Bayer's MRI Contrast Agent Gets Panel Endorsement, But Without Comparative Safety Claims

FDA's Peripheral and Central Nervous System Drugs Advisory Committee declines to deem gadobutrol "lower risk" for causing nephrogenic systemic fibrosis.

Gadobutrol injection, Bayer HealthCare's MRI gadolinium-based contrast agent for neurological MRI scans, garnered a unanimous 16-0 approval recommendation from an FDA advisory panel Jan. 21, which also voted 15-1 that it should not be contraindicated for high-risk kidney patients, while drawing the line at terming the drug "lower risk."

The risk in question is that of nephrogenic systemic fibrosis, an incurable, untreatable condition that causes skin lesions and a potentially fatal fibrotic process in multiple organs. This is a class-wide risk for GBCAs, which led FDA to require that these products bear a boxed warning on their label ('Boxed Warning Added To Gadolinium Contrast Agents For MRI On Risk Of Potentially Fatal Disease,' 'The Pink Sheet' DAILY, May 23, 2007).

FDA asked the panel whether gadobutrol could be deemed lower risk for causing this condition. In its background materials, the agency explained that like Bracco Diagnostics' GBCA ProHance Injection (gadoteridol), gadobutrol has a "macrocyclic" molecular structure that reportedly can "cage" gadolinium, preventing it from escaping to build up in tissue ('Bayer's MRI Contrast Agent Gadovist Could Get 'Lower-Risk' Designation,' 'The Pink Sheet' DAILY, Jan. 19, 2011).

The form of the question, however, discomfited several committee members. "Data from folks who are not in same category as the high-risk category is being presented to us as proof of low risk. I think it can't be classified as a high-risk drug," Emil Paganini, Cleveland Clinic Foundation, said. "However, its exposure time, and vague, soft, smoky-type data don't make me confident it can be labeled as lower risk."

FDA Says No To Comparative Safety Claims

"Higher and lower risk are proxies for contraindication or not; it's not a comparative safety claim," Rafel Rieves, director of FDA's Division of Medical Imaging Products, responded. "We don't want to have the misperception that this agent is a low-risk agent. We think all these agents have important risks for NSF."

There are five GBCAs approved for CNS scans: ProHance, Bracco's MultiHance Injection (gadobenate dimeglumine), Bayer's own MAGNEVIST Injection (gadopentetate dimeglumine), GE Healthcare's OMNISCAN Injection (gadodiamide) and Mallinckrodt's OptiMARK Injection (gadoversetamide injection). The last three are contraindicated for patients with chronic, severe kidney disease or acute kidney injury, while the boxed warning for ProHance and MultiHance only says the risk is higher in those patients.

Rieves agreed to reword the question about NSF risk to ask simply whether gadobutrol should be contraindicated for the kidney patients, leading to the 15-1 vote that this was not necessary. This was the second day in a row the committee had requested a rewrite of FDA's preset questions ('Avid Amyvid Could Be Approved If Training Improves, Advisory Committee Says,' 'The Pink Sheet' DAILY, Jan. 20, 2011).

The outlier on the contraindication question, Mark Green, Mount Sinai School of Medicine, said that he wanted to see more post-marketing data, and Pooja Khatri, University of Cincinnati Academic Health Center, agreed that a contraindication is not necessary although she voted with the majority.

Existing post-marketing data through October 2010 include reports of 10 NSF cases among patients who had received gadobutrol out of a total of 6 million doses administered. All but two of the NSF cases were "confounded" due to the fact that the patients had received at least one other GBCA. Both single-agent cases occurred before 2009, since which the use of gadobutrol has expanded greatly without generating new cases, Martin Rosenberg, Bayer's senior director of global clinical development for diagnostic imaging, said.

DMIP clinical reviewer Barbara Stinson provided comparable figures showing 438 single-agent NSF cases out of 47 million Omniscan doses administered worldwide through February 2009, plus another 90 confounded cases; for Magnevist, 135 single-agent cases and 276 confounded cases out of 95 million doses; for Optimark, seven single-agent cases and 11 confounded cases out of 800,000 doses; for MultiHance, no single-agent cases and eight confounded cases out of 6 million doses; and for ProHance, one single-agent case and 13 confounded cases out of 12.3 million doses.

But these numbers are still too small and the subject too controversial for FDA to adjudicate comparable safety claims, Rieves said.

Bayer apparently is renegotiating the proposed trade name of gadobutrol with FDA, as Stinson mentioned in her recommendation for approval.

The company did not use the trade name "Gadovist" in its presentation to the committee, although it explained in its background package that the drug is marketed in 62 countries under that name, sometimes with "1.0" added to emphasize its concentration, which is double that of other GBCAs. That dosing difference itself has generated some concern about the potential for medication errors, FDA noted in its briefing documents.

-Martin Berman-Gorvine ( [email protected] )