RESUMO
OBJECTIVE: To evaluate the prognostic value of programmed death ligand-1 (PD-L1) and programmed death-1 (PD-1) expression in patients with upper tract urothelial carcinoma (UTUC). PATIENTS AND METHODS: A retrospective multicentre study was conducted in 283 patients with UTUC treated with radical nephroureterectomy (RNU) between 2000 and 2015 at 10 French hospitals. Immunohistochemistry analyses were performed using 2 mm-core tissue microarrays with NAT105® and 28.8® antibodies at a 5% cut-off for positivity on tumour cells and tumour-infiltrating lymphocytes to evaluate PD-L1 and PD-1 expression, respectively. Multivariable Cox regression models were used to determine the independent predictors of recurrence-free (RFS), cancer-specific (CSS) and overall survival (OS). RESULTS: Overall, 63 (22.3%) and 220 (77.7%) patients with UTUC had PD-L1-positive and -negative disease, respectively, while 91 (32.2%) and 192 (67.8%) had PD-1-positive and -negative disease, respectively. Patients who expressed PD-L1 or PD-1 were more likely to have pathological tumour stage ≥pT2 (68.3% vs 49.5%, P = 0.009; and 69.2% vs 46.4%, P < 0.001, respectively) and high-grade (90.5% vs 70.0%, P = 0.001; and 91.2% vs 66.7%, P < 0.001, respectively) disease with lymphovascular invasion (52.4% vs 17.3%, P < 0.001; and 39.6% vs 18.2%, P < 0.001, respectively) as compared to those who did not. In multivariable Cox regression analysis adjusting for each other, PD-L1 and PD-1 expression were significantly associated with decreased RFS (hazard ratio [HR] 1.83, 95% confidence interval [CI] 1.09-3.08, P = 0.023; and HR 1.59, 95% CI 1.01-2.54, P = 0.049; respectively), CSS (HR 2.73, 95% CI 1.48-5.04, P = 0.001; and HR 1.96, 95% CI 1.12-3.45, P = 0.019; respectively) and OS (HR 2.08, 95% CI 1.23-3.53, P = 0.006; and HR 1.71, 95% CI 1.05-2.78, P = 0.031; respectively). In addition, multivariable Cox regression analyses evaluating the four-tier combination of PD-L1 and PD-1 expression showed that only PD-L1/PD-1-positive patients (n = 38 [13.4%]) had significantly decreased RFS (HR 3.07, 95% CI 1.70-5.52; P < 0.001), CSS (HR 5.23, 95% CI 2.62-10.43; P < 0.001) and OS (HR 3.82, 95% CI 2.13-6.85; P < 0.001) as compared to those with PD-L1/PD-1-negative disease (n = 167 [59.0%]). CONCLUSIONS: We observed that PD-L1 and PD-1 expression were both associated with adverse pathological features that translated into an independent and cumulative adverse prognostic value in UTUC patients treated with RNU.
RESUMO
OBJECTIVE: To assess PD-L1 expression in tumor (TC) and tumor infiltrating immune cells (IC) as a predictive factor of BCG therapy failure in high-risk NMIBC. MATERIALS AND METHODS: Patients treated with complete resection followed by bladder BCG instillation for high-risk NMIBC were included. Early recurrence (ER) was defined as tumor recurrence after BCG induction course. The association between ER and immuno-histochemistry PD-L1 (E1L3N clone) expression by tumors cells (TC) and tumor infiltrating immune cells (IC) was investigated using an exact Fisher test variant. RESULTS: A total of 186 patients were included, of whom 38 (20.4%) were ER, 35 (18.8%) were positive for TC PD-L1 expression and 60 (32.3%) were positive for IC PD-L1. ER was not significantly (p = 0.97) more frequent in the TC PD-L1 ≥ 1% group (n = 7, 20.0%) than in the TC PD-L1-negative group (n = 31, 20.5%). Patients with IC PD-L1 negative had ER in 15 (19.2%) cases and patients with IC PD-L1 ≥ 1% had ER in 23 (21.3%) cases. PD-L1-positive expression for IC (threshold > 1%) was correlated with immune infiltrate density (95.2% dense immune infiltrate vs 47.2% low immune infiltrate, p < 0.05), with increased expression of PD-L1 by IC after BCG therapy (p = 0.006). CONCLUSION: No association was observed between immuno-histochemistry PD-L1 positivity and ER after BCG therapy. Nevertheless, the relationship between immune infiltrate and PD-L1 positivity confirmed the interest of assessing the immune infiltrate density to define tumor's profile.