The use of cohorts of patients followed from disease onset has been invaluable in better understanding the natural history of treated PD, including the evolution of cognitive dysfunction in the disease. Using this approach we have described that there are two independent syndromes of cognitive impairment, one of which is a harbinger of a dementia and the other a consequence of disordered dopamine networks (Fig. 17.4). We and others have investigated genetic variations associated with these cognitive syndromes, and shown that PD patients, especially more elderly ones, who carry the H1/H1 MAPT genotype and/or a heterozygous mutation in GBA develop dementia earlier in their disease course—presumably because these genetic variants drive the disease process. Understanding how this occurs will be vital in better treating it, but these observations support the idea that PD can be better thought of not so much as a single disorder or a disorder that has many different subtypes, but as a disease process that can run at different speeds. In other words, it is likely that all patients with PD will eventually develop a dementia, with a final common pathophysiological pathway in everyone,
Fig. 17.4 Schematic representation of two distinct cognitive syndromes in early PD, and their hypothesized aetiological pathways. 'Frontal executive' impairments are caused by alterations in dopaminergic activity within frontostriatal networks, which are in turn influenced by dopaminergic medication and COMT genotype, and are not associated with global cognitive decline or dementia. In contrast, 'posterior cortical' type cognitive impairments do not have a dopaminergic basis, but reflect the early stages of an age-dependent dementing process influenced by genetic variations in MAPT and GBA, which is likely to involve cortical Lewy body deposition.
Williams-Gray CH, Evans JR, Goris A, et al, The distinct cognitive syndromes of Parkinson's disease: 5 year follow-up of the CamPaIGN cohort, Brain 2009, 132, 2958-69, by permission of Oxford University Press.
but in some it develops within 5 years of disease onset whereas in others it takes over 25 years. The key is in understanding the major determinants driving the rate of pathology and thus disease burden and expression.
However, as we have discussed, not all cognitive deficits in PD are linked to protein aggregation in cortical neurons and dementia, rather some are due to changes in neurotransmitter networks that arise as a compensatory response to the disease process. For example, as the nigrostriatal dopamine pathway fails other neighbouring dopaminergic pathways (mesolimbic, mesocortical) become upregulated as the whole dopaminergic network tries to maintain its level of activity. The consequence is that in the more intact dopaminergic projections there is a relative overdose of dopamine in cortical and ventral striatal areas which can be further exacerbated by the use of dopaminergic medication to treat the nigrostriatal loss. Understanding this, and how genetic variants (e.g. in COMT) affect the central handling of synaptic dopamine, may lead to a better use of pharmacological agents in patients, not only in terms of which patients to treat but also at what point in the disease process.
This complexity of pathophysiological processes underlying the different types of cognitive impairment in PD highlights the danger implicit in the recent adoption of the term PD-MCI. This concept has merit, but only if it is understood to be a short-hand way of saying that the patient does not have dementia but does have some cognitive problems. It is not defining a discrete entity either clinically or pathophysiologically; it does not provide any prognostic index to the patient’s problems; and it does not have a single therapeutic solution. In time it may be refined, in particular to specify subtypes and their prognosis, and may ultimately be restricted to describe those patients who are developing incipient PD-D; if so then the term will have more merit, especially as we move towards an era of disease-modifying therapies.
In summary we have tried to highlight the complex changes that arise within the central nervous system of treated patients with PD as a result of evolving pathology and compensatory mechanisms. All of this plays out in a brain that may have developed in subtly different ways as a result of common genetic variations, which fundamentally alters the expression of the pathological processes of PD. Important targets for future research will be to understand whether genes associated with disease states affect the brain developmentally, and how they drive the disease process at a cellular and network level. Only by carefully defining the different types of cognitive impairment in PD can we hope to start to do this, and by so doing we may be able to fundamentally change how we treat this common complication in patients.