Immunological Correlates of Response to Immune Checkpoint Inhibitors in Metastatic Urothelial Carcinoma.

BACKGROUND: The identification of prognostic and/or predictive biomarkers for response to immune checkpoint inhibitors (ICI) could help guide treatment decisions. OBJECTIVE: We assessed changes in programmed cell death-1 (PD1)/PD1 ligand (PDL1) expression in key immunomodulatory cell subsets (myeloid-derived suppressor cells [MDSC]; cytotoxic T lymphocytes [CTL]) following ICI therapy and investigated whether these changes correlated with outcomes in patients with metastatic urothelial carcinoma (mUC). PATIENTS AND METHODS: Serial peripheral blood samples were collected from ICI-treated mUC patients. Flow cytometry was used to quantify PD1/PDL1 expression on MDSC (CD33+HLADR-) and CTL (CD8+CD4-) from peripheral blood mononuclear cells. MDSC were grouped into monocytic (M)-MDSC (CD14+CD15-), polymorphonuclear (PMN)-MDSC (CD14-CD15+), and immature (I)-MDSC (CD14-CD15-). Mixed-model regression and Wilcoxon signed-rank or rank-sum tests were performed to assess post-ICI changes in immune biomarker expression and identify correlations between PD1/PDL1 expression and objective response to ICI. RESULTS: Of 41 ICI-treated patients, 26 received anti-PDL1 (23 atezolizumab/3 avelumab) and 15 received anti-PD1 (pembrolizumab) therapy. Based on available data, 27.5% had prior intravesical Bacillus Calmette-Guérin therapy, 42% had prior neoadjuvant chemotherapy, and 70% had prior cystectomy or nephroureterectomy. Successive doses of anti-PDL1 correlated with decreased percentage of PDL1+ (%PDL1+) M-MDSC, while doses of anti-PD1 correlated with decreased %PD1+ M- and I-MDSC. Although pre-treatment %PD1+ CTL did not predict response, a greater %PD1+ CTL within 9 weeks after ICI initiation correlated with objective response. CONCLUSIONS: Treatment with ICI correlated with distinct changes in PD1/PDL1-expressing peripheral immune cell subsets, which may predict objective response to ICI. Further studies are required to validate immune molecular expression as a prognostic and/or predictive biomarker for long-term outcomes in mUC.

Myeloid-derived suppressors cells (MDSC) correlate with clinicopathologic factors and pathologic complete response (pCR) in patients with urothelial carcinoma (UC) undergoing cystectomy.

BACKGROUND: Myeloid derived suppressor cells (MDSC) are heterogeneous immunosuppressive cells with potential predictive and prognostic roles in cancer. The association between MDSC, clinicopathologic factors, and pathologic response in patients with bladder urothelial carcinoma (UC) was explored. METHODS: Peripheral blood or tissue were collected from patients with UC undergoing definitive surgery. MDSCs levels were measured in peripheral blood mononuclear cells and fresh tumor tissue. MDSCs were identified by flow cytometry and defined as total MDSC (T-MDSC) CD33+/HLADR-. From this population, 3 subsets were identified: polymorphonuclear-MDSC (PMN-MDSC) defined as CD33+/HLADR-/CD15+/CD14-, monocytic-MDSC (M-MDSC) defined as CD33+/HLADR-/CD15-/CD14+, and immature-MDSC (I-MDSC) defined as CD33+/HLADR-/CD15-/CD14-. MDSC populations were presented as % of live nucleated blood cells. Spearman correlations (r) and Wilcoxon rank sum test were used to assess correlations between MDSC populations, clinicopathologic factors, and pathologic complete response (pCR). RESULTS: 85 patients scheduled to undergo cystectomy from February 2015 through Dec 2016 were included. All patients had blood drawn for analysis and 23 patients had residual tumor tissue collected for analysis at the time of surgery. Of these 85, 74 (87%) were men with a median age at diagnosis of 68 (range: 44-87). Pure UC was the most common histology (75%); 28 (35%) patients had prior treatment with intravesical therapy and 36 (42%) were treated with neoadjuvant chemotherapy, primarily gemcitabine plus cisplatin (n = 24). On surgical pathology, 18 (21%) of the patients had pCR, 11 (13%) had positive lymph nodes, and 20 patients (24%) had lymphovascular invasion. Statistically significant associations were found between circulating MDSC levels and pCR rates (P<0.01), absolute neutrophil-lymphocyte ratio (P = 0.008), and histology (P = 0.01). Tumor % M-MDSCs were negatively associated with lymphovascular invasion (P = 0.04). There were no significant correlations between peripheral blood mononuclear cells and tumor MDSC subtypes. CONCLUSIONS: Blood and tissue MDSC levels correlate with several clinicopathologic factors and may predict for pCR. Future studies are needed to highlight the role of MDSC in predicting long-term outcomes and to determine the clinical implications of these findings.