Scientific innovations like gene editing, artificial intelligence (AI) and robotic surgery have revolutionised the ways we find, understand and treat disease. One precision medicine technology in particular — biomarker testing — is helping doctors and patients identify the treatment plans most likely to work for them. This could increase the chances of individuals achieving functional cure for hepatitis B, controlling respiratory diseases, or even prevent tumours from reoccurring in those with cancer.
Biomarkers are characteristics of the body or disease that can be measured to provide information about a patient’s biological condition or specific disease. They can include genes, proteins, or other substances.
Once the unique characteristics of the patient’s disease have been identified, doctors can gain new insights, including what treatments to suggest, as some treatments may only work for people who have certain biomarkers.
“Biomarker testing can help doctors be more targeted in their treatment plans,” said Gordana Vlahovic, MD, vice president and medicine development leader in immuno-oncology at GSK.
“For example, many patients with cancer have tumours with a high chance of recurring. The precision we can gain from biomarker testing in these types of tumours can help identify patients most likely to respond to treatment in more profound ways and for lengthier periods of time. This ultimately can lead to better long-term outcomes for the patient.”
Critical insights
There is no “one-size-fits-all” answer for patient treatments, but today’s technological advances are revealing more “this-fits-this-patient” approaches.
Chronic hepatitis B, for example, affects around 300 million people worldwide. It is a heterogeneous condition, meaning that it can present in different ways in different patients depending on their genetic make-up or their immune systems.
During clinical trials for a potential new treatment, GSK scientists learned that not all patients responded to the virus and to treatment as they expected. To investigate why this was, they collected biomarker data, patient records and measurements of the virus before, during and after treatment. They used AI and machine learning algorithms to sift through the data and share the insights back to the research team. The team then tested these insights to validate the findings and understand whether factors they thought were impacting treatment responses from different patients were connected.
This allowed scientists to identify five specific sub-groups of patients, helping them gain a deeper understanding of how the virus manifested differently in different patients, with a view to identifying which potential sequence of treatments would work best for them. This process is part of a feedback loop that will help to inform future research in hepatitis B, and could be used in other disease areas as well.
“We’ve been working hard over the last few years to gather some of the richest and largest data sets on genetics, genomics and different biological factors that allow us to map out factors that give rise to disease,” Patrick Schwab, senior director of AIML engineering at GSK, says.
“We’re using artificial intelligence and machine learning to really try to understand this large map of health information where we are able to derive key insights that otherwise we couldn’t see with the human eye."
Understanding eosinophils
For 25 years, immunology experts at GSK have studied the complex and fascinating roles eosinophils play in health and disease – including their potential as biomarkers.
As a type of white blood cell, eosinophils contribute to our immune defence against infections. As a biomarker, identified via elevated levels in a simple blood test, they can also aid the recognition of treatable, chronic inflammatory conditions such as severe asthma with an eosinophilic phenotype (SA-EP), chronic rhinosinusitis with nasal polyps (CRSwNP) or Hypereosinophilic Syndrome (HES).
Over 80% of people with severe asthma have an elevated Blood Eosinophil Count (BEC) and up to 40% of COPD patients have abnormal eosinophilic inflammation, which is associated with an increased risk of exacerbations – a sustained worsening of their symptoms.
Treatable ‘traits’ such as the identification of elevated eosinophils can help healthcare professionals to find treatments for specific patients and improve their overall health outcomes.
“The traditional stepwise approach follows a route by which each patient receives a similar treatment following the initial diagnosis,” Ian Pavord, professor of respiratory medicine at the University of Oxford, UK, says.
“The treatable traits approach looks at the individual and characteristics or ‘traits’ they exhibit.
“The traits guide the choice of medicine to address the factors involved in the poor control of their condition.”
Identifying tumours with the dMMR biomarker
Mismatch repair deficient (dMMR) tumours are most commonly found in endometrial, colorectal and other gastrointestinal cancers. They can also appear in patients with Lynch syndrome, an inherited condition that increases the risk of many types of cancer. In the US, it is estimated that 14% of solid tumours contain the dMMR biomarker.
Lab tests can identify whether a tumour has the dMMR marker by examining samples from surgery, biopsy, or through a blood test.
If the dMMR biomarker is identified in the tumour, a doctor may be able to recommend immuno-oncology therapy. Immune checkpoint inhibitors are the only targeted class of medicines approved to date to specifically treat these biomarker-selected tumour types.
They work by blocking checkpoint proteins on immune cells called T-cells. Normally these checkpoints keep the body’s immune responses in order by allowing T-cells to kill harmful cells in the body while keeping healthy cells intact.
Cancer cells often take advantage of this pathway to avoid being targeted by the immune system. In cancer treatment, when a patient is prescribed a checkpoint inhibitor, the drug prevents T cells from “turning off,” thereby allowing them to kill cancer cells.
“The past decade has seen a significant increase in immuno-oncology therapies being developed, and thanks to the identification and use of biomarkers we can better understand which patients are most likely to benefit from these treatments,” Mansoor Raza Mirza, MD, chief oncologist at Copenhagen University Hospital in Denmark, says.
“The use of biomarkers has made it possible to help extend patient survival and improve quality of life. We’re even helping to reduce healthcare costs by not prescribing treatments that are unlikely to work.”
Precision partnerships
GSK researchers are also partnering with organisations like Tempus, a US-based precision medicine company, to find new therapies.
Tempus’ artificial intelligence-based platform uses de-identified data – anonymised information from patients – from more than 50% of academic medical systems in the US.
GSK scientists are using this information to explore how biomarker testing can support clinical trial design and how the results from studies could lead to the discovery and development of new precision medicines. It could also help select patients who could benefit from future candidate medicines in GSK’s portfolio.
“GSK’s access to Tempus’ vast database of patient data enables us to work together to advance our biomarker and patient outcomes research,” says Anne-Marie Martin, GSK’s senior vice president of precision medicine and GSK’s business owner for the Tempus collaboration.
“We are able to gain further value in our precision medicine work through their other services, which include test development, a network that helps match patients to applicable clinical trials, and technologies that advance the development of artificially-grown organ models that we can use in our basic research programs,” she says.
“All of these initiatives contribute to our broader work to use precision medicine to make advancements in the treatment of cancer—with the ultimate goal of providing patients with more personalised treatment, faster.”
***
Dr. Mirza reports consulting or advisory role at AstraZeneca, Biocad, GSK, Karyopharm, Merck, Roche, Zai Lab; speakers’ bureau fees from AstraZeneca and GSK; research funding (to institution) from Apexigen, AstraZeneca, Deciphera (trial chair), GSK, and Ultimovacs; and personal financial interest in Karyopharm (stocks/shares, member of Board of Directors).
