Third period achievment 





The criteria for patient pre-selection have been defined to optimize the safety of the phase I clinical trial. They include clinical, molecular, biochemical, immunological and general criteria and also take into account the nature of the anesthetic and surgical procedures. At present, a total of 58 patients affected with moderate or severe RDEB and reduced expression of type VII collagen have been identified for pre-selection. Additional patients are currently being investigated for enrollment in the pre-selection group. To date 21 patients have been enrolled in the GENEGRAFT EBGen study in the UK and in France. 

Pre-selected patients will be shortlisted according to their suitability for the clinical trial including their immunereactivity towards recombinant type VII collagen. The best 10 candidates will be tested for their keratinocyte proliferative capacity. This will allow to select 3 to 6 most suitable candidates for the trial.


Assessment of the immune response towards type VII collagen: To predict a possible immune reaction towards type VII collagen, we had previously set up highly sensitive and specific ELISA and ELISPOT assays (Pendaries et al., 2010). These assays have now been optimized and we have modified the purification process to enhance the yield of type VII collagen recovery and to shorten the procedure. The yield has been improved by 20 fold and the duration of the process has been reduced from 2 days to 8 hours. These substantial progresses made in type VII collagen production and purification are essential to ensure a better availability and lower costs for further testing. 




An optimized γ-retroviral SIN-vector expressing the type VII collagen (COL7A1) cDNA (transgene) needed to be developed, that allows efficient transduction and expression of type VII collagen in RDEB patient fibroblasts and keratinocytes. The therapeutic vector has to be generated by a packaging cell that is able to pseudotype the vector with an amphotropic envelope.

A suitable SIN-vector was characterized, termed pCMS-EF1.COL7A1.SIN1 (E890) that was able to transfer COL7A1 cDNA with very high efficiency into primary target cells. This therapeutic vector was integrated into a plasmid-based exchange construct, termed pbib-ETAR.fcvi-E890 that was used to target the viral genome into a retroviral packaging cell. This packaging cell (HA820) was developed on HEK293vecAMPHO cells (Ghani et al., 2007) and was preselected for stable high titer vector production by a “tagging” vector that provided Flp recombinase recognition sites at the proviral integration site. The targeted exchange of pbib-ETAR.fcvi-E890 was performed at this chromosomal locus. Selected clones were characterized, which at this stage of the process means that selectable markers of the tagging vector were removed and instead those of the targeting construct were implemented. Targeting-specific PCR was used to confirm precise targeting. The PCR-products were subcloned and sequence-verified. The most promising producer clones were subjected to a final clean-up step that was designed to remove almost all tagging vector sequences and thereby eliminates the possibility of any recombination between the tagging and the therapeutic vector. This polishing step was confirmed by locus specific PCR, and again confirmed by sequencing of the amplification product. The type VII collagen producer clone, HA820-E890#14.2, was expanded as a primary and secondary seed bank (PSB, SSB). SSB-cells were used to generate test batches of the therapeutic vector and to optimize the production process.


In parallel, we have developed a new functional titration method based on FACS detection of type VII collagen expression that allows for rapid and accurate quantification of transduced cells. This protocol was used to perform functional titration of the viral supernatants (pilot runs approx 1.5 Liter) and to precisely measure the level of transduction on primary keratinocytes and fibroblasts. We then tested the effect of different sequence modifications in the pCMS backbone to improve the viral titres of the supernatant.

Next, we tested whether a different internal promoter may improve expression in target cells which are null for type VII collagen expression (BeFa) or viral titer (in transient production) and a stable producer clone was developed by EUFETS. This new packaging cell line achieved the production of the SIN COL7A1 vectors with high titres (2.106 ip/ml up to 4.106 ip/ml) allowing for the use of non-concentrated and raw supernatants for further clinical use.

Two batches of raw (non-purified) viral supernatants were used to transduce primary RDEB keratinocytes and fibroblasts with a high level of efficiency, which validated the producer clone. The PSB-cells were expanded to a master cell bank (MCB) which was tested for safety (advantageous virus, RCR, etc.) and has been fully characterized and certified. 

The certified master cell bank was used to produce a clinical grade retroviral vector batch, which has subsequently been tested, certified  and released in March 2015 and is ready for use in the clinical trial.




Assessment of provirus integrity: The study of COL7A1 rearrangements was required to demonstrate the safety of the approach and for the quality control of transduced cells prior to grafting the skin equivalent onto patients. Preliminary observations led to the conclusion that rearrangements occurred during the reverse transcription step. Southern-blot and Western-blot analyses were used to characterize integrated COL7A1 provirus rearrangements. The presence of COL7A1 rearrangements could be detected at the genomic and/or protein levels, although abnormal bands at the protein level were difficult to distinguish from physiological proteolytic degradation products. For this reason, we concluded that provirus rearrangements had to be investigated at the DNA level. In addition, to estimate the frequency of these events and to try to get insights into the mechanism involved, we have set up a large scale experiment to isolate a larger number of rearrangement events. Our preliminary results pointed to the role of the viral reverse transcriptase and its template switching activity during the reverse transcription process that occurs after the infection of target cells. Analyses of the breakpoint sequences failed to reveal any rearrangement hotspot. These results have allowed to estimate the frequency of genomic rearrangement of the transgene after transduction. They are consistent with a mechanism involving the reverse-transcription step and they have led us to propose a detection method based on Southern-blot analysis of transduced bulked cell populations.


Bridging studies: In order to document the safety with the new vector that will be used in the clinic, we have designed protocols for bridging studies in mice including tumorigenicity, monitoring of the graft and biodistribution studies. The protocols and the strategy cover the gap between the first generation vector and the vector optimized during the first two periods of the project by EUFETS and INSERM. The protocols have been precisely defined and the studies will start by early 2016 with the validation of the engineered skin production process.



During this period, UNIMORE addressed the transfer and adaptation modifications required for the development of the gene-corrected skin equivalent under conditions which will be used during the GMP-production process. All the relevant information and knowledge from the research protocols developed by INSERM and UC3M were transferred to UNIMORE for the adaptation to GMP protocols. This includes the identification of the best GMP-certified reagents to achieve efficient transduction of primary keratinocytes and fibroblasts using the SIN-COL7A1 retroviral vector; analysis of the proliferative capacities of transduced keratinocytes; the production of skin equivalents using GMP-certified reagents; grafting of gene-corrected skin equivalents on immunotolerant mice and assessment of type VII collagen expression of the grafts. Each of these steps has been successfully achieved and is described in detail in section 3.3 of the report. It is important to note that these steps correspond to a development phase aiming at establishing GMP-quality protocols, but is not a GMP production phase.




In order to validate the skin equivalent production process, we needed to demonstrate our capability to produce, transport and graft large skin equivalent in condition similar to that will be used on the patients.

For that purpose we have produced large skin equivalent made of normal human keratinocytes and fibroblasts and grafted onto the facia 8 weeks-old pig and followed up during 1 month. The results demonstrated the formation of a well differentiated, pluristratified epidermis adherent to the underlying dermis. The surgeons and the INSERM team have so far validated the production of large area of skin equivalents the handling and transport procedure of large skin equivalents, the surgical procedure, the dressing (less adherence to the graft) and the glue.