Sprifermin (AS902330) in Cartilage Injury Repair (CIR): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Sprifermin (AS902330) in Cartilage Injury Repair (CIR) (NCT01066871), a randomized phase 2 study. The primary outcome is Percent Change From Baseline in Cartilage Defect Volume at Month 12.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group phase II trial of sprifermin for acute cartilage injury of the knee. The official title describes a multicenter, double-blind, placebo-controlled study. The registry record also identifies quadruple masking, but it does not specify which groups of people were included in that designation. Participants were allocated to one of three sprifermin dose groups or to placebo. The official title describes weekly intra-articular injections of 10, 30, or 100 µg for three consecutive weeks.

The registry record lists 74 participants enrolled. The primary outcome analysis used the modified intent-to-treat population, defined as participants from the intent-to-treat population who had at least one post-treatment MRI assessment. The number analyzed for the primary endpoint was 16 in the 10 µg group, 14 in the 30 µg group, 15 in the 100 µg group, and 16 in the placebo group. The record does not provide a participant flow explaining how the analyzed groups relate to the total enrollment.

The primary endpoint was percent change from baseline in cartilage defect volume at Month 12. Secondary outcomes included further MRI measures of defect volume and thickness, cartilage repair ratings, other structural knee scores, and participant-reported knee outcomes. The registry record lists 13 outcome measures and 12 secondary outcomes. It gives the study status as terminated; the reason for termination and the original recruitment target are not reported in the registry record. The trial start and primary completion are recorded as 2010-03 and 2013-04.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The modified intent-to-treat (mITT) analysis set included all participants from the ITT analysis set who had at least 1 post-treatment magnetic resonance imaging assessment. "N" (number of participants analyzed) signifies the participants who were evaluable for this outcome measure.

5. Baseline Characteristics

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

Population: The intention-to-treat (ITT) population included all participants randomized to a trial treatment.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline in Cartilage Defect Volume at Month 12, measured over Baseline, Month 12.

The primary endpoint measured the percent change in cartilage defect volume from baseline to Month 12, using central MRI. A percent change expresses the change relative to the baseline volume, rather than reporting only an absolute volume difference. The registry definition is the Month 12 volume minus the baseline volume, multiplied by 100 and divided by the baseline volume. A negative value therefore represents a lower measured defect volume at Month 12 than at baseline.

Using an imaging-based structural measure gives the trial a way to assess the cartilage defect itself. It is distinct from symptoms or function, which are assessed through other outcomes listed in the record, including knee scores. Structural and participant-reported outcomes address different aspects of the condition and need not move in parallel.

The endpoint is reported as a group mean with standard deviation. A group mean conveys the average observed percent change in that analysis group; it does not show the range of individual responses. The standard deviation describes how dispersed participant measurements were around the group mean. The primary analysis population required at least one post-treatment MRI assessment, so the endpoint result pertains to participants meeting that imaging criterion rather than automatically to every enrolled participant.

7. Results Posted in the Registry

• Sprifermin (AS902330) 10 mcg: mean -14.40 versus -17.00 on control; difference 2.60 (95% CI -16.05 to 21.25), standardized effect d = 0.10; z-test p = 0.785.
• Sprifermin (AS902330) 30 mcg: mean -17.80 versus -17.00 on control; difference -0.80 (95% CI -20.93 to 19.33), standardized effect d = -0.03; z-test p = 0.938.
• Sprifermin (AS902330) 100 mcg: mean -12.00 versus -17.00 on control; difference 5.00 (95% CI -16.92 to 26.92), standardized effect d = 0.16; z-test p = 0.655.

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.

Number of Participants With Acute Inflammatory Reactions

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Sprifermin (AS902330) 10 mcg: 28.6% (4/14) versus 16.7% (2/12) on control; risk difference 11.9 percentage points (95% CI -19.8 to 43.6), risk ratio 1.71 (0.38–7.78); unadjusted two-proportion z-test p = 0.473.
  • Sprifermin (AS902330) 30 mcg: 0.0% (0/14) versus 16.7% (2/12) on control; risk difference -16.7 percentage points (95% CI -37.8 to 4.4); unadjusted two-proportion z-test p = 0.112.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Sprifermin (AS902330) 10 mcg 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 Sprifermin (AS902330) 30 mcg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover a broad set of structural, imaging, and participant-reported measures. They include cartilage defect volume and thickness at several follow-up times, cartilage repair ratings, knee imaging features, cartilage repair grades, and knee symptom and function scores. Taken together, these outcomes could describe different dimensions of the injury and follow-up, but they are not interchangeable with the primary Month 12 volume measure.

The registry record lists 12 secondary outcomes, but detailed results are not reported in the registry record for most of those measures. A result is posted for the number of participants with acute inflammatory reactions, including comparisons for the 10 µg and 30 µg groups with placebo. The record does not give a corresponding comparison for the 100 µg group in that outcome. The secondary results therefore do not provide a complete set of estimates across all doses and outcomes.

Secondary outcomes can help contextualize a primary result, but they do not replace it. In this trial, the primary comparisons are already imprecise, and the available secondary results cannot establish a consistent pattern across the full range of prespecified outcomes. In particular, a finding on one secondary measure would not by itself resolve uncertainty about the primary MRI endpoint or demonstrate that different outcome domains agree.

Unadjusted secondary results carry less weight when many outcomes, time points, and group comparisons are considered. The chance of obtaining at least one apparently notable result by chance increases as more tests are examined. The registry record does not report a multiplicity adjustment or a hierarchy for interpreting secondary outcomes. The secondary findings should therefore be treated as exploratory evidence and interpreted with attention to the incomplete posting of results, rather than as independent confirmation of the primary finding.

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

• Percent Change From Baseline in Cartilage Defect Volume and Cartilage Defect Thickness in the Target Knee at Months 3 and 6
• Change From Baseline in Cartilage Defect Volume in the Target Knee at Months 3, 6 and 12
• Change From Baseline in Cartilage Defect Thickness in the Target Knee at Months 3, 6 and 12
• Number of Participants With Response to Magnetic Resonance Observation of Cartilage Repair Tissue (MOCART) Sub-scales
• Change From Baseline in Boston Leeds Osteoarthritis Knee Score (BLOKS) Sub-scale (Bone Marrow Lesion [BML] Size, Osteophyte Size, Meniscal Extrusion Score [MES], and Meniscal Tear Score [MTS]) Scores at Month 12
• Number of Participants With Shift From Baseline in BLOKS Sub-Scales (Cartilage 1, Synovitis, Effusion) Scores at Month 12
• Number of Participants With Change From Baseline in International Cartilage Repair Society (ICRS) Grade at Months 6 and 12
• Change From Baseline in Knee Injury and Osteoarthritis Outcome Score (KOOS) Sub-scale Scores and International Knee Documentation Committee (IKDC) Score at Months 3, 6 and 12

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is a continuous measure: percent change in cartilage defect volume. The record reports a mean and standard deviation for each group. A mean summarizes the central tendency of the observed measurements. A standard deviation describes their spread; it is not the same as a standard error or a confidence interval. In these groups, the standard deviations are much larger than the estimated between-group mean differences, indicating considerable participant-to-participant variation relative to the contrast estimates.

What does a mean difference represent?

Each primary comparison is the sprifermin group mean minus the placebo group mean. The estimate is expressed in percentage points of the endpoint’s percent-change scale. A positive difference means the sprifermin group’s mean is numerically higher than placebo’s; a negative difference means it is numerically lower. Because more negative endpoint values indicate a larger decrease in measured defect volume, the sign of a difference needs to be interpreted alongside the endpoint definition and both group means. A difference is not itself a within-group change from baseline.

How should the confidence interval be read?

A confidence interval gives a range of values reasonably compatible with the data and the statistical procedure used. It reflects uncertainty in the estimated group difference, not the range containing most individual participant outcomes. A narrow interval would indicate greater precision; a wide interval indicates less precision. All three primary intervals include zero and extend to values on both sides of zero. They therefore do not isolate a single direction of difference. The interval widths are consistent with the limited precision expected from the small analyzed groups and the variation in the outcome.

What does the p-value add?

The unadjusted p-value measures how incompatible the observed comparison is with a null model of no group difference, under the assumptions of the test used. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The reported values—0.785, 0.938, and 0.655—are all unadjusted. The registry record does not report a complete statistical analysis plan or the prespecified testing strategy, so the p-values should not be read as evidence that the trial was designed to make definitive confirmatory claims.

Why can small groups produce uncertain estimates?

When a group contains relatively few observations, its mean can be sensitive to the values of individual participants. If individual outcomes also vary substantially, uncertainty in the estimated mean and in a difference between means increases. This is why the point estimate alone is not enough: the confidence interval communicates how much the estimate could vary under repeated sampling and the method’s assumptions. The primary endpoint analysis involved 14 to 16 participants per group, and the reported standard deviations indicate appreciable dispersion.

What is the role of the analysis population?

The primary outcome used a modified intent-to-treat set, defined as participants in the intent-to-treat set with at least one post-treatment MRI assessment. This definition preserves an intent-to-treat starting point while requiring post-treatment imaging for the endpoint calculation. It also means that the primary estimate does not directly include participants without an evaluable post-treatment MRI. The registry record does not report the reasons for missing MRI assessments, how missing measurements were handled beyond the population definition, or whether sensitivity analyses were conducted.

Why can over-dispersion matter for binary outcomes?

For a binary outcome, such as whether a participant had an event, a simple model may assume that observations vary according to a specified probability and are independent. If the actual variation is greater than that model assumes, uncertainty can be understated; this is commonly described as over-dispersion. With small groups and few events, estimates of event differences can also be unstable. This is a general statistical consideration, not a claim that over-dispersion was found in this trial. The registry record does not describe a model for the secondary binary outcome or an adjustment for over-dispersion.

The posted primary comparisons are unadjusted recalculations from the summary statistics. The registry record does not report sponsor analyses for these comparisons or a detailed inferential analysis plan. No discrepancy is identified between the posted summaries and the recalculated comparisons. The recalculations therefore describe simple contrasts and should not be confused with a complete sponsor analysis or with adjustment for covariates, multiplicity, missing data, or other design features.

11. Confidence Intervals and Precision

Precision is best judged from the confidence intervals, not from the point estimates alone. For 10 µg sprifermin versus placebo, the estimated mean difference is 2.6, with a 95% confidence interval from -16.05 to 21.25. For 30 µg, the difference is -0.8, with a 95% confidence interval from -20.93 to 19.33. For 100 µg, it is 5.0, with a 95% confidence interval from -16.92 to 26.92. These intervals are broad and include zero, leaving substantial uncertainty about both the direction and size of each contrast.

A confidence interval is not a range of individual participant outcomes and does not guarantee that the true value lies within its bounds. It is a measure of sampling uncertainty under the statistical method. In this setting, each interval covers differences that would have different interpretations, including no difference and values on either side of zero. The point estimates are consequently not precise enough to define a clear comparative pattern.

The small analyzed group sizes and the reported standard deviations help explain the limited precision. The confidence intervals should not be reduced to a binary conclusion based only on whether they include zero: their breadth is important. The primary endpoint estimates are compatible with a range of possible group differences, and the registry record does not provide a prespecified threshold for interpreting the size of a meaningful 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 sample-size figures are planning illustrations based on the observed effect, not the original trial’s design calculations. For these continuous-outcome comparisons, the figures are participants per arm, rather than total participants across both arms. 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 set as the stated fraction of the observed effect. The original design assumptions and any sample-size justification are not reported in the registry record.

The 80% power column gives the estimated number of participants per arm for a hypothetical study to have an 80% probability of detecting the specified effect if that effect were real and the planning assumptions held. The 90% power column gives the corresponding number for a higher probability of detection under those same assumptions. These are design calculations, not statements about the probability that a treatment effect exists or about the probability that this completed trial detected one.

Rows planning for a smaller fraction of the observed effect require more participants per arm. Smaller differences are harder to distinguish from random variation, so achieving the same planned power generally requires a larger study. The observed-effect basis is especially uncertain when the estimate comes from small groups, because the observed difference may not represent the effect that a future study would see.

These numbers are not observed power. They are hypothetical planning values and should not be read as an explanation for the trial’s enrollment or as evidence that the original study was sized to detect the posted contrasts. The registry record does not report the trial’s original sample-size assumptions.

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 = 2.600; pooled standard deviation σ = 26.911.
2. n = 2 × (1.96 + 0.842)² × 26.911² / 2.600² = 1,681.7, rounded up to 1,682 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 <- 2.6                      # observed difference in means
sigma <- 26.9112560093356                     # 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(1066871)
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 = 1,682 (80% power), 2,252 (90% power)
 75% of observed effect: n per arm = 2,990 (80% power), 4,003 (90% power)
 50% of observed effect: n per arm = 6,727 (80% power), 9,006 (90% power)
power.t.test: 1682.7 
Simulated power at n = 1,682 per arm: 0.800 (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. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome and 12 secondary outcomes, with several outcomes measured at multiple time points or across multiple scales. This creates a multiplicity issue: when many statistical comparisons are made, some results can appear noteworthy by chance even if there is no underlying effect. Whether and how to control this issue depends on the planned testing strategy. The strategy, including any correction, hierarchy, or distinction between confirmatory and exploratory tests, is not reported in the registry record.

The primary comparisons are presented as unadjusted. The posted results do not establish that the three dose comparisons were controlled as a family of tests, and the secondary outcome comparisons should likewise not be interpreted as multiplicity-adjusted evidence. This matters when interpreting isolated p-values, especially when there are multiple doses, outcomes, and follow-up times.

The primary endpoint population was modified intent-to-treat: participants from the intent-to-treat set with at least one post-treatment MRI assessment. The analyzed group sizes are 16, 14, 15, and 16. The registry record does not provide a full account of participant flow, the number with missing endpoint data by reason, or the method used to address missing measurements beyond the stated analysis-population definition. It also does not report sensitivity analyses for missing data or describe how departures from the intended treatment schedule were handled.

The record gives the study status as terminated, but the reason for termination, whether recruitment or follow-up ended as planned, and the consequences for outcome availability are not reported in the registry record. These omissions limit assessment of whether incomplete data or early termination could have influenced the posted estimates. They do not show that such influences occurred; they mean that their extent cannot be evaluated from the registry record.

15. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: joint injury (5.6%), hypertension (5.6%), menorrhagia (5.3%).

The serious adverse event figures give the number of participants at risk and the number affected in each group. The at-risk counts were 19 for 10 µg sprifermin, 18 for 30 µg, 18 for 100 µg, and 19 for placebo. The corresponding counts affected were 1, 0, 2, and 0. These are descriptive counts, not estimates adjusted for differences in exposure or follow-up.

The registry also lists serious event terms with percentages: Joint injury, 5.6%; Hypertension, 5.6%; and Menorrhagia, 5.3%. The record does not provide an arm attribution alongside these terms in the posted summary. The percentages should not be treated as comparative rates between treatment groups.

The number of deaths is not reported in the registry record. Serious adverse events and other adverse events are distinct categories, and the serious-event counts do not summarize all adverse experiences. These data describe what was posted; small group counts and the absence of event-level context limit comparisons. They do not establish causation or provide a complete safety assessment.

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

The primary result is a set of group means and unadjusted comparisons for percent change in cartilage defect volume at Month 12. All groups had negative mean changes, and each estimated difference between a sprifermin dose and placebo had a confidence interval that included zero. The intervals are broad, so the posted data do not precisely determine the direction or size of a group difference.

That uncertainty should not be converted into a claim that the groups were equivalent. An imprecise estimate can be compatible with no difference and with differences in either direction. The record does not provide an equivalence or non-inferiority design, a prespecified margin, or a clinically meaningful threshold for the endpoint. Nor do the estimates establish that any observed within-group change was caused by the assigned treatment; changes from baseline can occur for reasons that the group comparisons are intended to help address.

The primary endpoint is an MRI-based structural measure, not a direct measure of pain, function, or overall health. The record lists knee symptom and function outcomes among the secondary measures, but comprehensive results for those outcomes are not reported in the registry record. The primary MRI result therefore should not be treated as a complete account of how participants felt or functioned.

The study status is terminated, and the reason is not reported in the registry record. The posted information also does not fully describe participant flow, missing-data handling, multiplicity procedures, or the complete statistical analysis plan. These limitations make it difficult to assess the full context of the estimates. The appropriate interpretation is confined to the posted results: small analyzed groups, uncertain primary comparisons, and incomplete information for assessing the wider trial evidence. No clinical recommendation follows from these statistical results alone.

17. Why This Trial Matters Statistically

This trial illustrates why an estimated treatment difference needs to be read together with its precision. The point estimates are modest in relation to the reported variability, and the confidence intervals extend across zero in every dose comparison. Looking only at the direction or size of a mean would obscure how much uncertainty remains.

It also demonstrates the distinction between within-group change and a randomized comparison. Each group’s mean percent change describes what was observed among its analyzed participants. The treatment contrast asks how that mean compares with the placebo mean. A negative mean in each group does not by itself show a treatment-specific effect; the randomized between-group comparison is central to that question.

The registry record also highlights challenges common to multi-outcome trials. A range of structural and participant-reported outcomes can provide complementary information, but multiple measures and follow-up times make prespecification and multiplicity handling important. Those analysis details are not reported in the registry record, and the secondary results are incomplete. Readers therefore need to distinguish the primary endpoint comparisons from exploratory or unavailable secondary evidence.

Finally, the planning illustrations show how strongly required sample size can depend on the effect a future study aims to detect. They are hypothetical calculations based on observed contrasts, not evidence of the original design or of power achieved in this trial. This distinction matters particularly when the observed estimates come from small analyzed groups. Taken together, the record is a useful example of the limits of drawing firm conclusions from uncertain estimates and incomplete statistical reporting.

18. Limitations

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

The primary endpoint was analyzed in small groups, with 14 to 16 participants per group. The observed standard deviations are substantial, and the resulting confidence intervals are broad. This limits the precision of the estimated differences and makes the point estimates vulnerable to sampling variation.

The primary analysis population required at least one post-treatment MRI assessment. The registry record does not report a complete participant flow, reasons for absent MRI data, or sensitivity analyses addressing missing outcomes. It is therefore not possible to determine from the posted results how much missing imaging may affect the estimated comparisons.

The trial is recorded as terminated, but the reason and its implications for recruitment, follow-up, or outcome availability are not reported in the registry record. The record also does not report the original sample-size justification, a full statistical analysis plan, or the approach to multiplicity. The posted comparisons are unadjusted, and the available secondary results do not cover all listed outcomes.

The primary endpoint is a structural MRI measure. It is not a substitute for symptom, function, or other clinical outcomes. Although such outcomes are listed, their detailed results are not reported in the registry record. The safety summary is also limited: serious adverse event counts are small, event terms lack arm attribution in the summary, and deaths are not reported in the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Sprifermin (AS902330) in Cartilage Injury Repair (CIR) — NCT01066871.
• Registry record last updated 2016-03-23; status last verified 2016-02.
• Results first posted 2016-03-23.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

The registry record describes a randomized, placebo-controlled, parallel-group phase II trial of three intra-articular sprifermin doses for acute knee cartilage injury. It lists 74 participants enrolled and gives the status as terminated. The primary endpoint was percent change in cartilage defect volume from baseline to Month 12, measured by central MRI.

For the primary endpoint, analyzed group sizes ranged from 14 to 16. The three unadjusted mean differences versus placebo had confidence intervals that included zero and were broad. The record lists 12 secondary outcomes, but detailed results are not reported in the registry record for most of them. The reason for termination, original sample-size assumptions, full analysis plan, and detailed handling of missing data are not reported in the registry record.