Ashish M Kamat1, Roger Li2, Michael A O'Donnell3, Peter C Black4, Morgan Roupret5, James W Catto6, Eva Comperat7, Molly A Ingersoll8, Wim P Witjes9, David J McConkey10, J Alfred Witjes11. 1. Department of Urology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA. Electronic address: akamat@mdanderson.org. 2. Department of Urology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA. 3. Department of Urology, University of Iowa, Iowa City, IA, USA. 4. Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada. 5. Department of Urology, Groupe Hospitalier Pitie-Salpetriere, Assistance Publique Hopitaux de Paris, Faculty of Medicine Pierre et Marie Curie, University Paris Sorbonne, Paris, France. 6. Royal Hallamshire Hospital and University of Sheffield, Sheffield, UK. 7. Department of Pathology, Tenon Hospital, AP-HP, University Pierre and Marie Curie, Paris, France. 8. Department of Immunology, Institut Pasteur and INSERM U1223, Paris, France. 9. EAU Research Foundation, EAU Central Office, Arnhem, The Netherlands. 10. The Greenberg Bladder Cancer Institute, Johns Hopkins School of Medicine, Baltimore, MD, USA. 11. Department of Urology, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands.
Abstract
CONTEXT: Bacillus Calmette-Guérin (BCG) is currently the most effective intravesical therapy for nonmuscle invasive bladder cancer, reducing not only recurrence rates but also preventing progression and reducing deaths. However, response rates to BCG vary widely and are dependent on a multitude of factors. OBJECTIVE: To review existing data on clinical, pathologic, immune, and molecular markers that allow prediction of BCG response. EVIDENCE ACQUISITION: PubMed and MEDLINE search of English language literature was conducted from its inception to July 2017 using appropriate search terms. Following systematic literature review and analysis of data, consensus voting was used to generate the content of this review. EVIDENCE SYNTHESIS: As seen in the EORTC and CUETO risk nomograms, clinicopathologic features, especially tumor stage and grade, are the most effective predictors of BCG response. Data are less robust with regards to the association of response with age, female sex, recurrent tumors, multiplicity of tumors, and the presence of carcinoma in situ. Single biomarkers, such as tumor p53 and urinary interleukin-2 expression, have had limited success in predicting BCG response, possibly due to the multifaceted nature of the generated immune response. More comprehensive biomarker panels (eg, urinary cytokines), have a more robust correlation with response, as do patterns of urinary cytologic fluorescent in-situ hybridization examination. Gene expression data correlate with disease progression, but studies examining potential associations with BCG response are limited. CONCLUSIONS: Currently, the best predictors of BCG response are clinicopathologic features such as tumor grade and stage. Panels of urinary cytokines, as well as fluorescent in-situ hybridization patterns of cytologic anomalies, appear to be promising biomarkers. The complexity of the immune response to BCG and the heterogeneity of bladder cancer suggest that future studies should amalgamate measures reflecting innate immune response and tumor/stromal gene expression before these can be adopted for clinical use. PATIENT SUMMARY: Bacillus Calmette-Guérin (BCG) immunotherapy is an effective treatment for many patients with nonmuscle invasive bladder cancer. An individual's response to BCG can be predicted by using various features of the cancer. In the future, predictive markers using molecular measures of the tumor type and the immune response to BCG may allow us to precisely know an individual's likely outcome after BCG treatment.
CONTEXT: Bacillus Calmette-Guérin (BCG) is currently the most effective intravesical therapy for nonmuscle invasive bladder cancer, reducing not only recurrence rates but also preventing progression and reducing deaths. However, response rates to BCG vary widely and are dependent on a multitude of factors. OBJECTIVE: To review existing data on clinical, pathologic, immune, and molecular markers that allow prediction of BCG response. EVIDENCE ACQUISITION: PubMed and MEDLINE search of English language literature was conducted from its inception to July 2017 using appropriate search terms. Following systematic literature review and analysis of data, consensus voting was used to generate the content of this review. EVIDENCE SYNTHESIS: As seen in the EORTC and CUETO risk nomograms, clinicopathologic features, especially tumor stage and grade, are the most effective predictors of BCG response. Data are less robust with regards to the association of response with age, female sex, recurrent tumors, multiplicity of tumors, and the presence of carcinoma in situ. Single biomarkers, such as tumor p53 and urinary interleukin-2 expression, have had limited success in predicting BCG response, possibly due to the multifaceted nature of the generated immune response. More comprehensive biomarker panels (eg, urinary cytokines), have a more robust correlation with response, as do patterns of urinary cytologic fluorescent in-situ hybridization examination. Gene expression data correlate with disease progression, but studies examining potential associations with BCG response are limited. CONCLUSIONS: Currently, the best predictors of BCG response are clinicopathologic features such as tumor grade and stage. Panels of urinary cytokines, as well as fluorescent in-situ hybridization patterns of cytologic anomalies, appear to be promising biomarkers. The complexity of the immune response to BCG and the heterogeneity of bladder cancer suggest that future studies should amalgamate measures reflecting innate immune response and tumor/stromal gene expression before these can be adopted for clinical use. PATIENT SUMMARY: Bacillus Calmette-Guérin (BCG) immunotherapy is an effective treatment for many patients with nonmuscle invasive bladder cancer. An individual's response to BCG can be predicted by using various features of the cancer. In the future, predictive markers using molecular measures of the tumor type and the immune response to BCG may allow us to precisely know an individual's likely outcome after BCG treatment.