Metered Dose Inhaler (MDI) and Dry Powder Inhaler (DPI) Drug Products--Quality Considerations

Administrative draft guidance, not enacted legislation. This captured publisher edition preserves the listed document and attachments; later developments are not independently consolidated.

15087 Dft
Metered Dose Inhaler
(MDI) and Dry Powder
Inhaler (DPI) Products -
Quality Considerations
Guidance for Industry
DRAFT GUIDANCE
This guidance document is being distributed for comment purposes only.
Comments and suggestions regarding this draft document should be submitted within 60 days of
publication in the Federal Register of the notice announcing the availability of the draft
guidance. Submit electronic comments to https://www.regulations.gov. Submit written
comments to the Division of Dockets Staff (HFA-305), Food and Drug Administration, 5630
Fishers Lane, rm. 1061, Rockville, MD 20852. All comments should be identified with the
docket number listed in the notice of availability that publishes in the Federal Register.
For questions regarding this draft document contact (CDER) Richard Lostritto 301-796-1697.
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
April 2018
Pharmaceutical Quality/CMC
Revision 1

Metered Dose Inhaler
(MDI) and Dry Powder
Inhaler (DPI) Products -
Quality Considerations
Guidance for Industry
Additional copies are available from:
Office of Communications, Division of Drug Information
Center for Drug Evaluation and Research
Food and Drug Administration
10001 New Hampshire Ave., Hillandale Bldg., 4 th Floor
Silver Spring, MD 20993-0002
Phone: 855-543-3784 or 301-796-3400; Fax: 301-431-6353
Email: druginfo@fda.hhs.gov
https://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
April 2018
Pharmaceutical Quality/CMC
Revision 1

Contains Nonbinding Recommendations
Draft — Not for Implementation
TABLE OF CONTENTS
I. 	INTRODUCTION............................................................................................................. 1
II. 	BACKGROUND ............................................................................................................... 2
A. General ............................................................................................................................................... 2
B. Regulatory Status ............................................................................................................................... 3
III. 	MDI AND DPI PRODUCT DEVELOPMENT .............................................................. 4
A. 	Quality Target Product Profile (QTPP)....................................................................................... 4
B. 	Critical Quality Attributes (CQAs) .............................................................................................. 4
1. MDI and DPI Products .................................................................................................................... 4
2. Drug Substance ................................................................................................................................ 5
3. Excipients ......................................................................................................................................... 6
4. Container Closure System (Including the Device Constituent Part) for MDIs................................ 6
5. Container Closure System (Including the Device Constituent Part) for DPIs ................................ 7
C. 	Product and Process Development ............................................................................................... 7
1. Product Development....................................................................................................................... 8
2. Process Development ..................................................................................................................... 10
D. 	Development of Control Strategy ............................................................................................... 12
1. Controls for Incoming Materials ................................................................................................... 12
2. In-process Measurements and Monitoring .................................................................................... 13
3. Release Testing of the MDI and DPI Product ............................................................................... 14
IV. 	INFORMATION TO BE SUBMITTED IN AN APPLICATION ............................. 15
A. 	Information on the Drug Substance ........................................................................................... 15
B. 	Description and Composition (P1) ............................................................................................. 16
1. MDIs .............................................................................................................................................. 16
2. DPIs ............................................................................................................................................... 17
C. 	Pharmaceutical Development (P2) ............................................................................................. 17
D. 	Manufacture (P3) ......................................................................................................................... 18
E. 	Control of Excipients (P4) ........................................................................................................... 19
F. 	Control of MDI and DPI Product (P5) ...................................................................................... 22
1. Description..................................................................................................................................... 23
2. Valve Delivery................................................................................................................................ 24
3. Delivered Dosage Uniformity (DDU) ............................................................................................ 24
4. Uniformity of Dosage Units ........................................................................................................... 24
5. Aerodynamic Particle Size Distribution (APSD) ........................................................................... 25
6. Spray Pattern ................................................................................................................................. 25
7. Foreign Particulates ...................................................................................................................... 26
8. Microbial Limits ............................................................................................................................ 26
G. 	Reference Standards or Materials (P6)...................................................................................... 26
H. 	Container and Closure System (P7) ........................................................................................... 26

Contains Nonbinding Recommendations
Draft — Not for Implementation
I. 	Stability (P8) ................................................................................................................................. 28
J. 	Labeling ........................................................................................................................................ 29
1. MDIs .............................................................................................................................................. 30
2. DPIs ............................................................................................................................................... 34
V. 	APPENDIX ...................................................................................................................... 38
A. 	Tables ............................................................................................................................................ 38
B. 	MDI and DPI Product Characterization Studies (P2) .............................................................. 39
1. General Considerations for Significant Change............................................................................ 40
2. Recommendations for Specific Characterization Studies .............................................................. 40
C. 	Approaches to Evaluating Delivered Dose Uniformity (DDU) ................................................ 43
1. Parametric Tolerance Interval Testing (PTIT) .............................................................................. 43
2. Counting Test ................................................................................................................................. 45

Contains Nonbinding Recommendations
Draft — Not for Implementation
1
Metered Dose Inhaler (MDI) and Dry Powder	1
Inhaler (DPI) Products – Quality Considerations	2
Guidance for Industry1
3
4
5
This draft guidance, when finalized, will represent the current thinking of the Food and Drug	6
Administration (FDA or Agency) on this topic. It does not establish any rights for any person and is not	7
binding on FDA or the public. You can use an alternative approach if it satisfies the requirements of the	8
applicable statutes and regulations. To discuss an alternative approach, contact the FDA staff responsible	9
for this guidance as listed on the title page.	10
11
12
13
I. 	INTRODUCTION	14
15
The purpose of this guidance is to provide recommendations to industry on the development and	16
manufacture of inhalation aerosols (also known as metered dose inhalers (or MDIs)) and	17
inhalation powders (also known as dry powder inhalers (or DPIs)). The recommendations in this	18
guidance can apply to MDI and DPI products intended for local or systemic effect.	19
20
This guidance describes points to consider to ensure product quality and performance for MDIs	21
and DPIs. It describes chemistry, manufacturing, and controls (CMC) information recommended	22
for inclusion in new drug applications (NDAs) and abbreviated new drug applications (ANDAs);	23
however, the principles are applicable to products used during clinical trials, and over the	24
product lifecycle as well. It also provides recommendations on certain aspects of labeling for	25
NDA and ANDA MDI and DPI products.	26
27
This guidance does not discuss aqueous-based nasal spray drug products and inhalation solution,	28
suspension, and spray drug products, or the manufacture of drug substances. However, some of	29
the principles of this guidance may be applicable to nasal delivery products. Also, this guidance	30
does not discuss considerations for when an MDI or DPI includes electronic components,	31
software, or novel inhaler components that might affect the performance or reliability of the	32
product. The applicant should refer to the applicable requirements and recommendations outlined	33
in the appropriate regulations and guidances, respectively, from the Center for Devices and	34
Radiological Health (CDRH).	35
36
FDA previously published a draft guidance on this topic on November 13, 1998.2 The present	37
guidance is a revision of the previous draft, updated to reflect current standards and requirements	38
1 This guidance has been prepared by the Office of Pharmaceutical Quality in the Center for Drug Evaluation and
Research, in collaboration with the Center for Devices and Radiological Health, at the Food and Drug
Administration.
2 We update guidances periodically. To make sure you have the most recent version of a guidance, check the FDA
Drugs guidance web page at
https://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm.

Contains Nonbinding Recommendations
Draft — Not for Implementation
2
to enhance understanding of appropriate development approaches for these products consistent	39
with the quality by design (QbD) paradigm.	40
41
In general, FDA’s guidance documents do not establish legally enforceable responsibilities.	42
Instead, guidances describe the Agency’s current thinking on a topic and should be viewed only	43
as recommendations, unless specific regulatory or statutory requirements are cited. The use of	44
the word should in Agency guidances means that something is suggested or recommended, but	45
not required.	46
47
48
II. 	BACKGROUND	49
50
A. General	51
52
MDIs and DPIs are products that deliver one or more drug substances to the site of action	53
through the inhalation route. Both types of products are used to treat lung diseases characterized	54
by obstruction of airflow and shortness of breath, including asthma and chronic obstructive	55
pulmonary disease (COPD), as well as respiratory infections and cystic fibrosis. The inhalation	56
route offers further potential for systemic drug delivery.	57
58
MDI products consist of a drug formulation (the drug constituent part) and a container closure	59
system. An MDI drug formulation contains the drug substance(s), either dissolved or suspended,	60
in a (1) propellant, (2) mixture of propellants, or (3) mixture of solvents, propellants, and/or	61
other excipients. An MDI container closure system consists of the device constituent part (i.e.,	62
the canister, the actuator, the metering valve), and any additional features (e.g., integrated spacer,	63
integrated dose counter), as well as protective secondary packaging (e.g., an overwrap). MDI	64
products use energy stored in a liquefied gas propellant under pressure for generating aerosols	65
suitable for pulmonary drug delivery.	66
67
DPI products also consist of a drug formulation (the drug constituent part) and a container	68
closure system. However, the designs of DPI products differ considerably from those for MDI	69
products. A DPI drug formulation contains the drug substance and excipients including a drug	70
carrier (e.g., lactose). A DPI container closure system consists of the device constituent part and	71
any protective secondary packaging (e.g., an overwrap). Current designs of DPI products	72
include pre-metered and device-metered DPIs, either of which can be driven by a patient’s	73
inspiration alone (passive) or with power-assistance of some type (active) for production of drug	74
particles intended for inhalation.	75
76
Pre-metered DPIs contain previously measured amounts of drug formulation in	77
individual containers (e.g., capsules, blisters, cartridges, dosing discs) that are inserted	78
into the device constituent part during manufacturing or by the patient before use. The	79
pre-metered dose can be inhaled directly or it can be transferred to a chamber before	80
being inhaled by the patient.	81
82

Contains Nonbinding Recommendations
Draft — Not for Implementation
3
Device-metered DPIs have an internal reservoir containing sufficient quantity of	83
formulation for multiple doses that are metered by the device constituent part during use	84
by the patient.	85
86
The performance of MDI and DPI products depends on many key aspects of the drug	87
formulation, container closure system (including the device constituent part), manufacturing, and	88
patient handling. Product and process understanding is therefore critical to: (1) the development	89
and manufacture of these products, (2) the maintenance of product quality and performance	90
through the expiration date under patient use conditions, and (3) the maintenance of product	91
quality and performance over the product life cycle, including continual improvement.	92
93
B. Regulatory Status	94
95
MDIs and DPIs are combination products (see 21 CFR 3.2(e)).3 As drug-device combination	96
products, they are subject to the current good manufacturing practice (CGMP) requirements for	97
drugs and devices (see 21 CFR part 4).4 Further information about the CGMP requirements for	98
combination products is available in the FDA guidance for industry and FDA staff Current Good	99
Manufacturing Practice Requirements for Combination Products, including an explanation of a	100
streamlined approach for demonstrating compliance with both drug and device CGMP	101
requirements.	102
103
In particular, design controls (21 CFR 820.30) apply to any combination product that includes a	104
device constituent part that is subject to them, including all MDIs and DPIs.5 Essentially, design	105
control activities confirm that there are no negative interactions between constituent parts, and	106
assure that their combined use results in a combination product that is safe and effective and	107
performs as expected. Guidance for industry on pharmaceutical development addresses product	108
design and development procedures, reflecting quality by design principles.6 While quality by	109
design and design controls share similar characteristics and goals, the device Quality System	110
regulation (21 CFR 820) includes specific requirements for design development that	111
manufacturers must satisfy.7	
112
113
It may be possible to leverage many aspects of pharmaceutical development as described in ICH	114
Q8(R2) to achieve compliance with design controls. For example, the Quality Target Product	115
3 A combination product is composed of two or more of the three types of medical products (i.e., drug, device, and
biological product), that are either physically, chemically, or otherwise combined into a “single-entity;” “co-
packaged” together; or under certain circumstances distributed separately to be used together as a “cross-labeled”
combination product. See 21 CFR 3.2(e).
4 See 21 CFR part 4 available at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/cfrsearch.cfm?cfrpart=4.
5 For single-entity and co-packaged combination products, design control requirements apply to the development of
the combination product as a whole. For cross-labeled combination product, design control requirements apply only
to the device constituent part but should ensure the safety and effectiveness of the device when used with the other
constituent part(s) of the combination product.
6 See FDA guidance for industry Q8(R2) Pharmaceutical Development, ICH.
7 For example, requirements under 21 CFR 820 for design control, purchasing controls, management responsibility
and corrective and preventive action must be met. See Current Good Manufacturing Requirements for Combination
Products at: https://www.fda.gov/RegulatoryInformation/Guidances/ucm126198.htm for additional information
regarding options for complying with the requirements of 21 CFR 820 for a combination product.

Contains Nonbinding Recommendations
Draft — Not for Implementation
4
Profile (QTPP) (see section III.A. below) is similar to “design inputs” (21 CFR 820.30(c)),	116
which ensure that design requirements are appropriate to address the intended use of the product.	117
Further, studies conducted to verify that the critical quality attributes (CQAs) are met in the	118
finished product may also address requirements for design “verification” and “validation” (21	119
CFR 820.30(f), (g)), which ensure that the product’s “design outputs” (21 CFR 820.30(d)) result	120
in a product that safely and effectively meets user needs and achieves its intended effects.8	
121
122
MDI and DPI applicants must ensure that their development and manufacturing procedures and	123
documentation satisfy all regulatory requirements applicable to their combination product,	124
including for design control (some of which may be satisfied by following Q8(R2) as previously	125
noted). This guidance offers recommendations for how to pursue product development and	126
manufacture in a compliant manner, generally using concepts and terminology familiar to drug	127
sponsors and manufacturers to do so.	128
129
130
III. 	MDI and DPI PRODUCT DEVELOPMENT	131
132
A. 	Quality Target Product Profile (QTPP)	133
134
Prior to the development of an MDI or DPI, the applicant should establish the desired quality	135
target product profile (QTPP). The QTPP is a prospective summary of the quality characteristics	136
of a drug product, and in this case, the combination product, that ideally will be achieved to	137
ensure the desired quality, taking into account safety and efficacy of the MDI or DPI (ICH	138
Q8(R2)).9 Examples of QTPP elements for MDIs and DPIs include: proposed dosage form and	139
delivery system, strength (e.g., targeted metered dose for DPIs, targeted delivered dose for	140
MDIs), purity, stability, and aerodynamic performance.	141
142
B. 	Critical Quality Attributes (CQAs)	143
144
1. 	MDI and DPI Products	145
146
Early in the development process of an MDI or DPI, the applicant should develop a list of	147
potential CQAs for the combination product. A CQA is a physical, chemical, biological, or	148
microbiological property or characteristic that should be within an appropriate limit, range, or	149
distribution to ensure the desired product quality (ICH Q8(R2)). Those aspects of the design of	150
the combination product that are essential for proper functioning of the product are also	151
considered part of the required design output (21 CFR 820.30(d)). Knowledge of the QTPP for	152
the product, in combination with information from prior knowledge, risk assessments, and/or	153
experimentation, can be used to develop the list of product CQAs. The list of product CQAs can	154
be modified as product development progresses and new knowledge is gained. CQAs for the	155
8 Additional requirements for design control include preparation of a design plan (21 CFR 820.30(b)) and holding
review meetings with specified personnel in attendance (21 CFR 820.30(e)). See Current Good Manufacturing
Requirements for Combination Products for additional information regarding design control requirements for
combination products and other CGMP requirements for combination products that include a device constituent part.
9 See FDA guidance for industry Q8(R2) Pharmaceutical Development, ICH.

Contains Nonbinding Recommendations
Draft — Not for Implementation
5
drug substance(s), excipients, and container closure system (including the device constituent	156
part) should also be developed (see below).	157
158
For MDIs, potential product CQAs typically include assay, impurities and degradants, delivered	159
dose, aerodynamic particle size distribution (APSD), spray pattern, leachables, alcohol/excipient	160
content, foreign particulate matter, moisture content, net content (drug substance and excipients),	161
microbial load and device constituent part characteristics such as component dimensions and	162
valve delivery (shot weight). The force and distance necessary to advance the dose counter10 and	163
the product actuation force (force to deliver the drug from the device constituent part) are CQAs.	164
If the MDI product is actuated by the patient’s inhalation, the air flow necessary to actuate the	165
device for drug release can be considered a CQA.	166
167
For DPIs, potential product CQAs typically include assay, impurities and degradants, delivered	168
dose, APSD, volatile/semi-volatile leachables content, foreign particulate matter, moisture	169
content, net content, microbial load, and device constituent part characteristics such as specific	170
resistance to air flow.	171
172
Each CQA, either alone or in concert with other CQAs, should relate to one or more elements of	173
the product QTPP. Some of the elements of the QTPP can be related to CQAs of the device	174
constituent part as well as to CQAs of the product formulation. For example:	175
176
• 	Delivered drug purity is usually related to the following CQAs: impurities and	177
degradants of the drug substance and excipients, foreign particulate matter, and	178
amount of leachables (e.g., from the device constituent part, container components, or	179
manufacturing environment).	180
181
• 	Targeted delivered dose (product strength) for MDIs is usually related to the	182
following CQAs: assay, metered dose, and net content.	183
184
• 	Aerodynamic performance for MDIs is usually related to the following CQAs:	185
delivered dose, APSD, spray pattern, moisture content, net content, device constituent	186
part CQAs, and drug substance CQAs.	187
188
• 	Targeted metered dose in a device-metered DPI is usually related to the following	189
CQAs: the device constituent part CQAs (e.g., dimensions of metering components)	190
and the physicochemical properties of the formulation.	191
192
Additional relationships between QTPP elements and CQAs for MDIs and DPIs are shown in	193
Table A, Table B, and Table C in the Appendix, section V.A.	194
195
2. 	Drug Substance	196
197
The physical, chemical, and microbiological properties of the drug substance that should be	198
within an appropriate limit, range, or distribution to ensure the desired product quality are	199
10 See FDA guidance for industry Integration of Dose-Counting Mechanisms into MDI Drug Products.

Contains Nonbinding Recommendations
Draft — Not for Implementation
6
considered CQAs of the drug substance. These should be identified and characterized early in	200
development. Additional characterization can take place throughout the development and life	201
cycle of the product.	202
203
For drug substances used in MDIs or DPIs, potential CQAs can include assay, particle size	204
distribution (PSD), moisture content, bulk density, flow properties, morphic form (e.g.,	205
amorphous, crystalline, hydrate), morphology of drug particles (e.g., shape, crystal habit, texture,	206
surface area, rugosity), residual solvent content, and impurities.	207
208
3. 	Excipients	209
210
The physical, chemical, and microbiological properties of the excipients that should be within an	211
appropriate limit, range, or distribution to ensure the desired product quality are considered	212
CQAs of the excipients. These should be identified and characterized early in development.	213
Prior knowledge can be particularly useful in identifying the CQAs of excipients since many	214
excipients are already used in similar products.	215
216
The potential impact of an excipient on product quality can depend on the intrinsic	217
characteristics and properties of the excipient chosen, and the amount of the excipient used in the	218
formulation. Examples of potential CQAs for excipients used in MDIs or DPIs can include:	219
assay, boiling point and vapor pressure, moisture content, density, impurity profile, particle	220
morphology (e.g., shape, crystal habit, texture, surface area, rugosity), flow properties,	221
amorphous content, microbial limits, pyrogens or bacterial endotoxins, and PSD.	222
223
4. 	Container Closure System (Including the Device Constituent Part) for MDIs	224
225
The container closure system for an MDI consists of the device constituent part (i.e., canister, the	226
actuator, the metering valve), including any additional features (e.g., integrated spacer, integrated	227
dose counter). It can also include protective secondary packaging. Critical device constituent	228
part components are those that may come into contact with the formulation or the patient, or are	229
necessary for device function.	230
231
The materials used to fabricate the MDI device constituent part may come into direct contact	232
with either the formulation or the patient, thereby potentially affecting product safety and	233
performance. For example, due to the presence of organic propellant/vehicle in MDI	234
formulations, leaching of compounds from the valve and/or canister components of the container	235
closure system into the formulation can occur, which is a potential safety or effectiveness	236
concern. Device constituent part materials also have the potential to affect the aerodynamic	237
performance of the MDI product. For example, delivered dose and APSD of the MDI product	238
can be affected by the surface properties of the device constituent part and/or its components.	239
Therefore, the properties of materials used in the fabrication of the device constituent part and	240
the quantitative compositions after fabrication should be considered CQAs.	241
242
The device constituent part (e.g., actuator orifice, mouthpiece, metering chamber) has an	243
important role in generating aerosol particles, determining the aerosol characteristics, and	244
controlling the amount of medication available to the patient. For instance, the actuator orifice	245

Contains Nonbinding Recommendations
Draft — Not for Implementation
7
size can affect the APSD, spray velocity, plume geometry, and spray pattern. Thus, the	246
dimensions of the device constituent part can be considered CQAs.	247
248
The secondary packaging (e.g., foil pouch) for MDIs can provide additional protection to the	249
product from humidity. Therefore, when such additional protection is important, the material	250
properties of the secondary packaging can be considered CQAs.	251
252
5. 	Container Closure System (Including the Device Constituent Part) for DPIs	253
254
The container closure system for a DPI consists of the device constituent part and protective	255
secondary packaging (e.g., overwrap, carton).	256
257
As with MDIs, the materials used to fabricate the DPI device constituent part may come in direct	258
contact with either the formulation or the patient, thereby potentially affecting product safety and	259
performance. For instance, drug particle-surface interactions, such as adhesion of drug onto	260
mouthpiece surfaces, can affect the delivered dose and APSD. Therefore, the properties of	261
materials used in the critical device constituent part components are important and the	262
quantitative compositions of the critical device constituent part components after molding should	263
be considered CQAs. Critical device constituent part components are those that may come into	264
contact with the formulation or the patient or are necessary for device function.	265
266
A DPI device constituent part acts as the delivery system of the drug. The design, geometry, and	267
dimensions of the device constituent part can influence the device resistance, air flow, shear, and	268
turbulence generated within the device constituent part, and thus the drug delivery of a DPI	269
product. Therefore, these device constituent part attributes can be considered CQAs.	270
271
The secondary packaging (e.g., foil pouch) for DPIs can provide additional protection to the	272
product from humidity. Therefore, when such additional protection is important, the material	273
properties of the secondary packaging can be considered CQAs.	274
275
C. 	Product and Process Development	276
277
Development of an MDI or DPI product should involve consideration of aspects such as aerosol	278
delivery characteristics, portability, ease of use, device constituent part robustness, inclusion of a	279
dose counter, appropriateness of a lockout, cleaning needs, and suitability to the patient	280
population.	281
282
The Agency recommends that applicants use prior knowledge specific to their formulation,	283
manufacturing process, and device constituent part design to identify QTPP, CQAs, and potential	284
risks to the product, and then initiate product and process development to define a control	285
strategy that eliminates or mitigates the risks. Applicants should consider using risk assessment	286
tools such as those listed in ICH Q911 or ISO 14971 Risk Management – Medical Devices12 (e.g.,	287
Failure Modes and Effects Analysis (FMEA), Failure Modes, Effects, and Criticality Analysis	288
11 See FDA guidance for industry Q9 Quality Risk Management, ICH.
12 For additional information on risk management for combination products, see Current Good Manufacturing
Requirements for Combination Products. See also ISO 14971 Risk Management – Medical Devices.

Contains Nonbinding Recommendations
Draft — Not for Implementation
8
(FEMCA), Fault Tree Analysis (FTA), Ishikawa diagram) starting from early product	289
development to identify factors (e.g., material attributes, process parameters) which have the	290
potential to impact product quality. The identified factors can be further studied (e.g.,	291
experimentally, by modeling) to define an appropriate control strategy that assures that the	292
manufacturing process consistently produces product of the desired quality.	293
294
Examples of some of the factors the applicant should consider, to understand potential impacts	295
on MDI or DPI product CQAs, include the following:	296
297
• 	Physiochemical properties of the drug substance(s) and excipients and their	298
interactions (e.g., densities, amorphous or crystalline forms, flow properties, adhesive	299
and cohesive properties).	300
301
• 	Lot-to-lot variability of drug substance and excipient properties (e.g., PSD, moisture	302
content, impurity profiles, surface morphology) and device constituent part	303
composition and properties (e.g., surface contamination, leachables content).	304
305
• 	Interaction of two or more drug substances when co-formulated.	306
307
• 	Potential for microbial growth.	308
309
Risk assessment and process development experiments can lead to an understanding of	310
univariate and multivariate relationships between material attributes and process parameters and	311
how they affect MDI or DPI CQAs. Experimentation and modeling can also help identify	312
appropriate ranges for these variables, within which consistent product quality can be achieved.	313
Identification of appropriate ranges can facilitate scale-up and technology transfer. Multivariate	314
combinations of appropriate ranges for material attributes and process parameters also can be	315
included in a design space.	316
317
Another factor to consider concerns the stage of development when pivotal clinical trials (i.e.,	318
phase 2b, phase 3) are conducted. Dose-ranging studies are considered pivotal trials, and the to-	319
be-marketed MDI should be used during dose-ranging studies to avoid potential therapeutic	320
differences. If an applicant completes optimization of the MDI or DPI product and	321
manufacturing process only after the pivotal clinical trials have been completed, the applicant	322
should consider establishing a relationship between the in vitro characterization of the product	323
and its in vivo performance. In the absence of such a relationship, additional in vivo studies	324
(e.g., clinical studies) might be warranted to determine whether the product manufactured for	325
clinical trials and the product proposed for commercial distribution have the same therapeutic	326
effect.	327
328
1. 	Product Development	329
330
a. 	MDIs	331
332
The following are examples of potential design and development issues that should typically be	333
considered during the development of an MDI:	334

Contains Nonbinding Recommendations
Draft — Not for Implementation
9
335
• 	The selection or design of the device constituent part (canister, valve components,	336
actuator, and dose counter) is generally informed by prior knowledge or experience,	337
and can be optimized during development as early as feasible and should be	338
completed prior to phase III study of the combination product if possible.	339
340
• 	The target fill volume of an MDI is usually established based on the number of	341
actuations required from the product, delivered dose, concentration of the drug	342
substance in the formulation, and metered volume. Unavoidable leakage of the	343
propellant over the shelf-life and the number of actuations required for priming	344
during testing and use should also be factored into the fill volume. Fill volume,	345
formulation homogeneity (for suspensions) and concentration, and fill weight are	346
likely to have a significant impact on the delivered dose of the product throughout the	347
life of the unit. The internal pressure of the device constituent part and vaporization	348
rate of the aerosol produced upon actuation are determined primarily by the properties	349
and amount of propellant(s) and cosolvent(s), because these constitute the majority of	350
the MDI formulation.	351
352
• 	For suspension based MDIs, the potential for settling, creaming, or aggregation of the	353
drug substance can be minimized if the drug substance and the propellants have	354
similar densities.	355
356
• 	A non-uniform dispersion of drug substance can also result from adhesion of the	357
suspended drug particles to various components of the device constituent part (e.g.,	358
valves, canister). This adhesion can contribute to changes in delivered dose and	359
APSD.	360
361
• 	Solution-based MDIs generally have better delivered dose uniformity (DDU)	362
compared to suspension based MDIs, but they may have more degradants, since the	363
drug substance is completely dissolved and is more susceptible to degradation	364
reactions.	365
366
• 	Organic cosolvents, which are often used to enhance the solubility of the drug	367
substance, may have the potential to solubilize the components of the device	368
constituent part. Thus, it is prudent to employ materials of construction in the device	369
constituent part that reduce the possibility of leachables in the product (e.g., plastics	370
and coatings less likely to be solubilized in the liquid phase of the formulation, pre-	371
extracted elastomers).	372
373
b. 	DPIs	374
375
The following are examples of potential design and development issues that should typically be	376
considered during the selection and development of a DPI:	377
378
• 	Carriers such as lactose can promote uniformity and flowability of a blend during	379
manufacturing. Carriers can also enhance the reproducibility of the metered,	380

Contains Nonbinding Recommendations
Draft — Not for Implementation
10
delivered, and fine particle dose of the DPI product (by reducing agglomeration of the	381
drug substance).	382
383
• 	Properties that can be important to consider for selection of carriers during product	384
development include: ratio of drug substance to excipient, physical and chemical	385
compatibility, and PSD. Interparticulate interactions between the drug substance and	386
excipients and with the container closure/device constituent part at a microscopic	387
level (e.g., cohesive and adhesive properties, surface activity, specific surface area,	388
static charge properties of the formulation) can also be important. These properties	389
and interactions can affect, for example, blend uniformity, powder flow, and	390
delivered dose.	391
392
• 	The stability of the formulation can be affected by ambient humidity. For example,	393
exposing hygroscopic excipients to moisture can result in a decrease in the fine	394
particle dose of the drug substance. If moisture ingress into the device constituent	395
part affects product performance, additional protective container closure components	396
(e.g., desiccants, foil overwraps) can be used.	397
398
2. 	Process Development	399
400
Process development should include the following:	401
402
• 	Selection of an appropriate manufacturing process (including manufacturing	403
equipment).	404
405
• 	Identification of factors or process variables that have a potential to impact MDI or	406
DPI product CQAs.	407
408
• 	Process optimization (which includes determination of appropriate ranges for the	409
process variables).	410
411
• 	Determination of in-process controls.	412
413
• 	Identification of an approach for scale-up (if applicable).	414
415
The crystallinity of the drug substance in MDIs and DPIs can be affected by mechanical	416
processing, including micronization. This can lead to the generation of amorphous particles that	417
are thermodynamically unstable, with a tendency to convert to a more stable crystalline state	418
with time. This recrystallization of micronized material could lead to uncontrolled particle	419
growth, thereby affecting the MDI or DPI product CQAs (e.g., APSD, DDU). Therefore, a	420
conditioning step should be considered following micronization to allow conversion of	421
amorphous to crystalline form under controlled conditions of temperature and humidity.	422
423
Evaluation of process monitoring data during the development of the manufacturing process can	424
enhance process understanding and support continual process development over the product life	425
cycle.	426

Contains Nonbinding Recommendations
Draft — Not for Implementation
11
427
a. 	MDIs	428
429
Typical manufacturing operations for an MDI are sequential mixing of the drug substance(s),	430
propellants and cosolvents, filling, device constituent part assembly, and packaging.	431
432
• 	For a suspension formulation, adequate mixing and circulation within the formulation	433
tank, filling tank, and filling heads is necessary during the filling process to achieve	434
uniformity of product fill into individual units.	435
436
• 	Filling processes are usually pressure fill, cold fill, or a combination of these	437
depending on the formulation characteristics, type of equipment available, and	438
manufacturing expertise and experience. MDI canisters can be filled with a pre-	439
specified calibrated amount of formulation in single or multiple steps.	440
441
• 	A better understanding of the filling process can be obtained by designing	442
experiments to study the impact of deliberate variations in the process parameters on	443
the MDI product assay, consistency of filling of both the drug substance and the	444
propellant, valve crimp measurements, weight checking, spray testing, etc. These	445
experiments could be used to optimize the filling operation and define an appropriate	446
design space for the MDI filling operation. For example, the filling operation of an	447
MDI can be optimized by evaluating the change in concentration of the drug	448
substance in the formulation tank during the filling process (due to the volatility of	449
the propellants) and determining the amount of propellant to be added to maintain the	450
concentration of the drug substance.	451
452
• 	Results from testing of product from trial runs can form the basis for further	453
optimization of the formulation or manufacturing process.	454
455
b. 	DPIs	456
457
Typical manufacturing operations for a DPI are dry powder blending or spray drying of the drug	458
substance(s) and excipients (carrier), blister or capsule filling (reservoir filling for device-	459
metered DPIs), device constituent part assembly, and packaging.	460
461
• 	The physical properties of the drug substance(s) and/or excipient(s) are usually	462
modified before they are used in the formulation. Particle generation or modification	463
processes can include spheronization, spray drying, and micronization.	464
465
• 	Increased drug substance particle cohesiveness resulting from the presence of very	466
small particles can adversely affect flowability, fillability, and dispersibility. Because	467
the concentration of the drug substance in the formulation is usually low, it can be	468
difficult to achieve a uniform distribution of the drug substance by direct blending.	469
These problems are typically minimized by blending drug particles with larger carrier	470
particles (e.g., lactose).	471
472

Contains Nonbinding Recommendations
Draft — Not for Implementation
12
• 	Blend uniformity can be measured using traditional methods such as High	473
Performance Liquid Chromatography (HPLC), where samples are obtained at the end	474
of blending using a sampling thief. Also, online technologies such as near infrared-	475
based procedures can be used to monitor blending in real time and ensure a	476
homogenous blend. The equipment (e.g., blender type), process parameters	477
associated with blending (e.g., blending time, blender speed, blender fill level), and	478
environmental conditions can impact blend uniformity and therefore the dose content	479
uniformity of the drug product. Final blend properties (e.g., bulk density, particle	480
size, flowability) can impact the process parameters for filling capsules, blisters, and	481
reservoirs.	482
483
• 	Other approaches to achieving a uniform distribution of the drug substance in the	484
formulation can include the use of spray drying or supercritical fluid technology.	485
486
• 	Typical filling methods for DPIs include dosator or tamp filling. Process parameters	487
can include blister/capsule filling speed, powder bed height in auger, and	488
encapsulator speed (for capsules). The capsules, blisters, reservoirs, or disks are	489
normally sealed to protect the formulation from environmental factors (e.g.,	490
humidity) which can affect product performance. Process parameters for sealing can	491
include the sealing temperature, dwell time, and machine speed. The effect of the	492
sealing process parameters on DPI product CQAs (e.g., impurities and degradants,	493
delivered dose, and APSD) should be investigated since the sealing process normally	494
involves heating.	495
496
D. 	Development of Control Strategy	497
498
As defined in ICH Q10,13 a control strategy is a “planned set of controls, derived from current	499
product and process understanding that assures process performance and product quality.” For	500
MDIs and DPIs, the overall purpose of the control strategy is to ensure that the CQAs are within	501
the appropriate range, limit, or distribution to assure drug substance and product quality.	502
503
The control strategy can include controls for incoming materials, in-process controls, and release	504
testing.	505
506
1. 	Controls for Incoming Materials	507
508
Appropriate controls for the drug substance(s), excipients, device constituent part(s), and	509
packaging materials should be established.14 If more than one drug substance is used in the	510
product formulation, controls should be in place for each of them, irrespective of the amount	511
present. If PSD of the drug substance or an excipient can affect the CQAs (e.g., APSD) of the	512
product, the PSD should be controlled. Similarly, other CQAs such as polymorphic form or	513
moisture content should be controlled if they can affect the quality of the product. If PSD of the	514
drug substance or an excipient is further modified by the product manufacturer as part of the	515
13 See FDA guidance for industry Q10 Pharmaceutical Quality System, ICH.
14 These controls must satisfy purchasing control requirements as described at 21 CFR 820.50. See Current Good
Manufacturing Requirements for Combination Products for additional information regarding this requirement.

Contains Nonbinding Recommendations
Draft — Not for Implementation
13
product manufacturing process, appropriate in-process controls and monitoring should be	516
established (see next section on in-process measurements and monitoring).	517
518
Excipients used in MDIs and DPIs are typically referenced to Drug Master Files (DMFs) and	519
have compendial monographs. For excipient CQAs that can impact the performance of the	520
finished MDI or DPI product, but are not included in a compendial monograph, appropriate	521
controls should be established.	522
523
Performance testing of the device constituent part (e.g., dimensions, valve functionality, dose	524
counter, actuator-orifice, extractables) is typically done by the vendors or fabricators of the	525
device constituent part and verified initially and on an annual basis by the applicant under their	526
internal quality system. The appropriateness of these tests and acceptance criteria should be	527
evaluated. For device constituent part components that will be in contact with the formulation or	528
the patient’s mouth, appropriate testing for extractables can be used as a substitute for leachables	529
testing in the product if a valid extractables-leachables correlation is established. Suitability of	530
the materials used for the device constituent part components can be addressed by their	531
compliance to biocompatibility testing standards (e.g., United States Pharmacopoeia (USP)	532
<87>, USP <88>, ISO 10993).	533
534
2. 	In-process Measurements and Monitoring	535
536
The in-process controls typically used for the manufacturing processes of MDIs can include:	537
formulation homogeneity, valve performance testing, heat stress testing, and weight checking.	538
Similarly, in-process controls for the manufacturing processes of DPIs can include: blend	539
uniformity, moisture content, fill weight, and sealing integrity, where applicable. Additional	540
monitoring of content uniformity using a stratified sampling approach during manufacturing15	
541
should be used for pre-metered DPIs with low drug loading.	542
543
Typically, drug substance or excipient manufacturers control the PSD of these materials before	544
they are provided for further manufacture. Alternatively, the manufacturer producing the	545
formulation for inclusion in the MDI or DPI can choose to adjust the PSD of these materials to	546
an appropriate range or distribution prior to using them. If micronization is used to adjust the	547
PSD, the in-process controls can include: total duration of micronization, PSD of the incoming	548
materials, feed rate, inlet air flow rate, air pressure, physical and mechanical properties of input	549
materials, number of times a lot is micronized, and re-introduction of carry-overs from previous	550
micronized lots. If a spray drying process is used, the in-process controls can include: solution	551
or suspension feed rate, inlet air and product temperatures, and air flow rate. If supercritical	552
fluid extraction is used, the in-process controls can include: concentration of solution, pressure,	553
temperature, and flow rates of carbon dioxide and drug solutions. For any of these three	554
technologies, it may be possible to develop mathematical models to predict PSD as a function of	555
process parameters or material attributes. In such cases, predictions from these models can be	556
used in lieu of actual measurement of PSD. Such models should be verified and updated.16	
557
15 The Use of Stratified Sampling of Blend and Dosage Units to Demonstrate Adequacy of Mix for Powder Blends,
PDA J Pharm Sci and Tech, 57, 64-74 (2003).
16 See FDA guidance for industry Q8, Q9, & Q10: Questions and Answers: Appendix: Q&As from Training
Sessions; (Q8, Q9, & Q10 Points to Consider), ICH.

Contains Nonbinding Recommendations
Draft — Not for Implementation
14
558
In situations where PSD cannot be directly measured (e.g., samples may not be accessible from	559
the formulation tank of an MDI until the canisters are filled), controls should be established to	560
ensure consistent manufacture of product with the desired PSD (e.g., monitoring and trending of	561
manufacturing process parameters that impact PSD).	562
563
When blister units or protective secondary packaging are used, controls should be established to	564
ensure that the seal area functions properly in terms of adhesion (e.g., heat seal, adhesive) and	565
mechanical seal. Appropriate integrity testing and acceptance criteria for seal completeness	566
(e.g., vacuum leak test) and seal strength (e.g., peel strength test) should be established to ensure	567
acceptable sealing properties within a batch and between batches.	568
569
3. 	Release Testing of the MDI and DPI Product	570
571
Release testing is performed on each batch of MDI or DPI product as part of the overall control	572
strategy. Each of the product attributes listed on the MDI or DPI product specification, most of	573
which are related to product CQAs, are normally tested at release. In some cases, if upstream	574
controls can be used to confirm that a batch of product meets a CQA related to an attribute on the	575
specification, that attribute does not need to be tested at release for every batch.	576
577
DDU and APSD should be included on the specifications for all MDIs and DPIs. Testing of	578
these attributes is performed on the assembled product using appropriate analytical procedures	579
(e.g., USP <601>). For DDU, the Agency also supports alternative statistical approaches using	580
parametric tolerance interval testing (PTIT),17,18 because these approaches are more relevant for	581
assuring the overall quality of the entire batch of an MDI or DPI.	582
583
APSD testing for an MDI or DPI confirms that the APSD profile of the product remains	584
consistent from the beginning of device constituent part use to the end. APSD testing is also	585
used to confirm that the product used in the clinical trials has similar drug delivery	586
characteristics to the to-be-marketed product. APSD is typically tested using an appropriate	587
cascade impactor and is dependent on both the formulation and the container closure system.	588
The measurement of the APSD is influenced by the characteristics of the MDI or DPI product	589
aerosol and is not solely determined by the size of the individual drug substance particles present	590
in the formulation. The impactor should have enough sizing stages to measure the total	591
distribution. The Agency recommends that all of the cascade impactors used to test the MDI or	592
DPI product throughout development should have the same design (e.g., Andersen Cascade	593
Impactor or Next Generation Impactor) and configuration. DPIs with low flow resistance require	594
high flow rates to achieve optimal pressure drop across the device constituent part. These device	595
constituent parts should be tested using impactors with alternative validated stage configurations.	596
It can be appropriate to refer to the current USP chapter for APSD procedures.	597
598
17 Presentation on “Parametric Tolerance Interval Test for Dose Content Uniformity (PTIT)” to the Advisory
Committee for Pharmaceutical Science (ACPS) on October 25, 2005. See
http://www.fda.gov/ohrms/dockets/ac/05/slides/2005-4187s1_Slide%20Index%20Day%201.htm.
18 Parametric Two-Tier Sequential Quality Assurance Test of Delivery Dose Uniformity of Multiple-Dose Inhaler
and Dry Powder Inhaler Drug Products, Journal of Biopharmaceutical Statistics, 18: 976-984, 2008.

Contains Nonbinding Recommendations
Draft — Not for Implementation
15
During development, the formulation and device constituent part components should be	599
examined microscopically. If the results indicate the formation of agglomerates, crystal growth,	600
the presence of large particles or foreign particulates, or changes in morphology of the drug	601
substance, appropriate controls for release and stability should be developed. In addition, if the	602
formulation supports microbial growth, appropriate controls for release and stability should be	603
considered.	604
605
606
IV. 	INFORMATION TO BE SUBMITTED IN AN APPLICATION	607
608
An applicant must provide technical data and information in sufficient detail to permit the	609
Agency to make a knowledgeable judgment about whether to approve the application or whether	610
grounds exist under section 505(d) of the Federal Food, Drug, and Cosmetic Act (FD&C Act) to	611
refuse to approve the application.19 This includes information about the drug substance20 and	612
information about the MDI or DPI product.21, 22	
613
614
The recommendations below are particularly relevant to MDIs and DPIs developed by following	615
traditional developmental approaches and are based on Agency experience with these products.	616
Information for more enhanced development could be different, although an applicant would be	617
expected to demonstrate enhanced knowledge and understanding. For example, alternative	618
control strategies to ensure product quality could be proposed. Applicants are encouraged to	619
discuss such proposals and their justification with the appropriate review division during	620
development.	621
622
The focus of this section is on aspects of MDIs and DPIs that are unique to these products. The	623
format of the submitted information should be based along the lines described in ICH M4Q.23	
624
625
A. 	Information on the Drug Substance	626
627
As described in section 3.2.S of ICH M4Q, the information submitted about the drug substance	628
should include information on General Properties, Manufacturer, Description of the	629
Manufacturing Process and Process Controls, Control of Materials, Controls of Critical Steps and	630
Intermediates, Manufacturing Process Development, Characterization, Control of Drug	631
Substance, Reference Standards, Container Closure System, and Stability.	632
633
Attributes typically included on the specifications for drug substances used in MDIs and DPIs	634
are listed in Table 1, below. Additional recommendations can be found in ICH Q6A,24 and other	635
applicable guidances.	636
19 See 21 CFR 314.50(d).
20 See 21 CFR 314.50(d)(1)(i).
21 See 21 CFR 314.50(d)(1)(ii).
22 For additional assistance on where to provide device constituent information using the eCTD format see eCTD
Technical Conformance Guide: Technical Specifications Document: guidance for industry Providing Regulatory
Submissions in Electronic Format —Certain Human Pharmaceutical Product Applications and Related Submissions
Using the eCTD Specifications.
23 See FDA guidance for industry M4Q: The CTD — Quality, ICH.
24 See FDA guidance for industry Q1A(R2) Stability Testing of New Drug Substances and Products, ICH.

Contains Nonbinding Recommendations
Draft — Not for Implementation
16
637
Table 1. Attributes Usually Tested at Release and on Stability	638
for Drug Substances Used in MDIs and DPIs	639
640
Attribute 	Release 	Stability
Color 	X
Appearance (visual and microscopic) 	X 	X
Identity 	X
Moisture Content 	X 	X
Residue on Ignition 	X
Specific Rotation 	X
Assay 	X 	X
Impurities 	X 	X
Microbial Limits 	X 	X
Melting Range 	X 	X
PSD 	X 	X
Morphology* 	X 	X
Amorphous Content 	X 	X
Individual Residual Solvents 	X
Heavy Metals** 	X
* Examples include shape, crystal habit, texture, surface area, and rugosity.	641 ** Can be replaced with elemental impurities.25	
642
643
B. 	Description and Composition (P1)	644
645
As described in ICH M4Q, section 3.2.P.1 of the application should include a list of all	646
components (i.e., ingredients) used in the manufacture of the MDI or DPI drug constituent part.	647
648
1. 	MDIs	649
650
The amount of each component in the final formulation should be expressed in terms of	651
concentration (i.e., amount per unit volume or weight), as well as amount per container and	652
amount delivered from the valve per actuation. The amount of drug delivered from the	653
mouthpiece and any associated features (e.g., integrated spacers) per actuation should be	654
provided. The mass of drug delivered from the mouthpiece per actuation is the specified target	655
delivered dose (TDD) and is used to denote the strength. In addition, for suspension	656
formulations, the density of the individual formulation components should be included. The	657
reported densities should be measured at the product storage temperature.	658
659
25 See FDA guidance for industry Q3D Elemental Impurities, ICH, USP General Chapter <232> Elemental
Impurities-Limits, USP General Chapter <233> Elemental Impurities-Procedures, and FDA guidance for industry
Elemental Impurities in Drug Products. When final, this guidance will represent the FDA’s current thinking on this
topic.

Contains Nonbinding Recommendations
Draft — Not for Implementation
17
2. 	DPIs	660
661
The amount of each drug should be expressed in terms of concentration (i.e., amount per unit	662
weight (e.g., micrograms per gram)) and as net content (in micrograms) per capsule or blister.	663
The metered amount and the mass of the drug delivered from the mouthpiece under defined test	664
conditions (i.e., flow rate, duration) should both be provided. The mass of drug delivered from	665
the mouthpiece is the specified target delivered dose (TDD). The metered amount of the drug	666
from a DPI is used to denote its strength, not the specified TDD.	667
668
For device-metered DPIs, the TDD, metered dose, and net formulation content should be	669
provided.	670
671
C. 	Pharmaceutical Development (P2)	672
673
As described in ICH M4Q, section 3.2.P.2 of the application should contain information on	674
studies conducted to establish that the dosage form, formulation, manufacturing process,	675
container closure system, microbiological attributes, and usage instructions specified in the	676
application are appropriate for the intended use of the MDI or DPI product. Because an MDI or	677
DPI is a combination product, this section should address the developmental process for the	678
entire product including the device constituent part. The applicant should consider including the	679
following:	680
681
• 	A description of the QTPP.	682
683
• 	A list of the CQAs of the MDI or DPI product, along with the limit, range, or	684
distribution associated with each CQA and appropriate justification.	685
686
• 	Identification of those aspects of drug substances, excipients, container closure	687
system (including the device constituent part), and manufacturing processes important	688
to attaining product quality.	689
690
• 	Rationale for the selection or design of the proposed container closure system	691
(including the device constituent part) and storage conditions, including a summary of	692
the changes in container closure components used throughout the development (e.g.,	693
in tabular form).	694
695
• 	Pilot scale or larger scale process development studies used to support the proposed	696
commercial scale control strategy. This could include:	697
698
o 	Summary of prior knowledge and risk assessment methodologies used to identify	699
the process parameters and material attributes that have the potential to impact	700
product CQAs.	701
702
o 	Summary of experimental studies used to identify operating ranges or design	703
space. If design of experiments (DOE) was used, a summary table should be	704
provided that includes input factors, ranges studied, results, and conclusions.	705

Contains Nonbinding Recommendations
Draft — Not for Implementation
18
706
o 	Appropriate scale-up correlations should be provided to justify proposed ranges at	707
commercial scale.	708
709
o 	Rationale for the selection of input materials and their proposed acceptance	710
criteria.	711
712
o 	Rationale for the selection of the manufacturing process, including in-process	713
controls.	714
715
o 	Justification for the formulation overfill per unit, needed to maintain the	716
performance of the MDI or DPI product throughout the labeled number of	717
actuations, as applicable.	718
719
• 	Summary data from MDI or DPI product characterization studies. These are used to	720
demonstrate the robustness and performance of the product and support labeling.	721
Studies recommended for MDIs and DPIs are listed in Table 2, below. The	722
applicability of each of the characterization studies outlined below for a given product	723
can be discussed with the responsible review division. Additional information on the	724
purpose and design of these characterization studies can be found in the Appendix,	725
section V.B.2.	726
Table 2. Characterization Studies	727
728
Studies 	MDI 	DPI
In-Use Period 	X 	X
Temperature Cycling 	X 	X
Priming and Repriming 	X
Effect of Patient Use 	X 	X
Effect of Storage and Shaking (suspension formulated MDIs
only)
X
Effect of Orientation of the Device on Delivered Dose 	X
Drug Deposition on Mouthpiece and/or Accessories 	X 	X
Cleaning Instructions 	X 	X
Profiling of Actuations Near Device Exhaustion 	X 	X
Effect of Varying Flow Rate on DPI Performance 	X
Effect of Flow Rate and Inhalation Delay on MDIs with
Spacers
X
Robustness 	X 	X
729
D. 	Manufacture (P3)	730
731
As described in ICH M4Q, section 3.2.P.3 of the application should contain information about	732
where and how the MDI or DPI product will be manufactured. This should include information	733
on the drug and device constituent parts and the final combination product assembly. In	734
addition, the application should contain information necessary to demonstrate compliance with	735

Contains Nonbinding Recommendations
Draft — Not for Implementation
19
21 CFR part 4. See the FDA guidance for industry and staff Current Good Manufacturing	736
Practice Requirements for Combination Products for more information on these requirements.	737
In addition, the FDA guidance for industry and staff Quality System Information for Certain	738
Premarket Application Reviews provides information regarding the quality system information	739
that should be included in a regulatory submission.	740
741
The complete street address and contact information (e.g., email, phone and fax numbers) should	742
be listed in the application form 356h for each facility involved in the manufacturing or testing of	743
the MDI or DPI product, including the testing of components of the product. If manufacturing	744
information is provided in a DMF, all sites that are described in the DMF should also be listed in	745
the application form 356h.	746
747
The batch formula and a description of the manufacturing process and process controls should be	748
provided. A detailed schematic diagram of the proposed production process, including	749
descriptions of the equipment, operating conditions, and process controls, should also be	750
provided.26	
751
752
If a drug substance or excipient is micronized after being received from a supplier, the process	753
parameters for micronization should be described as part of the product manufacturing process.	754
If a conditioning step follows micronization, the conditioning parameters and process controls	755
should also be described.	756
757
If the MDI manufacturing process involves filling a suspension into a canister, either by pressure	758
fill or cold fill, appropriate process parameters and in-process controls to assure the formulation	759
homogeneity should be provided in the application.	760
761
If the manufacturing process involves blending of drug or excipient particles, the process	762
parameters associated with blending (e.g., blender size, blending time, blender speed, blender	763
loading configurations, environmental conditions) and in-process controls for assuring blend	764
uniformity should be described.	765
766
Filling and packaging procedures (primary and protective secondary packaging) for the MDI or	767
DPI product should be described in the application, including relevant process controls for these	768
operations.	769
770
E. 	Control of Excipients (P4)	771
772
As described in ICH M4Q, section 3.2.P.4 of the application should provide the following	773
information on control of excipients:	774
775
• 	Manufacturer, supplier, characterization studies, certificate of analysis and other	776
specific information should be provided as appropriate, for all excipients.	777
778
• 	Specifications for excipients.	779
26 See 21 CFR 314.50(d)(1)(ii)(c).

Contains Nonbinding Recommendations
Draft — Not for Implementation
20
780
• 	Analytical procedures used for testing the excipients, when appropriate.	781
782
• 	Analytical validation information, when appropriate.	783
784
DMFs can also be referenced in the application for quality and toxicological information. For	785
additional guidance on pharmacological and toxicological considerations, the applicant should	786
consult available CDER guidance,27 or contact the responsible review division.	787
788
For certain compendial excipients, the specifications should include tests in addition to those	789
stated in the monograph. Typical examples are shown in Table 3, below.	790
791
Table 3. Examples of Tests in Addition to Compendial Excipient Tests	792
793
Dosage
Form Excipient 	Function 	Tests
MDI 	Dehydrated Alcohol,
USP
Cosolvent 	• Water content (e.g., Karl Fischer)
• Impurity profile
Lecithin, NF 	Surfactant 	• Tests that define the
compositional profile in detail
Oleic Acid, NF 	Surfactant 	• Impurity profile in detail
DPI 	Lactose
Monohydrate, NF
Carrier 	• Quantitative color and clarity
• Anomeric purity
• Elemental impurities
• Amorphous content
• Microbial limits
• Organic volatile impurities
• Related impurities
• Particle size distribution
• pH
• Assay
• Particle size and morphology
• Pyrogens and bacterial endotoxins
Anhydrous Lactose,
NF
Carrier 	• Quantitative color and clarity
• Anomeric purity
• Elemental impurities
• Amorphous content
• Microbial limits
• Organic volatile impurities
• Related impurities
27 See FDA guidance for industry Nonclinical Studies for the Safety Evaluation of Pharmaceutical Excipients.

Contains Nonbinding Recommendations
Draft — Not for Implementation
21
Dosage
Form Excipient 	Function 	Tests
• Particle size distribution
• pH
• Assay
• Monohydrate lactose content
• Particle size and morphology
• Pyrogens and bacterial endotoxins
794
For propellants (e.g., HFA-134a, HFA-227) specifications should include the following tests:	795
identity, appearance, assay, acidity, total residue, moisture content, related impurities, and	796
unrelated impurities (e.g., CO, N2 , O2 gases). Generally, the assay acceptance criterion should	797
not be less than 99.99 percent for propellants. The related impurities acceptance criteria for	798
HFA-134a and HFA-227, shown in Table 4 and Table 5, are typical of the limits that are	799
considered acceptable.	800
801
Table 4. Examples of Acceptance Criteria for Impurities in HFA-134a	802
803
Impurity 	Acceptance
Criteria (ppm)
Impurity 	Acceptance
Criteria (ppm)
HCC-40 	5 	HCFC-133a 	5
HFC-23 	5 	HCFC-161 	30
HFC-32 	5 	HCFC-1121 	5
HFC-125 	5 	HCFC-1122 	5
HFC-134 	1000 	HCFC-1122a 	5
HFC-143a 	20 	CFC-11 	5
HFC-152 	5 	CFC-12 	100
HFC-152a 	300 	CFC-12B1 	5
HFC-245cb 	5 	CFC-13 	5
HFC-1123 	5 	CFC-113 	5
HFC-1132 	5 	CFC-114 	5
HFC-1225ye 	5 	CFC-114a 	25
HFC-1234yf 	5 	CFC-115 	5
HFC-1243zf 	5 	CFC-1112a 	5
HFC-1336mzz 	5 	FC-1318my-T 	5
HCFC-22 	50 	FC-1318my-C 	5
HCFC-31 	5 Total unsaturates (including
HCFC-1122) 5
HCFC-123 	5 	Individual unidentified impurities 	5
HCFC-123a 	5 	Total unidentified impurities 	10
HCFC-124 	100 	Other organic impurities 	50
HCFC-124a 	5 	Any other identified saturated impurity 	5
HCFC-132b 	5 	Total impurities 	1000

Contains Nonbinding Recommendations
Draft — Not for Implementation
22
804
Table 5. Examples of Acceptance Criteria for Impurities in HFA-227	805
806
Impurity 	Acceptance Criteria (ppm)
P124 	3
P227 ca
(Saturated) 3
Unsaturated related impurities
P1216 hexafluoropropene 	3
P1225ye
1,1,1,2,3, pentafluoropropene 3
P1225zc
1,1,1,3,3,pentafluoropropene 3
P245cb 1,1,1,2,2-Pentafluoropropene
Hexafluorocyclopropane 	2
1 chloro-1,2,2,2,tetrafluoroethane 	10
Octafluoropropane 	2
Chloropentafluoroethane 	2
4-methylperfluoropentene-2 (isomer 1) 	2
2-methylperfluoropentene-2 (isomer 2) 	2
2-chloroheptafluoropropane 	2
Hexafluoropropane 	2
Heptafluorobutene 	2
2-H-2-methyperfluoropentane 	2
Water content 	10
Acidity as Hydrogen Chloride 	0.1
Non-volatile residue 	20
Total unsaturated 	5
Individual unidentified impurities 	3
Other organic impurities 	50
Any other identified saturated impurity 	5
Total impurities 	20
Purity by GC Assay 	>99.99
807
F. 	Control of MDI and DPI Product (P5)	808
809
As described in ICH M4Q, section 3.2.P.5 of the application should contain the following	810
information on control of MDI or DPI product:	811
812
• 	Specification.	813
• 	Analytical procedures.	814
• 	Validation of analytical procedures.	815
• 	Characterization of impurities.	816

Contains Nonbinding Recommendations
Draft — Not for Implementation
23
• 	Batch analyses.	817
• 	Justification for the proposed specification.	818
819
Typical release tests for MDIs and DPIs are provided in Table 6, below.	820
821
Table 6. Attributes Typically Included on Specifications for MDIs and DPIs	822
823
Attribute 	MDI 	DPI
Description 	X 	X
Identification 	X 	X
Assay 	X 	X
Impurities and Degradation Products 	X 	X
Valve Delivery (Shot Weight) 	X
Delivered Dose Uniformity (DDU) 	X 	X
Uniformity of Dosage Units 	X
Aerodynamic Particle Size Distribution (APSD) 	X 	X
Spray Pattern 	X
Foreign Particulate Matter 	X 	X
Microbial Limits 	X 	X
Water or Moisture Content 	X 	X
Alcohol/Antioxidants/Preservatives Content* 	X
Net Content (Fill) Weight 	X 	X
Leachables (Stability) 	X 	X
* When present	824
825
The proposed analytical procedures should be documented in sufficient detail28 that they can be	826
reviewed and reproduced in FDA laboratories. If any attribute is tested in-process during	827
manufacturing in lieu of release testing, it should be indicated as such on the specification.	828
829
The following information for specific attributes and criteria should be provided:	830
831
1. 	Description	832
833
MDIs and DPIs: The appearance of the contents of the container (i.e., formulation) and the	834
appearance of components of the container closure system should conform to their respective	835
descriptions as an indication of product integrity. For example, there should be no visible	836
evidence of drug substance surface deposition or corrosion of container closure system	837
components of an MDI, such as pitting or discoloration. If any color is associated with the	838
formulation (either present initially or from a known degradative process occurring during shelf	839
life), a quantitative color test with appropriate acceptance criteria should be established, unless	840
the impurity causing the color has been identified and its concentration will be monitored by	841
another analytical procedure.	842
28 See USP General Chapter <5> Inhalational and Nasal Drug Products-General Information and Product Quality
Tests, and USP General Chapter <601> Aerosols, Nasal Sprays, Metered-Dose Inhalers, and Dry Powder Inhalers,
for additional information, including analytical procedures for some of the attributes.

Contains Nonbinding Recommendations
Draft — Not for Implementation
24
843
2. 	Valve Delivery	844
845
MDIs: Valve delivery (amount of formulation released per actuation, shot weight) should be	846
measured for 10 units. The acceptance criteria should be not more than (NMT) ±15 percent for	847
individual actuations and NMT ±10 percent for the mean of the actuations relative to the target.	848
Acceptance testing for valve delivery on incoming valve lots can be substituted for the release	849
testing of valve delivery for the MDI product, if justified. However, the acceptance criteria for	850
valve delivery should be included in the MDI product specification.	851
852
3. 	Delivered Dosage Uniformity (DDU)	853
854
MDIs and DPIs: The test for DDU measures the amount of drug discharged from the	855
mouthpiece of the MDI or DPI and compares that measurement to the TDD.	856
857
Not more than two actuations per determination should be used for DDU. Where the number of	858
actuations per minimum dose specified in the product labeling is one, the number of actuations	859
per determination should be limited to one. The amount of drug substance discharged should be	860
expressed both as the actual weight and as a percent of the label claim from the actuator. The	861
USP Unit Spray <601> sampling apparatus can be used and containers should be primed	862
according to the instructions in the labeling (as appropriate). Testing should be carried out under	863
optimized conditions of air flow rate and total air volume (drawn through the device during the	864
test). For DPIs, inhalation aerosols, and inhalation aerosols with integrated spacers or similar	865
accessories, the volume of collection should not exceed 2 L at a constant flow rate.	866
867
Testing for each batch should be conducted on an appropriate number of representative units (at	868
least 10). For MDIs and device-metered DPIs, each MDI or device-metered DPI is considered a	869
unit and both the initial dose and the last of the labeled number of doses should be tested. For	870
pre-metered DPIs, each container (capsule, single blister, or single cartridge) is considered a unit.	871
The sampling approach (including the number of samples tested and the number of replicate	872
analyses performed per sample) should be included as part of the analytical procedure and	873
acceptance criteria.	874
875
The Agency recommends that applicants adopt a PTIT approach to measuring DDU. However,	876
alternative approaches can be used if appropriately justified. The Appendix (section V.C.)	877
includes two examples of approaches to measuring DDU, including the PTIT approach. MDIs	878
and DPIs: The test for DDU measures the amount of drug discharged from the mouthpiece of the	879
MDI or DPI and compares that measurement to the TDD.	880
881
4. 	Uniformity of Dosage Units	882
883
Pre-metered DPIs (i.e., each dose is separately packaged or segregated within a package): The	884
DPI product specification should include a test and acceptance criteria for the content uniformity	885
of pre-metered dosage units (e.g., as described in USP General Chapter <905> Uniformity of	886
Dosage Units).	887
888

Contains Nonbinding Recommendations
Draft — Not for Implementation
25
5. 	Aerodynamic Particle Size Distribution (APSD)	889
890
MDIs and DPIs: The equipment (e.g., multistage cascade impactor, vacuum source, valve,	891
timing system) used to characterize the APSD of the MDI or DPI product should be described.	892
Any accessories or modifications to the equipment (e.g., stage substitution, expansion chamber,	893
inlet stem, adaptors, collection plate surface coating) should be included in the description. The	894
qualification criteria for the equipment should be included in the description of the analytical	895
procedure.	896
897
Testing should be carried out under the same optimized conditions of air flow rate as is used in	898
the DDU test. Other important test parameters (e.g., air flow duration, relative humidity,	899
temperature) and information (e.g., cleaning of the equipment between runs, frequency of	900
mensuration) should be specified in the procedure. For DPIs, the volume per measurement	901
should not exceed 4 L.	902
903
APSD should be determined separately for each MDI or DPI. An appropriate minimum number	904
of MDI or DPI products (e.g., 5) should be tested individually and the determination for each	905
unit should be performed with the minimum number of actuations justified by the sensitivity of	906
the analytical procedure used to quantitate the deposited drug. The amount of drug deposited on	907
the critical stages of the cascade impactor should be sufficient for reliable assay, but not so	908
excessive as to bias the results by masking individual actuation variability. For MDIs, device-	909
metered DPIs, and pre-metered DPIs that contain enclosed ordered assemblies of individual dose	910
units, the APSD should usually be measured for the initial dose and also for the last of the	911
labeled number of doses. However, if there is no discernible APSD trend from beginning- to	912
end-of-unit life in the data from submission batches, routine testing for post-approval batches can	913
be performed only at the beginning-of-unit life.	914
915
It is not considered adequate to characterize the APSD in terms of the mass median aerodynamic	916
diameter (MMAD) and geometric standard deviation (GSD) alone, or to limit the	917
characterization only to fine particle mass or fine particle fraction. Acceptance criteria should be	918
proposed based on the amount of drug deposited on various stages of the equipment. Applicants	919
should propose acceptance criteria for groupings of consecutive stages rather than proposing an	920
acceptance criterion for each individual stage. In most cases, three or four groupings should be	921
sufficient to characterize the APSD adequately.	922
923
The mass balance (i.e., the amount of drug substance deposited on all surfaces from the valve to	924
the equipment filter) should be measured for each run. If the mass balance is not between 85 and	925
115 percent of TDD, the test result should be investigated under the applicant’s quality system.	926
The investigation should include evaluation of the suitability of the analytical procedure and	927
dose delivery testing of the units that failed APSD mass balance.	928
929
6. 	Spray Pattern	930
931
MDIs: The test procedure for spray pattern of the MDI product should include the following	932
information: collection distances between the mouthpiece and the measurement plane	933
(preferably at least two), number of actuations per spray pattern (preferably n = 1), position and	934

Contains Nonbinding Recommendations
Draft — Not for Implementation
26
orientation of the measurement plane relative to the mouthpiece, and visualization method. The	935
collection distances should provide adequate discriminatory capability. The acceptance criteria	936
at different distances should include the shape and size of the spray pattern with the ratio of the	937
longest to the shortest axes stated (e.g., 1.00 – 1.20).	938
939
Acceptance testing for spray pattern on incoming actuator lots with the specified valve can	940
substitute for the release testing of spray pattern for the MDI product, if justified. However, the	941
acceptance criteria for the spray pattern should be included in the MDI product specification.	942
943
7. 	Foreign Particulates	944
945
MDIs and DPIs: The MDI or DPI product specification should include tests and acceptance	946
criteria for foreign particulates. The acceptance criteria should include limits for less than 10	947
micrometers, 10 to 25 micrometers, and greater than 25 micrometers.	948
949
8. 	Microbial Limits	950
951
MDIs and DPIs: The MDI or DPI product specification should include tests and acceptance	952
criteria for total microbial count and specified indicator organisms. USP compendial methods	953
and criteria in General Chapters <610> Alternative Microbiological Sampling Methods for	954
Nonsterile Inhaled and Nasal Products and <1111> Microbiological Examination of Nonsterile	955
Products Acceptance Criteria for Pharmaceutical Preparations and Substances for	956
Pharmaceutical Use can be referenced.	957
958
9. 	Leachables	959
960
MDIs and DPIs: Additional information related to leachables and extractables can be found in	961
the following documents: USP Chapters <1663>, <1664>, and <1664.1>, and PQRI	962
Recommendations to FDA.29	
963
964
G. 	Reference Standards or Materials (P6)	965
966
As described in ICH M4Q, section 3.2.P.6 of the application should contain information on	967
reference standards or reference materials used for testing of the MDI or DPI product, if not	968
previously provided in 3.2.S.5, Reference Standards or Materials.	969
970
H. 	Container and Closure System (P7)	971
972
As described in ICH M4Q, section 3.2.P.7 of the application should contain the information for	973
the container closure system (which includes the device constituent part and the primary and	974
secondary packaging). The application for MDIs and DPIs should also include the following	975
information, as provided in Table 7, below.	976
977
29 See http://pqri.org/wp-content/uploads/2015/08/pdf/LE_Recommendations_to_FDA_09-29-06.pdf.

Contains Nonbinding Recommendations
Draft — Not for Implementation
27
Table 7. Information to be Submitted in Support of an	978
Application for an MDI or DPI Container Closure System	979
(Including the Device Constituent Part)	980
MDI 	MDI and
DPI
DPI
Canister 	Valves and
Components
Actuator/Mouthpiece
and Additional
Accessories
Protective
Packaging
Device
Constituent
Part and
Components
Fabricator(s) of Device
Constituent Part and
Components
• 	• 	• 	• 	•
Unique Identifier(s) 	• 	• 	• 	• 	•
Composition and Control of
Materials for Critical
Components
• 	• 	• 	• 	•
Engineering Drawings with
Precise Dimensions and
Tolerances
• 	• 	• 	•
Cleaning procedures and
reagents used •
Control Extraction
Procedures and Data • 	•
Control Procedures for
Residues1 	•
Qualitative and Quantitative
Extractable Profile(s) • 	•
Toxicological Evaluation of
Extracted Materials (and
Residues1 )
• 	•
Specification and Analytical
Sampling Plans 2 	• 	• 	• 	• 	•
CoA or Representative Test
Data
• 	• 	• 	• 	•
Functional and Performance
Characteristics3 	•
Identity, Composition, and
Treatment Procedures of
Elastomeric Components
•
Flow Resistance 4 	•
USP Biological Reactivity
Testing <87> and <88> and
Food Additive Regulation5
• 	•
1 Process contaminants (where appropriate)	981 2 Can include for example, dimensions, qualitative and quantitative extractables and residues, physicochemical	982 parameters, compositional controls, and/or performance characteristics	983 3 For example: valve actuation force, stroke length, valve delivery, and valve leakage of the assembled	984 valve/canister combination containing placebo formulation	985 4 Supportive information and data should be provided to characterize any dependence of the drug delivery and	986 formulation deagglomeration on the flow resistance of the device constituent part	987 5 If the components are not recognized as safe for food contact under appropriate regulations, extractables (e.g.,	988 organic solvent(s), water), obtained under defined experimental conditions, should be established analytically both	989 qualitatively and quantitatively. In addition to in vitro and in vivo tests and other safety data for these components	990

Contains Nonbinding Recommendations
Draft — Not for Implementation
28
not recognized as safe for food contact, extractables profiles with multiple solvents should be assessed	991 toxicologically and a rationale provided to support limits for extractables that can be applied on a routine basis.	992
993
Identity and concentration profiles of the leachables in the MDI or DPI product or placebo	994
formulation (i.e., MDI or DPI product formulation without drug substance) should be determined	995
for the primary stability batches and should include testing at multiple time-points to the end of	996
the proposed shelf life. These data should be correlated, if possible, with the extractables	997
profile(s) of the container closure system determined under the various control extraction study	998
conditions.	999
1000
For ANDAs, the applicant can compare the extraction profiles of the container closure system	1001
with the leachables profile(s) of the MDI or DPI product (or placebo) after storage under	1002
accelerated stability conditions as long as the applicant confirms that post-approval verification	1003
activities will include an assessment of initial production stability batches to confirm the results	1004
for the MDI or DPI product (or placebo) through the expiration dating period. If equilibrium is	1005
not reached by six months, real-time long-term data should be used to establish an appropriate	1006
expiration dating period. If the compared results are within the applicant’s acceptance criteria	1007
but there are qualitative differences, the results should be discussed with the responsible review	1008
division.	1009
1010
For additional information on container closure systems, refer to appropriate Agency guidance	1011
and available standards.	1012
1013
I. 	Stability (P8)	1014
1015
Stability studies should be conducted as recommended in ICH Q1A(R2), Q1C, Q1D, and	1016
Q1E.30,31 The MDI or DPI product should be packaged as intended for commercialization,	1017
including secondary packaging. Stability data collected during the clinical investigations phase	1018
on the MDI or DPI product packaged in a different container closure system configuration can be	1019
provided as supporting data, with appropriate justification.	1020
1021
If protective secondary packaging is used, the routine stability test storage conditions for the	1022
product in the presentation intended for distribution should include both long-term storage at	1023
25ºC/60 percent relative humidity (RH) and at 30ºC/65 percent RH for one-half of the proposed	1024
expiration dating period.	1025
1026
Table 8 below describes the attributes that should be tested during stability studies. During the	1027
conduct of stability studies, the MDI or DPI product should be stored in upright, horizontal, and	1028
inverted orientations. If sufficient data demonstrate that orientation does not affect the product	1029
quality, routine stability studies can be conducted on product stored in only one orientation.	1030
Alternatively, if data demonstrate a certain orientation is detrimental to product stability, that	1031
30 See FDA guidance for industry Q1A(R2) Stability Testing of New Drug Substances and Products, ICH, FDA
guidance for industry Q1C Stability Testing for New Dosage Forms, ICH, FDA guidance for industry Q1D
Bracketing and Matrixing Designs for Stability Testing of New Drug Substances and Products, ICH, and FDA
guidance for industry Q1E Evaluation of Stability Data, ICH.
31 See FDA guidance for industry ANDAs: Stability Testing of Drug Substances and Products and FDA guidance for
industry ANDAs: Stability Testing of Drug Substances and Products: Questions and Answers.

Contains Nonbinding Recommendations
Draft — Not for Implementation
29
orientation should be used in routine stability studies. However, shipping and storage of the	1032
marketed product should utilize the most favorable orientation.	1033
1034
Table 7. Attributes Normally Tested During Stability Studies	1035
1036
Attribute 	MDI 	DPI
Description 	X 	X
Assay 	X 	X
Impurities and Degradation Products 	X 	X
Valve Delivery (Shot Weight) 	X
Delivered Dose Uniformity (DDU) 	X 	X
Aerodynamic Particle Size Distribution (APSD)* 	X 	X
Spray Pattern 	X
Particulate Matter 	X 	X
Microbial Limits 	X 	X
Leachables 	X
Alcohol Content** 	X
Water or Moisture Content*** 	X 	X
Leak Rate 	X
* For suspension-based MDIs, device-metered DPIs, and multi-dose DPIs that contain enclosed	1037 ordered assemblies of individual pre-metered dose units, the stability studies on the primary stability	1038 batches should determine the effect of storage time and conditions on the APSD through unit life	1039 (determinations from the initial actuations and also for the last of the labeled number of actuations).	1040 If APSD changes through unit life, the proposed stability protocol should include APSD testing at the	1041 beginning and end of unit life.	1042 ** When present	1043 *** In addition to moisture present in the excipient	1044
1045
J. 	Labeling	1046
1047
The following information is recommended for the labeling of MDIs and DPIs, to help achieve	1048
consistency and uniformity in the content, product title, and format.32 In this section, the term	1049
“drug product” refers to the combination product (i.e., the MDI or DPI product) and is used for	1050
clarity because pertinent labeling regulations and requirements use the term “drug product.”	1051
These comments are directed mainly at labeling issues unique to prescription MDI and DPI	1052
products. Additional information regarding the labeling of drug products can be found in 21	1053
CFR part 201. See also FDA guidance for industry on Naming of Drug Products Containing	1054
Salt Drug Substances.33	
1055
32 As a general matter, ANDAs are required to include information to show that the labeling proposed for the
generic drug is the “same” as the RLD, with certain limited exceptions, such as for changes required because of
differences approved under a suitability petition (see section 505(j)(2)(c) of the FD & C Act and 21 CFR 314.93), or
because the generic drug and the RLD are produced or distributed by different manufacturers the (see section
505(j)(2)(A)(v) of the FD & C Act). Applicants intending to submit an ANDA covering an MDI or DPI may also
refer to FDA’s draft guidance entitled Comparative Analyses and Related Comparative Use Human Factors Studies
for a Drug-Device Combination Product Submitted in an ANDA for additional information. When final, this
guidance will reflect FDA’s current thinking on this topic.
33 In addition, see USP General Chapter <1121> Nomenclature for the Monograph Naming Policy for Salt Drug
Substances in Drug Products and Compounded Preparations.

Contains Nonbinding Recommendations
Draft — Not for Implementation
30
1056
1. 	MDIs	1057
1058
The labeling of oral MDIs should state the established name of the product as (Drug) Inhalation	1059
Aerosol and provide the strength as the amount delivered per actuation under defined in vitro	1060
conditions. For nasal MDIs, the product labeling should state the established name of the	1061
product as (Drug) Nasal Aerosol and provide the strength as the amount delivered per actuation.	1062
The established name and strength should be followed by a phrase such as “For oral inhalation	1063
only” or “For nasal inhalation only,” as appropriate.	1064
1065
In addition to the information typically required under Title 21 for the description of the drug	1066
substance and formulation (21 CFR part 201), the product labeling should include the following	1067
information specific to MDI products:	1068
1069
a. 	DESCRIPTION Section of the Prescribing Information	1070
1071
• 	A description of the appearance of the actuator and cap.	1072
1073
• 	The specified TDD from the mouthpiece per actuation should be expressed:	1074
1075
o 	For example: “Each actuation meters ‘x’ mcg of drug in ‘w’ mg of	1076
suspension (solution) from the valve and delivers ‘y’ mcg of drug,	1077
equivalent to ‘z’ mcg of drug substance (if applicable) from the actuator	1078
(i.e., mouthpiece or nasal adapter).”	1079
1080
o 	The term “approximately” should not be used to modify the medication	1081
amount delivered. If special circumstances warrant additional statements	1082
regarding the metered amount, this should be discussed with the	1083
appropriate review division.	1084
1085
• 	A statement should be included that the amount of drug delivered to the lung	1086
will depend on patient coordination of device actuation with the inhalation	1087
maneuver, as well as on patient factors such as inspiratory flow and peak	1088
inspiratory flow (PIF) through the delivery system, which may vary for	1089
asthma, COPD, and other patient populations.	1090
1091
• 	A list of all excipients should be included. Substances should be identified by	1092
their established names.	1093
1094
• 	If the drug substance that exits the mouthpiece is a hydrate, solvate, or	1095
complex, this information should be clearly specified with proper strength	1096
conversion for the active moiety.	1097
1098
• 	The number of usable actuations per container.	1099
1100

Contains Nonbinding Recommendations
Draft — Not for Implementation
31
• 	A statement should be included that the canister should be discarded when the	1101
labeled number of actuations has been used.	1102
1103
b. 	HOW SUPPLIED/STORAGE AND HANDLING Section of the	1104
Prescribing Information	1105
1106
• 	The net content (fill) weight of the container should be stated.	1107
1108
• 	The medication amount delivered (TDD) from the actuator.	1109
1110
• 	The number of actuations for each canister fill weight should be included.	1111
Qualifying terms such as “at least” and “approximately” should not be used.	1112
1113
• 	A description of the actuator and protective cap to be used with the container	1114
and valve, including the color and appearance, should be included.	1115
1116
• 	A statement should be provided that the canister should only be used with the	1117
accompanying actuator and that the actuator should not be used with any other	1118
inhalation drug product.	1119
1120
• 	A statement should be provided that the correct amount of medication in each	1121
inhalation cannot be ensured after the labeled number of actuations from the	1122
canister has been used, even though the canister may not be completely	1123
empty. Additionally, a statement should be included that the canister should	1124
be discarded when the labeled number of actuations has been used.	1125
1126
• 	Storage conditions should be clearly stated, including any warning statements	1127
regarding temperature and humidity.	1128
1129
• 	Any preferred storage orientation should be indicated.	1130
1131
• 	A statement should be included regarding the appropriate temperature of the	1132
MDI before use, as well as any requirements for shaking, if necessary. In	1133
addition, the impact of the cooling effect from multiple successive actuations	1134
on product performance should be described, if applicable.	1135
1136
• 	If protective secondary packaging (e.g., foil overwrap) is used, this should be	1137
clearly stated. In addition, appropriate statements should be included that the	1138
contents enclosed in the protective secondary packaging should not be used	1139
after a specified number of days (e.g., 2 weeks, 30 days) from the date the	1140
protective package was compromised (in-use period).	1141
1142
• 	Any warning statements required under 21 CFR 369.21 (e.g., storage above	1143
120ºF may cause bursting, keep out of reach of children, do not puncture, do	1144
not use or store near heat or open flame, never throw container into fire or	1145
incinerator, do not spray into eyes).	1146

Contains Nonbinding Recommendations
Draft — Not for Implementation
32
1147
• 	Information about shaking, priming, and repriming should be provided, and	1148
should be supported by data in the pharmaceutical development section of the	1149
application.	1150
1151
c. 	Instructions for Use34,
	1152
1153
• 	Detailed, step-by-step, appropriately illustrated instructions for patient use	1154
should be included. FDA recommends that the following information be	1155
incorporated into the instructions:	1156
1157
o 	A figure that displays the various elements of the MDI (e.g., actuator, cap,	1158
canister, sleeve, counter).	1159
1160
o 	A statement should be included that the canister should only be used with	1161
the specified accompanying actuator and that the actuator should not be	1162
used with any other inhalation drug product.	1163
1164
o 	A statement instructing the patient to confirm that the canister is fully	1165
seated in the actuator (i.e., mouthpiece or nasal adapter).	1166
1167
o 	A statement instructing the patient to confirm the absence of foreign	1168
objects in the mouthpiece before using the MDI and after removing the	1169
protective mouthpiece cap.	1170
1171
o 	Instructions for initial priming and repriming of the MDI units.	1172
1173
o 	Instructions to provide assurance of coordination of device actuation with	1174
patient inhalation.	1175
1176
o 	A statement cautioning against spraying the eyes with the formulation.	1177
1178
• 	Storage conditions should be stated, including any warning statements	1179
regarding temperature and humidity. A statement should be included	1180
regarding the appropriate temperature of the MDI at the time of use, as well as	1181
any requirements for shaking, if necessary (i.e., for suspension products).	1182
Any preferred storage orientation should be noted.	1183
1184
• 	If protective secondary packaging was used for the MDI product, appropriate	1185
statements should be included that the content of the protective secondary	1186
34 Instructions for Use: Typically these are developed as part of the Human Factors Engineering Design and Risk
Mitigation analysis. For additional information, see: FDA guidance for industry and staff, Applying Human Factors
and Usability Engineering to Optimize Medical Device Design, and FDA draft guidance for industry, Human
Factors Studies and Related Clinical Study Considerations in Combination Product Design and Development.
When final, this guidance will represent the FDA’s current thinking on this topic.

Contains Nonbinding Recommendations
Draft — Not for Implementation
33
packaging should not be used after a specified number of days (e.g., 2 weeks,	1187
30 days) from the date the protective package was opened (in-use period).	1188
1189
• 	Cleaning instructions should be included, if appropriate.	1190
1191
• 	A statement should be included that the canister should be discarded when the	1192
labeled number of actuations has been used.35 Also, a statement should be	1193
included that the correct amount of medication in each inhalation cannot be	1194
ensured after the labeled number of actuations even though the canister may	1195
not be completely empty.	1196
1197
• 	Warning statements required under 21 CFR 369.21 (e.g., storage above 120ºF	1198
may cause bursting, keep out of reach of children, do not puncture, do not use	1199
or store near heat or open flame, never throw container into fire or incinerator,	1200
do not spray into eyes).	1201
1202
d. 	Container Labels and Carton Labeling	1203
1204
In addition to the information typically required to be included on the container label and/or	1205
carton labeling under Title 21, the container label should include the following information	1206
specific to MDI products:	1207
1208
• 	Amount of the drug delivered per actuation from the mouthpiece/nosepiece	1209
and the valve.	1210
1211
• 	Number of usable actuations per container.	1212
1213
• 	Recommended storage conditions including any warning statements regarding	1214
temperature and humidity.	1215
1216
• 	Use period once the MDI product is removed from protective packaging (if	1217
applicable).	1218
1219
• 	The instruction “Shake well before using” for suspension formulations.	1220
1221
• 	A statement that the MDI product should only be used with the mouthpiece	1222
provided (e.g., “For oral inhalation with (Drug Product Name) actuator	1223
only”).	1224
1225
• 	Reference to the patient’s Instructions for Use and additional instructional	1226
statements (e.g., instructions for initial priming and repriming the MDI unit,	1227
inhalation instructions, instructions pertaining to protective caps, etc.).	1228
1229
35 See FDA guidance for industry Integration of Dose-Counting Mechanisms into MDI Drug Products.

Contains Nonbinding Recommendations
Draft — Not for Implementation
34
• 	Warning statements required under 21 CFR 369.21 (e.g., storage above 120°F	1230
may cause bursting, keep out of reach of children, do not puncture, do not use	1231
or store near heat or open flame, never throw container into fire or incinerator,	1232
do not spray into eyes).	1233
1234
In the case of small labels, only some of the information listed above must be included on the	1235
label (21 CFR 201.10(i)). However, all labeling information required by the FD&C Act and the	1236
regulations in Title 21 of the Code of Federal Regulations must be included on the carton, outer	1237
container, wrapper, and leaflet as appropriate.	1238
1239
2. 	DPIs	1240
1241
The labeling of oral DPIs should state the established name of the product as (Drug) Inhalation	1242
Powder and provide the strength as the amount per metered dose unit. For nasal DPIs, the	1243
product labeling should state the established name of the product as (Drug) Nasal Powder and	1244
provide the strength as the metered dose. The established name and strength should be followed	1245
by a phrase such as “For oral inhalation only” or “For nasal inhalation only,” as appropriate.	1246
1247
In addition to the information typically required under Title 21 for the description of the drug	1248
substance and formulation (21 CFR part 201), the product labeling should include the following	1249
information specific to DPI products:	1250
1251
a. 	DESCRIPTION Section of the Prescribing Information	1252
1253
• 	A description of the appearance of the actuator and cap.	1254
1255
• 	The metered amount of medication to be delivered to the patient should be	1256
expressed:	1257
1258
o 	For example: “Each metered amount of ‘x’ mg of formulation contains	1259
‘y’ mcg of drug equivalent to ‘w’ mcg of drug substance (if applicable)	1260
and ‘z’ mg of carrier excipient(s).”	1261
1262
o 	If special circumstances require additional statements regarding the	1263
metered amount, this should be discussed with the appropriate review	1264
division.	1265
1266
• 	Specified TDD from the mouthpiece under defined in vitro conditions should	1267
be stated:	1268
1269
o 	For example: “The drug product delivers ‘y’ mcg of drug with an in vitro	1270
flow rate of 60 L/min for a collection time of 2 seconds (2 L total	1271
volume).”	1272
1273
• 	A statement should be included that the amount of drug delivered to the lungs	1274
will depend on patient factors, such as inspiratory flow and PIF through the	1275

Contains Nonbinding Recommendations
Draft — Not for Implementation
35
delivery system, which may vary for asthma, COPD, and other patient	1276
populations.	1277
1278
• 	A list of all excipients should be included. Substances should be identified by	1279
their established names.	1280
1281
• 	If the drug substance that exits the mouthpiece is a hydrate, solvate, or	1282
complex, this information should be clearly specified with proper strength	1283
conversion for the active moiety.	1284
1285
• 	For DPIs that contain lactose, a statement should be included that the	1286
formulation may contain residual amounts of milk-related proteins.	1287
1288
• 	The number of usable actuations per container if appropriate.	1289
1290
b. 	HOW SUPPLIED Section of the Prescribing Information	1291
1292
• 	The net weight of the container contents should be stated for device-metered	1293
DPIs.	1294
1295
• 	The number of medication actuations expected throughout the shelf life of the	1296
drug product should be indicated. Qualifying terms such as “at least” and	1297
“approximately” should not be used.	1298
1299
• 	Protective secondary packaging (e.g., foil overwrap) should be described. In	1300
addition, appropriate statements should be included that the content of the	1301
secondary protective packaging should not be used after a specified number of	1302
days (e.g., 2 weeks, 30 days) from the date the protective package was	1303
compromised (in-use period).	1304
1305
• 	Storage conditions should be stated, including any warning statements	1306
regarding temperature, humidity, and light.	1307
1308
• 	A brief description of the appearance and color of the body, cap, and other	1309
markers of the device constituent part should be provided, particularly for ease	1310
of identification of different strengths of drugs delivered by the same device	1311
constituent part.	1312
1313
• 	A statement should be included that the DPI unit should be discarded when	1314
the labeled number of actuations has been used, if appropriate.	1315
1316

Contains Nonbinding Recommendations
Draft — Not for Implementation
36
c. 	Instructions for Use36	
1317
1318
• 	Detailed, step-by-step, appropriately illustrated instructions for patient use	1319
should be included. Important elements of the DPI (e.g., body, cap, other	1320
instructive markings such as arrows depicting direction or alignment, etc.)	1321
should be clearly identified with illustrations.	1322
1323
• 	Storage conditions should be stated, including any warning statements	1324
regarding temperature, humidity, and light.	1325
1326
• 	If secondary protective packaging (e.g., foil overwrap) is used for the DPI	1327
product, device constituent part, or unit dose container, this should be stated.	1328
Appropriate statements should be included that the content of the secondary	1329
protective packaging (e.g., device-metered DPIs, pre-metered DPIs) should	1330
not be used after a specified number of days (e.g., 2 weeks, 30 days) from the	1331
date the secondary protective packaging was opened (in-use period).	1332
1333
• 	For device-metered DPIs without a locking mechanism, a statement should be	1334
included stating that the correct amount of medication in each inhalation	1335
cannot be ensured after the labeled number of doses, even though the device-	1336
metered DPI may not be completely empty. A statement recommending that	1337
the device-metered DPI be discarded after the labeled number of doses has	1338
been delivered should be included as well.	1339
1340
• 	Cleaning instructions should be included, if appropriate.	1341
1342
d. 	Container Labels and Carton Labeling	1343
1344
In addition to the information typically required to be included on the container label and/or	1345
carton labeling under Title 21, the container label should include the following information	1346
specific to DPI products:	1347
1348
• 	Amount of the drug per metered unit.	1349
1350
• 	Number of usable actuations per container or device-metered DPI or for the	1351
device constituent part (if re-used), as appropriate.	1352
1353
• 	Recommended storage conditions including any warning statements regarding	1354
temperature, humidity, or light.	1355
1356
• 	Use period once the DPI product is removed from protective packaging, if	1357
applicable.	1358
1359
36 Ibid

Contains Nonbinding Recommendations
Draft — Not for Implementation
37
• 	A statement that the DPI product should only be used with the device	1360
constituent part provided, where applicable (e.g., “For oral inhalation with	1361
(Drug Product Name) actuator only”).	1362
1363
• 	Any special dispensing instructions for the pharmacist and additional	1364
statements for the physician, if applicable.	1365
1366
• 	Reference to the patient’s Instructions for Use and additional instructional	1367
statements (e.g., loading instructions for pre-metered DPIs, inhalation	1368
instructions, instructions pertaining to protective caps, etc.).	1369
1370
In the case of small labels, only some of the information listed above must be included on the	1371
label (21 CFR 201.10(i)). However, all labeling information required by the FD&C Act and the	1372
regulations in Title 21 of the Code of Federal Regulations must be included on the carton, outer	1373
container, wrapper, and leaflet as appropriate.	1374
1375

Contains Nonbinding Recommendations
Draft — Not for Implementation
38
V. 	APPENDIX	1376
1377
A. 	Tables	1378
1379
Table A. General Relationship Between QTPP Elements and CQAs for MDIs	1380
1381
CQA
QTPP Elements
Strength
(Emitted Dose) 	Purity
Aerodynamic
Performance 	Stability
Assay 	X 	X 	X
Purity Profile 	X 	X
Delivered Dose 	X 	X 	X 	X
Aerodynamic Particle
Size Distribution
(APSD)
X 	X
Spray Pattern/Plume
Geometry
X
Leachables 	X 	X
Amount of
Excipients/Formulation
X 	X
Foreign Particulate
Matter
X 	X
Moisture Content 	X 	X
Net Contents 	X 	X 	X
Device Constituent
Part (dimensions, etc.)
X 	X
1382
Table B. General Relationship Between QTPP Elements and CQAs for DPIs	1383
1384
CQA
QTPP Elements
Strength
(Metered Dose) 	Purity
Aerodynamic
Performance 	Stability
Assay 	X 	X 	X
Purity Profile 	X 	X
Delivered Dose 	X 	X 	X 	X
Aerodynamic Particle
Size Distribution
(APSD)
X 	X
Spray Pattern/Plume
Geometry
X
Volatile/Semi-volatile
Leachables
X 	X
Amount of
Excipients/Formulation
X 	X

Contains Nonbinding Recommendations
Draft — Not for Implementation
39
CQA
QTPP Elements
Strength
(Metered Dose) 	Purity
Aerodynamic
Performance 	Stability
Foreign Particulate
Matter
X 	X
Moisture Content 	X 	X
Net Contents 	X 	X 	X
Device Constituent
Part (dimensions, etc.)
X 	X
1385
Table C. Typical MDI and DPI Product Specifications, CQAs and Stability	1386
Attributes.	1387
1388
Attribute 	MDI
CQA for
MDIs 	DPI
CQAs for
DPIs 	Stability
Description 	X 	X 	X
Identification 	X 	X 	X
Assay 	X 	X 	X 	X 	X
Impurities and Degradation Products 	X 	X 	X 	X 	X
Valve Delivery (Shot Weight) 	X 	X
Delivered Dose Uniformity (DDU) 	X 	X 	X 	X 	X
Aerodynamic Particle Size
Distribution (APSD)
X 	X 	X 	X 	X
Spray Pattern 	X 	X 	X
Particulate Matter 	X 	X 	X 	X
Microbial Limits 	X 	X 	X 	X 	X
Leachables (Stability) 	X 	X 	X
Water or Moisture Content 	X 	X 	X 	X 	X
Alcohol Content* 	X 	X 	X
Net Content (Fill) Weight 	X 	X 	X 	X 	X
* When present	1389
1390
B. 	MDI and DPI Product Characterization Studies (P2)	1391
1392
As stated in section IV.C. (Pharmaceutical Development (P2)), summary data from various MDI	1393
or DPI product characterization studies should be provided in the application. Table 2.	1394
Characterization Studies lists the recommended studies. Some detail for each of these studies is	1395
provided below in section B.2. Unless otherwise indicated, the studies should be conducted on	1396
the to-be-marketed configurations and versions of MDI and DPI products. A minimum of three	1397
batches using the formulation and device constituent part of the to-be-marketed configuration	1398
and version should be studied to support the reliability of the manufacturing processes and the	1399
reproducibility of product performance.	1400
1401
1402

Contains Nonbinding Recommendations
Draft — Not for Implementation
40
1. 	General Considerations for Significant Change	1403
1404
For any of the characterization studies described in this section that involve stability testing,	1405
significant change should be considered:	1406
1407
• 	In general, failure to meet the acceptance criteria for any attribute normally tested	1408
during stability studies.	1409
1410
• 	For assay, a change from the initial value of five percent or more.	1411
1412
• 	For DDU, a change in the mass of the mean dose of 10 percent or more (determined	1413
separately on samples taken from the beginning and end of product life), or a failure	1414
to meet the acceptance criteria for first tier testing.	1415
1416
• 	For APSD, a change in the total mass of fine particles (e.g., particles less than five	1417
micrometers) more than 10 percent.	1418
1419
• 	For the description, changes such as: discoloration or other changes in the	1420
appearance of the contents, distortion of valve components, valve clogging or	1421
malfunction, canister corrosion, and adherence of the drug to the walls of the	1422
container or valve components.	1423
1424
2. 	Recommendations for Specific Characterization Studies	1425
1426
a. 	In-Use Period	1427
1428
The purpose of these studies is to demonstrate, for MDI or DPI products marketed in protective	1429
secondary packaging, that the product will perform in accordance with its specification for the	1430
entire length of the proposed in-use period after the protective secondary packaging is opened.	1431
1432
Study Design: Conduct stability studies under intermediate conditions (e.g., 30°C/65 percent	1433
RH) on samples of the MDI or DPI product with the protective secondary packaging opened.	1434
Measure appropriate parameters (e.g., DDU, APSD, water content). Include samples of product	1435
at the beginning and near the end of its proposed shelf life. It is recommended that the study	1436
duration period be twice the proposed in-use period.	1437
1438
b. 	Temperature Cycling	1439
1440
The purpose of these studies is to demonstrate that fluctuating changes in temperature and	1441
humidity (such as those encountered during shipping and handling) will not have an adverse	1442
effect on the quality and performance of the MDI or DPI product. This information, in	1443
conjunction with stability data, should support the proposed storage conditions in the product	1444
labeling.	1445
1446
Study Design: Conduct cycling studies for 3-4 weeks using two different storage conditions, one	1447
subzero (–10 to –20°C) and the other above room temperature (40ºC). Cycle between these	1448

Contains Nonbinding Recommendations
Draft — Not for Implementation
41
conditions every 12 hours. (Alternative conditions and durations can be used, if they can be	1449
justified.) Compare test results to results from control samples (stored under the proposed long-	1450
term storage conditions as opposed to the temperature cycling conditions) tested at the same	1451
intervals.	1452
1453
c. 	Priming and Repriming	1454
1455
The purpose of these studies is to support the instructions in the product labeling for priming	1456
(how many times a patient should actuate a unit before initial use) and repriming (how many	1457
times a patient should actuate a unit before using it again after defined periods of rest).	1458
1459
Study Design: Measure the delivered amount of drug substance from consecutive actuations of	1460
individual units after defined resting intervals (e.g., 0, 6, 12, 24, 48 hours, 3 days). After each	1461
resting interval, repeat actuations until the delivered amount of drug substance per actuation	1462
consistently meets the acceptance criteria for DDU. Test units at the beginning and near the end	1463
of the proposed shelf life. If resting orientation affects the results, test units in various resting	1464
orientations (upright and inverted, or upright and horizontal). Testing can be performed	1465
concurrently on separate samples with progressively longer resting periods.	1466
1467
d. 	Effect of Patient Use	1468
1469
The purpose of these studies is to confirm that the MDI or DPI product functions properly after	1470
repeated patient uses of the product.	1471
1472
Study Design: Collect a number (e.g., 50-100) of partially used product units (including units	1473
near the labeled number of actuations) from clinical studies and measure appropriate parameters	1474
(e.g., DDU and APSD) and dose counter function. Also collect and investigate any MDI or DPI	1475
products that were reported as malfunctioning.	1476
1477
e. 	Effect of Storage and Shaking (suspension formulated MDIs only)	1478
1479
The purpose of these studies is to confirm that shaking instructions in the product labeling for	1480
suspension formulated MDIs are adequate to assure satisfactory dose delivery performance.	1481
1482
Study Design: Measure appropriate parameters (e.g., DDU and APSD) on the MDI product	1483
stored for increasing periods of time subsequent to shaking. From the results, determine the	1484
maximum allowable use time after shaking. Include the effects of shaking duration and storage	1485
orientation if these factors affect the results.	1486
1487
f. 	Effect of Orientation of the DPI Product on Delivered Dose	1488
1489
The purpose of these studies is to support any statements made in the product labeling about the	1490
DPI product orientation during use.	1491
1492
Study Design: Measure appropriate parameters (e.g., DDU and APSD) for DPI products	1493
actuated while oriented at various angles.	1494

Contains Nonbinding Recommendations
Draft — Not for Implementation
42
1495
g. 	Drug Deposition on Mouthpiece and/or Accessories	1496
1497
The purpose of these studies is to determine the amount of drug deposited within the device	1498
constituent part during use, which can relate to cleaning requirements.	1499
1500
Study Design: Measure the mean amount of drug deposited per actuation on the mouthpiece or	1501
other device constituent part components (e.g., spacers or valved holding chambers).	1502
1503
h. 	Cleaning Instructions	1504
1505
The purpose of these studies is to confirm that any cleaning instructions (method and frequency)	1506
included in the product labeling for the device constituent part components (e.g., actuator or	1507
mouthpiece) will assure that the product maintains its ability to deliver the labeled dose of drug	1508
upon use.	1509
1510
Study Design: Measure appropriate parameters (e.g., DDU and APSD) for product actuated	1511
according to a schedule that simulates patient use, including cleaning (if required). Include units	1512
both at the beginning and near the end of shelf life.	1513
1514
i. 	Profiling of Actuations Near Device Exhaustion	1515
1516
The purpose of these studies is to confirm that the product delivers the labeled number of doses,	1517
and to characterize delivery performance if the product is used beyond the labeled number of	1518
actuations (for MDIs and device-metered DPIs that do not lock after the labeled number of	1519
actuations).	1520
1521
Study Design: Measure appropriate parameters (e.g., DDU and APSD) for product units that	1522
have already delivered the number of doses listed on the product labeling until no more drug is	1523
delivered upon actuation. Include units at both the beginning and end of shelf life. Include a	1524
graphical presentation of the findings as part of the study results.	1525
1526
j. 	Effect of Varying Flow Rate on DPI Performance	1527
1528
The purpose of these studies is to characterize the sensitivity of the delivery performance of DPIs	1529
to variation in inspirational flow rates that can be achieved by the patient population that is to use	1530
the product. This information is used to confirm the chosen design of the device constituent part.	1531
1532
Study Design: Using a flow rate range and volume consistent with the intended patient	1533
population, measure appropriate parameters (e.g., DDU and APSD) as a function of flow rate at	1534
the recommended constant volumes.	1535
1536
1537

Contains Nonbinding Recommendations
Draft — Not for Implementation
43
k. 	Effect of Flow Rate and Inhalation Delay on MDIs with Spacers	1538
1539
The purpose of these studies is to characterize the sensitivity of the delivery performance to	1540
variation in inspirational flow rates and inhalation delay on MDI products used with a spacer or	1541
holding chamber.	1542
1543
Study Design: Using a flow rate range and volume consistent with the intended patient	1544
population, measure appropriate parameters (e.g., DDU and APSD) as a function of flow rate at	1545
the recommended constant volumes. Also assess the effect of increasing waiting periods (e.g., 0,	1546
5, 10 seconds) between actuation and initiation of in-flow. For breath-activated MDI products,	1547
determine the ranges of flow rates that generate actuations containing the label claim amount of	1548
delivered dose and the corresponding acceptable APSD.	1549
1550
l. 	Robustness	1551
1552
The purpose of these studies is to confirm that the MDI or DPI product is of sufficiently robust	1553
design to withstand shipping conditions and typical patient usage.	1554
1555
Study Design: Subject a number of units to actions (e.g., dropping, agitation, shipping) that will	1556
simulate conditions the product could be exposed to after it is released, including during patient	1557
use. Determine the effect of these actions on MDI or DPI product performance by measuring	1558
DDU, APSD, and dose counter function.	1559
1560
C. 	Approaches to Evaluating Delivered Dose Uniformity (DDU)	1561
1562
1. 	Parametric Tolerance Interval Testing (PTIT)	1563
1564
FDA recommends that applicants establish test parameters (e.g., sample size, tolerance interval	1565
factor (k factor)) and acceptance criteria that will ensure, to a confidence level of 95 percent, that	1566
at least 90 percent of the units in a batch (i.e., the coverage) will meet the established upper and	1567
lower limits (i.e., 80-120 percent of TDD).	1568
1569
FDA recommends the use of a two one-sided tolerance interval (TOSTI) test and a two-tiered	1570
approach to setting acceptance criteria:	1571
1572
First tier acceptance criteria:	1573
1574
• 	For pre-metered DPIs, measure the amount of drug substance discharged from the	1575
mouthpiece as a percentage of TDD (X) from n units and calculate the mean (X¯) and	1576
standard deviation(s). For MDIs and device-metered DPIs, measure the initial dose	1577
and the last of the labeled doses for each of the n units for a total of 2*n	1578
measurements. The batch passes if:	1579
1580

Contains Nonbinding Recommendations
Draft — Not for Implementation
44
X	¯ + (k1 · s) ≤ 120
and
X	¯ - (k1 · s) ≥ 80
where k 1 is the one-sided tolerance interval factor,
which depends on the sample size (n for pre-metered
DPIs and 2*n for MDIs and device-metered DPIs),
coverage, and confidence level.
Second tier acceptance criteria:	1581
1582
• 	If the batch does not pass tier one testing, measure the amount of drug substance	1583
discharged from the mouthpiece as a percentage of TDD (X) from an additional m	1584
units for pre-metered DPIs (or the initial dose and the last of the labeled doses for	1585
each of the additional m units for MDIs and device-metered DPIs), combine the	1586
results with those from the first tier, and calculate the mean (X¯) and standard	1587
deviation(s). The batch passes if:	1588
X	¯ + (k1 · s) ≤ 120
and
X	¯ - (k1 · s) ≥ 80
where k 1 is the one-sided tolerance interval factor,
which depends on the sample size (n+m for pre-
metered DPIs and 2*(n+m) for MDIs and device-
metered DPIs), coverage, and confidence level.
Table D below shows k1 factors specific to a confidence level of 95 percent and 90 percent	1589
coverage for a range of sample sizes.	1590
1591
Table D. One-sided Tolerance Interval Factors (k 1 ) for 95	1592
Percent Confidence Level and 90 Percent Coverage37	1593
1594
Sample Size 	k1 Factor
10 	2.911
15 	2.566
20 	2.396
25 	2.292
30 	2.220
35 	2.167
40 	2.125
50 	2.065
60 	2.022
90 	1.940
120 	1.899
240 	1.819
480 	1.766
1595
FDA recommends that applicants adopt a 1:3 ratio for the change in sample size from tier 1 to	1596
tier 2 testing (i.e., if n = 10, then n+m = 30). Furthermore, FDA recommends that the tier 1	1597
sample size should not be smaller than ten units. Applicants can set sample sizes much larger	1598
than n = 10 and n+m = 30, but should consider the cost/benefit ratio before setting sample sizes	1599
37 For example, see Hahn and Meeker, Statistical Intervals – A Guide for Practitioners, John Wiley and Sons, Inc.
1991.

Contains Nonbinding Recommendations
Draft — Not for Implementation
45
that are very large. FDA believes that tier 2 sample sizes larger than about 90 (i.e., n+m = 90)	1600
will provide very little additional benefit.	1601
1602
2. 	Counting Test	1603
1604
The application of a counting test for evaluating DDU was in use before other approaches such	1605
as PTIT were developed. If an applicant chooses to evaluate DDU using a counting test, the	1606
Agency recommends a two-tiered approach to acceptance criteria, as described below:	1607
1608
First tier acceptance criteria:	1609
1610
• 	For pre-metered DPIs, the amount of drug substance measured by the test is not	1611
outside 80-120 percent of TDD for more than 1 determination (out of 10). For MDIs	1612
and device-metered DPIs, for each of 10 units, the initial dose and the last of the	1613
labeled doses are measured. The amount of drug substance is not outside 80-120	1614
percent of TDD for more than 2 determinations (out of 20).	1615
1616
• 	The amount of drug substance measured by the test is not outside 75-125 percent of	1617
TDD for any determination.	1618
1619
• 	For pre-metered DPIs, the mean is not outside 85-115 percent of TDD.	1620
1621
• 	For MDIs and device-metered DPIs, the mean of separate determinations made for	1622
the initial dose from each unit and the mean of separate determinations made for the	1623
last of the labeled number of doses for each unit are not outside 85-115 percent of	1624
TDD.	1625
1626
If the above acceptance criteria are met, the batch passes the test for DDU.	1627
1628
If the amount of drug substance is outside 80-120 percent of TDD in more determinations than	1629
are permitted by the first criterion, testing can be performed on 20 additional units to determine	1630
if the batch meets second tier acceptance criteria, provided that the other two criteria described	1631
above are met.	1632
1633
Second tier acceptance criteria:	1634
1635
• 	The amount of drug substance measured by the test is not outside 80-120 percent of	1636
TDD for more than 3 of 30 determinations for pre-metered DPIs or for more than 6 of	1637
60 determinations for MDIs and device-metered DPIs (30 for the initial dose from	1638
each unit and 30 for the last of the labeled number of doses for each unit).	1639
1640
• 	The amount of drug substance measured by the test is not outside 75-125 percent of	1641
TDD for any determination.	1642
1643
• 	For pre-metered DPIs, the mean is not outside 85-115 percent of TDD.	1644
1645

Contains Nonbinding Recommendations
Draft — Not for Implementation
46
• 	For MDIs and device-metered DPIs, the mean of separate determinations made for	1646
the initial dose from each unit and the mean of separate determinations made for the	1647
last of the labeled number of doses for each unit are not outside 85-115 percent of	1648
TDD.	1649