Revolutionising Oncology: How AI Is Transforming Cancer Research and Treatment
Resource person: Raman Singh, Founder and Chief Executive Officer of Juniper Biologics
Artificial Intelligence (AI) has undeniably become deeply entrenched in our lives, with one of its most promising domains being healthcare. Its integration into healthcare spans across diagnosis, treatment, and prediction. In diagnostics, AI has demonstrated remarkable proficiency in interpreting screening modalities to detect illnesses and irregularities.
In this exclusive Q&A with HealthcareAsia, Raman Singh, Founder and Chief Executive Officer of Juniper Biologics, sheds light on how AI-led technologies are enabling personalised treatment, accelerated drug development, and optimised treatment protocols for oncological advancement.
HCA: What differentiator does AI bring to progressing cancer research and treatment?
Raman Singh: Today, cancer is the leading cause of death in the world owing to the large number of cancer cases and mortality rates. There were 20 million 1 new cancer diagnoses in 2022 and approximately 9.7 million deaths. This is especially concerning for the Asian region. Research2 estimates approximately 10.6 million cancer cases will be detected in Asia alone by 2030.
When you consider one in five people will develop cancer within their lifetime, the need for developing early detection and treatment protocols is ever more critical. In Asia3, with ageing populations and lifestyle changes, cancer incidence rates might be lower than the West but mortality rates are higher.
In circumstances like these, AI advancement in the field of healthcare and oncology can be a key advantage, enabling early cancer detection and predictive analysis. AI algorithms can detect and analyse genomic data to allow speedy interventions and more informed decisions. Additionally, AI improves the scope for personalised treatment that reflects individual characteristics and the genetic makeup of the patient, improving effectiveness and minimising side effects.
Research estimates approximately 10.6 million cancer cases will be detected in Asia alone by 2030.
HCA: What does this progress look like in real time?
Raman Singh: Singapore is a great example to look toward to see how far we’ve come in leveraging AI in oncology. The National University of Singapore (NUS) developed a novel precision technique4 that can distinguish between cancerous and healthy cells in a much faster way. Together with the National University Cancer Institute (NCIS), NUS also developed the recommendation tool Curate.AI5, which builds on a patient’s clinical data and is able to find optimal doses for chemotherapy that turn out to be around 20% lower compared to standard dosages. Approaches like these are transforming the space.
HCA: Has leveraging AI enabled new treatment landscapes in oncology? Where can we go from here?
Raman Singh: Within the realm of cancer, Carcinoma of Unknown Primary (CUP) poses a rare and formidable challenge. In this unique condition, accounting for approximately 2-5% of all global cancers, doctors face challenges in tracing the cancer’s origin within the human body, posing a significant obstacle to effective treatment. This occurs when the original tumour is exceptionally small or has spontaneously disappeared, either due to the body’s immune response or self-elimination after spreading.
Here, technologies like artificial intelligence and machine learning are evolving the existing expertise of medical professionals and researchers in comprehending and treating CUP with greater effectiveness.
What is Cancer of unknown primary (CUP)?
CUP indicates that secondary cancer has spread throughout the body, yet the origin of the cancer, known as the primary tumour, remains unidentified by doctors, according to Cancer Research UK.
HCA: How does AI enable greater personalisation in cancer treatment protocols, as you mentioned?
Raman Singh: Advanced computer programmes, supported by AI—especially deep learning mode —efficiently analyse complex patterns in gene and DNA data. These AI tools are meticulously designed to process and interpret genetic information, offering crucial insights into tissue origin (lineage) and accurately classifying tumours based on genetic characteristics to determine the type of cancer.
For example, researchers at MIT and Dana-Farber Cancer Institute have developed a computational model backed by artificial intelligence. Described in a published paper6, this model can anticipate the origin of a patient’s cancer within the body by examining the sequence of approximately 400 genes. The AI-driven tool demonstrated its efficacy by accurately classifying over 40% of tumours with unknown origins in a dataset of around 900 patients, leading to a 2.2-fold increase in the number of patients eligible for genomically guided treatments.
Precision oncology is an exciting development, utilising advanced molecular profiling, as a treatment approach that considers individual variations in patients’ genes, environments, and lifestyles. This method enhances the delivery of targeted therapies, ultimately improving patient outcomes. Clinical trials incorporating molecularly guided therapies have demonstrated notable enhancements in overall survival.
HCA: What role does AI play in accelerating large scale cancer treatment medication?
Raman Singh: The investment period7 in getting effective treatments for a disease to market is considerable – anything between US$1-4 billion and a gestation period of 10-15 years. One can well imagine the human cost extracted by any major disease during this time. This also means expensive and time-consuming drug development programmes may not be pursued for less common diseases or those affecting poorer populations.
However, AI’s ability to rapidly absorb extensive data from biomedical databases is increasingly used to enhance computer-aided drug design, especially in oncology. This significantly reduces the time and cost involved in bringing a drug to market, especially in areas with unmet needs, such as in cancer diseases.
For instance, researchers can now discern indicators of a cancer’s potential vulnerability to new immunotherapy drugs, which aim to boost a patient’s immune system to better fight the tumour. This eliminates the necessity for doctors to locate and manage control groups, saving precious time for the patient in the treatment stage.
That said, there are still limitations we need to contend with. Genetic screening is currently either too expensive or imprecise for wider use, even with consumer-focused testing kits. More biotech firms are now working to reduce costs and enhance accuracy. This calls for trusted profiling partners to offer high-quality molecular information for precise and individualised treatment decisions.
HCA: What does the future for AI adoption in oncology and large-scale pharma look like?
Raman Singh: We’re only at the beginning of this journey and the future looks limitless. Even now, the development of anti-tumour drugs leveraging advanced AI technologies is a key focus of collaborations between major global pharmaceutical companies and smaller entities at the forefront of research. The AI in oncology market is expected to be US$1.78 billion in 2024 alone, reaching US$5.63 billion8 by 2028 with a CAGR of 33%. This surge is propelled by factors including integration with electronic health records, broader applications in treatment decision support, enhanced patient engagement and education and regulatory support.
Ultimately, improving therapies and making novel drugs more accessible is a priority. The progress in AI precision medicine, based on patients’ genetic, molecular, and clinical characteristics, is bringing about more accurate diagnoses and optimised treatment outcomes, especially for a disease that threatens to grow more prevalent regionally and globally.
SOURCES:
1. https://www.who.int/news/item/01-02-2024-global-cancer-burden-growing–amidst-mounting-need-for-services
2. https://www.tandfonline.com/doi/pdf/10.1080/03007995.2023.2231761
3. https://www.tandfonline.com/doi/pdf/10.1080/03007995.2023.2231761
4. https://www.duke-nus.edu.sg/medicus/2021-issue3/ai-to-speed-up-cancer-diagnosis
5. https://news.nus.edu.sg/nus-ai-platform-enables-doctors-to-optimise-personalised-chemotherapy-dose/
6. https://news.mit.edu/2023/ai-model-can-help-determine-where-patients-cancer-arose-0807
7. https://www.weforum.org/agenda/2024/01/ai-and-quantum-revolution-transform-drug-discovery/
8. https://www.globenewswire.com/news-release/2024/02/01/2822261/0/en/Artificial-Intelligence-In-Oncology-Market-Size-Is-Expected-To-Reach-5-6-Billion-By-2028-At-A-CAGR-Of-More-Than-33-As-Per-The-Business-Research-Company-s-Artificial-Intelligence-In.html
Category: Education, Top Story
