Abstract 4513: Impact of patient safety regulatory considerations on early oncology clinical investigations
Bibliographic record
Abstract
Introduction: Model assisted approaches such as mTPI2 designs are widely applied in early phase oncology trials to guide dose escalation decisions. Recent health authority interactions have highlighted the importance of design parameters such as the target dose limiting toxicity (DLT) rate and the equivalence interval (EI), in situations where lower toxicity is desired. As the dose escalation strategy may vary with different parameter configurations, the shift of target DLT and EI motivated us to explore the optimal cohort size for a given design. Methods: We simulated thousands of escalation trials using the mTPI2 design with five dose levels under five common scenarios with monotonal dose toxicity relationship, maximum total N=36-48 and maximum cohort size of n=12 in each simulated trial. The target DLT was set at 25% with an EI of (20%, 33%). Using simulation, we compared the mTPI2 performance with cohort sizes of three, four, and five, evaluating: 1) Reliability: Probability of identifying the true MTD; 2) Safety: Proportion of patients experiencing DLT or overdosed (receiving MTD or higher doses); 3) Dose Escalation Efficiency (DEE): Reliability adjusted by number of patients treated; and 4) Incorrect Dosing Decisions: Proportion of decisions escalating above or de-escalating below the true MTD. Results: Preliminary results show that a cohort size of four increased the sample size by 1.4-4.9 but improved reliability by 4.6%-10.8% in absolute value compared to a cohort size of three, resulting in the highest DEE among all cohort sizes tested in all scenarios. A cohort size of five also increased reliability but marginally diminishing. A cohort size of four had the lowest likelihood of incorrect dosing decisions (an average ∼4% absolute reduction) compared to a cohort size of three in all scenarios, which indicated that cohort size four generates the shortest dose escalation/de-escalation pathway to estimating MTD that ensures the highest DEE among the three cohort sizes tested. The simulation also indicated that, generally, safety performance improved when cohort size increased. When cohort size increased from three to four, the proportion of patients experiencing DLT reduced by 1-2% and proportion of overdosed patients reduced by 2.9-5.0% both in absolute values. Conclusions: When using the mTPI2 dose escalation design in early clinical trials, we strongly recommend that trial statisticians conduct simulations to assess various configurations across all potential scenarios. This approach optimizes the dose escalation process, increasing the likelihood of identifying the true MTD and minimizing patient exposure to toxicity. Increasing cohort size can offer operational benefits, such as reducing the need to replace non-evaluable participants. Enhanced reliability may also mitigate risks in further development, at the cost of a small number of additional patients treated at early stages. Citation Format: Baisong Huang, Cindy Lu, Danny Lu, Daniel Slade, Hazel Kurz, Thomas Jahn, Sergio Vicente, Bruno Medeiros, Jayne Marshall. Impact of patient safety regulatory considerations on early oncology clinical investigations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4513.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.002 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".