TumorGlow® is an investigational technique that has not been approved by the Food and Drug Administration for any indication. The information in the following article, including statements from individuals herein, should not be interpreted as an endorsement or recommendation by Penn Medicine or the Perelman School of Medicine.
The history of present innovation is driven by past innovation. And such was the case when Sunil Singhal, MD, began searching more than a decade ago for a better way to visualize rogue cancer cells.
Making Malignancies Visible
A thoracic surgeon at Penn Medicine, Dr. Singhal was no longer able to accept an ageless tenet of cancer surgery: that surgeons could only remove cancers they could see or feel. A limitation that made eradicating a patient’s cancer more a matter of luck than science.
“We were performing surgery in the dark,” Dr. Singhal says today. “Using the same techniques--palpation and visual recognition--that surgeons introduced a century ago.”
What Is Intraoperative Molecular Imaging?
As Director of the Thoracic Surgery Research Laboratory at the Perelman School of Medicine, Dr. Singhal knew that 40% of patients who left the operating room “cured” of cancer would go on to develop a recurrence. And that for a third of these patients, the cancers were recurring within 2 cm of the original site of resection.
But no one knew how to make these occult malignancies visible.
For Dr. Singhal, the answer would arrive from his familiarity with two earlier scientific discoveries:
- Tumor angiogenesis is the process by which tumors create feeder blood vessels. This has been known since the groundbreaking work of clinician researcher Dr. Judah Folkman in the 1970s. By the 1990s, investigators knew that tumor vessels were permeable, a result of their rapid, aberrant growth, and proposed that immunostaining leaking from tumors could provide a marker in tumor imaging.
- Although the technology to image tumors in real time was not available until the early 2000s, a component of the technology needed to see cancer cells has been in use since the 1950s. An intravenous protein-bound dye, indocyanine green (or ICG) has the capacity to glow brightly under infrared light, and employs an “optical window” to permit excellent penetration of tissue beneath the visible surface.
By combining the two discoveries, Dr. Singhal realized that he might be able to take advantage of the flawed structure of tumor vessels, since a systemic infusion of ICG drawn into tumors would leak out into surrounding tissue, defining their presence under infrared light.
In the decade since, he has continued to explore this hypothesis in clinical studies involving more than 1,000 patients at Penn Medicine. This investigational series may now constitute the largest clinical experience with intraoperative molecular imaging in the world.
Intraoperative molecular imaging is one of several names used to describe the technique Dr. Singhal developed (others include fluorescence-guided surgery, and second window ICG).
The combination of ICG dye and imaging technology involved in the technique are now known as TumorGlow®.
TumorGlow and the FDA
TumorGlow is considered a custom medical device by the Food and Drug Administration (FDA) because it is composed of tangible components (ICG and an imaging instrument), among other considerations.
To date, TumorGlow is in FDA feasibility and safety trials for use in lung cancer and ovarian cancer, and has been independently evaluated in pancreatic cancer, head and neck malignancies and mediastinal cancers.
TumorGlow in Lung Cancer Surgery
The motivation for using TumorGlow in lung cancer surgery is to reduce the incidence of recurrence.
Lung cancers can recur even when the algorithm for surgery involves meticulous resection at the tumor margins with intraoperative frozen sectioning and excision of involved lymph nodes and visible satellite lesions with the principal lesion.
As with the findings from later Phase II studies, those from the pilot studies of TumorGlow were promising.
“Our initial experience showed that TumorGlow was successful for locating ground glass opacities as well as adenocarcinomas,” Dr Singhal says, noting that the former are a potential precursor for malignant lung adenocarcinoma. In addition, in these studies, surgeons were able to accurately identify indeterminate pulmonary nodules and to locate nodules <1 cm that might otherwise have escaped CAT or PET scan identification.
Second Window ICG: TumorGlow in Neurosurgery
Soon after Dr. Singhal started investigating TumorGlow in thoracic surgery, Penn neurosurgeon John Y. K. Lee, MD, began exploring TumorGlow in clinical studies to locate and identify the margins of brain tumors before and during brain surgery. Over the next five years, he developed a technique he called “second window ICG,” to distinguish his technique from traditional ICG.
Second window ICG involves infusing ICG at much higher doses (2.5 to 5 mg∕kg) than that used in vascular angiography (0.35 mg∕kg), then allowing the initial optical window to pass over 24 hours while the dye is eliminated from the body.
At this point, the dye collected in the tumor permits a second window of opportunity for observation. Dr. Lee is an enthusiastic advocate for TumorGlow to delineate brain tumors.
“What's remarkable about near infrared light is that as a longer wavelength, I can see through things that block visible light,” Dr. Lee says of TumorGlow. “Just as radio waves—which have a very long length—can go through walls, near infrared can go through tissue. So even before I open the dura, I'm seeing the tumor, and know the direct route to reach it, which minimizes damage to the normal brain during surgery.”
Dr. Lee has now treated more than 300 patients with TumorGlow.
TumorGlow Studies in Head-and-Neck and Pancreatic Cancers
Recently, Drs. Lee and Singhal have been joined in clinical trials of TumorGlow imaging by clinicians studying the technique in head-and-neck and pancreatic cancers.
Jason Newman, MD, has investigated the use of ICG in squamous cell and salivary gland carcinomas of the head and neck, and recently completed a safety and feasibility pilot study using TumorGlow with transoral robotic surgery (TORS).
“TumorGlow has the potential to more clearly identify the location and margins of tumors in the limited space in which head and neck cancers appear,” Dr. Newman said recently. “The hope is that we can say with greater confidence that we both got around the tumor and left as much normal tissue as possible,” he continued.
Future studies of TumorGlow will begin to explore possible endpoints for the sensitivity and specificity of lymph node identification at surgery. Dr. Newman is currently studying other intraoperative molecular imaging techniques in head and neck cancers, as well.
Major Kenneth Lee, IV, MD, has investigated intraoperative TumorGlow imaging in the identification of neoplasms and margin assessment during surgery for pancreatic cancer, procedures traditionally hindered by high rates of local and distant recurrence from positive margins and unrecognized metastases.
“We hypothesized that intraoperative TumorGlow could serve as a useful adjunct in assessing margins and extent of disease during pancreatic resection,” Dr. Lee said of a recently completed feasibility study that he conducted with Dr. Singhal and colleagues from Penn Surgery using the second window ICG technique.
The trial had several important outcomes, according to Dr. Lee. First, the technique identified pancreatic tumors in the laparoscopic setting. Second, TumorGlow helped identify negative versus positive margins after a resection to predict whether certain masses were benign or malignant. The final finding was particularly important.
“For patients that get treated prior to surgery, we know that imaging can be somewhat inaccurate, because what we may think is tumor tissue is just radiation change or dead tumor,” Dr. Lee explains. “And when you can’t distinguish between live and dead tumors, it’s difficult to figure out who’s operable.”
In the trial, TumorGlow showed some promise in differentiating patients who had a dramatic response to prior therapy from those that had little or no response—however, Dr. Lee points out, at this point too few patients have been treated to prove this concept.
“We’re making progress,” he says. “We hope in the future to prove it’s useful in a wider population of patients with pancreatic cancer.”
Other Fluorescent Agents Under Investigation: OTL38 Studies in Ovarian Cancer
TumorGlow isn't the only intraoperative molecular agent being investigated at Penn Medicine. Janos Tanyi, MD, PhD, from the Perelman School of Medicine at the University of Pennsylvania, is conducting studies with OTL38, a contrast agent unrelated to ICG, in women with ovarian cancer.
OTL38 contains a ligand that targets folate receptor α (FRα), a folate-binding protein over-expressed in more than 70% of ovarian cancers. In a Phase II multicenter interventional trial for which Dr. Tanyi was the lead investigator at Penn Medicine, OTL38 discovered at least one additional pathology-confirmed lesion missed in the operating room in 48% of patients who received the agent.
Dr. Tanyi is now the lead investigator at Penn Medicine for a Phase III multicenter randomized trial of OTL38 in combination with fluorescent light to detect Folate Receptor-positive (FR+) ovarian cancer lesions not detected by palpation and visualization under normal light. Participants must have FR+ ovarian cancer and be scheduled to undergo primary surgical cytoreduction, interval debulking, or recurrent ovarian cancer surgery.
“Having the capacity to positively identify and remove cancerous tissue can significantly increase overall survival in ovarian cancer,” Dr. Tanyi said in a recent interview. “We look forward to further evaluating the potential of OTL38 in this study.”
The Future is Bright for TumorGlow
According to Dr. Singhal, the future of ICG-based intraoperative molecular imaging is bright at Penn Medicine.
“I think that the first generation of dye – the TumorGlow dye – that we developed at Penn Medicine has enormous potential” he says. “But in the future, we’re likely to see the development of new dyes with ever increasing capacity to see malignant tissue in an increasing number of cancers. The field is going to be dramatically evolving over the next decade.”