This article is more than 5 years old. Information may no longer be current.
Rivaroxaban reduces VTE among ambulatory patients with pancreatic cancer
Click Here to Manage Email Alerts
CHICAGO — Rivaroxaban thromboprophylaxis greatly reduced venous thromboembolism among ambulatory patients with pancreatic cancer, according to results of a prespecified subgroup analysis of the CASSINI study presented at ASCO Annual Meeting.
Researchers observed no increase in major bleeding during the intervention period.
“Pancreatic cancer is a known very high-risk malignancy associated with venous thromboembolism and, in fact, by itself, has been the subject of two prior randomized trials,” Alok A. Khorana, MD, professor of medicine at Cleveland Clinic Lerner College of Medicine, vice chair for clinical services at Taussig Cancer Institute and director of the gastrointestinal malignancies program at Cleveland Clinic, told HemOnc Today.
“So this was definitely a malignancy where we felt patients would benefit from thromboprophylaxis, and we wanted to focus on that benefit,” he added. “We, therefore, preplanned this subgroup analysis and even used pancreatic cancer diagnosis as a stratification factor in the trial to get a complete sense of the risk-benefit of rivaroxaban prophylaxis.”
The randomized, double-blind CASSINI study evaluated the safety and efficacy of rivaroxaban (Xarelto, Janssen) compared with placebo among ambulatory patients with cancer who were initiating new systemic cancer therapy.
Researchers enrolled 1,080 patients, including 362 with pancreatic cancer. All patients were at high risk for VTE, defined as a Khorana score of 2 or greater.
All study participants underwent screening ultrasound at baseline and every 8 weeks thereafter. Researchers reported VTE rates of 4.5% overall (n = 49) and 6.6% (n = 24) among the pancreatic cancer subgroup.
The final analysis included 841 patients (median age, 66 years; 57% male) randomly assigned 1:1 to rivaroxaban 10 mg daily or placebo for up to 180 days.
A composite of symptomatic deep vein thrombosis, asymptomatic proximal DVT, any pulmonary embolism and VTE-related death served as the primary efficacy endpoint. Major bleeding — per the definition of International Society on Thrombosis and Haemostasis — served as the primary safety endpoint.
In the entire study population, the primary efficacy endpoint occurred among 5.95% of patients assigned rivaroxaban and 8.79% of those assigned placebo (HR = 0.66; 95% CI, 0.4-1.09).
However, more than one-third (37.8%) of events observed in the rivaroxaban group occurred among patients who had discontinued the drug. On average, patients remained on the study drug for 4.5 months rather than the planned 6 months, yet researchers kept those patients in the active therapy group.
At ASCO, Khorana and colleagues presented results of a prespecified subgroup analysis that focused on 273 patients (32.6% of overall cohort) with pancreatic cancer (rivaroxaban, n = 135; placebo (n = 138).
During the intervention period, five patients (3.7%) assigned rivaroxaban and 14 (10.1%) assigned placebo experienced primary endpoint events (HR = 0.35; 95% CI, 0.13-0.97). Researchers calculated a number needed to treat of 16.
An analysis that included the primary efficacy endpoint and the secondary endpoint of arterial/visceral events showed further benefit with rivaroxaban, with six patients (4%) assigned the study drug and 17 (12%) assigned placebo experiencing events (HR = 0.34; 95% CI, 0.14-0.87). In this analysis, researchers calculated a number needed to treat of 13.
“We were aware of the high risk of VTE in pancreatic cancer, as well as prior randomized controlled trials showing benefit,” Khorana said. “So [the magnitude of benefit observed with rivaroxaban] was not surprising. [However, it was] encouraging that we have the ability to use a daily oral agent in this setting. Prior trials [evaluated] daily self-injection of heparin, which can be challenging for patients, as well as financially more burdensome.”
Two patients (1.5%) assigned rivaroxaban and three (2.3%) assigned placebo experienced major bleeding.
“[We were] pleasantly surprised to see that we did not appear to increase major bleeding rates, [which were] numerically lower in the rivaroxaban arm and overall quite low,” Khorana said.
Correlative biomarker studies of patients without VTE revealed a significant decline in D-dimer values at week 8 and week 16 among those assigned rivaroxaban compared with those assigned placebo (P < .01).
Additional subgroup analyses from CASSINI are planned. More data will be presented later this year on bleeding events, total reduction in thrombotic events — including arterial events, which were not included in the primary endpoint — and outcomes according to type of chemotherapy, Khorana said.
“Certainly rivaroxaban prophylaxis should be a consideration in high-risk malignancies based on risk-benefit,” Khorana told HemOnc Today. “This subgroup analysis shows that [patients with] pancreatic cancer ... are a clear subgroup where benefit is quite substantial and risks are minimized, so I believe this should be a standard approach.” – by Mark Leiser
Reference:
Vadhan-Raj S, et al. Abstract 4016. Presented at: ASCO Annual Meeting; May 31-June 4, 2019; Chicago.
Disclosures: Khorana reports research funding to his institution from, honoraria from, consultant/advisory roles with, or travel, accommodations or expenses from AngioDynamics, Array BioPharma, Bayer, Bristol-Myers Squibb, Halozyme, Janssen, Leap Oncology, Merck, Pfizer, Pharmacyclics, Pharmacyte Biotech and Seattle Genetics. Please see the abstract for all other authors’ relevant financial disclosures.