The researchers collected tumor samples from 218 patients who had metastatic urothelial carcinoma. Primary tumors were collected from most of these patients (87%), with the remaining samples were collected from the metastatic sites (13%). In terms of treatment, 49.1% of patients received chemotherapy, 72.9% received immune checkpoint inhibitors (ICIs), 11.9% received antibody-drug conjugates, and 6.9% received targeted therapy. The median follow-up from the time of metastatic disease diagnosis was 21.6 months. Researchers performed tumor-only targeted DNA sequencing and RNA sequencing. Although sequencing results suggested potential treatments for 69.3% of patients, only 5% received targeted therapy based on these results. Tumors were classified into RNA expression based on different molecular subtypes. Stroma-rich subtypes exhibited high expression of extracellular matrix genes, smooth muscle genes, and genes related to epithelial-to-mesenchymal transition. Basal/squamous and stroma-rich tumors showed increased expression of immune-related genes. Subtype distribution differed between metastatic and non-metastatic tumors, with a higher proportion of stroma-rich tumors in metastatic tumors.
Genomic analysis revealed a high prevalence of non-silent variants in TP53, RB1, chromatin-modifying genes, and DNA damage repair genes. In addition, there were many hotspot variants in kinase signaling genes. The tumor immune microenvironment was profiled based on immune gene expression signatures from bulk RNA sequencing data. Consistent with prior reports, these immune signatures were elevated in basal/squamous and stroma-rich subtypes but not in luminal subtypes. The basal/squamous subtype also exhibited enrichment of genes related to inflamed T cells and IFN-gamma. The stroma-rich subtype showed enrichment of stromal infiltration signatures, which has been associated with resistance to immune checkpoint inhibitors. Additional T cell receptor (TCR) clonality analysis revealed that TCR-alpha and TCR-beta were enriched. Most immune cell types were increased in the basal subtype, while luminal tumors showed specific enrichment of plasma cells, activated mast cells, memory B cells, and activated dendritic cells. Immune cell infiltration was further characterized by staining tumor samples for CD8. Overall, 60% of tumors exhibited CD8+ cells that were restricted to the stroma (“excluded”), 39.4% exhibited CD8+ cells within the stroma and tumor (“inflamed”), and only 0.6% of the tumors lacked CD8+ cells (“desert”). Tumors with CD8+ cells within the tumor were more likely to be of the basal/squamous subtype and had significantly higher immune gene expression scores. These findings were associated with clinical response to immune checkpoint inhibitors. Patients deriving the highest benefit from this immune checkpoint blockade had tumors more likely to be categorized as “inflamed” or “excluded.”
Fifty-five percent of patients responded to chemotherapy, with those having stroma-rich or luminal subtypes exhibiting a response greater than 50%. The response rate to immune checkpoint inhibitors was 36%. On the other hand, this treatment led to clinical benefit in 65% of patients, which was associated with higher TMB, T cell “inflamed” status, and immune gene expression score. ERCC2 mutations were associated with enhanced response to chemotherapy. In addition, patients with RB1, ATM, or FANCC mutations exhibited increased survival time and clinical benefit. In an integrated elastic net model of clinical and genomic parameters with 25 predictors, poor response to treatment was associated with stroma-rich subtype, B-cell gene signature, and claudin signature. Better response to treatment was associated with TMB, M1-macrophage signature, and baseline ECOG performance score.
This study proves that DNA and RNA sequencing can be combined with clinical parameters to predict response to treatment to guide precision medicine. As noted by the authors, a small proportion of patients received precision treatments based on genomic and molecular data. The publicly available dataset provides a valuable resource for future studies.
Written by: Bishoy M. Faltas, MD, Director of Bladder Cancer Research, Englander Institute for Precision Medicine, Weill Cornell Medicine
Reference:
- Damrauer JS, Beckabir W, Klomp J, et al. Collaborative study from the Bladder Cancer Advocacy Network for the genomic analysis of metastatic urothelial cancer. Nat Commun. 2022;13(1):6658. Published 2022 Nov 4. doi:10.1038/s41467-022-33980-9