The ClinicalTrials.gov record lists the study as NOT_YET_RECRUITING. The registry lists a start date of March 1, 2025, a primary completion date of July 31, 2025, and planned enrollment of 179 participants.
1. Trial at a Glance
LARC is a study of radiological markers in locally advanced rectal carcinoma. Its two registered primary endpoints are time to progression and overall survival, each stratified according to the Likert Score derived from diffusion-weighted MRI at restaging.
| Feature | LARC |
|---|---|
| Study acronym | LARC |
| Registry title | Radiological Markers in Locally Advanced Rectal Cancer |
| Condition | Locally Advanced Rectal Carcinoma |
| Planned enrollment | 179 |
| Registry status | NOT_YET_RECRUITING |
| Start date | March 1, 2025 |
| Primary completion date | July 31, 2025 |
| Lead sponsor | IRCCS San Raffaele |
| Sponsor type | OTHER |
| Primary endpoints | Time To Progression (TTP) Stratified by Likert Score; Overall Survival (OS) Stratified by Likert Score |
2. Clinical Question
The central statistical question is whether the Likert Score derived from diffusion-weighted MRI at restaging is associated with subsequent time to progression and overall survival in patients with locally advanced rectal cancer treated with total neoadjuvant therapy.
Population
Patients with locally advanced rectal carcinoma. The registry lists planned enrollment of 179 participants.
Clinical setting
The overall survival endpoint specifically describes patients with locally advanced rectal cancer treated with total neoadjuvant therapy (TNT).
Radiological marker
Patients are stratified based on the Likert Score derived from diffusion-weighted MRI (DWI) at restaging.
Primary question
Does the Likert Score provide prognostic information for progression and survival?
The registry information does not describe a randomized treatment comparison or specify intervention and comparator arms. Accordingly, the principal statistical contrast is between groups defined by the radiological marker rather than between randomized therapies.
3. Trial Design
The ClinicalTrials.gov record describes a study enrolling 179 participants and specifies two five-year time-to-event endpoints stratified by Likert Score. The available record does not specify randomized allocation, treatment arms, or a treatment-comparison framework.
4. Endpoints
| Endpoint | Registry definition | Time frame | Planned statistical approach |
|---|---|---|---|
| Time To Progression (TTP) Stratified by Likert Score | Time from diagnosis to disease progression, assessed through clinical, radiological, or pathological criteria. Patients are stratified based on the Likert Score derived from diffusion-weighted MRI at restaging. TTP is measured in months. | 5 years | Kaplan-Meier survival curves, log-rank comparison between groups, and Cox proportional hazards regression to evaluate prognostic significance of the Likert Score. |
| Overall Survival (OS) Stratified by Likert Score | Time from diagnosis to death from any cause in patients with locally advanced rectal cancer treated with total neoadjuvant therapy. Patients are stratified based on the Likert Score derived from diffusion-weighted MRI at restaging. OS is measured in months. | 5 years | Kaplan-Meier survival curves, log-rank comparison between groups, and Cox proportional hazards regression to assess prognostic value of the Likert Score. |
Time to progression
The registry defines TTP as the time from diagnosis to documented disease progression based on clinical, radiological, or pathological criteria. This time origin is important: it is not described as time from randomization or time from treatment initiation.
Overall survival
OS is defined as the time from diagnosis to death from any cause. Patients who are alive at the end of the study period or lost to follow-up are censored at their last recorded visit.
5. Statistical Methodology
Kaplan-Meier estimation
The registry specifies Kaplan-Meier survival curves for both primary endpoints. This is appropriate for estimating the probability of remaining event-free over time when some observations are right-censored.
Here, di represents the number of events at time ti, while ni is the number of individuals at risk immediately before that time.
For LARC, the resulting survival curves would describe how the probability of remaining free from progression, or remaining alive, changes over the five-year follow-up according to the Likert Score grouping.
Log-rank comparison
The registry specifies the log-rank test for comparisons between Likert Score groups. The log-rank test evaluates whether the observed event experience differs between survival curves over follow-up, giving greater weight to the accumulated pattern of events rather than focusing on a single time point.
Cox proportional hazards regression
The registry specifies Cox proportional hazards regression to evaluate the prognostic significance of the Likert Score for both TTP and OS. This produces a hazard ratio that describes the relative event rate associated with the predictor under the fitted model.
The coefficient β represents the association between the predictor and the log hazard. The corresponding hazard ratio is exp(β).
Prognostic versus treatment-effect analysis
This distinction is central to interpreting LARC. A prognostic analysis asks whether a characteristic is associated with future outcome. A randomized treatment analysis asks whether assigning one intervention rather than another changes outcome. The registered endpoints and statistical methods concern the prognostic value of the Likert Score, not a randomized treatment-effect estimate.
6. Statistical Methods Explained
Why use Kaplan-Meier curves?
Kaplan-Meier estimation is useful because TTP and OS occur at different times and some patients may not experience the event during follow-up. Rather than discarding patients without observed events, the method incorporates their available follow-up until censoring.
What does the log-rank test contribute?
The log-rank test provides a formal comparison of the survival experience across Likert Score groups. It evaluates the overall pattern of observed event times rather than simply comparing the proportion of events at the end of follow-up.
Why use Cox regression?
Cox regression allows the Likert Score to be evaluated through a hazard ratio. This provides a relative measure of the event rate associated with the prognostic marker and is complementary to the Kaplan-Meier curves and log-rank test.
What does a hazard ratio mean here?
A hazard ratio below 1 would indicate a lower estimated instantaneous event rate for the group represented in the numerator relative to the reference group, whereas a hazard ratio above 1 would indicate a higher estimated event rate. It is not a probability, a relative risk, or a percentage of patients who will experience the event.
Why does the time origin matter?
The registry defines both TTP and OS from diagnosis. Changing the time origin can change the interpretation of a survival analysis. A time measured from diagnosis answers a different question from one measured from treatment initiation, surgery, restaging, or randomization.
What does censoring mean?
Censoring means that the event has not been observed by the time the patient's available follow-up ends. For example, an individual alive at the end of the study contributes survival information through the last recorded observation without being counted as a death.
What assumption accompanies the Cox model?
The standard Cox proportional-hazards interpretation assumes that the relative hazard between groups is reasonably stable over time. If that assumption is substantially violated, a single hazard ratio can become an incomplete summary of the differences between survival curves.
7. Planned Analysis
The ClinicalTrials.gov record does not contain posted statistical analyses or outcome estimates. It specifies the endpoints and their intended statistical methods, so the planned analysis can be described without treating an unreported estimate as an observed result.
Planned TTP analysis
TTP will be measured in months from diagnosis to disease progression based on clinical, radiological, or pathological criteria. Patients will be stratified according to the Likert Score derived from DWI at restaging.
Step 1: Estimate curves
Kaplan-Meier curves will estimate progression-free probabilities over time for the Likert Score groups.
Step 2: Compare groups
The log-rank test will compare the observed time-to-progression distributions between groups.
Step 3: Model association
Cox proportional hazards regression will evaluate the prognostic significance of the Likert Score.
Step 4: Interpret censoring
Individuals without documented progression contribute follow-up information up to their applicable censoring time.
Planned OS analysis
OS will be measured in months from diagnosis to death from any cause. The registry specifies Kaplan-Meier estimation, log-rank comparison, and Cox proportional hazards regression according to Likert Score.
Kaplan-Meier
Estimate survival probabilities over the five-year time frame.
Log-rank test
Compare survival experience between Likert Score groups.
Cox regression
Assess the prognostic value of the Likert Score through a time-to-event model.
Censoring
Patients alive at the end of the study period or lost to follow-up are censored at their last recorded visit.
8. Understanding the Likert Score Analysis
The statistical structure of LARC is different from a conventional two-arm randomized clinical trial. The exposure of interest is the Likert Score derived from DWI at restaging. The principal objective is to determine whether this radiological classification is associated with subsequent clinical outcomes.
| Statistical element | Role in LARC |
|---|---|
| Predictor | Likert Score derived from diffusion-weighted MRI at restaging |
| Time origin | Diagnosis |
| TTP event | Documented disease progression based on clinical, radiological, or pathological criteria |
| OS event | Death from any cause |
| Time scale | Months |
| Follow-up | 5 years |
| Curve estimator | Kaplan-Meier |
| Group comparison | Log-rank test |
| Prognostic model | Cox proportional hazards regression |
Why a radiological marker can be studied statistically
A prognostic marker is valuable statistically when differences in the marker correspond to differences in subsequent outcome. Survival analysis is well suited to this question because the outcome is not simply whether progression or death occurred; the timing of the event is part of the endpoint.
For example, two groups could ultimately experience similar numbers of events but have very different event timing. Kaplan-Meier estimation preserves this temporal information, while the Cox model summarizes relative event rates through the hazard ratio.
Why restaging is important to interpretation
The Likert Score is derived from DWI at restaging, whereas the registered time origin for both TTP and OS is diagnosis. This creates an important temporal structure: the prognostic measurement occurs after the stated time origin. The analysis therefore needs to be interpreted in light of when the marker becomes available relative to the start of follow-up.
This timing does not invalidate the registry's planned analysis, but it is important when explaining what the resulting association represents. A marker assessed later in a patient's disease course can reflect information that was not available at diagnosis.
9. TTP: Statistical Interpretation
TTP is a time-to-event endpoint in which the event is disease progression. The registry permits progression to be established through clinical, radiological, or pathological criteria. This creates a clearly defined event process while allowing several forms of evidence to establish progression.
The Kaplan-Meier curves would describe the estimated probability of remaining without documented progression over time within the Likert Score groups. The log-rank test would provide a formal comparison of the groups, while Cox regression would quantify the association between the Likert Score and the hazard of progression.
A TTP hazard ratio would not be the probability that a patient progresses, nor would it mean that a corresponding percentage of patients are protected from progression. It would describe a relative event-rate measure from the fitted Cox model.
When a Cox hazard ratio is eventually reported, its confidence interval will quantify statistical uncertainty around the estimated association. A narrow interval indicates greater statistical precision than a wide interval, although precision is not the same thing as clinical importance.
10. OS: Statistical Interpretation
OS is defined as the time from diagnosis to death from any cause. Unlike a disease-specific mortality endpoint, death from any cause is counted as the event regardless of its underlying cause.
The Kaplan-Meier analysis would estimate survival probabilities over time within the Likert Score groups. The log-rank test would compare the survival distributions, and Cox regression would assess the prognostic value of the Likert Score.
Patients who remain alive at the end of the study period or who are lost to follow-up are censored at their last recorded visit. Such observations still contribute information before censoring; they are not treated as deaths merely because follow-up ended.
An association between Likert Score and OS would not, by itself, demonstrate that changing the radiological marker would change survival. Prognostic association and causal treatment effect are different statistical questions.
11. Censoring and Follow-Up
Right censoring is fundamental to both registered endpoints. The statistical analysis does not require every participant to be followed until progression or death, provided the assumptions underlying the censoring mechanism and survival analysis are reasonable.
| Endpoint | Event | Registry censoring information |
|---|---|---|
| TTP | Disease progression | Time is measured from diagnosis to documented progression; the registry describes the endpoint as a time-to-progression measure. |
| OS | Death from any cause | Patients alive at the end of the study period or lost to follow-up are censored at their last recorded visit. |
The validity of Kaplan-Meier and Cox analyses depends in part on the relationship between censoring and the underlying event process. If patients who are censored differ systematically from those remaining under observation in ways related to future event risk, the interpretation of the estimated survival function can become more complicated.
12. Cox Proportional-Hazards Assumption
The planned use of Cox regression introduces an important model assumption: the proportional-hazards condition. In its simplest interpretation, the hazard ratio between two groups is assumed to remain reasonably stable over time.
For a categorical Likert Score comparison, the estimated coefficient determines the hazard ratio relative to the selected reference group.
A hazard ratio is therefore a model-based summary rather than a direct description of the entire survival experience. If the curves separate early and later converge, or cross over time, a single hazard ratio may conceal important temporal features.
For that reason, the planned Kaplan-Meier curves remain important even if a Cox hazard ratio is eventually reported. The curves provide a visual description of the event experience that is not reducible to one regression coefficient.
13. Log-Rank Testing Versus Cox Regression
The registry specifies both a log-rank test and Cox proportional hazards regression. These methods are closely related but serve different reporting purposes.
| Method | Primary role | What it contributes |
|---|---|---|
| Kaplan-Meier | Descriptive estimation | Estimated survival or progression-free probability over time. |
| Log-rank test | Group comparison | Formal test of whether the survival distributions differ. |
| Cox regression | Prognostic modeling | Hazard ratio quantifying the relative event rate associated with the Likert Score. |
Using all three provides complementary information. A p-value from a log-rank test does not quantify the magnitude of an association. Conversely, a hazard ratio does not replace the need to understand the underlying survival curves.
14. Statistical Significance and Clinical Meaning
When the registry eventually contains results, statistical significance should be considered separately from the magnitude and precision of the estimated association.
Effect size
A hazard ratio describes the relative event rate under the Cox model. Its magnitude is distinct from whether a hypothesis test crosses a significance threshold.
Precision
The confidence interval describes uncertainty around the estimated association. Wider intervals indicate less precise estimation.
P-value
A p-value addresses evidence against a specified null hypothesis. It does not measure the size of the prognostic association.
Absolute outcomes
Kaplan-Meier estimates at clinically relevant time points provide an absolute perspective that a hazard ratio alone cannot provide.
15. Limitations
- No posted outcome estimates: the ClinicalTrials.gov record contains no posted statistical analyses, so there are no reported hazard ratios, confidence intervals, p-values, or observed event estimates to interpret.
- Registry status: the record lists the study as NOT_YET_RECRUITING, which is important context when interpreting the absence of posted results.
- Prognostic rather than randomized treatment comparison: the available registry information does not specify randomized treatment arms or a treatment comparator.
- Timing of the marker: the Likert Score is derived from DWI at restaging, while both primary endpoints use diagnosis as the time origin. This timing is important when interpreting the marker as a prognostic variable.
- Multiple primary endpoints: the registry specifies two primary endpoints, TTP and OS. The available record does not provide an alpha-allocation or multiplicity strategy.
- Cox model assumptions: the planned Cox analysis relies on the proportional-hazards framework. If proportional hazards do not hold, a single hazard ratio may not fully characterize the relationship between Likert Score and outcome.
- Censoring: survival estimates depend on appropriate handling and assumptions concerning patients whose outcomes are not observed during the available follow-up.
- Unreported analysis details: the available record does not specify additional modeling details such as covariate adjustment, missing-data methods, or formal sensitivity analyses.
16. What the Registry Does and Does Not Establish
The registry establishes a clear statistical framework for studying the relationship between a radiological marker and two clinically important time-to-event outcomes. It does not, in the available record, provide numerical evidence that the Likert Score is or is not associated with either endpoint.
| Question | Registry information |
|---|---|
| What is the marker? | Likert Score derived from diffusion-weighted MRI at restaging. |
| What is the TTP time origin? | Diagnosis. |
| What counts as progression? | Clinical, radiological, or pathological criteria. |
| What is the OS time origin? | Diagnosis. |
| What counts as an OS event? | Death from any cause. |
| How long are the endpoints followed? | 5 years. |
| How are survival probabilities estimated? | Kaplan-Meier curves. |
| How are groups compared? | Log-rank test. |
| How is prognostic significance assessed? | Cox proportional hazards regression. |
| Are numerical results posted? | No formal statistical analyses are posted in the ClinicalTrials.gov record. |
17. Why This Trial Matters Statistically
LARC is a useful teaching example because it connects a radiological measurement with two time-to-event outcomes while illustrating several important distinctions in survival analysis.
| Concept | How it appears in LARC |
|---|---|
| Prognostic marker | Likert Score derived from diffusion-weighted MRI at restaging. |
| Time-to-event outcome | Both primary endpoints measure the timing of clinical events. |
| Kaplan-Meier estimation | Specified for TTP and OS. |
| Log-rank testing | Specified for comparisons between Likert Score groups. |
| Cox regression | Specified to evaluate prognostic significance. |
| Censoring | Explicitly incorporated into the OS endpoint definition. |
| Time origin | Diagnosis is used for both primary endpoints. |
| Radiological assessment | DWI at restaging supplies the Likert Score. |
| Five-year follow-up | Both primary endpoints have a five-year time frame. |
From image to outcome
The conceptual chain is straightforward but statistically important:
The statistical task is to determine whether the information represented by the Likert Score is associated with subsequent differences in the time-to-event distributions. That is fundamentally different from asking whether an assigned treatment caused an improvement in outcome.
18. Reading a Future LARC Results Table
When numerical analyses are posted, the most informative presentation would combine the Kaplan-Meier estimates, log-rank result, Cox hazard ratio, confidence interval, and the number and timing of events. These quantities answer related but distinct questions.
| Reported quantity | Statistical question |
|---|---|
| Kaplan-Meier probability | What is the estimated probability of remaining event-free or alive at a given time? |
| Median event time | When does the estimated survival probability reach 50%, if it does during follow-up? |
| Hazard ratio | How does the estimated instantaneous event rate compare between groups? |
| 95% confidence interval | How precisely is the hazard ratio estimated? |
| Log-rank p-value | How much evidence is there against equality of the survival distributions under the specified test? |
| Event counts | How much observed event information underlies the time-to-event analysis? |
A statistically complete interpretation should not reduce all of these quantities to a single p-value. The relationship between the curves, the absolute survival probabilities, the hazard ratio, and its confidence interval together provide a more informative description of the evidence.
19. Statistical Interpretation vs Clinical Interpretation
Statistical interpretation
The registered analysis is designed to estimate and compare time-to-event distributions according to Likert Score and to quantify the marker's prognostic association using Cox regression.
Clinical interpretation
The clinical relevance of the marker would depend on whether differences associated with the Likert Score are sufficiently large, precise, and useful for understanding outcomes in locally advanced rectal cancer.
These are deliberately separate questions. Statistical evidence can establish an association without establishing that the marker itself causes the outcome. Conversely, a potentially useful clinical marker may require more than a statistically significant association to demonstrate practical value.
20. Data Interpretation Framework
A useful way to evaluate the eventual results is to move through the analysis in a fixed order.
Examine the Kaplan-Meier curves
Look at the direction, timing, and persistence of separation between Likert Score groups.
Review event information
Consider how many progression and death events occurred and how much censoring is present.
Read the hazard ratio
Interpret the relative event-rate estimate in relation to the reference group and the Cox model.
Read the confidence interval
Assess how precisely the prognostic association has been estimated.
Consider the clinical context
Determine whether the observed statistical association has a meaningful interpretation for patients with locally advanced rectal cancer.
21. Limitations of a Single Hazard Ratio
The planned Cox analysis is valuable because it produces a compact measure of association, but the hazard ratio should not be treated as a complete description of the outcome distribution.
If the proportional-hazards assumption is reasonable, the hazard ratio can provide a useful summary of the relative event rate. If the relative hazards vary substantially over time, however, one constant hazard ratio can obscure the temporal structure visible in the Kaplan-Meier curves.
This is why the registry's combination of Kaplan-Meier estimation, log-rank testing, and Cox regression is statistically complementary. The curve describes the time pattern, the log-rank test evaluates the group comparison, and the Cox model quantifies the relative event rate.
22. Two Primary Endpoints
LARC has two registered primary endpoints: TTP and OS. Both use the same diagnostic time origin and the same Likert Score-based prognostic framework, but they capture different clinical events.
| Feature | TTP | OS |
|---|---|---|
| Time origin | Diagnosis | Diagnosis |
| Event | Disease progression | Death from any cause |
| Time unit | Months | Months |
| Time frame | 5 years | 5 years |
| Marker grouping | Likert Score from DWI at restaging | Likert Score from DWI at restaging |
| Curve method | Kaplan-Meier | Kaplan-Meier |
| Group comparison | Log-rank | Log-rank |
| Regression | Cox proportional hazards | Cox proportional hazards |
TTP and OS should therefore not be treated as interchangeable. A radiological marker may have a different relationship with progression than with mortality, and the endpoints capture different stages of the patient's clinical course.
23. Study Timeline
Study start date
The ClinicalTrials.gov record lists March 1, 2025 as the study start date.
Primary completion date
The registry lists July 31, 2025 as the primary completion date.
Longitudinal outcome assessment
Both registered primary endpoints have a five-year time frame.
24. Sponsor and Registry Context
| Registry field | Recorded information |
|---|---|
| Study acronym | LARC |
| Official title | Radiological Markers in Locally Advanced Rectal Cancer |
| Lead sponsor | IRCCS San Raffaele |
| Sponsor type | OTHER |
| Condition | Locally Advanced Rectal Carcinoma |
| Enrollment | 179 |
| Registry status | NOT_YET_RECRUITING |
25. Sources
- ClinicalTrials.gov: NCT06843655 — Radiological Markers in Locally Advanced Rectal Cancer.
Continue through the Clinical Biostats clinical-trial library
Clinical trial analyses connect study design, endpoints, survival methods, regression, and interpretation into a practical statistical framework.
26. Record Summary
LARC is registered as a study of radiological markers in locally advanced rectal cancer with planned enrollment of 179 participants. Its two primary endpoints evaluate time to progression and overall survival, each measured from diagnosis over a 5-year time frame and stratified according to a Likert Score derived from diffusion-weighted MRI at restaging.
The registry specifies a coherent survival-analysis framework: Kaplan-Meier curves to estimate time-to-event probabilities, log-rank tests to compare groups, and Cox proportional hazards regression to evaluate the prognostic significance of the Likert Score. The OS definition explicitly incorporates censoring for patients alive at the end of the study period or lost to follow-up.
The most important statistical distinction is that this framework addresses prognostic association, not a randomized treatment effect. The eventual interpretation should therefore combine the survival curves, event information, hazard ratios, confidence intervals, and p-values rather than reducing the analysis to a single statistical threshold.