Background Since the approval of the first immune checkpoint inhibitor (ICI) targeting CTLA-4 in 2011 (ipilimumab), six others, targeting the PD-1/PD-L1, have been approved by FDA for a total of more than 19 indications,^6,7,8 and the number is growing. These approvals paved the way for rapid growth in the number of candidates in the pipelines. It is critical for these candidates to pursue the right development strategy to demonstrate their potential to regulatory authorities and reach patients without delay. Unexpected challenges in such a competitive field risks leading to expensive modifications and possible discontinuations. This is compounded by the lack of clarity in important development questions such as study design,⁵ the choice of endpoints and appropriate statistical methods.^1,2,3 In this regard, FDA’s guidance document⁴ provides a useful summary of the topics encountered by clinical development practitioners such as endpoints, clinical trial design and statistical analysis. However, it does not capture the unique challenges of the checkpoint inhibitor space, namely traditional phase I study designs and their ability to predict dosing and detect dose-related toxicities¹ and endpoint selection given the unconventional response patterns.₂

Methods The approval packages of the seven FDA-approved ICIs contain a wealth of information related to the focus areas, expectations and concerns of the agency. However, they run into thousands of pages, which renders manual analysis too time-consuming and/or incomplete. In this work, we use Regulatory Foresight, a proprietary AI software tool developed by Biotech Square Inc., that employs state-of-the-art techniques in Computer Vision, Natural Language Processing and Machine Learning to extract, standardize, and analyze interactions from drug and biologic applications reviewed by FDA.

Results Using Regulatory Foresight, we discovered (a) the major topics of interest and concerns of the FDA, (b) the commonalities and differences in topics between the individual ICIs, (c) the evolution of topics from the oldest to the most recently approved ICI, and (d) the unaddressed topics in official FDA guidance documents.

Conclusions This work successfully uncovers regulatory requirements in the development of immune checkpoint inhibitors using AI algorithms in order for sponsors to (a) optimize strategies for development of new drugs, (b) better understand regulatory expectations, and (c) adequately prepare for meetings and submissions to regulatory agencies. In addition this work discovers the current gaps in official FDA guidance documents so that they may be adequately addressed in future versions.

References

1. Jardim DL, de Melo Gagliato D, Kurzrock R. Lessons From the Development of the Immune Checkpoint Inhibitors in Oncology. Integr Cancer Ther 2018;17(4):1012–1015. doi:10.1177/1534735418801524

2. Ferrara R, Pilotto S, Caccese M, et al. Do immune checkpoint inhibitors need new studies methodology?. J Thorac Dis2018;10(Suppl 13):S1564–S1580. doi:10.21037/jtd.2018.01.131

3. ‘Impact Story: Determining the Clinical Benefit of Treatment Beyond Progression with Immune Checkpoint Inhibitors’ US FDA Regulatory Science Impact Story. Accessed on the 26th of July 2020: https://www.fda.gov/drugs/regulatory-science-action/impact-story-determining-clinical-benefit-treatment-beyond-progression-immune-checkpoint-inhibitors

4. ‘Clinical trial endpoints for the approval of cancer drugs and biologics’ Guidance for industry. US Department of Health and Human Services, FDA, Oncology Center of Excellence, CDER, CBER. December 2018

5. Gong, J., Chehrazi-Raffle, A., Reddi, S. et al. Development of PD-1 and PD-L1 inhibitors as a form of cancer immunotherapy: a comprehensive review of registration trials and future considerations. J. Immunotherapy cancer 2018;6:8. https://doi.org/10.1186/s40425-018-0316-z

6. Vaddepally RK, Kharel P, Pandey R, Garje R, Chandra AB. Review of Indications of FDA-Approved immune checkpoint inhibitors per NCCN guidelines with the level of evidence. Cancers (Basel) 2020;12(3):738. Published 2020 Mar 20. doi:10.3390/cancers12030738

7. PD1/PD-L1Landscape. Cancer Research Institute. Accessed on the 26th of July 2020. https://www.cancerresearch.org/scientists/immuno-oncology-landscape/pd-1-pd-l1-landscape

8. Immuno-Oncology Landscape. Cancer Research Institute. Accessed on the 26th of July 2020. https://www.cancerresearch.org/scientists/immuno-oncology-landscape