By KEVIN NEW and KIM COGHILL
BBI Washington Writers
GAITHERSBURG, Maryland Medical product manufacturers had a rewarding month before FDA advisory panels in February, going three-for-four in earning recommendations for approval. In the fourth panel session, no vote was taken despite general agreement that efficacy was demonstrated in clinical trials and that the product has potential.
Firms receiving backing from agency panels meeting during the month included:
R2 Technology (Sunnyvale, California), for its ImageChecker CT-LN 1000 computer-aided detection (CAD) system designed to identify smaller solid nodules in the lung during computed tomography (CT) scans.
Ophtec USA (Boca Raton, Florida), for its Artisan/Verisyse phakic intra-ocular lens (IOL) for correcting fairly extreme myopia.
Refractec (Irvine, California), for its Viewpoint CK system used in a conductive keratoplasty procedure for the improvement of near vision in those who require only reading glasses and those who require reading and distance glasses.
In a late-February session, members of the Neurological Devices Panel said that, given the absence of randomized control data and a bothersome mortality rate, they would have to review additional information before they would recommend approval of Concentric Medical's (Mountain View, California) MERCI Retriever to ensnare blood clots and restore blood flow to ischemic stroke patients.
While the Neurological Devices Panel did not take a formal vote on the MERCI Retriever, most members agreed that the company's clinical trial demonstrated efficacy on the revascularization endpoint and thus has potential. However, members also said they were troubled by the 38% death rate in the firm's single-arm, 114-patient trial. The panel's comments will be forwarded to the FDA, which makes the final decision on marketing approval.
Concentric is seeking expanded use of its Retriever to restore blood flow in the neurovasculature by removing thrombus in patients experiencing an ischemic stroke. The Concentric Retriever essentially identical to the MERCI Retriever currently is marketed in the U.S. as a tool for retrieving foreign bodies from the peripheral and neurovasculature. According to data provided by the company, ischemic stroke is the third-leading cause of death in the U.S., but the leading cause of adult disability. An ischemic stroke occurs when a blood clot or piece of plaque is lodged in a vessel in the brain, preventing blood from reaching and providing nourishment to part of the brain and it begins to die.
Clearance under the proposed indication is believed to be a plus for stroke patients because clinical trials demonstrate that in certain patients the device can reverse damage caused by a stroke. The retriever extends the window of treatment to eight hours, compared to the three-hour window afforded by standard drug therapy used to dissolve blood clots. Indeed, Andrew Ku, a panel member and radiologist with Allegheny Radiology Associates (Pittsburgh, Pennsylvania), said that drugs often do not dissolve blood clots. Therefore, having access to an alternative device would probably save lives. "This device may offer patients a difficult choice," Ku said. "If it works well, you end up doing better; if it doesn't work, you end up doing worse."
Commenting that the Retriever clearly has potential, panelist Colin Derdeyn, a radiologist at the Washington University School of Medicine (St. Louis, Missouri), said he would probably use it off-label.
Like the Concentric Retriever, the MERCI Retriever is made of a nitinol wire with a helical-shaped tip. The helix is covered with a platinum coil for radioplacity, and the nitinol core wire is coated with a hydrophilic coating, ending proximal to the helical-shaped tip. The tip of the retriever has five loops, which are designed to capture thrombus from the neurovasculature and foreign bodies from the neurovasculature, peripheral vascular and coronary vascular systems. The device is inserted into an artery in the groin, then guided up the body into the brain. For both foreign body retrieval and thrombus removal, a microcatheter is positioned beyond the obstruction and the retriever is deployed. The microcatheter and retriever are pulled back to engage the foreign body or thrombus. All are then withdrawn from the body.
As part of its application, privately held Concentric submitted data from its single-arm Mechanical Embolus Removal in Cerebral Ischemia I/II (MERCI) clinical trial and its PROACT II trial, which included a control arm. In the MERCI trial, the Retriever restored blood flow in 61 of 114 patients (54%). Patients who were not eligible for standard therapy were studied up to eight hours after initial stroke. The primary endpoint was revascularization in all of the major cerebral vessels immediately following retriever treatment while minimizing the occurrence of serious device-related adverse event. Serious device-related adverse events were target vessel perforation, target vessel intramural dissection or significant embolization in previously uninvolved arterial territory.
While panel members agreed that MERCI hit its primary endpoint, several questioned whether revascularization was an appropriate selection. "I'm concerned that this could go either way," said John Marler, a panel member and neurologist with the National Institute for Neurological Disorders and Stroke (Bethesda, Maryland). "This is very difficult because we don't have any way of knowing whether it will help or hurt patients because of the lack of data." An FDA spokesman said the MERCI trial probably didn't include a control group because it is difficult to randomize patients in such indications as stroke. However, he said that the PROACT trial did in fact include a control group, which was compared with MERCI.
In the MERCI trial, roughly 13% of participants experienced some sort of serious adverse events, including nine patients with symptomatic intracranial hemorrhage; six patients with vessel dissection or perforation; and two patients with embolization of clot into a previously uninvolved area. The mortality rate in the MERCI trial was 38% compared to 25% in the PROACT II treatment arm and 27% in the PROACT II placebo arm. Of the 114 patients treated in the MERCI trial, at 90-day follow-up 43 had died. Of the 61 patients who had successful revascularization, 15 patients had died (25%) and of the 53 patients who did not have successful revascularization, 28 had died (53%).
CT-LN seen helping earlier detection
R2's ImageChecker CT-LN 1000 device will assist radiologists in detecting lung nodules at an earlier stage, thereby giving oncologists the opportunity to treat lung cancer much earlier, the company said in its presentation. Lung cancer remains the most deadly form of cancer, accounting for 28% of all cancer deaths annually, according to the American Cancer Society (ACS; Atlanta, Georgia). Last year, the disease claimed 157,200 lives in the U.S., according to ACS data.
The ImageChecker CT-LN 1000 CAD system was recommended for approval by a unanimous vote of the Radiological Devices Advisory Panel of the Center for Radiological Health. The system works in conjunction with CT scanning equipment already in use in radiology departments, said Ron Castellino, MD, chief medical officer for R2 Technology. "Radiologists in the clinical setting increasingly are seeing instances of smaller solid lung nodules, and that's the rationale behind offering this system as a complement to CT systems," he said. Typically, solid lung nodules require a biopsy, which isn't feasible for lung nodules as small as four millimeters, he noted.
The panel asked the manufacturer to re-analyze its data to ensure certain patients are ideal for the adjunctive screening tool. It also recommended a post-approval surveillance study to be coordinated with the FDA.
R2's ImageChecker CAD system was the first such system approved by the FDA, in 1998, for use with film screening and diagnostic mammograms and digital mammography, and Castellino said the system for lung scanning works in much the same way. The CT-LN 1000 system identifies the nodules and automatically takes the radiologist to the CT image slice where the nodule is located, he said. The 3-D representation also shows the nodule's location in relation to arteries and veins within the lung, Castellino added.
In its clinical study, R2 Technology retrospectively reviewed CT scans from five clinical settings and used an expert panel of radiologists to determine if a nodule was "actionable" meaning required monitoring by a physician. The panel agreed that, because tissue biopsies are not feasible in these cases, an expert panel of three radiologists is appropriate for determining actionable nodules.
"Our hospital used the system for three months during a beta test, and it was extremely helpful in catching missed small nodules," said Pablo Delgado, MD, program director of radiology at the University of Missouri-Kansas City School of Medicine and St. Luke's Hospital (Kansas City, Missouri). "There was a rapid learning curve, and we had no malfunctions. I would recommend that any radiologist first conduct a review without CAD and then review with CAD, however," he said.
While the vote of approval was unanimous, some reservations about the system were raised. David Stark, MD, past chairman of the department of radiology at Downstate Medical Center (Brooklyn, New York) and lead reviewer for the panel, said, "This is more complicated than putting a rover on Mars. I love technology just as much as everyone with a PDA, but there are unintended consequences here if we move too quick, or conversely, too slow." Stark also expressed concern over the false-positive rates that could increase if the CAD system is used. "Some radiologists will be induced to call a positive and could cause a change in treatment, such as chemotherapy vs. biopsy," he said. "This technology forces radiologists to work for it, rather than the other way around."
The ImageChecker was launched for commercial sale in the European Union at the European Congress of Radiology in March 2003. In January, R2 received 510(k) clearance for temporal comparison tools on the Imagechecker CT workstation, which provide the ability to track nodule growth progression or regression over time.
The company said that if the ImageChecker CT CAD system wins FDA clearance, it will be the first such system for the detection of lung nodules on CT exams commercialized in the U.S.
Split decision on phakic IOL
The Ophthalmic Devices Advisory Panel, meeting in early February, was split over recommending approval for the Artisan/Verisyse phakic IOL. Panelists voted 6-6 to recommend agency approval of the company's premarket approval (PMA) application, with panel chair Jayne Weiss, MD, of Wayne State University's Kresge Eye Institute (Detroit, Michigan), breaking the tie with her vote to approve.
The IOL is manufactured by Ophtec, a subsidiary of Ophtec BV (Groningen, Netherlands), with exclusive distribution to be handled in North America and Japan by Advanced Medical Optics (Santa Ana, California). AMO is currently one of the distributors of the lens under the name Verisyse in Europe, South America and portions of the Asia-Pacific region.
The lens does not substitute for the natural lens of the eye but is implanted on top of the natural lens. The technology developed by Ophtec enables the lens to be attached to the non-reactive, mid-periphery of the iris in a way that does not inhibit the iris' function, according to the company. The lens is implanted in the anterior chamber between the iris and corneal endothelium and is designed to reduce the risk of endothelial cell loss and cataract formation associated with crystalline lens trauma. The technology was first developed in the late 1970s for the correction of aphakia the removal of the natural lens following cataract surgery. Ophtec reports that more than 400,000 aphakic Ophtec lenses have been implanted worldwide, and an additional 100,000 phakic IOLs have been implanted.
The application sought by the company was for reduction or elimination of myopia in adults with myopia ranging from greater than -5 diopters to -20 diopters, with less than two diopters of astigmatism at the spectacle plane, but the panel recommended limiting use of the device to patients with myopia ranging from -9 diopters to -20 diopters. It also recommended other conditions, foremost the limiting of use to older patients and others with the most severe nearsightedness.
While the device is designed to provide better vision for those who cannot benefit from LASIK surgery or who must wear extremely thick glasses which provide only a limited visual field, implantation of the device comes with a variety of risks: potential corneal damage, cataracts and the increase of natural cell loss, concerns that apparently led to the recommendation of use only for older patients.
The Artisan/Verisyse is offered in 5 mm and 6 mm versions to accommodate different pupil sizes, and should theoretically diminish halos and starbursts in patients with large pupils, the panel agreed. But Arthur Bradley, professor of optometry and visual sciences and adjunct professor of neural and cognitive science at Indiana University (Bloomington, Indiana), argued that "the data supplied by the sponsor does not support that assumption."
"I believe the device is not best-suited for the intended patients," said William Mathers, MD, lead reviewer for the panel and professor of ophthalmology at Oregon Health Sciences University's Casey Eye Institute (Portland, Oregon). "Those between 21 and 50 have different needs than those over 50. I would not want to implant this IOL in patients younger than 35."
Refractive IOL products such as the Artisan/ Verisyse "fill an unmet market need in the U.S.," said Doyle Stulting, MD, PhD, of the department of ophthalmology at the Emory University Eye Center (Atlanta, Georgia) and director of the university's Cornea Service. Stulting referred to a "successful 13-year track record" of the lens's use in Europe and the company's study results, saying that the Artisan/ Verisyse phakic IOL "provides excellent refractive outcomes -95% of study participants achieved manifest refraction spherical equivalents that were within 1 diopter of target."
Rick McCarley, president and chief executive officer of Ophtec USA, said that, assuming approval of the Artisan/Verisyse, it could be "one of the first phakic IOLs offered commercially in the U.S." And he said that a three-year study demonstrated that the lens "is a safe and effective vision correction alternative." The company said it is expecting a decision by the FDA in about six months.
System for CK procedure gets thumbs-up
The Ophthalmic Devices panel voted unanimously to recommend approval of Refractec's Viewpoint CK system for the conductive keratoplasty (CK) procedure for the improvement of near vision in patients suffering from presbyopia. The CK system is the first product ever reviewed by the FDA specifically for presbyopic patients who want to improve their near vision. The FDA first approved the system in April 2002 for those over 40 with farsightedness (hyperopia).
CK is a non-laser, three-minute procedure that requires no cutting and no removal of tissue, can be carried out in the doctor's office and requires only eye-drop anesthesia. The procedure is performed with a probe the size of a human hair that emits radio waves to shrink small amounts of collagen in the cornea. This creates a steeper shape to the cornea and, in turn, modifies a person's point of focus. The procedure works to improve patients' near vision without sacrificing binocular distance vision, according to Refractec.
The vote, while unanimous, came with what panel chair Weiss termed a "laundry list of labeling conditions." Weiss said, "What we are suggesting to the agency is that the sponsor makes it clear in the language that there is no eraser on the edge of the probe with this procedure, and that we can't guarantee where someone will end up once the wheel has been spun." CK can improve vision temporarily but may not completely rid a patient's need for reading glasses, and the panel felt that distinction was not made as clearly as it should be in the labeling.
"To treat patients with near or moderate presbyopia with CK appears to be safe, but the treatment has not been shown to be effective for less than one diopter," James Saviola, OD, chief of the Vitreoretinal & Extraocular Devices Branch of CDRH, told members of the panel. "Putting a tighter restriction on the indication would eliminate many patients from the applicable pool of candidates, however."
Safety and data support approval of the device for all indications, because the procedure is already commercialized and being used, the panel concluded. "The safety and efficacy of the device has been established because it's already available in the marketplace, and doctors will continue to use it for the off-label indication regardless of what we recommend," said Michael Grimmett, MD, assistant professor of ophthalmology at the Bascom Palmer Eye Institute at the University of Miami School of Medicine (Miami, Florida). "At least now data will be available that was previously unavailable for the procedure."
That data used clinical 12-month follow-up of patients and showed: 94% of patients could see magazine- and newspaper-sized print with the treated eye; 83% could see 20/20 and magazine-, newspaper- or phonebook-sized print; and there were no serious, sight-threatening or unanticipated safety events reported.
An estimated 90 million people in the U.S. have presbyopia and so must rely on reading glasses often multiple pairs. H.L. Milne, MD, a CK surgeon and director of The Eye Center (Columbia, South Carolina), said that with approval of the new application "we will be allowed to inform the wider baby boomer population that finally there is a safe, simple treatment available just for them. It will be welcome news to the millions who have been waiting to cut the chain on the hassle and age-stigma associated with reading glasses."
Privately held Refractec is studying other potential applications for CK technology such as astigmatism and post-cataract treatment.