Treatment of Canine Osseous Tumors with Photodynamic Therapy: A Pilot Study
- First Online:
- Cite this article as:
- Burch, S., London, C., Seguin, B. et al. Clin Orthop Relat Res (2009) 467: 1028. doi:10.1007/s11999-008-0678-5
- 318 Downloads
Photodynamic therapy uses nonthermal coherent light delivered via fiber optic cable to locally activate a photosensitive chemotherapeutic agent that ablates tumor tissue. Owing to the limitations of light penetration, it is unknown whether photodynamic therapy can treat large osseous tumors. We determined whether photodynamic therapy can induce necrosis in large osseous tumors, and if so, to quantify the volume of treated tissue. In a pilot study we treated seven dogs with spontaneous osteosarcomas of the distal radius. Tumors were imaged with MRI before and 48 hours after treatment, and the volumes of hypointense regions were compared. The treated limbs were amputated immediately after imaging at 48 hours and sectioned corresponding to the MR axial images. We identified tumor necrosis histologically; the regions of necrosis corresponded anatomically to hypointense tissue on MRI. The mean volume of necrotic tissue seen on MRI after photodynamic therapy was 21,305 mm3 compared with a pretreatment volume of 6108 mm3. These pilot data suggest photodynamic therapy penetrates relatively large canine osseous tumors and may be a useful adjunct for treatment of bone tumors.
Photodynamic therapy (PDT) uses a nonthermal laser delivered light through a fiber optic cable to stimulate a photoactivated chemotherapeutic agent (photosensitizer) that ablates tumor cells [10, 16, 17, 23, 26, 30, 33, 36]. PDT has limited systemic and local side effects [3, 32, 33]. Previous rodent studies suggest PDT can ablate various forms of bone cancer [4, 5, 31, 37], although our concern has been whether PDT can induce necrosis in larger osseous tumors given presumed limitations of light penetration in osseous tumors.
The purpose of this study was to determine if PDT could induce necrosis in large osseous tumors and to quantify the volume of tumor tissue ablated.
Materials and Methods
For this pilot study, we identified seven large-breed pet dogs (33–65.5 kg) with spontaneous osteosarcomas in the forelimb distal radius through the oncology clinic at the University of California, Davis, School of Veterinary Medicine. The radiographs and MR images of all the animals tested were consistent with osteosarcoma and later confirmed by histology. All tumors were classified as Stage IIb [12, 13] and all of the tumors in the dogs except one had areas of spontaneous necrosis present before therapy identified on MRI (T1 gadolinium sequences). Consent for treatment was obtained from the owners of the dogs. All procedures were approved by the Animal Care Committee at the University of California, Davis, School of Veterinary Medicine.
The affected limb was prepared and draped in sterile fashion. A 2-mm stainless steel guide rod was placed through a small stab incision over the proximal anterior surface of the affected radius and advanced through the intramedullary canal of the radius into the osteosarcoma under fluoroscopic guidance. A cannulated trochar was placed over the guide rod and advanced to the center of the tumor. The guide rod then was removed and a 0.94-mm fiber optic cable was advanced through the guide rod and placed in the tumor. The location of the cannulated trochar and placement of the treatment fiber in the tumor were confirmed by intraoperative fluoroscopy. The photosensitizer verteporfin (benzoporphyrin derivative monoacid; QLT, Inc, Vancouver, Canada) then was administered systemically (0.4 mg/kg) through an intravenous access portal over 10 minutes before light delivery. Light from a 690-nm diode laser was administered 5 minutes after completion of the transfusion of the photosensitizer. Light was delivered at 250 mW/cm over 1000 seconds through a 0.94-mm × 5-cm cylindrical diffuser (Medlight SA, Ecublens, Switzerland) for a total light dose of 500 J/cm. One animal was used as a control in which light was administered at 500 J/cm without administration of the photosensitizer (sham treatment). The fiber optic cable was removed after treatment. The animals were extubated and monitored in the intensive care unit for 48 hours where analgesia was maintained. Forty-eight hours posttreatment, the animals returned to the MRI suite anesthetized and the PDT-treated limbs were reimaged using the pretreatment MRI protocol. Immediately after imaging, the animals were taken to the procedure room and the affected forelimbs were amputated with wide margins, in conventional fashion. The dogs were followed clinically in the oncology unit postoperatively and received routine neoadjuvant cisplatinum.
The forelimb of the dog was amputated and sectioned from the wrist. Seventeen sections were cut in the axial plane to correlate with alternate axial MR images based on the distance from the distal radius. The specimens then were decalcified in 10% formic acid over 28 days and sectioned using a microtome. The sections from each amputated limb were fixed on microscopy slides, stained with hematoxylin and eosin, and analyzed under light microscopy by one of the authors (SB) and a dedicated musculoskeletal pathologist (RK). Anatomic features of the histology sections were used to confirm the correlation between MRI and histology sections. The tissue around the fiber optic cable tract was assessed microscopically.
Finally, the pattern and distribution of hypointense signal seen on the axial MRI were compared with the corresponding axial histology sections to assess whether the pattern and distribution of hypointense tissue posttreatment correlated with necrosis on histology. This was performed by comparing anatomic features of the histology sections to anatomic features of the MRI slices taken equidistant from the wrist.
Volume of necrosis in canine osteosarcomas after PDT
The distribution of hypointense signal seen on T1 gadolinium sequences matched the pattern and distribution of necrosis seen on the corresponding histology sections (Figs. 2, 3). Likewise, the axial dimension of hypointense tissue on the control’s posttreatment MRI was minimal and determined to be 3 mm, corresponding to the passage of the cannula.
PDT uses nonthermal coherent light delivered via fiber optic cable to locally activate a photosensitive chemotherapeutic agent that ablates tumor tissue. It is unknown whether PDT can treat large osseous tumors owing to the limitations of light penetration. In this pilot study, we determined if PDT can induce necrosis in large osseous tumors and quantified the volume of treated tissue in seven dogs with spontaneous osteosarcomas of the distal radius.
Osteosarcomas are common in dogs and commonly affect the radius. The characteristics of canine osteosarcomas are histologically similar to those of human osteosarcomas and we therefore presumed this was a good model to test our hypothesis. Limits to this study include the limited number of dogs that were treated and the limited number of control animals. However, this was a pilot study, and owing to ethical reasons, a limited number of animals were used with limited followup (48 hours) before amputation. The long-term clinical followup of the dogs was not reported because the dogs were lost to followup (<1 year), making it difficult to determine whether the treatment had either a positive or negative effect on the dog’s survival. Likewise, a potential downside of the technique used is potential seeding of tumor tissue into adjacent tissue during placement and withdrawal of the fiber optic cable and hemorrhage secondary to fiber placement. Our assessments were not blinded, although we did not believe this essential for this pilot study. Finally, our study was limited in that there was variability in lesion sizes produced in animals with the same type of tumor. This can be explained in part by the following: the osteosarcomas in this study were all Stage IIb and some dogs had areas of spontaneous necrosis noted preoperatively. Placement of fibers in an area of preexisting necrosis served to limit the measurable effect attributed to PDT and skewed the results negatively. For example, the cylindrical diffusing tip was placed in an area of preexisting necrosis in the tumor of one animal (Dog JR, Table 1). The measurable lesion in this animal’s osteosarcoma was limited on the post-PDT MRI because the preexisting necrosis was indiscernible in the mediolateral plane from the lesion created from the PDT, rather than because there was a poor response to treatment in that tumor. In contrast, in animals in which there was no or limited spontaneous necrosis preoperatively, large areas of necrosis from PDT treatment were determined easily.
We observed necrosis histologically and on MRI. Although necrosis was seen in all tumors except one before PDT, the extent of necrosis after PDT was substantial in all of the PDT-treated radii except for the control. The hypointense tissue on MRI matched the area of necrosis seen on histologic specimen, and we were able to obtain a volumetric measurement from the MRI to determine the volume of necrosis after PDT was substantially increased.
The effect of PDT is mediated through direct ablation and immune-mediated vascular stasis [11, 16]. We found a mean treatment effect size of approximately 16 cm3, with a consistent effect seen in the anteroposterior (2.5 cm) and mediolateral (2.6 cm) dimensions. The effect measured along the rostrocaudal (5.2 cm) plane correlated well with the length of the diffusing tip used for treatment in this study (5 cm).
Our results are consistent with those of several trials in rodents largely consisting of experiments in orthotopic osteosarcoma, chondrosarcoma, and fibrosarcoma models [8, 14, 20, 28, 31]. In these studies, PDT was able to ablate primary bone tumors, although the size of the tumors was small. Several other animal studies showed PDT can induce necrosis in metastatic adenocarcinomas in bone as well [4, 6]. Burch et al.  showed, with a bioluminescent metastatic rat model, PDT ablated human breast cancer in the axial and appendicular skeleton. Several other early-phase clinical trials have shown PDT is capable of ablating various primary human cancers, including breast, prostate, bladder, and lung cancer [2, 9, 10, 15, 17, 18, 24, 27, 29, 38]. However, the extent of ablation was limited in most of these studies because the light source had to be placed on the surface of the diseased tissue and not directly in the tumor. In contrast, the fiber can be placed in the center of the tumor in osseous tumors. It may be the skeleton is the ideal place for PDT since the fiber optic cable can be placed directly in the bone without compromising the native tissue.
Our data show PDT can induce tumor necrosis in relatively large bone tumors and the light penetration does not appear to be a limiting factor in the animals treated. The treatment protocol in our study used one 0.94-mm fiber. However, the treatment effect could be increased by using multiple fibers placed accordingly. Arguably, complete ablation of the osteosarcoma in a dog may have been obtained if several fibers were appropriately arranged in the tumor in the medullary canal. The technique also is applicable to the axial skeleton, and Burch et al.  also examined safe application of low-dose PDT in porcine vertebrae and canine vertebrae. The average volume of necrosis in dogs in the current study is approximately the size of large metastatic lesions seen in the thoracic and lumbar spine [1, 11]. PDT may have potential as an adjunctive therapy for treatment of spinal metastases along with radiation and bisphosphonates [22, 25, 34, 35, 39].
Our pilot data suggest PDT can ablate a considerable volume of large bone tumors and the light penetration in osseous tumors is not a limiting factor. The other benefits of PDT are that it is minimally invasive (the treatment fiber is less than 1 mm in diameter); because it is targeted, it has limited systemic and local side effects and repeat treatments can be made; the treatment time is, on average, extremely short (33 minutes); and it could easily be used as an adjunct with other strategies to treat osseous tumors.
We thank Cynthia Chin, MD, Neuroradiology, University of California, San Francisco, for reviewing the MR images of the animals treated in this study and Rita Kandel, MD, Pathology, Mount Sinai Hospital, University of Toronto, for reviewing the histology slides of the animals treated in this study. We also thank QLT, Inc, for providing pharmaceutical support for the project by supplying the photosensitizer verteporfin.
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.