Supported via arm's length funding from Bayer

Highlights from

American Society of Hematology

Annual meeting 2018

San Diego 1-4 December 2018

CASSINI: Rivaroxaban demonstrates thromboprophylactic efficacy in high-risk ambulatory patients with cancer

Take-home messages
  • VTE in patients with cancer leads to morbidity and mortality
  • Treatment with rivaroxaban provided a statistically significant reduction in thrombotic events versus placebo
  • Rivaroxaban was well tolerated with an overall adverse event profile similar to placebo
“We have long advocated that a risk-adapted approach using a validated risk assessment tool could be used to target prophylaxis for greater clinical benefit in cancer patients.”

Dr Alok Khorana, Professor of Medicine, Department of Hematology and Medical Oncology, Cleveland Clinic, Cleveland, US

Results from the phase III randomised CASSINI trial (NCT02555878) were presented at the late-breaker session of the American Society of Hematology (ASH) 60th Annual Meeting 2018. The data, presented by Dr Alok Khorana, Professor of Medicine, Department of Hematology and Medical Oncology, Cleveland Clinic, Cleveland, US, showed that rivaroxaban was effective in reducing venous thromboembolism (VTE) and VTE-related deaths in patients with various cancers.

VTE in cancer leads to an increase in mortality, morbidity, hospitalisations and treatment delays. Two prior randomised clinical trials have examined the use of (ultra) low-molecular-weight heparins for thromboprophylaxis in patients with cancer, PROTECHT (nadroparin vs placebo) and SAVE-ONCO (semuloparin vs placebo), the endpoints of which were arterial thromboembolism (ATE) and/or VTE (on-treatment analysis). Both trials showed benefits for prophylaxis, however event rates were low.

“We have long advocated that a risk-adapted approach using a validated risk assessment tool could be used to target prophylaxis for greater clinical benefit in cancer patients,” Dr Khorana explained. This hypothesis led to the CASSINI trial, the aim of which was to assess the efficacy and safety of rivaroxaban thromboprophylaxis versus placebo in ambulatory patients with cancer initiating a new systemic regimen and at high risk of VTE.

The trial was a multinational, multicentre, double-blind, phase IIIb superiority study. Adult patients with histologically confirmed solid tumours or locally advanced or metastatic lymphoma who were at a high risk of VTE (as defined by a Khorana score ≥2) were enrolled. All patients underwent a baseline compression ultrasonography (CUS) of the lower extremity to ensure that they did not have pre-existing VTE. A total of 841 patients with no prior VTE were then stratified according to the presence or absence of pancreatic cancer and then randomised to receive rivaroxaban 10 mg once daily (n=420) or placebo (n=421) once daily during the intended treatment period of 180 days. Patients presented for follow-up visits every 8 weeks where lower extremity CUS were recorded during treatment. The study ended after a follow-up period of 30 days.

Dr Khorana reported the results from the analysis of the primary endpoint, which was a composite of symptomatic or asymptomatic lower extremity proximal deep vein thromboembolism (DVT), symptomatic upper or lower extremity distal DVT, symptomatic or incidental pulmonary embolism or VTE-related death. There were fewer events in the rivaroxaban arm compared with the placebo arm, where the cumulative event rates were 5.95% and 8.79%, respectively (hazard ratio (HR): 0.66; 95% CI: 0.40-1.09; p=0.101). Dr Khorana explained that these results were not statistically significant, largely because 38.7% of the events occurred in patients who had already discontinued the drug. In the on-treatment period, there was a significant reduction of event rates from 6.41% with placebo to 2.62% with rivaroxaban (HR: 0.40; 95% CI: 0.20-0.80; p=0.007).

Dr Khorana went on to report that the secondary efficacy endpoints, which were confirmed ATE, visceral thrombosis and all-cause mortality were significantly lower in the rivaroxaban arm compared with the placebo arm. The secondary endpoint was a composite of the primary endpoint plus arterial and visceral events, and was observed in 6.90% of the patients treated with rivaroxaban versus 10.70% with placebo (HR: 0.62; 95% CI: 0.39-0.99; p=0.044). The other composite secondary efficacy endpoint was the primary endpoint plus asymptomatic lower-extremity distal DVT in 5.95% of patients in the rivaroxaban arm versus 9.74% in the placebo arm (HR: 0.59; 95% CI: 0.36-0.97; p=0.035).

During the primary analysis period, the composite secondary endpoint (primary endpoint plus all-cause mortality) was observed in 23.1% of the patients in the rivaroxaban arm compared with 29.5% of the patients in the placebo arm (HR: 0.75; 95% CI: 0.57-0.97; p=0.030). “In any primary prevention approach, safety is key and incidence of major and non-major bleeding was low,” added the oncologist. In the rivaroxaban and placebo arms, respectively, major bleeding occurred in 1.98% versus 0.99% of patients, and clinically relevant non-major bleeding in 2.72% versus 1.98% of patients. Adverse events were comparable between the placebo arm and the rivaroxaban arm and no additional safety signals were identified.

Overall, the CASSINI trial showed that rivaroxaban significantly reduced VTE and VTE-related death during the on-treatment period versus placebo but not during the full study period. “Consideration should be given to baseline screening in high-risk patients starting systemic therapy, an approach that we have already adopted at the Cleveland Clinic, my institution, and that we recently published,” Dr Khorana added.

Based on Khorana A A, Soff G A et al. Rivaroxaban thromboprophylaxis in high-risk ambulatory cancer patients receiving systemic therapy: results of a randomized clinical trial (CASSINI) (abstract LBA-1). Presented on Tuesday 4 December 2018.

Top image: © Racksuz

Article image: © Sturti

The content and interpretation of these conference highlights are the views and comments of the speakers/ authors.

Supported by