Correlation Between Pretreatment Neutrophil-to-Lymphocyte Ratio and Programmed Death-Ligand 1 Expression as Prognostic Markers in Non-Small Cell Lung Cancer

Background The neutrophil-to-lymphocyte ratio (NLR) at baseline treatment is an important marker of systemic inflammation, which is correlated with survival benefits in lung, breast, ovarian, bladder, and colorectal cancer. Programmed death-ligand 1 (PD-L1) expression is a biomarker with discording results regarding survival benefits in lung cancer. In our research, we studied the relationship between these two markers in patients with lung cancer. Methods Patients with stage I, II, III, and IV lung cancer (n = 80) were included in this retrospective study. The NLR baseline was recorded before the initiation of treatment. The NLR cut-off value was 4. PD-L1 expression was determined by immunohistochemical staining. Univariate and multivariate survival analyses were conducted to test their prognostic value. Results NLR proved to be a significant prognostic factor for progression-free survival (PFS) (p=0.002, Log Rank) with a mean PFS of 27.7 months for low NLR patients and 12.8 months for high NLR patients. It was also significant for overall survival (OS) (p=0.007, Log Rank) with a mean OS of 52 months for low NLR patients and 41.6 months for high NLR patients. The prognostic impact of PD-L1 expression on PFS and OS was not statistically significant with a mean PFS of 23.1 months for PD-L1-negative patients and 15.8 months for PD-L1-positive patients (p=0.422, Log Rank). Mean OS was 49 months for PD-L1-negative patients while for PD-L1-positive patients, it was 43.3 months (p=0.550 Log Rank). Regarding the correlation between PD-L1 expression and NLR value, PFS mean survival times were 13.1 months for PD-L1(+)/NLR>4, 15.1 months for PD-L1(-)/NLR>4, 16.4 months for PD-L1(+)/NLR<4 and 27.8 months for PD-L1(-)/NLR<4. This correlation between PFS and the combined PD-L1 and NLR prognostic factor was statistically relevant (p=0.04). For OS, the PD-L1/NLR combined prognostic factor was not statistically relevant (p=0.055). A mean PFS time of 27.8 months was reported for PD-L1(-)/NLR<4 group patients while for the other groups, the mean PFS was 14.9 months (p=0.045). In univariate analysis, the elevated NLR was significantly associated with a decreased PFS time (HR=2.31, 95% CI =1.323- 4.051, p=0.03) as well as OS (HR=3.555, 95% CI=1.310- 9.652, p=0.013). In multivariate analysis, NLR remained statistically significant for PFS (HR=2.160, 95% CI=1.148- 4.062, p=0.013) and OS (HR=4.364, 95% CI=1.474- 12.921, p=0.008) after adjusting for the factors of age, gender, tumor stage, lymph node stage, clinical stage, histology, and PD-L1 expression. PD-L1 expression was not a valid prognostic factor for progression or death in either univariate or multivariate analysis. We also stratified the disease control rate (DCR) depending on PD-L1/NLR combined factor expression. In the PD-L1(-)/NLR<4 group, we had the highest number of partial responses (PRs) and only one complete response (CR) compared to the other groups (p=0.006). Conclusions As the number of patients is limited in the present analysis, it is hypothesized that these two markers can be useful in dividing patients into two prognostic groups: the good prognostic group reunites PD-L1(+)/NLR<4 and PD-L1(-)/NLR<4 and the poor prognostic group reunites PD-L1(+)/NLR>4 and PD-L1(-)/NLR>4.