4. Special Protocol Assessments
1105
Special Protocol Assessment (SPA) is a process that ideally results in agreements between
1106
the sponsor of a drug or biological product
73
and the division responsible for reviewing the
1107
application. The SPA provisions of the Federal Food, Drug, and Cosmetic Act (FD&C Act)
1108
apply to clinical trial protocols intended to form the primary basis for demonstration of
1109
effectiveness in support of a new drug application (NDA), biologics license application
1110
(BLA), or efficacy supplements to approved NDAs or BLAs; the SPA provisions do not
1111
apply to IVD protocols.
74
1112
1113
In codevelopment programs, an SPA submission may include questions regarding certain
1114
clinical trial design elements related to a drug or biological product, including an IVD’s
1115
effect on interpretation of product data. However, a SPA submission should not include
1116
questions related to aspects of the IVD’s performance (i.e., IVD data collection that is
1117
independent of the drug or biological product). In general, questions about the drug or
1118
biological product should be directed to the therapeutic product review center, and questions
1119
about the IVD should be directed to the appropriate IVD review center. FDA expects that
1120
the therapeutic product and IVD review centers will consult each other on crossover issues.
1121
73
The SPA provisions apply to agreements between FDA and the sponsor of an investigation or an applicant for
approval for a drug under FD&C Act section 505(b) or for a drug that is also a biological product under section
351 of the Public Health Service Act. See 21 U.S.C. 355(b)(5).
74
See FD&C Act section 505(b)(5) and FDA’s “Guidance for Industry: Special Protocol Assessment” for
additional information on SPAs (
http://www.fda.gov/downloads/Drugs/.../Guidances/ucm080571.pdf
).
Contains Nonbinding Recommendations
Draft - Not for Implementation
32
1122
Sponsors should note that alterations to an IVD (e.g., changed cut-off value, altered scoring
1123
system, addition of analytes) or changes in the performance characteristics of an IVD (e.g.,
1124
sensitivity, specificity) may affect the type or interpretation of the data collected in the
1125
therapeutic product trial. In some cases, these IVD changes could negatively affect the
1126
ability to interpret the therapeutic product data and could necessitate amending or revising
1127
the terms of the SPA agreement, as described in the SPA guidance. For example, IVD
1128
alterations might change the characteristics of the enrolled patient population or could alter
1129
the threshold for a positive outcome used as a primary endpoint.
1130
1131
If an IVD is altered or replaced with a different technology after the trial has begun,
1132
interpretation of therapeutic product data may be negatively impacted. Under section
1133
505(b)(5)(C)(ii) of the FD&C Act, such changes may be considered a substantial scientific
1134
issue essential to determining the safety or efficacy of the therapeutic product, identified after
1135
the trial has begun, and may lead to rescission of the SPA agreement.
1136
F. Planning for Contemporaneous Marketing Authorizations
1137
When an IVD companion diagnostic is essential for the safe and effective use of a
1138
therapeutic product, FDA intends to make every effort to coordinate the review so that the
1139
therapeutic product and the companion diagnostic can receive marketing authorization at the
1140
same time. To achieve contemporaneous marketing authorizations, FDA recommends that
1141
the IVD and therapeutic product sponsors plan ahead to assure coordination of the
1142
therapeutic product and IVD submissions.
1143
1144
1. Coordinating Review Timelines
1145
To support contemporaneous marketing authorizations for the therapeutic product and
1146
IVD companion diagnostic, consideration should be given to the differences in review
1147
timelines for the different products. NDAs and BLAs (and their supplements) are
1148
reviewed under standard review timelines or under priority review timelines if the criteria
1149
for priority review are met.
75
Review times may be shortened even further for a
1150
marketing application of a breakthrough therapy-designated product.
76
In addition,
1151
rolling review may be available for applications for therapeutic products designated as
1152
fast track or breakthrough therapy.
77
Review of PMAs can be placed on hold if
1153
deficiencies are identified during review of the submission, e.g., if FDA determines that
1154
75
See FDA’s guidance for Industry “Expedited Programs for Serious Conditions – Drugs and Biologics”
(
http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm358301.pdf
).
76
See FDA’s Manual of Policies and Procedures: “Good Review Practice: Review of Marketing Applications
for Breakthrough Therapy-Designated Drugs and Biologics That Are Receiving an Expedited Review”
(http://www.fda.gov/downloads/AboutFDA/CentersOffices/OfficeofMedicalProductsandTobacco/CDER/Manu
alofPoliciesProcedures/UCM407009.pdf).
77
See note 75.
Contains Nonbinding Recommendations
Draft - Not for Implementation
33
supplemental testing is necessary.
78
Unless care is taken to assure that the submission for
1155
the IVD companion diagnostic is timely and complete, the sponsor may incur delay in the
1156
total time to marketing authorization of the IVD companion diagnostic, which may in
1157
turn affect the timing of the approval of the corresponding therapeutic product. The
1158
points discussed below are intended to help sponsors manage timing aspects for the
1159
separate submissions.
1160
1161
i. Modular PMA
1162
In most cases, the modular PMA approach will allow the most flexibility for IVD companion
1163
diagnostic submissions. For “traditional” PMAs, an applicant submits all components of a
1164
PMA, as outlined in 21 CFR 814.20, simultaneously. A “modular” PMA process allows an
1165
applicant to submit discrete sections, or modules, of the PMA as they are completed.
79
Using
1166
the modular PMA process, the IVD companion diagnostic sponsor submits analytical data,
1167
manufacturing data, and other information required under 21 CFR 814.20, while collecting,
1168
compiling and analyzing the clinical data. When the clinical data are complete, the data are
1169
submitted in the final module of the PMA, and the 180-day “PMA review clock,” under 21
1170
CFR Part 814, begins on that date.
80
1171
1172
When implemented appropriately, the modular PMA approach allows the applicant to resolve
1173
deficiencies identified by the IVD review center earlier in the review process, making the
1174
final review more likely to be completed concurrently with review of the therapeutic product.
1175
1176
ii. Premarket Review Submissions
1177
As with all medical devices, FDA will apply a risk-based approach to determine the
1178
appropriate regulatory pathway (e.g., a PMA or a premarket notification submission
1179
(510(k))) for a specific IVD companion diagnostic for its intended use. A Class III IVD
1180
companion diagnostic that obtains FDA approval is typically approved for a specimen type,
1181
target population and therapeutic product. If an approved IVD companion diagnostic is to be
1182
used for additional specimen types, target populations or therapeutics, the sponsor can submit
1183
a PMA supplement for the new intended use. Other types of changes may also require a
1184
PMA supplement.
81
The type of PMA supplement is dependent on the type of change and
1185
the nature of the review required. FDA recommends that sponsors consult with the
1186
appropriate IVD review division to discuss the appropriate type of submission.
1187
1188
If FDA has previously classified a legally marketed (predicate) IVD companion diagnostic
1189
78
See FDA’s guidance for industry and FDA staff “FDA and Industry Actions on Premarket Approval
Applications (PMAs): Effect on FDA Review Clock and Goals”
(
http://www.fda.gov/medicaldevices/deviceregulationandguidance/guidancedocuments/ucm089733.htm
).
79
See FDA guidance “Premarket Approval Application Modular Review”
(
http://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/ucm0897
67.pdf
) for additional information about the modular PMA review process, including instructions and
provisions for modular PMAs.
80
Upon receipt of the final module, FDA makes its filing decision based on whether the last module includes all
the information necessary to complete the PMA as required by 21 CFR 814.20. If FDA files the PMA, the
filing date is the date that the application became complete, typically the receipt date of the last module.
81
21 CFR 814.39.
Contains Nonbinding Recommendations
Draft - Not for Implementation
34
as a Class II (non-exempt) device, a new IVD companion diagnostic may obtain marketing
1190
authorization if FDA determines, through review of a premarket notification (510(k))
1191
submission, that the new IVD companion diagnostic is substantially equivalent to the
1192
predicate.
82,83
If no appropriate predicate is available, a new IVD companion diagnostic is
1193
considered Class III and subject to premarket approval by operation of law.
84
However, if
1194
FDA believes that a reasonable assurance of safety and effectiveness for a new IVD
1195
companion diagnostic may be provided by general controls or general and special controls,
1196
FDA may identify the test as eligible for the de novo process.
85
Devices eligible for the de
1197
novo process may obtain marketing authorization if FDA determines, through review of a de
1198
novo request for classification, that general controls or general and special controls provide a
1199
reasonable assurance of safety and effectiveness. Devices that are classified into Class I or
1200
Class II through the de novo process may be marketed and used as predicates for future
1201
510(k) submissions. Changes to a Class I or Class II device that could significantly affect the
1202
safety or effectiveness of the device or a major change or modification in its intended use
1203
require a new premarket submission (e.g., a 510(k) or in some instances a PMA).
86
1204
1205
iii. Bioresearch Monitoring Inspections and Manufacturing Inspections
1206
There are two types of inspections that can occur in the context of a PMA submission:
1207
bioresearch monitoring (BIMO) inspections and manufacturing inspections. The BIMO
1208
program conducts inspections of clinical investigations to ensure the protection of research
1209
subjects and the integrity of data submitted in support of the PMA. Sponsors should
1210
anticipate the Agency’s need to inspect clinical trial sites with respect to both the therapeutic
1211
product and the IVD companion diagnostic. When an IVD companion diagnostic PMA is
1212
reviewed, CDRH/CBER BIMO personnel have the authority to inspect the clinical trial
1213
enrollment sites; however, the inspections of clinical enrollment sites will usually be
1214
coordinated by the lead therapeutic product review center (i.e., CDER Office of Scientific
1215
Investigations or the CBER Division of Inspections and Surveillance) and may be performed
1216
by the FDA’s Office of Regulatory Affairs. Nonetheless, the IVD manufacturer should still
1217
submit information about the clinical testing sites, including clinical line data, to the PMA for
1218
BIMO review.
87
FDA will coordinate review and inspections of clinical sites, as needed,
1219
among the appropriate review center(s).
1220
1221
To facilitate IVD-related BIMO activities, PMA applicants should submit BIMO information
1222
that is organized together, in its own section, or otherwise easily identifiable. BIMO
1223
information typically includes lists of the clinical investigators with contact information, all
1224
82
21 U.S.C. 360c(i).
83
See FDA guidance “The 510(k) Program: Evaluating Substantial Equivalence in Premarket Notifications
[510(k)]”
(
http://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/UCM28
4443.pdf
) for more information about substantial equivalence.
84
See sections 513(f)(1) and 515(a) of the FD&C Act (21 U.S.C. 360c(f)(1) and 360e(a)).
85
See section 513(f)(2) of the FD&C Act (21 U.S.C. 360c(f)(2)).
86
21 CFR 807.81(a)(3).
87
A therapeutic product company can submit the data either in a master file or in the NDA/BLA, which can
then be referenced by the IVD manufacturer in the PMA, as a means to include the clinical line data in the PMA
review while maintaining confidentiality of the data; see Section III.F.1.iv and v.
Contains Nonbinding Recommendations
Draft - Not for Implementation
35
testing sites with relevant information about the analytical or clinical testing performed at
1225
each site, and the associated IRBs (see Appendix 3). BIMO will also confirm that the line
1226
data received in the submission matches the data obtained at the testing site. Therefore,
1227
information about the location of records should be included in the submission.
1228
1229
For IVD PMAs, submission of manufacturing information for review is required, and FDA
1230
will usually conduct manufacturing inspections at the IVD manufacturing site(s). For IVD
1231
companion diagnostics, FDA will attempt to schedule inspections as early as possible in the
1232
application review process so that inspection results are available to inform the IVD review
1233
division and to allow time for the sponsor/manufacturer to address any significant inspection
1234
findings.
1235
1236
To achieve timely inspections, FDA recommends that PMA applicants use the modular PMA
1237
process for premarket submission and discuss the contents and timeline for the components
1238
of the submission with the review division prior to the submission. Submission of the
1239
manufacturing module as early as possible helps to allow sufficient time for the review
1240
division to assist the manufacturer to assure that all necessary documentation is in place
1241
ahead of scheduling the manufacturing inspection. This is particularly important when the
1242
manufacturing of the IVD companion diagnostic is done outside the U.S., as inspections in
1243
other countries may take longer to schedule.
1244
1245
iv. Master Files
1246
For various reasons, such as to address a bridging study, additional information from the
1247
therapeutic product trial that is not included in the NDA or BLA (and is therefore not
1248
accessible through a letter of authorization (see Section III.F.1.v.)) may need to be sent to the
1249
appropriate IVD review center for review. If the therapeutic product sponsor does not want
1250
its data, or a subset of the data, to be shared with the IVD sponsor (i.e., the party that would
1251
normally submit IVD data and information), the therapeutic product sponsor has the option to
1252
submit the data directly into a master file (MAF), which is accessible to the IVD review
1253
center but not accessible to the IVD sponsor. A MAF allows the therapeutic product
1254
sponsor’s proprietary information to undergo confidential review by FDA, without sharing
1255
the information with the IVD sponsor.
1256
1257
When submitted in support of a PMA, the data in a MAF will be reviewed by FDA and the
1258
MAF holder (i.e., the therapeutic product sponsor) will receive, if appropriate, a MAF
1259
deficiency letter. Additionally, with the MAF holder’s consent, the PMA applicant will
1260
receive a major deficiency letter that states a MAF deficiency letter has been sent to the MAF
1261
holder. FDA will not conduct any additional PMA review until all deficiencies, including
1262
those in the MAF, have been addressed. The MAF holder should send its response to the
1263
deficiencies to the MAF. The PMA applicant should reference the MAF when sending its
1264
own response to its major deficiencies letter. For further information about MAFs, refer to
1265
information available from the FDA website or contact CDRH or CBER.
88
1266
88
See FDA website:
http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/HowtoMarketYourDevice/PremarketSubm
issions/PremarketApprovalPMA/ucm142714.htm.
Contains Nonbinding Recommendations
Draft - Not for Implementation
36
1267
v. Letters of Authorization
1268
In most cases, the marketing authorizations of the therapeutic product and the IVD
1269
companion diagnostic are dependent on each other. Therefore, the review staff from each
1270
center assigned to review the respective applications will consult with the other center on
1271
issues that may affect the review. For this reason, the therapeutic product and IVD sponsors
1272
may need to submit letters of authorization, authorizing the other applicant to refer to the
1273
corresponding NDA, BLA or PMA (or other IVD premarket submission if applicable) in
1274
support of the other applicant’s product. See Appendix 4 for sample letters of authorization.
1275
1276
vi. Priority Review
1277
IVD companion diagnostic submissions may qualify for priority review if the criteria in 21
1278
U.S.C. 360e(d)(5) are met. Generally, CDRH and CBER have granted priority review status
1279
to IVD companion diagnostic submissions, particularly when the IVD companion diagnostic
1280
is the first-of-a-kind. The IVD companion diagnostic sponsor may formally request priority
1281
review for the IVD or FDA may grant priority review on its own initiative. FDA review staff
1282
will manage the priority review of the submission through the mechanism outlined in FDA
1283
guidance “Priority Review of Premarket Submissions for Devices.”
89
Sponsors should
1284
consider their responsibilities for priority review as described in the same document.
1285
Although the guidance indicates that FDA will take most PMAs granted priority review to an
1286
advisory panel, FDA does not intend to take IVD companion diagnostic PMAs to panel
1287
unless the scientific issues associated with the candidate IVD companion diagnostic warrant
1288
panel review. Note that the current policies of CDER and CBER for advisory committee
1289
consideration of therapeutic product applications will remain in place.
1290
1291
For therapeutic products, priority review may be granted for a product that treats a serious
1292
condition and, if approved, would provide a significant improvement in safety or
1293
effectiveness.
90
This more rapid review process may make it difficult to achieve
1294
contemporaneous marketing authorization of the associated IVD companion diagnostic.
1295
Therapeutic product sponsors should ensure that their IVD companion diagnostic sponsor
1296
partners are aware of the potential for therapeutic product priority review and are prepared to
1297
submit their PMA in a timely fashion.
1298
1299
vii. Therapeutic Products Receiving Accelerated Approval
1300
FDA may decide to grant accelerated approval of a therapeutic product, if the therapeutic
1301
product treats a serious condition, generally provides a meaningful advantage over available
1302
therapies, and demonstrates an effect that either (1) is on a surrogate endpoint that is
1303
reasonably likely to predict clinical benefit, or (2) is on a clinical endpoint that can be
1304
measured earlier than irreversible morbidity or mortality (IMM) and that is reasonably likely
1305
to predict an effect on IMM or other clinical benefit (i.e., an intermediate clinical endpoint).
91
1306
1307
89
Available at:
http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/ucm089643.htm.
90
For additional information on priority review, see note 75.
91
For additional information, see note 75.
Contains Nonbinding Recommendations
Draft - Not for Implementation
37
If the therapeutic product sponsor intends to seek accelerated approval, the clinical trial
1308
intended to support approval should be designed in a way to appropriately validate the
1309
candidate IVD companion diagnostic.
1310
1311
For drugs and biological products granted accelerated approval, postmarketing confirmatory
1312
trials have been required to verify and describe the clinical benefit. For a therapeutic product
1313
(as described in this guidance) granted accelerated approval, it is likely that the
1314
postmarketing confirmatory trial(s) will also include the IVD companion diagnostic. If
1315
labeling claims are expanded based on such studies, the applicant should consider whether
1316
the intended use of the IVD companion diagnostic will require modification. A modification
1317
to the intended use of an IVD typically requires submission of a new device application or a
1318
supplement.
92
1319
1320
2. When Contemporaneous Marketing Authorization is Not Possible
1321
As stated in the IVD companion diagnostic guidance,
93
although there is an expectation
1322
for contemporaneous marketing authorizations for the therapeutic product and its IVD
1323
companion diagnostic, FDA recognizes there may be circumstances that prevent this.
1324
FDA will resolve each situation on a case-by-case basis, taking into account the specific
1325
circumstance surrounding the use of the therapeutic product and the characteristics of the
1326
IVD companion diagnostic.
1327
1328
3. Shipment and Verification of an IVD Companion Diagnostic
1329
Prior to Marketing Authorization
1330
In most cases, a laboratory will need time to set up and verify a new IVD before it can be
1331
used for routine clinical testing. As a result, there could be a significant delay before patients
1332
could benefit from an IVD that has just received marketing authorization. For an IVD
1333
companion diagnostic, such a delay could mean patients are unable to receive the
1334
corresponding therapeutic product during this period of time, even if both products receive
1335
contemporaneous marketing authorization.
1336
1337
To ensure immediate patient access to the therapeutic product upon approval, IVD
1338
companion diagnostic manufacturers may wish to ship the IVD to laboratories for setup and
1339
verification, after its design has been finalized and clinical trials have been completed but
1340
prior to its marketing authorization.
94
As long as use of the IVD companion diagnostic is
1341
limited to setup and verification only, is not otherwise used for diagnosing patients, and
1342
otherwise meets the criteria in 21 CFR 812.2(c)(3), FDA will consider it to be an exempt
1343
92
If the IVD companion diagnostic that was originally approved with the therapeutic product is used in the
postmarket studies, the type and content of the submission will depend on the specifics of the trial, see also
Section III. F.1.ii. of this guidance.
93
See note 3.
94
Note that changes to the IVD may occur during the premarket review process (e.g., manufacturing changes,
labeling changes, or other changes), such that a laboratory may need to perform additional verification activities
with the version of the IVD companion diagnostic that receives marketing authorization.
Contains Nonbinding Recommendations
Draft - Not for Implementation
38
investigational device per the IDE regulation. Sponsors should be aware that they are still
1344
subject to:
1345
· 21 CFR 809.10(c), requiring appropriate labeling of the IVD companion diagnostic as
1346
“Investigational Use Only.” Once the IVD is authorized, the manufacturer may
1347
provide new labeling consistent with the marketing authorization.
1348
· 21 CFR 812.119, governing the disqualification of clinical investigators.
1349
Laboratories that participate in these activities are considered study sites until the
1350
IVD companion diagnostic receives marketing authorization.
1351
1352
As an IVD companion diagnostic is considered investigational prior to marketing
1353
authorization, any use for diagnosis of patients outside of the scope of an investigation
1354
conducted according to 21 CFR Part 812 is generally not permitted. FDA may inspect study
1355
sites or take other appropriate action should it obtain information that the IVD companion
1356
diagnostic is being used for diagnosis outside of the scope of the investigation. FDA
1357
recommends that manufacturers communicate with laboratories about permitted uses of the
1358
IVD companion diagnostic and maintain records documenting the laboratories that have
1359
received it. FDA recognizes that laboratories may wish to determine whether setup and
1360
verification of a particular IVD companion diagnostic is a worthwhile activity, and does not
1361
consider speculative discussions about the price of the IVD for this purpose prior to
1362
marketing authorization to be commercialization or to otherwise violate 21 CFR 812.7.
1363
G. Labeling Considerations
1364
The labeling of a therapeutic product/IVD companion diagnostic pair should be consistent.
95
1365
The IVD companion diagnostic’s labeling should specify those particular analytes (e.g., gene
1366
variants, expression patterns, protein expression) that are specified in the therapeutic product
1367
labeling. For example, if a therapeutic product is indicated for a population that has a
1368
particular spectrum of gene variants, the IVD companion diagnostic generally should be
1369
indicated for the detection of all the variants in the spectrum.
1370
1371
1. Claims for IVD Companion Diagnostics Based on Use in Trial
1372
There are several types of claims that may be generated for an IVD companion diagnostic,
1373
based on how the IVD was used
96
in the major efficacy therapeutic product trial(s). The
1374
types of claims and the trial designs that support them are discussed below.
1375
1376
95
Appropriate labeling for an IVD companion diagnostic and the corresponding therapeutic product is further
described in the guidance “In Vitro Companion Diagnostic Devices”
(
http://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/UCM26
2327.pdf
).
96
Examples of uses of IVDs include: selection of the treatment population, exclusion of patients likely to suffer
severe adverse reactions, stratification of the various trial arms to ensure balanced representation of the
treatment/control arms, and selection of dose in treatment arms.
Contains Nonbinding Recommendations
Draft - Not for Implementation
39
i. Predictive Claims
1377
Predictive claims
97
for IVD companion diagnostics should be supported by evidence that
1378
clinical benefit accrues only to, or primarily to, a population defined by the IVD result (i.e.,
1379
only test-positive or test-negative patients), or that serious adverse reactions are confined to a
1380
population defined by the IVD result. The evidence to support a claim for prediction of
1381
clinical benefit is generally derived from studies in which both test-positive and test-negative
1382
subjects are enrolled. Each test-defined subset is split and then randomized to
1383
“investigational therapy” and “control/placebo therapy” arms (e.g., Figure 1A). This type of
1384
design will demonstrate whether the IVD result is predictive of therapeutic response. It may
1385
be possible, with appropriate pre-specification of the expected treatment by test result
1386
interaction, to support predictive claims using a prospective-retrospective trial design (see
1387
Section III.D.4.). Note that the evidence to support a claim for prediction of serious adverse
1388
reactions may require different approaches from that for prediction of effectiveness, if it is
1389
considered unethical to place subjects who are considered more likely to have a serious
1390
adverse reaction in the investigational therapeutic product arm.
1391
1392
It is not possible to support prediction claims for the IVD when only test-positive or test-
1393
negative subjects are selected for enrollment in a trial because there will be no information
1394
about safety and efficacy in the population that is not treated (e.g., Figure 1B).
1395
1396
ii. Selection Claims
1397
Trial designs in which only test-positive (or test-negative) subjects are selected for
1398
enrollment in a trial (e.g., Figure 1B) typically support IVD companion diagnostic claims for
1399
patient selection. For a selection claim, if the major efficacy trial demonstrates adequate
1400
safety and effectiveness of the therapeutic product within the population selected by the IVD,
1401
the IVD is considered to be “clinically validated” in that it selected a population that benefits
1402
from the therapeutic product.
1403
1404
iii. Monitoring Claims
1405
IVD companion diagnostics for patient monitoring help select the dosage of a therapeutic
1406
product during treatment, or indicate when therapy should be modified or discontinued to
1407
avoid harm. An IVD companion diagnostic for monitoring may be required because the
1408
therapeutic product demonstrates important safety issues and/or a lack of efficacy (that
1409
presents a risk of serious harm to the patient) when administered to a patient outside of the
1410
established therapeutic window. Monitoring to determine when to discontinue therapy (e.g.,
1411
when a patient is not expected to achieve any additional benefit but could incur harm) may
1412
also be an IVD companion diagnostic claim. Trial designs to support IVD companion
1413
diagnostic monitoring claims are beyond the scope of this guidance, and FDA recommends
1414
discussing such approaches with the Agency.
1415
97
In the context of this guidance document, the term “predictive” or “prediction” indicates whether the test
result can be used to predict a patient’s response to a therapeutic product. This is distinct from the term’s use in
other contexts, such as for microbiology tests.
Contains Nonbinding Recommendations
Draft - Not for Implementation
40
H. Postmarketing Considerations
1416
Under the Food and Drug Administration Amendments Act of 2007 (FDAAA),
1417
postmarketing requirements can be used to assess a therapeutic product’s safety in a given
1418
patient population. If a therapeutic product’s use in a patient population is determined by an
1419
IVD companion diagnostic, the therapeutic product and IVD sponsors should seek input from
1420
the appropriate centers to ensure that such postmarketing clinical trials are designed to meet
1421
stated objectives.
1422
1423
For adverse reactions that occur when an IVD companion diagnostic and a therapeutic
1424
product are used together, reportable events that can be reasonably attributed only to IVD
1425
performance problems must be reported in accordance with 21 CFR Part 803, while those
1426
reportable events that are reasonably attributed only to the therapeutic product must be
1427
reported to the therapeutic product center in accordance with 21 CFR 314.80 or 600.80. For
1428
reportable events that can be attributed to both products, or when it is not clear which product
1429
may have caused the problem, report the event in accordance with both regulations.
1430
1431
Contains Nonbinding Recommendations
Draft - Not for Implementation
41
APPENDIX 1: Critical Points of the Codevelopment
1432
Process
1433
Efficient codevelopment of a therapeutic product with an IVD companion diagnostic requires
1434
coordination of the development programs of the two products, including interactions with
1435
all relevant FDA review divisions (see Figure A1).
1436
1437
Figure A1.
1438
1439
1440
Therapeutic product development typically advances through a series of clinical trial phases
1441
and includes predictable points of interaction with the FDA (e.g., specified meetings and
1442
submissions).
98
IVD development, on the other hand, is typically not linear and many
1443
analytical validation studies may take place without prior FDA involvement. In
1444
codevelopment programs, the clinical validity of the IVD is typically assessed in the
1445
therapeutic product clinical trials.
1446
1447
Sponsors of developmental or candidate IVD companion diagnostics may use the Pre-Sub
1448
program at any point during IVD development, to discuss any aspect of the development
1449
program, including the appropriateness of analytical or clinical protocols and possible
1450
regulatory pathways, among other things.
99
1451
1452
98
See FDA guidance, “Formal Meetings between the FDA and Sponsors or Applicants”
(
http://www.fda.gov/downloads/Drugs/Guidances/ucm153222.pdf).
99
More information about the Pre-Sub program can be found in the FDA guidance “Requests for Feedback on
Medical Device Submissions: The Pre-Submission Program and Meetings with Food and Drug Administration
Staff”
(
www.fda.gov/downloads/medicaldevices/deviceregulationandguidance/guidancedocuments/ucm311176.pdf
).
Contains Nonbinding Recommendations
Draft - Not for Implementation
42
The Pre-IND, End-of-Phase 1 (EOP1) and End-of-Phase 2 (EOP2) meetings for a therapeutic
1453
product are critical times to discuss plans for a therapeutic product’s development. If the
1454
therapeutic product review center determines that an analytically validated test is necessary
1455
to meet the stated objectives of the clinical trial, FDA may not allow the trial to proceed
1456
without an adequately validated test. If the IVD sponsor has not initiated interaction with the
1457
appropriate IVD review center by the time the therapeutic product sponsor holds key
1458
milestone meetings, FDA strongly recommends that the IVD sponsor do so at that time.
1459
Contains Nonbinding Recommendations
Draft - Not for Implementation
43
1460
APPENDIX 2: Subject Specimen Handling
1461
Considerations
1462
An appropriate sample acquisition plan is critical to a successful codevelopment strategy.
1463
Sponsors may find it helpful to consider resources on biospecimen reporting, such as the
1464
Biospecimen Reporting for Improved Study Quality (BRISQ) recommendations.
100
1465
1466
1. Banking Samples
1467
FDA strongly recommends that sponsors collect and bank (where analytes are stable under
1468
banking conditions) the specimens from all subjects tested for participation in the trial when
1469
possible, regardless of whether a specific IVD companion diagnostic intended for
1470
commercialization will be used in the clinical trial. There are two primary reasons for
1471
banking specimens: (1) diagnostic indications with respect to a specific therapeutic product
1472
require a correlation between the candidate IVD companion diagnostic test results with
1473
subject specimens and the subject status; and (2) analytical performance of the IVD is
1474
demonstrated with subject specimens. For these reasons and others, it is important to
1475
consider a specimen banking plan when contemplating codevelopment programs.
1476
1477
2. Sample or Analyte Specifications
1478
An ideal specimen banking plan should be structured around obtaining specimens from all
1479
subjects who are tested for possible enrollment into a marker-driven or marker-stratified
1480
therapeutic product trial, whether or not the subjects were actually enrolled, with the
1481
exception of those who were excluded from the trial due to not meeting other inclusion
1482
criteria for the trial. The availability of samples from all subjects in the ITD population
1483
allows the test developer to meet analytical performance study requirements, such as
1484
determining accuracy of the test. Analytical validation with specimens also allows for an
1485
adequate evaluation of test performance with the variables present in the major efficacy
1486
therapeutic product trial(s) and likely to be present in clinical care when the therapeutic
1487
product is approved. This includes, but is not limited to, the mode of collection (e.g.,
1488
surgical resection, core needle biopsy), anatomical sites of collection (e.g., primary,
1489
metastatic), histology and stage. Additionally, if changes are made to the test, or the CTA is
1490
not the candidate IVD companion diagnostic, the samples will need to be retested with the
1491
candidate for the purpose of assessing efficacy of the therapeutic product based on the results
1492
obtained with the test version intended for commercialization.
1493
1494
Sponsors should plan to bank both the specimen and any processed specimen (e.g., DNA
1495
extractions) used for the initial testing. The banked tissue is useful for the analytical
1496
performance studies since most performance studies should include the preanalytic steps.
1497
The processed samples, such as DNA extractions, are useful in the event that the sample
1498
needs to be retested for a demonstration of concordance between the CTA and the candidate
1499
100
Moore, HM. Biospecimen reporting for improved study quality (BRISQ). Cancer Cytopathol. 2011.
119(2):92-101.
Contains Nonbinding Recommendations
Draft - Not for Implementation
44
IVD companion diagnostic. While having large amounts of homogeneous sample from each
1500
subject is ideal, it may not be achievable, especially where the sample collection method
1501
requires invasive procedures that are not part of standard clinical care for the disease or
1502
condition in question. In their sampling plan, sponsors should plan to obtain a sufficient
1503
sample volume to perform the necessary test, plus enough overage to enable retesting one or
1504
more times (where possible and ethical).
1505
1506
3. Foreign Countries
1507
Sample banking can be complicated when samples are obtained from subjects in countries
1508
that do not typically allow specimens to leave the country of origin. In designing a sample
1509
banking plan, this possibility should be carefully considered. If it is likely that a significant
1510
number of samples from a therapeutic product trial will be inaccessible due to country-
1511
specific export limitations, sponsors may try to establish a plan to both bank samples and
1512
retest in those countries.
1513
1514
4. Informed Consent
1515
The definition of human subject includes a subject’s specimens (21 CFR 812.3(p)), and thus,
1516
informed consent applies to the use of specimens. In the U.S., to use a human specimen in an
1517
investigation, legally effective informed consent must be obtained from the subject (or his
1518
legal representative).
101,102
It is good practice to outline the uses of the subject’s sample that
1519
may reasonably be anticipated, either in the therapeutic product clinical trial consent or in a
1520
separate document dedicated to the sample collection only, even if the laws and regulations
1521
in the country of origin do not specifically require it. It is also good practice to obtain
1522
samples from subjects who are not enrolled in the trial, so that the ITD population is properly
1523
represented in the banked samples. Informed consent may also be required for these
1524
samples, e.g., if the investigational IVD will be used on the samples.
1525
1526
5. Specimen Annotation
1527
Thorough sample annotation is critical to successful development of an IVD companion
1528
diagnostic. It is very important to adequately annotate specimens with relevant information
1529
that will inform both their use in the therapeutic product trial and potential later uses.
1530
Relevant information includes factors that may affect test performance and factors that may
1531
affect the therapeutic product evaluation. The latter are typically outlined as demographics
1532
and stratification factors in the clinical trial. These factors may also be evaluated as sources
1533
of bias in the event that there are missing samples in analysis of test performance that
1534
informs therapeutic product use.
1535
1536
Subject characteristics may include:
1537
101
See 21 CFR Part 50, 21 CFR Part 812, and 21 U.S.C. 360j(g)(3)(D).
102
Currently, FDA intends to exercise enforcement discretion with respect to the informed consent
requirements, see note 101, under certain circumstances for IVD investigations using leftover human specimens
that are not individually identifiable. See FDA guidance “Informed Consent for In Vitro Diagnostic Device
Studies Using Leftover Human Specimens that are Not Individually Identifiable”
(
http://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/ucm0712
65.pdf
).
Contains Nonbinding Recommendations
Draft - Not for Implementation
45
· Disease or condition grade, stage, severity, or other standardized measures of patient
1538
status
1539
· Previously administered therapies
1540
· Study stratification factors, e.g., age, sex, race/ethnicity, tumor size, geographical
1541
location, performance status
1542
1543
Sample characteristics may include:
1544
· Type of specimen, e.g., tumor, blood, serum, urine, plasma, tissue, saliva
1545
· If tumor sample, percent tumor/stromal/necrotic proportion
1546
· Content of potential inhibitory or cross-reactive substances, e.g., melanin
1547
· Anatomical site of collection
1548
· Collection method and container type
1549
· Primary, metastatic, normal, abnormal
1550
1551
Sample handling and preliminary preparative steps may include:
1552
· Biopsy, fine needle aspirate
1553
· Formalin-fixed paraffin embedded (FFPE), frozen, centrifuged, fractionated,
1554
extracted, macrodissected, etc.
1555
· Date of collection/handling/preparation
1556
· Storage conditions (e.g., temperature) including conditions associated with shipping
1557
to laboratory
1558
1559
6. Storage
1560
When specimens are stored for later use, the sponsor should consider the stability of the
1561
analyte(s) of interest. Some analytes are labile and require special handling or storage
1562
conditions, while others are more stable and can withstand a variety of handling and storage
1563
conditions. To the degree that the stability of the analyte in the matrix of choice is not well-
1564
defined, the sponsor should perform a thorough assessment of the anticipated handling and
1565
storage conditions to ensure that conditions are selected that will allow later informative use
1566
of the samples. It is acceptable to extract or purify the analyte(s) of interest if extraction or
1567
purification (or partial purification) is required to stabilize it. In this case, complete
1568
analytical studies will necessitate that the sponsor demonstrate that the extraction or
1569
purification can be consistently carried out in a way to assure expected test performance.
1570
FDA recommends that a single, uniformly implemented method be used in any sample
1571
handling or extraction procedures, as use of more than one method may introduce variables
1572
into the test performance that cannot be quantified.
1573
1574
Contains Nonbinding Recommendations
Draft - Not for Implementation
46
APPENDIX 3: BIMO Information to Submit in a PMA
1575
To facilitate the CDRH/CBER BIMO inspection of investigational testing sites in clinical
1576
trials, it is recommended that PMA applicants submit the following information, stratified by
1577
the type of study (analytical validation vs. clinical validation) from each of the testing sites:
1578
· Analytical studies for PMA (information provided by site for each study)
1579
o
Site information (including name, street address, city, state, zip code, name of
1580
contact, and telephone number)
1581
o
Location of source documents
1582
o
Statement of location of line data (e.g., at the site or with sponsor)
1583
o
Patient/subject information, unless the studies were conducted with leftover
1584
specimens that are not individually identifiable
1585
o
Sample Data Collection/Case Report Forms
1586
o
Investigator Agreements
1587
o
Conflict of Interest/ Financial Disclosure
1588
o
Informed Consent Document(s), unless the studies were conducted with
1589
leftover specimens that are not individually identifiable
1590
o
Protocol Deviations
1591
o
IRB information
1592
o
Monitoring Plan
1593
o
Line Listings (stratified by site and then subject)
1594
· Clinical testing by site (e.g., centralized testing for enrollment)
1595
o
Site information (including name, street address, city, state, zip code, name of
1596
contact, and telephone number)
1597
o
Statement of location of line data (e.g., at the site or with sponsor)
1598
o
Patient/subject information, if needed
1599
o
Location of source documents
1600
o
Case Report Forms
1601
o
Investigator Agreements
1602
o
Conflict of Interest/Financial Disclosure
1603
o
Informed Consent Document(s)
1604
o
Protocol Deviations
1605
o
Line Listings (stratified by site and then subject)
1606
Contains Nonbinding Recommendations
Draft - Not for Implementation
47
1607
APPENDIX 4: Letters of Authorization
1608
For efficient review of a therapeutic product and its corresponding IVD companion
1609
diagnostic, the therapeutic product sponsor and the IVD sponsor should send letters of
1610
authorization to FDA that authorize the other sponsor to cross-reference the premarket
1611
submission or incorporate the relevant content by reference.
1612
1613
The center reviewing the IVD (CDRH/CBER) needs permission from the therapeutic
1614
product sponsor to rely on the data in the NDA/BLA to support the PMA (or other device
1615
premarket submission if applicable). The letter authorizing this cross-reference should be
1616
sent to the Document Control Center of the center reviewing the IVD (CDRH/CBER) to
1617
the attention of the IVD reviewer. Also, the center reviewing the therapeutic product
1618
(CDER/CBER) needs permission from the IVD sponsor to rely on the data in the PMA
1619
(or other device premarket submission if applicable) to support the NDA/BLA. The letter
1620
authorizing this cross reference should be sent to the electronic gateway of the center
1621
reviewing the therapeutic product (CDER/CBER) to the attention of the therapeutic
1622
product reviewer.
1623
1624
Letters should clearly specify the product name, sponsor name and submission number(s)
1625
(e.g., PMA, BLA, or NDA numbers). Authorizing FDA to rely on information in the
1626
corresponding product premarket submission does not authorize FDA to share that
1627
information with the other company; the information remains confidential in accordance
1628
with the applicable laws.
103
1629
1630
Two examples of letters of authorization are provided below.
1631
1632
Example 1: An IVD sponsor authorizing CDER to refer to a PMA in support of an
1633
NDA
1634
1635
[IVD Sponsor Name]
1636
[Address]
1637
1638
[Date]
1639
1640
[CDER Reviewer]
1641
[Address]
1642
1643
Re: Authorization Letter to Cross Reference [PMA#] [IVD Name]
1644
1645
This letter authorizes CDER to refer to [IVD Sponsor Name]’s PMA [PMA number] for
1646
103
For information on FDA treatment of confidential information and what constitutes trade secret, confidential
commercial or financial information, and private personal identifier information, see the FDA regulations
implementing the Freedom of Information (FOI) Act in 21 CFR Part 20. See also FDA’s FOI web page at
http://www.fda.gov/ RegulatoryInformation/foi/ default.htm
.
Contains Nonbinding Recommendations
Draft - Not for Implementation
48
[IVD Name] in support of [Drug Sponsor Name]’s NDA application [NDA number] for
1647
[Drug Name and Indication] and [Drug Name and Indication 2 (if applicable)].
1648
1649
By copy of this letter, we authorize [Drug Sponsor Name] to incorporate information
1650
contained in the PMA by reference into their NDA submission(s) as necessary. (Optional
1651
if the IVD sponsor wishes to allow the drug sponsor to incorporate IVD information into
1652
the NDA submission.)
1653
1654
Please contact [Name] at [Phone Number] or [E-mail] with questions.
1655
1656
[Signature]
1657
[Name]
1658
[Title]
1659
1660
Example 2: A drug sponsor authorizing CDRH to refer to an NDA(s) in support of
1661
a PMA for an IVD companion diagnostic
1662
1663
[Drug Sponsor Name]
1664
[Address]
1665
1666
[Date]
1667
1668
[CDRH Reviewer]
1669
[Address]
1670
1671
Re: Authorization Letter to Cross Reference [NDA #] [Drug Name]
1672
1673
This letter authorizes the Center for Devices and Radiological Health to refer to [Drug
1674
Sponsor Name]’s New Drug Application [NDA number] for [Drug Name] in support of
1675
[IVD Sponsor Name]’s PMA [PMA number] for [IVD Name], which is intended to be
1676
used for [Intended Use].
1677
1678
By copy of this letter, we authorize [IVD Sponsor Name] to incorporate information
1679
contained in the NDA(s) by reference into their PMA submission as necessary. (Optional
1680
if the drug sponsor wishes to allow the IVD sponsor to incorporate drug information into
1681
the PMA submission.)
1682
1683
Please contact [Name] at [Phone Number] or [E-mail] with questions.
1684
1685
[Signature]
1686
[Name]
1687
[Title]
1688
Document Outline - General
- Regulation of Investigational IVDs and Therapeutic Products
- Risk Assessment and IDE Requirements
- Submission of Investigational IVD Information Related to Investigational Drugs or Biological Products
- IDE Applications for Investigational IVDs in Codevelopment Trials
- Planning Ahead for IVD Validation in Potential Codevelopment Programs
- Expectation for Analytical Validation Prior to Investigational IVD Use in Therapeutic Product Trials
- New Intended Uses for IVDs
- IVD Prototypes in Early-Phase Therapeutic Product Clinical Trials
- Using Research Use Only Components as Part of a Test System
- Prescreening for Eligibility for Therapeutic Product Clinical Trials
- Preanalytic Procedures and Testing Protocols
- Planning Ahead for Analytical Validation Studies
- Therapeutic Product Clinical Trial Design Considerations
- General Considerations for Early Therapeutic Product Development
- General Considerations for Late Therapeutic Product Development
- Prognostic and Predictive Markers
- Prospective-Retrospective Approaches
- Considerations for Identifying Intended Populations
- Considerations for IVD Development in Late Therapeutic Product Development
- Training Samples Sets versus Validation Samples Sets
- Effect of Changes to the Test Design
- IVD Bridging Studies
- Special Protocol Assessments
- Planning for Contemporaneous Marketing Authorizations
- Coordinating Review Timelines
- When Contemporaneous Marketing Authorization is Not Possible
- Shipment and Verification of an IVD Companion Diagnostic Prior to Marketing Authorization
- Labeling Considerations
- Claims for IVD Companion Diagnostics Based on Use in Trial
- Postmarketing Considerations
- APPENDIX 1: Critical Points of the Codevelopment Process
- APPENDIX 2: Subject Specimen Handling Considerations
- APPENDIX 3: BIMO Information to Submit in a PMA
- APPENDIX 4: Letters of Authorization
Dostları ilə paylaş: |