Abstract
Aim: Systemic inflammatory response (SIR) has been reported to be an important determinant of disease progression and survival in patients with colorectal cancer. This study investigated the prognostic relevance of changes in the platelet count on survival and the predictive value of changes in platelet/lymphocyte ratio (PLR) and neutrophil/lymphocyte ratio (NLR) on the pathological tumor response to preoperative chemoradiotherapy (CRT) in patients with locally advanced rectal cancer (LARC). Patients and Methods: From 2006 to 2015, 291 consecutive patients with LARC who were treated with preoperative CRT followed by curative surgery at the Kyungpook National University Medical Center (Daegu, Korea) were retrospectively analyzed. A cut-off value of 370×103/μl for the platelet count was used and a PLR ≥235 was defined as high. Any change in the PLR or NLR was calculated based on subtracting the pre-CRT PLR or NLR from the post-CRT values. Results: A total of 17.5% patients had stage II and 82.5% had stage III LARC. Initially high NLR and PLR were significantly associated with poor clinical outcomes. Patients who maintained a high platelet count after CRT also had an advanced pathological stage (p=0.028), low pathological complete response rate (p=0.048), and high relapse rate (p=0.021). For patients with an initially low PLR, the multiple logistic regression analysis revealed that a high PLR change (odds ratio (OR)=2.301, 95% confidence interval (CI)=1.269-4.174; p=0.006) and clinical stage II compared to stage III (OR=1.878, 95% CI=1.231-2.865; p=0.003) were significant independent markers predictive of a good response to CRT. Conclusion: The present results suggest that platelet and PLR change after preoperative CRT, along with the initial platelet count, can be used as prognostic and predictive markers for the oncological outcomes in patients with LARC.
Preoperative chemoradiotherapy (CRT) followed by total mesorectal excision is the standard treatment for locally advanced rectal cancer (LARC). Preoperative CRT leads to clinically meaningful tumor regression, including pathological complete response (CR) in 8-30% of patients (1). Several studies have found that a good pathological response after CRT improves the clinical outcome when compared to a poor response (2). However, since the tumor response to preoperative CRT varies among patients, predictive factors need to be identified to help clinicians develop risk-adapted treatment strategies.
The systemic inflammatory response (SIR), including C-reactive protein (CRP), Glasgow prognostic score (GPS) consisting of the CRP and serum albumin, neutrophil to lymphocyte ratio (NLR), and platelet to lymphocyte ratio (PLR), has been reported as an important determinant of disease progression and survival in patients with colorectal cancer (3). An elevated NLR has also been associated with a poor tumor response and unfavorable prognosis in patients with LARC (4). Plus, the PLR, reflecting both the inflammatory and the coagulation status, has been correlated with the progression of many tumors, including colorectal cancer (5-9). Platelets are already known to contribute to tumor growth, invasion, angiogenesis, and metastasis (10, 11). Thus, changes in the platelet count following chemotherapy were recently proposed as a potential marker for treatment response (12). However, the prognostic impact of changes in the platelet count and PLR after CRT in patients with rectal cancer has not yet been fully investigated.
Accordingly, this study investigated the association of the changes in the platelet count, PLR and NLR after CRT with the clinical outcomes in patients with LARC.
Patients and Methods
Patient eligibility. A retrospective database of 291 patients with pathologically confirmed, locally advanced rectal adenocarcinoma who completed preoperative CRT and underwent radical resection between January 2006 and October 2015 at Kyungpook National University Medical Center (Daegu, Korea) was analyzed.
The initial clinical stage was evaluated using total colonoscopy, computed tomography, pelvic magnetic resonance imaging, and whole-body positron emission tomography/computed tomography according to the seventh edition of the tumor, node, and metastasis (TNM) staging system 2010 (AJCC, 2010) (13). Longitudinal measurements of the neutrophil, lymphocyte, and platelet counts and the concentrations of serum carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9) were all assessed before and after CRT. The CRT consisted of 45 Gy delivered in 25 daily fractions over 5 weeks with concurrent 5-fluorouracil (425 mg/m2) and leucovorin (20 ng/m2) on days 1-4 and 29-32. Total mesorectal excision was performed 6-8 weeks after the completion of CRT.
Definition of laboratory parameters. Blood samples were obtained from the patients on two separate occasions: during their first visit to the clinic (pre-CRT) and at 2 to 4 weeks before surgery (post-CRT). For the platelet count, a cut-off value of 370×103/μl was selected according to a previous study (14), dividing the patients into low (<370×103/μl) and high (≥370×103/μl) groups. The NLR was calculated as the neutrophil count divided by the lymphocyte count and classified into two categories as low (<5) or high (≥5). When calculating the PLR, a PLR ≥235 was defined as being high. Any change in the NLR or PLR was calculated by subtracting the pre-CRT NLR or PLR from the post-CRT values.
Statistical analysis. Relapse-free survival (RFS) was estimated from the time of surgery until the time of disease recurrence or death. Overall survival (OS) was calculated from the date of diagnosis to death from any cause. Student's t-test and a Chi-square test were both applied to the qualitative variables. A logistic regression was used to determine a predictor of the pathologic tumor response. The survival analysis used the Kaplan–Meier method with a log-rank test. A multivariate analysis was performed using variables with a value of p<0.1 in a univariate analysis using Cox's proportional hazards model to derive a potentially suitable set of predictors. Two-sided p-values of <0.05 were considered to be significant. The statistical analyses were performed using SPSS software version 18.0 (SPSS, Inc., Chicago, IL, USA).
Patient characteristics (N=291).
Results
Patient characteristics. This study included 291 patients with LARC who received preoperative CRT followed by surgery. The baseline clinical demographics and laboratory findings are summarized in Table I. The median age of the patients was 59 years (range=25-88 years) and the ratio of males to females was approximately 7:3. The clinical stages before CRT were as follows: stage II (n=51, 17.5%) and stage III (n=240, 82.5%). After CRT, 41 patients (14.1%) achieved pathologic CR and 91 patients (31.3%) experienced relapse after CRT followed by surgical resection. Twenty-eight (9.6%) patients had an initially high NLR and 59 patients (18.6%) had a high PLR.
Association between clinicopathological features and the neutrophil/lymphocyte ratio (NLR): and platelet/lymphocyte ratio (PLR).
Association between initial NLR/PLR and clinical outcomes. A comparison between the clinicopathological characteristics of each group is shown in Table II. Initially high NLR and PLR were significantly associated with poor clinical outcomes, where the high-NLR group had a more advanced pathological T stage (p=0.006) and experienced a higher relapse rate [odds ratio (OR)=2.4, 95% confidence interval (CI)=1.09-5.27; p=0.025] than the low NLR group. In addition, the patients with an initially high PLR had a lower pathologic CR rate (OR=0.273, 95% CI=0.08-0.91; p=0.026) and higher rate of recurrence (OR=2.21, 95% CI=1.23-3.98; p=0.007). Notwithstanding this, the post CRT NLR and PLR were not associated with any oncological outcomes.
Association of change in platelet count with clinical outcomes. The patients were divided into three groups according to their pre- and post-CRT platelet count using a cut-off of 370×103/μl: low pre-CRT count, regardless of the post-CRT count (group A); high pre-CRT count and low post-CRT count (group B); high pre-CRT and post-CRT count (group C) (Table III). The patients who remained in the high platelet count group after CRT (group C) had advanced pathological stage (p=0.028), low CR rate (p=0.048), and high relapse rate (p=0.021). The pathologic CR rate after CRT was 0% in group C and 16% in group A, respectively. The estimated 5-year RFS and OS were also significantly different among the three groups (Table IV). The estimated 5-year RFS for groups A, B, and C was 67.1%, 53.8%, and 30.7%, respectively (p=0.003). The estimated 5-year OS rate for group C (42.4%) was the worst compared to that for groups A and B (78.8% and 63.3%, respectively, p=0.002) (Figure 1).
Association of change in platelet count with clinical outcomes in patients divided using a cut-off of 370×103/μl: low pre-chemoradiotherapy (CRT) count, regardless of the post-CRT count (group A); high pre-CRT count and low post-CRT count (group B); high pre-CRT and post-CRT count (group C).
PLR changes and tumor response after CRT in the group with initially low PLR. In order to evaluate the relationship between the PLR changes and the tumor response after CRT, the patients with an initially low PLR (N=232) were selected for further analysis (Table V). The patients were classified using a cut-off value of 115 obtained from an SAS simulation analysis (post-CRT PLR minus pre-CRT PLR, low <115; high ≥115). The changes in the PLR were significantly associated with pathological T-stage (p=0.009), yet not associated with the pathological CR rate (p=0.644) nor the relapse rate (p=0.799). The multiple logistic regression analysis indicated that a high PLR change (OR=2.301, 95% CI 1.269-4.174; p=0.006) and good clinical TNM stage (II vs. III: OR=1.878, 95% CI=1.231-2.865; p=0.003) were significant independent predictive markers of a good response to CRT. However, the estimated 5-year RFS (65% vs. 69.9%; p=0.999) and estimated 5-year OS (85.1% vs. 74%; p=0.408) were not significantly different between the two groups (Figure 2).
Multivariate analysis of prognostic factors in patients divided using a cut-off of 370×103/μl: low pre-chemoradiotherapy (CRT) platelet count, regardless of the post-CRT count (group A); high pre-CRT count and low post-CRT count (group B); high pre-CRT and post-CRT count (group C).
The relationship between clinicopathological characteristics and platelet/lymphocyte ratio (PLR) changes after chemoradiotherapy (N=232, initially low PLR).
Kaplan–Meier cumulative relapse-free (RFS) (A) and overall (OS) (B) survival curves according to change in platelet count. Patients were divided using a platelet count cut-off of 370×103/μl: low pre-chemoradiotherapy (CRT) count, regardless of the post-CRT count (group A); high pre-CRT count and low post-CRT count (group B); high pre-CRT and post-CRT count (group C). A: 5-Year RFS for group A was 67.1% vs. group B of 53.8% vs. group C of 30.7% (p=0.003). B: 5-Year OS group A was 78.8% vs. group B of 63.3% vs. group C of 42.4% (p=0.002).
Discussion
This study investigated the prognostic impact of platelet and PLR changes after CRT in a large population of patients with LARC, and found that changes in the platelet count and PLR did have a prognostic or predictive effect on the survival of the patients.
The association between inflammation and cancer was first described by Virchow in 1863 (15). Recognized as the ‘seventh hallmark of cancer’, inflammation contributes to tumor proliferation, angiogenesis, metastasis, defeats the adaptive immune responses, and alters the response to treatment agents (16). Moreover, since the SIR is known to be associated with clinical outcome in a variety of malignancies (17), the laboratory parameters of the SIR, including the NLR and PLR, have already been investigated as prognostic or predictive markers in different cancer populations. The present study also found that an initially high NLR and PLR were poor prognostic factors for patients with LARC, which is consistent with previous studies on the prognostic value of the pre-operative assessment of the NLR and PLR in colorectal cancer (3, 18).
Kaplan–Meier cumulative relapse-free (RFS) (A) and overall (OS) (B) survival curves according to change in platelet/lymphocyte ratio (PLR) in patients with an initially low PLR (N=232). Patients were divided using a cut-off 115 into low (<115) and high (≥115) change after chemoradiotherapy (CRT). A: 5-Year RFS for those with low PLR change was 65% vs. 69.9% for those with high PLR change (p=0.999). B: 5-Year OS for those with low PLR change was 85.1% vs. 74% for those with high PLR change (p=0.408).
The current study found that changes in the platelet count and PLR after CRT were prognostic or predictive marker of the pathological tumor response to preoperative CRT. Blood platelets are a key component linking the processes of hemostasis, inflammation and tissue repair (19). Pre-clinical animal models have shown that pharmacologically- or genetically-induced thrombocytopenia and platelet function defects are associated with reduced metastasis (20). In contrast, thrombocytosis is considered as an adverse prognostic factor in many types of solid cancers (21, 22), and previous studies demonstrated that initial thrombocytosis was a negative predictive factor for the pathological response and had an adverse impact on survival in rectal cancer (14). However, there have been relatively few studies that evaluated the clinical significance of a change in the platelet count after preoperative CRT in patients with rectal cancer, even though a change in the PLR after chemotherapy has already been correlated with the chemotherapy response in patients with gastric cancer (23). In the present study, the oncological outcomes were significantly associated with a change in the platelet count after preoperative CRT. For example, the patient group which maintained a high platelet count after CRT had worse estimated 5-year RFS (53.8% vs. 30.7%) and OS (63.3% vs. 42.4%) when compared to the group whose platelet count was low after CRT. Meanwhile, the group with initially low platelet count had better survival rates (5-year RFS=67.1%, 5-year overall survival=78.8%), regardless of their post-CRT platelet count. Therefore, a lower platelet count after preoperative CRT was correlated with an improved survival rate for the patients with LARC with an initially high platelet count.
While reduced T-lymphocyte count resulting in peripheral lymphopenia after radiotherapy was first described in the 1970s (24), the clinical significance of this phenomenon has not yet been fully evaluated. In this study, a considerable change in the PLR (≥115) after preoperative CRT was identified as a significant independent predictive factor for achieving a good pathological stage in the group with initially low PLR (OR=2.301, 95% CI=1.269-4.174; p=0.006). It would seem that the patients with a favorable response to CRT had a lower lymphocyte count after CRT due to the effect of radiation. Despite the decrease in the platelet count after CRT, the lymphocyte count dropped more than the platelet count. Thus, the post-PLR values for these patients decreased more after CRT than in the poor responders. To our knowledge, this is the first study evaluating the predictive impact of PLR changes on the tumor response after preoperative CRT in patients with LARC with an initially low PLR, although many studies have focused on the association of initially high levels of inflammatory markers with clinical outcomes in LARC.
The current study has certain limitations, including its retrospective nature and results that did not include both the SIR and the local inflammatory response. The SIR and local characteristics of a tumor, including the well-investigated peritumoral infiltrate (Klintrup criteria) (25) and tumor-infiltrating lymphocytes (TILs) (26), all have an influence on disease progression. Although there were no significant differences in the CRT response according to the density of TILs after CRT, CRT can enhance the local immune response by increasing the number of TILs, and high TIL densities before treatment have already been associated with a good response to neoadjuvant CRT and a favorable prognosis (26). Accordingly, further investigation of TILs in tumor tissues is needed to elucidate the precise contribution of the host immune reaction to the response to CRT (27).
In conclusion, the present results suggest that changes in platelet count and PLR after preoperative CRT, along with the initial platelet count, can be used as prognostic and predictive markers for the oncological outcome in patients with LARC.
Acknowledgements
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (2014R1A5A2009242).
Footnotes
↵* These Authors contributed equally to this study as first authors.
Conflicts of Interest
The Authors declare that they have no conflict of interest in regard to this study.
Ethical Approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.
- Received December 28, 2016.
- Revision received February 14, 2017.
- Accepted February 21, 2017.
- Copyright© 2017, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved