Скачать 1.23 Mb.
6 notice that the tests are sort of up here in the order that
7 they were implemented in the blood community.
8 So, the very first that was done was a syphilis
9 test. It is a required test at this time. It involves
10 donor notification, deferral and/or lookback.
11 Unfortunately, this one is the outlier. It has been
12 regulated by the 510(k) mechanism, probably largely because
13 it was the very first kit and the regulatory mechanisms
14 weren't as well defined at that time.
15 You will see that the next tests that came along
16 were tests for HBV. They were required. They involved
17 notification, deferral and/or lookback, and we have handled
18 those previously as PLA and now I am going to refer to them
19 all as BLA.
20 The same thing goes then for any other test that
21 has been required or recommended by FDA where we have
22 recommended notification, deferral or lookback. We have
23 generally had those come in as BLAs.
24 When we get down to ALT, you note that this test
25 is not a required test. It is a test that is done
60 1 voluntarily by the blood organizations. Some of them have
2 notification, deferral and lookback procedures; some of them
3 don't. This kit is handled through the 510(k) mechanism.
4 Also, similar to ALT tests are the CMV tests where
5 the testing is done voluntarily. It may not even be done on
6 all units. It may be done on selected units. There is
7 generally no notification, deferral or lookback. The unit
8 is simply not used for a particular intended recipient.
9 That, again, is handled as a 510(k).
10 The most recent addition to the list is the Parvo
11 B19 test kits. As Robin mentioned in his talk, they are not
12 required. There is no donor notification, different or
13 lookback, and they are an in-process control.
15 As an in-process control, these Parvo B19 test
16 kits -- normally the plasma fractionator develops and
17 manufactures the test kit in-house. They then supplement
18 their biologics license application for their fractionated
19 product to include an additional in-process test. FDA
20 reviews that BLA supplement for the scientific soundness of
21 the in-process control, and then has the opportunity to
22 review the test and the results during post-approval
23 inspections to ensure that the test has continued validity.
25 I also wanted to compare and contrast standards
61 1 that are applied under the various mechanisms. Under the
2 BLA mechanism we have a dual set of standards. Both the
3 biologic standards and the medical device standards apply.
4 The biologics standards include standards for safety,
5 purity, potency, sterility, identity and lot release. The
6 medical device standards include compliance with the quality
7 system regulations, or they are also known as the current
8 good manufacturing practices. There is a substantial number
9 of labeling requirements under the medical device standards
10 and there is a requirement for registration and product
13 If you look at the standards that are applied for
14 a PMA, we have only one set of standards that apply here,
15 and that is the medical device standards but more of them
16 apply than applied to the BLA. We have performance
17 standards, sometimes voluntary and sometimes not voluntary.
18 We have to sorry about safety, effectiveness, quality system
19 regulations, the labeling requirements and registration and
22 The standards that are applied under the 510(k),
23 if it is a Class I it is just general controls and safety,
24 effectiveness and the rest. If it is a Class II we add in
25 the special controls. So, the Class II 510(k) would be more
62 1 stringent than a Class I.
3 So, what does this mean in terms of the regulatory
4 burden both for FDA and for a manufacturer of the test kit?
5 With a BLA there would be extensive clinical trials. As I
6 mentioned before, those clinical trials would have to be
7 performed under an IND. There would be a pre-approval
8 inspection by Center for Biologics staff. There would be
9 post-approval inspections by the team Biologics staff, and
10 the kits would be subject to routine lot-by-lot lot release.
11 For a PMA you can have pretty much the same level
12 of clinical trial testing performed. There would also be a
13 pre-approval inspection, but those inspections would be done
14 by the field. Post-approval inspections would also be done,
15 and they would also be done by the field. But we have no
16 mechanism built in for lot release for PMA products,
17 although sometimes if we feel the need for it we can request
19 Finally, the 510(k) mechanism -- there can be
20 limited clinical trial data. There are no pre-approval
21 inspections for 510(k)s and post-approval inspections are,
22 at this point of time, only done on a for-cause basis. The
23 limitations on resources are such that the field has pretty
24 much said they will expend their resources on Class III
25 devices, which would be the PMA devices, and those other
63 1 devices for which there is a need to go out and inspect.
2 So, they would be for cause and, again, there is no lot
3 release for 510(k) products.
5 So, where does HAV testing right now fall? The
6 current HAV tests that are available have been regulated by
7 CDRH. Their current indication for use is detection of
8 antibodies to HAV in human serum or plasma. They are
9 regulated as Class III devices, which means that general and
10 specific controls are insufficient to assure the safety and
11 effectiveness of the device. So, they are reviewed under
12 the PMA mechanism. That was the middle one on the previous
15 For the purposes of how CBER might choose to
16 regulate HAV NAT tests, if for some reason we thought we
17 could review them by other than a PMA mechanism, we do have
18 the opportunity to utilize Section 207 of FDAMA 1997. This
19 is known as the evaluation of automatic Class III
20 designation or also known as do novo classification.
22 In this case the kit manufacturer would submit a
23 complete 510(k) for an HAV NAT test, and their specific
24 intended use would have to be different for it to qualify
25 for the de novo. We would recommend that the intent
64 1 indication for use be something along the lines of testing
2 plasma pools for the presence of HAV DNA. After receipt of
3 the 510(k), CDER would review it and determine is there a
4 predicate; are there deficiencies in the submission; is the
5 device considered low risk; and if there is no predicate,
6 could it qualify for de novo classification?
8 Under the current system there would not be a
9 predicate for a device that is intended for use in testing
10 of plasma pools for the presence of HAV DNA. So, CBER would
11 have to prepare an NSE letter, which means not substantially
12 equivalent. We would list any deficiencies that we may have
13 found in the review of the submission. We would state that
14 there is no predicate, and we would also state that the
15 device might qualify for the de novo classification. After
16 the firm receives that letter, they have 30 days to submit a
17 request to FDA for a Class II designation under the de novo
18 classification mechanism.
20 FDA then reviews that request, and a response to
21 that request must be made within 60 days. The things that
22 are reviewed at that time are the previous review of the NSE
23 510(k); whether or not there were deficiencies; whether or
24 not they can be addressed. They base it on the review of
25 the request for classification. If the firm has requested
65 1 Class II, does the FDA feel that it fits appropriately into
2 Class II, and there is a risk evaluation made.
4 If, for some reason, the FDA determines that the
5 new test kit has to be classified as a Class III, then there
6 would be a submission of a PMA or a PDP required. If, on
7 the other hand, FDA agreed that the test could be classified
8 as Class I or Class II on the basis of the fact that a
9 510(k) has already been submitted and reviewed, assuming it
10 is acceptable, it can be deemed to be cleared. It can go to
11 market immediately, and it then serves as the predicate for
12 future submissions of similar kinds of test kits.
14 Lastly, in all of these cases the kind of data
15 that will be needed would have to cover accuracy,
16 specificity, sensitivity, precision and stability. The data
17 requirements could be different though based on the kind of
18 submission that would come in. They would be most
19 burdensome for a BLA and the least burdensome for a 510(k).
20 In any case, the applicant would be advised to seek guidance
21 from CBER to know just how much testing needed to be done.
22 That is it.
23 DR. HOLLINGER: Is this clear to the committee,
24 how this works? Are you saying that the FDA prefers this to
25 be a 510(k) because it requires less extensive evaluation?
66 1 MS. KOCH: I don't think we have made that
2 decision, that we prefer it that way, but if we follow
3 current thinking it would make sense to review it as a
4 510(k). So, there was a little bit of a focus on 510(k) but
5 it is not a done deal.
6 DR. HOLLINGER: And what would be the predicate
7 device for that decision?
8 MS. KOCH: In this case, that is why I was
9 explaining the de novo classification. There is no
10 predicate for this device. So, we have to utilize the new
11 mechanism available to us under FDAMA, the de novo
12 classification. So, normally when there is no predicate the
13 device is automatically classified as Class III and
14 automatically has to come in as a PMA, but just to present
15 the fact that there is an option if the committee were to
16 recommend, because of the level of importance of the test,
17 that we could go with a lower regulatory mechanism, the
18 510(k) mechanism is available to us. It is just not a
19 straightforward one.
20 DR. HOLLINGER: Thank you. Jay?
21 DR. EPSTEIN: I think that Dr. Koch answered the
22 question, but basically what we are saying is that should
23 the committee recommend, and should we concur, that there is
24 not a need for routine donor screening, if you concur that,
25 nonetheless, when that is done on a minipool you should
67 1 identify the positive unit and the infected donor and inform
2 the donor, that is then medical testing.
3 Well, medical testing for hepatitis A has
4 precedence in the agency. It is reviewed as a Class III PMA
5 in CDRH. So, we wouldn't see a real difference between a
6 NAT test versus an antibody test versus an antigen test. It
7 should be treated as a medical diagnostic. So, what we are
8 trying to explain is that if we get that recommendation from
9 the committee and concur, we wanted the committee to
10 understand what was at stake with oversight of that test as
11 a medical diagnostic, and what we are saying is that the
12 current system would require that it be a Class III PMA but
13 that there is a legal mechanism under the FDA Modernization
14 Act for it to be, if you will, down-classified to a 510(k),
15 which would then make the oversight more consistent with the
16 way we deal with other non-required tests which are,
17 nonetheless, sometimes reported as medical information to
18 the donor, and that would include CMV, syphilis and ALT.
19 So, what we are saying is if you go that route and
20 concur that this need not be a routine donor screen, because
21 it should still be viewed as a medical diagnostic, we are
22 suggesting that there is a route for harmonizing it with the
23 other tests that are viewed similarly. Is that helpful,
25 DR. HOLLINGER: Yes. Dr. Koerper?
68 1 DR. KOERPER: Could you please just briefly
2 refresh my memory as to the difference between Class I,
3 Class II and Class III?
4 MS. KOCH: A Class I device is considered a low
5 risk device such that general controls, which would be
6 registration, listing and adherence to GMPs, are sufficient
7 to ensure the safety and effectiveness of the device. A
8 Class II device has a little bit higher risk, and it has
9 been deemed insufficient to have just general controls. We
10 think that there are special controls in addition to general
11 controls that may be necessary. A Class III device is
12 considered the highest risk device, and general controls and
13 special controls alone are inadequate or insufficient to
14 ensure the safety and effectiveness of the device so we
15 require the clinical trials and a pre-approval application,
16 and there are usually some sort of performance standards
17 that are developed along with that application.
|1 department of health and human services||Department of health and human services|
|Department of health and human services||Department of Health and Human Services|
|Department of health and human services||U. S. Department of health & human services|
|Department of health and human services||U. S. Department of health and human services|
|U. S. Department of health and human services||Department of health and human services|