Clinical Consults™: Hemophilia Perspectives: Current Insights on the Evolution of Gene Therapy


The following activity features highlights and faculty commentary from a satellite CME Dinner Symposium preceding the 60th American Society of Hematology Annual Meeting & Exposition (ASH) that address issues on the evolution of gene therapy in hemophilia.

Please answer all Knowledge Check questions found in the Introduction in order to move through the activity


Hemophilia is the most common severe inherited bleeding disorder.1 Hemophilia A and B are X chromosome–linked recessive diseases caused by a lack or deficiency of  coagulation cascade factor VIII (FVIII) or factor IX (FIX), respectively, which are required for thrombin generation.1,2 Both types of hemophilia cause bleeding into muscles and joints, which may be life-threatening, and can lead to joint damage and disability.1-3 Severity is dependent on plasma levels of FVIII or FIX activity: severe: <1%; moderate: 1%-5%; mild: >5%-50%.2 About 50% of people with hemophilia (PWH) have a severe form, and all racial and ethnic groups are affected.4

Hemophilia A and B are caused by genetic changes in the F8 and F9 genes, respectively, which are located on the X chromosome (Figure 1).1-3 Though hemophilia is inherited,5 one-third of PWH do not have a known family history of hemophilia (spontaneous mutation).

Figure 1. F8 and F9 are the genes that code for FVIII and FIX, respectively

Figure 1. F8 and F9 are the genes that code for FVIII and FIX, respectively

Among PWH, there is a large degree of allelic heterogeneity in the F8 and F9 genes.2

  • F8 mutations occur in all exons and intron/exon junctions. The majority of F8 variants involve point mutations (66.5%), and mutations occur most often in the A domains of FVIII. Intron 22 and Intron 1 are involved in pathologic inversions, caused by recombination with homologous regions outside of the F8 gene. Intron 22 and Intron 1 inversions are found in 45% and 1%-2% of PWH with severe hemophilia A, respectively.2
  • F9 point mutations are found through the gene, including the promoter and 3’ UTR, with the majority occurring within the serine protease domain (56%). Missense mutations are the most common single nucleotide variants across all disease severities. There is no counterpart of the F8 intron 22 and 1 inversions in hemophilia B.2

Although conventional diagnosis of hemophilia involves the more rapid laboratory measurement of circulating FVIII and FIX activity,2 identification of an individual’s causative mutation is important for several reasons6:

  • To inform clinical decisions
    • Genotypes correlate with the severity of bleeding7
    • Genotypes are a major determinant of risk for developing alloantibodies (inhibitors) to factor replacement, an important limitation in current treatment8-10
  • To determine familial hemophilia gene carrier status
  • Prenatal diagnosis in the fetus in a woman who is a carrier
  • To facilitate research in this rare disease.

Methods of molecular diagnosis used for hemophilia continue to evolve. A proposed algorithm for molecular diagnosis using next generation sequencing (NGS) is shown in Figure 2.11 However, other techniques such as multiplex ligation-dependent probe amplification (MLPA), array comparative genomic hybridization (aCGH), and mutation scanning may be used to identify F8 and/or F9 mutations.2 Only for severe hemophilia A, intron 22 and intron 1 inversion testing is done prior to sequencing. Fortunately, current molecular diagnostic technologies can detect 91%-99% of causative mutations in hemophilia A and B.2  

Figure 2. Proposed molecular diagnostic algorithm using next-generation sequencing (NGS)11

Figure 2. Proposed molecular diagnostic algorithm using next-generation sequencing

Adapted with permission from Bastida JM, et al. Thromb Haemost. 2017;117:66-74.

What strategies are used for molecular diagnosis in PWH?


Current Treatment

Management of hemophilia has changed dramatically with the availability of recombinant factor replacement products, extended half-life products, and non-factor therapies.12-14 Currently, the standard of care for severe hemophilia is prophylaxis with infusions of recombinant FVIII or FIX given twice a week for hemophilia A or once every 1 to 2 weeks for hemophilia B.13,14 For hemophilia A, the standard is in flux with the recent availability of a non-factor product, emicizumab─a bispecific monoclonal antibody─which is given once-weekly or every other week.15 As a group, PWH have a normal life span, and with current treatment, living with minimal bleeding is a reality.14 However, current treatments are not curative.

The high frequency of infusions and high cost of factor replacement products remain significant burdens to patients. In addition, PWH still face continuous bleeding risk and the potential of developing anti-factor alloantibodies (inhibitors).14 Access to factor replacement products is limited in some global regions and countries.

These unmet needs underpin the goals of gene therapy for PWH─a one-time therapeutic intervention that is potentially curative─providing a sustained level of clotting factor, preventing bleeding, and eliminating the need for lifelong factor replacement prophylaxis.16

Acquired hemophilia A should be considered if the actively bleeding patient has had recent onset or acute bleeding with no personal or family history of bleeding diatheses, particularly if the patient has cancer, even though bleeding can be the initial manifestation of a malignancy.1,3

Gene Therapy in Hemophilia

Gene therapy has the potential to address unmet needs in hemophilia with a one-time treatment producing sustained levels of FVIII or FIX to prevent bleeding.14 Figure 3 illustrates how gene therapy might work in a person with hemophilia.16

Figure 3. How might gene therapy work?16

Figure 3. How might gene therapy work?

Images adapted from NEJM Quick Take in George LA, et al. N Engl J Med. 2017;377:2215-2227. Available at: Accessed January 22, 2019.

How do we think gene therapy works?


Michael Recht, MD, PhD


Gerald Dolan, MD

The F8 and F9 genes are engineered and packaged in a viral vector for administration, with the goal of reversing the effects of deficient or missing clotting factor.17 There is no need to raise FVIII or FIX activity levels to 100%, as lower levels appear to be sufficient to prevent bleeding. Importantly, long-term efficacy and safety of gene therapies can be readily assessed in this population.

Adenovirus-associated virus (AAV) is the preferred vector because it can be engineered to recognize and deliver transgenes effectively to different tissues, such as the liver.17  AAV is a nonenveloped parvovirus that is widespread in humans but not associated with disease. A limitation is that some people exhibit preexisting immunity to AAVs, and these individuals are currently excluded from clinical studies. For hemophilia gene therapies, AAV5 and AAV8 are the most commonly used subtypes. Technological advances have made AAV and other vectors more efficient in various gene therapy applications.

Gene therapies are engineered to replace the genome of an AAV vector with an expression cassette (engineered F8 or F9) transgene.17

  • Hemophilia A: F8 is a large gene (186 kb) that encodes a 9-kb mRNA transcript.2 B-domain deleted (BDD) FVIII is smaller than full-length FVIII while retaining full activity and half-life.18 It is used for some rFVIII replacement products. It is also useful for gene therapy because its smaller transgene (4.4-kb transcript) can be packaged into rAAV cassettes, which are limited to 4.7 kb of space.17
  • Hemophilia B: The F9 gene is 34 kb in length and encodes a 2.8-kb mRNA transcript.2 The F9 gene is smaller than F8 and fits easily into AAV cassettes, which spurred earlier development of gene therapy for hemophilia B than for hemophilia A.17

The Padua variant is a naturally occurring F9 mutation coding for FIX that produces an 8-fold higher specific activity.19 The Padua variant is increasingly being used as the F9 transgene to optimize FIX activity levels produced by gene therapy.   

Clinical Review

Recent landmark studies have shown the groundbreaking potential of gene therapy in hemophilia─3 studies in hemophilia B and 1 study in hemophilia A (Figure 4).16,20-23

Figure 4. Breakthrough studies showing the promise of gene therapy for hemophilia16,20-23

Figure 4. Breakthrough studies showing the promise of gene therapy for hemophilia

1. Nathwani AC, et al. N Engl J Med. 2011;365:2357-2365; 2. Nathwani AC, et al. N Engl J Med. 2014;371:1994-2004; 3. Simioni P, et al. N Engl J Med. 2009;361:1671-1675; 4. George LA, et al. N Engl J Med. 2017;377:2215-2227; 5. Miesbach W, et al. Blood. 2018;131:1022-1031; 6. Rangarajan S, et al. N Engl J Med. 2017;377:2519-2530.

The first successful hemophilia gene therapy study was in hemophilia B (Figure 5).20,21 The Nathwani study (2011) demonstrated that AAV could successfully deliver F9 via peripheral administration. The transgene induced good expression of FIX in patients with hemophilia B, with significantly reduced use of FIX concentrate and bleeding episodes. In the high-dose group, a consistent increase in FIX levels (mean, 5.1%) was observed in all 6 patients. Transient elevations in liver enzymes occurred in some patients.20,21

Figure 5. Results of the first successful gene transfer study in hemophilia B20,21

Figure 5. Results of the first successful gene transfer study in hemophilia B

Reprinted with permission from Nathwani AC, et al. N Engl J Med. 2014;371:1994-2004.

In 2017, George et al reported production of sustained FIX levels using administration of an engineered AAV2 vector with the Padua IX variant as the transgene (fidanacogene elaparvovec [formerly SPK-9001]) (Figure 6).16 The study involved 10 patients and showed a mean steady-state FIX activity ranging from 14.3% to 76.8%. Participants 7 and 9 (not shown) experienced asymptomatic transient elevation in liver enzymes, or decline in FIX activity, potentially indicative of an immune response to the AAV capsid. Both patients received a tapering course of corticosteroid, after which their alanine aminotransferase (ALTs) returned to baseline within 1 week while FIX activity levels remained stable. SPK-9001 is currently being evaluated in a phase 3 study (NCT02484092).

Figure 6. SPK-9001 and sustained FIX activity levels in patients with hemophilia B.16

  1. Fidanacogene elaparvovec (formerly SPK-9001)
  2. Sustained FIX activity levels in patients with hemophilia B treated with SPK-9001
    Mean steady-state FIX activity: 14.3% to 76.8%

    Reprinted with permission from George LA, et al. N Engl J Med. 2017;377:2215-2227.

In another study for hemophilia B, patients were given AMT-060, an AAV5 vector with a liver-specific promoter driving expression of a codon-optimized, wild-type, human F9 gene (Figure 7).22 Sustained expression of FIX was observed, with FIX activity levels between 4.4 and 6.9 IU/dL. Going forward, AMT-061 (an AAV5 carrying a gene cassette with Padua-FIX transgene) is being studied in phase 3 (NCT03569891).

Figure 7. AMT-060 and sustained FIX levels in hemophilia B22

  1. AMT-060
  2. Sustained FIX activity levels in patients with hemophilia B treated with AMT-060

    Reprinted with permission from Miesbach W, et al. Blood. 2018;131:1022-1031.

For hemophilia A, a landmark study for gene therapy was reported in 2017.23 BMN 270 (valoctocogene roxaparvovec) uses an rAAV5 vector and a BDD-FVIII transgene (Figure 8). Administration of BMN 270 led to increased FVIII activity levels over the course of 52 weeks and FVIII activity levels that remained within the therapeutic range of 50 to 150 IU/dL for 2 years. Patients in this study experienced a 97% reduction in mean annualized bleeding rate and a 96% reduction in mean FVIII usage starting at 4 weeks after treatment (Figure 9).23 BMN 270 is currently being evaluated in 2 phase 3 studies (NCT03392974; NCT03370913).

Figure 8. BMN 270 (valoctocogene roxaparvovec) in hemophilia A23,24

  1. BMN 270 (valoctocogene roxaparvovec) rAAV5 –BDD-FVIII
  2. Sustained FVIII activity levels in patients with hemophilia A treated with BMN 270
    No FVIII activity above upper limit of normal at year 2

    Upper and lower box bounds represent 25th and 75th percentiles. The whisker lines represent the minimum and maximum values.

    Rangarajan S, et al. N Engl J Med. 2017;377:2519-30; 2. Rangarajan S, et al. WFH 2018; Abstract T-FPMED01-001.

Figure 9. Reduction in annualized bleeding rate (ABR) and FVIII usage starting from 4 weeks on BMN 27023,24

Figure 9. Reduction in annualized bleeding rate (ABR) and FVIII usage starting from 4 weeks on BMN 270

Rangarajan S, et al. N Engl J Med. 2017;377:2519-30; 2. Rangarajan S, et al. WFH 2018; Abstract T-FPMED01-001.

While short-term risks appear to be low, long-term safety is not yet established.

  • The most common complication was transient increase in liver enzymes. Transaminitis has been effectively managed with prednisolone, and in some recent studies, prednisolone was given prophylactically.16,20,23
  • The only serious adverse event was progression of preexisting chronic arthropathy in one participant in the hemophilia A gene therapy trial.23
  • There were no deaths, no development of inhibitors, no thrombosis, and no excess activation of coagulation.16
  • Vector genomes shed transiently into bodily fluids and then are cleared.16
  • Although high-titer antibodies to the AAV capsid have been observed after infusion, no significant T cell response to the transgenes were observed.16
  • To date, no genotoxic or gene-silencing events have been noted in human participants.16

Collectively, what do these landmark studies tell us about the status of gene therapy for hemophilia?

Michael Recht, MD, PhD

Click "Next" to review this patient case.