Biomarker assays might be letting down the precision oncology team due to a lack of standardisation between tests, platforms and positivity thresholds, pathologists say.
According to a Comment article in the Lancet Oncology, the current FDA process of approving drugs with the companion diagnostics used in clinical trials isn’t always helping with consistency of biomarker assays.
The paper’s authors, led by Dr Roberto Salgardo on behalf of the International Immuno-Oncology Biomarker Working Group, said companion assays are co-developed with a drug “without regard to the other assays being developed for the same biomarker”.
“For example, PD-L1 assay kits are approved by the FDA in 15 different cancer types but the PD-L1 staining patterns, scoring methods, and positivity thresholds are different in almost all of these cancer types.”
“Moreover, the various assays and scoring systems are not equivalent, despite being matched to the same specific drug. There are at least five non-equivalent assays for PD-L1, each with its own scoring system and tumour site indications.”
The authors said the absence of assay standardisation was an emerging issue for triple-negative breast cancer.
“In the IMpassion130 trial, 46% of patients with triple-negative breast cancer were deemed to be positive using the Ventana PD-L1 (SP142) assay; when using other assays (eg, the PD-L1 IHC 22C3 pharmDx assay) in the same patients, the PD-L1 positive prevalence increased to nearly 80%,” they said.
“The cause of these inconsistencies is multifactorial and includes reproducibility issues and variable antibody and assay sensitivity, even when different assays use the same antibody.”
They said lower sensitivity of an assay, with potentially false-negative results, could lead to fewer patients receiving therapy and benefit.
“Some oncologists might prefer their pathologist to use an FDA-approved assay, despite being unaware of the analytical validity of the assay and the fact that many laboratory-developed tests can perform as well as FDA approved companion diagnostics.”
“Others might prefer an assay with a higher positive prevalence to identify more patients that can be treated.”
“Industry, regulatory agencies, governments, clinicians, and patients also need to be aware that a positive phase 3 trial does not guarantee consistency, reproducibility, and practicality of the biomarker-specific assay used in the trial.”
They recommended that the current assay approval pathway should be updated to include mandating a detailed assessment of the analytical validity of an assay before it is considered as a companion diagnostic.
“Although PD-L1 is the latest diagnostic challenge, it is neither the first nor will it be the last such challenge to face the community unless focused efforts in a partnership between all stakeholders are directed towards standardisation of assay development for both current and future applications.”