CHICAGO – GE Healthcare is making a deeper move into precision medicine with a series of software-focused collaborations in oncology that the firm hopes will make advanced cancer care more widely available.
The firm used the occasion of the annual Radiological Society of North America meeting to publicize deals with Sophia Genetics, the University of Cambridge, and British startup Optellum. All three collaborations are meant to augment GE Healthcare's Edison integration platform, which brings together data from multiple sources including electronic health records, laboratory and radiology information systems, imaging equipment, and other medical devices.
Ben Newton, UK-based general manager for oncology at GE Healthcare, said that the imaging giant is focusing more on outcomes than in the past. "That's really where the future is," he said. "It's not just about providing equipment. It's actually providing service, solutions, and software to help with those outcomes."
Newton said that technology like this helps assure that only those patients who truly need follow-up care receive it after initial screening.
Newton said that GE is working on multiple levels to improve rapid diagnosis of breast cancer, lung cancer, prostate cancer, and other pathologies. "Where we can more readily triage patients from screening into diagnosis, there is this ability to transform outcomes," he said.
The three agreements that GE has in the last month or so are independent. Indeed, Sophia Genetics' chief medical officer, Philippe Menu, said that that company is not working directly with Cambridge and Optellum, and declined to comment on those organizations' partnerships with GE.
However, Newton said that GE could act as a hub to bring various parties together to support precision oncology care. He also said to expect additional agreements like these.
According to Newton, breaking down data silos to inform clinicians is the main purpose of the Cambridge partnership.
Specifically, GE Healthcare will work with the University of Cambridge and Cambridge University Hospitals in the UK to develop an artificial intelligence-based application for precision cancer care.
The collaboration, which builds on earlier work at Cambridge supported by the Mark Foundation for Cancer Research and Cancer Research UK, will apply machine learning and other advanced software engineering techniques to produce this app. The partners will seek to integrate clinical, imaging, and genomic data from multiple sources into a single interface for oncologists, surgeons, radiologists, pathologists, and clinical nurse specialists to create personalized treatment plans for each patient, initially in ovarian cancer.
The alliance between GE and Sophia Genetics, announced last month when Sophia released its third quarter financial results, seeks to develop AI-driven analytics and workflow technologies for oncology, with an emphasis on matching treatments based on genomic profiles and cancer types.
Initially, they will be working to integrate data between Sophia's flagship DDM platform and GE's Edison, but Menu said that this is the first of three areas of focus. The firms also will be engaging hospitals to bring together imaging and genomics and, later, looking to develop multimodal analytics applications that draw from both technology platforms.
Menu said that Sophia and GE are particularly concentrating on radiogenomics, but noted that the biggest early challenge will be prioritizing activities. For that reason, there is no specific timetable for rolling out features.
The two firms are working now to identify clinical partners and specific experiments to undertake, but no decisions have been made. "RSNA was a great opportunity for us to meet a lot of different customers jointly with GE and discuss collaboration opportunities," Menu said.
The Optellum deal, unveiled at RSNA in Chicago, will apply the startup's decision support in an effort to improve early diagnosis and optimize treatment of lung cancer, an area of medicine often plagued by unnecessary medical procedures on patients who falsely test positive.
GE and Optellum will combine their technology to refine the assessment of lung nodules.
Václav Potěšil, Optellum's chief business officer, said that the firm has a goal of diagnosing every cancer patient and getting those individuals into treatment as early as possible. "By early stage, we really mean tiny stage Ia tumors so patients can be cured and be disease-free for the rest of their life," he said.
The challenge, according to Potěšil, is that there are literally millions of people around the world with indeterminant lung nodules at any given time. Patients with false positives might be subject to unnecessary invasive procedures such as pulmonary resections for benign nodules.
Potěšil said that Optellum addresses this problem by casting a wide net with AI-based decision support software that reads radiology reports and digital images to look for nodules at an early stage so clinicians can test for malignancies — including with genomic sequencing of biopsied tissue — and then get patients who truly are positive into treatment as soon as possible.
Optellum, founded in 2016, applies what Potěšil called a "digital biomarker" trained on about 100,000 CT scans with verified cancer diagnoses, then applies its algorithm to the imaging data. The firm gained US Food and Drug Administration 510(k) clearance in March for this technology, called Virtual Nodule Clinic, to diagnose early-stage lung cancer.
He said that Virtual Nodule Clinic is the company's way of steering the right patients into molecular diagnostics and neoadjuvant immunotherapies. Potěšil said that Optellum will soon announce its first provider-side partnership with a major US academic medical center.
GE's Newton said that Optellum's software is "exquisite" in how it stratifies nodules according to their risk of being cancerous, even those that low-dose CT might turn up false positives for. Patients falsely diagnosed then take up valuable healthcare resources and might be subjected to unnecessary treatment and emotional strain around thinking they have cancer.
"Anything we can do to classify those nodules and detect them earlier through different services and rapid diagnostic models with this sort of AI technology that [Optellum has] developed could really transform outcomes and five-year survival and bring cancers like lung to the kind of levels of survivorship that we see" with prostate cancer, breast cancer, and skin cancer, Newton said.
He said that GE hopes to integrate Optellum's platform with Edison, though the first step in that partnership is to start working with some of Optellum's provider customers, then measure the impact GE has on service utilization, the patient experience, and patient outcomes.
Newton added that asymptomatic screening for cancers including lung, breast, cervical, and prostate has declined since the COVID-19 pandemic began. That has led to a drop-off in diagnosis as well as a rush for testing as symptoms appear.
"It's putting enormous strain on healthcare providers," both in terms of providing tests and for managing patients being diagnosed in later stages of cancer.
As therapies get more targeted and specialized, the imperative of finding the right mix of patients to test new drugs increases as well. A widely cited 2016 study from the American Society of Clinical Oncology found that fewer than 5 percent of adult cancer patients enroll in clinical trials related to their cancers even though closer to 20 percent are eligible.
"Closing that gap through earlier detection methods and, obviously, characterization methods is critical," Newton said.