AUA 2018: Selection of Appropriate Genitourinary Cancers for Immunotherapy
First, Ross highlighted the role of PD-L1 immunostaining. Notably, unlike in lung cancer where such testing is a required companion diagnostic, PD-L1 immunostaining is a complementary test in genitourinary oncology. Despite a shared target, different IHC antibodies are used and differing scoring systems exist (utilizing either staining of infiltrating immunocytes or tumor cells), depending on the planned immunotherapeutic agent. Taken together, the varying antibodies, platforms, and criteria for assessment present a complex assessment process. Regardless of approach, PD-L1 is only variably associated with clinical response rates, progression-free, and overall survival. Thus, PD-L1 has not been proven to robustly predict response to immunotherapy in genitourinary cancer.
Second, as an alternative to PD-L1 immunohistochemistry, Ross described quantification of PD-L1 gene amplification. The presence of this gene amplification finding is highly correlated with high levels (>50%) of PD-L1 IHC staining. Notably, PD-L1 gene amplification may occur in tumors with both high and low tumor mutation burden. While this is a very rare finding in genitourinary malignancies, PD-L1 gene amplification is highly predictive of benefit from immunotherapy.
Third, Ross highlighted microsatellite instability as a biomarker. While MSI is common in patients with colorectal cancer, it is an uncommon finding in genitourinary cancers (1-2%). However, as pembrolizumab has received blanket approval for all patients with MSI high status regardless of tumor histology, an MSI assay may significantly affect patients’ therapeutic options. Notably, almost all patients who have MSI high also have high levels of tumor mutational burden.
Thus, Ross transitioned to the fourth potential biomarker – TMB (tumor mutational burden). While most patients with MSI high status have high TMB, the converse was not true. TMB requires a sensitive assay and a large amount of DNA to identify these mutations. Key is that driver mutations, including germline variants, SNPs, and copy alterations are not included in the quantification of TMB. Ross cited data from Dr. Rosenberg demonstrating that TMB was correlated to the response to atezolizumab in the treatment of urothelial carcinoma. Citing the registry IMMvigor210 trial, he demonstrated that patients with higher levels of TMB had statistically significantly improved overall survival following administration of atezolizumab.
Ross then moved on to biomarkers which have not reached clinical practice. First, he presented on gamma-interferon. This represents a marker of “inflamed” cancer, associated with tumor infiltrating lymphocytes, which themselves are associated with PD-L1 positivity. However, no approved or commercialized assays are available.
Finally, he highlighted single gene predictive biomarkers which, depending on the specific biomarker, may predict efficacy, resistance or hyper-progression. With respect to markers of efficacy, BRAF, MET exon 14 splice site mutation, and PBRM1 were noted. Loss of function of PBRM1 was noted to be associated with significantly improved survival in renal cell carcinoma. Unlike the prior markers highlighted, PBRM1 is common in genitourinary cancers including 41% of patients with clear cell RCC and 23% of patients with papillary RCC. While JAK1/2 were mentioned as markers of resistance, he focused on STK11. This is an immunosuppressive phenotype which confers resistance to immunotherapeutic approaches. While data are limited in genitourinary cancers, he highlighted evidence of this associated in patients with non-small cell lung cancer. Last, he highlighted MDM2 amplification as a potential harbinger of hyper-progression. While the mechanism of effect remains to be elucidated and validation is required, such a finding raises the concern of immunotherapy driving progression.
Thus, in conclusion, indications for immunotherapy continue to expand in genitourinary oncology. There are a number of emerging biomarkers which may help to guide the provision of such therapy. However, further work is necessary to validate these and better align particular therapies with individual patients.
Presented by: Jeffrey Ross, MD, Mayo Clinic
Written by: Christopher J.D. Wallis, Urology Resident, University of Toronto @WallisCJD at the 2018 AUA Annual Meeting - May 18 - 21, 2018 – San Francisco, CA USA