Article Text
PDF
Abstract
Objective Venous thromboembolism is associated with significant patient morbidity, mortality, and can lead to delays in treatment for patients with cancer. The objectives of this study were to identify the incidence of venous thromboembolism in patients with advanced ovarian cancer receiving neoadjuvant chemotherapy, and identify risk factors for venous thromboembolism.
Methods A systematic literature search of biomedical databases, including Ovid Medline, Web of Science, Scopus, CINAHL, and Embase was performed on December 6, 2022 and updated on December 21, 2023 for peer reviewed articles. Studies were included if they were cohort studies or randomized controlled trials that evaluated the incidence of venous thromboembolism for patients with ovarian cancer receiving neoadjuvant chemotherapy. Risk of bias assessment was performed using the Newcastle Ottawa Scale for cohort studies and the Cochrane risk of bias tool for randomized controlled trials. Random effects meta-analysis was used to pool results across studies.
Results A total of 2636 studies were screened, and 11 were included in the review. Ten were retrospective cohort studies, and one was a randomized controlled trial. The incidence of venous thromboembolism in the included studies ranged from 0% to 18.9%. The pooled incidence rate of venous thromboembolism was 10% (95% confidence interval (CI) 7% to 13%). This remained significant when restricted to only studies with a low risk of bias (pooled incidence of 11%, 95% CI 9% to 14%). Body mass index of ≥30 kg/m2 was a significant risk factor for venous thromboembolism with a pooled odds ratio of 1.76 (95% CI 1.13 to 2.76)
Conclusions The results from this study demonstrated a 10% incidence of venous thromboembolism for patients with advanced ovarian cancer receiving neoadjuvant chemotherapy. This suggests that there may be a role for universal thromboprophylaxis in this population.
Trial registration PROSPERO CRD42022339602.
- Venous Thromboembolism
- Carcinoma, Ovarian Epithelial
- Gynecology
- Ovarian Cancer
- Preoperative Care
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Patients with ovarian cancer are at risk of venous thromboembolism, which is associated with significant morbidity, healthcare costs, and mortality.
WHAT THIS STUDY ADDS
This study identified a 10% pooled incidence of venous thromboembolism for patients with advanced ovarian cancer receiving neoadjuvant chemotherapy, based on available literature.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY
These findings suggest that there may be a role for universal thromboprophylaxis in this patient population.
Introduction
Patients with cancer are at an increased risk of venous thromboembolism, which results in significant morbidity, healthcare costs, and mortality.1–4 The reasons for this increased risk in patients with ovarian cancer are multifactorial and relate back to Virchow’s triad, including venous stasis from ascites or masses, procoagulant activity of cancer, and endothelial damage from cancer treatments.5 6 Studies have demonstrated that patients with ovarian cancer experiencing a venous thromboembolism have worse overall survival compared with those who do not.7–9
Thromboprophylaxis reduces the incidence of venous thromboembolism in the ambulatory cancer population.10–12 The Khorana score is endorsed by the American Society of Clinical Oncology guidelines to determine if thromboprophylaxis should be given to ambulatory cancer patients receiving chemotherapy.13 14 Using the Khorana score, patients with a gynecologic malignancy are assigned 1 point for high risk disease, giving them at least an 'intermediate risk score'. Depending on additional risk factors (body mass index, hemoglobin, and platelet and leukocyte counts) patients may not meet the criteria to be considered a high risk score. This prediction model has limitations in how accurately it can be applied to the ovarian cancer population. A study of patients receiving chemotherapy, including over 2700 gynecologic cancer patients, did not find that the score was predictive of venous thromboembolism within 6 months.15
For patients with advanced ovarian cancer, neoadjuvant chemotherapy is an alternative to primary cytoreductive surgery when disease or patient factors preclude surgery.16–18 This approach has been demonstrated to decrease perioperative risks, without compromising survival. Venous thromboembolism during neoadjuvant chemotherapy may be an independent prognostic factor, and may also contribute to treatment delays which could influence oncologic outcomes.19–21 Retrospective studies have demonstrated the incidence of venous thromboembolism in ovarian cancer patients during neoadjuvant chemotherapy to be upwards of 7%22 and potentially as high as 18%.23 Despite this, guidelines do not recommend routine thromboprophylaxis in this population, highlighting a need for higher quality evidence.21 24 The purpose of this study was to estimate the incidence of venous thromboembolism during neoadjuvant chemotherapy for patients with advanced ovarian cancer. Furthermore, to identify patient and disease related risk factors for venous thromboembolism, and the impact of venous thromboembolism on cancer treatment and survival.
Methods
Protocol and Search Strategy
This study was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines25 (online supplemental Appendix 2). Searches were conducted in collaboration with a health sciences librarian (CM), using the academic databases Ovid Medline, Web of Science, Scopus, CINAHL, and Embase. A combination of controlled vocabulary, including medical subject headings (MeSH), and keyword terms were used. Terms used in the search included neoadjuvant therapy, chemotherapy, pulmonary embolism, venous thromboembolism, venous thrombosis, venous thromboembolism, ovarian carcinoma, ovarian cancer, and ovarian neoplasms (online supplemental Appendix 1).
Supplemental material
Study Selection
Studies considered for inclusion were those evaluating the incidence of venous thromboembolism during neoadjuvant chemotherapy for patients with ovarian cancer. Studies were excluded that did not include ovarian cancer patients receiving neoadjuvant chemotherapy, or did not include venous thromboembolism as an outcome. Observational studies or randomized controlled trials were considered for inclusion. Conference abstracts, editorials, review articles, and non peer-reviewed publications were excluded. No language limitations were imposed. Titles and abstracts were screened using Covidence software (Covidence, Veritas Health Innovation, Melbourne, Australia) by two independent reviewers (KAB and SB). Duplicate studies were identified and removed using Covidence, and checked for accuracy by one reviewer (KAB). The full text of potentially relevant articles was screened by two independent reviewers (KAB and SB), and any discrepancies were resolved by consensus.
Data Collection
Data were extracted in duplicate by two independent reviewers (KAB and SB) using a data extraction form in Microsoft Excel that had been piloted by the reviewers. Any discrepancies were resolved by consensus. Extracted data included: author name, year, country, study setting, years of study, study design, participant demographics (age of study population, body mass index), disease characteristics (histology, stage, number of cycles of chemotherapy), study population size, number of patients who received neoadjuvant chemotherapy, number who developed a venous thromboembolism, venous thromboembolism incidence, treatment of venous thromboembolism, overall survival, and reported risk factors for venous thromboembolism, with odds ratio (OR) and 95% confidence interval (CI). A post hoc decision was made to include rates of interval debulking surgery and optimal cytoreduction after identifying that this information was presented in multiple studies. Risk of bias assessment was performed using the Newcastle Ottawa Scale for cohort studies26 and the Cochrane risk of bias tool for randomized controlled trials27 by two independent reviewers (KAB and SB).
Data Synthesis
A descriptive summary of the included studies was performed. When data on incidence of venous thromboembolism were not available, this information was calculated by dividing venous thromboembolism events during neoadjuvant chemotherapy from the total number of patients receiving neoadjuvant chemotherapy, removing any patients who had a venous thromboembolism at the time of diagnosis. Any studies where this information was not available were excluded. A random effects meta-analysis was performed to report the pooled incidence of venous thromboembolism and pooled OR of risk factors for venous thromboembolism. Heterogeneity between studies was assessed using the I2 statistic. Sensitivity analysis restricted to studies with a low risk of bias was planned. Analysis was performed using Stata SE V.14 (College Station, Texas, USA).
Results
Study Selection and Characteristics
Searches performed on December 6, 2022 and updated on December 21, 2023, resulted in 2636 studies for assessment (Figure 1). After excluding 484 duplicate studies, the titles and abstracts of 2152 studies were screened, and the full text of 71 studies were assessed for inclusion. Fifteen studies were initially included in the review; however, four studies were subsequently excluded as data on the number of venous thromboembolism diagnoses during neoadjuvant chemotherapy were not available separately from the remaining study cohort.28–31 This resulted in 11 studies included in the review.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram of study screening and selection. VTE, venous thromboembolism.
Study characteristics are outlined in Table 1. Ten of the included studies were retrospective cohort studies20 22 23 32–38 with one randomized controlled trial.39 Publication dates ranged from 201732 36 to 2023.38 Eight studies included only patients with ovarian cancer receiving neoadjuvant chemotherapy,20 22 23 34–36 38 and the remaining three studies included a mixed population of ovarian cancer patients, including some receiving neoadjuvant chemotherapy.32 33 37 The study populations were heterogeneous with varying definitions of neoadjuvant chemotherapy; the most common was patients receiving at least one cycle of chemotherapy before interval debulking surgery.22 35 36 38 Other definitions included receiving chemotherapy with planned debulking surgery,20 34 39 and some studies did not include a definition of neoadjuvant chemotherapy.23 32 33 37 Study sample sizes ranged from 6839 to 26 863.33 When restricted to only patients receiving neoadjuvant chemotherapy, the largest sample size was 29035 (Table 2). Five studies excluded patients with venous thromboembolism at diagnosis,20 23 35 36 38 and four studies excluded patients with preexisting anticoagulation use.22 23 32 34 Four studies reported the types of neoadjuvant chemotherapy regimens used; of those, carboplatin–paclitaxel was the most common regimen reported.33 34 39 40
Characteristics of included studies
Incidence of venous thromboembolism in included studies
Risk of Bias Assessment
The 10 retrospective studies were evaluated for risk of bias using the Newcastle Ottawa Scale (online supplemental Appendix 4) and the one randomized controlled trial was evaluated using the Cochrane risk of bias tool. Of the cohort studies, seven were rated as low risk of bias with a score of ≥7,20 21 31–34 37 while the remaining three were at higher risk of bias, with a score of <7.30 35 36 The randomized trial was assessed to have 'some concerns' for bias using the Cochrane risk of bias tool,38 mostly related to the risk of bias from the randomization process.
Venous Thromboembolism Incidence
The incidence of venous thromboembolism of the included studies ranged from 0%39 to 18.9%.23 Two studies did not include the incidence, and this was calculated from the information on venous thromboembolism events and eligible patients.32 33 The pooled incidence rate of venous thromboembolism was 10% (95% CI 7% to 13%) (Figure 2). There was moderate heterogeneity between studies, with an I2 of 67.8%. A sensitivity analysis restricted only to studies with a low risk of bias was performed. The pooled incidence rate of venous thromboembolism was 11% (95% CI 9% to 145%), with low to moderate heterogeneity between studies with an I2 of 37.1% (Figure 2). Three studies30 32 37 included treatments used for venous thromboembolism; in those studies, the rate of inferior vena cava filter use varied widely from 8.3%32 to 81.8%.30
(A) Pooled incidence of venous thromboembolism (VTE). (B) Pooled incidence of VTE restricted to studies with a low risk of bias. ES, effect size.
Risk Factors for Venous Thromboembolism
Information on risk factors for venous thromboembolism was available from six studies20 22 23 34 35 38 (online supplemental Appendix 3). There was significant variability in the risk factors included and the definitions used. Five studies evaluated body mass index as a risk factor; of these, three studies included body mass index ≥30 kg/m2,20 23 34 while two studies used 35 kg/m2.22 38 Six studies evaluated age as a risk factor, with varying definitions; two studies used an age cut-off of ≥70 years,22 35 one study used ≥60 years,34 and three studies considered age as a continuous variable.20 23 38 As a result, this was excluded from the meta-analysis. Two studies evaluated race as a risk factor using self-reported data, and these studies differed in how different races were grouped.22 23 Pooled results showed that body mass index >30 kg/m2 was the only statistically significant risk factor with a pooled OR of 1.76 (95% CI 1.13 to 2.76) (Figure 3). The remaining variables evaluated were not significant including: stage IV disease (pooled OR=1.29, 95% CI 0.78 to 2.16), CA125 level (pooled OR=1.24, 95% CI 0.68 to 2.27), and race (pooled OR=1.72, 95% CI 0.64 to 4.65). When restricted to only studies with a low risk of bias, body mass index >30 kg/m2 remained a significant risk factor (pooled OR=1.94, 95% CI 1.03 to 3.65).
Meta-analysis of risk factors for venous thromboembolism including (A) body mass index (BMI) >30 kg/m2, (B) CA125 level, and (C) stage IV disease.
Impact of Venous Thromboembolism on Cancer Treatment and Survival
Three studies reported on the relationship between venous thromboembolism and interval debulking surgery. Two studies reported that venous thromboembolism during neoadjuvant chemotherapy was associated with lower rates of debulking surgery (88% vs 60%,23 and 78.6% vs 58.3%20). The remaining study reported that venous thromboembolism during neoadjuvant chemotherapy was associated with decreased rates of optimal cytoreduction (80% vs 50%).22 This study also reported a significantly longer time from starting neoadjuvant chemotherapy to surgery for patients with venous thromboembolism (105.5 vs 143.9 days).22 One study reported an association between venous thromboembolism and survival, demonstrating a decreased overall survival in patients who had a venous thromboembolism (15 months vs 26.8 months).20
Discussion
Summary of Main Results
This systematic review identified 11 studies reporting the incidence of venous thromboembolism in patients with advanced ovarian cancer receiving neoadjuvant chemotherapy. The pooled incidence of venous thromboembolism in this population was 10% for all included studies, and 11% when restricted only to studies with a low risk of bias. A body mass index of ≥30 kg/m2 was the only significant risk factor for venous thromboembolism, with a pooled OR of 1.76 (95% CI 1.13 to 2.76). Other risk factors evaluated, including CA125, race, and stage, were not statistically significant. Two studies reported reduced rates of interval debulking surgery, and one study reported lower rates of optimal cytoreduction in patients with venous thromboembolism.
Results in the Context of Published Literature
Patients with ovarian cancer are at increased risk of venous thromboembolism compared with the general population. The addition of antiangiogenic agents, such as bevacizumab, has also been demonstrated to be a risk factor for venous thromboembolism, although this was not supported in the study by Garcia et al,39 potentially relating to the small sample size included in the study. None of the other included studies used or reported bevacizumab use; therefore, it is not possible to draw any conclusions about the influence of bevacizumab on venous thromboembolism in the neoadjuvant context based on the current study. Venous thromboembolism within a year of cancer diagnosis is a poor prognostic factor and is associated with a lower overall survival.2 9 Three of the included studies evaluated the association between venous thromboembolism, ovarian cancer treatment, and survival. Two studies demonstrated that fewer patients with a venous thromboembolism underwent interval debulking surgery,20 23 and one study demonstrated lower rates of optimal cytoreduction,22 all of which are known to be associated with poorer overall survival in patients with advanced ovarian cancer.
The American Society of Clinical Oncology13 and American Society of Hematology40 guidelines recommend thromboprophylaxis for ambulatory patients receiving chemotherapy based on whether patients are at high risk, defined as a Khorana score of ≥2. However, questions remain around whether this score is predictive of venous thromboembolism in the ovarian cancer population, as only 10% of the population used to derive and validate the Khorana score had a gynecologic malignancy.15 Furthermore, the Khorana score reports a 7.1% risk of venous thromboembolism in the high risk group; our current study has identified an overall venous thromboembolism incidence of 10% in ovarian cancer patients receiving neoadjuvant chemotherapy, which exceeds the high risk threshold modeled by the score. Two of the included studies evaluated Khorana score specific to the neoadjuvant chemotherapy context, and neither identified the score to be predictive of venous thromboembolism in this population.35 38 Some authors have advocated for the development of a predictive model for venous thromboembolism specific to patients with ovarian cancer,21 while others have proposed a universal approach to thromboprophylaxis. A recent quality improvement study by McLaughlin et al demonstrated that routine thromboprophylaxis in this population reduced the rate of venous thromboembolism from 21.3% to 8.4%, increasing the risk of bleeding.31 Patients in this study received thromboprophylaxis with either enoxaparin or apixaban.
Low molecular weight heparin and direct oral anticoagulants have been demonstrated to be effective for venous thromboembolism prevention in ambulatory patients receiving chemotherapy. The PROTECHT and SAVE-ONCO randomized clinical trials demonstrated a reduction in venous thromboembolism with prophylactic low molecular weight heparin compared with placebo.10 11 Both the AVERT and CASSINI trials have also demonstrated the safety and efficacy of direct oral anticoagulants on venous thromboembolism prevention; these trials randomized patients with a Khorana score of ≥2 to a direct oral anticoagulant versus placebo.41 42 Both trials demonstrated a significant reduction in venous thromboembolism in patients receiving direct oral anticoagulants versus placebo; however, there was also an increase in bleeding in patients with gynecologic malignancies. A systematic review comparing direct oral anticoagulants to either low molecular weight heparin or placebo for venous thromboembolism prevention or treatment demonstrated that direct oral anticoagulants are effective at prevention of venous thromboembolism with a low risk of major bleeding.43 However, questions remain as to the preferable regimen for venous thromboembolism prophylaxis for patients with gynecologic cancers.
Strengths and Weaknesses
Strengths of this systematic review included that it was conducted with rigorous methodology and reported in accordance with the PRISMA guidelines. The study methodology and data synthesis plan were established a priori to limit the risk of bias. Limitations of this study included that there were a limited number of studies evaluating venous thromboembolism during neoadjuvant chemotherapy available for inclusion. All of the included studies except one39 were retrospective in nature, increasing the potential for bias. There was significant heterogeneity in how studies defined neoadjuvant chemotherapy and only four of the included studies specified the indications for neoadjuvant chemotherapy,22 35 38 39 which may influence the generalizability of these results. There was insufficient information on the chemotherapy regimens, number of cycles of neoadjuvant chemotherapy, rates of interval debulking surgery, and survival to draw significant conclusions from the available data.
Implications for Practice and Future Research
Given the risk of venous thromboembolism for patients with advanced ovarian cancer undergoing neoadjuvant chemotherapy, routine thromboprophylaxis should be considered in this patient population. However, more research is needed to determine if a universal thromboprophylaxis strategy is safe and cost-effective, or if stratification by risk factors is preferable.21 A recent study concluded that routine direct oral anticoagulant use for venous thromboembolism prophylaxis in this population was more effective than no prophylaxis, but was not a cost effective strategy.44 With the approval of some generic direct oral anticoagulants, further research is needed to revisit the question of cost effectiveness. Another area of interest is the use of artificial intelligence and machine learning to predict venous thromboembolism, and whether this can replace traditional scoring systems with greater accuracy.45 It will also be important for future studies to consider patient reported outcomes and medication adherence when determining if routine thromboprophylaxis is acceptable from a patient perspective. Studies suggest that direct oral anticoagulants may be preferable from a patient perspective for the treatment of cancer associated venous thromboembolism,46 however patient reported outcome data on venous thromboembolism prophylaxis specific to the ovarian cancer population are limited.
Conclusions
This systematic review found that patients with advanced ovarian cancer undergoing neoadjuvant chemotherapy had a 10% incidence of venous thromboembolism during this time. Patient body mass index of ≥30 kg/m2 increased the risk of venous thromboembolism in this population. Furthermore, venous thromboembolism during neoadjuvant chemotherapy may impact rates and timing of interval debulking surgery and overall survival. Given the high incidence of venous thromboembolism, there may be a role for universal thromboprophylaxis in this population; however, further research is needed to confirm if this is a safe and cost effective strategy, and if this has an influence on overall survival in this population.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
X @kristinblackmd
Contributors KAB and SB performed the data collection. CM developed the review search strategy. KAB and SB performed the descriptive analysis and AM performed the statistical analysis. All authors (KAB, SB, AM, CM, SP, and PC) contributed to study conceptualization, writing and editing of the manuscript, and approved the final manuscript. KAB is the guarantor and responsible for the overall content and conduct of the study.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.
Competing interests SP has received consultancy fees from GSK.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.