12:45 PM
Dec 22, 2017
 |  BioCentury  |  Product Development

Beyond PD-L1 assays

How Genentech envisions gene signature test that could replace PD-L1 assays

Data from Genentech Inc.’s Phase III IMpower150 trial of Tecentriq atezolizumab in first-line non-small cell lung cancer provides the company with its first hard data showing what work lies ahead before it will be ready to deploy a gene signature biomarker to supplant PD-L1 assays.

In IMpower150, the Roche unit’s T effector (Teff) gene expression panel, did not show much difference on progression-free survival (PFS) compared with its PD-L1 assay. Previous studies in different populations, however, showed the Teff signature may be able to identify a bigger group of patients who would benefit from Tecentriq than the PD-L1 assay finds, without any loss in the degree of benefit.

Going forward, Genentech thinks more exploratory work is required to refine the panel, and to identify populations in which Teff can do a better job than PD-L1 assays at predicting who will and will not respond to Tecentriq.

Currently, immunohistochemistry (IHC) assays for PD-L1 expression are the most widely used tests for selecting patients to be treated with PD-1 or PD-L1 inhibitors. Three of the four available assays use a mAb targeting PD-L1 to stain tumor cells. Genentech’s SP142 antibody stains both tumor and immune cells.

The four assays run on different systems, which can produce different background staining, and all use different cutoff points for determining PD-L1 positivity. For all four, the cutoffs tend to be either very high or very low -- such as ≥1% or ≥50% -- because pathologists can’t visually distinguish between small differences in expression.

“The reason why gene expression is so attractive is because it allows you to dial up and down the threshold as a continuous variable.”

Priti Hegde, Genentech

“Pathologists, to be frank, are human,” Genentech’s Cancer Immunotherapy Biomarker Lead Priti Hegde told BioCentury. “It is really hard for them to decide between small differences.”

A gene expression test could improve sensitivity and simplify development.

“The reason why gene expression is so attractive is because it allows you to dial up and down the threshold as a continuous variable,” she said. “You don’t need to train a pathologist on 10 different cutoffs and then test those 10 different cutoffs.”

As a result, drug developers could look at any number of thresholds retrospectively to identify which...

Read the full 1843 word article

User Sign in

Trial Subscription

Get a 4-week free trial subscription to BioCentury

Article Purchase

$150 USD
More Info >