Prognostic Value of Inflammation-based Prognostic Scores in Patients with Colorectal Cancer

Abstract

Background

Anatomopathological staging is the primary method to determine the prognosis of patients with colorectal carcinoma (CRC). However, new tools have been developed that can complement it, such as the analysis of the elevation of systemic inflammatory markers.

Objective

To evaluate the impact of the elevation of scores based on inflammatory markers (the neutrophil-to-lymphocyte ratio [NLR], the Glasgow Prognostic Score [GPS], and isolated C-reactive protein [CRP]) in the prognosis of patients diagnosed with CRC and submitted to potentially curative surgery in Hospital de Braga, Portugal, between January 1st, 2005, and December 31st, 2010.

Methods

A retrospective analysis of the data of 426 patients was performed, with a collection of several clinico-pathological variables, as well as the levels of lymphocytes, neutrophils, albumin and CRP, in the pre- and postoperative periods, to apply the different scores to the sample.

Results

From the analysis of the survival curves, we concluded that patients with increased NLR in the pre- and postoperative periods present a lower cancer-related survival than patients with normal NLR (preoperative period: 93.7 versus 122 months; p<0.001; postoperative period: 112 versus 131 months; p=0.002). Patients with increased NLR in the pre- and postoperative periods also had a lower disease-free survival (preoperative period: 88.0 versus 122 months; p<0.001; postoperative period: 111 versus 132 months; p=0.002). In addition, increased pre- and postoperative NLR was associatedwith a higher risk of death due to CRC (preoperatively: hazard ratio [HR]=2.25; p<0.001; postoperatively: HR=2.18; p=0.003). However, the multivariate analysis shows that only postoperative NLR (ajusted HR =2.66; p=0.002) does so independently of the remaining variables.

Conclusion

Regarding the scores applied to the sample, the NLR was the one that most consistently related to the prognosis of the patients. However, it would be useful to develop a prospective study that could confirm this relationship.

Keywords:
colorectal cancer; scores of systemic inflammatory markers; NLR; GPS; CRP

Introduction

Recently, the impact of elevated systemic inflammatory markers such as the Glasgow Prognostic Score (GPS), the neutrophil-to-lymphocyte ratio (NLR) or isolated C-Reactive Protein (CRP) on the prognosis of different oncological diseases, namely in colorectal carcinoma (CRC) has been studied and recognized.^{1 2 3 4 5 6 7}

The GPS consists of the combination of CRP values and serum albumin, and its increase may reflect a systemic inflammatory response state associated with the neoplastic process. Hypoalbuminemia can also be related to cachexia present in advanced stages of the disease.^{8 9}

The increase in the NLR has been associated with a worse prognosis not only in several oncological diseases, but also in cases of cardiovascular diseases. The rationale that supports this association is also based on the fact that a greater systemic inflammatory response, often subclinical, may be related to a worse prognosis.¹⁰

Given these assumptions, it is extremely important to understand the relationship between these scales and the prognosis of patients operated due to CRC in order to be able to more quickly and easily establish the prognosis of patients and guide them in a more personalized and effective manner.

Materials and Methods

The present is an observational, retrospective and descriptive study.

The study population consisted of all patients with a diagnosis of CRC who underwent surgical treatment between January 1st, 2005, and December 31st, 2010. A non-probabilistic convenience sample was developed, applying several inclusion and exclusion criteria to the target population. Inclusion criteria: patients with postoperative histological diagnosis of colorectal adenocarcinoma, who underwent surgical resection at Hospital de Braga between January 1st, 2005, and December 31st, 2010. Exclusion criteria: patients undergoing surgical resection with palliative intent; patients undergoing urgent surgical resection; patients in whom it was not possible to assess the value of inflammatory parameters; patients with no result of histological staging; and patients who died in the postoperative period (30 days or less after surgery).

For each patient, we collected data on: gender; age at the date of surgery; personal history of neoplasia; family history of CRC; duration of the symptoms; tumor location; macroscopic aspect of the tumor; preoperative levels of carcinoembryonic antigen (CEA); tumor dimension; histological type; tumor staging; levels of lymphocytes, neutrophils, albumin and pre- and postoperative CRP; date of death; and date of onset of local recurrence and distant progression of the disease.

Cancer-related survival was defined as the period in months from the date of the first surgery to the date of cancer-related death.

Disease-free time was defined as the period from the first surgery until the diagnosis of local recurrence or distant progression of the disease.

As the present is a retrospective study, it was not possible to guarantee that the analyses had been carried out at the same pre- and postoperative periods. In order to minimize the impact of this situation, the data was collected at a interval-time within a maximum period of 15 days before and 3 days after surgery.

The statistical analysis was performed using the Statistical Package for the Social Sciences (IBM SPSS Statistics for Windows, IBM Corp., Armonk, NY, US) software, version 22.

For all tests performed, statistically significant results were defined as p < 0.05.

Initially, a descriptive analysis of the many variables collected was carried out, in order to characterize the sample under study. Then, three scales were chosen to characterize the systemic proinflammatory state of the patients: the NLR, the GPS, and elevation of isolated CRP. Each of the scales was applied to patients in the pre- and postoperative periods.

With regard to the NLR, which is the ratio between the absolute levels of neutrophils and lymphocytes, values above 5 eere considered high, according to a study by Urrejola et al.¹¹

The GPS, which results from the combination of the CRP and albumin levels, was categorized into 3 groups (0, 1 and 2), depending on whether they had 0 altered values, only 1 or 2 respectively, as can be seen in Table 1. For the calculation of this scale, CRP levels higher than 1 mg/dl and albumin lower than 3.5 g/dl were defined as altered, since these were the reference values of the laboratory responsible for the analytical evaluation of all patients.

Table 1
Categorization of the different groups on the Glasgow Prognostic Score (GPS)

GPS = 0	CRP ≤ 1 mg/dl and albumin ≥ 3.5 g/dl
GPS = 1	CRP > 1 mg/dl or albumin < 3.5 g/dl
GPS = 2	CRP > 1 mg/dl and albumin < 3.5 g/dl

• Abbreviation: CRP, C-reactive protein.

Finally, levels of isolated CRP was defined as increased when higher than 1 mg/dl, for the same reason mentioned for the calculation of the GPS.

In order to determine the correlation between the different scales and the clinico-pathological characteristics of the patients, crosstabs were performed, and the interdependence of the variables was analyzed using the Fisher exact test whenever the number of cells with a value lower than 5 was greater than 20% of the total cells, and the Pearson Chi-squared (χ²) test when it was lower than 20%. As a measure of effect size, the Phi coefficient (Φ) was used for 2 × 2 tables, as well as the Cramer V for larger tables. For both coefficients, small, medium or large effects were considered for values close to 0.1, 0.3 and 0.5 respectively. For the variables that showed a statistically significant correlation, adjusted standardized residuals were evaluated, and residuals with an absolute value greater than 1.96 were considered as statistically significant, in order to determine the contribution of each cell to the significance of the test and thus determine the direction of the association.

In order to determine the impact of the elevation of the different scales on the outcome of the patients, an attempt was made to compare the time from surgery to the occurrence of two events: death by CRC or recurrence of the disease. For this purpose, Kaplan-Meier survival curves were made, in order to check if there were differences between the groups with increased and normal inflammatory parameters in relation to the survival time and disease-free time. The comparison between the survival curves of the different groups was performed using the log-rank test.

Finally, a univariate analysis of the impact of all variables under study on the patients' survival time was performed, followed by a multivariate analysis using only those that, in the univariate analysis, significantly influenced the survival time. For this evaluation, the Cox proportional-hazards model was used.

The present study was carried out in accordance with the principles of the Declaration of Helsinki, the Convention on Human Rights and Biomedicine, the guidelines of the Council for International Organizations of Medical Sciences, and the Guide to Good Clinical Practice, and was approved by the Ethics Subcommittee on the Life and Health Sciences of Universidade do Minho (SECVS-063/2017) and by the Health Ethics Committee of Hospital de Braga (HB-70/2017).

Results

After applying the exclusion criteria, a sample of 426 patients was obtained.

The clinico-pathological characteristics and the distribution of the sample in relation to the different scales are described in Tables 2 and 3 respectively.

Table 2
Clinico-pathological characteristics of the sample

	n (%)
Gender
Male	260 (61.0)
Female	166 (39.0)
Age (years)
< 45	16 (3.8)
> 45	410 (96.2)
Previous neoplasia
Present	56 (13.1)
Absent	370 (86.9)
Family history of colorectal canrcinoma
Present	3 (7.7)
Absent	359 (83.3)
Unknown	34 (8.0)
Time with symptoms (months)
< 6	287 (67.4)
> 6	59 (13.8)
Asymptomatic	80 (18.8)
Location
Right colon	99 (31.7)
Left colon	213 (68.3)
Rectum	114 (26.8)
Macroscopic appearance
Polypoid	222 (52.1)
Ulcerated	106 (24.9)
Infiltrative	34 (8.0)
Exophytic	29 (6.8)
Indeterminate	35 (8.2)
Carcinoembryonic antigen (ng/ml)
≤ 10	307 (72.1)
> 10	52 (12.2)
Indeterminate	67 (15.7)
Tumor size (cm)
≤ 4.5	255 (59.9)
> 4.5	154 (36.1)
Indeterminate	17 (4.0)
Histological type
Adenocarcinoma	379 (89.0)
Mucinous adenocarcinoma	44 (10.3)
Signet-ring cell adenocarcinoma	3 (0.7)
Stage
I	79 (18.5)
II	187 (43.9)
III	160 (37.6)

Table 3
Distribution of patients in relation to the scales of systemic inflammatory markers

	Preoperative period	Postoperative period
Neutrophil-to-lymphocyte ratio
Normal (≤ 5)	282	104
Increased (> 5)	75	276
Indeterminate	69	46
Glasgow Prognostic Score
0	5	0
1	56	14
2	6	56
C-reactive protein
Normal (≤ 1mg/dl)	9	1
Increased (> 1mg/dl)	90	177
Indeterminate	327	248

Of the studied patients, 32.8% died due to CRC. The remaining 67.2% are either still alive, or died of other causes, or were lost to follow-up. The average cancer-related survival of the patients was of about 117 months.

The cancer-related survival curve obtained by the Kaplan-Meier method is shown in Figure 1.

Fig. 1
Kaplan-Meier curve for CRC survival.

Of the studied patients, 23.2% had local or distant recurrence of the disease. The average disease-free time was of about 114 months.

The survival curve related to disease-free time obtained by the Kaplan-Meier method is shown in Figure 2.

Fig. 2
Kaplan-Meier curve for the disease-free survival of patients.

Interdependence Analysis

The assessment of the interdependence of the different clinical variables under study and the scales under analysis is outlined in Tables 4, 5, 6.

Table 4
Analysis of interdependence between pre- and postoperative NLR and several clinico-pathological variables

	Preoperative NLR			Postoperative NLR
	≤ 5	> 5	Statistics	≤ 5	> 5	Statistics
Gender
Male	176	49	χ2 = 0.217Φ = −0.25 p = 0.641	57	184	χ2 = 4.579Φ = −0.110 p = 0.032
Female	106	26		47	282
Age (years)
≤ 45	10	2	p ≤ 0.99	6	9	χ2 = 1.253Φ = 0.057 p = 0.263
> 45	272	73		98	267
Previous tumor
Yes	66	9	χ2 = 0.317Φ = −0.30 p = 0.573	9	42	χ2 = 2.801Φ = 0.086 p = 0.094
No	241	41		95	234
Family history
Yes	26	4	p = 0.473	6	25	χ2 = 0.993Φ = 0.053 p = 0.319
No	239	61		88	230
Presentation
Asymptomatic	60	10	χ2 = 2.371Φ = 0.082 p = 0.124	21	52	χ2 = 0.089Φ = 0.015 p = 0.766
Symptomatic	222	65		83	224
Time with symptoms
< 6 months	182	56	χ2 = 0.618Φ = −0.046 p = 0.432	71	185	χ2 = 0.381Φ = 0.035 p = 0.537
> 6 months	40	9		12	39
Location
Colon	196	86	χ2 = 6.251Φ = −0.132 p = 0.012	76	199	χ2 = 0.036Φ = 0.010 p = 0.850
Rectum	63	12		28	77
Macroscopic appearance
Polypoid	154	33	χ2 = 1.677V = 0.071 p = 0.624	54	147	χ2 = 0.585V = 0.041 p = 0.900
Ulcerated	74	15		27	68
Infiltrative	22	8		10	21
Exophytic	21	4		6	19
Carcinoembryonic antigen (ng/ml)
≤ 10	220	27	χ2 = 7.041Φ = 0.152 p = 0.008	88	191	χ2 = 5.576Φ = 0.132 p = 0.018
> 10	44	14		6	37
Tumor size (cm)
≤ 4.5	170	102	χ2 = 1.356Φ = 0.063 p = 0.244	66	162	χ2 = 1.752Φ = 0.069 p = 0.186
> 4.5	39	32		31	106
Histological type
Adenocarcinoma	250	65	p = 0.818	97	239	p = 0.157
Mucinous adenocarcinoma	30	9		7	34
Signet-ring cell adenocarcinoma	2	1		0	3
Stage
I	63	8	χ2 = 8.823V = 0.157 p = 0.012	17	59	χ2 = 2.717V = 0.085 p = 0.257
II	120	28		41	119
III	99	39		46	98

• Abbreviations: NLR, Neutrophil-to-lymphocyte ratio; V, Cramer V; Φ, Phi coefficient; χ², Chi-squared.

Table 5
Analysis of interdependence between pre- and postoperative GPS and several clinico-pathological variables

	Preoperative GPS			Postoperative GPS
	0	1/2	Statistics	1	2	Statistics
Gender
Male	4	49	p ≤ 0.99	9	37	χ2 = 0.016Φ = −0.015 p = 0.900
Female	1	13		5	19
Age (years)
≤ 45	0	2	p < 0.99	0	1	p ≤ 0.99
> 45	5	60		14	55
Previous tumor
Yes	1	4	p = 0.330	1	3	p ≤ 0.99
No	4	58		13	53
Family history
Yes	2	6	p = 0.110	5	4	p = 0.010
No	3	54		8	51
Presentation
Asymptomatic	1	17	p ≤ 0.99	7	6	χ2 = 11.430Φ = 0.404 p < 0.001
Symptomatic	4	45		7	50
Time with symptoms
< 6 months	3	33	p ≤ 0.99	6	41	p ≤ 0.99
> 6 months	1	12		1	9
Location
Colon	2	50	p = 0.070	11	39	p = 0.742
Rectum	3	12		3	17
Macroscopic appearance
Polypoid	2	35	p = 0.369	10	25	p = 0.315
Ulcerated	3	15		2	18
Infiltrative	0	5		0	4
Exophytic	0	5		1	6
Carcinoembryonic antigen (ng/ml)
≤ 10	4	53	p ≤ 0.99	11	41	p ≤ 0.99
> 10	0	6		2	6
Tumor size (cm)
≤ 4.5	5	36	p = 0.148	9	31	p = 0.592
> 4.5	0	25		5	24
Histological type
Adenocarcinoma	4	54	p = 0.526	14	46	p = 0.194
Mucinous adenocarcinoma	1	8		0	10
Signet-ring cells adenocarcinoma	−	−		−	−
Stage
I	3	7	p = 0.017	4	7	χ2 = 2.187Φ = 0.177 p = 0.335
II	2	30		5	25
III	0	25		5	24

• Abbreviations: GPS, Glasgow Prognostic Score; Φ, Phi coefficient; χ², Chi-squared.

Table 6
Analysis of the interdependence between preoperative C-reactive protein and several clinico-pathological variables

	Preoperative C-reactive protein
	≤ 1	> 1	Statistics
Gender
Male	7	65	p ≤ 0.99
Female	2	25
Age (years)
≤ 45	0	4	p ≤ 0.99
> 45	9	86
Previous tumor
Yes	1	10	p ≤ 0.99
No	8	80
Family history
Yes	1	8	p = 0.568
No	7	78
Presentation
Asymptomatic	2	20	p ≤ 0.99
Symptomatic	7	70
Time with symptoms (months)
< 6	6	56	p ≤ 0.99
> 6	1	14
Location
Colon	4	74	p = 0.019
Rectum	5	16
Macroscopic appearance
Polypoid	4	48	p = 0.087
Ulcerated	2	22
Infiltrative	3	6
Exophytic	0	6
Carcinoembryonic antigen (ng/ml)
≤ 10	7	70	p ≤ 0.99
> 10	0	8
Tumor size
≤ 4.5 cm	9	55	p = 0.026
> 4.5 cm	0	33
Histological type
Adenocarcinoma	8	79	p ≤ 0.99
Mucinous adenocarcinoma	1	11
Signet-ring cell adenocarcinoma	Not available	Not available
Stage
I	3	12	p = 0.234
II	4	41
III	2	37

• Note: The value of the isolated CRP was only used in the preoperative period, since, in the postoperative period, only 1 patient had normal values.

The evaluation of the crosstab tables regarding the various scores used revealed the existence of a correlation between increased preoperative NLR and the tumor location in the colon/rectum (χ²⁽¹⁾ = 6.251; p = 0.012; Φ = −0.132). The value of Φ shows that it is a weak association. The analysis of standardized adjusted residuals confirms that colon injury is associated with increased preoperative NLR, as can be seen in Table 4.

The preoperative NLR also showed a correlation with CEA values (χ²⁽¹⁾ = 7.041; p = 0.008; Φ = −0.152) and more advanced stages of the disease (χ²⁽²⁾ = 8.823; p = 0.012; Cramer V [V] = 0.157). Both situations reflect a weak correlation. The analysis of the standardized adjusted residuals shows that increased NLR is positively related to an increase in the CEA value, and to more advanced stages of the disease, as can be seen in Table 4.

Regarding the postoperative NLR, its increase is related to the patient's gender (χ²⁽¹⁾ = 4.579; p = 0.032; Φ = −0.110), and, in this case, it was higher NLR in male patients. There is also a correlation with the increase in CEA (χ²⁽¹⁾ = 5.576; p = 0.018; Φ = −0.132), with higher values of NLR associated with higher values of CEA. Again, in both cases it is a weak association, as can be seen in Table 4.

The preoperative GPS was associated with the stage of the disease (Fisher exact test; p = 0.017), with a lower GPS associated with less advanced stages of the disease, as can be seen in Table 5.

The postoperative GPS, on the other hand, was associated with family history (Fisher exact test; p = 0.010), with a higher GPS with the presence of family history. There was also a correlation between the elevation of the preoperative GPS and the occurrence of symptoms prior to diagnosis (χ²⁽¹⁾ = 11.430; p < 0.001; Φ = 0.404), as can be seen in Table 5.

The increase in preoperative CRP, in turn, was associated with the location of the tumor in the colon (Fisher exact test; p = 0.019) and with its larger dimension (Fisher exact test; p = 0.026), as can be seen in Table 6.

Cancer-Related Survival Assessment

In order to assess the time of cancer-related survival in relation to the various scales under study, Kaplan-Meier curves were made for each group, and these were compared using the log-rank test.

Patients with increased NLR in the preoperative period had a significantly lower survival than patients with normal NLR (average survival of 122 months and 93.7 months respectively; p < 0.001), as shown in Figure 3A and Table 7.

Table 7
Average cancer-related survival time according to NLR in the preoperative and postoperative periods

	Average
	Estimate	Standard error	95% confidence interval
			Inferior limit	Superior limit
Preoperative NLR
< 5	122.248	3.009	116.350	128.145
> 5	93.675	7.332	79.305	108.045
Global	116.401	2.900	110.717	122.086
Postoperative NLR
< 5	131.792	4.100	123.757	139.827
> 5	112.022	3.408	105.343	118.701
Global	117.488	2.751	112.096	122.880

• Abbreviation: NLR, Neutrophil-to-lymphocyte ratio.

Fig. 3
Kaplan-Meier curves for survival according to the NLR in the preoperative (A) and postoperative (B) periods.

Also, patients with increased NLR in the postoperative period had a lower survival than patients with normal NLR (average survival of 132 months and 112 months respectively; p = 0.002), as shown in Figure 3B and Table 7.

The application of the GPS did not show significant differences between patients who scored 0, 1 or 2 on the scale, both in the pre- and postoperative periods (p = 0.092 and 0.254 respectively), as shown in Figures 4A and 4B.

Fig. 4
Kaplan-Meier curves for survival according to the GPS in the preoperative (A) and postoperative (B) periods.

The value of the isolated CRP was only used in the preoperative period, since, in the postoperative period, only one patient had values within normal limits. However, the increase in preoperative CRP showed a statistically significant decrease in patient survival (p = 0.038), which is illustrated in Figure 5.

Fig. 5
Kaplan-Meier curve for survival according to CRP levels.

Evaluation of Disease-Free Time

Similarly, the time from surgery to the occurrence of local and distant recurrences of the disease was also evaluated, using the comparison of the Kaplan-Meier curves of the patients according to the different scales.

The evaluation and comparison of the curves showed significant differences in disease-free time in patients with increased NLR in the preoperative period compared to patients with normal NLR (88.0 months and 123 months respectively; p < 0.001), as shown in Figure 6A and Table 8. In the postoperative period, these differences were also noticeable (132 months if normal NLR, and 111 months if increased NLR; p = 0.002), as shown in Figure 6B and Table 8.

Fig. 6
Kaplan-Meier curves for disease-free survival according to the NLR in the preoperative (A) and postoperative (B) periods.

Table 8
Average disease-free time according to NLR in the preoperative and postoperative periods

	Average
	Estimate	Standard error	95% confidence Interval
			Inferior limit	Superior limit
Preoperative NLR
< 5	122.710	3.201	116.436	128.984
> 5	88.004	7.685	72.942	103.065
Global	115.657	3.105	109.572	121.742
Postoperative NLR
< 5	132.010	4.377	123.430	140.590
> 5	110.784	3.720	103.493	118.075
Global	116.977	2.973	111.150	122.805

• Abbreviation: NLR, Neutrophil-to-lymphocyte ratio.

The application of the GPS did not show significant differences at any time with regard to the time until recurrence (p = 0.083 and 0.538 respectively), as shown in Figures 7A and 7B.

Fig. 7
Kaplan-Meier curves for disease-free survival according to the GPS in the preoperative (A) and postoperative (B) periods.

In this case, it was only possible to evaluate the CRP in the preoperative period, which showed a tendency towards significance between the two groups (p = 0.059), as shown in Figure 8.

Fig. 8
Kaplan-Meier curve for disease-free survival according to CRP levels in the preoperative period.

Uni- and Multivariate Analysis of the Impact on Cancer-Related Survival

In order to determine the impact of each variable on the cancer-related survival of the patients, a univariate analysis was performed according to the Cox regression model. Then, using the variables that showed statistical significance in the univariate analysis, a multivariate analysis was carried out in order to determine which variables influenced the survival time independently of the others, as shown in Table 9.

Table 9
Univariate and multivariate analyses of the impact of different variables on the time of cancer-related survival of patients, according to the Cox regression model

	Univariate analysis			Multivariate analysis
	Unadjusted hazard ratio	95% confidence interval	p-value	Adjusted hazard ratio	95% confidence interval	p-value
Gender (female)	1.012	0.689–1.486	0.953	−	−	−
Age (> 45 years)	1.534	0.487–4.830	0.465	−	−	−
Previous tumor (yes)	0.937	0.535–1.640	0.818	−	−	−
Family history (yes)	0.507	0.222–1.158	0.107	−	−	−
Presentation (symptomatic)	1.220	0.752–1.978	0.421	−	−	−
Time with symptoms (> 6 months)	0.757	0.412–1.388	0.368	−	−	−
Location (Rectum)	1.047	0.695–1.578	0.826	−	−	−
Macroscopic appearance				−	−	−
Polypoid	0.877	0.376–2.044	0.761
Ulcerated	1.070	0.445–2.572	0.880
Infiltrative	1.527	0.580–4.019	0.391
Carcinoembryonic antigen (> 10)	2.206	1.334–3.647	0.002	1.707	0.986–2.954	0.056
Tumor size (>4.5 cm)	1.116	0.757–1.645	0.580	−	−	−
Histological type (mucinous)	1.374	0.770–2.452	0.283	−	−	−
Stage
II	3.018	1.357–6.710	0.007	3.367	1.299–8.729	0.013
III	6.064	2.774–13.257	< 0.001	6.706	2.630–17.102	< 0.001
Preoperative NLR (> 5)	2.252	1.484–3.418	< 0.001	1.509	0.902–2.523	0.117
Postoperative NLR (>5)	2.178	1.310–3.621	0.003	2.656	1.412–4.998	0.002
Preoperative GPS (1 or 2)	2.47	0.042–8.028	0.352	−	−	−
Postoperative GPS (2)	2.000	0.593–6.474	0.264	−	−	−
Preoperative CRP (> 1 mg/dl)	2.481	0.217–2.841	0.184	−	−	−

• Abbreviations: GPS, Glasgow Prognostic Score; CRP, C-reative protein; NLR, Neutrophil-to-lymphocyte ratio.

An observation of the univariate analysis in Table 9 enables us to conclude that the risk of death from CRC among patients with levels of CEA greater than 10 ng/ml is 2.21 times higher than that of patients with normal CEA levels (unadjusted hazard ratio [HR] = 2.21; p = 0.002). In addition, the risk of dying due to CRC among patients in stage II is 3.02 times higher than that of patients in stage I (unadjusted HR = 3.02; p = 0.007). In turn, stage-III patients have a 6.06-fold higher risk (unadjusted HR = 6.06; p < 0.001). The univariate analysis of the impact of inflammatory markers on CRC-related survival confirms what had already been verified in the analysis of survival curves. The increase in the NLR in the preoperative period is associated with a 2.25-fold increased risk of death from CRC (unadjusted HR = 2.25; p < 0.001). In the postoperative period, the increase in this scale implies an increase of 2.18 in the risk (unadjusted HR = 2.18; p = 0.003).

The increase in the GPS in the pre- and postoperative periods was not associated with a statistically significant increase in the risk of death from CRC, as did the CRP score in the preoperative period.

The multivariate analysis enables the identification of the variables that influence the survival time related to CRC independently from the other variables. By observing the multivariate analysis in Table 9, we can conclude that patients in stages II (adjusted HR = 3.37; p = 0.013) and III (adjusted HR = 6.71; p < 0.001) present a risk of 3.37 times and 6.71 times higher of death from CRC respectively. Also, the increased postoperative NLR independently increases the risk of death from CRC by 2.66 times (adjusted HR = 2.66 and p = 0.002).

The levels of CEA and the value of the preoperative NLR do not significantly increase the risk of death from CRC independently of the other variables.

Discussion

Currently, staging is the main determinant of the prognosis of patients with CRC.^{1 12} Recently, the impact of the elevation of systemic inflammatory markers has gained increasing interest. Scales such as the GPS, the NLR and isolated CRP have been shown to influence the prognosis of various cancers, including CRC.^{2 3 4 5 6 7 11 13}

The study sample, composed of 426 patients, showed a predominance of male individuals, which is in agreement with what is described in the literature.¹⁴ Moreover, the median age at diagnosis, 71.5 years, with 96.2% of the diagnoses occuring after the age of 45 years, is in accordance with the bibliographic data, which place age as one of the main non-modifiable risk factors for the development of CRC.^{1 13 15} Regarding the location of the tumor, there was a predominance of tumors located in the left colon (50.0%), which is also in accordance with what is described in the literature.^{16 17 18}

Regarding the impact of the elevation of the different inflammatory markers on survival, we found that patients with increased NLR in the preoperative period (p < 0.001) had a significantly lower survival rate than patients with normal NLR. The same was true for patients with increased NLR in the postoperative period (p = 0.002).

In addition, the increase in the preoperative levels of isolated CRP was associated with a significant decrease in patient survival (p = 0.038).

Regarding the GPS, no significant differences in survival were detected either in the preoperative period (p = 0.092) or in the postoperative period (p = 0.254).

With regard to the local and distant recurrences of the disease, significant differences were detected in the disease-free time in patients with increased NLR in the preoperative period compared to patients with normal NLR (p < 0.001). This was also observed in the postoperative period (p = 0.002). However, neither the application of the GPS (preoperatively: p = 0.083; postoperatively: p = 0.538), nor the isolated CRP (p = 0.059 in the preoperative period) showed significant differences in relation to recurrence.

The univariate analysis of the data showed that the increase in the levels of CEA (unadjusted HR = 2.21; p = 0.002), more advanced stages of the disease (stage II: unadjusted HR = 3.02; p = 0.007; stage III: unadjusted HR = 6.06; p < 0.001), and an elevation of the NLR in the preoperative and postoperative periods (unadjusted HR = 2.25; p <0.001; unadjusted HR = 2.18; p = 0.003 respectively) are associated with a higher risk of death due to CRC.

The multivariate analysis, in turn, showed that, of these variables, only the advanced stage (stage II: adjusted HR = 3.37; p = 0.013; stage III: adjusted HR = 6.71; p <0.001) and the postoperative NLR (adjusted HR = 2.66; p = 0.002), increase the risk independently of the remaining variables.

The fact that the preoperative NLR did not show significant results in the multivariate analysis may be due to the existence of a correlation (p = 0.012) between increased preoperative NLR and more advanced stages of the disease, which is not the case for postoperative NLR, as can be confirmed in Table 4.

Thus, we can conclude that, in the study sample, the NLR would be the scale whose results most consistently influenced the patients' prognosis, regarding cancer-related survival and the time to relapse of the disease. However, only the postoperative NLR of the scales under analysis influences the prognosis independently of the remaining variables.

This conclusion can be supported by the understanding of the pathophysiological mechanisms in which these immune cells are involved, since the increase in the number of neutrophils has been associated with tumor progression, by the creation of cytokines that induce tumor growth, as well as by the creation of an proangiogenic environment, favorable to vascularization and tumor invasion.^{19 20 21} In contrast, lymphocytes play a primarily antitumor role, and their increase reflects the activation of the patient's immune system.^{22 23 24 25}

The first study to demonstrate a direct relationship between elevated inflammatory markers and decreased survival in patients with CRC was carried out in 2007 by McMillan et al.,²⁶ who showed that an increase in the preoperative GPS was associated with a shorter survival.

In 2012, Sugimoto et al.³ also demonstrated that an increased GPS was related to decreased survival, as did Guthrie et al.,²⁷ who, in 2013, obtained similar results. The latter study even concluded that the GPS was a scale superior to the NLR in assessing the impact on survival.

There are several studies demonstrating the association between patients and an increase in the preoperative inflammatory parameters; however, few are related to the postoperative period. Nevertheless, in 2015, Shibutani et al.²⁸ concluded that an elevated NLR both in the preoperative and postoperative periods correlated with a decrease in patient survival.

More recently, Rosi et al.²⁹ demonstrated that the GPS and the NLR are superior to other markers in assessing the survival of patients with CRC.

Much less frequent are the studies relating the increase in inflammatory parameters with the recurrence of the disease. However, in 2017, Balde et al.³⁰ demonstrated that an increase in the preoperative NLR is a strong predictor of shorter survival time without recurrence.

The present study has some limitations. The fact that this is a retrospective study makes it difficult to collect the data, and to ensure the homogeneity of the records. In addition, some patients were lost to follow-up. Regarding the comparison of the scales, there was less data available for the GPS and CRP than for the NLR, which may help to justify the fact that the former two did not show significant results.

Conclusion

Inflammatory markers have been increasingly associated with the prognosis of different neoplasias, namely CRC. The present study demonstrated an association between pre- and postoperative NLR and the survival and recurrence of patients with CRC. In addition, it showed that the postoperative NLR influences CRC mortality independently of the remaining variables.

These markers are routinely requested in the preoperative exams of patients, so they are accessible and do not represent an additional cost; therefore, their implementation in clinical practice is simple and will enable an assessment of additional surgical risk as well as the prognosis of the patient.

References

• ¹
  Chua TC, Saxena A, Chu F, Zhao J, Morris DL. Predictors of cure after hepatic resection of colorectal liver metastases: an analysis of actual 5- and 10-year survivors. J Surg Oncol 2011;103(08): 796-800
• ²
  Maeda K, Shibutani M, Otani H, et al. Prognostic value of preoperative inflammation-based prognostic scores in patients with stage IV colorectal cancerwho undergo palliative resection of asymptomatic primary tumors. Anticancer Res 2013;33(12):5567-5573
• ³
  Sugimoto K, Komiyama H, Kojima Y, Goto M, Tomiki Y, Sakamoto K. Glasgow prognostic score as a prognostic factor in patients undergoing curative surgery for colorectal cancer. Dig Surg 2012; 29(06):503-509
• ⁴
  Ishizuka M, Nagata H, Takagi K, Horie T, Kubota K. Inflammationbased prognostic score is a novel predictor of postoperative outcome in patients with colorectal cancer. Ann Surg 2007;246 (06):1047-1051
• ⁵
  Shibutani M, Maeda K, Nagahara H, et al. A high preoperative neutrophil-to-lymphocyte ratio is associated with poor survival in patients with colorectal cancer. Anticancer Res 2013;33(08): 3291-3294
• ⁶
  Chiang SF, Hung HY, Tang R, et al. Can neutrophil-to-lymphocyte ratio predict the survival of colorectal cancer patients who have received curative surgery electively? Int J Colorectal Dis 2012;27 (10):1347-1357
• ⁷
  Chung YC, Chang YF. Serum C-reactive protein correlates with survival in colorectal cancer patients but is not an independent prognostic indicator. Eur J Gastroenterol Hepatol 2003;15(04):369-373
• ⁸
  Wu J, Cai Q, Li H, et al. Abstract 102: Circulating C-reactive protein and colorectal cancer risk: a report from the Shanghai Men's Health Study. Cancer Res 2013;73(8, Supplement):102-102
• ⁹
  Zhang P, Xi M, Li QQ, et al. Themodified glasgow prognostic score is an independent prognostic factor in patients with inoperable thoracic esophageal squamous cell carcinoma undergoing chemoradiotherapy. J Cancer 2014;5(08):689-695
• ¹⁰
  Güragaç A, Demirer Z. The neutrophil-to-lymphocyte ratio in clinical practice. Can Urol Assoc J 2016;10(3-4):141
• ¹¹
  Urrejola GI, Bambs CE, Espinoza MA, et al. [An elevated neutrophil/lymphocyte ratio is associated with poor prognosis in stage II resected colon cancer]. Rev Med Chil 2013;141(05):602-608
• ¹²
  Gloeckler Ries LA, ReichmanME, Lewis DR,Hankey BF, Edwards BK. Cancer survival and incidence fromthe Surveillance, Epidemiology, and End Results (SEER) program. Oncologist 2003;8(06):541-552
• ¹³
  Hassan C, Gimeno-García A, Kalager M, et al. Systematic review with meta-analysis: the incidence of advanced neoplasia after polypectomy in patients with and without low-risk adenomas. Aliment Pharmacol Ther 2014;39(09):905-912
• ¹⁴
  Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015;136(05):E359-E386
• ¹⁵
  Chen K, Qiu JL, Zhang Y, Zhao YW. Meta analysis of risk factors for colorectal cancer. World J Gastroenterol 2003;9(07):1598-1600
• ¹⁶
  Lieberman DA, Prindiville S, Weiss DG, Willett WVA Cooperative Study Group 380. Risk factors for advanced colonic neoplasia and hyperplastic polyps in asymptomatic individuals. JAMA 2003;290 (22):2959-2967
• ¹⁷
  Tarver T. Cancer Facts & Figures 2017. American Cancer Society (ACS). J Consum Health Internet 2017;16(03):366-367
• ¹⁸
  Hemminki K, Santi I,Weires M, Thomsen H, Sundquist J, Bermejo JL. Tumor location and patient characteristics of colon and rectal adenocarcinomas in relation to survival and TNM classes. BMC Cancer 2010;10(01):688
• ¹⁹
  Neal CP, Mann CD, Sutton CD, et al. Evaluation of the prognostic value of systemic inflammation and socioeconomic deprivation in patients with resectable colorectal liver metastases. Eur J Cancer 2009;45(01):56-64
• ²⁰
  Terzic J, Grivennikov S, Karin E, Karin M. Inflammation and colon cancer. Gastroenterology 2010;138(06):2101-2114.e5
• ²¹
  Nozawa H, Chiu C, Hanahan D. Infiltrating neutrophils mediate the initial angiogenic switch in a mouse model of multistage carcinogenesis. Proc Natl Acad Sci U S A 2006;103(33):12493-12498
• ²²
  Kitayama J, Yasuda K, Kawai K, Sunami E, Nagawa H. Circulating lymphocyte is an important determinant of the effectiveness of preoperative radiotherapy in advanced rectal cancer. BMC Cancer 2011;11(01):64
• ²³
  Okano K, Maeba T, Moroguchi A, et al. Lymphocytic infiltration surrounding liver metastases fromcolorectal cancer. J Surg Oncol 2003;82(01):28-33
• ²⁴
  Chiba T, Ohtani H, Mizoi T, et al. Intraepithelial CD8+ T-cellcount becomes a prognostic factor after a longer follow-up period in human colorectal carcinoma: possible association with suppression of micrometastasis. Br J Cancer 2004;91(09): 1711-1717
• ²⁵
  Erdman SE, Sohn JJ, Rao VP, et al. CD4+CD25+ regulatory lymphocytes induce regression of intestinal tumors in ApcMin/+ mice. Cancer Res 2005;65(10):3998-4004
• ²⁶
  McMillan DC, Crozier JE, Canna K, Angerson WJ, McArdle CS. Evaluation of an inflammation-based prognostic score (GPS) in patients undergoing resection for colon and rectal cancer. Int J Colorectal Dis 2007;22(08):881-886
• ²⁷
  Guthrie GJ, Roxburgh CS, Farhan-Alanie OM, Horgan PG, McMillan DC. Comparison of the prognostic value of longitudinal measurements of systemic inflammation in patients undergoing curative resection of colorectal cancer. Br J Cancer 2013;109(01):24-28
• ²⁸
  Shibutani M,Maeda K, Nagahara H, et al. Theprognostic significance of a postoperative systemic inflammatory response in patients with colorectal cancer. World J Surg Oncol 2015;13(01):194
• ²⁹
  Rossi S, Basso M, Strippoli A, et al. Are Markers of Systemic Inflammation Good Prognostic Indicators in Colorectal Cancer? Clin Colorectal Cancer 2017;16(04):264-274
• ³⁰
  Balde AI, Fang S, He L, et al. Propensity score analysis of recurrence for neutrophil-to-lymphocyte ratio in colorectal cancer. J Surg Res 2017;219:244-252

Publication Dates

• Publication in this collection
  24 Jan 2022
• Date of issue
  Oct-Dec 2021

History

• Received
  12 Sept 2020
• Accepted
  18 Jan 2021