Cancer researchers are rapidly discovering new mutations that predict a particular patient’s prognosis, or how that patient’s cancer will respond to drugs. But doctors don’t routinely use even the most basic genetic tests—covering clinically relevant mutations that scientists have known about for years. “We are not applying the insights we have today,” says Jeff Elton, cofounder of the Kew Group, a cancer care startup in Cambridge, Massachusetts. “That will get worse as time goes on.”
The Kew Group aims to change that by reforming how cancer care is both delivered and paid for. The company is developing a network of community clinics across the country, which will implement new software to help oncologists there choose the right diagnostic tests—including a growing number of molecular and genetic tests—and treatments for individual patients.
“In my mind, this is the first real coherent vision of what personalized medicine means in oncology and how, on a practical level, to deliver this to the majority of practices,” says Robert Green, an oncologist who heads the West Palm Beach Cancer Institute, in Florida, part of the Cancer Clinics of Excellence. The Kew Group recently merged with the Cancer Clinics of Excellence to create a network of 22 oncology practices operating in 15 states across the country. The centers will see about 60,000 newly diagnosed cancer patients this year.
The company’s model is based on academic medical centers, which are often good at implementing the latest diagnostic tests. The problem is that 70 to 85 percent of cancer patients in the United States get their care at community clinics rather than at academic medical centers. “That kind of testing and decision making is not available within community oncology centers,” says Raju Kucherlapati, another of Kew’s cofounders. “The goal of this new enterprise is to change that, so that patients, wherever they get treated, get the most up-to-date care.” Kucherlapati previously cofounded Millennium Pharmaceuticals and is now a professor of medicine at Harvard Medical School.
In lung cancer alone, there are eight to 10 genetic mutations that can determine what drugs are most likely to work, says Elton. In one type, called non-small-cell lung cancer, genetic testing for mutations in a gene known as EGFR can predict whether the patient will respond to a certain class of drugs. But only about a quarter of the patients get tested, he says.
The company is creating a software platform that would help physicians use an increasingly complex array of molecular tests for cancer patients. Boards of oncologists and pathologists that specialize in specific organs or specific molecular pathways will update the recommendations with new discoveries. “Our goal is to accumulate data as quickly as possible,” says Elton, who was formerly the chief operating officer for the Novartis Institutes for Biomedical Research.
Some guidelines from professional associations and other sources already exist, but they’re not updated fast enough to deal with new discoveries, and they’re not user friendly. “They are like old phone books with a huge amount of information,” says Kucherlapati. “Clinicians want to be able to cull information from the guidelines and make them useful and accessible within the context of the patients they see.”
Kew has also been working with insurers to change how cancer care is paid for. Academic centers are often funded through grants, while physician and local hospital-owned clinics, even those designated as nonprofit, often make income based on profit from prescription drugs, says Elton. That means they have little incentive to implement genetic tests that would limit the number of patients that receive those drugs.
“If [physicians] do what they are supposed to do, they make less money,” says Elton. In one case study of different treatments for the same type of cancer, the best treatment for the patient lost money for the practice, Elton found.
Rather than paying per test or per drug, Kew aims to have insurers pay per episode of care, incorporating all aspects of treatment. “Everyone’s best interest is in treating the patients appropriately, and that’s where incentives ought to lie,” says Green. He adds that molecular testing could decrease the overall cost of care, because expensive medicines won’t go to people who are unlikely to respond. “As personalized medicine is becoming more accepted, payers are becoming open to new methods [of reimbursement],” says Elton.
Because the same platform will be implemented across its network of clinics, Kew will be able to collect large volumes of data, which can be mined to determine what treatments work best, what are most cost-effective, and whether physicians are following protocol. (This data appeals to insurers, since the company should ultimately be able to show that the new pricing structure is more cost-effective with equal or better quality of care.)
The software will also help physicians find clinical trials suitable for their patients; with the rapidly growing number of diagnostic tests and experimental drugs, the best treatment option for a particular patient may be in a clinical trial. And while about five to 10 percent of patients at academic medical centers participate in clinical trials, that number is only about 1.5 percent for community centers.
Elton says that the model Kew is creating could work for other diseases as well, such as neurological disorders. These fields don’t yet have the genetic knowledge that cancer does—Elton estimates that the state of molecular testing for neurodegenerative disease resembles that of cancer five years ago. However, given the rapid pace of sequencing technology, that could change quickly.
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
Data analytics reveal real business value
Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.
Driving companywide efficiencies with AI
Advanced AI and ML capabilities revolutionize how administrative and operations tasks are done.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.