Searches were conducted in Medline OVID, Embase, and PubMed to identify potential eligible trials and sources of literature on 17 June 2022. Search terms used for Medline and Embase included combinations of “Fanconi anemia” (and term variations) AND “gene editing”/”gene vector”/”hematopoietic stem cell transplantation”/”hematopoietic stem cell”/”genetic therapy”. Terms were limited to English language sources only.

Inclusion criteria were randomized control trials (RCTs), clinical trials, case studies, orcase–control studies, and any participants of any age with FA having been enrolled in any type of gene therapy or editing trial from January 1999 to June 2022.

Exclusion criteria were preclinical, opinion, and review articles, or articles not published in English. Although 40 relevant preclinical studies related to the topic were screened in full text and may be discussed in the discussion part of this article, they were not statistically analyzed.

Data were collected into a structured Microsoft Excel-designed data collection form and where available the following were collected

Two authors planned to work independently to assess for risk of bias using the Cochrane Risk of Bias tool for any RCTs and the Newcastle–Ottawa quality assessment scale for non-randomizedcase–control studies; however, none were present in this review.

Homogeneity between studies was assessed to determine whether a meta-analysis could be performed. After discussion with a senior statistician from the University of Melbourne, it was decided that a qualitative analysis was more appropriate.

Publication bias was to be assessed with a funnel plot if more than 10 studies were chosen to be included in the review (not in the case of this systematic review). Language bias was a possibility as only English studies were reviewed.

The PRISMA flowchart shows the results obtained from the search criteria. From 2,100 initial records selected from a literature search of Medline Ovid, Embase, and PubMed, 1,561 were screened for title and abstract once duplicates were removed. Sixty-five full records were sought for retrieval as relevant, and all were obtained. Out of these records, eight full studies were included. Seventeen opinion pieces were excluded, and 40 preclinical pieces were excluded (although relevant preclinical data and opinion articles by experts will be involved in the discussion).

The four selected trials ran from 1999 to 2022. Overall, 22 patients were enrolled for gene therapy: 15 were treated with lentiviral-based therapy from 2016 onwards (19–21), while the other seven received retroviral-based therapy from 1999 to 2007 (22, 23). Nine had the FANCA mutation and the rest were FANCC .

The youngest patient receiving therapy in any trial was 9 months old and the oldest was 29 years old. These studies are summarised in Table 1 .

Successful engraftment of patient cells was demonstrated in only one study (19) and its interim report (20), although resistance to mitomycin C of patient cells was shown in other studies without prolonged hematological responses. The number of reinfused CD34+ cells/kg did not differ significantly between studies, although the number was around 50-fold lower than that being used in other successful gene therapy trials for severe combined immunodeficiency (24). The lower number is likely because FA patients have fewer CD34+ stem cells available for correction (11) and because of their shorter expansion time prior to transplant (25).

Viral vectors differed amongst studies, with lentiviral vectors being favored in the later trials after evidence of safety concerns with retroviral vectors.

Early studies with gammaretroviral vectors showed low levels of success in many trials, and after a patient treated with retroviral gene therapy for x-linked severe combined immunodeficiency developed T-cell acute lymphoblastic leukemia, many groups switched to lentiviral vectors (26, 27). The International Fanconi Anemia Gene Therapy Working Group (28), formed in 2010, also recommended the use of lentiviral vectors over gammaretroviral vectors, as the former could be transduced with shorter protocols, which were less damaging to fragile FA HSCs (24). Subsequent human trials have used a specific lentiviral vector that is self-inactivating and pseudo-typed with vesicular stomatitis virus glycoprotein (VSV-g). This vector is regulated by the human phosphoglycerate kinase (PGK) promoter and encodes the full-length FANCA cDNA.

Two trials have reported on the use of HSC-mobilizing medications, which allow for peripheral blood stem cell collection in FA patients. GCSF at a dose of 6μg/kg or 12μg/kg 12 hourly for days 1–8 and plerixafor at a dose of 240ug/kg 24 hourly for days 4–7 induced mobilization of CD34+ cells; these treatments were safe and effective. Some side effects were observed, such as abdominal pain, vomiting, and fever, but no serious adverse events were reported (29). These methods are beneficial in FA patients because of the reduced number and quality of CD34+ stem cells they produce (11). In both mobilization trials, one-third of the patients—notably the oldest participants—were unable to meet the CD34+ threshold for stem cell collection. This highlights the importance of collecting stem cells early in life, before bone marrow failure develops.

The first FA gene therapy study was published in 1999 (22) by Liu et al. Four FANCC patients—two male and two female, aged 11–29 years—were included in the study. Hematopoietic progenitor cells were collected via bone marrow harvest and transduced ex vivo with the G1FASvNa.52 retroviral vector carrying the FANCC gene. Cells were returned in separate infusions. Patient 1 had 6 × 10⁵ bone marrow CD34+ cells/kg reinfused in four cycles. The FANCC transgene was detectable in peripheral blood (0.01% after infusion one) and bone marrow, for 16 months post infusion. Patient 2 received 56 × 10⁶ peripheral blood CD34+ cells/kg in four infusions and had FANCC vector sequences detected in 1%–3% of peripheral blood mononuclear cells; however, all other samples were negative. Patient 3 was reinfused with a total of 20.4 × 10⁶ peripheral blood CD34+ cells/kg in three infusions and FANCC vector sequences were detectable in <0.01% peripheral blood and bone marrow mononuclear cells, initially dropping to nil after infusion three.

Despite the detection of vectors, the patients only transient improvement in blood counts post each infusion and only short-lived increases in hematopoietic colony numbers in the presence of MMC with subsequent gene therapy cycles. To help determine if the transient improvements in blood counts were related to the gene transfer, the researchers harvested bone marrow from patient 1 to review for FANCC -marked cells at 16 months post-last transfusion: none were found. Again, the researchers reinfused transduced cells 7 months later but were unable to detect evidence of FANCC-marked cells post this infusion.

The only patient to show the sustained presence of FANCC transgene was patient 4. She received 1.13 × 10⁶ bone marrow-derived CD34+ cells/kg in one infusion but was diagnosed with squamous cell carcinoma of the vulva not long after. She underwent pelvic radiation of 45Gy, which induced pancytopenia, resulting in a delay in her continuing radiation treatment. Peripheral blood samples prior to and immediately post radiation did not show any FANCC-marked cells; however, at 220 days post-infusion, as pancytopenia started to recover, the transgene was present and this was confirmed by the RNA PCR assay in peripheral blood mononuclear cells on day 414 (22). This case demonstrated for the first time that the viral expression of a therapeutic transgene results in functional correction and a proliferative advantage over untransduced bone marrow cells in a FA patient.

In 2007, Kelly et al. (23) published their evaluation of the safety and efficacy of gene transfer in seven FANCA patients at Cincinnati Children’s Hospital Medical Center (CCHMC). While all patients had stem cells collected, only three went on to the gene transfer stage (FAAGT 1001–1003) and only two patients (FAAGT 1001 and 1003) had stem cells infused. The cells were collected, pre-stimulated for 36 h, and transduced twice with a murine stem cell virus (MSCV)- FANCA retroviral vector, spending 84 h in culture. The cells were tested via exposure to mitomycin C (MMC) and PCR for MSCV- FANCA vector sequences before reinfusion. Patient FAAGT 1001 received a total of 4.5 × 10 ⁵ cells nucleated cells/kg and patient FAAGT 1003 received 3.5 × 10 ⁵ total nucleated cells/kg. FAAGT 1002 had poor cell yield post cryopreservation and poor cell numbers post-transduction so could not be reinfused.

Post-transplant, the reinfusion was well tolerated, but in subsequent blood and bone marrow monitoring patient FAAGT 1001 did not have detectable provirus in blood or bone marrow at any time. In contrast, patient FAAGT 1003 had one to five copies of provirus/10⁵ detected in nucleated peripheral blood cells at 2 and 4 weeks post-infusion but transduced cells were never detected in bone marrow. Both patients had a transient increase in hemoglobin and platelet counts, but no molecular evidence of gene-corrected cells in vivo. No engraftment of transduced gene-corrected cells was shown.

In 2016, an abstract (21) was presented by Jennifer Adair at the First International Fanconi Anemia Working Group Meeting (30) of a Phase I clinical trial (NCT01331018) describing two male patients in Seattle with FANCA mutations who received lentiviral gene therapy. Both patients had low or declining platelets and neutrophils at the time of cell collection and transfusion. No conditioning protocol was included. The first patient was 22 years old and received 9.4 × 10⁶ CD34+ cells from the bone marrow after purification without mobilization. Cells were transduced with a lentiviral vector carrying the FANCA gene in the presence of N-acetylcysteine, resulting in a vector copy number of 0.33 per cell. The second patient was a 10-year-old male who received 30.6 × 10⁶ CD34+ cells from bone marrow. Notably, while the first patient’s cells underwent purification, it resulted in cellular loss, possibly due to low CD34 expression. Subsequently, the second patient underwent only red cell depletion, which led to a higher vector copy number in the transfused product, and higher CFC. This finding indicated an important consideration for FA patients: non-purified cells could more efficiently survive gene transfer, which had also been suggested in previous preclinical studies (31). The transplant was reportedly well tolerated by both patients, although while restoration of MMC resistance was shown in vitro, neither had persistent transduced cells post-infusion. Again, no engraftment of transduced gene-corrected cells was able to be shown in this trial.

The most recently published study investigating gene therapy in FA was published in 2019 by Río et al. (19). This was the first gene therapy trial to show the successful engraftment of gene-corrected cells. In this clinical trial (NCT03157804 FANCOLEN-1), which was part of the larger EUROFANCOLEN collective, four male FANCA patients in Spain underwent mobilized stem cell collection with GCSF 12ug/kg twice daily for 6–7 days and plerixafor 240ug/kg daily for 2–3 days. Cells from patients FA-02002 and FA-02004 were cryopreserved for 20 months and 22 months, respectively, until evidence of bone marrow failure necessitated reinfusion. The remaining two patients received fresh autologous cells that were transduced and reinfused. They collected between 160,000 and 410,000 CD34+ × 10⁶ cells/kg for these patients and calculated an estimated corrected colony-forming cell (CFC) number by calculating the mean vector copy number per cell and analyzing the proportion of MMC-resistant CFCs present. The vector used in this trial was the phosphoglycerate kinase- FANCA -mutated woodchuck hepatitis virus post-transcriptional regulatory element (PGK- FANCA -WPRE*) lentiviral vector (19). Cells were transduced with the virus ex vivo, and then patients were reinfused with between 7,300 cells/kg and 160,000 cells/kg. Patient FA-02002 was infused with 250,000 thawed corrected CD34+ cells/kg equating to 14,000 corrected CFCs/kg. Engraftment of gene-corrected cells was reviewed by analyzing peripheral blood samples every 3 months from 6 months post-infusion. Patient FA-02002 reached 55% of gene-marked cells in peripheral blood leukocytes and 70% in B lymphocytes at 30 months. In bone marrow, they reached 37%–60% of gene-marked cells at 2 years post-infusion and 43.5% of bone marrow CD34+ cells were gene marked at that time.

Patient FA-02004 had 160,000 CD34+ cells/kg equating to 7,300 corrected CFCs/kg infused. They had marked delayed engraftment of gene-marked cells and by 24 months post-infusion the number of corrected cells in the bone marrow was lower at 15%–25%.

The two patients with freshly transduced and infused cells were FA-02006, who received higher cell doses at 410,000 CD34+ cells/kg equating to 160,000 cCFCs/kg, and had 4%–8% of gene-marked cells in the bone marrow at 12 months post-infusion, and patient FA-02005, who had between 10% and 20% of gene-marked cells at 24 months post-infusion. This was slower than engraftment expected from a conditioned bone marrow transplant but was consistently rising in all patients.

Phenotypic correction was assessed ex vivo with survival after MMC exposure and the proportion of chromosomal breaks in peripheral blood T cells after diepoxybutane (DEB) administration. All patient-derived bone marrow samples had progressive increases in MMC resistance over time, and a strong correlation with this and the proportion of corrected bone marrow CD34+ cells, suggests that the level of gene correction was responsible for the phenotypic reversion. Cells collected from patient FA-02002 24 months post-infusion showed up to 70% survival post-MMC exposure, a number similar to healthy donor values. Blood counts also stabilized in three out of four patients (FA-02002, FA-02004, and FA-02006). None of the patients’ bone marrow showed cytogenetic abnormalities via FISH and no clonal cellular populations were found. All patients were alive at the time of follow-up.

In 2021 the interim results of the Phase II study RP-102 clinical trials for nine FA-A patients aged 2–6 years were reported at ASH (20). This followed on from the publication of the successful Phase I/II FANCOLEN-1 trial, with improvements in transduction enhancers, lentiviral vectors, and modifications in cellular processing. The eligibility criteria for the trial included age over or equal to 1 year, with no HLA-matched sibling donor and at least 30 CD34+ cells/μL in the bone marrow. The team used mobilization protocols including GCSF and plerixafor. The viral vector was the previously mentioned lentiviral vector, which also carries the FANCA gene (PGK- FANCA -WPRE). No conditioning was used. The team measured the engraftment of gene-corrected cells with insertion site analysis, including peripheral blood and bone marrow vector copy number (VCN) and MMC resistance in bone marrow colony-forming units as well as improvements in pancytopenia. Engraftment was shown in six patients with peripheral blood VCN at 6 months and two out of three patients who were followed up to 12 months showed increased resistance to MMC in bone marrow colony-forming units. The team reported adverse events: one patient developed further bone marrow failure and required a transplant after an influenza B infection and another patient had a grade 2 infusion reaction to the gene therapy product (20).

In addition to the gene therapy trials, two further studies (29, 32) evaluated the combined use of GCSF and plerixafor to augment stem cell collection in FA patients for the purpose of gene therapy. The first was a multicenter study named FANCOSTEM-1, which enrolled 13 Spanish FANCA patients aged 3 to 16 years. The patients were given GCSF 12 ug/kg twice daily and plerixafor 240 ug/kg up to four doses and apheresis was initiated when CD34+ numbers exceeded 5 CD34+ cells/μL in peripheral blood. Five patients withdrew: two because they did not meet the chromosomal aberration threshold (50%) via diepoxybutane testing for distinguishing FA patients and three who took part in the accompanying gene therapy trial. Mobilization was successful in the remaining cases, and no collections had to be prematurely ceased due to adverse events (32).

The second study (29) was a Phase I/II trial named EUDRACT. Five French FANCA -associated patients were recruited, aged from 2 to 12 years, but one patient was excluded due to cytogenetic abnormalities in their bone marrow. In the remaining four patients, GCSF was used at 6 ug/kg twice daily from days 1 to 8 and plerixafor 240μg/kg each day 2 h prior to apheresis from day 5 onwards. The two youngest patients with evidence of bone marrow failure were able to meet the minimum cellular requirements of 5 × 10⁶/kg CD34+ cells for collection, but the two older patients did not.

The survival number of all studies was 100% at the time of follow-up. Both studies concluded that the combination of GCSF and plerixafor use was safe in FA patients, although increased patient age and certain hematological factors could predict the lower mobilization of CD34+ cells.