AI in Oncology?- The Swiss Army Knife for Patients, Providers and Researchers

by?Ioanna Deni

[email protected]

For more insights on the intersection of AI and healthcare,?visit our website ?and?subscribe ?for regular updates.

Cancer: A burden since the dawn of man

A group of medically inclined Egyptians reported more than 5000 years ago in Edwin Smith Papyrus the first case of a cancer diagnosis. It was written: swelling in the breast is a grave disease, there is no cure [1]. Along with these manuscripts, there have been identified fossilized bone tumors in human mummies suggestive of the bone cancer called osteosarcoma. It is highly probable that cancers have been as old as humanity itself.?

Throughout history, significant progress has been achieved in detecting and treating cancers [2]. While the list of innovations is vast and ever-growing, a few exciting breakthroughs below to give you an idea of this progress:

• The widespread adoption of the Pap test has resulted in a 70% reduction in cervical cancer-related deaths.
• Following the association between smoking and lung cancer, there was a 50% reduction in smoking rates.??
• A screening mammography becomes the best tool for breast cancer detection saving thousands of women across the world [3].?

What are cancers and oncology?

In simple terms, cancer is characterized by cells that have lost their ability to regulate their growth, leading to uncontrolled cell division. This abnormal cell behavior happens at a genetic level and all cell types are susceptible to it;, such as cells in the blood, bones, and skin. Thus, cancers can be of various types; carcinomas, leukemia, myelomas. Cancers are so common that it is estimated that an unsettling high percentage of people (~40%) will receive a cancer diagnosis at some point in their lives [4]. As we appreciate the profound dangers posed by cancers, we allocate more time, funding, research and resources towards combating this formidable threat.?

In the medical world, oncology is the branch dedicated to the study, diagnosis, and treatment of cancer. This field is crucial because cancer is not a singular condition, but rather a complex array of diseases that manifest in different ways. In fact, there are more than 200 types of cancer, each requiring a unique approach for diagnosis and treatment.

Cancers can be classified based on where they begin in the body, such as breast cancer originating in the breast tissue or lung cancer originating in the lung. However, another essential way to categorize cancer is by identifying the type of cell where it starts. In this regard, there are five main groups [5]:

1. Carcinoma: This cancer begins in the skin or in tissues that line or cover internal organs. It has different subtypes like adenocarcinoma, basal cell carcinoma, squamous cell carcinoma, and transitional cell carcinoma.
2. Sarcoma: Originating in the connective or supportive tissues like bone, cartilage, fat, muscle, or blood vessels, sarcomas are relatively rare but often aggressive.
3. Leukaemia: This is cancer of the white blood cells and generally starts in the blood-forming tissues such as bone marrow.
4. Lymphoma and Myeloma: These cancers initiate in the cells of the immune system, affecting lymph nodes and other lymphatic tissues.
5. Brain and Spinal Cord Cancers: Also known as central nervous system cancers, these begin in the tissues of the brain and spinal cord.

Understanding these classifications is pivotal in oncology, as it aids healthcare professionals in diagnosing correctly and formulating targeted treatment plans. Armed with this knowledge, both clinicians and patients are better equipped to navigate the complex landscape of cancer treatment.

The Oncology market is anticipated to double in size over the next decade, expanding its total value to over 400 billion dollars. Specifically, the cancer diagnostic segment is predicted to account for almost ~50% of the total market making it a colossal industry [6].

AI: A Healthcare Companion

Artificial intelligence is a powerful tool for enhancing cancer care throughout the patient journey. AI applications in oncology encompass various areas, such as optimizing cancer research, and predictions of cancer patient outcomes and treatment responses [8].

AI has been integrated with ease in some areas of oncology and it is improving with tremendous speed. For instance:

• The 美国华盛顿大学 developed a novel imaging AI-powered technique using their Open-Top Light Sheet (OTLS) microscopy paired with AI. Their technology scans through multiple layers of imaged esophageal biopsy specimens, identifying regions with a high potential of neoplastic growth [10].?
• Azra AI , a healthtech company that is focusing on tools to help healthcare professionals monitor and manage cancer patient care, recently launched its Patient Journey module. The module is capable of harnessing the power of AI and machine learning to analyze vast amounts of patient data, providing valuable insights into treatment effectiveness, patient adherence, and population health trends [11].
• Etcembly Ltd designed an immunotherapy drug for cancer with ChatGPT-inspired generative AI. They trained their machine learning model on naturally-occurring molecules known as T cell receptors found in cancer patients. These receptors can bring immune cells into contact with cancer cells to destroy them but they bind with low affinity to cancer cells. A generative large language model similar to ChatGPT made a new T cell receptor that can be used to efficiently connect cancer cells to immune cells and thus it was used to design a potent cancer drug [12].?
• Insilico Medicine , another pioneer in oncology, used? genAI to produce a potent, orally available inhibitor for a cell division protein kinase, a well-known enzyme for driving cancers, particularly acute myeloid leukemia (AML) and advanced solid tumors [13]. In addition to their advances using GenAI, Insilico Medicine combined two rapidly developing technologies, quantum computing and generative AI. They aim to amplify their Generative Adversarial Networks (GANs), one of their most successful generative models, with quantum computing in drug discovery which already shows remarkable collaborative strength [14].?

Many pharmaceutical and tech giants have created partnerships over the years to address as many areas of oncology as they can, see below for select examples [15].

Forging the future: 5 key questions straight from our investing rulebook

At Intelligence Ventures , we focus on all aspects of disease and the patient journey. Our goal is to unearth intricacies within the lesser-explored realms of oncology and commit to investing in AI innovations that will propel humanity against cancer. To reach these objectives, we look for the following features in an AI-enabled oncology startup.

• Clinical use: One pivotal factor we examine is the likelihood of healthcare professionals, like doctors, nurses, or healthcare providers, embracing the technology. While impressive technology and robust data are essential, if it doesn't find practical use in the clinical setting, the investment's viability may be in question, and investors may opt not to proceed.

• Patient use: Another crucial aspect we consider is the potential for patients to derive benefits from the technology. Technology’s positive impact on patients is a priority. If the technology doesn't effectively lessen the burden of cancer for patients in any aspect of their journey; treatment, diagnosis, administrative tasks, its investment prospects could be reconsidered. We all try to improve the patient's experience.

• Pros vs risks: The third feature we seek in a company is solid foundations of quantitative and qualitative data demonstrating that the benefits of the technology outweigh the potential risks. All AI-powered healthcare startups have risks. Companies must not only be aware of these risks but also actively address and advocate for their solutions, providing evidence that the technology's advantages far surpass the associated risks. This approach ensures a strong case for an investment.

• Market potential vs. market saturation: When it comes to investments, do not underestimate the significance of the market. Companies must identify their competitors and distinguish their technology. To secure a worthwhile investment, companies need to either lead with innovation or excel in some aspect of AI. It's challenging to guarantee investment success when the market is saturated and overflowing. Therefore, understand the market and set your company apart.

• Technology integration: Finally, we consider the broader perspective. Can the technology seamlessly integrate into the existing framework of cancer diagnosis/treatment/research? If not, what are the chances of it surpassing the technology currently in use? If the technology cannot be smoothly incorporated into the patient's journey or doesn't offer a significant improvement over an existing, less effective process, its prospects as an investment may be reevaluated.

At Intelligence Ventures, we are dedicated to supporting emerging technologies. We strive to recognize and nurture the distinctive qualities of companies, assisting them in accentuating and evolving to succeed in reaching their own objectives and grow.

AI in Oncology: Paving the untravelled path

While AI has made remarkable progress in the field of oncology, there is always a need for further improvements. Reflecting on one of my interviews at the Cancer Research UK Cambridge Centre for Doctoral Training, I was presented with an intriguing questionscenario: If I had all of the resources — unlimited funds, time, and support—how could I contribute to advancing the ongoing battle against cancer and achieving victory? Revisiting this question, and considering the potential of AI-enabled technologies as well as personalized medicine, my response would be as follows:

• In Biotech: Cancer is an intricate disease, with genetic underpinnings, yet the exact triggers for these genetic changes remain unknown. Moreover, the genes involved are subject to a complex network of regulation through various pathways and signals. In the realm of biotechnology in oncology, I would wholeheartedly dedicate my efforts to the pursuit of a molecule capable of suppressing cancer. There are limited advancements in cancer prevention, but if I could I would design an inhibitor of oncogenes that has no other side effects. A small molecule that could be given periodically to everyone, effectively preventing cancer! AI algorithms could also model and predict how these molecules interact with specific oncogenes, enabling us to fine-tune the inhibitors for maximum efficacy and minimum side effects. The addition of AI could offer the computational power and predictive modeling required to make this vision a reality more quickly and effectively.

• In Medtech: In the field of medical technology, my investment would be on developing a discreet wearable device, akin to a stylish bracelet. This innovative technology would have the capability to detect potential malignancies throughout the entire body. When it identifies a possible threat, the device would promptly alert the wearer with a gentle vibration and simultaneously assess the level of risk. MRIs and numerous imaging technologies can be time-consuming and emotionally distressing for individuals. The small wearable device could be an opportunity for these procedures to be made non-invasive and even stylish. By incorporating AI algorithms into this wearable technology, we can ensure more accurate and faster detection of malignancies. The AI could process real-time health data to recognize abnormal patterns that may be indicative of cancer, allowing for early intervention. Additionally, machine learning models could continuously improve the device's accuracy by learning from a wealth of data over time, making the device even more reliable and effective in its diagnoses. Incorporating AI in this manner could drastically improve the device's detection capabilities and offer a genuinely groundbreaking approach to early cancer detection.

• In Saas: In the realm of Software as a Service, I envision the creation of a dedicated platform for cancer patients. While a multitude of tools support clinicians in managing the complexity of cancer, there's a need to empower patients facing this challenge. I would develop a platform, much like existing social media platforms, where cancer survivors and patients can connect. This platform would provide a personalized feed, recommending healthcare facilities, doctors, and treatment options based on the experiences and preferences of fellow patients and survivors on a person’s follower profile. Once a person is diagnosed with cancer, it can become a significant part of a person's identity. Therefore, there's a compelling need for a thoughtfully designed platform where individuals can embrace this aspect of their lives and continue their fight against it. By integrating AI into this platform, we can offer intelligent, personalized recommendations that go beyond simply matching patients with healthcare options. The AI could analyze various data points like medical history, lifestyle, and even genomic data to tailor suggestions for treatment paths, clinical trials, or supportive communities. Machine learning algorithms could also adapt and learn from user interactions, continuously refining the recommendations and becoming increasingly effective over time. The inclusion of AI could revolutionize the way patients navigate their cancer journey, offering a level of personalized guidance that was previously unattainable.

So, why bring up this anecdote and these speculations? The answer is straightforward: these are precisely the kinds of transformative technologies that we see as prime investment opportunities. In a world increasingly leaning towards personalization and data-driven solutions, the convergence of artificial intelligence with healthcare—particularly in the area of oncology—presents an untapped market ripe for innovation. These ideas are not just thought experiments; they represent potential leaps in our approach to one of humanity's most enduring challenges. As investors, we have both the opportunity and the responsibility to identify, nurture, and scale such groundbreaking ventures that have the potential to change the very fabric of healthcare and improve lives globally.

More about Intelligence Ventures

We are an emerging venture capital firm dedicated to cultivating innovation at the intersection of artificial intelligence and healthcare within the United States. Our commitment lies in the strategic investment and nurturing of pre-seed, seed, and Series A companies, fueling their growth and fostering the next generation of industry leaders.

Our initial fund, AI Health Fund I, is focused on companies that use artificial intelligence to increase efficiencies and/or solve computationally intractable problems that place a ceiling on our ability to develop new drugs, advance them through clinical trials, and ultimately diagnose and treat patients. We are industry vertical agnostic and believe that generative AI and more specific ML models can be used to accelerate innovation in biotech, pharma, medtech, and diagnostics.

For more information, visit our website at www.intelligencevc.com or reach out to [email protected] for any inquiries. Be sure to follow us on LinkedIn and Twitter , and subscribe for further installments of The Intelligence Report .

References

[1] Hajdu, Steven I. “A Note from History: Landmarks in History of Cancer, Part 1.” Cancer 117, no. 5 (2010): 1097–1102. https://doi.org/10.1002/cncr.25553

[2] American Cancer Society. https://www.cancer.org

[3] Kowalski, Amanda E. “Mammograms and Mortality: How Has the Evidence Evolved?” Journal of Economic Perspectives 35, no. 2 (2021): 119–40. https://doi.org/10.1257/jep.35.2.119

[4] National Cancer Institute. https://www.cancer.gov

[5] "Cancer Types: A Comprehensive Overview." Cancer Research UK, October 9, 2023. https://www.cancerresearchuk.org/about-cancer/what-is-cancer/how-cancer-starts/types-of-cancer

[6] Cusabio Life Science. (n.d.). What is Cancer? https://www.cusabio.com/cancer.html

[7] “Oncology Market - Global Industry Analysis, Size, Share, Growth, Trends, Regional Outlook, and Forecast 2023 – 2032.” Precedence Research. https://www.precedenceresearch.com/oncology-market

[8] “Current Applications of Artificial Intelligence in Oncology.” Targeted Oncology. https://www.targetedonc.com/view/current-applications-of-artificial-intelligence-in-oncology

[9] Farina, Eduardo, Jacqueline J Nabhen, Maria Inez Dacoregio, Felipe Batalini, and Fabio Y Moraes. “An Overview of Artificial Intelligence in Oncology.” Future Science OA 8, no. 4 (2022). https://doi.org/10.2144/fsoa-2021-0074

[10] “Ai Rides Shotgun in the Quest for Better Cancer Diagnostics.” Fred Hutch. https://www.fredhutch.org/en/news/spotlight/2023/10/ccsg-barner-modernpath.html

[11] “Azra AI Launches Groundbreaking Patient Journey Module for Comprehensive Cancer Patient Tracking.” Business Wire, July 27, 2023. https://www.businesswire.com/news/home/20230727674666/en/Azra-AI-Launches-Groundbreaking-Patient-Journey-Module-for-Comprehensive-Cancer-Patient-Tracking

[12] Joyce, Liam. “Etcembly Reveals World’s First Immunotherapy Drug Designed Using Generative AI Technology.” Harwell Campus, August 23, 2023. https://www.harwellcampus.com/etcembly-immunotherapy-designed-using-generative-ai

[13] Belli, Brita, Bio.News Staff, and Clary Estes. “How We’re Expanding Cancer Treatments with Generative AI.” Bio.News , April 14, 2023. https://bio.news/health/generative-ai-expanding-cancer-treatments-tumors-aacr-annual-meeting-2023-insilico-medicine

[14] Ramsey, Reviewed by Lily. “Combining Generative AI and Quantum Computing to Accelerate Drug Discovery.” News, May 20, 2023. https://www.news-medical.net/news/20230519/Combining-generative-AI-and-quantum-computing-to-accelerate-drug-discovery.aspx

[15] “Ai in Oncology: Deep Pharma Intelligence.” Deep Pharma. https://www.deep-pharma.tech/oncology-ai