Understanding the genetic mutations underlying cancer subtypes is a powerful tool in helping interpret MRIs, CT scans, and other imaging studies. Genetic sequencing is now routinely performed in some cancers to clarify the cancer’s biological behavior and identify the best drug targets.
Working with Sylvester Comprehensive Cancer Center colleagues, Francesco Alessandrino, M.D., assistant professor in clinical radiology at the University of Miami Miller School of Medicine, is working to extend the universe of cancers for which this analysis is routine practice.
“’The eye sees only what the mind is prepared to comprehend,’” Dr. Alessandrino says, quoting novelist Robertson Davies. “When we know what patterns are likely, we know where to focus our lens. We look more closely or more expansively to see what we might not have otherwise seen.”
In a 2020 article published in Radiology, Dr. Alessandrino and his then-colleagues at Harvard Medical School laid out correlations between genomic profiles and imaging patterns in urothelial cancer patients. The retrospective study demonstrated that knowing specific mutations can help highlight “hot spots” on the CT scan.
“We now know that certain mutations correlate with nodal or osseous metastases, while others correlate with peritoneeal metastases,” Dr. Alessandrino said. “This work makes a case for early molecular analysis of mutations and helps connect the dots for radiologists.”
Now at the Miller School of Medicine, Dr. Alessandrino is expanding his work to other cancers. In collaboration with Marilyn Huang, M.D., associate professor of gynecologic oncology and director of translational gynecologic oncology research at Sylvester, he is working to elucidate these correlations in endometrial cancer, with preliminary results presented at the Society of Gynecology Oncology Annual Meeting.
A Reverse Interrogation of Tumor Spread
In the rapidly evolving field of radiogenomics, features noted during imaging may later be linked to genomic factors through genetic sequencing. In the even newer reverse approach known as genomic radiology, genomic factors are identified first and – based on that knowledge – imaging features then isolated and followed.
“We are hoping that genomic radiology will shed light on the biological behavior of these cancers and help us identify metastases earlier,” Dr. Alessandrino said.
In the 2020 study, Dr. Alessandrino and his team analyzed 103 patients with urothelial cancer, who underwent genomic analysis and were also staged with cross-sectional imaging, including CT, MRI and PET/CTs. They examined the genomic mutations by likely types of metastases, finding that TP53 mutation was associated with 1.7-1.9 times the chance of osseous or nodal metastases, while patients with RB1 mutations were 5.9 times more likely to have peritoneal metastases.
They also subdivided the cases into a high-risk group (including TP53 and/or RB1 and/or KDM6A mutations), and a low-risk group (having ARID1A, FGFR3, PIK3CA, STAG2 or TSC1 mutations).
They found that the high-risk mutational group had a metastasis-free survival (time from diagnosis to the development of metastases) of 3.5 months, versus 17.3 months for the low-risk group. The overall survival was 19 months versus 24.4 months.
Brighter Prospects for Patients with Metastases
Peter Choyke, M.D., director of the Molecular Imaging Branch of the National Institutes of Health, commented on Dr. Alessandrino’s study in an editorial in the same issue. “Together with artificial intelligence, the arrival of genomic radiology will further increase the value of oncologic imaging,” he wrote. “By building on these archives (e.g. Cancer Imaging Archive and The Cancer Genome Atlas), it should be possible to provide genome-based predictions of radiologic findings for the broad spectrum of human cancers.”
While the cost of whole-genome cancer genome mapping is dropping to below $1,000, Dr. Choyke noted that much of this information is not currently used, least of all by radiologists. Further, its accumulation is expensive. “The potential improvements suggested by Alessandrino and others will require substantial investment in matched genomic and radiologic data sets for a wide spectrum of cancers,” he wrote.
“Patients with metastatic cancers have increasingly better prospects,” Dr. Alessandrino said. “More systemic therapies are available, and there is increased knowledge of the clinical impact of certain mutations.
“There is a lot yet to understand about why patients develop metastases and why a cancer metastasizes in a certain way. Our study and current work moves us in the direction of learning more.”