Chiara Nicolò (1), Mark Lovern (2), Nathan Teuscher (2), Daniel Roeshammar (1)
(1) InSilicoTrials Technologies SpA, Riva Grumula 2, 34123, Trieste, Italy (2) InSilicoTrials Technologies LLC, 1007 N Orange St., Wilmington, Delaware 19801 , USA
Introduction: In silico clinical trials are expected to play an increasingly important role in guiding the design of oncology phase I studies aiming at determining an optimal recommended dose for subsequent Phase II trials. While Project Optimus may change dose-finding studies [1], such trials typically involve a design where successive cohorts of patients receive increasing dose levels until a maximum tolerated dose (MTD) associated with a predefined level of acceptable toxicities is reached [2].
Here we present an application of the InSilicoTrials’ cloud platform to simulate an oncology phase I dose-escalation study. The implemented trial simulator is illustrated with the modeling of published exposure-response data of neutropenia grades from a first-in-human (FIH) study of a thymidylate synthase inhibitor [3].
Methods: The cloud-based InSilicoTrials.com platform provides advanced modeling and simulation tools to perform in silico trial analyses [4]. Model templates and simulations tools are seamlessly integrated with the technical components of the platform into user-friendly online applications, allowing users to automatically set up trial scenarios, run simulations and process outcomes. The proposed platform embeds the R programming language and a NONMEM simulation engine without requiring direct access to the solvers by the user. It is built on the Microsoft Azure cloud environment, in compliance with the highest standards of security and privacy (amongst others HIPAA Privacy and Security Rules; ISO/IEC 9001, 20000, 22301, 27017, 27018 and 27001; FDA 21 CFR Part 11 (GxP); Protection Directive 95/46/EC).
We have implemented a clinical trial simulator that allows simulations of FIH studies for cytotoxic oncology products. Such studies typically employ a 3+3 dose escalation approach. This strategy involves enrolling three subjects at the first dose level and conducting a safety assessment to determine the enrollment of the next three subjects, either at the same dose (if there is a safety signal) or at the next higher dose (if there is not a safety signal). This safety assessment is repeated after each three patients are enrolled to either escalate the dose, expand the current group, or stop the trial. At the end of the study each dosing cohort will consist of either three or six patients.
Utilizing the web-based tool for dose-escalation studies simulation requires the following inputs:
- A pharmacokinetic (PK) model or a dose-exposure model
- An exposure-response model for the safety signal
- A safety signal threshold
- The sequence of dose levels to be tested
- The number of study replications
A study replication is a full simulation of the dose escalation process. After running the simulation, the following outputs are available for each replication:
- Individual subject response information in terms of exposure and safety signal output
- Summary information at each dose level (e.g., number of subjects, average exposure, and safety signal rate)
- The identified MTD
Additionally, a meta-analysis of all replicates is provided, including the following outputs:
- Summary distribution of MTD levels across replicates
- Average number of subjects in each cohort across replicates
Results: The implemented trial simulator was successfully tested with the modeling of published exposure-response data of neutropenia grades from a FIH study of a thymidylate synthase inhibitor [3]. The dose-exposure relationship was described by fitting a linear model on the log-transformed area under the concentration-time curve (AUC) data. An ordinal logistic regression model was used to relate AUC to neutropenia grades. A 3+3 design study with dose levels as those evaluated in the real clinical study [3] was simulated. Of 1000 study replicates, the majority were in agreement with the MTD established in the real dose-escalation study [3], providing evidence of the simulator’s capability at reproducing real trial scenarios.
Conclusions: This example illustrates the application of InSilicoTrials platform to perform in silico dose-finding oncology studies. The simulator integrated into the cloud-based platform enables to setup and run in silico clinical trials in a user-friendly way by using a step-by-step integrated workflow, allowing to explore alternative dose escalation schemes, as well as the impact of variability and sample size on the probability of establishing a MTD, and to compare candidate drugs with different anticipated safety profiles.
References:
[1] U.S. Food & Drug Administration, “Project Optimus.” https://www.fda.gov/about-fda/oncology-center-excellence/project-optimus
[2] C. Le Tourneau, J. J. Lee, and L. L. Siu, “Dose Escalation Methods in Phase I Cancer Clinical Trials,” JNCI J. Natl. Cancer Inst., vol. 101, no. 10, pp. 708–720, May 2009, doi: 10.1093/jnci/djp079.
[3] D. A. Rinaldi et al., “Initial phase I evaluation of the novel thymidylate synthase inhibitor, LY231514, using the modified continual reassessment method for dose escalation,” J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol., vol. 13, no. 11, pp. 2842–2850, Nov. 1995, doi: 10.1200/JCO.1995.13.11.2842.
[4] C. Nicolò et al., “Accelerating Digitalization in Healthcare with the InSilicoTrials Cloud-Based Platform: Four Use Cases,” Ann. Biomed. Eng., vol. 51, no. 1, pp. 125–136, Jan. 2023, doi: 10.1007/s10439-022-03052-6.
Reference: PAGE 32 (2024) Abstr 10932 [www.page-meeting.org/?abstract=10932]
Poster: Methodology - Study Design