ProspectiveTrial of Proton Beam Combined With Anti-VEGF Therapy for Exudative Age-related Macular Degeneration (AMD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for ProspectiveTrial of Proton Beam Combined With Anti-VEGF Therapy for Exudative Age-related Macular Degeneration (AMD) (NCT01213082), a randomized phase 1, phase 2 study. The primary outcome is Percent of Eyes With Severe Ocular Adverse.

2. Clinical Question

The trial’s stated purpose was to test whether adding proton beam irradiation to intravitreal anti-VEGF therapy was safe and potentially more effective than anti-VEGF therapy alone for eyes with exudative age-related macular degeneration. The randomized groups compare two radiation doses, each paired with anti-VEGF therapy, against sham irradiation paired with anti-VEGF therapy. Thus, the design addresses the contribution of radiation when anti-VEGF therapy is part of each regimen; it is not a comparison of anti-VEGF therapy against no anti-VEGF therapy.

The primary outcome focuses on severe vision loss attributed to radiation retinopathy or papillopathy by Month 24. The registry record also lists the number of anti-VEGF injections administered as a secondary outcome. It does not post a result for that secondary outcome. The record therefore gives some descriptive information about the primary event in each group, but it does not provide a reported statistical comparison establishing whether the groups differ.

3. Trial Design

This was a randomized, triple-masked, parallel-group Phase I/II study. Its official title describes a prospective comparison of intravitreal anti-VEGF therapy combined with proton beam radiation versus sham irradiation. The registry lists three arms: 24GyE PBT, 16GyE PBT, and sham radiation. The intervention descriptions indicate that anti-VEGF therapy was combined with each of the radiation or sham approaches.

The registry reports enrollment of 34. The primary-outcome arm counts posted in the results are 6.0 for 24GyE PBT, 4.0 for 16GyE PBT, and 6.0 for sham radiation. Those are the counts associated with the displayed primary-outcome values; they should not be taken as an explanation of participant flow or of the analysis population. Details on screening, allocation procedures, masking roles, withdrawals, and the numbers assessed at each follow-up point are not reported in the registry record.

The study began in 2010-09, has a primary completion date of 2017-01-31, and is listed as completed. The sponsor is the University of California, Davis. The record classifies the study as Phase I and Phase II, but does not describe the rationale for the phases or explain whether different objectives applied to different stages.

The primary outcome is a binary event expressed as a percentage: severe ocular vision loss of three or more lines associated with radiation retinopathy or papillopathy, assessed at Month 24. The posted value is 0.0 in each of the three arms. One secondary outcome is listed, the number of anti-VEGF injections administered; a result for it is not reported in the registry record. Although the registry lists two outcome measures, it reports no analyses for the primary outcome comparison.

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:

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 Percent of Eyes With Severe Ocular Adverse, measured over Month 24.

The primary endpoint is the percentage of participants with severe ocular adverse vision loss at Month 24. The event description specifies vision loss of three or more lines associated with radiation retinopathy or papillopathy. This makes the endpoint a binary outcome: an individual either meets the event definition or does not. The emphasis is on a severe visual change with a stated association to radiation-related ocular conditions, rather than on a general measure of vision or every adverse event.

Tracking such an outcome can describe an important potential harm during follow-up. In a study comparing radiation doses with sham irradiation, an outcome tied to radiation retinopathy or papillopathy is relevant to evaluating the safety question. The endpoint’s timing also gives a specified point at which the event status is summarized; it does not describe the course of vision over the full follow-up period.

There is a wording inconsistency in the registry record: the endpoint title and unit refer to a percentage of participants, while the outcome description refers to vision loss in eyes. The record does not clarify how the participant-level percentage was formed if both eyes could contribute, or whether one eye per participant was assessed. It also does not report the outcome’s analysis population or additional assessment details. The posted arm values are 0.0 for all three groups, but that alone does not resolve the unit wording or establish how complete the Month 24 assessments were.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a binary outcome summarized as a percentage: whether the defined severe vision-loss event was reported at Month 24. Statistical methods for binary outcomes generally begin with the number meeting the event definition and the number assessed. A percentage expresses the event frequency relative to its denominator. In this record, the posted arm values are 0.0, but the registry does not clarify the endpoint’s participant-versus-eye wording or report a detailed denominator definition.

What does a zero event value tell us?

A zero observed value means that no event was recorded among the observations contributing to that result. It does not mean that the event is impossible, nor does it establish that the true event frequency is exactly zero. When events are rare and group sizes are small, the data can be compatible with a range of underlying event frequencies. The record does not provide confidence intervals to show that uncertainty, so the reader cannot assess a reported interval’s width or endpoints.

Comparing binary outcomes across groups usually requires a contrast, such as a difference in event proportions, together with an estimate of uncertainty. Here, all three posted values are 0.0. The registry notes that values are missing or not estimable for a two-arm comparison and lists no analyses. Accordingly, there is no reported effect estimate or statistical test to interpret. It would be inappropriate to infer a comparative treatment effect from the descriptive values or to reconstruct an inferential result that the record does not provide.

Why does uncertainty matter?

A confidence interval is a range of values generated by a statistical procedure to express the uncertainty around an estimated quantity. It is not a guarantee that the true value lies in that range. For a treatment comparison, an interval around an effect estimate helps distinguish a precise estimate from one that remains compatible with materially different possibilities. No confidence interval or comparison estimate is reported for this primary endpoint, so the record does not quantify the precision of any between-group contrast.

General methods for binary data may use exact or approximate procedures, and the choice can matter when event counts are sparse. Approximate methods rely on conditions that can be unreliable with very few events; exact approaches can avoid some approximations but may still yield wide uncertainty. These are general statistical considerations, not a description of the method used in this trial. The analysis method is not reported in the registry record.

What does the study design imply for analysis?

Randomization is intended to make assigned groups comparable on average, but it does not remove uncertainty, guarantee balance in a small trial, or resolve missing outcome data. The parallel design means participants are assigned to one study group rather than crossing between regimens. The registry identifies the study as randomized and triple masked, but does not describe the allocation sequence, allocation concealment, who was masked, or whether masking remained effective. Those procedural details should not be inferred from the design labels.

The study has three arms, which permits more than one scientifically distinct comparison. The active radiation arms can be considered against sham irradiation, or against one another, but each contrast addresses a different question. The registry does not report which comparisons were planned as primary statistical contrasts or how multiple comparisons would be handled. It also gives no adjusted analysis or covariate specification. Those omissions prevent a detailed reconstruction of the intended analysis.

How should planning calculations be read?

General sample-size calculations estimate how many participants might be needed under specified assumptions about an effect, variability or event frequency, allocation, and statistical error. They are planning tools, not outcomes observed in a trial. The registry record describes a planning illustration based on a fraction of the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It also states that the illustration is not the trial’s own design. No numerical planning result is included in the registry record for interpretation.

Such an illustration should not be called observed power. Observed power is not a substitute for an effect estimate and its uncertainty; a planning calculation instead explores a hypothetical design under chosen assumptions. In this case, the primary binary endpoint does not have an estimable two-arm comparison in the registry record, so no result from a planning illustration can establish that the trial was adequately powered to detect a particular contrast. The trial’s sample-size justification and its original statistical plan are not reported in the registry record.

What can be concluded from the available statistical record?

The record supports a descriptive statement: the posted primary outcome is 0.0 in each listed arm. It does not support a claim of statistical significance, equivalence, non-inferiority, or a quantified difference. It also does not report a confidence interval, the statistical procedure, or the handling of missing observations. Keeping the descriptive result separate from unreported inference is essential when interpreting a sparse binary endpoint.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("24GyE PBT", "16GyE PBT", "Sham Radiation")
n      <- c(6, 4, 6)
events <- c(0, 0, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  24GyE PBT  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0
ARM  16GyE PBT  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Sham Radiation  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0

11. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures and identifies one secondary outcome: the number of anti-VEGF injections administered. It does not report the result for that secondary outcome. The primary outcome is the Month 24 percentage with severe ocular vision loss associated with radiation retinopathy or papillopathy. No analysis is listed for the primary comparison.

Multiplicity arises when several outcomes, time points, or treatment contrasts are tested. Repeated testing can make it more likely that at least one result appears noteworthy by chance. Statistical plans may address this through a prespecified hierarchy, adjustment, or a clear distinction between primary and exploratory analyses. The registry record does not report a multiplicity strategy, nor does it specify the planned contrast hierarchy across the three arms. It is therefore not possible to determine how the study intended to manage multiple comparisons.

Missing-data handling is also not reported in the registry record. There is no description of how participants without a Month 24 assessment were handled, whether missing values were imputed, or whether sensitivity analyses were performed. The record does not report an analysis population definition, such as whether the primary analysis followed randomized assignment or included only participants with available outcome measurements.

These omissions limit interpretation of the zero values. The arm counts attached to the posted results do not by themselves describe participant flow or confirm completeness of follow-up. Without outcome-availability and analysis-population details, the extent to which the posted values represent all randomized participants cannot be established from the registry record.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports no serious adverse events in the posted safety groups. The serious-event counts are 0 among 6 at risk in the 24GyE plus anti-VEGF group, 0 among 4 at risk in the 16GyE plus anti-VEGF group, and 0 among 6 at risk in the sham irradiation plus anti-VEGF group. Deaths are reported as 0 in each group.

For other adverse events, the registry lists 0 affected among 6 at risk in the 24GyE group, 1 among 4 at risk in the 16GyE group, and 2 among 6 at risk in the sham group. These are reported counts, not evidence that the groups differ. No comparative safety analysis or uncertainty estimate is reported. The record does not give event descriptions for these other adverse events, their severity, timing, or whether an individual could contribute more than one event.

Zero reported serious events should not be interpreted as proof that serious harms cannot occur. The numbers at risk are limited, and the registry does not report safety follow-up completeness or additional detail about ascertainment. These figures describe what was recorded in the listed groups; they do not provide a precise estimate of safety in a broader population.

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

The registry results document a specific descriptive finding: the Month 24 primary endpoint value is 0.0 in all three posted arms. The endpoint concerns severe vision loss of three or more lines associated with radiation retinopathy or papillopathy. This tells the reader that the event was not recorded in the posted group results, not that the possibility of the event has been ruled out.

The results do not establish that proton beam radiation and sham irradiation have the same effect. Similar observed values, particularly when all are zero, are not a formal demonstration of equivalence. Equivalence or non-inferiority would require an appropriate prespecified framework, a defined margin, and an analysis with uncertainty estimates. Those details and analyses are not reported in the registry record. Nor does the endpoint alone establish comparative effectiveness on vision or other outcomes.

The trial’s stated purpose included safety and the possibility of greater effectiveness, but purpose is not itself a result. The posted primary outcome speaks to a particular severe event at a particular follow-up time. It does not summarize every safety outcome, the full course of visual change, or the burden of treatment. The listed secondary outcome, anti-VEGF injection count, has no result reported in the registry record.

The size and completeness of the analyzed groups are also important. The posted primary-outcome arm counts are 6.0, 4.0, and 6.0, and there is no reported confidence interval or comparison. Without further information on participant flow and missing outcome data, the registry values cannot show whether all enrolled participants contributed to the endpoint. The correct interpretation is therefore narrow: no primary endpoint events are recorded in the posted arm results, while the comparative effect and its precision remain unreported.

14. Why This Trial Matters Statistically

This trial raises a useful statistical problem: how should a reader interpret an apparently reassuring zero-event result from a small randomized study? A zero in each group can look decisive at first glance, but it does not provide evidence that the underlying risks are identical or absent. The amount of information depends not only on whether events occurred, but also on how many observations contributed and how completely outcomes were assessed.

The three-arm design also highlights why the comparison must be defined before interpreting results. The radiation-dose arms and the sham arm address different contrasts, even though anti-VEGF therapy is included in each regimen. A single label such as “treatment versus control” can obscure the distinct questions raised by comparing each radiation dose with sham or comparing the radiation doses with one another. The registry does not report a prespecified comparison hierarchy or estimates for these contrasts.

Finally, the record illustrates the difference between a study’s objective and the statistical evidence posted for that objective. It describes a safety and potential-effectiveness question, but provides only descriptive primary-outcome values, with no reported analysis or uncertainty interval. The mismatch between the endpoint’s participant-based unit and eye-based wording further shows how outcome definitions and denominators affect interpretation. Clear analysis populations, outcome definitions, and comparative estimates are central to judging what a trial can establish.

15. 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 main limitation is the lack of reported inferential results for the primary endpoint. The registry lists no analyses and states that values are missing or not estimable for a two-arm comparison. There is no comparative estimate, confidence interval, or test result with which to assess the magnitude or precision of a difference between groups.

The primary endpoint is described inconsistently: its title and unit refer to a percentage of participants, while its definition refers to vision loss in eyes. The registry record does not explain how eyes were selected or handled if both eyes were relevant. It also does not provide detailed event-assessment procedures, outcome denominators beyond the posted group counts, or information about completeness of Month 24 follow-up.

Further limitations include the absence of reported participant flow, analysis-population definitions, missing-data methods, allocation details, masking roles, and a statistical analysis plan. The record does not report how the study chose its enrollment target or whether a sample-size justification informed the trial design. A planning illustration based on an observed effect is not a substitute for that information or for an analysis of the randomized trial.

The safety information is similarly limited to aggregate affected counts. The record does not describe the other adverse events or provide comparative safety estimates. The listed anti-VEGF injection outcome has no result reported. These gaps constrain interpretation of both safety and treatment-use questions and prevent broad conclusions from the posted primary endpoint alone.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: ProspectiveTrial of Proton Beam Combined With Anti-VEGF Therapy for Exudative Age-related Macular Degeneration (AMD) — NCT01213082.
• Registry record last updated 2018-07-17; status last verified 2018-05.
• Results first posted 2018-07-17.
• 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.

19. Record Summary

The registry record describes a completed randomized, triple-masked, parallel Phase I/II study of proton beam radiation combined with intravitreal anti-VEGF therapy versus sham irradiation combined with anti-VEGF therapy for exudative age-related macular degeneration. Enrollment is listed as 34. The three arms are 24GyE PBT, 16GyE PBT, and sham radiation.

The primary outcome is the Month 24 percentage with severe vision loss of three or more lines associated with radiation retinopathy or papillopathy. Its posted value is 0.0 in each arm. The registry reports no serious adverse events or deaths in the listed groups, and lists other adverse-event counts of 0, 1, and 2 across the respective arms. No primary comparison analysis is reported, and the secondary outcome of anti-VEGF injection count has no posted result. The registry does not report confidence intervals, missing-data methods, or analysis-population details.