Hello everyone!
My post today brings a reflection on some concepts I’ve been pondering about how we define PD and how the use of multimodal data from large cohorts are essential for this development. This is a topic of special interest for me since it pertains to my PhD project, which aims to derive PD subtypes based on multimodal data.
Currently, we define PD primarily based on the clinico-pathological convergence of findings. In other words, a diagnosis of PD is made when a person exhibits typical motor and non-motor clinical characteristics, coupled with the pathological finding of abnormal intracellular alpha-synuclein accumulation and subsequent formation of Lewy bodies in certain brain regions (Poewe W, Seppi K, Tanner CM, Halliday GM, Brundin P, Volkmann J, Schrag AE, Lang AE. Parkinson disease. Nat Rev Dis Primers. 2017). This practical definition has been and remains very useful for us as clinicians to address the issue pragmatically and has led to the development of various useful treatments for the disease, such as levodopa and Deep Brain Stimulation. However, this definition has a serious problem: it is overly reductionist in a disease where we know heterogeneity is the rule (Greenland JC, Williams-Gray CH, Barker RA. The clinical heterogeneity of Parkinson’s disease and its therapeutic implications. Eur J Neurosci. 2019).
But one might wonder: how can each patient with PD, while being quite distinct, all converge to some common characteristics that allow us to diagnose them with the same disease? This question led me to a recently published article advocating an interesting concept that can be applied to this issue (Sadnicka A, Edwards MJ. Between Nothing and Everything: Phenomenology in Movement Disorders. Mov Disord. 2023). In the words and images of the article itself:
We suggest that the convergence of many diseases onto a limited number of phenotypic patterns may reflect the fact that the sensorimotor system can only ‘break’ in a limited number of ways. Therefore, if we can understand the system-level processes that underpin specific phenotypes, this could unlock novel phenotype-specific therapies regardless of the specific underlying disease process.
For the full figure legend, please visit the aforementioned article (too long to insert here)
In summary, the article suggests that a common phenomenology might be a reflection of a series of pathological processes that converge to a common phenotype. Some interventions (such as levodopa and DBS) treat patients at a phenotype level, while others could treat the disease at an etiology level. Could it be that various small processes are simultaneously occurring and, depending on which pathways are diseased in each specific patient, other pathways besides a common pathway causing bradykinesia, rigidity, and tremor could also be active, thereby causing a common phenotype associated with other variable and heterogeneous phenotypes?
This leads to the importance of multimodal data and high-quality cohorts we have access to today. Empirically, we have seen that clinico-pathological convergence has been ineffective so far regarding the discovery of new disease modifying treatments, as two high-quality studies directed at alpha-synuclein pathology recently failed (Jensen PH, Schlossmacher MG, Stefanis L. Who Ever Said It Would Be Easy? Reflecting on Two Clinical Trials Targeting α-Synuclein. Mov Disord. 2023). Based on this and the possibilities I mentioned above, it is highly plausible to think that new disease-modifying treatments will have a better chance of success if they identify the peculiarities of the metabolic pathways and specific pathological processes involved in each patient, rather than the common pathway of convergence.
In this sense, various authors advocate the inclusion of diverse data types (not only clinical data, which is subject to many biases) in understanding PD. These data can include elements such as different modalities of neuroimaging, data from electrophysiological studies, transcriptomics, genomics, proteomics, metabolomics, microbiome, among many others! I will exemplify this concept and conclude this post with a representative image from a very interesting article that inspired the title of this topic (Espay AJ, Lang AE. Parkinson Diseases in the 2020s and Beyond: Replacing Clinico-Pathologic Convergence With Systems Biology Divergence. J Parkinsons Dis. 2018).
For the full figure legend, please visit the aforementioned article (too long to insert here)
