Neoadjuvant Immunoradiation for Resectable Non-Small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Neoadjuvant Immunoradiation for Resectable Non-Small Cell Lung Cancer (NCT03237377), a non-randomized phase 2 study. The primary outcome is Toxicities as Measured by Number of Participants Experiencing Adverse Events.

2. Clinical Question

The trial concerns neoadjuvant treatment, meaning treatment before the planned operation. The registry summary describes thoracic radiation delivered concurrently with immunotherapy, followed by an immunotherapy dose in the pre-surgical window and surgical resection. The planned surgical procedure is lobectomy for participants whose disease is judged resectable in that way.

The study's stated clinical context is a pilot evaluation of immunoradiation. Its registry description makes the next treatment cohort conditional on preliminary safety of the durvalumab-and-radiation combination. The listed additional regimen includes tremelimumab with durvalumab and radiation. The record does not establish a comparison of that regimen with durvalumab and radiation: it records no participants at risk in the additional arm.

Accordingly, the primary question should be read as descriptive safety monitoring in the durvalumab-with-radiation group. The registry also lists surgical morbidity and mortality, pathologic response, radiologic response, and overall survival as secondary outcomes. Their outcome results are not reported in the registry record. The stated question does not, by itself, establish whether treatment changes tumor response, survival, or the likelihood of undergoing surgery.

3. Trial Design

The registry classifies NCT03237377 as a phase 2, non-randomized, unmasked, sequential study. Enrollment is recorded as 9, and the study status is terminated. The listed start date is 2017-12-12 and the primary completion date is 2021-11-04. The sponsor is Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins.

The brief description presents the study as a pilot for patients with stage IIIA non-small cell lung cancer considered resectable with lobectomy. The first regimen combines durvalumab with thoracic radiation. The summary describes immunotherapy every 4 weeks for 2 doses during radiation, followed by one dose of immunotherapy alone in the pre-surgical window. It describes radiation as 45Gy in 25 fractions. The arm description specifies durvalumab by IV infusion every 4 weeks for up to 3 doses or cycles, consistent with the summary's description of treatment before surgery.

A further cohort of durvalumab, tremelimumab, and thoracic radiation was to open if preliminary safety of durvalumab with radiation was established. The registry lists that arm, but records zero participants at risk in it. The record does not report why the study was terminated or whether the condition for opening the additional cohort was met. Those details should not be inferred from the arm counts.

After resection, participants may receive standard adjuvant chemotherapy at the treating investigator's discretion. This means post-surgical chemotherapy was not described as a uniform assignment for all participants. The registry also lists lobectomy as a procedure. Because allocation was non-randomized and the additional arm has no participants recorded at risk, the study does not support a like-for-like comparison between the two listed regimens. The registry record does not report a detailed statistical analysis plan, a sample-size justification, or whether interim analyses were planned.

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: The study was terminated due to low accrual on Arm A (Durvalumab with Radiation) so no participants were assigned to Arm B (Durvalumab and Trememlimumab with Radiation).

Overall Study

Analysis population for the primary outcome, as defined in the registry: all participants who were assigned to the durvalumab with radiation arm

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 protocol planned for a second cohort for patients to be treated with durvalumab and trememlimumab with radiation, however, due to low accrual in cohort 1, there was no second cohort.

6. Primary Endpoint

The registry lists the primary outcome as Toxicities as Measured by Number of Participants Experiencing Adverse Events, measured over 3 months post surgery.

The primary outcome is the count of participants experiencing adverse events defined by CTCAE v4.0. The outcome population is all participants assigned to the durvalumab-with-radiation arm, and the stated timeframe is 3 months post surgery. This is a participant count, not a count of individual adverse-event episodes: a participant who experiences multiple events is still a participant in the outcome count. The registry does not report a more detailed rule for how repeat events or events occurring at different times were handled.

Safety outcomes are used to describe the types and frequency of harms observed during treatment and follow-up. A participant count can communicate how many people experienced at least one event under the endpoint definition, but it does not on its own show event severity, duration, treatment relatedness, or whether an event was caused by the study regimen. Those distinctions require separate definitions and results.

The registry's adverse-event entries also separate participants with serious adverse events from those with other adverse events. These entries are useful for describing recorded safety experience, but they should not automatically be treated as the full primary endpoint result unless the registry explicitly identifies them as that result. A primary-outcome result reported with its analysis population and event definition is not separately presented in the registry record. The stated timeframe is after surgery; the record does not clarify how participants who did not undergo surgery would be handled for this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The central statistical distinction is between describing observed counts and estimating a comparative treatment effect. A count of participants with an event can summarize the safety experience in a defined group. A comparison requires an appropriate comparator group, consistent outcome definitions, and an analysis that accounts for how participants entered each group. This registry identifies a non-randomized sequential design, and it records zero participants at risk in the durvalumab-and-tremelimumab-with-radiation arm. The listed arm counts therefore do not form a meaningful treatment contrast.

What does a participant count represent?

A participant count answers how many people met a specified event definition during the stated observation period. It differs from an event count, which can include multiple events for the same person. The primary outcome is described as the number of participants experiencing adverse events defined by CTCAE v4.0, with a timeframe of 3 months post surgery. The available safety entries separately report serious adverse events, other adverse events, and deaths. Readers should preserve those distinctions rather than combine the categories or assume that the group-level entries are interchangeable with a separately reported primary-outcome analysis.

Counts are most interpretable alongside the population in which they were observed, the number at risk, the follow-up window, and clear event definitions. Here the registry identifies the assigned population for the primary endpoint and gives the at-risk count for the durvalumab-with-radiation group. It does not report all details needed to reconstruct an endpoint analysis, including how the endpoint was handled for participants who did not undergo surgery or how missing outcome information was treated. A count should not be read as a complete description of event timing, duration, seriousness, or causal attribution.

Why does allocation matter?

Random allocation is a design method intended to make treatment groups comparable, on average, with respect to measured and unmeasured characteristics. A non-randomized design does not provide that same basis for a causal comparison. Participants assigned to different regimens may differ for reasons related to treatment selection, clinical judgment, timing, or eligibility. The registry describes a sequential design, so the arms should not be treated as concurrent randomized groups. In addition, no participants are recorded at risk in the additional immunotherapy arm. The relevant information is thus descriptive experience in the arm with recorded participants, not a comparative estimate of regimen effects.

Masking is listed as none. In general, masking can reduce the influence of knowledge of treatment assignment on some assessments or decisions. Its absence does not invalidate observed counts, but it can matter when outcomes involve judgment or reporting. The registry record does not describe blinded outcome assessment or a specific process to reduce assessment bias.

What would uncertainty measures add?

A confidence interval is a statistical range intended to communicate the uncertainty around an estimate under a specified model and sampling process. A p value evaluates compatibility between data and a defined null hypothesis under an analysis model; it does not measure the probability that a treatment works. Neither type of result is reported for the outcomes in this record, and no comparative estimate or inferential test is listed. It would be inappropriate to construct new estimates from the posted counts or to imply that an unreported interval supports a particular conclusion.

Small studies commonly yield imprecise estimates because the observed data contain limited information about the range of outcomes in a wider population. This is a general statistical consideration, not a calculated precision result for this trial. A count observed in a small pilot can identify events that warrant careful description, but it cannot establish a stable rate or rule out uncommon outcomes. The registry does not report a sample-size justification that would clarify the intended precision of the safety estimate.

How should planning illustrations be understood?

Any sample-size illustrations associated with these results are planning exercises, not observed trial results and not observed power. The stated assumptions use a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect set as a stated fraction of the observed effect. Such calculations answer a hypothetical planning question conditional on their assumptions. They do not show that the trial was designed with those assumptions, and they cannot repair the lack of a like-for-like comparator in the registry record. The record does not report the sponsor's own sample-size rationale. The illustrations should therefore not be taken as evidence that the enrolled study had adequate power or as confirmation of a treatment effect.

For a binary outcome, a planning calculation typically depends on the event probability assumed in each group and the contrast to be detected. General methods may require different assumptions when outcomes are clustered, repeated, or more variable than a simple model allows. Over-dispersion is greater variability than a model expects; if ignored, it can make uncertainty appear smaller than it is. These are general concepts only. The registry does not report a model for over-dispersion, a cluster structure, or an adjusted analysis for this trial. The methods described in the registry are not detailed enough to reconstruct an inferential analysis.

The appropriate reading is consequently narrow: the posted record documents a pilot design and descriptive safety counts for the arm with participants recorded at risk. It does not provide a randomized estimate, an analysis of statistical uncertainty, or a basis for comparing outcomes across the listed regimens.

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("Durvalumab With Radiation")
n      <- c(9)
events <- c(9)
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  Durvalumab With Radiation  9  9  100.0  66.4  100.0  70.1  100.0  69.2  99.7

11. Multiplicity, Missing Data and Analysis Populations

The registry names a primary safety outcome and lists secondary outcomes of surgical morbidity and mortality, percentage of participants with pathologic response, percentage of participants with radiologic response, and overall survival. It records 6 outcome measures overall. The outcome names do not establish which analyses were prespecified, how each measure was defined in detail, or how any results were analyzed. No statistical analyses are listed in the record.

Multiplicity arises when several outcomes or comparisons are tested. Without a stated analysis plan, it is not possible to determine whether multiplicity adjustments were planned, whether the outcomes were ranked in a testing hierarchy, or whether the results were intended to be descriptive. The registry record does not report a multiplicity strategy. This matters particularly when many outcome estimates are considered together, because selecting or emphasizing results after observing them can make evidence seem more conclusive than it is.

The record does not describe missing-data counts, reasons for missingness, imputation, censoring rules, or sensitivity analyses. It does not report how the primary outcome was handled for participants who did not reach surgery, although the endpoint timeframe is defined relative to surgery. These details are important because an outcome unavailable for some participants may affect the population represented by the reported count. No missing-data procedure should be assumed.

The primary outcome population is stated as all participants assigned to the durvalumab-with-radiation arm. The record does not further explain whether the analysis followed that assignment regardless of treatment received, whether participants were excluded from the safety analysis, or how deviations were addressed. It also does not report a separate population for each secondary outcome. The registry's allocation and outcome descriptions should therefore be distinguished from a complete statistical analysis plan: key details about multiplicity, missingness, and analysis populations are not reported in 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: aspiration pneumonia (11.1%), thromboembolic event (11.1%), appendicitis (11.1%), pneumothorax (11.1%), atrial fibrillation (11.1%).

For the durvalumab-with-radiation group, the registry records 9 participants at risk, 3 participants affected by serious adverse events, and 9 affected by other adverse events. It also records 2 deaths. These entries describe recorded safety experience; they do not show whether any event or death was caused by treatment. The registry does not report causes of death or a treatment-relatedness assessment.

The serious-event terms each have a reported value of 11.1%: aspiration pneumonia, thromboembolic event, appendicitis, pneumothorax, atrial fibrillation, and aspiration. These terms should not be summed. The record does not indicate whether terms occurred in the same participant, whether events overlapped across categories, or whether each term represents a distinct person. The term list is not a substitute for event-level timing or severity information.

The second listed arm has zero participants recorded at risk and zero affected. That is not evidence of no harm in that regimen; it means the record supplies no participant experience in that arm for a safety comparison. The counts also do not provide a background rate or a concurrent control. Interpret them as descriptive counts from the recorded durvalumab-with-radiation group, within the registry's stated context and limits.

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

The registry record establishes that this was a phase 2 pilot study of neoadjuvant immunoradiation for resectable non-small cell lung cancer, with a primary safety outcome defined around adverse events by 3 months post surgery. It records participants at risk and adverse-event counts for the durvalumab-with-radiation arm. Those entries provide a descriptive account of reported safety events in that group.

The counts do not establish that durvalumab with radiation caused the reported events, nor do they show how the experience compares with usual care or another treatment. The study was non-randomized, unmasked, and sequential; the additional durvalumab-and-tremelimumab-with-radiation arm has no participants recorded at risk. There is therefore no like-for-like control comparison in the registry figures. The record also does not provide inferential analyses or uncertainty intervals for the safety counts.

The listed deaths cannot be interpreted as treatment-related from the information reported. The serious-event terms likewise do not establish causation, and the registry does not clarify whether terms overlapped within participants. The presence of a named outcome, such as pathologic response or overall survival, does not mean that an outcome result is available. Results for the named secondary outcomes are not reported in the registry record.

The study's terminated status should be reported as recorded, but the reason for termination is not reported in the registry record. Nor does the record establish whether the additional cohort opened or why no participants are recorded in it. These unknowns limit interpretation of the study's conduct and planned progression.

In statistical terms, this is evidence of what the registry records in a small pilot cohort, not a precise estimate of comparative safety or efficacy. It can be read as a descriptive safety record that identifies reported events and deaths. It cannot establish a treatment effect, quantify comparative risk, or support a conclusion about tumor response or survival. The registry's limitations apply to interpretation; they do not justify replacing the posted counts with assumptions about unreported outcomes.

14. Why This Trial Matters Statistically

Pilot studies can be useful for documenting feasibility and safety signals before a larger research program, but their role is different from that of a randomized comparative trial. This study's sequential structure reflects a plan in which a further regimen would be considered conditional on preliminary safety of durvalumab with radiation. The registry record is therefore important to read as a study of an early treatment strategy, not as a direct test of competing regimens.

The statistical point is that study design determines what questions the results can answer. A descriptive count can report observed events in a treatment group. It cannot, without an appropriate comparison and analysis, separate treatment effects from patient characteristics, disease course, clinical decisions, or other care. The non-randomized allocation and absence of recorded participants in the additional regimen arm make this distinction especially relevant.

The registry also illustrates why endpoint definitions and analysis populations matter. Its primary outcome is framed as a participant count for adverse events, with a specified clinical definition and follow-up timeframe. That wording helps identify what the study intended to monitor, while the absence of a separately reported primary analysis and details on missing outcomes limits reconstruction of the endpoint result.

Finally, a terminated study with modest enrollment may leave clinically relevant questions unresolved without making the recorded observations irrelevant. The reported safety counts, event terms, and deaths should remain visible as descriptive findings, while claims about comparative safety, response, survival, or causation require evidence not reported in the registry record. This distinction is central to responsible interpretation of early clinical research.

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 enrollment is recorded as 9, and the registry lists no statistical analyses. The record does not report confidence intervals, p values, adjusted estimates, or a comparative analysis. The counts consequently offer limited information about the precision of the safety experience and cannot support a treatment-effect estimate.

The study is non-randomized, unmasked, and sequential. These design features limit causal interpretation and comparability between treatment groups. The additional durvalumab-and-tremelimumab-with-radiation arm has zero participants recorded at risk, so it cannot serve as an observed control or regimen comparison. The record does not explain why the arm has no participants or whether it opened.

The primary outcome is specified relative to surgery, but the record does not describe how participants who did not undergo surgery would be handled. It also does not report missing outcome information, follow-up completeness, or analysis rules for the endpoint. The safety entries do not include event-level timing, duration, attribution, or causes of death. Serious-event terms are listed with percentages, but overlap among terms is not described.

The registry lists several secondary outcomes, but their results are not reported in the registry record. The number of outcome measures recorded exceeds the number of named secondary outcomes, and the record does not provide further analysis details to explain this. It does not report a multiplicity strategy, a detailed statistical plan, or the sponsor's sample-size justification. The reason for the terminated status is also not reported in the registry record. These gaps restrict what can be concluded about the study's execution, analysis, and results.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Neoadjuvant Immunoradiation for Resectable Non-Small Cell Lung Cancer — NCT03237377.
• Registry record last updated 2026-05-27; status last verified 2026-05.
• Results first posted 2023-10-16.
• 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

NCT03237377 is a phase 2, non-randomized, unmasked, sequential pilot study of neoadjuvant immunoradiation for resectable stage IIIA non-small cell lung cancer. Enrollment is recorded as 9. The study is listed as terminated, with a start date of 2017-12-12 and a primary completion date of 2021-11-04. The sponsor is Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins.

The primary outcome is the number of participants experiencing adverse events defined by CTCAE v4.0, in the durvalumab-with-radiation group, with a timeframe of 3 months post surgery. The registry records serious and other adverse-event counts and deaths for that group. The listed durvalumab-and-tremelimumab-with-radiation group has zero participants at risk. No statistical analyses are listed, and results for the named secondary outcomes are not reported in the registry record.