1038P - The impact of cancer-associated thrombosis and treatment related bleedings on patients' quality of life

Date 09 October 2016
Event ESMO 2016 Congress
Session Poster display
Topics Bioethics, Legal, and Economic Issues
Presenter Simon Noble
Citation Annals of Oncology (2016) 27 (6): 351-358. 10.1093/annonc/mdw377
Authors S. Noble1, A.J. Lloyd2, S. Dewilde3, E. Reimer4, A. Lee5
  • 1Institute Of Cancer & Genetics, Cardiff University School of Medicine, CF14 4YS - Cardiff/GB
  • 2Outcomes Research, Bladon Associates Ltd, Oxford/GB
  • 3She, Services in Health Economics, Brussels/BE
  • 4Patient Access Thrombosis, Leo Pharma A/S, Copenhagen/DK
  • 5Department Of Medicine, Gordon and Leslie Diamond Health Care Centre, Vancouver/CA



Venous thromboembolism (VTE), is common in cancer patients and its treatment is associated with a high risk of recurrent VTE (rVTE) and bleeding. The initial VTE is known to reduce health-related quality of life (HRQL) but little is known about the impact of later events and complications. The CATCH trial investigated the benefits of extended anticoagulation for the prevention of rVTE events in 900 patients with active cancer and acute VTE from 32 countries. We present analyses of the trial data here to describe the impact of rVTE and bleeding events on HRQL.


EQ-5D-3L data were collected at baseline and every month for seven months. EQ-5D scores range from 1.0 (full health) through 0 (dead) and down to -0.594. Analyses were designed to control for effects of covariates such as age, gender, metastatic status, primary site of cancer, ECOG status, and history of VTE, while estimating the specific impact of a rVTE or bleeding event. Mixed models for repeated measures were designed to accommodate correlations in the dataset through different specifications of the variance-covariance matrices. The impact of an event was reflected when it occurred within ±2 weeks from a planned data collection point.


HRQL data were available from 883 patients. A total of 76 rVTE and 141 bleeding events occurred during follow-up, which was reflected in 183 HRQL assessments.

Estimated HRQL scores

Number of events EQ-5D mean 95% CI
No event 4525 0.64 0.62-0.67
Non-fatal DVT 36 0.61 0.52-0.68
Non-fatal PE 3 0.62 0.48-0.73
Fatal PE 16 0.46 0.27-0.60
Recurrent VTE 55 0.57 0.49-0.64
Major bleed 15 0.59 0.46-0.69
Non-major bleed 113 0.62 0.57-0.67


The trial data shows that experiencing rVTE or bleeding significantly worsens HRQL. The data allow us to quantify the burden for patients and the value of secondary VTE prevention. Detecting VTE related HRQL signals against the background of many other influences on HRQL measurement, proved to be challenging when analysing the data. Not all events were captured in the planned HRQL assessments, which was a study limitation. Further work (possibly using qualitative methods) could help us to understand how and why patients' HRQL is affected.

Clinical trial identification


Legal entity responsible for the study

Leo Pharma A/S


Leo Pharma A/S


A.J. Lloyd: was paid a fixed fee for his work on the statistical analysis and report writing from this project S. Dewilde: was paid a fixed fee for her work analysing the data and supporting dissemination. E. Reimer: Employee of Leo Pharma A/S. A. Lee: Consultancy to Leo Pharma. Honoraria from Pfizer, Bayer and research funding from Bristol-Myers Squibb. All other authors have declared no conflicts of interest.