Body weight algorithm predicts humane endpoint in an intracranial rat glioma model

Humane endpoint determination is fundamental in animal experimentation. Despite commonly accepted endpoint criteria for intracranial tumour models (20% body weight loss and deteriorated clinical score) some animals still die before being euthanized in current research. We here systematically evaluated other measures as surrogates for a more reliable humane endpoint determination. Adult male BDIX rats (n = 119) with intracranial glioma formation after BT4Ca cell-injection were used. Clinical score and body weight were assessed daily. One subgroup (n = 14) was assessed daily for species-specific (nesting, burrowing), motor (distance, coordination) and social behaviour. Another subgroup (n = 8) was implanted with a telemetric device for monitoring heart rate (variability), temperature and activity. Body weight and clinical score of all other rats were used for training (n = 34) and validation (n = 63) of an elaborate body weight course analysis algorithm for endpoint detection. BT4Ca cell-injection reliably induced fast-growing tumours. No behavioural or physiological parameter detected deteriorations of the clinical state earlier or more reliable than clinical scoring by experienced observers. However, the body weight course analysis algorithm predicted endpoints in 97% of animals without confounding observer-dependent factors. Clinical scoring together with the novel algorithm enables highly reliable and observer-independent endpoint determination in a rodent intracranial tumour model.