There are not enough safety and efficacy data for the FDA to issue a low-risk Class II device classification for salvaged blood irradiators used to prevent metastasis during surgery, an advisory panel agreed on Tuesday.
Members of the Radiological Devices Advisory Panel said that these devices should carry a high-risk Class III designation, meaning that any sponsor seeking this indication will have to prove that the devices not only kill circulating tumor cells in salvaged blood, but they reduce the risk of metastasis and improve long-term patient outcomes without imposing undue risk.
Ahead of the meeting, FDA staff only found 10 studies on these devices, just two of which are cleared for use in the U.S. Furthermore, none of the studies provided high-quality or long-term data on the devices' efficacy.
In addition, because none of the studies looked at risks, FDA staff came up with their own list of potential problems, to which members of the advisory panel added concerns.
One of the most troubling, said Daniel Song, MD, of Johns Hopkins University in Baltimore, is that the radiation that would supposedly kill tumor cells could also kill lymphocytes, which are critically important in fighting cancers. In fact, similar devices cleared to prevent graft-versus-host disease (GVHD) actually work by killing immune cells.
"T cells are highly sensitive to radiation -- much more so than tumor cells," said Song. "We now understand that the immune system has a very important role in controlling cancers, and this could actually have a negative effect" on cancer prognosis, instead of a protective effect.
Additionally, while it's unclear how much blood would actually be exposed to radiation during the procedure, the volume would likely be small -- a liter or two at most, he noted. The number of circulating tumor cells varies with each cancer, with each stage, and with each patient, and while more cells certainly predict a higher risk, "I'm personally pessimistic" that killing a percentage of them in the small amount of treated and re-infused blood would have much effect on the metastatic risk, he added. "You're only watering down a percentage of cells in circulation."
This questionable benefit, added to the risks of extending surgical time to allow for the irradiation and the potential damage to immune cells, set off alarm bells for other panel members, too. Radiation itself provokes cell changes that can lead to malignancy, noted Andy Chen, MD, PhD, of Oregon Health & Science University in Portland.
It's also not clear where the irradiation procedure would take place if the device was used in this capacity, some members said.
The devices for GVHD are typically set up in the blood bank, explained Jonathan Waters, MD, of the University of Pittsburgh. During the procedure, runners take the salvaged blood from the operating room to the blood bank, where it's irradiated and repackaged. Every time blood is physically separated from the patient, there's a risk of error, he added.
"I have seen mismatching where the unit [of blood] went to the wrong place and was transfused after processing," Waters said. "There's a significant risk of removing the blood from the OR to a remote location."
The two devices cleared for use in the U.S. were regulated through the 510(k) pathway. Designating the irradiators as a Class III device would mean that any manufacturer who wanted this indication would have to submit a premarket approval. The panel was quite clear in what would be required to prove safety and efficacy for blood irradiation to prevent metastasis.
Studies would have to be randomized and include both short- and long-term safety data. Since the indication is an oncologic one, outcomes would have to include factors like time to disease progression, rates of metastasis, and overall survival. And since cancers are so wildly variable in progression and metastatic risk, studies should be specific to cancer type and stage, said Victor van Berkel, MD, PhD, of the University of Louisville in Kentucky.
"We would like a very clearly delineated clinical profile of the patients that would fit into this circumstance," he said. "We would want both short-term and long-term oncologic data suggesting benefit. We want to know exactly what the transfused blood volume is."
In a practical sense, he added, the discussion could largely be an academic exercise. FDA reviewers turned up no evidence that either of the two cleared devices are even being used for this indication in the U.S.
"I agree that we don't have enough information, and this should be a Class III device," van Berkel said. "However, I don't think, practically speaking, that there is going to be a free-market value application for any of these devices. I can't imagine that someone is going to design a trial to try to look at these things in such a small use patient population."
While the FDA is not required to follow the advice of its advisors, it typically does.
The agency will publish a proposed rule, which will be open for a public comment period. After consideration of all additional comments received, the FDA will proceed with issuance of a final rule, which will identify the final classification for this device type.