From targeted therapy to precision medicine
Harrison continued a low dose of chemo until 2015 and remained in remission until 2016, when cancer returned to his central nervous system. This time, targeted therapies had gained even more traction in the health care industry, and doctors found another new treatment protocol, CAR T-cell therapy.
“CAR T therapy empowers the immune system to recognize and kill cancer cells,” says McKinion. Your own white cells attack cancer cells the same way they would attack a bacteria.”
Today, thanks to an amazing combination of factors – including persistence, targeted therapies and immunotherapy – Harrison is an active, cancer-free teenager.
“At every stage, Harrison was just months beyond a cutoff date,” says McKinion. If he had presented with cancer earlier in either case, the treatment that saved his life would not have been available."
McKinion, who has a doctorate in divinity, has become an advocate for families affected by childhood cancer. He testified before Congress in 2016 to recommend more funding for childhood cancer research, and he is co-founder of Harrison’s House, a nonprofit in Wake Forest, NC, that provides camp experiences for teens with cancer, serious illnesses and special needs.
“If it wasn’t for data and the use of big data in research, Harrison would have been dead seven years ago," says McKinion. "Now he’s about to graduate from high school. He’s playing baseball and soccer, and he wants to go change the world.”
What career plans does Harrison have after college? “He wants to study data and analytics,” says McKinion. “His goal is to change the world through data because he knows what it’s meant to him and his life.”
SAS helps fund childhood cancer research
Cancer research is a costly endeavor. Identifying better treatments, building capabilities and developing support programs are all resource-intensive and time-consuming. To truly make an impact with its financial contributions, The Kids' Cancer Project uses SAS to improve the efficiency of its donor management process.
Cancer treatments get personal
“Because cancer is so individualized, treatment has to be correspondingly precise,” says McKinion. “Plus, precision medicine is less toxic. If you can’t be precise, you have to carpet bomb with chemotherapy, versus the laser bomb of targeted therapy or precision medicine.”
Especially in childhood cancer, the collateral damage is so great that precision medicine helps to prevent some of the most damaging side effects of transfusions, joint replacements, bone marrow transplants and more.
“Today’s therapies are developed to be more personalized, precise and data driven than ever,” explains Mark Lambrecht, PhD, Director of the Health and Life Sciences Global Practice at SAS.
Steve McKinion advocates for childhood cancer research.
As cancer research and cancer treatments evolve in this new world of precision medicine, analytics plays a crucial role. From treatment innovation to clinical research, and from medical imaging to early efficacy tests – the importance of analytics in cancer care and diagnosis cannot be overlooked.
“Analytics not only helps discover new therapies, but also measures their impact once they are in use, and it helps to financially reward the therapies that deliver the highest patient value,” says Lambrecht.
The type of therapy that eventually put Harrison in remission when his cancer relapsed, CAR T-cell therapy, works by enhancing a patient’s T-cells to fight off the most resistant cancer cells. The treatment uses a combination of cellular therapy, gene therapy and immunotherapy to boost a patient’s immune system to become an advanced cancer-fighting machine.
A history of SAS in cancer research
The FDA requires SAS v5 transport format for submissions, making SAS the primary data submission format for nearly all cancer clinical trials.
SAS is a founding member of the CEO Roundtable on Cancer, a nonprofit dedicated to helping eliminate cancer as a personal disease and public health problem.
SAS starts collaborating and working on the next generation of clinical data standards through CDISC.
The human genome is sequenced, ushering in the possibility of precision medicine and providing more data for SAS and others to analyze.
NIH uses SAS to develop risk assessment tools for the public to better understand the risks for certain cancers.
ClinicalStudyDataRequest.com launches, making data from 4,000 trials available to academic researchers.
CDC uses SAS to verify relative survival rates for cancer.
Joining forces to understand cancer
In Norway, the Oslo Cancer Cluster – a nonprofit organization with more than 90 globally connected organizations working in the cancer field – has a unique interest in personalized medicine. The members cover the entire oncology value chain, including hospitals, research centers, patient organizations, biotech startups, biopharmaceutical companies and technology providers like SAS.
One of the many projects underway at Oslo Cancer Cluster is research into CAR T-cell therapy and its effectiveness against solid cancer tumors.
Currently, CAR T-cell therapy is more effective in treating blood cancers like leukemia and lymphoma. However, the development of CAR T-cells for nonblood, or “solid,” cancers has been more difficult.
Researchers at Oslo Cancer Cluster have discovered a new way to model solid cancer cells in order to test CAR T-cells on those models
The ultimate goal of the consortium is to uncover the best treatment for each unique patient. Analytics is a crucial part of that effort.
“In cancer treatment right now, there is a digitization of oncology,” explains Ketil Widerberg, General Manager of Oslo Cancer Cluster. “Analytics can achieve two things for cancer research. One is to understand cancer better, to be able to see patterns we didn’t see before. Secondly, we use analytics to understand how we can treat patients better, to give the right treatment to the right patient at the right time.”
Data sharing offers hidden insights
While immunotherapy focuses on medicine at the personal level, many efforts to advance cancer care require data from a lot of different patients.
Clinical trials, including those for immunotherapy treatments, are intentionally and meticulously designed to study the specific effects of a specific therapy on a specific type of patient. This level of control is needed to ensure the safety and efficacy of the treatments being studied.
But researchers today are realizing that the answers to many complex questions lie outside of those controlled study environments – and one solution is to open clinical trial data for analysis across multiple clinical trials. Consider these three programs that combine data for advanced research:
- The Project Data Sphere cancer research platform sparks innovation by opening data to new research possibilities. To support these efforts, SAS hosts the research platform and provides access to analytics technology at no cost to researchers. The data in the platform is deidentified and consistent with industry requirements.
- BioGrid Australia, a nonprofit organization, combines data from hospitals and research institutions to improve clinical research into various cancers, like bowel cancer. The project brings together data from patient screenings at 19 sites across the country.
- The Latin American Cooperative Oncology Group (LACOG) explores data from thousands of cancer patients at more than 150 hospitals and 15 countries throughout the region. The project is helping identify known roadblocks to optimal care, such as poor access to treatment, medication and preventive care. The long-term goals are to better serve patients, facilitate the development of new techniques and technologies to improve cancer care, and even affect public policy.
Greater access to patient-level clinical data is good for patients, science and business. From a business perspective, pooling data can significantly reduce the cost of drug development while improving the efficiency of clinical trials.
Cancer detection with computer vision
The data gathered by the programs above has been used to validate the importance of early detection and treatment of cancer. As Harrison’s story shows, the first step in treating cancer successfully is first identifying the cancer. Likewise, being able to detect whether his treatment was working was crucial to Harrison’s treatment path.
“A large number of kids with immunotherapy respond immediately. But some do not. You have to figure out for some kids why it doesn’t take. The possible answers are so vast, there’s no way to do it without big data,” says McKinion.
Ultimately, the sooner doctors can identify what is or isn’t working, the sooner they can put patients on a path to a treatment that works for them.
At Amsterdam University Medical Center, a team of radiologists and surgeons recently began using computer vision, a form of artificial intelligence, to improve the process of identifying cancer tumors and measuring the changes in tumors after treatment. Their initial project uses object detection to identify and measure tumors in CT scans of livers from patients with colorectal cancer that spread to the liver.
Typically, radiologists measure the size of tumors manually in the scans before and after treatment. This is very time consuming and is prone to subjectivity, but it is crucial work. If the patient’s tumors are responding to treatment, that also makes the patient a good candidate for surgery.
Leaders at the hospital saw this as a perfect pilot project to test the capabilities of analytics and AI. Computer vision models are constructed to analyze the medical images in a fraction of the time, and object detection is used to recognize tumors and tumor sizes almost instantaneously. AI models are more objective and accurate than the radiologist’s measurements alone.
This use of AI not only frees up radiologists to do more hands-on work with patients, but also improves decisions and therefore could potentially save lives. By finding results faster and more accurately, computer vision can improve treatment strategies with the potential to lead to better outcomes of cancer patients.
Lambrecht predicts many other uses for AI in the fight against cancer, including the ability to generate biomarkers with deep learning that can better determine endpoints for treatment based on safety and efficacy.
Harrison McKinion, a high school senior, is healthy and in remission.
Advancements save lives
The entire landscape of cancer treatment and cancer research is changing.
Clinical trials, which used to be broken down into phases, will soon be replaced with umbrella studies that seek to study multiple candidate therapies in a single trial. This differs from traditional trials that were designed to investigate a single treatment across a standard population.
These more sophisticated studies often start by screening participants for hundreds of cancer-related genes and then assigning patients into a specific treatment path within the trial based on the genetic results.
“These studies also have the potential to greatly increase the number of patients who are eligible to receive certain drugs relative to other trials’ design,” explains Lambrecht.
Of course, umbrella studies and other newer research formats for precision medicine generate more data and more complex data than standard trials. But today’s advances in analytics – including new machine learning algorithms and faster processing speeds – are up to the task.
Cancer patients and their families, like the McKinions, are grateful.
- IoT in health care: Unlocking true, value-based careGiven the potential of IoT – and the challenges of already overburdened health care systems around the world – we can’t afford not to integrate IoT in health care.
- The Humanity in Artificial IntelligenceCould artificial intelligence be the change agent we need to solve many problems around the globe? Read how AI could accelerate our ability to have a a positive, lasting impact.