A team of neurosurgeons has, for the first time, combined real-time MRI technology with non-invasive cellular mapping techniques to develop a new biopsy approach that increases the accuracy of diagnosis for patients with brain cancer. Here, Practical Neurology TM speaks with Clark C. Chen, MD who is vice-chairman of research and academic affairs in the division of neurosurgery at the UC San Diego School of Medicine, who was part of the team behind the approach.
Why is there a need for this technology? How is this better than existing methods for biopsy?
For most solid tumors, including brain tumors, definitive diagnosis is required and defines subsequent treatment for our patients. And this definitive diagnosis requires meticulous securement of the abnormal tissues and careful pathological examination of these tissues. This process is particularly challenging for a number of cancers, including the most common form of primary brain tumors, glioblastoma multiforme. The reason this is particularly challenging is—as the word multiforme implies—that this disease is notoriously heterogeneous in terms of its histological appearance. In other words, I can sample one area of the tumor and then sample the next area. The histologic appearance of these areas would look different. So, if you don’t have adequate sampling of the entire tumor, you risk a false diagnosis that could lead to inadequate treatment for the patient. In the literature, the misdiagnosis occurs in as high as 30 percent of brain tumor patients. Our goal is to use the most updated technology to avoid misdiagnosis and minimize the risk of a biopsy.
Now, regarding the second question. Historically, brain biopsies are done in a “blind manner.” And what do I mean by that? As you know, the skull is an opaque structure. So, there’s no way for me to look inside the skull and see where my biopsy needle is precisely located in real time. The way we’ve learned to preform these biopsies is by using mathematical triangulation. Essentially, a frame is mounted on the patients head with four rigid posts. We localize the radiographic abnormality relative to those four posts. So, in a stereotactic needle biopsy, the surgeon does not have a real feel in terms of exactly where the biopsy needle is located or what is happening at the biopsy site. As a result, most surgeons are fairly conservative about biopsy of multiple regions in the tumor. And you can imagine if the tumor is heterogeneous and different regions contain different cellularity, different cell structure, a limited sampling increases the chance of misdiagnosis. On the other hands, biopsy of multiple areas increases the risk of the procedure. So, there’s a fine balance between sampling the entire tumor and the safety of the procedure using the traditional method.
What can you tell us about the technology?
There are two particular technologies that are married in this procedure. The first technology is an advanced MRI visualization technique called Restriction Spectrum Imaging. This method allows us to define the density of cells in different regions of the tumor. We refer to this as the cellularity density of the region. Restriction Spectrum Imaging does this by determining water molecule movement in the region. Basically, the more water movement there is, the less cells there are in the region. By examining the cellular density map of the tumor, we can intelligently select the biopsy target sites, as to minimize the number of the biopsies and still have a good sampling of the entire tumor. This is very similar to conducting a political poll in terms of surveying public opinions. When you take a poll, you don’t go out and survey every person in the United States. Instead, you use techniques to select a representative population and then you poll that population. In a very similar way, the Restriction Spectrum Imaging allows us to identify the regions that are sufficiently different, so that we can target our biopsies to these regions and get a good sense of the overall histology.
The second technology allows me to visualize in real time where the biopsy needle is located in the cerebrum, thereby bypassing the blind nature of this procedure. Essentially, MRI-compatible equipment has been developed so that I can perform the procedure in the MRI. And in doing so, I can visualize where my biopsy needle is in real time using the MRI—throughout the procedure. And that allows me to do two things. One, I can verify precisely where the biopsy is occurring in the tumor, allowing visual confirmation of the intended biopsy. The second thing that it allows me to do, is that it allows me to react to what actually happens during the surgery, very much like a conventional surgery. If I find evidence of bleeding, I could react accordingly by terminating the procedure. On the other hand if I’m assured the repeat biopsy isn’t causing bleeding, I am reassured that it is safe to secure additional biopsies as needed.
So, the combination of these two technologies allows me to preoperatively to define the pertinent regions of the tumor for biopsy and to confirm that the actual biopsies occur in these regions.
How exactly are you studying it? What do you hope to find out?
The technology is sufficiently mature that we’re moving it in to the clinical setting. We are now doing studies to correlate the different regions of cellularity that were defined by the restriction spectrum imaging to the histology of the biopsy specimens. In other words, we wanted to determine whether the theoretical benefit of this method translates into empiric, clinical benefit for our patients. And the way we’re doing that is through correlative studies with clinical outcome. What we’re finding is something very interesting. That is, if you sample the wrong regions of the tumor during the biopsy, you could very well end up with the wrong diagnosis. We had a very interesting case where biopsies from one particular region of the tumor that has low RSI signal yielded specimens suggestive of a low-grade tumor. But yet, a few millimeters away from this area was a region with high RSI signal. Specimens secured from this region yielded the diagnosis of a high-grade tumor, which is the correct diagnosis. We’re now in the process of expanding our clinical experience, so that we can better define how best to use this technology.
What’s next in Restriction Spectrum Imaging? How will define its success?
As a physician, I define success on patient-to-patient basis. That is, if I secure an accurate diagnosis for a particular patient, that’s a success. From an investigator’s perspective, I think a success is to demonstrate that the technology elevate the standard of care for all patients. In order to do this, we will need to compare the frequency of misdiagnosis when this technology is used versus when the technology is not used. By providing this sort of empiric data, we’ll be able to assure our patients in terms of the efficiency and efficacy of the procedure that we’re doing and assure them the best possible surgical procedure was performed.