Human Gene Therapy - April 2023 - 330

330
TROXELL ET AL.
Figure 1. Representative SEC-MALS UV chromatograms with molar mass information obtained with two separate columns. Samples were diluted fivefold in
mobile phase and 45 lL was injected. Flow rate was 0.3 mL/min and the multisampler was maintained at 4C. Ten separate volume transfers were performed
from a single vial of process confirmation material. Data shown are from a representative chromatogram and the total data from 10 independent injections are
shown in Supplementary Table S1. The black trace was obtained with the Wyatt 500A˚ column and the green trace was obtained with the SRT-1,000A˚ column.
Baseline-to-baseline integration of the monomer peak was performed in ASTRA software to obtain protein, DNA, and peak statistics. The test article was the
process confirmation material. The molar masses for the DNA, capsid, and total vector symbols are indicated in figure. SEC-MALS, size exclusion chromatography
with UV and multiangle light scattering.
performed. To determine linearity of a method, a minimum
of five concentrations distributed across the range of
detection is recommended.36 The mean CP/mL and VG/
mL values shown in Supplementary Table S1 obtained
with the Wyatt column were used to assess linearity and
accuracy by recovery. A plot of the theoretical CP/mL and
VG/mL against the measured values produced strong
linearity (R2 > 0.999) over two orders of magnitude
(Fig. 2A, B). Moreover, the mean molar mass data for both
vector and DNA were consistent across these sample
ranges (Fig. 2C). Accuracy was assessed by using an appropriate
number of determinations and levels that cover
the expected reportable range.36
Each replicate recovered within 20% of the spike value
across all levels (Fig. 3A; Table 1). There was an increase in
the variability for molar masses and concentrations below
1.0 ·1012VG/mL (Table 1). Since one of the goals ofSECMALS
is to detect and characterize aggregate species, it
was also evident that the off-target peaks were at low
concentrations (<4% of total mass) and required higher
sample concentrations (>2.5 · 1012 VG/mL; Fig. 3B). To
achieve accurate and precise measurements for all CQAs
Figure 2. Linearity of CP/mL, VG/mL, and molar mass SEC-MALS data obtained with the Wyatt 500A˚ column. Samples were diluted twofold followed by serial
1.7-fold dilutions to cover nine levels. Five volume transfers were performed at each level with singlicate injections (injection volume and flow rate were as in
Fig. 1). Baseline-to-baseline integration of peaks were performed in ASTRA software to obtain protein, DNA, and peak statistics. (A) Mean CP/mL concentrations
of injected material at each dilution level from five preparations at each level are shown. Error bars show the standard deviation, but are below
visibility on the graph. (B) Mean VG/mL concentrations of injected material at each dilution level from five preparations at each level are shown. Error bars
show the standard deviation, but are below visibility on the graph. (C) Mean molar mass data for vector and DNA content of injected material at each dilution
level from five preparations at each level are shown. Error bars show the standard deviation. Test article was the process confirmation material. CP, capsid
particles; VG, vector genomes.

Human Gene Therapy - April 2023

Table of Contents for the Digital Edition of Human Gene Therapy - April 2023