Hricak supports new MI subspecialty

Dr. Hedvig Hricak used her presidential address before the opening session of the 2010 RSNA meeting to express her desire for a new molecular imaging subspecialty while cautioning radiologists that MI is not appropriate for all clinical settings.

“We need a new breed of physicians with training in nuclear medicine, diagnostic radiology, radiochemistry, and molecular biology, which today is in nobody’s curriculum,” she told a full house in the Arie Crown Theater at McCormick Place in Chicago.

The requirements of the new subspecialty would include proficiency in PET/CT, MRI/PET, and optical imaging and expertise with new PET tracers.

Yet for now, not all radiologists need worry about becoming molecular imagers, Hricak conceded. Every imaging practice needs to reflect the community it serves, she said. Subspecialty MI would be as out of place in a primary care facility as general radiology would be inappropriate for imaging as it is now practiced in tertiary-care settings, she said.

The goal of personalized cancer care for every patient is still at least a decade away, Hricak said, but the radiology chair at New York’s Memorial Sloan-Kettering Cancer Institute described how imaging is helping lead medicine in that direction.

“There is no question that the ultimate goal is preemptive medicine, but the tools of molecular medicine are only emerging. For the next five to 10 years, we need to do better for prevention, early detection, and we need to engage personalized medicine,” she said.

Earlier cancer detection with cross-sectional imaging has already been responsible for significant improvements in survival, especially for breast and lung cancer patients, according to Hricak. Its effect on cancer detection overall can be seen in the virtual elimination of exploratory surgery for identifying the presence and extent of disease, she noted.

Moving forward, 3D anatomic imaging is addressing the limitation of RECIST criteria for measuring cancer progression. Accurate measures of tumor volume and mass are promising dramatic improvements in the use of diameter alone to measure tumor growth.

Modern cross-sectional imaging is not powerful enough alone to lead to personalized medicine. Along that road, quantified physiologic measures of cancer are needed to gauge its response to therapy, Hricak said. Early experience with fluorodeoxythymidine (FLT) F-18PET has shown that it can identify positive responders to a single session of image-guided radiation therapy just three weeks after treatment, compared with the three months needed to observe a response with anatomic imaging. In fact, changes in cell proliferation with FLT F-18 PET could be seen just one day after treatment, Hricak said.

“We are in an age of molecular medicine,” she said. “There has been a notable shift from one-size-fits-all to predictive, prognostic, and optically driven personalized medicine.”

The prognostic value of biomarkers cannot be overestimated, Hricak said. As an example, she pointed to a targeted treatment of colon cancer metastases with the anti-epidermal growth factor receptor (EGFR) antibody Cetuximab. Despite excellent trial results, researchers established a strong correlation between the presence of a KRAS gene mutation and the 36% of patients who did not respond to the expensive treatment. A subsequent economic analysis indicated that KRAS biomarker testing for patient selection would save more than $600 million in avoided drug costs in the U.S. alone, while significantly improving the drug’s overall clinical efficacy.

In this context, diagnostic imaging has never looked more promising, Hricak said. She predicted radiologists will excel in integrated diagnostics, biology-driven interventional radiology, and theranostics, as well as diagnostic imaging.

Integrated diagnostics will include imaging and pathology. Obtaining one or the other provides only half the information the oncologist needs. Hricak said.

For biology-driven interventional radiology, current practices already employ FDG PET/CT to precisely guide biopsy and RF ablation, but far more sensitive applications will be introduced soon.

“This is only the beginning,” she said. “The wealth of information that we can get from the molecular imaging of the future is absolutely tremendous.”

As illustrated with small animal imaging, MI will play an essential role in stem cell therapy by tracking the distribution of stem cells following their introduction to guide biopsy. In the example case, labeled stem cells were distributed unevenly, creating the risk of a negative needle biopsy if the wrong section of anatomy was sampled.

Receptor-based approaches promise to overcome the limitations of nuclear imaging posed for characterizing the heterogeneous nature of metastatic cancer, she said. Metastases differentiate themselves biologically from the primary tumor in nearly one-third of all cancer cases. This means the predictive biomarkers of metastatic lesions do not necessarily match the biomarkers of the primary tumor.

Such cases carry the risk of a mixed therapeutic response where the primary tumor may respond positively, but some metastases do not respond. A separate set of biomarkers to address this problem is needed, Hricak said. She showed examples of androgen receptor imaging with F-18 FDHT PET for metastatic prostate and breast cancers that could possibly play this role.

“For the first time, we can really see (the heterogeneous nature of different metastases) and we can use images as a predictive biomarker and find the right therapy or cocktail of therapies,” she said.

Such applications reflect the value of molecular imaging in the field of theranostics. It combines targeted imaging and with the development of targeted therapies. Imaging for these applications is crucial for finding new therapies and also for measuring their value in clinical trials.

“As we develop targeted therapies, we need to develop targeted imaging so we can select the right patient and we can follow the patient using that imaging,” she said.