Table 2.

Results from drug substance release testing (n = 66 batches derived from 7 donors)

Parameter	Specification	Result	Deviations
			absolute	%
ABCB5+ cell content	≥ 90%	97.72 ± 1.88%	1a	1.5a
Mycoplasma	Not detectable (< 10 CFU/ml)	Not detectable in 65/66 batches	1	1.5
Endotoxin level	≤ 2 EU/ml	≤ 2 EU/ml in 66/66 batches	0	0
Live cell count	n.a.b	275.33 × 106	n.a.b	n.a.b
Deviation of live cell count from expected valuec	≤ 30%	≤ 30% in 46/49 batchesd	3	6%d
Cell vitalitye	≥ 90%	98.60 ± 0.62%	0	0
Cell viabilityf	≥ 90%	99.45 ± 0.64%	0	0
CD90+ cell content	≥ 90%	99.45 ± 0.50%	0	0
Bead residues	≤ 0.5%	0.04 ± 0.05%	0	0
Microbiological control	No growth	No growth in 66/66 batches	0	0
IL-1RA secretion	> 125 pg/ml and ratiostim/unstim > 1	> 125 pg/ml and ratiostim/unstim > 1 in 66/66 batches	0	0
VEGF secretion	> 46.9 pg/ml	> 46.9 pg/ml in 66/66 batches	0	0
Angiogenic differentiation	Capillary structures in one or both seeded cell concentrations	positive in 63/66 batches	3	4.5%

Numeric results are given as mean ± SD

n.a. not applicable

^aFor one batch, measurement procedure was not fully GMP-conform. Although this did not affect the result, the batch was not released for drug production

^bAcceptance criteria for cell count are only defined for the aliquots into which the drug substance batches are divided for cryostorage, not for the batch as a whole

^cExpected value = sum of the live cell counts of all ABCB5⁺ cell isolates that were pooled to produce the drug substance batch

^dThe criterion “Deviation of live cell count from expected value” was implemented only from donor 4 on, which is the reason for the lower sample number (n = 49)

^eVitality was defined as percentage of live cells, defined as propidium iodide-excluding cells

^fViability was defined as percentage of metabolically active cells, defined as cells converting calcein acetoxymethylester to calcein