A Study Of Lapatinib Versus Placebo Followed By Chemoradiation In Patients With Locally Advanced Head And Neck Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Of Lapatinib Versus Placebo Followed By Chemoradiation In Patients With Locally Advanced Head And Neck Cancer (NCT00371566), a randomized phase 2 study. The primary outcome is Change From Baseline of the Apoptotic Index During Treatment Phase.

2. Clinical Question

3. Trial Design

The record describes a randomized, single-blinded, placebo-controlled, multi-centre, parallel-group Phase II study. It lists 107 participants as enrolled and identifies lapatinib oral tablets and placebo as the interventions. The title describes lapatinib or placebo followed by chemoradiation in patients with locally advanced head and neck cancer. The study status is completed, with a start date of 2006-03 and a primary completion date of 2007-12.

The primary outcome was the change from baseline in apoptotic index during the treatment phase, measured in tumour biopsy samples at baseline and Week 2. The primary population was intent-to-treat: all subjects randomised to study treatment, regardless of whether they actually received study medication. The arm-level primary outcome summaries list 27.0 participants for placebo and 57.0 for lapatinib. The registry record does not explain the difference between the overall enrollment and the numbers included in these outcome summaries, or give reasons for missing outcome data.

The registry lists several secondary outcomes, including tumour-cell proliferation measured by Ki-67, radiological response after treatment and follow-up, and circulating tumour cell measures. The record has 33 outcome measures and 32 posted secondary outcomes, while one analysis is listed for the primary outcome. The randomized design supports a comparison between the assigned groups, but the registry does not report enough detail to assess all implementation features, such as the randomization procedure, allocation concealment, the precise treatment schedule, or the chemoradiation regimen. Those details are not reported in the registry record.

The study description says it was designed to explore effects of lapatinib monotherapy on apoptosis and necrosis in pretreatment and post-treatment tumour tissue. This focus is consistent with a tissue-based biological endpoint, rather than a primary endpoint defined as a clinical event or a long-term patient outcome. The record does not report the trial's original sample-size justification or its planned statistical analysis in full.

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.

Treatment Phase

Reasons for not completing this period:

Chemoradiation Phase

Reasons for not completing this period:

Follow Up Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Intent-to-treat (ITT) population comprised of all subjects who were randomised to study treatment, regardless of whether they actually received study medication.

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline of the Apoptotic Index During Treatment Phase, measured over Baseline and Week 2.

The primary endpoint is change from baseline in the apoptotic index during the treatment phase, reported as a percentage of positive cells. The index was assessed using the TUNEL assay. As described in the registry, the assay counts at least 1000 neoplastic nuclei, identifies cells with morphological changes defining cell death, and uses 10 randomly chosen fields examined at 400x magnification. A responder was defined as having 20% cell death.

Apoptosis is a form of cell death. Measuring the proportion of tumour cells identified by this assay gives a tissue-based measure of biological activity during the short treatment phase. Comparing change from baseline can describe how the measured index differs over that interval between randomized groups. The primary summaries and comparison, however, concern mean change; the responder definition does not mean that the reported primary analysis is a responder-rate analysis. No responder counts are reported for this endpoint.

As a general statistical point, a continuous endpoint retains information about the range of observed measurements rather than reducing every participant to a responder or non-responder category. A change score also expresses the follow-up measurement relative to baseline. Its interpretation depends on how the index is measured and on the study's analysis choices. The registry does not provide individual measurements, detailed assay quality-control information, or a complete account of how missing baseline or Week 2 samples were handled.

7. Results Posted in the Registry

• Lapatinib: mean 4.20 versus 6.20 on control; difference -2.00 (95% CI -6.78 to 2.78), standardized effect d = -0.24; z-test p = 0.413.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline of Cell Proliferation Rate of the Ki-67 Proliferative Index in Tumour Biopsy Samples During Treatment Phase

• 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:
  • Lapatinib: mean -5.60 versus -1.20 on control; difference -4.40 (95% CI -8.81 to 0.01), standardized effect d = -0.39; z-test p = 0.051.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Lapatinib and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several kinds of measurement: change in tumour-cell proliferation measured by Ki-67, radiological response after treatment and follow-up, and circulating tumour cell measures at baseline and after treatment phases. This range indicates that the study assessed biological activity in tissue alongside imaging and circulating-cell outcomes. The registry lists 32 posted secondary outcomes, but the detail available for interpretation varies across outcomes.

The posted Ki-67 outcome is also a continuous change measure. Its arm summaries and comparison are reported separately in the results display. The remaining outcome titles include radiological response and circulating tumour cell measures, but numerical results for those outcomes are not reported in the registry record posted for this analysis. Their titles alone cannot establish the direction or size of any group difference.

The secondary outcomes can add context to the primary tissue-based endpoint, but they are not interchangeable with it. Ki-67 assesses proliferation, whereas the primary endpoint concerns apoptosis measured using TUNEL. Radiological response and circulating tumour cell measures are different outcome types again. Agreement or disagreement across these measures would not, by itself, prove that one result is correct; each has its own measurement properties, timing, and uncertainty.

As a general statistical issue, testing many outcomes increases the chance that some results will appear noteworthy by chance if each is interpreted in isolation. Unadjusted secondary findings therefore usually carry less confirmatory weight than a prespecified primary analysis, especially when a complete multiplicity plan is unavailable. The registry does not report a multiplicity adjustment or a hierarchy for interpreting secondary outcomes. The secondary results are best treated as exploratory context, not as independent confirmation of the primary finding or as a substitute for unreported outcomes.

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

• Change From Baseline of Cell Proliferation Rate of the Ki-67 Proliferative Index in Tumour Biopsy Samples During Treatment Phase
• Overall Radiological Response After Treatment Phase in mITT Population
• Overall Radiological Response After Follow-up Phase in mITT Population
• Overall Radiological Response After Treatment Phase in ITT Population
• Overall Radiological Response After Follow-up Phase in ITT Population
• Number of Circulating Tumor Cells at Baseline in mITT Population
• Number of Participants With Circulating Tumor Cells After Treatment Phase in mITT Population
• Number of Participants With Circulating Tumor Cells After Chemoradiotherapy Phase in mITT Population

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: the registry reports a mean and standard deviation in each randomized group. The posted placebo summary is n = 27.0, mean change 6.2, and standard deviation 12.1. The lapatinib summary is n = 57.0, mean change 4.2, and standard deviation 5.53. A mean describes the centre of a set of measurements, while the standard deviation describes how spread out measurements are around that mean. These arm summaries do not show the distribution of individual changes, so they cannot reveal whether the values are symmetric, clustered, or affected by outliers.

The recalculation compares the two posted group means directly. Its mean difference is -2.0 for lapatinib relative to placebo, and the reported standard error for each group is 2.329 for placebo and 0.733 for lapatinib. The pooled standard deviation used in the recalculation is 8.204, and the standardized difference is -0.244. These are unadjusted summary-statistic calculations; they do not reproduce an adjusted model or use participant-level data. Because the discrepancy flag is false, there is no flagged inconsistency in the registry's posted information requiring an alternative interpretation of the recalculation.

The registry also reports ANCOVA as its analysis method, with a net mean difference of -1.7, a 95% confidence interval from -5.50 to 2.19, and p = .394. ANCOVA is a regression-based approach that compares groups while accounting for selected covariates. In a baseline and follow-up setting, it can adjust the follow-up comparison for baseline measurement, although the registry record does not state which covariates were included in this particular model. The reported estimate should therefore be described as an ANCOVA result, without assuming the exact adjustment set or model specification.

As a general statistical concept, adjustment can improve precision when covariates explain variation in the outcome, but its effect depends on the covariates and the model. An adjusted analysis and an unadjusted comparison need not return identical estimates or uncertainty intervals. Here, the estimates are both negative and their confidence intervals both include no difference. Their numerical similarity does not prove that the analyses are interchangeable. The registry's reported ANCOVA analysis remains the sponsor's own analysis; the recalculation is a transparent, unadjusted comparison of posted summaries.

What does the p-value describe?

As a general explanation, a p-value measures how compatible the observed data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it does not quantify the probability that the result will replicate. The unadjusted recalculated p-value is 0.413, while the registry reports p = .394 for ANCOVA. These values are not evidence that the null is true or that the groups are identical. They indicate that the analyses do not provide strong statistical evidence against the null model under their respective assumptions.

A p-value also does not describe the range of effect sizes compatible with the data. The confidence interval is more useful for that purpose because it gives an interval of estimates consistent with the analysis at its stated confidence level. Neither the p-value nor the interval establishes clinical importance; the registry does not report an importance threshold for the primary mean change.

Why can variability matter?

As a general explanation, greater variability makes a group mean less precise when other features of the design are held fixed. Unequal standard deviations and unequal group sizes can also make the comparison's precision depend on the specific variance assumptions and method. The primary summaries show different standard deviations and different analysis counts between arms. The recalculation uses the posted summary statistics and its reported calculation method; it cannot check assumptions against participant-level distributions.

Over-dispersion is a general concern when a statistical model assumes less variability than the data contain. Underestimating variability can make standard errors too small and intervals too narrow. The primary endpoint is analyzed as a continuous measure, and the registry does not report a diagnostic assessment of model fit, distribution shape, or over-dispersion. It would therefore be inappropriate to claim that over-dispersion occurred or that a particular corrective method was used.

What does the endpoint definition add?

The TUNEL description specifies a tissue assay and a responder threshold of 20% cell death. These details explain how the biological measure was defined, but the posted primary analysis is based on mean change. No responder analysis is posted for this primary outcome, and the registry does not report how many participants met the responder definition. The threshold should not be mistaken for the mean difference or treated as a test of the group comparison.

General statistical interpretation also depends on prespecification, data completeness, and the analysis population. The registry defines the intent-to-treat population, but it does not give details about missing samples, exclusions from the outcome analysis, or sensitivity analyses. The full original statistical plan and the original sample-size assumptions are not reported in the registry record. Without those details, it is not possible to verify all analysis decisions or determine whether the posted outcome analysis followed a prespecified plan.

11. Confidence Intervals and Precision

The reported ANCOVA estimate is a net mean difference of -1.7, with a 95% confidence interval from -5.50 to 2.19. The unadjusted recalculation gives a mean difference of -2.0, with a 95% confidence interval from -6.785 to 2.785. Both intervals span zero, so both are compatible with no average difference as well as with differences in either direction under their respective methods. The endpoints of the intervals also show that the estimates remain uncertain in magnitude.

As a general explanation, a confidence interval describes the range of effect estimates that are reasonably compatible with the observed data and the analysis assumptions at the stated confidence level. It is not a guarantee that the true value lies in the interval, nor does it assign a probability to each possible value after observing the data. A wider interval indicates less precision than a narrower interval on the same scale. The two intervals should not be treated as directly interchangeable because one comes from the sponsor's ANCOVA and the other from an unadjusted recalculation.

The arm summaries include different numbers of observations and different standard deviations. These features contribute to the uncertainty of the mean comparison. The registry does not report individual participant data, reasons why the outcome-summary counts differ from overall enrollment, or a detailed account of missing outcome handling. It is therefore not possible to assess whether missing measurements could have affected the estimates or the intervals.

Precision is also distinct from importance. A confidence interval can be statistically compatible with zero while including effects that might be important in either direction; deciding importance requires a clinically meaningful threshold, which is not reported for this endpoint. The short-term tissue measure and the registry's responder definition do not, by themselves, supply such a threshold for the mean difference.

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 estimates are illustrations of how many participants per arm would be needed under stated assumptions, using the observed effect as a starting point. They are not the trial's original design calculations, and they are not observed power. The original sample-size justification and design assumptions are not reported in the registry record.

For continuous outcomes such as this primary endpoint, the figures are participants per arm, not a total across both arms and not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is a stated fraction of the observed effect. The planning rows correspond to planning for 100%, 75%, and 50% of the observed effect; the columns describe participants per arm for 80% power and participants per arm for 90% power.

In plain language, a power target is the probability, under the assumed effect and statistical model, of detecting a difference at the selected significance level. The 90% power column plans for a higher chance of detection than the 80% column, which generally requires more participants under otherwise similar assumptions. Planning for a smaller effect requires more participants because a smaller group difference is harder to distinguish from random variation. The figures should be read as conditional planning scenarios, not as a claim about what the completed study was able to detect.

The estimates are based on the observed effect, which is itself uncertain. Using an observed estimate as the planning input can make the resulting numbers sensitive to random variation in this trial. The planning display is therefore useful for illustrating how sample size changes as the target effect and power target change, but it does not recover the trial's actual planned sample size or justify conclusions about its original 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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -2.000; pooled standard deviation σ = 8.204.
2. n = 2 × (1.96 + 0.842)² × 8.204² / -2.000² = 264.1, rounded up to 265 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 <- -2                      # observed difference in means
sigma <- 8.20409653770603                     # 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(371566)
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 = 265 (80% power), 354 (90% power)
 75% of observed effect: n per arm = 470 (80% power), 629 (90% power)
 50% of observed effect: n per arm = 1,057 (80% power), 1,415 (90% power)
power.t.test: 265.1 
Simulated power at n = 265 per arm: 0.805 (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 the sponsor's primary comparison as an ANCOVA estimate: a net mean difference of -1.7, a 95% confidence interval from -5.50 to 2.19, and p = .394. That is the trial's reported analysis. The recalculation uses the posted group means and standard deviations to form an unadjusted comparison, giving a mean difference of -2.0, a 95% confidence interval from -6.785 to 2.785, and an unadjusted p-value of 0.413.

The recalculation is not a replacement for the ANCOVA. It does not account for the adjustment used in the sponsor's model, and the registry record does not state which covariates were included. The two estimates are close in direction and broadly similar in interpretation, but their small numerical differences are expected when methods and adjustment differ. Both confidence intervals include zero and extend to values on either side of zero.

The registry record does not flag a discrepancy between its reported result and the recalculated summary-statistic result. The sponsor's analysis should still be distinguished from the recalculation: the former is the reported model-based analysis, while the latter is a simple unadjusted check based on aggregate values. Neither should be interpreted as establishing clinical importance, since no threshold for a clinically important difference in the mean apoptotic-index change is reported.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies the intent-to-treat population for the primary outcome as all subjects randomised to study treatment, regardless of whether they actually received study medication. That definition is important because it preserves the randomized group assignment in the target analysis population. However, the outcome summaries list 27.0 participants for placebo and 57.0 for lapatinib, while total enrollment is listed as 107. The registry record does not explain the difference or state whether the outcome counts reflect unavailable samples, exclusions, or another analysis-specific reason.

The record does not report how missing baseline or Week 2 biopsy measurements were handled, whether any participants were excluded from the endpoint analysis, or whether sensitivity analyses were conducted. It also does not provide a detailed accounting of the number randomized to each group for every analysis population. These omissions limit assessment of the potential effect of incomplete outcome data. It is not possible to infer that missingness was harmless or that it biased the estimates in a particular direction.

The registry lists multiple secondary outcomes and 33 outcome measures. It does not report a multiplicity adjustment, a testing hierarchy, or a complete plan for controlling the chance of false-positive findings across outcomes. As a general statistical explanation, when many hypotheses are tested, the chance of at least one apparently unusual result can increase if each test is interpreted without regard to the others. This is why isolated secondary p-values generally need cautious interpretation.

The original protocol-level statistical analysis plan, including the prespecified covariates for ANCOVA, interim analyses, missing-data procedures, and multiplicity strategy, is not reported in the registry record. The absence of these details does not establish that the study lacked such procedures; it means they cannot be verified from the record. The sample-size illustration uses stated planning assumptions, but the trial's original sample-size justification is not reported.

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: pyrexia (5.6%), mucosal inflammation (5.6%), respiratory tract infection (5.6%), vomiting (5.6%), constipation (2.9%).

The serious adverse event summaries list 36 participants at risk in the placebo group, with 14 affected, and 69 at risk in the lapatinib group, also with 14 affected. The record also lists all 36 placebo participants and all 69 lapatinib participants as at risk for other adverse events, with 36 and 69 affected, respectively. These counts should be read with their stated denominators; the groups used for safety summaries are not the same counts as those shown for the primary outcome.

The listed serious adverse event terms and frequencies are pyrexia, mucosal inflammation, respiratory tract infection, and vomiting, each at 5.6%; constipation at 2.9%; and tumour haemorrhage at 2.8%. The registry summary does not assign these listed term frequencies to a particular treatment group in the information, so they should not be used to compare arms. Deaths are not reported in the registry record.

As a general point, adverse-event counts describe events recorded during the study and do not, by themselves, establish that treatment caused an event. These summaries also do not convey timing, severity beyond the serious-event classification, duration, or the full clinical context. They are descriptive registry figures, not a basis for medical advice or a complete safety assessment.

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

The primary result describes a short-term comparison of changes in a tumour-tissue apoptotic index between randomized lapatinib and placebo groups. The sponsor-reported ANCOVA estimate is -1.7, with a 95% confidence interval from -5.50 to 2.19 and p = .394. The unadjusted recalculation is -2.0, with a 95% confidence interval from -6.785 to 2.785 and an unadjusted p-value of 0.413. Both estimates are negative, but both intervals include zero and allow values in either direction.

This means the posted analysis does not precisely determine the direction or magnitude of the average difference in this measure. It does not establish that the groups are identical, nor does it establish that one group has a clinically meaningful advantage. The registry does not report a threshold for a clinically important difference in mean apoptotic-index change. The responder definition of 20% cell death is part of the endpoint description, but responder counts and a responder comparison are not reported.

The result concerns an assay-based tumour measurement at baseline and Week 2. It is not, by itself, a result about long-term tumour control, survival, symptoms, or quality of life. Although the record lists radiological response and circulating tumour cell outcomes, their numerical results are not reported in the registry record. No conclusions about those outcomes should be inferred from their titles.

As a general statistical principle, an estimate and its uncertainty should be considered together. A p-value alone cannot tell whether a possible effect is important, and a confidence interval does not resolve questions that the measured endpoint does not address. The registry's incomplete reporting of missing-data procedures, adjustment details, and the full analysis plan further limits the strength of conclusions that can be drawn from the posted results.

18. Why This Trial Matters Statistically

This study is statistically interesting because it uses a biological endpoint measured in tumour tissue, with a defined assay and a short measurement interval. The primary outcome is a continuous mean change, while the endpoint description also includes a responder definition. Keeping those concepts separate matters: the posted group comparison concerns average change, and the registry does not report how many participants met the responder threshold.

The registry presents both a sponsor-reported ANCOVA result and enough arm-level summary statistics for an unadjusted recalculation. Comparing them illustrates why analysis method matters. ANCOVA can account for covariates, whereas the recalculation uses only aggregate group summaries. The estimates are similar in direction, but their confidence intervals and p-values are specific to their respective methods. The record does not identify the ANCOVA covariates, so the model cannot be fully reconstructed from the registry information.

The confidence intervals also make the uncertainty visible. The posted estimates do not precisely determine the magnitude or direction of the mean difference. That is a more informative reading than treating a p-value as a binary verdict. The planning illustrations further show, in general, how assumed effect size and desired power influence required sample size, while remaining distinct from the study's original design assumptions.

Finally, the secondary outcomes span tissue, imaging, and circulating-cell measurements. Their differing definitions show why a multi-outcome study needs clear prespecification and careful interpretation. The registry does not report a multiplicity strategy or the complete original analysis plan. For clinical researchers, this record highlights the value of reporting analysis populations, missing-data handling, covariate choices, and outcome-level results alongside estimates and confidence intervals.

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 give detailed eligibility criteria, baseline participant characteristics, the allocation concealment procedure, or the specific method used to implement single blinding. It identifies the study as multi-centre but does not report the number of centres. The title describes treatment followed by chemoradiation, but the full regimen, dose schedule, and timing details are not reported in the registry record. These omissions limit appraisal of the treatment context and study conduct.

The outcome summaries include 27.0 placebo and 57.0 lapatinib participants, while the record lists enrollment of 107. Reasons for the difference between enrollment and endpoint-analysis counts are not reported. The registry does not describe the number or pattern of missing biopsy samples, how such missing data were handled, or whether sensitivity analyses were performed. Since the primary endpoint relies on tumour tissue, the availability and completeness of samples are relevant to interpretation.

The reported ANCOVA estimate is not accompanied by the covariates or full model specification. The original sample-size justification and complete statistical analysis plan are not reported. Nor does the record report the prespecified approach to multiplicity across the numerous listed outcomes. These gaps prevent a full assessment of whether the posted analyses match the original plan and how the secondary findings should be weighted.

The primary outcome is a short-term assay-based measure, and the registry does not provide a clinically important difference threshold for its mean change. Several secondary outcome titles are listed without numerical results in the record. The serious adverse event summary includes counts and selected terms, but the listed term frequencies are not linked to arms in the available details, and deaths are not reported. These are reporting limitations, not evidence that unreported procedures or outcomes did not exist.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study Of Lapatinib Versus Placebo Followed By Chemoradiation In Patients With Locally Advanced Head And Neck Cancer — NCT00371566.
• Registry record last updated 2010-04-06; status last verified 2010-04.
• Results first posted 2010-01-29.
• 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

NCT00371566 is a completed randomized, single-blinded, placebo-controlled, parallel-group Phase II study sponsored by GlaxoSmithKline. It enrolled 107 participants and studied lapatinib oral tablets versus placebo in locally advanced squamous cell carcinoma of the head and neck, with chemoradiation described in the title. The study started in 2006-03, and primary completion was 2007-12.

The primary endpoint was change from baseline in tumour apoptotic index during the treatment phase, measured at baseline and Week 2 as percentage of positive cells. The sponsor-reported ANCOVA estimate was -1.7, with a 95% confidence interval from -5.50 to 2.19 and p = .394. An unadjusted recalculation gave -2.0, with a 95% confidence interval from -6.785 to 2.785 and p = 0.413.

The intervals include no difference and allow estimates in either direction. The record lists multiple secondary outcomes, but complete analysis details, the original sample-size justification, and missing-data and multiplicity procedures are not reported in the registry record.