- Email: [email protected]
Combined modality treatment of bone metastases: response of the rhabdomyosarcoma R1H of the rat to postoperative irradiation combined with local release of daunorubicin from acrylic cement
Radiotherapy and Oncology 46 (1998) 317–320
Short communication
Combined modality treatment of bone metastases: response of the rhabdomyosarcoma R1H of the rat to postoperative irradiation combined with local release of daunorubicin from acrylic cement Gu¨nter Fro¨schle a, Jo¨rg O. Zieron b, Ulrich M. Carl b,c, Hans-Peter Beck-Bornholdt b,* a Department of Surgery, University of Hamburg, Hamburg, Germany Institute of Biophysics and Radiobiology, Martinistr. 52, University of Hamburg, D-20246 Hamburg, Germany c Department of Radiotherapy, University of Du¨sseldorf, Du¨sseldorf, Germany
b
Received 25 October 1996; revised version received 7 August 1997; accepted 11 August 1997
Abstract Reconstruction techniques for pathological fractures caused by bone metastases often include acrylic cement to stabilise defects. The aim of this study was to test in a rodent model, whether outcome could be improved by local chemotherapy from an anticancer drug added to the acrylic cement. Tumour excision 17 days after transplantation into the femur was followed by fractionated irradiation. Supplementation of acrylic cement with daunorubicin led to a considerable reduction of TCD37% from 72.9 Gy (62.2–82.8) to 29.3 Gy (21.8–37.5) (P = 0.0007). Local chemotherapy diffusing from bone cement combined with postoperative radiotherapy was highly effective in the experimental system studied. 1998 Elsevier Science Ireland Ltd. Keywords: Radiotherapy; Surgery; Chemotherapy; Experimental tumour
1. Introduction The development of bone metastases is a common and severe event for cancer patients. Bone metastases has an incidence of around 100 000 new cases per year in the United States, and is a frequent problem in patients with prostate, breast, and lung cancer [19]. Pain, pathological fractures, frequent neurologic deficits, and forced immobility caused by bone metastases significantly decrease the quality of life. They require treatment to palliate pain, prevent fractures of weight-bearing bones, and reconstruct bone defects. With few exceptions, a radical curative approach to the treatment of bone metastases is considered to be unrealistic and its attempt could risk treatment complications in the patients with incurable diseases [1,26]. However, some patients with bone metastases might have
* Corresponding author.
a relatively long life expectancy, for example in breast cancer or in case of solitary metastases [18,19,23]. It is therefore important not only to treat the metastases as safely and effectively as possible, but also to give treatment with long lasting effect [20]. A multidisciplinary approach is frequently required in management of bone metastases, because in most cases, no single method is sufficient in maintaining the patient symptom free, and a combination of systemic and local therapeutic modalities may be required. The use of chemotherapeutic agents is made difficult by their significant side effects which often complicates the management of these patients [22]. About 10% of bone metastases progress to pathological fractures with devastating consequences for the patient, particularly in long bones [3]. Patients are considered at high risk for fractures, if they show lytic lesions in weight-bearing bones with a diameter of more than 2.5 cm or where there is more than 50% cortical destruction [10]. Therefore, lytic lesions in weight-bearing bones require special attention. Operative fixation, radiation, or a combination of both
0167-8140/98/$19.00 1998 Elsevier Science Ireland Ltd. All rights reserved PII S0167-8140 (97 )0 0153-9
318
G. Fro¨schle et al. / Radiotherapy and Oncology 46 (1998) 317–320
is often considered. Surgical approach involves the reconstruction of fractured bones aiming to re-establish an acceptable quality of life. Since many reconstruction techniques use acrylic cement to bridge and stabilise defects, treatment might possibly be improved by local chemotherapy from an anticancer drug added to the acrylic cement. It is well established that numerous antibiotics added to acrylic cement diffuse into the surrounding tissues [7–9,12,17]. It has been shown that methotrexate [11,13] and daunorubicin [15,21] diffuse from polymethylmethacrylate implants and that polymerisation of the cement did not hamper drug activity. Experiments with tritiated daunorubicin have shown that relatively high concentrations of the drug were present at the implantation site. In contrast, low concentrations of the agent were observed in other organs. In particular, the high initial exposure of lung and heart, observed after i.v. administration of the drug, did not occur when the drug was added to the acrylic cement [16] or to hydroxylapatite [25]. The aim of the present study was to test the effect of a combination of postoperative radiotherapy with local chemotherapy in an experimental tumour system growing in the femur to simulate bone metastases.
2. Materials and methods The experiments were performed on the rhabdomyosarcoma R1H of the rat. The animals were provided with food and water ad libitum and were kept at periods of 12 h lights on and 12 h lights off. Animal care was in accordance with local and institutional guidelines. Further details of the tumour host system were reported previously [27]. Transplantation was performed by implanting a 1 mm3 piece of tumour tissue into the femur of male WAG/Rij albino rats (190-260 g). Animals were anaesthetised with 0.5 mg/100 g xylazin and 4.5 mg/100 g ketaminhydrochloride i.m. After shaving and local disinfection, an incision at the distal lateral femur of the rat was performed. Soft tissue and muscles were kept aside to trepanate above the lateral condyl with a 1.4 mm borer. Spongiosa was removed and the tumour piece placed in the medullary. The drill hole was closed with wax. Seventeen days after implantation tumours were excised aiming for a complete resection of all macroscopically recognisable tumour tissue without security margins, corresponding approximately to R1 resection. Histological examination revealed remaining tumour clusters at the resection zone. Median tumour mass at excision was 0.40 g (range 0.01–3 g). Because tumour resection reduces mechanical bone stability, polymethylmethacrylate was used to re-stabilise the rat bone, according to regular osteosynthetic practice. The defect was bridged and stabilised with either polymethylmethacrylate alone, or in combination with daunorubicin. The content of daunorubicin ranged from 0.9 to 1.4 mg, depending on the amount of bone cement, with content of 2% mannitol for optimal dau-
norubicin separation. Daunorubicin, mannitol, and the fluid component of polymethylmethacrylate were mixed prior to surgery, and the dry component was added during the operation. Solidification time of about 2 min was used to reconstruct the bone defect. The wound was closed with clamps. Measurements with tritiated daunorubicin at 24 and 72 h after implantation revealed high concentrations of the substance in the tumour, whereas normal tissues only showed low levels of the drug (Table 1). After 4 days of wound healing, the former tumour site was locally irradiated with 200 kVp X-rays (0.5 mm Cu filtering, 2 Gy/min dose rate) applying three fractions per week. Total doses of 50–80 Gy applied in 12 fractions were used for postoperative radiotherapy without chemotherapy (4.17–6.67 Gy per fraction), whereas animals also receiving daunorubicin were treated with total doses of 10–40 Gy applied in 5 fractions (2.0–8.0 Gy per fraction). Due to the nature of fractionation schedules it is not possible to keep constant all but one of the following five parameters: total dose, dose per fraction, overall treatment time, dose intensity and time interval between fractions. In any fractionation experiments, always at least two of these parameters are changed, leading to intrinsic problems in data interpretation. The present experiments were designed in order to apply comparable doses per fraction and dose intensity (i.e. dose per week) in both treatment arms. Earlier experiments have shown that dose per fraction has nearly no influence on response in this tumour system in the range of 0.4–12.5 Gy per fraction [4]. Unirradiated tumours with and without chemotherapy were used as controls. Animals were anaesthetised during irradiation. The following endpoints were used throughout this study: (1) Local tumour control, i.e. the percentage of animals that did not develop recurrences during the observation period of 12 months. Animals developing pulmonary metastases were censored at time of death for this endpoint. (2) Dissemination of metastases to the lungs, i.e. the percentage of animals that developed lung metastases. The animals were sacrificed Table 1 Daunorubicin concentrations in various tissues 24 and 72 h after implantation Organ
Tumoura Faeces Colon Spleen Kidney Lung Jejunum Heart Liver Brain a
mg daunorubicin per g of tissue 24 h after implantation
72 h after implantation
224 94 15 9 4 3.5 3 2.5 2 2
3005 6 4 5.5 2 2 2 1.5 2 1.5
Piece of tumour tissue was deliberately left in the femur.
G. Fro¨schle et al. / Radiotherapy and Oncology 46 (1998) 317–320
when showing distinct dyspnoea and pulmonary metastases were verified by section. TCD37% and confidence intervals were calculated applying the maximum likelihood method. Cox proportional hazards regression analysis was used to assess the independent importance of potential prognostic variables.
3. Results and discussion Histology revealed tumour growth along the bone marrow, destruction of the corticalis and infiltration of the soft tissue close to the trepanation hole. Fig. 1 shows the local tumour control obtained as a function of total dose. While for postoperative radiation alone no local tumour control was observed at doses of 50–65 Gy, 50% of the tumours receiving 72.5 or 80 Gy were locally controlled. A TCD37% of 72.9 Gy (95% confidence interval 62.2–82.8 Gy) was calculated. After combination of postoperative radiotherapy with local chemotherapy, the dose response curve is shifted to the left. At doses of 30 and 40 Gy, more than 50% of the tumours were locally controlled. TCD37% was reduced to 29.3 Gy (21.8–37.5 Gy), i.e. local chemotherapy allows the total dose to be reduced to 40%. The difference is highly significant (P = 0.0007, maximum likelihood method). While no local control (0/9) was observed with surgery alone (no radiotherapy, no local chemotherapy), one tumour (1/12) was locally controlled after local postoperative chemotherapy alone (no radiotherapy). No side effects were observed from local chemotherapy. Due to the experimental design there was a considerable heterogeneity in tumour mass at explantation. To exclude the potential pitfall of a systematic bias due to this parameter, a multivariate analysis was performed including the covariates total dose, daunorubicin (yes/no), and tumour mass at excision. The analysis verified the highly significant influence of total dose (x2 = 57.2, P , 0.00001) and of chemotherapy (x2 = 26.1, P , 0.00001) on local tumour
Fig. 1. Dose response relationship for postoperative radiotherapy of the rhabdomyosarcoma R1H of the rat growing in the cavity of the femur after radiotherapy alone (W), or after combination with local daunorubicin treatment (X). Local tumour control is plotted versus total dose. The bars indicate the 95% confidence intervals of the TCD37%. The difference is highly significant (P = 0.0007, maximum likelihood method). The number of animals at risk is given by the figures next to the data points.
319
control. Tumour mass at excision seems to play no role for recurrence in this experiment (x2 = 0.07, P = 0.8). In the experiment, not only tumour recurrence, but also metastatic dissemination into the lungs contributed considerably to disease related death of the animals. The probability of metastatic dissemination increased continuously with increasing tumour mass at excision. While no lung metastases were observed for small tumours (,0.1 g), the probability of dissemination reached 50% for large tumours (2–3 g). A multivariate analysis was performed including the same covariates as the former analysis. This analysis revealed that the probability for metastatic dissemination depended on tumour mass at excision (x2 = 4.41, P = 0.036), although this result is not significant if the significance level is adjusted for multiple testing [5]. Neither total radiation dose (x2 = 3.48, P = 0.07), nor local chemotherapy (x2 = 2.31, P = 0.13) showed a significant influence on pulmonary dissemination of the disease. Bone metastases causing pain or lytic destruction can be effectively treated by postoperative radiotherapy [26]. The aim of the present study was to investigate in a rodent tumour model the effect of a combination of these treatment modalities with a local chemotherapy allowing relatively high concentrations of the agent to be brought to the former tumour site for a sufficient duration but keeping the systemic dose low. R1H tumours growing in the femur were used to simulate bone metastases. Although this is a rather artificial system it might provide useful information concerning the treatment of such lesions. The results obtained with this tumour-host system show that local chemotherapy by daunorubicin added to polymethylmethacrylate allows the total dose required for local tumour control to be reduced considerably. The present results do not exclude the possibility that systemic chemotherapy at maximal tolerable doses could have had a similar effect on the tumour. The systemic toxicity connected with the local administration of daunorubicin with bone cement was found to be tolerable in rats with no remarkable side effects. Unfortunately, we were not able to quantify any clinically relevant side effects in our experimental system. This is obviously one of the main limitations of this study. Nevertheless, the result suggests possible new approaches in treatment of bone metastases that might allow the results to be improved by combining postoperative radiotherapy with the application of high local concentrations of chemotherapeutic agents. It has been advocated that a minority of patients with bone metastases require a higher dose of fractionated radiotherapy, particularly those where the main aim of treatment is to restore or maintain the structural integrity of bone. This issue is complicated by a possibly increased risk of pathological fracture, as shown by an 18% fracture rate in long bones of patients receiving a higher dose in the RTOG trial [2,6,24]. We hope that the application of local chemotherapy using agents diffusing from bone cement might help to solve this dilemma.
320
G. Fro¨schle et al. / Radiotherapy and Oncology 46 (1998) 317–320
The combination of surgery, local chemotherapy, and radiotherapy appears to be a promising therapeutic concept for the treatment of bone metastases or even primary tumours [14]. However, further experimental work is required to verify whether the positive results observed after local chemotherapy in the R1H tumour are a peculiarity of this tumour-host system or a more general effect. Acknowledgements The skilful technical assistance by Mrs. Maria Omniczynski is gratefully acknowledged. Supported by the Spierling-Stiftung.
References [1] Allen, K.L., Johnson, T.W. and Hibbs, G.G. Effective bone palliation as related to various palliation regimes. Cancer 37: 984–987, 1976. [2] Bates, T., Yarnold, J.R., Blitzer, P., Nelson, O.S., Rubin, P. and Maher, J. Bone metastasis consensus statement. Int. J. Radiat. Oncol. Biol. Phys. 23: 215–216, 1992. [3] Bates, T. A review of local radiotherapy in the treatment of bone metastases and cord compression. Int. J. Radiat. Oncol. Biol. Phys. 23: 217–221, 1992. [4] Beck-Bornholdt, H.-P., Maurer, T., Becker, S., Omniczynski, M., Vogler, H. and Wu¨rschmidt, F.: Radiotherapy of the rhabdomyosarcoma R1H of the rat: hyperfractionation - 126 fractions applied within 6 weeks. Int. J. Radiat. Oncol. Biol. Phys. 16: 701–705, 1989. [5] Beck-Bornholdt, H.-P. and Dubben, H.-H. Potential pitfalls in the use of P-values and in interpretation of significance levels. Radiother. Oncol. 33: 171–176, 1994. [6] Blitzer, P.H. Reanalysis of the RTOG study of the palliation of symptomatic osseous metastases. Cancer 55: 1468–1472, 1985. ¨ ber die Depotwirkung einiger [7] Buchholz, H.W. and Engelbrecht, H. U Antibiotica bei Vermischung mit dem Kunstharz Palacos. Chirurg 41: 511–515, 1970. [8] Carlsson, A.S., Josefsson, G. and Lindberg, L. Revision with gentamicin-impregnated cement for deep infections in total hip arthroplasties. J. Bone Joint Surg. 60A: 1059–1064, 1978. [9] Elson, R.A., Jephcott, A.E., McGechie, D.B. and Verettes, D. Antibiotics, loaded acrylic cement. J. Bone Joint Surg. 59B: 200–205, 1977. [10] Fidler, M. Incidence of fracture through metastasis in long bones. Acta Orthop. Scand. 52: 623–627, 1981. [11] Hernigou, Ph., Thie´ry, J.P., Benoit, J., Voisin, M.C., Leroux, P., Hagege, G., Delepine, G. and Goutallier, D. Methotrexate diffusion from acrylic cement: Local chemotherapy for bone tumours. J. Bone Joint Surg. 71B: 804–811, 1989. [12] Hoff, S.H., Fitzgerald, R.H. and Kelly, P.J. The depot administration of penicillin G and gentamicin in acrylic bone cement. J. Bone Joint Surg. 63A: 798–804, 1981. [13] Langendorff, H.-U., Jungbluth, K.H., Dingeldein, E., Wahlig, H.,
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
Delling, G. and Senokowitsch, R. Cytostaticahaltiger Knochenzement: Neue Aspekte in der Behandlung maligner Knochentumoren. Langenbecks Arch. Chir. 371: 123–136, 1987. Langendorff, H.-U. Cytostatic bone cement: experimental studies and first clinical experience. In: New Developments for Limb Salvage in Musculoskeletal Tumors, Kyoto Orthopaedic Symposium, pp. 623–632. Editor: T. Yamamuro. Springer-Verlag, Tokyo, 1989. Ma¨hlitz, J., Fro¨schle, G., Langendorff, H.-U. and Westendorf, J. In vivo sustained release of daunorubicin from acrylic cement implanted for treatment of bone metastases. Proc. Control Rel. Bioact. Mater. 21: 105, 1994. Ma¨hlitz, J., Fro¨schle, G., Westendorf, J. and Langendorff, U. Evaluation of a surgery/chemotherapy combination in the local treatment of bone metastases. Preclinical investigations in a nude mice model. J. Cancer Res. Clin. Oncol. 120 (Suppl.): 119, 1994. Marks, K.E., Nelson, C.L. and Lautenschlager, E.P. Antibioticimpregnated acrylic bone cement. J. Bone Joint Surg. 58A: 358– 364, 1976. Perez, J.E., Machiavelli, M., Leone, B.A., Romero, A., Rabinovitch, M.G., Vallejo, C.T., Bianco, A., Rodriguez, R., Cuevas, M.A. and Alvarez, L.A. Bone-only versus visceral-only metastatic pattern in breast cancer: analysis of 150 patients. A GOCS study. Am. J. Clin. Oncol. 13: 294–298, 1990. Porter, A.T. and Fontanesi, J. Palliative irradiation for bone metastasis - a new paradigm. Int. J. Radiat. Oncol. Biol. Phys. 29: 1199– 1200, 1994. Rasmusson, B., Vejborg, I., Jensen, A.B., Andersson, M., Banning, A.-M., Hoffmann, T., Pfeiffer, P., Nielsen, H.K. and Sjogren, P. Irradiation of bone metastases in breast cancer patients: a randomized study with 1 year follow-up. Radiother. Oncol. 34: 179– 184, 1995. Rudigier, J., Degreif, J., Rudig, L. and Wahlig, H. Die Reaktion des Knochenlagers auf methotrexathaltigen Knochenzement - Tierexperimentelle Untersuchungen. In: Hefte zur Unfallheilkunde 207, 268 pp. Editors: K.H. Jungbluth and A. Pannke. Springer-Verlag, Berlin, 1989. Serafini, A.N. Current status of systemic intravenous radiopharmaceuticals for the treatment of painful metastatic bone disease. Int. J. Radiat. Oncol. Biol. Phys. 30: 1187–1194, 1994. Toma, S., Venturino, A., Sogno, G., Formica, C., Bignotti, B., Bonassi, S. and Palumbo, R. Metastatic bone tumors. Non-surgical treatment. Outcome and survival. Clin. Orthop. 295: 246–251, 1993. Tong, D., Gillick, L. and Hendrickson, F.R. The palliation of symptomatic osseous metastases. Final results of the Radiation Therapy Oncology Group. Cancer 50: 893–899, 1982. Uchida, A., Shinto, Y., Araki, N. and Ono, K. Slow release of anticancer drugs from porous calcium hydroxylapatite ceramic. J. Orthop. Res. 10: 440–445, 1992. Ultmann, J.E., Phillips, T.L. and Mauch, P.M. Treatment of metastatic cancer to bone. In: Cancer. Principles and Practice of Oncology. pp. 1564–1568. Editors: V. DeVita, T. Hellman, S. Rosenberg. Lippincott, Philadelphia, PA, 1982. Vogler, H. and Beck-Bornholdt, H.-P. Radiotherapy of the rhabdomyosarcoma R1H of the rat: kinetics of cellular inactivation by fractionated irradiation. Int. J. Radiat. Oncol. Biol. Phys. 14: 317– 325, 1988.
Short communication
Combined modality treatment of bone metastases: response of the rhabdomyosarcoma R1H of the rat to postoperative irradiation combined with local release of daunorubicin from acrylic cement Gu¨nter Fro¨schle a, Jo¨rg O. Zieron b, Ulrich M. Carl b,c, Hans-Peter Beck-Bornholdt b,* a Department of Surgery, University of Hamburg, Hamburg, Germany Institute of Biophysics and Radiobiology, Martinistr. 52, University of Hamburg, D-20246 Hamburg, Germany c Department of Radiotherapy, University of Du¨sseldorf, Du¨sseldorf, Germany
b
Received 25 October 1996; revised version received 7 August 1997; accepted 11 August 1997
Abstract Reconstruction techniques for pathological fractures caused by bone metastases often include acrylic cement to stabilise defects. The aim of this study was to test in a rodent model, whether outcome could be improved by local chemotherapy from an anticancer drug added to the acrylic cement. Tumour excision 17 days after transplantation into the femur was followed by fractionated irradiation. Supplementation of acrylic cement with daunorubicin led to a considerable reduction of TCD37% from 72.9 Gy (62.2–82.8) to 29.3 Gy (21.8–37.5) (P = 0.0007). Local chemotherapy diffusing from bone cement combined with postoperative radiotherapy was highly effective in the experimental system studied. 1998 Elsevier Science Ireland Ltd. Keywords: Radiotherapy; Surgery; Chemotherapy; Experimental tumour
1. Introduction The development of bone metastases is a common and severe event for cancer patients. Bone metastases has an incidence of around 100 000 new cases per year in the United States, and is a frequent problem in patients with prostate, breast, and lung cancer [19]. Pain, pathological fractures, frequent neurologic deficits, and forced immobility caused by bone metastases significantly decrease the quality of life. They require treatment to palliate pain, prevent fractures of weight-bearing bones, and reconstruct bone defects. With few exceptions, a radical curative approach to the treatment of bone metastases is considered to be unrealistic and its attempt could risk treatment complications in the patients with incurable diseases [1,26]. However, some patients with bone metastases might have
* Corresponding author.
a relatively long life expectancy, for example in breast cancer or in case of solitary metastases [18,19,23]. It is therefore important not only to treat the metastases as safely and effectively as possible, but also to give treatment with long lasting effect [20]. A multidisciplinary approach is frequently required in management of bone metastases, because in most cases, no single method is sufficient in maintaining the patient symptom free, and a combination of systemic and local therapeutic modalities may be required. The use of chemotherapeutic agents is made difficult by their significant side effects which often complicates the management of these patients [22]. About 10% of bone metastases progress to pathological fractures with devastating consequences for the patient, particularly in long bones [3]. Patients are considered at high risk for fractures, if they show lytic lesions in weight-bearing bones with a diameter of more than 2.5 cm or where there is more than 50% cortical destruction [10]. Therefore, lytic lesions in weight-bearing bones require special attention. Operative fixation, radiation, or a combination of both
0167-8140/98/$19.00 1998 Elsevier Science Ireland Ltd. All rights reserved PII S0167-8140 (97 )0 0153-9
318
G. Fro¨schle et al. / Radiotherapy and Oncology 46 (1998) 317–320
is often considered. Surgical approach involves the reconstruction of fractured bones aiming to re-establish an acceptable quality of life. Since many reconstruction techniques use acrylic cement to bridge and stabilise defects, treatment might possibly be improved by local chemotherapy from an anticancer drug added to the acrylic cement. It is well established that numerous antibiotics added to acrylic cement diffuse into the surrounding tissues [7–9,12,17]. It has been shown that methotrexate [11,13] and daunorubicin [15,21] diffuse from polymethylmethacrylate implants and that polymerisation of the cement did not hamper drug activity. Experiments with tritiated daunorubicin have shown that relatively high concentrations of the drug were present at the implantation site. In contrast, low concentrations of the agent were observed in other organs. In particular, the high initial exposure of lung and heart, observed after i.v. administration of the drug, did not occur when the drug was added to the acrylic cement [16] or to hydroxylapatite [25]. The aim of the present study was to test the effect of a combination of postoperative radiotherapy with local chemotherapy in an experimental tumour system growing in the femur to simulate bone metastases.
2. Materials and methods The experiments were performed on the rhabdomyosarcoma R1H of the rat. The animals were provided with food and water ad libitum and were kept at periods of 12 h lights on and 12 h lights off. Animal care was in accordance with local and institutional guidelines. Further details of the tumour host system were reported previously [27]. Transplantation was performed by implanting a 1 mm3 piece of tumour tissue into the femur of male WAG/Rij albino rats (190-260 g). Animals were anaesthetised with 0.5 mg/100 g xylazin and 4.5 mg/100 g ketaminhydrochloride i.m. After shaving and local disinfection, an incision at the distal lateral femur of the rat was performed. Soft tissue and muscles were kept aside to trepanate above the lateral condyl with a 1.4 mm borer. Spongiosa was removed and the tumour piece placed in the medullary. The drill hole was closed with wax. Seventeen days after implantation tumours were excised aiming for a complete resection of all macroscopically recognisable tumour tissue without security margins, corresponding approximately to R1 resection. Histological examination revealed remaining tumour clusters at the resection zone. Median tumour mass at excision was 0.40 g (range 0.01–3 g). Because tumour resection reduces mechanical bone stability, polymethylmethacrylate was used to re-stabilise the rat bone, according to regular osteosynthetic practice. The defect was bridged and stabilised with either polymethylmethacrylate alone, or in combination with daunorubicin. The content of daunorubicin ranged from 0.9 to 1.4 mg, depending on the amount of bone cement, with content of 2% mannitol for optimal dau-
norubicin separation. Daunorubicin, mannitol, and the fluid component of polymethylmethacrylate were mixed prior to surgery, and the dry component was added during the operation. Solidification time of about 2 min was used to reconstruct the bone defect. The wound was closed with clamps. Measurements with tritiated daunorubicin at 24 and 72 h after implantation revealed high concentrations of the substance in the tumour, whereas normal tissues only showed low levels of the drug (Table 1). After 4 days of wound healing, the former tumour site was locally irradiated with 200 kVp X-rays (0.5 mm Cu filtering, 2 Gy/min dose rate) applying three fractions per week. Total doses of 50–80 Gy applied in 12 fractions were used for postoperative radiotherapy without chemotherapy (4.17–6.67 Gy per fraction), whereas animals also receiving daunorubicin were treated with total doses of 10–40 Gy applied in 5 fractions (2.0–8.0 Gy per fraction). Due to the nature of fractionation schedules it is not possible to keep constant all but one of the following five parameters: total dose, dose per fraction, overall treatment time, dose intensity and time interval between fractions. In any fractionation experiments, always at least two of these parameters are changed, leading to intrinsic problems in data interpretation. The present experiments were designed in order to apply comparable doses per fraction and dose intensity (i.e. dose per week) in both treatment arms. Earlier experiments have shown that dose per fraction has nearly no influence on response in this tumour system in the range of 0.4–12.5 Gy per fraction [4]. Unirradiated tumours with and without chemotherapy were used as controls. Animals were anaesthetised during irradiation. The following endpoints were used throughout this study: (1) Local tumour control, i.e. the percentage of animals that did not develop recurrences during the observation period of 12 months. Animals developing pulmonary metastases were censored at time of death for this endpoint. (2) Dissemination of metastases to the lungs, i.e. the percentage of animals that developed lung metastases. The animals were sacrificed Table 1 Daunorubicin concentrations in various tissues 24 and 72 h after implantation Organ
Tumoura Faeces Colon Spleen Kidney Lung Jejunum Heart Liver Brain a
mg daunorubicin per g of tissue 24 h after implantation
72 h after implantation
224 94 15 9 4 3.5 3 2.5 2 2
3005 6 4 5.5 2 2 2 1.5 2 1.5
Piece of tumour tissue was deliberately left in the femur.
G. Fro¨schle et al. / Radiotherapy and Oncology 46 (1998) 317–320
when showing distinct dyspnoea and pulmonary metastases were verified by section. TCD37% and confidence intervals were calculated applying the maximum likelihood method. Cox proportional hazards regression analysis was used to assess the independent importance of potential prognostic variables.
3. Results and discussion Histology revealed tumour growth along the bone marrow, destruction of the corticalis and infiltration of the soft tissue close to the trepanation hole. Fig. 1 shows the local tumour control obtained as a function of total dose. While for postoperative radiation alone no local tumour control was observed at doses of 50–65 Gy, 50% of the tumours receiving 72.5 or 80 Gy were locally controlled. A TCD37% of 72.9 Gy (95% confidence interval 62.2–82.8 Gy) was calculated. After combination of postoperative radiotherapy with local chemotherapy, the dose response curve is shifted to the left. At doses of 30 and 40 Gy, more than 50% of the tumours were locally controlled. TCD37% was reduced to 29.3 Gy (21.8–37.5 Gy), i.e. local chemotherapy allows the total dose to be reduced to 40%. The difference is highly significant (P = 0.0007, maximum likelihood method). While no local control (0/9) was observed with surgery alone (no radiotherapy, no local chemotherapy), one tumour (1/12) was locally controlled after local postoperative chemotherapy alone (no radiotherapy). No side effects were observed from local chemotherapy. Due to the experimental design there was a considerable heterogeneity in tumour mass at explantation. To exclude the potential pitfall of a systematic bias due to this parameter, a multivariate analysis was performed including the covariates total dose, daunorubicin (yes/no), and tumour mass at excision. The analysis verified the highly significant influence of total dose (x2 = 57.2, P , 0.00001) and of chemotherapy (x2 = 26.1, P , 0.00001) on local tumour
Fig. 1. Dose response relationship for postoperative radiotherapy of the rhabdomyosarcoma R1H of the rat growing in the cavity of the femur after radiotherapy alone (W), or after combination with local daunorubicin treatment (X). Local tumour control is plotted versus total dose. The bars indicate the 95% confidence intervals of the TCD37%. The difference is highly significant (P = 0.0007, maximum likelihood method). The number of animals at risk is given by the figures next to the data points.
319
control. Tumour mass at excision seems to play no role for recurrence in this experiment (x2 = 0.07, P = 0.8). In the experiment, not only tumour recurrence, but also metastatic dissemination into the lungs contributed considerably to disease related death of the animals. The probability of metastatic dissemination increased continuously with increasing tumour mass at excision. While no lung metastases were observed for small tumours (,0.1 g), the probability of dissemination reached 50% for large tumours (2–3 g). A multivariate analysis was performed including the same covariates as the former analysis. This analysis revealed that the probability for metastatic dissemination depended on tumour mass at excision (x2 = 4.41, P = 0.036), although this result is not significant if the significance level is adjusted for multiple testing [5]. Neither total radiation dose (x2 = 3.48, P = 0.07), nor local chemotherapy (x2 = 2.31, P = 0.13) showed a significant influence on pulmonary dissemination of the disease. Bone metastases causing pain or lytic destruction can be effectively treated by postoperative radiotherapy [26]. The aim of the present study was to investigate in a rodent tumour model the effect of a combination of these treatment modalities with a local chemotherapy allowing relatively high concentrations of the agent to be brought to the former tumour site for a sufficient duration but keeping the systemic dose low. R1H tumours growing in the femur were used to simulate bone metastases. Although this is a rather artificial system it might provide useful information concerning the treatment of such lesions. The results obtained with this tumour-host system show that local chemotherapy by daunorubicin added to polymethylmethacrylate allows the total dose required for local tumour control to be reduced considerably. The present results do not exclude the possibility that systemic chemotherapy at maximal tolerable doses could have had a similar effect on the tumour. The systemic toxicity connected with the local administration of daunorubicin with bone cement was found to be tolerable in rats with no remarkable side effects. Unfortunately, we were not able to quantify any clinically relevant side effects in our experimental system. This is obviously one of the main limitations of this study. Nevertheless, the result suggests possible new approaches in treatment of bone metastases that might allow the results to be improved by combining postoperative radiotherapy with the application of high local concentrations of chemotherapeutic agents. It has been advocated that a minority of patients with bone metastases require a higher dose of fractionated radiotherapy, particularly those where the main aim of treatment is to restore or maintain the structural integrity of bone. This issue is complicated by a possibly increased risk of pathological fracture, as shown by an 18% fracture rate in long bones of patients receiving a higher dose in the RTOG trial [2,6,24]. We hope that the application of local chemotherapy using agents diffusing from bone cement might help to solve this dilemma.
320
G. Fro¨schle et al. / Radiotherapy and Oncology 46 (1998) 317–320
The combination of surgery, local chemotherapy, and radiotherapy appears to be a promising therapeutic concept for the treatment of bone metastases or even primary tumours [14]. However, further experimental work is required to verify whether the positive results observed after local chemotherapy in the R1H tumour are a peculiarity of this tumour-host system or a more general effect. Acknowledgements The skilful technical assistance by Mrs. Maria Omniczynski is gratefully acknowledged. Supported by the Spierling-Stiftung.
References [1] Allen, K.L., Johnson, T.W. and Hibbs, G.G. Effective bone palliation as related to various palliation regimes. Cancer 37: 984–987, 1976. [2] Bates, T., Yarnold, J.R., Blitzer, P., Nelson, O.S., Rubin, P. and Maher, J. Bone metastasis consensus statement. Int. J. Radiat. Oncol. Biol. Phys. 23: 215–216, 1992. [3] Bates, T. A review of local radiotherapy in the treatment of bone metastases and cord compression. Int. J. Radiat. Oncol. Biol. Phys. 23: 217–221, 1992. [4] Beck-Bornholdt, H.-P., Maurer, T., Becker, S., Omniczynski, M., Vogler, H. and Wu¨rschmidt, F.: Radiotherapy of the rhabdomyosarcoma R1H of the rat: hyperfractionation - 126 fractions applied within 6 weeks. Int. J. Radiat. Oncol. Biol. Phys. 16: 701–705, 1989. [5] Beck-Bornholdt, H.-P. and Dubben, H.-H. Potential pitfalls in the use of P-values and in interpretation of significance levels. Radiother. Oncol. 33: 171–176, 1994. [6] Blitzer, P.H. Reanalysis of the RTOG study of the palliation of symptomatic osseous metastases. Cancer 55: 1468–1472, 1985. ¨ ber die Depotwirkung einiger [7] Buchholz, H.W. and Engelbrecht, H. U Antibiotica bei Vermischung mit dem Kunstharz Palacos. Chirurg 41: 511–515, 1970. [8] Carlsson, A.S., Josefsson, G. and Lindberg, L. Revision with gentamicin-impregnated cement for deep infections in total hip arthroplasties. J. Bone Joint Surg. 60A: 1059–1064, 1978. [9] Elson, R.A., Jephcott, A.E., McGechie, D.B. and Verettes, D. Antibiotics, loaded acrylic cement. J. Bone Joint Surg. 59B: 200–205, 1977. [10] Fidler, M. Incidence of fracture through metastasis in long bones. Acta Orthop. Scand. 52: 623–627, 1981. [11] Hernigou, Ph., Thie´ry, J.P., Benoit, J., Voisin, M.C., Leroux, P., Hagege, G., Delepine, G. and Goutallier, D. Methotrexate diffusion from acrylic cement: Local chemotherapy for bone tumours. J. Bone Joint Surg. 71B: 804–811, 1989. [12] Hoff, S.H., Fitzgerald, R.H. and Kelly, P.J. The depot administration of penicillin G and gentamicin in acrylic bone cement. J. Bone Joint Surg. 63A: 798–804, 1981. [13] Langendorff, H.-U., Jungbluth, K.H., Dingeldein, E., Wahlig, H.,
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
Delling, G. and Senokowitsch, R. Cytostaticahaltiger Knochenzement: Neue Aspekte in der Behandlung maligner Knochentumoren. Langenbecks Arch. Chir. 371: 123–136, 1987. Langendorff, H.-U. Cytostatic bone cement: experimental studies and first clinical experience. In: New Developments for Limb Salvage in Musculoskeletal Tumors, Kyoto Orthopaedic Symposium, pp. 623–632. Editor: T. Yamamuro. Springer-Verlag, Tokyo, 1989. Ma¨hlitz, J., Fro¨schle, G., Langendorff, H.-U. and Westendorf, J. In vivo sustained release of daunorubicin from acrylic cement implanted for treatment of bone metastases. Proc. Control Rel. Bioact. Mater. 21: 105, 1994. Ma¨hlitz, J., Fro¨schle, G., Westendorf, J. and Langendorff, U. Evaluation of a surgery/chemotherapy combination in the local treatment of bone metastases. Preclinical investigations in a nude mice model. J. Cancer Res. Clin. Oncol. 120 (Suppl.): 119, 1994. Marks, K.E., Nelson, C.L. and Lautenschlager, E.P. Antibioticimpregnated acrylic bone cement. J. Bone Joint Surg. 58A: 358– 364, 1976. Perez, J.E., Machiavelli, M., Leone, B.A., Romero, A., Rabinovitch, M.G., Vallejo, C.T., Bianco, A., Rodriguez, R., Cuevas, M.A. and Alvarez, L.A. Bone-only versus visceral-only metastatic pattern in breast cancer: analysis of 150 patients. A GOCS study. Am. J. Clin. Oncol. 13: 294–298, 1990. Porter, A.T. and Fontanesi, J. Palliative irradiation for bone metastasis - a new paradigm. Int. J. Radiat. Oncol. Biol. Phys. 29: 1199– 1200, 1994. Rasmusson, B., Vejborg, I., Jensen, A.B., Andersson, M., Banning, A.-M., Hoffmann, T., Pfeiffer, P., Nielsen, H.K. and Sjogren, P. Irradiation of bone metastases in breast cancer patients: a randomized study with 1 year follow-up. Radiother. Oncol. 34: 179– 184, 1995. Rudigier, J., Degreif, J., Rudig, L. and Wahlig, H. Die Reaktion des Knochenlagers auf methotrexathaltigen Knochenzement - Tierexperimentelle Untersuchungen. In: Hefte zur Unfallheilkunde 207, 268 pp. Editors: K.H. Jungbluth and A. Pannke. Springer-Verlag, Berlin, 1989. Serafini, A.N. Current status of systemic intravenous radiopharmaceuticals for the treatment of painful metastatic bone disease. Int. J. Radiat. Oncol. Biol. Phys. 30: 1187–1194, 1994. Toma, S., Venturino, A., Sogno, G., Formica, C., Bignotti, B., Bonassi, S. and Palumbo, R. Metastatic bone tumors. Non-surgical treatment. Outcome and survival. Clin. Orthop. 295: 246–251, 1993. Tong, D., Gillick, L. and Hendrickson, F.R. The palliation of symptomatic osseous metastases. Final results of the Radiation Therapy Oncology Group. Cancer 50: 893–899, 1982. Uchida, A., Shinto, Y., Araki, N. and Ono, K. Slow release of anticancer drugs from porous calcium hydroxylapatite ceramic. J. Orthop. Res. 10: 440–445, 1992. Ultmann, J.E., Phillips, T.L. and Mauch, P.M. Treatment of metastatic cancer to bone. In: Cancer. Principles and Practice of Oncology. pp. 1564–1568. Editors: V. DeVita, T. Hellman, S. Rosenberg. Lippincott, Philadelphia, PA, 1982. Vogler, H. and Beck-Bornholdt, H.-P. Radiotherapy of the rhabdomyosarcoma R1H of the rat: kinetics of cellular inactivation by fractionated irradiation. Int. J. Radiat. Oncol. Biol. Phys. 14: 317– 325, 1988.