coefficient of variation. The registry record does not report its numerical result or a statistical analysis result.

Safety counts are listed for the desipramine periods and continued-access combinations with gemcitabine and pemetrexed. They are descriptive and are not randomized comparisons. Analysis details, missing-data methods, multiplicity procedures, and the original sample-size justification are not reported in the registry record.