As a former scientific reviewer within ODE I was not at all surprised to see the internal survey results of how “intended use” and “indication for use” were inconsistently interpreted and applied. This issue has been around and debated internally since the mid to late 80’s and has never truly been addressed directly by the Agency. This inconsistency of interpretation can either cause a major surprise during a 510(k) review or can render a marketed device misbranded and adulterated for being promoted for an unapproved use. This inconsistency deserves the Agency’s highest attention since it can create a major problem for the manufacturer. At one end of the spectrum a finding of not substantially equivalent during a 510(k) review can end a product or a company. At the postmarketing phase of the product lifecycle, a misbranding/adulteration violation of the FD&C act can cause a suspension of sales of the medical device and litigation as a result of a serious injury/death due to use of the medical device for unapproved uses.
It is critically important for the reader to know that the phrase “indication for use” is not used with regard to medical devices in the FD&C Act, as Amended. This includes both Section 510/513 (where the 510(k) process is defined) and 515 (where the PMA process is defined). The law specifically uses “intended use” to indicate how, in part, a device is to be found equivalent via 510(k) of the Act. Section 515 specifically requires the applicant or sponsor of a PMA to provide the “intended use” of a device (not indication for use). However, when the law was codified into regulation the use of a new phrase for devices, “indication for use”, was incorporated. Indication for use is used multiple times in 21 CFR 801 to define requirements, for example, exemptions from adequate instructions for use (i.e., prescription devices, which most devices on the market today are). In addition, 21 CFR 814.20 define what an indication for use is within the context of a PMA.
This is a strange observation (no direct reference for indication for use in device law), although I am sure not unique when it comes to law being translated into regulation, and one must wonder why this occurred (sorry noconspiracy theories here). However, it is reasonable to assume that those writing the regulation at the time viewed PMAs to be analogous to New Drug Applications (NDAs). This based on the fact that both class III (PMA) devices and new drugs (NDA) pertain to new and novel applications of technology for human use with little history, thus high risk. The drug law is replete with references to indication or indication for use with respect to applications and labeling. Thus, it would appear the device regulators borrowed this terminology from drug law/regulation when writing the device regulations. However, it still remains the fact that Congress never used “indication for use” when passing the FD&C Act, as Amended.
Now for the practical effect of having these two different phrases. I believe there is really no real practical difference between these two phrases when one looks at the 510(k) review process implemented by FDA. Yes, I appreciate that a new “intended use” requires a 510(k) to be found “not substantially equivalent” and thus requiring a PMA. This is not a minor effect I grant you but FDA has been finding 510k submissions with a modified intended use to be equivalent for many years and continues to do so. Let me explain how.
The FDA has always considered “indication for use” to be equal to or at least a subset of “intended use”. This is supported by ODE’s Blue Book Memo “Guidance on the CDRH Premarket Notification Review Program 6/30/86 (K86-3)” and the first question asked in the flowchart contained in ODE’s Policy. The first question is “Does New Device Have Same Indication Statement?” (see below). Thus, in answering the question “does a device have a new intended use” (which is required to be answered during the review and would result in a not substantially equivalent decision by law) FDA first looks at indication for use. Following the flowchart if indication for use is the same the question of intended use is answered (i.e., no new intended use). There is no other part of the review flowchart that results in a new intended use decision.
If the indication is different, the FDA then looks at the impact of the indication difference on the intended effect. Thus, you must have both a change to indication (first) and then a change in intended effect (second) to achieve a “new intended use” decision. In other words, if a device cuts hepatic or liver tissue (and indicates nothing else in labeling) and compares itself to a scalpel, the reviewer would conclude there is no new indication for use or intended use. The fact that it may have a new exotic technology that allows cutting of hepatic tissue is evaluated later in the review flowchart (but unrelated to new indication for use or intended use).
While a reviewer, the approach I used, and was endorsed by management at that time, was the use of the following definitions. Please be forewarned that to assume this was consistent with all reviewers and divisions within ODE would be incorrect.
Indication for Use = the clinical condition it is intended to diagnose,
treat (devices seldom cure anything), reconstruct or improve upon.
Patient populations were considered within the review of indication for use. For example, pediatrics and adult populations were always considered separately. In addition, if a patient population was historically considered off limits and were now being considered to be part of the treatment population this too was considered (for example, to use an orthopedic implant in the osteoporotic patient population (or don’t exclude it)). In addition, claims made with respect to the treatment were also considered part of the indication for use review. For example, an laparoscopic or minimally invasive device that claimed a faster healing rate or fewer adverse events/serious complications compared to the predicate were either found not equivalent or were forced to provide clinical study evidence to support such claims (my personal preference was the latter option. Don’t believe that if a laparoscopic instrument is equivalent in all other ways that the device should be reviewed under a PMA.).
Intended Use = a device’s effect on the human body.
By effect I mean what does the device do to the human body, (e.g., cut, ablate, coagulate, replace a body part, image, inject/extract liquids/tissue, etc). During the 510k review process the technological methodology in achieving this effect was NOT part of intended use but did affect the decision as to whether the device was substantially equivalent.
After applying these definitions to the Blue Book Policy K86-3 to a few hundred 510(k) submissions over multiple therapeutic areas it made complete sense (to me of course) why you can review a device within the 510(k) process even if the intended use is different. If a device is already cleared to cut or ablate tissue in the liver and you want to modify its intended use to cut tissue in the colon or the heart it would be ridiculous to elevate the review of that device to the PMA process based on this new intended use (remember there is no such thing as an indication for use in the FD&C Act as Amended or the regulations governing the 510(k) process). However, if you claim that ablating tissue treats a specific clinical condition (e.g., atrial fibrillation) than you clearly have a new indication for use or intended use and a PMA should be required.
You must look at both intended use and indication for use to understand whether a device should be reviewed via the 510(k) process. Why? I am a firm believer that medical devices and pharmaceuticals are two different groups of products that deserve their own unique regulatory process. Pharmaceuticals are prepackaged and delivered in a consistent manner. This delivery is typically consistent across the treated population (oral/injected) and depends on a chemical/metabolic reaction not affected by human intervention. The need to review how a drug is delivered to a patient is not so important in minimizing risk to the patient. Devices, on the other hand, are typically tools that allow a clinician or surgeon to repair, inject/remove substances, reconstruct, image or diagnose a condition of the body. Most devices do not treat/cure a medical condition. They are tools to assist a clinician or surgeon. For example, a medical laser, although a complex medical device allows a surgeon to cut or coagulate tissue. When a medical device is labeled or indicated as a tool and nothing more a regulator needs to only assure the device can be safely used as a tool (i.e., that it can be consistently be used in a safe manner to achieve it utilitarian function as a tool). How effective the laser is in cutting or coagulating tissue is largely dependent on the skill of the surgeon. Thus, this is why I believe it is appropriate to find a medical laser to be equivalent to a scalpel for the purposes of cutting soft or fibrous tissue if the appropriate testing is provided in the 510(k) submission. The 510(k) approach would continue to be appropriate if the type of laser is changed (CO2 to Argon) if the intended use continues to be cutting soft or fibrous tissue. However, if a medical laser manufacturer wants to take the next step and label there product with an indication for use that includes keratoplasty (corneal sculpting) the medical laser continues to be a tool that is used to cut tissue BUT now it is indicated to correct vision, a new indication for use and, therefore, intended use.
Let’s provide another example. FDA has allowed radiofrequency ablation devices (RAD) to remain in the 510(k) process for cutting/coagulating tissue in multiple therapeutic areas (gastroenterology, cardiac, etc.). However, not all RADs are regulated through the 510(k) process. RAD devices are preamendment devices and are classified as class II devices. These devices are cleared via 510(k) to cut/coagulate/denature tissue, including liver and cardiac. This is also true when a manufacturer has modified there preamendment designs to include temperature control mechanisms or saline cooled systems. These technological changes were allowed through a 510(k) process. This, in the author’s opinion, makes complete sense. However, a RAD device only stays in this 510(k) process IF the manufacturer sticks with the general indication of ablating tissue regardless if the tissue is in the liver or the heart. However, the moment the manufacturer wants to promote or label the RAD for treatment of atrial fibrillation the device has a new indication for use, and therefore a new intended use, and becomes a class III device and a PMA is required.
So to wrap up this topic, although there is no reference to indication for use in the FD&C Act as Amended or the regulation governing the 510(k) process, FDA has incorporated indication for use as part of the definition of “intended use”. Furthermore, FDA has allowed changes in indication for use, thus intended use, via the 510(k) process. I believe this existing practice should be continued but clarified by guidance. Equally or possibly more important is to provide this guidance to reviewers and managers within ODE to assure consistent application during the review process.
It is recognized that the FDA has not been consistent on how they use intended use/indication for use over its greater than 30 years of regulating medical devices. A very good example of this is how general ablation devices for urology have experienced “intended use creep” over the years. In contrast to my previous device examples, as urological ablation devices evolved from ablating urological tissue to treating benign prostatic hyperplasia FDA management did not consider this intended use evolution to be considered a change in intended use. I believe this decision was not in harmony with the FD&C Act and contributes to the confusion experienced by private industry. For example, if one can move to a “treatment” intended use in urology in a 510(k) process why can you not do the same in cardiac indictions (atrial fibrillation)? However, my hope is that if they follow the approach I have outlined from a historical perspective that a more consistent and predictable review process will occur without a dramatic change in regulatory burden. In addition, manufacturers will be less likely to make significant changes to labeling that will be consider misbranded and adulterated.
Be sure to check out our RSS Feeds page for the latest news impacting the medical product industry!