Abstract
Objective
This study aims to investigate the risk factors associated with microvasculature invasion (MI) in patients with pancreatic cancer and evaluate its prognostic significance.
Methods
A retrospective analysis was conducted on the clinical data of 288 pancreatic cancer patients who underwent radical pancreatectomy between June 2012 and June 2024. The collected data included demographic characteristics, pathological findings, and laboratory results. Logistic regression analysis was performed to identify potential factors associated with the occurrence of MI. The Kaplan-Meier method was employed to estimate disease-free survival (DFS) and overall survival (OS). Univariate and multivariate Cox proportional hazards regression models were applied to assess the impact of various factors on patient prognosis.
Results
Among 288 patients, MI was detected in 93 patients (32.3%). Positive microvasculature biomarkers, positive regional lymph nodes (RNP), poor differentiation grade, and reduced MCV might be independent risk factors for MI. Multivariate Cox regression analysis showed that MI, tumor site, RNP, grade, chemotherapy, chloride (Cl) and thrombin time were independent risk factors for DFS and OS. Among 93 patients with MI, no statistically significant difference in prognosis was observed between the MVI and LVI subtypes.
Conclusion
Positive microvasculature biomarkers, positive RNP, poor histological grade, and reduced MCV levels might serve as independent risk factors for the development of MI. The presence of MI was independent risk factors for DFS and OS. However, no significant prognostic association was observed between the specific origin of invasion (whether MVI or LVI).
Supplementary Information
The online version contains supplementary material available at 10.1186/s12957-025-04131-3.
Keywords: Microvasculature invasion, MVI, LVI, Prognostic, Pancreatic cancer
Introduction
Pancreatic cancer is among the most lethal malignant tumors, with a five-year survival rate of merely 9% [1, 2]. In recent years, the incidence of pancreatic cancer has shown a rising trend, particularly in developed countries and regions [3]. There are numerous factors that influence the prognosis of pancreatic cancer, encompassing various aspects including genetics, plasma biomarkers, imaging findings, pathological characteristics, and therapeutic interventions [4, 5]. Among them, microvasculature invasion (MI) is a common pathological characteristic in pancreatic cancer and is widely recognized as one of the key factors influencing patient outcomes [6].
MI refers to the presence of clusters of cancer cells within the microvasculature of a tumor. It is commonly described as microvascular invasion (MVI) or lymphovascular invasion (LVI) in various research studies [6]. MVI typically refers to the invasion of tumor cells into microvascular, and is commonly identified using markers such as Elastica-Masson staining, CD31, and CD34 [7, 8]. In contrast, LVI refers to the invasion of tumor cells into lymphovascular, with D2-40 being the most representative marker [9]. Numerous studies have directly employed these terms without providing clear definitions of MVI or LVI [10, 11], and no research has yet investigated whether such an approach may contribute to reporting bias. Furthermore, existing studies exhibit inconsistencies in the definitions of MVI and LVI. Some consider MVI and LVI as distinct manifestations of a shared pathological feature [6], while others view them as separate entities with differing biological implications and prognostic significance [12]. Despite these ongoing debates, the majority of studies recognized either MVI or LVI as an independent predictor of adverse prognosis in patients with pancreatic cancer [6, 13, 14].
This study aims to investigate the risk factors associated with MI in pancreatic cancer, as well as the prognostic significance of it, with a particular focus on the relationship between the tissue origin of MI and patient outcomes. By integrating a comprehensive review of existing literature with the findings of this study, we aim to provide a more refined theoretical foundation for both clinical practice and mechanistic research related to MI in pancreatic cancer.
Methods
Study population
A total of 288 patients who underwent radical resection for pancreatic cancer between June 2012 and June 2024 at the Department of Hepatobiliary and Pancreatic Surgery, Jiaxing First Hospital, were enrolled in this retrospective study. The inclusion criteria were as follows: (1) histologically confirmed pancreatic ductal adenocarcinoma; (2) no prior history of anticancer treatment before surgery; and (3) availability of complete clinicopathological and follow-up data. The exclusion criteria were defined as follows: (1) concomitant diagnosis of other malignant tumors; (2) presence of positive surgical margins; (3) presence of distant metastases; and (4) postoperative mortality due to surgical complications. All enrolled patients met the predefined selection criteria. This study was approved by the Institutional Review Board of our institution (approval number: LS2021-KY-160).
The collected clinical data included the patient’s gender, age. Laboratory data included blood routine indicators, coagulation function, levels of CA199, CA50, CA242, and biochemical function, etc. All laboratory data were collected within one week prior to the surgery. The pathological data covered the tumor location, maximum tumor diameter (cm), number of regional lymph nodes examined during the surgery (RNE), number of positive regional lymph nodes detected during the surgery (RNP), tumor differentiation grade, TNM staging, MI status, and the immunohistochemical expression of D2-40, CD31, and CD34. The follow-up data included whether chemotherapy performed and metastasis occurred after surgery, disease-free survival (DFS) time (defined as the time interval from the surgery date to the recurrence date), and overall survival (OS) time (defined as the time interval from the surgery date to the death date). The cut-off time for the last follow-up was December 31, 2024.
Pathonulllogical evaluation
In this study, the pathological evaluation was jointly completed by two senior pathologists. If cancer cell clusters were detected in the microvasculature structure, it was defined as MI. Lymphovascular (LVI) was determined based on the expression of D2-40, and microvascular invasion (MVI) was identified according to the expression of CD31 and CD34. In the description of this study, unless otherwise specified, the term “MI” does not distinguish between MVI and LVI. Among them, MVI specifically refers to microvascular invasion, and LVI specifically refers to lymphovascular invasion.
Statistical analysis
The statistical analysis of this study was implemented using R (Version 4.4.2). Chi-square test was used to analyze the count data, while Mann-Whitney U test was applied to the measurement data. Missing values were addressed using multiple imputation with the “mice” package in R. Factors related to the occurrence of MI were included in the logistic regression model. DFS and OS were estimated by the Kaplan-Meier method, and the log-rank test was used to evaluate whether the differences in survival rates were statistically significant. Meanwhile, univariate and multivariate Cox proportional hazards regression models were adopted to assess the relationship between each variable and the prognosis of patients. When the P value < 0.05, the difference was considered to be statistically significant.
Result
Patient cohort
The demographic characteristics, clinical data, laboratory data and pathological findings were presented in Table 1. A more comprehensive overview of additional plasma detection parameters were shown in Supplementary Table 1. Between 2012 and 2024, a total of 288 patients who underwent radical pancreatic surgery were included in the study according to the predefined inclusion and exclusion criteria. The study detected MI in 93 patients (32.3%). The comparative results showed no significant statistic differences between the two groups of patients in terms of gender, age, tumor location, RNE, tumor diameter, T stage, and AJCC stage. However, significant statistic differences were observed in biomarker of MI, RNP, grade, postoperative metastasis, chemotherapy, neutrophil-to-lymphocyte ratio (NLR), neutrophil count, neutrophil percentage, lymphocyte percentage, mean corpuscular volume (MCV), activated partial thromboplastin time (APTT), carbohydrate antigen 19 − 9 (CA199), and carbohydrate antigen 50 (CA50) (p < 0.05).
Table 1.
Baseline characteristics of patients following radical resection for pancreatic cancer
| Variables | Overall N = 288a |
MI (-) N = 195 |
MI (+) N = 93 |
p-valueb |
|---|---|---|---|---|
| Sex | 0.447 | |||
| Male | 160.0 (55.6%) | 105.0 (53.8%) | 55.0 (59.1%) | |
| Female | 128.0 (44.4%) | 90.0 (46.2%) | 38.0 (40.9%) | |
| Age (years) | 67.0 (39.0, 87.0) | 68.0 (39.0, 87.0) | 66.0 (44.0, 82.0) | 0.214 |
| Biomarker | < 0.001 | |||
| Negative | 100.0 (34.7%) | 82.0 (42.1%) | 18.0 (19.4%) | |
| Positive | 188.0 (65.3%) | 113.0 (57.9%) | 75.0 (80.6%) | |
| Site | > 0.999 | |||
| Head | 171.0 (59.4%) | 116.0 (59.5%) | 55.0 (59.1%) | |
| Body & tail | 117.0 (40.6%) | 79.0 (40.5%) | 38.0 (40.9%) | |
| RNE | 9.0 (0.0, 41.0) | 9.0 (0.0, 33.0) | 10.0 (0.0, 41.0) | 0.395 |
| RNP | < 0.001 | |||
| Negative | 157.0 (54.5%) | 124.0 (63.6%) | 33.0 (35.5%) | |
| Positive | 131.0 (45.5%) | 71.0 (36.4%) | 60.0 (64.5%) | |
| Grade | 0.014 | |||
| Poorly | 177.0 (61.5%) | 110.0 (56.4%) | 67.0 (72.0%) | |
| Well | 111.0 (38.5%) | 85.0 (43.6%) | 26.0 (28.0%) | |
| Tumor size (cm) | 3.2 (0.6, 8.5) | 3.2 (0.6, 8.5) | 3.0 (1.5, 8.5) | 0.662 |
| T Stage | 0.573 | |||
| T1 + T2 | 210.0 (72.9%) | 140.0 (71.8%) | 70.0 (75.3%) | |
| T3 + T4 | 78.0 (27.1%) | 55.0 (28.2%) | 23.0 (24.7%) | |
| AJCC8th Stage | 0.469 | |||
| I + II | 248.0 (86.1%) | 170.0 (87.2%) | 78.0 (83.9%) | |
| III + IV | 40.0 (13.9%) | 25.0 (12.8%) | 15.0 (16.1%) | |
| Postoperative metastasis | 0.007 | |||
| Negative | 167.0 (58.0%) | 124.0 (63.6%) | 43.0 (46.2%) | |
| Positive | 121.0 (42.0%) | 71.0 (36.4%) | 50.0 (53.8%) | |
| Chemotherapy | 0.031 | |||
| Negative | 166.0 (57.6%) | 121.0 (62.1%) | 45.0 (48.4%) | |
| Positive | 122.0 (42.4%) | 74.0 (37.9%) | 48.0 (51.6%) | |
| NLR | 3.3 (1.1, 42.5) | 3.2 (1.1, 42.5) | 3.7 (1.2, 27.0) | 0.016 |
| Neutrophil Count (10^9/L) | 3.9 (1.3, 19.0) | 3.7 (1.3, 19.0) | 4.2 (2.1, 16.7) | 0.013 |
| Neutrophil Percentage (%) | 70.6 (45.3, 94.3) | 68.9 (45.3, 94.3) | 72.3 (49.0, 93.3) | 0.006 |
| Lymphocyte Percentage (%) | 20.9 (2.5, 44.8) | 21.9 (2.5, 44.8) | 19.6 (3.5, 41.4) | 0.018 |
| MCV (fl.) | 91.1 (71.8, 103.8) | 91.4 (71.8, 102.8) | 89.6 (71.9, 103.8) | 0.017 |
| APTT (S) | 34.5 (27.4, 46.6) | 34.8 (27.4, 46.6) | 34.0 (27.4, 46.3) | 0.033 |
| CA19-9 (U/ml) | 154.4 (0.6, 4,481.0) | 132.6 (0.6, 4,481.0) | 283.9 (0.6, 2,360.0) | 0.026 |
| CA50 (U/ml) | 70.0 (0.2, 154.0) | 59.8 (0.2, 154.0) | 86.1 (1.0, 154.0) | 0.027 |
APTT Activated Partial Thromboplastin Time, Biomarker immunohistochemical expression of D2-40, CD31, and CD34, CA50 Carbohydrate Antigen 50, CA19-9 Carbohydrate Antigen 19 − 9, MCV Mean Corpuscular Volume, NLR Neutrophil-to-Lymphocyte Ratio, RNE Number of Regional Lymph Nodes Examined, RNP Number of Positive Regional Lymph Nodes
aMedian (Min, Max); n (%)
bWilcoxon rank sum test; Chi-square test
Variables associated with microvasculature invasion (MI)
As shown in Table 1, several variables associated with MI, including the positive microvasculature biomarkers D2-40, CD31, and CD34 (Biomarker), RNP, tumor differentiation grade (Grade), postoperative metastasis status, chemotherapy, NLR, neutrophil count, neutrophil percentage, lymphocyte percentage, MCV, APTT, CA19-9 and CA50, demonstrated statistically significant differences between the MI-positive and MI-negative groups. Further logistic regression analysis of these variables (Table 2) indicated that positive of D2-40, CD31, and CD34 (Biomarker), RNP, a poor differentiation grade and reduced MCV might serve as independent risk factors for the development of MI.
Table 2.
Analysis of risk factors for microvasculature invasion (MI)
| Variables | Estimate | Std. Error | z value | OR | P Value |
|---|---|---|---|---|---|
| Biomarker (Positive) | 1.25 | 0.34 | 3.66 | 3.49 (1.82–6.99) | < 0.001 |
| RNP (Positive) | 1.32 | 0.30 | 4.38 | 3.74 (2.09–6.85) | < 0.001 |
| Grade (Well) | −0.78 | 0.32 | −2.45 | 0.46 (0.24–0.85) | 0.015 |
| MCV (fl.) | −0.06 | 0.03 | −2.14 | 0.94 (0.89–0.99) | 0.034 |
Biomarker Immunohistochemical expression of D2-40, CD31, and CD34, MCV Mean Corpuscular Volume, RNP Number of Positive Regional Lymph Nodes
Risk factors associated with the prognosis of patients following surgery
The follow-up period ranged from 180 to 3700 days. The median follow-up time for DFS was 274 days, and the median follow-up time for OS was 457 days. The DFS and OS curves were shown in Fig. 1. The median disease-free survival time was 227 days for the patients with MI and 426 days for the those without MI (p < 0.001). The overall median survival time was 413 days for the patients with MI and 630 days for those without MI (p=0.0018).
Fig. 1.
The Kaplan-Meier survival estimates for disease-free survival and overall survival after surgery grouped by microvasculature invasion. A DFS after surgery, (B) OS after surgery. MI Microvasculature Invasion
The study used the Schoenfeld residual plot to test the proportional hazards assumption of the Cox regression model. The result (Supplementary Fig. 1) showed that there was no significant time trend in the residuals, which was in line with the proportional hazards assumption.
The results of the univariate Cox regression analysis between all variables and DFS were summarized in Supplementary Table 2. Variables that demonstrated statistical significance were presented in Table 3. Multivariate Cox regression analysis revealed that MI, tumor site, RNP, grade, postoperative metastasis, AJCC 8th edition staging, chemotherapy, preoperative chloride ion (Cl-), thrombin time, and globulin were identified as independent prognostic factors for DFS (p < 0.05).
Table 3.
Univariate and multivariate Cox regression analyses of factors associated with disease-free survival after surgery
| Variables | Univariate Cox | Multivariate Cox | ||
|---|---|---|---|---|
| HR (95%CI) | P value | HR (95%CI) | P value | |
| MI (negative vs. positive) | 1.76 (1.31–2.36) | <0.001 | 1.50 (1.04–2.16) | 0.030 |
| Biomarker (negative vs. positive) | 1.49 (1.08–2.06) | 0.015 | ||
| Site (Head vs. Body & Tail) | 0.68 (0.50–0.91) | 0.01 | 0.67 (0.45–0.99) | 0.047 |
| RNP (negative vs. positive) | 1.83 (1.37–2.44) | < 0.001 | 1.40 (1.01–1.96) | 0.046 |
| Grade (poorly vs. well) | 0.58 (0.43–0.79) | < 0.001 | 0.56 (0.40–0.78) | 0.001 |
| AJCC8th (I + II VS III) | 2.10 (1.41–3.12) | < 0.001 | 1.76 (1.13–2.75) | 0.013 |
| Postoperative metastasis (negative vs. positive) | 3.19 (2.37–4.28) | < 0.001 | 2.60 (1.83–3.69) | < 0.001 |
| Chemotherapy (negative vs. positive) | 0.64 (0.48–0.85) | 0.002 | 0.70 (0.51–0.97) | 0.031 |
| NLR | 1.04 (1.01–1.06) | 0.001 | ||
| WBC (10^9/L) | 1.10 (1.04–1.16) | 0.001 | ||
| Neutrophil Count (10^9/L) | 1.10 (1.05–1.16) | < 0.001 | ||
| Neutrophil Percentage (%) | 1.03 (1.01–1.04) | < 0.001 | ||
| Lymphocyte Percentage (%) | 0.97 (0.95–0.98) | < 0.001 | ||
| Monocyte count (10^9/L) | 2.82 (1.10–7.24) | 0.031 | ||
| MCHC (g/L) | 1.01 (1.00–1.03.00.03) | 0.017 | ||
| MPV (fl.) | 1.15 (1.03–1.27) | 0.014 | ||
| PDW (%) | 1.11 (1.05–1.17) | < 0.001 | ||
| Indirect Bilirubin (µmol/L) | 1.01 (1.00–1.01.00.01) | 0.048 | ||
| Total Protein (g/L) | 0.98 (0.96–1.00.96.00) | 0.028 | ||
| Globulin (g/L) | 0.95 (0.92–0.98) | 0.004 | 0.92 (0.87–0.97) | 0.003 |
| Chloride (Cl) (mmol/L) | 0.93 (0.90–0.97) | < 0.001 | 0.93 (0.89–0.97) | 0.003 |
| Free Fatty Acids (mmol/L) | 1.45 (1.06–1.99) | 0.02 | ||
| Uric Acid (µmol/L) | 1.00 (1.00–1.00) | 0.015 | ||
| Fibrinogen (g/L) | 1.19 (1.04–1.36) | 0.009 | ||
| Thrombin Time (S) | 0.83 (0.74–0.93) | 0.002 | 0.79 (0.68–0.91) | 0.001 |
Biomarker Immunohistochemical expression of D2-40, CD31, and CD34, MCHC Mean Corpuscular Hemoglobin Concentration, MI Microvasculature Invasion, MPV Mean Platelet Volume, NLR Neutrophil-to-Lymphocyte Ratio, WBC White Blood Cell Count, PDW Platelet Distribution Width, RNP Number of Positive Regional Lymph Nodes
The results of all variables associated with OS from the univariate Cox analysis were summarized in Supplementary Table 2. Variables that demonstrated statistical significance were subsequently included in Table 4. The multivariate Cox regression analysis revealed that MI, age, tumor site, RNP, grade, chemotherapy, platelet distribution width (PDW), mean corpuscular hemoglobin concentration (MCHC), preoperative chloride ion (Cl-), preoperative sodium ion (Na+), thrombin time, and plasma osmolality were independent prognostic factors for OS (p < 0.05).
Table 4.
Univariate and multivariate Cox regression analyses of factors associated with overall survival after surgery
| Variables | Univariate Cox | Multivariate Cox | ||
|---|---|---|---|---|
| HR (95%CI) | P value | HR (95%CI) | P value | |
| MI (negative vs. positive) | 1.60 (1.19–2.15) | 0.002 | 1.78 (1.25–2.54) | 0.001 |
| Age (Years) | 1.02 (1.00–1.03.00.03) | 0.035 | 1.03 (1.01–1.05) | 0.007 |
| Site (Head vs. Body & Tail) | 0.72 (0.53–0.97) | 0.031 | 0.63 (0.43–0.94) | 0.023 |
| RNP (negative vs. positive) | 1.63 (1.23–2.16) | 0.001 | 1.56 (1.13–2.15) | 0.007 |
| Grade (poorly vs. well) | 0.51 (0.38–0.69) | < 0.001 | 0.48 (0.34–0.66) | < 0.001 |
| T stage (I + II VS III + IV) | 1.44 (1.06–1.97) | 0.021 | 1.94 (1.35–2.81) | < 0.001 |
| AJCC8th (I + II VS III) | 2.14 (1.45–3.15) | < 0.001 | ||
| Postoperative metastasis (negative vs. positive) | 1.74 (1.31–2.31) | < 0.001 | 1.36 (0.98–1.90) | 0.066 |
| Chemotherapy (negative vs. positive) | 0.47 (0.35–0.63) | < 0.001 | 0.49 (0.35–0.68) | < 0.001 |
| NLR | 1.02 (1.00–1.04.00.04) | 0.028 | ||
| WBC (10^9/L) | 1.06 (1.00–1.11.00.11) | 0.037 | ||
| Neutrophil Count (10^9/L) | 1.07 (1.02–1.12) | 0.008 | ||
| Neutrophil Percentage (%) | 1.03 (1.01–1.04) | < 0.001 | ||
| Lymphocyte Count (10^9/L) | 0.64 (0.47–0.88) | 0.006 | ||
| Lymphocyte Percentage (%) | 0.97 (0.95–0.98) | < 0.001 | ||
| MCHC (g/L) | 1.02 (1.01–1.03) | 0.001 | 1.02 (1.00–1.03.00.03) | 0.018 |
| MPV (fl.) | 1.13 (1.02–1.26) | 0.022 | ||
| PDW (%) | 1.11 (1.05–1.18) | < 0.001 | 1.10 (1.02–1.19) | 0.012 |
| Sodium (Na) (mmol/L) | 0.95 (0.90–0.99) | 0.016 | 1.11 (1.01–1.22) | 0.024 |
| Chloride (Cl) (mmol/L) | 0.93 (0.89–0.96) | < 0.001 | 0.93 (0.87–0.99) | 0.030 |
| Uric Acid (µmol/L) | 1.00 (1.00–1.00) | 0.013 | ||
| Osmolality (mOsm/L) | 0.96 (0.93–0.98) | 0.001 | 0.94 (0.90–0.98) | 0.006 |
| Fibrinogen (g/L) | 1.14 (1.01–1.30) | 0.035 | ||
| Thrombin Time (S) | 0.87 (0.77–0.98) | 0.018 | 0.82 (0.71–0.94) | 0.005 |
CA242 Carbohydrate Antigen 242, MCHC Mean Corpuscular Hemoglobin Concentration, MI Microvasculature Invasion, MPV Mean Platelet Volume, NLR Neutrophil-to-Lymphocyte Ratio, WBC White Blood Cell Count, PDW Platelet Distribution Width, RNP Number of Positive Regional Lymph Nodes
The relationship between the microvasculature origin of microvasculature invasion and prognosis in pancreatic cancer
The specific location and origin of microinvasion were determined by two pathologists through comprehensive analysis of the expression patterns of relevant markers (D2-40,CD31,CD34) and the morphological characteristics of microvascular structures. These cases were subsequently classified into either MVI (microvascular invasion) or LVI (lymphovascular invasion). Among the 93 patients with MI, 8 cases exhibited both MVI and LVI concurrently. Given that these overlapping cases represented a small proportion of the overall cohort, they were randomly assigned to either the MVI or LVI group to maintain analytical clarity without introducing significant bias. Specifically, 27 cases were identified as MVI and 66 as LVI. As illustrated in Fig. 2, the origin of tumor microinvasion couldn’t demonstrate a statistically significant association with prognosis.
Fig. 2.
The Kaplan-Meier survival estimates for DFS and OS after surgery group by origin of microvasculature invasion. A DFS after surgery, (B) OS after surgery. MVI Microvascular Invasion, LVI Micro Lymphovascular Invasion
Discussion
This study systematically investigated the relevant risk factors associated with MI and its potential impact on prognosis by conducting a retrospective analysis of the clinicopathological data of 288 patients who underwent radical surgery for pancreatic cancer. The findings revealed that positive microvasculature-specific biomarkers, positive of RNP, a low differentiation grade, and reduced MCV levels were independent risk factors for the occurrence of MI. In the multivariate Cox regression analysis, MI, tumor site, RNP, grade, chemotherapy, chloride (Cl) and thrombin time were identified as independent prognostic factor for DFS and OS. Notably, among the 93 MI-positive patients, no statistically significant difference in prognosis was observed between the MVI and LVI subtypes.
The biological mechanisms underlying MI in the context of pancreatic cancer progression
The results of this study indicated that MI was an independent risk factor for prognosis in pancreatic ductal adenocarcinoma, both in terms of DFS and OS. In the existing literature, while MVI typically denoted microvascular invasion and LVI referred to lymphovascular invasion, these terms were frequently used interchangeably in the context of pancreatic cancer. In the absence of clear differentiation via immunohistochemical techniques, MVI or LVI was commonly employed to describe the presence of cancer cell clusters within the lumen of microvasculature [6]. The formation of MI results from the coordinated action of multiple biological mechanisms, primarily encompassing the following processes: First, epithelial-mesenchymal transition (EMT) represents a crucial process through which tumor cells acquire migratory and invasive capabilities. During EMT, tumor cells lose epithelial polarity, degrade the extracellular matrix (ECM), and exhibit enhanced motility, enabling invasion into the vascular or lymphatic systems [15]. Second, dysfunction of the microvascular endothelial barrier increases vascular permeability, thereby facilitating the entry of tumor cells into the vascular lumen [16]. Third, under the influence of the tumor microenvironment, cancer-associated fibroblasts (CAFs) contribute to early tumor invasion by remodeling the ECM [17, 18]. Collectively, these mechanisms play significant roles in the progression and development of pancreatic cancer.
Risk factors associated with the occurrence of MI
The results of this study indicated that positive microvasculature biomarkers, positive reginal lymph nodes (RNP), poor differentiation grade, and reduced mean corpuscular volume (MCV) were significant risk factors closely linked to the occurrence of MI. Positive microvasculature biomarkers (such as D2-40, CD31, and CD34) typically reflect a high microvascular density within the tumor, which has been shown to correlate with early postoperative recurrence, metastasis, and adverse prognosis [19]. Given that pancreatic cancer cells often disseminate via microvasculature into the lymphatic or portal venous systems, leading to metastasis in the liver or other distant organs, RNP may represent a clinical manifestation following the development of MI [6]. Consequently, a notable association between MI and positive RNP can be observed. Poorly differentiated or undifferentiated pancreatic cancer exhibits greater invasiveness, with tumor cells more likely to penetrate the microvasculature endothelial barrier, thereby facilitating MI. A similar pattern had been documented in hepatocellular carcinoma research, where a low degree of differentiation frequently coexisted with MVI and contributes to unfavorable clinical outcomes [20]. Elevated MCV is commonly associated with macrocytic anemia and malnutrition, and had also been implicated in poorer prognoses according to other studies [21]. This study is the first to identify a potential association between increased MCV and MI, however, further research is required to validate its clinical relevance, as current literature remains limited.
The relationship between the origin of MI and prognosis
The relationship between the origin of microvascular invasion and its prognostic significance has long been a focus of research. The traditional perspective suggested that MVI is more likely to lead to hematogenous metastasis, whereas LVI primarily contributes to lymph node metastasis [16].In this study, based on the expression profiles of vascular markers and the morphological characteristics of microvascular structures, 93 patients with MI were classified into either the MVI group or the LVI group. The findings revealed no significant association between the origin type of tumor microinvasion and patient prognosis. This result aligns with previous literature reviews [6]. Therefore, this study proposes that in the pathological assessment of pancreatic cancer, the presence of cancer cell clusters within microvascular-like or micro lymphovascular structures should be defined as MI, without the necessity of strictly differentiating its vascular origin or nature.
A new perspective on the prognostic value of laboratory indicators
This study identified that preoperative decreases in serum chloride (Cl) levels, shortened thrombin time and reduced globulin levels are independent risk factors for DFS. Furthermore, increased preoperative platelet distribution width (PDW), elevated mean corpuscular hemoglobin concentration (MCHC), decreased Cl and sodium (Na) levels, shortened thrombin time, and reduced plasma osmolality are independent predictors of OS. Notably, both Cl and thrombin time were significantly associated with adverse outcomes in both DFS and OS. Although plasma Cl levels are influenced by dietary intake and renal excretion, accumulating evidence suggested that chloride channels play a crucial role in tumor progression [22, 23]. Chloride Voltage-Gated Channel 3 (CLCN3) is a key member of the voltage-gated chloride channel family, and it participates in the amplification of cervical cancer, breast cancer and lung cancer [24, 25]. Chloride intracellular channel 1 and 4 (CLIC1 and CLIC4) have been confirmed to play significant roles in various tumors, and compared with CLCN3, they may play a more crucial role in ion channel function [26]. However, the specific mechanisms underlying their involvement in pancreatic cancer remain unexplored. Thrombin time primarily reflects thrombin activity. In patients with Takayasu arteritis, shortened thrombin time has been linked to poorer clinical outcomes [27]. Within the tumor microenvironment, thrombin promotes tumor growth, angiogenesis, and metastasis through the activation of protease-activated receptors (PARs), particularly PAR-1 and PAR-2 [28]. In pancreatic cancer, the thrombin-PAR1 signaling pathway has been shown to contribute to immunosuppression and immune evasion [29].
Advantages and limitations of the study
The primary strength of this study is the large sample size and the systematic collection of comprehensive clinicopathological data. This robust dataset provides a reliable foundation for a more accurate analysis of factors associated with MI and their impact on prognosis. Nevertheless, this study has several limitations. First, as a retrospective and single-center study, it is inherently susceptible to selection bias and information bias. Secondly, the conclusions of this study, especially the impact of laboratory test results on clinical outcomes, have not yet been externally validated and still require confirmation through more independent research. Thirdly, certain potential confounding variables, such as treatment protocols and patients’ lifestyle factors, were not incorporated into the analysis. These omissions may influence the interpretation of prognostic outcomes.
In conclusion, this study identified factors associated with MI in pancreatic cancer and elucidated its correlation with patient prognosis. It further confirmed that MI is a critical pathological feature reflecting the invasive nature of pancreatic cancer, offering valuable insights for future research in this area. Nevertheless, due to the limitations in study design and sample size, further large-scale prospective studies are required to validate these findings.
Supplementary Information
Acknowledgements
None.
Authors’ contributions
Conception and design: Huangbao Li, Fengqing Zhao.Literature search: Yuyang Zhou, Chenyang Bian.Pathological slide reading: Jiayin Yu, Xiangyi Chen.Literature analysis and interpretation: Huangbao Li, Fengqing Zhao.Manuscript writing: All authors.Final approval of the manuscript: All authors.
Funding
The work was funded by grants from the Science and Technology Project of Zhejiang Medical and Health [2025KY1589], Key Discipline of Hepatobiliary and Pancreatic Surgery of Jiaxing City [2023-zc-005], Translational Therapy Center for Hepatobiliary Pancreatic Cancer [2021-YJZX-04], Science and Technology Project of Jiaxing [2022AD30066, 2023AD31059], Jointly Cultivate Disciplines of Jiaxing City and Zhejiang province [2023-PYXK-001], and National Clinical Key Specialty Construction Project [2023-GJZK-001].
Data availability
The data for statistical processing is available from the first author or corresponding author.
Declarations
Ethics approval and consent to participate
This study was performed in accordance with the World Medical Association Declaration of Helsinki and was approved by the Ethics Committee of Cancer Hospital of China Medical University. This study was a retrospective study and all the data used for the final analysis were anonymous. Therefore, the informed consent from the enrolled patients was waived. This study was approved by the Institutional Review Board at the affiliated hospital of Jiaxing University (approval number: LS2021-KY-160).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Uchihara D, Shimajiri S, Harada Y, Kumamoto K, Oe S, Miyagawa K, et al. Long-chain fatty acyl coa synthetase 4 expression in pancreatic cancer: a marker for malignant lesions and prognostic indicator for recurrence. Diagn pathol. 2025;20(1):59. 10.1186/s13000-025-01659-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Siegel RL, Kratzer TB, Giaquinto AN, Sung H, Jemal A. Cancer statistics, 2025. CA Cancer J Clin. 2025;75(1):10–45. 10.3322/caac.21871. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Stoffel EM, Brand RE, Goggins MP. Cancer: changing epidemiology and new approaches to risk assessment. Early Detect Prev Gastroenterol. 2023;164(5):752–65. 10.1053/j.gastro.2023.02.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ouyang S, Zhang J, Liu F, Jiang Q, Xing W, Chen J. Global research trends and hotspots in prognostic prediction models for pancreatic cancer: a bibliometric analysis. Front Oncol. 2025;15:1588735. 10.3389/fonc.2025.1588735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cai J, Chen H, Lu M, Zhang Y, Lu B, You L, et al. Advances in the epidemiology of pancreatic cancer: Trends, risk factors, screening, and prognosis. Cancer Lett. 2021;520:1–11. 10.1016/j.canlet.2021.06.027. [DOI] [PubMed] [Google Scholar]
- 6.Li H, Pan W, Xu L, Yin D, Cheng S, Zhao F. Prognostic significance of microvascular invasion in pancreatic ductal adenocarcinoma: a systematic review and meta-analysis. Med Sci Monit. 2021;27:e930545. https://doi.org/10.12659/msm.930545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gonzalez J, Bahmad HF, Ocejo S, Abreu A, Popp M, Gogola S, et al. The usefulness of elastin staining to detect vascular invasion in cancer. Int J Mol Sci. 2023. 10.3390/ijms242015264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rzepakowska A, Zurek M, Grzybowski J, Pihowicz P, Gornicka B, Niemczyk K, et al. Microvascular density and hypoxia-inducible factor in intraepithelial vocal fold lesions. Eur Arch Otorhinolaryngol. 2019;276(4):1117–25. 10.1007/s00405-019-05355-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Naito Y, Ishikawa H, Sadashima E, Okabe Y, Takahashi K, Kawahara R, et al. Significance of neoadjuvant chemoradiotherapy for borderline resectable pancreatic head cancer: pathological local invasion and microvessel invasion analysis. Mol Clin Oncol. 2019;11(3):225–33. 10.3892/mco.2019.1885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Takahashi H, Katsuta E, Yan L, Tokumaru Y, Katz MHG, Takabe K. Transcriptomic profile of lymphovascular Invasion, a known risk factor of pancreatic ductal adenocarcinoma Metastasis. Cancers (Basel). 2020;12(8):2033. 10.3390/cancers12082033. [DOI] [PMC free article] [PubMed]
- 11.Panaro F, Kellil T, Vendrell J, Sega V, Souche R, Piardi T, et al. Microvascular invasion is a major prognostic factor after pancreatico-duodenectomy for adenocarcinoma. J Surg Oncol. 2019;120(3):483–93. 10.1002/jso.25580. [DOI] [PubMed] [Google Scholar]
- 12.Javed AA, Mahmud O, Fatimi AS, Habib A, Grewal M, He J, et al. Predictors for long-term survival after resection of pancreatic ductal adenocarcinoma: a systematic review and meta-analysis. Ann Surg Oncol. 2024;31(7):4673–87. 10.1245/s10434-024-15281-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Zhang G, Li B, Yin X, Gao S, Shen S, Wang H, et al. Systemic therapy and perioperative management improve the prognosis of pancreatic ductal adenocarcinoma: a retrospective cohort study of 2000 consecutive cases. Int J Surg. 2022;104:106786. 10.1016/j.ijsu.2022.106786. [DOI] [PubMed] [Google Scholar]
- 14.Van Goor I, Schouten TJ, Verburg DN, Besselink MG, Bonsing BA, Bosscha K, et al. Predicting long-term disease-free survival after resection of pancreatic ductal adenocarcinoma: a nationwide cohort study. Ann Surg. 2024;279(1):132–7. 10.1097/sla.0000000000006004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Dudas J, Ladanyi A, Ingruber J, Steinbichler TB, Riechelmann H. Epithelial to mesenchymal transition: A mechanism that fuels cancer Radio/Chemoresistance. Cells. 2020;9(2):428. 10.3390/cells9020428. [DOI] [PMC free article] [PubMed]
- 16.Wong SY, Hynes RO. Lymphatic or hematogenous dissemination: how does a metastatic tumor cell decide? Cell cycle. 2006;5(8):812–17. 10.4161/cc.5.8.2646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ge F, Zeng C, Wang J, Liu X, Zheng C, Zhang H, et al. Cancer-associated fibroblasts drive early pancreatic cancer cell invasion via the SOX4/MMP11 signalling axis. Biochimica et Biophysica Acta (BBA). 2024;1870(1):166852. 10.1016/j.bbadis.2023.166852. [DOI] [PubMed] [Google Scholar]
- 18.Raaijmakers K, Adema GJ, Bussink J, Ansems M. Cancer-associated fibroblasts, tumor and radiotherapy: interactions in the tumor micro-environment. J Exp Clin Cancer Res. 2024;43(1):323. 10.1186/s13046-024-03251-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Annese T, Tamma R, Ruggieri S, Ribatti D. Angiogenesis in pancreatic cancer: Pre-Clinical and clinical Studies. Cancers (Basel). 2019;11(3):381. 10.3390/cancers11030381. [DOI] [PMC free article] [PubMed]
- 20.Xiong SP, Wang CH, Zhang MF, Yang X, Yun JP, Liu LL. A multi-parametric prognostic model based on clinicopathologic features: vessels encapsulating tumor clusters and hepatic plates predict overall survival in hepatocellular carcinoma patients. J Transl Med. 2024;22(1):472. 10.1186/s12967-024-05296-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jomrich G, Gruber M, Gruber ES, Mühlbacher J, Radosavljevic S, Wilfing L, et al. Prognostic significance of mean corpuscular volume in patients with pancreatic ductal adenocarcinoma and multimodal treatment. J visc Surg. 2024;161(2):99–105. 10.1016/j.jviscsurg.2023.06.004. [DOI] [PubMed] [Google Scholar]
- 22.Patel SH, Edwards MJ, Ahmad SA. Intracellular ion channels in pancreas cancer. Cell Physiol Biochem. 2019;53(S1):44–51. https://doi.org/10.33594/000000193. [DOI] [PubMed] [Google Scholar]
- 23.Luo Y, Liu X, Li X, Zhong W, Lin J, Chen Q. Identification and validation of a signature involving voltage-gated chloride ion channel genes for prediction of prostate cancer recurrence. Front Endocrinol (Lausanne). 2022;13:1001634. 10.3389/fendo.2022.1001634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ren L, Li Y, Feng Y, Zhang Z, Yang H, Li M. CLCN3 in mediating the proliferation of human ovarian cancer cells. Transl Cancer Res. 2024;13(3):1443–57. 10.21037/tcr-23-1272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Li Y, Yang Y, Ma Q, Cheng H, Wang H, Ma C, et al. HNRNPK/CLCN3 axis facilitates the progression of LUAD through CAF-tumor interaction. Int J Biol Sci. 2022;18(16):6084–101. 10.7150/ijbs.76083. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Peretti M, Angelini M, Savalli N, Florio T, Yuspa SH, Mazzanti M. Chloride channels in cancer: focus on chloride intracellular channel 1 and 4 (CLIC1 AND CLIC4) proteins in tumor development and as novel therapeutic targets. Biochim Biophys Acta. 2015;1848(10 Pt B):2523–31. 10.1016/j.bbamem.2014.12.012. [DOI] [PubMed] [Google Scholar]
- 27.Li X, Fang C, Gao S, Wu Z, Du L, Tian X et al. Association between Pre-operative Disease Activity and Long Term Revascularisation Outcomes in Patients with Takayasu’s Arteritis: A Multicentre Retrospective Cohort Study. Eur J Vasc Endovasc Surg. 2025;70(5):666-75. 10.1016/j.ejvs.2025.04.068. [DOI] [PubMed]
- 28.Aleksandrowicz K, Hempel D, Polityńska B, Wojtukiewicz AM, Honn KV, Tang DG, et al. The complex role of thrombin in cancer and metastasis: focus on interactions with the immune system. Semin Thromb Hemost. 2024;50(3):462–73. 10.1055/s-0043-1776875. [DOI] [PubMed] [Google Scholar]
- 29.Schweickert PG, Yang Y, White EE, Cresswell GM, Elzey BD, Ratliff TL. Thrombin‐PAR1 signaling in pancreatic cancer promotes an immunosuppressive microenvironment. J Thromb Haemost. 2021;19(1):161–72. 10.1111/jth.15115. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data for statistical processing is available from the first author or corresponding author.