Translation of a human induced pluripotent stem cell-derived ovarian support cell product to a Phase 3 enabling clinical grade product for in vitro fertilization treatment

Subject ages, ethics, and informed consent

This study was performed according to the ethical guidelines outlined in the Declaration of Helsinki. Oocyte donor participants were enrolled in the study at several fertility clinics using informed consent for donation of gametes for research purposes, with ethical approval from CNRHA 47/428973.9/22 (Spain), Western IRB No. 20225832 (USA), and Protocol No. GC-MSP-01 (Peru), respectively. These study centers performed the non-clinical research associated with oocyte maturation and/or embryo formation endpoints with no subsequent embryo transfer.

For the MOMA, all animal care and procedures were conducted according to protocols approved (reference number 10133) by the Institutional Animal Care and Use Committee of Barcelona Science Park de Barcelona (IACUC-PCB), Spain. The PCB Animal Facility is accredited and registered by Generalitat of Catalonia government (B-9900044) as a breeding and user center for laboratory animal research. All protocols carried out in the PCB Animal Facility comply with standard ethical regulations and meet quality and experimental requirements of current applicable National and European legislation (RD 53/2013 Council Directive; 2010/63/UE; Order 214/1997/GC).

For the assessment involving 40 patients treated as part of a two-phase clinical evaluation, patients were enrolled under informed consent at two fertility centers. Protocol Fertilo P1, approved by the ethics committee (CIEI-FMH-USMP, FWA No. 00015320, IRB No. 00003251) was utilized. This evaluation was prospectively registered before recruitment began ISRCTN36032472, with regular IRB updates throughout the two phases as safety and efficacy data became available, to guide treatment in the First in Human evaluation. The protocol, standard operating procedures, and design ensured rigorous oversight and monitoring and compliance with international ethical guidelines.

Source material

The RUO-hiPSC line was sourced from the laboratory of G. Church; this line is referred to as the F66 line in the reference text (8). The RUO-hiPSC line was derived from the NIA Aging Cell Repository fibroblast line AG07141 using Epi5 footprint-free episomal reprogramming. A female donor hiPSC line (VCT-37-F35) was sourced from Reprocell USA (9000 Virginia Manor Rd #207, Beltsville, MD 20705) to serve as the starting material for our CG cell line. The parental CG hiPSC starting material was derived from fibroblasts from the skin punch of an eligible United States donor, ID: RPC-VCT-37. Fibroblasts were isolated from a skin punch biopsy and expanded at Reprocell USA (9000 Virginia Manor Rd #207, Beltsville, MD 20705). Donor identity information for 16 loci was obtained by STR analysis performed on donor blood and fibroblasts to confirm sample identity prior to release of the fibroblasts. Fibroblasts were harvested and cryopreserved in cryopreservation containers (ThermoFisher Scientific, catalog number: 377267) in vapor-phased liquid nitrogen (−196°C) to enable testing and further manufacture. The hiPSC line was generated under GMP conditions using a non-integrative, mRNA-based reprogramming technology. The unmodified parental hiPSC line (VCT-37-F35) was characterized by sterility, endotoxin, STR analysis, Karyotype analysis, mycoplasma, viability, phenotype marker, cell purity and morphology, and differentiation testing.

Donor eligibility and cell line manufacturing were performed in line with relevant regulatory standards and guidelines. Regarding specific stages in the manufacturing process, proper controls were implemented for fibroblast derivation according to established guidelines, while reprogramming and cell expansion took place under fully GMP conditions in compliance with regulatory standards and guidelines of the FDA, EMA, and PMDA. Donor eligibility was determined to be in accordance with 21 CFR Part 1271 Subpart C and FDA Guidance for Industry: Eligibility Determination for Donors of Human Cells, Tissues, and Cellular and Tissue Based Products (HCT/Ps), 2021. All experiments involving human cells were performed according to ISSCR 2021 guidelines (International Society for Stem Cell Research) and approved by the IRB committees.

Plasmid manufacturing

Plasmids utilized for engineering were manufactured as previously described (8). Briefly, transcription factor cDNAs (NR5A1, RUNX2, and GATA4) were obtained as Gateway entry clones from the TFome (33) or the ORFeome (34). These sequences were cloned into a barcoded Dox-inducible expression vector (Addgene #175503) using MegaGate cloning (35). The final NR5A1, RUNX2, and GATA4 plasmids (pCK530_22_NR5A1, NM_004959.9; pCK530_33_RUNX2, NM_001024630.4; and pCK530_3_GATA4, NM_001308093) were subjected to whole plasmid sequencing via Plasmidsaurus using Oxford Nanopore Technology with custom analysis and annotation. All plasmids were screened for purity and stored at −20°C, while glycerol stocks of transformed bacteria were stored at −80°C.

Cell engineering

The parental GMP hiPSC clone VCT-37-F35 was thawed at 37°C and resuspended in NutriStem-XF media (Satorius, 05-100-1A) supplemented with Y-27632 (STEMCELL Tech, 72304). Cells were centrifuged and the supernatant was removed. Then, pelleted cells were again resuspended in NutriStem-XF media (Satorius, 05-100-1A) supplemented with Y-27632 (STEMCELL Tech, 72304) and cell count and viability were assessed. Cells were then seeded onto flasks pre-coated with iMatrix-511 (Nippi, 892005) and incubated for 6 days at 37°C. To adapt hiPSCs to the media and substrate, cells were cultured for 3 passages, during which cell quality was monitored based on morphology and confluency; for each round of expansion, Accutase (STEMCELL Tech, 7922) was used to facilitate cell detachment from flasks, and cell count and viability were assessed. Parental hiPSCs (VCT-37-F35) at passage 11 (day 16) were assessed for confluency, detached using Accutase (STEMCELL Tech, 7922), centrifuged, resuspended in TeSR-AOF supplemented with Y-27672 (STEMCELL Tech, 100-0403), and cell count and viability were assessed. Resuspended cells were centrifuged and resuspended in electroporation buffer (P3 solution, P3 Lonza Nucleofection Kit, V4XP-3024) with equimolar quantities of the 3 plasmids encoding the transgenes NR5A1, RUNX2, and GATA4 (pCK530_22_NR5A1, pCK530_33_RUNX2, and pCK530_3_GATA4) in addition to the PiggyBac transposase plasmid (pAMP470 pBASE-Super PiggyBac Transposase). Delivery of the plasmids into the cells was performed using the Lonza 4D-Nucleofector Unit set to the CM-113 program. Transfected cells were plated (100,000 cells/cm²) into a well of a 24-well plate pre-coated with Vitronectin-XF (STEMCELL Tech, 7180) and cultured with TeSR-AOF supplemented with Y-23672 (STEMCELL Tech, 100-0403). For the next four days (days 17-20), daily media changes with TeSR-AOF (STEMCELL Tech, 100-0402) supplemented with Puromycin (Sigma-Millipore, P9620-10ML) were performed for selection; cell quality was monitored by morphological and confluency checks. The following four days (days 20-24), resistant cells were maintained with TeSR-AOF with no supplementation (STEMCELL Tech, 100-0402). On day 24, expanded resistant cells were detached from the flasks using Accutase (STEMCELL Tech, 7922), centrifuged, and resuspended into TeSR-AOF supplemented with Y-23672 (STEMCELL Tech, 100-0403). The cells from one well of the 24-well plate (1.9 cm²) were further expanded into two wells of a 6-well plate (19.2 cm²). From days 25-27, daily media changes with TeSR-AOF (STEMCELL Tech, 100-0402) supplemented with Puromycin (Sigma-Millipore, P9620-10ML) were conducted, and cell quality was monitored by morphological and confluency checks. On day 28, clonal selection was performed by limiting dilution of cells on TeSR-AOF (STEMCELL Tech, 100-0402) supplemented with CloneR^TM2 (STEMCELL Tech, 100-0691). Single cells were plated onto Vitronectin-XF (STEMCELL Tech, 7180). During clonal expansion (days 29-34), individual wells were carefully monitored to assess colony morphology, growth/confluency, and confirm the presence of a single colony per well. TeSR-AOF (STEMCELL Tech, 100-0402) media changes were performed daily until ready for colony selection. Between days 35-46, once a clone had reached significant levels of expansion (< 10 days after seeding), the individual well was treated with Accutase (STEMCELL Tech, 7922) to promote cell detachment. Cells were centrifuged, resuspended in TeSR-AOF supplemented with Y-23672 (STEMCELL Tech, 100-0403), and seeded onto one well of a 24-well plate pre-coated with Vitronectin-XF (STEMCELL Tech, 7180). When a given clone was ready to be passed, the same procedure was repeated and expanded cells were plated onto a 6-well plate pre-coated with Vitronectin-XF (STEMCELL Tech, 7180). During this expansion, all clones were screened for integration of the three transcription factors (NR5A1, RUNX2, and GATA4). In addition, TF insertion sites were verified by Whole Genome Sequencing (Azenta). The analysis showed no evidence of mutations in 20 common proto-oncogenes. Expanded cells were treated with Accutase (STEMCELL Tech, 7922), centrifuged, and the supernatant was removed. Cells from each clone were resuspended in CryoStor® CS10 (STEMCELL Tech, 7930), gently mixed, and filled into cryovials. The finished vials were cryopreserved in a Mr. FrostyTM freezing container at −80°C freezer for a maximum of 24 hours. The cryopreserved cells were stored in cryopreservation containers (Azenta Life Sciences, catalog number: 67-0755-11) and transferred to vapor-phase liquid nitrogen (−196°C) with temperature monitoring for long-term storage.

Cell screening, selection, and preliminary characterization

After seed clones were assessed by genotyping PCR to verify the presence of the three TFs, nine clones harboring all the TFs were identified and each clone was subjected to a more in-depth screening process. To identify the lead candidate, each of the nine clones was individually differentiated and subjected to a series of assays to ensure identity (pluripotency markers, and genotyping), conformance (cell count and viability), and potency (OSC production and function). The leading candidate clone (2-D10) selected to be used as the starting material for the CG cell line was named CG-hiPSC.

hiPSC maintenance and OSC differentiation

hiPSCs were maintained in feeder-free conditions and cell culture plates were pre-coated with either Matrigel (Corning), Laminin-521 (StemCell Technologies), or Laminin-511 (Reprocell). Cells were maintained in either mTESR1 (StemCell Technologies) or TeSR-AOF (StemCell Technologies) without antibiotics at 37 °C in 5% CO₂. All hiPSCs were verified for expression of markers associated with pluripotency (TRA-1-60 and TRA-1-81), negative for mycoplasma or other human adventitious agents (performed by IDEXX Bioanalytics), and karyotypically normal (G-band karyotype test, performed by WiCell Research Institute, and Karyostat, performed by ThermoFisher). RUO-hiPSCs and CG-hiPSCs were authenticated by SNP array (CellID, performed by ThermoFisher). All OSC differentiations were initiated from previously characterized hiPSC working banks under passage 25. Characterization of hiPSC working cell banks includes assessment of marker expression, morphology, vial appearance, container closure integrity, cell count and viability post-thaw, differentiation potential, and testing for endotoxin, mycoplasma, sterility, adventitious agents, and genomic integrity. OSC differentiation was performed as previously described (8). In short, hiPSCs were exposed to the Rho kinase inhibitor, Y-27632, and the WNT activator, CHIR99021, to prime the cells into a mesodermal fate. Exposure to doxycycline throughout the process induced overexpression of the TFs NR5A1, RUNX2, and GATA4. The differentiation process required five days in culture. All images were taken with an ECHO Revolve Microscope (Discover ECHO).

Cell count and viability

Cell counts and viability assessment was performed using an Eve Automated Cell Counter (NanoEnTeck) and a NucleoCounter NC-202 (ChemoMetec).

Single cell RNA sequencing (scRNA-seq)

OSCs were cryopreserved in CryoStor CS10 (StemCell Technologies) prior to being processed for single-cell RNA sequencing by Genewiz (Azenta). Cells were loaded onto a Chromium Single Cell Chip (10x Genomics) and processed through the Chromium Controller to generate single-cell gel beads in emulsion. scRNA-seq libraries were generated using the Chromium Single Cell 3′ Library & Gel Bead Kit (10x Genomics). Target cell recovery was estimated to be 3,000 cells per sample. Files were generated following a split-set analysis workflow. First bcl2fastq was used to generate the fastq files. Next, a reference genome was generated using split-pipe and the Homo sapiens GRCh38 file, also using STAR (36) and Samtools (37). Then, split-pipe was run to process and align the files against the reference genome. The resulting files were generated in an mtx matrix format. Finally, the files were combined, and an Anndata object was made in h5ad file format. For the analysis, cells with less than 200 genes were filtered out, as well as genes found in less than 3 cells. Cell counts were normalized to 10,000 Unique molecular identifiers (UMIs) per sample and log (ln) plus 1 transformed. Principal component analysis was performed using the Scanpy package (v1.9.6) based on 30 PCA components, and using PCA results, nearest neighbor analysis was performed using n_neighbors = 20. Batch correction was performed using Scanpy’s ComBat method (38). The number of components used for batch correction was 30 and the data was then transformed using the Uniform Manifold Approximation and Projection (UMAP) method (39). Clusters were formed using the Leiden method (40) with a resolution of 0.25 for the complete dataset (containing all lots described in the manuscript), and projected into the subsetted objects. At first, 15 Leiden clusters were identified. Thereafter, clusters were combined based on biological similarities, which resulted in nine final clusters: Early GC I, Early GC II, Early GC III, GC I, GC II, GC III, Atresia/luteolysis, Mitochondrial gene enriched (Atresia/luteolysis), and Ribosomal gene enriched (Atresia/luteolysis). The marker genes per cluster, as well as certain granulosa cell markers (GJA1, MDK, BBX, HES4, PBX3, YBX3, BMPR2, CD46, COL4A1, COL4A2, LAMC1, ITGAV, and ITGB1), were analyzed to identify cluster cell types. Dot plots and feature plots were generated to assess the expression levels of certain genes per cluster or per sample. Based on downstream analysis, clusters 0 and 5 were subsetted from the original object and they were re-clustered using the Leiden method at a higher resolution of 0.3. Subclusters were combined based on biological similarities. Once all the subgroups were identified, the subclusters resulting from cluster 0 and cluster 5 were merged with the original object. Gene signatures based on genes in the folliculogenesis stage (18) were also analyzed and used in predicting cluster identification. Signature scores for the RUO-OSC-M subset were generated for two groups: Antral and Pre-Ovulatory genes. Based on the final object, 4 more subsets were generated: (1) a RUO-OSC-M object consisting of the following lots: lot 6, lot 7, lot 8, lot 29, lot 48, and lot 56; (2) a RUO-OSC-L object consisting of lot 77 and lot 86; and (3) a CG-OSC-L object consisting of lot 88 and lot 90. UMAPs of all the subsets were generated with the cluster names from the original object.

Bulk RNA-sequencing

Libraries for RNA sequencing were generated using the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina (NEB #7765L) in conjunction with NEBNext Multiplex Oligos for Illumina (Unique Dual Index UMI Adaptors RNA Set1, NEB #7416S) and NEBNext Poly(A) mRNA Magnetic Isolation Module (NEB #E7490L), according to the manufacturer’s instructions. The library pool was sequenced at Azenta using Illumina 2×150 bp, ~350 M PE reads (~105 GB), lightening package. Illumina sequencing files (bcl-files) were converted into fastq read files using Illumina bcl2fastq (v2.20) software deployed through BaseSpace using standard parameters. RNA-seq data gene transcript counts were aligned to Homo sapiens GRCH38 (v2.7.4a) genome using STAR (v2.7.10a) (36) to generate gene count files and annotated using ENSEMBL (36). Gene counts were combined into sample gene matrix files (h5). Computational analysis was performed using the Scanpy (v1.9.6) package. Two h5ad files were joined based on similar features and genes. The two merged files were created into one Anndata object which was normalized to 10,000 UMI per sample and log (ln) plus 1 transformed. Principal component analysis was performed using 30 PCA components. Projection into two dimensions was performed using the UMAP method (39). This analysis was performed to evaluate the CG-hiPSC sub-clones: 2-A7, 2-C9, 2-D10, 2-G1, 2-G11, 2-H10, 3-C7, 3-D3, and 3-E3.

Barcode-Seq identification of residual OSC presence in human oocytes and embryos

Low input bulk RNA sequencing of individual post-IVM-oocytes was performed using the NEB Low Input RNA-seq library preparation kit and sequenced on an Illumina MiSeq. Two reference positive control samples of RUO-OSCs were directly sequenced to serve as controls, using the same library preparation and sequencing kit. Additionally, two negative control samples of patient derived cumulus cells were used as a somatic cell negative control and likewise prepared using the same library preparation and sequencing kit. All samples were assessed for RNA integrity scores (RIN) of greater than 8 and sequenced to the same depth.

In order to determine the residual presence of OSCs using RNA-sequencing, a set of sequences unique to OSCs was utilized as a target for assessment. This unique sequence consists of a 20 base pair barcode that is expressed uniquely on the 3’ untranslated region (UTR) or the exogenous transcription factors (TF) used to manufacture Fertilo OSCs.

Barcode identification was performed on Fastq files using the following method. Briefly, a Kallisto reference genome library was built using Kallisto index, depending on target sequences, using kmers = 15 to kmers =31. An index on specific barcode sequences and flanking regions down stream of inserted TF sequences was included. Additionally, an index reference genes (KRT19 and EpCAM) known to be active in OSC lines and oocyte cells was included. Kallisto sudo alignment of transcriptomic read files (FASTQ) to kallisto genomic index was performed. Read counts were then collected and compared of barcode sequences found down stream of inserted synthetic transcription factor (TF) genes.

To ensure the oocytes were sequenced with enough depth and were not negative due to low read depth, two highly expressed housekeeping genes were likewise analyzed, KRT19 and EpCAM, to filter samples with unsuitably low read depth. These housekeeping genes were chosen for their conserved expression in both oocytes and reference somatic cell and Fertilo OSC controls. Relative quantification of the barcodes and gene sequences of interest were performed using kallisto pseudoalignment discovery (v 0.44.0) and expressed as a transcript per million (TPM) measure.

To select barcodes that are sensitive and specific to detection of Fertilo OSCs, three 20 bp barcodes, with 2 kmer lengths, as well as the endogenous gene were evaluated for sensitivity and specificity. Sensitivity was assessed as being detectable in the reference RUO OSC positive control sample. Specificity was assessed as being detectable in the reference RUO OSC positive control sample and undetectable in the cumulus cell somatic negative control.

For analysis of residual Fertilo OSCs in embryo samples, the preimplantation genetic testing for aneuploidy (PGT-A) of trophectoderm biopsies was utilized. The platform utilized for PGT-A utilized microarray analysis (Juno Genetics) on 4 to 8 cell biopsy samples. The analysis utilized single nucleotide polymorphisms (SNPs) and haploblock mapping to determine whether the biopsy sample is euploid, aneuploid, segmental aneuploid, or mosaic. Furthermore, for each embryo, a sample of maternal blood and paternal sperm were utilized to compare the SNPs of the embryo to the SNPs of the maternal and paternal donors. Furthermore, a sample of RUO OSCs was provided to determine the SNP profile of residual RUO OSCs. Therefore, through analysis of the embryo’s SNPs presence of maternal, paternal, and residual Fertilo OSC DNA could be accurately identified in parallel to aneuploid status. For all biopsies, day 3 laser-assisted zona drilling was performed, and biopsies were performed on embryos that reached transferable quality grade on day 5 through 7 of development, as determined by the Gardner Score of 3BB or greater.

Design of Experiments (DOE)

DOE was conducted using JMP software (JMP 17, SAS Institute Inc., Cary, NC, USA) to optimize experimental parameters, including media supplementation, substrates, and doxycycline treatment designs. Factors were selected from established modulators for OSC specification and factor ranges were determined through literature review (table S2). Responses were FOXL2 expression and viability. Viability was used to screen and remove experimental groups below 50% cell survival prior to the final statistical analysis. A custom design was employed and optimized for D-optimality to investigate main effects and interactions. JMP software facilitated generation of the experimental design matrix, as well as statistical analysis through response surface methodology (RSM), ANOVA, and optimization of conditions to maximize response desirability for FOXL2 expression.

Proteomics

Liquid chromatography followed by tandem mass spectrometry (LC-MS/MS) analysis was conducted on a series of samples, including RUO-hiPSC (n=1), CG-hiPSC (n=1), RUO-OSC-M lot 56 at time 0h (n=1) and after 24h (n=1) of culture with supplemented IVM media, and CG-OSC-L lot 88 at time 0h (n=1) and after 24h (n=1) of culture with supplemented IVM media. Each condition involved the analysis of 2 million cells. Supplemented IVM media consisted of IVM media (Origio) supplemented with 75 mIU/mL of recombinant FSH (Millipore), 100 mIU/mL recombinant hCG (Sigma), 500 ng/mL of androstenedione (Sigma), 1 ug/mL of doxycycline (StemCell Tech) and 10 mg/mL of human serum albumin (HSA; Life Global). Conditioned media derived from RUO-OSC-M lot 56 (n=1) and CG-OSC-L lot 88 (n=1) were also analyzed. To generate conditioned media, 2 million OSCs were cultured in 2 mL of supplemented IVM media for 24 hours, maintaining the ratio of the intended clinical cell dose of 1,000 OSC cells per 1 µL of media. The 24-hour culture was performed with CO₂ set for a pH of 7.2-7.4. Following culture, OSCs and conditioned media were separated and processed independently. Supplemented IVM media without OSCs was used as a media control. The conditioned media were subjected to consecutive centrifugations (300 x g, 1,200 x g, and 3,000 x g) to remove cellular remnants, and then passed through albumin depletion columns (AVK-50, AlbuVoid Albumin Depletion Columns Biotech Support Group) to eliminate HSA-derived albumin. Proteins from cells and conditioned media were precipitated using acetone, re-suspended in 0.1% RapiGest and 25 mM ammonium bicarbonate, reduced with DTT, and alkylated with iodoacetamide, before undergoing in-solution trypsin digestion overnight at 37°C. The resulting peptides were desalted using C18 stage-tip columns prior to analysis using a Thermo Fisher Scientific EASY-nLC 1200 coupled online to a Fusion Lumos mass spectrometer (Thermo Fisher Scientific). For LCMS/MS analysis, buffer A (0.1% FA in water) and buffer B (0.1% FA in 80 % ACN) were used as mobile phases for gradient separation. For peptide separation, a packed in-house 75 µm x 15 cm chromatography column (ReproSil-Pur C18-AQ, 3 µm, Dr. Maisch GmbH, German) was used. Peptides were separated with a gradient of 5–40% buffer B over 30 min, and 40%-100% B over 10 min at a flow rate of 400 nL/min. Fusion Lumos mass spectrometer operated in a data independent acquisition (DIA) mode, collecting MS1 scans in the Orbitrap mass analyzer from 350-1400 m/z at 120K resolutions. The instrument was set to select precursors in 45 x 14 m/z wide windows with 1 m/z overlap from 350-975 m/z for HCD fragmentation. MS/MS scans were collected in the orbitrap at 15K resolution. Data analysis involved searching against human Uniprot database (8/7/2021) using DIA-NN v1.8 with filtering for 1% false discovery rate (FDR) for both protein and peptide identifications. Protein intensities were normalized and log transformed for relative quantitation, and multiple hypothesis correction of p-values was performed using the Benjamini-Hochberg method. Proteomic analyses were conducted at the Proteomics and Metabolomics Core Facility at Weill Cornell Medicine (New York, USA). GO Chord graphs generated with the free online platform, SRplot (41).

Immunostaining (immunofluorescence and flow cytometry)

Immunofluorescence staining was conducted on fixed hiPSCs adhered to a slide, following the protocol recommendations from the New York Stem Cell Foundation (NYSCF). Briefly, hiPSCs were fixed in 4% paraformaldehyde (PFA), followed by a blocking step in a blocking buffer (3% donkey normal serum and 0.1% Triton-X). The primary antibodies were mouse monoclonal antibody against OCT3/4 (1:200; sc5279, Santa Cruz Biotechnology), goat polyclonal antibody against SOX2 (1:50; AF2018, R&D systems), goat polyclonal antibody against NANOG (1:50; AF1997, R&D systems), and Alexa Fluor 488 mouse monoclonal antibody against TRA-1-60 (1:100; 560173, BD Biosciences). The secondary antibodies were Alexa Fluor 555 donkey anti-mouse IgG (A32773, Invitrogen), Alexa Fluor 488 donkey anti-goat IgG (A32814, Invitrogen), and Alexa Fluor 647 donkey anti-goat IgG (A32849, Invitrogen). All antibody dilutions were prepared in a blocking buffer and incubated at room temperature (RT) for 1 hour. After incubations, samples underwent three washes of 30 min each with PBS containing 0.1% Tween-20 (PBST). Subsequently, samples were mounted with Prolong Gold mounting medium prior to imaging using an ECHO Revolve microscope.

Flow cytometry analyses were conducted on live and fixed RUO-OSC and CG-OSC samples. For the analysis of live cells, cells were incubated with a PE-conjugated mouse monoclonal antibody against CD82 (1:50 dilution; 342104, BioLegend) in FACS wash (dPBS with 5% fetal bovine serum (FBS)). After incubation, cells were washed with FACS wash, stained with propidium iodide (1:20 dilution; P4864, Millipore Sigma) for live/dead staining, and subsequently analyzed using a CytoFlex Flow Cytometer. For the analysis of fixed cells, cells were fixed with 4% PFA for 15 minutes at RT and then washed with dPBS. After, cells were permeabilized using FACS wash solution containing 0.1% Triton X-100 (A16046.AE, Thermo Fisher Scientific). The primary antibodies were mouse monoclonal antibody against OCT3/4 (1:50 dilution; sc5279, Santa Cruz Biotechnology) and rabbit polyclonal antibody against FOXL2 (1:100 dilution; A16244, ABclonal). The secondary antibodies were Alexa Fluor 555 donkey anti-mouse IgG (A32773, Invitrogen), and Alexa Fluor 488 donkey anti-rabbit IgG (A32790, Invitrogen). Following incubations, cells were washed with FACS wash containing Triton X-100 and then analyzed using a CytoFlex Flow Cytometer. Unstained cells (negative controls) were used to determine the gating strategy.

For analytical testing, hiPSCs were stained with Alexa Fluor 488-conjugated TRA-1-60 antibody (1:100 dilution; TRA-1-60 Monoclonal Antibody Alexa Fluor 488, BD Pharmigen™, Cat.#: 560173) for 1 hour on ice. Subsequently, cells were stained with propidium iodide (1:20 dilution; Sigma, Cat.#: P4864) right before analysis to distinguish live/dead cells. A total of ≥ 30,000 TRA-1-60+ live cells were acquired on a CytoFLEX S Flow Cytometer and the data was analyzed using FlowJo 10.10.0. The gating strategy was defined using an unstained hiPSC sample.

RT-qPCR and genotyping PCR

For genotyping PCR, DNA extraction from various hiPSC clones was carried out using the QuickExtract DNA Extract Solution (Epicentre), following the manufacturer’s instructions. PCR amplification was performed using Q5 High-Fidelity 2X Master Mix (New England Biolabs) for 35 cycles with a 20-second extension time. Subsequently, PCR was performed to validate the integration of the 3 TFs NR5A1, GATA4, and RUNX2. The PCR protocol involved an initial denaturation step at 98°C for 30 seconds, followed by 35 cycles of denaturation at 98°C for 10 seconds, annealing at 66-70°C (NR5A1, 67°C; RUNX2, 66°C; GATA4, 70°C) for 10 seconds, and extension at 72°C for 20 seconds, with a final extension step of 2 minutes. The reaction was then held at 4°C. Gel electrophoresis (2% agarose) was performed to confirm the presence of TFs.

RT-qPCR was performed to assess the gene expression of POU5F1 and NANOG, following protocol recommendations from the New York Stem Cell Foundation (NYSCF). RNA extraction was performed using the Quick-RNA Microprep Kit (Zymo Research) following the manufacturer’s instructions. cDNA synthesis was carried out with the LunaScript RT SuperMix Kit (New England Biolabs). Quantification of RNA and cDNA was performed using Nanodrop. PowerUp SYBR Green Master Mix (Applied Biosystems) was used for RT-qPCR. For RT-qPCR, the thermocycler program consisted of a primer annealing stage at 25°C for 2 minutes, followed by cDNA synthesis at 55°C for 10 minutes, concluded with heat inactivation at 95°C for 1 minute. The RT-qPCR protocol involved an initial denaturation step at 95°C for 2 minutes, followed by 40 cycles of denaturation at 95°C for 3 seconds, annealing at 60°C for 30 seconds, and an analysis step.

Functional assessment (oocyte maturation)

The oocyte maturation-stimulating potential of various OSC batches was used to evaluate potency. Briefly, human immature oocytes surrounded by cumulus cells, known as cumulus-oocyte complexes (COCs), were retrieved from subjects that underwent minimal stimulation protocols. Subsequently, these immature COCs were co-cultured with different batches of OSCs for 24-30 hours to facilitate IVM. After IVM, oocytes were evaluated for their maturation state and categorized into immature stages (GV oocyte or MI oocyte) or mature stage (MII oocytes). MII oocyte maturation rate (%) was calculated by dividing the total number of mature MII oocytes by the initial number of immature oocytes; MII oocyte maturation rate was used as the potency readout. For a comprehensive description of the methods follow, refer to our recent publication (11). Sibling oocytes were used for initial comparisons described in fig. S2B. For all the other comparisons, the control and OSC-IVM groups consist mostly of oocytes from non-overlapping donor cohorts. All analyses were performed using One-way ANOVA (Dunnett’s multiple comparison test). The number of oocyte donor participants and the total number of oocytes used per OSC batch assessed are detailed in table S7.

Control conditions for hiPSC manufacturing

The controlled conditions used for manufacturing were established based on the current Good Manufacturing Practice (cGMP) conditions outlined in 21 CFR Parts 210 and 211 with a focus on use of suitable materials, implementation of contamination and cross contamination controls, and adherence to standard operating procedures. Briefly, a quality control unit was established at Gameto and all personnel and consultants met qualifications for assigned functions; all buildings and facilities were deemed suitable for manufacturing; equipment was determined to be appropriate for use with detailed maintenance and cleaning procedures; raw materials were received, stored, handled, and tested appropriately; standard operating procedures for production and processing were written and executed, and deviations were reported; packaging, labeling, holding, and distribution procedures were put in place; and protocols for salvaged drug products were developed. In addition, suitable laboratory controls, as well as procedures for maintaining records and generating reports, were established.

Safety testing

Mycoplasma testing was performed on OSCs in duplicate. Mycoplasma testing included detection of the presence of Mycoplasma pulmonis and Mycoplasma sp. DNA was extracted and real-time quantitative polymerase chain reaction (RT-qPCR) was used for the direct detection of nucleic acid sequences for the mycoplasma targets. In addition, mycoplasma testing was also performed on cell culture supernatant collected during harvesting day. Cell culture supernatant was inoculated into a single flask containing sterile complete mycoplasma broth. The flask was incubated under microaerophilic conditions at 36°C ± 1°C under 5-10% CO₂ in nitrogen for 7 days. Broth culture was collected, RNA was isolated, and RT-qPCR was then performed to amplify and detect nucleic acid sequences from organisms within the class Mollicutes, including mycoplasma species and related Acholeplasma laidlawii.

Endotoxin testing was performed on thawed hiPSC or OSC samples using a quantitative kinetic chromogenic LAL (Limulus Amebocyte Lysate) assay in accordance with the USP <85> compendial method.

Sterility testing was performed in accordance with the USP <71> compendial method to detect bacteria and fungi. Samples are aseptically directly inoculated onto vessels containing Fluid Thioglycollate Medium (FTM) and Soybean Casein Digest Medium (SCDM)/Tryptic Soy Broth (TSB). Two vessels containing uninoculated FTM and two vessels containing uninoculated SCDM/TSB serve as negative controls. FTM vessels are placed at 31°C – 35°C for 14 days, whereas SCDM/TSB vessels are placed at 21°C – 25°C for 14 days. All vessels are observed for evidence of growth on days 3, 4 or 5, day 7 or 8, and day 14. The test is considered valid if no growth is observed in the negative controls.

Adventitious agents (human test panel) testing was performed on hiPSCs and OSCs. Adventitious agents (human test panel) testing included detection of the presence of the most common human adventitious agents: Hepatitis A, Hepatitis B, Hepatitis C, HTLV Type I & II and STLV Type 1, Human Immunodeficiency Virus Type 1, Human Immunodeficiency Virus Type 2, Epstein Barr Virus, Human Cytomegalovirus, Human Herpesvirus Type 6, Human Herpesvirus Type 7, Human Herpesvirus Type 8, Parvovirus B19, JC Virus, and BK Virus. DNA is extracted, and real-time quantitative polymerase chain reaction (qPCR) is used for the direct detection of nucleic acid sequences for Hepatitis B, Epstein Barr Virus, Human Cytomegalovirus, Human Herpesvirus Type 6, Human Herpesvirus Type 7, Human Herpesvirus Type 8, Parvovirus B19, JC Virus, and BK Virus. RNA is extracted, and real-time reverse transcription quantitative polymerase chain reaction (RTqPCR) is used for the direct detection of nucleic acid sequences for Hepatitis A, Hepatitis C, HTLV Type I & II and STLV Type 1, Human Immunodeficiency Virus Type 1, and Human Immunodeficiency Virus Type 2.

Genomic integrity of hiPSCs and OSCs is assessed through whole genome sequencing (WGS). Sample processing is performed using the PacBio Nanobind PanDNA Kit, followed by library preparation with the SMRTbell® Prep Kit 3.0. Sequencing is then conducted on the PacBio Revio, a long-read sequencing system that utilizes single-molecule real-time (SMRT) sequencing technology. To evaluate genome-wide DNA alterations that may have arisen during the manufacturing of hiPSC banks and Fertilo OSC lots, variant calling analysis is performed. The effects of detected variants on transcripts are annotated and predicted using Ensembl. Comparisons are made between the test sample DNA and the GRCh38 reference genome, with intersection tests conducted for single nucleotide polymorphisms (SNPs), insertions/deletions (INDELs), and structural variants (SVs).

The presence of residual hiPSCs in the test article (Fertilo) is assessed by RT-qPCR targeting the pluripotent stem cell markers OCT4 and NANOG. GAPDH serves as a housekeeping gene for normalization. The acceptance criteria for Residual hiPSC testing is “OCT4 ≤ 0.1% and NANOG ≤ 0.1%”.

Residual doxycycline was performed on OSC samples. Exposure to doxycycline throughout the hiPSC differentiation process induces overexpression of the transcription factors, NR5A1, RUNX2, and GATA4, directing differentiation of hiPSCs towards OSCs. To confirm that doxycycline is not carried over into the final product, an ELISA against doxycycline (Abcam, Cat. #ab285232) was performed on conditioned media.

Embryotoxicity is tested using a mouse embryo assay (MEA) to confirm that Fertilo does not cause harm to embryos or gametes, following the FDA guidance recommendations on conducting “Mouse Embryo Assay for Assisted Reproduction Technology Devices”. At the completion of embryo culture, blastocysts formation rate (BFR) is assessed, and the acceptance criteria is ≥ 80% embryos successfully developed into blastocysts.

Murine oocyte maturation assay (MOMA)

Hybrid (B6/CBA) females, and male mice from the same genetic backgrounds, were purchased from Janvier Labs. Upon arrival, all mice were quarantined and acclimated to the PCB Animal facility (PRAAL) for approximately 1 week prior to use. Mice were housed with a 12-hr light/dark cycle (lights on at 7:00 A.M.) with ad libitum access to food and water. All procedures involving mice, e.g., handling, administration of hormones for superovulation of females were conducted at the PRAAL Animal Facility, while procedures involving oocyte manipulation, e.g., oocyte collection, oocyte insemination and embryo culture up to blastocyst were performed at Embryotools’ laboratories.

A total of 10 mL of Fertilo IVM media, consisting of IVM Base Media (Cooper), 10 mg/mL human serum albumin (HSA) (Cooper), 75 mIU/mL recombinant follicle stimulating hormone (rFSH) (Sigma), 100 mUI/mL human chorionic gonadotropin (hCG) (Sigma), and 500 ng/mL androstendione (Sigma), and 10 mL of IVM control media, consisting of IVM Base Media (Cooper), 10 mg/mL HSA (Cooper), 75 mIU/mL rFSH (Sigma), 100 mUI/mL hCG (Sigma), and 500 ng/mL androstendione (Sigma), were prepared and stored at 4°C until use. Plating dishes were prepared 16-24 hours prior to oocyte culture to allow for equilibration. The wells of a universal GPS dishes (UGPS-010, Cooper) were prepared with 50 µL LAG media, 100 µL Fertilo IVM media, 100 µL IVM control media, or 50 µL Fertilo IVM media/ IVM control media for wash droplets. Then, 12 mL of mineral oil (Kitazato) was used to overlay droplets. All dishes were incubated overnight at 37°C with CO2 set for a pH of 7.2-7.4 and 5% O2.

On the day of oocyte culture, Fertilo OSCs were seeded 2-5 hours before co-culture. To thaw Fertilo OSCs, one OSC cryovial was removed from liquid nitrogen storage and placed upright into a 37°C water bath or dry bead bath to incubate for 2-3 minutes or until ice dissolved. Then, 400 µL of pre-warmed Fertilo IVM media was added to the vial and the cells were mixed by pipetting. After, 500 µL of the cell suspension was added to a 1.5 mL Eppendorf tube and centrifuged at 300 x G for 5 minutes at room temperature (RT). The supernatant was then removed and the cells were gently resuspended in another 400 µL pre-warmed Fertilo IVM media. Following centrifugation at 300 x G for 5 minutes at RT, most supernatant was removed while ensuring not to disturb the pellet. The cells were then resuspended in pre-warmed Fertilo IVM media to 100,000 OSCs per 50 µL according to batch specifications provided in the Fertilo Certificate of Analysis. Then, the plating dishes were removed from the incubator and 50 µL of media was removed from the wells with Fertilo IVM media. After gently pipetting the cell suspension, 50 µL of OSCs were added to the wells with Fertilo IVM media, for a concentration of 100,000 OSCs in 100 µL media. The plate was then placed back into the 37°C incubator for at least 2 hours until oocyte seeding.

Fresh immature mouse oocytes at the germinal vesicle (GV) stage were collected from the ovaries of hybrid (B6/CBA) females between 6 and 8 weeks of age stimulated with 7.5 IU pregnant mare serum gonadotropin (PMSG) 48 hours before retrieval. GVs with cumulus cells were then randomly split between the Fertilo IVM group or the IVM control group. Oocytes were then plated and the plates were incubated at 37°C for 18 hours for IVM. After IVM, oocytes were stripped using hyaluronidase (H3884-50MG, Sigma-Aldrich) and assessed for maturity. Mature oocytes with first polar body extrusion were selected for intracytoplasmic sperm injection (ICSI). Fresh sperm from B6/CBA hybrid strain mice was obtained in a microdroplet of culture medium and cultured for 15 minutes at 37°C, 5% CO2, and 5% O2. After incubation, 3 µL of the concentrated sperm solution was further diluted in a 150 µL droplet of the same culture medium. ICSI was performed using a piezo drill-based protocol optimized for the mouse species. Briefly, mouse sperm heads were isolated from the tail and injected into the mature oocytes. Following injection, the oocytes were thoroughly washed and cultured until day 5 (120 hours) in benchtop incubators at 37°C, 5% CO2, and 5% O2. Embryo development was monitored until the last day of culture, at which point images of each embryo were acquired.

Clinical Treatment using Fertilo and Traditional IVM

Once informed consent was obtained, subjects were screened and evaluated for participation. If required, patients were pre-treated for vitamin D deficiency or insulin resistance using calcifediol or metformin respectively before beginning stimulation. Eligible subjects that met all of the inclusion criteria were started on oral contraceptive pills (OCPs) for 10-15 days with a 5-day washout period. For evaluative purposes, the final day of the washout period was considered Day 1 of the menstrual cycle. On Day 2, transvaginal ultrasound (TVU) was performed to assess the antral follicle count (AFC) and follicle diameter for monitoring; in addition, baseline evaluation of serum levels of E2, P4, LH, and FSH were performed via a local laboratory. On Day 2, controlled ovarian stimulation (COS) began via administration of 100 mg clomiphene citrate (Clomid) once daily for 5 days (Day 2-Day 6). Ultrasound was performed on Day 7 to determine the number of 150 IU rFSH (Gonal-F) injections that would be administered. If the leading follicle was 6 mm or less, three once daily injections of rFSH were administered on Days 7, 8, and 9. If follicles were 7 mm to 9 mm on Day 7, 150 IU (Gonal-F) was administered once daily on Day 7 and 8, for a total of two injections. If the leading follicle was greater than 9 mm on Day 7, a single injection of 150 IU (Gonal-F) was administered on Day 7.

The ideal leading follicle diameter range was 10-12 mm. On Day 8, 9 or 10, a recombinant hCG (Ovidrel) trigger of 250 µg was administered and the oocyte retrieval (OR) was scheduled for 34-36 hours later. In the event of rapid follicular growth, if a follicle was 13 mm or greater at the Day 2 or Day 7 ultrasound, the cycle was cancelled and repeated in a subsequent menstrual cycle.

In the observational safety phase, all twenty subjects were assigned to OSC-IVM (Fertilo) treatment. For the control comparator phase, patients were randomized at a 1:1 ratio to OSC-IVM (Fertilo) or Traditional IVM (Media Only IVM) on the day of hCG trigger. Oocyte retrieval was performed according to standard clinical practice using a 17-gauge single lumen needle with 70 mm Hg suction pressure. Rapid needle movement (curettage) was performed to improve oocyte extraction from the follicular wall. All oocyte retrievals were performed under anesthesia. Oocytes were collected into warmed collection media containing heparin in order to minimize clotting. Further, tubing length was minimized and the needle was flushed every 4-5 punctures to minimize clotting and temperature fluctuations during retrieval. No follicular flushing was performed in any retrieval cycle.

Oocyte search was conducted according to standard IVM procedures. Briefly, follicular aspirate was kept warm and passed over a 70 micron cell strainer to separate and filter blood from potential cumulus oocyte complexes (COCs). The filter was then overturned and the COCs were released into a search media lacking heparin (ASP) via dipping and pipette expulsion. COCs were identified under stereomicroscope and transferred to a warmed holding droplet until all oocytes had been found. Typical IVM oocytes were highly compact, containing one to a few layers of cumulus enclosure. Expanded COCs were rarely identified, due to the use of an hCG trigger, and proper monitoring during stimulation was implemented to keep the leading follicle under 12 mm.

All COCs were cultured in either Fertilo or Traditional IVM for 30 hours. IVM culture was performed the same for both conditions. Briefly, IVM was performed in Birr 4+8 dishes using microdroplet culture. Culture droplets were 100 µL and were utilized for the culture of up to 5 COCs. Traditional IVM utilized the IVM basal media containing supplementation with 75 mIU/mL rFSH (Gonal-F), 100 mIU/mL hCG (Ovidrel), and 10 mg/mL HSA. For OSC-IVM, the same IVM basal media was utilized and performed as previously described. In OSC-IVM, 100,000 viable OSCs were supplemented into each 100 µL droplet the day of oocyte retrieval. All culture media plates were prepared the day before oocyte retrieval to allow for equilibration under IVF grade oil in 37°C incubators with CO2% to yield a pH of 7.2 to 7.4 and 5% O2. For OSC-IVM, the OSCs (Fertilo) were delivered to the clinic under liquid nitrogen vapor shipment and stored in liquid nitrogen until use. For each application, at least two vials of OSCs were prepared to enable treatment of up to 30 COCs across 6 culture droplets, with each droplet containing 100,000 OSCs. Briefly, the day of oocyte retrieval OSCs were thawed at 37°C for 2-3 minutes, gently resuspended using IVM basal media, and centrifuged at 300 x G for 5 minutes to pellet the cells. Supernatent was fully removed to eliminate residual cryoprotectant (Cryostore CS10) and the cells were resuspended using a defined volume of equilibrated complete IVM media described above. A cell count was performed using a cell counting chamber and 100,000 viable OSCs were supplemented into each droplet by replacement of the corresponding volume with the cell suspension. The dishes were incubated for at least 30 minutes prior to oocyte co-culture to allow for re-equilibration. All COCs were washed in IVM basal media and moved into the culture droplets of either OSC-IVM or Traditional IVM and cultured for 30 hours in a 37°C incubator at a %CO2 to achieve a pH of 7.2 to 7.4 and 5% O2.

Mature oocytes were fertilized using ICSI and cultured in group or single culture using single-step embryo culture medium. Assessment of the fertilization rate was performed at 16-18 hours post-ICSI, then at Days 3, 5, 6, and 7 post IVM. A standardized embryo grading system based on the classification system by Gardner & Schoolcraft (Gardner classification) was used on Days 5-7. Embryos that did not meet their respective checkpoints were discarded and blastocysts of freezable quality (Gardner grade 3BB or greater) underwent vitrification (Kitazato). Embryos were hatched using laser-assisted hatching and underwent trophectoderm biopsy prior to vitrification on the day of freezing. Preimplantation genetic testing for aneuploidy (PGT-A) was performed using microarray-based or SNP-based NGS analysis in local laboratories. Only embryos considered euploid were utilized for transfer. Low grade mosaicism, after consultation with a geneticist, were utilized for transfer depending on patient preference. The highest quality embryo was thawed later for frozen embryo transfer (FET) within 2 months of cryopreservation. Only single blastocyst transfers were allowed and all transfers were FET cycles.

For uterine priming, starting on the first day of menses, subjects received 2 mg oral estradiol treatment twice daily with an optional 100 µg patch replaced every three days until the endometrial thickness was ≥7 mm. Progesterone was then administered at 50 mg intramuscular once daily to prepare for ET or via a combination of intravaginal (800 mg) and subcutaneous progesterone. Progesterone could be increased to 100 mg daily if clinically indicated due to bleeding or a low progesterone level.

Following ET, serum beta human chorionic gonadotropin (BHCG) levels were measured 10-14 days post-transfer and followed until an appropriate rise was confirmed (2-3 days following 1st confirmation), indicating a biochemical pregnancy.

Transvaginal ultrasounds were performed at 4-6 weeks gestation to establish the presence of a gestational sac, indicating clinical pregnancy, and a follow-up ultrasound was performed at 8-12 weeks of gestation to confirm ongoing pregnancy with a normal, healthy heartbeat. If a subject became pregnant, P4 and E2 administration were continued for up to 8-12 weeks of gestation.

The site staff followed-up with subjects who had an ongoing pregnancy after discharge from the clinic to obtain a second trimester pregnancy ultrasound record (Week 20-24 Anomaly scan performed by an OBGYN) and serious adverse events (SAE) collection via a second trimester in person visit. Additionally, the clinic obtained birth data from the patient. Adverse events of special interest (AESI) such as response to stimulation and retrieval, ovarian hyperstimulation syndrome (OHSS), pregnancy loss, fetal anomalies, pregnancy complications, birth weight, birth defects, and neonatal complications were collected. For the purposes of this evaluation, only the outcomes of the first embryo transfer were considered in the rates.

Data Analysis

Significant differences in oocyte maturation rates were evaluated using unpaired t-tests compared to control with a p-value <0.05 considered statistically significant in all cases. For the MOMA, a Fisher’s exact test was used to compare maturation, ICSI survival, cleavage, and blastocyst formation rates. A p-value <0.05 was considered statistically significant in all cases. Data analyses were conducted using GraphPad Prism software (GraphPad Software, San Diego, CA, USA). In the clinical evaluation second phase (comparative), for all demographic comparisons between arms, an unpaired, two-tailed T-test was performed. The comparative statistical analysis between Traditional IVM and Fertilo groups for embryological outcomes was performed using logistic regression. Analysis was performed by logistic regression on outcomes per COC per cycle comparing Fertilo versus Traditional IVM treatment. Significance was based on the Fertilo distribution compared to the Traditional IVM mean for each development point. A p-value of less than 0.05 was considered statistically significant in all tests. Both phases of the evaluation were designed as a small cohort, observational analysis to provide preliminary evidence of safety and efficacy at the embryological level and was not powered for determination of superiority or non-inferiority for in vivo endpoints after embryo transfer.