Every year, the week after Thanksgiving, there is the world’s largest medical meeting in Chicago, where more than 55,000 specialists from throughout the world converge to present new and updated research in all areas of medicine.
I had just returned from that meeting and thought, instead of a question-and-answer posting this month, let’s summarize the breast imaging and breast cancer relevant results of the Radiological Society of North America’s 103rd Scientific Assembly. I was able to attend over 25 sessions, and here are some of the topics I found interesting and applicable to our specialty in years to come:
Tumor biology: Breast cancer has been classified pathologically into several types, with invasive ductal carcinoma(IDC) the most common, invasive lobular carcinoma(ILC) the second most common and several more pathological types that are infrequent. Now researchers are using additional data to classify these into sub-groups, and each sub-group will behave differently, some growing rapidly, others very slowly. This methodology is already being used somewhat in local cancer treatment, and you may have heard the term “estrogen responsive” or “triple negative.” The hope is that by developing more accurate classification of the “tumor biology,” treatment can be tailored to each individual, avoiding the more aggressive treatments in some women.
Deep machine learning: This topic was the buzz word in all fields of radiology. Using the increased data from digital imaging and the increased computing power available each year, computer programs (called AI, or artificial intelligence) are being developed to analyze images beyond what has been possible in the past. The idea is to be able to tell if a “spot” on a mammogram is likely to be cancer or not, thus decreasing the number of false alarm additional evaluations, while still detecting all of the cancers. When these computer programs become commercially available, it appears that the combination of AI and a trained radiologist will increase our accuracy in the early detection of breast cancer.
Pre-cancerous lesions: There are several lesions in the breast that are either associated with cancer or pre-cancerous, which in the past have been surgically excised. Research was presented that supports a less aggressive treatment for two of these lesions: radial scar and intraductal papilloma. We can treat these lesions with what is called “vacuum-assisted biopsy,” which allows us to remove essentially all of the lump with a needle-based technique, thus avoiding conventional surgery.
Michael J. Ulissey, M.D., is a partner at the Breast Diagnostic Centers of Auburn and Federal Way. In addition to taking care of patients locally, he continues to participate in research as an adjunct professor at the University of Texas Health Sciences Center. You can reach him at firstname.lastname@example.org.