Advances in human genome research and Artificial Intelligence (AI) are shaping a new era in medicine, with the potential to revolutionize healthcare. Personalized medicine (PM) is at the forefront of this transformation, by moving away from the traditional ‘one-size-fits-all’ treatment model where the same medication is prescribed for all patients with a particular disease and instead embracing a tailored approach for each patient, based on their unique individual characteristics. This shift enables more precise and effective care, optimizing treatment outcomes, reducing healthcare costs, saving time and improving both the quality and longevity of life, ultimately enhancing overall patient well-being.
This article provides comprehensive information on personalized medicine and how it transforms current treatment practices.
What is Personalized Medicine?
Personalized medicine, sometimes referred to as “precision medicine” or “stratified medicine”, is a medical paradigm that aims to improve individual health by carefully tailoring treatments based on each person’s unique prognostic or genomic information.1 It is not limited to the treatment of diseases but also plays a key role in the early detection and prevention of diseases.2 Modern advancements in personalized medicine are heavily reliant on technologies like Genome Sequencing methods that analyze a patient’s fundamental biology such as DNA, RNA and proteins, to diagnose diseases more accurately.3 These techniques can identify genetic variants that were previously unknown, offering opportunities to explore alternative treatment options. By examining changes in DNA, RNA or protein levels, sequencing can reveal mutations in cancer signaling pathways or DNA repair mechanisms.4
For example, RNA-sequencing identifies specific RNA molecules involved in diseases. Unlike DNA, RNA levels can fluctuate in response to environmental factors, making RNA-seq an important tool for gaining a deeper understanding of a person’s health status.3 Additionally, comparing transcriptome data (the complete set of RNA molecules [transcripts] produced in a cell or organism at a specific time) from patients and healthy individuals allows for a more precise identification of genetic profiles, helping to narrow down potential genetic causes of disease. As a result, early diagnosis becomes possible, facilitating the development of personalized treatment strategies tailored to individual patients.4
In the current treatment strategy, when prescribed medication proves ineffective, patients are typically switched to an alternative drug, resulting in a trial-and-error approach. This often leads to suboptimal outcomes, such as adverse drug reactions, drug interactions, disease progression, delayed effective treatment and overall patient dissatisfaction. 5 Personalized medicine offers a solution to this challenge. By increasing the utilization of molecular stratification (such as assessing mutations that cause resistance to certain treatments, gene expression profiles etc), medical professionals can obtain clear, evidence-based insights to tailor treatment strategies for individual patients, improving outcomes and patient satisfaction. 5
Key Components of Personalized Medicine
- Genomic analysis:
However, in an increasing number of clinical scenarios, broader genetic investigations such as gene panels, exome sequencing and whole genome sequencing (WGS) are proving to be more advantageous. These approaches are particularly beneficial in cases where the phenotype is variable or non-specific, and the potential number of causative genes is extensive.7 For example, GLUT1 deficiency syndrome diagnosed through whole genome sequencing (WGS), helped a child to undergo specialized treatment with a low-carbohydrate diet, leading to significant symptomatic improvement.7 Using a patient’s genomic data, it is possible to determine individual drug metabolism rates, identifying fast and poor metabolizers. These variations can greatly affect treatment efficacy and influence the risk of drug toxicity.8
- Pharmacogenomics (PGx):
Pharmacogenomics is the study of interindividual variations in DNA sequences that influence drug responses, enabling its integration into routine clinical practice for more personalized and effective treatments.9 Genetic differences have been found to account for 20% to 95% of the variability in how patients respond to medications.6 This is because some genes are key regulators of absorption, distribution, metabolism and excretion (ADME) of drugs. These processes determine the level of drug exposure in a patient, influencing both its therapeutic effects and potential side effects. 10
There is another group of genes, distinct from those involved in ADME, that directly affects how patients respond to medications. Some of these genes can predict severe and potentially life-threatening immune-mediated toxicities and identifying a patient’s genetic susceptibility to such reactions allows healthcare providers to avoid certain medications, enhancing treatment safety and effectiveness.10
For example, the variant alleles HLA-B15:02 and HLA-A31:01 are genetic markers linked to an increased risk of severe drug reactions in patients taking specific medications.
a. HLA-B*15:02: This genetic variation is strongly associated with a higher risk of Stevens-Johnson Syndrome (SJS) and Toxic Epidermal Necrolysis (TEN), both of which are severe, life-threatening skin reactions. These reactions can occur in patients treated with carbamazepine or oxcarbazepine, which are anticonvulsant medications commonly prescribed for conditions like epilepsy and bipolar disorder.
b. HLA-A*31:01: This allele is linked to a higher risk of other adverse drug reactions, including maculopapular exanthema (a type of skin rash), drug reaction with eosinophilia and systemic symptoms (DRESS), and also SJS/TEN in patients taking carbamazepine.11
Considering this, clinical guidelines suggest that patients who carry these genetic variants may need to avoid certain medications like carbamazepine or oxcarbazepine or be closely monitored if they are prescribed these drugs. Genotyping for these alleles can help in tailoring medication choices, thereby reducing the risk of severe adverse reactions Pharmacogenomics thereby enhances patient care by tailoring selection of medication and dosage based on an individual’s genetic profile. This approach minimizes the risk of adverse reactions while improving treatment effectiveness, ultimately leading to better patient outcomes and increased satisfaction for both patients and healthcare providers.12
- Biomarker testing:
a. Prognostic biomarkers can predict how a disease is likely to develop or run its course which thereby helps physicians to estimate how the disease will progress over time, the risk of recurring and overall survival rates. This all helps to manage patients and aid in the prescription of treatments.13
b. Predictive biomarkers help to determine the potential effectiveness of a specific treatment for an individual patient aiding in clinical decision-making.
c. Pharmacodynamic biomarkers evaluate a drug’s impact on the disease.
d. Diagnostic biomarkers are used to identify the specific disease present in a patient sample which assist in the early recognition and diagnosis of disease.14 For example, C Reactive protein (CRP) is used as a diagnostic marker in cancer that is, high levels of C Reactive protein in blood could signify cancer is present in the body.15
- Artificial Intelligence:
The advancement of personalized medicine heavily depends on Artificial Intelligence (AI), which enables the extraction of valuable insights from vast datasets. Specifically, an AI tool known as Machine Learning (ML) is utilized to identify correlations and patterns in data through statistical inference. Machine Learning algorithms are broadly categorized into supervised and unsupervised learning. In supervised learning, data is divided into a training set, where correct outputs are labeled by a human “supervisor” to develop a predictive model, and a validation set, which is used to assess the model’s accuracy. In contrast, unsupervised learning identifies patterns and clusters within data without human labeling. A specialized subset of unsupervised learning, known as Deep Neural Networks (DNNs), can extract hidden patterns from vast datasets, making them particularly useful in biomedical research, disease classification, and drug discovery.16
AI-based Clinical Decision Support Systems (CDSSs) have been shown to enhance clinical decision-making by delivering patient-specific insights and evidence-based recommendations. Integrating AI into CDSSs can significantly improve patient outcomes by increasing diagnostic accuracy, optimizing treatment selection and minimizing medical errors.17
Based on effectiveness reports from numerous selected studies, Support Vector Machine (SVM) and Random Forest (RF) methods have proven to be highly effective in predicting and diagnosing cancer using genomic data. These machine learning techniques can analyze complex genomic datasets to identify patterns that are indicative of cancer, enabling early detection and personalized treatment strategies.18
Another powerful algorithm employed in radiomics-focused research is the Convolutional Neural Network (CNN), a deep learning technique designed to process and classify visual data. CNNs excel at handling image inputs, making them especially valuable in medical imaging. These networks improve the accuracy of automatic labeling and classification, which is critical for diagnosing diseases like cancer from radiological images.18 In some studies reviewed, researchers utilized CNNs to propose cancer diagnoses and predictions based on radiogenomics (the combination of radiological and genomic data) and histology images (microscopic images of tissue samples).18 This approach merges imaging data with genomic information to provide a more comprehensive understanding of cancer, potentially improving diagnostic precision and treatment outcomes.
Conclusion:
Personalized medicine offers a transformative approach to healthcare by tailoring treatments to a patient’s unique genetic and clinical profile, which can enhance treatment effectiveness and reduce side effects. While it holds great promise, several challenges hinder its widespread adoption and approval. Key obstacles include obtaining approval for routine use from regulatory agencies, high costs associated with genetic testing and targeted therapies, concerns about the privacy of genetic data, ethical issues, and the need for broader acceptance among healthcare stakeholders such as physicians, healthcare executives, insurance companies, and patients. However, with advancements in technology and increased collaboration, personalized medicine has the potential to become more accessible and improve patient outcomes in the future.
References:
- Kapelner A, Bleich J, Levine A, Cohen ZD, DeRubeis RJ and Berk R.Evaluating the Effectiveness of Personalized Medicine With Software. Big Data. 2021;4:572532. Available from: https://www.frontiersin.org/journals/big-data/articles/10.3389/fdata.2021.572532/full
- Goetz LH, Schork NJ. Personalized medicine: motivation, challenges, and progress. Fertil Steril. 2018 Jun;109(6):952-963.
- Khan, Abas & Mir, Mohammad. PERSONALISED MEDICINE. The Lancet Oncology. 2022;1.456.
- Eren K, Taktakoğlu N, Pirim I. DNA sequencing methods: From past to present. Eurasian J Med., 2022;54(Suppl. 1):S47-S56. Available from: https://www.eajm.org/Content/files/sayilar/224/8.pdf
- Mathur, S., Sutton, J.”Personalized medicine could transform healthcare (Review)”. Biomedical Reports 7.1 (2017): 3-5.
- Odekunle, Florence. Genomics in Personalized Medicine. International Journal of Health Sciences and Research.2016; 6. 311-315.
- Brittain HK, Scott R, Thomas E. The rise of the genome and personalised medicine. Clin Med (Lond). 2017 Dec;17(6):545-551.
- Odekunle, Florence. Genomics in Personalized Medicine. International Journal of Health Sciences and Research. 2016;6. 311-315.
- Kabbani D, Akika R, Wahid A, Daly AK, Cascorbi I and Zgheib NK (2023) Pharmacogenomics in practice: a review and implementation guide. Pharmacol.14:1189976.
- Thomas M Polasek, Kym Mina, Graeme Suthers. Pharmacogenomics in general practice: The time has come. March 2019;48(3)
- Phillips EJ, Sukasem C, Whirl-Carrillo M, Müller DJ, Dunnenberger HM, Chantratita W, Goldspiel B, Chen YT, Carleton BC, George AL Jr, Mushiroda T, Klein T, Gammal RS, Pirmohamed M. Clinical Pharmacogenetics Implementation Consortium Guideline for HLA Genotype and Use of Carbamazepine and Oxcarbazepine: 2017 Update. Clin Pharmacol Ther. 2018 Apr;103(4):574-581.
- Jennifer K Hockings, Amy L Pasternak, Angelika L Erwin, Neil T Mason, Charis Eng, J Kevin Hicks. Pharmacogenomics: An evolving clinical tool for precision medicine. Cleveland Clinic Journal of Medicine. Feb 2020;87(2):91-99.
- Drugan T, Leucuța D. Evaluating Novel Biomarkers for Personalized Medicine. Diagnostics. 2024; 14(6):587
- Ong FS, Das K, Wang J, Vakil H, Kuo JZ, Blackwell WL, Lim SW, Goodarzi MO, Bernstein KE, Rotter JI, Grody WW. Personalized medicine and pharmacogenetic biomarkers: progress in molecular oncology testing. Expert Rev Mol Diagn. 2012 Jul;12(6):593-602.
- Hart PC, Rajab IM, Alebraheem M, Potempa LA. C-Reactive Protein and Cancer-Diagnostic and Therapeutic Insights. Front Immunol. 2020 Nov 19;11:595835.
- Muhammad Wildan Gifari, Pugud Samodro, Dhadhang Wahyu Kurniawan. Artificial Intelligence toward Personalized Medicine. Pharmaceutical Sciences and Research (PSR).2021;8(2): 65 -72.
- Khaled Ouanes& Nesren Farhah. Effectiveness of Artificial Intelligence (AI) in Clinical Decision Support Systems and Care Delivery. Journal of Medical Systems. 2024;48:74.
- Rezayi S, R Niakan Kalhori S, Saeedi S. Effectiveness of Artificial Intelligence for Personalized Medicine in Neoplasms: A Systematic Review. Biomed Res Int. 2022 Apr 7;2022:7842566.
