Prognostic value of hemoglobin-to-red cell distribution width ratio and inflammation markers in colorectal cancer

Abstract

BACKGROUND

The hemoglobin-to-red cell distribution width ratio (HRR) is a recently introduced, easily accessible marker that provides insights into inflammation and the tumor vascular microenvironment. It has been suggested to have prognostic value for overall survival in various types of cancer, including urothelial carcinoma, lung cancer, and hepatocellular carcinoma. It has not yet been sufficiently investigated in colorectal cancers (CRC).

AIM

To investigate the prognostic significance of the HRR and other inflammation-based hematological markers in patients with metastatic CRC. Additionally, the study evaluated the impact of surgical interventions, particularly metastasectomy, and multiple clinical and laboratory parameters on overall survival. By identifying low-cost, accessible prognostic indicators, this research seeks to support clinicians in optimizing treatment strategies and risk stratification for patients with CRC.

METHODS

In this retrospective study, patients diagnosed with CRC between January 2020 and December 2024 were analyzed. The impact of HRR in conjunction with inflammatory markers and a total of 22 different clinical and laboratory parameters on overall survival were evaluated using univariate Cox regression and a multivariate model. Survival curves were visualized using Kaplan-Meier analysis.

RESULTS

A total of 155 patients with CRC were included in the study. The median age was 60 years, and 61.9% presented with de novo metastasis. In the receiver operating characteristic curve and area under the curve analysis performed to determine the optimal cutoff, the values were found to be 6.10 for carcinoembryonic antigen (CEA) (P = 0.036), 18.85 for platelet-to-red cell distribution width ratio (P = 0.028), and 10.87 for platelet distribution width-to-lymphocyte ratio (P = 0.028). For neutrophil-to-lymphocyte ratio, systemic immune-inflammation index (SII), platelet-to-lymphocyte ratio (PLR), monocyte-to-lymphocyte ratio, HRR, and carbohydrate antigen 19-9, an optimal cutoff could not be determined using the receiver operating characteristic-area under the curve analysis. Therefore, the median values were adopted as the cutoffs (3.09, 835.96, 177.50, 0.380, 0.824, and 21.6, respectively). Univariate analysis identified male gender (P = 0.045), being under 65 years of age (P = 0.001), history of metastasectomy (P = 0.001), low serum CEA level (P = 0.010), low PLR (P = 0.024), low SII (P = 0.010), and high HRR (P = 0.025) as favorable prognostic factors for overall survival. In the multivariate model, being under 65 years of age [hazard ratio (HR) = 1.59, 95% confidence interval (CI): 1.06-2.39, P = 0.025], metastasectomy (HR = 0.49, 95%CI: 0.29-0.85, P = 0.011), CEA (HR = 1.51, 95%CI: 1.0-2.28, P = 0.048), and PLR (HR = 1.63, 95%CI: 1.09-2.44, P = 0.018) emerged as independent prognostic factors for overall survival, whereas gender, SII, and HRR did not retain statistical significance.

CONCLUSION

In conclusion, low HRR alone was a prognostic indicator. However, when modelled with other inflammatory and clinical parameters, it did not provide a sufficiently strong marker feature.

Key Words: Colon cancer; Neutrophil-to-lymphocyte ratio; Hemoglobin-to-red cell distribution width ratio; Systemic immune-inflammation index; Inflammation markers

Core Tip: Colorectal cancer is a leading cause of cancer-related deaths. While survival can be estimated with special tests, using inexpensive and validated methods such as hemogram parameters may be beneficial for clinicians. Blood parameter scales are currently used for some cancers, but there is not enough data on colorectal cancer. Hemoglobin/red cell distribution width ratio, which combines many parameters such as nutrition and inflammation in the same ratio, may be useful in predicting overall survival.

INTRODUCTION

An estimated 2000000 new cancer cases and approximately 618000 cancer-related deaths are expected in the United States by 2025. Colorectal cancer (CRC) is the third most common type of cancer in males and females and is the leading cause of cancer-related deaths[1,2]. Stage 4 CRC was historically treated with surgery, chemotherapy, and currently targeted therapy methods. However, it still has a poor prognosis[3,4]. Determining prognostic indicators in advance guides clinicians in choosing intensive triplet treatment regimens for select patients[5,6].

Tumor-associated macrophages are known to be regulated by the cancer microenvironment and play a significant role in treatment failures[7]. Factors such as transforming growth factor-β and interleukin-10 have the potential to regulate monocytes and polarize them into M2 macrophages, which subsequently promote tumor cell proliferation, invasion, metastasis, angiogenesis, and matrix deposition and remodeling[7]. Chronic inflammation is thought to influence both cancer development and prognosis through DNA damage-induced mutations and abnormal DNA methylation[8]. Complete blood count is a routine assessment performed by clinicians in individuals with cancer.

Recently, various complete blood count parameters, such as the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), systemic immune-inflammation index (SII), and monocyte-to-lymphocyte ratio (MLR), have been identified as potential prognostic biomarkers in CRC[9-11]. These markers, which reflect both the tumor microenvironment and systemic inflammation, stand out as accessible and cost-effective prognostic tools. Additionally, carcinoembryonic antigen (CEA) levels at the time of diagnosis and Eastern Cooperative Oncology Group have been shown to be prognostic markers[12,13].

Low hemoglobin levels in cancer patients may be associated with malnutrition, decreased bone marrow reserves due to inflammation, and tumor activity[14,15]. Red cell distribution width (RDW) is another important routine hemogram parameter. Traditionally, RDW has been used for the diagnosis and classification of anemia subtypes[16]. Recent studies suggest that RDW, similar to other inflammatory markers, may serve as an indicator of inflammation[17,18]. Moreover, changes in RDW have been shown to play a role in tumor etiology[19]. Based on this premise, RDW has been identified as a prognostic marker in various cancers, including breast cancer and non-small cell lung cancer[20,21]. Hemoglobin-to-red cell distribution width ratio (HRR) has been investigated as a prognostic marker in different types of cancer[22-24]. However, the relationship between HRR and prognosis in CRC remains insufficiently elucidated.

This study aimed to investigate the effects of surgical procedures, such as metastasectomy, on the prognosis of patients with metastatic CRC along with the impact of HRR and other pathological and hemogram parameters on survival.

MATERIALS AND METHODS

Material method

Study design and patient selection: This retrospective study included a total of 155 outpatients diagnosed with metastatic CRC between January 2020 and December 2024. The inclusion criteria were as follows: (1) Patients aged ≥ 18 years; (2) Those who had received and completed at least 2 months of treatment (chemotherapy with or without biological agents) following diagnosis; (3) Patients with organ metastases confirmed through CT, magnetic resonance imaging, or other imaging modalities; and (4) Those without accompanying renal dysfunction. Patients with a history of malignancy, active smokers, had concomitant rheumatologic diseases, known hereditary anemia, multiple synchronous cancers, active infectious disease, or immunosuppressive drug use were excluded from the study.

Data collection: The demographic data, clinicopathological characteristics, and serum laboratory parameters of the patients before the first chemotherapy were recorded from the electronic medical records system. The pre-treatment laboratory parameters included NLR, PLR, MLR, HRR, platelet-to-red cell distribution width ratio (PRR), platelet distribution width-to-lymphocyte ratio (PDWLR), SII, CEA, and carbohydrate antigen 19-9. The SII was calculated as platelet × neutrophil/lymphocyte and recorded.

Statistical analysis

Statistical analyses were performed using SPSS Statistics software version 24. Categorical variables were presented as counts and percentages. To determine the optimal cutoff values for the analyses, a receiver operating characteristic (ROC) curve and area under the curve (AUC) analysis were initially performed. ROC-AUC analysis was preferred because it gave accurate results in similar biomarker studies[25]. Sensitivity and specificity analyses were conducted using the ROC curve. For laboratory parameters where a cutoff could not be determined using the ROC-AUC analysis, the median cutoff value was used. This cutoff value was utilized to classify the patients into “low” and “high” groups. Survival analysis was conducted using the Kaplan-Meier method, and group comparisons were performed using the log-rank test. Factors influencing survival were assessed using univariate and multivariate analyses based on the Cox regression model. The “Forward: LR” method was applied for multivariate analyses. Hazard ratios (HRs) were reported with corresponding 95% confidence intervals (95%CI). Overall survival (OS) was calculated from the date of metastasis diagnosis to the date of death or last follow-up. A P value of < 0.05 was considered statistically significant.

The values for NLR, PLR, SII, MLR, HRR, PRR, and PDWLR were calculated using the following formulas: NLR = neutrophil-to-lymphocyte ratio; SII = (platelet × neutrophil)/lymphocyte; PLR = platelet-to-lymphocyte ratio; MLR = monocyte-to-lymphocyte ratio; HRR = hemoglobin-to-RDW ratio; PRR = platelet-to-RDW ratio; and PDWLR = platelet distribution width-to-lymphocyte ratio.

OS was defined as the time from the date of metastasis diagnosis to either the date of death or the last follow-up. Survival analyses were conducted using the Kaplan-Meier method and the Cox regression model. Variables found to be significant in the univariate Cox regression analysis were included in the multivariate analysis. The “Forward: LR” method was applied for multivariate analysis. A P value of < 0.05 was considered statistically significant for all analyses.

RESULTS

The study was completed with 155 patients who met the inclusion criteria. The median age was 60 years (range: 24-83 years), and 99 patients (63.9%) were female. A total of 96 patients (61.9%) belonged to the de novo metastatic group, with the liver being the most common site of metastasis (69.7%). By the end of the study, 104 patients (67.1%) had died due to cancer-related causes. The median OS for all patients was 25.1 months (95%CI: 19.1-31.2). The general characteristics and laboratory data of the patients are presented in Table 1.

Table 1 Clinicopathological features of the patients.

Patients’ characteristics	Number	%
Age
< 65	94	60.6
≥ 65	61	39.4
Sex
Male	99	63.9
Female	56	36.1
ECOG PS
< 2	55	64.5
≥ 2	100	35.5
Tumor location
Right	31	20.0
Left	124	80.0
Tumor grade
< 2	22	14.2
≥ 2	133	85.8
Body mass index
< 25	68	43.9
≥ 25	87	56.1
Metastasis status
De novo	96	61.9
Recurrence	59	38.1
Liver metastasis
No	47	30.3
Yes	108	69.7
Lung metastasis
No	93	60.0
Yes	62	40.0
Peritoneal metastasis
No	119	76.8
Yes	36	23.2
Metastasectomy
No	123	79.4
Yes	32	20.6
Treatment lines
< 3	112	72.3
≥ 3	43	27.7
Laboratory parameters	Low1	High1
CEA (ng/dL)	71 (45.8)	84 (54.2)
CA19-9 (U/mL)	78 (50.3)	77 (49.7)
NLR	78 (50.3)	77 (49.7)
PLR	78 (50.3)	77 (49.7)
SII	78 (50.3)	77 (49.7)
MLR	77 (49.7)	78 (50.3)
HRR	77 (49.7)	78 (50.3)
PRR	81 (52.3)	74 (47.7)
PDWLR	75 (48.4)	80 (51.6)

¹Low and high values were calculated according to optimal cutoffs.

Data are presented as n (%). ECOG PS: Eastern Cooperative Oncology Group-performance score; CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19-9; NLR: Neutrophil-lymphocyte ratio; PLR: Platelet-to-lymphocyte ratio; SII: Systemic immune-inflammation index; MLR: Monocyte-to-lymphocyte ratio; HRR: Hemoglobin-to-red cell distribution width ratio; PRR: Platelet-to-red cell distribution width ratio; PDWLR: Platelet distribution width to lymphocyte ratio.

The ROC-AUC analysis was conducted to determine the following optimal cutoff values: 6.10 for CEA (AUC = 0.604, 95%CI: 0.51-0.70, P = 0.036); 18.85 for PRR (AUC = 0.609, 95%CI: 0.52-0.70, P = 0.028); and 10.87 for PDWLR (AUC = 0.609, 95%CI: 0.52-0.70, P = 0.028). Since an optimal cutoff could not be determined using the ROC-AUC analysis for NLR, SII, PLR, MLR, HRR, and carbohydrate antigen 19-9, the median values were used as the cutoff points (3.09, 835.96, 177.50, 0.380, 0.824, and 21.6, respectively).

To identify factors associated with OS, univariate analysis revealed that male gender (P = 0.045), under 65 years old (P = 0.001), history of metastasectomy (P = 0.001), low serum CEA levels (P = 0.010), low PLR (P = 0.024), low SII (P = 0.010), and high HRR (P = 0.025) were favorable prognostic factors (Table 2). Using the parameters that were significant in the univariate analysis, a multivariate model was constructed to accurately evaluate prognostic factors for OS. In this model, being under 65 years of age (HR = 1.59, 95%CI: 1.06-2.39, P = 0.025), metastasectomy (HR = 0.49, 95%CI: 0.29-0.85, P = 0.011), CEA (HR = 1.51, 95%CI: 1.0-2.28, P = 0.048), and PLR (HR = 1.63, 95%CI: 1.09-2.44, P = 0.018) emerged as independent prognostic factors for OS. In the multivariate analysis, gender, SII, and HRR did not retain statistical significance (Table 2).

Table 2 Cox regression analysis for overall survival values.

Variable	Category	Univariate analysis		Multivariate analysis
		HR (95%CI)	P value	HR (95%CI)	P value
Patients’ characteristics
Sex	Male/female	0.66 (0.44-0.99)	0.0451
ECOG PS	< 2/≥ 2	1.25 (0.84-1.87)	0.270
Body mass index	< 25/≥ 25	1.42 (0.95-2.13)	0.084
Age	< 65/≥ 65	1.51 (1.02-2.25)	0.0401	1.59 (1.06-2.39)	0.0251
Tumor grade	< 2/≥ 2	0.97 (0.56-1.65)	0.897
Tumor location	Right/left	1.51 (0.87-2.63)	0.142
Metastasis status	De novo/recurrence	1.09 (0.72-1.66)	0.693
Liver metastasis	No/yes	0.99 (0.64-1.54)	0.976
Peritoneal metastasis	No/yes	1.32 (0.81-2.15)	0.273
Lung metastasis	No/yes	0.90 (0.60-1.35)	0.607
Metastasectomy	No/yes	0.42 (0.24-0.71)	0.0011	0.49 (0.29-0.85)	0.0111
Treatment lines	< 3/≥ 3	0.86 (0.70-1.07)	0.168
Laboratory parameters
CEA (ng/dL)	< 6.1/≥ 6.1	1.70 (1.14-2.55)	0.0101	1.51 (1.00-2.28)	0.0481
CA19-9 (U/mL)	< 21.6/≥ 21.6	1.34 (0.91-1.98)	0.138
NLR	< 3.09/≥ 3.09	1.38 (0.93-2.04)	0.108
PLR	< 177.5/≥ 177.5	1.57 (1.06-2.33)	0.0241	1.63 (1.09-2.44)	0.0181
SII	< 835.96/≥ 835.96	1.67 (1.13-2.46)	0.0101
MLR	< 0.38/≥ 0.38	1.00 (0.67-1.48)	0.988
HRR	< 0.83/≥ 0.83	0.64 (0.43-0.95)	0.0251
PRR	< 18.85/≥ 18.85	1.19 (0.80-1.77)	0.384
PDWLR	< 10.87/≥ 10.87	1.05 (0.71-1.57)	0.794

¹Significant values. ECOG PS: Eastern Cooperative Oncology Group-performance score; CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19-9; NLR: Neutrophil-lymphocyte ratio; PLR: Platelet-to-lymphocyte ratio; SII: Systemic immune-inflammation index; MLR: Monocyte-to-lymphocyte ratio; HRR: Hemoglobin-to-red cell distribution width ratio; PRR: Platelet-to-red cell distribution width ratio; PDWLR: Platelet distribution width to lymphocyte ratio; HR: Hazard ratio; CI: Confidence interval.

Kaplan-Meier survival curves were generated for the prognostic factors. The corresponding median OS values according to sex, age, metastasectomy, CEA levels, PLR, SII, and HRR were 21.7 months (95%CI: 16.3-27.1) vs 32.5 months (95%CI: 23.6-41.4) (log-rank P = 0.043), 31.8 months (95%CI: 22.1-41.5) vs 24.6 months (95%CI: 18.9-30.2) (log-rank P = 0.039), 24.6 months (95%CI: 20.9-28.3) vs 48.3 months (95%CI: 15.7-81.0) (log-rank P = 0.001), 34.8 months (95%CI: 22.0-47.6) vs 21.0 months (95%CI: 15.7-26.3) (log-rank P = 0.009), 34.6 months (95%CI: 31.0-38.2) vs 21.7 months (95%CI: 17.8-25.6) (log-rank P = 0.022), 33.7 months (95%CI: 25.0-42.4) vs 21.7 months (95%CI: 17.0-26.4) (log-rank P = 0.010), and 21.7 months (95%CI: 16.8-26.6) vs 32.5 months (95%CI: 21.8-43.2) (log-rank P = 0.024), respectively, with each comparison showing a significant difference (Figure 1).

Figure 1 Kaplan-Meier survival curves for overall survival. A: Carcinoembryonic antigen; B: Sex; C: Age; D: Metastasectomy status; E: Platelet-to-lymphocyte ratio; F: Systemic immune-inflammation index; G: Hemoglobin-to-red cell distribution width ratio; H: All patients. CEA: Carcinoembryonic antigen; PLR: Platelet-to-lymphocyte ratio; SII: Systemic immune-inflammation index; HRR: Hemoglobin-to-red cell distribution width ratio.

DISCUSSION

In this study, which included 155 patients with metastatic CRC, being under 65 years of age and female and having a CEA level of ≥ 6.1, high PLR, high SII, and low HRR were found to be associated with poor prognosis. Additionally, patients eligible for liver metastasectomy at diagnosis and follow-up were identified as having a favorable prognostic indicator. Age, metastasis, CEA, and PLR collectively demonstrated the potential to serve as a low-cost prognostic marker model that could be utilized by clinicians.

The role of inflammation in cancer progression has been widely studied, with evidence linking immune dysregulation to tumor development and survival outcomes. The uncontrolled increase in the immune system leads to a decrease in immunity and results in tumoral progression[26]. The identification of inflammation using inexpensive and easily accessible markers plays a crucial role in survival assessment and consequently in clinicians’ decision-making. Kim et al[27] concluded that high PLR and high NLR are indicators of poor survival outcomes in patients with CRC. Similarly, Erstad et al[28], in a study involving 151 patients with resectable colorectal liver metastases, found that patients with PLR ≥ 220 and NLR ≥ 5 at diagnosis could have poor long-term survival outcomes. Tan et al[29] demonstrated through their analysis that SII, one of the inflammation markers, is a predictor for OS, progression-free survival, and distant metastasis-free survival. In our study, we concluded that patients with high NLR, high PLR, and high SII had poor prognoses, and these results were consistent with the literature.

RDW is a parameter that reflects the variation in the size and volume of erythrocytes in the circulatory system, indicating erythrocyte heterogeneity. Along with the type of anemia, it serves as a marker of patients’ nutritional statuses. RDW has been reported as an indicator of inflammation and has been associated with cardiovascular mortality, the clinical course of necrotizing pancreatitis, hepatitis B, and other inflammatory diseases[30-33]. Low hemoglobin levels, indicating anemia, have been correlated with poor prognosis in various tumors. This association in cancer patients may be attributed to cancer-related hypoxia[34]. Hypoxia can lead to increased angiogenesis in tumor tissues by upregulating molecules such as vascular endothelial growth factor and epidermal growth factor, which are involved in tumor angiogenesis and metabolism. Increased angiogenesis, in turn, can result in reduced treatment tolerance and poorer survival outcomes[35].

A decreased HRR has been identified as a poor prognostic marker in different cancer types due to factors such as poor nutritional status, increased inflammation, and enhanced angiogenesis[23,36]. In a study conducted by Tuncel et al[37], an HRR value below 0.89 was associated with poor prognosis in patients with rectal cancer[37]. In the study by Wu et al[38], low HRR was shown to be a poor prognostic indicator for small cell lung cancer. In our study, we determined the HRR cutoff value as 0.83 using ROC curve analysis and demonstrated that a low HRR was associated with poor prognosis, consistent with the literature.

Our study had certain limitations. One of the important limitations was the retrospective design of the study. Another limitation was the intra-patient variability of hemogram parameters due to factors such as bleeding and acute infections. However, a key strength of our study was the analysis of 22 different clinical and laboratory parameters using homogeneous survival data from a single center.

CONCLUSION

Our study was among the few that have demonstrated the prognostic significance of HRR as a marker reflecting both anemia and inflammation in patients with metastatic CRC. HRR, in combination with CEA, gender, metastasectomy, and other inflammation markers (PLR, SII), was found to be a prognostic indicator. Future prospective studies are needed to validate our findings and develop stronger predictive models by incorporating additional inflammation markers and using more robust analyses, including innovative machine learning-based risk models[39]. Furthermore, it is recommended that the external validation of these results be conducted using independent datasets.