This page separates information reported in the ClinicalTrials.gov record from statistical interpretation. The registry reports that results have been posted, but the ClinicalTrials.gov record contains no formal statistical-analysis records or numerical primary-endpoint estimates.
1. Trial at a Glance
HERTHENA-Lung01 is a randomized, open-label, parallel-group phase 2 trial evaluating two patritumab deruxtecan dosing approaches in subjects with metastatic or locally advanced EGFR-mutated non-small cell lung cancer. The ClinicalTrials.gov record lists objective response rate, assessed by blinded independent central review, as the single registered primary endpoint.
| Feature | HERTHENA-Lung01 |
|---|---|
| Trial name | HERTHENA-Lung01 |
| Phase | Phase 2 |
| Population | Subjects with metastatic or locally advanced EGFR-mutated non-small cell lung cancer |
| Design | Randomized, parallel-group |
| Masking | None |
| Allocation | Randomized |
| Primary purpose | Treatment |
| Enrollment | 277 |
| Interventions | Patritumab Deruxtecan (Fixed dose) and Patritumab Deruxtecan (Up-Titration) |
| Lead sponsor | Daiichi Sankyo |
| Sponsor type | Industry |
| ClinicalTrials.gov | NCT04619004 |
2. Clinical Question
The trial addresses a dosing-related clinical question within subjects with metastatic or locally advanced EGFR-mutated non-small cell lung cancer: how does patritumab deruxtecan perform when administered using the two registered dosing approaches, a fixed-dose approach and an up-titration approach?
Population
Subjects with metastatic or locally advanced EGFR-mutated non-small cell lung cancer.
Intervention
Patritumab deruxtecan administered using the fixed-dose approach.
Comparator
Patritumab deruxtecan administered using the up-titration approach.
Primary question
The registered primary endpoint is objective response rate as assessed by blinded independent central review.
This is therefore not a conventional placebo-versus-active-treatment comparison. Both randomized groups receive patritumab deruxtecan, with the registered intervention difference being the dosing approach. That distinction matters when interpreting any eventual treatment-group contrast: a difference would primarily inform the comparison of the two dosing strategies represented by the trial, rather than establish an effect of patritumab deruxtecan versus no patritumab deruxtecan.
3. Trial Design
Patritumab Deruxtecan — Fixed Dose
- Patritumab deruxtecan
- Fixed-dose approach
- Serious adverse events: 90 affected out of 225 at risk
Patritumab Deruxtecan — Up-Titration
- Patritumab deruxtecan
- Up-titration approach
- Serious adverse events: 16 affected out of 50 at risk
4. Trial Timeline and Registry Status
Trial start
The ClinicalTrials.gov record identifies February 2, 2021 as the study start date.
Primary completion
The ClinicalTrials.gov record identifies November 21, 2022 as the primary completion date.
Active, not recruiting
The ClinicalTrials.gov status is ACTIVE_NOT_RECRUITING.
The trial is therefore presented here as a completed-primary-endpoint study with a current registry status of active, not recruiting. The ClinicalTrials.gov record does not provide additional follow-up dates or a separate long-term efficacy data cutoff.
5. Primary Endpoint
| Endpoint | Registry definition | Time frame | Type |
|---|---|---|---|
| Objective Response Rate (ORR) as Assessed by Blinded Independent Central Review (BICR) | ORR is defined as the proportion of participants with a best overall response (BOR) of confirmed complete response (CR) or confirmed partial response (PR) as assessed by BICR per Response Evaluation Criteria in Solid Tumors (RECIST) v1.1. CR was defined as a disappearance of all target lesions and PR was defined as at least a 30% decrease in the sum of diameters of target lesions based on RECIST v1.1. | Data collected from screening until time of disease progression by BICR, death, lost to follow up, study discontinuation, whichever occurs first, assessed up to approximately 21 months | Binary |
The primary endpoint is therefore a participant-level binary response outcome summarized as a proportion. Each participant is classified according to the prespecified best overall response framework, and the resulting proportion is the ORR.
Why blinded independent central review matters
The use of BICR introduces an important separation between treatment administration and efficacy assessment. The registry identifies the response endpoint as being assessed by an independent central review process rather than relying solely on an investigator's local assessment. Statistically, this can help standardize radiologic classification across treatment groups.
The trial is open-label, so participants and treating investigators are not masked to assignment. That makes the independent central review particularly relevant to the measurement of a radiologic endpoint: the endpoint assessment has an additional centralized review layer even though the overall trial is not masked.
6. Statistical Methodology
Binary endpoint analysis
Because the registered primary endpoint is binary, the fundamental efficacy quantity is the proportion of participants who meet the ORR definition. In a two-arm randomized trial, a formal comparison would typically examine the response proportions in the two randomized groups and quantify uncertainty around the between-group difference or another prespecified effect measure.
The exact analysis population and the exact inferential procedure should follow the prespecified statistical analysis plan. The ClinicalTrials.gov record does not provide those additional specifications.
Confidence intervals for a response proportion
A confidence interval around ORR quantifies uncertainty in the estimated response proportion. In a randomized comparison, confidence intervals can also be constructed for a treatment-group contrast, such as a risk difference or risk ratio, depending on the prespecified analysis.
The ClinicalTrials.gov record contains no numerical ORR estimates or confidence intervals. It would therefore be inappropriate to reverse-engineer an estimate from the enrollment or safety counts.
Between-arm comparison
For two randomized groups with a binary endpoint, a formal analysis could use a method such as a two-sample comparison of proportions, a stratified test if randomization strata were prespecified, or a regression model for binary outcomes. The correct method depends on the statistical analysis plan, including whether any stratification factors were used and which effect measure was prespecified.
Time-to-event terminology does not automatically apply
The primary endpoint is ORR rather than overall survival or progression-free survival. The registry's time frame describes the period during which response information is collected, but that does not convert ORR into a time-to-event endpoint. A participant can have a best overall response classification without ORR itself being analyzed as a Kaplan-Meier event time.
7. Results Status
Registry results status
The ClinicalTrials.gov record reports that results have been posted and that 9 outcome measures are posted.
No formal statistical-analysis records were posted on ClinicalTrials.gov for the registered primary endpoint.
The ClinicalTrials.gov record therefore support a precise statement about results availability, but not about the numerical efficacy result. There is no ORR percentage, treatment-group response count, confidence interval, p-value, risk difference, risk ratio, odds ratio, or other formal comparison in the ClinicalTrials.gov record.
What would normally be analyzed?
| Question | Typical statistical treatment | What is available here |
|---|---|---|
| What proportion responded? | Estimate ORR as the proportion with confirmed CR or confirmed PR. | The endpoint definition is available; the numerical ORR is not reported. |
| Do the randomized groups differ? | Compare response proportions using the prespecified binary-endpoint method. | No formal comparison is reported. |
| How precise is the estimate? | Report a confidence interval around the proportion or treatment contrast. | No confidence interval is reported. |
| Is there statistical evidence against a null hypothesis? | Report the prespecified hypothesis test and p-value. | No p-value is reported. |
This distinction is central to responsible interpretation. "Results posted" does not mean that every numerical result is available in the ClinicalTrials.gov record. The analysis should report only the statistical quantities actually supported by the source data.
8. Planned Analysis of the Primary Endpoint
Because the registry provides a binary ORR endpoint but no corresponding statistical-analysis record, the appropriate educational framework is to describe how such an endpoint is normally analyzed without assigning an outcome to HERTHENA-Lung01.
Step 1: Define response
Each participant is classified according to the registered BICR assessment. A confirmed complete response requires disappearance of all target lesions. A confirmed partial response requires at least a 30% decrease in the sum of diameters of target lesions based on RECIST v1.1.
Step 2: Calculate ORR
The number of participants with confirmed CR or confirmed PR is divided by the relevant analysis denominator. The resulting proportion is the objective response rate.
Step 3: Compare randomized groups
Because the study is randomized and has two parallel groups, an inferential comparison can be made between the fixed-dose and up-titration strategies if specified by the statistical analysis plan. The choice of effect measure and test should be taken from that prespecified plan rather than selected retrospectively.
Step 4: Quantify uncertainty
A confidence interval should accompany the estimated effect. The interval communicates the statistical precision of the observed estimate and should be interpreted separately from the p-value.
9. Safety Results
The ClinicalTrials.gov record provides serious adverse-event counts by arm. These are the only safety results available in the ClinicalTrials.gov record.
| Safety measure | Patritumab Deruxtecan 5.6 mg/kg | Patritumab Deruxtecan Up-Titration |
|---|---|---|
| Serious adverse events | 90/225 affected/at risk | 16/50 affected/at risk |
The counts describe the number affected and the corresponding at-risk denominator reported by the registry data. They should not be interpreted as an efficacy finding or as evidence about the primary ORR endpoint.
The two denominators also differ substantially. This is expected to affect the precision of arm-specific safety estimates: a smaller at-risk population generally provides less statistical information about an adverse-event frequency than a larger one.
10. Interpreting the Binary Primary Endpoint
ORR is the proportion of participants whose best overall response is a confirmed complete response or confirmed partial response under the registered BICR and RECIST v1.1 framework. It is therefore a response endpoint based on tumor-assessment classifications.
ORR is not the same as overall survival, progression-free survival, duration of response, or the probability that an individual participant will remain alive for a particular period. A response proportion summarizes one prespecified dimension of treatment outcome.
The registry specifically requires a confirmed CR or PR. That requirement distinguishes the primary endpoint from a one-time observation of tumor shrinkage and makes the response classification dependent on the registered assessment framework.
Blinded independent central review provides a centralized assessment mechanism for the radiologic endpoint. Because the overall trial is unmasked, this distinction is important when considering how the primary endpoint was assessed.
11. Statistical Methods Explained
Why is ORR a binary endpoint?
The registry defines the endpoint as the proportion of participants meeting a response criterion: confirmed CR or confirmed PR. Each participant can therefore be represented by a response indicator for the purpose of the primary analysis. The endpoint is summarized across participants as a proportion.
Why is the endpoint based on best overall response rather than a single scan?
The registered definition uses best overall response and requires confirmation. This makes the endpoint a summary of the participant's response classification over the specified assessment period rather than a measurement from one isolated tumor assessment.
Why does the registry's time frame not make ORR a survival endpoint?
The time frame specifies when response information is collected: from screening until disease progression by BICR, death, loss to follow-up, or study discontinuation. The endpoint itself remains a binary response classification. Time-to-event methods such as Kaplan-Meier estimation are not automatically required merely because an endpoint has a longitudinal assessment period.
How would a confidence interval help?
A confidence interval describes statistical uncertainty around an estimated response proportion or treatment contrast. It is different from the observed estimate itself: two studies could have the same estimated ORR but different interval widths if their information differs.
What would a p-value tell us?
If a prespecified hypothesis test compared the randomized groups, the p-value would quantify how compatible the observed data are with the null hypothesis under that test. It would not measure the magnitude of the response difference, and it would not tell us the probability that either dosing strategy is superior.
Why should the fixed-dose and up-titration groups not be described as placebo versus treatment?
Both registered arms receive patritumab deruxtecan. The intervention descriptions differ in dosing approach rather than in the presence versus absence of the drug. Therefore, any between-arm analysis would address the randomized dosing comparison represented by the trial.
12. Randomization and Causal Interpretation
Randomization is one of the most important statistical features of HERTHENA-Lung01. In a randomized parallel-group trial, assignment is determined by the trial's allocation mechanism rather than by an investigator's choice based on an observed participant characteristic. This supports comparison of outcomes between randomized groups under the assumptions of the study design.
However, randomization does not make the two groups numerically identical at baseline, nor does it eliminate uncertainty. It provides the foundation for causal comparison by reducing systematic differences in treatment assignment, while the statistical analysis quantifies the remaining sampling uncertainty.
Randomized comparison
The trial's allocation is explicitly listed as randomized and its design model as parallel.
Open-label design
Masking is listed as none. Treatment assignment was therefore not masked at the trial level.
Central review
The primary ORR endpoint is assessed by blinded independent central review.
Endpoint focus
The single registered primary endpoint is binary ORR rather than a time-to-event endpoint.
13. Analysis Populations and Denominators
One of the most easily overlooked statistical issues in a clinical-trial report is the denominator. The overall enrollment reported in the ClinicalTrials.gov record is 277, while the serious-adverse-event results use 225 and 50 as arm-specific at-risk denominators.
For ORR, the appropriate denominator is determined by the prespecified efficacy analysis population and response-assessment rules. The ClinicalTrials.gov record does not specify that denominator. It would therefore be incorrect to assume that 277 is automatically the denominator for the primary ORR calculation.
| Quantity | registry value | Statistical significance |
|---|---|---|
| Overall enrollment | 277 | Describes the registered study enrollment |
| Serious AE denominator — fixed-dose arm | 225 | Defines the registry-reported safety at-risk population |
| Serious AE denominator — up-titration arm | 50 | Defines the registry-reported safety at-risk population |
| Number of posted outcome measures | 9 | Indicates that multiple outcomes are available in the registry results record |
Keeping these denominators separate is essential. An analysis can be numerically correct yet misleading if an outcome is reported using a denominator that does not correspond to its prespecified analysis population.
14. Limitations
- No formal statistical-analysis records reported: no formal statistical analyses were posted, so numerical effect estimates, confidence intervals, and p-values cannot be reported from the ClinicalTrials.gov record.
- Results are posted but incompletely represented in the ClinicalTrials.gov recordset: the registry data state that results are posted and list 9 posted outcome measures, but the numerical primary-endpoint results are not included in the registry-reported material.
- Open-label design: masking is listed as none. This is relevant to interpretation of outcomes that could potentially be influenced by knowledge of treatment assignment, although the primary ORR assessment uses BICR.
- Unknown prespecified inferential method: the ClinicalTrials.gov record identifies ORR as binary but does not identify the exact formal test or effect measure used for the primary comparison.
- Unknown missing-data rules: the ClinicalTrials.gov record does not state how missing or unevaluable response assessments were handled in the primary analysis.
- Unknown stratification: the ClinicalTrials.gov record does not identify randomization strata, so a stratified versus unstratified comparison cannot be attributed to this trial from the ClinicalTrials.gov record.
- Safety denominators differ from overall enrollment: the serious-adverse-event denominators posted on ClinicalTrials.gov for the two arms total 275 rather than the overall enrollment of 277; the source data provided here do not explain the discrepancy.
- No subgroup results reported: the ClinicalTrials.gov record does not provide subgroup estimates, interaction tests, or forest-plot results.
- No survival estimates reported: the registry-reported primary endpoint is ORR, and the ClinicalTrials.gov record does not provide median survival, hazard ratios, or other time-to-event results.
15. Design Features Not Specified in the ClinicalTrials.gov record
Several statistical design topics are important in a complete trial analysis but cannot be attributed to HERTHENA-Lung01 without supporting information. The ClinicalTrials.gov record does not specify a non-inferiority margin, crossover rules, factorial structure, multiplicity strategy, interim-analysis procedure, missing-data or imputation method, randomization stratification factors, or Bayesian methodology.
| Design topic | What the ClinicalTrials.gov record establishes |
|---|---|
| Non-inferiority margin | Not specified in the ClinicalTrials.gov record. |
| Crossover | Not specified in the ClinicalTrials.gov record. |
| Factorial design | The design is identified as parallel; no factorial structure is reported. |
| Multiplicity | No multiplicity procedure is reported. |
| Interim analysis | No interim-analysis method is reported. |
| Missing-data / imputation | No method is reported. |
| Stratification | No stratification factors are reported. |
| Bayesian methods | No Bayesian method is reported. |
This is an important methodological boundary. A statistical analysis page should distinguish between methods that are appropriate in general for a particular endpoint and methods that are documented as having been used in a particular trial.
16. Why This Trial Matters Statistically
HERTHENA-Lung01 is a useful teaching example because it illustrates how the statistical interpretation of a randomized oncology trial depends on the exact question being randomized and the precise endpoint definition.
| Concept | How it appears in HERTHENA-Lung01 |
|---|---|
| Randomization | The allocation is randomized. |
| Parallel-group design | The design model is parallel. |
| Open-label treatment | Masking is listed as none. |
| Binary endpoint | ORR is identified as the primary endpoint type. |
| Objective response | The primary endpoint is the proportion with confirmed CR or confirmed PR. |
| BICR | Primary ORR assessment is performed by blinded independent central review. |
| RECIST v1.1 | Response classification is defined using RECIST v1.1. |
| Confidence intervals | Would normally quantify uncertainty around an estimated response effect, although none are reported here. |
| Hypothesis testing | A formal between-arm test would normally be prespecified for a randomized binary endpoint, but the registry-reported test is not reported. |
| Analysis populations | The appropriate denominator can differ by endpoint; safety denominators reported here differ from total enrollment. |
17. Clinical Interpretation vs Statistical Interpretation
Statistical interpretation
The trial is randomized and parallel, with a binary primary endpoint assessed by BICR. The ClinicalTrials.gov record does not contain the numerical ORR result or a formal statistical comparison, so no treatment-effect estimate can be assigned from the ClinicalTrials.gov record.
Clinical interpretation
The trial compares two patritumab deruxtecan dosing approaches in subjects with metastatic or locally advanced EGFR-mutated non-small cell lung cancer. The primary clinical outcome selected by the registry is objective tumor response.
These two levels of interpretation should remain separate. The clinical question concerns response under two dosing strategies; the statistical question concerns how that response is measured, compared, and quantified with uncertainty.
18. A Deeper Look at BICR and RECIST
Confirmed complete response
The registry definition states that CR is a disappearance of all target lesions. This creates a categorical endpoint definition rather than a subjective statement that a tumor "looked better."
Confirmed partial response
The registry-reported definition states that PR requires at least a 30% decrease in the sum of diameters of target lesions based on RECIST v1.1. The threshold therefore provides a standardized rule for classifying tumor shrinkage.
Best overall response
ORR is based on the participant's best overall response during the registered assessment period. Consequently, the endpoint is not simply the response status at one arbitrary time point.
Independent central review
The use of BICR adds a centralized assessment mechanism. For statistical interpretation, this matters because the endpoint is constructed from a prespecified review process rather than being left entirely to local investigator assessment.
This simplified representation illustrates the binary structure of the primary endpoint. It does not specify the trial's missing-data or unevaluable-response rules, which are not reported here.
19. What a Complete Primary Analysis Would Report
A complete statistical presentation of the primary endpoint would normally allow the reader to answer four separate questions: how many participants responded, what proportion responded in each randomized group, how different the groups were, and how precise that difference was.
Descriptive estimate
ORR in each randomized dosing group.
Effect estimate
A prespecified between-group measure such as a risk difference, risk ratio, or another binary-outcome contrast.
Uncertainty
A confidence interval around the response estimate or treatment contrast.
Hypothesis test
A prespecified p-value if formal hypothesis testing was part of the statistical analysis plan.
None of these numerical inferential quantities is contained in the ClinicalTrials.gov record. Providing them from another source would violate the trial-data restriction for this page, so they are intentionally not presented.
20. Related Tutorials
Learn more about the methods used in this trial:
21. Related Statistical Calculators
22. Sources
- ClinicalTrials.gov: HERTHENA-Lung01, NCT04619004.
- Linked PubMed record: PubMed record for PMID 38958845.
- Linked PubMed record: PubMed record for PMID 37689979.
- Linked PubMed record: PubMed record for PMID 37212796.
The numerical content of this page is restricted to the registry-reported HERTHENA-Lung01 trial data. The linked PubMed records are provided as the publication-navigation sources associated with the trial record; no numerical results from those publications are added here.
Continue through the Clinical Biostats statistical pathway
Build from the trial's core concepts: randomization, binary endpoints, response assessment, confidence intervals, and comparisons of proportions.
23. Record Summary
HERTHENA-Lung01 is a randomized phase 2, parallel-group, unmasked trial with an enrollment of 277 and two patritumab deruxtecan dosing approaches: fixed dose and up-titration. Its single registered primary endpoint is objective response rate assessed by blinded independent central review, with ORR defined as the proportion of participants whose best overall response is a confirmed complete response or confirmed partial response according to the registered RECIST v1.1 definition.
The most important statistical feature of the ClinicalTrials.gov record is the distinction between results availability and available statistical estimates. ClinicalTrials.gov reports that results are posted and that 9 outcome measures are posted, but the ClinicalTrials.gov record contains no formal analysis records.
The ClinicalTrials.gov record reports serious adverse events in 90 of 225 participants in the 5.6 mg/kg fixed-dose group and 16 of 50 participants in the up-titration group. Those denominators are endpoint-specific safety populations in the ClinicalTrials.gov record and should not automatically be substituted for the overall enrollment of 277 when considering the efficacy endpoint.