AIM PD-L1-NSCLC: Artificial intelligence-powered PD-L1 quantification for accurate prediction of tumor proportion score in diverse, multi-stain clinical tissue samples

Patients with NSCLC can benefit from treatment with PD-L1-targeting immunotherapy, and current guidelines recommend quantification of the
PD-L1 biomarker for patient tissue samples.¹ Currently, PD-L1 expression is assessed by pathologist using an approved PD-L1 immunohistochemistry assay.² However, manual assessment is challenging because the four different FDA approved different PD-L1 immunohistochemical assays have different scoring criteria.2 Additionally, pathologist inter- and intra-variability can affect scoring.³

Here, we report the development and validation of machine learning (ML) models for the quantification of PD-L1 in non-small cell lung cancer
(NSCLC) that are clone agnostic and can be incorporated into clinical trials with varied workflows.

References:

1.Bironzo P, Di Maio M. A review of guidelines for lung cancer. J Thorac
Dis. 2018;10(Suppl 13):S1556-S1563. doi:10.21037/jtd.2018.03.54
2. Kim H, Chung JH. PD-L1 Testing in Non-small Cell Lung Cancer: Past,
Present, and Future [published correction appears in J Pathol Transl
Med. 2020 Mar;54(2):196]. J Pathol Transl Med. 2019;53(4):199-206.
doi:10.4132/jptm.2019.04.24
3. Rimm DL, Han G, Taube JM, Yi ES, Bridge JA, Flieder DB, Homer R,
West WW, Wu H, Roden AC, Fujimoto J, Yu H, Anders R, Kowalewski
A, Rivard C, Rehman J, Batenchuk C, Burns V, Hirsch FR, Wistuba II. A
Prospective, Multi-institutional, Pathologist-Based Assessment of 4
Immunohistochemistry Assays for PD-L1 Expression in Non-Small Cell
Lung Cancer. JAMA Oncol. 2017 Aug 1;3(8):1051-1058