We spoke with Dr. G. Larry Maxwell, Chair of the Department of Obstetrics and Gynecology at Inova Fairfax Hospital, about a potential new biomarker to help distinguish metastatic versus nonmetastatic disease in endometrial cancer.
As part of our coverage of the 2017 Society of Gynecologic Oncology annual meeting, being held March 12th to March 15th in National Harbor, Maryland, we are speaking with Dr. G. Larry Maxwell, MD, a gynecologic oncologist, the chairman of obstetrics and gynecology at the Inova Fairfax Hospital in Virginia. Dr. Maxwell gave a talk at the meeting on a potential new biomarker that could be used to distinguish between endometrial cancer with or without metastatic disease (abstract 18).
-Interviewed by Anna Azvolinsky, PhD
OncoTherapy Network: First, how common is it for patients to present with advanced endometrial cancer, so metastatic disease? And how is metastatic disease in these patients generally detected?
Dr. Maxwell: About 25% of the 60,000 endometrial cancer cases diagnosed in the US each year are going to be in patients who have advanced-stage disease where the cancer has spread beyond the confines of the uterus. So, the methods that we use for determining stage are removal of lymph nodes and biopsies obtained at the time of surgery. And the lymph nodes that are removed are often difficult to distinguish the total number and they will be an aggregate of nodal tissue and fat that is removed at the time of surgery, and then the pathologist will ultimately determine the number of nodes in that aggregate of tissue. But that determination can vary according to the pathologist and literally they are taking the mass of tissue and applying pressure on the surface of their bench top to find areas that feel a little bit more firm than others, and that is literally how sophisticated it is in terms of identifying the number of nodes and then in turn, taking those nodes and looking at them under a microscope.
So, when you have different pathologists looking at one side and a different one looking at the other, the same aggregates of tissue can lead to different node counts. So, the system we have for identifying metastasis seems a bit antiquated when you consider all of the advances we have in molecular profiling and testing these days. So, that has been an opportunity where we could use the primary tumor biopsy to try and discern who has the highest likelihood of having metastatic disease because that in turn helps focus the referral to the subspecialist who may need to do more extensive surgery than what perhaps what a general ob/gyn could provide with just a hysterectomy.
OncoTherapy Network: Okay. So, you and your colleagues evaluated a potential noninvasive diagnostic biomarker to detect metastatic disease, so a very different approach from the standard one you just described that is currently used. Can you tell us about what this diagnostic biomarker panel is?
Dr. Maxwell: So, what we are trying to discern is whether there is a particular signature that could be evaluated in the tumor from inside the uterus that could predict the likelihood that the cancer has spread into the lymph nodes or other areas of the abdominal cavity. So, we had used some of the extensive genetic data that is out, publically available from The Cancer Genome Atlas [TCGA] project. This was a large $400 or $500 million dollar genomics study done by the NIH several years ago. And what we had identified using that big dataset is that there appears to be different genes that were turned on or off in tumors that ultimately had metastasis at the time of surgery.
So, what is really important in identifying these kinds of actionable biomarker panels is that you have to have independent validation in other sample sets and you also need to have some analytical validity so that you are using different types of analytical approaches and platforms to ultimately come up with the same biomarker. So, that is what has been lacking in reports to date and even in our own research. So, in this study we took that original TCGA data, we took samples that we analyzed in our lab, we took samples that we had acquired from the gynecologic oncology group at NCI and we built models that on discovery, in turn, we took forward and validated in independent sample sets and validated just using different types of analytical approaches.
So, the biomarker panel that we identified is a 7-gene panel that ultimately provides improved predictive accuracy over what we currently use to make those predictions which is looking at those tumors under the microscope and characterizing what they look like architecturally. So, it's exciting to think that we may have molecular panels that may improve if not, in time, potentially replace some of the conventional methods we use in terms of looking at the appearance of tumors under the microscope.
OncoTherapy Network: What are some of the genes that are part of the 7-gene panel you identified?
Dr. Maxwell: There are several genes that have been previously described in conjunction with other cancers-these are new in terms of being described in the context of endometrial cancer. In terms of their exact biological function, that is something that we'll need to investigate as part of the ongoing efforts that we have to really understand, not only the clinical utility of this as a prediction tool, but also to acknowledge whether or not it could be used for development of targeted therapeutic strategies.
The panel not only predicts metastasis, but the genes when turned on or off in extreme directions can ultimately be associated with poorer prognosis or outcomes in cancer patients. So, we think that this provides yet another layer of justification for why we have identified something that may truly be quite relevant in the context of clinical management of endometrial cancer.
OncoTherapy Network: You mentioned that you used multiple data sets to identify and partly validate this 7-gene marker panel. So, what’s next as far as trying to develop and/or validate this biomarker for endometrial cancer?
Dr. Maxwell: The next steps will be identifying this in still yet more independent datasets and particularly determining whether or not we can identify these in preoperative biopsies. I think that ultimately we have developed a biomarker panel in terms of making predictions off of hysterectomy specimens, but the next step will be to confirm that this can be reproduced in biopsies and that it can be potentially done using paraffin blocks that are typically archived after someone’s surgery.
So, when someone has a hysterectomy or a cancer operation, the tissue is actually fixed in formalin and then embedded in paraffin blocks. And then these wax blocks containing the tissue can be stored 10 to 20 years or indefinitely such that if there is every a need to pull that tissue back to evaluate it, that is an option patients have in working with their doctors. This also makes it a lot easier in terms of having clinical utility if we can just be able to test the paraffin block that is in storage as opposed to having fresh-frozen tissue for some of these assays, and that makes it a little more difficult and maybe somewhat less than ideal in terms of an economically viable option for doing these kind of testing.
OncoTherapy Network: Thank you so much for joining us today, Dr. Maxwell.
Dr. Maxwell: Thank you.