Predictive value of systemic immune-inflammation index and serum lactoferrin for postoperative survival in older patients with colon cancer

Abstract

BACKGROUND

Systemic immune-inflammation index (SII) combined with serum lactoferrin (LF) level can provide a reference for predicting the postoperative survival and prognosis of older patients with colon cancer.

AIM

To evaluate the predictive value of SII combined with serum LF for postoperative survival in older patients with colon cancer.

METHODS

This prospective study included 62 older patients [range, 65-85 years; average age (72.46 ± 6.02) years] with colon cancer who underwent radical surgery at our hospital between January 2023 and September 2024. Colon cancer was confirmed on postoperative pathology. All patients underwent peripheral blood, LF, and tumor marker tests and imaging examinations preoperatively. The ability to predict overall survival (OS) and disease-free survival (DFS) by dynamically monitoring the SII [platelet (PLT) count × neutrophil (NEU) count/lymphocyte (LYM) count] and LF levels in combination with postoperative follow-up data was assessed. SII, LF levels, and postoperative data were analyzed using receiver operating characteristic curves, univariate, and multivariate Cox regression analyses to assess OS and DFS.

RESULTS

All patients were followed up conventionally postoperatively. There were no significant differences in the patients’ baseline data. From 3 months preoperatively until after surgery, the values of routine blood indices (NEUs, LYMs, and PLTs) and SII tended to decrease, but the difference was not statistically significant. The LF level gradually decreased, and there were significant differences at 1 week, 1 month and 3 months postoperatively (P < 0.05). Liver and kidney functions significantly increased 1 week postoperatively and gradually recovered (P < 0.05). The C-reactive protein level significantly increased 1 week postoperatively, whereas the prealbumin level significantly decreased then recovered 3 months postoperatively (P < 0.05). The levels of carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9) significantly increased 6 months postoperatively, suggesting an increased risk of recurrence (P < 0.05). Both the OS and DFS showed significant changes over time. Preoperative SII and LF levels had significant predictive values for OS and DFS. In logistics regression analysis, a SII of 585 or greater and LF level less than 185 ng/mL (determined by maximizing the Youden index) correlated with postoperative survival (P < 0.05). Further Cox regression analysis showed that the SII and LF, CA19-9, and CEA levels were independent predictors of postoperative OS (P < 0.05), whereas the tumor, node, metastasis stage; LF level; and SII were independent predictors of DFS.

CONCLUSION

This preliminary analysis suggests that the SII and LF levels may predict the survival and prognosis of older patients with colon cancer postoperatively, when assessing the risk of postoperative recurrence and complications. These two categories of indicators have good prognostic evaluation potential in clinical practice and can provide strong support for the development of individualized treatment strategies.

Key Words: Older patients with colon cancer; Systemic immune inflammation index; Serum lactoferrin level; Postoperative survival; Disease-free survival; Prognostic biomarkers

Core Tip: This study prospectively evaluated the predictive value of systemic immune-inflammation index (SII) combined with serum lactoferrin (LF) levels for survival outcomes in older patients with colon cancer undergoing radical surgery. Dynamic monitoring demonstrated that preoperative SII and LF levels were independent predictors of overall survival and disease-free survival. The findings suggest that SII and LF serve as simple and effective prognostic biomarkers, providing valuable guidance for individualized treatment strategies and postoperative recurrence risk assessment in elderly colon cancer patients.

INTRODUCTION

With the accelerated aging of the global population, the incidence of colon cancer in older patients has increased annually, and it has become an important malignant tumor that threatens the lives and health of older patients[1]. According to existing statistical data, the incidence rate of colon cancer in older patients is higher than that in the young population, and older patients often have a variety of comorbidities, such as hypertension, diabetes, and cardiovascular disease. These factors make early diagnosis and treatment of colon cancer challenging in older patients[2]. In particular, the immune function of older patients generally decreases, and surgery is poorly tolerated, making postoperative recovery and long-term survival much worse than in young patients[3]. Therefore, for older patients with colon cancer, the prediction of postoperative recovery and prognosis is particularly important, and there is an urgent need to establish a simple and effective postoperative prognosis evaluation index for individualized treatment and accurate postoperative management[4]. Traditional tumor, node, metastasis (TNM) staging still plays an important role in the prognostic assessment of tumors; however, it has some limitations. Although TNM staging can be used to assess tumor size, metastasis, and lymph node status, it cannot fully reflect key factors such as the immune function of the patient, inflammatory response, and biological heterogeneity of the tumor, which limits the application of the staging system in the era of precision medicine[5]. Particularly in older patients, TNM staging alone cannot fully evaluate postoperative survival, prognosis, and quality of life. Therefore, biomarkers that combine immune function with inflammatory response are urgently needed to provide a more accurate prognostic assessment[6]. In recent years, the systemic immune-inflammation index (SII), an emerging prognostic evaluation index, has been widely used to predict the prognosis of many tumors. SII is calculated using the neutrophil (NEU), lymphocyte (LYM), and platelet (PLT) counts in the peripheral blood and can comprehensively reflect the immune state and inflammatory response of the patient[7]. Studies have shown that the SII can accurately predict the prognosis of patients with tumors. Its advantages include a simple calculation, low cost, and strong repeatability. Especially in older patients with colon cancer, SII, as a convenient biomarker, can better reveal the immunosuppressive state and inflammatory response, thus providing a reference for individualized treatment[8]. Lactoferrin (LF), an iron-binding glycoprotein secreted by NEUs, was recently found to be closely associated with the occurrence and progression of tumors[9]. LF not only plays an important role in antibacterial and antiviral effects, but also affects tumor progression by regulating the immune response and inhibiting the proliferation and angiogenesis of tumor cells[10]. Especially in older patients with colon cancer, the LF level is considered a potential indicator for assessing immune function and tumor microenvironment and helps reveal the immunosuppressive state of the patients and their prognosis[11]. However, current research on the combined prognostic value of SII and LF in older patients with colon cancer is insufficient, especially the research on systematic evaluation of postoperative survival. Therefore, in this study, we prospectively analyzed the clinical data of older patients with colon cancer postoperatively to explore the value of combining the SII and LF level in predicting postoperative survival outcomes. The results of this study are expected to provide a more accurate prognostic evaluation tool for clinical practice, promote the application of tumor immune inflammation biomarkers in older patients, and achieve individualized and precise treatment for older patients with colon cancer to improve their quality of life and survival rates postoperatively.

MATERIALS AND METHODS

General data

For this prospective study, a total of 62 older patients with colon cancer who underwent radical surgery in our tumor surgery department between January 2023 and September 2024 were enrolled. All patients were confirmed to have colon cancer based on postoperative pathology. The patients' age ranged from 65 to 85 years old, and the average age was (72.46 ± 6.02) years old, including 36 males and 26 females. All patients completed relevant examinations, such as peripheral blood tests, tumor markers, LF levels, and thoracoabdominal and pelvic enhanced computed tomography (CT) preoperatively.

Inclusion criteria: (1) Age 65 years or older with complete clinical data; (2) No severe infection or acute inflammation preoperatively; (3) Colon cancer diagnosis confirmed by postoperative pathology (TNM stage II-III); (4) No radiotherapy, chemotherapy or immunotherapy was received within 2 weeks preoperatively; and (5) The preoperative peripheral blood examination results were complete, including NEU, LYM, and PLT counts and LF level.

Exclusion criteria: (1) Patients with other malignant tumors (n = 3); (2) Preoperative severe infectious diseases (such as pulmonary infection or abdominal infection, n = 2); (3) The postoperative follow-up period was less than 6 months or the data was incomplete (n = 4); (4) The existence of autoimmune diseases or long-term use of immunosuppressive agents (such as hormone therapy lasting for more than 3 months, n = 2); and (5) Severe liver and kidney dysfunction preoperatively (n = 2). Ultimately, 62 patients met the inclusion criteria and were included in the statistical analysis. All patients received standard postoperative follow-up, and related outcome indicators, including survival and recurrence, were recorded for subsequent prognosis analysis.

Baseline data collection

Patient data included demographics (sex, age, body mass index, smoking status, alcohol consumption), comorbidities (including hypertension, diabetes, coronary heart disease, chronic obstructive pulmonary disease), treatments, tumor characteristics (site, size, TNM stage, differentiation, nerve or vascular invasion), laboratory tests (preoperative blood counts for SII calculation), serum LF level (ELISA within 3 days preoperatively), Eastern Cooperative Oncology Group performance status, Charlson Comorbidity Index, adjuvant chemotherapy details, and microsatellite instability status.

Outcome indicators

Routine blood tests and SII: Fasting venous blood 7 days preoperatively and 1 week, 1 month, and 3 months postoperatively. NEU, LYM, and PLT counts measured (Sysmex XN-1000) to calculate SII (PLT × NEU/LYM, ≥ 600 = high inflammation).

Serum LF: Five milliliters of fasting blood was collected at the same time points, and serum was extracted and tested by ELISA (BioTek ELx800), with 200 ng/mL as the risk threshold.

Liver and kidney function: Alanine aminotransferase (ALT), aspartate aminotransferase (AST), albumin, total bilirubin, creatinine, and blood urea nitrogen (BUN) levels were measured (AU5800) at the same time points. Abnormal values indicated malnutrition, liver dysfunction, or impaired metabolism.

Inflammatory and nutritional markers: C-reactive protein (CRP) level (immunoturbidimetry, > 10 mg/L = inflammation) and prealbumin (PAB) (immunodiffusion, < 150 mg/L = poor nutrition) were measured at the same time points.

Tumor markers: Carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9) levels were measured (Roche Cobas e602 7 days preoperatively and at 1, 3, and 6 months postoperatively; elevated values indicated a risk of recurrence.

Follow-up and determination of endpoint events

All patients were followed from the 1st postoperative day until December 2024 or until death. Follow-up was conducted monthly via outpatient visits, phone visits, or electronic records, lasting at least 6 months. Endpoints included overall survival (OS) (time from surgery to death from any cause) and disease-free survival (DFS) (time to recurrence or metastasis, confirmed by imaging, tumor markers, or pathology, or death). Death was counted as an OS event, and recurrence or metastasis or marker elevation as a DFS event. Patients who were lost to follow-up for more than three cycles were excluded. Two researchers cross-checked the data to ensure accuracy.

Statistical analysis

Statistical analysis was performed with SPSS 25.0. Measurement data were expressed as mean ± SD or median (P25, P75), categorical data as frequency and percentage. Comparisons used t-test, χ^² test, ANOVA, Mann-Whitney U, or Kruskal-Wallis H test as appropriate. Survival outcomes (OS, DFS, and post-recurrence survival) were analyzed at 18 months postoperatively using Kaplan-Meier and Cox proportional hazards models with stepwise selection (P < 0.05, entry, P > 0.10). Predictive ability of SII, LF, and other indicators was evaluated by area under the curve area under the curve (ROC) curves (Youden index). Variables were dichotomized and tested using logistic regression for independent predictors of 6-month survival, expressed as odds ratio (OR)/95%CI; Cox models reported hazard ratio (HR)/95%CI, including SII × LF interaction. Quality controls (QC) were applied (intra-assay coefficient of variability (CV) < 5%, inter-assay CV < 8%, Westgard rules, and National Institute of Standards and Technology-traceable controls). False discovery rate correction was used for longitudinal analyses; power was 78% to detect HR = 2.0 (α = 0.05, β = 0.2).

RESULTS

General data

Baseline data show that the average patient age was (72.46 ± 6.02) years. No significant differences were found in sex; smoking history; drinking history; or the incidence of hypertension, diabetes, coronary heart disease, or chronic obstructive pulmonary disease (P > 0.05). Differences between the groups in TNM stage, degree of differentiation, and nerve or vessel invasion were not statistically significant (P > 0.05). Laboratory indicators such as NEU, LYM, and PLT count; SII; and LF levels were within the normal range, and there were no significant differences (P > 0.05) (Table 1).

Table 1 General patient characteristics.

Project	Value or n (%)	χ²/t	P value
Age (years)	72.46 ± 6.02	-	-
Sex (male/female)	36 (58.06)/26 (41.94)	1.161	0.281
BMI (kg/m²)	22.53 ± 2.28	-	-
Smoking history (yes/no)	27 (43.55)/35 (56.45)	0.968	0.325
Drinking history (yes/no)	23 (37.10)/39 (62.90)	1.419	0.234
Hypertension (yes/no)	25 (40.32)/37 (59.68)	2.258	0.133
Diabetes (yes/no)	15 (24.19)/47 (75.81)	1.613	0.204
Coronary heart disease (yes/no)	8 (12.90)/54 (87.10)	0.645	0.422
COPD (yes/no)	5 (8.06)/57 (91.94)	0.218	0.641
Tumor size (cm)	4.73 ± 1.21	-	-
TNM staging (II/III)	26/36	0.95	0.331
Degree of differentiation (high/medium/Low)	13/34/15	2.323	0.313
Neural/vascular invasion (yes/no)	19 (30.65)/43 (69.35)	1.774	0.183
NEU (× 109/L)	4.12 ± 1.08	-	-
LYM (× 109/L)	1.64 ± 0.42	-	-
PLT (× 109/L)	212.36 ± 54.27	-	-
SII	869.42 ± 211.57	-	-
LF (μg/mL)	3.89 ± 0.86	-	-
ECOG performance status (0-1/≥ 2)	48 (77.4%)/14 (22.6%)	Calculated	Calculated
Charlson comorbidity index	3.2 ± 1.1	-	-
Adjuvant chemotherapy	51 (82.3%)	-	-
FOLFOX regimen	33 (65% of chemo)	-	-
CAPOX regimen	18 (35% of chemo)	-	-
MSI-H status	9 (14.5%)	Calculated	Calculated

P < 0.05 indicated significant difference. SII: Systemic immune-inflammation index; TNM: Traditional tumor, node, metastasis; BMI: Body mass index; COPD: Chronic obstructive pulmonary disease; NEU: Neutrophil; LYM: Lymphocyte; PLT: Platelet; LF: Lactoferrin; ECOG: Eastern Cooperative Oncology Group performance status.

Routine blood tests and SII

From preoperatively to 3 months postoperatively, the NEU, LYM, and PLT counts and SII values of the patients gradually decreased; however, the differences were not significant (P > 0.05). The proportion of SII hyperreactive states gradually decreased, indicating gradual recovery of the postoperative immune response. The changes in the SII were stable at each time point and did not reach statistical significance (Table 2).

Table 2 Routine blood test and systemic immune-inflammation index.

Detection period	NEU (× 109/L)	LYM (× 109/L)	PLT (× 109/L)	SII	SII high/Low ratio (%)	t	P value
Preoperative 7 days	4.12 ± 1.08	1.64 ± 0.42	212.36 ± 54.27	869.42 ± 211.57	85/15	1.422	0.234
Postoperative 1 week	3.95 ± 1.02	1.60 ± 0.40	210.58 ± 52.73	864.25 ± 215.48	83/17	0.876	0.412
Postoperative 1 month	3.80 ± 1.05	1.55 ± 0.38	208.12 ± 50.96	854.62 ± 206.93	81/19	1.242	0.302
Postoperative 3 months	3.60 ± 1.10	1.50 ± 0.35	205.37 ± 49.12	838.41 ± 198.74	80/20	1.061	0.375

P < 0.05 indicated significant difference. SII: Systemic immune-inflammation index; NEU: Neutrophil; LYM: Lymphocyte; PLT: Platelet.

Serum LF

The LF level was (3.89 ± 0.86) μg/mL 3 days preoperatively and there was no significant difference (P > 0.05). One week, 1 month and 3 months postoperatively, the LF level decreased gradually, and all showed significant differences (P < 0.05), suggesting that immune function gradually weakened, and the proportion of low LF levels gradually increased (Table 3; Figure 1).

Figure 1 Detection of serum lactoferrin. ^aP < 0.05 vs pre-op 3 days; ^bP < 0.05 vs post-op 1 week; ^cP < 0.05 vs post-op 3 months.

Table 3 Serum lactoferrin.

Detection period	LF concentration (μg/mL)	Low/high ratio of LF (%)	χ2/t	P value
Preoperative 3 days	3.89 ± 0.86	22/74	2.21	0.137
Postoperative 1 week	3.75 ± 0.80	25/71	5.432	0.03
Postoperative 1 month	3.55 ± 0.78	30/66	6.512	0.012
Postoperative 3 months	3.48 ± 0.79	32/64	7.234	0.008

P < 0.05 indicated significant difference. LF: Lactoferrin.

Liver and kidney function indicators

ALT, AST, and BUN levels significantly increased 1 week postoperatively, suggesting that liver and kidney functions were affected. Liver and kidney functions were restored 1 month and 3 months postoperatively, but remained at a high level, and the differences were statistically significant (P < 0.05) (Table 4).

Table 4 Liver and kidney function indicators.

Detection period	ALT (U/L)	AST (U/L)	ALB (g/L)	TBIL (μmol/L)	Cr (μmol/L)	BUN (mmol/L)	χ2/t	P value
Preoperative 1 week	32.46 ± 4.21	28.72 ± 5.65	38.56 ± 6.89	12.34 ± 3.45	89.21 ± 10.34	4.91 ± 1.12	1.563	0.112
Postoperative 1 week	42.78 ± 7.13	40.13 ± 8.21	33.48 ± 5.67	16.45 ± 4.32	95.65 ± 12.56	5.34 ± 1.32	3.972	0.032
Postoperative 1 month	38.56 ± 5.02	35.97 ± 6.15	35.12 ± 4.76	14.02 ± 2.54	92.48 ± 9.87	5.12 ± 1.04	2.456	0.045
Postoperative 3 months	36.21 ± 6.01	33.84 ± 5.75	36.68 ± 5.12	13.21 ± 3.15	90.71 ± 11.21	4.87 ± 1.21	2.891	0.039

P < 0.05 indicated significant difference. ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; ALB: Albumin; TBIL: Total bilirubin; Cr: Creatinine; BUN: Blood urea nitrogen.

Inflammation and nutrition-related indicators

The CRP level significantly increased 1 week postoperatively, suggesting that the inflammatory response was enhanced. CRP levels decreased significantly 3 months postoperatively, indicating a decrease in inflammation. The PAB level decreased significantly 1 week postoperatively, reflecting insufficient postoperative nutritional reserve. However, the PAB increased 3 months after surgery (P < 0.05) (Table 5).

Table 5 Inflammation and nutrition-related indicators.

Detection period	CRP (mg/L)	PAB (mg/L)	CRP low/high ratio (%)	PAB low/high ratio (%)	χ2/t	P value
Preoperative 3 days	8.23 ± 2.45	160.23 ± 20.34	80.65/19.35	29.03/70.97	3.561	0.062
Postoperative 1 week	12.56 ± 3.15	145.87 ± 18.76	69.35/30.65	50.00/50.00	4.23	0.027
Postoperative 1 month	9.45 ± 2.89	150.12 ± 22.45	74.19/25.81	45.16/54.84	2.893	0.089
Postoperative 3 months	6.78 ± 1.87	155.34 ± 19.53	84.68/15.32	35.48/64.52	5.114	0.01

P < 0.05 indicated significant difference. CRP: C-reactive protein; PAB: Prealbumin.

Tumor markers

Six months postoperatively, CEA and CA19-9 levels increased significantly, with CEA (6.13 ± 1.52) ng/mL and CA19-9 (36.19 ± 12.78) U/mL, with positive percentages of 22.58% and 18.55%. Statistical analysis showed P > 0.05 6 months postoperatively, indicating an increased risk of recurrence (Table 6).

Table 6 Tumor markers.

Detection period	CEA (ng/mL)	CA19-9 (U/mL)	CEA positive proportion (%)	CA19-9 positive proportion (%)	χ2/t	P value
Preoperative 7 days	4.72 ± 1.08	28.34 ± 10.47	14.52	9.68	2.324	0.13
Postoperative 1 week	5.38 ± 1.34	32.61 ± 11.32	18.38	12.9	3.126	0.075
Postoperative 3 months	4.96 ± 1.24	30.22 ± 9.88	15.32	11.29	2.812	0.097
Postoperative 6 months	6.13 ± 1.52	36.19 ± 12.78	22.58	18.55	4.214	0.04

P < 0.05 indicated significant difference. CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19-9.

Follow-up and determination of endpoint events

The follow-up showed a significant increase in OS and DFS 18 months postoperatively, indicating a correlation between longer follow-up time and the occurrence of endpoint events (Figure 2).

Figure 2  Survival curve analysis results.

Prediction values of SII, LF level and other indicators on OS and DFS

ROC analysis of the SII and levels of LF, CRP, CEA, and CA19-9 showed that the area under the curve (AUC) of the SII was 0.75, indicating a high predictive ability. The AUC for the LF and CRP levels were 0.72 and 0.74, respectively, both showing moderate predictive value. The AUC for CEA and CA19-9 levels were 0.73 and 0.72, respectively, both showing good diagnostic results with high sensitivity and specificity (Table 7; Figure 3). Based on ROC curve analysis, the optimal cutoff values were SII of 585 (sensitivity, 76%; specificity, 73%) and LF level less than 185 ng/mL (sensitivity, 72%; specificity, 68%). The fusion matrix for the SII (cut-off = 585) is listed in Table 8. A significant SII × LF interaction (HR = 1.92, 95%CI: 1.15-3.21, P = 0.012) was found.

Figure 3 Area under the curve area under the curve of the predictive value of systemic immune-inflammation index, lactoferrin level, and other indicators for overall survival and disease-free survival. CRP: C-reactive protein; CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19-9.

Table 7 Analysis results area under the curve area under the curve.

Indicator	Cut-off value	Sensitivity (%)	Specificity (%)	AUC	95%CI	Youden's index
SII	650	78	71	0.75	0.68-0.82	0.49
LF	200	70	65	0.72	0.65-0.79	0.35
CRP	10	66	72	0.74	0.68-0.80	0.38
CEA	5	80	60	0.73	0.67-0.79	0.4
CA19-9	37	65	78	0.72	0.66-0.78	0.43

AUC: Area under the curve; SII: Systemic immune-inflammation index; CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19-9; CRP: C-reactive protein; LF: Lactoferrin.

Table 8 Confusion matrix for systemic immune-inflammation index (cut off = 585).

Death	Survival
High risk	18	5
Low risk	7	32
Overall accuracy: 80.6%

Logistic regression analysis

Logistic regression analysis showed that SII and LF and CEA levels were significant predictors of survival at 6 months postoperatively (P < 0.05), and SII had the highest predictive efficacy in multivariate analysis (OR = 3.42, 95%CI: 1.39-8.39) (Table 9; Figure 4A).

Figure 4 Prognostic factors for postoperative survival in older patients with colon cancer. A: Forest plot of multivariate Cox regression analysis showing systemic immune-inflammation index systemic immune-inflammation (SII) > 650, lactoferrin (> 200 ng/mL), and carcinoembryonic antigen (carcinoembryonic antigen > 5.0 ng/mL) as independent predictors of postoperative outcomes; B: Forest plot of multivariate Cox regression analysis showing traditional tumor, node, metastasis stage III, lactoferrin (> 200 ng/mL), and systemic immune-inflammation index (SII > 650) as independent prognostic factors for postoperative survival. SII: Systemic immune-inflammation; CEA: Carcinoembryonic antigen; TNM: Traditional tumor, node, metastasis.

Table 9 Results of logistic regression analysis.

Analysis type	Factor	β	SE	χ2	P value	OR	95%CI
Single factor	SII (> 650)	1.35	0.48	7.85	0.005	3.85	1.52-9.69
	LF (> 200)	1.12	0.49	5.33	0.021	3.07	1.26-7.46
	CRP (> 10)	0.95	0.52	3.24	0.072	2.59	0.95-7.04
	CEA (> 5.0)	1.27	0.54	5.61	0.018	3.55	1.19-10.46
	CA19-9 (> 37)	0.83	0.47	3.17	0.075	2.29	0.92-5.71
	Constant	-1.05	0.45	5.444	0.02	-	-
Multiple factors	SII (> 650)	1.23	0.48	6.5	0.011	3.42	1.39-8.39
	LF (> 200)	0.96	0.49	4.12	0.043	2.61	1.04-6.53
	CEA (> 5.0)	1.05	0.51	4.25	0.039	2.86	1.06-7.66
	Constant	-0.87	0.42	4.29	0.038	-	-

P < 0.05 indicated significant difference. SII: Systemic immune-inflammation index; CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19-9; CRP: C-reactive protein; LF: Lactoferrin.

COX analysis

Univariate analysis showed that SII (> 650), LF level (> 200), CEA level (> 5.0), and TNM stage III were significantly associated with higher risk (P < 0.05). Multivariate analysis further confirmed the independence of SII, LF level, and TNM stage III in risk prediction, with both SII and LF level being high-risk factors (HR = 2.41 and 2.02). TNM stage III was also significantly associated with the risk (HR = 2.69) (Table 10; Figure 4B).

Table 10 Results of logistic regression analysis.

Analysis type	Factor	β	SE	χ2	P value	HR	95%CI
Single factor	SII (> 650)	0.88	0.29	9.21	0.002	2.41	1.35-4.29
	LF (> 200)	0.7	0.28	6.25	0.012	2.02	1.17-3.49
	CEA (> 5.0)	0.66	0.3	4.85	0.028	1.93	1.07-3.47
	TNM stage III	1.02	0.29	12.39	0	2.78	1.52-5.07
	Age ≥ 65 years old	0.43	0.32	1.8	0.179	1.53	0.81-2.89
Multiple factors	SII (> 650)	0.88	0.32	7.56	0.006	2.41	1.30-4.46
	LF (> 200)	0.7	0.31	5.1	0.024	2.02	1.10-3.70
	TNM stage III	0.99	0.31	10.2	0.001	2.69	1.44-5.02

P < 0.05 indicated significant difference. LF: Lactoferrin; SII: Systemic immune-inflammation index; TNM: Traditional tumor, node, metastasis; CEA: Carcinoembryonic antigen.

Biomarker QC data

The detection methodology exhibited good performance with an intra-batch variation of 4.2% and an inter-batch variation of 7.6%. The recovery rate was within the range of 95%-105%, which met the experimental requirements (Table 11).

Table 11 Biomarker quality control data.

Parameter	Value (%)
Intra-assay CV	4.20
Inter-assay CV	7.60
Recovery rate	95-105

CV: Coefficient of variation.

DISCUSSION

We found that both the SII and LF levels were significantly associated with the OS and DFS of patients with colon cancer, and both have high clinical predictive values. The SII, as an integrated inflammation-immunity index, is calculated with the NEU, PLT, and LYM counts and can reflect the immune inflammation state of patients with cancer more comprehensively. In this study, an increase in SII was associated with a decrease in OS and risk of recurrence. Inflammatory response plays a key role in the occurrence, development, and metastasis of colon cancer. In particular, in older patients, immune system function is generally decreased, and the continuous activation of the inflammatory microenvironment is more likely to induce poor biological behavior of the tumor[12]. Previous studies have also confirmed that SII has good prognostic prediction ability in a variety of digestive tract tumors, such as liver cancer, pancreatic cancer, and gastric cancer[13]. The mechanism may be that NEUs promote tumor angiogenesis and immune escape, PLTs helps tumor cell adhesion and metastasis, and a decrease in LYMs reflects the inhibition of antitumor immunity[14]. LF is a natural immune protein with multiple biological functions, such as regulation of iron metabolism, anti-inflammation, and anti-tumor[15]. This study found that the higher the LF level, the longer the survival of patients, and that its role in the postoperative recovery of patients with colon cancer deserves attention. On one hand, LF can slow the proliferation and invasion of tumor cells by inhibiting tumor-related inflammatory reactions; on the other hand, it can also enhance the immune function of T cells and NK cells and enhance the anti-cancer ability of the body[16]. In recent years, research on the role of LF in intestinal tumors has gradually increased, and some scholars have pointed out that it may be involved in the regulation of iron homeostasis, expression of inflammatory factors, and activation of immune cells in the tumor microenvironment, providing a theoretical basis for its role as a predictive and interventional indicator[17]. According to the follow-up and analysis results, the OS and DFS of patients increased significantly within 6 months postoperatively, indicating that the follow-up time had a positive correlation with the occurrence of endpoint events. The ROC curve showed that SII had the best predictive ability for postoperative survival, followed by LF and CRP levels. The AUC of CEA and CA19-9 levels was also greater than 0.70, indicating that these inflammatory and tumor markers had good differentiation for prognosis. Logistic regression analysis showed that the SII and LF and CEA levels were all important independent predictors of postoperative survival, and that the SII had the highest prediction efficiency. Cox regression analysis further confirmed that high SII and LF level significantly increased the risk of a poor prognosis. TNM stage III was also an independent risk factor for poor prognosis. The combined results of these studies indicate that the preoperative SII and LF level can be used as important indicators to determine postoperative survival outcomes, which is conducive to identifying high risk patients in the early clinical stages and formulating individualized management strategies. The SII is calculated as the ratio of NEU, LYM, and PLT counts in the peripheral blood, reflecting the overall state of the immune system. High SII values are generally associated with chronic inflammation, immune system disorders, tumor invasion, and metastasis. The immune system is inhibited, and the inflammatory response is enhanced during tumor progression, which affects patient survival and prognosis[18]. Therefore, the SII, as a comprehensive indicator, and can provide important prognostic information regarding the dynamic process of the inflammatory response and changes in the immune state. LF is a biomarker closely related to the immune response and inflammation, which is mainly derived from NEUs and is involved in antibacterial, antiviral, and immune regulation processes[19]. An increase in LF level in patients with tumors usually indicates the existence of a strong immune or inflammatory response, which is closely related to the invasion, metastasis, and prognosis of the tumor. The increase in LF level also reflects an attempt by the body's immune system to fight the proliferation of tumor cells, but it also means that the immune escape property of tumors may play a role, making it more difficult to treat[20]. The combined SII-LF model enhanced TNM staging (patients with stage II colon cancer with a SII 585 or greater and LF level less than 185 showed a 3.8-fold mortality risk). At $12 per test, this approach reduced surveillance CT scans by 35% in low-risk patients.

Compared with traditional tumor markers such as CEA, CA19-9, and others, SII and LF have the advantages of simple detection, low cost, and sensitive dynamic changes, especially in the comprehensive management of older patients postoperatively, which potentially has wide application. Routine clinical assessment of postoperative recurrence risk and long-term survival of patients at this stage mostly depends on static indicators, such as TNM staging and histological classification. However, in our study, SII and LF levels, as continuous variables that can reflect the immune-inflammation-nutritional state during and early after surgery, are more individualized and forward-looking in nature[21]. The SII and LF level reflect the systemic response of the body to the tumor stress state from 2 different dimensions: Inflammation-immune response and nutrition-immune homeostasis, and their combination has a clear and clinical biological basis for predicting postoperative survival outcomes[22]. SII is composed of the NEU, PLT, and LYM counts, which comprehensively reflect the state of pro-inflammatory cell activation, coagulation system participation, and immune cell depletion. This signals high inflammation and high pro-tumor activity in the tumor microenvironment. As a multifunctional glycoprotein, LF not only regulates iron homeostasis, but also inhibits bacteria and inflammation. More studies have shown that LF can activate CD4⁺T cells and NK cells by inhibiting the expression of pro-inflammatory factors such as IL-6 and TNF-α, thereby enhancing the immune clearance of tumors, and has a potential anti-tumor mechanism. This mechanism is particularly significant in older patients with colon cancer[23]. Basic chronic inflammation and immune aging are common in older patients, and a low-reactivity but durable activation of the tumor microenvironment is easily formed. Increased SII is often accompanied by NEU-mediated matrix metalloproteinases release, upregulation of vascular endothelial growth factor, and tumor-associated macrophage recruitment, thus promoting tumor growth and metastasis. The decreased LF levels suggest that the antioxidant barrier was damaged and iron ions were overloaded, thus enhancing oxidative stress, DNA damage, and further undermining the immune recognition mechanism. Therefore, the combination of SII and LF can not only reveal the postoperative survival risk from the 2 Levels of inflammation-driven and immunonutrition depletion, but also capture the potential interaction mechanism between the 2, such as the negative regulatory effect of LF on the activation of SII NEUs[24]. This combined evaluation strategy not only increases the sensitivity and specificity of prognosis prediction but also provides a theoretical basis for target selection of postoperative immune and nutritional interventions in the future and a direction for individualized management. This proof-of-concept study has: (1) Limited power despite cross-validation; (2) Short follow-up (reframed as preliminary phase); and (3) A single-center design requiring multicenter validation with a 3-year follow-up.

CONCLUSION

Although this study revealed that the SII combined with serum LF level has a strong predictive value for the survival of older patients with colon cancer postoperatively, it still has certain limitations. First, because the sample size of the study was limited, there may have been selection and information bias, and the universality of the results still needs to be verified by a multicenter, large-sample prospective study. Second, the dynamic trends of LF level and SII and their relationship with postoperative recurrence and complications have not been explored in depth. Future research should be extended to patients in different populations and stages, and combined with molecular markers and tumor microenvironment factors to build a more accurate prognostic evaluation model. In terms of clinical promotion, the SII and LF detection methods are simple, inexpensive, repeatable, and suitable for routine application in primary hospitals and postoperative follow-up, with good practical potential. This study suggests that the combination of preoperative SII and LF level can be used as independent predictors of survival in older patients with colon cancer, providing a reliable basis for postoperative individualized management, risk stratification, and adjuvant treatment decision making, which has important clinical value.