Evaluation of Safety and Effectiveness of Fostamatinib Compared to Placebo in Patients in Asia With Rheumatoid Arthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of Safety and Effectiveness of Fostamatinib Compared to Placebo in Patients in Asia With Rheumatoid Arthritis (NCT01569074), a randomized phase 2 study. The primary outcome is Proportion of Patients Achieving ACR20 at Week 12, Comparison Between Fostamatinib and Placebo.

2. Clinical Question

3. Trial Design

This phase 2 study was a randomized, parallel-group comparison of four oral fostamatinib dosing regimens with oral placebo. Its official title describes a multicentre, double-blind, placebo-controlled, dose-ranging study in Asia. The registry records quadruple masking. In a masked trial, treatment knowledge is withheld from designated parties; the specific parties covered by the registry’s masking designation are not reported in the registry record.

The study enrolled 163 participants. The stated purpose was to evaluate effectiveness and safety in people with active rheumatoid arthritis who were taking methotrexate but not responding adequately. The stated study duration was 12 weeks. The primary endpoint was assessed at 12 weeks, and the secondary outcomes included response measures at week 1 and week 12, as well as disease activity, function, and health-status measures.

The registry lists four fostamatinib regimens and a placebo regimen. The primary outcome results report 26 participants for the 100 mg twice-daily group, 29 for the regimen beginning at 100 mg twice daily and changing after 4 weeks, 27 for 75 mg twice daily, 28 for 50 mg twice daily, and 28 for placebo. The number contributing to each outcome need not represent the total enrolled population; the reason for the differences is not reported in the registry record.

The registry records the study as terminated. The reason for termination, the intended enrollment target, the randomization procedure, and details of any changes to the planned study are not reported in the registry record. The listed start was 2012-04, and primary completion was 2013-07.

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 135 patients failed screening.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The full analysis set includes those patients who received at least 1 dose of investigational product. Patients were analysed by randomised treatment in accordance with the intention to treat principle.

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 Proportion of Patients Achieving ACR20 at Week 12, Comparison Between Fostamatinib and Placebo, measured over 12 weeks.

The primary endpoint was the proportion of patients achieving ACR20 at 12 weeks. ACR20 means meeting the American College of Rheumatology 20% response criteria. The registry’s description says the criteria use swollen and tender joint counts out of 28 joints, blood measures of inflammation such as C-reactive protein, and physician and patient assessments of disease activity, pain, and physical function.

This is a binary response outcome: each participant is classified as meeting or not meeting the response criterion at the specified assessment. The arm-level result is therefore a percentage of responders rather than an average change on a continuous scale. The registry identifies the outcome unit as percentage of responders and its measure type as a count.

As a general statistical point, a composite response criterion can summarize several dimensions of disease status in a single endpoint. It can make the comparison interpretable as the proportion reaching a defined threshold, while not describing every component separately. ACR20 indicates that the specified response threshold was reached; it does not by itself show the size of change in each component, the durability of a response, or the experience of every participant.

The full analysis set included patients who received at least one dose of investigational product. The registry says patients were analyzed according to randomized treatment under the intention-to-treat principle. The detailed handling of missing endpoint assessments is not reported in the registry record.

7. Results Posted in the Registry

• FOSTA 100 MG BID PO: 53.8% (14/26) versus 32.1% (9/28) on control; risk difference 21.7 percentage points (95% CI -4.1 to 47.5), risk ratio 1.68 (0.88–3.19); unadjusted two-proportion z-test p = 0.107.
• FOSTA 100 MG BID (4 WKS) THEN 150 MG QD PO: 55.2% (16/29) versus 32.1% (9/28) on control; risk difference 23.0 percentage points (95% CI -2.0 to 48.1), risk ratio 1.72 (0.91–3.22); unadjusted two-proportion z-test p = 0.080.
• FOSTA 75 MG BID PO: 25.9% (7/27) versus 32.1% (9/28) on control; risk difference -6.2 percentage points (95% CI -30.1 to 17.7), risk ratio 0.81 (0.35–1.86); unadjusted two-proportion z-test p = 0.612.
• FOSTA 50 MG BID PO: 46.4% (13/28) versus 32.1% (9/28) on control; risk difference 14.3 percentage points (95% CI -11.0 to 39.6), risk ratio 1.44 (0.74–2.82); unadjusted two-proportion z-test p = 0.274.

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.

Proportion of Patients Achieving ACR20 at Week 1, Comparison Between Fostamatinib and Placebo

• 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:
  • FOSTA 75 MG BID PO: 21.2% (7/33) versus 15.2% (5/33) on control; risk difference 6.1 percentage points (95% CI -12.5 to 24.6), risk ratio 1.40 (0.49–3.97); unadjusted two-proportion z-test p = 0.523.
  • FOSTA 50 MG BID PO: 18.2% (6/33) versus 15.2% (5/33) on control; risk difference 3.0 percentage points (95% CI -14.9 to 21.0), risk ratio 1.20 (0.41–3.55); unadjusted two-proportion z-test p = 0.741.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between FOSTA 75 MG BID PO 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 FOSTA 50 MG BID PO and the control. The interval gives the range of effects that remain compatible with the data.

Proportion of Patients Achieving ACR50 at Week 12, Comparison Between Fostamatinib and Placebo

• 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:
  • FOSTA 100 MG BID PO: 30.8% (8/26) versus 14.3% (4/28) on control; risk difference 16.5 percentage points (95% CI -5.5 to 38.5), risk ratio 2.15 (0.74–6.31); unadjusted two-proportion z-test p = 0.145.
  • FOSTA 100 MG BID (4 WKS) THEN 150 MG QD PO: 34.5% (10/29) versus 14.3% (4/28) on control; risk difference 20.2 percentage points (95% CI -1.4 to 41.8), risk ratio 2.41 (0.86–6.81); unadjusted two-proportion z-test p = 0.077.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between FOSTA 100 MG BID PO 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 FOSTA 100 MG BID (4 WKS) THEN 150 MG QD PO and the control. The interval gives the range of effects that remain compatible with the data.

Proportion of Patients Achieving ACR70 at Week 12, Comparison Between Fostamatinib and Placebo

• 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:
  • FOSTA 100 MG BID PO: 11.5% (3/26) versus 0.0% (0/28) on control; risk difference 11.5 percentage points (95% CI -0.7 to 23.8); unadjusted two-proportion z-test p = 0.064.
  • FOSTA 100 MG BID (4 WKS) THEN 150 MG QD PO: 13.8% (4/29) versus 0.0% (0/28) on control; risk difference 13.8 percentage points (95% CI 1.2 to 26.3); unadjusted two-proportion z-test p = 0.042.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between FOSTA 100 MG BID PO and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for FOSTA 100 MG BID (4 WKS) THEN 150 MG QD PO than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry lists secondary outcomes spanning early ACR20 response, more stringent ACR response thresholds at week 12, disease activity, physical function, and health status. The posted secondary results include nine outcomes, while the registry records 10 outcome measures and 37 analyses overall. The secondary endpoints broaden the questions beyond the primary ACR20 comparison, but they do not replace it.

Across the posted results, response estimates do not move uniformly across every dose regimen or endpoint. Some week-12 response outcomes show higher observed proportions in particular regimens than in placebo, while other regimens have lower or similar proportions. The pattern should be interpreted in the context of outcome definitions, assessment times, group sizes, and the sponsor’s reported analysis for each comparison. The secondary results do not establish a single consistent pattern across all measures.

In general, secondary outcomes can help describe whether findings extend to other response thresholds or domains, but they are often less decisive than a prespecified primary endpoint. When many outcomes and comparisons are examined, the chance of observing an apparently notable result somewhere among them increases. This is the general multiplicity problem. The registry does not report whether the secondary comparisons were adjusted for multiplicity or how they were ordered for confirmatory interpretation.

Accordingly, secondary p-values should be treated as exploratory unless the prespecified analysis plan says otherwise; that plan is not reported in the registry record. The secondary findings can add context and identify patterns for further study, but unadjusted secondary results carry less weight than a clearly prespecified primary analysis. This interpretation does not require repeating each endpoint’s result: the central point is that breadth of outcomes adds information while also increasing the need for caution.

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

• Proportion of Patients Achieving ACR20 at Week 1, Comparison Between Fostamatinib and Placebo
• Proportion of Patients Achieving ACR50 at Week 12, Comparison Between Fostamatinib and Placebo
• Proportion of Patients Achieving ACR70 at Week 12, Comparison Between Fostamatinib and Placebo
• ACRn - Comparison Between Fostamatinib and Placebo at Week 12
• Proportion of Patients Achieving DAS28-CRP<=3.2 at Week 12, Comparison Between Fostamatinib and Placebo
• Proportion of Patients With DAS28-CRP EULAR Response at Week 12, Comparison Between Fostamatinib and Placebo
• Proportion of Patients With HAQ-DI Response at Week 12, Comparison Between Fostamatinib and Placebo
• SF-36 - Comparison of the Change in PCS From Baseline Between Fostamatinib and Placebo at Week 12

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: each participant either met the ACR20 response criterion at week 12 or did not. The registry reports the proportion of responders in each randomized group. Comparisons are therefore expressed as differences between proportions, rather than differences between mean scores.

What does a difference in proportions describe?

In general, a difference in proportions compares the observed frequency of an outcome between groups. A positive difference means the outcome proportion is higher in the first group named in the comparison than in the reference group, while a negative difference means it is lower. The value is an estimate from the analyzed participants and is subject to sampling variation. The arm-level percentages help show the underlying response pattern; the contrast summarizes the comparison.

For the registry’s primary comparisons, the sponsor reports Mantel-Haenszel weighted differences in proportions. In general, Mantel-Haenszel methods combine comparisons across strata using weighting. The registry record does not identify the stratification variables or provide further details of the implementation. Its reported estimates are therefore best described using the method and measure stated in the record, without inferring what factors were used in the weighting.

How should the reported p-values be understood?

In general, a p-value measures how incompatible the observed data are with a specified null hypothesis, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it is not a measure of the magnitude or importance of a difference. The registry lists p-values for its reported comparisons, but the statistical analysis plan and the decision rule used for interpreting those values are not reported in the registry record.

The registry’s four primary p-values are 0.059, 0.054, 0.570, and 0.262 for the four fostamatinib comparisons, in the order presented in the results. The confidence intervals accompanying those analyses are identified as 80% intervals. Because the record also contains recalculations with different values, the sponsor’s reported analysis should be kept distinct from the unadjusted recalculation.

Why are there unadjusted recalculations?

The recalculations use the posted responder summaries to make simple, unadjusted comparisons. They do not reproduce the sponsor’s weighted Mantel-Haenszel analysis. The corresponding unadjusted p-values are 0.107, 0.08, 0.612, and 0.274. Their purpose is to offer a direct comparison based on the posted summary counts, not to replace or correct the sponsor’s analysis. Since the registry record flags a discrepancy, the sponsor’s own analysis takes precedence for the reported trial inference.

In general, changing the method or adjustment can change an estimate, its uncertainty, or its p-value. Differences between analyses should be described rather than blended into a single result. The record does not provide enough detail to identify every reason for the difference between the sponsor’s analysis and the recalculation.

What does a confidence interval add?

In general, a confidence interval gives a range of values compatible with the data under a particular method and confidence level. Wider intervals indicate less precision; narrower intervals indicate more precision, conditional on the analysis assumptions. A confidence interval is not a guarantee that the true value lies in the interval, nor does it describe the range of individual participant outcomes. Here, the sponsor reports 80% intervals, while the recalculated comparisons have intervals of their own. The confidence level for those recalculated intervals is not reported in the registry record, so it should not be inferred.

Why can a binary outcome be uncertain?

In general, the proportion of responders is estimated from the number of participants with the event and the number assessed. When group sizes are limited, a small change in the event count can materially change the estimated proportion. This is one reason to consider the number of participants, the interval around a contrast, the analysis method, and the study context together rather than focusing on a single estimate.

The registry identifies the full analysis set as patients who received at least one dose and says they were analyzed by randomized treatment according to the intention-to-treat principle. It does not report a detailed missing-data method, the handling of intercurrent events, model diagnostics, or the complete statistical analysis plan. Those omissions limit what can be said about implementation beyond the information explicitly recorded.

11. Confidence Intervals and Precision

The primary endpoint estimates are proportions based on relatively small outcome groups, with reported group sizes ranging from 26 to 29 in the primary results. The observed ACR20 percentages span from 25.9% to 55.2% across the four fostamatinib regimens, compared with 32.1% for placebo. The spread of those point estimates alone does not describe their precision.

The sponsor’s reported analyses give 80% confidence intervals for weighted differences in proportions. For 100 mg twice daily, the estimate is 0.23 with an interval from 0.07 to 0.39; for the regimen that changed dose after 4 weeks, the estimate is 0.24 with an interval from 0.08 to 0.39. For 75 mg twice daily, the estimate is -0.06 with an interval from -0.20 to 0.08; for 50 mg twice daily, the estimate is 0.14 with an interval from -0.02 to 0.29.

In general, an interval that includes zero is compatible with no difference under the stated interval method, as well as with values on either side of zero. An interval that does not include zero indicates a narrower set of compatible values under that method, but it does not establish the clinical meaning or reproducibility of the estimate. The sponsor intervals are 80% intervals, not 95% intervals. The confidence level for recalculated intervals is not reported in the registry record.

The group sizes and intervals make clear that the estimates should not be read as exact effects. The uncertainty is especially relevant when comparing several dosing regimens and when considering estimates based on a limited number of responders. Precision concerns the uncertainty in a group-level estimate; it does not indicate the degree of variation between individual patients.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on observed effects, not the trial’s original design justification. All figures are participants per arm for the binary outcomes in this analysis; they are not total event counts. 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 specified as a fraction of the observed effect.

The 80% power column gives an illustrative number of participants per arm under which the calculation would have an 80% chance of detecting the specified effect if the assumptions and effect were correct. The 90% power column describes the corresponding target for a higher probability of detection. In general, increasing the target power requires a larger planned sample, and smaller effects require more participants because they are harder to distinguish from random variation.

Rows planning for 100% of the observed effect use the effect estimated in the trial as the target for the illustration. Rows planning for 75% or 50% of the observed effect use smaller target effects. Those estimates can produce very different sample-size requirements, particularly when the observed effect is small. The figures are not observed power: they are hypothetical planning calculations and do not describe the probability that the completed trial “had power” after its results were known.

The registry’s own original assumptions for sample-size planning, including its intended power, significance level, effect size, allocation assumptions, and any allowance for missing outcomes, are not reported in the registry record. These illustrations should therefore not be used to infer why the study enrolled 163 participants or whether the completed trial met its original design goals.

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 = 0.321; treatment proportion p₁ = 0.538; average p̄ = 0.430.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7001 = 1.3722.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6831 = 0.5749.
4. n = (1.3722 + 0.5749)² / (0.2170)² = 80.5, rounded up to 81 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 <- 0.321428571428571                          # control proportion
p1 <- 0.538461538461538                          # 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(1569074)
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 = 81 (80% power), 108 (90% power)
 75% of observed effect: n per arm = 142 (80% power), 189 (90% power)
 50% of observed effect: n per arm = 312 (80% power), 417 (90% power)
power.prop.test: 80.5 
Simulated power at n = 81 per arm: 0.797 (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. Analyses Reported by the Sponsor

The sponsor’s primary outcome analyses are reported as Mantel-Haenszel weighted differences in proportions, with 80% confidence intervals and p-values. The simple recalculations use the posted response summaries for unadjusted comparisons. These are distinct analyses and produce different p-values: the sponsor reports 0.059, 0.054, 0.570, and 0.262, while the recalculated values are 0.107, 0.08, 0.612, and 0.274.

The registry record flags a discrepancy. Accordingly, the sponsor’s own analysis takes precedence when describing the trial’s reported inference. The recalculations remain useful as a separate, transparent check based on the posted summaries, but they should not be substituted for the analysis reported by the sponsor.

In general, differences between an adjusted or weighted analysis and a simple unadjusted calculation can arise because they use different procedures and may target different estimates. The registry does not give enough detail to attribute the discrepancy to a specific implementation choice. The stratification details and complete statistical analysis plan are not reported in the registry record. It is therefore most accurate to present each result with its method, rather than treating one set of values as a numerical correction to the other.

15. Multiplicity, Missing Data and Analysis Populations

The registry records multiple primary and secondary analyses, including four primary comparisons against placebo and a range of secondary outcomes. It lists 10 outcome measures and 37 analyses overall. When several hypotheses are tested, the chance of at least one apparently unusual result can increase even if no underlying effect exists. This is the general multiplicity concern.

Whether multiplicity was controlled, and if so by what method, is not reported in the registry record. The record does not state a testing hierarchy, a family of primary comparisons, a multiplicity-adjusted significance threshold, or whether the secondary findings were intended as confirmatory or exploratory. The reported p-values should therefore be read as the values listed for the individual analyses, not assumed to have been adjusted across outcomes or doses.

The primary population is described as the full analysis set: patients who received at least one dose of investigational product, analyzed according to randomized treatment under the intention-to-treat principle. The registry does not provide a detailed account of how missing week-12 assessments were handled, whether every randomized participant had an evaluable primary outcome, or whether sensitivity analyses were conducted. The reasons for any missing data and their distribution by treatment group are not reported in the registry record.

In general, missing outcomes can affect a binary endpoint if the probability of response among participants without an assessment differs from that among those assessed. The direction and size of any such effect depend on the missing-data pattern and the assumptions used to address it. Without the handling method and supporting information, the impact cannot be quantified from the registry results alone.

The registry identifies the analysis population and the general intention-to-treat principle, but the detailed analysis plan, interim analyses, and protocol-specific decision rules are not reported in the registry record. No assumptions about them should be inferred from the outcome summaries.

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: laryngitis viral (3.2%), hypertension (3.2%), diarrhoea (3.0%), osteoporosis (3.0%), rheumatoid arthritis (3.0%).

The registry lists serious adverse events for the four fostamatinib groups and placebo. The numbers at risk were 31, 33, 33, 33, and 33, respectively. Serious adverse events were recorded for 1 participant in the 100 mg twice-daily group, 1 in the regimen that changed dose after 4 weeks, 3 in the 50 mg twice-daily group, 0 in the 75 mg twice-daily group, and 1 in the placebo group.

These are counts of participants with serious adverse events, not proof that an event was caused by the assigned treatment. The registry also reports counts with other adverse events: 21, 17, 17, 16, and 15 in the same order. Those counts describe the posted safety results but do not, by themselves, summarize severity, duration, or relationship to treatment.

The listed serious-event terms include viral laryngitis and hypertension at 3.2%, and diarrhoea, osteoporosis, rheumatoid arthritis, and carpal tunnel syndrome at 3.0% each. The registry record does not report deaths. Such event summaries are limited by the number of participants and the duration of follow-up. They should not be used to infer comparative safety beyond the events and counts actually recorded.

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

The results describe ACR20 response proportions at 12 weeks among participants in the analyzed groups. They show variation across the four fostamatinib regimens and placebo, with the observed percentages ranging from 25.9% to 55.2% for fostamatinib and 32.1% for placebo. The registry’s reported weighted analyses and the simple unadjusted recalculations are not identical; the sponsor’s reported analysis takes precedence because the record flags a discrepancy.

In general, these data estimate group-level outcomes in this particular study. They do not describe the response of every participant, and an observed group difference does not establish why the groups differed. The confidence intervals convey uncertainty under their respective methods, while p-values describe compatibility with a statistical null hypothesis under analysis assumptions. Neither alone establishes clinical importance.

The study concerns Asian patients with active rheumatoid arthritis who were not responding adequately to methotrexate, and the primary endpoint is measured at 12 weeks. The findings should be understood within that population, endpoint, and follow-up period. The registry’s termination status is recorded, but the reason for termination is not reported in the registry record.

The data do not establish long-term outcomes, effects on every component of rheumatoid arthritis, or what would happen under circumstances not studied. They do not show that every person in a group experienced the group average or response proportion. Secondary outcomes provide additional context but involve multiple comparisons, and multiplicity handling is not reported in the registry record.

This statistical interpretation is not medical advice and does not establish that a regimen should or should not be used. It explains what the posted results estimate, the uncertainty attached to them, and the limits of conclusions that can be drawn from the registry record.

18. Why This Trial Matters Statistically

This trial is statistically informative because it compares several dosing regimens against a placebo group using a response endpoint with a clear threshold. A dose-ranging design can reveal whether observed outcomes vary across regimens, while keeping the primary question focused on a defined response at a stated assessment time.

It also illustrates why the analysis method matters. The posted arm-level response proportions permit simple unadjusted comparisons, while the sponsor reports weighted Mantel-Haenszel estimates. Those approaches are not interchangeable, and the registry flags a discrepancy. Readers need to preserve the distinction between the sponsor’s analysis and the recalculation rather than selecting whichever result appears more compelling.

The study is also a useful example of precision and multiplicity concerns. Several dose groups and secondary endpoints create a broad set of comparisons. Small outcome groups can make estimates uncertain, while many tests increase the importance of a prespecified plan and appropriate multiplicity handling. Those details are not reported in the registry record.

Finally, planning illustrations based on observed effects highlight the relationship between target effect size and the sample needed to detect it. They are not evidence about the study’s original design rationale. Keeping planning calculations separate from the completed trial results prevents retrospective sample-size illustrations from being mistaken for observed power or protocol assumptions.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The analyzed group sizes for the primary endpoint are small, and the percentages are estimates with consequential uncertainty. The sponsor’s reported primary analyses and unadjusted recalculations differ; the registry flags this discrepancy but does not give enough detail to account for every difference. The sponsor analysis takes precedence, but the missing implementation details limit independent interpretation of the weighted method.

The registry lists the study as terminated, but the reason for termination is not reported in the registry record. The duration is stated as 12 weeks, so the posted results do not describe longer-term outcomes. The detailed enrollment target and the reasons the outcome group sizes differ from the overall enrollment are not reported in the registry record.

The registry does not report the original sample-size justification, the full statistical analysis plan, adjustment for multiple comparisons, interim analyses, or detailed missing-data methods. It also does not identify the stratification variables used for the Mantel-Haenszel analyses. Those omissions restrict assessment of how closely the posted analyses follow the intended protocol and how robust the findings are to alternative assumptions.

The primary outcome is a binary composite response, which does not show each component’s results separately. Secondary endpoints add breadth but also raise multiplicity concerns; adjustment status is not reported. Safety counts are based on the listed follow-up and participant groups, and deaths are not reported in the registry record. These limitations do not invalidate the posted findings, but they narrow what can responsibly be concluded from them.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Evaluation of Safety and Effectiveness of Fostamatinib Compared to Placebo in Patients in Asia With Rheumatoid Arthritis — NCT01569074.
• Registry record last updated 2014-04-07; status last verified 2014-02.
• Results first posted 2014-04-07.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT01569074 is a phase 2 randomized, quadruple-masked, parallel-group study of four oral fostamatinib regimens and placebo in Asian patients with active rheumatoid arthritis and an inadequate response to methotrexate. The study enrolled 163 participants and was intended to last 12 weeks. Its primary endpoint was the proportion achieving ACR20 at week 12.

The registry reports arm-level responder percentages and Mantel-Haenszel weighted comparisons with 80% confidence intervals. Simple unadjusted recalculations yield different values; the record flags a discrepancy, so the sponsor’s analysis takes precedence. The study status is terminated, and the reason is not reported in the registry record. Details of multiplicity control, missing-data handling, and the original sample-size rationale are also not reported.