br Statistical analyses br To calculate the significance in
2.13. Statistical analyses
To calculate the significance in the efficacy studies, the area under the curve was calculated for each tumor and the mean values of the groups were compared using the Welch unequal variance. To determine whether there is a significant difference between the frequencies of the tumor regression in the different treatment groups, we used the Fischer's exact test. If not stated otherwise, an unpaired two-sided Student t-test without Welch correction was used. The statistical ana-lyses were performed using either GraphPad Prism (version 7.00 for Windows, GraphPad Software, La Jolla, California, US) or Microsoft Excel.
3.1. Characterization of PEBCA particles
The particle size, polydispersity index (PDI) and zeta potential for the batches used, were in the range of 148–227 nm (z-average), 0.04–0.19 and - (0.6–2.4) mV, respectively. The drug content in the final particles was 6.0–8.6% (w/w), giving 2.0–3.4 mg CBZ/ml in the NP stock solutions (Table S1). The size of PEBCA NPs increased when adding the drug CBZ (Table S1). The size distribution curves for the two batches used for efficacy studies in the MAS98.12 tumor model and that of CBZ (forming clusters in solution) are shown in Fig. 1.
3.2. PEBCA-CBZ inhibits tumor growth more efficiently than free CBZ in the MAS98.12 mice model
To test the efficacy of the PEBCA NPs with incorporated CBZ in a PDX model, MAS98.12 tumors were implanted into the mammary fat pad of nude mice and treated with the drug-loaded particles (PEBCA-CBZ), empty particles (PEBCA), non-encapsulated (free) CBZ and saline as control (Fig. 2A). The injected dose was 2 × 15 mg CBZ/kg body weight, which corresponds to a particle dose of 2 × 175 mg/kg (Fig. 2A). Tumor growth was not affected by empty PEBCA NPs, but markedly inhibited by CBZ. The effect was significantly more pro-nounced when the drug was encapsulated in the NPs compared to free
Fig. 1. Size distribution of the batches used in the MAS98.12 efficacy study. The size distributions for PEBCA-CBZ (the batch with size z-average of 215 nm in Table S1) is shown in dark blue and for PEBCA (without drug; the batch with z-average of 156 nm in Table S1) is shown in light blue. The size distribution of CBZ, which forms clusters in solution, is shown in red. Intensity (%) on the y-axis means percent intensity of the total scattered light. (For interpretation of the references to colour in this Laurdan figure legend, the reader is referred to the web version of this article.)
drug (Fig. 2A). Notably, in the PEBCA-CBZ treated group 6 out of 8 tumors went into complete remission while this was the case in only 2 out of 9 CBZ-treated tumors and none in the negative control groups (Fisher's exact test (1-sided): p = 0.04; comparing PEBCA-CBZ with CBZ).
To further evaluate differences in tumor growth between the four groups, we calculated the area under the curve (AUC) for each in-dividual tumor and compared the mean values of each treatment group. The efficacy of PEBCA-CBZ is significantly better than treatment with free CBZ (p = .02; Fig. S1). The toxicity of the different treatments measured as body weight relative to the weight at treatment start is shown in Fig. 2B. The two treatments with CBZ (non-encapsulated or encapsulated as PEBCA-CBZ) caused a decline in body weight by ap-proximately 15%, but one week after the last injection the effect of the body weight was reversed. This tolerable body weight loss and the re-covery time were comparable for PEBCA-CBZ and CBZ. Administration of empty PEBCA NPs did not cause any toxic effect as estimated from the body weights.
The efficacy of PEBCA-CBZ and non-encapsulated CBZ was also studied in MDA-MB-231 tumor-bearing mice. In this model we did not detect a significant difference between the two CBZ formulations, but a delay in tumor growth was observed (Fig. S2). When compared to MAS98.12, the CBZ (free and encapsulated drug) was less effective in
the MDA-MB-231 tumors. It was not possible to improve the efficacy by increasing the dose, due to limiting systemic toxicity .
3.3. In vivo biodistribution of PEBCA particles
The biodistribution of the PEBCA NPs in MAS98.12 bearing mice was studied by fluorescence imaging up to 96 h after injection of par-ticles containing the fluorescent dye NR668. The mice were imaged using the IVIS® Spectrum scanner after 1, 4, 24 and 96 h (Fig. 3A), and then sacrificed such that organs could be harvested and visualized ex vivo. The images of all organs harvested 24 h after injections are shown in Fig. 3B, and the mean radiant efficiency relative to the pixel size of the region of interest per organ is plotted in Fig. 3C. The images of organs obtained at 1, 4 and 96 h are shown in Fig. S3. Injection of the free NR668 dye did not give any detectable fluorescence with the wa-velengths used (data not shown), thus indicating that the detected signals are from PEBCA-bound NR668. The images shown in Fig. 3 and Fig. S3 demonstrate a rapid uptake in all tissues shown; 24 h after in-jection the strongest signals were observed in liver, spleen and lymph nodes, although fluorescence also was easily detectable in tumors, kidneys, hearts and lungs (Fig. 3C).