For whom?

Over the last 25 years, the regulatory requirements for the approval of contrast agents have been greatly increased by Medicine Agencies, in a quest to “elevate” the standards for the approval of these agents by treating them like all other drugs. That quest may have been well-intentioned, but it ignores the fact that, very unlike all other drugs, contrast media are designed to: (i) have no therapeutic effect; (ii) be administered in one dose, under medical supervision; and (iii) provide their effect under the principles of physics, not those of pharmacology. So, given that the ultimate purpose of Medicine Agencies is to promote public health, a relevant question to ask is: Are these policy changes beneficial for patients?

Body area approvals for extracellular agents

A notable example of the way these changes have worked is the change in the way gadolinium-based contrast agents (GBCA) have been indicated for use. When gadopentetate dimeglumine was approved in the 1980s, it was simply approved for intravenous (i.v.) use. Its safety at a given i.v. dose was demonstrated, and its distribution and elimination kinetics were defined both in man and animals. With that, its effect was understood to be (and actually is) that it provided, under the physics of magnetic resonance imaging (MRI), contrast enhancement of blood and tissues while present in them. There were no restrictions for the placement of the MR coil – it did not matter if one imaged the liver, kidneys, breasts, brain, heart, musculoskeletal structures, et cetera. The effect, and the physics behind it, was the same.

When the next generation of gadolinium-based contrast media (GBCA) came on the market, efficacy got a greater regulatory emphasis. Maybe the authorities had been inspired by the change in their approach to antibiotics, where the first agents were approved for infections, but later, with spreading bacterial resistance, newer ones were approved for individual species. Of course, this example would ignore the fact that, in regard to the efficacy of contrast agents, one does not develop resistance to the laws of physics. Anyway, that did not deter the authorities.

So, it became that newer GBCA would typically start with a limited approval for imaging of a single body area (usually the CNS), and their use in imaging the rest of the body would be “off-label” until each new region could be specifically studied by the manufacturer with a dedicated clinical trial program, each such program then being subject to intense regulatory review, and sometimes actual denial – regardless of the product efficacy that is inherent in the laws of physics.

Does this make sense? For example, abdominal, musculoskeletal and breast MRI may (or may not!) be covered under an approval for “body MRI”, but such an approval does not include the heart. Is the heart not part of the body? As mentioned before CNS is already approved at an earlier stage and is not covered by “body MRI”. If a marketing authorization holder (MAH) gets an approval for liver MRI, what about the right kidney, which is difficult to avoid when imaging the liver. Imaging of the right kidney (and left kidney) is off-label. I am still waiting for the innovation of having law enforcement cameras positioned in the femoral arteries, telling authorities that GBCA is illegally entering the lower extremities of the patient, because it was approved by the Medicines Agency only for renal MRA, not for peripheral MRA.

Is this approach scientifically valid? Clearly not to the imaging field, where, in the interest of patient care, we use GBCA for liver, kidneys, breasts, abdomen, blood vessels (MRA), heart, musculoskeletal structures, et cetera, providing great value to diagnosis. How could it be in the interest of public health for regulators to ignore the principles of physics, and the extensive clinical experience that informs the global standards of medical imaging practice? Furthermore, I am still waiting for an explanation as to why there should be any difference regarding safety (adverse reactions) dependent on whether, for a GBCA, the coil is placed around the knee or the brain.

Does this make sense economically, by adding to the cost of current contrast agents, and making the cost prohibitive to develop future ones? The expense of meeting onerous regulatory requirements that were designed for therapeutic drugs but are arbitrarily applied to contrast agents consumes the R&D budget in several companies. At the same time generic iodine- and gadolinium-based contrast agents appear on the market; the generic companies do not invest in new indications for the contrast agent. And finally, we – radiologists – always want lower prices to the benefit of our patients. Contrast media business does not have the same financial size as, for example, drugs for antihypertension, which are taken daily. If current trends continue, we will come to the end of contrast agent innovation. Is that in the best interests of public health?

Peripheral vs. central reading

In the past, for approval of contrast agents, images from developmental trials were evaluated by the investigational sites, and their findings were considered sufficient support for a marketing authorization application. Considering that is how contrast agents are used in clinical practice, this made sense. But, it is no longer the case. All images must now be sent for central reading by three “blinded” readers who are not affiliated with the clinical trial sites, and have no other knowledge about the cases. This can certainly have advantages in multicenter studies, in regard to standardizing readings, and is more scientifically “pure”. But the isolation of these readers from realistic clinical context, and their assignment to exhausting dedicated reading sessions, can have some unusual consequences, particularly when authorities focus on statistical rigor rather than on clinical merit or physical principles. For one GBCA that was approved for MRA of some vascular territories, approval for use in a major vascular territory was denied because one of the three central readers did not obtain a sufficient sensitivity for stenosis in imaging one of the major vascular territories. The regulators did not look at the images. What will happen the day when local investigators find the images to be beneficial for the treatment of patients and central reading shows conflicting results among readers? Will a marketing authorization be denied? Will the company withdraw the application in order to avoid a “no”?

The efficacy of technology

Despite the desire of authorities to hammer the square peg of contrast agents into the round hole of therapeutics, diagnostic efficacy is very different from therapeutic efficacy. For example, when evaluating an antihypertensive, blood pressure is a simple measurement of the intended effect; either the drug decreases blood pressure or not, and it does so in at least the same percentage of patients as the comparator, or not. But in radiology, the efficacy of a contrast-enhanced scan performed tends to vary much more according to the features and settings of the scanner, and the skill and mindset of the radiologist who performed the reading. However, the authorities mostly ignore the primary roles of the imaging equipment and interpretive skill in determining the efficacy of imaging, and attribute all issues with the efficacy results to the imaging agent being tested. Does that make sense?

Money

The budgets of the various Medicines Agencies are covered at least in part by the fees paid by the companies. The employees of these agencies have an interest in securing their agencies' incomes. Recently, in Denmark we have seen that the Health Authorities have laid off employees at the regulatory office due to fewer applications for new drug approvals. For a regulatory authority, the logic is that the more applications the better, for they bring more work and more fees, which justify more employees and bigger budgets. So, if regulators do not get many new drugs to review, one way to maintain their income and staffing is to divide the applications they do get into multiple body areas. Is this why they take this “indication splitting” approach to contrast agents? Is that in the best interests of patients? I have never been able to get an answer to my questions. Maybe the Medicines Agencies should be financed by the governments and not directly by the industry in the future.

Confused?

There are no logical scientific explanations for the increased bureaucracy around the testing of contrast agent efficacy. It does not bring safer or better imaging agents to market, when the focus is on splitting approvals for the various body regions, instead of on testing special patient populations, such as children, or patients with allergy or impaired renal function. So why is it done this way? In the case of nephrogenic systemic fibrosis the clinical testing did not reveal the serious adverse very late reaction; a more careful review of the literature, including the preclinical studies, might have been more beneficial for the patients than splitting the body up into sections (1).

As it works now, a contrast agent is approved for some indications; then, based on data in the peer-reviewed literature, the experience of colleagues, and ultimately professional guidelines, many radiologists use the agent off-label, in the interests of patient care. This use has increased over the years. Few severe adverse reactions are seen, with no difference between approved and unapproved body area indications. Radiologists prefer that the contrast agent has a general approval, because they are concerned that some patients (or even some local authorities) will question their judgment, when they see that the contrast agent being used for a scan is not approved for that particular body area by the Medicines Agency. To get an approval, the MAH has to perform expensive phase 3 studies, which they are not motivated to do, because they already sell a lot of the agent for that use, and the potential increase in sales from an approval will never cover the costs of the program.

What to do? With one imaging agent, radiologists collected safety data from more than 5000 off-label examinations. They then tried to convince the Medicines Agency to approve the agent for the indication. “No” was the answer. Why? One plausible explanation is that a “yes” opens an uncontrolled door, to a situation of companies saving money, and Agencies losing money. That does not work for the Agencies, of course. But would it work for patients?

Summary of product characteristics

The summary of product characteristics (SmPC) is the legal document for the use of drugs. Recently, I read in the SmPC of an extracellular GBCA that the agent was contraindicated in patients with a pacemaker. In my >30 years in radiology and contrast media research, I have never heard about an interaction between GBCA and a pacemaker. If the contraindication refers to the effect of magnetic fields, why was it applied to the GBCA? Would an unenhanced MRI not have the same effect? In another SmPC for a GBCA, I read that it was recommended that the X-ray equipment was in good condition, even though no GBCA has ever been approved for radiography or CT. There seems to be space for improvements.

Conclusion

We have a system for approval of imaging agent indications that has gone too far into bureaucracy, with efficacy data, which is driven by the laws of physics, now playing a greater role than safety data. A GBCA with a certain relaxivity enhances blood to the same extent in the lower and upper extremity, as long as it circulates. How can it be approved for one and not the other? Instead of focusing on dividing the body into various arbitrary regions (e.g. based on the placement of the coil in MR), regulators should focus on evaluating safety data for children, patients with reduced renal function or on dialysis, and patients with asthma, vs. that in those with no risk factors. That is what truly matters to patient care. The SmPC should be made readable.

By the way, most iodine-based contrast agents are approved for general use where ultrasound agents are only approved for a few parts of the body.

I am left to ask: Who is in charge, and what are they thinking? And I am left to think: The system needs an update.