Open AccessChange in neutrophil-lymphocyte ratio after initiation of chemotherapy can be a prognostic indicator in de novo stage IV breast cancer patients.
Author(s) -
Hee Bong Shin,
Ji Sun Kim,
Hee Jeong Kim,
Jong Won Lee,
Beom Seok Ko,
Byung Ho Son,
Sei Hyun Ahn
Publication year - 2019
Publication title -
journal of global oncology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.002
H-Index - 17
ISSN - 2378-9506
DOI - 10.1200/jgo.2019.5.suppl.70
Subject(s) - medicine , hazard ratio , stage (stratigraphy) , chemotherapy , breast cancer , neutrophil to lymphocyte ratio , confidence interval , gastroenterology , multivariate analysis , cancer , oncology , lymphocyte , paleontology , biology
70 Background: The de novo stage IV breast cancer has poor prognosis, predicting response to treatment in the affected patients is difficult. We investigated whether the initial neutrophil to lymphocyte ratio (NLR) at diagnosis and NLR change after the first palliative chemotherapy cycle can be a prognostic indicators. Methods: We retrospectively reviewed 218 de novo stage IV breast cancer patients with available NLR values who underwent palliative chemotherapy as an initial treatment. We analyzed cancer specific survival (CSS) according to initial NLR (iNLR), NLR change after the first chemotherapy cycle (ΔNLR), and a combination of these two. Results: The mean patient age was 47.2 years; the median follow-up period was 29.8 months. The mean iNLR and ΔNLR values were 2.83 ± 2.19 and 0.39 ± 3.74, retrospectively, amd were used as cut off points. There was no significant difference between low and high iNLR groups (p = 0.431); however, there was a significant correlation between ΔNLR and CSS (p = 0.031). The 1-, 3-, and 5- year CSS rates of patients in the increased ΔNLR group were significantly lower than those of patients in the stationary or decreased group. (78.4%, 35.4%, 20.8% vs 88.9%, 52.6%, 27.1%; p = 0.031). Multivariate analysis suggested that ΔNLR was an independent prognostic factor (hazard ratio (HR) = 1.748, 95% confidence interval (CI) = 1.084 - 2.818). The analysis of the combination of iNLR and ΔNLR showed that patients in the high iNLR and increased ΔNLR group had poorer prognosis than those in the low iNLR and stationary or decreased ΔNLR group (HR = 4.294, 95% CI = 1.586 - 11.629). Conclusions: Initial NLR alone was not a prognostic indicator among de novo stage IV breast cancer patients. However, patients with increased NLR after palliative chemotherapy exhibited worse CSS. Patients with high initial NLR and increased NLR after treatment might be a non responder to treatment.