CENTER FOR DRUG EVALUATION, NMPA

　　October 2024

1. INTRODUCTION

　　In recent years, lentiviral vectors have been widely used in clinical practice, which can be directly used as vector products in human , or can mediate the transfer and expression of target genes in target cells (such as T cells, NK cells, stem cells, etc.) to prepare in vitro genetically modified cells for human use. The lentiviral vector systems currently used in clinical practice have been modified and optimized, such as deleting nonessential genes, deleting long terminal repeat sequences and transcription initiation related regions, reducing homology between transfer plasmids and packaging plasmid sequences, reducing homology between plasmid vector / helper gene sequences and packaging cell DNA, and separating helper gene sequences into different plasmids to avoid homologous or non-homologous recombination to generate replication competent lentiviruses (RCL). Although these modifications greatly reduce the risk of RCL generation, considering the insufficient research on the generation mechanism and structure of RCL, as well as the potential risks to product quality, clinical risks to patients, and public biosafety risks that RCL may generate, regulatory agencies still consider RCL detection as an important an important aspect of lentiviral vector quality research and control.

　　Stable lentiviral vector packaging cell lines are difficult to prepare and obtain. Typically, multiple plasmids are transiently transfected into cells for lentiviral vector production. Currently, lentiviral vectors based on the HIV-1 viral backbone and using vesicular stomatitis virus glycoprotein (VSV-G) for envelope replacement are widely used in clinical applications. The following content aims to propose general technical requirements for RCL detection in such lentiviral vector systems, but does not include technical requirements for RCL detection in ex vivo transduced cells.

　　Based on the current situation of domestic declarations of cell and gene therapy products, most of the RCL (replication competent lentivirus) tests are entrusted to third-party testing agencies, while some applications adopt in-house developed methods. Since targeted technical guidelines for RCL testing have not yet been promulgated in China, there are many frequently asked questions in aspects such as materials for testing, amounts for testing, assays for testing, and method validations. The Center for Drug Evaluation has formulated the technical requirements for these questions by combining the preliminary research findings and the requirements of relevant foreign regulatory technical guidelines, for the reference of applicants and testing agencies.

　　This technical requirement only reflects the current views and cognition of regulatory authorities. With the development of science and technology and the accumulation of regulatory knowledge and experience, relevant content will be continuously improved and updated.

2. FREQUENTLY ASKED QUESTIONS AND ANSWERS

2.1 Materials for Testing Question 1: During the production stage of lentiviral vectors, should the viral vector supernatant and the end of production cells be tested simultaneously or should one of them be selected for testing? Is it required to test each production batch of the viral vector supernatant and the end of production cells for RCL?

Answer:

Lentiviral vectors are usually produced by transient transfection of cells with multiple plasmids. According to the generation mechanism of RCL, during the packaging stage of viral vectors, the transfer plasmid and the packaging plasmid may undergo homologous sequence recombination in the cells to form a virus with replication competence. In addition, the plasmid vector and the genome of potential endogenous viral particles in the cells may also undergo homologous or non-homologous recombination to generate a virus with replication competence. Considering that the risk of RCL generation due to gene recombination may vary between batches, and studies have found that RCL is not always detected consistently in the viral vector supernatant and end of production cells. Therefore, to ensure that there is no RCL risk during the viral vector stage, it is recommended that each batch of the viral vector supernatant and the end of production cells for clinical trials and commercial production use should be tested for RCL using the cell culture assay.

Question 2: What are the general considerations for selecting materials for RCL testing during the production stage of lentiviral vectors?

Answer:

Generally, materials that is most likely to detect RCL should be selected for testing. Considering that downstream purification processes may destroy any RCL present in the sample, when conditions permit, the unprocessed viral vector supernatant is preferred for testing when conditions permit.

　　According to the actual commercial production process of lentiviruses and the requirements for sample amount calculation, it may happen that the sampling volume of the unprocessed viral vector supernatant is relatively large, and the existing testing conditions and capabilities cannot meet the requirements. Therefore, samples at appropriate stages (such as the concentrated viral vector supernatant) or the final viral products can be selected for RCL testing based on the characteristics of the viral vector production process. However, evaluation and verification should be carried out according to specific situations to ensure the reliability and accuracy of the testing methods. It should be noted with special attention that the concentrated viral vector supernatant or the final viral products with a relatively high titer may be toxic to the amplifying cells. Therefore, a reasonable preliminary experiment should be designed in the test to evaluate the interference of the test samples on the testing methods.

2.2 Amounts for Testing Question 1: How to calculate the sampling amount of viral vector supernatant?

Answer:

Based on the early experience of using retroviral vectors (including production experience and clinical application experience), it is recommended to take at least 5% of the unprocessed viral vector supernatant for RCL testing. However, with the optimization and development of the process and the continuous accumulation of clinical application experience, it is recommended to test sufficient amount of the viral vector supernatant to ensure a 95% probability of detection of RCL if present at a concentration of 1 RCL/dose equivalent. According to the calculation formula in the current FDA guidelines, when the viral vector is used as an in vivo gene therapy product, the volume of the final viral product to be tested needs to be calculated by substituting the titer of the final viral product into the formula. When the viral vector is used as an ex vivo gene modification system, the volume of the final viral product to be tested needs to be calculated by substituting the titer of the final viral product, the multiplicity of infection, the number of ex vivo transduced cells, etc. into the formula. If the samples to be tested are the unprocessed viral vector supernatant or the concentrated viral vector supernatant, since there are differences between the titers of the above samples and the titer of the final viral product, a larger volume of the unprocessed viral vector supernatant or the concentrated viral vector supernatant needs to be taken to ensure that the sampling amount meets the possibility of detecting 1 RCL per dose with a probability of 95%. The actual testing volume is recommended to be calculated by substituting the actually measured titer of the unprocessed viral vector supernatant or the concentrated viral vector supernatant into the formula. The formula is as below:

　　Vt = -(1 / (1 RCL / dose equivalent)) ln(1 - 0.95)

　　Vt represents the sample testing volume. Dose equivalent: For the viral vector directly used as an in vivo gene therapy product, it is defined as the as the maximum amount of vector expected to be administered at one time; For ex vivo genetically modified cells, a dose equivalent is the amount of vector used to transduce the maximum number of target cells for each production lot.

　　Considering that in the early stage of product research and development, the information about the maximum dose of lentiviruses transducing ex vivo cells may not have been fully confirmed, and it may not be possible to obtain the maximum volume of the viral vector supernatant to be tested based on the calculation formula. In this case, it is also possible to consider using 5% of the unprocessed viral vector supernatant for testing.

　　The testing samples, such as the unprocessed viral vector supernatant, the concentrated viral vector supernatant, and the final viral product, should be clearly stated in the application documents. The application documents should also provide in detail the process and basis for calculating the sample testing amount to ensure that the sampling amount meets a 95% probability of detection of RCL if present at a concentration of 1 RCL/dose equivalent.

Question 2: How to calculate the amount of end of production cells for testing?

Answer:

The end of production cells refer to the cells isolated during the stage of harvesting the viral vector supernatant or the cells finally obtained after multiple harvests of the viral vector supernatant. The testing amount is usually calculated as 1% of the total cell number or 1 × 10? cells (whichever is less).

2.3 Assays for Testing Question 1: Is it feasible to detect RCL by cell culture assay just through continuous amplification culture without adding an indication phase after the amplification phase?

Answer:

Viruses lacking accessory genes may have a slower growth rate compared to wild-type viruses. Moreover, the residual p24 protein and plasmid residues (such as VSV-G) in lentiviral vectors may cause false positive results. In addition, in the absence of RCL, low-level sequence transfer (such as psi-gag) may occur in cells during the amplification phase, resulting in false positive results. Therefore, adding an indication phase of culture after the amplification phase for further testing can ensure the accuracy and reliability of the test results as much as possible. If there is sufficient data during the amplification phase to prove that continuous subculture can eliminate the false positive risks caused by residual proteins (such as p24 protein) and residual plasmid , or if the end-point detection method is not affected by plasmid and protein residues, or if cytopathic effects can be observed in the sample groups during the amplification phase, then the indication phase may not be necessary. If sufficient data cannot be provided, it is recommended to co-culture the viral vector supernatant and end of production cells with a permissive cell line (such as C8166 cells) for at least 5 passages to amplify any potential RCL. After the amplification culture, centrifuge the co-culture, and then co-culture the collected supernatant with C8166 cells again as the indication phase. Finally, collect the samples from the indication phase for RCL testing.

Question 2: Is it feasible to select only one endpoint detection method for RCL testing by cell culture method?

Answer:

Due to the complex generation mechanism of RCL and unclear structure, several common endpoint detection methods currently available are all based on the theoretically assumed structure of RCL to select endpoint indicators to indirectly reflect the presence of RCL. Since different endpoint detection methods have different applicability, applicants are encouraged to use at least two endpoint detection methods with different principles for RCL detection, to mutually verify the authenticity and reliability of the detection results. Common endpoint detection methods include sequence detection (such as psi-gag, VSV-G), reverse transcriptase activity detection (PERT), and functional protein detection (such as p24 protein), etc., It should be noted that if both endpoint detection results are positive, it can generally be judged as RCL positive. If one endpoint result is positive and the other is negative, sufficient investigation and research documents need to be provided to confirm the reliability of the detection results.

Question 3: What are the general considerations for selecting positive control viruses in detection methods? What research data is required for a positive virus?

Answer:

Most of the envelope proteins of lentiviruses used in clinical practice are VSV-G. Theoretically, it is more reasonable to use HIV with the envelope protein with VSV-G as the positive virus than using wild-type HIV or attenuated HIV. However, the HIV positive virus with the VSV-G envelope protein poses a relatively large hazard to personnel and the environment. At present, there is not sufficient evidence to show that RCL has similar growth and physicochemical characteristics to HIV with the VSV-G envelope protein. Therefore, it is recommended to select wild-type HIV, attenuated HIV, or recombinant conditional replication lentiviral vectors that meet the requirements for the sensitivity of RCL verification as positive control viruses, ensuring the representativeness of the positive control viruses.

　　Positive viruses are crucial for confirming the sensitivity of detection methods. Therefore, it is recommended that detailed information on the source, preparation process, main functional elements, and biological titer of the positive virus should be provided in the application materials, and attention should be paid to the stability of the genetic characteristics, growth characteristics, and biological titer of the positive virus during storage.

Question 4: What are the requirements for establishing the experimental group? Is it necessary to establish an inhibitory control group in the experimental group?

Answer:

When testing the sample using a validated method, it is usually necessary to establish a positive control group, a negative control group, an inhibitory control group, and a sample group. The positive control group and the negative control group are crucial for evaluating the specificity and reproducibility of the method. Therefore, reasonable positive control and negative control groups should be established in the experimental group.

　　Since the samples used in the validation/verification of the RCL method may differ from the samples used in the actual test in terms of virus titer, matrix components, etc., and research has found that high-titer lentivirus supernatant may have an inhibitory effect on the amplification of co-cultured cells, thereby affecting the sensitivity of RCL detection. Meanwhile, components other than the virus in the test samples may also affect the detection of RCL. Therefore, it is recommended to establish an inhibitory control group during actual sample testing to prove that the samples to be tested have no impact on the detection of the positive virus.

Question 5: What are the general considerations for selecting co-cultured cells in cell culture methods? Is it necessary to establish a cell bank and conduct characterization for co-cultured cells?

Answer:

The selection of co-cultured cells is related to the envelope of the virus. Meanwhile, it is also necessary to consider whether the cells can support the replication of the virus. Usually, cell lines that are susceptible to and easy to replicate the virus are selected as co-cultured cells. Currently, the C8166 cell line is a relatively recognized cell line that is susceptible to HIV viruses with the envelope protein replaced by VSV-G. It is recommended to preferentially select the C8166 cell line. If other cell lines are selected, comprehensive methodological studies should be carried out, and the equivalence of the method should be verified before they can be considered as co-cultured cells.

　　Comprehensive research materials in accordance with the relevant requirements for cells used in testing in the Chinese Pharmacopoeia, should be provided for co-cultured cells, such as documentary evidence of legal sources, information on the process of cell banking, and reports of cell line characterization.

2.4 Method Validation Question: What are the requirements for the method validation of RCL detection?

Answer:

The detection of RCL is divided into two experimental stages: co-culture and endpoint detection, and these two stages are both needed to be taken into consideration for method validation. In the co-culture stage, reasonable positive control, negative control, and inhibitory control groups should be established, and the final test results should be judged according to the endpoint detection results. The validation items usually carried out include specificity, detection limit, and robustness. For specificity, the detection rates of the positive control, negative control, and inhibitory control groups should be considered, and the setting standards for the detection rates should have a reasonable basis. The detection limit validation usually involves setting up positive controls with different dilutions and ensuring that the detection rate meets a certain standard. The robustness validation should be set up in combination with factors that may affect the method, such as passage number, passage time, cell density, etc. The endpoint detection methods of RCL are usually quantitative or qualitative methods based on biochemistry and molecular biology, and the corresponding validation can be carried out according to the requirements of ICH Q2 and the relevant guidelines of the Chinese Pharmacopoeia on analytical method validation. Generally, for quantitative detection methods, the validation items such as specificity, quantitation limit, detection limit, accuracy, precision, linearity, range, and robustness should be carried out, and for qualitative detection methods, the validation items such as specificity, detection limit, and robustness should be carried out.

3. REFERENCES

　　FDA. Guidance for Industry –Supplemental guidance on testing for replication competent retrovirus in retroviral vector based gene therapy products and during follow-up of patients in clinical trials using retroviral vector. 2006.

　　FDA. Testing of Retroviral Vector-Based Human Gene Therapy Products for Replication Competent Retrovirus During Product Manufacture and Patient Follow-up. 2020.

　　EMA. Guideline on Development and Manufacture of Lentiviral Vectors. 2004.

　　EMA. Gene transfer medicinal products for human use. 2019.

　　Technical Guidelines for Pharmaceutical Research and Evaluation of in vitro Gene Modification Systems (Trial). 2022.