Radiation therapy for giant cell tumor of bone

0360-3016/93 $6.00 + .oO Copyright Q 1993 Pergamon Press Ltd.

Inr. J Radiation Oncologv Rio/. Phys , Vol. 26, PP. 299-304 Printed in the UXA All rights reserved

0 Brief Communication RADIATION C. JOSEPH BENNETT,

JR.,

THERAPY

FOR GIANT

M.D.,* ROBERT B. MARCUS, AND WILLIAM

CELL JR.,

F. ENNEIUNG,

TUMOR

OF BONE

M.D.,* RODNEY M.D.+

R. MILLION,

M.D.*

University of Florida College of Medicine, Gainesville, FL Purpose: Giant cell tumor of hone is usually treated with surgical curettage. For recurrent tumors, tumors that are inoperable because of location, and tumors that would require amputation or another radical procedure limiting function, does radiotherapy provide an alternative for local control? Methods and Materials: Sixteen patients with histologically confirmed, giant cell tumor of bone were treated at the University of Florida with irradiation between March 1973 and September 1988. Minimum follow-up was 32 months; 63% of the patients had follow-up for at least 5 years, 44% for greater than 10 years. All sites received doses of 35 Gy or more, and all were treated with megavoltage irradiation. Results: In 12 (75%) of 16 patients, the tumor was controlled locally with irradiation. The four failures occurred at 8, 13, 13, and 25 months following initiation of treatment. Surgical salvage was successful in all four failures for an overall local control rate of 100%. One patient developed pulmonary metastasis 1 month after surgical salvage and is alive without evidence of disease after multiple courses of chemotherapy, surgical resection, and whole-lung irradiation. All patients tolerated the treatment well with no severe or chronic complications. No secondary soft-tissue sarcomas have occurred within the irradiated areas. Conclusion: Giant cell tumor of bone is not a radioresistant tumor as once believed, and complications seen with modern treatment regimens are minor. Bone neoplasms, Giant cell tumor, Radiotherapy.

malignant fibrous histiocytoma (7). This high rate of transformation (20%) was seen in the era before megavoltage therapy (4, 7). With the use of megavoltage irradiation, this incidence seems to be extremely low, but further follow-up is needed. Giant cell tumors have traditionally been managed with surgery alone. Local recurrence rates range from 40% to 60% for patients treated with simple surgical curettage (5, 7, 10, 14, 19). Many patients undergo multiple surgical procedures in an attempt to obtain local control. Over the past 10 years, reports in the literature have shown that radiation therapy is effective in obtaining local control of these lesions with minimal side effects or complications (1, 6, 8, 17, 18). This paper presents the University of Florida experience with both primary and recurrent giant cell tumors of bone treated with irradiation.

INTRODUCTION

Giant cell tumor of bone is a locally aggressive tumor with a high rate of local recurrence and a low potential for distant metastasis. The incidence of giant cell tumor is low, representing approximately 5% of all tumors of the skeleton (7). In the Orient, this frequency increases to 20% of all primary skeletal neoplasms (7, 19). Giant cell tumor of bone usually occurs in individuals 18 years old or older, with a peak incidence between 20 and 40 years (4, 10). There is a slight predominance of women (52% to 55%), with this margin increasing for patients less than 20 years old (75%) (7, 10, 16). A malignant or highgrade form of giant cell tumor is occasionally seen in untreated lesions. This appears to have no prognostic significance (7). A sarcoma may also occur in conjunction with a giant cell tumor (5). A more frequent concern regarding radiation treatment of low-grade lesions is the documentation of “malignant transformation” of these tumors in the site of previous irradiation. When this occurs, it is referred to as a malignant giant cell tumor, usually in the form of an osteosarcoma, fibrosarcoma, or a

METHODS

AND

MATERIALS

Between March 1973 and September 1988, 16 patients were treated for monostotic giant cell tumor of bone with

* Dept. of Radiation Oncology.

Radiation Oncology, University of Florida Health Science CenPO Box 100385, Gainesville, F’L 32610-0385. Accepted for publication 3 December 1992.

t Dept. of Orthopaedic Surgery. Reprint requests to: Robert B. Marcus, Jr., Department of

ter,

299

300

I. J. Radiation Oncology 0 Biology 0 Physics

radiation therapy at the University of Florida. None had received prior irradiation or chemotherapy. The University of Florida Department of Pathology reviewed the pathologic material and histologically confirmed giant cell tumor for all patients. Work-up of the primary lesion included skeletal roentgenogram, angiography, computed tomography (CT) (after 1979), magnetic resonance imaging (MRI) (after 1984) and bone scan. A chest roentgenogram was also obtained in all patients. Patients ranged in age from 16 to 75 years (mean age, 35 years). Ten patients were women, six were men (1.67: 1). Twelve patients were white, four were black. No cases were associated with hyperparathyroidism or Paget’s disease. The distribution of treated sites is shown in Figure 1. It is our policy that surgical curettage with or without methacrylate cementation, or en bloc resection with a wide surgical margin, depending on the stage of the lesion, is appropriate treatment for previously untreated giant cell tumor. Thus, patients referred for radiation therapy are those with recurrent disease or lesions located such that an appropriate surgical procedure would entail an unacceptable amount of disability. In addition to an overall analysis of local control, patients were grouped according to four factors for localcontrol comparisons: (a) tumor in membranous versus long bones, (b) dose of radiation, (c) number of previous surgical attempts, and (d) gender. Six of the patients were treated for recurrent tumor, and ten were previously untreated. Tumor was confined within the cortex of the primary site in five patients, and 11 patients had soft tissue extension. Patients with recurrent tumor had undergone one to three previous operative procedures, and had documented regrowth of the tumor before irradiation. All recurrences were in the site of original bulky disease. One patient (case 3) developed a giant

Volume 26, Number 2, 1993

cell tumor of the ilium following curettage and iliac bone harvest for a giant cell tumor of the right distal radius. This case probably represents surgical seeding. The tumor was recurrent at the iliac site after surgical excision. Patients were treated 5 days a week with continuouscourse radiation therapy using a variety of energies, including 6oCo, 8-MV X rays, and 17-MV X rays. Electrons were added when appropriate. One patient was treated twice a day, with a 6-hr interval between fractions. An attempt was made to cover the entire operative bed or known gross tumor with a 3- to 5-cm margin. Shrinking fields were used when possible, and a strip of skin was spared in extremity treatments to reduce the risk of lymphedema. Tumor doses ranged from 35 Gy to 54 Gy (mean 42.75 Gy). Fraction sizes ranged from 169 cGy to 233 cGy once a day. One patient was treated twice a day with 125 cGy per fraction. Follow-up times were calculated from the first day of radiation treatment. The minimum follow-up was 32 months, with 63% of patients having a follow-up of 5 or more years, and 44%, 10 or more years. At the time of this review 15 of 16 patients were alive. The one death was secondary to a myocardial infarction 10 years after irradiation of a right tarsal navicular lesion with no evidence of disease after surgical salvage. No patients were lost to follow-up. Any interval increase in the size of the tumor or appearance of new disease within or adjacent to the treated site after radiotherapy was considered a local failure. With respect to measurable gross disease, the tumor was considered locally controlled if clinical or radiographic followup demonstrated partial regression with no subsequent enlargement, or complete regression after radiotherapy with no regrowth.

RESULTS

Fig. 1. Distribution

of treated sites.

The clinical features, details of treatment, and current status of each patient are noted in Table 1. The results of treatment are shown in Tables 2 through 5. Local control of disease was obtained in 12 (75%) of 16 patients. Previously untreated disease was controlled in eight (80%) of ten patients, and recurrent disease in four (67%) of six patients (see Table 2). All treatment failures occurred in sites of previous irradiation at 8, 13, 13, and 25 months following initiation of radiotherapy. No geographic or marginal failures were noted. On histologic study, all recurrences revealed giant cell tumor with no evidence of malignant transformation. A dose analysis is shown in Table 3. All membranous lesions received at least 40 Gy, and only one failure was noted. A trend toward increased local control was seen in the membranous bone lesions (Table 4) although with a p value of 0.11, it did not reach statistical significance. This decrease in local control of long-bone lesions does not appear to be related to total dose or dose per fraction.

301

Giant cell tumor of bone ??C. J. BENNETT, JR. et al. Table 1. Summary

No.

Site

Age/Sex

Status

u

Previous surgery

of cases Radiation cCy/Fx/ Days

LF

DM

1 2

30 M 51 F

Right proximal tibula Right tarsal navicular

Marginal resection Biopsy only

3500/l 5/2 1 3500/20/29

N

N

U

Y

N

3 4

22 F 24 F

Right iliac crest Right proximal humerus

R R

Marginal resection Curettage

5500132142 4000/32/22

N

Y

N N

5 6 7 8 9 10 11 12

19 F 15 M 39 M 16 F 43 M 23 F 29 F 41 F

Lumbar spine Right temporal fossa Sacrum Left proximal tibia Thoracic spine Left proximal fibula Left maxillary sinus Cervical spine

Curettage X 2 Biopsy only Curettage Curettage X 2 Biopsy only Marginal resection Curettage X 2 Intracapsular resection

4500/25/35 5000/28/40 5000/28/44 3500/15/22 4500/25/37 3500/20/28 5400/32/59 4500/25/37

N N N N N N N Y

N N N N N N N Y

13

74 F

Left proximal tibia

R

Curettage X 3

3500/15/21

Y

N

14 15 16

42 M 21 M 17 F

Sacrum Sacrum Cervical spine

U U U

Biopsy only Biopsy only Biopsy only

4000/23/38 4000/20/30 4500/25/35

N N

N N N

N

Present status

Surgical Salvage

Right, below-knee amputation Right forequarter amputation

Anterior cervical corpectomy Proximal tibia1 resection

NED 9 yrs 6 mos DID 1 I yrs 2 mos’ NED 14yrs9mos NED 8 yrs 9 mos* NED NED NED NED NED NED NED NED

4 yrs 9 mos II yrs7mos 3 yrs 4 mos 9 yrs 5 mos 2 yrs 8 mos 12 yrs 1 mo 18 yrs 1 mo 4 yrs 8 mos*

NED 4 yrs 7 mos* NED 10 yrs 5 mos NED13yrs4mos NED4yrs 11 mos

* After salvage therapy. LF = Local failure; DM = Distant metastasis; U = untreated; R = Recurrent; NED = No evidence of disease: DID = Dead of intercurrent disease.

All four failures occurred in women. The trend seems to hold true in other reports, with 10 of 13 failures seen in female patients (see Table 4). This does not appear to be related to the site of origin, with 65% (26/40) of tumors in females occurring in membranous sites (see Table 5). Whether hormonal influence plays a role in local recurrence has not been demonstrated. Local control of tumors confined to the site of origin was 3 of 5 (60%) compared with 9 of 11 (82%) for tumors with soft tissue extension. Analysis of local control with regard to the number of previous surgical attempts showed no correlation. Complications No severe complications occurred in our series. Two patients with sacral lesions developed severe pain in the sacral area during irradiation, peaking 1 to 2 months after completion of treatment. Both patients were treated conservatively with ultimate resolution of the pain. There has been no development of a soft tissue sarcoma within the fields of previous irradiation.

DISCUSSION Giant cell tumor of bone has long been considered a benign lesion with a low metastatic potential. It is our belief that giant cell tumor is a malignant lesion with a high rate of local recurrence and a low potential for distant metastasis. Curettage, with or without methacrylate cementation, or wide excision are considered the treatment methods of choice for giant cell tumors of bone. Enneking has described three staging categories for tumors of bone: Stage I, latent (remains static or heals spontaneously); Stage II, active (progressive growth but limited by natural barriers); and Stage III, locally aggressive (progressive growth not limited by natural barriers) (10). Enneking has also described four basic types of resection and margins of resection: intracapsular (piecemeal debulking or curettage), marginal (shell out en bloc through pseudocapsule or reactive zone), wide (intracompartmental en bloc resection with a cuff of normal tissue), and radical (extracompartmental en bloc resection of the entire compartment) (10).

Table 3. Local control

accordine. to radiation

Table 2. Local control with radiation therapy for previously untreated vs. recurrent giant cell tumors

Surgical Previously Recurrent

untreated

Local control

salvage

Ultimate local control

8/10 416

212 212

lO/lO 616

therarrv dose

DoseC&Y)

Long bone Membranous Total

bone

3500

4000

4500

z 5000

Total

315 315

O/l 212 213

314 314

414 414

3/6 (50%) 9/ 10 (90%) 12/16 (75%)

302

I.J.

Radiation Oncology 0 Biology0 Physics

Volume 26, Number 2, 1993

Table 4. Review of literature: Radiotherapy series Local control Local control No. of patients

Long bone

Princess Margaret Hospital, 1983 (I) Beijing, 1986 (6)

15 35

818 -

M.D. Anderson Hospital, 1986 (18) Copenhagen, 1987 (8) Massachusetts General Hosp., 1989 (17) University of Florida, 1991 Total

10 10 11 16 97

l/4 314 316 15122 (68%)+

Institution

Membranous bone

Total

Developed sarcoma

Distant metastases

617 -

14/15* 26135

1 0

7110 616 617 9110 34140 (85%)+

7110 7110 9/11 12/16 75197 (77%)

Female

Male

0 0

lO/lO

415

0 0 0 0

3 2 2

(I&

(88%)

63 1/1 415 315 416 515 6110 616 30140 19122 (75%)’ (86%)$

1

* Failure due to geographic miss. ‘p = 0.11. +p = 0.24.

The current surgical policy at our institution is curettage alone for Stage I lesions, curettage with methacrylate cementation for Stage II lesions, and wide en bloc excision for Stage III lesions. Local control rates with this policy are 85% to 90% for Stage I, II, and III lesions (5, 9, 10). Occasionally, the surgeon is faced with recurrent tumors, tumors that are inoperable because of location, or tumors requiring procedures that would produce unacceptable postoperative functional deficits. These are indications for radiotherapy. A review of recent radiotherapy literature (see Table 4) shows an average local-control rate consistent with our own findings and an 8% incidence of pulmonary metastasis, consistent with previous surgical reports (5). Of note is the extremely low incidence of secondary soft-tissue sarcoma development within fields of previous irradiation (one of 97, or 1%). Patients treated with orthovoltage prior to the megavoltage era had approximately a 20% incidence (4, 7). Cahan and coworkers noted that these sarcomas may develop up to 20 years after irradiation (3). In our literature review of 24 patients with IO-year minimum follow-up, only one patient developed a soft-tissue sarcoma in the previously irradiated field. This patient developed a fibrosarcoma 2 1 years after radiotherapy in Italy (the energy used and dose given are unknown) (1). With modern techniques and higher energies, the incidence ap-

Table 5. Review of literature: Control by gender and bone type Local control Female Membranous Long bone Total *p = .ll. ’p = .24.

Male

Total

bone 21/26 (81%) 13/14 (93%) 34/40 (85%)* 9/14 (64%) 6/8 (75%) 15/22 (68%)* 30/40 (75%)+ 19/22 (86%)’

pears to be very low, although further follow-up is needed to address this issue fully. Local control of membranous bone lesions was achieved in 9 (90%) of 10 patients in our study. Control is vital because of the difficulties in resecting these lesions should they recur. Our literature review reveals a local control rate of 85% for membranous bone lesions compared with 68% for long-bone lesions. A knowledge of normal postirradiation changes seen on follow-up radiographic studies is imperative (13). Figure 2 shows preirradiation, postirradiation, 2-year followup, and 9-year follow-up computed tomography (CT) scans of a 42-year-old man with a sacral lesion (case 14). Extensive involvement of the right ilium with a remaining sclerotic rim is seen in the pretreatment study (Fig. 2A). Following irradiation (Fig. 2B), the sclerotic rim disappears. In a surgically treated patient, this would herald a recurrence, but in the postirradiation patient, this is a normal finding. Follow-up studies show reappearance of the sclerotic rim (Fig. 2C), continued calcification of the ilium, and resolution of the mass with an almost symmetrical pelvis 9 years after irradiation (Fig. 2D). These changes were seen in all 16 patients. Knowledge of these normal radiographic changes can prevent unnecessary and expensive evaluations of suspected recurrences. The hormonal influence of giant cell tumors has been extensively studied with regard to parathyroid hormone and calcitonin (11, 12, 20). Malawer and associates identified cytoplasmic staining with estradiol and progesterone in five analyzed cases of giant cell tumor of bone (15). The hormonal response of giant cell tumors when exposed to estrogen or progesterone has not been well documented. The greater percentage of failures occurring in female patients suggests a direct relationship. One failure in our series (case 4) was in a 24-year-old woman treated for a recurrent tumor of the right proximal humerus following curettage. The patient received 4000 cGy in 32 fractions (twice-a-day treatments) using 6oCo. Twenty-four months after treatment, the patient was found to have p,n intra-

303

Giant cell tumor of bone 0 C. J. BENNETT, JR. et al.

a

d b Fig. 2. Computed tomography scans of a 42-year-old man with a sacral lesion (case 14). (a) Preirradiation scan, showing sclerotic rim (arrow). (b) Postirradiation, the sclerotic rim has disappeared (arrow). (c) 2-year follow-up. (d) 9-year follow-up.

uterine pregnancy. One month later, she had rapid, painful enlargement of the right proximal humerus. Recurrence was documented radiographically, and the patient had a right forequarter amputation. The specimen contained giant cell tumor. Berman also reported the development of a giant cell tumor of the fourth thoracic vertebral body associated with an intrauterine pregnancy (2). Further investigation into hormonal influence is needed. Pulmonary metastasis was seen in 8% of patients in our literature review (see Table 4). One patient in our study (6%) developed pulmonary metastases. After multiple courses of chemotherapy, surgical resection, and wholelung irradiation, she remains alive 3 years 7 months after the diagnosis of metastasis. She has had no further recurrence in the bone.

Treatment recommendations We recommend surgical curettage, with or without methacrylate cementation, or wide resection alone to patients for whom these procedures can be carried out without producing disabling loss of function. If surgical margins are intracapsular with gross residual disease remaining, or if the lesion is recurrent, radiation therapy should be considered. For patients with inoperable lesions or for those in whom surgical treatment will cause a severe limitation of function, we recommend radiation therapy alone. Because of the poor local control seen in long-bone lesions, we currently use a tumor dose of at least 40 Gy. It is clear that giant cell tumor of bone is not a radioresistant tumor as once believed, and that complications seen with modern treatment regimens are minor.

REFERENCES 1.

Bell, R. S.; Harwoocl, A. R.; Goodman, S. B.; Fornasier, V. L. Supervoltage radiotherapy in the treatment of difficult giant cell tumors of bone. Clin. Orthop. 174:208-2 16; 1983.

2. Berman, H. L. The treatment of benign giant-cell tumors of the vertebrae by irradiation. Radiology 83:202-207;1964. 3. Cahan,

W. G.; Woodward,

H. Q.; Higinbotham,

N. L.;

304

4.

5. 6.

7.

8.

9. 10. 11.

12.

13.

1. J.

Radiation Oncology ??Biology0 Physics

Stewart, F. W. Sarcomas arising in irradiated bone. Cancer 9:753;1956. Campanacci, M.; Baldini, N.; Boriani, S.; Sudanese, A. Giant-cell tumor of bone. J. Bone Joint Surg. 69-A:106114;1987. Carrasco, C. H.; Murray, J. A. Giant cell tumors. Orthop. Clin. North Am. 20:395-405;1989. Chen, Z. X.; Gu, D. Z.; Yu, Z. H.; Qian, T. N.; Huang, Y. R.; Hu, Y. H.; Gu, X. Z. Radiation therapy of giant cell tumor of bone: Analysis of 35 patients. Int. J. Radiat. Oncol. Biol. Phys. 12:329-334;1986. Dahlin, D. C.; Unni, K. K. Bone tumors: General aspects and data on 8542 cases, 4th edition. Springfield, IL: Charles C. Thomas; 1986. Daugaard, S.; Johansen, H. F.; Barfod, G.; Laustein, G.; Schicadt, T.; Lund, B. Radiation treatment of giant-cell tumour of bone (osteoclastoma). Acta. Oncol. 26:4 l-43; 1987. Eckardt, J. J.; Grogan, T. J. Giant cell tumor of bone. Clin. Orthop. 204:45-58; 1986. Enneking, W. F. Musculoskeletal tumor surgery. New York, NY: Churchill Livingstone; 1983. Goldring, S. R.; Dayer, J. M.; Rosenblatt, M. Factors regulating the response of cells cultured from human giant cell tumors of bone to parathyroid hormone. J. Clin. Endocrinol. Metab. 53:295-300; 198 1. Goldring, S. R.; Schiller, A. L.; Mankin, H. J.; Dayer, J. M.; Krane, S. M. Characterization of cells from human giant cell tumors of bone. Clin. Orthop. 204:59-75;1986. Harwood, A. R.; Fomasier, V. L.; Rider, W. D. Supervoltage

Volume 26, Number 2, 1993

14.

15.

16.

17.

18.

19.

20.

irradiation in the management of giant cell tumor of bone. Radiology 125:223-226;1977. Larsson, S. E.; Lorentzon, R.; Boquist, L. Giant-cell tumor of bone: A demographic, clinical, and histopathological study of all cases recorded in the Swedish Cancer Registry for the years 1958 through 1968. J. Bone Joint Surg. 57-A: 167-173;1975. Malawer, M.; Bray, M.; Kass, M. Fluorescent histochemical demonstration of estrogen and progesterone binding in giant cell tumors of bone: Preliminary observations. J. Surg. Oncol. 25:148-152;1984. McDonald, D. J.; Sim, F. H.; McLeod, R. A.; Dahlin, D. C. Giant-cell tumor of bone. J. Bone Joint Surg. 68-A: 235-242; 1986. Schwartz, L. H.; Okunieff, P. G.; Rosenberg, A.; Suit, H. D. Radiation therapy in the treatment of difficult giant cell tumors. Int. J. Radiat. Oncol. Biol. Phys. 17: 10851088;1989. Seider, M. J.; Rich, T. A.; Ayala, A. G.; Murray, J. A. Giant cell tumors of bone: Treatment with radiation therapy. Radiology 161:537-540;1986. Sung, H. W.; Kuo, D. P.; Shu, W. P.; Chai, Y. B.; Liu, C. C.; Li, S. M. Giant-cell tumor of bone: Analysis of two hundred and eight cases in Chinese patients. J. Bone Joint Surg. 64-A:755-761;1982. Wientroub, S.; Binderman, I.; Somjen, D.; KIetter, Y.; Salama, R.; Weissman, S. L. Response to parathyroid and calcitonin of human giant cell tumor of bone cultured in patients’ serum. Clin. Orthop. 167:277-279; 1982.