ESMO 2022: Biomarkers for Advanced Urothelial Carcinoma: Are We There Yet?

( The 2022 ESMO annual meeting featured a urothelial carcinoma session, including a discussant presentation by Dr. Rana McKay discussing biomarkers for advanced urothelial carcinoma. Dr. McKay discussed two abstracts including “Genomic biomarkers in peripheral blood from patients enrolled in the JAVELIN Bladder 100 trial of avelumab first-line maintenance in advanced urothelial carcinoma” presented by Dr. Tom Powles and “PD-L1 expression on immune cells by SP142 co-localizes with dendritic cells and is associated with improved OS with atezolizumab in patients with untreated metastatic urothelial cancer” presented by Dr. Enrique G. Pulido. Dr. McKay started by noting that the treatment landscape of advanced urothelial carcinoma includes expanding treatments with novel mechanisms of action and drug delivery mechanisms:

JAVELIN Bladder 100-0.jpg

However, treatment selection is becoming complex and based on clinical features and biomarkers:

JAVELIN Bladder 100-1.jpg

To date, multiple trials have investigated frontline immunotherapy options with variable results, ranging from ORRs of 10% to 55%, primary progressive disease rates ranging from 11% to 43%, and ORRs in PD-L1 positive populations ranging from 28% to 57%:

JAVELIN Bladder 100-1.jpg

Treatment expansion raises questions regarding selection, whereby can biomarkers be used to guide clinical management. Dr. McKay notes that biomarkers can be prognostic (associated with an outcome irrespective of treatment) and/or predictive (associated with an outcome in the context of specific treatment). To formally establish a biomarker as predictive, there should be a comparison by treatment arm (experimental treatment versus control) and biomarker status (positive versus negative). There have been many biomarkers explored but there has been limited formal testing outside of PD-L1 expression:

JAVELIN Bladder 100-3.jpg

Dr. McKay notes that exploratory biomarker analysis work from JAVELIN Bladder 100 suggests that multiple biomarkers may identify patients who benefit from therapy.1 However, this prior work utilized immunohistochemistry and whole-exome sequencing and did not investigate epigenetic biomarkers. Epigenetic modifications play an important role in cancer, the term first coined in 1942 by embryologist Conrad Waddington who described the processes between the genotype and phenotype. In 1990, Robin Holliday, a molecular biologist, developed a more contemporary definition stating a “study of changes in gene function that do not entail change in DNA sequence.” Epigenetic modifications include DNA methylation, histone modification, and nucleosome positioning.

Chromosome conformation capture technologies are next generation techniques to allow chromatin conformation signatures to be captured. Chromatin is a highly organized complex of DNA and proteins, with multilayered chromosomal loop interactions regulating transcription. Chromosome conformation capture was first introduced in 2002 to allow capture of these chromosomal loop interactions (termed chromatin conformation signatures). In the study by Powles et al. presented at ESMO 2022, peripheral blood from 496 patients in the JAVELIN Bladder 100 trial underwent EpiSwitch analysis by PCR assay of 3D genomic templates prepared from fixed intact nuclei:

JAVELIN Bladder 100-4.jpg

Tumor immune activity was scored using a gene expression signature (JAVELIN Renal 101 Immune). Chromatin conformation signatures association with tumor JAVELIN Renal 101 Immune scores was determined via univariate and multivariate analyses of a training set of 80 specimens. This study found that chromatin conformation signature POU2F2 was positively associated with JAV-Immuno and signatures of tertiary lymphoid structures, and expression in tissues was associated with survival benefit to avelumab (with no association in the blood). Furthermore, the absence of POU2F2 in blood in patients with low tumor mutation burden was associated with a survival benefit to avelumab. Based on this data, Dr. McKay notes that there is very little published on POU2F2 and that questions remain such as (1) What is the relevance of POU2F2 and interplay with T cell immunity? (2) What is the biologic rationale for these findings? Previous work suggests that POU2F2 is a key mediator of CBX7-induced down-regulation of PD-L1 in bladder cancer cells. CBX7 proteins epigenetically repress the expression of the downstream genes and are involved in regulating cancer progression. Additionally, CBX7 repressed PD-L1 expression, which is mediated through downregulation of POU2F2:

JAVELIN Bladder 100-5.jpg

Dr. McKay notes that the strength of this study is the novel assay integrating assessment of tumor epigenetic variations. Limitations include (i) the biologic rationale of the study findings, (ii) understanding the study findings in the context of other urothelial cancer states (ie. localized disease, advanced disease pre-treatment, and dynamic changes on treatment), and (iii) validation testing and application clinically.

Moving to the study presented by Pulido et al., Dr. McKay highlighted that multiple PD-L1 assays are available, with assays differing by antibody, scoring algorithm, and cut off to determine positivity:

JAVELIN Bladder 100-6.jpg

In the work presented at ESMO 2022, the SP142 and 22C3 assays select for different patients, defining patients with PD-L1 positive tumors from IMvigor 130 differently based on the assay:

JAVELIN Bladder 100-7.jpg

This study found that the SP142 immunohistochemical assay preferentially co-localized with dendritic cells, and that PD-L1 on dendritic cells is essential for the immune response. Dendritic cells are the most efficient antigen presenting cells, are important for T-cell priming, essential for the response to PD-L1 blockade, and are upregulated by IFNy and T cells in the tumor microenvironment. SP142 scoring appeared to have better prognostic ability than 22C3 CPS scoring in patients receiving atezolizumab monotherapy:

JAVELIN Bladder 100-8.jpg

Additionally, arms B and C who were selected by the 22C3 assay had shorter median OS than those selected by the SP142 assay:

JAVELIN Bladder 100-9.jpg

Perhaps the most important result is the prognostic and predictive ability of SP142 in the cisplatin ineligible population, with high SP142 PD-L1 expression associated with better prognosis and was predictive of atezolizumab benefit:

JAVELIN Bladder 100-10.jpg

Dr. McKay highlighted several strengths of this trial, including (i) demonstrating that defining PD-L1 positive patients is variable depending on the assay utilized, and (ii) the SP142 assay was able to identify a population of patients most likely to derive benefit from atezolizumab in the frontline testing. The main limitation of this study is the limited description of other patient clinical parameters that may have contributed to the observed findings.

To conclude, Dr. McKay highlighted that data utilizing POU2F2 as a biomarker is not yet ready for integration into clinical practice, whereas utilization of the SP142 assay is ready (and is already being used) for clinical practice.

Presented by: Rana R. McKay, MD, University California – San Diego, La Jolla, CA

Written by: Zachary Klaassen, MD, MSc – Urologic Oncologist, Assistant Professor of Urology, Georgia Cancer Center, Augusta University/Medical College of Georgia, @zklaassen_md on Twitter during the 2022 European Society of Medical Oncology (ESMO) Annual Hybrid Meeting, Paris, FR, Fri, Sept 9 – Tues, Sept 13, 2022.


  1. Powles T, Sridhar SS, Loriot Y, et al. Avelumab maintenance in advanced urothelial carcinoma: Biomarker analysis of the phase 3 JAVELIN Bladder 100 trial. Nat Med. 2021 Dec;27(12):2200-2211.