Osteogenic The UCLA Rufus J.
Sarcoma of the Head and Neck
Experience
Mark, MD; Joel A. Sercarz, MD; Luu Tran, MD; Leslie G. Dodd, MD; Michael Selch, MD; Thomas C. Calcaterra, MD
\s=b\ Eighteen cases of osteogenic sarcoma of the head and neck were treated at our institution between 1955 and 1987. The patients' ages ranged from 5 to 73 years, with a median age of 28 years. The sex distribution was equal. Follow-up ranged from 1 to 276 months, with a median of 79 months. The primary site of the tumor was the mandible in nine cases, maxilla and paranasal sinuses in six, skull in two, and orbit in one. Six of 18 patients were free of disease with greater than 5 years of follow-up. Four of the six received combined surgery, radiation therapy, and chemotherapy as their primary treatment. Of the five patients treated with surgery alone, four suffered recur-
rences, one of whom was salvaged with further surgery and chemotherapy. Five patients were treated initially without surgery. They received radiation therapy with or without chemotherapy; all five developed local recurrence. We conclude that osteogenic sarcoma of the head and neck is an aggressive tumor, prone to both local and distant failure. Based on our series and from published experience involving the extremities, osteogenic sarcoma of the head and neck should be managed with
Osteogenic tumor,
adults
sarcoma
common
(OS) is
an un¬
accounting for
7400 new cases and 4200 deaths in the United States annually.1 Osteogenic sarcoma occurs most frequently in the long bones of adolescents and young
Accepted for publication January 8,1991. From the Department of Radiation Oncology (Drs Mark, Tran, and Selch), the Department of Surgery, Division of Head and Neck (Drs Sercarz and Calcaterra), and the Department of Pathology (Dr Dodd), UCLA Medical Center. Presented at the 32nd Annual Meeting of the American Society for Head and Neck Surgery, Palm Beach, Fla, May 1,1990. Reprint requests to the Department of Radiation Oncology, UCLA Medical Center, Los Angeles, CA 90024 (Dr Mark).
multimodality therapy. (Arch Otolaryngol 1991;117:761-766)
Head
Neck
Surg.
the period of maximal about 10% occur in the head and neck, accounting for only about 900 new cases per year in the United States.2 Several epidemiologie risk factors have been related to the development of OS, including a history of ionizing radiation exposure, fibrous dysplasia,
during
growth.2 Only
retinoblastoma, or prior exposure to thorium oxide, a radioactive scanning agent. Four percent of all patients with OS had a history of previous radiation therapy for other tumors or
conditions.2"4
The recent head and neck surgery
literature contains only infrequent case reports and small series of OSs. The prognostic factors and optimal management of OS of the head and neck are poorly defined. The largest published series of head and neck OS are from the Armed Forces Institute of Pathology, Wash¬ ington, DC (1967),4 and Memorial Sloan-Kettering, New York, NY (1971). The authors concluded that radical resection is the treatment of choice. The 5-year survival rate was 23.3% to 35%.4'5 Although radiation therapy was used in some patients, no definitive recommendation was made regarding its effectiveness. Skull le¬ sions had the worst prognosis, while there was disagreement whether man¬ dibular or maxillary tumors were more
lethal.4,5
Because of the rarity of head and neck OS, there have not been adequate data in the literature to evaluate the role of adjuvant therapy for head and neck OS. This study is a retrospective review of the UCLA experience in OS of the head and neck and includes a review of recent developments in the treatment of OS at other sites.
PATIENTS AND METHODS Of 2500 patients with sarcomas seen at UCLA between 1955 and 1987, 18 patients
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
Table 1.—Clinical Characteristics of Patients in the Series* Patient No./
Sex/Age, 1/M/18
2/M/14
Site
y
Grade
Stage
Maxilla
Nasal
cavity, pansinus,
T-3
Treatment S + R + C, extended
R
Outcome
Salvage
NED
maxillectomy 10/81, postoperative R + C (6000 cGy in 30 fractions) + C 3/70 (2400 cGy in 10
Status
A, NED 2/89
LDM
R 1/71
NED
A, NED 2/89
LR 11/79
R + C 1/80
D, LDM 3/71
fractions)
anterior fossa 3/F/32
4/M/18
Mandible
Orbit, zygoma,
T-1
T-3
brain
S + R + C.
hemimandibulectomy 11/81, preoperative R + C (5040 cGy in 28 fractions) S + C, radical gross resection 10/78, postoperative C for positive
5/M/31
Mandible
T-2
SM C + S, hemimandibulectomy 4/88, C preoperatively
6/F/5
Orbit, sphenoid,
T-3
C 8/78
T-3
R 10/78
7/F/73
ethmoid Parietal occiput, brain, neck
(2000 cGy in fractions)
10
D, LD 6/80
(6120 cGy in 34 fractions) A, NED 2/89 Persistent disease Persistent
D, LD 12/78 D, LD 11/78
disease
node
8/M/35
Mandible
T-1
S 3/57, negative SM
LR 10/64
RC 12/64
A, LD 11/65
(3500 cGy in 10 fractions) 9/M/28 10/F/56
Mandible Frontal bone,
T-3
S 9/81, hemimandibulectomy S + R + C 7/68 (5000 cGy in 25 fractions),
T-3
R 3/67
LR 11/67
C 11/67
A, LD 1/£
T-1
S 1/65
LR 11/65
S 12/65, S 2/67 (2400 cGy in 21 fractions 9/67, 3000 cGy in 20 fractions 11/67)
A, LD 5/68
S + C + R 9/64, R+C
NED
T-1
brain
11/M/24
Mandible
13/F/25
orthovoltage (6500 cGy in 35 fractions)
NED LR 1/69
A, NED 3/90
A, NED 12/84
preoperatively (4000 cGy fractions) partial hemimandibulectomy
LR 1/73
R + C2/61 S 8/65, gross resection
DM 3/66 LR 12/65
C + R
LR (6000
in 20
14/M/30
Mandible
T-1
15/M/22
Mandible
T-1
16/F/58
Maxilla, maxillary
T-2
S 7/72,
CS 2/73, hemiman¬
A, NED 4/85
dibulectomy, preoperative C C3/66 SR 12/65 (6000 cGy in 30
sinus
D, DM 10/66 D, LDM 4/66
fractions)
17/F/58
Maxilla, maxillary sinus
18/F/25
Mandible
30 T-2
cGy in fractions)
D, LD 5/66
S + R + C 9/64,
A, NED 4/81 hemimandibulectomy, preoperative C + R (6000 cGy in 30 fractions) * S indicates surgery; R, radiation therapy; C, chemotherapy; NED, no evidence of disease; A, alive; LDM, local and distant metastasis; D, dead; LR, local recurrence; LD, local disease; SM, surgical margins; and DM, distant metastasis.
with OS of the head and neck were identified in the UCLA tumor and pathology regis¬ tries. Sex distribution was equal, with nine cases in male and nine in female patients. Patient ages ranged from 5 to 73 years, with a median age of 28 years. The location of the primary tumors in the series was as follows: nine cases arose in the mandible, four in the maxilla, two in the or¬ bit, and one in the frontal bone. Two cases had extensions into multiple sinuses. Four of the patients in the series had a
prior history of radiation therapy. Three of these patients underwent radiation therapy for retinoblastoma. Their postirradiation OS developed with latency periods of 4, 11, and 18 years following radiation. The fourth pa¬ tient underwent radiation therapy for multi¬ ply recurrent fibrous dysplasia of the mandi¬ ble. Osteogenic sarcoma was diagnosed in this patient 4 months after completion of ra¬ diation therapy. Before treatment, patients were assessed by complete history and physical examina-
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
tion,
routine laboratory tests, biopsy, and chest roentgenography. Computed tomogra¬ phy of the head and neck was performed in nine cases. Histologie features of the primary tumor were reviewed by a UCLA patholo¬ gist in each case to confirm the diagnosis. There were no cases of parosteal OS. Tumors were classified as conventional OS and as¬ signed a grade based on histologie appear¬ ance. To evaluate local extent of disease and tumor bulk, patients' tumors were retro¬ spectively staged according to the TNM
staging system for sarcomas.
Treatment was individualized based on the location of the primary lesion, the extent of disease, and the judgment of the treating physicians. The mode of initial therapy is categorized in the left column of Table 1. Surgery consisted of excision of the lesion with an appropriate margin. One patient, who had lymph node metastasis at the time of diagnosis, also had brain involvement at the time of diagnosis and was treated with radia¬ tion. None of the other 17 patients had lymph node involvement, and none underwent pro¬ phylactic neck dissection. Radiation was delivered by a cobalt 60, 6MeV or 10-MeV linear accelerator. Patients received doses ranging from 2000 to 6500 cGy. The neck was not electively irradiated. Chemotherapeutic agents used included doxorubucin hydrochloride, cyclophospha¬ mide, methotrexate sodium, and vincristine sulfate; doxorubucin was used most
Table 2.—Results by Treatment Treatment
Surgery (n 5) Surgery + radiation (n 5)
Modality* NEDS
NED
=
+
chemotherapy
=
Radiation with
or without chemotherapy (n 5) Surgery + chemotherapy (n 2) *NED indicates no evidence of disease; LD, local disease; DM, distant metastasis; LDM, local and distant metastasis; and NEDS, no evidence of disease after salvage. =
=
100
frequently. Patient outcome was assessed by physical examination whenever possible. Those pa¬ tients not examined directly were assessed by chart review, review of radiologie study results, and telephone calls to primary physi¬ cians. The follow-up time ranged from 1 to 276 months, with a median of 79 months.
Overall Survival
Relapse-Free Survival
RESULTS
2
Each patient's case is summarized in Table 1. The most frequent presenting complaint leading to the diagnosis of OS was jaw mass, which occurred in seven cases. Three patients presented with a cheek mass, three with bone pain, and one each with a mass in the orbit, forehead, and occiput. Results
Stage Five (63%) of eight patients with by
stage T-1 lesions showed
3
.
no
evidence
of disease
beyond the 5-year follow-up. Only two (20%) of 10 patients with
stage T-2/3 lesions have been rendered free of disease, and the follow-up for one of the two patients is only 10 months.
by Treatment Modality Four of the five patients treated with surgery alone developed a recur¬ rence. One of these patients was sal¬ vaged with subsequent chemotherapy and surgery. The only patient ren¬ Results
dered disease free with surgery alone had been treated with a hemimandibu¬ lectomy for a low-grade lesion. Five patients received combined surgery, radiation therapy, and che-
10
0
12
_i_i_
_i_
36
60
-I_I_I_L
84
_l_
_l_
108
Time, mo Fig 1. Kaplan-Meier life table of patients in the series. —
motherapy. Four (80%) of these five patients were rendered free of disease. The patient in whom multiple treat¬ ment modalities failed
was treated with orthovoltage radiation in which 5000 cGy of radiation was delivered to the skin. The treatment at the site of the tumor was likely suboptimal. Of the two patients receiving chemo¬ therapy and surgery, one was disease free following treatment. Five patients were treated with radiation with or without chemotherapy. All five of the patients who received radiation thera¬ py had local recurrence. Four of the five had stage T-2 to T-3 lesions, and none was salvaged. One patient was treated with chemotherapy alone for an unresectable sinus OS and died 4
months after diagnosis. Treatment mo¬ dality vs patient outcome is summa¬ rized in Table 2.
Salvage Results Eight patients underwent salvage attempts for recurrent disease, and only one survived. This patient re¬ ceived preoperative chemotherapy fol¬ lowed by hemimandibulectomy for a
local recurrence and is free of disease 12 years after salvage therapy. The other seven patients died despite at¬
tempted salvage.
Outcome and Survival At last follow-up, eight patients in series had local recurrences, one with distant and two with local and our
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
distant metastasis. Of the eight pa¬ tients with local recurrence, six have died of their disease. Death was most often due to brain extension. The pa¬ tient in the series who died before treatment similarly had direct exten¬ sion to the brain. Consistent with OS in other sites, distant métastases to the lung were common. All three pa¬ tients with distant metastasis had pul¬ monary involvement. In addition, one patient developed métastases to the brain and one patient had disease in the regional lymphatics. All three pa¬ tients who developed distant metasta¬ sis died of their disease. Nine (81%) of 11 recurrences have occurred within 2 years. Seven (47%) of 15 assessable patients have enjoyed 5 years or more disease-free survival. Actuarial survival curves were calcu¬ lated with use of the Kaplan-Meier life table method (Fig 1).
Fig 2.—Low-power photomicrograph of an osteogenic sarcoma with irregular, neoplastic osteoid and cellular sarcomatous stroma (hematoxylin-eosin, original magnification 25).
COMMENT The patient profile in this series is consistent with those in prior reviews. The sex distribution is equal, and most lesions in the head and neck arise in the mandible and maxilla in patients in the third and fourth decades of life. Patients usually present with a mass over the cheek or jaw, which can be painful.4'8 The tendency for uncontrol¬ lable local failure as a cause of death has also been emphasized in previously
published series.6 In the UCLA
series, four of the 18
patients had a history of prior irradia¬ tion. Three of these patients had previ¬ ously undergone radiation therapy for retinoblastoma and developed OS at
intervals of 4, 11, and 18 years there¬ after. Interestingly, patients with re¬ tinoblastoma have an increased risk for developing OS independent of radia¬ tion exposure. Among patients with hereditary retinoblastoma, the inci¬ dence of OS is increased 500 times over the expected incidence.2 Similarly, de¬ letion of the long arm of chromosome 13, which occurs with increased fre¬ quency in patients with retinoblas¬ toma, is also associated with develop¬ ment of OS.12 Benedict et al9 have recently reported the cloning of a re¬ tinoblastoma gene that is thought to be responsible for the development of re¬ tinoblastoma and OS.
Fig 3.—Higher-power photomicrograph of osteogenic sarcoma illustrating osteoblasts amid irregular osteoid (hematoxylin-eosin, original magnification The fourth patient who developed OS after radiation therapy was treated for recurrent fibrous dysplasia of the mandible. One percent of patients with fibrous dysplasia will eventually devel¬ op OS.10 In this case, the patient devel¬ oped OS 4 months after irradiation, which suggests that the sarcoma exist¬ ed before radiation therapy. Histologically, OS comprises a ma-
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
cytologically atypical 100).
lignant, undifferentiated stroma and neoplastic osteoid or bone. The stromal component is characterized by dense cellularity, pleomorphism, and cytolog¬ ically atypical osteoblasts. The mitotic rate of the proliferating stromal cells is often increased. Neoplastic osteoid and bony spicules may be sparsely scat¬ tered throughout the tumor or ar¬ ranged in a fine, complex (filigree)
2 and 3 are photomi¬ crographs that show the typical fea¬ tures of OS. Osteogenic sarcoma can be subclassified into different categories based on the predominant characteristic of the stroma. Osteoblastic, chondroblastic, and fibroblastic types are recog¬ nized. There are also special variants of OS worth mention because of impor¬ tant prognostic implications and often deceptive histologie features. One such variant is the intraosseous well-differ¬ entiated osteosarcoma initially de¬ scribed by Unni et al." Interestingly, this parosteal OS-like tumor is most often misdiagnosed as fibrous dyspla¬ sia. The patient in this series initially treated for fibrous dysplasia may, in fact, have had this prognostically fa¬ vorable variant of OS.
pattern. Figures
Osteogenic
grade based
assigned a anaplasia component. Low-grade
sarcomas are on
the relative
of the stromal lesions (grade 1) may stimulate paros¬ teal OS or fibrous dysplasia. Increas¬
ingly, anaplastic tumors are assigned higher grades, the most cytologically bizarre being the grade 4 lesions. Ma¬ lignant osteogenic neoplasms of the jaw are more often of lower grade and have been reported to have a more favorable prognosis than those of the extremities.12 The largest series of head and neck OS showed no correla¬ tion between histologie character and
prognosis.4
The optimal management for pa¬ tients with OS of the head and neck is unclear. The rarity of the lesion pre¬ cludes adequate patient enrollment in prospective randomized studies. Some authors have favored radical surgery alone. Forteza et al7 reported 85% long-term survival in a series of nine patients managed with radical sur¬ gery. Three patients received radia¬ tion therapy, two preoperatively and one postoperatively. Their data com¬ pare favorably with other series that report a 5-year survival of 20% to
33%.4'5·7'13
Recent small series and case reports suggest that combined therapy em¬
ploying surgery, radiation therapy, an¬ d/or chemotherapy improves results in osteosarcoma.14'17 Although the data are retrospective, the results reported
in the present series tend to support
aggressive
multimodality
therapy.
None of the patients in the UCLA series treated with surgery alone re¬ mained disease free. Of the five longterm disease-free survivors with ade¬ quate follow-up, four received adjuvant radiation therapy and chemo¬ therapy. Because of the high frequen¬ cy of local recurrence, the use of radia¬ tion therapy would be expected to
improve results.18"20 More extensive experience with OS has been gained in the extremities. In
the recent past, the accepted manage¬ ment of OS of the extremities was amputation. Despite radical surgery, the overall 2-year survival had been reported to be between 5% and 20%.17'21"24 The development of effective chemotherapy against osteosarcoma in the 1970s permitted less radical sur¬ gery. Since then, limb-sparing surgery has become the preferred treatment.25 Several studies have demonstrated that neither survival nor the risk of local recurrence is compromised with the limb-sparing approach.22"26 Eilber et al25 reported a 2.7% local recurrence rate in 183 patients with malignant skeletal or high-grade softtissue sarcomas treated according to a limb-salvage protocol. The regimen consisted of preoperative doxorubucin therapy and radiation therapy followed by en bloc resection and postoperative chemotherapy. With a median followup of 32 months, overall survival among patients treated with limb sal¬ vage was identical to that of those undergoing amputation and chemo¬
therapy.26
Two randomized trials of OS of the extremities have reported preliminary data showing improved disease-free survival in patients receiving adjuvant therapy. In a series by Link et al,27114 patients were enrolled to study the advantages of adjuvant chemotherapy. Among the randomized patients, the 2year actuarial disease-free survival rate in the chemotherapy arm of the study was 56% compared with 18% in the surgery alone group (P .001). Among the patients who refused ran¬ domization, the survival rate was 50% in those who received chemotherapy postoperatively vs 10% in those receiv¬ =
ing no adjuvant treatment.27 In another trial, 59 patients
with
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
extremity OS were randomized to re¬ ceive aggressive adjuvant postopera¬
tive chemotherapy or no treatment after receiving preoperative doxorubu¬ cin therapy and radiation therapy. With a median follow-up of 2 years, both disease-free survival and overall survival rates were statistically higher in the patients receiving aggressive adjuvant therapy. In the trials of both Link et al27 and Eilber et al25 , the follow-up period was too short to know whether the advantages will persist.28 Several studies have shown im¬ proved local control and survival when adjuvant radiation therapy with or without chemotherapy is added. Chamber and Mahoney18 reported a 73% 5-year survival in 33 patients re¬ ceiving high-dose preoperative radia¬ tion therapy followed by wide surgical resection. Akbiyik and Alexander19 and de Fries et al2" reported similar results with preoperative irradiation, resec¬ tion, and postoperative chemotherapy. Our data, while based on a small series examined retrospectively, tend to sup¬ port these aggressive approaches. There is no proof that the foregoing results from the extremities are rele¬ vant to head and neck OS. Except for its pattern of failure, OS of the facial bones has a natural history similar to that of OS of the extremities. There is a much lower risk of distant metastasis in the head and neck but a high rate of local recurrence.3 This may be due to the difficulty in achieving wide surgical margins in the head and neck for ana¬ tomic and cosmetic reasons. The litera¬ ture suggests that radical surgery alone frequently results in local fail¬ 3-6'20 Because of the rarity of head ure. and neck OS, it may never be possible to directly study the effect of adjuvant therapy in a prospective, randomized fashion. Data from the extremities are therefore probably the best way to judge the effectiveness of adjuvant treatment.
The
ogies
development of new technol¬ procedures, such as mono¬
and
clonal antibodies, may become rele¬ vant to the treatment of OS in the future. A monoclonal osteonectin anti¬ body has been developed that binds to OS cells.29 In the future, it may be possible to link radioactive isotopes to monoclonal antibodies, permitting spe-
cific cell killing. However, clinical tri¬ als employing monoclonal antibodies in solid tumors have thus far been
disappointing.
In conclusion, OS of the head and neck is a formidable tumor demanding
aggressive treatment. Adequate surgi-
cal resection is the mainstay of treat¬ ment. Patients should receive adju¬ radiation and vant therapy based accumulated on chemotherapy, experience in head and neck OS and recently published data concerning OS of the extremities. Although this se-
ries cannot compare different types of adjuvant therapy, the data support the conclusion that surgery alone is unlike¬ ly to cure most patients with OS. Fu¬ ture studies are required to determine the correct sequence of treatment.
References 1. Antman KH, Eilber FR, Shiu MH. Soft tissue current trends in diagnosis and manage-
sarcomas:
ment. Curr Probl Cancer. 1989;13:337-367. 2. Goorin AM, Abelson HT, Frei E. Osteosarcoma: fifteen years later. N Engl J Med.
1985;313:1637-1643. 3. Malawer MM, Link MP, Donaldson LL. Sarcomas of bone. In: Devita VT, Hellman S, Rosenberg SA, eds. Cancer: Principals and Practice of Oncology. Philadelphia, Pa: JB Lippincott; 1989:1442-1453. 4. Garrington GE, Scofield HH, Cornyn J, et al. Osteosarcoma of the jaws: analysis of 56 cases. Cancer. 1967;20:377-391. 5. Caron AS, Jahder SI, Strong EW. Osteogenic sarcoma of the facial and cranial bones: a review of 43 cases. Am J Surg. 1971;122:719-725. 6. Pritchard DJ. Sarcomas of bone. Cancer Treat Symp. 1983;2:247-250. 7. Forteza G, Colmenero B, Lopez-Barea F. Osteogenic sarcoma of the maxilla and mandible. Oral Surg Oral Med Oral Pathol. 1986;62:179-184. 8. Clark L, Unni KK, Dahlin DC. Osteosarcoma of the jaw. Cancer. 1983;51:2311-2316. 9. Benedict W, Fung Y, Murphee L. The gene responsible for the development of retinoblastoma and osteosarcoma. Cancer. 1988;62:1691-1694. 10. Wick MR, McLeod RA, Siegal GP, Greditzer HG, Unni KK. Sarcomas of bone complicating osteitis deformans (Paget's disease): fifty years experience. Am J Surg Pathol. 1981;5:47-51. 11. Unni KK, Dahler DC, Beabout JW, et al. sarcoma. Parosteal Cancer. osteogenic
1976;378:2466-2475. 12. de Fries HO, Perlin E, Leibel SA. Sarcoma of the mandible. Arch Otolaryngol. 1979;105:358\x=req-\
359. 13. Mirra JH. Bone Tumors: Clinical, Radiologic and Pathologic. Philadelphia, Pa: Lea & Febiger; 1989. 14. McKenna WG, Barnes MM, Kinsella TJ, et al. Combined modality treatment of adult soft tissue sarcomas of the head and neck. Int J Radiat Oncol Biol Phys. 1987;13:1127-1133. 15. Beziat JL, Tabone E, Bailly C, Gerard JP. Osteogenic sarcoma of the tongue. J Oral Maxillo-
fac Surg. 1989;47:524-528. 16. Link MP. Adjuvant therapy in the treatment of osteosarcoma. In: DeVita VJ, Hellman S, Rosenberg SA, eds. Important Advances in Oncology, 1986. Philadelphia, Pa: JB Lippincott; 1986:193\x=req-\ 207. 17. Marcove
RC, Mike V, Hajack JV, et al. Osteogenic sarcoma under the age of 21. J Bone Joint Surg Am. 1970;52:411-423. 18. Chambers RG, Mahoney WD. Osteogenic sarcoma ofthe
J
mandible: current management. Am
Surg. 1970;36:463-471. 19. Akbiyik N, Alexander LL.
Osteosarcoma of the maxilla treated with radiation therapy and surgery. J Natl Med Assoc. 1981;73:355-356. 20. deFries HO, Perlin E, Leibel SA. Treatment of osteogenic sarcoma of the mandible. Arch Oto-
laryngol. 1979;105:358-359. 21. Sweetman R. Surgical management of primary osteosarcoma. Clin Orthop. 1975;111:57-64.
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
Campanacci M, Bacci G, Betoni F, et al. The osteogenic sarcoma of the extremities: twenty years experience at the Institukto Orthope22.
treatment
dico Rizzoli. Cancer. 1981;48:1569-1581. 23. Eilber FR, Eckhardt J, Morton DL. Advances in the treatment of sarcomas of the extremities: current status of limb salvage. Cancer.
1984;54:2695-2701.
24. Simon MA, Achliman MA, Thomas N, et al. Limb salvage treatment versus amputation for osteosarcoma of the distal end of the femur. J Bone Joint Surg Am. 1986;68:1331-1337. 25. Eilber FR, Morton DL, Eckardt J, et al. Limb salvage for skeletal and soft tissue sarcomas: multidisciplinary preoperative therapy. Cancer.
1984;53:2579-2584.
26. Enneking WF. A system for the functional evaluation for the surgical management of musculoskeletal tumors. In: Enneking WF, ed. Limb Salvage in Musculoskeletal Oncology. New York, NY: Churchill Livingstone Inc; 1987:5-16. 27. Link M, Goorin A, Miser A, et al. The role of adjuvant chemotherapy in the treatment of osteosarcoma of the extremities: preliminary results of the multiinstitutional osteosarcoma study. Proc Am Soc Clin Oncol. 1985;4:237. 28. Eilber FR, Guiliano A, Eckardt J, et al. Adjuvant chemotherapy for osteosarcoma: a randomized prospective trial. J Clin Oncol. 1987;5:21-26. 29. Jundt G, Schultz A, Berghauser KH, et al. Immunocytochemical identification of osteogenic bone tumors by osteonectin antibodies. Virchows Arch A. 1989;414:345-353.
Sarcoma of the Head and Neck
Experience
Mark, MD; Joel A. Sercarz, MD; Luu Tran, MD; Leslie G. Dodd, MD; Michael Selch, MD; Thomas C. Calcaterra, MD
\s=b\ Eighteen cases of osteogenic sarcoma of the head and neck were treated at our institution between 1955 and 1987. The patients' ages ranged from 5 to 73 years, with a median age of 28 years. The sex distribution was equal. Follow-up ranged from 1 to 276 months, with a median of 79 months. The primary site of the tumor was the mandible in nine cases, maxilla and paranasal sinuses in six, skull in two, and orbit in one. Six of 18 patients were free of disease with greater than 5 years of follow-up. Four of the six received combined surgery, radiation therapy, and chemotherapy as their primary treatment. Of the five patients treated with surgery alone, four suffered recur-
rences, one of whom was salvaged with further surgery and chemotherapy. Five patients were treated initially without surgery. They received radiation therapy with or without chemotherapy; all five developed local recurrence. We conclude that osteogenic sarcoma of the head and neck is an aggressive tumor, prone to both local and distant failure. Based on our series and from published experience involving the extremities, osteogenic sarcoma of the head and neck should be managed with
Osteogenic tumor,
adults
sarcoma
common
(OS) is
an un¬
accounting for
7400 new cases and 4200 deaths in the United States annually.1 Osteogenic sarcoma occurs most frequently in the long bones of adolescents and young
Accepted for publication January 8,1991. From the Department of Radiation Oncology (Drs Mark, Tran, and Selch), the Department of Surgery, Division of Head and Neck (Drs Sercarz and Calcaterra), and the Department of Pathology (Dr Dodd), UCLA Medical Center. Presented at the 32nd Annual Meeting of the American Society for Head and Neck Surgery, Palm Beach, Fla, May 1,1990. Reprint requests to the Department of Radiation Oncology, UCLA Medical Center, Los Angeles, CA 90024 (Dr Mark).
multimodality therapy. (Arch Otolaryngol 1991;117:761-766)
Head
Neck
Surg.
the period of maximal about 10% occur in the head and neck, accounting for only about 900 new cases per year in the United States.2 Several epidemiologie risk factors have been related to the development of OS, including a history of ionizing radiation exposure, fibrous dysplasia,
during
growth.2 Only
retinoblastoma, or prior exposure to thorium oxide, a radioactive scanning agent. Four percent of all patients with OS had a history of previous radiation therapy for other tumors or
conditions.2"4
The recent head and neck surgery
literature contains only infrequent case reports and small series of OSs. The prognostic factors and optimal management of OS of the head and neck are poorly defined. The largest published series of head and neck OS are from the Armed Forces Institute of Pathology, Wash¬ ington, DC (1967),4 and Memorial Sloan-Kettering, New York, NY (1971). The authors concluded that radical resection is the treatment of choice. The 5-year survival rate was 23.3% to 35%.4'5 Although radiation therapy was used in some patients, no definitive recommendation was made regarding its effectiveness. Skull le¬ sions had the worst prognosis, while there was disagreement whether man¬ dibular or maxillary tumors were more
lethal.4,5
Because of the rarity of head and neck OS, there have not been adequate data in the literature to evaluate the role of adjuvant therapy for head and neck OS. This study is a retrospective review of the UCLA experience in OS of the head and neck and includes a review of recent developments in the treatment of OS at other sites.
PATIENTS AND METHODS Of 2500 patients with sarcomas seen at UCLA between 1955 and 1987, 18 patients
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
Table 1.—Clinical Characteristics of Patients in the Series* Patient No./
Sex/Age, 1/M/18
2/M/14
Site
y
Grade
Stage
Maxilla
Nasal
cavity, pansinus,
T-3
Treatment S + R + C, extended
R
Outcome
Salvage
NED
maxillectomy 10/81, postoperative R + C (6000 cGy in 30 fractions) + C 3/70 (2400 cGy in 10
Status
A, NED 2/89
LDM
R 1/71
NED
A, NED 2/89
LR 11/79
R + C 1/80
D, LDM 3/71
fractions)
anterior fossa 3/F/32
4/M/18
Mandible
Orbit, zygoma,
T-1
T-3
brain
S + R + C.
hemimandibulectomy 11/81, preoperative R + C (5040 cGy in 28 fractions) S + C, radical gross resection 10/78, postoperative C for positive
5/M/31
Mandible
T-2
SM C + S, hemimandibulectomy 4/88, C preoperatively
6/F/5
Orbit, sphenoid,
T-3
C 8/78
T-3
R 10/78
7/F/73
ethmoid Parietal occiput, brain, neck
(2000 cGy in fractions)
10
D, LD 6/80
(6120 cGy in 34 fractions) A, NED 2/89 Persistent disease Persistent
D, LD 12/78 D, LD 11/78
disease
node
8/M/35
Mandible
T-1
S 3/57, negative SM
LR 10/64
RC 12/64
A, LD 11/65
(3500 cGy in 10 fractions) 9/M/28 10/F/56
Mandible Frontal bone,
T-3
S 9/81, hemimandibulectomy S + R + C 7/68 (5000 cGy in 25 fractions),
T-3
R 3/67
LR 11/67
C 11/67
A, LD 1/£
T-1
S 1/65
LR 11/65
S 12/65, S 2/67 (2400 cGy in 21 fractions 9/67, 3000 cGy in 20 fractions 11/67)
A, LD 5/68
S + C + R 9/64, R+C
NED
T-1
brain
11/M/24
Mandible
13/F/25
orthovoltage (6500 cGy in 35 fractions)
NED LR 1/69
A, NED 3/90
A, NED 12/84
preoperatively (4000 cGy fractions) partial hemimandibulectomy
LR 1/73
R + C2/61 S 8/65, gross resection
DM 3/66 LR 12/65
C + R
LR (6000
in 20
14/M/30
Mandible
T-1
15/M/22
Mandible
T-1
16/F/58
Maxilla, maxillary
T-2
S 7/72,
CS 2/73, hemiman¬
A, NED 4/85
dibulectomy, preoperative C C3/66 SR 12/65 (6000 cGy in 30
sinus
D, DM 10/66 D, LDM 4/66
fractions)
17/F/58
Maxilla, maxillary sinus
18/F/25
Mandible
30 T-2
cGy in fractions)
D, LD 5/66
S + R + C 9/64,
A, NED 4/81 hemimandibulectomy, preoperative C + R (6000 cGy in 30 fractions) * S indicates surgery; R, radiation therapy; C, chemotherapy; NED, no evidence of disease; A, alive; LDM, local and distant metastasis; D, dead; LR, local recurrence; LD, local disease; SM, surgical margins; and DM, distant metastasis.
with OS of the head and neck were identified in the UCLA tumor and pathology regis¬ tries. Sex distribution was equal, with nine cases in male and nine in female patients. Patient ages ranged from 5 to 73 years, with a median age of 28 years. The location of the primary tumors in the series was as follows: nine cases arose in the mandible, four in the maxilla, two in the or¬ bit, and one in the frontal bone. Two cases had extensions into multiple sinuses. Four of the patients in the series had a
prior history of radiation therapy. Three of these patients underwent radiation therapy for retinoblastoma. Their postirradiation OS developed with latency periods of 4, 11, and 18 years following radiation. The fourth pa¬ tient underwent radiation therapy for multi¬ ply recurrent fibrous dysplasia of the mandi¬ ble. Osteogenic sarcoma was diagnosed in this patient 4 months after completion of ra¬ diation therapy. Before treatment, patients were assessed by complete history and physical examina-
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
tion,
routine laboratory tests, biopsy, and chest roentgenography. Computed tomogra¬ phy of the head and neck was performed in nine cases. Histologie features of the primary tumor were reviewed by a UCLA patholo¬ gist in each case to confirm the diagnosis. There were no cases of parosteal OS. Tumors were classified as conventional OS and as¬ signed a grade based on histologie appear¬ ance. To evaluate local extent of disease and tumor bulk, patients' tumors were retro¬ spectively staged according to the TNM
staging system for sarcomas.
Treatment was individualized based on the location of the primary lesion, the extent of disease, and the judgment of the treating physicians. The mode of initial therapy is categorized in the left column of Table 1. Surgery consisted of excision of the lesion with an appropriate margin. One patient, who had lymph node metastasis at the time of diagnosis, also had brain involvement at the time of diagnosis and was treated with radia¬ tion. None of the other 17 patients had lymph node involvement, and none underwent pro¬ phylactic neck dissection. Radiation was delivered by a cobalt 60, 6MeV or 10-MeV linear accelerator. Patients received doses ranging from 2000 to 6500 cGy. The neck was not electively irradiated. Chemotherapeutic agents used included doxorubucin hydrochloride, cyclophospha¬ mide, methotrexate sodium, and vincristine sulfate; doxorubucin was used most
Table 2.—Results by Treatment Treatment
Surgery (n 5) Surgery + radiation (n 5)
Modality* NEDS
NED
=
+
chemotherapy
=
Radiation with
or without chemotherapy (n 5) Surgery + chemotherapy (n 2) *NED indicates no evidence of disease; LD, local disease; DM, distant metastasis; LDM, local and distant metastasis; and NEDS, no evidence of disease after salvage. =
=
100
frequently. Patient outcome was assessed by physical examination whenever possible. Those pa¬ tients not examined directly were assessed by chart review, review of radiologie study results, and telephone calls to primary physi¬ cians. The follow-up time ranged from 1 to 276 months, with a median of 79 months.
Overall Survival
Relapse-Free Survival
RESULTS
2
Each patient's case is summarized in Table 1. The most frequent presenting complaint leading to the diagnosis of OS was jaw mass, which occurred in seven cases. Three patients presented with a cheek mass, three with bone pain, and one each with a mass in the orbit, forehead, and occiput. Results
Stage Five (63%) of eight patients with by
stage T-1 lesions showed
3
.
no
evidence
of disease
beyond the 5-year follow-up. Only two (20%) of 10 patients with
stage T-2/3 lesions have been rendered free of disease, and the follow-up for one of the two patients is only 10 months.
by Treatment Modality Four of the five patients treated with surgery alone developed a recur¬ rence. One of these patients was sal¬ vaged with subsequent chemotherapy and surgery. The only patient ren¬ Results
dered disease free with surgery alone had been treated with a hemimandibu¬ lectomy for a low-grade lesion. Five patients received combined surgery, radiation therapy, and che-
10
0
12
_i_i_
_i_
36
60
-I_I_I_L
84
_l_
_l_
108
Time, mo Fig 1. Kaplan-Meier life table of patients in the series. —
motherapy. Four (80%) of these five patients were rendered free of disease. The patient in whom multiple treat¬ ment modalities failed
was treated with orthovoltage radiation in which 5000 cGy of radiation was delivered to the skin. The treatment at the site of the tumor was likely suboptimal. Of the two patients receiving chemo¬ therapy and surgery, one was disease free following treatment. Five patients were treated with radiation with or without chemotherapy. All five of the patients who received radiation thera¬ py had local recurrence. Four of the five had stage T-2 to T-3 lesions, and none was salvaged. One patient was treated with chemotherapy alone for an unresectable sinus OS and died 4
months after diagnosis. Treatment mo¬ dality vs patient outcome is summa¬ rized in Table 2.
Salvage Results Eight patients underwent salvage attempts for recurrent disease, and only one survived. This patient re¬ ceived preoperative chemotherapy fol¬ lowed by hemimandibulectomy for a
local recurrence and is free of disease 12 years after salvage therapy. The other seven patients died despite at¬
tempted salvage.
Outcome and Survival At last follow-up, eight patients in series had local recurrences, one with distant and two with local and our
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
distant metastasis. Of the eight pa¬ tients with local recurrence, six have died of their disease. Death was most often due to brain extension. The pa¬ tient in the series who died before treatment similarly had direct exten¬ sion to the brain. Consistent with OS in other sites, distant métastases to the lung were common. All three pa¬ tients with distant metastasis had pul¬ monary involvement. In addition, one patient developed métastases to the brain and one patient had disease in the regional lymphatics. All three pa¬ tients who developed distant metasta¬ sis died of their disease. Nine (81%) of 11 recurrences have occurred within 2 years. Seven (47%) of 15 assessable patients have enjoyed 5 years or more disease-free survival. Actuarial survival curves were calcu¬ lated with use of the Kaplan-Meier life table method (Fig 1).
Fig 2.—Low-power photomicrograph of an osteogenic sarcoma with irregular, neoplastic osteoid and cellular sarcomatous stroma (hematoxylin-eosin, original magnification 25).
COMMENT The patient profile in this series is consistent with those in prior reviews. The sex distribution is equal, and most lesions in the head and neck arise in the mandible and maxilla in patients in the third and fourth decades of life. Patients usually present with a mass over the cheek or jaw, which can be painful.4'8 The tendency for uncontrol¬ lable local failure as a cause of death has also been emphasized in previously
published series.6 In the UCLA
series, four of the 18
patients had a history of prior irradia¬ tion. Three of these patients had previ¬ ously undergone radiation therapy for retinoblastoma and developed OS at
intervals of 4, 11, and 18 years there¬ after. Interestingly, patients with re¬ tinoblastoma have an increased risk for developing OS independent of radia¬ tion exposure. Among patients with hereditary retinoblastoma, the inci¬ dence of OS is increased 500 times over the expected incidence.2 Similarly, de¬ letion of the long arm of chromosome 13, which occurs with increased fre¬ quency in patients with retinoblas¬ toma, is also associated with develop¬ ment of OS.12 Benedict et al9 have recently reported the cloning of a re¬ tinoblastoma gene that is thought to be responsible for the development of re¬ tinoblastoma and OS.
Fig 3.—Higher-power photomicrograph of osteogenic sarcoma illustrating osteoblasts amid irregular osteoid (hematoxylin-eosin, original magnification The fourth patient who developed OS after radiation therapy was treated for recurrent fibrous dysplasia of the mandible. One percent of patients with fibrous dysplasia will eventually devel¬ op OS.10 In this case, the patient devel¬ oped OS 4 months after irradiation, which suggests that the sarcoma exist¬ ed before radiation therapy. Histologically, OS comprises a ma-
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
cytologically atypical 100).
lignant, undifferentiated stroma and neoplastic osteoid or bone. The stromal component is characterized by dense cellularity, pleomorphism, and cytolog¬ ically atypical osteoblasts. The mitotic rate of the proliferating stromal cells is often increased. Neoplastic osteoid and bony spicules may be sparsely scat¬ tered throughout the tumor or ar¬ ranged in a fine, complex (filigree)
2 and 3 are photomi¬ crographs that show the typical fea¬ tures of OS. Osteogenic sarcoma can be subclassified into different categories based on the predominant characteristic of the stroma. Osteoblastic, chondroblastic, and fibroblastic types are recog¬ nized. There are also special variants of OS worth mention because of impor¬ tant prognostic implications and often deceptive histologie features. One such variant is the intraosseous well-differ¬ entiated osteosarcoma initially de¬ scribed by Unni et al." Interestingly, this parosteal OS-like tumor is most often misdiagnosed as fibrous dyspla¬ sia. The patient in this series initially treated for fibrous dysplasia may, in fact, have had this prognostically fa¬ vorable variant of OS.
pattern. Figures
Osteogenic
grade based
assigned a anaplasia component. Low-grade
sarcomas are on
the relative
of the stromal lesions (grade 1) may stimulate paros¬ teal OS or fibrous dysplasia. Increas¬
ingly, anaplastic tumors are assigned higher grades, the most cytologically bizarre being the grade 4 lesions. Ma¬ lignant osteogenic neoplasms of the jaw are more often of lower grade and have been reported to have a more favorable prognosis than those of the extremities.12 The largest series of head and neck OS showed no correla¬ tion between histologie character and
prognosis.4
The optimal management for pa¬ tients with OS of the head and neck is unclear. The rarity of the lesion pre¬ cludes adequate patient enrollment in prospective randomized studies. Some authors have favored radical surgery alone. Forteza et al7 reported 85% long-term survival in a series of nine patients managed with radical sur¬ gery. Three patients received radia¬ tion therapy, two preoperatively and one postoperatively. Their data com¬ pare favorably with other series that report a 5-year survival of 20% to
33%.4'5·7'13
Recent small series and case reports suggest that combined therapy em¬
ploying surgery, radiation therapy, an¬ d/or chemotherapy improves results in osteosarcoma.14'17 Although the data are retrospective, the results reported
in the present series tend to support
aggressive
multimodality
therapy.
None of the patients in the UCLA series treated with surgery alone re¬ mained disease free. Of the five longterm disease-free survivors with ade¬ quate follow-up, four received adjuvant radiation therapy and chemo¬ therapy. Because of the high frequen¬ cy of local recurrence, the use of radia¬ tion therapy would be expected to
improve results.18"20 More extensive experience with OS has been gained in the extremities. In
the recent past, the accepted manage¬ ment of OS of the extremities was amputation. Despite radical surgery, the overall 2-year survival had been reported to be between 5% and 20%.17'21"24 The development of effective chemotherapy against osteosarcoma in the 1970s permitted less radical sur¬ gery. Since then, limb-sparing surgery has become the preferred treatment.25 Several studies have demonstrated that neither survival nor the risk of local recurrence is compromised with the limb-sparing approach.22"26 Eilber et al25 reported a 2.7% local recurrence rate in 183 patients with malignant skeletal or high-grade softtissue sarcomas treated according to a limb-salvage protocol. The regimen consisted of preoperative doxorubucin therapy and radiation therapy followed by en bloc resection and postoperative chemotherapy. With a median followup of 32 months, overall survival among patients treated with limb sal¬ vage was identical to that of those undergoing amputation and chemo¬
therapy.26
Two randomized trials of OS of the extremities have reported preliminary data showing improved disease-free survival in patients receiving adjuvant therapy. In a series by Link et al,27114 patients were enrolled to study the advantages of adjuvant chemotherapy. Among the randomized patients, the 2year actuarial disease-free survival rate in the chemotherapy arm of the study was 56% compared with 18% in the surgery alone group (P .001). Among the patients who refused ran¬ domization, the survival rate was 50% in those who received chemotherapy postoperatively vs 10% in those receiv¬ =
ing no adjuvant treatment.27 In another trial, 59 patients
with
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
extremity OS were randomized to re¬ ceive aggressive adjuvant postopera¬
tive chemotherapy or no treatment after receiving preoperative doxorubu¬ cin therapy and radiation therapy. With a median follow-up of 2 years, both disease-free survival and overall survival rates were statistically higher in the patients receiving aggressive adjuvant therapy. In the trials of both Link et al27 and Eilber et al25 , the follow-up period was too short to know whether the advantages will persist.28 Several studies have shown im¬ proved local control and survival when adjuvant radiation therapy with or without chemotherapy is added. Chamber and Mahoney18 reported a 73% 5-year survival in 33 patients re¬ ceiving high-dose preoperative radia¬ tion therapy followed by wide surgical resection. Akbiyik and Alexander19 and de Fries et al2" reported similar results with preoperative irradiation, resec¬ tion, and postoperative chemotherapy. Our data, while based on a small series examined retrospectively, tend to sup¬ port these aggressive approaches. There is no proof that the foregoing results from the extremities are rele¬ vant to head and neck OS. Except for its pattern of failure, OS of the facial bones has a natural history similar to that of OS of the extremities. There is a much lower risk of distant metastasis in the head and neck but a high rate of local recurrence.3 This may be due to the difficulty in achieving wide surgical margins in the head and neck for ana¬ tomic and cosmetic reasons. The litera¬ ture suggests that radical surgery alone frequently results in local fail¬ 3-6'20 Because of the rarity of head ure. and neck OS, it may never be possible to directly study the effect of adjuvant therapy in a prospective, randomized fashion. Data from the extremities are therefore probably the best way to judge the effectiveness of adjuvant treatment.
The
ogies
development of new technol¬ procedures, such as mono¬
and
clonal antibodies, may become rele¬ vant to the treatment of OS in the future. A monoclonal osteonectin anti¬ body has been developed that binds to OS cells.29 In the future, it may be possible to link radioactive isotopes to monoclonal antibodies, permitting spe-
cific cell killing. However, clinical tri¬ als employing monoclonal antibodies in solid tumors have thus far been
disappointing.
In conclusion, OS of the head and neck is a formidable tumor demanding
aggressive treatment. Adequate surgi-
cal resection is the mainstay of treat¬ ment. Patients should receive adju¬ radiation and vant therapy based accumulated on chemotherapy, experience in head and neck OS and recently published data concerning OS of the extremities. Although this se-
ries cannot compare different types of adjuvant therapy, the data support the conclusion that surgery alone is unlike¬ ly to cure most patients with OS. Fu¬ ture studies are required to determine the correct sequence of treatment.
References 1. Antman KH, Eilber FR, Shiu MH. Soft tissue current trends in diagnosis and manage-
sarcomas:
ment. Curr Probl Cancer. 1989;13:337-367. 2. Goorin AM, Abelson HT, Frei E. Osteosarcoma: fifteen years later. N Engl J Med.
1985;313:1637-1643. 3. Malawer MM, Link MP, Donaldson LL. Sarcomas of bone. In: Devita VT, Hellman S, Rosenberg SA, eds. Cancer: Principals and Practice of Oncology. Philadelphia, Pa: JB Lippincott; 1989:1442-1453. 4. Garrington GE, Scofield HH, Cornyn J, et al. Osteosarcoma of the jaws: analysis of 56 cases. Cancer. 1967;20:377-391. 5. Caron AS, Jahder SI, Strong EW. Osteogenic sarcoma of the facial and cranial bones: a review of 43 cases. Am J Surg. 1971;122:719-725. 6. Pritchard DJ. Sarcomas of bone. Cancer Treat Symp. 1983;2:247-250. 7. Forteza G, Colmenero B, Lopez-Barea F. Osteogenic sarcoma of the maxilla and mandible. Oral Surg Oral Med Oral Pathol. 1986;62:179-184. 8. Clark L, Unni KK, Dahlin DC. Osteosarcoma of the jaw. Cancer. 1983;51:2311-2316. 9. Benedict W, Fung Y, Murphee L. The gene responsible for the development of retinoblastoma and osteosarcoma. Cancer. 1988;62:1691-1694. 10. Wick MR, McLeod RA, Siegal GP, Greditzer HG, Unni KK. Sarcomas of bone complicating osteitis deformans (Paget's disease): fifty years experience. Am J Surg Pathol. 1981;5:47-51. 11. Unni KK, Dahler DC, Beabout JW, et al. sarcoma. Parosteal Cancer. osteogenic
1976;378:2466-2475. 12. de Fries HO, Perlin E, Leibel SA. Sarcoma of the mandible. Arch Otolaryngol. 1979;105:358\x=req-\
359. 13. Mirra JH. Bone Tumors: Clinical, Radiologic and Pathologic. Philadelphia, Pa: Lea & Febiger; 1989. 14. McKenna WG, Barnes MM, Kinsella TJ, et al. Combined modality treatment of adult soft tissue sarcomas of the head and neck. Int J Radiat Oncol Biol Phys. 1987;13:1127-1133. 15. Beziat JL, Tabone E, Bailly C, Gerard JP. Osteogenic sarcoma of the tongue. J Oral Maxillo-
fac Surg. 1989;47:524-528. 16. Link MP. Adjuvant therapy in the treatment of osteosarcoma. In: DeVita VJ, Hellman S, Rosenberg SA, eds. Important Advances in Oncology, 1986. Philadelphia, Pa: JB Lippincott; 1986:193\x=req-\ 207. 17. Marcove
RC, Mike V, Hajack JV, et al. Osteogenic sarcoma under the age of 21. J Bone Joint Surg Am. 1970;52:411-423. 18. Chambers RG, Mahoney WD. Osteogenic sarcoma ofthe
J
mandible: current management. Am
Surg. 1970;36:463-471. 19. Akbiyik N, Alexander LL.
Osteosarcoma of the maxilla treated with radiation therapy and surgery. J Natl Med Assoc. 1981;73:355-356. 20. deFries HO, Perlin E, Leibel SA. Treatment of osteogenic sarcoma of the mandible. Arch Oto-
laryngol. 1979;105:358-359. 21. Sweetman R. Surgical management of primary osteosarcoma. Clin Orthop. 1975;111:57-64.
Downloaded From: http://archotol.jamanetwork.com/ by a University of Manitoba User on 06/18/2015
Campanacci M, Bacci G, Betoni F, et al. The osteogenic sarcoma of the extremities: twenty years experience at the Institukto Orthope22.
treatment
dico Rizzoli. Cancer. 1981;48:1569-1581. 23. Eilber FR, Eckhardt J, Morton DL. Advances in the treatment of sarcomas of the extremities: current status of limb salvage. Cancer.
1984;54:2695-2701.
24. Simon MA, Achliman MA, Thomas N, et al. Limb salvage treatment versus amputation for osteosarcoma of the distal end of the femur. J Bone Joint Surg Am. 1986;68:1331-1337. 25. Eilber FR, Morton DL, Eckardt J, et al. Limb salvage for skeletal and soft tissue sarcomas: multidisciplinary preoperative therapy. Cancer.
1984;53:2579-2584.
26. Enneking WF. A system for the functional evaluation for the surgical management of musculoskeletal tumors. In: Enneking WF, ed. Limb Salvage in Musculoskeletal Oncology. New York, NY: Churchill Livingstone Inc; 1987:5-16. 27. Link M, Goorin A, Miser A, et al. The role of adjuvant chemotherapy in the treatment of osteosarcoma of the extremities: preliminary results of the multiinstitutional osteosarcoma study. Proc Am Soc Clin Oncol. 1985;4:237. 28. Eilber FR, Guiliano A, Eckardt J, et al. Adjuvant chemotherapy for osteosarcoma: a randomized prospective trial. J Clin Oncol. 1987;5:21-26. 29. Jundt G, Schultz A, Berghauser KH, et al. Immunocytochemical identification of osteogenic bone tumors by osteonectin antibodies. Virchows Arch A. 1989;414:345-353.
