No single quality of life parameter can predict survival across all cancers

An EORTC study has shown that health related quality of life parameters provide prognostic information regarding the survival of patients with cancer. However, no single health related quality of life scale or single quality of life parameter can predict survival across all cancers, and each cancer site requires careful examination.

The study pooled data from 30 EORTC randomized, controlled trials and investigated quality of life and symptoms as prognostic factors for survival in eleven different cancer sites: brain, breast, colorectal, esophageal, head and neck, lung, ovarian, pancreatic, prostate, and testicular cancers, and melanoma. In total, 15 health related quality of life parameters were assessed using the EORTC QLQ-C30 questionnaire, a standardized and validated patient self-assessment tool.

EORTC QLQ-C30 questionnaires were completed prior to randomization by 7417 patients. The prognostic value (p<0.05) of 15 health related quality of life parameters for overall survival was evaluated for each cancer site by means of univariate and multivariate Cox proportional hazard modeling. Models were adjusted for age, gender, distant metastasis, World Health Organization performance status and stratified by clinical study.

According to cancer site, the following health related quality of life parameters were found to have prognostic value:

• brain cancer: cognitive functioning (hazard ratio (HR) =0.95; p<.0001);
• breast cancer: nausea and vomiting (HR=1.17; p=0.0011);
• colorectal cancer: physical functioning (HR=0.93; p<.0001),nausea and vomiting (HR=1.07; p<.0001) and appetite loss (HR=1.07; p<.0001) ;
• esophageal cancer: physical functioning (HR=0.88; p=0.0072);
• head and neck cancer: nausea and vomiting (HR=1.14; p=0.0097);
• lung cancer: physical functioning (HR=0.94; p=0.0006) and pain (HR=1.08; p<0.0001);
• melanoma: dyspnea (HR=1.06; p<.0001);
• ovarian cancer: nausea and vomiting (HR=1.2; p<.0001);
• pancreatic cancer: global quality of life (HR=0.83; p=0.0073);
• prostate cancer: role functioning (HR=0.96; p=0.006) and appetite loss (HR=1.07; p<.0001);
• testicular cancer: role functioning (HR=0.81; p=0.0144).

The results of this study demonstrate that health related quality of life parameters provide prognostic information and improve predictive accuracy in cancer clinical trials. Physical functioning and nausea/vomiting were found to provide prognostic information in three and four cancer sites respectively, but no unique HRQOL cancer domain with predictive value for overall survival could be identified.

Dr. Andrew Bottomley, EORTC Assistant Director, says ““These results build on well over a decade of research from clinical trials and on systematic reviews, and they show QOL is prognostic for survival across a wide set of disease sites. What we have now is significant individual patient data, pulling apart the idea that perhaps one single QOL scale may be prognostic across all disease sites. Cleary, no magic bullet has been found here where we hoped we would hit QOL with one scale acting as a rigorous prognostic factor for all the disease sites. Clearly this is a far more complex situation than many of us had imagined. A more individualized and disease site approach is needed when looking at the value of QOL as a prognostic factor. As we continue to build an awareness and understanding of these issues, we hope to make better use of the patient’s perspective in future clinical trials and treatment.”

About the EORTC

The EORTC is a unique organization – a vibrant example of the fact that academic science and research know no national boundaries. Established in 1962, the EORTC is a non-profit European research organization operating as an international association under Belgian law.

The EORTC currently links a network of more than 2,500 pre-clinical scientists and oncologists in more than 300 hospitals in over 30 countries. It encompasses all aspects of cancer research, from translational research and new drug development to large phase III clinical trials and meta-analyses.

The 170 members of the EORTC Headquarters staff handle some 6,000 new patients enrolled each year in cancer clinical trials, approximately 30 protocols that are permanently open to patient entry, over 50,000 patients who are in follow-up, and a database of more than 180,000 patients.

The ultimate goal of the EORTC is to improve the future of cancer therapy by developing new agents and innovative approaches and to test more effective treatment strategies using commercially available drugs, or surgery and radiotherapy.

John Bean