NinePoint Medical Looks Deep For Barrett’s Esophagus

NinePoint Medical Inc. is looking to shed some insights on the mysteries of the esophagus. The venture-backed start-up launched NvisionVLE, an imaging system that can produce real-time images of the esophagus, enabling gastroenterologists to see where they need to cut when performing biopsies of tests of worrisome esophageal tissue including Barrett’s esophagus.

• The simple esophagus remains one of the least understood parts of the human anatomy. This veil of secrecy can be deadly for patients with Barrett’s esophagus.
• A byproduct of gastroesophageal reflux disease, Barrett’s changes the lining of the esophagus at a cellular level. This change may, or may not, eventually lead to esophageal cancer.
• The uncertainty requires Barrett’s patients to undergo regular screenings. Every year or two, they must submit to an endoscopic procedure that examines the esophagus and takes multiple biopsies.
• NinePoint’s NvisionVLE system could provide a sharper pair of eyes for gastroenterologists, allowing them to pinpoint pre-cancerous and cancerous changes not only on the lining of the esophagus but also 3 mm deep into the tissue.
• Eventually, NinePoint’s images could be used to develop more precise methods of determining when benign Barrett’s is about to go bad.

The esophagus remains somewhat of a mystery to gastroenterologists, particularly regarding the treatment of Barrett’s esophagus, a cellular change in the lining of the esophagus caused by chronic injury due to gastroesophageal reflux disease (GERD). Barrett’s is potentially a precursor to deadly esophageal cancer. But the progression from Barrett’s to cancer can be slow or even stalled. Physicians can’t be certain which patients will become the sickest, so protocol requires regular endoscopic screening with white-light probes and many biopsies taken from the esophageal wall to determine whether the disease is progressing.

Imaging companies and researchers have worked to lift the veil a bit. A report by Charles J. Lightdale, MD, a gastroenterologist with Columbia University and [New York-Presbyterian Hospital ], notes that high-resolution white-light endoscopy and digital chromoendoscopy provide clear images of the surface of the esophageal wall. Endoscopic ultrasonography (EUS) can capture images from below the surface, but not with the resolution necessary to detect layers of dysplasia or cancer. Another imaging technology, Confocal laser endomicroscopy (CLE) can get clear pictures from beneath the surface. But the focus of the images is tight, so a developing problem in an adjacent area that hasn’t been scanned may be missed. The next logical step would be a way to scan the entire esophagus with a system capable of capturing high-resolution images of both the surface and subsurface.

NinePoint Medical Inc. is hoping to make that next step possible. In the spring, the venture-backed start-up launched NvisionVLE, an imaging system that can produce real-time images of the esophagus. It uses an advanced form of optical coherence tomography that is capable of capturing tissue images at a depth of 3 mm (or 3,000 microns) circumferentially, in 6-cm lengths at a time. Charles Carignan, MD, CEO of NinePoint, says by comparison technologies like confocal endomicroscopy can’t see beyond 250 microns. Gastroenterologists using NvisionVLE are seeing things they previously might have missed. NinePoint’s mission and objectives were set early on by Third Rock Ventures, a relatively young venture capital firm that focuses primarily on biopharmaceutical companies. The firm maintains a smaller interest in medical devices if the investment is built on a platform that could upend a standard therapy. Third Rock Ventures and co-investor Prospect Venture Partners hope they’ve found that in NinePoint. The pair invested $33 million in a Series A in 2010. [See Deal]

Crafting A Company Around A Need

Third Rock doesn’t tend to wait for the good ideas. (See (Also see "Third Rock Ventures Rolls Its Own" - Scrip, 18 Jun, 2013.).) In 2008, the firm hired Jessica Duda, a former associate at Polaris Partners who had left the venture world to go to Cyberkinetics Neurotechnology Systems Inc., where she was director of corporate development and operations. In 2008, Duda returned to venture capital, taking a project director role at Third Rock Ventures’ Convergence group. Her mission was to find an important unmet medical need and build a company around it. “This is where Third Rock has an incredibly unique model,” Duda says. “They are true company creators. They leave their offices, go out there into the hospitals and the academic institutions and say, ‘OK, where are there areas of need in the clinic and with the patients, and what technologies are at the forefront of being ready for commercialization to meet those needs?’” In biotech, Third Rock invests in early-stage research, but in devices, the technology has to be close to some type of commercialization.

Duda walked the halls of universities and chatted up physicians in hospitals, trying to identify their biggest bottlenecks. She says Third Rock wanted to identify a platform technology in an area where a privately held start-up could secure a leadership position. The start-up, it decided, should remove the many, sometimes redundant, steps that lie between symptom and treatment. In cancer, for example, the initial detection of a lump or abnormal scan is followed by a biopsy, then a staging meeting. “At that point maybe you’ll have an initial treatment plan, which may or may not work. You’ll have a follow-up, and then you have a second treatment plan, and more follow-up, Duda says. “It’s really not smart medicine.”

“Our thesis was if you could take an image and then automatically go back to that location in a repeatable and reliable fashion, then you would be able to extract cancers” without removing healthy tissue.” – Jessica Duda, NinePoint

Third Rock considered pursuing a personal diagnostics technology. Indeed, that remains an interest for the firm. But the venture company saw the best opportunity in in vivo pathology. Duda says an imaging system that could capture a high-resolution, real-time image of the body could open the door for more effective surgical treatment. This isn’t a new concept. Neurosurgeons use computed tomography (CT) scans during surgery, but the three-dimensional (3-D) image is static. “Our thesis was if you could take an image and then automatically go back to that location in a repeatable and reliable fashion, then you would be able to extract cancers” without removing healthy tissue. This is particularly important in an area such as the esophagus, where the removal of tissue can impede function.

Third Rock’s search led it first to Raman spectroscopy, an imaging modality that is based on the analysis of scattered light. Raman is used widely in the chemistry and physics fields, but it is slowly gaining traction in health care, particularly in cancer diagnosis. Duda places Raman imaging in the realm of molecular diagnostics, and the distinction between a diagnostic and an imaging modality is an important and recurring theme in the NinePoint story. She says Raman goes well beyond imaging, providing the ability to analyze tissue and determine whether cancer is present.

The FDA would likely have required a pre-market approval application for Raman, “making it a five- to 10-year development proposition. It would have taken hundreds of millions of dollars,” Duda explains, “and it doesn’t really make sense for a diagnostic to take that long in light of the type of payback that a venture firm would need.”

Dune Medical Devices Inc. proved that very point earlier this year. The FDA issued a PMA for the company’s MarginProbe, an intraoperative, tissue assessment tool for early-stage breast cancer surgery. The agency approved the device based on a 664-patient prospective, multicenter, randomized, double arm study to evaluate the effectiveness of MarginProbe in identifying cancerous tissue along the margins of removed breast tissue during initial lumpectomy procedures, seven years after Dune Medical had obtained CE mark for the imaging system.

Establishing Guidelines

When Third Rock formed NinePoint, it established some guidelines to avoid any regulatory briar patches. (See (Also see "NinePoint Medical Inc. " - Medtech Insight, 1 Feb, 2011.).) Duda says any product would have to meet two milestones: clinical trials with data within 12 months of being licensed and the company would have to put a product on the market within three years of acquiring the technology. “We had lots of tracks running in parallel,” she says. “The technology was really important. The clinical problem was really important, as were the infrastructure and the business case that we were putting together.”

NinePoint initially conceived of a handheld device – like a pen – that was capable of determining whether a certain spot of tissue was cancerous or benign, an approach that would be similar to Dune’s device. But conversations with physicians refined the idea. According to Duda, physicians wished for an imaging device that could tell them where to put the pen. “They were consistently saying, “With my naked eye, with my current endoscope, with my current high-definition white light, I still don’t see disease as well as I need to.”

Duda says NinePoint went to [Massachusetts General Hospital] (MGH) to talk about the Raman technology and its vision of in vivo pathology, or taking the microscope into the body to identify precise points that needed to be treated. The meetings at MGH led the company to Gary Tearney, MD, PhD, associate director of MGH’s Wellman Center for Photomedicine and the optical diagnostics program leader at the Center for Integration of Medicine and Innovative Technology. Duda and other NinePoint reps laid out their vision for in vivo pathology. But Tearney came to the meeting with his own ideas. “Gary stood up and he gave his technology description of OFDI (optical frequency domain imaging) and how it allows you to see areas of abnormality, it directs your pen to where it needs to go; that’s when we said, ‘Oh, there you go. There’s that convergence.’” NinePoint licensed the OFDI technology from MGH in 2010, establishing the largest such agreement in the hospital’s storied history. [See Deal]

Blind Man’s Biopsy

OFDI is a form of optical coherence tomography, OCT, which was first put to use in medicine in the mid-1990s. Lincoln Labs and James Fujimoto, PhD, a professor of electrical engineering at Massachusetts Institute of Technology, began testing the imaging technology as a way to peer more deeply into the eye and retina. The research bore a company, Advanced Ophthalmic Devices, which would eventually be purchased by Carl Zeiss. LightLab Imaging Inc., another firm centered on cardiovascular and endoscopic OCT, was spun out and eventually acquired by Goodman Co. Ltd.[See Deal] (See (Also see "A Bright Future For Optical Imaging" - Medtech Insight, 16 Dec, 2011.)). In addition to companies, Fujimoto’s MIT lab spun out entrepreneurially minded scientists and engineers including Tearney and Brett E. Bouma, PhD. The pair went to MGH and continued work on the technology, creating the application that NinePoint ultimately would license.

In the aforementioned paper, Lightdale says OCT is analogous to ultrasonography, but instead of sound waves it employs infrared light from a laser and optical scattering. The light bounces off the tissue creating two-dimensional (2-D) images that can be constructed into a 3-D image. The OCT sensor doesn’t come into contact with tissue. NinePoint’s probe sits within a balloon that centers the probe inside the esophagus. The sharper images OCT provides found footing in ophthalmology. But the GI tract has gone largely unexplored.

Bouma, who sits on NinePoint’s Technology Advisory Board, says he began working with gastroenterologist Norman S. Nishioka, MD, a gastroenterologist at MGH and an associate professor of medicine at Harvard Medical School, to develop an application for the GI tract. “We got our prototypes into the GI-Endo unit quickly and got a good sense of what we could and couldn’t see,” Bouma recalls. They both felt the technology had potential, but the image rate was too slow. “It only allowed us to place a probe at discrete locations in the esophagus and take a kind of snapshot, one picture, two pictures, that type of thing,” he explains. “We originally thought to provide a form of what we called optical biopsy, but we realized there wasn’t a compelling need for that in endoscopy. Biopsies are cheap; they’re pretty innocuous and pretty easy to perform. The biopsy is an industry standard and to replace it wouldn’t have been a winning gambit.”

Instead, Bouma says, the pair began looking at what biopsies can’t do. The typical biopsy involves a medical blind man’s bluff, with an endoscope snapping up small tissue samples of the esophageal lining hoping to find any troubling tissue that needs attention. Bouma says he and Nishioka realized that screening the entire esophagus would shed some necessary light on the biopsy process.

Around 2003, the pair developed their first prototype, changing up the light source and the receiver to create optical frequency domain imaging. The new system could manage hundreds of images per second. This enabled Bouma, Nishioka, and eventually NinePoint to use the technology to image the entire esophagus in one to two minutes. Bouma says OFDI had previously only been theorized, but was thought to require a laser with “very unique characteristics.” He says, “It turned out that there was no such laser in existence. And so we invented a new class of laser that emits a very narrow band frequency, a single wavelength of light, but that wavelength scans rapidly in time. So it’s what we call a wavelength-swept laser.” The new laser opened the door for optical frequency domain imaging.

Bouma says the idea received quick acceptance from the GI community. At the same time, he and Nishioka pursued an intravascular application for imaging the inside of coronary arteries and that pursuit gained traction more quickly. Terumo Corp. licensed the technology as a vascular imaging device. [See Deal] However, the transition was time-consuming. “It takes a huge effort on the academic side to transfer that technology, to train the engineers in industry how to make that device,” he explains.

Once the Terumo relationship was up and running, Bouma says, “we then felt like our bandwidth was open a little bit more, affording us an opportunity to work on the endoscopy application.” At some point the two talked GI applications with Pentax Corp. and Olympus Corp.’s Olympus Medical Systems Group, but Bouma says he really preferred working with a start-up, where he could have a direct hand in product development. “We were delighted when the opportunity came from NinePoint,” he says. “The relationship has been just fabulous for me, participating in a company that only launched a few years ago and now has over 60 employees and a product released.”

Solving Barrett’s

Nishioka says NinePoint’s particular focus will be Barrett’s esophagus, a particularly vexing condition of the GI tract that can be a precursor to esophageal cancer, a particularly deadly form of the disease that must be caught early. As noted, Barrett’s, a changing in the lining of the esophageal wall, can be brought on by gastroesophageal reflux disease or GERD, which hits 20% to 30% of the population. But finding that segment of the population can be challenging. Physicians can’t run an endoscope down the esophagus of every GERD patient. Following the diagnosis of Barrett’s, a patient generally undergoes an endoscopic evaluation every two to three years. Biopsies are taken, and the tissue is tested to see if the Barrett’s is changing, moving first to Barrett’s esophagus with dysplasia. This isn’t a cancer diagnosis, but it’s seen as a step toward cancer. “There is low-grade dysplasia, high-grade dysplasia, and the next step would be actual cancer,” Nishioka says. “If you start moving down that path, then obviously we need to keep a much tighter eye on you in terms of repeat endoscopy and so forth. It becomes a surveillance issue.” However, surveillance is far from perfect. It relies on an endoscopy, during which the physician takes biopsies – small pinches of tissue – which are then sent off to be analyzed under a microscope by a pathologist. Nishioka says the endoscopic cutting retrieves just a small percentage of the potential disease so testing is imprecise. “You could be misled into thinking someone doesn’t have any dysplasia or abnormality, when in fact they do; you just didn’t happen to sample it at the right spot,” he says.

Enter NinePoint. The NvisionVLE system’s OFDI imaging can scan fairly rapidly the entire Barrett’s segment. The scans might not replace biopsy, at least not initially, but could pinpoint areas where samples should be taken. Nishioka says the idea really is touse this technology as an adjunct to endoscopy to improve the efficacy of the surveillance procedure. NvisionVLE’s imaging could identify points where the tissue should be removed. “The protocol typically is to take four quadrant biopsies,” NIshioka explains. “Your esophagus is like a cylinder, so we typically take a biopsy in each of the four quadrants, roughly at 12, 3, 6, and 9 o’clock. We do that every couple of centimeters.” With NinePoint’s system, instead of randomly taking biopsies up and down the esophagus, the clinician can actually hone in on where things may be particularly suspicious and, hopefully, yield an accurate result. “It’s not designed to replace the biopsies but to make them more efficient and to reduce that sampling error.” Nishioka says preliminary data suggest OFDI could more easily identify suspicious areas, but it’s too early to pinpoint a success rate.

“It’s not designed to replace the biopsies but to make them more efficient and to reduce that sampling error.” – Norman Nishioka, MD

Quick Progress

Early reports are promising. In May, NinePoint released the findings of two clinical trials at the Digestive Disease Week 2013 conference in Orlando. The research came from clinical trials performed at the Mayo Clinic in Rochester, MN, and Jacksonville, FL, and the Kansas City Veteran’s Affairs Hospital, and in conjunction with MGH. The study at the Mayo Clinic explored the correlation between dysplasia detected by advanced OCT and histopathological analysis in patients suffering from Barrett’s. The data demonstrated a dysplasia detection rate in patients with Barrett’s esophagus of 95% using NinePoint’s NvisionVLE imaging system.

In the second study, researchers evaluated the feasibility and safety of VLE imaging with Barrett’s. No adverse effects were reported in the 74 patients in the study. The procedure took an average of 17 minutes, and every case of high-grade dysplasia was detected.

A month earlier, NinePoint had announced the commercial launch of the NvisionVLE imaging system in the US. The company had obtained 510(k) notification in December 2011 for use as an imaging tool in the evaluation of human tissue microstructure by providing 2-D, cross-sectional, real-time depth visualization. In April, the FDA expanded the use to include imaging of esophageal tissue microstructure. Carignan says the company has secured three notifications overall, including one in 2012 that allowed the company to increase the pressure inside the balloons it uses to deploy the probe.

The third and most recent approval for esophageal imaging didn’t come as easily as the others. Carignan says FDA review got tighter as NinePoint concentrated on the esophagus. The concern centers around the function of the imaging system. Is it a diagnostic, or is it used for imaging? Carignan says NinePoint successfully argued that the NvisionVLE simply provides sharper images. “The physician still has to review the images and decide what he thinks is happening on the image,” he explains. The line could blur in the future as NinePoint begins incorporating algorithms and other methods of defining tissue. Long term, NvisionVLE could be a diagnostic, but the system has a long way to go before gaining broad enough acceptance.

Carignan could point to the FDA’s approval of Cellvizio, put out by Mauna Kea Technologies, which has GI and pulmonary approvals. Carignan says Mauna Kea’s technology offers higher resolution than the NinePoint system, but it doesn’t provide the sweeping, deeper views of NvisionVLE. “Their concern is whether you are going to change the practice of medicine,” Carignan says. “I always push back and say, ‘No, we are not changing the practice of medicine. The practice of medicine may change as a result of our technology.’ That’s what happens with breakthrough technologies all the time.”

“The practice of medicine may change as a result of our technology.’ That’s what happens with breakthrough technologies all the time.” – Charles Carignan, MD, NinePoint

Starting Sales

Kurt Heine, vice president of sales for NinePoint, has been selling endoscopy tools for more than a decade. Prior to joining NinePoint, he managed sales teams at Boston Scientific Corp. Heine says the gastrointestinal field has “evolved quite a bit” in recent years, moving away from more invasive surgeries. NinePoint salespeople are pitching directly to the gastroenterologists, who are accustomed to having new technologies brought to them. “There is a lot of excitement around the desire to get imaging below the surface, because right now there’s an unmet need clinically in the esophagus and in the GI tract for different disease states,” he says. “We’re kind of in that sweet spot of that 1 to 3 millimeters where there’s a huge need to be able to see, but no current modality really gets you there.” Heine says NinePoint isn’t competing against other technologies; rather it’s trying to build a market around an unmet clinical need. NvisionVLE is selling for $150,000 per capital unit. The optical probes, disposed after each procedure, cost $1,095. NinePoint is selling primarily in the US.

NinePoint has many possible outcomes for investors. Third Rock partner Mark Levin says initially Third Rock envisioned creating a company that could both pinpoint cancer, or other conditions, and treat them by removing or ablating them. NinePoint hasn’t emphasized the second part of that formula yet, but there is potential to take the first part to new levels. One potential application would be developing a scan that could register genomic sequences of cancer, Barrett’s or various forms of dysplasia. This would enhance the ability to diagnose the stages of esophageal disease, making it easier to develop a treatment plan. “There are lots of cancer genome products but very little has been done in esophageal genomics,” Levin says. “Very little is known about why normal tissue changes to Barrett’s or esophageal cancer.”

Long term, NinePoint will be able to compile a database of esophageal images. Levin says the company believes the compilation will give gastroenterologists valuable insights on the esophagus, creating another potential line of revenue for NinePoint. The firm is exploring other methods of imaging as well, with a plan for a “pill cam” under development. The pill could be administered in a doctor’s office, saving the time and expense of an endoscopic procedure. The device would be tethered and pulled back through the esophagus, taking images on the way up and down. Early trials show that four scans of the esophagus – two down and two back up – took only six minutes to complete.

NinePoint is positioning itself within a growing area of interest for device companies. (See (Also see "Buyer Appetite Might Appeal To GI Device Investors" - Medtech Insight, 27 Dec, 2011.)and (Also see "Covidien Still Hungry For GI After Barrx Buy" - In Vivo, 27 Dec, 2011.).) The most notable acquisition in this space remains Covidien Ltd.’s purchase of Barrx Medical Inc. in 2011. [See Deal] Covidien paid out $325 million in cash but with earn-outs the purchase price could hit $391 million. Barrx had positioned its HALO ablation system as an alternative for current methods of treating and monitoring the various forms of Barrett’s. (See (Also see "Barrx Medical: Making A Case For Ablation" - In Vivo, 1 Oct, 2010.).) Barrx is making a case for ablating tissue in people who aren’t currently treated. Of the six million people worldwide suffering from Barrett's, only 11% suffer from dysplasia and only 1% register with high-grade dysplasia, which requires immediate medical intervention, according to Barrx.

The two companies would seem to run in parallel. Indeed, Covidien should be considered a potential acquirer. But in some ways they’re offering countering options. Barrx is suggesting that all Barrett’s sufferers undergo ablation to eliminate the risk of cancer, which is small but on par with the risk of a polyp in the colon developing into cancer. Yet, polyps are removed quickly upon detection. NinePoint’s NvisionVLE could provide a third option for gastroenterologists. Rather than ablating all patients with dysplasia, doctors could perform a quick, complete scan looking for any sign of trouble. In the end, NinePoint could help be a light in the tunnel of the esophagus.