Long-term safety and efficacy following systemic administration of a self-complementary AAV vector encoding human FIX pseudotyped with serotype 5 and 8 capsid proteins

Abstract

Adeno-associated virus vectors (AAV) show promise for liver-targeted gene therapy. In this study, we examined the long-term consequences of a single intravenous administration of a self-complementary AAV vector (scAAV2/ 8-LP1-hFIXco) encoding a codon optimized human factor IX (hFIX) gene in 24 nonhuman primates (NHPs). A dose-response relationship between vector titer and transgene expression was observed. Peak hFIX expression following the highest dose of vector (2 × 10(12) pcr-vector genomes (vg)/kg) was 21 ± 3 µg/ml (~420% of normal). Fluorescent in-situ hybridization demonstrated scAAV provirus in almost 100% of hepatocytes at that dose. No perturbations of clinical or laboratory parameters were noted and vector genomes were cleared from bodily fluids by 10 days. Macaques transduced with 2 × 10(11) pcr-vg/kg were followed for the longest period (~5 years), during which time expression of hFIX remained >10% of normal level, despite a gradual decline in transgene copy number and the proportion of transduced hepatocytes. All macaques developed serotype-specific antibodies but no capsid-specific cytotoxic T lymphocytes were detected. The liver was preferentially transduced with 300-fold more proviral copies than extrahepatic tissues. Long-term biochemical, ultrasound imaging, and histologic follow-up of this large cohort of NHP revealed no toxicity. These data support further evaluation of this vector in hemophilia B patients.

Figure 1

Clearance of self-complementary AAV vector (scAAV)2/8-LP1-hFIXco vector after peripheral vein administration of 2 × 10¹² pcr-vector genomes (vg)/kg. Clearance of the vector from rhesus plasma, urine, saliva, and stool was determined using a quantitative PCR (qPCR) assay on samples collected following peripheral vein administration of 2 × 10¹² pcr-vg/kg scAAV2/8-LP1-hFIXco. Standards consisting of serial dilutions of scAAV2/8-LP1-hFIXco in rhesus plasma were used to define the sensitivity of the assay. Results are expressed as mean transgene copy number (vg)/ml ± SE of sample obtained from three animals in this dose cohort and indicate that the vector genomes have cleared from the body fluids by day 10.

Figure 2

Expression of human factor IX (hFIX) in rhesus macaques transduced with different doses of adeno-associated viral (AAV) vectors. Plasma hFIX levels (mean ± SEM) measured by enzyme-linked immunosorbent assay (ELISA) at the stated time points following peripheral vein administration of scAAV-LP1-hFIX of (a) 2 × 10¹² pcr-vg/kg, (b) 2 × 10¹¹ pcr-vg/kg, (c) 6 × 10¹⁰ pcr-vg/kg, and (d) 2 × 10¹⁰ pcr-vg/kg. (e) The relationship between vector dose and plasma hFIX levels and scAAV transgene copy number in the liver at 4 weeks after gene transfer.

Figure 3

Biodistribution of vector following peripheral vein administration of 2 × 10¹² pcr-vg/kg scAAV2/8-LP1-hFIXco. Results of quantitative PCR (qPCR) analysis of genomic DNA, isolated from the indicated organs at 8 weeks after administration of 2 × 10¹² pcr-vg/kg of scAAV2/8 particles via the peripheral venous route using primers unique to hFIXco. Shown is transgene copy number per diploid genome ± SE corrected for variation in loading and amplification efficiency using GAPDH primers. BM, bone marrow.

Figure 4

Humoral immune response after peripheral vein administration of self-complementary AAV vector (scAAV) vector. Plasma obtained from macaques after peripheral administration of scAAV2/8-LP1-hFIXco at the stated doses and at different time points after vector administration was analyzed for the presence of adeno-associated virus (AAV)8-specific immunoglobulin G (IgG) by enzyme-linked immunosorbent assay (ELISA) ± SE. Assays were performed in triplicate.

Figure 5

Long-term expression of human factor IX (hFIX) following a single bolus infusion of research grade self-complementary adeno-associated viral vector (scAAV)-LP1-hFIXco. (a) Plasma human FIX (hFIX) levels (mean ± SEM) measured by enzyme-linked immunosorbent assay (ELISA) at the stated time points following a single bolus infusion of scAAV-LP1-hFIXco into the peripheral vein using serotype 8 pseudotyped vector (arrows show points when liver biopsies were obtained). Macaques in these studies have been described previously.^, The bottom panel shows Southern blot analysis of genomic DNA extracted from the liver obtained at the stated time points after peripheral vein administration of scAAV2/8-LP1-hFIXco and digested with BsrDI followed by hybridization with an hFIXco-specific probe at the stated time. Arrow shows the expected ~1.5 kb hybridization band. (b) The relationship between plasma hFIX levels and transgene copy number in the liver. Plasma hFIX levels (mean ± SEM) following a single bolus infusion of scAAV-LP1-hFIXco into the (c) mesenteric vein using serotype 8 pseudotyped vector, and (d) peripheral vein using serotype 5 pseudotyped vector. Gray arrows with “R” in Figure 5c,d denote administration of the combination of rituximab (325 mg/m²) and cyclophosphamide (250 mg/m²) as a bolus weekly infusion for a period of 4 weeks.

Figure 6

Expression of human factor IX (hFIX) following in-utero delivery of self-complementary adeno-associated viral vector (scAAV)2/8-LP1-hFIXco in mice. Correlation between mean body mass (weight in grams) and plasma human FIX (hFIX) levels (mean ± SEM) measured by enzyme-linked immunosorbent assay (ELISA) at the stated time points following a single in-utero administration of scAAV2/8-LP1-hFIXco.