Assessment of PD-L1 Expression by Digital Pathology demonstrates prognostic significance for overall survival in Triple Negative Breast Cancer and a novel association with overall survival in ER-/HER+ breast cancer.

Research output: Contribution to conferencePoster

Abstract

The clinical utility of PLD1 is evidenced by its application as a prognostic and predictive biomarker in skin, lung and bladder cancer. Several agents targeting PDL1 (atezolizumab, durvalumab, and avelumab) have been approved for use in the clinic. However, the subjective assessment of PDL1 by IHC has resulted in a lack of consensus of how best to deliver PDL1 as a robust clinical biomarker. Currently the methodology requires identification of PDL1 expression only within the epithelial component of the patient tissue, discounting expression within the inflammatory cells.
To bring a degree of consistency and reproducibility we demonstrate the usefulness of our digital pathology platform, QuPath, in the high throughput assessment of PDL1 expression in triple negative breast cancer (TNBC). Our results demonstrate, consistent with the previously reported findings, that PLD1 expression is significantly associated with clinical outcome in TNBC (95%CI 1.08-6.95) p= 0.02). Furthermore, we report a novel association of PDL1 expression in ER-/HER+ breast cancer (HR 2.45 (1.22-4.91) p= 0.009). Quantification of PDL1 protein expression with the antibody clone SP142 demonstrated significant patient stratification by outcome, employing a clinically relevant expression threshold of 1%. Additional approaches, for example, RNAScope and array-based gene expression analysis produced an association with clinical outcome yet this was not statistically significant.
Our data demonstrates accurate and reproducible biomarker quantification using our digital pathology platform. Moreover, we provide robust evidence for the application of digital pathology to the assessment of PDL1 in TNBC and ER-/HER+ breast cancer. These findings warrant the confirmation of these results, perhaps in the context of a clinical trial, which may inform treatment stratification within this subset of breast cancer, which continues to have an unmet clinical need.
Original languageEnglish
Publication statusUnpublished - 2018
EventIrish Association for Cancer Research Annual Meeting - Dublin, Ireland
Duration: 21 Feb 201824 Feb 2018

Conference

ConferenceIrish Association for Cancer Research Annual Meeting
CountryIreland
CityDublin
Period21/02/201824/02/2018

Fingerprint Dive into the research topics of 'Assessment of PD-L1 Expression by Digital Pathology demonstrates prognostic significance for overall survival in Triple Negative Breast Cancer and a novel association with overall survival in ER-/HER+ breast cancer.'. Together they form a unique fingerprint.

Cite this