Aims Huntington’s disease (HD) is a devastating genetic neurodegenerative condition typically manifesting clinically in the fourth or fifth decade. With the advent of genetic therapies there is increased need to identify the earliest changes associated with carrying the HD gene. In this study we sought to determine the earliest functional imaging differences between HD gene carriers and matched controls. Based on previous work, we hypothesised that as compared to controls, HD gene carriers decades from onset would show a neural ‘reward bias’ – an exaggerated striatal response to gains as compared to losses.

Methods We recruited 35 HD gene carriers, estimated to be on average 26 years from motor onset, and 35 controls. Groups were well matched for age, gender and education level.

Participants completed a reinforcement learning task in a fMRI scanner using a sequence optimised for orbitofrontal and striatal signal. In this task participants were required to learn to choose between stimuli with the aim of maximise rewards and avoiding losses. Task behaviour was modelled using a computational model and computational variables from the best fitting model was used to probe fMRI data.

Results As hypothesised, we found that, in comparison to matched controls, gene carriers over 25 years from motor onset showed exaggerated striatal responses to gain as compared to loss predicting stimuli (p=0.003) in a reinforcement learning task. Using computational analysis, we also found group differences in striatal representation of stimulus value (p=0.0007).

Conclusion These represent the earliest functional imaging differences between HD gene carriers and controls. Behaviourally gene carriers, 9 years from predicted onset, have shown enhanced learning from gains as compared to losses. Importantly, we found no group differences in behaviour, or caudate volumes. Our data suggests a therapeutic window exists whereby HD- related functional neural changes are detectable 25 years before predicted onset.