Skip to main content

Table 4 PD-L1 (SP263) assay performance test results

From: Development of a programmed cell death ligand-1 immunohistochemical assay validated for analysis of non-small cell lung cancer and head and neck squamous cell carcinoma

Study outline

Design

NSCLC results, % (95 % CI)

Intra-day: performance within a single run on a BenchMark ULTRA instrument

10 cases

5 PD-L1 high

5 PD-L1 low/negative

n = 100

PPA 100.0 (88.6–100.0)

NPA 100.0 (83.9–100.0)

OPA 100.0 (92.9–100.0)

Inter-day: performance on the same BenchMark ULTRA instrument over 5 non-consecutive days for a minimum of 20 days

10 cases

5 PD-L1 high

5 PD-L1 low/negative

n = 50

PPA 100.0 (92.9–100.0)

NPA 96.0 (86.5–98.9)

OPA 98.0 (93.0–99.4)

Intermediate precision: Inter-antibody lot (n = 3 lots)

18 cases

9 PD-L1 high

9 PD-L1 low/negative

PPA & NPA n = 243

OPA n = 486

PPA 99.2 (97.0–99.8)

NPA 97.5 (94.7–98.9)

OPA 98.4 (96.8–99.2)

Intermediate precision: Inter-detection kit lot (n = 3 lots)

18 cases

9 PD-L1 high

9 PD-L1 low/negative

PPA & NPA n = 243

OPA n = 486

PPA 99.2 (97.0–99.8)

NPA 97.9 (95.3–99.1)

OPA 98.6 (97.1–99.3)

Intermediate precision: Intra-platform (n = 3 instruments)

18 cases

9 PD-L1 high

9 PD-L1 low/negative

PPA & NPA n = 243

OPA n = 486

PPA 99.2 (97.0–99.8)

NPA 97.5 (94.7–98.9)

OPA 98.4 (96.8–99.2)

  1. CI confidence interval, NPA negative percentage agreement, NSCLC non-small cell lung cancer, OPA overall percentage agreement, PD-L1 programmed cell death ligand-1, PPA positive percentage agreement