Assessment of Karnofsky (KPS) and WHO (WHO-PS) performance scores in brain tumour patients: the role of clinician bias.

Inclusion of brain tumour patients in oncological protocols may be hampered by their neurological impairment. The goal of this study was to assess the reliability of Karnofsky Performance Scale (KPS) and WHO Performance Scale (WHO-PS) scores in this population. A cross-sectional survey was conducted through the Association des Neuro-Oncologues d'Expression Française (ANOCEF) and European Neuro-Oncology Association (EANO) networks. Clinicians were asked to write a text defining their operative definition of a patient with ≥ 70 KPS and to assess KPS and WHO-PS in six different clinical case vignettes. Two hundred seventy-six clinicians sent a response. The operative definition mentioned a normal life (89%), what patients were able (26%) or unable (29%) to do, normal cognitive processing (8%) and caregivers (6%). Older physicians mentioned more often what patients were unable to do (p = 0.005). The two scales were homogeneous in less severely handicapped patients only. More patients were excluded for hemiplegia than for expressive aphasia. Older physicians significantly excluded more patients for KPS and WHO-PS. Speciality of the physician significantly influenced scoring. On multivariable analysis, age and speciality of the physicians were correlated with KPS and WHO-PS rating even if adjusted on cases. Discordant scoring increased with severity of the deficit: in nearly all cases, the KPS would have denied, while WHO-PS would have allowed, access to a trial. Performance scores assigned to brain tumour patients are clinician and score dependant. WHO-PS would allow more access to a trial. Specific criteria should be developed for patients with neurological deficits to facilitate their access to trials.