A Study to Compare Subcutaneous (SC) Versus Intravenous (IV) Administration of Herceptin (Trastuzumab) in Women With Human Epidermal Growth Factor Receptor (HER) 2-Positive Early Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Compare Subcutaneous (SC) Versus Intravenous (IV) Administration of Herceptin (Trastuzumab) in Women With Human Epidermal Growth Factor Receptor (HER) 2-Positive Early Breast Cancer (NCT00950300), a randomized phase 3 study. The primary outcome is Observed Serum Trough Concentration (Ctrough) of Trastuzumab Prior to Surgery.

2. Clinical Question

3. Trial Design

This was a phase III, randomized, open-label, parallel-group study sponsored by Hoffmann-La Roche. The registry lists an enrollment of 596 participants and a completed status. The record gives a start date of 2009-10-16 and a primary completion date of 2011-07-12. It describes a multicenter trial, although the number and locations of centers are not reported in the registry record.

Participants were assigned to subcutaneous or intravenous trastuzumab, each alongside chemotherapy. The regimen described eight 21-day cycles before surgery. Docetaxel was given during the first four cycles, followed by 5-fluorouracil, epirubicin, and cyclophosphamide during cycles five through eight. After surgery, participants were to receive ten more cycles of trastuzumab by their assigned route. The registry summary describes follow-up for safety and efficacy for up to five years after study treatment or until disease recurrence, whichever occurred earlier.

The primary endpoint was pharmacokinetic rather than a clinical event endpoint: observed trastuzumab serum trough concentration before surgery, measured pre-dose on day one of cycle eight. Its primary analysis population was a per-protocol pharmacokinetic population, not simply every enrolled participant. The endpoint population included participants with at least one measurable trastuzumab concentration and excluded those with major protocol violations related to pharmacokinetic sampling.

The registry does not report the randomization method, allocation concealment procedures, the intended sample-size justification, or the detailed statistical analysis plan. It does not report how the open-label design was managed for endpoint measurement. Although the record describes randomization and parallel groups, further operational details should not be inferred from those labels.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: A total of 833 participants were screened, out of which, 596 participants were enrolled into the study.

Neoadjuvant/Adjuvant Treatment Periods

Reasons for not completing this period:

TFFU Period

Reasons for not completing this period:

SFU Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Primary Pharmacokinetic (PK) Per Protocol (PP) Population: All participants with at least one measurable trastuzumab serum concentration and who did not have any major protocol violations related to PK sampling for the primary endpoint.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Intent-to-Treat (ITT) Population: All participants with at least one efficacy assessment after administration of the first treatment cycle.

6. Primary Endpoint

The registry lists the primary outcome as Observed Serum Trough Concentration (Ctrough) of Trastuzumab Prior to Surgery, measured over Pre-dose (0 hours) on Day 1 of Cycle 8 (cycle length of 21 days).

The primary endpoint was the observed serum trough concentration of trastuzumab before surgery. A trough is a drug concentration measured near the end of a dosing interval, just before the next dose. In this study, samples were taken pre-dose at 0 hours on day one of cycle eight, with a 21-day cycle length. Concentration was expressed in μg/mL, and the registry identifies the endpoint summary as a mean with standard deviation.

As a pharmacokinetic measure, trough concentration describes drug exposure at a particular point in the dosing cycle. Comparing this measure across administration routes can characterize whether the routes produce different measured concentrations under the trial regimen. It is not itself a measure of tumor response, recurrence, survival, or quality of life. The registry lists other pharmacokinetic outcomes as secondary measures, including trough concentrations at other time points and additional concentration measures.

The posted primary analysis population was the primary pharmacokinetic per-protocol population. Participants needed at least one measurable serum concentration and could not have major protocol violations related to sampling for the primary endpoint. That restriction matters when interpreting the analyzed group: the reported result describes the eligible pharmacokinetic analysis population, rather than necessarily every randomized or enrolled participant. The number screened, the number randomized, and detailed reasons for exclusion from this endpoint population are not reported in the registry record.

7. Results Posted in the Registry

• Herceptin SC + Chemotherapy: mean 78.70 versus 57.80 on control; difference 20.90 (95% CI 14.07 to 27.73), standardized effect d = 0.55; 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.

Observed Ctrough of Trastuzumab After Surgery

• 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 deviations.
• Result:
  • Herceptin SC + Chemotherapy: mean 90.40 versus 62.10 on control; difference 28.30 (95% CI 20.99 to 35.61), standardized effect d = 0.71; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Herceptin SC + Chemotherapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Predicted Ctrough of Trastuzumab Prior to Surgery

• 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 deviations.
• Result:
  • Herceptin SC + Chemotherapy: mean 80.30 versus 51.40 on control; difference 28.90 (95% CI 24.38 to 33.42), standardized effect d = 1.06; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Herceptin SC + Chemotherapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Predicted Ctrough of Trastuzumab After Surgery

• 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 deviations.
• Result:
  • Herceptin SC + Chemotherapy: mean 80.60 versus 51.70 on control; difference 28.90 (95% CI 23.93 to 33.87), standardized effect d = 1.05; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Herceptin SC + Chemotherapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary pharmacokinetic outcomes cover trough concentrations at additional time points, predicted concentrations, concentrations above a stated threshold, and other concentration-profile measures. Twenty secondary outcomes are posted, while the registry lists 22 outcome measures overall. The available continuous secondary concentration comparisons generally have the same direction as the primary result: the posted subcutaneous-arm means are higher, and the sponsor-reported geometric mean ratios are above one. This overall pattern is consistent with the primary pharmacokinetic contrast, without making the secondary outcomes independent confirmation of a clinical effect.

Secondary outcomes can provide context about whether the pharmacokinetic pattern is seen across different measures and assessment times. They remain distinct endpoints, with their own definitions, populations, and uncertainty. The overall pattern does not show that the treatment routes differ in efficacy or patient-important outcomes; the outcomes are pharmacokinetic measures. Results for secondary outcomes that are not posted are not reported in the registry record.

Secondary findings generally carry less inferential weight than a pre-specified primary analysis, particularly when many outcomes are evaluated. Without a reported multiplicity strategy, nominal or unadjusted significance levels across multiple secondary comparisons can overstate how much evidence any single result contributes. The registry record does not report whether multiplicity adjustment was planned or applied, nor does it provide a hierarchy or confirmatory testing sequence for the secondary outcomes.

Accordingly, the secondary findings are best treated as supportive descriptive context. Their directional consistency with the primary pharmacokinetic result can be noted, but it cannot resolve uncertainty about clinical outcomes, the sponsor’s non-inferiority margin, or the statistical handling of multiple comparisons.

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

• Observed Ctrough of Trastuzumab After Surgery
• Predicted Ctrough of Trastuzumab Prior to Surgery
• Predicted Ctrough of Trastuzumab After Surgery
• Number of Participants With Ctrough of Trastuzumab >20 μg/mL Prior to Surgery
• Number of Participants With Ctrough of Trastuzumab >20 μg/mL After Surgery
• Maximum Serum Concentration (Cmax) of Trastuzumab Prior to Surgery
• Time of Maximum Serum Concentration (Tmax) of Trastuzumab Prior to Surgery
• Area Under the Concentration-Time Curve From 0 to 21 Days (AUC21d) of Trastuzumab Prior to Surgery

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous: each analyzed participant contributes a serum concentration measured in μg/mL, and the registry summarizes the groups using means and standard deviations. The primary endpoint is observed trough concentration before surgery. The reported comparison is between subcutaneous trastuzumab plus chemotherapy and intravenous trastuzumab plus chemotherapy.

What does a mean difference express?

A mean difference subtracts the average in one group from the average in the other, using the stated direction of comparison. It remains in the original measurement units, so the primary recalculated difference is expressed in μg/mL. A difference is an absolute contrast, not a relative percentage or ratio. The posted arm means and the recalculated mean difference provide complementary descriptions: one gives each group’s average, and the other gives the estimated separation between those averages.

The simple recalculation uses the posted group means, standard deviations, and group sizes. It is unadjusted, meaning it does not account for covariates or other features of the trial beyond those summary statistics. The registry does not report an adjusted mean-difference model for this result. Nor does it provide enough detail to infer whether the sponsor’s geometric-mean-ratio procedure used a particular transformation, model, or adjustment.

What does a confidence interval convey?

A confidence interval describes uncertainty around an estimated effect under the statistical method used. For the recalculated mean difference, the interval gives a range of values compatible with the observed data under that method. It is not a range containing a stated proportion of individual participants’ concentrations, and it is not a guarantee that the true effect lies inside the particular interval. A narrower interval generally indicates greater precision, while a wider interval indicates more uncertainty. The interval’s interpretation depends on the assumptions of the method and the data.

The registry’s sponsor analysis reports a geometric mean ratio with a 90% confidence interval. A ratio compares quantities on a relative scale, unlike the absolute mean difference. The record labels this analysis non-inferiority, but does not report the non-inferiority margin, its rationale, or enough analysis detail to reconstruct the decision. A confidence interval for a ratio can be compared with a pre-specified margin only when that margin and the applicable decision rule are known. They are not reported in the registry record.

What does the p-value add?

The unadjusted recalculation reports p < 0.001. In general, a p-value measures how incompatible the observed data are with a specified null hypothesis, under the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not quantify the clinical importance of a difference. The registry does not report the statistical test used for the sponsor analysis or a multiplicity-adjusted p-value for the primary comparison. The recalculated p-value should therefore be understood as a simple unadjusted analysis, not as a complete account of the sponsor’s inferential procedure.

The primary endpoint population is also part of the method. The registry describes a pharmacokinetic per-protocol population requiring measurable serum concentration data and no major protocol violations related to primary-endpoint sampling. The posted group sizes for the primary summary are 235 and 234. Those analysis counts differ from the overall enrollment figure, and the registry does not report a full participant-flow account or reasons for every difference in denominators.

General statistical context: standard deviations describe variability among participant-level measurements, whereas standard errors describe uncertainty in estimated group means. A standard deviation should not be interpreted as uncertainty in the mean. The posted endpoint summaries include standard deviations; the recalculation uses the summarized data to estimate a group contrast and its uncertainty. Individual-level distributions are not available in the registry record, so the shape of concentrations within either arm cannot be assessed from the reported means and standard deviations alone.

General statistical context: pharmacokinetic measures can be analyzed on an arithmetic or geometric scale. The arithmetic mean gives an average in the original units; a geometric mean ratio describes a relative contrast and commonly arises when values are analyzed on a logarithmic scale. These scales answer related but distinct questions. The registry supplies both a recalculated arithmetic mean-difference summary and a sponsor geometric mean ratio, but does not report the sponsor’s detailed calculation method. Neither measure should be silently substituted for the other.

For secondary continuous outcomes, the same broad distinction applies: an unadjusted mean comparison describes absolute separation on the measurement scale, while a geometric mean ratio describes relative separation. Multiple secondary outcomes also raise a multiplicity issue. If many hypotheses are examined, the chance of at least one apparently unusual result can increase unless an analysis plan addresses that issue. Whether and how such adjustment was planned or applied is not reported in the registry record. Statistical significance for an individual secondary result should therefore be interpreted in the context of the endpoint’s secondary status and the missing multiplicity details.

11. Confidence Intervals and Precision

The primary recalculated mean difference is 20.9 μg/mL, with a 95% confidence interval from 14.07 to 27.73 μg/mL. The interval gives an indication of precision for this unadjusted estimate: it includes values representing a smaller or larger separation than the point estimate. It is not a prediction interval for individual participants, and it does not capture every possible source of bias or uncertainty.

The interval is based on the posted group summaries and analyzed group sizes. It reflects sampling uncertainty under the assumptions of the recalculation, but does not account for unreported features such as protocol deviations beyond the stated per-protocol definition, missing-data handling, or model choices. The registry does not provide individual-level data or a full analysis specification, so those aspects cannot be evaluated from the record.

The sponsor’s result is reported on another scale: the geometric mean ratio is 1.33, with a 90% confidence interval from 1.24 to 1.44. A ratio interval and a mean-difference interval are not directly comparable in width because they use different scales and confidence levels. The registry labels the sponsor result as a non-inferiority analysis, but does not report the margin or enough methodological detail to judge the interval against a decision threshold.

In general, confidence intervals help distinguish a point estimate from its uncertainty. They do not determine whether a result is clinically important, and statistical precision does not resolve questions about the relevance of a pharmacokinetic endpoint to clinical outcomes. Here, the interval describes precision of the concentration comparison; clinical implications require evidence beyond this endpoint.

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 planning figures are participants per arm for 80% power and participants per arm for 90% power. They are planning illustrations based on the observed primary effect, not the original trial’s design justification. The registry record does not report the trial’s own sample-size assumptions or justification.

The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The assumed effect to detect is expressed as a stated fraction of the observed effect. The rows for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect therefore represent progressively smaller assumed effects relative to the observed primary comparison.

The 80% power column indicates the participants per arm that the planning calculation associates with an 80% chance of detecting the assumed effect under those assumptions, if the effect and model assumptions hold. The 90% power column gives the corresponding per-arm planning figure for a 90% chance. These are design calculations, not evidence about the probability that the observed trial result is true. This is not observed power.

For the primary endpoint, the figures range from 52 participants per arm for 80% power and 69 for 90% power when planning for 100% of the observed effect, to 205 and 274 when planning for 50% of the observed effect. The 75% row gives intermediate planning figures. Smaller effects need more participants to detect under the same assumptions, which is why the required arm sizes increase as the assumed effect decreases. The curve follows that same relationship: moving toward smaller target effects corresponds to larger planned group sizes.

These calculations are illustrations, not a retrospective judgment about whether the trial was adequately designed. They assume the effect sizes and statistical conditions and do not incorporate unreported design features, analysis populations, or missing-data procedures. The figures are per arm, not total participants across both arms and not total events.

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 = 20.900; pooled standard deviation σ = 37.704.
2. n = 2 × (1.96 + 0.842)² × 37.704² / 20.900² = 51.1, rounded up to 52 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 <- 20.9                      # observed difference in means
sigma <- 37.7037076656192                     # 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(950300)
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 = 52 (80% power), 69 (90% power)
 75% of observed effect: n per arm = 91 (80% power), 122 (90% power)
 50% of observed effect: n per arm = 205 (80% power), 274 (90% power)
power.t.test: 52.1 
Simulated power at n = 52 per arm: 0.803 (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 registry reports a sponsor analysis for the primary endpoint as a geometric mean ratio of 1.33, with a 90% confidence interval from 1.24 to 1.44. It labels that analysis non-inferiority. The recalculated result is a simple unadjusted difference between arithmetic means, reported as 20.9 μg/mL with a 95% confidence interval from 14.07 to 27.73 μg/mL and p < 0.001. These are different summaries on different scales, so their estimates and intervals should not be compared as if they were duplicate calculations of the same quantity.

The sponsor analysis is the trial’s reported inferential analysis; the recalculation is a transparent, unadjusted comparison derived from the posted summary statistics. The registry does not report the sponsor’s specific model, transformation, or non-inferiority margin, so the full analysis cannot be reconstructed from the record. The sponsor’s non-inferiority label should be reported as such, while recognizing that the decision threshold and its application are not available for independent assessment.

The record does not flag a discrepancy between the sponsor result and the recalculation. Their different numerical forms are expected from their different scales and methods, rather than evidence by themselves of a contradiction. The recalculation should be read as a complementary description of the posted group means, not as a replacement for the sponsor’s own analysis.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports one primary pharmacokinetic endpoint and numerous secondary outcomes. Multiple testing matters because evaluating many outcomes can increase the chance of obtaining at least one apparently statistically significant result even when there is no underlying effect for the outcomes being tested. General statistical practice may address this through a pre-specified testing hierarchy, adjustment of significance levels, or other procedures, but the registry record does not report whether any such approach was specified or used.

The primary recalculation is explicitly unadjusted. The registry also gives a sponsor geometric mean ratio for the primary endpoint and labels it as a non-inferiority analysis, but does not report the non-inferiority margin, its rationale, or the detailed decision procedure. The method used for the sponsor analysis, the treatment of multiplicity, and any protocol-defined sequence of hypotheses are not reported in the registry record. Without those details, the sponsor’s label and estimate can be described, but the decision framework cannot be independently reconstructed.

The primary analysis population is defined as a pharmacokinetic per-protocol group with at least one measurable concentration and no major protocol violations related to sampling. The record does not report how many participants were excluded for each reason, how missing concentrations were handled, whether any values were imputed, or whether an intention-to-treat analysis was also conducted. The registry lists 235 and 234 participants for the primary endpoint summaries, but does not provide a complete participant-flow account linking enrollment, randomization, and endpoint analysis.

General statistical context: analysis-population choices and missing-data handling can affect estimates when the analyzed participants differ from those randomized or when missingness is related to outcomes. The available record supports describing the defined per-protocol analysis and the posted denominators, but not judging the impact of exclusions or missing measurements. Details on interim analyses, stopping rules, protocol deviations beyond the endpoint definition, and sensitivity analyses are also not reported in the registry record.

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: febrile neutropenia (4.4%), neutropenia (3.0%), pneumonia (1.3%), endometrial cancer (0.7%), cardiac failure congestive (0.7%).

The registry lists serious adverse events among participants at risk in each arm. In the intravenous trastuzumab plus chemotherapy group, 45 of 298 participants at risk were recorded as affected by a serious adverse event; in the subcutaneous group, 65 of 297 were recorded as affected. These are counts of participants, not counts of all individual event episodes. The registry also lists other adverse events affecting 277 of 298 participants in the intravenous group and 283 of 297 in the subcutaneous group.

The listed serious adverse-event terms include febrile neutropenia at 4.4%, neutropenia at 3.0%, pneumonia at 1.3%, endometrial cancer at 0.7%, congestive cardiac failure at 0.7%, and tonsillitis at 0.7%. The record does not provide the event-specific denominators or further detail about severity, timing, relatedness, or recurrence in the information available for these results. Death counts are not reported in the registry record.

These figures are descriptive safety counts. The groups’ at-risk denominators and the reported counts should be read together, while recognizing that the record does not supply enough detail for a full comparative safety analysis. The counts alone do not establish causation or explain whether events were attributed to trastuzumab, chemotherapy, or other factors.

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

The primary result describes a difference in measured trastuzumab trough concentrations between administration routes before surgery. The posted arithmetic means and the unadjusted recalculation indicate a higher mean concentration in the subcutaneous arm for the analyzed pharmacokinetic population. The sponsor’s separate geometric mean ratio also summarizes the route comparison, on a relative scale, and is labeled as a non-inferiority analysis in the registry.

What the result does mean is limited to the endpoint and analysis population: it provides a pharmacokinetic comparison at the specified pre-dose time point during the described treatment regimen. Confidence intervals communicate uncertainty around estimates, and the recalculated interval is compatible with a range of mean differences. The sponsor’s interval similarly communicates uncertainty for its geometric mean ratio, but the registry does not report the non-inferiority margin required to assess the sponsor’s decision framework.

What the result does not mean is that either route produces better or worse clinical outcomes. A serum concentration is not a direct measure of cancer control, survival, recurrence, quality of life, or the full safety profile. The registry lists safety outcomes and secondary pharmacokinetic measures, but the primary endpoint is not a clinical efficacy endpoint. Nor can the results alone establish equivalence or non-inferiority without knowing the margin, analysis methods, and pre-specified decision criteria.

The randomized, parallel-group design supports a direct comparison between assigned routes, but open-label conduct and the per-protocol pharmacokinetic analysis are relevant context. The record does not provide enough detail to assess how all deviations, missing measurements, or potential sources of bias were handled. These limits call for careful interpretation of the posted concentration results without extending them to claims beyond the measured outcomes.

18. Why This Trial Matters Statistically

This trial illustrates how a clinical comparison can be framed around a pharmacokinetic endpoint rather than a clinical event. Trough concentration is a continuous measure, and its interpretation depends on the measurement time, units, analysis population, and summary scale. Those features make it important to distinguish the arithmetic mean difference from the sponsor’s geometric mean ratio rather than treating them as one result.

The record also illustrates why the analysis population matters. The primary result comes from a defined pharmacokinetic per-protocol group, not automatically from everyone enrolled. The posted endpoint denominators are smaller than the overall enrollment figure, but the record does not fully explain participant flow or missing measurements. This limits what can be concluded about the effect of exclusions or missing data.

Non-inferiority analyses require a decision threshold that is set and justified in advance. The registry identifies the sponsor’s primary analysis as non-inferiority, but does not report the margin or the detailed statistical plan. The reported ratio and confidence interval are useful statistics, yet the record does not contain all the information needed to assess the non-inferiority criterion independently.

Finally, the record includes several secondary pharmacokinetic outcomes. Their broad directional pattern may add descriptive context, but multiplicity procedures are not reported. This is a useful example of why a collection of consistent secondary findings should not be mistaken for multiple independent confirmations, particularly when many outcomes are evaluated and their inferential status is unclear.

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 does not report the sample-size justification for the original trial, the detailed randomization procedure, allocation concealment, or the full statistical analysis plan. The planning illustrations based on the observed effect do not fill those gaps. The record also does not report the sponsor’s non-inferiority margin or its justification, which limits independent assessment of the non-inferiority analysis.

The primary endpoint is a pharmacokinetic measure at one pre-dose time point before surgery. It does not directly measure clinical efficacy or patient experience. The registry describes an open-label trial, but does not give details about whether endpoint assessment or laboratory measurement was blinded. The available record does not permit evaluation of the potential consequences of open-label conduct.

The primary result uses a pharmacokinetic per-protocol population. Although the arm denominators are reported, a complete flow from enrollment through randomization and endpoint analysis is not available. Reasons for missing or excluded concentrations, imputation procedures, sensitivity analyses, and an intention-to-treat analysis are not reported in the registry record.

Many secondary outcomes are listed, but the handling of multiplicity and the pre-specified status of individual secondary analyses are not reported. Details about interim analyses and stopping rules are also not reported in the registry record. The record gives serious and other adverse-event counts, but does not provide enough detail to assess event attribution, timing, or a comprehensive comparative safety model. These limitations constrain interpretation of the posted statistical results and should not be replaced by assumptions about the protocol.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Compare Subcutaneous (SC) Versus Intravenous (IV) Administration of Herceptin (Trastuzumab) in Women With Human Epidermal Growth Factor Receptor (HER) 2-Positive Early Breast Cancer — NCT00950300.
• Registry record last updated 2018-01-23; status last verified 2017-12.
• Results first posted 2017-01-23.
• 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

NCT00950300 is a completed phase III randomized, open-label, parallel-group study sponsored by Hoffmann-La Roche. It enrolled 596 participants and compared subcutaneous with intravenous trastuzumab, each administered with chemotherapy, in women with HER2-positive early breast cancer. The described regimen included eight cycles before surgery and ten further trastuzumab cycles after surgery.

The primary endpoint was observed serum trough concentration before surgery, measured pre-dose on day one of cycle eight. The posted group means were 57.8 μg/mL for intravenous treatment and 78.7 μg/mL for subcutaneous treatment. The unadjusted recalculated mean difference was 20.9 μg/mL, with a 95% confidence interval from 14.07 to 27.73 μg/mL and p < 0.001. The sponsor reported a geometric mean ratio of 1.33 with a 90% confidence interval from 1.24 to 1.44 and labeled the analysis non-inferiority.

The registry does not report the non-inferiority margin, full statistical plan, original sample-size justification, or detailed missing-data procedures. The results describe pharmacokinetic concentrations and should not be interpreted as direct evidence about clinical efficacy or patient-important outcomes.