PHILADELPHIA—When surgery sampling and traditional imaging modalities come up short in assessing lymph node stage in cancer, where should physicians turn? The answer resides in molecular imaging, according to Mukesh G. Harisinghani, MD, from the department of abdominal MRI at the Massachusetts General Hospital in Boston, who spoke Wednesday at the Molecular Summit.
Current imaging techniques, such as CT, MRI and PET, do a good job in detecting cancer and metastases. But they are not always up to the task of providing accurate nodal staging, Harisinghani said.
“What we need are imaging techniques that can detect earlier cellular mechanisms that signal the beginning of cancer,” he added.
Harisinghani first detailed to conference attendees that FDG-PET imaging has advantages for detecting molecular activity compared with CT’s anatomic imaging. He showed a CT image taken two months after a patient with a gastrointestinal stromal tumor (GIST) had begun therapy with Gleevec (Novartis). CT showed that the tumor nearly doubled in size, but the PET scan revealed little FDG uptake, meaning that the metabolic activity of the cancer had dramatically diminished.
FDG-PET has its drawbacks, however, including false positives where there is inflammation and false negatives for very small lymph nodes, on the order of 2 mm to 3 mm.
Surgery is the current gold standard for staging lymph nodes, but it too has drawbacks. At times, surgeons cannot sample all nodal areas in question because of concerns of exacerbating morbidity.
“Having an imaging technique that accurately predicts the nodal stage is important because nodal staging has therapeutic and prognostic implications,” he said.
For example, for prostate cancer, a surgeon must first excise lymph nodes and then rush them to pathology. If the nodes are positive for cancer, the surgery is canceled and another treatment option is employed. The same is true for many other types of cancers. Patients with colorectal cancer and no lymph node metastases have an 80 percent chance of survival. With positive nodes, survival drops to half.
Harisinghani explained, with slight bemusement, the current method for characterizing nodes. “If they are oval and less than 10 mm, we call them benign. If they are oval and greater than 10 mm, we call them malignant.” The problem with this method is that there is a significant overlap in size between benign and malignant nodes. “It’s worse than flipping a coin,” he said.
Developing molecular nanoprobes
When developing a molecular probe, researchers first want to attempt to modify currently available imaging techniques, Harisinghani said, with an eye toward sensitivity and spatial resolution.
PET, SPECT and optical imaging have the necessary sensitivity, while CT, MR and ultrasound have the spatial resolution. But “structure without function is a corpse, and function without structure is a ghost,” he said. “We will need to have a combination of structural and functional imaging to achieve our goal.”
At MGH, researchers determined that MRI is much better than CT to detect small lymph nodes. They then examined the underlying biological processes involved and decided to develop a probe that would target macrophages, which are present in healthy tissue but not in cancerous tumors, and that would be visible with MRI.
The nanoparticle they developed binds to macrophages. On MRI, noncancerous lymph nodes appear dark, while cancerous nodes appear white. Harisinghani and colleagues successfully used the probe to characterize nodes as small as 2 mm. To date, they have shown success with the probe in various cancers, including head and neck, breast, prostate, bladder, cervical and testicular.
How does it translate into patient care? Studies have shown that using nanoparticle MRI can improve nodal staging and surgical sampling. They have also shown it is useful for radiation therapy planning, predicting response to therapy, and in detecting recurrence.
“Now we have an elegant way to show which nodes are affected with cancer,” he said.