Running title: AAV seroprevalence in DMD
Authors: Stella N. Nwosu1, MD, Thomas H. Vanderford2, PhD, Raj Razdan1, MS, Saila R. Upadhyayula1, MD, Han C. Phan3, MD, David Lee2 BS, Sumit Verma1, MD
Affiliations: 1Department of Pediatrics Neurologyand Neurosciences,Children’s Healthcare of Atlanta, and Emory University School of Medicine, Atlanta, Georgia 30307. 2Yerkees Division of Microbiology and Immunology, Yerkees Primate Center, Emory University, Atlanta, Georgia 30322. 3Department of Pediatric Neurology, University of Alabama, Birmingham, Alabama.
Word count: abstract 192; manuscript 1497, figures 3, table 1, supplemental tables 3
Corresponding author:
Sumit Verma, M.D.
Associate Professor of Pediatrics and Neurology,
Emory University School of Medicine,
Address: Emory Children’s Center, Room 161, 2015 Uppergate Dr., Atlanta, GA 30322.
Email: sumit.verma@emory.edu, Telephone number 404-712-4685, Fax number 404-785-4750
Ethical Publication Statement: We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
Disclosure of Conflicts of Interest: None of the authors have any conflict of interest to disclose.
Abstract
Adeno-associated virus (AAV) based gene therapies are emerging in Duchenne muscular dystrophy (DMD). Exposure to wild-type AAV can lead to development of neutralizing antibodies (Nab) and block AAV transduction, thereby limiting the delivery of AAV vector based gene therapy. Hence it is imperative to check the presence of AAV Nab in a patient who is a candidate for gene therapy. We prospectively enrolled 89 genetically confirmed boys with DMD (median age 10.5 years, 59% ambulatory, 54% on steroids) and performed AAV neutralization assays. AAV neutralizing titers were reported as the estimated dilution at which 50% of AAV infection has been inhibited (IC50). Low, moderate and high positive Nab titers were defined as <1:20, 1:20-1:240 and >1:240. AAV seroprevalence was detected in 54% (48/89) boys. Sub-type analysis showed AAV9 (75%, n=36) seroprevalence was lower compared to AAV2 (98%, n=47). Majority seropositive boys were older than 5 years of age and steroid exposure versus naïve status did affect the seroprevalence status. Although more Caucasian boys were enrolled in the study, the seroconversion rates were highest in the African American and Hispanic patients. We anticipate this data will lead to clinical preparedness for future gene therapy.
Key words: Duchenne muscular dystrophy, adeno associated virus, neutralizing antibodies titers, steroids, seroprevalence, seroconversion
Introduction:
Duchenne muscular dystrophy is an X-linked chronic progressive muscular disease caused by a mutation in the dystrophin (DMD) gene and is the most common form of muscular dystrophy in young boys. The lack of the dystrophin gene leads to a breach in the integrity of the sarcolemma, resulting in leaky muscle fibers, which attract inflammatory cells, which contribute(and contribute) to eventual muscle degeneration. Clinically, the disease manifests in childhood as progressive muscle weakness, joint contractures, scoliosis, gait difficulty, non-ambulatory status by the end of first decade and premature death from cardio-pulmonary failure in young adulthood1.
In the last decade, the gene enhancing [Food Drug Administration (FDA) approved exon 51, 53 and 45 skipping medications] that result in increased dystrophin expression have shown promise2, 3.Recently, the gene replacement trials using micro-dystrophin construct packed in adeno-associated virus (AAV) was conducted4. Although promising, there are several considerations before systemic gene delivery can be performed5, 6. Most notable of these considerations includes the presence of pre-existing neutralizing antibodies (Nab) to AAV serotypes,which will negatively impact the efficacy of the in vivo gene therapy uptake7. Furthermore, the real-world experience in spinal muscular atrophy (SMA) babies screened for intravenous gene therapy (Onasemnogene abeparvovec) showed 5.6 to 7.7% to have pre-existing AAV9 Nab, henceforth were deemed in-eligible for AAV based gene therapy8, 9.
While the AAV-based gene therapies are being developed in muscular dystrophy it may be prudent to screen the boys with DMD for AAV seroprevalence rate. On review of literature, we found a single AAV seroprevalence study in a French DMD cohort10. This study aims to determine the seroprevalence rate of AAV serotypes in a large cohort of DMD boys from the South Eastern United States.
Methods:
In this prospective study, we consecutively recruited genetically confirmed boys with DMD, followed in the muscular dystrophy clinic at the Children’s Healthcare of Atlanta. The study duration was from June 2016 to July 2020. Study participants age, genetic mutation, ethnic background, steroid use (type and duration), ambulatory status, parents education level and crowding status11 (According to the American crowding Index, crowding occurs when there is more than 1 person living in a room) were recorded. The study was approved by the local Institutional Review Board. Written informed consents and age-appropriate assents were obtained from the parents or legal guardians and patients. Spanish only speaking patients received certified Spanish translated IRB consent and assent forms.
Study procedure included one-time collection of 5 ml of blood for the anti-AAV Nab titers during the standard of care blood draw for 25 hydroxy vitamin D levels. Patient serum was separated and stored at -80 C research freezer. Study samples were transferred to Yerkees Primate Center Vaccine Biomarker Laboratory, Emory University (D.L., T.H.V.) for AAV neutralization assays. The serum was diluted in complete media (VWR; DMEM supplemented with 10% FBS, Pen/Strep, and L-glutamine) at ratios of 1:5, 1:20, 1:80, and 1:240. Serum dilutions were combined with ~ 1010-1011 AAV2 or AAV9 viral particles containing a firefly luciferase transgene under control of the CMV promoter (University of North Carolina Viral Vector Core) and incubated at 37°C in 5% CO2 for 1 hour. During incubation, confluent monolayers of HeLa cells (ATCC) were infected with viral particles of Adenovirus-5 (ATCC). After incubation, the antibody-virus mixtures were aliquoted onto the Adenovirus-5 HeLa cell monolayers and incubated at 37°C in 5% CO2. After 2 days, plates were removed from the incubator and washed once with PBS before addition of the firefly luciferase substrate (Bright-Glo assay kit, Bio-Rad). Plates were read on a BioTek luminometer within 15 minutes of substrate addition. AAV2 and 9 neutralizing titers were reported as the estimated dilution at which 50% of AAV2 and 9 infection has been inhibited (IC50). Detection of AAV2 and 9 IC50 titers >1:5 were considered seropositive. Low, moderate and high positive Nab titers were defined as <1:20, 1:20-1:240 and >1:240 dilution. Five study subjects had repeat blood draw for AAV neutralization titers during the study period. Descriptive statistics were used for data analysis.
Results:
Eighty-nine boy’s with DMD, median age of 10.5 years (Range 2-20 years), were enrolled. There were seven pairs of brothers in the study cohort. All study participants had a confirmed pathogenic mutation in the dystrophin gene (73%, n=65, out of frame deletion, 20%, n=18, point mutation and 7%, n=6, duplication). At the time of study enrollment, 59% (52/89) were ambulatory, 41% (37/89) non-ambulatory, and 54% (48/89) were receiving oral steroids (69%, n=34, deflazacort and 29%, n=14, prednisone). Ethnic background of the study participants was as follows: the Caucasians 48% (43/89), the Hispanics 29% (26/89), the African Americans 18% (16/89) and the South East Asian 4% (4/89). Parental education and the crowding status was recorded for 42% (37/89) of the study participants. Educational background was divided into 3 sub-groups; High School 37.8% (n=14), Graduate 43.2% (n=16), Post-graduate 19% (n= 7). Seventy percent (n=26) of the boy’s with DMD lived in a crowded housing structure.
Fifty-four percent (n=48/89) participants in the study were AAV seropositive (Table 1). Study participants were stratified based on age (<5, 5-12, >12 years). Ninety-three percent (83/89) participants were age 5 years and older. Boy’s younger than 5 years (n=6) were twice more likely to be seronegative (4%, n=4) then being seropositive (2%, n=2). Equal number of steroid exposed (51%, n=25) and naïve (48%, n=23) patients were seropositive. There was no statistical significant difference in the length of steroid use between the seropositive (36 ± 23.85 months, n=9) versus the seronegative (39 ± 19.41 months, n=7) patients. Seroprevalence rates were highest in the African Americans 63% (10/16) followed by the Hispanics 54% (14/26), the Caucasians 51% (22/43) and the South-East Asian Indians 50% (2/4). There was no difference in seroprevalence rates between the ambulatory (52%, n=25) versus the non-ambulatory (48%, n=23) patients. In addition, equal number of the study participants with the AAV seropositive versus the seronegative status lived in a crowded housing structure.
Seropositive patients were divided into three groups based on AAV2, 9 and 2&9 seroprevalence (Supplemental tables 1-3). Ninety-eight percent (47/48) seropositive subjects had AAV2 Nab, followed by 75% (36/48) had AAV9 Nab and 73% (35/48) had cross-reactivity between AAV 2 and 9 Nab. In the AAV2 seropositive sub-group, 49% (23/45) had high (>1:240) positive Nab titers whereas in the AAV9 seropositive sub-group, 50% (18/36) had moderate (1:20-1:240) high Nab titers (Figure 1). Sub-group analysis of the clinical variables showed 64% (23/36) AAV9 seropositive participants to be ambulatory and 49% (23/47) AAV2 seropositive participants were steroid naïve (Figure 2).
Five participants had repeat AAV Nab titers performed at the median interval of 2.5 years (Figure 3). Fifty percent (2/4) participants were seroconverted and the remaining one study subject with seropositive status remained unchanged. The two seroconverted participants were age 7 and 13 years, respectively. One of seroconverted subject was on oral steroids and other was steroid naïve.
Discussion
Inherited neuromuscular disorders are at the forefront of in vivo gene therapies5. Onasemnogene abeparvovec is one such example that is FDA approved for SMA8. This gene therapy treatment and several others in development employ AAV4, 6. AAV is a small virus (25nm) from the Parvoviridae family which naturally infects humans but are not associated with disease. AAV does not integrate in the patient genome and has low immunogenicity. Studies healthy human donors and SMA gene therapy candidates have shown that prior exposure to the wild-type AAV can lead to presence of pre-existing serum Nab titers7, 9. The AAV Nab blocks gene therapy transduction efficiency and therefore screening patients before delivering AAV vector gene therapy is mandatory. AAV micro-dystrophin is currently being developed for boy’s with DMD, however, the published studies on the AAV seroprevalence in this population is limited11. In this study we examined serum samples for AAV seroprevalence rate in a large cohort of DMD. Our goal was to screen, prepare and inform the DMD patients with regards to upcoming AAV based gene therapies.
Atlanta boys with DMD (n=89) showed 54% AAV seroprevalence rate. Leborgne et al, 2019, examined 130 French boys with DMD and found similar AAV seroprevalence rate. The French group showed 30-80% AAV seroprevalence in healthy subjects tested, however, our study lacked the control arm to know the AAV seroprevalence in age-matched healthy boy’s in the Atlanta metro. Follow-up AAV testing for limited number of participants (n=5) over the mean duration of 2.5 years showed 50% seroconversion rate. However, the French cohort had 6 year follow-up and only 11% (n=2/19) seroconversion rate.
AAV seroprevalence sub-group analysis showed significantly lower rate of AAV9 Nab titers compared to AAV2 Nab in the study population. To the best of our knowledge, this is the first study to analyze AAV9 Nab in a large DMD cohort in the United States. The distribution of Nab titers (low, moderate and high) showed high (>1:240) positive titers for AAV2 as compared to low to moderate titers (1:20-1:240) for AAV9. The lower seroprevalence and low to moderate Nab titers for AAV9 as noted in the above mentioned results is further evidence that AAV9 is a more suitable vector compared to AAV2 for gene delivery.
Study patient characteristics included majority of seropositive boys in the school-going age of 5 to 12 years. In addition, a significant number of AAV9 seropositive boys were ambulatory. Together, this make us postulate that wild-type AAV exposure likely occurred during school and or outdoor activities. Furthermore, controlling for the house crowding we did not find a difference between seropositive versus seronegative participants. Although we attempted to enroll successive DMD boys seen in the muscular dystrophy clinics, disproportionately higher Caucasian patients entered the study compared to minority population.
Steroid use did not differ between AAV seropositive versus the seronegative boys in this study, but among steroid naïve patients higher number were seropositive. Those on long-term steroids, the length of steroid treatment did not differ between the seropositive versus the seronegative group. Comparing the seropositive rates in steroid versus the steroid naïve patients there appeared to be a protective effect of steroid use, however, this observation was not statistically significant. In the large French DMD cohort, similar to our study results, steroids versus no steroids use did not differ in the seropositive and seronegative groups.
Our study results highlight that patients with DMD belonging to minority ethnic background (African American, Hispanics) were more likely to be seropositive compared to Caucasians and South East Asian DMD population. Detailed socio-economic status influencing the AAV seroconversion rate was beyond the scope of this study.
Study limitations included lack of longitudinal follow-up with repeat AAV testing for all participants and lack of the control arm. Other study limitations included limited AAV serotypes that were analyzed (only AAV 2 and AAV9), and a single-center experience.
In conclusion, our study results indicate significant (>50%) AAV seroprevalence rate in Atlanta DMD cohort. AAV9 seroprevalence was lower compared to AAV2. Majority seropositive boys were older than 5 years of age and steroid exposure versus naïve status did affect the seroprevalence status. Although more Caucasian boys were enrolled in the study, the seroconversion rates were highest in the African American and Hispanic patients. Future multi-center studies estimating AAV seroprevalence are needed.
References:
- Flanigan KM. Duchenne and Becker muscular dystrophies. Neurol Clin 2014;32:671–688.
- Mendell JR, Rodino-Klapac LR, Sahenk Z, et al. Eteplirsen for the treatment of Duchenne muscular dystrophy. Ann Neurol 2013;74:637–647.
- Frank DE, Schnell FJ, Akana C, et al. Increased dystrophin production with golodirsen in patients with Duchenne muscular dystrophy. Neurology 2020;94:e2270–e2282.
- Mendell JR, Sahenk Z, Lehman K, et al. Assessment of Systemic Delivery of rAAVrh74.MHC7.micro-dystrophin in children with Duchenne Muscular Dystrophy: A nonrandomized controlled trial. JAMA Neurol. 2020;77(9):1122-1131.
- Mendell JR, Al-Zaidy SJ, Rodino-Klapac LR, et al. Current Clinical Applications of In Vivo Gene Therapy with AAVs. Mol Ther 2021;29(2);464-488.
- Byrne BJ, Corti M, Muntoni F. Considerations for Systemic Use of Gene Therapy. Mol Ther. 2020;29(2):422-423.
- Kruzik, A, Fetahagic D, Hartlieb B, et al. Prevalence of Anti-Adeno-Associated Virus Immune Responses in International Cohorts of Healthy Donors. Prevalence of Anti-Adeno-Associated Virus Immune Responses in International Cohorts of Healthy Donors. Mol Ther Methods Clin Dev 2019;14:126-133.
- Hoy SM. Onasemnogene Abeparvovec: First Global Approval. Drugs 2019;79:1255-1262.
- Day JW, Finkel RS, Mercuri E, et al. Adeno-associated virus serotype 9 antibodies screened for treatment with onasemnogene abeparvovec. Mol Ther Methods Clin Dev.2021;21:76-82.
- Leborgne C, Latournerie V, Boutin S, et al. Prevalence and long-term monitoring of humoral immunity against adeno-associated virus in Duchenne Muscular Dystrophy patients. Cell Immunol. 2019;342:103780.
- WHO Housing and Health Guidelines. Geneva: World Health Organization; 2018. Table 3.1, Measures of crowding. Available from: https://www.ncbi.nlm.nih.gov/books/NBK535289/table/ch3.tab2/
Acknowledgement: Author’s would like to thank Raj Razdan2, MS and Kerry Dibernado, LPN, CCRP, research coordinators at Children’s Healthcare of Atlanta, Atlanta, Atlanta for study consent, sample collection and processing.
Figure 1: Anti-AAV IC50 Nab titers in the study participants.
Figure 2: AAV seroprevalence in steroid versus steroid-naïve study participants.
Figure 3: Follow-up AAV testing in five study participants.
Table 1: Study cohort with positive and negative anti-AAV Nab titers.
| All study participants (n=89) | Anti-AAV Nab positive (n=48) | Anti-AAV Nab negative n=(41) |
| Age | ||
| <5 years | 4%(n=2) | 9%(n=4) |
| 5 – 12 years | 58%(n=28) | 61%(n=23) |
| >12 years | 38%(n=18) | 37%(n=14) |
| Steroid use | ||
| Deflazacort | 35%(n=17) | 42%(n=17) |
| Prednisone | 16%(n=8) | 15%(n=6) |
| Steroid naive | 48%(n=23) | 44%(n=18) |
| Ethnic background | ||
| Caucasians | 46%(n=22) | 59%(n=21) |
| Hispanics | 29%(n=14) | 29%(n=12) |
| African American | 21%(n=10) | 15%(n=6) |
| South East Asians | 4%(n=2) | 5%(n=2) |
| Ambulatory status | ||
| Ambulatory | 50%(n=25) | 65%(n=26) |
| Non- ambulatory | 50%(n=23) | 39%(n=15) |
Supplemental table 1: Study participants with positive anti-AAV2 IC50 Nab titers stratified based upon age, steroid use, race and ambulatory status
| Low titers (<1:20) | Moderate titers (1:20-1:240 | High titers (>1:240) | |
| Anti-AAV2 Nab positive patients (n=47) | 11 | 13 | 23 |
| Age | |||
| < 5 years | 0 | 0 | 1 |
| 5-12 years | 6 | 7 | 14 |
| >12 years | 5 | 6 | 8 |
| Steroid use | |||
| Deflazacort | 5 | 3 | 9 |
| Prednisone | 0 | 5 | 2 |
| Steroid naive | 6 | 5 | 12 |
| Ethnic background | |||
| Caucasians | 6 | 5 | 10 |
| African American | 1 | 5 | 4 |
| Hispanics | 4 | 2 | 8 |
| South East Asians | 0 | 1 | 1 |
| Ambulatory status | |||
| Ambulatory | 4 | 4 | 16 |
| Non-ambulatory | 7 | 6 | 10 |
Supplemental table 2: Study participants with positive anti-AAV9 IC50 Nab titers stratified based upon age, steroid use, race and ambulatory status
| Low titers (<1:20) | Moderate titers (1:20-1:240) | High titers (>1:240) | |
| Anti-AAV9 Nab positive patients (n=36) | 9 | 18 | 9 |
| Age | |||
| < 5 years | 1 | 0 | 1 |
| 5-12 years | 4 | 12 | 6 |
| >12 years | 4 | 6 | 2 |
| Steroid use | |||
| Deflazacort | 1 | 6 | 4 |
| Prednisone | 2 | 6 | |
| Steroid naive | 6 | 6 | 5 |
| Ethnic background | |||
| Caucasians | 5 | 10 | 2 |
| African American | 4 | 3 | 1 |
| Hispanics | 0 | 4 | 5 |
| South East Asians | 0 | 1 | 1 |
| Ambulatory status | |||
| Ambulatory | 5 | 11 | 7 |
| Non-ambulatory | 4 | 7 | 2 |
Supplemental table 3: Study participants’ positive for both anti-AAV 2&9 IC50 Nab titers stratified based upon age, steroid use, race and ambulatory status.
| Variables | Low titers (<1:20) | Moderate titers (1:20-1:240) | High titers (>1:240) |
| Both anti-AAV2&9 NAB positive patients (n=35) | 2 | 6 | 27 |
| Age | |||
| < 5 years | 0 | 0 | 1 |
| 5-12 years | 2 | 5 | 16 |
| >12 years | 0 | 1 | 10 |
| Steroid use | |||
| Deflazacort | 1 | 1 | 11 |
| Prednisone | 0 | 3 | 5 |
| Steroid naive | 1 | 0 | 14 |
| Race | |||
| Caucasians | 1 | 4 | 10 |
| African American | 1 | 4 | 4 |
| Hispanics | 0 | 1 | 8 |
| South East Asians | 0 | 1 | 1 |
| Ambulatory status | |||
| Ambulatory | 2 | 5 | 13 |
| Non-ambulatory | 0 | 6 | 9 |