By Rose Higgins and Mimi Huizinga, MD, MPH, FACP
The introduction of genomics was a significant advance in the battle against cancer. The ability to understand how specific genes predict the likelihood of acquiring cancer or the success of a particular therapy added a new degree of precision and reliability to diagnosis and treatment.
Radiomics is poised to be the next great advance. This relatively new technology enables healthcare and life science organizations to take traditional images, such as MRIs and PET scans, and use artificial intelligence to extract hundreds of data points about the biology of a tumor or lesion. By comparing this newly available data to past images, as well as the biology of healthy organs, clinicians can gain a much deeper, ongoing understanding of how the tumor or lesion is responding to a specific therapy, informing their care and treatment decisions along every step. Here’s how it works throughout a patient’s cancer journey.
Early identification and diagnosis phase
Traditional images offer a point-in-time view of a tumor or lesion. When a radiologist characterizes a tumor as something to watch, that statement is not definitive because there isn’t enough data about it. A second reading at a later date will show whether the tumor has grown along the long and short axes, but still may not offer enough information to determine whether something should be done—and if so, what. In the meantime, valuable time is lost.
Radiomics provides quantitative details about the tumor’s biology that traditional images lack. Researchers and clinicians gain information about its volume, texture, and many other factors in a noninvasive, low-cost way. When radiomic data is combined with clinical and genomic information, oncologists are better able to personalize a course of treatment to produce the best outcomes.
An example of this type of predictive capability comes from research by the Moffitt Cancer Center in Tampa. When screening patients with low-dose computed tomography (LDCT), researchers discovered two biomarkers for lung cancer that can help stratify patients into risk groups related to five-year survival rates: Statistical Root Mean Square and Neighborhood Grey Tone Difference Matrix Busyness. Based on these features, they were able to accurately identify patients as high risk (0% five-year overall survival) or low risk (78% five-year overall survival).
Armed with this type of predictive information, clinicians may recommend more “aggressive follow-up and/or adjuvant therapy to mitigate [high-risk patients’] poor outcomes,” according to the study.
The value of radiomics during treatment begins with using the tumor’s biology to define a novel set of quantifiable patterns. These imaging biomarkers provide clinical signatures that are as unique as each patient, enabling clinicians to create an individualized treatment plan.
Once the patient goes into treatment, the traditional approach often becomes a matter of “wait and see.” Yet since standard images do not always provide enough immediate detail about whether the current treatment is producing the desired result, another biopsy may be ordered. These subsequent biopsies greatly increase the risk to the patient, especially for cancers such as glioblastoma that can settle deep into vital organs, so oncologists may hesitate to order them—leaving them in the dark as to the effectiveness of the treatment.
Radiomics delivers a similar view to a biopsy but is noninvasive, making it far easier for oncologists to track the treatment’s progress. Oncologists can gain the precise, detailed information they need to determine whether the treatment is producing the desired response while eliminating the risk to the patient and lowering the cost of care.
If the treatment is not proceeding as expected, oncologists can examine which key parameters, such as texture and heterogeneity, are not trending in a positive direction. They can then quickly change treatment courses based on the growing body of knowledge about cancer treatments. For example, they may elect to discontinue chemotherapy if it is determined to not be having the desired effect, saving the patient the distress that accompanies it. Or they may choose a different therapy combination based on the literature and how the tumor or lesion is reacting to the current treatment.
Radiomics can also significantly impact the way life sciences organizations develop and monitor their clinical trials, especially as more evidence is created and additional imaging biomarkers are identified.
It can be used in the initial stages of a clinical trial to identify patients who are more likely to be helped by a specific therapy and therefore should be included in the trial. It can also identify those who should be excluded based on a comparison between their tumor or lesion biology and known outcomes for similar cases.
During the clinical trial, radiomics can identify previously undiscoverable changes that point more precisely to the efficacy of a treatment course without the need for a biopsy. This information can help researchers come to a faster conclusion regarding whether to move to the next phase, whether certain participants should continue to be part of the trial, and other decisions.
The key factors enabled by radiomics are the ability to monitor safely at a deeper level and make course corrections quickly so no time is lost, helping ensure better decisions and/or outcomes.
In this phase, again, radiomics plays an important role. Cancer survivors routinely obtain medical images to enable their doctors to monitor their recovery. Radiomics can reveal the tumor’s biology and compare it to prior images, ensuring all key factors are examined and no key clues to tumor response/behavior are missed.
If the radiomics show there have been no changes since the last image, patients can be reassured to a far greater degree than with a traditional image. On the other hand, if there is a developing issue, oncologists will know right away so they can address it well before they might have otherwise.
Winning the battle
The more we understand about various types of cancer, the better chance patients have of surviving their encounters with it. Radiomics offers a powerful new weapon in the battle against one of the most prolific chronic conditions we face.
Rose Higgins is CEO of HealthMyne, a pioneer in applied radiomics, which is the cutting-edge field of extracting novel data and biomarkers from medical images. Mimi Huizinga, MD, MPH, FACP, serves on the HealthMyne board of directors and is the vice president and head of U.S. Oncology Medical, Novartis. Dr. Huizinga is a physician executive with deep experience achieving clinical excellence across the care continuum driven by evidence-based care, data, and analytics.