Research progress on substitution of in vivo method(s) by in vitro method(s) for human vaccine potency assays

ABSTRACT

Introduction

Potency is a critical quality attribute for controlling quality consistency and relevant biological properties of vaccines. Owing to the high demand for animals, lengthy operations and high variability of in vivo methods, in vitro alternatives for human vaccine potency assays are extensively developed.

Areas covered

Herein, in vivo and in vitro methods for potency assays of previously licensed human vaccines were sorted, followed by a brief description of the background for substituting in vivo methods with in vitro alternatives. Based on the analysis of current research on the substitution of vaccine potency assays, barriers and suggestions for substituting were proposed.

Expert opinion

Owing to the variability of in vivo methods, the correlation between in vivo and in vitro methods may be low. One or more in vitro method(s) that determine the vaccine antigen content and functions, should be established. Since the substitution involves with the change of critical quality attributes and specifications, the specifications of in vitro methods should be appropriately set to maintain the efficacy of vaccines. For novel vaccines in research and development, in vitro methods for monitoring the consistency and relevant biological properties, should be established based on reflecting the immunogenicity of vaccines.

KEYWORDS:

• Human vaccines
• potency
• in vivo potency
• in vitro potency
• method substitution
• 3Rs

1. Introduction

Vaccines are the most economical and effective tools to prevent and control infectious diseases, and their roles in the prevention and control of major infectious diseases have been highlighted during the recent coronavirus disease 2019 (COVID-19) pandemic and the monkeypox outbreak. To date, vaccines that aid in preventing infectious diseases (e.g. smallpox, poliomyelitis, and influenza) have been licensed globally (Table 1). According to ingredients and manufacturing processes, vaccines can be divided into live-attenuated, inactivated, subunit, nucleic acid, viral vector, recombinant, conjugate, and combined vaccines.

Table 1. Classification of vaccines licensed for human use [1–3].

Classification	Bacterial vaccines	Viral vaccines
Inactivated vaccines	Pertussis vaccine (whole cell, absorbed); Leptospira vaccine; Typhoid vaccine; Cholera vaccine	Japanese encephalitis vaccine; Tick-borne encephalitis vaccine; Hemorrhagic fever with renal syndrome bivalent vaccine; Influenza vaccine; EV71 vaccine; Hepatitis A vaccine; rabies vaccine for human use; Poliomyelitis vaccine
Live-attenuated vaccines	Plague vaccine; Anthrax vaccine; Brucellosis vaccine; BCG vaccine; Typhoid vaccine; Cholera vaccine	Japanese encephalitis vaccine; Measles vaccine; Rubella vaccine; Mumps vaccine; Varicella vaccine; Rotavirus vaccine; Hepatitis A vaccine; Poliomyelitis vaccine; Yellow fever vaccine; Herpes zoster vaccine; Influenza vaccine; Smallpox vaccine; Dengue Tetravalent Vaccine; Ebola zaire vaccine
Subunit vaccines	Pertussis vaccine (acellular, absorbed); Typhoid polysaccharide vaccine; Meningococcal polysaccharide vaccines; Pneumococcal polysaccharide vaccine; Diphtheria vaccine (adsorbed); Tetanus vaccine (adsorbed)	Influenza vaccine
Recombinant vaccines	Meningococcal Group B Vaccine	Hepatitis B vaccine; Hepatitis E vaccine; Human papillomavirus vaccine; COVID-19 vaccine; Influenza vaccine; Herpes zoster vaccine
Nucleic acid vaccines	None	COVID-19 vaccine (mRNA); COVID-19 vaccine (DNA)
Viral vector vaccines	None	COVID-19 vaccine (adenovirus vector)
Conjugate vaccines	Haemophilus type b conjugate vaccine; Meningococcal group A, C, W135 and Y conjugate vaccine; Meningococcal group C conjugate vaccine; Pneumococcal polysaccharide conjugate vaccine	None
Combined vaccines	Typhoid and paratyphoid A combined vaccine; Typhoid and paratyphoid A & B combined vaccine; Diphtheria and tetanus vaccine(adsorbed); Diphtheria and pertussis vaccine (adsorbed); Diphtheria, tetanus and pertussisvaccine(adsorbed);Diphtheria, tetanus, pertussis and haemophilus influenzae type b combined vaccine; Haemophilus type b and meningococcal group C conjugatevaccine	Hepatitis A (inactivated) and hepatitis B (recombinant)vaccine; Measles and mumps combined vaccine; Measles and rubella combined vaccine; Measles, mumps and rubella vaccine; Measles, mumps, rubella and varicella vaccine;
	Diphtheria, tetanus, pertussis and hepatitis B (recombinant)vaccine; Diphtheria, tetanus and poliomyelitis vaccine; Diphtheria, tetanus, pertussis and poliomyelitis vaccine; Diphtheria, tetanus, pertussis, hepatitis B (recombinant), poliomyelitis (inactivated) and haemophilus type b conjugated vaccine (absorbed); Diphtheria, tetanus, pertussis, poliomyelitis (inactivated) and haemophilus type b conjugate vaccine (adsorbed); Hepatitis A (inactivated, adsorbed) and typhoid polysaccharide vaccine

COVID-19, coronavirus disease 2019.

Potency is ‘The measure of the biological activity using a suitably quantitative biological assay, based on the attribute of the product which is linked to the relevant biological properties’ [4,5]. The potency of a final lot, which is directly associated with its efficacy, is a critical quality attribute (CQA) for vaccine quality control and release [6]. Typically, vaccine potency assays include in vivo and in vitro methods. In vivo assays require the assessment of protection against a challenge or antibody detection after immunization of animals. In vitro assays generally require the determination of the vaccine immunogen content. Vaccine potency assays are selected depending on vaccine types. In general, the in vitro virus titration method or bacterial content determination is employed for attenuated live vaccines; in vivo assays are adopted for assessing inactivated vaccines; the contents of major immunogenic components are determined for subunit vaccines; in vitro or in vivo assays are utilized for nucleic acid or viral vector vaccines [7–10]. During the initial stage of recombinant vaccine, in vivo assays are used, while confirming the correlation between in vitro and in vivo assays, in vivo assays can be substituted by in vitro assays.

In vivo assays, commonly used to evaluate the quality of the vaccine batch release, are subject to limitations such as high animal demand, lengthy operations, and high variability [11,12]. In 2018, in vivo potency assays utilized 10.8 million laboratory animals, and the number of laboratory animals subjected to in vivo potency assays was 8% of the total number of laboratory animals, which greatly exceeded the number of laboratory animals for whom safety assays were used in the lot release testing in Europe [13]. As the principles of reduction, refinement, and replacement (3 Rs) are extensively promoted and increasingly recognized, the World Health Organization (WHO), regulatory authorities, and vaccine manufacturers remain committed to substituting in vivo methods for vaccine assessment [14–18]. Specifically, the chapter entitled ‘Substitution of in vivo method(s) by in vitro method(s) for the quality control of vaccines’ was adopted in the European Pharmacopoeia (10th Edition) [19], and relevant chapters were introduced in the British Pharmacopoeia 2019 and Indian Pharmacopoeia 2022 (draft).

Owing to the concerted efforts exerted by vaccine manufacturers and regulatory authorities, in vitro assays can be substituted for in vivo potency assays for assaying the potency of batch-manufactured vaccines (i.e. recombinant hepatitis B vaccines [20–23], inactivated hepatitis A vaccines [24], inactivated poliomyelitis vaccines (IPV) [25], and recombinant human papillomavirus (HPV) vaccines [26]). Considering rabies vaccines, inactivated Japanese encephalitis vaccines, and enterovirus 71 (EV71) vaccines, studies examining the reliability of substituting in vivo assays with in vitro assays are ongoing [27–32]. Nonetheless, the progress toward substituting in vivo assays with in vitro assays is slow, and the major underlying factors are described below: 1. The conventional use of and maintenance of the habitual application of in vivo methods for vaccine potency assays; 2. The high variability of in vivo assays hinders the establishment of a good correlation between in vivo assays and in vitro assays; 3. The assay substitution causes changes in CQAs and specifications, and related guidance documents on designing experiments to obtain scientifically relevant data are lacking; 4. Manufacturing strains and processes vary among enterprises for the same variety.

The present review elaborates on the human vaccine potency assays and method characteristics, along with a systematic description of the research background and progress. Based on barriers concerning the substitution of in vivo assays by in vitro assays, substitution-related suggestions are proposed finally.

2. In vivo and in vitro methods for human vaccine potency assays

2.1. In vivo potency assays

In vivo potency assays include two categories, the protection assessment of animals from the challenge and the serum antibody evaluation for the immunized animals [7].

The protection assessment of animals from the challenge, a classical method for testing vaccine potency, involves challenging immunized animals with virulent organism; the potency is calculated by the number of immunized animals surviving the challenge relative to surviving animals using a reference vaccine. This method can reflect the combined effects of vaccine-induced humoral and cellular immune responses. According to the European Pharmacopoeia (10.0^th Edition), the challenge test is used for the potency test of rabies vaccines, inactivated tick-borne encephalitis vaccines, and whole-cell pertussis vaccines (Table 2) [33–35]. Nonetheless, the challenge test is only applicable to vaccines for which a suitable animal model is available, and the consistency between the experimental strains used for challenge and the clinically prevalent bacterial strains should be monitored and validated. In addition, it is difficult for some vaccine-related pathogens to build sensitive and easy-to-handle animal model.

Table 2. Summary of in vivo methods for vaccine potency assay [2,3].

Potency assay items		Vaccine varieties
		Bacterial vaccines	Viral vaccines
In vivo potency	Protection assessment of animals from the challenge	Tetanus vaccine; Plague vaccine; Anthrax vaccine; Brucellosis vaccine; Pertussis vaccine (whole cell); Leptospirosis vaccine; Diphtheria vaccine	Rabies vaccine for human use; Tick-borne encephalitis vaccine
	Serum antibody evaluation	Haemophilus influenzae type b conjugate vaccine; Diphtheria vaccine; Pertussis vaccine (Acellular); Group A and group C meningococcal conjugate vaccine	Japanese encephalitis vaccine (inactivated); Hemorrhagic fever with renal syndrome bivalent vaccine; Hepatitis B (recombinant) vaccine; Hepatitis E vaccine; Herpes zoster vaccine; HPV vaccines (recombinant); EV71 vaccine (inactivated); COVID-19 (recombinant) vaccine; COVID-19 (adenovirus) vaccine; Hepatitis A (inactivated) vaccine; IPV;

COVID-19, coronavirus disease 2019; HPV, human papillomavirus; IPV, inactivated poliomyelitis vaccines.

Since some vaccines are investigated to induce neutralizing antibodies that can inhibit viral entry and infection, the potency of vaccines can be assessed by determining the titer of neutralizing antibodies. The potency of inactivated Japanese encephalitis vaccine can be determined using the aforementioned methods (Table 2).

2.2. In vitro potency assays

Vaccine antigens, capable of stimulating the body to produce immune responses, underpin vaccine efficacy and safety [36–39]. Based on the defined neutralizing epitopes on pathogens and corresponding conformational antibodies with neutralizing activities, relevant in vitro test methods (i.e. enzyme-linked immunosorbent assay (ELISA)) can be established to determine the antigen content for assessing the potency. According to the European Pharmacopoeia (10.0^th Edition), except in vivo potency assays, the corresponding in vitro methods can be established using the vaccine antigen as the testing indicator to determine the potency of recombinant hepatitis B vaccines, inactivated hepatitis A vaccines, IPV, and recombinant HPV vaccines.

Bacterial capsular polysaccharide, the material basis for bacterial invasion and for exerting pathogenic effects, can serve as a major bioactive compound after purification, allowing vaccines to stimulate antibacterial immune responses within the body. As a vaccine component, bacterial capsular polysaccharides can elicit B cell responses and induce the body to produce specific protective antibodies [40,41]. According to the European Pharmacopoeia (10.0^th Edition), the polysaccharide content serves as an indicator to examine the immunogenicity of typhoid polysaccharide, pneumococcal polysaccharide, and meningococcal polysaccharide vaccines [42–44].

3. Research background for substituting in vivo assays by in vitro assays

3.1. Characteristics of in vivo and in vitro assays

The laboratory animal-based in vivo assay, capable of directly reflecting in vivo vaccine protection, is a traditional method for assessing the vaccine potency [45], whereas in vivo assays can encounter several limitations, such as a prolonged testing period, high variability, low power of discrimination, and noncompliance with animal ethics requirements. One or more in vitro method(s), scientifically designed and validated based on immunological, molecular, and physicochemical methods, can afford advantages such as no use of laboratory animals, a short testing period, markedly reduced need for resources, and low variability.

With advances in technology and the strict application of Good Manufacturing Practice (GMP), the classical manufacturing process and quality control level of vaccines have dramatically improved [46,47]. To monitor consistency of production and assess the potential impact of manufacturing changes, the appropriately designed in vitro assays are more suitable than the inherent variability of in vivo assays. Accordingly, the scientific values and relevance of in vivo assays should be continuously investigated, and the method substitution should be performed promptly.

3.2. Establishment and development of 3Rs principles

Laboratory animals, employed as human surrogates, play an irreplaceable role in progressing scientific development. With a surge in the number of laboratory animals employed in the field of life sciences, the suffering and rights of laboratory animals have received considerable attention. In 1959, Russell and Burch systematically proposed the 3Rs principles for the first time in the book title, The Principles of Humane Experimental Technique [48]. The 3Rs are fundamentally defined as replacing animal methods with non-animal, reducing numbers used where possible, and refining experimental methods to eliminate animal suffering. In 1978, the 3Rs were collectively described as ‘alternatives’ by Smyth, which was then extensively accepted [49].

Currently, 3Rs principles have profound impact on the establishment of laws regarding laboratory animals and the implementation of scientific research plans in biomedical research. Specifically, the WHO ‘Guideline for Independent Lot Release of Vaccines by Regulatory Authorities, Annex 2, TRS No. 978’ recommends to use validated in vitro alternatives in the testing process during the lot release of vaccines [50]. In 1986, the European Union issued ‘The European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes (ETS No. 123)’ to reduce the number of animal experiments, as well as the number of animals employed, encouraging the application of alternative methods to animal experiments [51].

3.3. Regulatory guidelines for substituting in vivo assays with in vitro assays

According to ETS No. 123, the European Pharmacopoeia Commission is committed to reduce the use of animals wherever possible in pharmacopoeial testing. A new chapter entitled ‘Substitution of in vivo method(s) by in vitro method(s) for the quality control of vaccines’ was adopted in the European Pharmacopoeia (10.0^th Edition) when compared with the European Pharmacopoeia (9.0^th Edition). The chapter states that inherently variable in vivo assays are inferior to the scientifically designed and validated in vitro methods when the production consistency and potential effects of production changes are assessed. Based on the 3Rs principles, in vivo assays that found to be limited value should be substituted by in vitro assays. Nonetheless, the inherent variability of in vivo assays and/or the lack of systematically validated data render(s) it challenging to conduct a one-to-one comparative study of in vivo and in vitro assays during method substitution. Based on a good consistency of the manufacturing process, the chapter describes the following suggestions on simplifying the substitution process for potency methods: (1) In vitro assays can scientifically and appropriately monitor consistency of production, and data are needed to confirm that alternatives can control the CQAs of vaccines and maintain the quality of vaccines to be released; (2) In vitro assays should be capable of determining the antigen content and functionality. If a single in vitro method is adopted, it is advisable to select neutralizing monoclonal antibodies targeting the conformational epitopes; if a single in vitro method cannot fully reflect the antigen content and functionality, multiple in vitro methods can be used; (3) During the initial evaluation for establishing an in vitro method, samples at different concentrations and subjected to different types of stress conditions should be used to ensure the quantitative accuracy and stability-indicating potential of the new method; (4) In vitro methods are more suitable for monitoring the production consistency owing to their high sensitivity.

The United Kingdom was directly introduced this chapter into the British Pharmacopoeia 2019 [52]. According to the vaccine situation, Indian Pharmacopoeia 2022 (draft) revised the chapter [53].

4. Research progress on potency assays substitution for human vaccines

4.1. Vaccines for which substitution of potency methods has been completed

4.1.1. Recombinant hepatitis B vaccine

In the initial marketing stage, the in vivo potency assay was conducted in mice to examine the quality of batch-manufactured recombinant hepatitis B vaccines. Given the shortcomings of animal tests, Abbott et al. developed an in vitro assay based on ELISA to quantify the content of recombinant hepatitis B surface antigen which comprises the vaccine and induces the immune response. The results confirm that the in vitro assay is superior to in vivo assay in sensitively detecting changes in sample potency [21–23,54]. According to WHO guidelines and the European and Chinese Pharmacopoeias, in vitro potency assays that monitor the vaccine quality consistency and discriminate reduced-potency vaccines can be employed to determine the quality of vaccine during the lot release process [55].

4.1.2. Inactivated hepatitis A vaccine

A study conducted by Poirier et al. based on reduced-potency inactivated hepatitis A vaccines suggested that the in vitro and in vivo potency results appear related, and the in vitro assay could detect changes in vaccine potency that the in vivo assay failed to detect [56]. According to monographs of the current European Pharmacopoeia, the potency of inactivated hepatitis A vaccines should be determined using in vivo assays that detect the specific antibodies induced in mice or in vitro assays that examine the antigen content in the lot release process [24].

4.1.3. IPV

IPV plays a crucial role in preventing poliovirus. During the initial after licensure, the in vivo assays of IPV were generally performed in the lot release process, namely, the post-immunization neutralizing antibody titers in sera of animals (e.g. Guinea pigs, chickens, and rats) were detected [57]. D antigen, the protective immunogen of IPV, can stimulate the production of protective neutralizing antibodies within the body. Therefore, a D antigen-based in vitro assays can be established and used to ensure consistency throughout production. Based on a well-established production process, in vitro assays can more sensitively detect the degradation of poliomyelitis antigens and are more suitable for evaluating the consistency of batch-manufactured products [58].

According to the European Pharmacopoeia Supplement 5.4, the in vivo assay can be substituted by the D antigen assay, provided that the evaluation results of passing/failing the vaccine batch testing obtained by the D antigen testing and the in vivo assays are consistent. To guide IPV manufacturers and regulatory authorities regarding the complete substitution of in vivo assay by in vitro assay, the guideline for validating D antigen assay was added in Supplement 5.6 [59]. The guideline defines the prerequisites for conducting substitution research, research samples, and result.

4.1.4. Recombinant HPV vaccine

HPV vaccines are composed of L1 proteins expressed in recombinant yeast and insect cells or Escherichia coli, which undergo extraction, separation, purification, and adjuvant incorporation [60]. Merck initially performed in vivo assays in mice to test the quality of clinical trial samples, subsequently establishing and validating in vitro relative potency (IVRP) assays based on the functional epitopes of vaccines and their neutralizing monoclonal antibodies to reduce the number of laboratory animals used [61]. A study assessing normal and reduced-potency vaccines revealed that the in vivo assay in mice and the IVRP assay could detect differences in vaccine potency, and the two assay results were well correlated. Additionally, the reduced-potency samples that failed the IVRP assay, are generally less immunogenic in the human body. Based on the aforementioned results, the IVRP assay is considered suitable for substituting in vivo potency assay in mice and can test the effectiveness of HPV vaccines in the lot release process. Therefore, in vitro assays were included in the European Pharmacopoeia and WHO guidelines [26,62].

4.2. Vaccines for which substitution of potency assays are under investigation

4.2.1. Inactivated rabies vaccine

The National Institutes of Health (NIH) test, a traditional method for testing the potency of rabies vaccines, requires the serial dilution of rabies vaccines to immunize mice and perform the test of protection against the effects of a lethal dose of rabies virus, during which numerous laboratory mice are employed and experience inexplicable suffering and death [63,64]. The European Pharmacopoeia (9.0^th Edition) recommends that the content of glycoprotein in vaccines should be determined by in vitro immunochemical methods substituting potency assays for in vivo challenge upon completion of the comparative study [65].

In vitro alternatives for inactivated rabies vaccines have been investigated for decades, and immunochemical methods that can quantitatively determine the antigen content in vaccines, including antibody binding assay [66,67], single radial immunodiffusion (SRID) assay [68–70], and ELISA [27,45,71,72], have been established. Among them, SRID assay and ELISA are performed by vaccine manufacturers to establish the antigen content in vitro during the manufacturing process. Nonetheless, it has been reported that the SRID assay fails to differentiate highly immunogenic native trimeric glycoproteins from soluble or denatured glycoproteins that are poorly immunogenic. While SRID assay is poorly correlated with the NIH test and inferior to ELISA in terms of the sensitivity [68,69,73]. Multiple studies have confirmed that monoclonal antibodies used in the ELISA can detect natively conformational glycoproteins in a highly sensitive and consistent way and are well correlated with NIH test results [71,74–76]. Although ELISA could be substituted for the NIH test, no one has been approved for human used rabies vaccine by the competent authority. The major reasons are as follows: (1) The NIH test, which can directly reflect the immune responses in the body against vaccines, is the gold standard for testing inactivated rabies vaccines and is recognized by vaccine manufacturers and regulatory authorities; (2) The applications of various types of rabies vaccine viruses and different manufacturing processes can result in different epitopes in the vaccines, which will hinder the establishment of an ELISA suitable for detecting multiple virus types [28].

4.2.2. Inactivated Japanese encephalitis vaccine

During the lot release process, the assay of neutralizing antibodies in immunized mice is typically conducted to assess the potency of inactivated Japanese encephalitis vaccines [77]. An ELISA method has been established based on the E antigen of the Japanese encephalitis virus and corresponding monoclonal antibodies for substituting in vivo methods [32,78]. Reduced-potency samples that obtained by treating at different temperatures were tested using ELISA and in vivo assay in mice, and the observed results were found to be correlated. The in vitro assay (i.e. ELISA) is an alternative to the in vivo assay, and its effects on different virus species need to be further verified.

4.2.3. Inactivated EV71 vaccine

During the R&D stage, in vivo and in vitro assays were performed for quality control of EV71 vaccines. In the lot release stage, EV71 vaccines were tested using in vivo and in vitro assays, and the antigen content in the vaccines positively correlated with the neutralizing antibody titer. Thus, in vitro assays can feasibly substitute in vivo assays. According to the requirements for the potency as say of EV71 vaccines in ‘Recommendations to assure the quality, safety and efficacy of EV71 vaccines (inactivated)’ published by the WHO, in vivo assays may be omitted, provided that the vaccine production consistency has been established for a suitable number of consecutive final bulks and in vitro assay yields consistent results to the in vivo assay [79].

5. Barriers on substituting in vivo methods by in vitro methods for potency assays

5.1. The conventional use and maintenance of habitual use of in vivo assays

The in vivo assay, which can directly reflect complex immune responses in the body against vaccines, is considered the gold standard for vaccine quality evaluation and is extensively applied by vaccine manufacturers and regulatory authorities for the lot release of vaccines [56]. Establishing a new method, including experimental development, pre-validation, validation, independent assessment, and approval by regulatory authorities, is a prolonged process [11]. Considering previously licensed vaccines, enterprises tend to maintain the use of the currently existing in vivo assays.

5.2. High variability associated with in vivo assays

The method substitution requires a one-to-one comparison between a new method and the previously existing method to ensure that the testing standards of the new method are not less rigorous than those of the previously existing method. The in vivo assays results fluctuate considerably, and the in vitro assays are superior to the in vivo method considering the result consistency [80]. Therefore, it can be challenging to conduct a one-to-one comparative study of in vivo and in vitro assays, and international multi-center cooperative studies may fail, given the high variance of the in vivo assays.

5.3. Substituting in vivo assays with in vitro assays can alter CQAs and specifications

In vivo assays examine the serum antibodies in immunized animals or the protection against challenge, while in vitro assays determine the content of the vaccine antigen or bacterial capsular polysaccharide. Therefore, the vaccine quality attributes will be assessed differently if in vivo assays are substituted with in vitro assays. The substitution may hinder by the lack of related guidance documents on designing experiments to obtain scientifically relevant data.

5.4. Challenges involved in developing an in vitro assay suitable for products of the same variety

Given the considerable number of vaccine manufacturers for the same variety and marked variations in strains and processes, establishing an in vitro assay suitable for all products using a single monoclonal antibody can be challenging. The biological reference materials and test methods established by enterprises according to the strain and process characteristics of in-house products encounter considerable challenges for approval by regulatory authorities.

6. Expert opinions

Potency is a CQA throughout vaccine release and post-marketing surveillance. Owing to the vaccine characteristics and history, in vivo assays are predominantly used to assess vaccine potency. With notable advances in technology and the strict application of GMP for vaccine production, novel vaccines are developed using the quality by design approach and the quality lifecycle management approach, with substantial improvements in the classical manufacturing process and quality control level of vaccines [81,82]. Moreover, vaccine production consistency can be ensured [83]. In addition, various novel testing techniques and the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use guidelines for method development and validation are continuously developed and updated [6,84–90]. Considering this setting, in vivo assays with known precision can be challenging to meet the quality control requirements of existing vaccines that demonstrate a high production consistency. Currently, substituting in vivo assays by more accurate and appropriately designed in vitro assays is pivotal for improving the level and accuracy of human vaccine quality control.

Regarding traditional vaccines, establishing a well correlation between highly variable in vivo assays and highly consistent in vitro assays presents to be a considerable challenge. Therefore, the well correlation between in vitro methods and in vivo methods is not mandatory, while one or more scientifically appropriate, antigenicity-related in vitro method(s) focusing on reflection and control of the production consistency should be established. Furthermore, during the process of in vitro methods established, comprehensive validation of the new methods can enhance the success rate of method substitution. In particular, the precision-indicating potential of the new methods should be assessed with samples at different concentrations, and the stability-indicating potential should be assessed with samples subjected to different types of stress conditions. For new vaccines in R&D, relevant studies should be conducted when possible, the in vitro methods that reflect the vaccine content and conformational integrity should be established, and the correlation between in vitro and in vivo methods based on the manufacturing process and immunogenicity should be investigated. If the methods are confirmed to be related to immunogenicity and can effectively monitor the manufacturing process changes, in vitro methods should be further validated with sufficient clinical study data. Additionally, biological reference materials, which are stable for a prolonged period and can preferably be traced to international biological reference materials, should be prepared to ensure the quality of vaccine potency assays. For vaccines containing adjuvants, the dissociation rate and stability of vaccines should be considered. If vaccine dissociation affects in vitro potency assay, a reference vaccine can be included to suitably control the effects of dissociation on potency assays.

Finally, the substitution of in vivo methods with in vitro methods for human vaccine potency assays involves changing testing principles, methods, CQAs, and specifications. Therefore, scientifically appropriate specifications for in vitro methods should be formulated by comprehensively considering variations and stability both in the manufacturing process and testing methods. Clinical trial results should also be considered for developing specifications of in vitro method. Thus, the consistency of vaccine lots could be accurately monitored, and the safety and efficacy of vaccines could be ensured.

Article highlights

• During the process of vaccine batch release, potency is a critical quality attribute (CQA) to control quality consistency and relevant biological properties. Potency assays of vaccines may be conveniently divided into in vivo methods (animal-based) and in vitro methods (non-animal-based).

• Owing to the characteristics and tradition of vaccines, in vivo methods which have high variability are widely adopted for vaccine potency assays.

• With the implementation of GMP and 3Rs principles, in vivo methods with inherent variability, may be considered inferior to monitoring the consistency of production. Therefore, it is imperative to substitute the in vivo methods.

• A single in vitro method based on neutralizing antibodies or multiple in vitro methods can improve the sensitivity and reduce the time required.

• The substitution of in vivo methods with in vitro methods accompanies with the change of CQAs and specifications. The specifications of in vitro methods should be appropriately set to maintain the efficacy of vaccines.

• For novel vaccines in research and development (R&D), in vitro methods for monitoring the consistency and relevant biological properties should be established based on reflecting the immunogenicity of vaccines.

Declaration of interests

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions

X Zhang and X Wu conceived and drafted the manuscript; Q He and J Wang provided valuable discussion; M Xu, Z Liang and Q Mao revised the manuscript. All authors have read and approved the article.

Additional information

Funding

This manuscript was funded by the CAMS Innovation Fund for Medical Sciences (2021-I2M-5–005).