We will each produce the equivalent of more than 300 million books’ worth of personal and health-related data over the course of our lifetimes, which may provide new insights on how to live longer and healthier lives. How can this data help? It can help create personalized healthcare. The term “personalized medicine” refers to the tailoring of medical care to each patient’s unique characteristics. This process eventually results in a change in the clinical treatment paradigm to “the right drug, for the right patient, at the right time.”
Today, personalization has expanded beyond therapy selection to include drug discovery, care planning and implementation, and increasingly, how we as consumers interact with businesses aiming to improve health. This development has been facilitated by public investment, biotechnology advancement, and digitization of health profiles. Understanding which mutation causes which disease by retrospective study and using that algorithm to prospectively identify mutations in a new population were at the core of the initial concept of precision medicine. This was known as univariate analysis. The incorporation of genomes has powerful implications in oncology.
Challenges in Cancer Care
Making the best treatment decisions will become increasingly challenging due to the intricacy of Immuno-oncology (IO), the classification of malignancies, and the proliferation of biomarkers. IO is a sophisticated, large-scale, and diverse experiment. For instance, between 2011 and 2018, the number of businesses funding trials increased by 70% annually, and the diversity of major tumor indications stays high each month, with new cohorts being started for roughly 43% of the major tumor types.
The Rise of Precision Oncology
Sadly, cancer becomes apparent only after a certain stage. If it could be detected earlier it would save the patient so much discomfort associated with general treatments like chemotherapy and essentially enhance the quality of life post-treatment. This is where AI comes in. Artificial Intelligence / Machine Learning are Oncology superheroes. First, a simple blood test is taken and analyzed for gene expressions. Genetic sequencing detects changes in the human body through the identification of genetic variations and can guide diagnosis and treatment. Doctors and patients require medical solutions that are more robust and trustworthy, with easy, standard operating procedures, shorter turnaround times, fewer testing samples, and cheaper treatment costs. Personalized Oncology checks all these boxes.
Along with tissue biopsy, emerging technologies like liquid biopsy provide minimally invasive, recurrent testing during the course of treatment. In the end, these technologies might enable early diagnosis and screening for high-risk patients with recognized biomarkers. Precision medicine has advanced with the recent FDA approvals of biomarker-based, indication-agnostic therapies and liquid biopsy companion diagnostics in oncology, such as the Epidermal Growth Factor Receptor (EGFR) detection test.
Additionally, measurable residual disease (MRD) detection allows for increased sensitivity in determining treatment response, detects relapse, and can hasten decision-making. Last but not least, there is a significant continuous effort to compile data and produce insights by building larger, more thorough, and longitudinal data sets of oncology patients. Numerous oncology analytics collaborations have already shown how unrelated efforts around clinical or genetic data may be combined to produce insightful information. Additionally, with patient authorization, large provider networks and academic organizations have started building positions for aggregated data.
Cancer is a highly complicated illness of the human genome that is influenced by numerous genetic and epigenetic variables. These include somatic mutations, which are mutations we acquire throughout life, as well as hereditary germline mutations in DNA. Understanding a tumor’s mutational profile is crucial for helping with diagnosis, directing therapeutic choices, and identifying patients who might not respond to treatment. The capacity to match the appropriate therapy to the appropriate patient to assist improve patient outcomes is the primary benefit of precision medicine in oncology.
They balanced the patient’s possible remaining life expectancy against the potential advantages of genetic information for the patient and their family when deciding whether to recommend testing. Testing can potentially identify hereditary variations that increase the risk of cancer for family members while also opening opportunities for promising clinical trials and experimental medicines. On the other hand, getting hold of investigational pharmaceuticals can be a laborious and uncertain process. Even when there are available targeted medicines, their therapeutic effects may be transient and their adverse effects may be challenging to deal with.
With applications in the detection of infectious diseases, the choice of cancer treatments, and non-invasive prenatal testing, genomic testing has emerged as a ground-breaking medical tool. A trend that is anticipated to accelerate will make genetic testing a standard in the clinic by including biomarkers and companion diagnostics in the FDA labelling of medications to guide therapy selection. Biomarker-based diagnostics and electronic medical records (EMR) are commonplace in oncology now. Healthcare is simultaneously producing, storing, analyzing, and consuming an unprecedented amount of data. Patients, providers, pharmaceutical companies, and payers are just a few of the sources of this data. For cancer patients alone, there are more than 13 million EMRs in the US.
Genetic Testing in Oncology
When patients’ traditional medicines were no longer working for them, oncologists frequently suggested tumor DNA testing. The goal is to identify molecular characteristics that we can target with medications in the hopes of extending their lives or, at the very least, easing their symptoms and halting disease progression. The timing was essential and morally charged for practitioners when deciding whether to suggest genetic testing to their patients. They were especially concerned about avoiding the terrible scenario when test findings would be delivered too late to assist a patient.
By introducing the clinical use of tumor mutational load, biomarker testing is essential for immunotherapy (TMB). The outcome and effectiveness of immunotherapy are highly dependent on TMB. In general, a high percentage of TMB would increase immunotherapy’s survival rates. Carcinogenesis has been greatly influenced by lifestyle choices and air pollution. To build parameters for predictions that are more accurate, mutation, methylation, and tumor driver gene data must be combined. The gathering of clinical data and carrying out many tests are both excellent approaches to improve the precision of the treatment when it comes to the therapeutic application of cancer biomarkers. In order to provide clinicians with a better, more individualized cancer therapy, the future of cancer therapy will concentrate on integrating DNA and RNA tests for several biomarkers, including known gene mutations, fusion genes, and various forms of human leukocyte antigens.
How Machines Learn from Oncologists
Oncology patients’ outcomes can be improved by data-driven decisions, and the entire oncology community should collaborate to quickly reap these benefits. In particular, four measures might be quite effective.
- Effectively and transparently use biomarker data
The field of oncology is predicted to require the use of biomarkers, which have been at the forefront of the discipline’s study and development. Complex genetic signatures connected to patient responses could be found by fusing biomarker data with clinical information from EMRs. Larger sample numbers will ultimately result in phase IV-quality data, allow algorithms to be trained in a patient-care environment, and allow results to be reported to payers and regulatory bodies. To define and standardize the collection, analysis, and reporting of real-world biomarker data, rigorous yet doable methodologies and practices are required.
- Connect the EMR with oncology decision support
Building a more comprehensive analytics platform that integrates cancer decision support with EMRs will require a variety of features, most importantly real-time data ingestion. Health Level Seven International (HL7)-compliant interfaces and EMR-specific programs must be used to examine clinical data. Integration would give current information and expertise for decisions, reducing or eliminating unnecessary data entry.
- Useful information can be gleaned from patient-provider discussions.
When connected to physician-decision support (PDS) algorithms, portals can be effective data tools. Data from portals would nevertheless need a link between communications and the PDS, much like EMR data. To extract useful information from talks, effective natural language processing (NLP) technologies would also be required. After being effective, a variety of detailed data would be accessible, such as modifications to the regimen, drug adherence, patient involvement, negative side effects, and qualitative therapeutic impact.
- Link post-approval management and payer compensation with data-driven systems
Systems for making decisions that are linked to the EMR should track the effectiveness and safety of commercially available treatments in real-world settings in addition to supporting medical decisions. The launch of new products is becoming more complex, with everything from the more varied patient and provider usage patterns to drug-device combos to cutting-edge coating materials. Numerous studies conducted over the past two years have questioned the long-term effects of medicines on survival and quality-of-life outcomes in the real world. Once approved, it was discovered that drugs approved by the FDA and the European Medicines Agency had an insufficient follow-up. Despite having obvious limitations, these trials demonstrated the importance of continuing to monitor legally prescribed medications.
Precision Oncology Successes
Cancer was the primary or second major cause of death before the age of 70 in 112 of 183 countries between 2000 and 2019, according to the World Health Organization. Treating Cancer can get complicated. It has thousands of genetic and epigenetic variants. AI-based algorithms have a lot of potential in terms of detecting genetic abnormalities and abnormal protein interactions early on, sometimes before symptoms manifest. It can detect even the smallest of tumors which might go unnoticed by human eyes.
Treatments for leukemia and lymphoma include Yescarta and Kymriah. When a patient receives these CAR-T therapies, their own immune system is genetically engineered to combat cancer cells. In the next five to ten years, we anticipate that more and more individual medicines will be approved thanks to the active development of further gene therapy methods, most notably CRISPR.
Consider the treatment of lung cancer. The world’s first non-smoker study, the Iressa Pan-Asian Study (IPASS), which revealed that the EGFR gene mutation is one of the most prevalent biomarkers in Asian patients, led to a major advancement in target therapy. Since then, early detection of lung cancer has been crucial to treatment success, with a cure rate of over 93 per cent in the zero, first, and second stages.
Can all Cancer-types be Cured?
Not yet. Unfortunately, more research is needed to cover all cancer types. The good news is that targeted cancer therapy, and in this case, immunotherapy, has transcended the purview of fundamental science. In 2020, there were about 3,500 novel medications in the R&D pipeline, an increase of 75% from 2015. More than 150 novel cancer medications have been approved by the FDA in the past 15 years, most of which are used to treat tumors with a particular genetic mutation. However, only a small portion of patients (about 10–30%) respond to immunotherapies, and it is impossible to anticipate who will and who won’t benefit from them. Immunotherapy might not work for a sizable portion of cancer patients. Combination regimens with chemotherapies or other forms of targeted cancer medications are currently being extensively studied in clinical trials, along with the creation of new cancer medications.
Precision medicine and artificial intelligence (AI) working together could revolutionize healthcare. Precision medicine techniques isolate patient phenotypes with less frequent responses to therapy or particular medical requirements. Through complex computing and inference, AI helps develop insights, allows the system to reason and learn, and enhances clinical decision-making. Recent studies suggest that translational research examining this convergence will aid in resolving the most challenging issues facing precision medicine, particularly those where non-genomic and genomic determinants will facilitate individualized diagnosis and prognostication when combined with data from patient conditions, clinical history, and lifestyles.