Safety and efficacy of apixaban versus enoxaparin in gynecologic cancer patients undergoing open surgery: an open-label randomized controlled trial

Abstract

Background

Women undergoing surgery for gynecologic malignancies are at high risk for postoperative venous thromboembolism (VTE). Although subcutaneous enoxaparin is commonly used for prophylaxis, concerns about cost, self-injections, and adherence have prompted investigation of oral alternatives such as apixaban. In this clinical trial, we compared the safety and efficacy of apixaban versus enoxaparin for postoperative VTE prophylaxis in patients undergoing open surgery for gynecologic cancers.

Methods

In this single-center, randomized, open-label trial, 450 patients with confirmed gynecologic malignancies undergoing open abdominal surgery were allocated in a 3.5:1 ratio to receive either apixaban (2.5 mg orally twice daily) or enoxaparin (40 mg subcutaneously once daily) for 14–28 days. The primary outcomes were the incidence of VTE and major bleeding within 30 days post‑surgery, corresponding to the active prophylaxis period. Data were analyzed using intention-to-treat principles.

Results

No significant difference was observed in the baseline characteristics of the two study groups. Within 30 days, VTE occurred in 2.0% of patients in the apixaban group and 3.0% in the enoxaparin group (p = 0.891), confirming non-inferiority. Major bleeding events were significantly lower in the apixaban group (1.4%) compared to the enoxaparin group (9.0%) (p = 0.001). No other adverse effects were reported in any of the study groups.

Conclusion

Apixaban demonstrated comparable efficacy and superior safety to enoxaparin for postoperative thromboprophylaxis in women undergoing open gynecologic cancer surgery. These findings support apixaban as a viable alternative to enoxaparin in this population.

Clinical trial number

IRCT20220211053991N1.

Background

Women with gynecological cancers, such as ovarian, uterine, and cervical cancer, face a significantly increased risk of venous thromboembolism (VTE) following surgery. The incidence of VTE in these patients ranges from 3% to 25%, influenced by factors like the type of cancer and surgical modality [1]. Ovarian cancer patients, in particular, exhibit the highest rates of VTE, with a 1.8 times greater likelihood of developing deep vein thrombosis compared to those with uterine cancer [2]. Current guidelines recommend extended-duration thromboprophylaxis after gynecologic cancer surgery. While Low molecular weight heparin (LMWH) has historically been the standard of care, recent updates—including the 2023 ASCO guideline—now endorse direct oral anticoagulants such as apixaban and rivaroxaban as acceptable alternatives in appropriately selected patients [3]. This shift is largely driven by the challenges like high costs, daily self-injections, and patient adherence issues have prompted interest in alternatives like direct oral anticoagulants (DOACs) [4]. Studies suggest that DOACs may offer similar efficacy with improved patient compliance [5]. Apixaban, an oral direct factor Xa inhibitor, has emerged as a promising alternative for preventing VTE in women with gynecological cancers. Several studies have compared apixaban with enoxaparin for postoperative thromboprophylaxis in gynecologic oncology; however, most available evidence is derived from retrospective cohorts, which limits causal inference and introduces potential selection bias. For example, Diamond et al. and Knisely et al. both reported comparable VTE rates between apixaban and enoxaparin, but their retrospective designs and heterogeneous patient populations restrict generalizability [6, 7]. Similarly, Boo et al. [8] and subsequent meta-analyses [9] rely largely on retrospective data. The only prospective randomized trial to date—conducted by Guntupalli et al. [10]—included a substantial proportion of minimally invasive procedures and benign gynecologic surgeries, making its findings less directly applicable to patients undergoing open oncologic surgery, who carry the highest postoperative VTE risk [11, 12]. These limitations underscore the need for a prospective, randomized study focused exclusively on women with gynecologic malignancies undergoing open abdominal surgery, which is the evidence gap addressed by the present trial.

The present study aimed to provide robust, prospective, and randomized evidence on the safety and efficacy of apixaban compared to enoxaparin in women undergoing open surgery for gynecologic malignancies.

Methods

Study design

This was a randomized, parallel-group, open-label clinical trial conducted at Firoozgar Hospital, Tehran, Iran, from August 2023 to August 2025. The trial compared two thromboprophylactic interventions in patients undergoing surgery for gynecologic malignancies. Patients were randomized in a 3.5:1 ratio to receive either apixaban or enoxaparin. The study protocol was approved by the Ethics Committee of Iran University of Medical Sciences (Ethics code: IR.IUMS.FMD.REC.1400.190), and written informed consent was obtained from all participants prior to enrollment. This clinical trial was registered in the Iranian Registry of Clinical Trials (IRCT20220211053991N1) on July 12, 2023.

Participants

Women aged 18 years or older with histologically confirmed gynecologic malignancies (ovarian, uterine, or cervical cancer) who were scheduled to undergo open abdominal surgery were eligible for inclusion. Exclusion criteria were: severe renal impairment (e.g., creatinine clearance < 30 ml/min), significant hepatic dysfunction, known hypersensitivity to apixaban or enoxaparin, current anticoagulation or antiplatelet therapy, active bleeding or high risk of bleeding, pregnancy, or breastfeeding. No patients in this cohort underwent ileostomy formation at the index operation.

Setting

This single-center clinical trial was conducted at Firoozgar Hospital, a tertiary teaching hospital affiliated with Iran University of Medical Sciences in Tehran, Iran. Eligible participants were identified in two settings: during hospitalization in the gynecologic oncology ward and during outpatient visits to the hospital’s gynecology clinic. A trained gynecologic oncology fellow assessed eligibility based on predefined criteria. All surgical procedures and follow-up assessments were performed at Firoozgar Hospital and, in some cases, through private consultations at the principal investigator’s gynecology office.

Randomization and masking

Randomization was carried out using permuted block randomization with blocks of size 9 to maintain balanced allocation throughout the enrollment process. The random sequence was generated online using the Sealedenvelope.com service. The allocation sequence was managed by a third-party team member not involved in patient recruitment or clinical care. As each participant became eligible, the corresponding group assignment was disclosed individually by this independent coordinator, ensuring allocation concealment at the point of enrollment. Due to the difference in the route of administration between the two interventions—oral tablets versus subcutaneous injections—blinding of patients and treating physicians was not feasible. As a result, the study was conducted in an open-label manner.

Sample size

The sample size was based on detecting a clinically meaningful difference in major bleeding rates between groups. Assuming an expected rate of 8% in the enoxaparin group and 1.5% in the apixaban group, and using a 3.5:1 allocation ratio, a total of 450 participants (100 in the enoxaparin group and 350 in the apixaban group) would be required to achieve 80% power at a two-sided alpha of 0.05. This estimation accounted for an anticipated dropout rate of approximately 10%. The ratio of 3.5:1 is determined to minimize the need for the intramuscular injection route and make drug prescription easier orally.

Participants

A total of 518 patients were screened for eligibility, of whom 450 were enrolled in the study. Participants were randomized into two groups: 100 patients received enoxaparin, and 350 received apixaban (Fig. 1). All enrolled patients had a confirmed diagnosis of gynecologic cancer, with the following distribution: 41.6% ovarian cancer (n = 187), 43.6% uterine cancer (n = 196), and 14.7% cervical cancer (n = 66). Three patients deviated from their assigned treatment: one patient initially allocated to apixaban received enoxaparin, and two patients in the enoxaparin group were treated with apixaban.

Fig. 1.

Flow diagram of the study

Perioperative care

All patients received standardized perioperative care at our institution, which incorporates key elements of Enhanced Recovery After Surgery (ERAS) [13], including early mobilization, early oral intake, and multimodal analgesia. Routine postoperative management includes ambulation within the first 24 h and removal of the urinary catheter on postoperative day 1 unless contraindicated. Median length of hospital stay was 2–3 days, with most patients discharged within this timeframe; the overall range was 2–5 days. Length of stay did not differ meaningfully between the apixaban and enoxaparin groups.

Interventions

Participants were randomly assigned in a 3.5:1 ratio to receive either oral apixaban or subcutaneous enoxaparin for thromboprophylaxis following gynecologic cancer surgery. All patients received standard preoperative unfractionated heparin (5,000 units subcutaneously) as part of routine institutional prophylaxis, administered uniformly across both groups. Patients in the apixaban group received 2.5 mg orally (Tehran-Shimi, Tehran, Iran) twice daily, starting 24 h after the surgical procedure. Those in the enoxaparin group were administered 40 mg subcutaneously enoxaparin (Actoverco, Karaj, Iran) once daily, initiated either 12 h before or 12 h after surgery, according to standard institutional protocols and physician discretion. The intended duration of prophylaxis in both groups ranged from 14 to 28 days, based on individual clinical factors and routine postoperative management.

Outcome measures

The primary outcomes of the study were major bleeding (safety outcome) and VTE (efficacy outcome). Major bleeding was defined according to the International Society on Thrombosis and Haemostasis (ISTH) criteria, which include any fatal bleeding, bleeding in a critical organ or space (such as the gastrointestinal tract or retroperitoneum), or bleeding resulting in a hemoglobin drop of at least 2 g/dL or requiring transfusion of two or more units of red blood cells [14].

VTE was defined as the occurrence of symptomatic or asymptomatic deep vein thrombosis (DVT) or pulmonary embolism (PE), confirmed by appropriate imaging techniques—Doppler ultrasonography for DVT and computed tomography pulmonary angiography (CTPA) for PE [15]. Events were recorded both during hospitalization and following discharge, provided they occurred within the designated prophylaxis period.

Data collection

Data on demographic variables (age, BMI, cancer type, disease stage) were recorded before the randomization. Clinical outcomes—including VTE, bleeding events, and adverse effects—were assessed at 7, 30, and 90 days postoperatively. Because prophylactic anticoagulation is completed within the first month and chemotherapy is typically initiated or resumed thereafter, primary analyses focused on events occurring within 30 days to avoid confounding.

Statistical analysis

Data were analyzed using IBM SPSS Statistics for Windows, Version 22.0 (IBM Corp., Armonk, NY, USA). Categorical variables were compared using the Chi-square test, and continuous variables were analyzed using the independent samples t-test. A p-value of less than 0.05 was considered statistically significant. All analyses followed the intention-to-treat (ITT) principle. Additionally, a post hoc non-inferiority analysis was conducted to compare apixaban with enoxaparin, using an absolute non-inferiority margin of 1.1% for venous thromboembolism (VTE) incidence. Confidence intervals for event rates were calculated using the Clopper–Pearson exact method.

Results

Baseline characteristics

The groups were comparable in baseline characteristics, including age (apixaban: 55.10 ± 15.37 years, enoxaparin: 57.29 ± 14.17 years; p = 0.201), BMI (apixaban: 27.92 ± 2.96 kg/m², enoxaparin: 28.00 ± 2.96 kg/m²; p = 0.812), cancer type (p = 0.810), and disease stage (Stage I: 27.3%, Stage II: 36.7%, Stage III: 22.2%, Stage IV: 13.8%; p = 0.819) (Table 1). All VTE and major bleeding outcomes reported below reflect events occurring within 30 days after surgery.

Table 1.

Comparative analysis of demographic and clinical characteristics between apixaban and enoxaparin groups

Variable		GROUP		P VALUE
		Apixaban ( N  = 350)	Enoxaparin ( N  = 100)
Age		55.10 ± 15.37	57.29 ± 14.17	0.201
BMI		27.92 ± 2.96	28.00 ± 2.96	0.812
Cancer Type	Ovarian • Stage I • Stage II • Stage III • Stage IV	143 (40.9%) 32 (22%) 40 (28%) 42 (29%) 29 (20%)	44 (44%) 10 (23%) 12 (27%) 12 (27%) 10 (23%)	0.810
	Uterine • Stage I • Stage II • Stage III • Stage IV	155 (44.3%) 20 (13%) 65 (42%) 55 (35%) 15 (10%)	41 (41%) 5 (12%) 17 (41%) 14 (34%) 5 (12%)
	Cervical • Stage I • Stage II • Stage III • Stage IV	52 (14.8%) 52 (100%) 0 0 0	15 (15%) 15 (100%) 0 0 0
Diabetes Mellitus		138 (39.4%)	38 (38%)	0.813
Hypertension		142 (40.6%)	41 (41%)	0.945
Cardiovascular Disease		115 (32.9%)	33 (33%)	0.982

Efficacy evaluation

Among the 450 patients included in the study, VTE occurred in 2.2% of cases (10/450). In the apixaban group, the incidence of VTE was 2.0% (7 cases), while in the enoxaparin group, it was 3.0% (3 cases) (Table 2). This difference was not statistically significant (p = 0.891). To further evaluate the non-inferiority of apixaban compared to enoxaparin, a post hoc analysis was performed using an absolute non-inferiority margin of 1.1% for VTE incidence (Fig. 2). The observed VTE rate in the apixaban group was 2.0% (7/350), with a 95% confidence interval of 0.81% to 4.08%, calculated using the Clopper–Pearson method. The point estimate for the enoxaparin group was 3.0% (3/100); adding the prespecified margin yielded a non-inferiority threshold of 4.1%. Since the upper bound of the confidence interval for apixaban remained below this threshold, apixaban was considered non-inferior to enoxaparin in this population.

Table 2.

Distribution of venous thromboembolism (VTE) by treatment group

Treatment Group	No VTE	VTE Occurrence	P value
Apixaban	343 (98%)	7 (2%)	0.891
Enoxaparin	97 (97%)	3 (3%)
Total	440 (97.8%)	10 (2.2%)

Fig. 2.

Non-inferiority assessment of apixaban versus enoxaparin

Safety evaluation

The major bleeding events were significantly lower in the apixaban group compared to the enoxaparin group (p = 0.001). In the enoxaparin group, 9 patients (9%) had a major bleeding event, including four gastrointestinal bleeding events. In the apixaban group, five patients (1.4%) had major bleeding events (Table 3). No other adverse effect was recorded in any of the study groups.

Table 3.

Major bleeding events by treatment group

Treatment Group	No Bleeding	Major Bleeding	P value
Apixaban	345 (98.6%)	5 (1.4%)	0.001
Enoxaparin	91 (91%)	9 (9%)
Total	436 (96.9%)	14 (3.1%)

Discussion

This study evaluated the efficacy and safety of apixaban compared to enoxaparin in patients undergoing surgery for gynecologic cancers. The incidence of venous thromboembolism (VTE) was comparable between the two groups, with rates of 2% in the apixaban group and 3% in the enoxaparin group. Although both agents demonstrated similar efficacy in VTE prevention, apixaban exhibited a more favorable safety profile, with a significantly lower incidence of major bleeding events (1.4% vs. 9%).

Overview of international surgical guidelines in gynecologic oncology

Surgical management in gynecologic oncology is guided by international consensus recommendations. European Society of Gynaecological Oncology (ESGO)/ European Society for Medical Oncology (ESMO) [16–19] and National Comprehensive Cancer Network (NCCN) guidelines [20] emphasize comprehensive staging and cytoreductive surgery for advanced ovarian cancer, and recommend minimally invasive approaches for appropriately selected early‑stage endometrial cancer, whereas open surgery remains standard for many advanced or high‑risk cases. American College of Obstetricians and Gynecologists (ACOG) [21] and Society of Gynecologic Oncology (SGO) [22] similarly endorse tailoring surgical approach to disease site, stage, and patient factors. In alignment with these guideline‑based principles, all patients in our cohort underwent open surgery, reflecting the predominance of advanced disease and the need for extensive staging or cytoreductive procedures during the study period.

Efficacy profile of apixaban versus enoxaparin

Diamond et al. [6] conducted a retrospective cohort study at a single-site urban academic medical center to evaluate the safety and efficacy of apixaban compared to enoxaparin for postoperative VTE prophylaxis in women undergoing laparotomy for gynecologic cancers. Among 215 patients, 65 received enoxaparin and 150 received apixaban. The study found no significant difference in VTE incidence within 90 days (3.33% vs. 4.61%; p = 0.6). Knisely et al. [7] conducted a retrospective cohort study at a comprehensive cancer center to assess the efficacy, safety, and feasibility of apixaban versus enoxaparin for postoperative VTE prophylaxis in patients undergoing open gynecologic cancer surgery. Among 452 patients, 348 received apixaban and 104 received enoxaparin. The 30-day VTE rate was significantly lower in the apixaban group (0.6% vs. 6.2%), while the 90-day VTE rate showed a non-significant trend favoring apixaban (2.7% vs. 6.2%). Guntupalli et al. [23] conducted a multicenter, randomized clinical trial to evaluate the safety and efficacy of apixaban versus enoxaparin for postoperative thromboprophylaxis in women undergoing surgery for gynecologic malignancies. In their study of 400 patients, apixaban demonstrated comparable rates of VTE (1.0% vs. 1.5%). Consistent with the study by Diamond et al. [6] and Guntupalli et al. [23], our results showed comparable VTE rates between apixaban and enoxaparin, further supporting the emerging evidence that direct oral anticoagulants offer similar efficacy to low‑molecular‑weight heparin in this setting.

Safety profile of apixaban versus enoxaparin

In the study by Diamond et al. [6] no significant difference was found in the major bleeding rate of the apixaban and enoxaparin group (1.31% vs. 3.08%; p = 0.38). Although non-significant, their safety data showed a trend toward greater safety with apixaban. In the study by Knisely et al. [7] Major and minor bleeding complications were similar between the two groups. In the study by Guntupalli et al. [23] major bleeding rate was 0.5% in both groups, confirming no statistically significant differences. The study by Guntupalli et al. was the only clinical trial available comparing apixaban with enoxaparin thromboprophylaxis in gynecological cancer. While our study was conducted at a single center, theirs was a multicenter trial, which enhances the generalizability of their findings. In contrast to earlier studies, we observed a significantly lower incidence of major bleeding events in the apixaban group. This discrepancy warrants further investigation in future trials to clarify potential factors influencing bleeding risk. Nonetheless, our findings—along with those from other studies— still support the broader adoption of apixaban for thromboprophylaxis in gynecologic malignancies, given its favorable efficacy and ease of use.

Cost-effectiveness of apixaban versus enoxaparin

While the efficacy and safety profiles of apixaban support its use as an alternative to enoxaparin in gynecologic cancers, many studies have also evaluated its cost-effectiveness [24, 25], patient adherence [10, 26], and satisfaction [10, 27]. These studies consistently show that apixaban is associated with lower out-of-pocket costs, improved treatment compliance due to oral administration, and higher patient satisfaction compared to injectable anticoagulants. These additional advantages further support the broader adoption of apixaban as a patient-centered and economically viable option for postoperative thromboprophylaxis in this high-risk population.

Generalizability of the findings

A notable feature of this cohort is that all endometrial cancer surgeries were performed via laparotomy. This reflects institutional practice and patient characteristics during the study period, including limited access to minimally invasive platforms and a high proportion of patients requiring comprehensive staging or cytoreductive procedures. Because surgical approach influences postoperative recovery and VTE risk, this pattern of care may affect generalizability to centers where minimally invasive surgery is routinely used for endometrial cancer. Also, given the pharmacokinetic considerations of both agents, we excluded patients with significant renal impairment. Therefore, our findings may not be generalizable to individuals with reduced renal function.

Limitations of the study

This study has several important limitations. It was conducted at a single tertiary care center, which may restrict the generalizability of the findings to broader or more diverse populations. The open‑label design introduces the possibility of observer or performance bias, particularly for outcomes such as bleeding assessment. Although clinical events were monitored through 90 days postoperatively, asymptomatic VTE may have gone undetected in the absence of routine imaging. Patient‑reported outcomes—including satisfaction, quality of life, and preferences regarding oral versus injectable prophylaxis—were not evaluated, despite their relevance to long‑term adherence and patient experience. The 3.5:1 randomization ratio increased exposure to apixaban but may have reduced statistical balance and limited the power of subgroup analyses. In addition, patients with significant renal impairment were excluded, which limits the applicability of these findings to individuals with reduced renal function. Finally, adherence to the assigned prophylactic regimen was not formally assessed through pill counts or injection logs, and the absence of structured compliance monitoring may influence the accuracy of both safety and efficacy estimates.

Future multicenter, double-blind randomized controlled trials are warranted to confirm our findings and enhance external validity. Incorporating patient-reported outcomes and cost-effectiveness analyses will provide a more comprehensive assessment of apixaban’s role in routine clinical practice. In addition, exploring the safety and efficacy of apixaban in high-risk subgroups—such as patients with advanced-stage disease—would help tailor prophylaxis strategies.

Conclusion

For women undergoing surgery for gynecologic malignancies, apixaban demonstrated comparable efficacy to enoxaparin in preventing VTE. Also, apixaban was associated with a significantly lower rate of major bleeding events, highlighting its favorable safety profile. These findings, combined with the known advantages of oral administration, support the use of apixaban as a safe, effective, and patient-friendly alternative to low molecular weight heparin for postoperative thromboprophylaxis in this high-risk population.

Author contributions

Soheila Aminimoghaddam: Study design and critically review the manuscript. Arash Mohazzab: Statistical analysis. Roghayeh Pourali: Patient recruitment. Hoda Zendehdel: Data collection and writing the first draft of the manuscript.

Funding

No funds were received for this study.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethical approval

This study was approved by the ethics committee of the Iran University of Medical Sciences, under the code IR.IUMS.FMD.REC.1400.190. All the research methods used in a study involving human subjects followed the ethical guidelines outlined in the Declaration of Helsinki.

Consent to participate

All patients provided a written informed consent before participation in the study.

Consent for publication

The patient provided comprehensive consent to use their medical data anonymously for publication.

Competing interests

The authors declare no competing interests.