Validation of analytical methods in GMP: the disposable Fast Read 102® device, an alternative practical approach for cell counting

Abstract

Background: The quality and safety of advanced therapy products must be maintained throughout their production and quality control cycle to ensure their final use in patients. We validated the cell count method according to the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use and European Pharmacopoeia, considering the tests' accuracy, precision, repeatability, linearity and range.

Methods: As the cell count is a potency test, we checked accuracy, precision, and linearity, according to ICH Q2. Briefly our experimental approach was first to evaluate the accuracy of Fast Read 102® compared to the Bürker chamber. Once the accuracy of the alternative method was demonstrated, we checked the precision and linearity test only using Fast Read 102®. The data were statistically analyzed by average, standard deviation and coefficient of variation percentages inter and intra operator.

Results: All the tests performed met the established acceptance criteria of a coefficient of variation of less than ten percent. For the cell count, the precision reached by each operator had a coefficient of variation of less than ten percent (total cells) and under five percent (viable cells). The best range of dilution, to obtain a slope line value very similar to 1, was between 1:8 and 1:128.

Conclusions: Our data demonstrated that the Fast Read 102® count method is accurate, precise and ensures the linearity of the results obtained in a range of cell dilution. Under our standard method procedures, this assay may thus be considered a good quality control method for the cell count as a batch release quality control test. Moreover, the Fast Read 102® chamber is a plastic, disposable device that allows a number of samples to be counted in the same chamber. Last but not least, it overcomes the problem of chamber washing after use and so allows a cell count in a clean environment such as that in a Cell Factory. In a good manufacturing practice setting the disposable cell counting devices will allow a single use of the count chamber they can then be thrown away, thus avoiding the waste disposal of vital dye (e.g. Trypan Blue) or lysing solution (e.g. Tuerk solution).

Figure 1

Cell count validation protocol. According to ICH Q2, the test was performed three times under the same operating conditions by Op1 and Op2. The concentration of the cell suspension was previously quantified using the Bürker chamber. Both operators then evaluated the accuracy of the method comparing the Bürker chamber with the Fast Read 102® chamber, performing a total (non-viable) cell count. In order to evaluate the precision and the repeatability of the method, intra and inter operator CV% was calculated using a viable cell count by Trypan Blue vital dye. The acceptance criteria were: inter and intra operator CV% < 10 % (total cell count); intra and inter operator CV% < 5 % (viable cell count). For linearity, we tested serial dilutions, in the following range: undiluted, 1:2, 1:4, 1:8, 1:16, 1:32, 1:64, 1:128. Op1 and Op2 performed a total cell count three times for each dilution. On the basis of the results, the regression line was calculated and the optimal range of dilutions was determined.

Figure 2

Bürker chamber and Fast Read 102® cell count method. The Bürker chamber has 9 large squares (1 mm² each), divided by double lines (0.05 mm apart) into 16 group squares. The double lines form small 0.0025 mm² squares. The Chamber depth is 0.1 mm. The cells were counted in each of the 4 large squares (identified by the triple line and shaded in the figure). At the end of the procedure the operators calculate the average of the 4 readings (from 4 large squares) and calculate the cell concentration as follows:. $\left[\frac{\text{Cell}}{ml}\right]=\left(\frac{\sum \text{Cells}\text{counted}in4\text{squares}}{4}\right)\times \text{dilution}\text{factor}\times {10}^4$. Fast Read 102® chamber, a plastic device with a slide divided into 10 chambers. Each chamber contains a grid with 10 squares, subdivided into 16 small squares. When the chamber was filled, the cells distributed in the 5 squares (black lines) were counted, taking into consideration, for each chamber, a size of 1 x 1 mm, a depth of 0.1 mm and a volume of 0.1 μl per square, the cell concentration (cells/ml) was determined by the formula: $\left[\frac{\text{Cell}}{ml}\right]=\left(\frac{\sum \text{cells}\text{counted}in5\text{squares}}{5}\right)\times \text{dilution}\text{factor}\times {10}^4$

Figure 3

Accuracy. Accuracy data demonstrated that the cell count obtained by the disposable Fast Read 102® cell count device is comparable to the Bürker cell count. Each Operator calculated the average and SD of its three counts for both counting devices to obtain the intra operator CV%. Op1 and Op2 then calculated intra operator CV% between Bürker vs Fast Read102® (a: Op1 raw data; b: Op2 raw data). Finally, inter operator CV% (Bürker vs Fast Read 102®) was reported (c). CV = coefficient of variation, Op1 = operator 1, Op2 = operator 2.

Figure 4

Precision and Repeatability on MNCs. On the basis of accuracy data, the following experiments were made to test precision, repeatability, linearity and range only by using Fast Read 102®. The assay was performed only using the Fast Read 102® chamber on MNCs. Operators 1 and 2 tested their cell counts three times. Cell viability was evaluated using Trypan Blue vital dye with a 1:2 dilution. To obtain the intra operator CV%, each operator calculated the average and SD of their three counts (a): Op1 raw data; (b): Op2 raw data. The inter operator CV% viability was then calculated (c). CV = coefficient of variation, MNC = mononuclear cells, Op1 = operator 1, Op2 = operator 2.

Figure 5

Precision and Repeatability on MSCs. As previously reported, also for MSCs, the assay was performed only using the Fast Read 102® chamber on MSCs. Operators 1 and 2 tested their cell counts three times. Cell viability was evaluated using Trypan Blue vital dye with a 1:2 dilution. To obtain the intra operator CV%, each operator calculated the average and SD of their three counts (a: Op1 raw data; b: Op2 raw data). The inter operator CV% viability was then calculated (c). CV = coefficient of variation, MSC = mesenchymal stem cells, Op1 = operator 1, Op2 = operator 2.

Figure 6

Linearity and range on wPB and wBM. Linearity and range was evaluated by Fast Read 102 ® device, as above described. The assay was performed on wPB (A) and wBM (B). The Operators tested different dilutions of cell suspension with Tuerk solution (undiluted, 1:2, 1:4, 1:8, 1:16, 1:32, 1:64, 1:128) in order to verify the best range of dilution to use. To avoid any potential interference of the Tuerk lysing solution and the Trypan Blue vital dye, in this case, the test only considers the total cell count and not cell viability. The regression line was calculated using the average of values obtained from Op1 and Op2’s counts at each dilution. The best range of dilution, in order to obtain a slope line value very similar to 1, is between 1:8 and 1:128. a1) wPB dilution range: 1:4, 1:8, 1:16, 1:32, 1:64, 1:128; a2) wPB dilution range: 1:8, 1:16, 1:32, 1:64, 1:128; b1) wBM dilution range: 1:4, 1:8, 1:16, 1:32, 1:64, 1:128; b2) wBM dilution range: 1:8, 1:16, 1:32, 1:64, 1:128. Op1 = operator 1, Op2 = operator 2.

Figure 7

Linearity and range on MNCs and MSCs. The assay was performed on MNCs (A) and MSCs (B), using Fast Read 102 ®, as previously reported. The Operators tested different dilutions of cell suspension (undiluted, 1:2, 1:4, 1:8, 1:16, 1:32, 1:64, 1:128). The test was performed without considering cell viability, as described above, to compare all the “linearity and range” data. The regression line was calculated by using the average of values obtained from Op1 and Op2’s counts at each dilution. The best range of dilution, in order to obtain a slope line value very similar to 1, is between 1:8 and 1:128. a1) MNCs dilution range: undiluted, 1:2, 1:4, 1:8, 1:16, 1:32, 1:64, 1:128; a2) MNCs dilution range: 1:4, 1:8, 1:16, 1:32, 1:64, 1:128; a3) MNCs dilution range: 1:8, 1:16, 1:32, 1:64, 1:128; b1) MSCs dilution range: undiluted, 1:2, 1:4, 1:8, 1:16, 1:32, 1:64, 1:128; b2) MSCs dilution range: 1:4, 1:8, 1:16, 1:32, 1:64, 1:128; b3) MSCs dilution range: 1:8, 1:16, 1:32, 1:64, 1:128. Op1 = operator 1, Op2 = operator 2.