← Clinical Trials
Solid Tumors Phase 1/2 Recruiting NCT05473156

APT: Complete Statistical Analysis of AP203 in Locally Advanced or Metastatic Solid Tumors

An independent statistical review of the phase 1/2 APT study evaluating the safety, pharmacokinetics, and clinical activity of AP203 in patients with locally advanced or metastatic solid tumors, with expansion to selected malignancies.

APT  ·  NCT05473156  ·  AP Biosciences Inc.
Registry status

This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.

1. Trial at a Glance

APT is a non-randomized, parallel, unmasked phase 1/2 study of AP203 in patients with locally advanced or metastatic solid tumors. The registry identifies dose-escalation and dose-expansion safety and clinical-activity endpoints, including dose-limiting toxicity, adverse events and serious adverse events, and investigator-assessed overall response rate.

168
Planned enrollment
ClinicalTrials.gov record
4
Arms
Parallel design
1/2
Phase
Phase 1/2
2026-12
Primary completion
Registry date
FeatureAPT
Study acronymAPT
PhasePhase 1/2
StatusRecruiting
PopulationPatients with locally advanced or metastatic solid tumors, with selected expansion populations including NSCLC, HNSCC, and ESCC
DesignNon-randomized, parallel, no masking
Primary purposeTreatment
InterventionAP203
Number of arms4
Enrollment168
Lead sponsorAP Biosciences Inc.
Sponsor typeIndustry
ClinicalTrials.govNCT05473156

2. Clinical Question

The study is designed to investigate the safety, pharmacokinetics, and clinical activity of AP203 in patients with locally advanced or metastatic solid tumors, followed by expansion to selected malignancies.

Population

Patients with locally advanced or metastatic solid tumors. The registry specifically identifies NSCLC, HNSCC, and ESCC among the conditions addressed by the study.

Intervention

AP203. The registry lists AP203 as the study drug and identifies four study arms.

Comparator

The study is registered as non-randomized. A randomized comparator is therefore not part of the registered allocation design.

Primary question

What are the dose-limiting toxicities and overall safety profile of AP203, and what clinical activity is observed during dose expansion?

3. Trial Design

01
Enrollment 168 participants
02
Dose escalation DLT evaluation
03
Safety AEs and serious AEs
04
Dose expansion Selected malignancies
05
Clinical activity Investigator-assessed ORR
Allocation
Non-randomized. Participants are not assigned to study groups through random allocation.
Design model
Parallel. The registry identifies a parallel study design with four arms.
Masking
None. The study is registered as unmasked.
Primary purpose
Treatment. The registered primary purpose is treatment.

The design has two important statistical features. First, the study combines dose-escalation and dose-expansion objectives, so not every endpoint applies to every phase in the same way. Second, the non-randomized design means that comparisons among arms, if made descriptively, do not have the same causal interpretation as comparisons generated by randomization.

Four-arm structure: the registry reports 4 arms and lists AP203 twice among the interventions. The registry information available here does not provide additional arm-specific dose or treatment details, so the statistical interpretation should remain at the level of the registered study structure and endpoints.

4. Study Timeline

2023-07-24

Study start

The registry lists July 24, 2023 as the study start date.

2026-12

Primary completion

The registry lists December 2026 as the primary completion date.

The study is registered as Recruiting. Because the registry record does not contain posted statistical analyses, the appropriate focus is the prespecified measurement framework and the statistical interpretation of the endpoints rather than an efficacy or safety conclusion.

5. Endpoints

EndpointStudy phaseRegistry definition / assessmentTime frame
Dose-Escalation: Number of Participants With Dose Limiting Toxicity (DLT) Dose-escalation DLT is defined as toxicity, meaning an adverse event at least possibly related to AP203, occurring during the DLT evaluation period. The initial 28 days and completion of the first 4 doses of treatment
Both Dose-Escalation and Dose-Expansion: Number of Participants With Adverse Events (AEs) and Serious AEs Dose-escalation and dose-expansion Safety assessed using incidence, nature, and severity of AEs and SAEs. At screening (≤ 28 days before the first dose of AP203), up to 90 days after the last dose
Dose-Expansion: Overall response rate (ORR) Dose-expansion ORR is defined as the proportion of participants with CR or PR, assessed by the Investigator according to RECIST version 1.1. At screening (≤ 28 days before the first dose of AP203), up to 90 days after the last dose

Why the endpoint definitions matter

The three registered primary endpoints represent different statistical objects. DLT is a participant-level safety event assessed over a specified early treatment window. Adverse events and serious adverse events are safety outcomes summarized by their incidence, nature, and severity over a longer observation period. ORR is a binary tumor-response outcome defined from CR and PR assessments under RECIST version 1.1.

These endpoints should therefore not be collapsed into one overall measure of treatment effect. Each answers a different question about AP203.

6. Statistical Methodology

Dose-limiting toxicity

The DLT endpoint is a count of participants who experience a qualifying toxicity during the defined DLT evaluation period. Statistically, this is naturally represented as the number of participants with DLT and the number evaluated for DLT. The corresponding proportion describes the frequency of DLT within the specified evaluation window.

Conceptual DLT proportion
DLT proportion = participants with DLT ÷ participants evaluated for DLT

The denominator is important because the endpoint is defined over a particular evaluation period rather than over indefinite follow-up.

Adverse events and serious adverse events

The registry specifies safety assessment using the incidence, nature, and severity of AEs and SAEs. Incidence is fundamentally a frequency measure: how many participants experience an event within the defined safety observation period. Nature and severity add clinical classification to that frequency.

For a non-randomized phase 1/2 study, descriptive safety summaries are especially important because the principal objective is characterization of the observed safety profile rather than estimation of a randomized treatment effect.

Overall response rate

ORR is explicitly defined as the proportion of participants with complete response (CR) or partial response (PR), assessed by the Investigator according to RECIST version 1.1.

Conceptual ORR
ORR = (number with CR + number with PR) ÷ number assessed for response

The resulting proportion summarizes the fraction of participants meeting the registry's response definition. It is not a time-to-event measure and does not by itself describe duration of response or survival.

Non-randomized analysis

Because allocation is registered as non-randomized, a difference between two arms would not automatically estimate a causal treatment effect. Arm composition can differ because participants are not balanced by randomization. This is particularly relevant when interpreting any descriptive comparison across doses, disease groups, or other study arms.

7. Planned Analysis

The registry defines the outcomes to be measured but does not post formal statistical analyses for the study. Results have not been posted on ClinicalTrials.gov.

DLT analysis

The DLT endpoint would ordinarily be summarized as the number and proportion of participants experiencing a DLT during the initial 28 days and completion of the first 4 doses of treatment. Because DLT is a binary participant-level outcome, an exact or binomial confidence interval can be used to describe uncertainty around an observed proportion when appropriate.

What the DLT analysis asks

How frequently does a qualifying toxicity occur during the prespecified DLT evaluation period?

What it does not establish

A DLT proportion does not by itself establish comparative efficacy or a causal difference between treatment groups.

AE and SAE analysis

The registry states that safety will be assessed using incidence, nature, and severity. A typical analysis would therefore present participant counts and proportions by adverse-event category, together with the relevant severity classification and distinction between AEs and serious AEs.

The registered observation period begins at screening, defined as ≤ 28 days before the first dose of AP203, and extends up to 90 days after the last dose. This time window is important when interpreting the denominator and event frequency.

ORR analysis

ORR would ordinarily be analyzed as a binary response proportion, with CR and PR counted as responses and the response denominator defined by the applicable response-assessment population. Confidence intervals can quantify uncertainty around the observed proportion.

Because the study is non-randomized, an ORR observed in one arm should not automatically be interpreted as a treatment effect relative to another arm. Differences may reflect differences in participant characteristics, disease composition, exposure, or other factors.

8. Statistical Methods Explained

Why is DLT evaluated over a defined early treatment period?

DLT is intended to capture treatment-related toxicity during a prespecified window. The APT registry defines that window as the initial 28 days and completion of the first 4 doses of treatment. Restricting the endpoint to a defined period makes the safety measure operationally consistent across participants.

What does a DLT proportion represent?

A DLT proportion is the fraction of evaluated participants who experience a qualifying DLT during the specified evaluation period. It is a frequency measure, not a hazard ratio and not a measure of relative treatment effect.

Why are AEs and SAEs reported separately from DLT?

DLT focuses on qualifying toxicity during a defined early period. The AE and SAE endpoint has a broader safety window and explicitly considers incidence, nature, and severity. The two endpoints therefore provide complementary descriptions of safety rather than interchangeable measures.

What does ORR measure?

ORR measures the proportion of participants whose best response meets the registry definition of CR or PR according to RECIST version 1.1. It is a categorical tumor-response endpoint and should not be interpreted as the percentage of participants who remain progression-free or alive.

Why does the non-randomized design matter for statistical interpretation?

Randomization balances measured and unmeasured characteristics in expectation. In a non-randomized study, participants entering different arms can differ systematically. Consequently, an observed difference between arms can describe the groups but cannot automatically be attributed to AP203 or to a particular dose without stronger assumptions and appropriate adjustment.

Why are confidence intervals useful for ORR and DLT?

A point estimate gives one observed proportion, while a confidence interval describes statistical uncertainty around that estimate. Wider intervals indicate less precision, which is especially relevant when an endpoint is evaluated in a relatively small subgroup or arm.

9. Understanding ORR Statistically

Binary response endpoint

ORR reduces tumor-response assessment to a binary outcome for analysis: a participant either meets the CR-or-PR definition or does not. This makes ORR straightforward to summarize, but the reduction also means that ORR does not capture the full longitudinal pattern of tumor response.

What ORR does not mean

An ORR of a particular percentage would not mean that the same percentage of participants were cured, survived for a specified period, or experienced a specified reduction in mortality. Those are different clinical and statistical endpoints.

RECIST v1.1 matters

The registry specifies Investigator assessment according to RECIST version 1.1. The response definition therefore depends on a structured radiologic assessment framework rather than an informal judgment that a tumor has improved.

10. Safety Analysis in a Phase 1/2 Study

Safety is central to the registered primary-endpoint framework. The study includes a specific DLT endpoint during dose escalation and a broader AE/SAE endpoint covering both dose-escalation and dose-expansion.

Safety endpointWhat is measuredStatistical perspective
DLTParticipants with qualifying toxicity during the DLT evaluation periodParticipant-level binary event and proportion
AEsIncidence, nature, and severity of adverse eventsDescriptive frequency and clinical classification
SAEsIncidence, nature, and severity of serious adverse eventsDescriptive frequency and clinical classification

The different observation windows should be retained when interpreting these measures. A DLT is specifically tied to the early DLT evaluation period, whereas AEs and SAEs are assessed from screening through up to 90 days after the last dose.

For safety data, participant-level incidence is generally more informative than simply counting the total number of events because one participant can experience more than one event. The registry's wording specifically emphasizes incidence, nature, and severity, supporting a multidimensional safety summary.

11. Dose Escalation and Dose Expansion

The endpoint structure distinguishes dose escalation from dose expansion. DLT is a dose-escalation endpoint, while ORR is a dose-expansion endpoint. AE and SAE assessment spans both phases.

Dose escalation
The registered primary DLT endpoint evaluates toxicity during the initial 28 days and completion of the first 4 doses of treatment.
Dose expansion
The registered primary clinical-activity endpoint is Investigator-assessed ORR according to RECIST version 1.1.

This separation is statistically important because the endpoints are designed for different stages of development. A DLT analysis addresses early tolerability, while an ORR analysis addresses observed antitumor activity in the expansion setting.

The registry does not provide additional details here on the exact dose levels, dose-selection algorithm, stopping boundaries, sample-size allocation by arm, or formal decision rules. Those details would determine how individual observations are translated into dose-escalation decisions and should not be inferred from the endpoint names alone.

12. Pharmacokinetics in the Study Context

The study title identifies pharmacokinetics as one of its objectives. Pharmacokinetic analysis generally concerns how drug concentrations change over time, including measures such as exposure and concentration-time behavior. However, the registered primary endpoints listed for this record are DLT, AE/SAE assessment, and ORR.

This distinction matters statistically: a study can have pharmacokinetic objectives without pharmacokinetic measures being listed among the registered primary endpoints. The primary-endpoint framework therefore should not be treated as a complete description of every exploratory measurement that may occur during the study.

13. Multiplicity and Endpoint Interpretation

The registry lists three primary endpoints spanning dose escalation, dose expansion, and both phases. These endpoints address different questions rather than representing repeated measurements of the same outcome.

EndpointPhaseStatistical role
Number of participants with DLTDose-escalationEarly toxicity characterization
Number of participants with AEs and serious AEsDose-escalation and dose-expansionBroad safety characterization
ORRDose-expansionClinical-activity characterization

Because the study is non-randomized and combines several development objectives, the statistical interpretation of these endpoints should remain endpoint-specific. A finding on ORR does not replace the safety evidence, and a safety finding does not provide evidence about tumor response.

The registry information does not specify a multiplicity-adjustment strategy for these endpoints. Therefore, the presence of several registered primary endpoints should not be interpreted as establishing a particular familywise error-control procedure.

14. Analysis Populations and Denominators

One of the most important practical issues in interpreting phase 1/2 safety and response data is the denominator. The number enrolled in the overall study is not necessarily the same as the number evaluable for every endpoint.

Overall enrollment

The registry reports enrollment of 168 participants.

DLT denominator

DLT is evaluated during a defined early treatment period, so its denominator should correspond to participants appropriately evaluated for that endpoint.

Safety denominator

AE and SAE summaries require a clearly defined safety observation population and follow-up window.

ORR denominator

ORR requires a defined response-assessment population under the RECIST v1.1 framework.

For this reason, an ORR or DLT proportion should not be interpreted simply as a percentage of the total enrolled population unless the analysis explicitly uses that denominator. Clear denominators are essential for reproducible clinical-trial reporting.

15. Missing Data and Follow-Up

The registry specifies endpoint observation windows but does not report a formal missing-data or imputation strategy in the information available here. That distinction is important because missing response assessments, incomplete DLT observation, or incomplete safety follow-up can affect the denominator and therefore the resulting proportions.

For a binary endpoint such as ORR, different approaches to unevaluable participants can produce different estimates. A rigorous statistical analysis therefore needs to define who is included in the response denominator and how participants without an evaluable response are handled.

Similarly, the DLT endpoint is tied to a defined period. Participants who do not complete the relevant observation window may require explicit rules to determine whether they are evaluable for DLT.

16. Non-Randomized Design: What Can Be Learned?

Descriptive evidence

A non-randomized phase 1/2 study can provide important information about the frequency and characteristics of toxicities and about observed tumor responses in the studied population.

Causal inference

Without randomization, differences between study arms can reflect differences in the participants entering those arms. An observed difference therefore does not automatically isolate the causal contribution of AP203 or a particular treatment level.

Precision

Even a large-looking response proportion can be statistically uncertain when based on a small number of response-evaluable participants. Confidence intervals provide the appropriate quantitative context for that uncertainty.

17. Limitations

18. Why This Trial Matters Statistically

APT illustrates an important class of clinical-trial statistics in which the central analytical problem is not simply comparing two randomized groups. The study combines early toxicity characterization, broader safety surveillance, pharmacokinetic objectives, and response assessment within a non-randomized phase 1/2 framework.

Statistical conceptHow it appears in APT
Non-randomized designAllocation is registered as non-randomized.
Parallel designThe registry identifies a parallel study model with 4 arms.
DLT analysisParticipant-level toxicity assessed during a defined early evaluation period.
Safety incidenceAEs and SAEs assessed using incidence, nature, and severity.
Binary response endpointORR is the proportion with CR or PR.
RECIST assessmentORR is assessed by the Investigator according to RECIST version 1.1.
Denominator definitionDLT, safety, and ORR each have endpoint-specific evaluation requirements.
Confidence intervalsUseful for quantifying uncertainty around observed proportions.
Phase-specific endpointsDLT is assigned to dose escalation, while ORR is assigned to dose expansion.
Safety follow-upAE and SAE assessment extends from screening through up to 90 days after the last dose.

19. Statistical Concepts in This Trial

APT provides a useful framework for understanding how statistical analysis changes when a study is designed primarily for early safety and clinical-activity characterization rather than randomized comparative efficacy.

20. Sources

Continue through Clinical Biostats

Explore statistical methods, clinical-trial analysis, and practical biostatistical workflows.

21. Record Summary

APT is a recruiting phase 1/2, non-randomized, parallel study evaluating AP203 in patients with locally advanced or metastatic solid tumors, with expansion to selected malignancies. The registry reports an enrollment of 168 participants and 4 arms, with dose-escalation DLT assessment, broader AE and SAE safety assessment, and investigator-assessed ORR according to RECIST version 1.1 as the registered primary endpoints.

Statistically, the study is best understood as a development-stage characterization study. DLT is a defined early binary toxicity endpoint; AE and SAE assessment provides broader safety characterization; and ORR is a binary response proportion. Because allocation is non-randomized, comparisons among arms require particular caution: differences can reflect the underlying composition of the participants rather than a randomized treatment effect.

Clinical Biostats methodology: The statistical story of an early-phase study depends on clearly defining the endpoint, observation window, analysis population, denominator, and uncertainty. For APT, those elements are especially important because the study combines dose escalation, safety assessment, and dose-expansion clinical activity within a non-randomized design.