ENHANCING INTERPRETATION OF PATIENT-REPORTED OUTCOME SCORES THROUGH INTUITIVE METRICS- AN EXAMPLE FROM PROSTATE CANCER
Author(s)
Naidoo S1, Paty J2, Fickley C2, Gwaltney C3, Hawryluk E2, Phung D4, Holmstrom S4
1Astellas Medical Affairs Global, Chertsey, UK, 2Quintiles Advisory Services, New York, NY, USA, 3Quintiles Advisory Services, Cambridge, MA, USA, 4Astellas Pharma Inc., Leiden, The Netherlands
OBJECTIVES: Regulatory agencies have highlighted the value of patient-reported outcome (PRO) endpoints in understanding treatment benefit. However, interpreting statistically significant health-related quality of life (HRQoL) benefits of new treatments can be challenging. Although the original scoring approaches must be used in trials, it may also be useful to generate complementary, intuitive metrics for PRO instruments that are easier to interpret by non-experts. In metastatic castrate-resistant prostate cancer, enzalutamide delays deterioration of various aspects of HRQoL versus placebo. Our objective was to develop a methodology for ‘translating’ quantitative PRO findings into simple metrics and language that describes the patient experience of disease and treatment impact thereon. METHODS: We utilized data from four enzalutamide clinical trials in different prostate cancer populations. A 0–100% metric for total scores and domains was computed through a linear transformation of results, in order to facilitate interpretation across trials and instruments (FACT-P, BPI, EORTC-QLQ C30/PR25). The 0–100% metric indicates the responses’ position relative to maximum scale length (e.g. a score of 119 on a 0–156 scale would equal 76%). This linear transformation to a common metric facilitated comparisons across trials and instruments. We then worked to align the metric with simple language that maps onto the patient’s experience. We summarized our findings across trials, instruments, and patient populations to identify consistent effects. RESULTS: Transformation of PRO data to another metric helps in understanding and interpreting previous findings. An example is the overall FACT-P, where patients entering one of the enzalutamide trials reported at baseline that they are functioning at 77%, suggesting a relatively high level of functioning with respect to their HRQoL. CONCLUSIONS: The changes over time captured in these trials, to be presented in detail, are more readily interpreted by non-PRO experts using this approach to scoring and the related, simple terminology.
Conference/Value in Health Info
2016-05, ISPOR 2016, Washington DC, USA
Value in Health, Vol. 19, No. 3 (May 2016)
Code
PCN145
Topic
Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes
Disease
Oncology