A Real-world, Comparative Study of FDA- approved Diagnostic Assays PD-L1 IHC 28-8 and 22C3 in Lung Cancer and Other Malignancies →

Between different types of diagnostic tests, different versions of a diagnostic assay, statistical agreement assessments are often needed, in order to bridge the assay performance or the clinical utility from one test to another, or to upgrade an outdated diagnostic test. The statistical methods involved in agreement assessments include Positive Percent Agreement (PPA), Negative Percent Agreement (NPA), Overall Percent Agreement (OPA) for qualitative assays, and Bland-Altman plots, Passing Bablok or Deming regression for quantitative assays, etc.

A paper that addressed a high degree of agreement between 2 FDA-approved PD-L1 diagnostic assays, 28-8 and 22C3 showed some examples of statistical agreement assessments.