By Deborah Borfitz
July 24, 2025 | A surge in research and development around fluorophores, various types of glowing molecules used to enhance visualization during surgical procedures, is driving excitement about their potential to improve the performance of human surgeons and “pave the way” for even better autonomous robotic surgery. One of the enthusiasts, particularly when it comes to finding the margins of soft-tissue sarcomas, is Eric Henderson, M.D., an orthopedic surgeon at Dartmouth Health and a sarcoma specialist at Dartmouth Cancer Center.
Literally dozens of these fluorescent probes are currently being investigated in clinical trials, including compounds that are targeted, non-targeted, and enzyme-activated, he says. In the targeted category are antibody-based probes and small molecule-based probes, both of which share the goal of attaching to a marker on a tumor but have markedly different pharmacokinetic profiles.
Henderson himself has multiple fluorescence-guided projects underway, notably one using a synthetic affibody peptide known as ABY-029 to improve soft-tissue sarcoma surgery. In a phase 0/1 clinical trial with 12 patients, the fluorophore recently demonstrated a high correlation with the expression of the epidermal growth factor receptor (EGFR) and contrast values were encouraging enough for translation to clinical practice (Molecular Cancer Therapeutics, DOI: 10.1158/1535-7163.MCT-24-0378).
Two other trials employing ABY-029 were run in parallel by his colleagues for glioblastoma and head and neck cancer that also tend to overexpress EGFR, says Henderson, a recognized leader in the field of fluorescence image guidance. He also has separate studies underway using a small nerve-specific molecule, developed by a Dartmouth graduate, for diagnosing necrotizing soft-tissue infections and quantitative determination of human cartilage health during routine arthroscopy.
In a rodent model, he and his team administered a pair of fluorophores with different mechanisms but the same wavelength to show that combining the probes provides greater brightness than any one fluorophore alone. This and other work led to a $31 million Moonshot award from the Advanced Research Projects Agency for Health (ARPA-H) last year to test a two-fluorophore combination in a human study.
In the Moonshot project, researchers will excite fluorophores using multiple wavelengths of light and simultaneously assess tissue optical properties, allowing them to reconcile the presence and depth of cancers, blood vessels, and nerves from the tissue’s surface, and simultaneously enhance different types of tissues, Henderson says. But his longer-term hope is to see the advent of autonomous robotic surgery.
To that end, Henderson participated in a day-long session with ARPA-H last summer to discuss how to get to that future state. Naysayers of fluorescence-guided surgery should note that the technology is “enabling machines to identify tissue,” a key enabler of autonomous robotic systems that can operate partially if not completely independent of a surgeon's direct control. “That will be game-changing because it will allow us to deploy the best surgical care to any place in the world.”
In the meantime, Henderson and his team plan to forge ahead with a “true phase 1 study” of ABY-029 in soft-tissue sarcoma surgery patients while also exploring other potential fluorescent probes. The goal is to identify the most promising molecular markers for tumor labeling purposes.
Henderson, who has 17 medical patents to his name and more on the way, says he has always been interested in trying to “push the edge of technology to make surgery better.” His work in fluorescence began shortly after he arrived at Dartmouth in 2013 and gave a presentation at a research meeting of its Center for Surgical Innovation.
He spoke about problems related to surgical navigation and assessing tumor margins and the promising possibilities with tumor-labeling fluorophores in the decades to come. Brian Pogue, a legendary researcher at Dartmouth who happened to be in the audience, told Henderson that a lot of that future state could be realized now, and the two began a collaboration that continues to this day.
Their research has been a true multi-center effort, Henderson says. Affibody Medical AB (Sweden) produced the ABY-029 targeted molecule and Bachem AG (Switzerland) the overall molecule. The finish-and-fill work was done at the University of Alabama at Birmingham, and preclinical toxicology testing and clinical testing were both conducted at Dartmouth.
Fluorescence-guided surgery is seeing growing adoption across various surgical disciplines but is much more widespread in neurosurgery, particularly for glioblastoma, says Henderson. This is the model disease on which the effectiveness of fluorescent guidance was first demonstrated in Germany back in the late 1990s.
The latest trio of early-phase trials with ABY-029 started patients on a microdose, as defined by the Food and Drug Administration (FDA), with escalation to three and then six times that dose. In the sarcoma study, the average “biological variance ratio”—a measure of the fluorescence of the tumor versus background tissue (the probe’s performance)—was found to be on par with that of antibody agents, but with a substantially reduced imaging-to-resection time, he reports.
Synthetic affibody peptides like ABY-029 are fragments of the binding region of an antibody. Like all small molecule probes, they act much faster than the antibody-based variety, says Henderson. ABY-029 is generally injected into patients on the day of surgery since fluorescence peaks between four and eight hours after administration.
Antibody-based probes, on the other hand, bind “very robustly” to tumors, he continues. Two EGFR-targeted monoclonal antibodies, cetuximab and panitumumab, have been successfully used by his friend and colleague Eben Rosenthal, M.D., at Vanderbilt University Medical Center.
Far more fluorescent probes are being used in a research context than clinical practice, Henderson notes. The FDA has to date approved only six such agents, one (fluorescein) of which has been used clinically since the late 19th century to identify corneal abrasions in ophthalmology. Another fluorophore (indocyanine green) got the agency’s nod in the late 1950s, followed by a long wait for the next four (methylene blue, pafolacianine, pegulicianine, 5-aminolevulinic acid), each targeting a different cancer or group of cancers.
Sarcomas are one of the five fundamental types of cancer, derived from the middle germ layer of developing embryos (mesoderm), and are rare, says Henderson. They account for less than 1% of all cancers, which equates to about 18,000 new cases in the U.S. each year. Soft-tissue sarcomas outnumber those affecting the bone eight to one, meaning between 15,000 and 16,000 cases annually in the country.
“Some sarcomas are very well behaved and have very clean margins,” he says. “If you take it out, it’s cured.” But other types of sarcomas are bad actors—myxofibrosarcoma sarcoma being the misbehaving poster child—and “have a knack for sending microscopic tendrils into the surrounding soft tissues.”
Even if a curative resection looks likely under the microscope or based on a visual assessment by the surgeon, there may be residual disease left behind, he explains. “One of the goals for fluorescence-guided surgery, particularly for sarcoma, is to be able to detect that microscopic positivity and remove it at the time of surgery”—not discover positive margins weeks later or when the patient has local recurrence two years later.
The current standard of practice is for surgeons to “grossly visually” identify areas with close or positive margins and send specimens of those areas to the pathology team to assess via frozen section microscopy. “The process takes approximately 25 to 35 minutes per section, so it can be time-consuming,” says Henderson. “It is also expensive, and it is worth noting that during that time the patient remains under anesthesia.”
Fluorescence-guided surgery requires a probe that is delivered into the body, as well as an imager, which complicates the regulatory picture, says Henderson, who was a reviewer on draft guidance on developing drugs for optical imaging issued by the FDA earlier this year. The document is to provide recommendations for designing clinical trials for intraoperative aids that involve the combined use of imaging drugs and devices.
While contrast agents are medicines, they are “given at the lowest possible dose that achieves the effective contrast to highlight the structure of interest,” Henderson says. “No one, including investigators, want to see a therapeutic effect; that is not the goal.”
“The FDA acknowledges that it has to approach these differently,” he continues. Each probe has its own fluorescence peak, which is a crucial factor in determining its suitability for certain applications, and must be compatible with a fluorescence imager—hardware that can sometimes be quite bulky.
Fluorophores are molecules that are stimulated by a wavelength of light and are brightest at their excitation peak. The wavelength at which fluorophores emit light as they return to their ground state, or emission peak, necessarily has less energy, he explains. Ideally, the difference between the excitation and emission peak (Stokes shift) is broad “because you don’t want to be stimulating in the same frequency you’re getting back.”
Imagers and fluorophores are for the most part being developed independently, he notes. Having a pharma company make the hardware as well as the software might not be a “recipe for success,” given the required technical complexity and the fact they may not inherently possess familiarity with medical device regulations.