Preoperative neutrophil to lymphocyte ratio >5 is a prognostic factor for recurrent colorectal cancer

Authors


  • Presented at the Association of Surgeons in Training (ASiT) Annual Conference, Sheffield, April 2011; Abstract published in Br J Surg 2011; 98: 1–80. Also presented at the ASGBI Annual Conference, Bournemouth, May 2011.

Miss Sreelakshmi Mallappa, Clinical Research Fellow, The Polyposis Registry, St Mark’s Hospital, Watford Road, Harrow HA1 3UJ, UK.
E-mail: s.mallappa@imperial.ac.uk

Abstract

Aim  Previous studies have demonstrated that raised preoperative neutrophil to lymphocyte ratio (NLR) is associated with poor prognosis in colorectal cancer (CRC). The aim of this study was to assess whether preoperative NLR could predict patients at risk of recurrence of CRC.

Method  All consecutive patients who underwent surgical resection for CRC over a 2-year period at our institution were analysed. Demographic data including CRC recurrence were prospectively collected from our institutional cancer database. CRC recurrence was diagnosed on radiological and endoscopic histopathological data. Preoperative NLR was calculated on baseline blood results, with a value >5 being a poor prognostic factor. Parametric survival analysis was used to identify risk factors for CRC recurrence. Hazard ratios (HRs) were calculated for gender, CRC stage using Jass score, preoperative NLR and CRC site. < 0.05 was considered statistically significant.

Results  In all, 297 patients (157 men) underwent CRC resection at a median age of 70 years (range 23–93); 164 patients had colon cancer, 111 rectal cancer and 22 recto-sigmoid cancer. The distribution by stage of CRC was 30.2% for stage 1, 23.8% for stage 2, 19.5% for stage 3 and 26.5% for stage 4. Over a median follow-up period of 3.35 (0.1–8) years, 59 (19.8%) patients had recurrent CRC. Multivariate analysis revealed CRC stage (HR 8.69, 95% CI 3.85–19.6, < 0.0001) and NLR >5 (HR 1.81, 95% CI 1.07–3.07, = 0.028) to be significant and independent risk factors predictive of recurrent CRC.

Conclusion  These data suggest that preoperative NLR >5 is predictive of CRC recurrence.

What is new in this paper?

At present, there is a relative paucity of data regarding the influence of systemic inflammatory markers on disease-free survival in patients undergoing elective, potentially curative surgical resection for colorectal cancer. We aimed to examine the role of preoperative neutrophil to lymphocyte ratio in predicting recurrence in this patient group.

Introduction

Colorectal cancer (CRC) is the second most common cause of cancer-related death in the UK [1]. The link between cancer and inflammation was first proposed by Virchow in 1863 when he noted the presence of leucocytes in neoplastic tissues [2,3]. There is increasing evidence that a complex interaction between the local characteristics of the tumour and the host systemic inflammatory response determines disease progression in cancer [4–6].

It has been postulated that markers of systemic inflammation may provide useful information for prognosis. Various markers of systemic inflammatory response including cytokines, C-reactive protein (CRP), Glasgow Prognostic Score, modified Glasgow Prognostic Score, neutrophil to lymphocyte ratio (NLR) and platelet to lymphocyte ratio have a proven prognostic role in various common solid tumours such as gastric cancer, non-small-cell lung cancer, ovarian cancer, intrahepatic cholangiocarcinoma, hepatocellular carcinoma, pancreatic cancer, CRC, nasopharyngeal cancer and malignant melanoma [7–11].

The value of systemic inflammatory response has been most extensively examined in CRC [7,9,11–22]. An elevated CRP level in response to the systemic inflammation has particularly been noted to be an important aetiological factor in the nutritional and functional decline of the patient with advanced cancer [6]. Recent definitions of cancer cachexia include elevation of CRP and hypoalbuminaemia [23,24]. The NLR has been shown to be associated with progression and metastasis of gastric cancer, non-small-cell lung cancer, ovarian cancer, intrahepatic cholangiocarcinoma, hepatocellular carcinoma, pancreatic cancer, CRC and nasopharyngeal cancer [25–32]. At present, there is little information on the influence of these prognostic markers on disease-free survival in patients undergoing elective, potentially curative surgical resection for CRC. The aim of the present study was to examine the predictive role of preoperative NLR for disease-free survival following elective potentially curative resection for CRC.

Method

All consecutive patients who underwent surgical resection for CRC over a 2-year period (2003–2004) at our institution were analysed. For the purpose of this study, only patients who had a potentially curative resection of the primary tumour and survived for more than 30 days postoperatively were included. The operation was classified as potentially curative if both the surgeon and the pathologist reported tumour-free margins of the specimen and no synchronous distant metastases were detected. The reason for this selection was to avoid inclusion of patients with locally residual tumour, where persisting tumour growth is likely. In all, 297 patients underwent elective, potentially curative resection of CRC between January 2003 and December 2004 at the North West London Hospitals NHS Trust. The tumours were staged using the Jass histopathology scoring system [33]. Demographic data including CRC recurrence were prospectively collected from our institutional cancer database.

Patients with coexistent haematological disorders or a known chronic inflammatory disorder, patients who had received preoperative radiotherapy and those who required emergency surgical resection for CRC were excluded from the study to ensure that the white cell count was representative of the normal baseline value. CRC recurrence was diagnosed on radiological and endoscopic histopathological data. The NLR was calculated from the preoperative blood sample by dividing the absolute neutrophil count by the absolute lymphocyte count. We chose an NLR cut-off of 5 as used in other studies [25,31,34]. This value is simple to use in clinical practice.

Statistical analysis

Univariate analysis of disease-free survival was performed using the Kaplan–Meier method (Fig. 1). Univariate and multivariate survival analysis based on Cox proportional hazards regression methodology was undertaken to identify individual risk factors related to CRC recurrence. Hazard ratios (HRs) are given with 95% CI. All statistical tests were two sided, with significance assumed at P < 0.050. Statistical analysis was carried out using Stata se 10.1 for Macintosh (StataCorp, College Station, Texas, USA). HRs were calculated for gender, CRC stage using Jass score [33], preoperative NLR and CRC site.

Figure 1.

 Kaplan–Meier survival analysis.

Results

Over a median follow-up period of 3.35 (0.1–8) years, 59 (19.9%) patients had recurrent CRC (Table 1). CRC recurrence at the anastomosis, the peritoneum, the retroperitoneum or in intra-abdominal lymph nodes was classified as loco-regional recurrence. Recurrence in the liver and outside the abdominal cavity was classified as distant. CRC recurrence was diagnosed by a combination of endoscopic histopathological (colonoscopy) data and radiology (CT to identify loco-regional and distant metastases) during routine postoperative follow-up of patients.

Table 1. Baseline data for 297 patients undergoing elective potentially curative resection for colorectal cancer (CRC).
 Number of patients*
  1. *With percentages in parentheses unless indicated otherwise.

  2. †Values are median (range).

Sex ratio (M:F)157:140
Age at surgery (years)†70 (23–93)
Follow-up (years)†3.35 (0.1–8)
Recurrence59
 Loco-regional8
 Distant51
Site of CRC
 Colon cancer164 (55)
 Rectal cancer111 (37.24)
 Recto-sigmoid cancer22 (7.38)
Stage of CRC based on Jass score, %
 Stage 130.2
 Stage 223.8
 Stage 319.5
 Stage 426.5

Eight patients developed loco-regional recurrence whilst 51 patients developed distant metastases. In the 59 patients, the median lymphocyte value was 1.4 (0.3–3.5) and the median neutrophil value was 6 (1.98–19.2).

Cox analysis of variables predicting CRC recurrence after elective potentially curative resection

Jass stage and preoperative NLR were found to predict CRC recurrence in univariate analysis (Table 2). Multivariate analysis identified Jass stage and preoperative NLR >5 as independent predictors of CRC (Table 3).

Table 2. Results of univariate Cox proportional regression analysis.
VariableHazard ratio95% confidence interval P
  1. NLR, neutrophil to lymphocyte ratio; CRC, colorectal cancer.

  2. *Variable evaluated in multivariate analysis (P < 0.250).

Gender
 Female1
 Male0.840.51–1.400.511
Jass stage*
 11
 21.650.67–4.070.275
 33.481.52–7.980.003
 47.863.58–17.2<0.0001
Preoperative NLR*
 <51
 >51.941.14–3.270.014
CRC site
 Colon1
 Rectum0.660.38–1.160.148
 Recto-sigmoid0.480.15–1.560.221
Table 3. Results of multivariate Cox proportional regression analysis.
VariableHazard ratio95% confidence interval P
  1. NLR, neutrophil to lymphocyte ratio.

Jass stage
 11
 21.660.68–4.100.268
 33.501.53–8.040.003
 48.693.85–19.6<0.0001
Preoperative NLR
 <51
 >51.811.07–3.070.028

Discussion

In the current study, we have demonstrated that elevated preoperative NLR is an independent factor predicting CRC recurrence following elective curative resection. The study has several strengths. Data on the cohort of patients included were recorded prospectively, including follow-up data related to local recurrence, distant metastasis and death. We included only elective surgical resection and excluded patients who required emergency surgical resection of CRC due to obstruction or perforation which may have skewed the preoperative NLR.

The study, to our best knowledge, is the first of its kind to demonstrate that an elevated preoperative NLR is predictive of tumour recurrence following elective potentially curative surgical resection of CRC where all stages of CRC were included. Over the past decade, it has been widely established that it is not only the tumour characteristics that determine its progression but also the host immune and inflammatory response. Systemic inflammatory response is an important tumour-stage-independent predictor of outcome in various cancer types [6]. Neutrophils and leucocytes play a crucial role in this host systemic inflammatory response.

Increasing levels of circulating neutrophils that ensue following a non-specific systemic inflammatory response to a tumour result in a relative lymphocytopenia and elevated NLR which were clearly demonstrated in our study. NLR has been shown to be associated with progression and metastasis of gastric cancer, non-small-cell lung cancer, ovarian cancer, intrahepatic cholangiocarcinoma, hepatocellular carcinoma, pancreatic cancer, CRC and nasopharyngeal cancer.

The association between elevated NLR and poor prognosis in patients with cancer is not clearly defined. Several possible theories have been put forward in an attempt to explain the relationship between elevated NLR and poor oncological outcome. Inhibition of apoptosis, promotion of angiogenesis and damage of DNA in response to the systemic inflammatory response could promote tumour metastasis and progression, as has been demonstrated by recent studies [31,32,35–38]. Elevated NLR is a relative lymphocytopenia, which indicates a significant depression of innate cellular immunity suggested by an increase in T8 suppressor lymphocytes and a marked decrease in T4 helper lymphocytes [39].

Lymphocytes play a key role in cytotoxic cell death and cytokine production that inhibit proliferation and metastatic activity of tumour cells [40]. Anti-tumour immune response of the host to the tumour is lymphocyte dependent. Elevated levels of lymphocytes in the peripheral blood and within the primary tumour have been linked with favourable prognosis in nasopharyngeal cancer and breast cancer patients [40–43]. Systemic inflammation results in the release of a myriad of inhibitory immunological mediators, which can result in a significant immunosuppression with resultant impaired lymphocyte function thus impairing the host anti-tumour immune response [44].

Neutrophils on the other hand have a pro-tumour effect by being the major contributors to tumour-related angiogenesis as they are recognized to be the primary source of circulating angiogenesis regulating chemokines, growth factors and proteases [31,35,36]. NLR can therefore be considered as the balance between pro-tumour inflammatory status and anti-tumour immune status [32]. Clearly, patients with elevated NLR have the balance tipped in favour of pro-tumour inflammatory status, which is associated with poor oncological outcome as established by recent studies. The results of the present study add further evidence to the importance of the systemic inflammatory response and the value of preoperative NLR in patients who undergo elective potentially curative resection of CRC.

The study has certain limitations. Preoperative NLR was evaluated retrospectively. We chose to dichotomize NLR as greater or <5 in order to be consistent with previous studies that have used the same cut-off value [25,29–31,37,45,46].

In conclusion, the study demonstrated that elevated preoperative NLR is associated with decreased disease-free survival in patients who undergo elective potentially curative resection for CRC. NLR is a simple, cost-effective, easily reproducible and widely available routine preoperative evaluation blood test. It is a tumour-stage-independent biomarker which can be used as a well-defined therapeutic target for future clinical trials. Preoperative NLR may also have a role when counselling patients undergoing elective potentially curative resection of CRC about the possible outcome.

Authors’ declaration

The authors declare that the submitted work is their own and that copyright has not been breached in seeking its publication. They confirm that the article is an original work, has not been published before, and is not being considered for publication elsewhere in its final form, in either printed or electronic media. There are no conflicts of interest.

Ancillary