FIGURE 9.

Distributions of $f_s$ and ${\gamma }_d$ obtained from fitting the multiple-dose model to the experimental data for a varying number of doses with a 2-week inter-treatment interval. The parameter distributions are represented as boxplots and were obtained from fitting the multiple-dose model to the 2-week inter-treatment interval datasets from Experiment 3 (Table 1), in which the model with varying $f_s$ and ${\gamma }_d$ was used. Outliers are represented as circles. Panel (A) shows the distributions of the fraction of surviving cells, such that a new value for $f_s$ is defined for each drug dose ( $f_s^{\mathit{1}},f_s^{\mathit{2}},\dots ,f_s^{\mathit{5}}$ ). We observe an increasing trend in the first four $f_s$ parameters, which suggests an increasing chemoresistance with each dose. However, $f_s^{\mathit{5}}$ takes on significantly lower values than $f_s^{\mathit{4}}$ ( $p=\mathrm{0.0015}$ ), which suggests that adding more doses may limit the trend towards chemoresistance. Panel (B) shows the distributions of the doxorubicin-induced death delay rate of irreversibly damaged cells, such that a new value of ${\gamma }_d$ is defined for each drug dose ( ${\gamma }_d^{\mathit{1}}$ , ${\gamma }_d^{\mathit{2}}$ ,…, ${\gamma }_d^{\mathit{5}}$ ). The values for ${\gamma }_d^{\mathit{2}}$ are significantly lower than those of ${\gamma }_d^{\mathit{1}}$ ( $p=\mathrm{5.6}\times {\mathrm{10}}^{-\mathrm{19}}$ ). Hence, the second irreversibly damaged subpopulation shows a slower transition to treatment-induced death. However, the subsequent doxorubicin doses induce irreversibly damaged subpopulations exhibiting an increasing ${\gamma }_d$ , with ${\gamma }_d^{\mathit{5}}$ being significantly larger than ${\gamma }_d^{\mathit{4}}\left(p=\mathrm{0.0093}\right).$ This observation further suggests that past a certain number of doses, initial chemoresistance appears to be reverted. An asterisk (*) indicates $p<$ 0.05 (two-sided Wilcoxon rank sum test).