Personalised med

Question 1

Intro

Answer 1

Personalised medicine involved tailoring therapy to patients depending on pt-specific information about their genotype, phenotype and environment
Has become a reality through advances in genomics and advances in biotech – high throughput analyses
Arguments on what true ‘personalised’ medicine is – subcat or only for 1 patient
Sometimes having to personalise every treatment is also a limitation – looking for off-the-shelf CAR T cells

Question 2

Van’t Veer, 2002

Answer 2

microarray of 78 BC tumours, identified a GEP of 70 genes that could predict the risk of future distant metastasis, and so could identify in advance which patients would and would not respond to chem
• The group with a ‘good’ as opposed to ‘poor’ prognosis did not suffer metastases in 5y
Computer software can now compare a patients mRNA profule to the GEP from this study and predict whether they will benefit from chemo
5 years after the pubilication of this study (07), FDA approved this test, named the MammaPrint test which helps clinicals identify high and low risk pts w early BC
• However, the clinical utility of this test does remain controversial, as Cardoso et al. (2016) published a phase III randomised study in the NEJM involving 6693 women with early-stage breast cancer, which found that only ~46% of patients considered to be a low genomic risk by the MammaPrint test but a high clinical risk (by looking at conventional histology) might not require chemotherapy. Thus more RCTs are required to validate its utility

Question 3

Banchereau et al. (2016)

Answer 3

For example, Banchereau et al. (2016) attempted to sub-type systemic lupus erythematosus (SLE), an autoimmune disease for which there is no curative treatment, but extreme heterogeneity at the molecular and clinical level
• They took blood samples from hundreds of paediatric SLE patients and analysed the whole blood transcriptome
• They then used weighted gene co-expression network analyses to identify patient-specific or co-expressed molecules that associated with clinical traits over time
• Based on this data, they stratified patients into seven groups, however this concept has not yet been translated into clinical practice so that clustering patients into groups in this manner can be used to identify potential therapeutics tailored to these patient groups or methods of disease-monitoring
This could be because information on the whole blood transcriptome performed only partly reflects the manifestations of SLE – could be overlooking other disease-specific changes that are important  hence there is a need to integrate genomics, metabolomics and other layers of data

Question 4

• Faurholt-Jepsen et al. (2015)

Answer 4

showed that smartphone data can be used as an electronic biomarker to gain insight into illness activity in bipolar disorder
• Their study involved 61 patients aged 18-60 diagnosed with bipolar disorder (BD)
• The authors used a software for smartphones (called the MONARCA I system) that collects automatically generated objective data, as well as self-monitored data on illness - activity
• Results:
• They found over time significant positive correlations between BD scores (for both mania and depression) on different classifications and the number of incoming/outgoing calls per day, the duration of calls, as well as the number of outgoing text messages per day
• They also reported significant negative correlations between self-monitored data (i.e. mood and activity) and depressive scores, and significant positive correlations between self-monitored data (i.e. mood and activity) and mania scores
• Moreover, using the automatically-generated objective data, they were able to discriminate between affective states
• There were some caveats to this study, including the small sample size of 33 patients (although patients were followed up for a median 310 days), there was also an larger proportion of participants with type II bipolar disorder compared to type I, and patients received different types, doses and combinations of psychopharmacological treatments throughout.
• May serve as a biomarker for illness activity for each patient
• Could be used to subcategorise disease and generate new approaches to behavioural therapy
• Ethical challenges here – too invasive/data leakage?

Question 5

• Tixier et al. 2020

Answer 5

identified a new exciting use for FDG-PET images of tumours: they showed that PET radiomics can reveal cancer transcriptomics, thereby providing us with enhanced insight into cancer pathophysiology in patients by analysing PET data
• Their study included 45 patients with locally advanced head and neck cancer who had undergone FDG-PET scans at the time of diagnosis and transcriptome analysis using RNAs from both cancer and healthy tissues on microarrays
• Using Genomica software, they identified relationships between PET radiomics and genes involved in the cell cycle, disease, DNA repair, the immune system, metabolism, extracellular matrix organisation and signal transduction pathways
• This highlights the potential of FDG-PET radiomic features to infer tissue gene expression and the activity of various cellular pathways in head and neck cancers, and potentially other cancers too!
• The results of their study also reveal the exciting promise of radiomics to personalise treatments through targeting specific molecular pathways