How to discuss gene therapy for haemophilia? A patient and physician perspective

Abstract Gene therapy has the potential to revolutionise treatment for patients with haemophilia and is close to entering clinical practice. While factor concentrates have improved outcomes, individuals still face a lifetime of injections, pain, progressive joint damage, the potential for inhibitor development and impaired quality of life. Recently published studies in adeno‐associated viral (AAV) vector‐mediated gene therapy have demonstrated improvement in endogenous factor levels over sustained periods, significant reduction in annualised bleed rates, lower exogenous factor usage and thus far a positive safety profile. In making the shared decision to proceed with gene therapy for haemophilia, physicians should make it clear that research is ongoing and that there are remaining evidence gaps, such as long‐term safety profiles and duration of treatment effect. The eligibility criteria for gene therapy trials mean that key patient groups may be excluded, eg children/adolescents, those with liver or kidney dysfunction and those with a prior history of factor inhibitors or pre‐existing neutralising AAV antibodies. Gene therapy offers a life‐changing opportunity for patients to reduce their bleeding risk while also reducing or abrogating the need for exogenous factor administration. Given the expanding evidence base, both physicians and patients will need sources of clear and reliable information to be able to discuss and judge the risks and benefits of treatment.

factor concentrates (CFCs) prepared from pooled plasma with HIV and hepatitis viruses, however, blighted many lives. 2 Safety improved with the introduction of effective viral inactivation measures followed by recombinant DNA technology in the 1980s. 2,3 Since then, CFCs have evolved with the development of extended half-life (EHL) versions that improve the QoL by reducing dosing frequency 4,5 and increase protection by enabling higher trough levels. Despite this, haemophilia continues to impose multiple complications including joint damage, functional impairment, acute and chronic pain, mental health/anxiety issues, reduced QoL, as well as impaired social participation, reduced educational attainment and diminished work productivity (Table 1).

| Unmet needs in haemophilia treatment
The limitations of current options highlight the need for less burdensome and more cost-effective treatment that limits the longer-term complications experienced by PWH (Table 2). Preliminary evidence in haemophilia A and B indicates that GT may offer the potential to address these limitations.

| WHAT IS G ENE THER APY
GT refers to the treatment of a disease through introducing a functional copy of a disease-causing gene, inactivation of the gene's effects through addition of novel or modified genes, or editing of a host gene to correct a congenital mutation. 6 GT strategies that are currently approved, or approaching approval, are largely aimed at treating diseases that are caused by a defect in a single gene, such as haemophilia, lipid disorders, retinal diseases and spinal muscular atrophy. The most common way to introduce therapeutic genes is via a viral vector. Unlike earlier approaches using adenoviral and retroviral vectors which insert the transgene into the genome of the host, 7  Joint damage Can result in chronic pain, disability and joint deformity at an early age 1,54,55 Poor health-related quality of life Closely linked to the extent of joint damage 54 Functional impairment More likely to suffer from arthropathy/arthritis, more likely to require knee/hip replacement compared with the general population. 1,56 Poor mobility, self-care issues, and inability to perform usual daily activities 57,58 Social isolation Inability to participate in social or sporting activities 59 Pain Higher pain levels and functional impairment associated with anxiety, depression and unemployment. 60,61 Pain/discomfort is an area where most individuals report experiencing 'extreme' issues. 54 Individuals may experience anger and frustration due to the pain, inconvenience and erratic nature of bleeds 62 Psychological Anxiety/depression are the areas where most individuals report experiencing 'extreme' issues 54 Personal productivity Adverse impact on educational achievement and work productivity due to absence and difficulties due to functional impairments and pain 57,63,64

Unmet need Impact
Treatment convenience Lifetime treatment, frequent injections. 65,66 Prophylaxis is timeconsuming, contributing to poor adherence 67

Joint damage despite factor prophylaxis
Indicates that prophylaxis is failing to control some subclinical bleeding 55,68 Inhibitor development Occurs in approximately one-third of patients with severe haemophilia A and <5% of those with haemophilia B and increases treatment cost and morbidity risks 69 High lifetime-treatment costs High factor concentrate costs, 1,70-72 means availability of factor prophylaxis is limited in many countries Pain See Table 1 Limits on activity and social participation See  18 These levels were associated with a reduction in median annualised bleed rate from 16 to 1 with cessation in the need for FVIII replacement by week 22 (Table 3A). 18 .
Vector-mediated GT in haemophilia B has demonstrated that it is possible to convert patients with severe disease (<1% FIX activity) to a 'mild' phenotype, that is endogenous FIX levels of 5% or more with vectors that carry wild-type FIX such as AMT-060 or scAAV2/8-LP1-hFIXc (Table 3B). [20][21][22][23] In addition, GT utilising wild-type FIX is associated with the cessation of factor prophylaxis in most participants, the reduction in exogenous factor usage by 73%-96% and a reduction in annualised bleed rates of between 70% and 94% in those groups who achieved mean FIX activity >5%. 20,23 In order to increase FIX expression, several groups have used other variants, such as the naturally occurring FIX Padua variant (eg SPK-9001 and AMT-061) and a variant with a novel lysine to arginine substitution at position 301 (FLT180a), which enhance FIX activity (Table 3C). With these approaches, FIX activities in the range of 30% to >40% have been reported along with reductions in annualised bleeds and exogenous FIX use of approximately 90% to 100%. In addition, these variants may allow a lower dose of GT to be used, which may be useful if vector dose is a factor in the development of capsid-specific immune responses.

| Safety
The safety profile of AAV vectors reflects the fact they are related to naturally occurring AAV, which are generally non-pathogenic in humans. As has been discussed, recombinant AAV only rarely integrates into host DNA, 8 (Table 3), which has also been observed in previous GT trials utilising intramuscular injection. 25 While ALT elevations are not a safety issue per se, as these events were generally asymptomatic and were treated with a course of corticosteroids, in some cases they have been associated with a reduction in factor activity (Table 3). 18,[20][21][22][23] However ALT elevations, along with worse than expected FIX activity, 26 (Table 3B). 18 Long-term safety is uncertain as the length of follow-up in published studies is generally from 1 year up to a maximum of eight years. 28,29 There are, however, positive safety reports from longer-term follow-up in animals. 30 .   Another advance in treatment has been the development of humanised bispecific antibody technology, which by binding to both activated FIX and factor X can mimic the action of FVIII (emicizumab).

| Unmet needs
Subcutaneous (SC) emicizumab therapy for FVIII deficiency with and without inhibitors has been licenced by the FDA, 32 which will offer a degree of freedom from fluctuating factor levels by conferring a constant level of protection while also removing the burden of IV infusion. 33,34 The level of protection conferred appears to be in the range which will prevent most bleeds, but does not confer a normal or nearnormal level of protection, so treatment for breakthrough bleeds and surgery with FVIII clotting factor will continue to be required.
Expectations of GT have changed significantly over the previous five years as we have seen FIX expression increase from a modest 4.4%-7% in early trials [20][21][22] to 33% more recently, 23 with the current hope being sustained expression of factor level in the normal range. Chronic pain impacts most PWH due to a combination of target joints, pre-existing haemophilic arthropathy and subclinical bleeds. With severe bleeding phenotype. c 3 and 11 bleeds pre-and post-therapy, respectively. d SAE was defined as any untoward medical occurrence that at any dose: results in death; is life-threatening; requires in-patient hospitalisation or prolongation of existing hospitalisation; results in persistent or significant disability or incapacity; is a congenital anomaly or birth defect; or is judged medically important by the investigator.
e The severity of AEs was defined as: Mild: Awareness of symptoms, sign, illness or event that is easily tolerated; Moderate: Discomfort sufficient to cause interference with usual activity; or Severe: Incapacitating with inability to work or undertake further normal activities. These tables included trials that were listed as 'Active' on https ://clini caltr ials.gov on the 18 December 2018. Therefore, trials that were listed as 'Terminated' at that time, or those that were not listed on ClinicalTrial.gov were not included. The trial of SB-FIX, a zinc finger nuclease that is delivered by an AAV vector and which inserts a functional FIX gene into hepatocytes, was not included in the table.

TA B L E 3 (Continued)
Gene therapy with a factor expression in the normal range would free PWH from their mental burden and may lead to a real reduction in the levels of anxiety and depression. 36 With greater levels of protection, the frequency of bleeds should decrease even in the presence of higher levels of physical activity. Freedom to carry out normal everyday activities, taken for granted by those without haemophilia, such as walking, running, cycling, swimming and potentially riskier sports participation would all become more attainable.

| Efficacy
From a physician perspective, it will be key to manage patient expec-   29,38 Patients should also be made aware of the fact that the apparent lack of clinically relevant integration of the AAV vector also means that any benefits from GT will not be passed on to children. Therefore, following the initial meeting with their physician, individuals should write down any questions they have and ensure they are answered. They should decide what outcomes they would consider to be acceptable, in terms of factor activity, duration of factor expression and the potential level of bleed reduction, while appreciating that there are still uncertainties in terms of the level and duration of factor expression.
Given that GT is still in the investigational stage, however, individuals should be prepared for the possibility of a poor outcome (such as low expression, no expression or early loss of expression).

| Safety
GT is a relatively new technology that is starting to enter clinical A key worry is GT safety, particularly due to the serious safety concerns in the early trials using integrating vectors. 40 43 In other animal studies, sperm cells appear to be refractory to AAV transduction, lessening the risks of vertical transmission. 44 However, while the risk of thirdparty infection is limited, physicians should recommend barrier contraception for up to 12 months as a precaution.
From a patient perspective concerns may include the risk of mutagenesis due to vectors insertional events (Table 4). Other concerns may include vector shedding and the risk of infecting family members and close contacts. Patients may also be worried about whether GT may increase the risk of inhibitor induction. There may also be anxiety related to potential trade-offs between increasing the vector dose, the level of factor activity that can be achieved and safety.

| Questions regarding trial participation
In terms of the physician perspective, setting patient expectations, both for clinical trials and ultimately for gene therapy as an approved treatment option for haemophilia, will be important, as some PWH will not qualify for treatment. The standard inclusion/exclusion criteria employed in clinical trials to date have limited participation to adult patients with severe or moderately severe haemophilia, 33,36,37 with exposure to factor treatment for defined periods of time (≥50 days minimum in published trials), and for most trials, normal liver and kidney function including absence of liver fibrosis. Trials generally exclude those with inhibitors, which would include approximately one-third of people with haemophilia A. All of the published lifted that exclusion criterion due to lack of evidence for associated reduced efficacy or immune responses due to pre-existing low-titre neutralising antibodies to AAV5. 45 Exclusion of key populations such children and adolescents, women with haemophilia, 46 and those with a history of inhibitors to factor replacement is consistent across Phase 1-3 haemophilia GT trials, but a GT trial in people with haemophilia A and inhibitors has been announced. 47 If GT does become available in children and adolescents, there may be ethical questions in terms of gaining informed consent. There has been an initial report of an adverse event potentially associated with a concomitant anti-HIV drug, 48 so if these kind of interactions are confirmed, care may need to be taken in treating some individuals with GT.
Beyond these factors, there are currently no characteristics that can be used to identify those who are likely to respond better or worse to GT, although this will likely become clearer as the evidence base grows. Importantly, the majority of newly approved treatments typically will likely gain indications specific to the populations studied in the clinical trials; thus, it will be important to set expectations for excluded populations on timeline of treatment availability and the necessity to gather evidence in these groups once GT becomes more established.
For patients who participate in clinical trials, there is a practical burden of frequent study visits in the short-term, as well as longterm (5 years on average) follow-up, which may be underestimated by potential participants. A typical trial may require weekly or up to tri-weekly visits in the first 6 months, monthly or quarterly visits up to 12 months and quarterly or bi-annual visits for the remaining follow-up despite participants potentially having normal or near-normal factor levels. Participants will generally need to record factor use and bleeds using an e-diary or similar approach, which will then be reviewed at each visit. During visits, body fluid samples will be required for vector shedding analysis; blood will be required for determining factor activity, inhibitors, liver enzymes, anti-factor antibodies, or AAV antibodies/neutralising antibodies, inflammatory markers, T-cell responses and other trial outcomes. Given that trials usually take place in specialist centres, study-related visits will likely involve travel and TA B L E 4 Typical questions PWH may have before deciding to enter a GT trial

Question
Which trial should I participate in?
What are the results, if any, from earlier phases of the trial?
What is the reputation of the trial team?
What vector is being used and what is the prevalence of pre-existing vector antibodies?
Will pre-existing antibodies automatically rule out trial participation or have strategies been developed to address this issue?
What vector dose is being infused and what is the anticipated range of factor expression? Is a higher vector dose worthwhile if the objective is higher factor expression?
Am I comfortable taking a prophylactic course of steroids if that is part of the protocol?
What duration of transgene expression is expected? What is the lower limit of duration of expression which would be persuasive to you in agreeing to participate in a trial or treatment? While lifetime expression is desirable, would I agree to treatment if expression was for 10 y? What about 1 y?
What is the potential for integration with an AAV vector? What is the likelihood of insertional mutagenesis and the risk of developing cancer in the future?
Is there a risk of inhibitor development?
Am I comfortable with the degree of monitoring and commitment required, especially in the first year, and with annual follow-up for up to 15 y? potentially overnight accommodation. Other potential logistical issues include limitations on travel, the need for abstinence from alcohol in some trials, and potentially limitations on physical exercise to avoid muscle-related transaminitis elevations. Even when GT is approved, it is likely that treated patients will need additional follow-up to confirm sustainable transgene expression, clinical efficacy and safety.
From a patient perspective, with so many trials recruiting and ongoing, it is worthwhile to review the key outcomes and the types of questions PWH might have ( Table 4). As discussed in the previous section, key outcomes for PWH include factor level, duration of factor expression, reduction in chronic pain, healthcare resource utilisation, impact on mental health and bleed frequency. 35 Therefore, it is likely that individuals will choose to participate in trials based on the factors which are most important to them.
In an ideal scenario, the consent process should involve an independent person separate from the haemophilia treatment centre team who will clearly set out the potential risks to ensure full informed consent; however, this is not a requirement for consent in most trials. When introducing trials to PWH, it makes sense for initial communications to take place in small groups as such meetings often develop into discussion forums, which prompts questions that some attendees may not have thought about. Individuals should also familiarise themselves with relevant information from their national, regional or global haemophilia patient organisations. As GT becomes more established, it will be important for these organisations to provide patient friendly educational materials including video and to facilitate education sessions/lectures/conferences for their members.
PWH should familiarise themselves with the trial protocol.

| Sources of information on gene therapy
Given the gaps in the evidence base, it will be important to embrace a shared decision-making approach. 49

| The future of haemophilia gene therapy
It is an exciting time in GT, when the long-heralded promise is starting to yield treatments that are entering the clinic. Most ongoing haemophilia GT trials utilise the process of gene addition, that is, infusing a healthy copy of a clotting factor gene (VIII or IX) via an AAV vector into a patient without altering their own DNA. There is at least one trial exploring gene editing for haemophilia, a process by which a zinc finger nuclease (sometimes referred to as 'DNA scissors') is used to insert the therapeutic transgene into a so-called safe harbour or area with high-transcriptional activity. 51 Other types of GT are also being explored and it will be important for physicians to educate patients and families on the different options and discuss which approaches meet individual needs. It is likely that when GT initially become available that they will be prescribed through a limited number of expert centres, which should be fully able to discuss and educate patients about treatment options.
As haemophilia GT enters the clinic and is subject to surveillance, the longer-term safety and efficacy profiles will become clearer.
Theoretically, at least, it would make sense to initiate GT before joint damage is manifest, which may start between 1 and 2 years of age. 52 While current trials in haemophilia are confined to adults, as the safety profile of GTs becomes more established, it will be important to include adolescents and children, so that treatments can be opened up to this important population. Treatment in younger populations, however, may present additional challenges in terms of the potential impacts of hormonal and developmental changes as well as liver growth on long-term GT effectiveness. The development of inhibitors is a major problem that limits clotting factor treatment options and efficacy, so it will also be of great interest whether GT, either alone or when combined with other approaches such as immune tolerance induction, can benefit such patients, and the future results from the recently announced GT trial in haemophilia A with inhibitors will be awaited with interest. 47 A number of different approaches have been studied in animals including classical immune tolerance induction with repeated exposure to antigens to therapies specifically targeting T or B cells. 53 While the treatment of haemophilia has improved, it is costly and burdensome. Despite CFC, haemophilia still has major adverse impacts on the QoL of PWH including functional impairment, pain, and psychosocial issues. Longer-term evidence is needed to confirm whether haemophilia GT offers durable efficacy precluding the need for factor replacement. Experience from clinical trials so far suggests that it offers a life-changing opportunity for PWH to reduce their bleeding risk while also reducing or abrogating the need for exogenous factor administration.