Thromboembolic Events after Surgery for Localized Prostate Cancer: Incidence,Risk Factors,and Survival Outcomes

Abstract

Background

The incidence of thromboembolism (TE) is lower in prostate cancer than in other malignancies, but its sequelae are associated with increased mortality. This study investigated the incidence, risk factors, and survival impact of TE in Asian population with localized prostate cancer undergoing surgical management.

Methods

This was a single-center retrospective cohort study of 4,880 localized prostate cancer patients who underwent surgical resection between March 1, 2002, and December 31, 2019. TE, venous thromboembolism (VTE), survival outcome, and related clinical variables were analyzed.

Results

The 3-month TE cumulative incidence was 0.33% and the 3-month VTE incidence was 0.16%. The 2-year TE and VTE incidence were 0.81% and 0.29%, respectively. Older age (≥75) (Subdistribution Hazard Ratio (SHR) 2.16, P = 0.0155) and an elevated C-reactive protein levels (≥0.35 mg/dL) (SHR 2.91, P = 0.0001) were significant risk factors for an increased incidence of TE. Older age (SHR 3.67, P = 0.0113) and pathological venous invasion (SHR 5.13, P = 0.0195) were significant risk factors for an increased incidence of VTE. The occurrence of VTE, older age, a history of diabetes, and high Gleason scores (≥8) were independently associated with poor survival.

Conclusions

The incidence of TE was low in this East Asian cohort of surgically treated patients with localized prostate cancer, however, VTE was an independent predictor of worse overall survival. These findings suggest that pharmacological thromboprophylaxis should be reserved for patients identified as being at high-risk for TE, whereas mechanical measures appear sufficient for the general low-risk cohort.

Keywords

thromboembolism venous thromboembolism localized prostate cancer prostatectomy survival outcome East Asian people

Introduction

Venous thromboembolism (VTE) is a prevalent complication in oncology, with cancer patients experiencing approximately 4- to 6- fold increased risk compared to the general population.¹ The occurrence of VTE is associated with poor survival, increased in-hospital mortality, increased rates of hospitalization, and interruptions in anticancer therapy. Consequently, preventing VTE has been a major concern in cancer patients.^2,3 While primary thromboprophylaxis can reduce these risks, routine anticoagulation is not universally recommended due to bleeding concerns.^4,5 Recent guidelines emphasize a risk-stratified approach, recommending tailored thromboprophylaxis strategies based on individual patient factors and the type of surgery performed.^6-8

The global incidence of prostate cancer is rising, with Asian populations demonstrating upward trends similar to those established in Western and African countries.^9,10 The incidence of thromboembolism (TE) in prostate cancer is lower than in other malignancies, with a recent nationwide database study reporting a rate of approximately 3.3%.¹¹ The rate is even lower in locally advanced disease, with a 2-year cumulative VTE incidence of about 1%.¹² However, this risk profile is largely derived from Western cohorts. VTE incidence is generally considered to be lower in Asians compared to White populations.^13-15 As a result, perioperative risk models developed using Western databases may not accurately predict thrombotic risk in Asian patients, potentially leading to inappropriate use of anticoagulation therapies.

Despite the growing burden of prostate cancer in Asia, granular data regarding VTE in East Asian patients with localized disease remains scarce. Determining the precise incidence and specific risk factors in this population is crucial for optimizing thromboprophylaxis protocols. In this study, we aimed to investigate the incidence, risk factors, and survival impact of TE in East Asian patients with localized prostate cancer who underwent surgery.

Methods

Patient Characteristics and Data Collection

This was a single-center retrospective cohort study of patients treated with surgical resection for localized prostate cancer. A total of 4880 patients who underwent open, laparoscopic, or robotic-assisted prostatectomy between March 1, 2002, and December 31, 2019, were included in the study. The study period was concluded in 2019 to ensure adequate follow-up for survival analysis and to exclude potential confounding effects associated with the COVID-19 pandemic. TE was defined as either VTE—includes deep venous thrombosis (DVT) and pulmonary embolism (PE)—or arterial embolism. All TE events were confirmed by imaging. Patients with a history of TE events prior to prostatectomy were excluded to identify only incident cases.

The results of laboratory testing that was performed within 1 month before surgery were collected. We used the Khorana Score cutoff values for hemoglobin levels, leukocyte counts, and platelet counts to predict TE.¹⁶ Surgical specimens were reviewed by our institution’s pathologist to confirm the diagnoses. Gleason scores were used to assess tumor grades. Score of 8 or higher were classified to be high Gleason scores. As there is no universal consensus for the optimal neutrophil-to-lymphocyte ratio (NLR) cutoff value, this study used a cutoff of 1.5, which was defined using maximally selected rank statistics.¹⁷ This approach determines the optimal cutpoint that provides the maximum standard log-rank statistic for predicting time-to-event outcomes. Conversely, the thresholds for C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) were defined empirically based on the established clinical upper limits of normal (ULN) utilized by our institution’s laboratory.

Statistical Analysis

Overall survival (OS) was analyzed using the Kaplan-Meier method, and survival curves were compared using the log-rank test. OS was defined as the time between the date of surgery and the date of death from any cause. Patients who were alive at the last follow-up were censored. A Cox proportional hazards model was applied to assess the associations between prognostic factors and survival, and hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated. A two-sided P value of less than 0.05 was considered statistically significant. For multivariable analysis, candidate covariates were initially reviewed based on clinical relevance and prior knowledge. Following univariable analysis, variables were further assessed to prevent the inclusion of conceptual overlapping domains. The final multivariable models were constructed by prioritizing clinically important covariates and potential confounders rather than relying solely on statistical significance. To confirm that the remaining covariates in the final models were mutually independent, Variance Inflation Factor (VIF) diagnostics were performed to assess and exclude multicollinearity.

A Fine-Gray subdistribution hazards model was employed to analyze the incidence of TE events, treating death as a competing risk.¹⁸ While the traditional Cox model is accurate for overall survival estimation,¹⁹ it treats competing events as independently censored observations, which frequently overestimates true TE incidence. The Fine-Gray model resolves this by estimating the subdistribution hazard, directly accounting for death as a competing event rather than simply censoring it. Subdistribution HRs (SHRs) with 95% CIs were calculated for relevant variables. P-values for each covariate were determined using a Wald test.

To verify the robustness of our primary findings, additional stratification and sensitivity analyses were performed. To evaluate the consistency of the overall survival models, stratified survival analyses were conducted across baseline tumor burden (pathological T-stage) and histological grade (Gleason score) subgroups. For the thromboembolism and venous thromboembolism analyses, a Cause-Specific Cox model was utilized as a competing-risk sensitivity framework to validate the primary Fine-Gray subdistribution hazard models.

Statistical analysis was performed using R version 4.2.2 (The R Foundation for Statistical Computing, Vienna, Austria. https://www.r-project.org). The study protocol was reviewed and approved by the Institutional Review Board (IRB) of Seoul National University Hospital (IRB No. H-2109-097-1255). The study was conducted in accordance with the Declaration of Helsinki pertaining to biomedical research involving human subjects.

Results

Patient Baseline Characteristics

A total of 4,880 patients underwent surgery from March 2002 to December 2019. Open prostatectomy was performed in 2010 (41.2%) patients, laparoscopic prostatectomy in 78 (1.6 %) patients, and robotic-assisted prostatectomy in 2778 (56.9%) patients. The baseline characteristics of the patients are summarized in Table 1.

Table 1.

Baseline Patient Characteristics

Characteristic	Criteria	Value
Age, median (range), years		68 (37-89)
BMI, n (%)	≥23	2042 (41.8)
BMI, n (%)	<23	2838 (58.2)
DM, n (%)	Diagnosed	697 (14.3)
	Negative	3356 (68.8)
	Not evaluated	827 (16.9)
Hypertension, n (%)	Diagnosed	2288 (46.9)
Hypertension, n (%)	Negative	2592 (53.1)
Dyslipidemia, n (%)	Diagnosed	695 (14.2)
Dyslipidemia, n (%)	Negative	4185 (85.8)
Gleason’s score, n (%)	5	5 (0.1)
	6	1103 (22.6)
	7	2955 (60.6)
	8	291 (6.0)
	9	311 (6.4)
	10	8 (0.2)
	Not evaluated	207 (4.2)
Pathologic T stage, n (%)	2	2742 (56.2)
	3	2012 (41.2%)
	4	4 (0.1%)
	Not evaluated	122 (2.5%)
Extraprostatic extension, n (%)	Positive	1926 (39.5)
	Negative	2792 (57.2)
	Not evaluated	162 (3.3)
Seminal vesicle invasion, n (%)	Positive	673 (13.8)
	Negative	3959 (81.1)
	Not evaluated	248 (5.1)
Pathologic N stage, n (%)	1	213 (4.4)
	0	1262 (25.9)
	X	3155 (64.7)
	Not evaluated	250 (5.1)
Surgical margin positivity, n (%)	Positive	1818 (37.3)
	Negative	1547 (31.7)
	Not evaluated	1515 (31.0)
Hb, median (range), g/dl		14.2 (8.2-17.9)
Platelet, median (range), 10⁹/L		210 (36 – 815)
WBC, median (range), 10⁹/L		5.97 (2.03-18.44)
NLR, median (range)		1.80 (0.25-898.00)
CRP, median (range), mg/dl		0.07 (0.01 – 30.65)
ESR, median (range), mm/hr		10 (1-120)
PSA, median (range), ng/ml		7.840 (0.002-332.400)

Abbreviations: BMI, body mass index; DM, diabetes; Hb, hemoglobin; WBC, white blood cell count; NLR, neutrophil-to-lymphocyte ratio; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; PSA, prostate-specific antigen.

Incidence and Risk Factors Associated With TE

For the entire cohort of 4,880 patients, the median follow-up duration for TE incidence was 43.1 months (95% CI, 41.7–44.8 months). The 3-month, and 2-year cumulative incidences of TE were 0.33% and 0.81%, respectively (Figure 1A). For VTE, the 3-month and 2-year cumulative incidences were 0.16% and 0.29%, respectively (Figure 1B).

Figure 1.

Cumulative incidence of localized prostate cancer (A) cumulative incidence of thromboembolism (B) cumulative incidence of venous thromboembolism

In the univariable analysis, older age (≥75 years), dyslipidemia, elevated NLR (≥1.5), elevated C-reactive protein (CRP) (≥0.35 mg/dL), and elevated erythrocyte sedimentation rate (ESR) (≥15 mm/hr) were associated with a significantly increased incidence of TE (Table 2 and Figure 2). To prevent redundancy between markers of systemic inflammation, NLR was excluded from the multivariable model in favor of CRP. ESR values were excluded from the multivariable analysis due to a high proportion of missing values. VIF values for remaining variables were below the threshold of 5 (Range: 1.002 – 1.005, Table S1). In the multivariable analysis, older age (SHR 2.16, P = 0.0155) and an elevated CRP level (SHR 2.91, P = 0.0001) were significant risk factors for an increased incidence of TE. In a sensitivity analysis using a cause-specific hazards model with the same covariates as the primary Fine–Gray multivariable analysis, the results were directionally consistent and showed no meaningful change in overall interpretation.

Table 2.

Fine–Gray Subdistribution Hazard Analysis of TE

Variable	Reference category	3-month TE events (%)	2-year TE events (%)	SHR	95% CI	P-value
Baseline
Age	<75	0.91	2.19	2.16	1.32–3.55	0.0023
BMI	<23	0.29	0.83	1.07	0.71–1.62	0.7514
Comorbidity
DM	Negative	0.14	0.72	0.99	0.51–1.92	0.9650
Hypertension	Negative	0.44	0.82	1.44	0.95–2.17	0.0837
Dyslipidemia	Negative	0.86	1.94	1.71	1.04–2.81	0.0331
Pathology
Gleason score	<8	0.49	0.70	0.55	0.25–1.20	0.1335
T staging	Stage 2	0.50	0.98	1.16	0.75–1.77	0.5066
Pathology EPE	Negative	0.52	1.02	1.29	0.84–1.97	0.2406
Pathology SVI	Negative	0.59	1.14	0.74	0.38–1.44	0.3805
N staging	Negative	1.30	1.83	1.13	0.45–2.82	0.7932
Lymphatic invasion	Negative	0.59	1.05	0.57	0.28–1.17	0.1236
Venous invasion	Negative	1.45	3.18	2.37	0.73–7.70	0.1501
Perineural invasion	Negative	0.38	0.67	1.01	0.63–1.62	0.9619
Margin involved	Negative	0.50	1.15	1.20	0.64–2.26	0.5641
Clinical variable
Hb	≥10 g/dl	3.33	3.33	2.68	0.66–10.88	0.1688
Platelet	<350 x10⁹/L	0.00	1.22	2.14	0.79–5.84	0.1355
WBC	≤11 x10⁹/L	0.00	0.00	<0.01	<0.01–<0.01	<0.0001
NLR	<1.5	0.52	1.11	1.99	1.16–3.42	0.0130
CRP	<0.35mg/dl	0.53	2.44	2.90	1.68–5.00	0.0001
ESR	<15mm/hr	0.35	1.63	2.17	1.24–3.79	0.0064
PSA	<20ng/ml	0.54	1.28	1.40	0.83–3.55	0.1416
Adjuvant ADT	Not prescribed	0.91	0.91	1.19	0.55–2.59	0.6622
Recurrence	Negative	0.29	0.62	0.53	0.31–0.91	0.0225
Distant metastasis	Negative	1.12	1.12	0.85	0.26–2.79	0.7823
Open prostatectomy	Laparoscopic or Robotic surgery	0.35	0.94	0.76	0.48–1.21	0.2495

Variable	Reference category	SHR	95% CI	P-value
Multivariable analysis
Age	<75	2.16	1.16-4.04	0.0155
Dyslipidemia	Negative	1.44	0.77-2.69	0.2516
CRP	<0.35mg/dl	2.91	1.68-5.03	0.0001

Abbreviations: TE, thromboembolism; SHR, subdistribution hazard ratio; CI, confidence interval; BMI, body mass index; DM, diabetes; Hb, hemoglobin; WBC, white blood cell count; NLR, neutrophil-to-lymphocyte ratio; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; PSA, prostate-specific antigen; ADT, androgen deprivation therapy.

Figure 2.

Cumulative incidence of thromboembolism (TE) among risk factors (A) comparison by age (≥75, <75) (B) comparison by dyslipidemia (positive, negative) (C) comparison by CRP (≥0.35, <0.35)

In addition, older age was significantly associated with an increased risk of VTE. Pathological venous invasion, distant metastasis, and elevated prostate-specific antigen levels (≥20 ng/mL) were associated with significant increases in VTE events (Table 3 and Figure 3). In the subsequent multivariable analysis, older age (SHR 3.67, P = 0.0113), and pathological venous invasion (SHR 5.13, P = 0.0195) remained significant risk factors for VTE. VIF values for remaining variables were below the threshold of 5 (Range: 1.041 – 1.195, Table S1). Likewise, in the above TE model, findings from a separate sensitivity analysis using a cause-specific hazards model with the same covariates as the VTE Fine–Gray multivariable analysis were generally consistent and did not materially change the overall interpretation.

Figure 3.

Cumulative incidence of venous thromboembolism (VTE) among risk factors

No clear association was found between the use of androgen deprivation therapy (ADT) or the type of surgery and the risk of either TE or VTE. Pathological venous invasion was associated with the highest SHR and a steep increase in cumulative VTE incidence (Figure 3B).

Table 3.

Fine–Gray Subdistribution Hazard Analysis of VTE

Variable	Reference category	3-month VTE events (%)	2-year VTE events (%)	SHR	95% CI	P-value
Baseline
Age	<75	0.30	0.69	2.97	1.30–6.76	0.0098
BMI	<23	0.24	0.47	1.78	0.85–3.74	0.1275
Comorbidity
DM	Negative	0.14	0.14	0.58	0.14–2.50	0.4693
Hypertension	Negative	0.22	0.32	1.37	0.65–2.88	0.4140
Dyslipidemia	Negative	0.43	0.78	1.66	0.67–4.07	0.2727
Pathology
Gleason score	<8	0.49	0.70	1.15	0.39–3.34	0.8034
T staging	Stage 2	0.35	0.48	1.63	0.75–3.50	0.2149
Pathology EPE	Negative	0.36	0.50	1.73	0.80–3.73	0.1647
Pathology SVI	Negative	0.59	0.97	1.61	0.64–4.03	0.3079
N staging	Negative	1.30	1.30	2.36	0.71–7.89	0.1630
Lymphatic invasion	Negative	0.39	0.61	0.70	0.20–2.50	0.5857
Venous invasion	Negative	1.45	3.18	5.08	1.17–22.02	0.0297
Perineural invasion	Negative	0.23	0.37	1.36	0.61–3.05	0.4572
Margin involved	Negative	0.33	0.46	1.79	0.64–5.04	0.2684
Clinical variable
Hb	≥10 g/dl	3.33	3.33	5.01	0.64–39.36	0.1255
Platelet	<350 x10⁹/L	<0.01	<0.01	1.88	0.25–14.12	0.5400
WBC	≤11 x10⁹/L	<0.01	<0.01	<0.01	<0.01-<0.01	<0.0001
NLR	<1.5	0.20	0.32	1.78	0.67–4.74	0.2495
CRP	<0.35mg/dl	0.53	1.16	2.04	0.65–6.48	0.2243
ESR	<15mm/hr	0.23	0.62	2.92	0.96–8.83	0.0583
PSA	<20ng/ml	0.54	1.28	3.41	1.17–9.96	0.0247
Adjuvant ADT	Not prescribed	0.91	0.91	2.42	0.84–6.97	0.1022
Recurrence	Negative	0.10	0.20	0.92	0.40–2.12	0.8377
Distant metastasis	Negative	1.12	1.12	3.71	1.06–13.02	0.0409
Open prostatectomy	Laparoscopic or Robotic surgery	0.10	0.31	0.74	0.31–1.74	0.4845

Variable	Reference category	SHR	95% CI	P-value
Multivariable analysis
Age	<75	3.67	1.34–10.02	0.0113
Venous invasion	Negative	5.13	1.30–20.21	0.0195
Distant metastasis	Negative	1.87	0.23–15.34	0.5610
PSA	<20ng/ml	2.56	0.74–8.86	0.1382

Abbreviations: VTE, venous thromboembolism; SHR, subdistribution hazard ratio; CI, confidence interval; BMI, body mass index; DM, diabetes; Hb, hemoglobin; WBC, white blood cell count; NLR, neutrophil-to-lymphocyte ratio; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; PSA, prostate-specific antigen; ADT, androgen deprivation therapy.

Survival Outcomes

At a median follow-up of 47.8 months (95% CI, 47.2–48.2) for the survival analysis, the 5-year OS was 92.7% (95% CI 91.7%–93.7%). In univariable analysis older age, diabetes, hypertension, Gleason scores ≥8, T stages of 3 or 4, pathological extraprostatic extension, pathological seminal vesicle invasion, positive N stage, pathological perineural invasion, low hemoglobin levels (<10 g/dL), elevated NLR, elevated CRP, elevated ESR, the use of adjuvant ADT, distant metastasis, the occurrence of TE, and the occurrence of VTE were prognostic factors for poor OS (Table 4, Figure 4). The 5-year OS rate was significantly lower in patients who developed VTE compared to those who did not (63.3% vs. 92.9%, respectively), representing a 29.6 percentage point decrease.

Table 4.

Cox Proportional Hazards Regression Model for Overall Survival

Variable	Reference category	5-year OS (vs. reference %)	HR	95% CI	P-value
Baseline
Age	<75	83.1 vs 93.9	3.38	2.72-4.19	<0.0001
BMI	<23	93.4 vs 92.2	0.78	0.64-0.94	0.0108
Comorbidity
DM	Negative	86.5 vs 93.8	1.85	1.41-2.43	<0.0001
Hypertension	Negative	91.8 vs 93.5	1.40	1.17-1.68	0.0002
Dyslipidemia	Negative	92.4 vs 92.7	1.12	0.86-1.44	0.4067
Pathology
Gleason score	<8	86.9 vs 94.0	1.55	1.22-1.96	0.0003
T staging	Stage 2	90.8 vs 94.5	1.48	1.23-1.77	<0.0001
Pathology EPE	Negative	90.9 vs 94.5	1.46	1.21-1.75	0.0001
Pathology SVI	Negative	86.7 vs 94.3	1.86	1.49-2.33	<0.0001
N staging	Negative	88.4 vs 92.9	1.52	1.06-2.20	0.0234
Lymphatic invasion	Negative	90.6 vs 93.1	0.88	0.69-1.11	0.2805
Venous invasion	Negative	82.9 vs 93.1	1.49	0.82-2.70	0.1949
Perineural invasion	Negative	91.7 vs 94.4	1.40	1.14-1.73	0.0016
Margin involved	Negative	91.6 vs 93.1	0.84	0.61-1.16	0.2812
Clinical variable
Hb	≥10 g/dl	84.9 vs 92.9	3.18	1.70-5.96	0.0003
Platelet	<350 x10⁹/L	84.3 vs 93.0	1.29	0.77-2.17	0.3377
WBC	≤11 x10⁹/L	92.7 vs 92.9	0.95	0.45-2.01	0.8837
NLR	<1.5	92.0 vs 94.3	1.33	1.07-1.65	0.0103
CRP	<0.35mg/dl	90.0 vs 93.3	1.43	1.04-1.96	0.0270
ESR	<15mm/hr	89.5 vs 95.1	1.77	1.36-2.31	<0.0001
PSA	<20ng/ml	88.9 vs 93.4	1.31	0.91-1.87	0.1470
Adjuvant ADT	Not prescribed	90.5 vs 92.8	1.49	1.07-2.08	0.0191
Recurrence	Negative	90.9 vs 93.3	1.10	0.91-1.33	0.3404
Distant metastasis	Negative	78.2 vs 93.0	1.53	1.08-2.17	0.0179
Open prostatectomy	Laparoscopic or Robotic surgery	90.9 vs 95.2	1.10	0.87-1.37	0.4203
Thrombosis
TE	Negative	80.5 vs 93.0	2.36	1.71-3.27	<0.0001
VTE	Negative	63.3 vs. 92.9	4.20	2.46-7.15	<0.0001

Variable	Reference category	HR	95% CI	P-value
Multivariable analysis
Age	<75	3.66	2.49-5.38	<0.0001
DM	Negative	1.69	1.10-2.60	0.0172
Hypertension	Negative	1.01	0.71-1.43	0.9607
Gleason score	<8	1.69	1.01-2.83	0.0458
T staging	Stage 2	1.29	0.85-1.96	0.2370
N staging	Negative	1.37	0.65-2.91	0.4066
Perineural invasion	Negative	0.86	0.57-1.30	0.4725
Hb	≥10 g/dl	1.69	0.73-3.92	0.2186
CRP	<0.35mg/dl	1.50	0.95-2.35	0.0793
Adjuvant ADT	Not prescribed	0.62	0.26-1.48	0.2823
Distant metastasis	Negative	0.91	0.38-2.18	0.8354
VTE	Negative	5.72	1.98-16.59	0.0013

Abbreviations: TE, thromboembolism; VTE, venous thromboembolism; HR, hazard ratio; CI, confidence interval; BMI, body mass index; DM, diabetes; Hb, hemoglobin; WBC, white blood cell count; NLR, neutrophil-to-lymphocyte ratio; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; PSA, prostate-specific antigen; ADT, androgen deprivation therapy.

Figure 4.

Kaplan-meier analysis of overall survival following thromboembolic events (A) overall survival stratified by the occurrence of any thromboembolism (B) overall survival stratified by the occurrence of venous thromboembolism

Prior to multivariable modeling, variables were clinically evaluated to prevent the inclusion of redundant, overlapping domains. Consistent with the TE analysis, to avoid redundancy among systemic inflammatory markers, NLR was excluded instead of CRP, which showed higher HR in univariable analysis and is the more universally standardized laboratory test. Similarly, TE was excluded to prevent clinical overlap with VTE, as VTE demonstrated a higher HR and represents a more specific clinical focus. Pathological extraprostatic extension positivity, and pathological seminal vesicle invasion were excluded due to their definitional clinical overlap with T stages 3 or 4 tumors variable. ESR variable was not included due to a high proportion of missing values. All included variables demonstrated VIF values well below the threshold of 5 (Range: 1.047 – 1.497, Table S1). As a result, older age, diabetes, high Gleason scores, and the occurrence of VTE were independently associated with worse survival outcomes (HR 3.66, P < 0.0001 for age; HR 1.69, P = 0.0172 for diabetes; HR 1.69, P = 0.0458 for Gleason scores; and HR 5.72, P = 0.0013 for VTE; Table 4).

To determine whether the prognostic impact of VTE on overall survival persists independently of baseline tumor burden, an additional stratified survival analysis was performed. VTE remained a strong, independent predictor of mortality across oncologic subgroups. Specifically, the deleterious effect of VTE was highly significant when stratifying by pathological T-stage (both <pT3 and ≥pT3) and by histological grade (both Gleason <8 and Gleason ≥8) (Table S2, Figure S1).

Discussion

In this large cohort study, we demonstrated that the 3-month and 2-year cumulative incidences were 0.33% and 0.81%, respectively, for TE and 0.16% and 0.29%, respectively, for VTE. Regarding VTE only, there were 18 cases of DVT (0.36%), and 10 cases of PE (0.20%). VTE was associated with inferior OS. Older age and elevated CRP levels were significant risk factors for TE in prostate cancer patients undergoing surgery. In addition, older age and pathological venous invasion were associated with an increased risk of VTE.

Prostate cancer is generally considered a malignancy with a relatively low incidence of TE compared to other cancers.¹⁵ The total incidence in our cohort was low, remaining below the approximately 1% rate reported in recent studies.^12,15,20,21 TE incidence varies largely among different ethnic and racial groups, with a lower prevalence of VTE in Asians.¹⁴ While the etiology is likely multifactorial, this disparity is largely attributed to the scarcity of thrombophilic genetic polymorphisms in East Asian populations compared to Western cohorts. Specifically, Factor V Leiden and Prothrombin G20210A mutations, which are prevalent in Western cohorts, are virtually absent in this demographic.²² However, caution regarding TE events is warranted in patients with risk factors. Numerous studies have reported that TE is associated with increased mortality in cancer patients,^23-25 and our study similarly confirmed this relationship. This underscores the importance of careful monitoring for TE, particularly in this high-risk population.

Previous studies conducted in prostate cancer patients undergoing surgery have mainly examined differences in TE events according to the type of surgery. LAPPRO cohort study of 3,544 prostate cancer patients undergoing surgery found that lymph node dissection during radical prostatectomy was associated with a significantly increased risk of DVT and PE, particularly after open surgery. Gleason score ≥8, pT4 stage, and a history of VTE was reported as critical independent predictors of thrombosis.²⁶ In contrast, our study found no significant difference in TE events between open and minimally invasive prostatectomy.

Substantial evidence supports that systemic inflammation increases thrombotic risk.²⁷ Aligning with our findings, CRP is associated with a higher risk of thrombosis in patients with cancer.^28-30 Elevated CRP provides a key connection between inflammation and thrombosis by activating platelets, inducing tissue factor (TF) expression, modulating the coagulation cascade, and inhibiting fibrinolysis.³¹ Furthermore, proinflammatory cytokines such as tumor necrosis factor–alpha (TNF-α) and interleukins exacerbate this prothrombotic state by inducing the expression of TF and von Willebrand factor (vWF) from vascular endothelial cells.^32,33

Advanced age and high tumor burden are well known risk factors for TE in various cancers.^14,34,35 Accordingly, our study demonstrated that age above 75 years significantly increased the risk of both TE and VTE. While previous prostate cancer studies have linked conventional tumor burden markers—such as metastasis and high Gleason score—to increased TE risk,³⁶ our analysis did not reflect this association. Instead, VTE occurred significantly more frequently in patients with pathological venous invasion. Mechanistically, direct tumor cell invasion into the vasculature likely initiates thrombosis via endothelial injury and the immediate systemic release of procoagulant factors, including tissue factor-bearing microparticles (MPs) and vWF.^33,37-40 To our knowledge, this is the first study to report an association between venous invasion on pathological review and an increased risk of VTE in prostate cancer.

Our study has several limitations. First, as a single-center retrospective study, inherent selection bias and missing data for certain variables may have influenced the outcomes. Second, the low incidence of TE and VTE limited the statistical power of our analyses. Despite these limitations, our study is meaningful in that it represents a large cohort study conducted in an East Asian population. Also, we focused on patients with localized prostate cancer who underwent curative surgery. The findings suggest that even in this narrowly defined patient population, the occurrence of venous thrombosis significantly impacts OS.

In conclusion, TE occurred in a relatively small proportion of surgically treated East Asian prostate cancer patients; however, it was associated with increased mortality. Given the 0.4% to 1.4% risk of major bleeding associated with pharmacological interventions, mechanical prophylaxis is a reasonable approach for the majority of low-risk patients.⁴¹ Nevertheless, patients with risk factors such as advanced age or elevated CRP levels had a significantly higher incidence of TE, indicating the need for careful postoperative monitoring in these high-risk individuals. Further studies integrating these baseline clinical predictors with advanced data science and machine learning algorithms may facilitate the development of highly accurate, individualized prognostic models for predicting thromboembolic outcomes.^42,43

Supplemental Material

Supplemental Material - Thromboembolic Events After Surgery for Localized Prostate Cancer: Incidence, Risk Factors, and Survival Outcomes

Supplemental Material for Thromboembolic Events After Surgery for Localized Prostate Cancer: Incidence, Risk Factors, and Survival Outcomes by Yongho Noh, Hyunkyung Park, Dong-Yeop Shin, Junshik Hong, Hyunsoo Cho, Ja Min Byun, Youngil Koh, Seung-Hwan Jeong, Hyeongdong Yuk, Sung‐Yong Cho, Chang Wook Jeong, Ja Hyeon Ku, Cheol Kwak and Inho Kim in Clinical and Applied Thrombosis/Hemostasis.

Supplemental Material

Supplemental Material - Thromboembolic Events After Surgery for Localized Prostate Cancer: Incidence, Risk Factors, and Survival Outcomes

Supplemental Material

Supplemental Material - Thromboembolic Events After Surgery for Localized Prostate Cancer: Incidence, Risk Factors, and Survival Outcomes

Footnotes

ORCID iD

Yongho Noh

Ethical Considerations

The study protocol was reviewed and approved by the Institutional Review Board (IRB) of Seoul National University Hospital (IRB No. H-2109-097-1255).

Consent to Participate

Consent to participate has been waived by Seoul National University Hospital IRB.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data used to support the findings of this study are available from the corresponding author upon reasonable request.*

Supplemental Material

Supplemental material for this article is available online.

References

Heit

Silverstein

Mohr

Petterson

O'Fallon

Melton

3rd . Risk factors for deep vein thrombosis and pulmonary embolism: a population-based case-control study. Arch Intern Med. 2000;160(6):809-815. doi:10.1001/archinte.160.6.809.

Sørensen

Mellemkjaer

Olsen

Baron

. Prognosis of cancers associated with venous thromboembolism. N Engl J Med. 2000;343(25):1846-1850. doi:10.1056/nejm200012213432504.

Lyman

Culakova

Poniewierski

Kuderer

. Morbidity, mortality and costs associated with venous thromboembolism in hospitalized patients with cancer. Thromb Res. 2018;164(Suppl 1):S112-s118. doi:10.1016/j.thromres.2018.01.028.

Bergqvist

Agnelli

Cohen

, et al. Duration of prophylaxis against venous thromboembolism with enoxaparin after surgery for cancer. N Engl J Med. 2002;346(13):975-980. doi:10.1056/NEJMoa012385.

Khorana

Kuderer

McCrae

, et al. Cancer associated thrombosis and mortality in patients with cancer stratified by khorana score risk levels. Cancer Med. 2020;9(21):8062-8073. doi:10.1002/cam4.3437.

Lyman

Carrier

, et al. American Society of Hematology 2021 guidelines for management of venous thromboembolism: prevention and treatment in patients with cancer. Blood Adv. 2021;5(4):927-974. doi:10.1182/bloodadvances.2020003442.

Kirschner

do Ó Hartmann

Parmentier

, et al. Primary Thromboprophylaxis in Patients with Malignancies: Daily Practice Recommendations by the Hemostasis Working Party of the German Society of Hematology and Medical Oncology (DGHO), the Society of Thrombosis and Hemostasis Research (GTH), and the Austrian Society of Hematology and Oncology (ÖGHO). Cancers (Basel). 2021;13(12):2905. doi:10.3390/cancers13122905.

Network

NCC

. NCCN Clinical Practice Guidelines in Oncology: Cancer-Associated Venous Thromboembolic Disease, Version 3.2025. National Comprehensive Cancer Network. https://www.nccn.org/guidelines/category_3. Accessed 8 January 2026.

Bray

Laversanne

Sung

, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229-263. doi:10.3322/caac.21834.

10.

Tsodikov

Gulati

Heijnsdijk

EAM

, et al. Reconciling the Effects of Screening on Prostate Cancer Mortality in the ERSPC and PLCO Trials. Ann Intern Med. 2017;167(7):449-455. doi:10.7326/m16-2586.

11.

Devarkonda

Ali

Kumar

, et al. Thromboembolic incidents in prostate cancer patients: A nationwide propensity-matched study. American Society of Clinical Oncology; 2024.

12.

Timp

Braekkan

Versteeg

Cannegieter

. Epidemiology of cancer-associated venous thrombosis. Blood. 2013;122(10):1712-1723. doi:10.1182/blood-2013-04-460121.

13.

White

Zhou

Murin

Harvey

. Effect of ethnicity and gender on the incidence of venous thromboembolism in a diverse population in California in 1996. Thrombosis and haemostasis. 2005;93(02):298-305.

14.

White

Keenan

. Effects of race and ethnicity on the incidence of venous thromboembolism. Thromb Res. 2009;123(Suppl 4):S11-S17. doi:10.1016/s0049-3848(09)70136-7.

15.

Mahajan

Brunson

Adesina

Keegan

THM

Wun

. The incidence of cancer-associated thrombosis is increasing over time. Blood Adv. 2022;6(1):307-320. doi:10.1182/bloodadvances.2021005590.

16.

Khorana

Kuderer

Culakova

Lyman

Francis

. Development and validation of a predictive model for chemotherapy-associated thrombosis. Blood. 2008;111(10):4902-4907. doi:10.1182/blood-2007-10-116327.

17.

Hothorn

Lausen

. On the exact distribution of maximally selected rank statistics. Computational Statistics & Data Analysis. 2003;43(2):121-137.

18.

Fine

Gray

. A Proportional Hazards Model for the Subdistribution of a Competing Risk. Journal of the American Statistical Association. 1999;94(446):496-509. doi:10.2307/2670170.

19.

Kraujalis

Ruzgas

Milonas

. Mortality rate estimation models for patients with prostate cancer diagnosis. Baltic journal of modern computing. 2022;10(2):170-184.

20.

Wani

Al-Mitwalli

Mukherjee

, et al. Venous Thromboembolism (VTE) in Post-Prostatectomy Patients: Systematic Review and Meta-Analysis. J Clin Med. 2023;12(12):3979. doi:10.3390/jcm12123979.

21.

Tikkinen

Craigie

Agarwal

, et al. Procedure-specific risks of thrombosis and bleeding in urological cancer surgery: systematic review and meta-analysis. European urology. 2018;73(2):242-251.

22.

Kang

Zhang

Lesmana

, et al. Venous thromboembolism laboratory testing (factor V Leiden and factor II c.∗97G>A), 2025 revision: A technical standard of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2025;27(8):101466. doi:10.1016/j.gim.2025.101466.

23.

Falanga

Di Nisio

, et al. Venous thromboembolism in cancer patients: ESMO Clinical Practice Guideline. Ann Oncol. 2023;34(5):452-467. doi:10.1016/j.annonc.2022.12.014.

24.

Sørensen

Pedersen

van Es

Büller

Horváth-Puhó

. Impact of venous thromboembolism on the mortality in patients with cancer: a population-based cohort study. Lancet Reg Health Eur. 2023;34:100739. doi:10.1016/j.lanepe.2023.100739.

25.

Chaturvedi

Sidana

Elson

Khorana

McCrae

. Symptomatic and incidental venous thromboembolic disease are both associated with mortality in patients with prostate cancer. PLoS One. 2014;9(8):e94048. doi:10.1371/journal.pone.0094048.

26.

Tyritzis

Wallerstedt

Steineck

, et al. Thromboembolic complications in 3,544 patients undergoing radical prostatectomy with or without lymph node dissection. J Urol. 2015;193(1):117-125. doi:10.1016/j.juro.2014.08.091.

27.

Ding

Yue

Tian

Liao

Meng

Zou

. Association between inflammatory biomarkers and venous thromboembolism: a systematic review and meta-analysis. Thromb J. 2023;21(1):82. doi:10.1186/s12959-023-00526-y.

28.

Guan

Gan

, et al. C-Reactive Protein, Genetic Susceptibility, and the Long-Term Risk of Venous Thromboembolism in Patients with Past Cancer. Thrombosis and Haemostasis. 2024.

29.

Moik

Riedl

Barth

, et al. Early change in C-reactive protein and venous thromboembolism in patients treated with immune checkpoint inhibitors. Cardio Oncology. 2024;6(6):965-975.

30.

Kröger

Weiland

Ose

, et al. Risk factors for venous thromboembolic events in cancer patients. Ann Oncol. 2006;17(2):297-303. doi:10.1093/annonc/mdj068.

31.

Fay

. Linking inflammation and thrombosis: Role of C-reactive protein. World J Cardiol. 2010;2(11):365-369. doi:10.4330/wjc.v2.i11.365.

32.

Bevilacqua

Pober

Majeau

Fiers

Cotran

Gimbrone

Jr . Recombinant tumor necrosis factor induces procoagulant activity in cultured human vascular endothelium: characterization and comparison with the actions of interleukin 1. Proc Natl Acad Sci U S A. 1986;83(12):4533-4537. doi:10.1073/pnas.83.12.4533.

33.

Hamza

Mousa

. Cancer-Associated Thrombosis: Risk Factors, Molecular Mechanisms, Future Management. Clin Appl Thromb Hemost. 2020;26:1076029620954282. doi:10.1177/1076029620954282.

34.

Fimognari

Repetto

Moro

Gianni

Incalzi

. Age, cancer and the risk of venous thromboembolism. Critical reviews in oncology/hematology. 2005;55(3):207-212.

35.

Zhao

Liang

Ling

, et al. Prevalence and treatment of venous thromboembolism in patients with solid tumors. Experimental and Therapeutic Medicine. 2022;24(6):743.

36.

Ording

Horváth‐Puhó

Lash

, et al. Prostate cancer, comorbidity, and the risk of venous thromboembolism: A cohort study of 44,035 Danish prostate cancer patients, 1995‐2011. Cancer. 2015;121(20):3692-3699.

37.

Tatsumi

. The pathogenesis of cancer-associated thrombosis. International journal of hematology. 2024;119(5):495-504.

38.

Piazza

Venous Thromboembolism and Cancer. Circulation. 2013;128(24):2614–2618. doi:doi:10.1161/CIRCULATIONAHA.113.002702

39.

Canonico

Santoro

Avvedimento

, et al. Venous thromboembolism and cancer: a comprehensive review from pathophysiology to novel treatment. Biomolecules. 2022;12(2):259.

40.

Thomas

Brill

Mezouar

, et al. Tissue factor expressed by circulating cancer cell-derived microparticles drastically increases the incidence of deep vein thrombosis in mice. J Thromb Haemost. 2015;13(7):1310-1319. doi:10.1111/jth.13002.

41.

Klaassen

Wallis

Lavallee

Violette

. Perioperative venous thromboembolism prophylaxis in prostate cancer surgery. World Journal of Urology. 2020;38(3):593-600.

42.

Subrahmanya

SVG

Shetty

Patil

, et al. The role of data science in healthcare advancements: applications, benefits, and future prospects. Ir J Med Sci. 2022;191(4):1473-1483. doi:10.1007/s11845-021-02730-z.

43.

Jin

Qin

Liang

, et al. Machine learning predicts cancer-associated deep vein thrombosis using clinically available variables. Int J Med Inform. 2022;161:104733. doi:10.1016/j.ijmedinf.2022.104733.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB

0.06 MB