A Study of Olaratumab (LY3012207) in Participants With Advanced Soft Tissue Sarcoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Olaratumab (LY3012207) in Participants With Advanced Soft Tissue Sarcoma (NCT02659020), a randomized phase 1, phase 2 study. The primary outcome is Phase 1b: Number of Participants With Dose Limiting Toxicity (DLT).

2. Clinical Question

The central statistical distinction is between the phase 1b safety assessment and the randomized phase 2 comparisons. The primary outcome is a count of participants meeting the dose-limiting toxicity definition during the first cycle, rather than a measure of tumor response or survival. Phase 2 comparisons address outcomes over longer follow-up and contrast the olaratumab combination with the placebo combination, with results separated by prior olaratumab treatment where reported.

The trial included 310 participants overall. That total does not itself describe how many participants contributed to each outcome comparison: the phase 1b and phase 2 results have their own analysis group sizes. The registry record does not report additional eligibility criteria or the original statistical rationale for the total enrollment.

3. Trial Design

The study is listed as phase 1b and phase 2, with randomized allocation, double masking, and a parallel design. Its official title characterizes phase 1b as open-label and phase 2 as randomized and double-blinded. The registry record reports enrollment of 310 participants and a completed status. The recorded study start is 2016-03-01, and primary completion is listed as 2020-07-28.

Phase 1b evaluated dose-limiting toxicity in cohorts receiving olaratumab with gemcitabine and docetaxel. One cohort received olaratumab 15 mg/kg on days 1 and 8. The 20 mg/kg dose was assessed in a cohort and an expansion cohort. The chemotherapy schedule in these descriptions was gemcitabine 900 mg/m^2 on days 1 and 8 and docetaxel 75 mg/m^2 on day 8 of a 21-day cycle. Treatment continuation criteria are described in the registry, but the primary safety outcome focuses on cycle 1.

Phase 2 compared olaratumab plus gemcitabine and docetaxel with placebo plus the same chemotherapy. The olaratumab regimen used a 20 mg/kg loading dose on days 1 and 8 of cycle 1, followed by 15 mg/kg on days 1 and 8 of subsequent cycles. The registry reports phase 2 outcomes separately for olaratumab-naive and olaratumab pre-treated participants for several endpoints.

The randomized, masked phase 2 design supports a more direct comparison between assigned treatment groups than the non-placebo phase 1b dose cohorts. The phase 1b counts should not be interpreted as a randomized treatment effect. The registry record does not report the detailed randomization procedure, allocation ratio, stratification factors, or the original design assumptions for enrollment.

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: Completers included participants who died from any cause.

Only the first 6 of 7 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Phase 1b: All participants who received at least one dose of Olaratumab.

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 from phase 1b/2.

Only the first 6 of 8 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Phase 1b: Number of Participants With Dose Limiting Toxicity (DLT), measured over Cycle 1 (Up To 21 Days).

The primary endpoint is the number of phase 1b participants with dose-limiting toxicity during cycle 1, up to 21 days. A dose-limiting toxicity is defined as an adverse event considered likely related to the study medication or combination that meets at least one specified severity criterion. The criteria include febrile neutropenia with documented grade 3 or higher infection or sepsis; grade 4 neutropenia lasting 7 days or longer; grade 4 thrombocytopenia or grade 3 thrombocytopenia complicated by hemorrhage; and certain nonhematologic grade 3 or higher toxicities, with stated exceptions for events manageable within 48 hours.

The analysis population is described as all participants who received at least one dose of olaratumab. The endpoint is a participant count, so each person is counted according to whether the defined event occurred during the specified period. It is not a count of all adverse events, and it does not convey how many events an individual experienced.

As a general design principle, a dose-limiting toxicity endpoint is used in early-phase research to describe whether a treatment combination can be administered at a studied dose without an unacceptable frequency of specified early toxicities. The limited cycle 1 window provides a defined period for this assessment. Here, the registry gives the definition and cohort counts, but does not report a formal dose-selection rule or a statistical decision threshold for choosing a dose. The phase 1b endpoint also does not, by itself, answer the phase 2 questions about survival, progression, or response.

7. Results Posted in the Registry

• Phase 1b: Cohort 2 - 20 mg/kg Olaratumab + Gemcitabine + Docetaxel: 22.2% (4/18) versus 4.8% (1/21) on control; risk difference 17.5 percentage points (95% CI -3.8 to 38.7), risk ratio 4.67 (0.57–38.07); unadjusted two-proportion z-test p = 0.104.
• Phase 1b: Cohort 2 Expansion - 20 mg/kg Olaratumab + Gemcitabine + Docetaxel: 20.0% (3/15) versus 4.8% (1/21) on control; risk difference 15.2 percentage points (95% CI -7.0 to 37.4), risk ratio 4.20 (0.48–36.56); unadjusted two-proportion z-test p = 0.151.

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.

Phase 2: Overall Survival (Olaratumab Pre-Treated)

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Log Rank); it was not recalculated.
• Result:
  • Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab Pre-treated) vs Phase 2: Placebo + Gemcitabine + Docetaxel (Olaratumab Pre-treated): hazard ratio 0.67 (95% CI 0.39–1.16), reported p = 0.148 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab Pre-treated) vs Phase 2: Placebo + Gemcitabine + Docetaxel (Olaratumab Pre-treated) and the control. The interval gives the range of effects that remain compatible with the data.

Phase 2: Progression Free Survival (PFS)

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Log Rank); it was not recalculated.
• Result:
  • Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab-naive) vs Phase 2: Placebo + Gemcitabine + Docetaxel (Olaratumab-naive): hazard ratio 0.69 (95% CI 0.48–1.01), reported p = 0.055 (as reported in the registry).
  • Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab Pre-treated) vs Phase 2: Placebo + Gemcitabine + Docetaxel (Olaratumab Pre-treated): hazard ratio 0.83 (95% CI 0.49–1.40), reported p = 0.482 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab-naive) vs Phase 2: Placebo + Gemcitabine + Docetaxel (Olaratumab-naive) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab Pre-treated) vs Phase 2: Placebo + Gemcitabine + Docetaxel (Olaratumab Pre-treated) and the control. The interval gives the range of effects that remain compatible with the data.

Phase 2: Percentage of Participants With a Complete or Partial Response (Objective Response Rate [ORR])

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab-naive): 32.1% (26/81) versus 23.3% (20/86) on control; risk difference 8.8 percentage points (95% CI -4.7 to 22.4), risk ratio 1.38 (0.84–2.27); unadjusted two-proportion z-test p = 0.201.
  • Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab Pre-treated): 30.4% (14/46) versus 23.3% (20/86) on control; risk difference 7.2 percentage points (95% CI -8.8 to 23.2), risk ratio 1.31 (0.73–2.34); unadjusted two-proportion z-test p = 0.369.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab-naive) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Phase 2: Olaratumab + Gemcitabine + Docetaxel (Olaratumab Pre-treated) and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes broaden the statistical picture beyond the phase 1b safety count. They include time-to-event measures, tumor response, and pharmacokinetic measures. For the phase 2 outcomes with numerical results described in the registry, the olaratumab groups have numerically higher response estimates and lower estimated hazard ratios than their corresponding placebo groups. That directional pattern does not translate into consistently precise evidence: the intervals are broad, and the reported tests do not exclude the null comparisons.

The phase 1b dose-limiting toxicity endpoint addresses an early safety question in small cohorts, while phase 2 progression-free survival, overall survival, and response address different clinical outcomes over longer windows. They are related in the broad sense that they assess the same treatment combination, but they do not measure the same construct. A difference in response percentage does not establish a difference in survival, and a safety count from cycle 1 does not predict a later time-to-event outcome.

Secondary results generally carry less confirmatory weight when many outcomes or subgroup comparisons are considered, particularly if no adjustment for multiple testing or prespecified hierarchy is reported. The registry record does not report a multiplicity strategy for these outcomes. A nominal p-value should therefore be considered alongside the estimate, its interval, the endpoint definition, the analysis method, and the broader set of analyses. The simple unadjusted recalculations are useful descriptive checks, but they are not substitutes for the registry’s reported analyses or a prespecified analysis plan.

Pharmacokinetic outcomes are listed, including measures of olaratumab, gemcitabine, and docetaxel concentrations and olaratumab clearance. Numerical results for these outcomes are not reported in the registry record information considered in this summary, so their direction and precision cannot be characterized. The count of listed outcome measures alone does not show whether results were favorable, unfavorable, or inconclusive.

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

• Phase 1b: Pharmacokinetics (PK): Maximum Serum Concentration (Cmax) of Olaratumab
• Phase 1b: PK: Minimum Serum Concentration (Cmin) of Olaratumab
• Phase 1b: PK: Elimination Half-Life (T1/2) of Olaratumab
• Phase 1b/2: PK: Cmax of Gemcitabine
• Phase 1b/2: PK: Area Under the Concentration-Time Curve From Time Zero to Infinity (AUC[0-∞]) of Gemcitabine
• Phase 1b/2: PK: Cmax of Docetaxel
• Phase 1b/2: PK: Area Under the Concentration Versus Time Curve From Time Zero to Infinity (AUC [0-∞]) of Docetaxel
• Phase 1b/2: Population PK: Clearance of Olaratumab

9. Statistical Interpretation

10. Statistical Methods Explained

Several kinds of statistics appear in the registry results, and they answer different questions. The primary phase 1b measure is a count of participants with a defined event. The phase 2 time-to-event outcomes are summarized with medians and hazard ratios. Objective response is a binary outcome summarized as a percentage and compared with odds ratios in the registry’s reported analyses. General statistical explanations below describe what these measures mean; they do not add information about analyses that the registry record does not describe.

What does a participant count show?

A count identifies how many participants in a cohort met the outcome definition, while the group size identifies how many participants were assessed. The phase 1b endpoint counts participants with at least one qualifying dose-limiting toxicity during cycle 1. It does not count repeated episodes separately. Because the cohorts are limited in size, a small change in the number of participants with an event could materially alter the estimated event frequency. The registry record does not provide a confidence interval for the primary count itself.

For the dose-cohort comparisons, the unadjusted recalculations express the difference in observed event proportions as a risk difference. A risk difference near zero indicates similar observed proportions; a positive value indicates a higher observed proportion in the second cohort in the comparison. These recalculations are descriptive and unadjusted. Their confidence intervals represent uncertainty in the estimated difference under the calculation method; intervals crossing zero do not rule out either direction of difference. The odds ratios and relative risks are alternative ways to express the same binary-outcome comparisons, but they are not interchangeable with a risk difference.

How should time-to-event outcomes be read?

Overall survival measures time from baseline to death from any cause, and progression-free survival measures time from baseline to objective disease progression or death from any cause. A median is the time by which half of the relevant observed survival distribution has experienced the event, under the usual interpretation of a time-to-event median. It is not the average follow-up time and does not describe every participant’s experience.

The hazard ratio summarizes the relative event rate over follow-up under the assumptions of the analysis. A value below 1 means that the estimated event rate is lower in the named olaratumab group relative to its comparator as coded in the reported comparison. It does not directly state how many months an individual gains, and it is not the same quantity as the ratio of the two medians. The registry reports log-rank analyses for these comparisons. The detailed model specification and assessment of assumptions are not reported in the registry record.

For the pre-treated overall-survival comparison, the registry reports median values and a hazard ratio. For progression-free survival, median values are reported for both treatment-history strata, along with hazard ratios. Differences between median estimates and hazard-ratio estimates are not inherently contradictory: they summarize different features of the time-to-event distributions. General interpretation of a hazard ratio also relies on the time-to-event analysis and its assumptions, which are not fully described in the record.

What do response percentages and odds ratios describe?

Objective response rate is the percentage of participants recorded as having a complete or partial response during the stated assessment window. A response percentage is a group-level proportion, not a measure of response duration or survival. The confidence intervals around the group percentages express uncertainty in those proportions. The recalculated risk difference compares the proportions directly. The registry’s reported comparison uses an exact Mantel-Haenszel test and reports an odds ratio with a confidence interval and p-value. An odds ratio above 1 indicates higher estimated odds in the first group under the reported ordering, but it should not be interpreted as a risk difference or as a direct percentage-point change.

General statistical principle: an exact test can be useful for binary outcomes when group counts are limited, because its calculation does not rely on the same large-sample approximation used by some methods. The registry identifies the method used for its reported response comparisons, but does not provide additional implementation details.

Why do confidence intervals and p-values differ in what they convey?

A confidence interval gives a range of estimates compatible with the observed data under a specified method and repeated-sampling interpretation. It communicates precision as well as the direction and size of estimates. A p-value measures how incompatible the data are with a particular null hypothesis under the test assumptions. It does not give the probability that the null hypothesis is true, and it does not measure clinical importance.

For the time-to-event results, intervals that include 1 do not exclude a null hazard ratio of 1 at the stated confidence level. For risk differences, intervals that include zero do not exclude equal proportions. These are statistical readings of the interval boundaries, not proof that groups are identical. Broad intervals indicate that a wide range of effects remains compatible with the data. The registry’s reported and unadjusted recalculated analyses should be kept distinct, since their methods and estimates can differ.

Why does over-dispersion matter in general?

General explanation: over-dispersion occurs when observed variability exceeds what a chosen statistical model expects. It can make standard errors too small if ignored, leading to intervals that are too narrow and tests that appear more decisive than warranted. The endpoints are counts of participants, binary response outcomes, and time-to-event measures; the registry record does not report an over-dispersion assessment or a model for the primary count. It would therefore be inappropriate to assert that over-dispersion was present or to infer how it was handled.

The registry also lists pharmacokinetic outcomes, including concentration and clearance measures, but numerical results for those measures are not included in the outcome information. The existence of a listed outcome does not establish what its estimate was or how it was analyzed. Similarly, the record does not report enough detail to reconstruct the full statistical analysis plan, missing-data procedures, or analysis populations for every secondary outcome.

11. Confidence Intervals and Precision

Precision is conveyed most directly by the confidence intervals, not by the point estimates alone. In the phase 1b comparisons, the recalculated risk-difference intervals are -0.038 to 0.387 for the 20 mg/kg cohort comparison and -0.070 to 0.374 for the expansion cohort comparison. Both include zero and span values on either side of it. The observed cohort counts are therefore not precise estimates of the difference between these small groups.

For progression-free survival in olaratumab-naive participants, the reported hazard-ratio interval is 0.476 to 1.007. The pre-treated comparison has a wider interval, 0.490 to 1.398. For pre-treated overall survival, the reported interval is 0.385 to 1.158. Each includes 1, although the width differs. The response risk-difference intervals are -0.047 to 0.224 for olaratumab-naive participants and -0.088 to 0.232 for pre-treated participants; both include zero.

General interpretation: a 95% confidence interval is a range generated by a method designed to cover the true parameter in 95% of repeated samples under its assumptions. It is not a guarantee that a particular interval contains the true value. An interval crossing the null value indicates that the data do not exclude that value at the corresponding confidence level, not that the groups have been shown equivalent.

Precision is limited by the size of the groups contributing to each comparison and by the number of observed events. The registry does not report every confidence-interval bound for every measure: for example, one overall-survival median has a listed lower bound but no upper bound in the reported summary. The record also does not provide a confidence interval for the primary phase 1b participant count itself. Such gaps should be distinguished from evidence of no uncertainty.

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 the observed effect, not the trial’s original design calculations. For binary outcomes, the figures refer to participants per arm. For time-to-event outcomes, they refer to the total number of events, not the number of participants enrolled. This distinction matters because an event target and an enrollment target are different quantities.

The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each scenario uses a stated fraction of the observed effect: 100%, 75%, or 50%. The 80% power column describes the number of participants per arm, or total events for a time-to-event comparison, that the planning calculation associates with an 80% chance of detecting the chosen effect if that effect and the assumptions were correct. The 90% column describes the corresponding target for a 90% chance.

Smaller effects require more information to detect under these assumptions, so the planning estimates increase as the effect fraction decreases. These figures do not indicate the probability that the observed trial result is correct, and they do not establish the power the completed trial actually had. This is not observed power. The original trial design assumptions and sample-size justification are not reported in the registry record. The calculations should therefore be used only as illustrations of how target size changes with the assumed effect and desired power, not as a reconstruction of the trial’s planned design.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.048; treatment proportion p₁ = 0.222; average p̄ = 0.135.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4832 = 0.9470.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4671 = 0.3931.
4. n = (0.9470 + 0.3931)² / (0.1746)² = 58.9, rounded up to 59 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
p0 <- 0.0476190476190476                          # control proportion
p1 <- 0.222222222222222                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2659020)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 59 (80% power), 79 (90% power)
 75% of observed effect: n per arm = 91 (80% power), 121 (90% power)
 50% of observed effect: n per arm = 170 (80% power), 227 (90% power)
power.prop.test: 58.9 
Simulated power at n = 59 per arm: 0.831 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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. Multiplicity, Missing Data and Analysis Populations

The registry record lists multiple outcomes and comparisons, including separate phase 2 analyses by prior olaratumab treatment for some endpoints. When many outcomes and subgroups are tested, the chance of at least one apparently notable result arising by chance can increase. General statistical practice may address this through a prespecified hierarchy or multiplicity adjustment, but neither should be presumed without documentation. A multiplicity strategy is not reported in the registry record.

The registry gives methods for some reported analyses: log-rank analyses for the time-to-event comparisons and exact Mantel-Haenszel tests for objective response. It also includes unadjusted recalculations of selected comparisons. These methods and estimates should be distinguished. The record does not report a complete analysis plan, a prespecified primary phase 2 endpoint, adjustment across endpoints, or a rationale for the selection of all reported comparisons.

Information about missing observations and censoring is also limited in the outcome summaries. Time-to-event analyses ordinarily account for participants whose event has not been observed by the end of follow-up through censoring, but the detailed censoring rules and follow-up completeness are not reported in the registry record. Missing-data methods for response and other outcomes are not reported in the registry record. It is therefore not possible to determine from these summaries how missing assessments were handled or how sensitive results were to alternative assumptions.

The primary phase 1b population is described as participants who received at least one dose of olaratumab. The analysis populations for all phase 2 outcomes, including whether analyses followed randomized assignment or another population definition, are not fully reported in the registry record information. The randomization procedure, allocation concealment, interim analyses, and detailed handling of protocol deviations are also not reported in the registry record. These omissions limit independent assessment of how analyses were selected and applied, but they do not justify assuming that a particular method was or was not used.

15. 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: pyrexia (14.0%), pneumonia (11.6%), pneumonitis (9.3%), febrile neutropenia (7.4%), thrombocytopenia (6.7%).

The serious adverse event summary gives the number at risk and the number affected in each listed group. In phase 1b, serious adverse events affected 6 of 21 participants in the 15 mg/kg cohort, 9 of 18 in the 20 mg/kg cohort, and 9 of 15 in the expansion cohort. For phase 2, the corresponding counts are reported separately by treatment and prior olaratumab status. These are participant counts, not event counts, and the registry summaries do not establish that every event was caused by study treatment.

The most frequently listed serious adverse event terms are pyrexia at 14.0%, pneumonia at 11.6%, pneumonitis at 9.3%, febrile neutropenia at 7.4%, thrombocytopenia at 6.7%, and atrial tachycardia at 6.7%. These percentages describe the listed terms, not the probability that a particular participant will experience any serious event. Terms may also represent different clinical events and should not be combined into a new total.

Deaths are separately counted in the group summaries, but the registry information does not specify causes or establish attribution. Safety comparisons across groups require caution because the treatment-history categories and group sizes differ. These summaries describe reported adverse event counts; they are not, by themselves, a causal comparison of safety.

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

The results describe observed outcomes in participants enrolled in this study and the uncertainty around selected comparisons. In phase 1b, the registry reports dose-limiting toxicity counts for three cohorts receiving olaratumab with gemcitabine and docetaxel. In phase 2, it reports comparisons with placebo plus the same chemotherapy for selected survival and response outcomes. The numerical estimates are not equally precise, and several confidence intervals include the null value for the corresponding measure.

A point estimate is not a verdict. Hazard ratios below 1 and numerically higher response percentages in the olaratumab groups indicate the direction of the estimates under the comparisons as reported. Their intervals and p-values show that uncertainty remains. Conversely, a confidence interval that includes the null does not establish that the treatments are identical. It means the data and analysis do not rule out the null value at the stated level; other values within the interval may also remain compatible with the observations.

The phase 1b safety cohorts are not the same kind of comparison as the randomized, double-blinded phase 2 groups. The dose-cohort counts are descriptive and should not be used to infer a randomized treatment effect. Phase 2 findings apply to the participants and outcomes represented in the registry results; they do not automatically extend to populations or settings not described in the record. The registry summaries also do not provide enough detail to assess every eligibility, missing-data, or analysis-population issue.

These results alone do not establish that olaratumab improves or fails to improve outcomes, and they should not be used as medical advice. They provide statistical estimates whose interpretation depends on the study design, definitions, analysis methods, uncertainty, and the full set of outcomes. The registry record does not report a multiplicity strategy or all original design assumptions, further limiting how strongly any individual secondary finding should be treated as confirmatory.

17. Why This Trial Matters Statistically

This trial illustrates why the design stage and the analysis stage answer different statistical questions. The phase 1b portion reports early dose-limiting toxicity counts across dose cohorts, while the randomized phase 2 portion compares a treatment combination with a placebo combination. Keeping those questions separate prevents descriptive safety counts from being mistaken for randomized evidence about later outcomes.

It also demonstrates the value of reading estimates together with their uncertainty. The survival and response point estimates have directions that can appear suggestive when viewed alone, but the confidence intervals show substantial uncertainty for several comparisons. A test result or point estimate without its interval would leave out important information about precision and the range of effects compatible with the data.

The study further shows how outcome choice shapes interpretation. A cycle 1 toxicity count, progression-free survival, overall survival, and objective response are distinct measures with different time windows and statistical scales. They should not be collapsed into a single claim of efficacy or safety. The separate olaratumab-naive and pre-treated results also make clear that subgroup estimates can have different precision, especially when group sizes are limited.

Finally, the record makes transparent some limits on retrospective statistical interpretation. It reports methods for selected endpoints, but does not report a full multiplicity strategy, detailed missing-data rules, or the original sample-size assumptions. Planning illustrations based on observed effects can help explain the scale of information needed under stated assumptions, but cannot recover the original trial design. This is a useful case for separating what the posted data quantify from what remains undocumented.

18. Limitations

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

The primary endpoint is a short-window phase 1b safety measure, while the principal phase 2 outcomes address survival and response. The endpoint that defines the primary result therefore does not directly answer the longer-term phase 2 comparisons. Phase 1b cohort counts are modest, and the dose cohorts do not provide a placebo-controlled comparison. Their observed differences are imprecise and descriptive.

The phase 2 results are reported for particular treatment-history groups and endpoints, but the registry record does not describe all analysis-population details or the full eligibility criteria in the outcome information. The pre-treated overall-survival result is reported for that subgroup; an overall-survival comparison for the olaratumab-naive subgroup is not included in the listed numerical results. Such absence should not be interpreted as a null finding.

Confidence intervals are broad for several comparisons, and the registry does not report a multiplicity strategy despite the presence of several outcomes and subgroup comparisons. The results may therefore be difficult to interpret as a single confirmatory set. Details of missing-data handling, censoring rules, interim analyses, and the original statistical analysis plan are not reported in the registry record.

Numerical results for the listed pharmacokinetic outcomes are not reported in the registry record information. In addition, some summaries provide incomplete interval bounds, including the reported median overall survival for the olaratumab pre-treated treatment group. These limitations constrain assessment of precision and consistency across all outcomes. They do not imply that unreported analyses were not conducted; they mark what cannot be established from the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study of Olaratumab (LY3012207) in Participants With Advanced Soft Tissue Sarcoma — NCT02659020.
• Registry record last updated 2022-05-09; status last verified 2022-04.
• Results first posted 2021-10-11.
• 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.

22. Record Summary

The registry record describes a completed phase 1b and phase 2 study of olaratumab with gemcitabine and docetaxel in participants with advanced soft tissue sarcoma. The phase 1b primary outcome is the number of participants with a dose-limiting toxicity during cycle 1. Phase 2 compares olaratumab plus chemotherapy with placebo plus chemotherapy and reports selected survival and objective response outcomes, including results by prior olaratumab treatment for several measures.

The primary safety counts are small and descriptive. Phase 2 point estimates show numerical differences in some outcomes, but the reported confidence intervals leave substantial uncertainty and do not establish a clear between-group distinction. The registry record reports methods for selected analyses but does not report a complete multiplicity strategy, original sample-size rationale, or detailed missing-data procedures. Interpretation should therefore remain tied to the specific outcomes, groups, and statistical methods documented in the record.