If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Good Manufacturing Practice–compliant change of raw material in the manufacturing process of a clinically used advanced therapy medicinal product–a comparability study
Fraunhofer ISC - Translational Center Regenerative Therapies, Würzburg, GermanyDepartment of Tissue Engineering and Regenerative Medicine, University Hospital Würzburg, Würzburg, Germany
Correspondence: Prof. Andrea Barbero, Department of Biomedicine, University of Basel, University Hospital Basel, Hebelstrasse 20, 4031 Basel, Switzerland.
The development of medicinal products often continues throughout the different phases of a clinical study and may require challenging changes in raw and starting materials at later stages. Comparability between the product properties pre- and post-change thus needs to be ensured. Here, we describe and validate the regulatory compliant change of a raw material using the example of a nasal chondrocyte tissue-engineered cartilage (N-TEC) product, initially developed for treatment of confined knee cartilage lesions. Scaling up the size of N-TEC as required for the treatment of larger osteoarthritis defects required the substitution of autologous serum with a clinical-grade human platelet lysate (hPL) to achieve greater cell numbers necessary for the manufacturing of larger size grafts. A risk-based approach was performed to fulfill regulatory requirements and demonstrate comparability of the products manufactured with the standard process (autologous serum) already applied in clinical indications and the modified process (hPL). Critical attributes with regard to quality, purity, efficacy, safety and stability of the product as well as associated test methods and acceptance criteria were defined. Results showed that hPL added during the expansion phase of nasal chondrocytes enhances proliferation rate, population doublings and cell numbers at passage 2 without promoting the overgrowth of potentially contaminant perichondrial cells. N-TEC generated with the modified versus standard process contained similar content of DNA and cartilaginous matrix proteins with even greater expression levels of chondrogenic genes. The increased risk for tumorigenicity potentially associated with the use of hPL was assessed through karyotyping of chondrocytes at passage 4, revealing no chromosomal changes. Moreover, the shelf-life of N-TEC established for the standard process could be confirmed with the modified process. In conclusion, we demonstrated the introduction of hPL in the manufacturing process of a tissue engineered product, already used in a late-stage clinical trial. Based on this study, the national competent authorities in Switzerland and Germany accepted the modified process which is now applied for ongoing clinical tests of N-TEC. The described activities can thus be taken as a paradigm for successful and regulatory compliant demonstration of comparability in advanced therapy medicinal products manufacturing.
], are innovative medicines for human use that are based on genes, tissues or cells and expected to play an important role especially in addressing yet-untreatable diseases and unmet clinical needs. As all medicinal products, ATMPs must comply with stringent European regulatory requirements for their production and use in clinical trials as well as for marketing authorization [
Guide to Good Manufacturing Practice for medicinal products annexes. Annex 2A Manufacture of advanced therapy medicinal products for human use, P.I.C.-O. SCHEME, Editor. 2022: https://picscheme.org/docview/4590.
European Pharmacopeia chapter 5.2.12. Raw materials of biological origin for the production of cell-based and gene therapy medicinal products., E.D.f.t.Q.o.M.H. (EDQM), Editor. 2022: https://www.edqm.eu/en/european-pharmacopoeia.
Directive 2001/83/EC of the European Parliament and of the Council of 6 November 2001 on the Community code relating to medicinal products for human use, E. Commission, Editor. 2001: https://eur-lex.europa.eu/homepage.html.
]. During the course of product development, including the generation of pre-clinical in vitro and in vivo data, the manufacturing process and critical quality attributes of the Investigational Medicinal Product (IMP) must be defined [
]. Moreover, before applying for a clinical trial authorization, the safety of the IMP must be established and first data on the mode of action and efficacy collected to demonstrate the suitability of the IMP for use in patients.
However, development often continues during the clinical phases, requiring implementation of changes to the manufacturing process or to the raw materials. These changes must maintain or improve product quality, efficacy and safety based on comparability studies, using a risk-based approach. Comparability studies are fundamental to ensure that the pre-clinical and clinical safety and/or efficacy data gathered as well as the benefit/risk balance of the product are valid throughout development [
]. The critical quality attributes defined during product development play an important role in the demonstration of comparability and assessment of the potential impact of the changes. Comparison by critical quality attributes demands a thorough characterization of the product beyond the defined release criteria including all controls of manufacturing process parameters and furthermore, evaluate the stability of the final product [
ICH Topic Q 5 E; Comparability of Biotechnological/Biological Products; note for guidance on biotechnological/biological products subject to changes in their manufacturing process.
]. With this N-TEC, a phase 1 (Nose to Knee I, ClinicalTrials.gov Identifier: NCT01605201) as well as a phase 2 (Nose to Knee II, ClinicalTrials.gov Identifier: NCT02673905) clinical trial have been performed to evaluate the safety and efficacy of N-TEC for the repair of focal cartilage defects in the knee. For the manufacturing of such N-TEC, nasal chondrocytes (NC) isolated from a nasal septum cartilage biopsy of 6-mm diameter, were cultured in two phases, i.e., expansion culture in two dimensions (to get a minimal number of 5.0 × 107 cells) and differentiation culture in a three-dimensional collagen membrane (of a size of 12 cm2) [
]. Autologous serum was initially used as supplement for both culture phases to supply essential mediators for growth and differentiation of the chondrocytes. After repair of focal cartilage defects, clinical indications were planned to be extended to the treatment of more challenging osteoarthritic (OA) cartilage lesions. The general safety and feasibility of this method has recently been demonstrated in two case report studies [
] where N-TEC were implanted in small OA cartilage lesions (4.5–6.5 cm2). However, the affected area in OA is usually larger, requiring greater cell numbers to generate larger N-TEC. In turn, this requires a substantial amount of blood to generate the autologous serum and thus would increase the burden for the patient. Therefore, the manufacturing protocols used in the former clinical studies had to be adapted.
A large number of studies have demonstrated that human platelet lysate (hPL) supplemented during the expansion culture significantly increases the growth of different mesenchymal cells, including chondrocytes [
Human platelet lysate successfully promotes proliferation and subsequent chondrogenic differentiation of adipose-derived stem cells: a comparison with articular chondrocytes.
] evaluated hPL as an alternative to autologous serum for the manufacturing of articular cartilage. They confirmed a greater proliferation rate of articular chondrocytes expanded in medium containing hPL (versus bovine serum and autologous serum) in cell culture and demonstrated that the chondrogenic capacity of articular chondrocytes was preserved after monolayer expansion. However, hPL has not been extensively studied for clinical applications using NC.
In this manuscript, we focus on the regulatory compliant implementation of a change in raw materials consisting in substitution of autologous serum by hPL during the culture of NC for the production of N-TEC for large cartilage defects (up to 40 cm2). In particular, we aimed this study at comparing the performance of NC once expanded with autologous serum (standard process as used in the clinical trial) or with virally inactivated clinical grade hPL (modified process).
Materials and Methods
Risk analysis
In a first step, a risk analysis using Failure Mode and Effects Analysis (FMEA) has been performed to evaluate the criticality of the changes to the standard manufacturing process (see supplementary Figure 1A) as required by European guidelines [
ICH Topic Q 5 E; Comparability of Biotechnological/Biological Products; note for guidance on biotechnological/biological products subject to changes in their manufacturing process.
]. The focus lies on those steps that are most appropriate to identify an impact of the change. The risk-based approach includes the definition of all necessary tests to demonstrate that the safety and efficacy profile of the clinical product manufactured with the previously developed standard and the modified manufacturing process to generate two N-TEC of a size of 20 cm2 each was not negatively affected. The main change in the process and associated risks are summarized in Table 1.
Table 1Main changes in the manufacturing process and potential risks for product quality, purity, efficacy, safety and stability using FMEA analysis.
Process step
Standard process
Modified process
Potential risk
Expansion phase
AS (5%)
hPL (5%)
•
Adventitious agents (low)
•
Tumorigenic potential (high)
•
Effect on chondrogenic potential (low)
•
Overgrowth of contaminating cells favored by the new medium composition (medium)
Chondrogenic phase
AS (5%)
HSA (1.25 mg/mL)
•
Reduced chondrogenic differentiation (high)
Transport medium composition
AS (5%)
HSA (1.25 mg/mL)
•
Stability of product not guaranteed (low)
In parenthesis are reported the degrees of potential risks (from low to high).
AS: autologous serum; FMEA: failure modes and effects analysis; hPL: human platelet lysate; HSA: human serum albumin.
The risk of adventitious agents can be minimized by choosing virus-inactivated hPL product. Thorough supplier qualification is necessary to ensure that the hPL product meets all quality requirements. The potential increase of tumor formation capacity has been assessed by performing a karyotyping of NC expanded beyond the foreseen culture time (i.e., passage 4, P4). Moreover, during cell expansion phase, cell numbers, population doublings (PDs) and proliferation rates have been quantified.
Potential negative effects on the chondrogenic potential were assessed by characterization of the extent of cartilage-specific markers of the generated tissue through immunohistochemical, histological and biochemical analysis. These analyses have been performed in N-TEC generated with P2 as well as P4 expanded cells, to simulate worst-case conditions.
The overgrowth of low amounts of potentially contaminating cells from the perichondrium has been investigated by comparing population doublings of both cell types (isolated from nasal septal cartilage biopsies) in the culture medium with fetal bovine serum as substitute of autologous serum (that could not be obtained) and hPL, respectively. A stability study has been carried out to determine the shelf-life of N-TEC during storage and transport under defined conditions (2–8°C) in medium containing human serum albumin (as replacement of autologous serum) (Table 2).
Table 2Summary of critical quality attributes with regard to quality, purity, safety, efficacy and stability as well as the respective test methods and acceptance criteria to be compared for the products manufactured with the standard (clinical batches) and modified process (see supplementary Figure 1B for details on processing steps and sampling strategy).
Critical quality attribute
Test method
Acceptance criteria
Quality
Cell viability in 2D
Percentage of viable cells
Validated cell counting with hemocytometer/trypan blue
>95%
Cell number and proliferation rate
Total no. of cells after expansion
Validated cell counting with hemocytometer
≥1.67 × 108 (for the generation of two N-TEC)
Proliferation rate
Calculation
<1.5 doublings/day
Purity
Contaminant cells
Low contamination by PC
Growth rate test
Difference in proliferation rate: PC/NC (FBS vs hPL) ≤10%
Efficacy
Cell viability in 3D
Percentage of viable cells
Histological assessment (H&E staining)
>70%
Cartilage specific proteins
GAG
Histological assessment (Safranin-O-staining and MBS)
MBS >3
GAG/wet weight
Biochemical analysis
>0.5 μg/mg
Type II collagen
Immuno-histochemistry
staining intensity at least “+”
Cell morphology
Cells with round morphology and embedded within lacunae
Histological assessment
MBS >3
Chondrogenic genes
Upregulation of cartilage and downregulation of fibroblastic markers
qPCR
gene expression levels at least similar
Safety
Tumorigenicity
Tumor formation potential
Risk analysis
No increased risk of tumorigenicity
Literature search
Karyotyping
No chromosomal abnormalities
Stability
Cell viability in 3D
Percentage of viable cells
Histological assessment (H&E staining)
>70%
Potency/Identity
Cartilage specific proteins (GAG), Cell morphology
A total of eight nasal septal cartilage samples were harvested pseudo-anonymously from patients, during routine rhinoplasty intervention at the University Hospital Basel (USB) and University Hospital Wurzburg (UKW) after informed consent of the patients. Nasal septal cartilage samples from five donors (four female, one male, age range: 29–53 years) were used to perform the comparison runs (standard versus modified process) in two laboratories (three at USB, two at UKW). The cells from the same five donors were used for expansion until passage four and karyotyping. Nasal septal cartilage samples from the other three donors (one female, two males age range 17-33 years) were used to isolate perichondrial cells (PCs) (purity study). For the standard process, seven patients from the ongoing phase 2 study were randomly selected. Similar analyses were performed as for the modified process.
Chondrocyte and PCs isolation, expansion and culture in 3-dimensional collagen scaffolds
Isolation of NC and PC and their culture in monolayer and in collagen type I/III membranes (Chondro-Gide®; Geistlich Pharma AG, Wolhusen, Switzerland) were performed using previously described protocols, hereby named standard [
] or modified protocols. Brief descriptions of these processes are reported in the sections that follow.
Isolation and expansion of PC and NC
Remnants of perichondrium were removed from the three cartilage biopsies with sterile forceps. PC were isolated and expanded for two passages as previously described [
] in Expansion Media (EM) containing 5% fetal bovine serum (FBS; EM_serum) or hPL (EM_hPL) (see media composition in supplementary Table 1). FBS was used in this study (instead of autologous serum) because no blood could be collected from the patients from whom the nasal septal cartilage samples (remnants of normal operations) were taken.
Cartilage samples, cleaned from perichondrium, were cut in small pieces using a sterile scalpel and incubated with 0.15% Good Manufacturing Practice–grade NB6 collagenase (Serva; Nordmark Biochemicals, Uetersen, Germany) for 22 h at 37°C, under agitation. Isolated NC were plated in tissue culture flasks at a density of 1 × 104 cells/cm2 and cultured in media containing 5% FBS or hPL for two passages [
]. To provide more surface for the cells to grow, P1 to P2 expansion of NC in medium containing hPL was performed in 5xT300 flaks (area = 1500 cm2) instead of 4xT150 flasks (area = 600 cm2) as for the standard process. In selected experiment NC were expanded in medium containing hPL up to four passages.
Manufacturing of N-TEC and stability study
NC expanded for two or four passages in medium containing hPL were centrifuged, resuspended in 1 mL of medium and seeded on 4 × 5-cm collagen type I/III membranes (Chondro-Gide®; Geistlich Pharma AG) at a density of 4.17 × 106 cells/cm2 by manually pipetting small drops (i.e., 30 drops of 33-μL cell suspension/drop) on the cell-permeable layer of the Chondro-Gide® membrane. During the seeding, the membranes were placed in low-adhesion 10-cm diameter culture dishes (Nunclon Sphera 90 mm; Thermo Fisher Scientific, Waltham, MA, USA) for better cell retention. The resulting cell-seeded scaffolds were incubated for 1.5 h in the incubator to promote cell adhesion. Then, 40 mL of Chondrogenic Medium (see medium composition in supplementary Table 1) was added in the dish and constructs were cultured for 2 weeks at 37°C and 5% CO2 with media changes twice/week. Critical quality attributes of the N-TEC manufactured with the modified protocol were compared with those of the N-TEC manufactured with the phase 2 clinical trial using the standard protocol. To verify whether the modified process influenced the stability of the manufactured N-TEC, the TEP were analyzed immediately and after 3 days of storage at 2–8°C in the defined transport medium (DMEM containing 2 mmol/L GlutaMax, 1 mmol/L sodium pyruvate, 10 mmol/L HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) buffer, 1.2 mg/mL human serum albumin). To mimic transport and storage (stress test), N-TEC were kept under agitation for 7 h on a shaker at 4°C or shipped from Würzburg to Basel (about 6 h) under the previously validated conditions for transport of the TEP in the Nose to Knee II clinical trial and subsequently stored statically for up to 3 days at 4°C. Critical quality attributes were compared at time of release and after 3 days.
Analytical methods
Proliferation rate and cell viability
Proliferation rate was calculated as the ratio of log2(N/N0) to T, where N0 and N are the numbers of cells, respectively, at the beginning and end of the expansion phase, log2(N/N0) is the number of cell doublings, and T is the time required for the expansion. Cell viability was estimated using the trypan blue exclusion test. In brief, cells suspensions were stained with 0.4% trypan blue and the number of dead cells (i.e., stained cells) and viable cells (i.e., total cells-dead cells) were counted in the hemocytometer, a previously validated method. Cell viability was calculated as a percentage ratio of the number of viable cells to the total number of cells.
Histology
N-TEC were fixed overnight in 4% formalin and embedded in paraffin. Sections of 5 μm in thickness were stained with hematoxylin and eosin (H&E) or Safranin-O (Saf-O) for glycosaminoglycans (GAG) as described [
]. Cell viability was assessed on H&E-stained histological images by experienced pathologists at the Institute of Pathology, University of Würzburg. Histological scoring via the Modified Bern Score (MBS) was performed on Saf-O–stained histological images as previously described [
]. To summarize, the MBS has two rating parameters that each receive a score between 0 and 3. First, the intensity of Saf-O staining (0 = no stain; 1 = weak staining; 2 = moderately even staining; 3 = even dark stain), and second, the morphology of the cells (0 = condensed/necrotic/pycnotic bodies; 1 = spindle/fibrous; 2 = mixed spindle/fibrous with rounded chondrogenic morphology; 3 = majority rounded/chondrogenic, embedded within lacunae). The two values are summed resulting in a maximum possible MBS of 6. Saf-O histological section was divided in nine predefined areas that were graded through a microscope (Widefield Microscope Olympus IX83) using a 10× objective. MBS were then derived averaging the nine obtained values. For the release of the N-TEC, cell viability >70% and an MBS >3.0 are defined as acceptance criteria (Table 2).
Immunohistochemistry against collagen type II (No. 0863171, MP Biomedicals, Santa Ana, CA, USA; 1:1000) was performed on paraffin sections using the Vectastain ABC Kit (Vector Labs, Burlingame, CA, USA) with hematoxylin counterstaining as in standard protocols [
]. Staining intensities were qualitatively assessed (collagen type II staining intensity: –: not detectable, +: weak/moderate, ++: intense) (Table 2).
Quantification of GAG and DNA content
N-TEC were weighted (wet weight, ww) and further digested with proteinase K (1 mg/mL proteinase K in 50 mmol/L Tris with 1 mmol/L ethylenediaminetetraacetic acid, 1 mmol/L iodoacetamide and 10 mg/mL pepstatin A) for 16 h at 56°C. The GAG content was determined as follow: samples were incubated with 1 mL of dimethylmethylene blue assay (341088; Sigma-Aldrich, St. Louis, MO, USA) solution (16 mg/L dimethylmethylene blue, 6 mmol/L sodium formate, 200 mmol/L GuHCL, pH 3.0) on a shaker at room temperature for 30 min. Precipitated dimethylmethylene blue assay–GAG complexes were centrifuged and supernatants were discarded. Complexes were dissolved in decomplexion solution (4 mol/L GuHCL, 50 mmol/L Na-acetate, 10% propan-1-ol, pH 6.8) at 60°C, absorption was measured at 656 nm and GAG concentrations were calculated using a standard curve prepared with purified bovine chondroitin sulfate. DNA content was measured using the CyQUANT Cell Proliferation Assay Kit (Molecular Probes Inc., Eugene, OR, USA) accordingly to the instructions of the manufacturer. GAG and DNA contents were reported as GAG/ww and DNA/ww.
Total RNA was extracted from N-TEC with the Quick RNA Miniprep Plus Kit (Zymo Research, Irvine, CA, USA) and quantitative gene expression analysis was performed as previously described [
]. Reverse transcription into cDNA was done from 3 μg of RNA by using 500 μg/mL random hexamers (Promega, Dübendorf, Switzerland) and 0.5 μL of 200 UI/mL SuperScript III reverse transcriptase (Invitrogen, Waltham, MA, USA). Assays on demand (Applied Biosystems, Waltham, MA, USA) were used with TaqMan Gene Expression Master Mix to amplify the gene of interest (see supplementary Table 2) and glyceraldehyde 3-phosphate dehydrogenase. The threshold cycle (CT) value of the reference gene, glyceraldehyde 3-phosphate dehydrogenase, was subtracted from the CT value of the gene of interest to derive ΔCT values. All displayed gene expression levels are the ΔCT values used for statistical analyses.
Karyotyping
NC were expanded in monolayer culture up to passage 4 (16–20 PDs) and 60%–70% confluence, before arresting cell cycle in metaphase using colcemid (KaryoMAX Colcemid Solution in phosphate-buffered saline, #15212012, Gibco by Life Technologies, Carlsbad, CA, USA; 10 µg/mL) at 37°C for 1 h 45 min for chromosome preparation. Subsequently, cells were trypsinized and suspended in KCl (4 g/L) at 37°C for 45 min. The cell suspension was centrifuged at 200g for 9 min, supernatant discarded, and KCl added dropwise while vortexing. Afterwards, cells were incubated at 37°C for 23 min and subsequently fixed at –20°C in a pre-chilled 3:1 methanol:glacial acetic acid solution, centrifuged as described before and dropped onto frozen slides. The slides were dried at 90°C for 90 min and cooled down before staining. Chromosomes were stained with 0.4% filtered Giemsa solution (Sigma-Aldrich). Twenty-seven (22 female, five male) metaphases were analyzed for numeric anomalies and to exclude structural chromosome abnormalities at 350-band resolution. Genetic analysis was conducted at the Institute for Human Genetics at the University of Würzburg with GTG banding [
] and visualized with GenASis BandView (ASI, Applied Spectral Imaging, Carlsbad, CA, USA; approved by the Food and Drug Administration, clinical use).
Statistical analyses
Statistical analysis was performed with the software GraphPad Prism 9, Version 9.3.1 (GraphPad, San Diego, CA, USA). The normality of the population was first assessed using the Shapiro–Wilk test and either parametric t test for normal populations or the non-parametric Mann–Whitney U test for non-normal populations were used. The following definitions were used for significance values: *P < 0.05, **P < 0.005, ***P < 0.0005, ****P < 0.00005, and ns = no significant difference
Results
The comparability study was performed at the two manufacturing sites (USB and UKW), both responsible for the manufacturing of the N-TEC in the phase 2 clinical trial. For the modified process, a total of eight N-TEC were manufactured: four N-TEC at USB (with cells from three donors) and four N-TEC at UKW (with cells from two donors).
Quality
The quality of the NC generated during culture under the standard and modified processes was assessed using the following critical quality attributes: cell viability, proliferation rate and total cell number. For the standard process, these parameters were previously collected during the clinical study. Cell viability was similarly high (more the 98%) for NC cultured with the standard or modified process. As expected, due to the mitogenic properties of the hPL, population doubling and proliferation rate were significantly increased in the modified process (1.3-fold [PD] and 1.2-fold (proliferation rate), respectively) (Figure 1). Importantly, proliferation rate of NC expanded with hPL was in the pre-defined range (i.e., <1.5 doublings/day) (Table 2). Because of the greater proliferation induced by hPL and the larger surface areas (1500 vs 600 cm2), a larger number of NC could be generated at P2 in the modified (2.3 × 108 ± 5.3 × 107) vs standard (6.2 × 107 ± 1.1 × 107) process. As planned, all cell cultures reached the required 1.67 × 108 cells after 2 weeks culture with hPL.
Fig. 1Cell viability (A), PDs (B) and proliferation rate (C) and cell number (D) of NC expanded in medium containing autologous serum (standard process) or hPL (modified process). Values are mean ± standard deviation of parameter generated from cells cultured in the standard (n = 7 donors) or modified (n = 5 donors) process. *P < 0.05, **P < 0.005, ***P < 0.0005, ****P < 0.00005. ns = no significant difference.
We have recently demonstrated that nasal septal PCs do not have the capacity to form GAG- and collagen type II–rich engineered tissues and that increasing amounts of perichondrium in the starting material decreases the quality of the N-TEC. Although care is taken during the harvesting of the starting material to leave the perichondrium intact in the patient, residues of the perichondrium may still be present on the collected nasal septal cartilage biopsy. The amount is so low that in the standard process no impact on the quality of the final product can be detected [
]. The results of the growth tests showed that hPL induced a similar increase (versus FBS) in the proliferation rates of PC and NC (percentage ratio PC/NC equal to 2.8% and 3.8%, respectively, for FBS and hPL) (Figure 2), indicating that both cell types benefit from the use of hPL in the same extent, thus excluding a potential overgrowth of PC in the modified process.
Fig. 2Proliferation rate of NC and PC expanded for two passages in medium containing 5% FBS or 5% hPL. Values are mean ± standard deviation of three independent experiment using cells from three different donors.
Efficacy or potency of the N-TEC is postulated to be related to the viability of the cells and content of cartilage-specific proteins. Release criteria selected for the clinical trial include the histological assessment of cell viability in H&E-stained tissue sections, GAG amounts and cell morphology, in Safranin-O–stained tissue sections, using the Modified Bern Score (MBS).
Additional analyses for potency were performed including characterization of samples by immunohistochemistry (collagen type II), biochemistry (GAG and DNA contents) and by qPCR (gene expression) (Figure 3). The (immuno)histological results showed that staining intensity for GAG and type II collagen were similarly high in the N-TEC generated with the standard and modified process independently of the passage number (Figure 3A,B). These findings were confirmed in the MBS evaluation (Figure 3C), where no significant differences could be seen between N-TEC generated with standard or modified protocol using cells expanded until P2 or until P4 (MBS = 4.9 ± 0.9 [standard], 5.6 ± 0.4 [modified P2], and 5.6 ± 0.2 [modified P4]). Importantly, subscores related to morphology (min: 0, max 3) were similarly high (2.49 ± 0.79 [standard], 2.81 ± 0.14 [modified P2] and 2.76 ± 0.03 [modified P4]), indicating that cells within the N-TEC manufactured with the two processes exhibit a differentiated chondrocytic phenotype. Biochemical analysis of the N-TEC generated using the standard and modified process showed no significant differences in the GAG/ww or the DNA/ww (Figure 3D,E). Cell viability of N-TEC generated with the standard or modified process using P2 or P4 cells was beyond the release target value of 70% (i.e., 90 ± 4.1 (standard), 94 ± 1.8 (modified P2) and 80 ± 0.0 [modified P4]). Interestingly, viability was slightly but statistically significant greater in N-TEC generated with the modified process (Figure 3F).
Fig. 3Histological analysis (Safranin-O, Saf-O, top pictures and type II collagen, Col2, bottom pictures) of the N-TEC generated with the modified process using cells from five donors (A). Staining of worst- and best-quality N-TEC generated with the standard process and representative staining of N-TEC generated with P2 or P4 cells with the modified process. Histological slides are oriented in the way to have on the top view the newly deposited cartilage matrix (in the seeded cell-permeable layer of the Chondro-Gide® membrane; the lower part of the N-TEC (appearing in bluish green in the Saf-O-pictures) represents the cell occlusive layer). Scale bars = 100 μm (B). Modified Bern Score (MBS, range 0–6) (C), GAG/wet weight (ww) (D) and DNA/wet weight (DNA/ww) (E) and cell viability (F) of N-TEC generated with the standard or modified process. Values are mean ± standard deviation of eight independent experiment using cells from five different donors for the modified process and seven independent experiments from seven donors for the standard process. *P < 0.05, **P < 0.005, ***P < 0.0005, ****P < 0.00005.
N-TEC were analyzed by RT-qPCR to quantify the expression of specific genes encoding for extracellular matrix component (COMP, ACAN, VCAN, type I and II collagen), differentiation markers (Sox-9, SerpinA1, HAPLN1 and MFAP5, receptors for FGF (FGFR3), BMP (BMPR1B), IGF (ITGFB4) and Wnt (SFRP1) signaling pathways, the neuroectodermal factor nestin and the degrading factors MMP-12. As reported in Figure 4, N-TEC generated with the modified protocol expressed statistically significant greater levels of ACAN (12.9-fold), Col2 (8.1-fold), ACAN/VCAN (107.4-fold), Col2/Col1 (18.8-fold), Sox-9 (8.1-fold), Serpin A1 (5.5-fold), HAPLN1 (8.2-fold), HAPLN1/MFAP5 (11.4-fold), FGFR3 (14.5-fold), BMR1B (4.9-fold) and SFRP1 (4.5-fold). These results demonstrated a superior chondrogenic redifferentiation state of the cells within the N-TEC generated with the modified (versus the standard) protocol. As shown in supplementary Figure 2, MBS scoring correlates with the expression levels of the cartilage indexes Col2/Col1 and ACAN/VCAN mRNA.
Fig. 4Quantitative reverse-transcription polymerase chain reaction analyses of N-TEC generated with the modified (gray bars) and standard protocol (white bars); gene expression is relative to glyceraldehyde 3-phosphate dehydrogenase levels. Values are mean ± standard deviation of samples from five different donors. *P < 0.05, **P < 0.005, ***P < 0.0005, ****P < 0.00005.
An extensive supplier qualification was performed to ensure that the chosen hPL product complied with the regulatory requirements for raw materials of biological origin for the production of cell-based medicinal products [
European Pharmacopeia chapter 5.2.12. Raw materials of biological origin for the production of cell-based and gene therapy medicinal products., E.D.f.t.Q.o.M.H. (EDQM), Editor. 2022: https://www.edqm.eu/en/european-pharmacopoeia.
]. The results of the supplier qualification with regard to the most important parameters are summarized in Table 3. Overall, the risk analysis allows us to conclude that the use of hPL for N-TEC generation is safe for patients.
Table 3Summary of the most important parameters addressed during supplier qualification.
Parameter
Specification
Origin of platelets
•
Platelets are sourced from German blood banks and tested according to regulations of blood donation EP0853 and Annex III 2006/17/EC (certified), identity confirmed
•
Number of donors per batch is approximately 60
Safety
•
Gamma irradiation after pooling of sera (Ph Eur 5.1.7) including viral clearance study with enveloped and non-enveloped viruses (at least a 4 log10 reduction)
•
Absence of antibiotics and validated aseptic processing procedures
•
0.2-μm sterile filtration
•
TSE safety is guaranteed by following Annex III of 2004/33/EC implementing Directive 2002/98/EC excluding donors with family history of developing TSE or Creutzfeldt–Jakob disease
Production process and in-process-controls/release testing
•
Quality management system and GMP requirements (ISO13485:2016)
•
General quality tested according to Ph Eur (pH [2.2.3], osmolality [2.2.35], sterility [2.6.1], endotoxin [2.6.14])
•
Mycoplasma tests are not according to Ph Eur, but sufficient (additional tests during N-TEC manufacturing processes)
•
Volume and turbidity specified
•
Contents of growth factors measured by ELISA by the manufacturer
•
Biological activity of each batch confirmed by cell growth test with MSCs by the manufacturer (three different batches have been tested on human NC by the authors of this paper to verify batch to batch consistency)
Storage conditions
•
Shelf life and storage conditions defined
•
Correct labeling containing all information attached
Documentation
•
Certificate of conformity issued for each batch
•
Supplier agreement closed between company and manufacturers
Removal of hPL from end-product
•
Use of 5% hPL only during expansion (2 weeks) leads to subsequent dilution and washing out of hPL from the final product during the differentiation phase (2 weeks)
•
Product is already used in clinical trials
Detailed information on specific points (e.g., content of growth factors) was obtained by the authors only under a non-disclosure agreement during supplier qualification.
One of the challenges in tissue engineering is the ability to generate a sufficient number of cells from a small donor sample by monolayer expansion. The presence of growth factors contained in hPL in the expansion medium significantly enhances the proliferation rate of chondrocytes and, consequently, may enhance the risk of genetic instability and aberrations and thus the potential for tumor formation. Therefore, classical karyotyping (Figure 5) of NC expanded for four passages with expansion medium containing hPL was performed as a prospective risk minimization measure. Overall, 27 metaphases were analyzed, and no pathologic changes have been detected in any of the samples.
Fig. 5Representative karyogram from NC expanded according to modified process of a female patient. Expansion was performed until passage 4 (corresponding to a total of 18.1 PDs).
The stability of the N-TEC after the end of production (shelf life) under pre-defined conditions is important for storage and transport to the respective clinical site. Therefore, the impact of changes in the manufacturing on the shelf life of the product has to be carefully assessed.
After production, each N-TEC was cut into two halves: one half was analyzed on the day of release (day 0, fresh), the other one was stored under different conditions and analyzed after 3 days of storage (day 3). Intensity and uniformity of staining for GAG and type II collagen were similar in fresh N-TEC or in N-TEC after 3 days of stress/storage condition (Figure 6A). Quantitative assessments demonstrated no statistically significant differences in MBS (5.63 ± 0.37 and 5.47 ± 0.79), cell viability (94.38 ± 1.77% and 90.83 ± 4.92%), GAG (4.70 ± 2.62 μg/mg and 3.73 ± 2.00 μg/mg) and DNA (3.70 ± 4.29 μg/mg and 3.78 ± 2.30 μg/mg) contents. Thus, this study confirms that even in the worst-case scenario (e.g., a rough transport) the quality of the N-TEC remains stable up to 72 h comparable to the shelf-life previously validated in the phase 2 study (standard process) (Figure 6B-E).
Fig. 6Representative Safranin-O (Saf-O) and type II collagen (Col2) images of N-TEC generated with the modified process at day of release (fresh) as well as 3 days after stress/storage conditions (stressed/stored). Histological slides are oriented in the way to have on the top view the newly deposited cartilage matrix (in the seeded cell-permeable layer of the Chondro–Gide membrane; the lower part of the N-TEC (appearing in bluish green in the Saf-O-pictures) represents the cell occlusive layer). Scale bars = 100 μm (A). MBS (B), viability (C), GAG/wet weight (D) and DNA/wet weight (E) of the N-TEC. Values are mean ± standard deviation of samples from different donors (n = 8 for release, n = 6 for stress test).
We demonstrated the regulatory-compliant changes in a critical raw material for an ATMP under clinical investigation according to the current guidelines using a comparability study. In particular, we evaluated the impact of the changes on quality, purity, efficacy, safety and stability of N-TEC. Based on pre-defined critical quality attributes, associated test methods and acceptance criteria, we could demonstrate comparable quality of the N-TEC manufactured with the standard and modified processes, which main differences rely on the use of autologous serum or hPL for the culture of NC.
We decided to use hPL as supplement for the expansion of NC for its known high mitogenic properties [
] that would allow to obtain greater number of cells for the production of large N-TEC necessary for the repair of extensive cartilage defects (e.g., in OA joint). The designed process allowed to maintain unaltered the initial size of the starting material and the time needed for cell expansion. Our strategy was justified by the fact that alternative approaches that could have been used to achieve larger number of cells, e.g., by increasing the initial size of biopsy from the patient or the number of cell passaging during the expansion phase have important drawbacks, like increased risk of donor-site morbidity or greater manufacturing costs, and lower cartilage quality (due to the reduction of the chondrogenic capacity of chondrocytes with increasing monolayer culture time) [
], respectively. Although the use of autologous hPL is possible, allogenic hPL is easier to standardize, has lower batch-to-batch variability and is more economical [
Gaps in the knowledge of human platelet lysate as a cell culture supplement for cell therapy: a joint publication from the AABB and the International Society for Cell & Gene Therapy.
]. However, careful selection of the product and supplier qualification is required, as hPL products may differ significantly in terms of quality and applicability within a clinical setting and have to fulfill the requirements for the use as a raw material in cell expansion processes for clinical application as described in the European Pharmacopeia chapter 5.2.12 [
European Pharmacopeia chapter 5.2.12. Raw materials of biological origin for the production of cell-based and gene therapy medicinal products., E.D.f.t.Q.o.M.H. (EDQM), Editor. 2022: https://www.edqm.eu/en/european-pharmacopoeia.
]. Among the factors influencing the quality, safety and efficacy of the hPL are the source of the platelets, the pathogen inactivation method, the manufacturing process as well as the final application and pool size [
Gaps in the knowledge of human platelet lysate as a cell culture supplement for cell therapy: a joint publication from the AABB and the International Society for Cell & Gene Therapy.
]. To minimize lot-to-lot variation, pooling of hPL from 50 to 250 donors is generally applied, although a recent mathematical approach has demonstrated that pooling of 16 donors is sufficient to minimize batch variations [
European Pharmacopeia chapter 5.2.12. Raw materials of biological origin for the production of cell-based and gene therapy medicinal products., E.D.f.t.Q.o.M.H. (EDQM), Editor. 2022: https://www.edqm.eu/en/european-pharmacopoeia.
Gaps in the knowledge of human platelet lysate as a cell culture supplement for cell therapy: a joint publication from the AABB and the International Society for Cell & Gene Therapy.
]. Several strategies are possible, but efficacy and safety of these methods may differ depending on the viruses tested. We opted for a product in which gamma irradiation is used to inactivate viruses in hPL. This technique was reported (i) to induce an efficient inactivation of several viruses including human immunodeficiency virus and hepatitis A virus, (ii) not to affect the content of key biochemical factors for cell culture or of most growth factors with the exception of FGF-2 (that, however, is supplemented in the expansion medium in our study) and (iii) not to affect the proliferation, clonogenic and differentiation potential, as well as immunosuppressive properties of human mesenchymal stromal/stem cells [
Viral inactivation of human platelet lysate by gamma irradiation preserves its optimal efficiency in the expansion of human bone marrow mesenchymal stromal cells.
Human platelet lysate successfully promotes proliferation and subsequent chondrogenic differentiation of adipose-derived stem cells: a comparison with articular chondrocytes.
] their subsequent cartilage-forming capacity upon three-dimensional culture in chondrogenic media not containing hPL, this aspect was thoroughly investigated in our comparability study.
N-TEC generated with both the standard and modified processes displayed high cell viability and exhibited cartilaginous properties as evidenced by large amounts of GAG and type II collagen matrix. The qPCR analyses demonstrated that several key chondrogenic genes (Sox9, Aggrecan, Col2), showed a significantly greater expression in N-TEC manufactured under hPL supplementation. In addition, also genes encoding for proteins which release is enhanced during the chondrogenesis of mesenchymal stromal cell (Serpin A1 [
Molecular validation of chondrogenic differentiation and hypoxia responsiveness of platelet-lysate expanded adipose tissue-derived human mesenchymal stromal cells.
]) were observed to be expressed at greater levels by N-TEC manufactured with the modified process. Furthermore, genes encoding for the FGF (e.g., FGFR3), BMP (e.g., BMPR1B) and Wnt (e.g., SFRP1) signaling pathways displayed greater expression levels in N-TEC manufactured with hPL-expanded NC. FGFR3 is abundantly expressed in articular chondrocytes in adulthood [
]. Overall, our results suggest that NC expanded with hPL as compared with autologous human serum were capable to re-differentiate into cartilaginous tissue but with a gene expression signature more similar to mature/healthy cartilage further confirming the efficacy and potency of the products. Although an extensive molecular characterization of the N-TEC is beyond the scope of this study, additional investigations (e.g., extend the analyses toward other signalling pathways regulators and confirm the qPCR data at protein levels) would be required to better investigate the molecular mechanisms accounting for those differences.
Regarding the potency test we demonstrated that N-TEC with greater MBS scores (5-6 versus 3-5) expressed greater levels of the differentiation indexes Col2/Col1 mRNA and ACAN/VCAN mRNA [
Specific growth factors during the expansion and redifferentiation of adult human articular chondrocytes enhance chondrogenesis and cartilaginous tissue formation in vitro.
]. These results indicate that MBS is capable of reliably capturing the chondrogenic status of the NC in the N-TEC and can be used as a potency marker. Moreover, this “easy-to-use” release criteria can be extended for the assessment of the quality of other cell-based products for cartilage repair, instead of the more time-consuming and costly analytical assays (e.g., qPCR) typically performed.
The preferred approach for a comparability study is a side-by-side testing using the same raw materials and especially a split-based approach for the biological starting material to reduce the source of variability. This is especially important for autologous cell-based products where the inter donor variability can represent a highly influencing factor for demonstrating comparability [
]. However, such approach could not be followed because of (i) limited amount of cartilage remnants available during operations and (ii) unacceptable burden on patient due to blood harvesting without clinical justification. An alternative approach accepted by EMA if a side-by-side study is not possible is the comparison of post change data (modified process) with historical data obtained during the Nose to Knee II clinical study [
The N-TEC manufacturing process using hPL was approved by competent authorities in Switzerland and Germany (Swissmedic and Paul-Ehrlich-Institute) and is currently used in a phase 1 study for repair of nasal septum perforations (clinicaltrials.gov NCT04633928) and for repair of cartilage lesions in the knee and ankle (temporary authorization issued by Swissmedic (TA-2020-00007).
The present comparability study demonstrated the successful change in a raw material for an ATMP manufacturing in terms of quality attributes of the product and of subsequent regulatory approval for use in clinical studies by national competent authorities. The approach followed in this comparability study is here offered as an example for the regulatory-compliant change of raw materials in late-stage clinical trials.
Funding
This work was supported by European Union's Horizon 2020 research and innovation program (grant agreement no. 681103) to Prof. Dr. Ivan Martin. The funding agency has no role in the design of the study or collection, analysis and interpretation of the data, writing of the report or submission of the manuscript.
Author Contributions
Conception and design of the study: AB, AW, SM, OP, PB, MH, StH, WK. Acquisition of data: FW, SF. Analysis and interpretation of data: AW, AB, SM, SeH, OP. Drafting or revising the article: AW, AB, SM, OP, SeH, IM, WK. All authors have approved the final article.
Declaration of Competing Interest
The authors have no commercial, proprietary or financial interest in the products or companies described in this article.
Acknowledgments
The authors thank Dr. Elena Gerhard-Hartmann, Dr. Stefan Kircher and Prof. Dr. Andreas Rosenwald, Pathological Institute University Würzburg, for the assessment of the viability of the N-TEC; Prof Dr. Eva Klopocki and Mrs Andrea Hörning from the Institut für Humangenetik, Biozentrum, Würzburg, Geistlich Pharma AG, Wolhusen, Switzerland for providing Chondro-Gide® membranes and MacoPharma International GmbH for providing human platelet lysate.
Guide to Good Manufacturing Practice for medicinal products annexes. Annex 2A Manufacture of advanced therapy medicinal products for human use, P.I.C.-O. SCHEME, Editor. 2022: https://picscheme.org/docview/4590.
European Pharmacopeia chapter 5.2.12. Raw materials of biological origin for the production of cell-based and gene therapy medicinal products., E.D.f.t.Q.o.M.H. (EDQM), Editor. 2022: https://www.edqm.eu/en/european-pharmacopoeia.
Directive 2001/83/EC of the European Parliament and of the Council of 6 November 2001 on the Community code relating to medicinal products for human use, E. Commission, Editor. 2001: https://eur-lex.europa.eu/homepage.html.
ICH Topic Q 5 E; Comparability of Biotechnological/Biological Products; note for guidance on biotechnological/biological products subject to changes in their manufacturing process.
Human platelet lysate successfully promotes proliferation and subsequent chondrogenic differentiation of adipose-derived stem cells: a comparison with articular chondrocytes.
Gaps in the knowledge of human platelet lysate as a cell culture supplement for cell therapy: a joint publication from the AABB and the International Society for Cell & Gene Therapy.
Viral inactivation of human platelet lysate by gamma irradiation preserves its optimal efficiency in the expansion of human bone marrow mesenchymal stromal cells.
Molecular validation of chondrogenic differentiation and hypoxia responsiveness of platelet-lysate expanded adipose tissue-derived human mesenchymal stromal cells.
Specific growth factors during the expansion and redifferentiation of adult human articular chondrocytes enhance chondrogenesis and cartilaginous tissue formation in vitro.
00005-1&id=gr1.jpg){kind=link}
00005-1&id=gr2.jpg){kind=link}
00005-1&id=gr3.jpg){kind=link}
00005-1&id=gr4.jpg){kind=link}
00005-1&id=gr5.jpg){kind=link}
00005-1&id=gr6.jpg){kind=link}