Comprehensive Comparison of 22C3 and SP263 PD-L1 Expression in Non-Small-Cell Lung Cancer Using Routine Clinical and Conditioned Archives

Abstract

PD-L1 harmonization studies revealed a strong correlation between the 22C3 and SP263 assays in non-small-cell lung cancer (NSCLC). However, the assays’ characteristics have yet to be validated in a variety of clinical and analytical settings. The results of 431 NSCLC samples tested concurrently in routine clinical practice with the PD-L1 22C3 and SP263 assays were reviewed, and both assays were performed on 314 archives of surgically resected NSCLCs to assess PD-L1 expression in relation to variables such as FFPE block age and FFPE section storage condition. In routine clinical samples, 22C3 showed the highest concordance rate with 94.5% of SP263 tumor proportion score (TPS) ≥50% and 92.3% of SP263 TPS ≥1%, while SP263 showed a concordance rate with 79.6% of 22C3 TPS ≥50% and 89.9% of 22C3 TPS ≥1%. In the archival analysis, the high TPS of 22C3 and SP263 (versus TPS 1%) were significantly associated with a more recent block (<3 years versus ≥3 years) (p = 0.007 and p = 0.009, respectively). Only the TPS of 22C3 was reduced when FFPE sections were stored at room temperature compared to SP263. However, when stored at 4 °C, the storage duration had no effect on expression in either assay. For 22C3 TPS 1–49 percent and ≥50 percent (OR = 1.73, p = 0.006 and OR = 1.98, p = 0.002, respectively). There was a considerably larger chance of preserved 22C3 expression in recent room-temperature paraffin section storage, although SP263 demonstrated preserved expression in prolonged room-temperature section storage. Despite the good association between PD-L1 22C3 and SP263 in routine clinical samples, FFPE blocks older than 3 years and sections held at room temperature for more than 1 week may result in an underestimation of PD-L1 status, particularly for the 22C3 test. However, the SP263 assay was more sensitive under these conditions.