EMA 发布鼻用制剂治疗等效性要求指南草案并公开征求意见
Draft guideline on the requirements for demonstrating therapeutic equivalence for nasal products
EMA 发布鼻用制剂治疗等效性要求指南草案（EMA/220888/2026），于 2026 年 10 月 8 日启动公开征求意见。草案针对含相同活性部分的鼻用制剂，按体外、药代动力学以及药效学与临床的递进方式说明治疗等效性或生物等效性的论证要求，并列出体外比较参数及接受标准。CHMP 于 2026 年 10 月 5 日通过该草案供征求意见，意见截止 2027 年 2 月 28 日。
草案把鼻用制剂治疗等效性判定分为体外、PK 与临床递进路径，并列出各体外参数及接受标准，便于对照征求意见稿定位章节。
PDF 文字版；图形和原始排版请参阅官方 PDF。
第 1 页
17 5 October 2026
18 EMA/220888/2026
19 Committee for Medicinal Products for Human Use (CHMP)
20
21
22 Guideline on the requirements for demonstrating
23 therapeutic equivalence for nasal products
24 Draft
Draft agreed by Immunologic and Inflammatory Diseases Working
July 2026
Party (IIWP)
Adopted by CHMP for release for consultation 5 October 2026
Start of public consultation 8 October 2026
End of consultation (deadline for comments) 28 February 2027
25
26
Comments should be provided using this EUSurvey form. For any technical issues, please contact
the EUSurvey Support.
27
Keywords Nasal, equivalence
28
29
30
Official address Domenico Scarlattilaan 6 ● 1083 HS Amsterdam ● The Netherlands
Address for visits and deliveries Refer to www.ema.europa.eu/how-to-find-us
Send us a question Go to www.ema.europa.eu/contact Telephone +31 (0)88 781 6000 An agency of the European Union
© European Medicines Agency, 2026. Reproduction is authorised provided the source is acknowledged.
第 2 页
31
32 Table of contents
33 Table of contents......................................................................................... 2
34 Executive summary ..................................................................................... 3
35 1. Introduction (background) ................................................................. 3
36 2. Scope .................................................................................................. 3
37 3. Legal basis and relevant guidelines .................................................... 4
38 4. General considerations ....................................................................... 4
39 5. In vitro comparison............................................................................. 6
40 6. Pharmacokinetics ................................................................................ 9
41 6.1. Locally acting drugs ......................................................................... 9
42 6.1.1. Equivalence regarding systemic safety .......................................... 9
43 6.1.2. Equivalence regarding efficacy ...................................................... 9
44 6.2. Systemically acting drugs............................................................... 10
45 7. Pharmacodynamic and clinical studies .............................................. 12
46 7.1. Locally acting drugs ....................................................................... 12
47 7.2. Systemically acting drugs............................................................... 13
48 7.3. Local tolerability............................................................................. 13
49 8. Children and adolescents .................................................................. 14
50 9. Usability studies................................................................................ 14
51 10. Definitions ........................................................................................ 15
52 11. List of abbreviations ......................................................................... 16
53
54
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 2/16
第 3 页
55
56 Executive summary
57 This guideline is intended as a complement to the “Guideline on the pharmaceutical quality of
58 inhalation and nasal products” (EMEA/CHMP/QWP/49313/2005 rev. 1). It addresses the requirements
59 for demonstration of therapeutic equivalence (TE) between nasally administered products containing
60 the same active moiety(ies).
61 Demonstration of TE between locally acting nasal products is based on a stepwise approach, where TE
62 could be demonstrated in vitro, if all in vitro requirements are fulfilled (pharmaceutically equivalent) or
63 else preferably by means of pharmacokinetics (PK) if equivalent systemic exposure (as a surrogate
64 marker for safety) and equivalent local absorption/deposition (as a surrogate marker for efficacy) is
65 demonstrated in spite of some in vitro differences. It is generally difficult to demonstrate TE using
66 pharmacodynamic (PD)or clinical endpoints as these are seldom sensitive enough.
67 The guideline also addresses the requirements for demonstrating bioequivalence (BE) for systemically
68 acting nasal products containing the same active moiety(ies) via BE studies and describes when these
69 studies can be waived based on in vitro data.
70
71 1. Introduction (background)
72 The aim of this guideline is to detail the data requirements for demonstrating TE between nasal
73 medicinal products with local action in the nasal mucosa containing the same active moiety(ies). In
74 addition, data requirements for demonstrating BE for nasal medicinal products with systemic action
75 containing the same active moiety(ies) are covered together with criteria for biowaivers.
76 The guideline is intended to be used together with the guideline on the pharmaceutical quality of
77 inhalation and nasal medicinal products (EMEA/CHMP/QWP/49313/2005 rev. 1) to cover aspects
78 related to demonstration of TE. Thereby, regarding locally acting formulations it is a sister guideline
79 to the Guideline on the requirements for demonstrating therapeutic equivalence between orally
80 inhaled products (OIP) for asthma and chronic obstructive pulmonary disease (COPD)
81 (CPMP/EWP/4151/00 Rev. 2) which covers TE related aspects for OIP products. Regarding
82 systemically acting formulations, reference is made to guidelines for demonstration of BE (Guideline
83 on the investigation of bioequivalence (CPMP/EWP/QWP/1401/98 Rev 1/Corr) and ICH M13A).
84
85 2. Scope
86 This document provides guidance on the requirements for demonstrating equivalence between nasal
87 formulations with the same active moiety(ies), including both single active substance products and
88 combination products. The guideline covers both medicinal products with local action in the nasal
89 mucosa (e.g. xylometazoline, fluticasone, azelastine) and with systemic action (e.g. naloxone,
90 desmopressin, fentanyl, ketamine, zolmitriptan).
91 The guideline focuses on abridged applications, but the principles described may be applicable for all
92 other applications that are based on demonstration of TE compared to a reference product, such as
93 line extensions, variation submissions or during product development. Also, in the case that there is a
94 need to confirm similarity to a product for which literature data is available (e.g., well-established use
95 applications), the same principles apply.
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 3/16
第 4 页
96
97 3. Legal basis and relevant guidelines
98 This guideline should be read in conjunction with the introduction and general principles, part I and II
99 of the Annex I to Directive 2001/83/EC and other pertinent elements outlined in the EU and the
100 International Council for Harmonisation (ICH) guidelines, especially those on:
101 • CPMP/EWP/239/95: Note for guidance on the clinical requirements for locally applied, locally
102 acting products containing known constituents;
103 • EMEA/CHMP/QWP/49313/2005 rev. 1: Guideline on the pharmaceutical quality of inhalation
104 and nasal products;
105 • EMA/CHMP/QWP/BWP/259165/2019: Guideline on quality documentation for medicinal
106 products when used with a medical device;
107 • CPMP/EWP/QWP/1401/98 Rev.1/Corr**: Guideline on the investigation of bioequivalence;
108 • ICH M13A Guideline on bioequivalence for immediate-release solid oral dosage forms
109
110 4. General considerations
111 TE implies that the efficacy and safety profile of the test and reference products are sufficiently
112 comparable so that clinically relevant differences between products can be reliably excluded. The
113 expression is applicable to either in vitro or in vivo data collection.
114 The demonstration of TE between nasally administered locally acting products is based on a stepwise
115 approach. In vitro data is to be provided (see Section 5) with a set of requirements to be fulfilled to
116 allow a conclusion on TE. If similarity cannot be demonstrated in vitro, PK data may be used to
117 support TE for locally acting medicinal products (see section 6). Demonstration of TE based on PD
118 data may be an option, but it is not recommended as it is deemed difficult to ensure assay sensitivity
119 (see Section 7).
120 For nasal medicinal products with systemic action containing the same active moiety(ies),
121 demonstration of bioequivalence of the test product to the reference product is required, generally via
122 the conduct of an in vivo BE study. However, for solutions, a waiver based on in vitro data is possible
123 if the criteria in section 5 described for demonstration of TE are fulfilled. Use of PD or clinical data is
124 not acceptable in case BE cannot be confirmed.
125
126 A schematic overview of the concept is presented in Figure 1.
127
128
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 4/16
第 5 页
129 Figure 1 Flow chart – demonstration of equivalence between nasally administered products
Locally Systemically
acting acting
Solution Non- solution
Compare the test and
reference products in
vitro
Compare the test
Are test and reference product
and reference
equivalent by means of in vitro data?
products in vitro
Are test and reference
product equivalent by means
NO of in vitro data?
Conduct PK studies
Studies should cover total exposure as a Conduct BE study(ies)
surrogate for safety, and local exposure (in a
Is BE demonstrated?
setting with charcoal if GI tract contribution to
absorption is not negligible) as a surrogate for
efficacy.
Is TE demonstrated? equivalent by means of
PK data for all active substances?
YES NO
NO YES
Reformulation or
Reformulation or YES stand-alone (i.e.
stand-alone (i.e. non- non-bridging)
YES
bridging) clinical data clinical data
likely required required
PD/clinical bridging studies may PD/clinical bridging
be used but is not recommended studies may not be used
due to low assay sensitivity. to support TE
EQUIVALENCE DEMONSTRATED
130
131
132
133
134
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 5/16
第 6 页
135
136 5. In vitro comparison
137 The characterisation of the in vitro properties is the first step in the evaluation and demonstration of TE
138 between the test and reference products. In vitro criteria, as specified below, should be studied and, if
139 not all criteria are fulfilled, progression to in vivo studies is needed. For systemically acting products, in
140 vivo BE studies may be waived only for solutions and in case all criteria for demonstration of TE as
141 described below are fulfilled.
142 The in vitro characterisation and comparison are essential and should always be performed irrespective
143 of whether it has already been decided to conduct a PK study. The in vitro comparability exercise
144 should be performed and evaluated as predefined in a study protocol, which should include similarity
145 conditions, methods of comparison and acceptance criteria. Test procedures for the in vitro comparison
146 should be state-of-the-art and demonstrated to be suitable for the intended use. In cases where TE is
147 demonstrated based solely on in vitro data, at least three consecutive batches of the test product and
148 three batches of the reference product should be tested with a minimum of ten units of each batch. If
149 there is a high variability observed in a critical quality attribute (CQA), a larger number of batches
150 and/or more units per batch needs to be tested. To limit variability and maximize reproducibility, it is
151 advised to use automated actuation for testing spray device related parameters. Test product batches
152 should be representative of the product to be marketed and the manufacturing process – i.e. batches
153 at or near production scale. Alternatively, pilot scale batches, at least 1/10 production scale may be
154 used for characterisation and comparative purposes, if there are no changes in the manufacturing
155 process and equipment, however evidence should be provided that scale-up does not affect product
156 quality. The batches of the reference product used in the in vitro equivalence comparison should be
157 representative of the product on the market including consideration of different ages. In case in vitro
158 comparison is only supportive for TE, and in vivo study(ies) are required, a smaller data set is
159 considered sufficient, e.g., three batches and five units of test respectively reference product.
160 For a nasal medicinal product, claiming TE to a reference medicinal product, the package of studies
161 required depends on the pharmaceutical dosage form. The CQAs that may have an impact on the
162 efficacy and safety of the drug product, and consequently need to be compared, should be defined and
163 justified. Also, the attributes not considered critical (and therefore do not need to be compared) should
164 be discussed and justified. Parameters specified in Table 5.1 below should be considered, but other
165 parameters may also be relevant to use depending on the specific medicinal product characteristics.
166 TE is sufficiently demonstrated if the test product fulfils all relevant in vitro criteria as compared to the
167 reference product.
168 Table 5.1. In vitro parameters for demonstration of TE between test and reference nasal products.
In vitro parameter for Nasal spray Nasal drops
Nasal powder
demonstration of TE Solution Suspension Solution Suspension
i. Qualitative and quantitative
composition, and Yes Yes Yes Yes Yes
pharmaceutical form
ii. Handling of the device Yes Yes Yes Yes Yes
iii. Particle size distribution of
No Yes No Yes Yes
active substance
iv. Other physical characteristics
No Yes No Yes Yes
of the active substance
v. Delivered dose Yes Yes Yes Yes Yes
vi. Physicochemical properties of
Yes Yes Yes Yes Yes
the formulation
vii. Droplet/particle size
Yes Yes No No Yes
distribution of the product
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 6/16
第 7 页
viii. Spray pattern and plume
Yes Yes No No Yes
geometry
ix. Mass in droplets/particles
Yes Yes No No Yes
smaller than 10 µm
169
170 i. Qualitative and quantitative composition, and pharmaceutical form
171 The test product should contain the same active substance(s) as the reference product (i.e., same salt,
172 ester, hydrate or solvate). The excipients should preferably be qualitatively the same and quantitatively
173 similar. In case of any qualitative and/or quantitative difference, it must be adequately justified that
174 this difference does not influence the relevant CQAs and/or any aspect of product performance,
175 including residence time, absorption, safety and local tolerability.
176 The pharmaceutical dosage form should be the same (e.g., nasal spray, nasal drops, nasal powder)
177 with the same dispersion state of the active substance (e.g., solution, suspension).
178 ii. Handling of the device
179 The handling of the devices for the test and reference products in order to release the required amount
180 of the active substance should be similar.
181 iii. Particle size distribution of the active substance
182 If the active substance is in the solid state, the particle size distribution should be similar. If next to the
183 active substance, other substances in the solid state are part of the formulation, then a method should
184 be used that is able to discriminate between active substance particles and other particles. D10, D50,
185 D90 and span ((D90-D10)/D50) should be evaluated. In case of a normal distribution curve, comparison
186 of D50 and span would be sufficient. If aggregates/agglomerates are present, the study should also be
187 conducted on the product after actuation. To conclude on similarity the 90% confidence interval (CI)
188 for the geometric mean ratio of the test and reference products should be contained within the
189 acceptance criteria of ±15%, assuming log-normal distribution of data (85-118%).
190 iv. Other physical characteristics of the active substance
191 If the active substance is in the solid state (powder, suspension), any difference e.g., in crystalline
192 structure and/or polymorphic form should not influence the performance of the product. If applicable,
193 the percentage of amorphous material in crystalline active substance and the rugosity of the particles
194 surface should be compared.
195 v. Delivered dose
196 The mean delivered dose should be similar. For solutions, delivered mass may be sufficient for
197 comparison. For the method of evaluation of similarity, see point iii.
198 vi. Physicochemical properties of the formulation
199 Depending on the pharmaceutical form, different physicochemical properties may be critical and
200 relevant to compare. For example, viscosity enhancers may be included to prolong the residence time
201 in the nasal cavity. For a suspension, wetting agents may be used to reduce surface tension and
202 thereby enhancing particle dispersion. Key parameters to be considered for evaluation and comparison
203 may be rheological properties (e.g., thixotropy, viscosity), surface tension, pH, density, osmolality and
204 buffer capacity. The physicochemical properties should be thoroughly discussed and justified.
205 For quantitative quality physicochemical characteristics, the 90% CI for the geometric mean ratio of
206 the test and reference products should be contained within the acceptance criteria of ±10%, assuming
207 log-normal distribution of data (90-111%). In case the reference product variability is higher than
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 7/16
第 8 页
208 10%, the acceptance range might be widened up to ±20% (80-125%) depending on the reference
209 product variability, using scaled-average-equivalence according to [U, L] = exp [±k·sR], where U is the
210 upper limit of the acceptance range, L is the lower limit of the acceptance range, k is the regulatory
211 constant set to 1.056 and sR is the standard deviation of the log-transformed values of the reference
212 product parameters. Table 5.2 gives examples of how different levels of variability lead to different
213 acceptance limits using this methodology.
214 Table 5.2. Acceptance limits depending on reference product coefficient of variation (CV).
CV Acceptance range
10.0 ±10% 90-111%
12.5 ±12.3% 87.7-114%
15.0 ±14.6% 85.4-117%
17.5 ±16.8% 83.2-120%
20.0 ±18.9% 81.1-123%
21.365 ±20% 80-125%
215
216 Alternatively, for quantitative quality physicochemical characteristics not related to strength, a CQA-
217 range (based on the underlying distribution for the reference product) may be proposed, using the
218 principle that all reference product batches on the market represent acceptable quality. For
219 physicochemical CQAs, demonstrating similarity using the overlap of the test distribution with this
220 reference range may be a more appropriate approach than using a mean with a fixed percentage
221 variation as the range of acceptance. Applicants are recommended to apply for scientific advice from a
222 competent authority and seek endorsement of their strategy if this is their approach of choice.
223 vii. Droplet/particle size distribution of the product
224 For a nasal spray, the droplet size distribution, and for nasal powders, the particle size distribution of
225 the product, should be compared using a non-aerodynamic method, for example, laser diffraction. D10,
226 D50, D90 and span ((D90-D10)/D50) should be evaluated separately and tested at two different, justified
227 distances from the laser beam (e.g., between 2 and 7 cm), and at least 3 cm apart. In case of a
228 normal distribution curve, comparison of D50 and span would be sufficient. For the method of
229 evaluation of similarity, see point iii.
230 viii. Spray pattern and plume geometry
231 Spray pattern and plume geometry describes characteristics of the emitted spray plume in terms of
232 size and shape. The spray device properties such as valve, pump and actuator may have a significant
233 impact on the spray pattern. The spray pattern should be determined using a single actuation and is
234 normally tested at two distances (e.g., between 3 and 7 cm) at least 3 cm apart. Dmin, Dmax, ovality
235 ratio (Dmax/Dmin) and area should be evaluated for the spray pattern measurements. Plume angle and
236 plume width should be evaluated for the plume geometry measurements. The analysis of the plume
237 geometry should be performed during the fully developed phase of the spray and preferably when the
238 spray is still in contact with the tip.
239 For all parameters, a statistical evaluation should be performed. The 90% CI for the geometric mean
240 ratio of the test and reference products should be contained within the acceptance criteria of ±20%,
241 assuming log-normal distribution of data (80-125%).
242 ix. Mass of droplets/particles smaller than 10 µm
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 8/16
第 9 页
243 Small droplets/particles (less than 10 µm) are more likely to be swallowed or inhaled and thereby
244 deposited in the gastrointestinal (GI) tract or the lungs rather than in the nasal cavity. Therefore, the
245 mass of active substance in droplets/particles smaller than 10 µm is considered a CQA for a nasal
246 spray/powder and should be compared. A cascade impactor, or an abbreviated impactor with a cut-off
247 plate insert for 10 µm, configured for nasal use, is recommended. For solutions, laser diffraction may
248 be used. For the test product, the mass of active substance in droplets/particles smaller than 10 µm,
249 or for solutions the fraction of droplets smaller than 10 µm, should not be more than +10% compared
250 to the reference product. The 90% CI for the geometric mean ratio of the test and reference products
251 should not be above +10% (one sided), assuming log-normal distribution of data (not larger than
252 111%).
253 6. Pharmacokinetics
254 6.1. Locally acting drugs
255 For locally acting medicinal products, if TE has not been demonstrated based on in vitro data as
256 discussed in section 5, TE may be demonstrated using in vivo PK studies as described below. See
257 Section 6.1.2 for general recommendations regarding the PK study (study design, study population,
258 test and reference products, dose or strength to be tested, sampling schedule considerations).
259
260 It may however be necessary to administer more than one dose due to low plasma concentrations and
261 analytical limitations, provided that participant-safety is adequately ensured and justified.
262
263 The same batches should be used for the efficacy and safety PK studies, whenever feasible.
264
265 6.1.1. Equivalence regarding systemic safety
266
267 i. Study design
268 In order to investigate systemic safety, the total systemic exposure for the test and reference product
269 should be compared in a PK study as outlined in Section 6.2.
270
271 ii. Primary PK parameters and acceptance criteria
272 To support safety, it is sufficient to demonstrate that the systemic exposure is not higher for the test
273 product than for the reference product, i.e., the upper limit of the 90% CI for the ratio of the test and
274 reference product for AUC(0-t) and Cmax should not exceed the upper BE acceptance limit of 125.00%.
275 For Cmax the recommendations in the Guideline on the investigation of BE regarding widening
276 acceptance criteria for Cmax based on high intra-individual variability can be followed.
277
278 iii. Additional considerations
279 Safety assessments including monitoring of adverse events should be included in the PK study. For
280 additional considerations on local tolerability, refer to section 6.3.
281
282 6.1.2. Equivalence regarding efficacy
283 i. Study design
284 In case the contribution from the GI tract to the total systemic exposure following nasal administration
285 is negligible (<5%) a PK study without charcoal blockade can be used for both efficacy and safety
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 9/16
第 10 页
286 comparisons. A low oral absolute bioavailability per se is, however, not synonymous with a negligible
287 systemic contribution from GI absorption, since the contribution from the GI tract depends on the
288 fraction of the dose being deposited in the nose and being swallowed, respectively, as well as on the
289 fraction absorbed into the systemic circulation from each site. For investigation of efficacy, a PK study
290 with activated charcoal should thus be performed, unless it can be justified that the contribution from
291 the GI tract to the total systemic bioavailability following nasal administration is negligible (<5%).
292
293 It should be noted that the contribution from the GI tract is generally larger for nasally administered
294 locally acting products than for orally inhaled products due to larger fraction of the dose being
295 swallowed. Therefore, it is not possible to claim negligible contribution from the GI tract following nasal
296 administration based on conclusions on negligible GI tract contribution following oral inhalation.
297
298 It is acknowledged that systemic exposure in a study with activated charcoal may not fully reflect local
299 nasal availability (and nasal efficacy), e.g. since absorption could partly occur via the lung from small
300 particles/droplets. However, the site of action and main site of absorption will be the nasal mucosa and
301 similar PK exposure is expected to imply comparable behaviour of the two formulations. Since use of
302 PK is more sensitive than a clinical endpoint study to detect differences between test and reference
303 product, this approach is recommended to demonstrate TE regarding efficacy.
304
305 The charcoal blockade efficiency in terms of binding, charcoal dose and frequency of administration
306 needs to be demonstrated (e.g., by in vitro binding studies and using a method that has been shown to
307 be effective in the literature).
308
309 In case the study with activated charcoal would not demonstrate measurable absorption, a PK study
310 would not be relevant in order to conclude on TE regarding efficacy.
311 If systemic exposure is known or expected to be non-measurable, PK data would not provide relevant
312 information. In such cases, it is recommended to seek scientific advice to discuss possible alternative
313 ways to demonstrate TE.
314
315 ii. Primary PK parameters and acceptance criteria
316 TE with regard to efficacy can be concluded if the 90% CI for the ratio of the test and reference
317 products is contained within the acceptance interval of 80.00-125.00 for AUC(0-t) and Cmax in the study
318 with activated charcoal blockade. A widening of the acceptance criteria for Cmax based on high intra-
319 individual variability in line with the recommendations in the Guideline on the investigation of
320 bioequivalence, may be possible. Any deviations from these acceptance criteria must be thoroughly
321 justified and can never be accepted for results below the lower limit of the acceptance range or when
322 data on safety is generated from a study using different test and/or reference batches.
323
324 6.2. Systemically acting drugs
325 For medicinal products intended for systemic action, demonstration of BE of the test product to the
326 reference product is required, generally via the conduct of an in vivo BE study. However, for solutions,
327 a waiver based on in vitro data is possible if the criteria in section 4 are fulfilled.
328 The guidance below is mainly applied for cases where both test and reference products are nasally
329 administered formulations. The same general principles may apply also in case the reference product
330 has a different route of administration (e.g. oral or intramuscular). However, in those cases there may
331 be a need for additional data relevant for the new route of administration compared to the approved
332 route.
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 10/16
第 11 页
333 i. Study design
334 Regarding study design considerations, primary PK parameters and acceptance criteria etc, reference
335 could, for most aspects, be made to the Guideline on the investigation of bioequivalence
336 (CPMP/EWP/QWP/1401/98 Rev 1/Corr) and ICH M13A Guideline on bioequivalence for immediate-
337 release solid oral dosage forms. However, it should be kept in mind that the latter GL is mainly
338 dedicated for drugs absorbed by the gastro-intestinal tract, while absorption is expected to occur
339 mainly at the nasal mucosa for systemically acting nasally administered drugs. Therefore, some items
340 in relation to the particularity of the oral route may not be relevant here (e.g. a potential need for both
341 a fasted and a fed study).
342 An open (bioanalytical laboratory blinded) randomized single-dose cross-over study is recommended
343 and deemed to be appropriate in most cases. However, alternative designs, such as replicate crossover,
344 parallel group or repeated-dose administration, could be accepted if justified as described in the BE
345 guidelines referred to above.
346
347 ii. Study population
348 To reduce variability, BE studies should be conducted in healthy volunteers. The outcome of the
349 investigations in healthy volunteers could be extrapolated to patients for which the drug is intended. If
350 the investigated active substance is known to have adverse effects and the pharmacological effects or
351 risks are considered unacceptable for healthy subjects, the study may instead be conducted in a
352 targeted patient population under suitable precautions and supervision. Healthy volunteers should be
353 understood here as subjects with normal general health conditions but also with no chronic or transient
354 nasal affections.
355 If the product is self-administered by subjects, it is critical that all subjects included are properly
356 trained to administer the product correctly in line with the product information and also to confirm
357 during the study that subjects administer the product correctly. Also, in case of administration by study
358 staff, correct administration should be confirmed. If administration is not correctly performed, subjects
359 should be excluded. Decision on exclusion should be made before bioanalysis.
360
361 iii. Investigational test and reference products
362 The batch of the reference product should be representative of the commercial batches available on
363 the market. How the representative batch(es) is chosen should be fully justified. Characterisation of
364 several batches of the reference product should be performed. A minimum of five batches may be
365 sufficient if suitably justified. However, if the reference product shows great variability or quality
366 changes over storage, a larger number of batches is needed. The delivered dose of the reference
367 batch(es) chosen for the in vivo study(ies) should be as close as possible to the calculated median of
368 the observed reference product batches. A deviation within ±15% is reasonable.
369 The test product should be representative of the product to be marketed. Production scale batches are
370 preferred but pilot scale batches may be used, if there are no changes in the manufacturing process
371 and if it is demonstrated that scale-up does not affect product quality.
372
373 iv. Dose or strength to be tested
374 The dose to be administered should follow the recommendations of the reference product as
375 presented in the product information.
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 11/16
第 12 页
376 If two or more strengths are available, it may be sufficient to include only one strength in the BE
377 study. The following requirements must be met:
378 a) The different strengths of the test product are manufactured by the same manufacturing
379 process.
380 b) The different strengths of the test product have the same qualitative and similar quantitative
381 composition. For example, the lower strength is achieved by a formulation with lower
382 concentration without any changes in actuator, metering valve or pump.
383 c) For solutions, the droplet size distribution should be similar for the test product strengths. For
384 suspensions and powders, the droplet size and the particle size distribution should be similar
385 for the test product strengths. For similarity criteria, see section 5.
386
387 v. Sampling schedule considerations
388 It is critical that the sampling schedule is planned so that Cmax can be reliably estimated and that it
389 can be avoided that Cmax is observed in the first sample post-dose. The sampling schedule should
390 cover the plasma concentration - time curve long enough to provide a reliable estimate of the extent
391 of exposure, which is achieved if AUC(0-t) covers at least 80% of AUC(0-∞).
392 If justified, a suitably truncated AUC can be used instead of AUC(0-t) for drugs with long terminal
393 elimination half-life. When such an approach is claimed, the decision for sampling cutoff time should
394 be defined prospectively in the study protocol and take into account the local residence time of the
395 drug before being cleared from the nasal mucosa. It should be noted that the truncation of AUC at 72
396 hours mentioned in ICH M13A is based on GI transit time. However, in most cases, this sampling time
397 would likely be sufficient also for nasally administered products.
398
399 vi. Primary PK parameters and acceptance criteria
400 The Cmax and AUC(0-t) should be evaluated. BE can be concluded if the 90% CI for the ratio of the test
401 and reference products is contained within the acceptance interval of 80.00-125.00 for AUC(0-t) and
402 Cmax. Intraindividual variability may be an issue for drug administered by nasal route. For Cmax the
403 recommendations in the Guideline on the investigation of bioequivalence regarding widening of
404 acceptance criteria for Cmax based on high intra-individual variability can be followed.
405 For drugs where the early onset of action is clinically relevant, an additional primary PK parameter
406 such as area under the concentration vs. time curve between two specific time points (pAUC) or tmax
407 should be used to establish BE as recommended in ICH M13A.
408
409 7. Pharmacodynamic and clinical studies
410 7.1. Locally acting drugs
411 As the objective is to demonstrate TE with the reference product rather than efficacy and safety of the
412 test product, PK endpoints are considered more sensitive to detect differences between test and
413 reference product than PD or clinical endpoints. For locally acting medicinal products, if TE has not
414 been demonstrated based on in vitro data (see section 5) or in an in vivo PK study (see section 6), it is
415 generally recommended to reformulate the product.
416
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 12/16
第 13 页
417 However, if other approaches with PD or clinical endpoints are considered, the study design must be
418 such that assay sensitivity is clearly shown at an acceptable level. Appropriate efficacy endpoints for TE
419 may include objective measures of nasal congestion/nasal airway resistance and/or inflammation,
420 depending on the drug’s mechanism of action. Subjective endpoints may be included as co-primary or
421 secondary endpoint, but use of only subjective endpoints is not acceptable. Safety assessments
422 including monitoring of adverse events should always be included in the efficacy studies regardless of
423 the design.
424
425 Regardless of the objective of the study, it is necessary to demonstrate that the sensitive part of the
426 dose/response curve for the PD/clinical parameter under investigation has been studied. To allow
427 estimating assay sensitivity, it is essential to include at least one non-zero dose level besides the dose
428 level(s) primarily investigated.
429
430 As for the PK studies (see section 6.2.2), the same batch of reference product should be used for
431 safety and efficacy PD studies, unless adequately justified, and should be representative of the product
432 on the market (see section 6.1). When feasible, it is of value to have access to PK data from the PD
433 studies.
434
435 It is recommended that the statistical analysis allows calculating relative potency. The relative potency
436 of the test product to the reference product is defined as the dose of the test product that produces the
437 same biological response as one unit of the dose of the reference product. This analysis should be
438 conducted based on the approach by Finney (1964)1 for the primary efficacy variable, unless
439 otherwise justified. The acceptance criteria for the 90% CI of the relative potency should be
440 prespecified and normally retained within 0.67 to 1.50. To support TE, it should be clearly shown that a
441 certain strength of the test product is more similar to the same strength of the reference product than
442 the closest adjacent differing higher or lower strength (anticipated to differ by a factor 2 irrespective of
443 whether there is an approved such strength or not). Any other choice of statistical approach must be
444 sensitive enough to ensure assay sensitivity at this level.
445
446 7.2. Systemically acting drugs
447 In line with the principles for generic development of oral formulations, for generic nasal products of
448 systemically acting drugs, PD or clinical data are not acceptable in case of deviation from BE. If a
449 waiver based on in vitro data, as discussed in section 5, is not applicable and BE has not been
450 demonstrated in an in vivo BE study, it is generally recommended to reformulate the product, or a full
451 stand-alone clinical data package would need to be provided to support a MAA instead of a BE
452 approach.
453
454 7.3. Local tolerability
455 Local tolerability of nasal medicinal products is an important consideration as some drugs or
456 formulations can irritate the nasal mucosa, leading to discomfort, sneezing, or nasal discharge. In
457 general, local tolerability may be ascertained by knowledge of the active substance and the choice of
458 well-established excipients.
1
Finney DJ. Statistical methods in biological assay. London: 104:1057–61. Griffin, 1964
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 13/16
第 14 页
459 Equivalence with respect to quality (see section 5 above) provides assurance of safety and local
460 tolerability if the excipients are qualitatively the same and quantitatively similar. However, if the
461 excipients are not qualitatively the same or not quantitatively similar, impact on the local safety profile
462 should be addressed in a local tolerability study in humans or thoroughly justified based on literature,
463 including e.g., irritation, changes in nasal secretions (discharge) or breathing (congestion), and nasal
464 mucosal appearance. Consideration should also be given to particular aspects of medicinal product use
465 e.g., posology and duration of treatment.
466
467 8. Children and adolescents
468 In case of a new nasal delivery device, not previously approved for children but where the reference
469 product has a paediatric indication, data on usability needs to be provided (see section 9). The
470 characteristics of the delivery device may be such that the device is more difficult to use for a child
471 than for an adult and, therefore, the child or their caregiver is less able to use the device correctly, or
472 the child/caregiver may use the device differently from an adult. Such differences in the handling of
473 the product by a child/caregiver may result in a different risk/benefit relationship in the child compared
474 with that seen in the adult. In case it has been shown that the device can be correctly handled and
475 emptied by children/caregivers and TE/BE has been demonstrated (see sections 5 and 6 above), the
476 lowest approved age for the test product could be set at the same as the reference product without
477 further data or justification.
478
479
480 9. Usability studies
481 For medicinal products for which the medical device and/or device part and the medicinal product form
482 an integral product that is not reusable (hereafter called integral), a formal usability study (also named
483 human factor study) may be required to demonstrate safe and effective use of the integral medicinal
484 product by the intended user population as stated in the ‘Guideline on quality documentation for
485 medicinal products when used with a medical device’ (EMA/CHMP/QWP/BWP/259165/2019), section
486 5.4.
487
488 For such studies, study participants should be recruited to include a number of distinct user groups
489 including patients (adults, and where appropriate children and adolescents) and caregivers, within
490 which both reference product-naïve and experienced users should be included. A minimum of 15
491 participants should be recruited in each distinct user group. Participants selection for these studies
492 should ensure representativeness of the intended users population incorporating general population
493 trends (e.g., left handedness, elderly, patient with manual coordination difficulties, such as arthritic
494 patients). The study protocol should direct participants in simulating the use of the new nasal delivery
495 device to deliver doses as per normal use using an empty or placebo device, unless a different study
496 setting is justified. The exercise should include the unpacking of a new device from the patient pack,
497 simulated delivery of the first dose, as well as the intended storage of the device. Participants should
498 be asked to simulate the delivery of further doses to assess the user interface with the device
499 throughout its life. Areas of focus should allow ensuring that the user understands key features of the
500 device. Clear acceptance criteria should be pre-specified together with an accompanying rationale in
501 the protocol. The outcome of this summative usability study should be reported in the form of a
502 usability report that should include details such as the intended use, observed risks, and study results
503 as well as its corresponding appendices, including the study protocol.
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 14/16
第 15 页
10. Definitions
Abridged application Application (generic, hybrid, biosimilar)
submitted under Articles 10(1), 10(3) and
10(4) of Directive 2001/83/EC
Active moiety The molecule or ion responsible for the
physiological or pharmacological action of the
drug substance
Assay sensitivity Ability of a clinical trial to distinguish an
effective treatment from a less effective
treatment or ineffective treatment.
Delivered dose The quantity of drug substance that is available
to the user, ex-device in one spray or actuation.
Dose/Single dose Amount of drug administered on a single
occasion. One dose may consist of several
actuations.
Metered dose The quantity of drug substance contained in the
delivery device metering chamber.
Reference product A product against which therapeutic
equivalence is claimed/assessed.
Relative potency The relative potency of the test product to the
reference product is defined as the dose of the
test product that produces the same biological
response as one unit of the dose of the
reference product (i.e., comparative outcomes
for different doses).
Single dose study Study involving a single administration of each
dose level under investigation.
Strength/dose Strength refers to the amount of active
substance metered or delivered in a single
actuation (e.g., 125 microgram per actuation of
a nasal spray). Dose, by contrast, refers to the
total amount of active substance administered
on one occasion. For example, one actuation in
each nostril (e.g., 2 x 125 microgram).
Therapeutic equivalence The performance of the test and reference
products is sufficiently comparable (within the
acceptance criteria outlined in this guideline) to
ensure negligible impact on efficacy or safety.
For systemically acting formulations, the term
bioequivalence is normally used, but in this
guideline, therapeutic equivalence may refer
also to systemically acting formulations.
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 15/16
第 16 页
11. List of abbreviations
AUC Area Under the Curve
BE Bioequivalence
CI Confidence Interval
Cmax Peak concentration
D10, D50, D90 Droplet/particle diameter at which 10%, 50% or
90% are smaller
Dmin, Dmax Minimum or maximum diameter of the spray
plume area
GI Gastrointestinal
ICH International Conference on Harmonisation
PD Pharmacodynamic
PK Pharmacokinetic
TE Therapeutic equivalence
tmax Time to peak concentration
Guideline on the requirements for demonstrating therapeutic equivalence for nasal
products
EMA/220888/2026 Page 16/16
来源：EMA Scientific Guidelines · ema.europa.eu