The efficacy of rosuvastatin to reduce circulating tissue factor extracellular vesicles after ovarian cancer surgery Open Access

TO THE EDITOR:

Ovarian cancer is among the most prothrombotic tumor diagnoses.¹ The postsurgical period in patients with ovarian cancer represents a confluence of risk factors for venous thromboembolism (VTE) including major surgery, reduced mobility, high-risk primary malignancy, along with the administration of adjuvant chemotherapy.^2,3 Extended pharmacologic thromboprophylaxis for 30 days is considered standard of care for females undergoing laparotomy for ovarian cancer.^4,5 

Tissue factor (TF)–bearing extracellular vesicles (EVs; TF+EVs) have been explored as a biomarker to predict thrombosis, specifically in cancer.^6-8 Statins have been shown reduce EV generation by interfering with intracellular cytosol to membrane trafficking.^9,10 Moreover, phase 3 trials support the efficacy of rosuvastatin to reduce VTE.^11,12 Rosuvastatin is an attractive antithrombotic agent specifically in cancer, considering it does not carry inherent risks of bleeding. We hypothesized that rosuvastatin could reduce circulating EV after ovarian cancer surgery. We conducted a pilot randomized clinical trial to evaluate dual rosuvastatin-enoxaparin thromboprophylaxis strategy for patients undergoing surgery for presumed ovarian cancer (including primary peritoneal and fallopian tube carcinoma).

This study was approved by Dana-Farber/Harvard Cancer Center Institutional Review Board, and this trial was registered at www.ClinicalTrials.gov (identifier: NCT03532139). We consented adult females (aged >18 years) undergoing pelvic or abdominal surgery with a histologic diagnosis of ovarian, fallopian, or primary peritoneal cancers. The full inclusion and exclusion criteria are presented in the supplemental Methods. Enrolled patients randomized to arm A received enoxaparin 40 mg subcutaneous daily for 30 days after surgery starting postoperative day 1. Patient randomized to arm B received 40 mg rosuvastatin orally daily from day 15 to day 60 after surgery (supplemental Figure 1). Patients declining randomization, were enrolled in arm C and received thromboprophylaxis per clinician discretion. Patients were to begin treatment in the postoperative setting as per hemostasis, no later than postoperative day 10.

Blood draws were collected on day 1 and after surgery (postsurgical days 15, 30, and 60). A bilateral lower extremity ultrasound was performed on days 30 and 60 for all participants. The primary outcome was to evaluate whether rosuvastatin in combination with enoxaparin lowers levels of circulating TF+EV compared with enoxaparin alone or standard-of-care thromboprophylaxis after laparotomy for ovarian cancer at 60 days. Secondary outcomes included estimates for rates of VTE, along with changes in total EV, D-dimer, and C-reactive protein (CRP).

Platelet-free plasma was stored at −80°C until analysis as detailed in the supplemental Methods.⁷ Plasma samples (100 μL) were incubated with Phycoerythrin (PE)-conjugated TF antibody (Clone IIID8, Biomedical Diagnosis) at room temperature for 30 minutes. Total and TV+EV levels were assessed using semiconductor-based flow cytometry.¹³ Nanoparticle tracking analysis (NTA) were also used to capture total EV levels.¹⁴ TF EV activity was performed using a validated activity-based assay¹⁵ (see supplemental Methods for technical assay details).

A total of 24 females were included, with a median age of 60 years (interquartile range, 54-72). Nine females (37.5%) previously received systemic cancer–directed therapy and most had an Eastern Cooperative Oncology Group performance status score of 1 (87.5%) at enrollment (Table 1). Fifteen patients were randomized, including 8 to enoxaparin alone (arm A) and 7 to enoxaparin and rosuvastatin (arm B). Nine patients participated but elected to receive thromboprophylaxis as per standard of care (arm C). There were no documented thrombotic events at day 30 or day 60. Similarly, there were no recorded major hemorrhagic events.

The addition of rosuvastatin to enoxaparin did not affect the sequential profile of total EV levels as measured by flow cytometry or NTA over the 60-day postsurgical period (P = .8; Figure 1A). In parallel with the stable levels of total EV, the quantification of TF+EVs were not significantly altered at 15 days after surgery, nor did they change from days to 30 or 60 days (P = .5; Figure 1B). Treatment with rosuvastatin starting at day 15 did not reduce the number of measured TF+EVs or total EVs at days 30 or 60 (P = .6). Similar to flow cytometry results, we did not observe a significant difference in TF+EV activity levels after surgery (P = .6). The addition of rosuvastatin therapy did not significantly decrease TF+EV activity (P = .8). We observed poor correlation between circulating TF+EVs as measured by flow cytometry and activity (R² = 0, P = 1; Figure 1C).

There was a decline in D-dimer levels over the 60-day postoperative period, which was not influenced by the addition of rosuvastatin (P = .9; Figure 1D). Similarly, CRP levels were elevated at baseline and decreased by day 30 after surgery (P < .05 in both rosuvastatin and control group). There was no significant difference between CRP levels over the 60-day period with or without administration of rosuvastatin (P = .8; Figure 1E).

Although the risk of VTE can be increased up to 80-fold in the postsurgical period up to 6 weeks after cancer surgery,¹⁶ similar to other contemporary studies, we did not observe postoperative VTE events. This could be potentially because of a reduction on length of hospitalization or increased use of laparoscopic techniques.¹⁷ 

EVs, including those harboring TF, have been shown to increase after surgery.^18-20 In a systematic review of patients with cancer that included TF+EVs (both activity and quantity), there was a significant correlation between TF+EV elevation and thrombotic risk, however this signal was driven by elevated quantitative increases rather than activity assays.²¹ A challenge with the development of plasma-based EVs as biomarkers (including TF+EVs) has been the lack of standardized quantification methods.²² We thus used 3 different approaches: flow cytometry (for total and TF+EVs); TF+EV activity levels; and NTA quantification based on Brownian motion analysis (for total EVs). We found that, at all 4 time points, the estimates of total EVs were overall similar for both antigen-based assays (ie, flow cytometry and NTA). In contrast, there was no correlation between the TF+EV measured by flow cytometry and TF activity in our samples. Previous studies comparing EV detection by flow cytometry and NTA similarly did not identify correlation between the 2 methodologies.^23,24 These differences may be related to varying sensitivities of the 2 methods to enumerate EVs of different size ranges.²⁵ A recent study compared specificity and sensitivity of different EV functional and antigen techniques found large interassay variability, highlighting the challenges of measuring this biomarker.²⁶ 

Statins have been shown to reduce generation of monocyte-derived TF-bearing microparticles¹⁰ and endothelial microparticles⁹ in preclinical and clinical studies. Inhibition of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase leads to downstream modulation of key regulators of membrane trafficking and exosome release including Ras and Rho kinases.^27,28 In our study, we did not observe a significant reduction in EVs after the administration of rosuvastatin (40 mg for 45 days). Given the small number of patients who received rosuvastatin in this study, we cannot exclude smaller effects in circulating EVs although we similarly did not observe a reduction in D-dimer arguing against a significant impact on thrombin generation in vivo. Although we used standardized procedures and 3 separate techniques to isolate and estimate circulating EVs, we acknowledge that these techniques are sensitive to preanalytical factors and differences in the assay that might affect the results and interpretation. In a randomized trial of 245 patients that tested the effects of statin use after anticoagulation cessation in patients with VTE, statin use was associated with a reduction in endogenous thrombin potential and peak thrombin generation²⁹ but no reduction in EV as measured by flow cytometry.³⁰ In a smaller crossover trial of 38 patients with advanced cancer receiving chemotherapy, rosuvastatin 20 mg daily for 3 to 4 weeks compared with placebo did not affect prothrombotic biomarkers including D-dimer,³¹ however it did not include patients with gynecological malignancy or measure EVs.

In females undergoing abdominopelvic resection for ovarian cancer, the addition of rosuvastatin to standard-of-care enoxaparin did not affect the levels of circulating total EVs or TF+EVs. Although this study was not designed to assess clinical benefit of postoperative rosuvastatin, any efficacy in this setting is unlikely to be mediated by a reduction of circulating EVs.

Acknowledgments: This study was supported through R34 HL135226 from the National Heart, Lung and Blood Institute, National Institutes of Health. This research was funded, in part, through the National Institutes of Health/National Cancer Institute Cancer Center Support Grant P30 CA008748 to Memorial Sloan Kettering Cancer Center (J.I.Z.). G.S. receives grant support through R01AA020744 from the National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health.

Contribution: R.P., L.G., M.S., and J.I.Z. designed protocol; S.O., M.B., R.P.R., Y.H., C.E., P.E., M.A.P., A.B., A.R., L.G., G.S., R.F., N.M., M.S., and J.I.Z. performed research; S.O., M.B., Y.H., S.R., and D.N. analyzed data; and R.P., S.O., and J.I.Z. authored the paper that was reviewed and approved by all coauthors.

Conflict-of-interest disclosure: R.P. reports serving as a consultant for Merck Research; and reports research funding from Conquer Cancer Foundation. R.P.R. received research funding from Bristol Myers Squibb (BMS) and Janssen; consultancy/advisory board participation fees from Abbott, BMS, Dova, Janssen, Inari, Inquis, and Penumbra; is the National Lead Investigator for Penumbra, STORM-PE trial; and is the President of The Pulmonary Embolism Response Team Consortium. G.S. is a consultant for Durect, Evive, Pandion, Surrozen, Pfizer, Novo Nordisk, Boehringer Ingelheim, Cyta Therapeutics, Resolution, and Intercept; reports stock options with Glympse, Satellite Bio, and Ventyx; and receives royalties from Springer and UpToDate. M.S. receives consultancy fees from GlaxoSmithKline Eisai, and AstraZeneca. J.I.Z. received data safety monitoring board fees from Sanofi and CSL Behring; and reports consultancy fees Perceptive, Incyte, BMS, and Regeneron. The remaining authors declare no competing financial interests.

Correspondence: Jeffrey I. Zwicker, Memorial Sloan Kettering Cancer Center, 1275 York Ave, New York, NY 10021; email: zwickerj@mskcc.org.