Fig 1.
In vitro evaluation of DCH-paclitaxel system.
A. On a per weight basis, paclitaxel is 500,000 times more potent than temozolomide at suppressing the growth of human gliomaspheres over a 7-day period in vitro. Growth rate (GR) inhibition calculations27 were used to quantify the relative effectiveness of various concentrations of each drug and normalized to untreated controls. B. Cumulative release curve for paclitaxel loaded into the DCH over a 6 week incubation in vitro. A 1000 fold sink of 5% FBS in 1XPBS was used. An initial burst release of paclitaxel of over the first three days was noted and 96% of drug was recovered at the end of the 6 week incubation period. C. HPLC chromatograms show that paclitaxel recovered from the DCH at the end of the 6 week in vitro incubation was chemically identical to freshly prepared drug. D. 2ul of DCH control and paclitaxel loaded were placed in 3ml of media and 100,000 HK308 cells. Gels dissolved in the media. Photos were taken at 24 and 72 hours.
Fig 2.
DCH-paclitaxel exhibits good biocompatibility in healthy CNS in contrast to paclitaxel in Cremophor EL vehicle.
A-D. Images of caudate putamen at 1 week after injection into healthy, uninjured tissue of Cremophor EL vehicle (A), Cremophor EL + paclitaxel (B), DCH only (C), or DCH + paclitaxel (D), showing single channel and merged multichannel immunofluorescence for multiple markers of inflammation and gliosis, CD 68, IBA-1 and GFAP. Scale bar, 200 μm for all images, D = DCH depot. E. Quantification of immunofluorescence intensity for each treatment group across a radial area of 1 mm originating from the center of the injection (n = 3 mice per stain per treatment). F. Area Under the Curve (AUC) calculations for the various immunofluorescence intensity traces from E. provide a single measure of cumulative staining within the 1 mm radial field. The DCH-paclitaxel system showed markedly and significantly less staining intensity, indicative of a more favorable foreign body response, compared with paclitaxel administered using the Cremophor EL vehicle * p < 0.01 or ** p < 0.001 (ANOVA with post-hoc Tukey’s multiple comparisons test).
Fig 3.
Paclitaxel-DCH depots persist locally for up to 4 weeks in vivo and are superior at preserving viable adjacent healthy CNS tissue compared to paclitaxel in Cremophor EL vehicle.
A. CD68 positive cells are present at the DCH-paclitaxel interface for up to 4 weeks post injection but there is minimal diminution of the depot size, minimal material resorption and minimal infiltration of inflammatory cells into the DCH-paclitaxel depot. B. NeuN positive, viable neurons are present in normal density and intermingled with mildly reactive astrocytes in close proximity with the DCH-paclitaxel depot (DCH-Pac). C. In contrast, there is pronounced depletion of NeuN positive neurons and severe proliferative reactive astrogliosis in tissue adjacent to injection of paclitaxel in Cremophor EL vehicle (Crem-Pac). Scale bar, 200 μm for all images. N = 3 mice per stain per treatment.
Fig 4.
DCH-paclitaxel reduces hGBM growth progression rates monitored by luciferase bioluminescence imaging in vivo and prolongs survival.
A. Schematic of the mouse brain identifying the anatomical location of the gliomasphere and then DCH injections within the caudate putamen (CP) region of the striatum. B and C. Graphs showing Luciferase Total Flux (p/s = photons/second) progression as a function of time as measured by bioluminescence IVIS imaging for each animal receiving either DCH only or DCH-paclitaxel in cohort 1 (n = 4 animals per group) (B) or cohort 2 (n = 10 animals per group) (C). An exponential growth regression was applied to each treatment group in B and C which is represented by a straight line on the semi-log plot. The delta t (Δt) is the time between the same absolute average flux value between the two treatment groups and was calculated to be approximately 1.5 weeks for cohort 2 D. Kaplan Meyer survival curve for cohort 2 animals which demonstrated that DHC-paclitaxel conferred a significant median survival time increase of 2.5 week (or a 23% extension of life) compared to the DCH alone (p = 0.0063). Statistic calculated by log rank test calculated by Prism 6 software. E. In a third cohort, animals were injected with gliomasphere cells followed by DHC two weeks later. The effect was smaller and the survival curve did not meet statistical significance.
Fig 5.
The DCH-paclitaxel system substantively ablates locally residing hGBM cells but does not prevent the migration of tumor cells within brain parenchyma.
A. Schematic of mouse forebrain showing the location of hGBM and DCH injections in the caudate putamen (CP). Box of dashed lines delineates the area quantitatively analyzed for the presence of GFP labeled cells in 5-week post injection tissue and presented in D and E. B,C. Survey immunofluorescent images of staining for GFP (hGBM cells) and GFAP show the location of hGBM cells in forebrain at 5 weeks post injection. B. Mouse that received hGBM cells and DCH-only. Note the high density of GFP-positive hGBM cells immediately above and below the injection site. C. Mouse that received hGBM cells and DCH-paclitaxel. Note the essential absence of GFP-positive hGBM cells immediately around the persisting DCH-paclitaxel depot (D). Note also the presence of hGBM cells that have migrated away from the injection site (box shown at higher magnification in 7C). D. Graph of quantification of GFP signal from hGBM cells. GFP staining intensity was measured as a function of brain depth in serial linear units in the boxed area shown in A. As expected, animals receiving DCH-only (blue) exhibited a high intensity of GFP signal in and immediately above the injection region (0.5 to 3 mm). In contrast, animals receiving DCH-paclitaxel (DCH-Pac, red) exhibited little or no GFP signal in this area, but did exhibit substantive signal at deeper levels (around 4mm). E. Bar graphs quantifying area under the Curve (AUC) for GFP staining intensity. Over the entire depth of the brain, DCH-paclitaxel treated animals (DCH-Pac, red) exhibited an over 72% reduction in hGBM-derived GFP signal compared with DCH-only (p<0.001). In the area immediately around the hGBM and DCH injections (surface to 3 mm), DCH-paclitaxel treated animals exhibited an over 97% reduction in hGBM-derived GFP signal (p<0.001). In contrast, in the deeper areas away from DCH depots (around 4 mm), DCH-paclitaxel treated animals exhibited a 65% greater hGBM-derived GFP signal (p<0.001). n = 5 per group for all measures.