hInGeTox: A human-based in vitro platform to evaluate lentivirus contribution to genotoxicity

hInGeTox: A human-based in vitro platform to evaluate lentivirus contribution to genotoxicity | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article h InGeTox: A human-based in vitro platform to evaluate lentivirus contribution to genotoxicity Mike Themis, Saqlain Suleman, Sharmin Al Haque, Andrew Guo, Huairen Zhang, and 16 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3837253/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 15 Jul, 2025 Read the published version in Gene Therapy → Version 1 posted 11 You are reading this latest preprint version Abstract Lentivirus vectors are effective for treatment of genetic disease and cancer, however, vector related insertional mutagenesis related genotoxicity is of concern and currently available safety models are not reliably predictive of safety in humans. We have developed h InGeTox as the first human in vitro platform that uses induced pluripotent stem cells and their hepatocyte like derivatives to further understand LV host interaction for vector safety evaluation and design. To characterise LV for genotoxic association, we used LTR and SIN configuration LV infected cells for a multi-omics analysis on data that included LV integration sites in cancer genes and their associated differential expression, clonal tracking of IS, novel vector/host fusion transcripts and methylated cancer genes with altered gene expression after infection. We present h InGeTox as a useful pre-clinical tool to identify lentivirus contributory factors mediating genotoxicity to use for improving LV design to provide gene therapy. Biological sciences/Stem cells/Stem-cell differentiation Biological sciences/Molecular biology Biological sciences/Cell biology Biological sciences/Cancer Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Introduction Cell and gene therapy clinical trials have increased over the past 10 years from 1,800 to 5000 listed in the National Institutes of Health that include new technologies involving CAR-T and gene edited cells. Currently, 40% of all trials are industry sponsored. In the US, each investigational new drug application (IND) requires FDA review, which primarily is based on safety considerations and may result in clinical hold ranging from 2 to 19 months. Recently, INDs have increased enormously with gene therapy product development and diversity and with commercial interest [1]. Between Jan 2020 and Dec 2022, 33 clinical holds were announced that concerned CAR-T therapies (15%) and lentivirus vector-based therapies (45%) [2]. Several gene therapy trials have successfully used gamma retrovirus (γ-RV) and lentivirus (LV) vectors for therapeutic gene delivery that offer permanent gene delivery to the host genome following virus integration. Integration, however, risks insertional mutagenesis and differences in γ-RV and LV integration site (IS) selection is known to influence their genotoxic potential [3]. Genotoxic effects by γ-RV includes host protooncogene expression changes driven by the LTR enhancer, as identified in X-SCID, WAS and CGD trials where oncogene upregulation was caused by γ-RV integration in either orientation to the gene locus [4–7]. Promoter activation involving RV in the same orientation to gene transcription is also known to cause gene upregulation [8]. As a result of this, modification of the LTR to SIN configuration in LV to abrogate promoter activity has proven to be a major improvement limiting host gene activation upon integration [9]. However, readthrough from the internal promoter used to replace the modified LTR has been reported to drive local host gene expression [8]. With SIN configuration, integration preference in the gene transcription unit and the developments of 3rd generation design to avoid the emergence of replication competence, LV vectors are considered much safer than RV. Hence, LV have become the vectors of choice and are currently used to treat a number of rare genetic diseases and for CAR-T cell immunotherapy. Unfortunately, genotoxicity concerns still remain and with the appropriate configuration LV has been shown capable of oncogenesis in tumour prone mice [10,11]. LV splicing with host cancer genes has also been shown to generate novel gene fusions with the potential to drive clonal expansion[12,13] as identified in a β−thalassemia clinical trial, where integration of a SIN LV caused 3’ end substitution of the HMGA2 gene, which abolished let7 microRNA control of this protooncogene, reduced HMGA2 degradation and clonal proliferation [14–16]. The role of epigenetics in cancer progression is also well known [17–20], and LV infection has also been found associated with hepatocellular carcinoma development in mice, that was suspected to be caused by methylation changes of protooncogene promoters under the control of the E2F transcription factor [21]. More recently, LV have been used successfully to generate CAR-T cells carrying anti-CD19 CAR cassettes for cancer immunotherapy. However, in a clinical trial against chronic lymphocytic leukaemia, CAR-T cells were found to persist as a result of gene inactivation by the CAR carrying vector. In this case, intronic insertion resulted in removal of control of TET2 after splicing with the LV vector. In a CD22 CAR-T trial, LV mediated CBL oncogene activation has also been suspected to have caused CAR-T cell persistence [22,23]. To understand RV and LV mediated genotoxicity more clearly and predict potential vector related side effects, in vitro and in vivo murine-based models have been developed. These include Cdkn2a null mice with inactivated p53 and pRb pathways that have been valuable in showing that the risk of tumour development by LV vectors is approximately 10-fold lower than RV. In a fetal/neonatal murine model LV delivery has also been found associated with high frequency liver cancer [21]. In vitro , the immortalization (IVIM) model that uses murine hematopoietic stem cells (HSC) has also been useful for vector risk assessment. This model demonstrates differences in RV and LV IS preference and integration in Evi1 and Prdm16 proto-oncogenes is responsible for cell transformation with RV greater than LV by a factor of 3:1. As a consequence, IVIM has been accepted by several regulatory agencies for pre-clinical evaluation of RV and LV safety. More recently data from this model has been used to provide transcriptomic signatures of leukaemogenesis supporting its use as a surrogate assay for genotoxicity assessment (SAGA) [24]. Although the models currently used to understand and assess LV genotoxicity have proven valuable, they are still considered potentially bias or over sensitive. Furthermore, no test can reliably predict long term safety in humans with widely variable predisposition to cancer. As a human based safety model is clearly needed, we chose to develop an alternative strategy to use for LV safety in which factors known to be contributory to genotoxicity are identified associated with LV design. For this, human induced pluripotent stem cells (iPSC) and their hepatocyte-like cell derivatives were used. iPSCs have been used widely to model human diseases and for pharmacotoxicological studies of disease treatment [25–27]. iPSC can be reprogrammed from a variety of patient cells and offer a personalised approach to risk assessment by considering the genetic background of the host. We have previously shown iPSC can be reliably reprogramed to 3D hepatocyte like cells[28] and to be true liver surrogates matching primary hepatocytes at the transcriptional level (submitted manuscript). In this report, we used positive and negative control LV, carrying native and SIN configuration LTR, respectively, to infect iPSC and their 3D HLC derivatives [29]. Data from vector/host interactions were subjected to multi-omics analysis to characterise vector/host interactions believed to support pre-malignancy and oncogenesis. Data obtained from this analysis was then aligned to transcriptional signatures of a range of cancers to profile the genotoxicity potential of each LV. We consider this to be the first human-based model that can provide essential information valuable to identify vector/host interactions indicative of genotoxicity to support improved safe vector design. h InGetox may also be considered useful as a decision-making tool to support LV product approval for gene therapy. Results iPSC derived 3D HLC are true surrogates of primary hepatocytes Bulk cultures of male JHU106i iPSC were differentiated to 3D HLC as previously described [30] and used for RNASeq to confirm iPSC pluripotency and HLC 3D spheroid characteristics. HLC gene expression was also compared with normal and several cancer cell phenotypes using unsupervised clustering, principal component analysis and powerful machine learning that showed HLCs align closely to primary hepatocytes (submitted manuscript). iPSC and 3D HLC cultures were infected with 2nd generation HIV-1 based LV vectors that differ by their LTR configuration and represent positive (pHV) and negative (pHR) controls carrying native LTR and SIN LTR, respectively. Infection of cells used an optimised MOI of 20 as detected by flow cytometry for GFP expression in iPSC at 90% and in 3D HLC spheroids at 85% following dissipation to single cells as previously shown [31]. Vector copy number (VCN), measured via TaqMan™ q-RTPCR (n = 3) next to standard curves generated for absolute vector copy quantification ranged between 1.45–2.54 vg/cell. Culture viability after infection showed no significant difference from untreated cells. LV insertion site analysis Insertion site (IS) analysis of LV was performed by EPTS/LM-PCR [32] following infection. Of 412,786 IS, insertions appeared in introns (60.9%), 3’ (19.1%) or 5’ (16.7%) untranslated regions (UTR) and exons (3.3%) (Figure S1 A) . To investigate the distance between IS and transcription start sites (TSS), the mean distance across each data set was plotted in the scale of 0–1 (exon or intron; normalised against gene length) or Log10 (3’ or 5’ UTR; value in base pairs). While inserts were found to be evenly distributed in introns, those identified in exons mainly congregated at 3’ end of genes (median insertion sites at 82.8% of averaged gene length). On average, inserts identified in 3’ or 5’ UTR were 27.9 kbp or 29.6 kbp away from the protein-coding regions. We then profiled IS gene targeting in iPSC through quantification of gene number in the context of different regulatory regions at two time points (3 and 30 day) after infection (Figure S1 B) . Between these times, gene number with pHR insertions reduced from 10,890 to 7,955 and with pHV insertions from 13,906 to 7,706 suggesting potential enrichment or clonal selection favouring particular IS. Focussing on oncogenes and tumour suppressor genes, this reduction was observed for IS in introns, exons and UTRs, however, at the late time point insertions in oncogenes and tumour suppressor genes remained higher in introns (Figure S1 C and D). IS profiling of gene density, chromosome location, proximity to CpG islands and GC content and position within the gene transcription unit in both iPSC and HLC genomes were identified as expected and as previously reported for HIV-1 LV integration [33]. Pathway analysis of LV IS associate with cellular proliferative potential Hallmark pathway analysis of IS genes identified for each time point of investigation is shown in Fig. 1. IS identified 3 days post iPSC infection were found mainly in exons associated with cell cycle eg. E2F targets, G2M checkpoint and DNA damage eg. c-Myc targets and genes involved in DNA Repair. IS isolated at the later 30 day time point in iPSC were mainly in introns and UTR regions of genes that, in addition to those found at the early time points analysis, were identified in genes associated with the PI3K-AKT/MTOR pathway and epithelial mesenchymal transition. Enriched pathways associated with pHR or pHV IS genes were characteristic of pathways of the inflammatory response and hypoxia, respectively. The inserted genes in each pathway mostly targeted were tumour suppressor genes that included TP53, NBN, POLD1, BRCA1/2, CHEK1, ATRX, BMPR1A, TGFBR2, SPOP, RUNX1, INPP4B, PTEN, PIK3R3, SMAD2, TSC2. For the tumour suppressor genes found with pHV inserts, these were in introns and UTR regulatory regions and present at both the early 3 day and late 30 day time points rather than restricted only to the early time point as observed for pHR insertions. For both vectors, oncogene IS were mainly in introns that included EIF4E, RAF1, GSK3B, CALR, MAPK1, EGFR, LCK, RAC1, RIT1 , and RPTOR associated with the PI3K/AKT/mTOR signalling pathway. Conversely, in infected HLC cells fewer IS genes were found associated with oncogenes or tumour suppressor genes with enriched genes being only members of the TNFα signalling pathway (Fig. 2). IS clonal tracking in iPSC identifies genes associated with clonal outgrowth We next clonally tracked IS in cancer genes in infected iPSC via their sequence count changes (SCC) between the 3- and 30-day time points and in HLC derived from infected iPSC. We focussed on differential absolute SCC IS of ≥ 2-fold (p < 0.05) represented only in significantly enriched genes and over-represented in cancer related biological pathways. Clonal tracking was investigated for IS that resided either in identical or non-identical locations in cancer genes to identify possible expansion of IS in proliferating clones with integrations that may influence gene expression. For SCC of IS identified at the same location at day 3 and 30, these always appeared in introns or UTR regions and in genes associated with eukaryotic translation, cell cycle regulation, kinases associated with protein phosphorylation and RNA export from nucleus ( Table S1 ). IS in genes with increasing SCC found at day 30 but not day 3 were suspected to be representative of IS below the limit of EPTS/LM-PCR detection in the bulk cell populations at the early time point. Seven hundred and seventeen targeted genes that included several oncogenes and tumour suppressor genes were found ( Table S1 ). Most interestingly, although IS clonal tracking found several increased SSC in cancer genes, these appeared most prevalent in HLC derived from iPSC infected by pHV than pHR (n = 29 vs n = 8, respectively). Included in these genes were SET , BRAF and MECOM , previously shown to influence clonal selection. Isolation and analysis of iPSC clones following infection To determine the effect of LV integration on IS gene expression, 7 single cell clones from pHR and 4 from pHV infections were isolated and expanded for DNA and RNA extraction for IS analysis and for q-RT-PCR analysis of gene expression. Following IS identification, for clones with IS that appeared, qRT-PCR analysis of the inserted gene was used for comparison with its expression in non-infected iPSC. For pHR and pHV, of 27 and 21 IS genes, respectively, each were identified > 2 fold upregulated in their expression compared to uninfected iPSC ( Table S2 ). Differential gene expression aligns with unique signatures representative of biological processes critical for oncogenesis To gain molecular insights into potential changes in transcription associated with each LV, RNASeq on infected iPSC at the early and later time points was used to provide an unbiased transcriptome-wide profiling of differentially expressed genes (DEG) against control uninfected cells. At the early time point both pHR and pHV LV DEGs are associated with strong immune signatures with iPSC displaying active cytokine production after infection. For all DEG, regardless of timepoint of harvest, significantly represented DEGs were identified for pHV (n = 1011) and pHR (n = 871) with increases in 14 oncogenes and decreases in 14 tumour suppressor and increases in 10 oncogenes and decreases in 20 tumour suppressor genes, respectively (Fig. 3A and B) . GO term analysis of these genes for annotated biological functions showed signalling pathways involving RNA transcription, protein modification, cell cycle, tyrosine kinase, and NF-kB common to both LV, however, unique enriched pathways for also evident to each LV. Upregulated DEGs in pHV infected iPSC characteristic of oncogenesis were implicated in methylation (n = 11). This contrasted with unregulated DEG associated with pHR infection particular to protection against oncogenesis involving a response to DNA damage (n = 20) and GTPase activity (n = 10) (Fig. 3C and D). Thirty eight (pHR) and forty nine (pHV) genes were identified with commonly shared as DEG (Log2FC > 1 (p 10, p,0.05). GO term pathways were enriched for RNA transcription, WNT signalling and cell differentiation. Between early and late time points DEG were identified for oncogenes and tumour suppressor genes implicated in tyrosine kinase receptor signalling pathways and cellular senescence in pHR infected iPSC in contrast to p13K and MAPK signalling pathway activation in pHV infected cells. DEG associated with pHR (n = 419) comprised of 22 oncogenes and 10 tumour suppressor genes and for pHV (n = 472) DEG comprised of 20 were oncogenes and 13 tumour suppressor genes. For both LV, DEG included DEG for MECOM and LMO2 genes. At the late time point, the major difference between pHR and pHV infected iPSC was characterised by the p53 pathway responding to DNA damage in pHR infected cells compared to an inflammatory response in pHV infected iPSC (Fig. 4A-F). Hallmark analysis of enriched signalling pathways of these gene sets for IS genes also with DEG are shown (Fig. 5 ). For infected HLC (harvested three days after LV transduction) compared to uninfected cells, DEG were also identified. DEG associated with pHR (n = 569) comprised of 37 oncogenes and 51 tumour suppressor genes. This contrasted with pHV (n = 3762) DEG of which 81 were oncogenes, including DEG for MECOM , LMO-2 and BRAF not found with pHR DEG and 82 tumour suppressor genes (Fig. 6). There were nearly 7 fold more upregulated DEG associated with pHV infected HLC than pHR suggesting a difference imposed by the native LTR configuration. GO term analysis of these showed genes mainly associated with tyrosine kinase signalling (n = 23) and protein phosphorylation (n = 19) and other related pathways such as ERK1/2 cascade and PI3K/AKT/mTOR pathway. Focussing on the upregulated DEG used for GO term analysis, pHV infected HLCs exhibit pathways involving chemotaxis (n = 23) and cancer signalling pathways (n = 36). including NF-kB (n = 69), MAPK (n = 54), Wnt (n = 39), JNK (n = 38), and PI3K/AKT This contrasted with pHR associated pathways that were characterised by groups of genes protective against viral infection (n = 36) such as interferon-associated genes IFI27, IFI44L, IFIT5, IRF7, ISG20), DNA damage (n = 35) such as repair proteins DCLRE1C and RAD50, p53-mediated apoptotic proteins BCL3, BCL6, TOPORS, zinc finger proteins ZC3H12A, ZDHHC16, ZBTB4), and autophagy (n = 25). (n = 25) such as ULK2, ATG4B, ATG9A, DEPP1, EPG5, VPS18, VPS13A). Weighted correlation network analysis reveals distinct LV associated gene expression To profile unique gene expression programmes across iPSCs or HLCs particular to pHR and pHV infection, we performed weighted correlation network analysis (WGCNA) and identified 9 significant functional modules associated with distinct biological processes (Fig. 7A). These modules are indicated by colour codes, gene numbers (g), and percentage of shared inserted target genes ( Table S3 ). Modules coloured in brown, turquoise, and green show higher proportions of shared inserted target genes and tumour suppressor genes associated with pHR infected iPSC or HLC (Fig. 7B). Through GO term analysis (Fig. 7C), these modules are associated with critical biological processes. critical for protein modification (brown), cellular metabolism (turquoise), synaptic signalling (blue), stimulus and immune response (green), epithelial cell differentiation (red), RNA metabolism (magenta), phagocytosis (pink), cellular respiration (yellow). The co-expressed genes that are associated with IS (Insert), oncogenes (Onco), or tumour suppressor genes (TSG) have been quantified ( Table S3 ) . Gene splicing with LV and the human genome in infected cells Gene splicing and readthrough is also known to occur between LV and the host genome. We used RNASeq data to identify a total of 69 vector/host fusion transcripts and aligned these with genes in each co-expression module (Fig. 8A). Thirty-eight of these that were found as fusions were also identified as differentially expressed and with LV insertions (Fig. 8B). Triple positive genes for IS, DEG and fusions found in infected HLC were present as a smaller proportion (46%) to those in infected iPSC (79%), most likely due to iPSC gene expression being significantly greater than HLC. Once again, these fusion genes were mainly identified in the turquoise (n = 39), brown (n = 14), and green (n = 6) module categories (Fig. 8C). Alignment of transcriptome changes in infected cells with cancer-specific gene signatures suggest LV associated genotoxicity To investigate further the probability of carcinogenesis associated with pHR or pHV infection, we firstly defined cancer-specific signatures using differential analysis of cancer genes highly expressed in several cancer types, compared with their respective normal tissue. Through pathway analysis, using GO terms or hallmark gene sets, we found these signatures associated with enriched pathways involving nucleic acid synthesis/metabolism, active transcription, cell proliferation, E2F targets and the G2M checkpoint. These signatures were then used to score against the transcriptomes of infected iPSC or HLC from early or late harvest data analysis (Fig. 9A). In general, infected iPSCs are characterised by higher cancer scores than infected HLC as expected. At the early harvest time point, pHR infected iPSC or HLC have higher cancer scores than pHV infected cells, in agreement with the IS and DEGs we identified earlier. However, after continued iPSCs culturing and sample harvesting at day 30, pHV infected cells clearly showed higher cancer scores than pHR cultures (Fig. 9B ). Epigenetic analysis reveals unique lentivirus-induced methylation profiles We then determined whether epigenetic changes via DNA methylation profiling of global methylation of the host genome or of selected genes in infected cells associates with LV infection. We characterised the DNA methylation landscape in host cells by identification of differentially methylated regions across infected samples. Next, we used CpG island methylation changes for cross referencing with IS and RNASeq data, to identify consensus genes for distinct biological processes and highlighted these for oncogenes and tumour suppressor genes. In general, pHV infected iPSC showed greater hypomethylation compared with pHR infected cells. The numbers of CpG island together with other regulatory elements including open-sea, shelf, and shore were found to peak at the gene body and were reduced elsewhere. These remained higher in hypermethylated regions than hypomethylated regions for both LV suggesting LVs general hypermethylation. We then focused on the CpG islands in hyper or hypo-methylated promoter regions (TSS1500 or TSS200) and found a marked increase in the number of hypermethylated genes corresponding to pHV infection of iPSC (n = 210) than pHR infection (n = 24) or HLC (n = 28). Through pathway analysis, we found hypermethylated signalling molecules included ABR , JAK3 , RASA3 , DGKZ , DEF8 , PRKCG , PRKCZ , SPSB4 that are associated with intracellular signal transduction that were enriched in pHV infected iPSC. Multi-omics analysis shows cancer-related genes shared between data sets To cross compare data from differentially methylated genes (DMR) with DEG and IS targeted genes, we identified common genes shared between different data sets from iPSC infected with pHR or pHV (Fig. 10). Whilst none of the genes were identified in all three data sets, genes were shared between methylomic and genomic (n = 3), methylomic and transcriptomic (n = 15) and genomic and transciptomic (n = 166). By comparing genes specifically in methylomic and transcription and in methylomic and IS data sets we identified similar pathways for cell signalling regarding cell adhesion (e.g., MARVELD3 and TBCD for tight junction assembly; FARP2, CDH11, CTNNA2, CDON, DST, ITGAV, NLGN1, PCDH10, PCDH18, ROBO2, SEMA5A), double-strand break repair (e.g., RTEL1, SPIDR) and apoptotic processes (e.g., FAF1, OPA1, TRAF2, CADM1, HTT, ITCH, MAGI3, RABEP1, SEMA3A, STK4) . Because altered sequence, methylated state, and expression levels of tumour suppressor genes are thought to be implicated in carcinogenesis, we focused on these genes. Firstly, we performed GO term or KEGG enrichment analysis of the tumour suppressor genes in genomic data (n = 717) and found that DNA repair pathways are mainly enriched in iPSC infected with pHR compared to pHV iPSC or HLC. In pHV infected iPSC, genes characteristic of positive regulation of autophagy (early harvest) or negative regulation of cell proliferation (late harvest) were found enriched, suggestive of a protective role against carcinogenesis. On the contrary, hepatocellular carcinoma-associated tumour suppressor genes were found to be enriched in pHV infected HLC. Pathway analysis of tumour suppressor genes in transcriptomic data (n = 4283) showed a number of intracellular signalling pathways such as Foxo, PI3K-Akt, mTOR, JAK-STAT, and FCERI were upregulated in pHR and pHV infected iPSC. Only a few signalling pathways such as p53 were identified across different conditions in HLC (Figure S2 ). Discussion Insertional mutagenesis can result in genotoxicity in the host by altering the expression of genes important to cellular proliferation. This has been identified both in the clinic and non-clinically associated with retrovirus (RV) and lentivirus (LV) vectors [14,34–38]. Limited information has been gained from clinical data and much of our understanding of vector associated genotoxicity has been provided by in vitro and in vivo models. These models have, however, been viewed as oversensitive, bias and difficult to transfer between laboratories and, therefore, deemed potentially unreliable and possibly restrictive to gene therapy clinical progress. None the less, models have revealed several vector derived factors that are believed to contribute towards genotoxicity [39–43]. Recent discussion between leaders in the gene therapy field considering safety and long-term prediction of carcinogenesis in March 2023 in London has resulted in publication of a consensus document aimed at bringing the field together to find ways to circumvent this clear limitation by improving our understanding of vector associated genotoxicity and assay standardisation. As part of the consensus, the need for a human-based model was agreed. However, any such model, although partly useful to identify significant vector related side effects that would support oncogenesis or pre-malignancy, would still not provide for long-term prediction of vector safety. In this report, our approach was to investigate further, the factors suspected to support cancer development associated with LV vectors, which now superseed RV, in a fully characterised novel human-based platform. We describe here the development of h InGetox as an alternative in vitro human-based model that offers a pre-clinical tool for improving LV safety design using mechanistic outreads characterising LV contribution to genotoxicity. These factors include IS gene selection in the host genome, their effect on the expression of these genes, global effects on differential gene expression, identifying truncations between the vector and nearby host cancer genes and their novel fusion transcripts that have arisen from vector/host splicing or readthrough from the internal promoter used to drive gene expression. Finally, included in this anaylsis, epigenetic modifications in form of methylation changes in the host genome following infection were measured. Because the vector long terminal repeat (LTR) is considered a major contributor to genotoxicity due to its promoter and enhancer activities, we chose to compare LV vectors that carry either the native LTR (pHV) or self-inactivating (SIN) LTR configuration (pHR) for their genotoxic effects on host cells. Rather than use immortal cell lines with mutated cancer gene associated pathways that we considered already carry bias towards oncogenicity, we chose human induced pluripotent stem cells (iPSC) for reprogramming to 3D liver-like cell (HLCs) derivatives. These cells were chosen because they present highly proliferative and quiescent states, respectively, and the liver is considered useful to examine drug pharmacotoxicological kinetics. We previously demonstrated iPSC and their HLC derivatives express markers of pluripotency and of the liver, respectively [28], and more recently further characterised these cells transcriptomically via unsupervised clustering, principal component analysis and powerful machine learning for comparison with several normal and cancer cell types. These analyses showed HLC gene expression align closely with primary hepatocytes and are not predisposed to oncogenesis making them genuine surrogate liver cells (publication submitted) to support the development of h InGetox. Both iPSC and HLC were highly permissive to infection enabling controllable vector copy number of between 1 and 2, with cell survival and viability close to that of untreated cells. By characterising 412,786 LV integrations in regulatory regions, both LV configurations were found distributed in the order of introns > UTR regions > exons. Interestingly, because the IS number reduced by 50% between the early (d3) and late (d30) time points, with increasing prevalence of IS in cancer genes (CG), this suggested enrichment of particular IS that favour cell survival or proliferation. This was also supported by increased gene expression in genes with IS for each LV in cloned iPSC and increasing sequence count changes in IS found in cancer genes that were clonally tracked both in iPSC and differentiated HLC bulk cultures. Although we found similar numbers of cancer genes with SSC carrying pHV and pHR LV insertions in iPSC over time, SSC appeared more than 3-fold higher in pHV infected HLC than pHR suggesting IS associated outgrowth continues following differentiation. Interestingly, in these cells, IS were identified in SET , BRAF and MECOM genes that have previously been found associated with genotoxicity studies [44]. To understand further genes chosen by each LV that were potentially supporting cellular outgrowth, we investigated, in more detail the IS identified in proliferating iPSC by IS gene positioning and hallmark pathway analysis. Differences between pHR and pHV LV were identified where enriched pathways characteristic of inflammation and hypoxia, respectively, were apparent and these IS were mainly in tumour suppressor genes. Interestingly, for pHR these appeared only at the early time point, whereas IS for these enriched pathways remained also at the later 30 day time point in pHV infected iPSC. Far fewer IS in oncogenes or tumour suppressor genes were found in HLC infected by each vector, most likely because of low expression of these genes. Targeted oncogenes in these cells were identified mainly in PI3K/AKT/mTOR signalling pathways. To understand the potential mechanisms that could be supporting genotoxicity, RNASeq was used to measure global differential gene expression (DEG) in bulk infected cultures by each LV. Upon infection of iPSC, at the early time point, upregulated DEG for both LV were found associated with signalling pathways and cytokine production characteristic of the innate immune response to infection. However, pHV appeared with a higher number of oncogenes with DEG. DEG associated with pHR in oncogenes and tumour suppressor genes were implicated DNA damage, for activated tyrosine kinase receptor pathway signalling and cell senescence pathways suggesting protection against oncogenesis [45]. This contrasted to pHV where activated PI3K signalling, and MAPK signalling pathways were prominent. In addition, GO analysis of pHV DEG revealed unique pathways associated with epigenetic changes involving methylation that are known to be characteristic to oncogenesis [46]. Importantly, network analysis of potential gene interactions for both LV at early and late time points highlighted TP53 as a central transcription factor in infected iPSC, however, a major difference between the pHR and pHV at the late time point, showed once again, using GO terms, pHR infected cells associated with the p53 pathway responding to DNA damage rather than an inflammatory response in pHV infected cells. In HLC, at the early time point of infection, we found pHV associated DEG nearly 7 fold that of pHR suggesting HLC infected with this vector dramatically alter their expression profiles. This DEG increase also included upregulated oncogenes and tumour suppressor genes compared to the SIN configuration vector. GO analysis showed these genes mainly associated with tyrosine kinase signalling, protein phosphorylation and other ERK1/2 cascade and P13K/AKT/mTOR related pathways and included MECOM , LMO-2 and BRAF genes. Interestingly, similar to our finding in infected iPSC and in contrast to pHV infected HLCs, DEG in SIN pHR vector infected HLC concerned genes that characteristically protect against viral infection. Weighted correlation network analysis (WGCNA) enabled profiling of unique gene expression programmes across iPSC or HLC infected by pHV and pHR. Several significant functional modules associated with distinct biological processes with shared inserted target genes by both vectors were identified, enabling co-expressed gene association and quantification with IS in oncogenes and tumour suppressor genes. Once again correlations differed between each vector with pHR more associated with modules representing protein modification, cellular metabolism and immune response. RNASeq analysis also enabled fusion genes resulting between each LV and host loci to be identified. Correlated, co-expressed genes and IS in oncogenes and tumour suppressor genes were then aligned with these. Fusions are suspected to have arisen either due to vector host gene splicing or readthrough from the internal promoter into host genes past the vector 3’ polyadenylation sequences. Although fewer fusions appeared in iPSC than HLC, a greater number of fusions appeared associated with pHR than pHV in iPSC. Interestingly, an equal number of fusions were observed in HLC for each LV. Fusions appeared to be more prominent in differentiated cells with more than two fold in HLC than iPSC. Of the 69 fusions identified, nearly fifty percent (n = 38) were identified in common with DEG and IS representing a potentially significant contribution to genotoxicity by both LV. This suggests modification important to avoid fusions occurring with host cancer genes. Further analysis of each fusion by mapping each the vector backbone would determine whether the fusions arose as a result of gene splicing or readthrough. Following virus infection, methylation is believed to be an innate mechanism used to prevent successful virus establishment and propagation [47]. We sought to determine whether this response to infection also altered the methylation profile in the host. To do this, we profiled CpG methylation in iPSC and HLC genomes of specific genes and crossed referenced these with IS and DEG data to identify consensus cancer genes and their ontologies. Overall, pHV infected iPSC were 10 times more hypermethylated than pHR infected cells and enriched genes involved in signal transduction that are associated with cellular proliferation were found indicating the LTR has a major influence on the epigenetic response by cells following infection. This was shown also in mice where following infection by LV, where changes to methylation profiles altered the expression of cancer genes under the control of the E2F transcription factor [21]. To use the contributory factors identified relating to genotoxicity for a comparative assessment of the safety of each LV, multi-omics was applied to LV IS, DEG and differential methylomics data. Whilst none of the genes were identified in all three data sets, data from suspected genotoxic factors were found overlapping with several sharing cell signalling pathways regarding cell adhesion. Distinct pathways between transcription and methylomic data sets were found for genes involved in DNA damage response and transcription and genomic sets for genes involved in the apoptotic response, respectively. We then focused on the tumour suppressor genes identified amongst these data sets, since altered sequence, methylated state, and or expression levels of these genes are thought to be implicated in carcinogenesis. Firstly, we performed GO term or KEGG enrichment analysis of the tumour suppressor genes in genomic data and found that DNA repair pathways are mainly enriched in pHR infected iPSC compared to pHV-infected iPSC or HLC. In pHV infected iPSC, genes characteristic of positive regulation of autophagy (Day 3) or negative regulation of cell proliferation (Day 30) were found to be enriched, suggestive of a protective role against carcinogenesis. On the contrary, hepatocellular carcinoma-associated tumour suppressor genes were found to be enriched in pHV-infected HLC. Pathway analysis for tumour suppressor genes in transcriptomic data showed a number of intracellular signalling pathways were upregulated in pHR and pHV infected iPSC and only a few signalling pathways like p53 were identified across different conditions in HLC. To infer the probability of carcinogenesis can be induced by pHR or pHV LV in iPSC or HLC, cancer signatures associated with enriched pathways regardless of cancer types were used for alignment with signatures scored from transcriptomes of infected iPSC or HLC at early and later time points compared with respective controls. Our finding that infected iPSC are characterised by higher cancer scores than infected HLC is most likely due to the nature of iPSC that are rapidly proliferating compared to differentiated cells. In agreement of the insertion sites and differentially expressed genes we identified at the early time point, pHR infected iPSC or HLC had higher cancer scores than pHV infected cells, however after prolonged culture (30d), the genotoxicity of the pHV vector was more apparent with higher cancer scores than pHR infected cells. In conclusion, we have developed h InGetox as a modular series of assays to measure vector interactions with human cells that are considered contributory to oncogenesis. We found both native LTR and SIN configuration LV carry genotoxic risk being capable of altering the expression of cancer genes. In this study, SIN LTR LV appears with less genotoxic risk than LV that carries the native LTR promoter and enhancer. As murine-based models may be considered unreliable to predict oncogenesis in humans and limited data can be obtained through clinical observations, h InGetox represents a novel platform that examines vector and host genotoxic interactions in a human genetic background. Although h InGetox may be considered predictive of safety, this is still limited to events that occur early on following gene therapy and, therefore, we propose this model beneficial to identify vector related genotoxicity that may be mitigated by improved vector design. For instance, this could involve modifications to the vector backbone such as removal of unnecessary splice donor or acceptor sites or 3’ sequences to reduce promoter readthrough. Furthermore, of value would be a reduction in vector CpG sequences responsible for promoting host innate immune recognition that causes epigenetic changes in the host that can lead to cancer. Finally, we propose h InGetox useful as a pre-clinical tool to screen the safety of several complex LV intended for gene therapy such as those used to carry CAR cassettes to generate CAR-T cells. Hence, by improving LV design and h InGetox screening, safer vectors will become available for regulator approval and safe therapeutic outcome. Online methods Growth and characterisation iPSC pluripotency and differentiation to hepatocyte-like cells A human iPSC line (JHUP106i) was cultured routinely on laminin 521 (BioLamina, France) coated plates in serum-free mTeSR™1 medium (STEMCELL Technologies, Cambridge) as previously described [48]. The cell was monitored regularly for infection and was propagated in antibiotic free medium. Bulk cultures of these cells were used for differentiation and infection experiments in these studies. These iPSCs were washed with 2ml PBS without calcium chloride and magnesium chloride. The cells were incubated with 1ml of Gentle Cell Dissociation Reagent (Stemcell Technologies) for 6 minutes until the cells transformed into single cells. Single hiPSCs were collected and resuspended in FACS-PBS (PBS supplemented with 0.1% BSA and 0.1% sodium azide), counted and resuspended at 1×10 6 cells/ml for use. Tubes containing 100,000 cells were incubated for 30 minutes at 4°C with fluorochrome conjugated antibodies. Following incubation, cells were then washed once with PBS, removing any unbound antibodies and centrifuged at 1500 rpm for 5 minutes. Antibody binding to the surface of the cells was measured using the optimum concentration of an appropriate fluorochrome conjugated isotype specific antibody. In this study, unstained cells were used as a negative control. Measurement was carried out by using an electronic live gate on forward scatter and side scatter parameters. Data was acquired for 20,000–50,000 gated live events for each sample using a Novocyte flow cytometer (Agilent Technologies) equipped with a 488 nm laser and analysed using Novoexpress software. Formation of self-aggregated 3D hiPSCs spheroids Agarose microplates were generated in 256-well format using the 3D Petri Dish® mould (Sigma Aldrich, Dorset) following the manufacturer instructions. These microplates were transferred to 12 well plates (Corning, Germany) as previously described [49,50]. hiPSCs were expanded on laminin coated plates, were incubated with 1 ml of Gentle Dissociation Buffer (Stemcell Technologies) for 7–10 minute at 37°C. The single cell suspension was centrifuged at 0.2 rcf for 5 minute and resuspended in mTeSR™1 supplemented with 10 µM Y-27,632 (Calbiochem, Watford) at a density of 2.0 x 10 6 live cells/ml. The prewarmed agarose microplates were seeded by transferring 190 ul of resulted cell suspension. After 2 hours, 1 ml mTeSR™1 supplemented with 10 uM Y-27,632 was gently added to each well of 12-well plate and incubated overnight at 37 C. Hepatic induction of self-aggregated hiPSCs spheroids Differentiation to HLC was performed as previously described [48,51]. Differentiation was initiated by replacing mTeSR™1 with endoderm differentiation medium: RPMI1640 containing 1x B27 (Life Technologies), 100ng/ml Activin A (PeproTech, Hammersmith), and 50 ng/ml Wnt3a (R&D Systems, Abingdon). The medium was changed every 24 hours, for 72 hours. On day 5, endoderm differentiation was substituted with hepatoblast differentiation medium. This medium was changed every second day for a further 5 days. This medium was composed of knockout-DMEM (Life Technologies), knockout serum replacement (KOSR-Life Technologies), 0.5% Glutamax (Life Technologies), 1% non-essential amino acids (Life-Technologies), 0.2% b-mercaptoethanol (Life Technologies), and 1% DMSO (Sigma Aldrich). On day 10, hepatoblast medium was replaced with hepatocyte maturation medium HepatoZYME (Life Technologies) containing 1% Glutamax (Life-Technologies), supplemented with 10 ng/ml hepatocyte growth factor (HGF, PeproTech) and 20ng /ml oncostatin M (OSM, PeproTech) as described previously [48,51]. On day 21 of differentiation, cells were cultured in maintenance medium containing William’s E media (Life Technologies), supplemented with 10ng/ml EGF (R&D systems), 10 ng/ml VEGF (R&D Systems), 10 ng/ml HGF (PeproTech), 10ng/ml bFGF (PeproTech), 10% KOSR, 1% Glutamax, and 1% penicillin-streptomycin (Thermo Fisher Scientific) for the remining study, as previously described [49]. Histology and immunofluorescence of 3D hepatospheres 3D spheroids were fixed in ice-cold methanol for 1 hour, washed in PBS and embedded in agarose. Agarose-embedded spheroids were embedded in paraffin and 4um sections were prepared. Antigen retrieval was performed using 1 x Tris-EDTA buffer solution for 15 minutes. Paraffin-embedded sections were also stained with Eosin and Hematoxylin and mounted in Pertex before microscopy. Brightfield images were taken using a Nikon Eclipse e600 microscope equipped with a Retiga 2000R camera (Q-imaging) and Image-Pro Premier software. In order to stain sectioned hepatospheres, tissue was blocked with 10% BSA in PBS-tween (PBST) and incubated with primary antibody overnight at 4 C. Species-specific fluorescent -conjugated secondary antibody were used (Alexa Flour 488/Alexa Flour 568; Invitrogen). Sections were counterstained with DAPI (4’6-diamidino-2-phenylin-dole) and mounted with Fluoromount-G (SouthernBiotech) before microscopy. qRT-PCR RNA was extracted from 3D hepatospheres using RNAeasy Mini RNA Extraction Kit (Qiagen) according to manufacturer’s instructions. RNA quantity and quality were evaluated using Nanodrop TM 200c. Following this step, cDNA was amplified using the RT2 First Strand Kit (Qiagen) following the manufacturer’s instruction. qPCR was performed with TaqMan Fast Advance Mastermix and primer pairs, using a Roche LightCycler 480 real-time PCR system. Gene expression was normalised to housekeeping gene; glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and expressed as relative expression over 3D hepatospheres on day 0 of differentiation as control sample. qPCR was conducted in triplicate and data was analysed using Roche LightCycler 480 software. Hepatocyte phenotyping To evaluate Cyp3A activity, 50 uM of Luciferin-PFBE substrate (Promega, Southampton) was incubated with 3D hepatospheres in HepatoZYME medium supplemented with 10 ng/ml HGF. Cytochrome P450 activity was measured 24 hour later using the P450-Glo assay kit (Promega) following manufacturer’s instruction. To measure AFP and ALB secretion, the supernatant was collected after 24 hour and quantified using commercial ELISA kits (Alpha Diagnostics International, Texas). Data were normalised with the total protein content measured using bicinchonic acid (BCA) assay (Thermo Fisher Scientific). Generation of high titre LV vectors HEK293T cells were grown in DMEM GlutaMAX supplemented with 10% foetal bovine serum and 1% Penicillin Streptomycin (Fisher Scientific, Loughborough), at 37ºC with 5% CO 2 . Cells were passaged regularly upon confluency. pHR'SIN-cPPT-SEW (pHR) and it’s native LTR counterpart (pHV) LV were generated as previously described [52]. Briefly, 1.5x10 7 HEK293T cells were seeded per T175 flask and incubating at 37°C, 5% CO2 overnight. Cells were transfected with 16µg eGFP transgene, 12µg pCMVR8.74 and 4µg pMD2.G with a transfection reagent in serum free medium. Medium was replaced 24 hours post transfection and supernatant harvested every 24 hours for 72 hours post replacement. Conditioned medium was filtered through 0.45µM filters (Fisher Scientific) and stored at 4°C for future use. Conditioned medium was concentrated via ultracentrifugation at 23,000 rpm at 4°C for 2 ½ hours, using an SW32Ti rotor and Optima XPN ultracentrifuge (Beckman Coulter, High Wycombe). Viral pellet was resuspended in 200µl serum free medium and stored at -80°C for future use. LV titration Infectious LV titre was calculated by as previously reported [53]. Briefly, 2 x 10 5 HEK293T cells were seeded and incubated at 37°C, 5% CO 2 overnight to adhere. Serial dilutions of concentrated LV were prepared and incubated in complete cell culture medium with 5µg/ml polybrene (Sigma Aldrich), for 20 minutes at room temperature before addition to cells. Medium was replaced after 24 hours incubation and incubated for a further 48 hours before analysis using a Novocyte flow cytometer (ACEA Biosciences Inc, San Diego) and data analysis using NovoExpress software. Dilutions expressing 1–30% GFP expression were analysed as accurate representations of viral titre (TU/ml), calculated as below. ((Cell count x (Percentage GFP expression/100))/Volume) * Dilution factor Optimisation of iPSC and HLC gene transfer One day prior to transduction, 3 x 10 5 iPSCs were seeded in pre-coated laminin plates. The following days, one well of the cells were dissociated using Gentle Cell Dissociation reagent (StemCell Technology). For lentiviral transduction, mTeSR1 medium containing 10 µM Y-27,632 (Calbiochem) and 5µg/ml polybrene reagent (Sigma-Aldrich, UK) was prepared. The virus added to the medium and the mixture was incubated for 20 minutes at room temperature. The medium from the cells removed and replaced with the medium containing the virus and rock inhibitor. The plate was incubated at 37°C for 24 hours. Following day, the transduction medium was replaced with fresh complete mTesr1 medium. This step was continued for three days and medium refreshed daily. The cells were extracted for flow cytometry analysis to determine the number of GFP positive cells. For 3D hepatospheres, William’s E medium (Life Technology) supplemented with 10 µM Y-27,632, 10 ng/ml EGF (R&D Sytems), 10 ng/ml VEGF (R&D Systems), 10 ng/ml HGF (PeproTech), 10 ng/ml bFGF (PeproTech), 5µg/ml polybrene (Sigma-Aldrich) and the virus was prepared. The mixture was incubated for 20 minutes at room temperature before adding to the cells. The transduction medium added to the cells and incubated for 24 hours before with complete William’s E medium with essential growth factor supplements. Following transduction, transduced 3D heps were kept 3 to 7 days until fluorescent cells were appeared. Cloning iPSC Prior to single cell cloning, iPSCs were transduced with lentiviral vectors as previously described. On day of SCC complete mTeSR1 (StemCell Technology) medium with conditioned medium at a ratio of 1:1 was prepared. The medium was supplemented with 10 µM Y-27,632 (Calbiochem) to enhance the cell survival. Transduced positive GFP cells were washed with PBS (Sigma-Aldrich) once and gently dissociated using Gentle Cell Dissociation reagent (StemCell Technology) for 15 minutes. The single cells were resuspended in mTeSR1 and Y-27632 and counted by haemocytometer. To make the final 2 cells/ml, 4.8 µl of cell solution was transferred to 12 ml of complete/conditioned medium and 500 µl from the cell suspension was dispensed per well of a 24-well plate. This was to ensure the plate was seeded at a density of 1 cell/well. Following seeding, the cells were undisturbed for 7 to 10 days. After 7 days, the plate was scanned for colonies. The cells from each colony were expanded and harvested for DNA/RNA extraction. 33 (pHR LV) and 7 (pHV LV) clones were isolated from infected samples respectively Nucleic acid isolation DNA and RNA samples were isolated from transduced samples according using DNeasy Blood & Tissue Kit and RNeasy Mini Kit (Qiagen, Manchester) respectively, according to manufacturer’s instructions. RNA was treated with DNAase I to remove contaminants, according to manufactures instructions (Qiagen). Nucleic acid concentration and purity was analysed using NanoDrop™ 2000c spectrophotometer (ThermoFisher Scientific, Hemel Hempstead). Nucleic acid integrity analysis DNA samples were initially analysed for vector presence. Therefore, a vector specific primer pair was designed and used in a standard PCR with 10ng DNA. Products were applied to 2% gel electrophoresis to visualize PCR amplicons and the detection of expected bands. RNAseq sample preparation for fusion transcript and expression analysis RNA quality was assessed on TapeStation 2200 system using TapeStation RNA ScreenTape & Reagents (Agilent, Santa Clara). Up to 1000 ng total RNA per sample were applied to SureSelect Strand-Specific mRNA Library Preparation for Illumina (Agilent) and TruSeq Stranded mRNA Library Prep Kit (Illumina) according to manufacturer’s instructions for library preparation. Libraries were sequenced in 150PE mode on Illumina HiSeq System. Sequencing data was analysed using GENE-IS (Afzal et al., 2017) for the detection of fusion transcripts and DEseq2 for the assessment of differentially expressed genes. Vector copy number analysis Vector copy number in the samples were determined by quantitative TaqMan™ Universal PCR Master Mix. Briefly, 10 ng of DNA were applied in triplicate analysis on CFX96 Touch Real-Time PCR Detection System (Bio-Rad, Hercules). A standard curve of 8 calibration standards (10 7 – 5 copies) was generated for absolute quantification of vector copies in the samples. LV specific primers and probes were designed and ordered from Sigma-Aldrich (Munich, Germany) and IDT Technologies (Coralville, USA), respectively. Primers were used at a final concentration of 720nM, probes at 140nM in a total reaction volume of 20µl. Analysis of vector integration sites Viral vector integration sites analysis was performed using Sharing-Extension Primer Tag Selection / Linear-Mediated Polymerase Chain Reaction (S-EPTS/LM-PCR), which is a shearing DNA based integration site analysis method, followed by next-generation sequencing. Raw sequencing data were trimmed based on quality (Phred) and filtered for containing both molecular barcodes at full identity. Remaining reads were further analysed using GENE-IS. Briefly, sequences were further trimmed and only sequences containing the expected vector-specific stretch were considered for following steps. First, sequences were aligned to the human genome (UCSC assembly release number hg38) by Burrows-Wheeler Aligner (BWA) MEM algorithm. Potential integration sites were then mapped with BLAST at a minimum alignment identity percentage of 95%. Adjacent genes and other features were annotated according to RefSeq database. For each detected integration site, the relative sequence count compared to all sequences was calculated. Analysis of common integration sites Biologically relevant IS clusters, called common integration sites (CIS), were analysed using a graphs based approach. Any IS detected was considered as node that contained the IS locus. If the distance of two nodes was less than 50kb, the nodes were connected and resulting nodes sets considered as CIS. Integration sites in proximity to cancer-related genes A list of over 700 well-defined cancer genes was compiled from the Cancer Gene Census database ( https://cancer.sanger.ac.uk/census ). Cancer gene data was obtained from Ensembl human genes ( http://www.ensembl.org/biomart/martview/ ; version GRCh38.p10). Relative frequencies of integration sites, that were detected within a 100kb window of a TSS of a cancer-related gene, were analysed. Methylome analysis using the Illumina Epic Kit Up to 250ng DNA per sample were applied to Infinium Methylation EPIC Kit (Illumina, San Diego) according to manufacturer’s instructions and arrays were scanned on Illumina’s iScan System. Generated data was analysed using ‘The Chip Analysis Methylation Pipeline’ in order to determine differentially methylated regions. Bioinformatic analysis pipeline We established an analytical pipeline in R (v.4.0.2). Raw data was curated into feature-sample matrices and streamlined into ExpressionSet objects using BioBase (v.2.50.0). The subsequent analysis mainly includes differential analysis, gene set enrichment analysis, signature score assessment, and weighted gene co-expression network analysis. Differential Analysis : We used limma (v3.46.0) for differential analysis on count matrices. First, we normalised and conducted log2-transformation on count matrices. Next, we constructed design matrices using phenotypic data and fitted these to the processed matrices. Following empirical Bayes moderation, we identified differentially expressed genes (DEGs) and retained those with an absolute log2 fold-change greater than 1 and BH-adjusted p-values less than 0.01. Lastly, we visualised these DEGs using EnhancedVolcano (v.1.8.0). Gene Set Enrichment Analysis (GSEA) : We used clusterProfiler (v3.18.1) for GSEA. First, we ranked gene expression levels from highest to lowest by comparing transcriptomic samples from study groups against controls. Next, we used Hallmark gene sets or GO terms from MSigDB (v.7.5.1) to perform GSEA, resulting in BH-adjusted p values, ratio of genes from each set, and other attributes. Lastly, we used treeplot or cnetplot to visualise the results. Signature Score Analysis : We defined molecular signatures for specific cancers using TCGA RNA-seq datasets. Differential analysis criteria for these signature genes were set at log2 fold-change > 2 and BH-adjusted p-values < 0.01. We then determined the scaled mean expression levels of these genes and visualised them using pheatmap (v1.0.12). Weighted Gene Co-expression Network Analysis (WGCNA) : We used WGCNA (v.1.70-3) for the analysis. We began with normalised and transposed count matrices and performed network topology analysis, resulting in an optimal soft threshold power. We then constructed topological overlap matrix using the blockwiseModules command and determine traits associated with each module by calculating hypothetical central genes. Lastly, we identified potential key drivers in selected modules by using the intramodularConnectivity command. Declarations Acknowledgments: Experimental procedures: SS, SA, MZ, SF, MSK. Data generation: SS, SA, MZ, SF, MSK, SP. Data interpretation: SS, SA, AG, AZ, AP, MZ, RF, WW, OSF, AD, YT, SNW, IGF, SP, MS, DH, MT. Data analysis: SS, SA, AG, AZ, AP, MZ, SF, DH, MF, RF, WW, OSF, AD, YT, SNW, IGF, SP, MS, MT. Paper preparation: SS, AG, AZ, AP, MT. Paper editing: SS, AG, AZ, MT. Supervision: DH, WW, OSF, AD, YT, SNW, IGF, MS, MT. Conceptualisation: DH, MS, MT. Methodology: DH, MS, YT, SNW, MS, MT. Paper review: SJ, AG, MT. Paper editing: SS, MT Conflict of interest statement: DH is founder, director and shareholder in Stemnovate Limited and Stimuliver ApS. All other authors declare no conflict of interest. Funding statement: This work was funded by an NC3Rs CRACK IT Challenge 21: InMutagene award, sponsored by GSK and Novartis. References Zufferey R, Dull T, Mandel RJ, Bukovsky A, Quiroz D, Naldini L et al. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery 1998;72:9873–80. Zhang S, Chen S, Li W, Guo X, Zhao P, Xu J et al. Rescue of ATP7B function in hepatocyte-like cells from Wilson’s disease induced pluripotent stem cells using gene therapy or the chaperone drug curcumin 2011;20:3176–87. https://doi.org/10.1093/hmg/ddr223 . Jahner D, Jaenisch R. Retrovirus-induced de novo methylation of flanking host sequences correlates with gene inactivity 1985;315:594–7. https://doi.org/10.1038/315594a0 . Saez-Cirion A, Manel N. Immune Responses to Retroviruses. 2018;36:193–220. https://doi.org/10.1146/annurev-immunol-051116-052155 . Lucendo-Villarin B, Rashidi H, Alhaque S, Fischer L, Meseguer-Ripolles J, Wang Y et al. Serum Free Production of Three-dimensional Human Hepatospheres from. Pluripotent Stem Cells 2019: 10.3791/59965 . https://doi.org/10.3791/59965. Waddington SN, Nivsarkar MS, Mistry AR, Buckley SM, Kemball-Cook G, Mosley KL et al. Permanent phenotypic correction of hemophilia B in immunocompetent mice by prenatal gene therapy 2004;104:2714–21. https://doi.org/10.1182/blood-2004-02-0627 . Almarza D, Bussadori G, Navarro M, Mavilio F, Larcher F, Murillas R. Risk assessment in skin gene therapy: viral-cellular fusion transcripts generated by proviral transcriptional read-through in keratinocytes transduced with self-inactivating lentiviral vectors 2011;18:674–81. https://doi.org/10.1038/gt.2011.12 . Cavazzana-Calvo M, Payen E, Negre O, Wang G, Hehir K, Fusil F et al. Transfusion independence and HMGA2 activation after gene therapy of human beta-thalassaemia 2010;467:318–22. https://doi.org/10.1038/nature09328 . Schroder AR, Shinn P, Chen H, Berry C, Ecker JR, Bushman F. HIV-1 integration in the human genome favors active genes and local hotspots 2002;110:521–9. https://doi.org/S0092867402008644 [pii]. Wu X, Li Y, Crise B, Burgess SM. Transcription start regions in the human genome are favored targets for. MLV Integr. 2003;300:1749–51. https://doi.org/10.1126/science.1083413 . Esteve J, Blouin JM, Lalanne M, Azzi-Martin L, Dubus P, Bidet A et al. Generation of induced pluripotent stem cells-derived hepatocyte-like cells for ex vivo gene therapy of primary hyperoxaluria type 1 2019;38:101467. https://doi.org/S1873-5061(19)30097-2 [pii]. Howe SJ, Mansour MR, Schwarzwaelder K, Bartholomae C, Hubank M, Kempski H et al. Insertional mutagenesis combined with acquired somatic mutations causes leukemogenesis following gene therapy of SCID-X1 patients 2008;118:3143–50. https://doi.org/10.1172/JCI35798 . Gonin P, Buchholz CJ, Pallardy M, Mezzina M. Gene therapy bio-safety: scientific and regulatory issues 2005:S146-52. https://doi.org/3302629 [pii]. Schmidt M, Hoffmann G, Wissler M, Lemke N, Mussig A, Glimm H et al. Detection and direct genomic sequencing of multiple rare unknown flanking DNA in highly complex samples 2001;12:743–9. https://doi.org/10.1089/104303401750148649 . Suleman S, Payne A, Bowden J, Al Haque S, Zahn M, Fawaz S et al. HIV- 1 lentivirus tethering to the genome is associated with transcription factor binding sites found in genes that favour virus survival 2022. Wills CA, Drago D, Pietrusko RG. Clinical holds for cell and gene therapy trials: Risks, impact, and lessons learned 2023;31:101125. https://doi.org/10.1016/j.omtm.2023.101125 . Stein S, Ott MG, Schultze-Strasser S, Jauch A, Burwinkel B, Kinner A et al. Genomic instability and myelodysplasia with monosomy 7 consequent to EVI1 activation after gene therapy for chronic granulomatous disease 2010;16:198–204. https://doi.org/10.1038/nm.2088 . Deichmann A, Hacein-Bey-Abina S, Schmidt M, Garrigue A, Brugman MH, Hu J et al. Vector integration is nonrandom and clustered and influences the fate of lymphopoiesis in SCID-X1 gene therapy 2007;117:2225–32. https://doi.org/10.1172/JCI31659 . Corcoran LM, Adams JM, Dunn AR, Cory S. Murine T lymphomas in which the cellular myc oncogene has been activated by retroviral insertion 1984;37:113–22. https://doi.org/10.1016/0092-8674(84)90306-4 . Shalginskikh N, Poleshko A, Skalka AM, Katz RA. Retroviral DNA methylation and epigenetic repression are mediated by the. antiviral host protein Daxx. 2013;87:2137–50. https://doi.org/10.1128/JVI.02026-12 . Wang Y, Alhaque S, Cameron K, Meseguer-Ripolles J, Lucendo-Villarin B, Rashidi H et al. Defined and Scalable Generation of Hepatocyte-like Cells from Human. Pluripotent Stem Cells 2017: 10.3791/55355 . https://doi.org/10.3791/55355. Lucendo-Villarin B, Rashidi H, Alhaque S, Fischer L, Meseguer-Ripolles J, Wang Y et al. Serum Free Production of Three-dimensional Human Hepatospheres from. Pluripotent Stem Cells 2019: 10.3791/59965 . https://doi.org/10.3791/59965. Bushman FD. Retroviral Insertional Mutagenesis in Humans: Evidence for Four Genetic Mechanisms Promoting Expansion of. Cell Clones. 2020;28:352–6. https://doi.org/10.1016/j.ymthe.2019.12.009 . Li Z, Modlich U, Anjali M. Leukemia diagnosis in murine bone marrow transplantation models 2009;506:311–29. https://doi.org/10.1007/978-1-59745-409-4_21 . Cesana D, Sgualdino J, Rudilosso L, Merella S, Naldini L, Montini E. Whole transcriptome characterization of aberrant splicing events induced by lentiviral vector integrations 2012;122:1667–76. https://doi.org/10.1172/JCI62189 . Hacein-Bey-Abina S, Hauer J, Lim A, Picard C, Wang GP, Berry CC et al. Efficacy of gene therapy for X-linked severe combined immunodeficiency 2010;363:355–64. https://doi.org/10.1056/NEJMoa1000164 . Cavazzana M, Bushman FD, Miccio A, Andre-Schmutz I, Six E. Gene therapy targeting haematopoietic stem cells for inherited diseases: progress and challenges 2019;18:447–62. https://doi.org/10.1038/s41573-019-0020-9 . Gay V, Moreau K, Hong SS, Ronfort C. Quantification of HIV-based lentiviral vectors: influence of several cell type parameters on vector infectivity 2012;157:217–23. https://doi.org/10.1007/s00705-011-1150-5 . Schwarzer A, Talbot SR, Selich A, Morgan M, Schott JW, Dittrich-Breiholz O et al. Predicting genotoxicity of viral vectors for stem cell gene therapy using gene expression-based machine learning 2021;29:3383–97. https://doi.org/10.1016/j.ymthe.2021.06.017 . Baum C, Kustikova O, Modlich U, Li Z, Fehse B. Mutagenesis and oncogenesis by chromosomal insertion of gene transfer vectors 2006;17:253–63. https://doi.org/10.1089/hum.2006.17.253 . Schambach A, Galla M, Maetzig T, Loew R, Baum C. Improving transcriptional termination of self-inactivating gamma-retroviral and lentiviral vectors. 2007;15:1167–73. https://doi.org/S1525-0016(16)31676-8 [pii]. Nobles CL, Sherrill-Mix S, Everett JK, Reddy S, Fraietta JA, Porter DL et al. CD19-targeting CAR T cell immunotherapy outcomes correlate with genomic modification by vector integration 2020;130:673–85. https://doi.org/10.1172/JCI130144 . Wu C, Dunbar CE. Stem cell gene therapy: the risks of insertional mutagenesis and approaches to minimize genotoxicity 2011;5:356–71. https://doi.org/10.1007/s11684-011-0159-1 . Koyama S, Ishii KJ, Coban C, Akira S. Innate immune response to viral infection 2008;43:336–41. https://doi.org/10.1016/j.cyto.2008.07.009 . Cameron K, Tan R, Schmidt-Heck W, Campos G, Lyall MJ, Wang Y et al. Recombinant Laminins Drive the Differentiation and Self-Organization of hESC-Derived Hepatocytes. 2015;5:1250–62. https://doi.org/S2213-6711(15)00313-6 [pii]. Rashidi H, Luu NT, Alwahsh SM, Ginai M, Alhaque S, Dong H et al. 3D human liver tissue from pluripotent stem cells displays stable phenotype in vitro and supports compromised liver function in vivo 2018;92:3117–29. https://doi.org/10.1007/s00204-018-2280-2 . Therapy AS of, G, Gene C. Cell & RNA Therapy Landscape: Q1 Quarterly Data Report. 2023. Fraietta JA, Nobles CL, Sammons MA, Lundh S, Carty SA, Reich TJ et al. Disruption of TET2 promotes the therapeutic efficacy of CD19-targeted T cells 2018;558:307–12. https://doi.org/10.1038/s41586-018-0178-z . Touw IP, Erkeland SJ. Retroviral insertion mutagenesis in mice as a comparative oncogenomics tool to identify disease genes in human leukemia. 2007;15:13–9. https://doi.org/S1525-0016(16)31245-X [pii]. Peng Y, Laser J, Shi G, Mittal K, Melamed J, Lee P et al. Antiproliferative effects by Let-7 repression of high-mobility group A2 in uterine leiomyoma 2008;6:663–73. https://doi.org/10.1158/1541-7786.MCR-07-0370 . Doerfler W. Impact of foreign DNA integration on tumor biology and on evolution via epigenetic alterations 2012;4:41–9. https://doi.org/10.2217/epi.11.111 . CJ.Boztug CB, Paruzynski K, Witzel A, Schwarzer M, Rothe A, Modlich M, Beier U, Göhring R, Steinemann G, Fronza D, Ball R, Haemmerle CR, Naundorf R, Rose SKühlckeK, M.Fraser C, Mathias L, Ferrari RA, Al-Herz MR, W, Kalle. .Kondratenko, I.Maródi, L.Glimm, H.Schlegelberger, B.Schambach, A.Albert, MH.Schmidt, M.von, C.Klein,. Gene therapy for Wiskott-Aldrich syndrome–long-term efficacy and genotoxicity 2014;12. Siler U, Paruzynski A, Holtgreve-Grez H, Kuzmenko E, Koehl U, Renner ED, et al. Successful Combination of Sequential Gene Therapy and Rescue Allo-HSCT in Two Children with X-CGD -. Importance of Timing. 2015;15:416–27. https://doi.org/10.2174/1566523215666150515145255 . Weber EL, Cannon PM. Promoter choice for retroviral vectors: transcriptional strength versus trans-activation potential 2007;18:849–60. https://doi.org/10.1089/hum.2007.067 . Montini E, Cesana D, Schmidt M, Sanvito F, Bartholomae CC, Ranzani M et al. The genotoxic potential of retroviral vectors is strongly modulated by vector design and integration site selection in a mouse model of HSC gene therapy. 2009;119:964–75. Alhaque S, Themis M, Rashidi H. Three-dimensional cell culture: from evolution to revolution 2018;373: 10.1098/rstb.2017.0216 . https://doi.org/20170216 [pii]. Montini E, Cesana D, Schmidt M, Sanvito F, Ponzoni M, Bartholomae C et al. Hematopoietic stem cell gene transfer in a tumor-prone mouse model uncovers low genotoxicity of lentiviral vector integration 2006;24:687–96. Biffi A, Montini E, Lorioli L, Cesani M, Fumagalli F, Plati T, et al. Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy. 2013;341:1233158. https://doi.org/10.1126/science.1233158 . Aiuti A, Cossu G, de Felipe P, Galli MC, Narayanan G, Renner M et al. The committee for advanced therapies’ of the European Medicines Agency reflection paper on management of clinical risks deriving from insertional mutagenesis 2013;24:47–54. https://doi.org/10.1089/humc.2013.119 . Condiotti R, Goldenberg D, Giladi H, Schnitzer-Perlman T, Waddington SN, Buckley SM et al. Transduction of fetal mice with a feline lentiviral vector induces liver tumors which exhibit an E2F activation signature 2014;22:59–68. https://doi.org/10.1038/mt.2013.193 . Hacein-Bey-Abina S, Garrigue A, Wang GP, Soulier J, Lim A, Morillon E et al. Insertional oncogenesis in 4 patients after retrovirus-mediated gene therapy of SCID-X1 2008;118:3132–42. https://doi.org/10.1172/JCI35700 . Ramaswamy S, Tonnu N, Menon T, Lewis BM, Green KT, Wampler D et al. Autologous and Heterologous Cell Therapy for Hemophilia B toward Functional Restoration of Factor IX. 2018;23:1565–80. https://doi.org/S2211-1247(18)30497-2 [pii]. Cesana D, Ranzani M, Volpin M, Bartholomae C, Duros C, Artus A, et al. Uncovering and dissecting the genotoxicity of self-inactivating lentiviral vectors in vivo 2014;22:774–85.https://doi.org/10.1038/mt.2014.3. Additional Declarations Yes there is potential conflict of interest. DH is founder, director and shareholder in Stemnovate Limited and Stimuliver ApS. All other authors declare no conflict of interest. Supplementary Files FigureS1.jpg Figure S1 A-D. Evaluation of LV insertion sites (IS) in iPSC and HLC FigureS2.jpg Figure S2. Pathway analysis showing enriched GO terms or KEGG across different comparisons. TableS1.xlsx Table S1. LV insertions in iPSC over time and in differentiated HLC. TableS2.xlsx Table S2. Relative gene expression of iPSC clones transduced with LV. TableS3.xlsx Table S3. Co-expression modules reveal key gene subsets with diverse functional implications across infected iPSCs and HLCs. Cite Share Download PDF Status: Published Journal Publication published 15 Jul, 2025 Read the published version in Gene Therapy → Version 1 posted Editorial decision: revise 13 Feb, 2024 Review # 1 received at journal 11 Feb, 2024 Review # 2 received at journal 15 Jan, 2024 Review # 3 received at journal 10 Jan, 2024 Reviewer # 3 agreed at journal 10 Jan, 2024 Reviewer # 2 agreed at journal 10 Jan, 2024 Reviewer # 1 agreed at journal 10 Jan, 2024 Reviewers invited by journal 10 Jan, 2024 Editor assigned by journal 09 Jan, 2024 Submission checks completed at journal 09 Jan, 2024 First submitted to journal 05 Jan, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3837253","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":266220587,"identity":"94908e70-dbd5-4980-90fb-31a2e6a6ef9d","order_by":0,"name":"Mike Themis","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABFElEQVRIiWNgGAWjYDACZgY2IGnBwCABpBIYbBLAog/AJBs+LRIMPBAtaRAtCfi0MCBrYWA4TFiLbjvzswc/KiQY7KWbj2548Od8nvmM3IMPEhjs5BkkoFaiAbPDbOaGPWeAtsgcS7uR2Ha7WOZGXrJBAkOyYYNE2gHsWnjYJHjbQA7LMbuR2HA7cQaQIZHAwJzAIJHegEuL5N9/IC35324k/DkH0mL+I4GhHq8Wad4GsC1sNxLYDoBtAXofGA44HcZmJi1zTIKH50Ya0GFtycUSPG+MJRIMjhu28TzD7v3zh59JvqmxkWOfkfzs5o8/dnkS7DmGHz5UVMvzs6cZYNMCAzxofAM8ETkKRsEoGAWjgCAAAG1SV3+hHZpdAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-6719-4271","institution":"Brunel University London, Gene Therapy and Genotoxicity Research Group","correspondingAuthor":true,"prefix":"","firstName":"Mike","middleName":"","lastName":"Themis","suffix":""},{"id":266220588,"identity":"5f24762a-7308-46fa-9acd-1499e0b47f04","order_by":1,"name":"Saqlain Suleman","email":"","orcid":"","institution":"Brunel University London","correspondingAuthor":false,"prefix":"","firstName":"Saqlain","middleName":"","lastName":"Suleman","suffix":""},{"id":266220589,"identity":"4a84caac-8c2e-4cd5-b7cd-1434bb3a8cf7","order_by":2,"name":"Sharmin Al Haque","email":"","orcid":"","institution":"Brunel University London","correspondingAuthor":false,"prefix":"","firstName":"Sharmin","middleName":"Al","lastName":"Haque","suffix":""},{"id":266220590,"identity":"3fbb7e70-3fdb-4504-a152-d8961563c61a","order_by":3,"name":"Andrew Guo","email":"","orcid":"","institution":"Biomavericks","correspondingAuthor":false,"prefix":"","firstName":"Andrew","middleName":"","lastName":"Guo","suffix":""},{"id":266220591,"identity":"106463bc-3f94-42ec-8b79-95b72db8cffd","order_by":4,"name":"Huairen Zhang","email":"","orcid":"","institution":"Biomavericks","correspondingAuthor":false,"prefix":"","firstName":"Huairen","middleName":"","lastName":"Zhang","suffix":""},{"id":266220592,"identity":"c57fb112-650d-4bc5-900b-9e4821a483b6","order_by":5,"name":"Annette Payne","email":"","orcid":"https://orcid.org/0000-0002-2009-0141","institution":"Brunel University London","correspondingAuthor":false,"prefix":"","firstName":"Annette","middleName":"","lastName":"Payne","suffix":""},{"id":266220593,"identity":"27375275-b3b8-48d0-b9cd-7c7d1f124082","order_by":6,"name":"Marco Zhang","email":"","orcid":"","institution":"Protagene","correspondingAuthor":false,"prefix":"","firstName":"Marco","middleName":"","lastName":"Zhang","suffix":""},{"id":266220594,"identity":"3c5e928c-02da-4bcb-a756-b688ddbd3ed3","order_by":7,"name":"Serena Fawaz","email":"","orcid":"","institution":"Brunel University London","correspondingAuthor":false,"prefix":"","firstName":"Serena","middleName":"","lastName":"Fawaz","suffix":""},{"id":266220595,"identity":"c21838ac-5519-4ef8-aad3-f92f47a71bb4","order_by":8,"name":"Mohammad Khalifa","email":"","orcid":"","institution":"Brunel University London","correspondingAuthor":false,"prefix":"","firstName":"Mohammad","middleName":"","lastName":"Khalifa","suffix":""},{"id":266220596,"identity":"8a6c0a5a-37c8-478b-92a1-36e7b0777a22","order_by":9,"name":"Susan Jobling","email":"","orcid":"","institution":"Brunel University London","correspondingAuthor":false,"prefix":"","firstName":"Susan","middleName":"","lastName":"Jobling","suffix":""},{"id":266220597,"identity":"a22f2c3e-4116-48ec-92e9-485c62a9a1a1","order_by":10,"name":"David Hay","email":"","orcid":"","institution":"University of Edinburgh","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Hay","suffix":""},{"id":266220598,"identity":"41bdd702-ee45-4895-b438-3d8837417262","order_by":11,"name":"Matteo Franco","email":"","orcid":"https://orcid.org/0000-0002-1267-4083","institution":"ProtaGene US Inc","correspondingAuthor":false,"prefix":"","firstName":"Matteo","middleName":"","lastName":"Franco","suffix":""},{"id":266220599,"identity":"d4f66982-f008-43ca-a64f-cec029b99343","order_by":12,"name":"Raffaele Fronza","email":"","orcid":"","institution":"Protagene","correspondingAuthor":false,"prefix":"","firstName":"Raffaele","middleName":"","lastName":"Fronza","suffix":""},{"id":266220600,"identity":"ab546142-0742-4076-b5e8-fa044b3778a7","order_by":13,"name":"Wei Wang","email":"","orcid":"","institution":"Protagene","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Wang","suffix":""},{"id":266220601,"identity":"5ef96d5a-5a43-47b0-b862-4e15857cf598","order_by":14,"name":"Olga Strobel-Freidekind","email":"","orcid":"","institution":"Protagene","correspondingAuthor":false,"prefix":"","firstName":"Olga","middleName":"","lastName":"Strobel-Freidekind","suffix":""},{"id":266220602,"identity":"b45db14d-1f99-43ef-b724-c405f5d61c86","order_by":15,"name":"Annette Deichmann","email":"","orcid":"","institution":"Protagene","correspondingAuthor":false,"prefix":"","firstName":"Annette","middleName":"","lastName":"Deichmann","suffix":""},{"id":266220603,"identity":"bcc64b09-c399-495f-bf72-1f7d5a7d664e","order_by":16,"name":"Yasuhiro Takeuchi","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yasuhiro","middleName":"","lastName":"Takeuchi","suffix":""},{"id":266220604,"identity":"9e507fe3-42c3-4afc-b92f-ac53dd993ef4","order_by":17,"name":"Irene Gil-Farina","email":"","orcid":"","institution":"Protagene","correspondingAuthor":false,"prefix":"","firstName":"Irene","middleName":"","lastName":"Gil-Farina","suffix":""},{"id":266220605,"identity":"6365c8fa-07b0-49e6-8621-cff1e02ab7d5","order_by":18,"name":"Jan Klapwijk","email":"","orcid":"","institution":"Cornelis Consulting Ltd.","correspondingAuthor":false,"prefix":"","firstName":"Jan","middleName":"","lastName":"Klapwijk","suffix":""},{"id":266220606,"identity":"c3302cc2-3ae2-4f06-9c42-707ecc452bed","order_by":19,"name":"Stefany Perera","email":"","orcid":"","institution":"Brunel University London","correspondingAuthor":false,"prefix":"","firstName":"Stefany","middleName":"","lastName":"Perera","suffix":""},{"id":266220607,"identity":"ca453e75-189b-496a-831e-4aac59070ab5","order_by":20,"name":"Manfred Schmidt","email":"","orcid":"","institution":"Protagene","correspondingAuthor":false,"prefix":"","firstName":"Manfred","middleName":"","lastName":"Schmidt","suffix":""}],"badges":[],"createdAt":"2024-01-05 12:42:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3837253/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3837253/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41434-025-00550-9","type":"published","date":"2025-07-15T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":49479850,"identity":"8060a180-a60c-4d58-a2fe-8cb9297cd8ba","added_by":"auto","created_at":"2024-01-11 14:39:35","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":408668,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/4d4d493ceedf4bbe295d6842.jpg"},{"id":49480198,"identity":"1fd4d233-b624-42a4-a851-47984c63fd9f","added_by":"auto","created_at":"2024-01-11 14:47:35","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":404947,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/74d9fe55eec27fde6ff4801e.jpg"},{"id":49479613,"identity":"d8b425c2-3288-456a-a27b-f397f1232ed8","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":475391,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/657bdbf4545624bd52230fd9.jpg"},{"id":49479617,"identity":"a7ef28be-bb3a-454e-b980-e1abfc58251c","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":554036,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure4X.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/754a6b3880e852ffed22829a.jpg"},{"id":49479851,"identity":"1626516f-2677-44b2-b3ab-c1aa07f195f5","added_by":"auto","created_at":"2024-01-11 14:39:35","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":212470,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/ea81babf8af69eb5b017690f.jpg"},{"id":49479619,"identity":"2a1a4932-cbe3-49f8-a6fa-439ccaf547e1","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":474502,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/3ddbe235eb03a68e9eed6705.jpg"},{"id":49479618,"identity":"4b45b37b-a049-415e-9cd0-7f343ef7ac11","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":500495,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/84a121bdedbeeace43010d29.jpg"},{"id":49479621,"identity":"ae242739-c5e8-4a31-8d7d-1a45ff57b2d7","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":520055,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/30fe009ca5e00bce21f9c6bb.jpg"},{"id":49479854,"identity":"2800d5cb-fe52-448e-a0d5-812447e23fdd","added_by":"auto","created_at":"2024-01-11 14:39:35","extension":"jpg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":458604,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/59d437f4953fa61e1878bb68.jpg"},{"id":49479622,"identity":"dbf02cc9-b381-45de-a8ef-a3521ad59fc6","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"jpg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":377323,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure10.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/35f9ba9b4c818bd86c231c58.jpg"},{"id":86835950,"identity":"57485d20-2def-456b-a398-c0ae7e2ef9b4","added_by":"auto","created_at":"2025-07-16 07:10:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5848455,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/9585dc3b-3609-4a75-a596-2c5a2b31694c.pdf"},{"id":49479853,"identity":"ba057b7d-3793-4783-b18b-27a8ba347469","added_by":"auto","created_at":"2024-01-11 14:39:35","extension":"jpg","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":350094,"visible":true,"origin":"","legend":"\u003cp\u003eFigure S1 A-D. Evaluation of LV insertion sites (IS) in iPSC and HLC\u003c/p\u003e","description":"","filename":"FigureS1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/873d88a30cd4a773fd61fae8.jpg"},{"id":49479609,"identity":"fda495ac-3b6b-4b51-a2fc-d4ab5ac42ff7","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":417784,"visible":true,"origin":"","legend":"\u003cp\u003eFigure S2. Pathway analysis showing enriched GO terms or KEGG across different comparisons.\u003c/p\u003e","description":"","filename":"FigureS2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/184dc709d2803a2c2cddf5e2.jpg"},{"id":49479608,"identity":"9814b944-3ffc-4de3-b00f-1b3fd5b2e93a","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":13419,"visible":true,"origin":"","legend":"\u003cp\u003eTable S1. LV insertions in iPSC over time and in differentiated HLC.\u003c/p\u003e","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/88c064b979b6e86c6c3ac75f.xlsx"},{"id":49479610,"identity":"3c13879e-684b-4e00-afbb-54411a3b9c55","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":13011,"visible":true,"origin":"","legend":"\u003cp\u003eTable S2. Relative gene expression of iPSC clones transduced with LV.\u003c/p\u003e","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/460c44e33215326a6cd5b1b2.xlsx"},{"id":49479611,"identity":"fe7a9428-effa-4658-bc03-bac40ec6d21f","added_by":"auto","created_at":"2024-01-11 14:31:35","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":12886,"visible":true,"origin":"","legend":"\u003cp\u003eTable S3. Co-expression modules reveal key gene subsets with diverse functional implications across infected iPSCs and HLCs.\u003c/p\u003e","description":"","filename":"TableS3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3837253/v1/ee5d68830d05e6b7f95fd63a.xlsx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential conflict of interest.\nDH is founder, director and shareholder in Stemnovate Limited and Stimuliver ApS. All other authors declare no conflict of interest.","formattedTitle":"\u003cp\u003e\u003csup\u003eh\u003c/sup\u003eInGeTox: A human-based in vitro platform to evaluate lentivirus contribution to genotoxicity\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCell and gene therapy clinical trials have increased over the past 10 years from 1,800 to 5000 listed in the National Institutes of Health that include new technologies involving CAR-T and gene edited cells. Currently, 40% of all trials are industry sponsored. In the US, each investigational new drug application (IND) requires FDA review, which primarily is based on safety considerations and may result in clinical hold ranging from 2 to 19 months. Recently, INDs have increased enormously with gene therapy product development and diversity and with commercial interest [1]. Between Jan 2020 and Dec 2022, 33 clinical holds were announced that concerned CAR-T therapies (15%) and lentivirus vector-based therapies (45%) [2].\u003c/p\u003e \u003cp\u003eSeveral gene therapy trials have successfully used gamma retrovirus (γ-RV) and lentivirus (LV) vectors for therapeutic gene delivery that offer permanent gene delivery to the host genome following virus integration. Integration, however, risks insertional mutagenesis and differences in γ-RV and LV integration site (IS) selection is known to influence their genotoxic potential [3]. Genotoxic effects by γ-RV includes host protooncogene expression changes driven by the LTR enhancer, as identified in X-SCID, WAS and CGD trials where oncogene upregulation was caused by γ-RV integration in either orientation to the gene locus [4\u0026ndash;7]. Promoter activation involving RV in the same orientation to gene transcription is also known to cause gene upregulation [8]. As a result of this, modification of the LTR to SIN configuration in LV to abrogate promoter activity has proven to be a major improvement limiting host gene activation upon integration [9]. However, readthrough from the internal promoter used to replace the modified LTR has been reported to drive local host gene expression [8]. With SIN configuration, integration preference in the gene transcription unit and the developments of 3rd generation design to avoid the emergence of replication competence, LV vectors are considered much safer than RV. Hence, LV have become the vectors of choice and are currently used to treat a number of rare genetic diseases and for CAR-T cell immunotherapy. Unfortunately, genotoxicity concerns still remain and with the appropriate configuration LV has been shown capable of oncogenesis in tumour prone mice [10,11]. LV splicing with host cancer genes has also been shown to generate novel gene fusions with the potential to drive clonal expansion[12,13] as identified in a β\u0026minus;thalassemia clinical trial, where integration of a SIN LV caused 3\u0026rsquo; end substitution of the \u003cem\u003eHMGA2\u003c/em\u003e gene, which abolished let7 microRNA control of this protooncogene, reduced \u003cem\u003eHMGA2\u003c/em\u003e degradation and clonal proliferation [14\u0026ndash;16].\u003c/p\u003e \u003cp\u003eThe role of epigenetics in cancer progression is also well known [17\u0026ndash;20], and LV infection has also been found associated with hepatocellular carcinoma development in mice, that was suspected to be caused by methylation changes of protooncogene promoters under the control of the E2F transcription factor [21].\u003c/p\u003e \u003cp\u003eMore recently, LV have been used successfully to generate CAR-T cells carrying anti-CD19 CAR cassettes for cancer immunotherapy. However, in a clinical trial against chronic lymphocytic leukaemia, CAR-T cells were found to persist as a result of gene inactivation by the CAR carrying vector. In this case, intronic insertion resulted in removal of control of \u003cem\u003eTET2\u003c/em\u003e after splicing with the LV vector. In a CD22 CAR-T trial, LV mediated CBL oncogene activation has also been suspected to have caused CAR-T cell persistence [22,23].\u003c/p\u003e \u003cp\u003eTo understand RV and LV mediated genotoxicity more clearly and predict potential vector related side effects, \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e murine-based models have been developed. These include \u003cem\u003eCdkn2a\u003c/em\u003e null mice with inactivated p53 and pRb pathways that have been valuable in showing that the risk of tumour development by LV vectors is approximately 10-fold lower than RV. In a fetal/neonatal murine model LV delivery has also been found associated with high frequency liver cancer [21]. \u003cem\u003eIn vitro\u003c/em\u003e, the immortalization (IVIM) model that uses murine hematopoietic stem cells (HSC) has also been useful for vector risk assessment. This model demonstrates differences in RV and LV IS preference and integration in \u003cem\u003eEvi1\u003c/em\u003e and \u003cem\u003ePrdm16\u003c/em\u003e proto-oncogenes is responsible for cell transformation with RV greater than LV by a factor of 3:1. As a consequence, IVIM has been accepted by several regulatory agencies for pre-clinical evaluation of RV and LV safety. More recently data from this model has been used to provide transcriptomic signatures of leukaemogenesis supporting its use as a surrogate assay for genotoxicity assessment (SAGA) [24]. Although the models currently used to understand and assess LV genotoxicity have proven valuable, they are still considered potentially bias or over sensitive. Furthermore, no test can reliably predict long term safety in humans with widely variable predisposition to cancer.\u003c/p\u003e \u003cp\u003eAs a human based safety model is clearly needed, we chose to develop an alternative strategy to use for LV safety in which factors known to be contributory to genotoxicity are identified associated with LV design. For this, human induced pluripotent stem cells (iPSC) and their hepatocyte-like cell derivatives were used. iPSCs have been used widely to model human diseases and for pharmacotoxicological studies of disease treatment [25\u0026ndash;27]. iPSC can be reprogrammed from a variety of patient cells and offer a personalised approach to risk assessment by considering the genetic background of the host. We have previously shown iPSC can be reliably reprogramed to 3D hepatocyte like cells[28] and to be true liver surrogates matching primary hepatocytes at the transcriptional level (submitted manuscript). In this report, we used positive and negative control LV, carrying native and SIN configuration LTR, respectively, to infect iPSC and their 3D HLC derivatives [29]. Data from vector/host interactions were subjected to multi-omics analysis to characterise vector/host interactions believed to support pre-malignancy and oncogenesis. Data obtained from this analysis was then aligned to transcriptional signatures of a range of cancers to profile the genotoxicity potential of each LV. We consider this to be the first human-based model that can provide essential information valuable to identify vector/host interactions indicative of genotoxicity to support improved safe vector design. \u003csup\u003eh\u003c/sup\u003eInGetox may also be considered useful as a decision-making tool to support LV product approval for gene therapy.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eiPSC derived 3D HLC are true surrogates of primary hepatocytes\u003c/h2\u003e \u003cp\u003eBulk cultures of male JHU106i iPSC were differentiated to 3D HLC as previously described [30] and used for RNASeq to confirm iPSC pluripotency and HLC 3D spheroid characteristics. HLC gene expression was also compared with normal and several cancer cell phenotypes using unsupervised clustering, principal component analysis and powerful machine learning that showed HLCs align closely to primary hepatocytes (submitted manuscript). iPSC and 3D HLC cultures were infected with 2nd generation HIV-1 based LV vectors that differ by their LTR configuration and represent positive (pHV) and negative (pHR) controls carrying native LTR and SIN LTR, respectively. Infection of cells used an optimised MOI of 20 as detected by flow cytometry for GFP expression in iPSC at 90% and in 3D HLC spheroids at 85% following dissipation to single cells as previously shown [31]. Vector copy number (VCN), measured via TaqMan\u0026trade; q-RTPCR (n\u0026thinsp;=\u0026thinsp;3) next to standard curves generated for absolute vector copy quantification ranged between 1.45\u0026ndash;2.54 vg/cell. Culture viability after infection showed no significant difference from untreated cells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eLV insertion site analysis\u003c/h2\u003e \u003cp\u003eInsertion site (IS) analysis of LV was performed by EPTS/LM-PCR [32] following infection. Of 412,786 IS, insertions appeared in introns (60.9%), 3\u0026rsquo; (19.1%) or 5\u0026rsquo; (16.7%) untranslated regions (UTR) and exons (3.3%) \u003cb\u003e(Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA)\u003c/b\u003e. To investigate the distance between IS and transcription start sites (TSS), the mean distance across each data set was plotted in the scale of 0\u0026ndash;1 (exon or intron; normalised against gene length) or Log10 (3\u0026rsquo; or 5\u0026rsquo; UTR; value in base pairs). While inserts were found to be evenly distributed in introns, those identified in exons mainly congregated at 3\u0026rsquo; end of genes (median insertion sites at 82.8% of averaged gene length). On average, inserts identified in 3\u0026rsquo; or 5\u0026rsquo; UTR were 27.9 kbp or 29.6 kbp away from the protein-coding regions. We then profiled IS gene targeting in iPSC through quantification of gene number in the context of different regulatory regions at two time points (3 and 30 day) after infection \u003cb\u003e(Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eB)\u003c/b\u003e. Between these times, gene number with pHR insertions reduced from 10,890 to 7,955 and with pHV insertions from 13,906 to 7,706 suggesting potential enrichment or clonal selection favouring particular IS. Focussing on oncogenes and tumour suppressor genes, this reduction was observed for IS in introns, exons and UTRs, however, at the late time point insertions in oncogenes and tumour suppressor genes remained higher in introns \u003cb\u003e(Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eC and D).\u003c/b\u003e IS profiling of gene density, chromosome location, proximity to CpG islands and GC content and position within the gene transcription unit in both iPSC and HLC genomes were identified as expected and as previously reported for HIV-1 LV integration [33].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003ePathway analysis of LV IS associate with cellular proliferative potential\u003c/h2\u003e \u003cp\u003eHallmark pathway analysis of IS genes identified for each time point of investigation is shown in \u003cb\u003eFig.\u0026nbsp;1.\u003c/b\u003e IS identified 3 days post iPSC infection were found mainly in exons associated with cell cycle eg. E2F targets, G2M checkpoint and DNA damage eg. c-Myc targets and genes involved in DNA Repair. IS isolated at the later 30 day time point in iPSC were mainly in introns and UTR regions of genes that, in addition to those found at the early time points analysis, were identified in genes associated with the PI3K-AKT/MTOR pathway and epithelial mesenchymal transition. Enriched pathways associated with pHR or pHV IS genes were characteristic of pathways of the inflammatory response and hypoxia, respectively. The inserted genes in each pathway mostly targeted were tumour suppressor genes that included \u003cem\u003eTP53, NBN, POLD1, BRCA1/2, CHEK1, ATRX, BMPR1A, TGFBR2, SPOP, RUNX1, INPP4B, PTEN, PIK3R3, SMAD2, TSC2.\u003c/em\u003e For the tumour suppressor genes found with pHV inserts, these were in introns and UTR regulatory regions and present at both the early 3 day and late 30 day time points rather than restricted only to the early time point as observed for pHR insertions. For both vectors, oncogene IS were mainly in introns that included \u003cem\u003eEIF4E, RAF1, GSK3B, CALR, MAPK1, EGFR, LCK, RAC1, RIT1\u003c/em\u003e, and \u003cem\u003eRPTOR\u003c/em\u003e associated with the PI3K/AKT/mTOR signalling pathway. Conversely, in infected HLC cells fewer IS genes were found associated with oncogenes or tumour suppressor genes with enriched genes being only members of the TNFα signalling pathway \u003cb\u003e(Fig.\u0026nbsp;2).\u003c/b\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eIS clonal tracking in iPSC identifies genes associated with clonal outgrowth\u003c/h2\u003e \u003cp\u003eWe next clonally tracked IS in cancer genes in infected iPSC via their sequence count changes (SCC) between the 3- and 30-day time points and in HLC derived from infected iPSC. We focussed on differential absolute SCC IS of \u0026ge;\u0026thinsp;2-fold (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) represented only in significantly enriched genes and over-represented in cancer related biological pathways. Clonal tracking was investigated for IS that resided either in identical or non-identical locations in cancer genes to identify possible expansion of IS in proliferating clones with integrations that may influence gene expression. For SCC of IS identified at the same location at day 3 and 30, these always appeared in introns or UTR regions and in genes associated with eukaryotic translation, cell cycle regulation, kinases associated with protein phosphorylation and RNA export from nucleus (\u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). IS in genes with increasing SCC found at day 30 but not day 3 were suspected to be representative of IS below the limit of EPTS/LM-PCR detection in the bulk cell populations at the early time point. Seven hundred and seventeen targeted genes that included several oncogenes and tumour suppressor genes were found (\u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). Most interestingly, although IS clonal tracking found several increased SSC in cancer genes, these appeared most prevalent in HLC derived from iPSC infected by pHV than pHR (n\u0026thinsp;=\u0026thinsp;29 vs n\u0026thinsp;=\u0026thinsp;8, respectively). Included in these genes were \u003cem\u003eSET\u003c/em\u003e, \u003cem\u003eBRAF\u003c/em\u003e and \u003cem\u003eMECOM\u003c/em\u003e, previously shown to influence clonal selection.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eIsolation and analysis of iPSC clones following infection\u003c/h2\u003e \u003cp\u003eTo determine the effect of LV integration on IS gene expression, 7 single cell clones from pHR and 4 from pHV infections were isolated and expanded for DNA and RNA extraction for IS analysis and for q-RT-PCR analysis of gene expression. Following IS identification, for clones with IS that appeared, qRT-PCR analysis of the inserted gene was used for comparison with its expression in non-infected iPSC. For pHR and pHV, of 27 and 21 IS genes, respectively, each were identified\u0026thinsp;\u0026gt;\u0026thinsp;2 fold upregulated in their expression compared to uninfected iPSC (\u003cb\u003eTable \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eDifferential gene expression aligns with unique signatures representative of biological processes critical for oncogenesis\u003c/h2\u003e \u003cp\u003eTo gain molecular insights into potential changes in transcription associated with each LV, RNASeq on infected iPSC at the early and later time points was used to provide an unbiased transcriptome-wide profiling of differentially expressed genes (DEG) against control uninfected cells. At the early time point both pHR and pHV LV DEGs are associated with strong immune signatures with iPSC displaying active cytokine production after infection. For all DEG, regardless of timepoint of harvest, significantly represented DEGs were identified for pHV (n\u0026thinsp;=\u0026thinsp;1011) and pHR (n\u0026thinsp;=\u0026thinsp;871) with increases in 14 oncogenes and decreases in 14 tumour suppressor and increases in 10 oncogenes and decreases in 20 tumour suppressor genes, respectively (Fig.\u0026nbsp;3A \u003cb\u003eand B)\u003c/b\u003e. GO term analysis of these genes for annotated biological functions showed signalling pathways involving RNA transcription, protein modification, cell cycle, tyrosine kinase, and NF-kB common to both LV, however, unique enriched pathways for also evident to each LV. Upregulated DEGs in pHV infected iPSC characteristic of oncogenesis were implicated in methylation (n\u0026thinsp;=\u0026thinsp;11). This contrasted with unregulated DEG associated with pHR infection particular to protection against oncogenesis involving a response to DNA damage (n\u0026thinsp;=\u0026thinsp;20) and GTPase activity (n\u0026thinsp;=\u0026thinsp;10) \u003cb\u003e(Fig.\u0026nbsp;3C and D).\u003c/b\u003e Thirty eight (pHR) and forty nine (pHV) genes were identified with commonly shared as DEG (Log2FC\u0026thinsp;\u0026gt;\u0026thinsp;1 (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) against uninfected cells and with SCC count changes (\u0026gt;\u0026thinsp;10, p,0.05). GO term pathways were enriched for RNA transcription, WNT signalling and cell differentiation.\u003c/p\u003e \u003cp\u003eBetween early and late time points DEG were identified for oncogenes and tumour suppressor genes implicated in tyrosine kinase receptor signalling pathways and cellular senescence in pHR infected iPSC in contrast to p13K and MAPK signalling pathway activation in pHV infected cells.\u003c/p\u003e \u003cp\u003eDEG associated with pHR (n\u0026thinsp;=\u0026thinsp;419) comprised of 22 oncogenes and 10 tumour suppressor genes and for pHV (n\u0026thinsp;=\u0026thinsp;472) DEG comprised of 20 were oncogenes and 13 tumour suppressor genes. For both LV, DEG included DEG for \u003cem\u003eMECOM\u003c/em\u003e and \u003cem\u003eLMO2\u003c/em\u003e genes. At the late time point, the major difference between pHR and pHV infected iPSC was characterised by the p53 pathway responding to DNA damage in pHR infected cells compared to an inflammatory response in pHV infected iPSC (Fig.\u0026nbsp;4A-F). Hallmark analysis of enriched signalling pathways of these gene sets for IS genes also with DEG are shown (Fig.\u0026nbsp;5\u003cb\u003e).\u003c/b\u003e\u003c/p\u003e \u003cp\u003eFor infected HLC (harvested three days after LV transduction) compared to uninfected cells, DEG were also identified. DEG associated with pHR (n\u0026thinsp;=\u0026thinsp;569) comprised of 37 oncogenes and 51 tumour suppressor genes. This contrasted with pHV (n\u0026thinsp;=\u0026thinsp;3762) DEG of which 81 were oncogenes, including DEG for \u003cem\u003eMECOM\u003c/em\u003e, \u003cem\u003eLMO-2\u003c/em\u003e and \u003cem\u003eBRAF\u003c/em\u003e not found with pHR DEG and 82 tumour suppressor genes (Fig.\u0026nbsp;6). There were nearly 7 fold more upregulated DEG associated with pHV infected HLC than pHR suggesting a difference imposed by the native LTR configuration. GO term analysis of these showed genes mainly associated with tyrosine kinase signalling (n\u0026thinsp;=\u0026thinsp;23) and protein phosphorylation (n\u0026thinsp;=\u0026thinsp;19) and other related pathways such as ERK1/2 cascade and PI3K/AKT/mTOR pathway.\u003c/p\u003e \u003cp\u003eFocussing on the upregulated DEG used for GO term analysis, pHV infected HLCs exhibit pathways involving chemotaxis (n\u0026thinsp;=\u0026thinsp;23) and cancer signalling pathways (n\u0026thinsp;=\u0026thinsp;36). including NF-kB (n\u0026thinsp;=\u0026thinsp;69), MAPK (n\u0026thinsp;=\u0026thinsp;54), Wnt (n\u0026thinsp;=\u0026thinsp;39), JNK (n\u0026thinsp;=\u0026thinsp;38), and PI3K/AKT This contrasted with pHR associated pathways that were characterised by groups of genes protective against viral infection (n\u0026thinsp;=\u0026thinsp;36) such as interferon-associated genes IFI27, IFI44L, IFIT5, IRF7, ISG20), DNA damage (n\u0026thinsp;=\u0026thinsp;35) such as repair proteins DCLRE1C and RAD50, p53-mediated apoptotic proteins BCL3, BCL6, TOPORS, zinc finger proteins ZC3H12A, ZDHHC16, ZBTB4), and autophagy (n\u0026thinsp;=\u0026thinsp;25). (n\u0026thinsp;=\u0026thinsp;25) such as ULK2, ATG4B, ATG9A, DEPP1, EPG5, VPS18, VPS13A).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eWeighted correlation network analysis reveals distinct LV associated gene expression\u003c/h2\u003e \u003cp\u003eTo profile unique gene expression programmes across iPSCs or HLCs particular to pHR and pHV infection, we performed weighted correlation network analysis (WGCNA) and identified 9 significant functional modules associated with distinct biological processes \u003cb\u003e(Fig.\u0026nbsp;7A).\u003c/b\u003e These modules are indicated by colour codes, gene numbers (g), and percentage of shared inserted target genes (\u003cb\u003eTable \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e\u003c/b\u003e). Modules coloured in brown, turquoise, and green show higher proportions of shared inserted target genes and tumour suppressor genes associated with pHR infected iPSC or HLC (Fig.\u0026nbsp;7B). Through GO term analysis (Fig.\u0026nbsp;7C), these modules are associated with critical biological processes. critical for protein modification (brown), cellular metabolism (turquoise), synaptic signalling (blue), stimulus and immune response (green), epithelial cell differentiation (red), RNA metabolism (magenta), phagocytosis (pink), cellular respiration (yellow). The co-expressed genes that are associated with IS (Insert), oncogenes (Onco), or tumour suppressor genes (TSG) have been quantified (\u003cb\u003eTable \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e)\u003c/b\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eGene splicing with LV and the human genome in infected cells\u003c/h2\u003e \u003cp\u003eGene splicing and readthrough is also known to occur between LV and the host genome. We used RNASeq data to identify a total of 69 vector/host fusion transcripts and aligned these with genes in each co-expression module (Fig.\u0026nbsp;8A). Thirty-eight of these that were found as fusions were also identified as differentially expressed and with LV insertions (Fig.\u0026nbsp;8B). Triple positive genes for IS, DEG and fusions found in infected HLC were present as a smaller proportion (46%) to those in infected iPSC (79%), most likely due to iPSC gene expression being significantly greater than HLC. Once again, these fusion genes were mainly identified in the turquoise (n\u0026thinsp;=\u0026thinsp;39), brown (n\u0026thinsp;=\u0026thinsp;14), and green (n\u0026thinsp;=\u0026thinsp;6) module categories (Fig.\u0026nbsp;8C).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eAlignment of transcriptome changes in infected cells with cancer-specific gene signatures suggest LV associated genotoxicity\u003c/h2\u003e \u003cp\u003eTo investigate further the probability of carcinogenesis associated with pHR or pHV infection, we firstly defined cancer-specific signatures using differential analysis of cancer genes highly expressed in several cancer types, compared with their respective normal tissue. Through pathway analysis, using GO terms or hallmark gene sets, we found these signatures associated with enriched pathways involving nucleic acid synthesis/metabolism, active transcription, cell proliferation, E2F targets and the G2M checkpoint. These signatures were then used to score against the transcriptomes of infected iPSC or HLC from early or late harvest data analysis (Fig.\u0026nbsp;9A). In general, infected iPSCs are characterised by higher cancer scores than infected HLC as expected. At the early harvest time point, pHR infected iPSC or HLC have higher cancer scores than pHV infected cells, in agreement with the IS and DEGs we identified earlier. However, after continued iPSCs culturing and sample harvesting at day 30, pHV infected cells clearly showed higher cancer scores than pHR cultures \u003cb\u003e(Fig.\u0026nbsp;9B\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eEpigenetic analysis reveals unique lentivirus-induced methylation profiles\u003c/h2\u003e \u003cp\u003eWe then determined whether epigenetic changes via DNA methylation profiling of global methylation of the host genome or of selected genes in infected cells associates with LV infection. We characterised the DNA methylation landscape in host cells by identification of differentially methylated regions across infected samples. Next, we used CpG island methylation changes for cross referencing with IS and RNASeq data, to identify consensus genes for distinct biological processes and highlighted these for oncogenes and tumour suppressor genes.\u003c/p\u003e \u003cp\u003eIn general, pHV infected iPSC showed greater hypomethylation compared with pHR infected cells. The numbers of CpG island together with other regulatory elements including open-sea, shelf, and shore were found to peak at the gene body and were reduced elsewhere. These remained higher in hypermethylated regions than hypomethylated regions for both LV suggesting LVs general hypermethylation. We then focused on the CpG islands in hyper or hypo-methylated promoter regions (TSS1500 or TSS200) and found a marked increase in the number of hypermethylated genes corresponding to pHV infection of iPSC (n\u0026thinsp;=\u0026thinsp;210) than pHR infection (n\u0026thinsp;=\u0026thinsp;24) or HLC (n\u0026thinsp;=\u0026thinsp;28). Through pathway analysis, we found hypermethylated signalling molecules included \u003cem\u003eABR\u003c/em\u003e, \u003cem\u003eJAK3\u003c/em\u003e, \u003cem\u003eRASA3\u003c/em\u003e, \u003cem\u003eDGKZ\u003c/em\u003e, \u003cem\u003eDEF8\u003c/em\u003e, \u003cem\u003ePRKCG\u003c/em\u003e, \u003cem\u003ePRKCZ\u003c/em\u003e, \u003cem\u003eSPSB4\u003c/em\u003e that are associated with intracellular signal transduction that were enriched in pHV infected iPSC.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eMulti-omics analysis shows cancer-related genes shared between data sets\u003c/h2\u003e \u003cp\u003eTo cross compare data from differentially methylated genes (DMR) with DEG and IS targeted genes, we identified common genes shared between different data sets from iPSC infected with pHR or pHV (Fig.\u0026nbsp;10). Whilst none of the genes were identified in all three data sets, genes were shared between methylomic and genomic (n\u0026thinsp;=\u0026thinsp;3), methylomic and transcriptomic (n\u0026thinsp;=\u0026thinsp;15) and genomic and transciptomic (n\u0026thinsp;=\u0026thinsp;166). By comparing genes specifically in methylomic and transcription and in methylomic and IS data sets we identified similar pathways for cell signalling regarding cell adhesion (e.g., MARVELD3 and TBCD for tight junction assembly; FARP2, CDH11, CTNNA2, CDON, DST, ITGAV, NLGN1, PCDH10, PCDH18, ROBO2, SEMA5A), double-strand break repair (e.g., \u003cem\u003eRTEL1, SPIDR)\u003c/em\u003e and apoptotic processes (e.g., \u003cem\u003eFAF1, OPA1, TRAF2, CADM1, HTT, ITCH, MAGI3, RABEP1, SEMA3A, STK4)\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eBecause altered sequence, methylated state, and expression levels of tumour suppressor genes are thought to be implicated in carcinogenesis, we focused on these genes. Firstly, we performed GO term or KEGG enrichment analysis of the tumour suppressor genes in genomic data (n\u0026thinsp;=\u0026thinsp;717) and found that DNA repair pathways are mainly enriched in iPSC infected with pHR compared to pHV iPSC or HLC. In pHV infected iPSC, genes characteristic of positive regulation of autophagy (early harvest) or negative regulation of cell proliferation (late harvest) were found enriched, suggestive of a protective role against carcinogenesis. On the contrary, hepatocellular carcinoma-associated tumour suppressor genes were found to be enriched in pHV infected HLC. Pathway analysis of tumour suppressor genes in transcriptomic data (n\u0026thinsp;=\u0026thinsp;4283) showed a number of intracellular signalling pathways such as Foxo, PI3K-Akt, mTOR, JAK-STAT, and FCERI were upregulated in pHR and pHV infected iPSC. Only a few signalling pathways such as p53 were identified across different conditions in HLC (Figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eInsertional mutagenesis can result in genotoxicity in the host by altering the expression of genes important to cellular proliferation. This has been identified both in the clinic and non-clinically associated with retrovirus (RV) and lentivirus (LV) vectors [14,34\u0026ndash;38]. Limited information has been gained from clinical data and much of our understanding of vector associated genotoxicity has been provided by \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e models. These models have, however, been viewed as oversensitive, bias and difficult to transfer between laboratories and, therefore, deemed potentially unreliable and possibly restrictive to gene therapy clinical progress. None the less, models have revealed several vector derived factors that are believed to contribute towards genotoxicity [39\u0026ndash;43]. Recent discussion between leaders in the gene therapy field considering safety and long-term prediction of carcinogenesis in March 2023 in London has resulted in publication of a consensus document aimed at bringing the field together to find ways to circumvent this clear limitation by improving our understanding of vector associated genotoxicity and assay standardisation.\u003c/p\u003e\n\u003cp\u003eAs part of the consensus, the need for a human-based model was agreed. However, any such model, although partly useful to identify significant vector related side effects that would support oncogenesis or pre-malignancy, would still not provide for long-term prediction of vector safety.\u003c/p\u003e\n\u003cp\u003eIn this report, our approach was to investigate further, the factors suspected to support cancer development associated with LV vectors, which now superseed RV, in a fully characterised novel human-based platform. We describe here the development of \u003csup\u003eh\u003c/sup\u003eInGetox as an alternative \u003cem\u003ein vitro\u003c/em\u003e human-based model that offers a pre-clinical tool for improving LV safety design using mechanistic outreads characterising LV contribution to genotoxicity. These factors include IS gene selection in the host genome, their effect on the expression of these genes, global effects on differential gene expression, identifying truncations between the vector and nearby host cancer genes and their novel fusion transcripts that have arisen from vector/host splicing or readthrough from the internal promoter used to drive gene expression. Finally, included in this anaylsis, epigenetic modifications in form of methylation changes in the host genome following infection were measured.\u003c/p\u003e\n\u003cp\u003eBecause the vector long terminal repeat (LTR) is considered a major contributor to genotoxicity due to its promoter and enhancer activities, we chose to compare LV vectors that carry either the native LTR (pHV) or self-inactivating (SIN) LTR configuration (pHR) for their genotoxic effects on host cells.\u003c/p\u003e\n\u003cp\u003eRather than use immortal cell lines with mutated cancer gene associated pathways that we considered already carry bias towards oncogenicity, we chose human induced pluripotent stem cells (iPSC) for reprogramming to 3D liver-like cell (HLCs) derivatives. These cells were chosen because they present highly proliferative and quiescent states, respectively, and the liver is considered useful to examine drug pharmacotoxicological kinetics. We previously demonstrated iPSC and their HLC derivatives express markers of pluripotency and of the liver, respectively [28], and more recently further characterised these cells transcriptomically via unsupervised clustering, principal component analysis and powerful machine learning for comparison with several normal and cancer cell types. These analyses showed HLC gene expression align closely with primary hepatocytes and are not predisposed to oncogenesis making them genuine surrogate liver cells (publication submitted) to support the development of \u003csup\u003eh\u003c/sup\u003eInGetox.\u003c/p\u003e\n\u003cp\u003eBoth iPSC and HLC were highly permissive to infection enabling controllable vector copy number of between 1 and 2, with cell survival and viability close to that of untreated cells. By characterising 412,786 LV integrations in regulatory regions, both LV configurations were found distributed in the order of introns\u0026thinsp;\u0026gt;\u0026thinsp;UTR regions\u0026thinsp;\u0026gt;\u0026thinsp;exons. Interestingly, because the IS number reduced by 50% between the early (d3) and late (d30) time points, with increasing prevalence of IS in cancer genes (CG), this suggested enrichment of particular IS that favour cell survival or proliferation. This was also supported by increased gene expression in genes with IS for each LV in cloned iPSC and increasing sequence count changes in IS found in cancer genes that were clonally tracked both in iPSC and differentiated HLC bulk cultures. Although we found similar numbers of cancer genes with SSC carrying pHV and pHR LV insertions in iPSC over time, SSC appeared more than 3-fold higher in pHV infected HLC than pHR suggesting IS associated outgrowth continues following differentiation. Interestingly, in these cells, IS were identified in \u003cem\u003eSET\u003c/em\u003e, \u003cem\u003eBRAF\u003c/em\u003e and \u003cem\u003eMECOM\u003c/em\u003e genes that have previously been found associated with genotoxicity studies [44].\u003c/p\u003e\n\u003cp\u003eTo understand further genes chosen by each LV that were potentially supporting cellular outgrowth, we investigated, in more detail the IS identified in proliferating iPSC by IS gene positioning and hallmark pathway analysis. Differences between pHR and pHV LV were identified where enriched pathways characteristic of inflammation and hypoxia, respectively, were apparent and these IS were mainly in tumour suppressor genes. Interestingly, for pHR these appeared only at the early time point, whereas IS for these enriched pathways remained also at the later 30 day time point in pHV infected iPSC. Far fewer IS in oncogenes or tumour suppressor genes were found in HLC infected by each vector, most likely because of low expression of these genes. Targeted oncogenes in these cells were identified mainly in PI3K/AKT/mTOR signalling pathways.\u003c/p\u003e\n\u003cp\u003eTo understand the potential mechanisms that could be supporting genotoxicity, RNASeq was used to measure global differential gene expression (DEG) in bulk infected cultures by each LV. Upon infection of iPSC, at the early time point, upregulated DEG for both LV were found associated with signalling pathways and cytokine production characteristic of the innate immune response to infection. However, pHV appeared with a higher number of oncogenes with DEG. DEG associated with pHR in oncogenes and tumour suppressor genes were implicated DNA damage, for activated tyrosine kinase receptor pathway signalling and cell senescence pathways suggesting protection against oncogenesis [45]. This contrasted to pHV where activated PI3K signalling, and MAPK signalling pathways were prominent. In addition, GO analysis of pHV DEG revealed unique pathways associated with epigenetic changes involving methylation that are known to be characteristic to oncogenesis [46]. Importantly, network analysis of potential gene interactions for both LV at early and late time points highlighted \u003cem\u003eTP53\u003c/em\u003e as a central transcription factor in infected iPSC, however, a major difference between the pHR and pHV at the late time point, showed once again, using GO terms, pHR infected cells associated with the p53 pathway responding to DNA damage rather than an inflammatory response in pHV infected cells.\u003c/p\u003e\n\u003cp\u003eIn HLC, at the early time point of infection, we found pHV associated DEG nearly 7 fold that of pHR suggesting HLC infected with this vector dramatically alter their expression profiles. This DEG increase also included upregulated oncogenes and tumour suppressor genes compared to the SIN configuration vector. GO analysis showed these genes mainly associated with tyrosine kinase signalling, protein phosphorylation and other ERK1/2 cascade and P13K/AKT/mTOR related pathways and included \u003cem\u003eMECOM\u003c/em\u003e, \u003cem\u003eLMO-2\u003c/em\u003e and \u003cem\u003eBRAF\u003c/em\u003e genes. Interestingly, similar to our finding in infected iPSC and in contrast to pHV infected HLCs, DEG in SIN pHR vector infected HLC concerned genes that characteristically protect against viral infection.\u003c/p\u003e\n\u003cp\u003eWeighted correlation network analysis (WGCNA) enabled profiling of unique gene expression programmes across iPSC or HLC infected by pHV and pHR. Several significant functional modules associated with distinct biological processes with shared inserted target genes by both vectors were identified, enabling co-expressed gene association and quantification with IS in oncogenes and tumour suppressor genes. Once again correlations differed between each vector with pHR more associated with modules representing protein modification, cellular metabolism and immune response. RNASeq analysis also enabled fusion genes resulting between each LV and host loci to be identified. Correlated, co-expressed genes and IS in oncogenes and tumour suppressor genes were then aligned with these. Fusions are suspected to have arisen either due to vector host gene splicing or readthrough from the internal promoter into host genes past the vector 3\u0026rsquo; polyadenylation sequences. Although fewer fusions appeared in iPSC than HLC, a greater number of fusions appeared associated with pHR than pHV in iPSC. Interestingly, an equal number of fusions were observed in HLC for each LV. Fusions appeared to be more prominent in differentiated cells with more than two fold in HLC than iPSC.\u003c/p\u003e\n\u003cp\u003eOf the 69 fusions identified, nearly fifty percent (n\u0026thinsp;=\u0026thinsp;38) were identified in common with DEG and IS representing a potentially significant contribution to genotoxicity by both LV. This suggests modification important to avoid fusions occurring with host cancer genes. Further analysis of each fusion by mapping each the vector backbone would determine whether the fusions arose as a result of gene splicing or readthrough.\u003c/p\u003e\n\u003cp\u003eFollowing virus infection, methylation is believed to be an innate mechanism used to prevent successful virus establishment and propagation [47]. We sought to determine whether this response to infection also altered the methylation profile in the host. To do this, we profiled CpG methylation in iPSC and HLC genomes of specific genes and crossed referenced these with IS and DEG data to identify consensus cancer genes and their ontologies. Overall, pHV infected iPSC were 10 times more hypermethylated than pHR infected cells and enriched genes involved in signal transduction that are associated with cellular proliferation were found indicating the LTR has a major influence on the epigenetic response by cells following infection. This was shown also in mice where following infection by LV, where changes to methylation profiles altered the expression of cancer genes under the control of the E2F transcription factor [21].\u003c/p\u003e\n\u003cp\u003eTo use the contributory factors identified relating to genotoxicity for a comparative assessment of the safety of each LV, multi-omics was applied to LV IS, DEG and differential methylomics data. Whilst none of the genes were identified in all three data sets, data from suspected genotoxic factors were found overlapping with several sharing cell signalling pathways regarding cell adhesion. Distinct pathways between transcription and methylomic data sets were found for genes involved in DNA damage response and transcription and genomic sets for genes involved in the apoptotic response, respectively.\u003c/p\u003e\n\u003cp\u003eWe then focused on the tumour suppressor genes identified amongst these data sets, since altered sequence, methylated state, and or expression levels of these genes are thought to be implicated in carcinogenesis. Firstly, we performed GO term or KEGG enrichment analysis of the tumour suppressor genes in genomic data and found that DNA repair pathways are mainly enriched in pHR infected iPSC compared to pHV-infected iPSC or HLC. In pHV infected iPSC, genes characteristic of positive regulation of autophagy (Day 3) or negative regulation of cell proliferation (Day 30) were found to be enriched, suggestive of a protective role against carcinogenesis. On the contrary, hepatocellular carcinoma-associated tumour suppressor genes were found to be enriched in pHV-infected HLC. Pathway analysis for tumour suppressor genes in transcriptomic data showed a number of intracellular signalling pathways were upregulated in pHR and pHV infected iPSC and only a few signalling pathways like p53 were identified across different conditions in HLC.\u003c/p\u003e\n\u003cp\u003eTo infer the probability of carcinogenesis can be induced by pHR or pHV LV in iPSC or HLC, cancer signatures associated with enriched pathways regardless of cancer types were used for alignment with signatures scored from transcriptomes of infected iPSC or HLC at early and later time points compared with respective controls. Our finding that infected iPSC are characterised by higher cancer scores than infected HLC is most likely due to the nature of iPSC that are rapidly proliferating compared to differentiated cells. In agreement of the insertion sites and differentially expressed genes we identified at the early time point, pHR infected iPSC or HLC had higher cancer scores than pHV infected cells, however after prolonged culture (30d), the genotoxicity of the pHV vector was more apparent with higher cancer scores than pHR infected cells.\u003c/p\u003e\n\u003cp\u003eIn conclusion, we have developed \u003csup\u003eh\u003c/sup\u003eInGetox as a modular series of assays to measure vector interactions with human cells that are considered contributory to oncogenesis. We found both native LTR and SIN configuration LV carry genotoxic risk being capable of altering the expression of cancer genes. In this study, SIN LTR LV appears with less genotoxic risk than LV that carries the native LTR promoter and enhancer. As murine-based models may be considered unreliable to predict oncogenesis in humans and limited data can be obtained through clinical observations, \u003csup\u003eh\u003c/sup\u003eInGetox represents a novel platform that examines vector and host genotoxic interactions in a human genetic background. Although \u003csup\u003eh\u003c/sup\u003eInGetox may be considered predictive of safety, this is still limited to events that occur early on following gene therapy and, therefore, we propose this model beneficial to identify vector related genotoxicity that may be mitigated by improved vector design. For instance, this could involve modifications to the vector backbone such as removal of unnecessary splice donor or acceptor sites or 3\u0026rsquo; sequences to reduce promoter readthrough. Furthermore, of value would be a reduction in vector CpG sequences responsible for promoting host innate immune recognition that causes epigenetic changes in the host that can lead to cancer.\u003c/p\u003e\n\u003cp\u003eFinally, we propose \u003csup\u003eh\u003c/sup\u003eInGetox useful as a pre-clinical tool to screen the safety of several complex LV intended for gene therapy such as those used to carry CAR cassettes to generate CAR-T cells. Hence, by improving LV design and \u003csup\u003eh\u003c/sup\u003eInGetox screening, safer vectors will become available for regulator approval and safe therapeutic outcome.\u003c/p\u003e"},{"header":"Online methods","content":"\u003cdiv id=\"Sec16\" class=\"Section3\"\u003e\n\u003ch2\u003eGrowth and characterisation iPSC pluripotency and differentiation to hepatocyte-like cells\u003c/h2\u003e\n\u003cp\u003eA human iPSC line (JHUP106i) was cultured routinely on laminin 521 (BioLamina, France) coated plates in serum-free mTeSR\u0026trade;1 medium (STEMCELL Technologies, Cambridge) as previously described [48]. The cell was monitored regularly for infection and was propagated in antibiotic free medium.\u003c/p\u003e\n\u003cp\u003eBulk cultures of these cells were used for differentiation and infection experiments in these studies. These iPSCs were washed with 2ml PBS without calcium chloride and magnesium chloride. The cells were incubated with 1ml of Gentle Cell Dissociation Reagent (Stemcell Technologies) for 6 minutes until the cells transformed into single cells. Single hiPSCs were collected and resuspended in FACS-PBS (PBS supplemented with 0.1% BSA and 0.1% sodium azide), counted and resuspended at 1\u0026times;10\u003csup\u003e6\u003c/sup\u003e cells/ml for use. Tubes containing 100,000 cells were incubated for 30 minutes at 4\u0026deg;C with fluorochrome conjugated antibodies. Following incubation, cells were then washed once with PBS, removing any unbound antibodies and centrifuged at 1500 rpm for 5 minutes. Antibody binding to the surface of the cells was measured using the optimum concentration of an appropriate fluorochrome conjugated isotype specific antibody. In this study, unstained cells were used as a negative control. Measurement was carried out by using an electronic live gate on forward scatter and side scatter parameters. Data was acquired for 20,000\u0026ndash;50,000 gated live events for each sample using a Novocyte flow cytometer (Agilent Technologies) equipped with a 488 nm laser and analysed using Novoexpress software.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n\u003ch2\u003eFormation of self-aggregated 3D hiPSCs spheroids\u003c/h2\u003e\n\u003cp\u003eAgarose microplates were generated in 256-well format using the 3D Petri Dish\u0026reg; mould (Sigma Aldrich, Dorset) following the manufacturer instructions. These microplates were transferred to 12 well plates (Corning, Germany) as previously described [49,50]. hiPSCs were expanded on laminin coated plates, were incubated with 1 ml of Gentle Dissociation Buffer (Stemcell Technologies) for 7\u0026ndash;10 minute at 37\u0026deg;C. The single cell suspension was centrifuged at 0.2 rcf for 5 minute and resuspended in mTeSR\u0026trade;1 supplemented with 10 \u0026micro;M Y-27,632 (Calbiochem, Watford) at a density of 2.0 x 10\u003csup\u003e6\u003c/sup\u003e live cells/ml. The prewarmed agarose microplates were seeded by transferring 190 ul of resulted cell suspension. After 2 hours, 1 ml mTeSR\u0026trade;1 supplemented with 10 uM Y-27,632 was gently added to each well of 12-well plate and incubated overnight at 37 C.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n\u003ch2\u003eHepatic induction of self-aggregated hiPSCs spheroids\u003c/h2\u003e\n\u003cp\u003eDifferentiation to HLC was performed as previously described [48,51]. Differentiation was initiated by replacing mTeSR\u0026trade;1 with endoderm differentiation medium: RPMI1640 containing 1x B27 (Life Technologies), 100ng/ml Activin A (PeproTech, Hammersmith), and 50 ng/ml Wnt3a (R\u0026amp;D Systems, Abingdon). The medium was changed every 24 hours, for 72 hours. On day 5, endoderm differentiation was substituted with hepatoblast differentiation medium. This medium was changed every second day for a further 5 days. This medium was composed of knockout-DMEM (Life Technologies), knockout serum replacement (KOSR-Life Technologies), 0.5% Glutamax (Life Technologies), 1% non-essential amino acids (Life-Technologies), 0.2% b-mercaptoethanol (Life Technologies), and 1% DMSO (Sigma Aldrich). On day 10, hepatoblast medium was replaced with hepatocyte maturation medium HepatoZYME (Life Technologies) containing 1% Glutamax (Life-Technologies), supplemented with 10 ng/ml hepatocyte growth factor (HGF, PeproTech) and 20ng /ml oncostatin M (OSM, PeproTech) as described previously [48,51]. On day 21 of differentiation, cells were cultured in maintenance medium containing William\u0026rsquo;s E media (Life Technologies), supplemented with 10ng/ml EGF (R\u0026amp;D systems), 10 ng/ml VEGF (R\u0026amp;D Systems), 10 ng/ml HGF (PeproTech), 10ng/ml bFGF (PeproTech), 10% KOSR, 1% Glutamax, and 1% penicillin-streptomycin (Thermo Fisher Scientific) for the remining study, as previously described [49].\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\n\u003ch2\u003eHistology and immunofluorescence of 3D hepatospheres\u003c/h2\u003e\n\u003cp\u003e3D spheroids were fixed in ice-cold methanol for 1 hour, washed in PBS and embedded in agarose. Agarose-embedded spheroids were embedded in paraffin and 4um sections were prepared. Antigen retrieval was performed using 1 x Tris-EDTA buffer solution for 15 minutes. Paraffin-embedded sections were also stained with Eosin and Hematoxylin and mounted in Pertex before microscopy. Brightfield images were taken using a Nikon Eclipse e600 microscope equipped with a Retiga 2000R camera (Q-imaging) and Image-Pro Premier software.\u003c/p\u003e\n\u003cp\u003eIn order to stain sectioned hepatospheres, tissue was blocked with 10% BSA in PBS-tween (PBST) and incubated with primary antibody overnight at 4 C. Species-specific fluorescent -conjugated secondary antibody were used (Alexa Flour 488/Alexa Flour 568; Invitrogen). Sections were counterstained with DAPI (4\u0026rsquo;6-diamidino-2-phenylin-dole) and mounted with Fluoromount-G (SouthernBiotech) before microscopy.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\n\u003ch2\u003eqRT-PCR\u003c/h2\u003e\n\u003cp\u003eRNA was extracted from 3D hepatospheres using RNAeasy Mini RNA Extraction Kit (Qiagen) according to manufacturer\u0026rsquo;s instructions. RNA quantity and quality were evaluated using Nanodrop\u003csup\u003eTM\u003c/sup\u003e200c. Following this step, cDNA was amplified using the RT2 First Strand Kit (Qiagen) following the manufacturer\u0026rsquo;s instruction. qPCR was performed with TaqMan Fast Advance Mastermix and primer pairs, using a Roche LightCycler 480 real-time PCR system. Gene expression was normalised to housekeeping gene; glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and expressed as relative expression over 3D hepatospheres on day 0 of differentiation as control sample. qPCR was conducted in triplicate and data was analysed using Roche LightCycler 480 software.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\n\u003ch2\u003eHepatocyte phenotyping\u003c/h2\u003e\n\u003cp\u003eTo evaluate Cyp3A activity, 50 uM of Luciferin-PFBE substrate (Promega, Southampton) was incubated with 3D hepatospheres in HepatoZYME medium supplemented with 10 ng/ml HGF. Cytochrome P450 activity was measured 24 hour later using the P450-Glo assay kit (Promega) following manufacturer\u0026rsquo;s instruction.\u003c/p\u003e\n\u003cp\u003eTo measure AFP and ALB secretion, the supernatant was collected after 24 hour and quantified using commercial ELISA kits (Alpha Diagnostics International, Texas). Data were normalised with the total protein content measured using bicinchonic acid (BCA) assay (Thermo Fisher Scientific).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\n\u003ch2\u003eGeneration of high titre LV vectors\u003c/h2\u003e\n\u003cp\u003eHEK293T cells were grown in DMEM GlutaMAX supplemented with 10% foetal bovine serum and 1% Penicillin Streptomycin (Fisher Scientific, Loughborough), at 37\u0026ordm;C with 5% CO\u003csup\u003e2\u003c/sup\u003e. Cells were passaged regularly upon confluency.\u003c/p\u003e\n\u003cp\u003epHR'SIN-cPPT-SEW (pHR) and it\u0026rsquo;s native LTR counterpart (pHV) LV were generated as previously described [52]. Briefly, 1.5x10\u003csup\u003e7\u003c/sup\u003e HEK293T cells were seeded per T175 flask and incubating at 37\u0026deg;C, 5% CO2 overnight. Cells were transfected with 16\u0026micro;g eGFP transgene, 12\u0026micro;g pCMVR8.74 and 4\u0026micro;g pMD2.G with a transfection reagent in serum free medium. Medium was replaced 24 hours post transfection and supernatant harvested every 24 hours for 72 hours post replacement. Conditioned medium was filtered through 0.45\u0026micro;M filters (Fisher Scientific) and stored at 4\u0026deg;C for future use.\u003c/p\u003e\n\u003cp\u003eConditioned medium was concentrated via ultracentrifugation at 23,000 rpm at 4\u0026deg;C for 2 \u0026frac12; hours, using an SW32Ti rotor and Optima XPN ultracentrifuge (Beckman Coulter, High Wycombe). Viral pellet was resuspended in 200\u0026micro;l serum free medium and stored at -80\u0026deg;C for future use.\u003c/p\u003e\n\u003cdiv id=\"Sec23\" class=\"Section3\"\u003e\n\u003ch2\u003eLV titration\u003c/h2\u003e\n\u003cp\u003eInfectious LV titre was calculated by as previously reported [53]. Briefly, 2 x 10\u003csup\u003e5\u003c/sup\u003e HEK293T cells were seeded and incubated at 37\u0026deg;C, 5% CO\u003csup\u003e2\u003c/sup\u003e overnight to adhere. Serial dilutions of concentrated LV were prepared and incubated in complete cell culture medium with 5\u0026micro;g/ml polybrene (Sigma Aldrich), for 20 minutes at room temperature before addition to cells. Medium was replaced after 24 hours incubation and incubated for a further 48 hours before analysis using a Novocyte flow cytometer (ACEA Biosciences Inc, San Diego) and data analysis using NovoExpress software. Dilutions expressing 1\u0026ndash;30% GFP expression were analysed as accurate representations of viral titre (TU/ml), calculated as below.\u003c/p\u003e\n\u003cp\u003e((Cell count x (Percentage GFP expression/100))/Volume) * Dilution factor\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\n\u003ch2\u003eOptimisation of iPSC and HLC gene transfer\u003c/h2\u003e\n\u003cp\u003eOne day prior to transduction, 3 x 10\u003csup\u003e5\u003c/sup\u003e iPSCs were seeded in pre-coated laminin plates. The following days, one well of the cells were dissociated using Gentle Cell Dissociation reagent (StemCell Technology). For lentiviral transduction, mTeSR1 medium containing 10 \u0026micro;M Y-27,632 (Calbiochem) and 5\u0026micro;g/ml polybrene reagent (Sigma-Aldrich, UK) was prepared. The virus added to the medium and the mixture was incubated for 20 minutes at room temperature. The medium from the cells removed and replaced with the medium containing the virus and rock inhibitor. The plate was incubated at 37\u0026deg;C for 24 hours. Following day, the transduction medium was replaced with fresh complete mTesr1 medium. This step was continued for three days and medium refreshed daily. The cells were extracted for flow cytometry analysis to determine the number of GFP positive cells.\u003c/p\u003e\n\u003cp\u003eFor 3D hepatospheres, William\u0026rsquo;s E medium (Life Technology) supplemented with 10 \u0026micro;M Y-27,632, 10 ng/ml EGF (R\u0026amp;D Sytems), 10 ng/ml VEGF (R\u0026amp;D Systems), 10 ng/ml HGF (PeproTech), 10 ng/ml bFGF (PeproTech), 5\u0026micro;g/ml polybrene (Sigma-Aldrich) and the virus was prepared. The mixture was incubated for 20 minutes at room temperature before adding to the cells. The transduction medium added to the cells and incubated for 24 hours before with complete William\u0026rsquo;s E medium with essential growth factor supplements. Following transduction, transduced 3D heps were kept 3 to 7 days until fluorescent cells were appeared.\u003c/p\u003e\n\u003cdiv id=\"Sec25\" class=\"Section3\"\u003e\n\u003ch2\u003eCloning iPSC\u003c/h2\u003e\n\u003cp\u003ePrior to single cell cloning, iPSCs were transduced with lentiviral vectors as previously described. On day of SCC complete mTeSR1 (StemCell Technology) medium with conditioned medium at a ratio of 1:1 was prepared. The medium was supplemented with 10 \u0026micro;M Y-27,632 (Calbiochem) to enhance the cell survival. Transduced positive GFP cells were washed with PBS (Sigma-Aldrich) once and gently dissociated using Gentle Cell Dissociation reagent (StemCell Technology) for 15 minutes. The single cells were resuspended in mTeSR1 and Y-27632 and counted by haemocytometer. To make the final 2 cells/ml, 4.8 \u0026micro;l of cell solution was transferred to 12 ml of complete/conditioned medium and 500 \u0026micro;l from the cell suspension was dispensed per well of a 24-well plate. This was to ensure the plate was seeded at a density of 1 cell/well. Following seeding, the cells were undisturbed for 7 to 10 days. After 7 days, the plate was scanned for colonies. The cells from each colony were expanded and harvested for DNA/RNA extraction. 33 (pHR LV) and 7 (pHV LV) clones were isolated from infected samples respectively\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec26\" class=\"Section3\"\u003e\n\u003ch2\u003eNucleic acid isolation\u003c/h2\u003e\n\u003cp\u003eDNA and RNA samples were isolated from transduced samples according using DNeasy Blood \u0026amp; Tissue Kit and RNeasy Mini Kit (Qiagen, Manchester) respectively, according to manufacturer\u0026rsquo;s instructions. RNA was treated with DNAase I to remove contaminants, according to manufactures instructions (Qiagen). Nucleic acid concentration and purity was analysed using NanoDrop\u0026trade; 2000c spectrophotometer (ThermoFisher Scientific, Hemel Hempstead).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec27\" class=\"Section3\"\u003e\n\u003ch2\u003eNucleic acid integrity analysis\u003c/h2\u003e\n\u003cp\u003eDNA samples were initially analysed for vector presence. Therefore, a vector specific primer pair was designed and used in a standard PCR with 10ng DNA. Products were applied to 2% gel electrophoresis to visualize PCR amplicons and the detection of expected bands.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec28\" class=\"Section2\"\u003e\n\u003ch2\u003eRNAseq sample preparation for fusion transcript and expression analysis\u003c/h2\u003e\n\u003cp\u003eRNA quality was assessed on TapeStation 2200 system using TapeStation RNA ScreenTape \u0026amp; Reagents (Agilent, Santa Clara). Up to 1000 ng total RNA per sample were applied to SureSelect Strand-Specific mRNA Library Preparation for Illumina (Agilent) and TruSeq Stranded mRNA Library Prep Kit (Illumina) according to manufacturer\u0026rsquo;s instructions for library preparation. Libraries were sequenced in 150PE mode on Illumina HiSeq System. Sequencing data was analysed using GENE-IS (Afzal et al., 2017) for the detection of fusion transcripts and DEseq2 for the assessment of differentially expressed genes.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec29\" class=\"Section2\"\u003e\n\u003ch2\u003eVector copy number analysis\u003c/h2\u003e\n\u003cp\u003eVector copy number in the samples were determined by quantitative TaqMan\u0026trade; Universal PCR Master Mix. Briefly, 10 ng of DNA were applied in triplicate analysis on CFX96 Touch Real-Time PCR Detection System (Bio-Rad, Hercules). A standard curve of 8 calibration standards (10\u003csup\u003e7\u003c/sup\u003e \u0026ndash; 5 copies) was generated for absolute quantification of vector copies in the samples. LV specific primers and probes were designed and ordered from Sigma-Aldrich (Munich, Germany) and IDT Technologies (Coralville, USA), respectively. Primers were used at a final concentration of 720nM, probes at 140nM in a total reaction volume of 20\u0026micro;l.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch2\u003eAnalysis of vector integration sites\u003c/h2\u003e\n\u003cp\u003eViral vector integration sites analysis was performed using Sharing-Extension Primer Tag Selection / Linear-Mediated Polymerase Chain Reaction (S-EPTS/LM-PCR), which is a shearing DNA based integration site analysis method, followed by next-generation sequencing.\u003c/p\u003e\n\u003cp\u003eRaw sequencing data were trimmed based on quality (Phred) and filtered for containing both molecular barcodes at full identity. Remaining reads were further analysed using GENE-IS. Briefly, sequences were further trimmed and only sequences containing the expected vector-specific stretch were considered for following steps. First, sequences were aligned to the human genome (UCSC assembly release number hg38) by Burrows-Wheeler Aligner (BWA) MEM algorithm. Potential integration sites were then mapped with BLAST at a minimum alignment identity percentage of 95%. Adjacent genes and other features were annotated according to RefSeq database. For each detected integration site, the relative sequence count compared to all sequences was calculated.\u003c/p\u003e\n\u003cdiv id=\"Sec31\" class=\"Section2\"\u003e\n\u003ch2\u003eAnalysis of common integration sites\u003c/h2\u003e\n\u003cp\u003eBiologically relevant IS clusters, called common integration sites (CIS), were analysed using a graphs based approach. Any IS detected was considered as node that contained the IS locus. If the distance of two nodes was less than 50kb, the nodes were connected and resulting nodes sets considered as CIS.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec32\" class=\"Section2\"\u003e\n\u003ch2\u003eIntegration sites in proximity to cancer-related genes\u003c/h2\u003e\n\u003cp\u003eA list of over 700 well-defined cancer genes was compiled from the Cancer Gene Census database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://cancer.sanger.ac.uk/census\u003c/span\u003e\u003c/span\u003e). Cancer gene data was obtained from Ensembl human genes (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.ensembl.org/biomart/martview/\u003c/span\u003e\u003c/span\u003e; version GRCh38.p10). Relative frequencies of integration sites, that were detected within a 100kb window of a TSS of a cancer-related gene, were analysed.\u003c/p\u003e\n\u003cdiv id=\"Sec33\" class=\"Section3\"\u003e\n\u003ch2\u003eMethylome analysis using the Illumina Epic Kit\u003c/h2\u003e\n\u003cp\u003eUp to 250ng DNA per sample were applied to Infinium Methylation EPIC Kit (Illumina, San Diego) according to manufacturer\u0026rsquo;s instructions and arrays were scanned on Illumina\u0026rsquo;s iScan System. Generated data was analysed using \u0026lsquo;The Chip Analysis Methylation Pipeline\u0026rsquo; in order to determine differentially methylated regions.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec34\" class=\"Section3\"\u003e\n\u003ch2\u003eBioinformatic analysis pipeline\u003c/h2\u003e\n\u003cp\u003eWe established an analytical pipeline in R (v.4.0.2). Raw data was curated into feature-sample matrices and streamlined into ExpressionSet objects using BioBase (v.2.50.0). The subsequent analysis mainly includes differential analysis, gene set enrichment analysis, signature score assessment, and weighted gene co-expression network analysis.\u003c/p\u003e\n\u003cp\u003e\u003cspan class=\"Underline\"\u003eDifferential Analysis\u003c/span\u003e: We used limma (v3.46.0) for differential analysis on count matrices. First, we normalised and conducted log2-transformation on count matrices. Next, we constructed design matrices using phenotypic data and fitted these to the processed matrices. Following empirical Bayes moderation, we identified differentially expressed genes (DEGs) and retained those with an absolute log2 fold-change greater than 1 and BH-adjusted p-values less than 0.01. Lastly, we visualised these DEGs using EnhancedVolcano (v.1.8.0).\u003c/p\u003e\n\u003cp\u003e\u003cspan class=\"Underline\"\u003eGene Set Enrichment Analysis (GSEA)\u003c/span\u003e: We used clusterProfiler (v3.18.1) for GSEA. First, we ranked gene expression levels from highest to lowest by comparing transcriptomic samples from study groups against controls. Next, we used Hallmark gene sets or GO terms from MSigDB (v.7.5.1) to perform GSEA, resulting in BH-adjusted p values, ratio of genes from each set, and other attributes. Lastly, we used treeplot or cnetplot to visualise the results.\u003c/p\u003e\n\u003cp\u003e\u003cspan class=\"Underline\"\u003eSignature Score Analysis\u003c/span\u003e: We defined molecular signatures for specific cancers using TCGA RNA-seq datasets. Differential analysis criteria for these signature genes were set at log2 fold-change\u0026thinsp;\u0026gt;\u0026thinsp;2 and BH-adjusted p-values\u0026thinsp;\u0026lt;\u0026thinsp;0.01. We then determined the scaled mean expression levels of these genes and visualised them using pheatmap (v1.0.12).\u003c/p\u003e\n\u003cp\u003e\u003cspan class=\"Underline\"\u003eWeighted Gene Co-expression Network Analysis (WGCNA)\u003c/span\u003e: We used WGCNA (v.1.70-3) for the analysis. We began with normalised and transposed count matrices and performed network topology analysis, resulting in an optimal soft threshold power. We then constructed topological overlap matrix using the blockwiseModules command and determine traits associated with each module by calculating hypothetical central genes. Lastly, we identified potential key drivers in selected modules by using the intramodularConnectivity command.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eExperimental procedures: SS, SA, MZ, SF, MSK. Data generation: SS, SA, MZ, SF, MSK, SP. Data interpretation: SS, SA, AG, AZ, AP, MZ, RF, WW, OSF, AD, YT, SNW, IGF, SP, MS, DH, MT. Data analysis: SS, SA, AG, AZ, AP, MZ, SF, DH, MF, RF, WW, OSF, AD, YT, SNW, IGF, SP, MS, MT. Paper preparation: SS, AG, AZ, AP, MT. Paper editing: SS, AG, AZ, MT. Supervision: DH, WW, OSF, AD, YT, SNW, IGF, MS, MT. Conceptualisation: DH, MS, MT. Methodology: DH, MS, YT, SNW, MS, MT. Paper review: SJ, AG, MT. Paper editing: SS, MT\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest statement:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDH is founder, director and shareholder in Stemnovate Limited and Stimuliver ApS. All other authors declare no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding statement:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was funded by an NC3Rs CRACK IT Challenge 21: InMutagene award, sponsored by GSK and Novartis.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003e\u003cspan\u003eZufferey R, Dull T, Mandel RJ, Bukovsky A, Quiroz D, Naldini L et al. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery 1998;72:9873\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eZhang S, Chen S, Li W, Guo X, Zhao P, Xu J et al. Rescue of ATP7B function in hepatocyte-like cells from Wilson\u0026rsquo;s disease induced pluripotent stem cells using gene therapy or the chaperone drug curcumin 2011;20:3176\u0026ndash;87. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/hmg/ddr223\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eJahner D, Jaenisch R. Retrovirus-induced de novo methylation of flanking host sequences correlates with gene inactivity 1985;315:594\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/315594a0\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eSaez-Cirion A, Manel N. Immune Responses to Retroviruses. 2018;36:193\u0026ndash;220. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1146/annurev-immunol-051116-052155\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eLucendo-Villarin B, Rashidi H, Alhaque S, Fischer L, Meseguer-Ripolles J, Wang Y et al. Serum Free Production of Three-dimensional Human Hepatospheres from. Pluripotent Stem Cells 2019:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3791/59965\u003c/span\u003e\u003c/span\u003e. https://doi.org/10.3791/59965.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eWaddington SN, Nivsarkar MS, Mistry AR, Buckley SM, Kemball-Cook G, Mosley KL et al. Permanent phenotypic correction of hemophilia B in immunocompetent mice by prenatal gene therapy 2004;104:2714\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1182/blood-2004-02-0627\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eAlmarza D, Bussadori G, Navarro M, Mavilio F, Larcher F, Murillas R. Risk assessment in skin gene therapy: viral-cellular fusion transcripts generated by proviral transcriptional read-through in keratinocytes transduced with self-inactivating lentiviral vectors 2011;18:674\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/gt.2011.12\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eCavazzana-Calvo M, Payen E, Negre O, Wang G, Hehir K, Fusil F et al. Transfusion independence and HMGA2 activation after gene therapy of human beta-thalassaemia 2010;467:318\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature09328\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eSchroder AR, Shinn P, Chen H, Berry C, Ecker JR, Bushman F. HIV-1 integration in the human genome favors active genes and local hotspots 2002;110:521\u0026ndash;9. https://doi.org/S0092867402008644 [pii].\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eWu X, Li Y, Crise B, Burgess SM. Transcription start regions in the human genome are favored targets for. MLV Integr. 2003;300:1749\u0026ndash;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.1083413\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eEsteve J, Blouin JM, Lalanne M, Azzi-Martin L, Dubus P, Bidet A et al. Generation of induced pluripotent stem cells-derived hepatocyte-like cells for ex vivo gene therapy of primary hyperoxaluria type 1 2019;38:101467. https://doi.org/S1873-5061(19)30097-2 [pii].\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eHowe SJ, Mansour MR, Schwarzwaelder K, Bartholomae C, Hubank M, Kempski H et al. Insertional mutagenesis combined with acquired somatic mutations causes leukemogenesis following gene therapy of SCID-X1 patients 2008;118:3143\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1172/JCI35798\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eGonin P, Buchholz CJ, Pallardy M, Mezzina M. Gene therapy bio-safety: scientific and regulatory issues 2005:S146-52. https://doi.org/3302629 [pii].\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eSchmidt M, Hoffmann G, Wissler M, Lemke N, Mussig A, Glimm H et al. Detection and direct genomic sequencing of multiple rare unknown flanking DNA in highly complex samples 2001;12:743\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1089/104303401750148649\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eSuleman S, Payne A, Bowden J, Al Haque S, Zahn M, Fawaz S et al. HIV- 1 lentivirus tethering to the genome is associated with transcription factor binding sites found in genes that favour virus survival 2022.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eWills CA, Drago D, Pietrusko RG. Clinical holds for cell and gene therapy trials: Risks, impact, and lessons learned 2023;31:101125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.omtm.2023.101125\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eStein S, Ott MG, Schultze-Strasser S, Jauch A, Burwinkel B, Kinner A et al. Genomic instability and myelodysplasia with monosomy 7 consequent to EVI1 activation after gene therapy for chronic granulomatous disease 2010;16:198\u0026ndash;204. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nm.2088\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eDeichmann A, Hacein-Bey-Abina S, Schmidt M, Garrigue A, Brugman MH, Hu J et al. Vector integration is nonrandom and clustered and influences the fate of lymphopoiesis in SCID-X1 gene therapy 2007;117:2225\u0026ndash;32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1172/JCI31659\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eCorcoran LM, Adams JM, Dunn AR, Cory S. Murine T lymphomas in which the cellular myc oncogene has been activated by retroviral insertion 1984;37:113\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0092-8674(84)90306-4\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eShalginskikh N, Poleshko A, Skalka AM, Katz RA. Retroviral DNA methylation and epigenetic repression are mediated by the. antiviral host protein Daxx. 2013;87:2137\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1128/JVI.02026-12\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eWang Y, Alhaque S, Cameron K, Meseguer-Ripolles J, Lucendo-Villarin B, Rashidi H et al. Defined and Scalable Generation of Hepatocyte-like Cells from Human. Pluripotent Stem Cells 2017:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3791/55355\u003c/span\u003e\u003c/span\u003e. https://doi.org/10.3791/55355.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eLucendo-Villarin B, Rashidi H, Alhaque S, Fischer L, Meseguer-Ripolles J, Wang Y et al. Serum Free Production of Three-dimensional Human Hepatospheres from. Pluripotent Stem Cells 2019:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3791/59965\u003c/span\u003e\u003c/span\u003e. https://doi.org/10.3791/59965.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eBushman FD. Retroviral Insertional Mutagenesis in Humans: Evidence for Four Genetic Mechanisms Promoting Expansion of. Cell Clones. 2020;28:352\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ymthe.2019.12.009\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eLi Z, Modlich U, Anjali M. Leukemia diagnosis in murine bone marrow transplantation models 2009;506:311\u0026ndash;29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-1-59745-409-4_21\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eCesana D, Sgualdino J, Rudilosso L, Merella S, Naldini L, Montini E. Whole transcriptome characterization of aberrant splicing events induced by lentiviral vector integrations 2012;122:1667\u0026ndash;76. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1172/JCI62189\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eHacein-Bey-Abina S, Hauer J, Lim A, Picard C, Wang GP, Berry CC et al. Efficacy of gene therapy for X-linked severe combined immunodeficiency 2010;363:355\u0026ndash;64. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1056/NEJMoa1000164\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eCavazzana M, Bushman FD, Miccio A, Andre-Schmutz I, Six E. Gene therapy targeting haematopoietic stem cells for inherited diseases: progress and challenges 2019;18:447\u0026ndash;62. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41573-019-0020-9\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eGay V, Moreau K, Hong SS, Ronfort C. Quantification of HIV-based lentiviral vectors: influence of several cell type parameters on vector infectivity 2012;157:217\u0026ndash;23. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00705-011-1150-5\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eSchwarzer A, Talbot SR, Selich A, Morgan M, Schott JW, Dittrich-Breiholz O et al. Predicting genotoxicity of viral vectors for stem cell gene therapy using gene expression-based machine learning 2021;29:3383\u0026ndash;97. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ymthe.2021.06.017\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eBaum C, Kustikova O, Modlich U, Li Z, Fehse B. Mutagenesis and oncogenesis by chromosomal insertion of gene transfer vectors 2006;17:253\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1089/hum.2006.17.253\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eSchambach A, Galla M, Maetzig T, Loew R, Baum C. Improving transcriptional termination of self-inactivating gamma-retroviral and lentiviral vectors. 2007;15:1167\u0026ndash;73. https://doi.org/S1525-0016(16)31676-8 [pii].\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eNobles CL, Sherrill-Mix S, Everett JK, Reddy S, Fraietta JA, Porter DL et al. CD19-targeting CAR T cell immunotherapy outcomes correlate with genomic modification by vector integration 2020;130:673\u0026ndash;85. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1172/JCI130144\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eWu C, Dunbar CE. Stem cell gene therapy: the risks of insertional mutagenesis and approaches to minimize genotoxicity 2011;5:356\u0026ndash;71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11684-011-0159-1\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eKoyama S, Ishii KJ, Coban C, Akira S. Innate immune response to viral infection 2008;43:336\u0026ndash;41. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cyto.2008.07.009\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eCameron K, Tan R, Schmidt-Heck W, Campos G, Lyall MJ, Wang Y et al. Recombinant Laminins Drive the Differentiation and Self-Organization of hESC-Derived Hepatocytes. 2015;5:1250\u0026ndash;62. https://doi.org/S2213-6711(15)00313-6 [pii].\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eRashidi H, Luu NT, Alwahsh SM, Ginai M, Alhaque S, Dong H et al. 3D human liver tissue from pluripotent stem cells displays stable phenotype in vitro and supports compromised liver function in vivo 2018;92:3117\u0026ndash;29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00204-018-2280-2\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eTherapy AS of, G, Gene C. Cell \u0026amp; RNA Therapy Landscape: Q1 Quarterly Data Report. 2023.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eFraietta JA, Nobles CL, Sammons MA, Lundh S, Carty SA, Reich TJ et al. Disruption of TET2 promotes the therapeutic efficacy of CD19-targeted T cells 2018;558:307\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41586-018-0178-z\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eTouw IP, Erkeland SJ. Retroviral insertion mutagenesis in mice as a comparative oncogenomics tool to identify disease genes in human leukemia. 2007;15:13\u0026ndash;9. https://doi.org/S1525-0016(16)31245-X [pii].\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003ePeng Y, Laser J, Shi G, Mittal K, Melamed J, Lee P et al. Antiproliferative effects by Let-7 repression of high-mobility group A2 in uterine leiomyoma 2008;6:663\u0026ndash;73. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1158/1541-7786.MCR-07-0370\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eDoerfler W. Impact of foreign DNA integration on tumor biology and on evolution via epigenetic alterations 2012;4:41\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2217/epi.11.111\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eCJ.Boztug CB, Paruzynski K, Witzel A, Schwarzer M, Rothe A, Modlich M, Beier U, G\u0026ouml;hring R, Steinemann G, Fronza D, Ball R, Haemmerle CR, Naundorf R, Rose SK\u0026uuml;hlckeK, M.Fraser C, Mathias L, Ferrari RA, Al-Herz MR, W, Kalle. .Kondratenko, I.Mar\u0026oacute;di, L.Glimm, H.Schlegelberger, B.Schambach, A.Albert, MH.Schmidt, M.von, C.Klein,. Gene therapy for Wiskott-Aldrich syndrome\u0026ndash;long-term efficacy and genotoxicity 2014;12.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eSiler U, Paruzynski A, Holtgreve-Grez H, Kuzmenko E, Koehl U, Renner ED, et al. Successful Combination of Sequential Gene Therapy and Rescue Allo-HSCT in Two Children with X-CGD -. Importance of Timing. 2015;15:416\u0026ndash;27. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2174/1566523215666150515145255\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eWeber EL, Cannon PM. Promoter choice for retroviral vectors: transcriptional strength versus trans-activation potential 2007;18:849\u0026ndash;60. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1089/hum.2007.067\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eMontini E, Cesana D, Schmidt M, Sanvito F, Bartholomae CC, Ranzani M et al. The genotoxic potential of retroviral vectors is strongly modulated by vector design and integration site selection in a mouse model of HSC gene therapy. 2009;119:964\u0026ndash;75.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eAlhaque S, Themis M, Rashidi H. Three-dimensional cell culture: from evolution to revolution 2018;373:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1098/rstb.2017.0216\u003c/span\u003e\u003c/span\u003e. https://doi.org/20170216 [pii].\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eMontini E, Cesana D, Schmidt M, Sanvito F, Ponzoni M, Bartholomae C et al. Hematopoietic stem cell gene transfer in a tumor-prone mouse model uncovers low genotoxicity of lentiviral vector integration 2006;24:687\u0026ndash;96.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eBiffi A, Montini E, Lorioli L, Cesani M, Fumagalli F, Plati T, et al. Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy. 2013;341:1233158. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.1233158\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eAiuti A, Cossu G, de Felipe P, Galli MC, Narayanan G, Renner M et al. The committee for advanced therapies\u0026rsquo; of the European Medicines Agency reflection paper on management of clinical risks deriving from insertional mutagenesis 2013;24:47\u0026ndash;54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1089/humc.2013.119\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eCondiotti R, Goldenberg D, Giladi H, Schnitzer-Perlman T, Waddington SN, Buckley SM et al. Transduction of fetal mice with a feline lentiviral vector induces liver tumors which exhibit an E2F activation signature 2014;22:59\u0026ndash;68. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/mt.2013.193\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eHacein-Bey-Abina S, Garrigue A, Wang GP, Soulier J, Lim A, Morillon E et al. Insertional oncogenesis in 4 patients after retrovirus-mediated gene therapy of SCID-X1 2008;118:3132\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1172/JCI35700\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eRamaswamy S, Tonnu N, Menon T, Lewis BM, Green KT, Wampler D et al. Autologous and Heterologous Cell Therapy for Hemophilia B toward Functional Restoration of Factor IX. 2018;23:1565\u0026ndash;80. https://doi.org/S2211-1247(18)30497-2 [pii].\u003c/span\u003e\u003c/li\u003e\n \u003cli\u003e\u003cspan\u003eCesana D, Ranzani M, Volpin M, Bartholomae C, Duros C, Artus A, et al. Uncovering and dissecting the genotoxicity of self-inactivating lentiviral vectors in vivo 2014;22:774\u0026ndash;85.https://doi.org/10.1038/mt.2014.3.\u003c/span\u003e\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"gene-therapy","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"gt","sideBox":"Learn more about [Gene Therapy](http://www.nature.com/gt/)","snPcode":"41434","submissionUrl":"https://mts-gt.nature.com/cgi-bin/main.plex","title":"Gene Therapy","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-3837253/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3837253/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLentivirus vectors are effective for treatment of genetic disease and cancer, however, vector related insertional mutagenesis related genotoxicity is of concern and currently available safety models are not reliably predictive of safety in humans. We have developed \u003csup\u003eh\u003c/sup\u003eInGeTox as the first human \u003cem\u003ein vitro\u003c/em\u003e platform that uses induced pluripotent stem cells and their hepatocyte like derivatives to further understand LV host interaction for vector safety evaluation and design. To characterise LV for genotoxic association, we used LTR and SIN configuration LV infected cells for a multi-omics analysis on data that included LV integration sites in cancer genes and their associated differential expression, clonal tracking of IS, novel vector/host fusion transcripts and methylated cancer genes with altered gene expression after infection. We present \u003csup\u003eh\u003c/sup\u003eInGeTox as a useful pre-clinical tool to identify lentivirus contributory factors mediating genotoxicity to use for improving LV design to provide gene therapy.\u003c/p\u003e","manuscriptTitle":"hInGeTox: A human-based in vitro platform to evaluate lentivirus contribution to genotoxicity","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-11 14:31:30","doi":"10.21203/rs.3.rs-3837253/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"revise","date":"2024-02-13T09:46:45+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"This content is not available.","date":"2024-02-11T14:42:17+00:00","index":1,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2024-01-15T08:12:47+00:00","index":2,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2024-01-10T14:58:11+00:00","index":3,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2024-01-10T08:14:53+00:00","index":3,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2024-01-10T07:45:15+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2024-01-10T07:26:46+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2024-01-10T07:08:49+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-01-09T09:41:02+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-01-09T09:40:49+00:00","index":"","fulltext":""},{"type":"submitted","content":"Gene Therapy","date":"2024-01-05T12:39:53+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"gene-therapy","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"gt","sideBox":"Learn more about [Gene Therapy](http://www.nature.com/gt/)","snPcode":"41434","submissionUrl":"https://mts-gt.nature.com/cgi-bin/main.plex","title":"Gene Therapy","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"2186231e-e599-4bbe-a316-37ec2e97f16e","owner":[],"postedDate":"January 11th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":28087623,"name":"Biological sciences/Stem cells/Stem-cell differentiation"},{"id":28087624,"name":"Biological sciences/Molecular biology"},{"id":28087625,"name":"Biological sciences/Cell biology"},{"id":28087626,"name":"Biological sciences/Cancer"}],"tags":[],"updatedAt":"2025-07-16T07:10:08+00:00","versionOfRecord":{"articleIdentity":"rs-3837253","link":"https://doi.org/10.1038/s41434-025-00550-9","journal":{"identity":"gene-therapy","isVorOnly":false,"title":"Gene Therapy"},"publishedOn":"2025-07-15 04:00:00","publishedOnDateReadable":"July 15th, 2025"},"versionCreatedAt":"2024-01-11 14:31:30","video":"","vorDoi":"10.1038/s41434-025-00550-9","vorDoiUrl":"https://doi.org/10.1038/s41434-025-00550-9","workflowStages":[]},"version":"v1","identity":"rs-3837253","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3837253","identity":"rs-3837253","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}