Commentary on ‘Lentivirus-mediated gene therapy for Fabry disease: 5-year end-of-study results from the Canadian FACTS trial’

Fabry disease is an X-linked lysosomal storage disorder of glycosphingolipid catabolism caused by pathogenic variants in the GLA gene encoding α-galactosidase A (α-Gal A). It is a progressive disorder with involvement of the renal, cardiac, neurologic and cerebrovascular systems, leading to reduced life expectancy.¹ α-Gal A deficiency causes lysosomal accumulation of glycosphingolipids such as globotriaosylceramide (Gb3) and globotriaosylsphingosine (lyso-Gb3) in the vascular endothelium, epithelial cells, podocytes, cardiomyocytes, mesangial cells and renal tubular cells. Currently available treatments for Fabry disease [enzyme replacement therapy (ERT) and chaperone therapy] are effective in mitigating the decline of renal and cardiac functions and increasing survival.^{2, 3} However, ERT is hampered by the enzyme's short half-life, incomplete tissue penetration, infusion-related reactions, and the development of antidrug antibodies. On the other hand, oral chaperone therapy has broad tissue distribution and penetration, but its efficacy is restricted to patients carrying specific GLA variants.⁴ Therefore, novel and more effective treatments are needed. The Fabry Disease Clinical Research and Therapeutics (FACTs) study aimed to overcome these limitations using ex vivo lentiviral vector (LV)-mediated gene transfer into autologous haematopoietic stem/progenitor cells (HSPCs) harvested from peripheral blood after mobilisation. Following non-myeloablative conditioning, genetically corrected HSPCs were infused into each patient to engraft and proliferate with all progeny cells carrying the therapeutic gene. After LV-mediated HSPC gene therapy, tissue-resident immune cells and circulating blood cells express the therapeutic gene if they are the progeny of the genetically corrected cells. The lysosomal enzyme secreted by the cells derived from genetically corrected HPSCs is then internalised by nearby and distant uncorrected cells and directed into their lysosomes for cross-correction. LV-mediated HSPC gene therapy has been used in several clinical trials including various inherited immunodeficiencies, haematological disorders, X-linked adrenoleukodystrophy, as well as other lysosomal storage disorders, such as metachromatic leukodystrophy and mucopolysaccharidosis type I.⁵

The FACTs trial involved five adult male individuals with classic Fabry disease. The interim results of the trial showed sustained reductions in Gb3 and lyso-Gb3.^{6, 7} In their follow-up article, Khan and colleagues presented longer 5-year data of the FACTs study.⁸ Importantly, no additional adverse events attributable to the gene therapy were observed and long-term α-Gal A expression along with sustained reductions in lyso-Gb3 and Gb3 continue to be observed in all study participants. Plasma and white blood cell α-Gal A activity was detected below the reference range in all except one patient who showed α-Gal A activity within the normal range for up to 5 years after gene therapy. Although the biomarker data are encouraging, the efficacy in preventing renal function loss, as evaluated by estimated glomerular filtration rate (eGFR), showed mixed results. While stabilisation of the eGFR was observed in a patient with mild renal disease at the baseline, only a mild reduction in eGFR was detected in three patients including one individual with plasma α-Gal A within the normal range, and progression of renal disease with steeper decline in the eGFR slope was observed in one patient who had Fabry-associated kidney disease at the baseline. Collectively these data and the lack of a control group prevent to draw conclusions about the efficacy of ex vivo gene therapy in halting the progression of renal disease, despite the sustained enzyme expression and improved biomarkers. Biomarkers can be reliable substitutes for clinical outcomes and their use has been proposed to improve the efficiency of clinical trials, especially in lysosomal storage disorders that require long-term evaluations.⁹ However, the correlation between lyso-Gb3 and clinical endpoints in Fabry disease is unsatisfactory¹⁰ and Fabry disease patients on ERT still lose renal function and have cardiac complications, albeit fewer and later compared to untreated patients.¹¹

Fabry disease has been associated with tissue remodelling and irreversible organ damage.^{12, 13} Patients in the FACTs trial were all adults. To prevent the decline in renal function, gene therapy may need timely intervention prior to the onset of stromal changes, fibrosis and cell death. It is also possible that ex vivo LV-mediated gene therapy might fail to achieve full cross-correction of renal cells such as podocytes, mesangial cells, or tubular cells, which contribute to the progression of the renal disease.

The FACTs trial used a reduced-intensity non-myeloablative conditioning regimen of low melphalan dose instead of the commonly used busulfan. This approach was used to improve the safety and to reduce the complexity of the transplant procedures in Fabry patients, who have considerable comorbidities that could increase the risk of transplant-related morbidity. However, gene therapy performed using a myeloablative conditioning regimen to achieve higher or supraphysiologic α-Gal A activity might provide greater disease correction. In the absence of positive selection, factors permitting high expression of the therapeutic gene, such as strong promoters or high vector copy number, might result in better disease outcomes,¹⁴ but may also increase the risk of genotoxicity. Recently, seven cases out of a total of 67 paediatric patients with cerebral adrenoleukodystrophy developed haematologic cancer (one case of acute myeloid leukaemia and six cases of myelodysplastic syndromes) several months after LV-mediated ex vivo gene therapy. All seven patients were found to have clonal LV insertions in tumour cells in or near the MECOM gene (in six patients) or the PRDM16 gene (in one patient).¹⁵ The cancer risk might be related to the strong viral promoter.¹⁵ The FACTs study used a recombinant LV with a self-inactivating LTR design delivering a human codon-optimised α-Gal A transgene under the control of the ubiquitous human elongation factor 1 alpha (EF1α) promoter. Therefore, the cancer risk might differ between LV-treated patients with Fabry disease and those with cerebral adrenoleukodystrophy. Persistent polyclonal haematopoiesis without evidence of clonal dominance in all five patients of the FACTs trial^{6, 7} is reassuring. Nevertheless, these troubling events raise the bar for ex vivo gene therapy by LV in Fabry disease, at least until the exact mechanisms underlying cancer development are unravelled and evidence of clinical benefit of the gene therapy over currently available therapies is provided.⁴

AR and NB-P wrote the manuscript.

The authors have no conflicts of interests.

Not applicable.