Tumor Biol. DOI 10.1007/s13277-016-4874-8
ORIGINAL ARTICLE
Prognostic impact of perioperative lymphocyte–monocyte ratio in patients with bladder cancer undergoing radical cystectomy Takashi Yoshida 1 & Hidefumi Kinoshita 1 & Kenji Yoshida 1 & Takao Mishima 1 & Masaaki Yanishi 1 & Hidekazu Inui 1 & Yoshihiro Komai 1 & Motohiko Sugi 1 & Takaaki Inoue 2 & Takashi Murota 2 & Katsuya Fukui 3 & Jiro Harada 4 & Gen Kawa 4 & Tadashi Matsuda 1
Received: 25 November 2015 / Accepted: 15 January 2016 # International Society of Oncology and BioMarkers (ISOBM) 2016
Abstract Various systemic inflammatory response biomarkers are associated with oncological outcome. We evaluated the superiority of prognostic predictive accuracy between neutrophil–lymphocyte ratio (NLR) and lymphocyte–monocyte ratio (LMR), and the prognostic significance of their perioperative change in patients with bladder cancer undergoing radical cystectomy (RC). We retrospectively analyzed 302 patients who had undergone RC in four institutions. Comparison of predictive accuracy between NLR and LMR was performed using receiver operating characteristic curve analysis. Overall survival (OS) and cancer-specific survival (CSS) were assessed with the Kaplan–Meier method and Cox regression analysis. Preoperative and postoperative LMR showed higher predictive accuracy for OS than NLR did (p = 0.034). Applying a cutoff of 3.41, change in perioperative LMR stratified patients into three groups (low, intermediate, and high risk), showing a significant difference in OS and CSS (p < 0.001, each), and pathological outcomes.
Electronic supplementary material The online version of this article (doi:10.1007/s13277-016-4874-8) contains supplementary material, which is available to authorized users. * Hidefumi Kinoshita [email protected]
1
Department of Urology and Andrology, Hirakata Hospital, Kansai Medical University, 2-3-1 Shin-machi, Hirakata, Osaka 573-1191, Japan
2
Department of Urology and Andrology, Takii Hospital, Kansai Medical University, Hirakata, Osaka, Japan
3
Department of Urology and Andrology, Kori Hospital, Kansai Medical University, Hirakata, Osaka, Japan
4
Department of Urology, Saiseikai Noe Hospital, Osaka, Japan
Multivariable analyses for OS and CSS showed that poor changes in LMR (high risk) were an independent prognostic factor (hazard ratio 5.70, 95 % confidence interval 3.49–9.32, p < 0.001; hazard ratio 4.53, 95 % confidence interval 2.63– 7.82, p < 0.001; respectively). Perioperative LMR is significantly associated with survival in patients with bladder cancer after RC, and it is possibly superior to NLR as a prognostic factor. Keywords Bladder cancer . Perioperative change . Lymphocyte–monocyte ratio . Neutrophil–lymphocyte ratio . Prognosis . Radical cystectomy
Introduction Radical cystectomy (RC) with pelvic lymph node dissection represents the mainstay of definitive therapy for patients with muscle-invasive and persistent non-muscle-invasive tumors. However, this aggressive therapy has a high risk of surgical complications [1], and 5-year overall survival (OS) is only 40–60 % after RC [2]. Therefore, RC with lymph node dissection as well as urinary diversion should be performed in carefully selected patients who might benefit. According to current guidelines, neoadjuvant chemotherapy is established for patients with invasive bladder cancer, whereas the indications for adjuvant chemotherapy are still under debate and advised within clinical trials [3]. When perioperative chemotherapy is recommended, estimating survival after RC plays an important role in patient counseling. Adjuvant chemotherapy is provided to patients according to surgical pa thology, such as pT stage, pN stage, lymphovascular invasion (LVI), and surgical margin [4]. To estimate more specific oncological outcome, however, novel prognostic markers are also required.
Tumor Biol.
There is accumulating evidence to support the role of the inflammatory response as an important prognostic factor in human cancer development and progression [5]. Several studies have demonstrated that elevated preoperative neutrophil– lymphocyte ratio (NLR) is related to poor prognosis in patients with several malignancies including bladder cancer [6, 7]. Despite only a few published articles, the clinical implication of preoperative lymphocyte–monocyte ratio (LMR) as a prognostic predictor has also been demonstrated in bladder cancer [8, 9]. Nevertheless, it remains uncertain which of the two markers is more related to survival of patients with bladder cancer treated with RC. Moreover, it is likely that tumor microenvironment as well as systemic inflammatory response might be changed because of surgery [10, 11]. However, only a few studies have assessed the relationship between perioperative inflammatory markers and survival in such patients. In the present study, we compared NLR and LMR, before and after RC, with regard to predictive accuracy for survival. Furthermore, we evaluated the prognostic significance of perioperative changes in NLR and LMR in patients with bladder cancer treated with RC.
Patients and methods Patients After obtaining Institutional Review Board approval, we reviewed 360 consecutive patients with bladder cancer treated with RC and lymph node dissection from January 1995 to December 2013 at our four affiliated institutions. Fifty-eight patients were excluded because of insufficient clinical data, loss to follow-up, non-bladder cancer, or persistent inflammatory diseases. Therefore, we analyzed 302 patients with data available for this retrospective study. The study population partially overlapped that described in our previous report [8]. Pathological diagnosis of urothelial carcinoma was confirmed by transurethral resection before performing RC. RC was performed in patients with muscle-invasive tumors without evidence of distant metastasis, bacille Calmette–Guérin (BCG)resistant carcinoma in situ, and recurrent multifocal superficial disease refractory to repeat transurethral resection with BCG. Histological subtype was reported according to the WHO criteria of 1973, and pathological stage was assessed by the TMN classification (seventh edition, 2010) approved by the American Joint Committee on Cancer. All clinical and pathological data were obtained from the patients’ medical records. These preoperative peripheral blood cell counts were obtained within 1 month before RC, and postoperative values were taken during early recovery time at 1–3 months after RC. No patient had significant perioperative infection or any other inflammatory disease at the time of examination. If patients received perioperative chemotherapy,
their laboratory data were obtained closest to the time of surgery after neoadjuvant chemotherapy, or just before adjuvant chemotherapy without evidence of bone marrow suppression. Postoperative follow-up comprised routine physical examination, blood evaluation, computed tomography, and cytological assessment. Patients were generally seen postoperatively every 3 months for 2 years, then biannually for 2 years, and annually thereafter. Survival data were retrieved from the patients’ medical records, and missing data were obtained by mail. Statistical analysis The end points of this study were OS and cancer-specific survival (CCS). All continuous data are shown as median values and interquartile ranges (IQRs). The NLR and LMR were calculated as the neutrophil count divided by lymphocyte count, and the lymphocyte count divided by the monocyte count, respectively. To compare the predictive accuracy of NLR and LMR for survival, we compared the area under the curve (AUC) of the receiver operating characteristic (ROC) analysis using the method of DeLong et al. [12]. Then, using either marker that had greater advantage in prognostic prediction, we analyzed the association of the perioperative changes in NLR or LMR with OS and cancer-specific survival (CSS). The best cutoff point for this marker was determined using ROC analysis, testing all possible cutoffs that could discriminate OS and CSS by log rank test [13]. According to the cutoff value, all patients were divided into four groups: HH, patients with both high preoperative and postoperative values; HL, patients with high preoperative and low postoperative values; LH, patients with low preoperative and high postoperative values; and LL, both low preoperative and postoperative values. The association of these groups with other clinicopathological variables was determined using Kruskal– Wallis tests followed by post hoc Mann–Whitney U tests with Bonferroni’s correction for multiple comparisons. Survival curves were described using the Kaplan–Meier method. Univariate and multivariate analyses were performed using a Cox proportional hazards model to determine associations between survival and potential prognostic factors. Hazard ratios (HRs) estimated from the Cox analyses are reported as relative risk with corresponding 95 % confident intervals (CIs). All statistical analyses were performed using SPSS version 21.0, and EZR version 1.29 [14]. A two-sided p value
ORIGINAL ARTICLE
Prognostic impact of perioperative lymphocyte–monocyte ratio in patients with bladder cancer undergoing radical cystectomy Takashi Yoshida 1 & Hidefumi Kinoshita 1 & Kenji Yoshida 1 & Takao Mishima 1 & Masaaki Yanishi 1 & Hidekazu Inui 1 & Yoshihiro Komai 1 & Motohiko Sugi 1 & Takaaki Inoue 2 & Takashi Murota 2 & Katsuya Fukui 3 & Jiro Harada 4 & Gen Kawa 4 & Tadashi Matsuda 1
Received: 25 November 2015 / Accepted: 15 January 2016 # International Society of Oncology and BioMarkers (ISOBM) 2016
Abstract Various systemic inflammatory response biomarkers are associated with oncological outcome. We evaluated the superiority of prognostic predictive accuracy between neutrophil–lymphocyte ratio (NLR) and lymphocyte–monocyte ratio (LMR), and the prognostic significance of their perioperative change in patients with bladder cancer undergoing radical cystectomy (RC). We retrospectively analyzed 302 patients who had undergone RC in four institutions. Comparison of predictive accuracy between NLR and LMR was performed using receiver operating characteristic curve analysis. Overall survival (OS) and cancer-specific survival (CSS) were assessed with the Kaplan–Meier method and Cox regression analysis. Preoperative and postoperative LMR showed higher predictive accuracy for OS than NLR did (p = 0.034). Applying a cutoff of 3.41, change in perioperative LMR stratified patients into three groups (low, intermediate, and high risk), showing a significant difference in OS and CSS (p < 0.001, each), and pathological outcomes.
Electronic supplementary material The online version of this article (doi:10.1007/s13277-016-4874-8) contains supplementary material, which is available to authorized users. * Hidefumi Kinoshita [email protected]
1
Department of Urology and Andrology, Hirakata Hospital, Kansai Medical University, 2-3-1 Shin-machi, Hirakata, Osaka 573-1191, Japan
2
Department of Urology and Andrology, Takii Hospital, Kansai Medical University, Hirakata, Osaka, Japan
3
Department of Urology and Andrology, Kori Hospital, Kansai Medical University, Hirakata, Osaka, Japan
4
Department of Urology, Saiseikai Noe Hospital, Osaka, Japan
Multivariable analyses for OS and CSS showed that poor changes in LMR (high risk) were an independent prognostic factor (hazard ratio 5.70, 95 % confidence interval 3.49–9.32, p < 0.001; hazard ratio 4.53, 95 % confidence interval 2.63– 7.82, p < 0.001; respectively). Perioperative LMR is significantly associated with survival in patients with bladder cancer after RC, and it is possibly superior to NLR as a prognostic factor. Keywords Bladder cancer . Perioperative change . Lymphocyte–monocyte ratio . Neutrophil–lymphocyte ratio . Prognosis . Radical cystectomy
Introduction Radical cystectomy (RC) with pelvic lymph node dissection represents the mainstay of definitive therapy for patients with muscle-invasive and persistent non-muscle-invasive tumors. However, this aggressive therapy has a high risk of surgical complications [1], and 5-year overall survival (OS) is only 40–60 % after RC [2]. Therefore, RC with lymph node dissection as well as urinary diversion should be performed in carefully selected patients who might benefit. According to current guidelines, neoadjuvant chemotherapy is established for patients with invasive bladder cancer, whereas the indications for adjuvant chemotherapy are still under debate and advised within clinical trials [3]. When perioperative chemotherapy is recommended, estimating survival after RC plays an important role in patient counseling. Adjuvant chemotherapy is provided to patients according to surgical pa thology, such as pT stage, pN stage, lymphovascular invasion (LVI), and surgical margin [4]. To estimate more specific oncological outcome, however, novel prognostic markers are also required.
Tumor Biol.
There is accumulating evidence to support the role of the inflammatory response as an important prognostic factor in human cancer development and progression [5]. Several studies have demonstrated that elevated preoperative neutrophil– lymphocyte ratio (NLR) is related to poor prognosis in patients with several malignancies including bladder cancer [6, 7]. Despite only a few published articles, the clinical implication of preoperative lymphocyte–monocyte ratio (LMR) as a prognostic predictor has also been demonstrated in bladder cancer [8, 9]. Nevertheless, it remains uncertain which of the two markers is more related to survival of patients with bladder cancer treated with RC. Moreover, it is likely that tumor microenvironment as well as systemic inflammatory response might be changed because of surgery [10, 11]. However, only a few studies have assessed the relationship between perioperative inflammatory markers and survival in such patients. In the present study, we compared NLR and LMR, before and after RC, with regard to predictive accuracy for survival. Furthermore, we evaluated the prognostic significance of perioperative changes in NLR and LMR in patients with bladder cancer treated with RC.
Patients and methods Patients After obtaining Institutional Review Board approval, we reviewed 360 consecutive patients with bladder cancer treated with RC and lymph node dissection from January 1995 to December 2013 at our four affiliated institutions. Fifty-eight patients were excluded because of insufficient clinical data, loss to follow-up, non-bladder cancer, or persistent inflammatory diseases. Therefore, we analyzed 302 patients with data available for this retrospective study. The study population partially overlapped that described in our previous report [8]. Pathological diagnosis of urothelial carcinoma was confirmed by transurethral resection before performing RC. RC was performed in patients with muscle-invasive tumors without evidence of distant metastasis, bacille Calmette–Guérin (BCG)resistant carcinoma in situ, and recurrent multifocal superficial disease refractory to repeat transurethral resection with BCG. Histological subtype was reported according to the WHO criteria of 1973, and pathological stage was assessed by the TMN classification (seventh edition, 2010) approved by the American Joint Committee on Cancer. All clinical and pathological data were obtained from the patients’ medical records. These preoperative peripheral blood cell counts were obtained within 1 month before RC, and postoperative values were taken during early recovery time at 1–3 months after RC. No patient had significant perioperative infection or any other inflammatory disease at the time of examination. If patients received perioperative chemotherapy,
their laboratory data were obtained closest to the time of surgery after neoadjuvant chemotherapy, or just before adjuvant chemotherapy without evidence of bone marrow suppression. Postoperative follow-up comprised routine physical examination, blood evaluation, computed tomography, and cytological assessment. Patients were generally seen postoperatively every 3 months for 2 years, then biannually for 2 years, and annually thereafter. Survival data were retrieved from the patients’ medical records, and missing data were obtained by mail. Statistical analysis The end points of this study were OS and cancer-specific survival (CCS). All continuous data are shown as median values and interquartile ranges (IQRs). The NLR and LMR were calculated as the neutrophil count divided by lymphocyte count, and the lymphocyte count divided by the monocyte count, respectively. To compare the predictive accuracy of NLR and LMR for survival, we compared the area under the curve (AUC) of the receiver operating characteristic (ROC) analysis using the method of DeLong et al. [12]. Then, using either marker that had greater advantage in prognostic prediction, we analyzed the association of the perioperative changes in NLR or LMR with OS and cancer-specific survival (CSS). The best cutoff point for this marker was determined using ROC analysis, testing all possible cutoffs that could discriminate OS and CSS by log rank test [13]. According to the cutoff value, all patients were divided into four groups: HH, patients with both high preoperative and postoperative values; HL, patients with high preoperative and low postoperative values; LH, patients with low preoperative and high postoperative values; and LL, both low preoperative and postoperative values. The association of these groups with other clinicopathological variables was determined using Kruskal– Wallis tests followed by post hoc Mann–Whitney U tests with Bonferroni’s correction for multiple comparisons. Survival curves were described using the Kaplan–Meier method. Univariate and multivariate analyses were performed using a Cox proportional hazards model to determine associations between survival and potential prognostic factors. Hazard ratios (HRs) estimated from the Cox analyses are reported as relative risk with corresponding 95 % confident intervals (CIs). All statistical analyses were performed using SPSS version 21.0, and EZR version 1.29 [14]. A two-sided p value
