Abstract 4337: Predicting PD-L1 status in solid tumors using transcriptomic data and artificial intelligence algorithms

• Published in: Cancer research (Chicago, Ill.) Ill.), 2023-04, Vol.83 (7_Supplement), p.4337-4337
• Main Authors: Charifa, Ahmad, Lam, Alfonso, Zhang, Hong, Ip, Andrew, Pecora, Andrew, Waintraub, Stanley, Graham, Deena, McNamara, Donna, Gutierrez, Martin, Jennis, Andrew, Sharma, Ipsa, Estella, Jeffrey, Ma, Wanlong, Goy, Andre, Albitar, Maher
• Format: Article
• Language: English
• ISSN: 1538-7445; 1538-7445
• DOI: 10.1158/1538-7445.AM2023-4337

PD-L1 immunohistochemistry (IHC) is routinely used to predict the clinical response to immune checkpoint inhibitors (ICIs); however, multiple assays and antibodies have been used. This study aimed to evaluate the potential of targeted transcriptome and artificial intelligence (AI) to determine PD-L1 RNA expression levels and predict the ICI response compared to traditional IHC. RNA from 396 solid tumors samples was sequenced using next-generation sequencing (NGS) with a targeted 1408-gene panel. RNA expression and PD-L1 IHC were assessed across a broad range of PD-L1 expression levels. The geometric mean naïve Bayesian (GMNB) classifier was used to predict the PD-L1 status. PD-L1 RNA levels assessed by NGS demonstrated robust linearity across high and low expression ranges, and those assessed using NGS and IHC (Tumor cells (TC), Tumor Proportion Score (TPS) and tumor-infiltrating immune cells (ICs) were highly correlated (Tables 1). RNA sequencing provided in-depth information on the tumor microenvironment and immune response, including CD19, CD22, CD8A, CTLA4, and PD-L2 expression status. Sub-analyses showed a sustained correlation of mRNA expression with IHC (TPS and ICs) across different solid tumor types. Machine learning showed high accuracy in predicting PD-L1 status, with the area under the curve varying between 0.988 and 0.920. Targeted transcriptome sequencing combined with AI is highly useful for predicting PD-L1 status. Measuring PD-L1 mRNA expression by NGS is comparable to measuring PD-L1 expression by IHC for predicting ICI response. RNA expression has the added advantages of being amenable to standardization and avoidance of interpretation bias, along with an in-depth evaluation of the tumor microenvironment. Correlation between PD-L1 expression levels and PD-L1 IHC results IHC test results Variable Cases (N) Mean Median Range Lower Quartile Upper Quartile 10th Percentile 90th Percentile Std. Dev. TC1% CD274 90.00 14.87 10.11 0.62 - 77.53 4.60 18.62 2.95 32.35 15.75 TC10% CD274 46.00 20.81 14.03 2.90 - 77.53 8.67 26.32 4.21 51.02 17.33 IC1% CD274 129.00 9.38 5.43 0.29 - 77.53 3.21 10.31 1.74 19.05 12.13 IC10% CD274 36.00 12.50 8.52 0.29 - 77.53 4.72 13.80 4.18 31.83 13.9