Presentation Authors: Koichiro Ogihara*, Eiji Kikuchi, Tokyo, Japan, Takashi Okabe, Saitama, Japan, Seiya Hattori, Kanagawa, Japan, Ryo Yamashita, Shizuoka, Japan, Syunsuke Yoshimine, Suguru Shirotake, Saitama, Japan, Kazuhiro Matsumoto, Ryuichi Mizuno, Tokyo, Japan, Satoshi Hara, Kanagawa, Japan, Masafumi Oyama, Takeshi Masuda, Saitama, Japan, Masashi Niwakawa, Shizuoka, Japan, Mototsugu Oya, Tokyo, Japan
Introduction: No reliable biomarker is available for predicting worse clinical outcome in chemo-resistant urothelial carcinoma (UC) patients treated with pembrolizumab. We focused on the neutrophil-lymphocyte ratio (NLR), which is reported to be a simple index of systemic inflammation and a possible biomarker for predicting prognosis in various malignancies, including UC. Our aim was to evaluate whether a high pretreatment NLR level (pre-NLR) could predict subsequent clinical outcomes in chemo-resistant UC patients treated with pembrolizumab.
Methods: We identified 50 cases treated with pembrolizumab for chemo-resistant UC between December 2017 and August 2018 at our 5 institutions. We investigated the association between pre-NLR levels and their prognosis. We defined patients with NLR of >3.90 as the high pre-NLR group according to a calculation by receiver-operating curve analysis.
Results: The high pre-NLR group consisted of 13 cases (26.0%). In the 45 cases who had measurable lesions, at the point of maximum effect the sum of the target lesion longest diameter (SLD) was decreased in 17 cases (37.8%) compared to baseline. SLD was decreased in 2 cases (15.4%) in the high pre-NLR group, which was significantly lower than that in the low pre-NLR group (15 cases, 46.9%, p=0.048). The disease control rate defined by RECIST ver. 1.1 at best response in the high pre-NLR group was 15.4%, which was significantly lower than that in their counterpart (45.9%, p=0.049). No association was observed between pre-NLR levels and the occurrence of adverse events. The 6-month progression-free survival (PFS) rate for the high pre-NLR group was 10.3Â±9.4%, which was significantly lower than that for their counterpart (31.9Â±9.2%, p=0.001). Multivariate Cox regression analysis revealed that pre-NLR level was the only independent indicator for disease progression (p=0.004). Furthermore, the 6-month cancer-specific survival (CSS) rate for the high pre-NLR group (53.7Â±15.7%) was significantly lower than that for their counterpart (90.0Â±6.7%, p < 0.001). Multivariate analysis revealed that an elevated pre-NLR was the only independent indicator for cancer-specific death (p=0.007). The change of NLR level before and after pembrolizumab was elevated in 28 cases (56.0%). However, there was no association between clinical outcome of PFS and CSS and the change of NLR level.
Conclusions: Elevated pre-NLR could identify a population with a poor response to pembrolizumab treatment among chemo-resistant UC patients.