Evaluation of Effectiveness of Two Dosing Regimens of Fostamatinib Compared to Placebo in Patients With Rheumatoid Arthritis (RA) Who Are Taking Methotrexate and Have Had Inadequate Response to Single TNF-alpha Antagonist: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of Effectiveness of Two Dosing Regimens of Fostamatinib Compared to Placebo in Patients With Rheumatoid Arthritis (RA) Who Are Taking Methotrexate and Have Had Inadequate Response to Single TNF-alpha Antagonist (NCT01197755), a randomized phase 3 study. The primary outcome is Proportion of Patients Achieving ACR20 at Week 24, Comparison Between Fostamatinib and Placebo.

2. Clinical Question

The clinical question is comparative: it concerns the observed proportion meeting a defined response threshold in each randomized group, rather than the change in a single symptom or laboratory measure. The two active regimens are distinct comparisons with placebo; their results should not be treated as a direct comparison of one fostamatinib regimen against the other.

The registry describes a population with prior inadequate response to a single TNF-alpha antagonist and ongoing methotrexate treatment. The findings therefore relate to the enrolled population and the study period, not automatically to other rheumatoid arthritis populations or treatment settings. The trial was intended to evaluate two dosing regimens against placebo over approximately six months, with the primary outcome assessed at 24 weeks.

3. Trial Design

This was a completed phase III randomized, parallel-group study sponsored by AstraZeneca. The registry records enrollment of 323 and lists three oral treatment groups: FOSTA 100 MG BID PO, FOSTA 100 MG BID for 4 weeks followed by 150 MG QD PO, and placebo PO. The official title describes the study as double-blind, while the registry’s masking classification is quadruple. The record does not explain which roles were included in that classification.

The study began in 2010-09, and primary completion was recorded in 2013-02. Its brief description says the study would last approximately six months; the primary outcome timeframe is 24 weeks. The primary outcome population is the full analysis set, defined as patients who received at least one dose of investigational product. Patients were analysed according to randomized treatment, consistent with the intention-to-treat principle as described in the record.

The three groups allow each fostamatinib regimen to be compared with placebo. The reported primary analyses use the placebo group as the comparator for each active regimen. These are not head-to-head tests between the two dosing regimens, and the registry does not report a direct statistical comparison between them.

The record does not report the original sample-size justification, the intended allocation ratio, or further detail about randomization procedures. It also does not report the trial’s original statistical analysis plan or whether interim analyses were planned or conducted. Those details should not be inferred from the posted enrollment or from planning illustrations based on observed results.

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 315 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 24, Comparison Between Fostamatinib and Placebo, measured over 24 weeks.

The primary endpoint is the proportion of patients achieving ACR20 at week 24. ACR20 is a threshold-based response measure: it classifies whether a patient meets the American College of Rheumatology 20% response criteria. The registry description says that the criteria use counts of swollen and tender joints out of 28 joints, blood test measures of inflammation such as C-reactive protein, and physician and patient assessments of disease activity, pain, and physical function.

Using a composite response measure can capture several dimensions of rheumatoid arthritis rather than relying on a single symptom or laboratory result. A threshold also gives the analysis a direct interpretation as a responder proportion: the posted group percentages describe the share of each group meeting the stated criteria at the assessment time. The endpoint is therefore a binary outcome for each participant, summarized as a percentage for each group.

The threshold does not convey the size of improvement among people who meet it, nor does it describe every aspect of an individual’s condition. Someone who does not reach ACR20 may still have changed on one or more component measures, while the response classification alone does not show the component-level pattern. The registry record does not report those component-level primary outcome results.

For the primary endpoint, the full analysis set includes patients who received at least one dose and assigns their analysis according to randomized treatment. The record does not report additional details about handling missing week-24 endpoint data. Thus, the posted proportions should be read with the stated population definition and with the registry’s limitations in reporting missing-data procedures in mind.

7. Results Posted in the Registry

• FOSTA 100 MG BID PO: 36.2% (38/105) versus 21.1% (23/109) on control; risk difference 15.1 percentage points (95% CI 3.1 to 27.1), risk ratio 1.72 (1.10–2.67); unadjusted two-proportion z-test p = 0.015.
• FOSTA 100 MG BID (4 WKS) THEN 150 MG QD PO: 27.8% (30/108) versus 21.1% (23/109) on control; risk difference 6.7 percentage points (95% CI -4.7 to 18.1), risk ratio 1.32 (0.82–2.11); unadjusted two-proportion z-test p = 0.252.

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:
  • Dosing Group A and B Combined PO: 25.4% (54/213) versus 3.7% (4/109) on control; risk difference 21.7 percentage points (95% CI 14.9 to 28.5), risk ratio 6.91 (2.57–18.57); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Dosing Group A and B Combined PO than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Proportion of Patients Achieving ACR50 at Week 24, 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: 18.1% (19/105) versus 8.3% (9/109) on control; risk difference 9.8 percentage points (95% CI 0.8 to 18.8), risk ratio 2.19 (1.04–4.62); unadjusted two-proportion z-test p = 0.033.
  • FOSTA 100 MG BID (4 WKS) THEN 150 MG QD PO: 13.0% (14/108) versus 8.3% (9/109) on control; risk difference 4.7 percentage points (95% CI -3.5 to 12.9), risk ratio 1.57 (0.71–3.47); unadjusted two-proportion z-test p = 0.260.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for FOSTA 100 MG BID PO than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • 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 24, 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: 14.3% (15/105) versus 2.8% (3/109) on control; risk difference 11.5 percentage points (95% CI 4.2 to 18.9), risk ratio 5.19 (1.55–17.41); unadjusted two-proportion z-test p = 0.002.
  • FOSTA 100 MG BID (4 WKS) THEN 150 MG QD PO: 2.8% (3/108) versus 2.8% (3/109) on control; risk difference 0.0 percentage points (95% CI -4.3 to 4.4), risk ratio 1.01 (0.21–4.89); unadjusted two-proportion z-test p = 0.991.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for FOSTA 100 MG BID PO than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • 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.

The secondary outcomes cover several aspects of response, including ACR response thresholds at different assessment times, a continuous ACRn measure, DAS28-CRP thresholds and response, and HAQ-DI response. Across the posted secondary outcomes, the overall pattern is not uniform: some comparisons show a separation between groups, while others are less clearly separated or have little observed difference. Results also vary by the response definition and assessment time.

That broad pattern can be considered alongside the primary ACR20 result, but it should not be treated as a series of independent confirmations. Secondary outcomes are related to the primary question in that they describe other response thresholds or dimensions of rheumatoid arthritis; they are not the same endpoint as ACR20 at week 24. Differences among outcomes can arise because they measure different constructs, use different thresholds, or are assessed at different times.

Unadjusted secondary results carry less weight as confirmatory evidence when many outcomes and comparisons are examined. With multiple statistical tests, the chance of at least one apparently notable result can increase even when there is no underlying effect for any particular comparison. The registry lists multiple outcome measures and analyses, but a multiplicity adjustment plan or hierarchy for interpreting the secondary results is not reported in the registry record.

Accordingly, the secondary findings are most useful as supporting or exploratory context. They should be interpreted with attention to their definitions, timeframes, intervals, and the overall number of analyses, rather than selecting isolated p-values. The registry record does not establish that these outcomes were adjusted for multiplicity or that they were intended to provide independent confirmatory evidence.

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 24, Comparison Between Fostamatinib and Placebo
• Proportion of Patients Achieving ACR70 at Week 24, Comparison Between Fostamatinib and Placebo
• ACRn - Comparison Between Fostamatinib and Placebo at Week 24
• Proportion of Patients Achieving DAS28-CRP <2.6 at Week 24, Comparison Between Fostamatinib and Placebo
• Proportion of Patients Achieving DAS28-CRP <=3.2 at Week 12, Comparison Between Fostamatinib and Placebo
• Proportion of Patients Achieving DAS28-CRP EULAR Response at Week 24, Comparison Between Fostamatinib and Placebo
• Proportion of Patients With a HAQ-DI Response at Week 24 - Comparison Between Fostamatinib and Placebo

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is a binary responder measure: each patient either meets ACR20 criteria at the stated assessment or does not. Group results are expressed as percentages of responders. A comparison of proportions is a natural way to describe how frequently the defined response occurred in the different randomized groups. It does not show the full distribution of component outcomes or the degree of change among responders.

What does a risk difference describe?

A risk difference is the difference between two event proportions, expressed on the proportion scale. In this trial, the event is achieving ACR20. A positive contrast in a fostamatinib-versus-placebo comparison indicates a higher observed responder proportion in the active-treatment group under the ordering used for the calculation. The contrast is an absolute measure, not a relative ratio. The sponsor’s reported primary analysis is described as a Mantel Haenszel weighted difference in proportion.

Weighting can combine information across strata rather than treating every stratum as though it contributed equally regardless of its size. The registry names the Mantel Haenszel method but does not report the strata or additional adjustment details for these analyses. It is therefore appropriate to identify the reported method without inferring which variables were used or how stratification was implemented.

What do the p-values say?

A p-value describes how compatible 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 does not measure the probability that the result will replicate. The sponsor-reported p-values for the primary comparisons are 0.004 and 0.168. They should be read alongside the estimated differences and confidence intervals, rather than as stand-alone measures of importance.

The separately calculated, unadjusted p-values are 0.015 and 0.252. These are simple recalculations from posted summary statistics, not a replacement for the sponsor’s analysis. Because the record identifies a discrepancy, the sponsor’s reported analysis takes precedence. The registry does not report enough detail to determine why the analyses differ.

Why do analysis populations matter?

The record defines the full analysis set as patients who received at least one dose of investigational product and states that patients were analysed by randomized treatment in accordance with the intention-to-treat principle. This describes the population and treatment assignment used for the primary analysis. It is not identical to a complete account of how every randomized patient’s missing endpoint data were handled. The record does not report the specific missing-data method, the number of missing week-24 observations, or sensitivity analyses for missing outcomes.

Randomized assignment is important because, in a well-conducted randomized trial, it supports comparisons that are less vulnerable to systematic baseline differences than comparisons based on treatment received or chosen. However, randomization alone does not remove uncertainty, address missing outcome data, or establish that all relevant analyses were conducted as planned. Details of the original analysis plan are not reported in the registry record.

Why are recalculations treated cautiously?

A recalculation from group-level counts and denominators can be useful as a check on the posted summary, but it may not reproduce a sponsor analysis that uses weighting or other analysis details. Here the registry reports sponsor analyses using the Mantel Haenszel method and separately flags a discrepancy with the simple recalculations. The sponsor’s results therefore provide the primary inferential interpretation. The recalculated results are best treated as a check on the posted summaries, not as a competing definitive analysis.

More generally, no statistical method can resolve questions that the endpoint and study design did not measure. The primary endpoint summarizes a threshold response at week 24. It does not, on its own, describe durability, every symptom, or component-level changes. The registry’s 12 outcome measures and 23 analyses indicate that the record includes outcomes beyond this primary comparison, but the multiplicity implications and any adjustment strategy are not reported.

11. Confidence Intervals and Precision

The sponsor-reported primary comparisons include 95% confidence intervals. For FOSTA 100 MG BID PO versus placebo, the weighted difference in proportion is 0.17, with an interval from 0.05 to 0.28. For FOSTA 100 MG BID for 4 weeks followed by 150 MG QD PO versus placebo, it is 0.07, with an interval from -0.03 to 0.18.

A confidence interval describes the uncertainty around an estimate under the analysis method. It is not a range containing a known fixed percentage of individual patient outcomes, nor does it assign a probability to the true effect after the data have been observed. In general, a narrower interval indicates greater precision, while a wider interval indicates more uncertainty about the range of plausible values under the method.

The first reported interval is entirely above zero. The second extends from a negative value to a positive value and therefore includes zero. That second interval is compatible with a range of contrasts, including no difference on the weighted-difference scale. Its width also indicates that the estimate does not pin down a precise contrast. The reported p-values, 0.004 and 0.168, are consistent with the distinction between the intervals’ relation to zero, but neither the interval nor the p-value conveys clinical importance by itself.

Precision depends on the amount and distribution of information available and on the analysis. The registry reports group sizes and event proportions, but the sponsor’s weighted analysis details are not reported in full. The discrepancy with the simple recalculations is another reason not to substitute the recalculated intervals for the sponsor-reported ones. The registry does not report a prespecified clinically important difference against which to judge whether the intervals are sufficiently narrow for a clinical decision.

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 illustrations are planning calculations based on the observed effect, not the original trial design calculation. For these binary outcomes, the displayed figures are participants per arm, not total participants across groups and not total events. Read each row as a planning scenario for detecting a stated fraction of the observed effect; the columns describe the participants per arm for 80% power and for 90% power.

The stated calculation 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 the stated fraction of the observed effect. The original trial’s sample-size assumptions and justification are not reported in the registry record. These illustrations are not observed power.

Power is the probability, under specified assumptions and for a specified effect, that a statistical test will reject the null hypothesis when that effect is present. An 80% power planning target and a 90% power planning target represent different design goals: the latter generally requires more participants per arm for the same assumed effect. The figures are conditional on the observed effect being a useful basis for planning, the stated allocation and statistical assumptions, and the comparison being considered.

Rows planning for a smaller fraction of the observed effect require more participants because smaller differences are harder to distinguish from random variation. This relationship is important when reading the curve: it is a planning illustration of how the required group size changes as the effect to be detected changes, not a description of the enrollment or observed results of this trial. It should not be read as evidence that a future study with the listed group size would necessarily reproduce the observed difference.

These calculations are pairwise illustrations and do not establish an appropriate design for a new study. They do not incorporate any adjustment for multiple outcomes or comparisons, and no such adjustment is specified in the assumptions. The registry’s actual enrollment was 323, but that fact does not reveal the original sample-size justification or whether the planned target was met.

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 = 0.211; treatment proportion p₁ = 0.362; average p̄ = 0.286.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6394 = 1.2531.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6304 = 0.5306.
4. n = (1.2531 + 0.5306)² / (0.1509)² = 139.7, rounded up to 140 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 <- 0.211009174311927                          # control proportion
p1 <- 0.361904761904762                          # 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(1197755)
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 = 140 (80% power), 187 (90% power)
 75% of observed effect: n per arm = 240 (80% power), 320 (90% power)
 50% of observed effect: n per arm = 515 (80% power), 688 (90% power)
power.prop.test: 139.7 
Simulated power at n = 140 per arm: 0.806 (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 registry reports the sponsor’s primary analyses as Mantel Haenszel weighted differences in proportion. For FOSTA 100 MG BID PO versus placebo, the reported difference is 0.17, with a 95% confidence interval from 0.05 to 0.28 and p = 0.004. For the regimen that changed from 100 MG BID to 150 MG QD, the reported difference is 0.07, with a 95% confidence interval from -0.03 to 0.18 and p = 0.168.

Simple unadjusted recalculations based on the posted summary statistics give different p-values: 0.015 and 0.252, respectively. The record flags this discrepancy. The sponsor’s own analysis takes precedence for interpreting the trial’s reported inferential results; the recalculation is not a substitute for it. The reported sponsor analyses and recalculations should not be blended or selectively chosen based on which result appears more favorable.

The registry does not explain the reason for the discrepancy, and the details needed to reproduce the sponsor analysis are not fully reported. In particular, the registry names the Mantel Haenszel method but does not provide enough information to reconstruct all analysis choices. The discrepancy should therefore be acknowledged without attributing it to a specific adjustment, data-handling decision, or computational issue.

The distinction matters because a difference in analysis can change estimates, uncertainty, or p-values. When communicating the primary result, the sponsor-reported estimate, interval, and p-value form the appropriate set of results to describe, while the separate recalculation can be noted as a consistency check that did not match.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists 12 outcome measures and 23 analyses. Assessing many outcomes or making several comparisons creates a multiplicity issue: even when each individual statistical test is conducted at a conventional threshold, the chance of at least one apparently notable result across a collection of tests can be greater than for a single test. This is a general statistical concern, not a claim that any particular result in this trial is false.

The registry record does not report a multiplicity adjustment strategy, a hierarchy assigning confirmatory status to particular outcomes, or a detailed prespecified testing sequence. It is therefore not possible to determine from the registry record whether the reported secondary p-values were adjusted or how the sponsor intended to control the overall error rate. The primary endpoint is identified, but that alone does not establish the multiplicity procedure for the other outcomes and comparisons.

The full analysis set is described as patients receiving at least one dose, analysed by randomized treatment. That provides useful information about the analysis population. However, the registry record does not report the number of patients with missing primary endpoint data, the specific method used to handle missing outcomes, or sensitivity analyses evaluating alternative assumptions about missingness. It also does not report how missing data were handled for each secondary endpoint.

These reporting gaps limit the ability to assess whether missingness could have affected the responder proportions or whether the analysis was robust to different assumptions. They do not establish that missing data were mishandled; they mean that the relevant procedures and checks are not reported in the registry record. The original statistical analysis plan, including any plans for interim analyses, is also not reported in the registry record. No conclusions about those procedures should be inferred from the posted results alone.

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: gastroenteritis (1.9%), arteriosclerosis coronary artery (1.0%), gastric ulcer (1.0%), gastritis (1.0%), non-cardiac chest pain (1.0%).

The registry reports serious adverse events by treatment group. The serious-event figures are 7 affected patients among 105.0 at risk in the FOSTA 100 MG BID PO group, 7 among 108.0 in the group that changed from 100 MG BID to 150 MG QD, and 6 among 109.0 in the placebo group. These are counts of patients affected and the corresponding group sizes at risk; they are not a measure of the severity or duration of each event.

The record also lists other adverse events affecting 51 patients in the first fostamatinib group, 49 in the second, and 46 in the placebo group. Serious and other adverse events are separate categories in the posted results, so they should not be combined. The listed serious-event terms include gastroenteritis at 1.9%, and arteriosclerosis coronary artery, gastric ulcer, gastritis, non-cardiac chest pain, and cholecystitis at 1.0% each.

The registry does not report death counts in the posted safety data; deaths are not reported in the registry record. These figures are descriptive summaries. They do not, by themselves, establish whether a treatment caused an event or provide a complete assessment of comparative safety. The record’s safety information should be read within the study’s population, follow-up, and reporting context.

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

The primary result describes a difference in the proportion of participants meeting ACR20 criteria at week 24 across the randomized groups. The sponsor-reported analysis estimates a positive weighted difference for FOSTA 100 MG BID PO versus placebo, with a 95% confidence interval from 0.05 to 0.28 and p = 0.004. For the regimen that changed from 100 MG BID to 150 MG QD, the estimate is 0.07, with a 95% confidence interval from -0.03 to 0.18 and p = 0.168. The first interval is above zero; the second includes zero and leaves more uncertainty about the contrast.

These results do not establish that every participant responded, that all aspects of rheumatoid arthritis changed in the same way, or that a response was sustained beyond week 24. ACR20 is a threshold-based composite measure. It does not show the separate changes in joint counts, inflammation measures, pain, disease activity assessments, or physical function, and the registry record does not report those component-level primary results.

The results also do not directly compare the two fostamatinib regimens with each other. Each reported primary comparison is against placebo. A difference between the two estimates is not itself a formal test of one active regimen against the other, and a direct comparison is not reported in the registry record.

Statistical evidence is not the same as clinical importance. The record does not report a prespecified clinically important difference for the primary contrast. Nor do the primary results alone establish the balance of benefits and harms, long-term outcomes, or applicability to patients outside the described trial population. The safety summaries and secondary outcomes add context but have their own reporting and interpretation limitations.

The findings should therefore be understood as evidence about the posted week-24 responder proportions and the sponsor’s stated comparison method in this trial. They do not, by themselves, answer every question about treatment choice, durability, individual response, or safety.

18. Why This Trial Matters Statistically

This trial illustrates several issues that matter when interpreting clinical results. First, a responder percentage is an accessible summary, but it reduces a composite set of patient and clinician measures to a threshold classification at a particular time. Knowing the response proportions helps describe the trial outcome, while the confidence intervals show that the estimated contrasts retain uncertainty.

Second, the registry presents both an analysis method and enough summary information for simple recalculations, but the recalculated p-values do not match the sponsor’s reported p-values. Recognizing and reporting that discrepancy is important: a straightforward recalculation can be a useful check, yet it cannot automatically replace an analysis that uses a different method. The registry does not report enough detail to identify the reason for the difference, so the sponsor’s analysis takes precedence without speculation about the cause.

Third, the trial has more than one outcome and comparison. Secondary results provide a broader view of measured responses, but their interpretation depends on their definitions, timing, and the handling of multiplicity. Without a reported multiplicity strategy, isolated secondary findings should not be treated as equivalent to a single prespecified primary result.

Finally, planning calculations based on an observed effect are not evidence that the trial had a particular achieved power. They illustrate how assumed effect size and a planning target relate to required participants per arm under stated assumptions. The original sample-size justification is not reported. Keeping design planning, observed results, and inferential uncertainty distinct is essential for a careful reading of this record.

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.

Several limitations are specific to the information reported for this trial. The primary endpoint is measured at week 24, and the brief study description gives an approximate duration of six months. The registry record does not report longer-term primary outcome results, so the posted endpoint cannot establish whether the observed response pattern persisted beyond the assessment period.

The enrolled population was defined by rheumatoid arthritis, methotrexate use, and inadequate response to a single TNF-alpha antagonist. Results should not automatically be generalized to patients with different treatment histories or other populations. The registry does not report detailed baseline characteristics in the information summarized for these results, which limits assessment of how closely the groups and enrolled population resemble other settings.

The full analysis set definition is reported, but specific missing-data procedures, counts of missing endpoint observations, and sensitivity analyses are not reported in the registry record. The sponsor’s reported Mantel Haenszel results also differ from simple unadjusted recalculations, and the record does not explain why. The sponsor-reported analyses take precedence, but the lack of detail limits independent reconstruction.

Multiplicity procedures for the numerous outcomes and analyses are not reported. This makes the confirmatory status of secondary findings difficult to assess. The registry also does not report a prespecified threshold for clinical importance, the original sample-size justification, or the detailed original statistical analysis plan. These are limitations in what can be concluded from the registry record, not evidence that the study necessarily failed to address these matters in its conduct.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Evaluation of Effectiveness of Two Dosing Regimens of Fostamatinib Compared to Placebo in Patients With Rheumatoid Arthritis (RA) Who Are Taking Methotrexate and Have Had Inadequate Response to Single TNF-alpha Antagonist — NCT01197755.
• 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

NCT01197755 is a completed phase III randomized parallel-group trial sponsored by AstraZeneca. It evaluated two oral fostamatinib regimens against oral placebo in patients with rheumatoid arthritis taking methotrexate after an inadequate response to a single TNF-alpha antagonist. The registry records enrollment of 323 and primary completion in 2013-02.

The primary outcome was the proportion achieving ACR20 at week 24. Sponsor-reported Mantel Haenszel analyses gave a weighted difference in proportion of 0.17 for FOSTA 100 MG BID versus placebo and 0.07 for the regimen that changed from 100 MG BID to 150 MG QD versus placebo. The registry flags a discrepancy between those analyses and simple unadjusted recalculations, so the sponsor’s analyses take precedence.

The record describes the full analysis set and reports secondary outcomes and safety summaries. Detailed multiplicity procedures, missing-data methods, the original sample-size justification, and the reason for the recalculation discrepancy are not reported in the registry record.