28MO - Adapting the Time-to-Event Continual Reassessment Method (TiTE-CRM) to include consolidation immunotherapy in a phase I drug-radiotherapy platform trial

Background

CONCORDE is the first phase I platform study in non-small cell lung cancer testing drug-radiotherapy combinations to identify the Recommended Phase II Dose of different DNA Damage Response inhibitors in combination with radiotherapy (RT). Dose escalation for each combination is informed by the TiTE-CRM, and a randomised calibration arm receiving RT alone aids the attribution of toxicities. Two experimental arms (PARP and ATM inhibitors) are currently recruiting, with a third arm (ATR inhibitor) opening soon. Durvalumab as consolidation immunotherapy treatment following RT will be included in future platform arms. However not all participants will receive durvalumab, for example due to progression or toxicity. How can one TiTE-CRM model accommodate both groups of participants (with and without durvalumab), with potentially different toxicity profiles?

Methods

We detail the novel design and implementation of the TiTE-CRM within CONCORDE, including the development and selected adaptations to the TiTE-CRM statistical model for CONCORDE-C (the third platform arm), and the simulation work undertaken to assess the robustness of these adaptations. Two separate populations were simulated and evaluated across various scenarios with different toxicity profiles and combined in the same adapted TiTE-CRM model. These simulations were performed to assess the model’s operating characteristics, utilising modified code from the R package ‘dfcrm’.

Results

Our simulations demonstrated that the proposed TiTE-CRM model for CONCORDE-C can answer the research question under a wide range of scenarios. However model performance decreases when the observed toxicity in participants receiving durvalumab consolidation treatment differs from our original assumptions.

Conclusions

These results show how the TiTE-CRM methodology may be used in practice in a complex drug-RT combination dose finding platform study and adapted to incorporate consolidation immunotherapy. This work also highlights the importance of a randomised calibration arm, to inform the attribution of toxicities and toxicity assumptions in this population.

Clinical trial identification

EudraCT 2020-000206-28, ISRCTN: 10142971.

Editorial acknowledgement

N/A

Legal entity responsible for the study

University of Leeds.

Funding

AstraZeneca and Cancer Research UK.

Disclosure

C. Faivre-Finn: Financial Interests, Personal, Advisory Role: AstraZeneca; Financial Interests, Personal, Speaker’s Bureau: AstraZeneca; Financial Interests, Institutional, Other, Support for attending meetings and/or travel: AstraZeneca, Elekta; Non-Financial Interests, Personal, Advisory Board: AstraZeneca. A. Greystoke: Financial Interests, Personal, Advisory Board: BMS, AstraZeneca, Boehringer Ingelheim, Takeda, Roche, Novartis; Financial Interests, Personal, Invited Speaker: Merck, MSD, Pfizer, Janssen; Financial Interests, Institutional, Invited Speaker: MSD, Pfizer, AstraZeneca, Novartis, Achilles; Non-Financial Interests, Personal, Advisory Role: National Institue for Health and Clinical Excellence; Other, Personal, Other, Clinical Lead for Cancer (paid position): North East Englad and Yorkshire Genomic Laboratory Hub. P.H. Shaw: Financial Interests, Institutional, Other, Educational support: Takeda; Financial Interests, Personal, Other, Honoraria: Takeda, BMS, AstraZeneca. C. Hiley: Financial Interests, Personal, Member of the Board of Directors, UK Clinical Director of Lung Cancer: GenesisCare. A. Chalmers: Financial Interests, Institutional, Research Grant: AstraZeneca; Financial Interests, Personal, Other, Honoraria: AstraZeneca. All other authors have declared no conflicts of interest.