Medical Device Daily Washington Editor

FDA's interest in the data generated by clinical trials for PMAs of medical devices seems to grow more "granular," as they say, and a new guidance from the agency is the second this year that mandates clinical trial design for a specific device in a specific application.

Earlier this year, FDA published the guidance for drug-eluting stents (DES) (Medical Device Daily, March 28, 2008), but the interest on the part of industry in bioresorbables and the fact that the DES guidance does not apply to bioresorbables suggest that the DES guidance might soon be obsolete. Stent makers also sometimes are heard to grouse about FDA's notion that a stent PMA is specific to blood vessels.

The latest guidance deals with catheter ablation for atrial flutter and at least in one respect, differentiates between radio frequency ablation and cryogenic ablation. However, at least one industry voice says the separation of flutter and fibrillation as indications, and hence as PMAs, makes sense.

As always, the agency says it will consider study designs other than randomized, controlled trials – including a single-arm study using performance goals – but the agency has often criticized such trials in advisory committee hearings. For atrial flutter trials, FDA recommends a primary efficacy endpoint using a chronic success measure such as freedom from recurrent flutter at six months. However, the document says that "there is sufficient scientific evidence to indicate that acute procedural success (defined as the creation of bi-directional block at the cavo-tricuspid isthmus) is predictive of chronic success," and hence may be used in lieu of a chronic primary efficacy endpoint.

However, this provision is said to apply only to radio frequency energy ablation, apparently leaving out makers of cryoablation equipment. Even in the case of RF ablation, the surrogate endpoint is of no help if the patient has to go on anti-arrhythmic drugs "in the follow-up period," in which case that patient should be counted as a study failure.

Suggested safety endpoints include freedom from cardiac perforation/tamponade, death, infarction and pulmonary embolism. Sponsors should base enrollment on the larger of the numerical requirement imposed by safety and efficacy, and minimum follow-up requirements are at seven days on the phone, and office visits at 30 days and six months. Any patient who is likely to experience flutter at the time of treatment should be on anti-coagulant therapy for three to four weeks prior unless an echocardiogram of the trans-esophageal area presents no evidence of left atrial appendage thrombus.

As always, the agency describes the guidance as not establishing "legally enforceable responsibilities."

Jean-Pierre Desmarais, the chief scientific officer at Cryocath Technologies (Montreal) told Medical Device Daily that he did not see it as unduly burdensome for FDA to require a separate PMA for atrial fibrillation and atrial flutter. The syndromes are "so different that FDA is right to require a separate PMA." He pointed out that doctors sometimes see both flutter and fibrillation in the same patient and that "flutter can induce fibrillation," but they are sufficiently distinct in terms of treatment and prognosis to warrant separate indications.

As for the rationale behind allowing a surrogate endpoint for RF devices and not cryogenic units, Desmarais said "I believe there are four RF devices approved for atrial flutter," and only one cryogenic unit, and as a result, a sponsor "can put forward a case that RF devices" can be evaluated with a surrogate endpoint.

As for the size of a trial, Desmarais speculated that for a single-arm trial with a performance criterion (rather than an RTC): "Typically when you look at a trial with a success rate at 90% ... a sponsor will need 150 to 170 patients." On the opposite extreme, an RCT designed to demonstrate superiority might require as many as 700 enrollees. Hence, he said, there is "not much incentive to go for superiority" rather than non-inferiority in an RTC, and "its less risky as well" to aim for non-inferiority if the data are at all equivocal.

OIG weighs in on gainsharing arrangement

Gainsharing might not draw a lot of press these days, but a recent opinion by the Office of Inspector General at the Department of Health & Human Services indicates the idea is not completely dead.

The July 31 OIG opinion, which did not surface on the Office's web site until Aug. 7, details an arrangement between an academic medical center and two physician groups, one neurosurgery group and the other a spinal surgery practice. OIG says that it found the gainsharing arrangement problematic, but the document also says that the hospital had agreed to pay the two physician practices the gainsharing fees only upon receipt of a favorable opinion from OIG, which may have played a role in OIG's decision not to take action.

The arrangement dealt with the use of devices and supplies during spinal fusion procedures based on a study of the operating room costs by the hospital, which identified "36 specific cost-savings opportunities." This list includes one "use-as-needed biologics," bone-morphogenic protein (BMP), and 35 standardized products. The use of BMP was said to have been reduced from 15% of spinal fusion patients to 11%. According to the OIG document, the arrangement "contained several safeguards intended to protect against inappropriate reductions in services." The physician groups were to receive a fee equal to half the savings.

The opinion found that the arrangement did a reasonable job of ensuring that the contract did not induce inappropriate financial incentives for referral and that the changes to OR practice were in line with objective historical data. OIG nonetheless expressed concern that such arrangements "could be vehicles to disguise payments for referrals," and that any such arrangement "that cannot be adequately and accurately measured for quality of care would pose a high risk of fraud or abuse." OIG also says that an agreement "structured so as to pose a heightened potential for patient steering and unfair competition would be considered suspect." OIG redacted the names of the institutions in question from the document.