Q&A: Developing lung cancer detection technology
Václav Potěšil, chief business officer of Optellum tells Med-Tech Innovation about its technology for lung cancer patients and the support it has had from EIT Health.
What is Optellum? What does it do?
Optellum’s software identifies and tracks at-risk patients through AI-powered biomarkers, assigning a lung cancer prediction score to specific lung nodules. This is important as it could transform life expectancy through much earlier diagnosis, and it uses the largest data set of CT scans in the world.
You have recently received government funding for AI-based early diagnostic lung cancer software, what made you go down the AI route?
We realised early on that machine learning applied to huge datasets of CT scans would help doctors determine patient’s risk of lung cancer in a far more accurate, consistent and less harmful way than the existing methods.
Lung cancer is currently the most common type of cancer and the leading cause of cancer deaths in the world, killing approximately 1.8 million people worldwide die each year.
The current five-year survival rate is only 20%, primarily due to most patients being diagnosed after the disease has progressed to an advanced stage (Stage III or IV). However, the survival rate for small tumours treated at Stage IA is as high as 90%. Early diagnosis, while the disease is still at a curable stage, is key to changing it from a deadly sentence into a disease that is survived by most patients.
The only way such pre-symptomatic lung tumours can be identified and intervened on at an early stage is through Computed Tomography Imaging available as the standard of care in every modern hospital. The Chest CT enables early detection of lung nodules, small lesions in the lung that may be early-stage tumours or harmless findings. Those are frequently found incidentally in scans acquired for other reasons (cardiac scans, emergency scans) or through preventative CT lung cancer screening of patients with history of heavy smoking.
Understanding the value of applying AI to this data, led to the forming of Optellum in 2016, to develop a commercial product and bring the benefits to clinicians and patients.
The software assigns a lung cancer prediction score, can you tell us what informs this score, and how accurate it is?
The Lung Cancer Prediction (LCP) score is powered by the world’s first FDA-cleared imaging AI/Radiomics based digital biomarker for lung cancer. The score is computed from full patterns of 3D pixels in standard images captured by CT scanners, using a neural network trained using the world’s largest curated database of scans.
With support provided through EIT Health (and other partnerships), our researchers have been using CT images to train a machine-learning neural network to recognise the signs of the deadly disease.
Our solution was designed and validated to help pulmonologists and radiologists make more accurate and consistent assessment of malignancy risk. This means, working together with the clinicians for the benefit of the patient, rather than replacing a clinician.
The LCP AI accuracy goes significantly beyond any non-invasive method available in today’s clinical practice. It has been extensively validated in multi-centre studies led by co-authors of clinical guidelines, and shown to consistently outperform conventional risk prediction models, currently recommended in clinical guidelines, and considered state-of-the-art in classifying nodules as low, intermediate, or high risk.
In one independent validation study led by physicians from Vanderbilt and Oxford, the AI was shown to correctly reclassify indeterminate nodules into high- and low-risk categories in more than a third of cancers and benign nodules, illustrating the potential to speed up lung cancer diagnosis and reduce invasive biopsies and surgeries on patients without lung cancer, compared to the current standard of care.
How will your product innovate the detection of lung cancer? What impact will it have on healthcare?
The successful performance of the Optellum AI could have a huge impact on the diagnostic workup in lung cancer screening, leading to fewer false positives and directing medical resources to those with the greatest need, and so facilitate a resource-efficient roll-out of CT lung cancer screening programs across Europe.
Our product will allow patients to benefit from earlier diagnosis and treatment of cancerous nodules, leading to better survival rates and outcomes.
Healthcare providers benefit from decrease waste of resources, focusing expensive tests and treatment procedures for the patients who need them. They also benefit from reduced care variation and the provision of the highest possible quality of lung cancer diagnosis and treatment, for every patient in every hospital.
You’ve worked closely with EIT Health, NIHR, Innovate UK and lately struck a partnership with Johnson & Johnson, can you briefly tell us about the role these organisations have played in the product’s development?
EIT Health, National Institute of Health Research (NIHR) and Innovate UK are all organisations that have supported us on our journey. Their research, advice and innovation grant funding has allowed us to collaborate with leading hospitals, to develop the software and validate it on diverse data sets from patient populations.
More specifically, how has EIT Health supported Optellum’s journey?
EIT Health has been an incredible supporter since day one. In 2016 we won our first EIT Health grant, where we were amongst those chosen to take part in the EIT Health Catapult programme.
We also collaborated on the Early Lung Cancer Diagnosis Using Artificial Intelligence and Big Data (LUCINDA) project which leveraged the expertise of the top radiologists and helped develop and validate the Optellum Lung Cancer Prediction (LCP) AI as a clinical tool. It allowed us to showcase how the LCP has robust performance across various patient populations, clinical pathways and CT scanner types.
Throughout all the projects we were involved in, we were very impressed by the entrepreneurial and dynamic nature of EIT Health team. They provided invaluable support and guidance while developing the grant applications and during the execution of the projects, without imposing high administrative burden (typical of many other government fundings) which can be distracting for a rapidly growing start-up company.
From our first proof of concept, before Optellum was founded, to product development and clinical trials, EIT Health support has been essential to see this work realise its potential. Today, that support has taken us all the way to having a cleared medical device, which is going to help real patients receive better care and save lives.
What plans do you have going forward? What is the next big step for Optellum?
Our Virtual Nodule Clinic is already used in clinical care by a first wave commercial customers and leading medical centres across the USA and we look forward to rolling out in other key markets soon.
We aim to expand Optellum’s portfolio to transform early lung cancer therapy, enabling optimal treatment decisions with the aim of treating patients earlier, potentially at a pre-cancerous stage.
At the core of our plans is an “Optellum Inside” strategy. We are working with partners to harness the power of Optellum’s AI software, to drive the right patients into molecular tests, interventional and therapy devices such as robotics, and drug therapies. Optellum recently announced strategic partnership with Johnson & Johnson and GE Healthcare.
Where do you want to be in 10 years’ time?
Our goal in ten years' time is to be considered the standard of care in early diagnosis and treatment of lung cancer and other deadly diseases of the lungs such as Interstitial Lung Diseases and COPD.
We hope the routine use of Optellum technology will help clinicians around the world to drive the right patients into the early diagnostic and treatment procedures. And that eventually they will be able to intervene at a pre-disease state, to prevent the disease before it has the chance to develop any clinical symptoms.
Anything else that you’d like to add?
Our success is due to a culture of collaboration and excellence, which places quality, clinicians and our patients at the heart of everything we do. We would like to thank everyone – our team of scientists, developers and other colleagues, our clinical partners at leading hospitals, our partners such as EIT Health - for working together with us to give patients with lung cancer and other deadly diseases of the lungs, the best chance to fight back.