New Frontiers: Molecular Imaging & the OR

Molecular imaging is spurring dramatic shifts in medicine. The latest venue to witness the revolution may be the surgical suite. The Advanced Multimodality Image-Guided Operating (AMIGO) suite at Brigham and Women’s Hospital (BWH) at Harvard Medical School in Boston weds intraoperative imaging and navigation systems in a surgical environment to set the stage for intraoperative applications of molecular imaging.  

Angela Kanan, RN, describes the 5,700-square foot, three-room suite as a laboratory. “[AMIGO] has every type of imaging modality you can have—ultrasound, MRI, CT, PET/CT, x-ray fluoroscopy. We can fuse MR and PET images.” Unlike conventional specialized suites, AMIGO is available for use by surgeons across the enterprise, contributing to the interdisciplinary laboratory model.

The rationale

Twenty years ago, BWH ushered in the era of intraoperative imaging with an intraoperative MRI system. Today, more than 150 intraoperative MRI systems are deployed globally. However, MRI does not represent a complete intraoperative solution. “We are entering the decade of molecular imaging. If these tools aren’t available in the OR, we will miss the revolution,” says Ferenc A. Jolesz, MD, co-principal investigator of the National Institutes of Health-supported National Center for Image-Guided Therapy at BWH.

The rationale for implementing PET/CT intraoperatively is simple—the diagnostic imaging modality should be the same method used during surgery or therapy. That’s because the anatomy shifts during surgery. Pre-operative images do not represent the same anatomy that surgeons visualize in the OR.
Abdominal procedures, for example, are prone to intraoperative anatomical shifts. The transition to laparoscopic procedures hinges on the use of carbon dioxide to enlarge the abdomen and accommodate surgical instruments. “This displaces everything. Anatomical shifts may be as much as 10 cm,” confirms Jolesz.  

In both cases, pre-operative images no longer provide an accurate roadmap to malignancies. However, surgeons face binary decisions: either remove the tissue or leave it in. Cancerous tissue must be removed, while normal tissue remains intact. “Without accurate intraoperative imaging data, surgeons don’t know where the cancer is during surgery. Surgeons need sensitive and specific imaging information,” says Jolesz.

Applications in action

Jolesz et al aim to demonstrate that molecular imaging can aid surgical performance. BWH is pioneering the model; AMIGO was installed in May 2011, and its PET/CT system launched in May.

One early application is image-guided thermal ablation. Jolesz envisions PET/CT filling an information gap. Currently, surgeons rely on an indirect measure of tumor coagulation necrosis by ablation. They assume heat-induced necrosis occurs when the tissue has reached the critical temperature of 60 degrees Celsius. Similarly, after cryoablation, surgeons assume that the malignant tissue has reached a freezing temperature inconsistent with survival.

“PET/CT offers a new opportunity to determine whether or not tissue is viable after the treatment. Molecular imaging can show if the target is metabolically active,” explains Jolesz.  “After ablation, the surgeon can use molecular imaging for live updates to determine if the target has been killed by ablation. If the tissue no longer shows uptake of the tracer, it provides direct evidence of tissue death.”

The researchers continue to refine procedures. One challenge is tumor 18F-FDG activity does not immediately dissipate with percutaneous radiofrequency ablation or cryoablation. Thus, surgeons can use 18F-FDG PET/CT for targeting or demonstrating complete treatment, not both. If 18F-FDG is injected after ablation, then areas with coagulative necrosis will not take up the tracer, while residual, untreated tumor parts will be seen on the PET image.

One solution is to use tracers specific to the tumor labeled with short half-life isotopes, such as Carbon-11, before the procedure, for target definition, as well as after the procedure to ensure that the ablation was complete and involved the entire tumor volume. The first injection shows the untreated tumor target. After that isotope decays and ablation is complete, a second injection is administered; any viable remaining tumor will show uptake of the tracer.
The BWH team also envisions using PET/CT as a GPS-like navigation tool to locate malignant lymph nodes or metastatic lesions intraoperatively. “There are thousands of lymph nodes in the abdomen. [Without intraoperative PET/CT], finding malignant lymph nodes during surgery is like finding a needle in a haystack,” says Jolesz.

In these cases, time can become an issue, so the surgeon may be forced to conclude the procedure prior to locating malignant lymph nodes. Or he or she may remove an incorrect node. The plan is to leverage intraoperative PET/CT to accurately identify the correct lymph nodes and guide the endoscope to the exact location, resulting in a shorter, more accurate procedure.   

Another early application is brachytherapy for cervical cancer. “Success requires accurate positioning of the radioactive seeds and avoidance of sensitive structures such as the ureter and rectum,” explains Jolesz. Currently, surgeons use intraoperative MR guidance to position the seeds, which has resulted in an 80 percent cure rate in AMIGO. Jolesz believes intraoperative FLT PET/CT could provide additional gains by allowing surgeons to identify areas of the tumor that are already dead and don’t require treatment. The strategy also could be used to confirm that a tumor is malignant.

Complementary technology

AMIGO is equipped with an array of systems and tools to optimize operations. One key system is a small intraoperative mass spectrometer. In the conventional surgical model, when tissue is removed, it’s sent to the lab and results arrive one week later. “This doesn’t help the surgeon intraoperatively,” says Jolesz.

In AMIGO, surgeons can remove a tissue sample and place it in the spectrometer. Within five minutes, the surgical teams knows if the sample is malignant or not, and can use the information to inform intraoperative decision-making. Currently, the approach is used in brain tumors and breast cancers, which are characterized by ill-defined boundaries and normal tissue infiltration.

Another essential ingredient is an onsite cyclotron that produces short-lived radioisotopes such as Carbon-11, Nitrogen-13 and Fluorine-18.
Finally, AMIGO is equipped with a sliding board to move patients from the OR to the PET suite. When a patient requires transfer from the OR to the PET/CT scanner for imaging, the sliding doors between rooms are opened, and a bridge is positioned between the OR table and the PET/CT table. The patient is shuttled into the scanner on an MRI/PET/CT/x-ray-compatible transfer board, which he or she never leaves throughout the procedure.

Looking ahead

Jolesz envisions additional surgical applications for AMIGO, and the team aims to streamline research focused on the validation of molecular imaging agents.

The current validation model for molecular imaging agents is arduous and expensive. It requires multi-center trials and thousands of patients, each of whom are injected and biopsied to produce a single data point—false negative, false positive or correct.

“With AMIGO, we can provide the same data after 20 procedures,” says Jolesz. Each surgery yields multiple samples of tumors that are heterogeneous in nature. Thus, the researchers can measure the absolute level of the molecular imaging agent at multiple points and compare them with MR data to provide an accurate assessment of the agent’s sensitivity and specificity.

As Jolesz and colleagues ponder the future, they also are building a model that brings the capabilities of molecular imaging into surgical suite. Just as intraoperative MRI ushered in a new era in neurosurgery, intraoperative PET/CT also will deliver new applications and improved patient care.