Comparison of red and green light for treating non-muscle invasive bladder cancer in rats using singlet oxygen-cleavable prodrugs with PPIX-PDT

Abstract

It has been 30 years since Photofrin-PDT was approved for the treatment of bladder cancer in Canada. However, Photofrin-PDT failed to gain popularity due to bladder complications. The PDT with red light and IV-administered Photofrin could permanently damage the bladder muscle. We have been developing a new combination strategy of PpIX-PDT with singlet oxygen-cleavable prodrugs for NMIBC with minimal side effects, avoiding damage the bladder muscle layer. PpIX can be excited by either green (532 nm) or red (635 nm) light. Red light could be more efficacious in vivo due to its deeper tissue penetration than green light. Since HAL preferentially produces PpIX in tumors, we hypothesized that illuminating PpIX with red light might spare the muscle layer. PpIX-PDT was used to compare green and red laser efficacy in vitro and in vivo. The IC₅₀ of PpIX-PDT was 18 mW/cm² with the red laser and 22 mW/cm² with the green laser in vitro. The efficacy of the red laser with 50, 75, and 100 mW total dose was similar to the same dose of green laser in reducing tumor volume. Combining PpIX-PDT with prodrugs methyl-linked mitomycin C (Mt-L-MMC) and rhodamine-linked SN-38 (Rh-L-SN-38) significantly improved efficacy (tumor volume comparison). PpIX-PDT or PpIX-PDT + prodrug combination did not cause muscle damage in histological analysis. Overall, a combination of PpIX-PDT and prodrug with 635 nm laser is promising for non-muscle invasive bladder cancer treatment.

Graphical Abstract

Both 532 and 635 nm lasers could produce similar antitumor efficacy in both in vitro and in vivo for the treatment of non-muscle invasive bladder cancer (NMIBC). The combination of 635 nm laser in PpIX-PDT and photo-activatable prodrugs for the treatment of NMIBC has been shown to be safe and promising.

INTRODUCTION

Bladder cancer takes about 17,000 lives each year making it the fourth most common cancer in men and ninth in women (1–3). In addition, it is the most expensive cancer to treat, with the lifetime cost of managing bladder cancer being as high as $200,000 due to the need for continuous visits to the clinic in order to diagnose recurrences long after the initial course of treatment (4). The recurrence rate of bladder cancer (5 years) can be as much as 15 times higher than that of breast cancer (5, 6). Currently, as a standard form of treatment, patients usually receive Bacillus Calmette-Guérin (BCG) immunotherapy or chemotherapy, instilled directly in the bladder, after the surgical removal of the tumor. Some of the patients who received chemotherapy or BCG had to stop treatment due to the side effects (7, 8).

Photodynamic therapy (PDT) can be a suitable alternative to BCG after surgery, especially for the BCG-unresponsive non-muscle invasive bladder cancers (NMIBC) (9). It has been 30 years since PDT using Photofrin was approved in Canada for NMIBC (10). However, intravenous (IV) Photofrin-PDT with red light illumination caused muscle damage for some patients; hence it did not gain popularity (11). PpIX-PDT, where PpIX is formed by bladder-instilled hexaminolevulinate (HAL), can be a better alternative to Photofrin-PDT for NMIBC. Unlike non-specific diffusion of IV-administered Photofrin throughout the bladder tissue, HAL instilled in the bladder preferentially produces PpIX in the bladder tumor and thus improves preferential formation of PpIX in the tumor over normal bladder muscle (12). PpIX can be excited with either green (532 nm) or red (635 nm) laser due to its soret and Q-bands (13). Both 532 nm and 635 nm light have their own pros and cons. Shorter wavelength 532 nm light can be safer for the muscle layer due to its limited penetration depth but 635 nm can be suitable for treating a tumor thicker than 1 mm (14). Also, red light diffusion can be less affected by the instillation solution. We hypothesized that with the preferential formation of PpIX in the tumor, PpIX-PDT with red light could be effective for NMIBC without damaging the muscle layer.

Our main goal here is to compare 532 nm and 635 nm lasers to determine a suitable illumination condition for treating orthotopic NMIBCs in a preclinical rat model. Light diffusion was evaluated with both green and red lasers in instillation media containing prodrugs (Figure 1). We also determined the in vitro IC₅₀ values of both lasers and the in vivo efficacy with an orthotopic rat bladder tumor model. Additionally, to aid in the antitumor efficacy determination, we used a simple detection tool for bladder tumors using a portable LED light and a handheld smartphone camera with an emission filter, modified from a similar technique introduced earlier (15). Lastly, we evaluated a prodrug-PpIX-PDT combination efficacy compared to PpIX-PDT only. We have previously demonstrated the in vitro cytotoxicity of several singlet oxygen-activatable prodrugs of paclitaxel, Combrestatin A4, SN-38, and mitomycin C (16, 17). We have further evaluated the in vivo efficacy of the paclitaxel prodrug (Rh-L-PTX) as well as mitomycin C prodrug (Mt-L-MMC) (18). To improve the drug diffusion in the bladder wall and to improve the in vivo efficacy of Rh-L-PTX, a more hydrophilic and smaller molecular weight Rh-L-SN-38 was developed. So, in this study, we have used mitochondria-targeted rhodamine-linked SN-38 (Rh-L-SN-38) and non-targeted (non-specific cellular localization) methyl-linked mitomycin C (Mt-L-MMC) prodrugs along with PpIX-PDT in an orthotopic rat bladder tumor model. A schematic presentation of the mechanism of PDT and the cytotoxicity effect of PpIX-PDT with 532 nm or 635 nm light with SO-cleavable prodrugs is presented in Figure 1.

Figure 1:

Mechanism of PDT and cytotoxicity effect of PpIX-PDT with 532 nm or 635 nm light with SO-cleavable prodrugs

MATERIALS AND METHODS

Cell line and animal model:

AY-27 cells (obtained from Dr. Steven Salman at the University of Toledo) were grown and maintained in a complete RPMI-1640 growth medium as described previously (17). Female Fischer F344 rats (150–180 g) aged 8–12 weeks were obtained from Envigo Global Services Inc. and housed for 7 days for acclimation prior to any procedure. An orthotopic rat tumor model was established as described previously with some modifications (19). Briefly, rats were anesthetized and maintained using (2.5%) isoflurane. The bladder was washed with 400 μl of 0.1 M hydrochloric acid solution for 15 seconds, then neutralized with 400 μl of 0.1 M NaOH for 15 seconds, and then washed with Phosphate Buffered Saline (PBS) three times. Subsequently, 400,000 cells in 400 μl plain media (without FBS and antibiotic) were instilled in the bladder for 60 min, with rats rotated 90° every 15 min. All animal procedures were conducted in compliance with the protocol approved by the University at Buffalo Institutional Animal Care and Use Committee.

Prodrug synthesis and formulation:

Prodrugs (Mt-L-MMC and Rh-L-SN-38) were synthesized, characterized, and purified as described previously (17). Prodrug formulation was prepared in two steps. In the first step, a stock solvent was prepared by mixing 5% (100 μl) ethanol and 5% (100 μl) Cremophor EL (EMD Millipore Corp., #238470–25GM) and diluted to a final volume (2 ml) with 90% PBS (1800 μl). In the second step, the required amount of prodrug was dissolved in 1% ethanol (20 μl), then 10% solvent prepared in step 1 (200 μl) was added and mixed well then diluted to final volume (2ml) with 89% PBS (1780 μl). To prepare HAL (8 mM) only or HAL + prodrug (HAL, 8 mM + prodrug), a stock of 20 mM HAL was prepared in PBS. To prepare HAL only, 20 mM HAL in PBS was diluted to 8 mM HAL formulation, keeping the formulation consistent with prodrugs as described above. To prepare the HAL + prodrug combination, the required volume of 20 mM HAL (in PBS) was added to the prodrug formulation in the final step (89% PBS minus the required volume of 20 mM HAL). Prodrug concentration in the final formulation was 200 μM and 1 mM for Rh-L-SN-38 and Mt-L-MMC, respectively.

Light diffusion in instillation media:

A frontal light distributor (model FD1) and isotropic probe (model IP85) from Medlight S.A. (Switzerland) were used for light diffusion experiments. For these experiments, a power meter (PM100D, Thorlabs) with a sensor (S121C, Thorlabs) and diode lasers (532 nm, MDL-III-532, and 635 nm, MDL-III-635, CNI laser, China) with a total power output set to 50 mW were used. Images were taken with a smartphone camera in manual mode. HAL (8 mM) and prodrug (Mt-L-MMC, 0.25, 0.5, and 1 mM and Rh-L-SN-38, 50, 100 and 200 μM) solutions were prepared as described above. A 5 ml glass vial with an internal diameter of 14 mm was used as the instillation chamber, mimicking a rat bladder (10–15 mm diameter). To measure the light propagation in various media conditions, the sensor was placed touching the wall of the flask (7 mm from the light source) and the light intensity was measured for each of the media (air, PBS, and prodrug formulation). To determine the influence of blood present in the urine, we collected blood from rats and then mixed it with PBS in various percentages to determine the light intensity using the same instrument setup described above. UV absorbances of the prodrugs in the formulation were also taken with a UV spectrophotometer scanning from 300 – 700 nm.

Phototoxicity and dark-toxicity:

A brief treatment scheme was presented in Figure 2. Phototoxicity was determined using AY-27 cells with a fixed concentration of HAL (0.5 mM) without or with illumination for 20 min with 532 nm and 635 nm diode lasers at a range of 2.5 to 100 mW/cm². Cells were grown and maintained in RPMI (Roswell Park Memorial Institute) 1640 medium supplemented with 10% fetal bovine serum, L-glutamine, 50 μg/mL streptomycin, 50 units/mL Penicillin G and 1.0 μg/mL fungizone (complete medium). 10,000 AY-27 cells per well were seeded with complete medium (200 μl) onto a 96-well plate and incubated for 24 hr at 37°C in 5% CO₂. HAL stock solution (8 mM in complete medium) was diluted to have a final HAL concentration of 0.5 mM. The cell growth media was replaced by the HAL solution (200 μl) and the plates were kept in the incubator for 2 hr. Then, the medium was again replaced with a fresh complete medium (200 μl). The plates were illuminated with either 532 nm or 635 nm diode laser at 2.5 – 100 mW/cm² for 20 min and returned to the incubator. After 48 hr, cell viability was determined using an MTT assay. Briefly, MTT stock solution in PBS buffer was added to each well to achieve a final concentration of 0.5 mg/mL, and the plate was incubated for 4 hr. After removing the MTT solution, cells were dissolved in 200 μl DMSO. The absorbance was measured at 570 nm, with a background reading at 650 nm. The cell survival was determined by comparing the absorbance of the treated well with that of the control wells. The in vitro experimental conditions such as administration of HAL for 2 hr were determined based on reference data (HAL, 0.5 mM) (20, 21) and the clinical protocol of blue light cystoscopy (22).

Figure 2:

Treatment scheme with 532 nm or 635 nm light with PpIX-PDT and PpIX-PDT + prodrugs combination for an orthotopic rat bladder tumor model with AY-27 cells.

Bladder imaging using PpIX-PDD (photodynamic diagnosis):

To visualize the whole bladder using PpIX-PDD, rats were instilled with 400 μl of 8 mM HAL solution and kept for 2 hr prior to euthanasia. The HAL solution was removed and the bladder was washed twice with PBS. After euthanasia, the bladder was excised and placed under illumination with a 405 nm portable LED light. To capture the fluorescent signal, a smartphone camera equipped with a 630–660 nm emission filter was used to take the images using a manual setting to keep the image acquisition parameters the same for all the images.

In vivo PDT:

To determine the antitumor efficacy of PpIX-PDT only or prodrug + PpIX-PDT combination, NMIBC was established in the bladder of female rats as described previously (19). On the treatment day (day 7 of cancer cell instillation) 400 μl of either vehicle, SN-38 (200 μM), MMC (1 mM), HAL (8mM) or HAL (8mM) + prodrug (200 μM of Rh-L-SN-38 and 1mM of Mt-L-MMC) combination were instilled into the bladder. After two hours of instillation, the drug solution was drained out and the bladder was washed 3 times with 400 μl PBS each time. The laser power was set to 50, 70, or 100 mW (total power as measured using an FD1 fiber, frontal diffuser). Afterward, FD1 fiber was replaced with an IP-85 fiber and inserted through the catheter, and connected to a 532 nm or 635 nm laser source for illumination of the bladder for 20 min. For, PpIX-PDT and prodrug combination treatment groups, only 635 nm was used with 50 mW total power. Following the illumination, the rats were returned to their cages.

Histology and imaging:

To prepare tumor histology samples for bladder muscle damage analysis, rats were euthanized two days after treatment, and their bladders were filled with O.C.T. compound (Fisher Scientific) before removal. The excised bladders were placed in cryomolds (Tissue-Tek, Sakura, CA) and frozen in isopentane chilled with liquid nitrogen. 10 μm cryosections were cut and kept at −80°C until further processing. After H&E staining, sections were protected with a coverslip mounted with Prolong Diamond (Fisher Scientific) mounting media. H&E-stained sections were imaged using the Aperio imagescope.

Data collection and statistical analysis:

After bladder extraction, a digital slide caliper was used to measure tumor volume (length, height, and width). The bladder area and tumor-covered area were measured using ImageJ. Bladder weight was taken on a digital scale after quickly drying with cotton balls and absorbent napkins. GraphPad Prism (software version 9.5.0) was utilized for data analysis. Student T-test was used to compare between groups (i.e., HAL only vs. HAL + Rh-L-SN-38; HAL only vs. HAL + Mt-L-MMC, etc.).

RESULTS

Photodynamic tumor imaging:

In the preclinical study of orthotopic bladder cancer treatment, often it is difficult to determine the bladder tumor area compared to the healthy bladder area. To aid in the quantitative analysis of bladder cancer efficacy, we made some modifications to a previously reported technique by Khan et. al. (23), and successfully imaged tumors using 405 nm excitation light and a 630–660 nm emission filter on a consumer smartphone (Figure 3). This imaging method was sensitive enough to detect tumors as small as 1 mm in the rat bladder (Figure 3D), following incubation with 8 mM HAL for 2 hr. HAL preferably produces PpIX in the tumor which helps in the tumor detection by the PpIX excitation light, 405 nm.

Figure 3:

Utilizing consumer smartphone to determine bladder tumor using fluorescence imaging. After 2 hr of incubation with HAL images were taken by illuminating with a 405 nm lamp and placing a 610–660 nm emission filter on the camera. (a,c,e) bladder images with white light without filter (b,d,f) bladder images with 405 nm light and emission filter (g) bladder image with emission filter but without illumination (h) smartphone with emission filter (i) 405 nm led and emission filter used in the imaging. (e and f) Circle indicates the presence of fat tissue where the fluorescence signal is not produced.

Light diffusion simulation in the pig bladder matrix:

We tried to find a suitable illumination condition for PpIX-PDT in the preclinical study as well as to determine the light diffusion when the depth of the tumor is greater than 1 mm (human bladder tumor).

In COMSOL Multiphysics, Helmholtz modification of the steady-state diffusion equation was used to simulate light propagation from a continuous wave light source placed adjacent to the bladder tissue (24). The diffusion equation is as follows,

$$-\nabla \cdot \text{D}\nabla \phi +{\mu }_a\phi =\text{q}$$

Here, q is the strength of the source (point source) light (100 mW) and φ represents the light fluence rate. μ_a denotes the optical absorption coefficient and D is the light diffusion coefficient which is defined as,

$$\text{D}=1/3{{\mu }_{\text{s}}}^{\prime }$$

Where μ_s^’ represents the reduced scattering coefficient of light.

To simulate light diffusion in the bladder tissue, previously reported values of absorption and scattering coefficient were used to simulate light diffusion in the tissue matrix (25). Light diffusion with 635 nm was dropped by 36% within the distance of 5 mm in the bladder tissue while the maximum light intensity of 532 nm laser was reduced by >80% within the same distance (Figure 4). Optical penetration depth is defined by the depth at which light intensity drops by 37% of the initial intensity (26). Accordingly, 635 nm optical penetration depth was calculated to be 5 mm whereas 532 nm penetration depth was estimated to be 3.7 mm.

Figure 4:

Light diffusion simulation using optical properties (absorption and scattering coefficient) of pig bladder with 532 nm (a) and 635 nm (b) light. (c) light intensity across the cut line. Scale bar: 2 mm.

Light attenuation by contaminated blood:

In our experience, after an acid-base wash of the bladder to develop a bladder tumor, sometimes rat bladders don’t completely heal by day 7 (treatment day). Apart from the preclinical model, in a clinical setting, there might be cases of bleeding if PDT is employed after surgical tumor removal. Hence, due to the probability of bleeding in the bladder while doing PDT, we determined the light diffusion difference in both light wavelengths. The light power difference (50 – 100 mW) did not affect the diffusion in both 532 and 635 nm lasers (Figure 5). However, the green light was more dramatically attenuated than the red light by the solution with blood. The green laser intensity decreased by 99% with less than the solution with 1% blood while at least 50% light intensity was retained with 635 nm light with up to 5% blood in the urine. IC₅₀ of % blood in the urine to decrease the light intensity was found to be 0.5% and 6% with 532nm and 635 nm light, respectively.

Figure 5:

Attenuation of light propagation by blood (a) and IC₅₀ of % blood in light attenuation (b) with 532 nm and 635 nm light.

Light attenuation by the prodrug:

It was demonstrated previously that tumors can be illuminated without washing out the prodrug solution instilled in the bladder to achieve a better light delivery (27). To determine if the light propagation is affected by the instilled solution with prodrugs, we determined the light diffusion in different prodrug solutions (Figure 6). It was found that with the 532 nm laser, light intensity was reduced by up to 40% and 66% with Mt-L-MMC (1 mM) and Rh-L-SN-38 (200 μM), respectively, when compared to PBS only. In contrast, for 635 nm laser, light intensity was reduced by just 12 and 7% with Mt-L-MMC and Rh-L-SN-38, respectively. Both Mt-L-MMC and Rh-L-SN-38 attenuated light with the 532 nm laser in a concentration-dependent manner. With increasing prodrug concentrations, no statistically significant light attenuation was observed with the 635 nm laser. No significant light attenuation was observed for either laser when only PBS or HAL was used.

Figure 6:

Light diffusion of 635 nm or 532 nm laser in the instillation media with prodrugs, Rh-L-SN-38 (50, 100, 200 μM) and Mt-L-MMC (250, 500, 1000 μM). (A) Rh-L-SN-38 with 635 nm light, (B) Mt-L-MMC with 635 nm light, (D) Rh-L-SN-38 with 532 nm light, (E) Mt-L-MMC with 532 nm light. (C) Experimental setup and (F) the scatter plot showing the impact of prodrugs on the fluence rate (mW/cm²) at the detector. Statistical significance (paired one-tailed t-test, **** P < 0.0001).

In vitro PpIX efficacy comparison:

In vitro PpIX-PDT efficacy was determined with 532 nm or 635 nm laser. A fluence rate of 2.5 to 100 mW/cm² was used to determine the cell survival at 48 hr after PDT. While at a low fluence rate (2.5 – 10 mW/cm²) there was no significant difference between the 532 nm and 635 nm laser, we have observed an increase in efficacy with the 635 nm laser with higher fluence rates (≥ 15 mW/cm²) (Figure 7). IC₅₀ value of PpIX-PDT with 532 nm and 635 nm was determined to be 22 and 18 mW/cm², respectively.

Figure 7:

In vitro PpIX-PDT efficacy with 532 nm or 635 nm light. Scatter plot of cell viability (a) and non-linear fit of cell viability with 532 nm or 635 nm light (b).

In vivo PpIX-PDT efficacy and toxicity:

In vivo efficacy was determined using the green and red lasers with a total power of 50, 70, and 100 mW. Fluorescence images were utilized to determine the tumor area. Both laser groups achieved significantly reduced tumor volume compared to vehicle only group. Our goal was to determine if there is any difference or benefit using either green or red laser. We observed that the 635 nm laser achieved better tumor control than 532 nm and this was observed with increasing light doses (Figures 8 and 9). Since total bladder area increases (27) with lower tumor burden we compared the total bladder area between treatment groups. The total bladder area was increased in both 50 mW and 70 mW power with 635 nm light compared to 532 nm (Figure 9A). The fraction of the bladder area covered with tumor was also significantly lower with 635 nm light with 50 mW and 70 mW (Figure 9C). Tumor volume was significantly lower for rats treated with 635 nm light at 70 mW and 100 mW (Figure 9D). Since the local toxicity, such as bladder muscle damage, can be a concern with deeper light penetration with a 635 nm laser, we performed histological analysis of bladder muscle. Bladder muscle was analyzed after rats were treated with only light: 532 nm or 635 nm laser with 50 mW, 70 mW, and 100 mW total power (Figure 10). No obvious muscle damage was observed with either 532 nm or 635 nm laser up to 100 mW total power for 20 minutes.

Figure 8:

White light and fluorescence images of the rat bladders treated with 532 nm or 635 nm light with 50 mW, 70 mW, and 100 mW total power. The top bars indicate the experiment groups.

Figure 9:

Scatter plot showing in vivo antitumor efficacy of PpIX-PDT with 532 nm or 635 nm laser observed in figure 8. (a) Total bladder area (b) tumor covered area (c) fraction of bladder covered with tumor (d) tumor volume comparison. Statistical significance (paired one-tailed t-test, * P < 0.05).

Figure 10:

H & E images of bladder muscle 48 hr after PpIX-PDT with 532 nm and 635 nm laser. (A) vehicle control tumor bladder. (B, C, D) 532 nm laser with 50, 70, and 100 mW total power, respectively. (E, F, G) 635 nm laser with 50, 70, and 100 mW total power, respectively. Scale bar: 100 μm. BL indicates the luminal side of the bladder (not necessarily inside the bladder lumen, A, D, E, G).

In vivo antitumor efficacy of the combination treatment:

In vivo antitumor efficacy was determined using a prodrug combination to determine if PpIX-PDT efficacy can be improved with the site-specific chemotherapy using the SO-cleavable prodrugs. Due to the improved light diffusion (Figures 4–6) and better antitumor effect (Figure 9), we chose 635 nm laser with 50 mW power for 20 minutes (84 J) for the combination treatment. Mitochondria-targeted prodrug (Rh-L-SN-38, 200 μM) and non-targeted prodrug (Mt-L-MMC, 1 mM) were used for the combination study. Figure 11 represents the white light images of the excised bladders. Overall bladder area was increased with PpIX-PDT alone and in combination with the prodrugs compared to the vehicle control groups (Figure 12A). PpIX-PDT + Mt-L-MMC and PpIX-PDT + Rh-L-SN-38 achieved a significant reduction of bladder area covered in tumor (48 ± 18 and 49 ± 7 mm², respectively) compared to PpIX-PDT alone (72 ± 19 mm²) (Figure 12B). Although not statistically significant, PpIX-PDT and prodrugs combination could also reduce the mean fraction of bladder area covered with tumor and the mean bladder weight compared to PpIX-PDT alone (Figure 12C, D). Since tumor burden increases bladder weight and decreases bladder area (27), we calculated weight/area as one of the treatment efficacy indicators. Even though not statistically significant, PpIX-PDT achieved significantly lower bladder weight per unit area (1.0 ± 0.2 mg/mm²) compared to the vehicle control group (1.9 ± 0.5 mg/mm²) (Figure 12E). Both prodrugs (Rh-L-SN-38 and Mt-L-MMC) in combination with PpIX-PDT achieved significantly lower tumor volume (49 ± 15 and 88 ± 78 mm³ for PpIX-PDT + Mt-L-MMC and PpIX-PDT + Rh-L-SN-38, respectively) compared to PpIX-PDT only (176 ± 79 mm³) (Figure 12F).

Figure 11:

White light and fluorescence images of the rat bladders following the combination treatment. The top bars indicate the experiment groups (vehicle only, SN-38 only, MMC only, PpIX-PDT only, PpIX-PDT + Rh-L-SN-38, PpIX-PDT + Mt-L-MMC).

Figure 12:

Scatter plot showing in vivo efficacy with 635 nm laser with PpIX-PDT and prodrug (Rh-L-SN-38 and Mt-L-MMC) combination. (A) Total bladder area (B) tumor covered area (C) fraction of bladder covered with tumor (D) bladder weight (E) area normalized bladder weight (F) tumor volume comparison (G) mean ± standard deviation presented in figures A-F. Statistical significance (paired one-tailed t-test, **P < 0.01 and * P < 0.05).

Local or systemic toxicity:

Systemic toxicity was not expected from these organ-specific local therapies with either PDT alone or PDT-prodrug combination. We did not see significant body weight loss or signs of morbidity (appearance and behavior) after treatment. Bladders treated with 635 nm laser with PpIX-PDT and prodrug combination were evaluated for muscle damage (Figure 13). As observed with the PDT-only group (Figure 10), no apparent bladder muscle damage was observed with the combination treatment.

Figure 13:

H & E images of bladder muscle 48 hr after PpIX-PDT + prodrug treated with 635 nm laser. (A) bladder muscle of vehicle control. (B) PpIX-PDT + Rh-L-SN-38. (C) PpIX-PDT + Mt-L-MMC. Scale bar: 100 μm. BL indicates the luminal side of the bladder (not necessarily inside the bladder lumen, A, C).

DISCUSSION

To develop an orthotopic bladder tumor model, acid and base washing of the bladder is used. Once rats are euthanized and the bladder is excised, often the bladder lumen is not as clean/transparent as a fresh bladder. This creates confusion as to whether a certain area has a tumor or just a lesion or mucus or an adherent fat tissue on the apical side of the bladder. Tumor area can be confirmed with histology but that sacrifices the quantitative analysis of the antitumor efficacy such as determining tumor-covered area and tumor volume. Fluorescence imaging, developed by Khan et. al. (23) helps with bladder tumor imaging in a very simple and cost-effective manner. Using this method, we were able to detect very small tumors (~1 mm) as well as larger tumors (~10 mm or larger). Even though the emission filter is on, if illuminated with white light, tumor does not emit fluorescence (Figure 3G), confirming that the tumor fluorescence (Figure 3B) is a result of 405 nm excitation of PpIX generated by instilled HAL. Also, sometimes when excising bladders adjacent fat or other tissue may still be attached to the bladder. This technique can also distinguish between fat tissue and tumor tissue (Figure 3F). This low maintenance (only USB charging is required) tool can be a promising tool in effective tumor detection in bladder treatment efficacy determination. This method is a modification of the previously developed oral cancer detection tool using PpIX fluorescence (23).

Due to the complication from Photofrin-PDT, it is rational to use 532 nm for PpIX-PDT because 532 nm-light propagation will be limited to the bladder epithelium and lamina propria and thus would not damage the muscle layer. While 532 nm light would be safer to the muscle layer compared to 635 nm light due to its limited penetration depth (< 1 mm), in the clinic, 635 nm laser was used for PpIX-PDT in the bladder cancer (28–30). Compared to the human bladder (3 +/− 1 mm thick) (31), fresh rat bladder thickness is much smaller (~ 100 – 200 μm). Rat bladder thickness can increase (>1 mm) when there is tumor formation in the bladder epithelium (27). We utilized previously reported absorption and scattering coefficients in pig bladder tissue to estimate the light diffusion differences between 532 nm and 635 nm light. As expected, 635 nm light retains up to 64% of the incident intensity while 532 nm light intensity drops to 17% only within 5 mm depth.

In the clinic, there could be bleeding from the surgical removal of the tumor resulting in blood in the urine. Based on our experiences in the rat orthotopic tumor model, due to acid-base washing rat bladders do not seem to be fully healed on day 7 and therefore bleeding may occur when the bladder is inflated during the HAL or prodrug instillation. Thus, we wanted to evaluate the impact of blood on light propagation within the instillation solution. Due to smaller urine or instillation volume (~ 400 μl) in the rat bladder, additional bleeding can contribute to ~10 % of blood in the urine. Blood in urine significantly reduces green laser light intensity because light intensity drops by 50% with only 0.5% blood in urine. As expected, the red laser is comparatively less affected by the blood in the urine, with intensity retained as much as 50% with up to 6% blood in the urine. Thus, 635 nm laser could be better than 532 nm laser in minimizing light attenuation potentially by blood in the urine. Since whole blood used in the experiment was collected through rat heart puncture it can be a mixture of both oxygenated and deoxygenated blood. The difference between light attenuation can be attributed to the absorption coefficient of deoxygenated whole blood at 532 and 635 nm which is 18.66 mm⁻¹ and 2.10 mm⁻¹, respectively. For oxygenated whole blood absorption coefficient is about 22.48 mm⁻¹ and 0.28 mm⁻¹ for 532 and 635 nm, respectively (32).

Additionally, illumination can be done with or without washing the instillation media, while without washing can be beneficial as it provides additional instillation time and avoids drug washout. Green laser intensity dropped by 40 – 66% while red laser intensity was affected by only 7 – 12% when illuminated without removing the instillation solution. Similarly, the light attenuation for both wavelengths correlates with the concentration of each prodrug. Rh-L-SN-38 showed more pronounced attenuation at similar concentrations (200 μM for Rh-L-SN-38 vs. 250 μM for Mt-L-MMC) for 532 nm light because Rh-L-SN-38 has higher absorption than Mt-L-MMC at 532 nm (UV spectra, Figure S1). Given that both prodrugs absorb less at 635 nm than at 532 nm, the observed attenuation at 635 nm was minimal. Hence, 635 nm can provide better light diffusion in the instillation media.

While light diffusion in the instilled solution is better with the red laser, PpIX has a much smaller molar extinction coefficient (EC, ε, (M−1·cm−1)) at 635 nm, EC at 635 nm = 4,152, compared to that at 532 nm, EC at 532 nm = 11,565 (13). Also, photobleaching of PpIX was similar for both 532 nm and 635 nm. Based on the photobleaching study (Fig S2), at 20 minutes with 10 mW/cm², PpIX fluorescence intensity was reduced by 20% with both lasers. Thus, we wanted to determine if 635 nm laser can achieve at least similar efficacy compared to 532 nm light. Using an NMIBC model with AY-27 cells, we compared in vitro PpIX-PDT efficacy when using a 532 nm or 635 nm laser. Keeping the HAL concentration fixed (0.5 mM) we determined cytotoxicity with a range of light doses (2.5 to 100 mW/cm²). With lower light doses (2.5 – 10 mW/cm²) cytotoxicity was similar for both light wavelengths, but overall, 635 nm produced better efficacy as seen in IC₅₀ calculation (Fig 7B). A similar trend was also observed in the case of in vivo PpIX-PDT efficacy. In the rat orthotopic bladder tumor model, we used a range of light doses (50 – 100 mW, total power). We have observed that both light wavelengths can effectively reduce the tumor volume (Figure 9D). While 635 nm appeared to be producing better tumor volume reduction compared to 532 nm with higher light doses, we did not observe a dose-dependent increase in efficacy with 635 nm light from 50 mW to 70 mW or 100 mW total power. We also observed total bladder area increased with 635 nm laser (Figure 9A) compared to the control bladder area, which indicates there was no considerable muscle damage with 635 nm laser.

Based on the efficacy study of 532 nm and 635 nm using HAL (PpIX-PDT) only, we wanted to further improve the efficacy by combining prodrugs with PpIX-PDT. Since 635 nm can achieve better diffusion in bladder tissue and instillation media, we decided to further investigate the combination antitumor efficacy with 635 nm laser. From the comparative efficacy study, 635 nm performed comparably to 532 nm laser. Also, there was no significant improvement observed with increased light doses. Hence, we treated the PpIX-PDT + prodrugs combination groups with 635 nm laser with a total power of 50 mW. We have previously (17) observed that mitochondria-targeted prodrugs achieve better cytotoxicity than non-targeted prodrugs of the same parent drug. Therefore, Rh-L-SN-38 was used at a concentration of 200 μM in this study, while Mt-L-MMC was used at a concentration of 1 mM. Although not statistically significant, both of the prodrugs with PpIX-PDT could achieve lower mean bladder weight, lower mean fraction bladder area covered with tumor and lower mean bladder weight per unit area compared to PpIX-PDT only. Both Mt-L-MMC and Rh-L-SN-38 with PpIX-PDT could also achieve significantly better antitumor efficacy in lowering tumor-covered area and tumor volume, compared to the PpIX-PDT-only group.

Since one of the potential concerns with using a 635 nm laser in PpIX-PDT is muscle damage due to deeper penetration of red laser, we have evaluated the muscle damage after PpIX-PDT and combination treatment. Since PDT exerts rapid damage with singlet oxygen, a histology study was performed 48 hours after the treatment to detect any immediate effect on the muscle layer. No significant muscle damage was observed with either 532 nm or 635 nm lasers up to 100 mW total power (Fig 10). No apparent muscle damage or additional toxicity was observed when the prodrug was combined with PpIX-PDT (Fig 13). This was also confirmed by the total bladder area as PpIX-PDT or the combination treatment did not reduce the total bladder area (Fig 9A and 12A), which could indicate muscle damage.

CONCLUSION

In the preclinical bladder cancer study, PpIX-PDT was demonstrated to be an effective PDT strategy against NMIBC due to its milder side effects compared to Photofrin-PDT. Although 532 nm light can be safer for the muscle layer due to the limited penetration of light itself, 635 nm light could be a better choice for a more efficacious antitumor effect without damaging the muscle layer presumably due to the preferential PpIX production in tumors over bladder muscle. Furthermore, the addition of SO-cleavable prodrugs to PpIX-PDT improved the antitumor effects of PpIX-PDT. Overall, PpIX-PDT with 635 nm laser combined with prodrug was proven to be a safe and promising treatment modality for NMIBC treatment.