Venous Thromboembolism in Women with Breast and Gynaecological Cancers: What Do We Know and What Should We Do?

Abstract

Cancer-associated thrombosis (CAT) is a major cause of morbidity and mortality in oncology. Female-specific cancers—including breast, ovarian, endometrial, and cervical malignancies—demonstrate distinct thrombotic profiles influenced by hormonal mechanisms, anatomical features, and treatment strategies. This narrative review synthesizes epidemiological evidence, mechanistic pathways, and clinical trial data to characterize CAT risk across these cancer types. Key modulators of thrombosis include hormonal exposures, tumor histology, inflammatory signaling, tissue factor expression, and therapeutic interventions such as surgery, chemotherapy, endocrine therapy, targeted agents, and immunotherapy. Despite growing understanding, current CAT risk prediction models insufficiently reflect the heterogeneity of female cancers and fail to integrate bleeding risks specific to gynecologic disease and treatment. Optimizing CAT prevention and management will require improved risk stratification tools, individualized thromboprophylaxis strategies, and incorporation of survivorship considerations, including fertility and contraception counseling. A multidisciplinary approach involving oncology, gynecology, and thrombosis specialists is essential to reduce the burden of CAT and improve outcomes for women with cancer.

Introduction

Female-specific cancers refer to malignancies that primarily or exclusively affect women due to their reproductive anatomy, hormonal environment, and genetic susceptibility. ¹ These include four of the ten most common cancers affecting women worldwide: breast, endometrial, cervical, and ovarian cancers. ¹ The latter three fall under the category of gynecological malignancies. According to the GLOBOCAN 2022 report from the International Agency for Research on Cancer, breast cancer ranks as the second most frequently diagnosed cancer globally, representing more than 10% of all cancer diagnoses. ¹ Among women, breast cancer is the most prevalent malignancy and remains the leading cause of cancer-related mortality (approximately 15% of cancer deaths), while cervical cancer ranks fourth in both incidence and mortality worldwide. ¹

Each of these cancers demonstrates distinct biological behavior, epidemiological patterns, risk factors, clinical presentations, and screening pathways, emphasizing the need for tailored prevention and early detection strategies. Although breast cancer is the most frequently diagnosed cancer in women, uterine and ovarian cancers represent major health challenges due to their frequent late-stage diagnosis and poorer prognosis. ² Improving awareness, early screening, and timely access to care may substantially reduce the burden of these malignancies.

Patients with cancer are significantly more likely to develop venous thromboembolism (VTE) than individuals without cancer.^3-6 Venous thromboembolism occurring in the context of cancer—referred to as cancer-associated thrombosis (CAT)—is associated with poor prognosis and reduced overall survival. ⁷ In both medical and surgical settings, CAT remains one of the leading causes of death in patients with malignancy, second only to cancer progression.^4,8 The risk of VTE in cancer patients is increasing steadily from 1.0% in 1997 to 3.4% of cancer patients in 2017. CAT generally represent thrombosis in the venous circulation in the first place, and regarding the risk of arterial thrombosis the latter is rather low in the population of female specific cancer. ⁹ As a result, this narrative review will focus on venous thrombosis.

In a study by Mulder et al, the hazard ratios for the development of CAT within 12 months following cancer diagnosis were reported as follows: 4.5 (3.9-5.1) for breast cancer, 10.8 (7.8-14.9) for uterine cancer, 20.6 (11.8-36.1) for cervical cancer, and 30.7 (21.0-45.1) for ovarian cancer. ¹⁰ More recent epidemiological analyses suggest a continued rise in CAT incidence, with evidence of a doubling of CAT risk during the course of treatment for breast, uterine, and ovarian cancers. ¹¹

While the role of biological sex and hormonal status in thrombosis is well established in non-cancer settings, this association remains insufficiently characterized among cancer patients, despite sex now being recognized as a key determinant of cardiovascular disease. ¹² Hormonal factors—particularly estrogen—appear to contribute substantially to these differences. Therefore, biological sex likely influences the pathogenesis, clinical presentation, and therapeutic response of thrombotic events in cancer.

In addition, treatment strategies used specifically in female cancers—including hormone therapy and extensive laparotomic surgery (notably in ovarian cancer)—may further increase thrombotic risk. These treatment modalities differ substantially from those used in non–female-specific malignancies, underscoring the unique clinical considerations associated with these cancers.

To better explore this complexity, we will first examine the physiopathology of CAT using a conceptual framework incorporating biological sex, cancer subtype, and treatment-specific factors. We will then discuss the current clinical management of CAT in the setting of female cancers and finally highlight the remaining gaps in knowledge and future research priorities.

A Multidimensional Model of Thrombotic Risk in Cancer: Integrating Sex-Related, Cancer-Related, and Treatment-Related Factors

Risk factors for thrombosis in cancer can be categorized into patient-related, cancer-related, treatment-related, and biomarker-based contributors. ¹³

Sex-Related Patient Factors

Hormonal mechanisms play a critical role in the development of endocrine-dependent malignancies such as breast, endometrial, and ovarian cancers. ¹³ Estrogens and progestogens regulate cellular proliferation and apoptosis within hormone-sensitive tissues, and prolonged unopposed estrogen exposure is strongly associated with carcinogenesis. ¹³ Risk factors contributing to sustained estrogenic stimulation include early menarche, late menopause, nulliparity, obesity, hormone replacement therapy, and polycystic ovary syndrome.

Genetic susceptibility also influences cancer risk. BRCA1 and BRCA2 mutations markedly increase the risk of breast and ovarian cancers, whereas Lynch syndrome represents a major hereditary predisposition for endometrial cancer.^14,15 Breast cancer incidence rises sharply after age 50 and peaks beyond age 65, ¹⁶ while endometrial cancer most commonly presents in postmenopausal individuals, peaking between ages 55 and 64. ¹⁷

Obesity significantly increases circulating estrogen levels and is strongly linked to both cancer development and cancer-associated thrombosis.^13,18 Indeed, 9% and 40% of patients with breast and endometrial cancer are obese.^13,18 A history of polycystic ovary syndrome, infertility, or prolonged unopposed estrogen therapy also contributes to risk.

Cervical cancer typically affects younger women, with most cases presenting between ages 35 and 44. ¹⁹ While obesity has minimal influence on cervical cancer risk, immunosuppression—including HIV infection—increases susceptibility. ²⁰ Persistent infection with high-risk HPV genotypes remains the strongest risk factor, with smoking acting as an additional modifiable contributor. Long-term oral contraceptive use is a confounding but relevant factor due to its association with both cervical cancer and thrombosis risk. ²⁰

Ovarian cancer primarily affects women over age 50. ²¹ Familial predisposition, particularly BRCA1/2 mutations, nulliparity, and late maternal age, significantly increases risk. ²² Obesity is associated with certain ovarian cancer subtypes and may influence outcomes. ²¹

Except for cervical cancer, age >60 years and BMI ≥30 kg/m² are shared risk factors for cancer-associated thrombosis. Meta-analyses report an OR of 1.03 (95% CI 1.00-1.05) for age >60 and 1.31 (95% CI 1.20-1.50) for obesity. ²³ In elderly patients, increased thrombosis risk is attributed to reduced mobility, vascular dysfunction, increased blood viscosity, and prolonged postoperative recovery. Obesity similarly promotes a procoagulant state, characterized by reduced venous return and increased inflammatory and coagulation biomarkers, including elevated fibrinogen and D-dimer levels. However, studies specifically focused on gynecologic malignancies have shown inconsistent associations between BMI and thrombosis risk.^24-26

Cancer-Related Factors

Although men and women have comparable overall CAT incidence, risk varies markedly by cancer type.^10,27 Female-specific cancers—particularly ovarian, uterine, and breast cancers treated with chemotherapy—carry a disproportionately high burden of CAT. ¹⁰ Ovarian cancer consistently ranks among the malignancies with the highest CAT incidence.^8,22,27,28

Cancer-related determinants include tumor site, histological subtype, grade, stage, and time since diagnosis. ¹³ Advanced and metastatic disease confer the highest risk. ²⁹ The clinical status of cancer—active, remission, or recurrent—modifies risk; active or recurrent disease induces systemic inflammation, release of procoagulant mediators, and endothelial disruption, all contributing to a heightened thrombotic phenotype.^30,31

Breast Cancer

In a cohort study by Londero et al, the 3-month cumulative incidence of thrombosis was 0% in benign or in situ cases and 0.4% (95% CI 0.19-0.61%) in invasive breast cancer. ³² Among invasive cancer cases, cumulative incidence increased over time: 0.6% at 1 year, 0.9% at 2 years, 1.2% at 5 years, and 1.7% at 10 years. Approximately 15% of all CAT events occur in breast cancer patients. ³²

Risk increases in the presence of a central venous catheter and metastatic disease. In the SEER-Medicare cohort of 89 841 women, catheter presence was independently associated with VTE risk (OR 2.71; 95% CI 2.43-3.02). ³³ The CAVECCAS randomized study reported no significant benefit of thromboprophylaxis in preventing catheter-related thrombosis over six months. ³⁴

Cancer stage is also an independent risk factor, with progressively increasing odds ratios versus stage I disease (Stage II: OR 1.22 [95% CI 1.10-1.35]; Stage III: OR 1.39 [95% CI 1.20-1.62]; Stage IV: OR 1.98 [95% CI 1.68-2.33]). ³³

Ovarian Cancer

Ovarian cancer carries one of the highest thrombotic risks, with prevalence estimates ranging from 5.2% to 8.1% within two years of diagnosis.^24,27 Clear cell carcinoma subtype shows the highest risk (2.5 to 4 times more risk than other subtypes). ³⁵ CAT is associated with significantly reduced survival.^36-38 Advanced stage, high tumor grade, and the presence of ascites—reflecting peritoneal carcinomatosis—are key predictors. ³⁸

Cervical Cancer

Tumor burden strongly correlates with CAT. Tumors >50 mm confer a nine-fold higher risk of thrombotic events compared with smaller lesions (10% vs 1% incidence). ³⁸ CAT incidence reaches 28% in FIGO stage IV, compared with 3% in Stages I–III. ³⁸ Bulky nodal disease likely contributes through pelvic venous compression.

Endometrial Cancer

Endometrial cancer demonstrates a strong relationship between tumor histology and CAT. High-grade endometrioid subtypes exhibit a substantially higher 6-month thrombosis incidence compared with low-grade disease. ³⁹ Comorbidities such as obesity, diabetes, hypertension, and advanced age further magnify risk. ³⁹

Treatment-Related Factors

Several therapeutic interventions used in the management of female-specific cancers have been associated with an increased risk of cancer-associated thrombosis (CAT).^40,41

Surgery: Laparotomy and Laparoscopy

Laparotomy remains a cornerstone in the treatment of gynecologic cancers and is widely recognized as a major risk factor for CAT.^42-44 The risk of postoperative pulmonary embolism is at least twice as high in cancer patients compared with patients undergoing surgery for benign disease, and in gynecologic malignancies the risk may reach a 14-fold increase. ⁴⁴ Despite the systematic implementation of thromboprophylaxis, postoperative CAT still occurs in approximately 6%–7% of patients treated surgically for gynecologic cancer. ⁴⁴ Prolonged operative duration—particularly procedures exceeding 180 min—has been consistently identified as a risk-enhancing factor. ²⁴ Minimally invasive techniques are associated with lower CAT incidence, suggesting that surgical approach contributes meaningfully to risk stratification. ⁴⁴ Evidence from the ENOXACAN II trial demonstrated that extending postoperative thromboprophylaxis from 7–10 days to 28 days after major abdominal or pelvic cancer surgery reduced venous thromboembolism by more than half, with an absolute risk reduction of 7%–8% and a number needed to treat of approximately 12–14, underscoring the substantial clinical benefit of prolonged prophylaxis. ⁴⁵

Hyperthermic Intraperitoneal Chemotherapy (HIPEC), increasingly used in advanced ovarian cancer, combines cytoreductive surgery with heated intraperitoneal chemotherapy. Although survival benefits have been reported, the procedure may contribute to thrombotic risk by triggering a systemic inflammatory response resembling SIRS and promoting endothelial activation—key mechanisms in thrombogenesis. ⁴⁶ Surgical trauma, prolonged anesthesia, and perioperative immobility amplify this prothrombotic state. Current evidence suggests a modest but clinically relevant increase in postoperative thrombosis following HIPEC, underscoring the importance of individualized anticoagulation strategies ⁴⁶ (5.6% of CAT after HIPEC).

Chemotherapy

Chemotherapy remains a foundational component of treatment for breast, ovarian, and endometrial cancers. The mechanisms linking chemotherapy to thrombosis include endothelial injury, suppression of natural anticoagulant pathways, and activation of platelets and coagulation cascades. ⁴⁰ Platinum-based agents are particularly associated with elevated thrombotic risk.^40,41

In breast cancer, the Walker cohort demonstrated a 10-fold increase in CAT among patients receiving chemotherapy (HR 10.8; 95% CI 8.2-14.2). ⁴⁷ In metastatic endometrial cancer, chemotherapy was similarly associated with elevated risk (HR 2.33; 95% CI 1.38-3.95; p = .002). ³⁹

While HER2-targeted monoclonal antibodies such as trastuzumab and pertuzumab do not significantly increase CAT risk, agents targeting angiogenesis—particularly the VEGF pathway—pose substantial thrombotic concerns. ⁴⁸

Anti-VEGF Therapy

Bevacizumab, the only currently approved anti-VEGF agent for ovarian cancer, improves progression-free survival but increases both thrombotic and hemorrhagic complications.^48,49 A meta-analysis of 3608 patients demonstrated a statistically significant increase in CAT among those receiving bevacizumab in addition to standard chemotherapy. ⁴⁸ Proposed mechanisms include endothelial dysfunction, reduced nitric oxide bioavailability, and increased platelet adhesion, fostering a prothrombotic milieu. However, when considering all anti-VEGF agents across broader indications, the incremental thrombotic risk appears attenuated, suggesting potential trial-related confounding. ⁵⁰

Immune Checkpoint Inhibitors

Immune checkpoint inhibitors targeting PD-1, PD-L1, or CTLA-4 have transformed the therapeutic landscape in breast, cervical, and endometrial cancers. Current evidence does not demonstrate a significant increase in thrombotic risk with checkpoint blockade.^51,52 Unlike chemotherapy, these agents may indirectly mitigate thrombosis by reducing tumor burden and systemic inflammatory signaling. Nevertheless, individualized assessment remains essential given the multifactorial nature of CAT risk.^40,52,53

PARP Inhibitors

PARP inhibitors—including olaparib, niraparib, and rucaparib—are now standard of care in ovarian cancer, particularly in BRCA-mutated disease. ⁵⁴ Available data suggest a mild to moderate increase in CAT risk, although toxicity profiles are dominated by hematologic adverse events such as thrombocytopenia, which may obscure thrombotic risk signals. An ESMO Open analysis reported a discrete increase in CAT incidence (OR < 1.5); however, absolute rates remain low. Further investigation is needed to better define high-risk phenotypes and appropriate thromboprophylaxis strategies. ⁵⁵

Antibody-Drug Conjugates (ADCs)

ADCs represent a rapidly expanding therapeutic class. In breast cancer, trastuzumab deruxtecan and sacituzumab govitecan are established treatments, while mirvetuximab soravtansine and tisotumab vedotin are emerging options in ovarian, cervical, and endometrial cancers.^56,57 Although thrombotic events have not been systematically characterized, biological plausibility exists through endothelial injury, cytokine-driven inflammation, and immune activation. Ongoing clinical trials will clarify whether routine thromboprophylaxis should be considered in selected patients.^56,57

Endocrine Therapy

Endocrine therapy is standard in hormone-receptor–positive breast cancer, representing almost 80% of all breast cancer cases. ¹⁴ Adjuvant therapy is prescribed for a minimum of 5 years, and may extend to 10 years in high-risk disease. ¹⁴ In metastatic settings, duration may be indefinite.

Tamoxifen significantly increases CAT risk, particularly in women >50 years and in the postmenopausal setting.^58-61 “In a large Danish cohort of 16 289 women with early-stage, estrogen-receptor–positive breast cancer, the 5-year cumulative incidence of deep-vein thrombosis or pulmonary embolism (DVT/PE) was 1.2% in patients receiving Tamoxifen versus 0.50% in those not treated — corresponding to a more than twofold increase in absolute risk. ⁶⁰ Due to this elevated risk, aromatase inhibitors such as anastrozole are now preferred in postmenopausal patients. Randomized trials—including the ATAC, TARGET, and NSABP B-35 trials—demonstrated consistently lower CAT incidence with anastrozole (range 0.8-4.8%) compared with tamoxifen (2.7-8.2%).^59,61

CDK4/6 inhibitors (abemaciclib, palbociclib, ribociclib), used in combination with endocrine therapy, have been associated with increased thrombosis.^62,63 In a multicenter observational cohort of 266 patients treated with CDK4/6 inhibitors, the 1-year incidence of any thrombotic event was approximately 10.4% overall; by agent, the incidence was 10.9% with palbociclib, 8.3% with ribociclib, and 4.8% with abemaciclib.^62,63 A meta-analysis of seven Phase-III trials (n = 4415) demonstrated a significantly elevated CAT risk (OR 2.90; 95% CI 1.07-7.87). ⁶¹

In endometrial cancer, endocrine therapy remains limited to select metastatic settings using megestrol acetate or aromatase inhibitors. Megestrol acetate carries a particularly elevated thrombotic risk; one study reported an increased CAT rate (HR 2.07; 95% CI 1.36-3.17; p = .001). ³⁹

CAT-Associated Biomarkers

Beyond BRCA mutations, additional genomic and molecular biomarkers contribute to CAT predisposition. ⁶⁴ Cancer-derived inflammatory cytokines, neutrophil extracellular trap activation, platelet activation, and release of microparticles promote a hypercoagulable phenotype. ³¹ Tissue factor (TF) represents a central mediator, and its expression may be driven by oncogenic pathways including ALK, ROS1, HER2, KRAS, STK11, EGFR, and IDH1/2. ³¹ Elevated TF expression—particularly in ovarian clear cell carcinoma and endometrioid subtypes—has been strongly correlated with increased CAT incidence. ⁶⁵

Management of CAT in Female-Specific Cancers

Thromboprophylaxis

Current guidelines recommend pharmacological thromboprophylaxis with low-molecular-weight heparin (LMWH) for most hospitalized medical and surgical cancer patients, including those with gynecologic malignancies.^66-69 In the outpatient setting, prophylaxis with LMWH or direct oral anticoagulants (DOACs) such as apixaban or rivaroxaban (where approved) may be considered for patients with a Khorana score ≥2 prior to initiating a new systemic chemotherapy regimen.^66,68-70

Hospitalized patients with cancer have approximately twice the risk of CAT compared with non-cancer patients.^68,69,71 Major oncology guidelines uniformly support prophylaxis with LMWH, unfractionated heparin, or fondaparinux during hospitalization, with extension after discharge in selected high-risk individuals. ⁶⁷

Patients undergoing surgery for gynecologic cancers are at particularly high risk of postoperative CAT. Postoperative prophylaxis with LMWH for at least 10 days is strongly recommended, with evidence supporting extended prophylaxis up to 4 weeks, especially after major abdominal or pelvic procedures.^42,44,45,72 Although LMWH remains the preferred agent, emerging data on direct factor Xa inhibitors for extended prophylaxis are encouraging but not yet definitive and require further validation.

In gynecologic oncology, systematic reviews suggest that combining LMWH with mechanical methods—such as intermittent pneumatic compression—provides the most effective risk reduction while maintaining an acceptable bleeding profile.^44,73 Extended prophylaxis may not be universally required, particularly for patients undergoing minimally invasive surgery, who generally exhibit lower thrombotic risk. Further research is needed to refine risk stratification and determine the optimal duration and modality of prophylaxis in this population.^42,73,74

Clinical Presentation of CAT in Female Cancer Patients

An analysis from the RIETE registry including 11 055 patients with active cancer found no major sex differences in CAT recurrence (deep vein thrombosis or pulmonary embolism) or major bleeding after approximately 5 months of anticoagulant therapy. ⁷⁵ However, women experienced lower rates of fatal bleeding and overall mortality compared with men. ⁷⁵

Superficial vein thrombosis may occur in cancer patients, particularly in women with breast cancer, where frequent venous access and local treatments can contribute to thrombotic events in superficial veins. ⁷⁵ Women with ovarian cancer may be at increased risk of unusual-site thrombosis, including splanchnic vein thrombosis, often related to external compression by the primary tumor, lymphadenopathy, or peritoneal metastases. ²⁴

CAT, irrespective of anatomical site, is associated with increased all-cause mortality across all malignancies. ⁷⁶ In particular, women with uterine or breast cancer who develop CAT have a markedly increased risk of death. In one study, CAT was associated with a 4.8-fold increased mortality risk in uterine cancer (HR 4.79; 95% CI 4.57-5.03) and a 3.9-fold increased risk in breast cancer (HR 3.93; 95% CI 3.80-4.06), compared with patients without CAT. ¹¹ These findings underscore the importance of early recognition, appropriate prophylaxis, and prompt treatment of CAT in women with female-specific cancers.

Anticoagulation and Bleeding Burden

International clinical practice guidelines, based on large randomized controlled trials, recommend LMWH or direct factor Xa inhibitors as first-line therapy for CAT.^66-69 Among DOACs, only factor Xa inhibitors—not direct thrombin inhibitors such as dabigatran—have been systematically evaluated in cancer-associated thrombosis.^77-81

Anticoagulant selection in newly diagnosed CAT is complex and should integrate patient-related, cancer-related, and treatment-related factors. Most prospective clinical trials did not stratify patients by cancer type, thrombotic presentation (symptomatic vs incidental), or key clinical variables such as age, renal function, or baseline bleeding risk. As a result, many recommendations rely on post hoc analyses, which offer useful but not definitive guidance. ⁸²

Hematologic toxicities may further complicate treatment. Thrombocytopenia is a well-recognized adverse effect of niraparib, a PARP inhibitor used in ovarian cancer.^54,55 In breast cancer, anemia, thrombocytopenia, and pancytopenia are frequently observed in patients treated with chemotherapy, antibody–drug conjugates, and CDK4/6 inhibitors.^40,83 These conditions may increase bleeding risk and limit the choice or intensity of anticoagulant therapy. Shared decision-making is therefore crucial to align therapeutic choices with patient values, tolerance, and overall treatment goals.^54,55

A meta-analysis of individual patient data from prospective studies evaluated CAT recurrence and bleeding in patients treated with tinzaparin. ⁷⁹ Among 1413 patients (median age 65 years; 51.9% female), the 6-month cumulative incidence of recurrent CAT was 5.4% (95% CI 4.3-6.7%), while major bleeding occurred in 3.4% (95% CI 2.6-4.6%). The highest recurrence rates were observed in genitourinary, lung, and gynecologic cancers, whereas major bleeding events were most frequent in breast, gastrointestinal, and genitourinary malignancies. ⁷⁹

Women with cancer are at increased risk of abnormal uterine bleeding both before and during anticoagulant therapy. ⁸⁴ Data from the RIETE registry showed that 51% of women who experienced abnormal uterine bleeding while on anticoagulation had an underlying malignancy, with uterine cancer present in 63% of these cases. ⁸⁴ In the Hokusai VTE Cancer trial, major bleeding occurred in 4.3% of patients with gynecologic cancer treated with edoxaban, compared with 1.6% in those receiving dalteparin. ⁵¹ In the SELECT-D trial, a substantial proportion of clinically relevant non-major bleeding episodes in patients treated with rivaroxaban originated from the genitourinary tract. ⁸¹ These findings highlight the need for particular caution when prescribing DOACs to women with endometrial or other gynecologic cancers.

In the absence of robust, cancer-type–specific bleeding risk models to guide the choice between LMWH and DOACs, pharmacokinetic properties may influence decision-making. ⁸⁵ In clinical situations where, rapid reversibility of anticoagulation is desirable—such as in patients at high risk of requiring urgent surgical or gynecologic interventions—LMWH, with its shorter half-life and established reversal strategies, may be preferred over oral agents.

The development of cancer-type–specific risk models for anticoagulant-associated bleeding is a major unmet need, particularly in women with endometrial and other gynecologic cancers.^67,68 Such tools could refine therapeutic decisions and improve outcomes in this high-risk population.

What are the Remaining Issues?

Risk Assessment for Thromboprophylaxis in Gynecologic Cancer Across the Patient Care Timeline

Current recommendations emphasize the importance of pharmacological prophylaxis for CAT in hospitalized cancer patients, including those undergoing surgical or medical treatment. Prophylaxis may also be appropriate in carefully selected ambulatory patients, based on validated clinical risk scores and individualized assessment.^67-69,86 Preventive strategies should integrate patient-specific, cancer-specific, and treatment-related variables to balance the risk of thrombosis against bleeding. Extended thromboprophylaxis beyond hospital discharge may be warranted in high-risk individuals, as the treatment of established CAT in cancer patients remains challenging due to the dual risk of recurrence and anticoagulation-associated bleeding.^68,71

The Khorana score classifies all patients with gynecologic malignancies as intermediate risk for CAT, limiting its discriminatory value within this subgroup. ⁶⁹ Furthermore, most predictive models developed to assess CAT risk include fewer than 10% gynecologic cancer cases, restricting external validity and clinical applicability in gynecologic oncology.

To address this gap, Wang et al proposed a nomogram in 2020 specifically tailored to gynecologic cancers. The model incorporated age, D-dimer level, body mass index, and surgical approach and demonstrated reliable predictive performance (c-index 0.721; 95% CI 0.60-0.70). ⁸⁷ Building on this work, the same group developed a second model in 2022 targeting epithelial ovarian cancer. This updated tool added progesterone receptor status and Ki-67 positivity to the predictive variables and demonstrated strong discrimination using AUC and calibration metrics. ⁴³ However, both nomograms require multicenter external validation before widespread clinical adoption.

Another model, the Thrombogyn score, was specifically developed for women undergoing surgery and chemotherapy for gynecologic cancers. This score stratifies risk based on BMI, chemotherapy exposure, and hemoglobin level.^88,89 It demonstrated good discriminatory ability in both derivation and validation cohorts (AUC 0.714; 95% CI 0.645-0.780 and 0.699; 95% CI 0.605-0.792, respectively), and predictive accuracy improved further when biomarker data were incorporated. Nevertheless, external prospective validation remains lacking.

Risk Assessment for Therapeutic Anticoagulation in Cancer Patients

Current guidelines recommend long-term anticoagulation for CAT as long as the malignancy remains active.^68-71 However, some clinical scenarios support shorter treatment durations, such as postoperative thrombosis, catheter-associated thrombosis, or CAT occurring during breast cancer endocrine therapy initiated more than five years prior, without evidence of recurrence. ⁵²

The development of accurate prediction models for CAT recurrence remains limited, particularly among postoperative patients with gynecologic malignancies. The nomogram by Wang et al demonstrated promising performance (c-index 0.721; 95% CI 0.644-0.797) in postoperative cohorts,^88,89 but requires further validation.

Despite multiple randomized studies, no anticoagulant has demonstrated clear superiority in preventing catheter-related thrombosis.^34,67,68 Likewise, no significant differences in major bleeding or mortality have been reported between anticoagulated and non-anticoagulated patients with catheter-associated thrombosis, leading most guidelines to advise against routine anticoagulation in this setting.^67,68

Endocrine therapy contributes variably to thrombosis risk across cancer types, and guideline interpretation remains inconsistent.^{59,61,67,68,90} Better consensus is needed to determine whether endocrine therapy should influence decisions regarding anticoagulant initiation, continuation, or discontinuation.

Anticoagulation duration should align with ongoing cancer activity, defined by residual disease or active systemic therapy. ⁹¹ Breast cancer presents a unique challenge because recurrence can occur more than a decade after diagnosis and treatment cessation. ⁹¹ Currently, no consensus exists on anticoagulation resumption in patients with late recurrence.

Pregnancy and Contraception

Pregnancy increases thrombosis risk approximately four- to five-fold, with fluctuations according to gestational stage. ⁹² A prior history of pregnancy-associated thrombosis markedly increases recurrence risk in subsequent pregnancies. ⁹³

Women diagnosed with malignancy during pregnancy face substantially elevated thrombotic risk. Odds ratios range from 6.6 for cervical cancer to 10.6 for ovarian cancer, while no statistically significant elevation has been observed in breast cancer (OR 2.5; 95% CI 0.3-17.4). ⁹⁴ LMWH is the recommended first-line therapy for prophylaxis and management of thrombosis during pregnancy due to its inability to cross the placental barrier. ⁹⁵ In contrast, DOAC safety data remain insufficient, and these agents are currently contraindicated in pregnancy. ⁹⁵

Breast cancer is the most common malignancy in women of reproductive age, and improved survival has increased the importance of reproductive health in survivorship care.^96,97 While guidance exists for fertility preservation and premature ovarian insufficiency management, structured contraceptive counseling remains underdeveloped.^85,98 Nearly 6% of patients in the CANTO study continued hormonal contraception despite contraindications and increased thrombosis risk. ⁸⁵ This highlights the need for proactive contraceptive counseling and individualized reproductive planning for patients with hormone receptor–positive disease.^99,100

Conclusion

Female-specific cancers—including breast, cervical, ovarian, and endometrial malignancies—represent a substantial global health burden. Their association with cancer-associated thrombosis underscores the need for tailored prevention and management strategies. Risk stratification remains challenging due to the limited discriminatory capacity of existing risk models and the underrepresentation of gynecologic cancer patients in validation cohorts.

LMWH remains the preferred agent for thromboprophylaxis and treatment, while the use of DOACs requires cautious evaluation, particularly in patients at high risk for bleeding or reproductive tract involvement. Reproductive health—including pregnancy considerations, fertility preservation, and contraception—must be integrated into clinical decision-making.

Ongoing research is needed to refine predictive tools, validate gynecologic cancer–specific risk models, and establish individualized treatment algorithms capable of reducing both thrombosis- and bleeding-related morbidity and mortality. A multidisciplinary approach involving oncologists, thrombosis specialists, hematologists, and gynecologic surgeons will be essential to optimize care for this complex patient population.