Case 2-2004 — A 32-Year-Old Man with Pain and Swelling of the Jaw
http://www.100md.com
《新英格兰医药杂志》
Presentation of Case
Dr. H. Daniel Clark (Oral and Maxillofacial Surgery): A 32-year-old man was admitted to the hospital because of a mandibular sarcoma. He had been well until two months before admission, when he had tenderness over the chin and saw a pea-size lump when he pulled his lower lip down. A dentist in another state confirmed the presence of a swelling on the anterior aspect of the mandibular alveolus, in the area of the right mandibular canine. Two or three weeks later, the patient consulted an oral and maxillofacial surgeon, who also found the firm mass, which was now 1 cm in diameter, labial to the lower incisors. A biopsy was performed, and a diagnosis of chondroblastic osteogenic sarcoma was made.
Dental periapical radiographs that had been obtained nine years before presentation showed that the lower right canine and left lateral incisor were tipped and slightly displaced. The patient had been aware of having "crooked" teeth for at least eight years and thought in recent months that the position of the splayed teeth was becoming worse. He was referred to this hospital.
The patient was well otherwise. His only medication was finasteride. He was an executive, and he drank alcohol but had never smoked. There was no family history of malignant tumors.
The pulse was 65 beats per minute and the blood pressure 135/65 mm Hg. On physical examination, there was no facial asymmetry. The mucosa over the mandible had a deep reddish-blue color over the region from the right canine to the left lateral incisor (Figure 1). Slight swelling and tenderness were observed on the lingual side of the mandible near the biopsy site.
Figure 1. Clinical Photograph of the Mandible on Admission.
Swelling and displacement of mandibular teeth are evident.
Differential Diagnosis
Dr. Thomas B. Dodson: May we review the radiologic studies?
Dr. Paul A. Caruso: Periapical dental radiographs that were obtained nine years before the current presentation show splaying of the right lateral incisor and the canine (Figure 2A). The lamina dura along the margin of the teeth is intact, but the periodontal ligament space is slightly enlarged.
Figure 2. Radiographic Studies.
A periapical radiograph obtained nine years before admission (Panel A) shows splaying of the roots of the lateral incisor and canine of the right mandible (arrow). The periodontal ligament space is slightly widened along the mesial aspect of the canine, but the lamina dura, depicted by a radiodense line along the periodontal ligament space, appears to be intact. A panoramic radiograph (Panel B) taken shortly before the biopsy of the lesion shows progressive splaying of the lateral incisor and canine (arrow), loss of definition of the lamina dura along the margin of the canine, and widening of the periodontal ligament space. An axial CT scan obtained after the biopsy (Panel C) shows postoperative changes in the mandible (arrow) and a sclerotic lesion that extends from the labial to the lingual aspect of the mandible, with enlargement of the periodontal ligament space adjacent to the canine. There is irregularity of the cortical margin and splaying of the lateral incisor and canine inferior to the biopsy site.
A panoramic view obtained elsewhere two years before the current presentation showed an increase in the splaying of the teeth, a widening of the periodontal ligament space, and a loss in the definition of the lamina dura relative to earlier images. A panoramic radiograph obtained at the time of presentation and before the biopsy shows progressive splaying of the teeth and widening of the periodontal ligament space (Figure 2B), both findings that can be observed in cancer.
Computed tomographic (CT) scanning of the paranasal sinuses, mandible, and neck, performed after the biopsy and without the use of intravenous contrast material, shows postoperative changes in the right side of the mandible and a sclerotic lesion that extends from the labial to the lingual aspect of the mandible (Figure 2C), with internal sclerosis and enlargement of the periodontal ligament space adjacent to the canine. There is irregularity of the cortical margin and splaying of the lateral incisor and the canine. There were no enlarged lymph nodes. The paranasal sinuses were normal. CT scans of the thorax, abdomen, and pelvis, obtained after the oral and intravenous administration of contrast material, revealed no evidence of metastatic disease.
Magnetic resonance imaging (MRI) examination showed a lesion, 1 cm in diameter, in the right anterior portion of the mandible; there was abnormal low signal on the T1-weighted images and high signal on the T2-weighted images, findings that may have indicated the presence of postbiopsy edema.
In summary, the imaging findings were progressive splaying of the lateral incisor and the canine, widening of the periodontal ligament space, heterogeneous sclerosis, slight cortical irregularity, and (on MRI) infiltration of the marrow cavity, a finding that was somewhat difficult to interpret, given that a biopsy of the area had recently been performed.
The differential diagnosis of a single, primarily sclerotic lesion of the jaw includes infection and certain dysplasias, such as fibrous or mature cemento-osseous dysplasia. Benign tumors, such as cementoblastomas and ossifying fibromas, can also be manifested as isolated sclerotic lesions in the jaw, as can malignant tumors, such as osteosarcomas.
Dr. Dodson: May we review the biopsy findings?
Pathological Discussion
Dr. G. Petur Nielsen: Examination of the biopsy specimen obtained at the other hospital showed several different histologic patterns. In some areas, the tumor cells are spindle shaped and embedded in an extracellular myxoid matrix (Figure 3A). In other areas, the tumor shows chondroblastic differentiation, with the formation of areas mimicking normal hyaline cartilage (Figure 3B). Focally, the neoplastic cells have produced an extracellular, eosinophilic matrix known as osteoid, which is a precursor to bone, and this finding confirms the diagnosis of osteosarcoma (Figure 3C). The relatively low cellularity of the tissue and the large areas of cartilaginous differentiation suggest that the lesion is a grade 1 or 2 (low-to-intermediate grade) chondroblastic osteosarcoma.
Figure 3. Biopsy Specimen of the Mandibular Lesion (Hematoxylin and Eosin, x500).
The tumor has a variety of histologic patterns. In some areas, the tumor cells are spindle shaped and embedded in an extracellular myxoid matrix (Panel A). In other areas, there is chondroblastic differentiation, with areas that mimic normal cartilage (Panel B). In still other areas, the neoplastic cells have produced an extracellular, eosinophilic matrix known as osteoid, which is a precursor to bone (Panel C) — a finding that confirms the diagnosis of osteosarcoma.
Discussion of Management
Dr. Dodson: Osteosarcoma involving the maxillofacial region is a challenge that requires the attention of many clinical specialists. Because of its rarity, it is hard for a single clinician or institution to have a large amount of experience with it. To formulate a treatment plan for this patient, I applied the principles of evidence-based clinical practice.
Evidence-based clinical practice is a paradigm for clinical decision making that integrates findings from the clinical literature with the clinicians' and the patients' preferences and values to provide informed, compassionate treatment.1 The prospect of performing an extensive literature search can be intimidating, but with practice and experience it is possible to become efficient in this task.
A valuable first step when addressing an unfamiliar topic or unusual clinical condition is to perform a search to obtain background information. The most efficient source for background-level information is secondary publications that abstract and summarize high-quality articles.2 In this case, I used an online service available through the library at this hospital. I searched using the key word "osteosarcoma" and found the heading "head and neck sarcomas." I was quickly able to find and review the following background material.
Osteosarcoma of the jaw is a rare and highly malignant bone tumor. It represents less than 10 percent of all osteosarcomas. There are approximately 110 new cases in the United States each year. The mean age of patients is 36 years (range, 5 to 78), and the age distribution is bimodal, with one peak in the third decade and a smaller peak after the age of 50.3 The condition occurs slightly more often in men than in women. The mandible is involved more frequently than the maxilla. Risk factors include previous radiotherapy, Paget's disease, previous diagnosis of an atypical fibro-osseous lesion, trauma, exposure to chemical agents such as beryllium oxide, viral infection, and hereditary predisposition.3,4
Signs and symptoms on presentation include swelling, sometimes pain, and paresthesia, which is an ominous finding. Patients may report changes in tooth position, loose teeth, or a change in the fit of a dental prosthesis. Many of these signs and symptoms are nonspecific, and there is often a delay of three to four months before the diagnosis is made.5,6 Factors associated with a better prognosis include young age, negative margins after resection, small tumor size (diameters smaller than 10 cm are associated with better survival, and diameters smaller than 4 cm are associated with a decreased risk of local recurrence), and low-to-intermediate histologic grade.5,6,7,8 The long-term prognosis is bleak. In one review, the 2-, 5-, and 10-year survival rates were 58, 37, and 20 percent, respectively.9 Fifty-six percent of the patients had recurrent disease at one or more sites. Local recurrence occurred in 50 percent and distant metastases in 20 percent.
The rarity of the disease and the lack of prospective studies make it difficult to render well-supported treatment recommendations. Most authors agree that the best outcomes are associated with en bloc resection with adequate margins (larger than 5 mm).5 For osteosarcoma of the jaw, unlike osteosarcoma of the appendicular skeleton, for which chemotherapy has revolutionized management, the role of adjuvant radiation therapy or chemotherapy is unclear.
In an attempt to develop a treatment plan, I applied a tool from evidence-based clinical management known as the critical appraisal exercise.10,11,12 The first step is to convert information needs into an answerable question. The method of formulating a good clinical question is known as PICO,13,14,15 where P denotes patient or population, I intervention, C control or comparison group, and O outcome. In this case, the clinical question was: "Among patients (P) with osteosarcoma of the jaw, do those receiving adjuvant therapy (I), as compared with those undergoing surgery alone (C), survive longer (O)?"
The second step in the critical appraisal exercise is to search the literature. An effective practitioner of evidence-based clinical management needs to hone his or her skills in this regard and search the literature with an efficiency bordering on the ruthless. The most efficient way to complete the search process is to search and review citations from secondary publications in which critical appraisal exercises or other synopses of various clinical questions and issues have already been completed. I searched Medline with the help of a search engine available at our hospital library, one that incorporated various filters to assist in the identification of high-quality articles related to natural history, prognosis, diagnosis, etiology, and therapy. These features make it possible to carry out complex searches in a relatively short time. Full-text versions of the key citations were obtained electronically.
The third step is to appraise the quality of the evidence in the literature, select the articles with the highest level of evidence to support clinical decision making, and abstract the articles. To select the best articles for review, I applied what is known as the "hierarchy of evidence" (Table 1).16 Studies derived from randomized clinical trials are considered level 1 evidence and provide the strongest support for clinical decision making. Cohort studies, case–control studies, case series, and "expert opinions" (levels 2 through 5, respectively) are considered progressively weaker sources of support for clinical decision making.
Table 1. Levels of Evidence to Support Treatment Decisions.
For osteosarcoma of the jaw, the quality of evidence that I could find to support clinical decision making was poor. The studies I identified were retrospective case series enrolling a few patients each year from single institutions or studies derived from combining data from many institutions (level-4 evidence). Because of the large time spans of these studies (as long as 37 years) — periods during which diagnostic, grading, and treatment schemes have evolved — it was difficult to evaluate the outcomes.17
Fortunately, one article had abstracted and tabulated the findings of 23 studies, and that article became the prime resource for this critical appraisal exercise.9 I briefly scanned the remaining articles to identify clinical "pearls" that may help guide or support treatment recommendations. Kassir et al.9 identified 23 studies for which treatment and survival data were reported; the studies comprised 163 patients, and the authors added 15 additional patients from their own institution. The median duration of survival among the patients treated with resection alone was 60 months, whereas that among the patients treated with surgery and adjuvant therapy was 24 months. There is a negative treatment bias, since patients with poor prognostic factors, such as positive surgical margins or large or high-grade tumors, are disproportionately selected to receive adjuvant therapy, whereas patients with smaller or lower-grade tumors are treated only with resection. However, there was no compelling evidence that adjuvant therapy contributed any incremental benefit in terms of survival in this or any of the other articles I reviewed. The role of neoadjuvant therapy in this disease, in contrast to osteosarcoma of the appendicular skeleton, is also unclear.
After carefully searching and reviewing the literature, I found that the clinical question remains unanswered. Nonetheless, we must proceed to the fourth step and recommend a treatment strategy. I would make the following recommendations for this patient, who has favorable prognostic factors, since he is young, and his tumor is of low-to-intermediate grade and small. The primary treatment should be surgical, with en bloc resection of the lesion with adequate margins (more than 5 mm). After the operation, I would reassess the prognostic variables on the basis of the findings on pathological examination of the surgical specimen. If the margins are negative, I would consider adjuvant radiotherapy to ensure local control and to decrease the risk of local recurrence. If the margins are positive, I would recommend additional surgery followed by adjuvant radiation therapy for local control and adjuvant chemotherapy for prevention of distant metastatic disease. If the lesion is large (more than 10 cm in diameter) or is a high-grade tumor, I would consider the use of chemotherapy even if margins are negative.
At this point, I would like to invite Dr. McIntyre to comment on the role of radiation therapy.
Dr. James F. McIntyre (Radiation Oncology): Osteosarcoma has a high rate of local recurrence. Since radiotherapy has been shown to control microscopic disease in long-bone osteosarcomas, it seems reasonable to assume that radiation can sterilize microscopic disease in head and neck osteosarcomas. Therefore, I would treat this young patient with aggressive local therapy: wide local resection and, depending on the pathological findings, postoperative radiotherapy.
Dr. Dodson: Dr. Clark will tell us about the subsequent management.
Dr. Clark: Dr. Leonard B. Kaban performed en bloc resection of the mandible, including the overlying muscle and mucosa, from the left premolar region to the right molar region to achieve 1.5-cm margins. After the jaw had been immobilized, a rigid titanium fixation plate was adapted to the contour of the mandible and fixed in place with three screws on either side of the defect. The immobilized jaw was released, and occlusion was verified. Intraoral wound closure was achieved without tension. The postoperative course was uneventful.
Dr. Dodson: Dr. Nielsen, will you comment on the pathological examination of the surgical specimen?
Dr. Nielsen: The resected specimen was a segment of the mandible, 5.5 by 5.5 by 1.3 cm (Figure 4A). On gross examination, there was diffuse thickening of the bone, with no discrete mass identified. Microscopically, however, there was residual osteosarcoma extensively involving the entire specimen (Figure 4B and Figure 4C), with a microscopical appearance identical to that of the biopsy specimen. The bony margins at either end were free of tumor; the distance of the tumor from the resection margin could not be determined but was believed to be less than 0.5 cm. The buccal surface of the specimen was negative for tumor; on the lingual aspect, the tumor had infiltrated the cortex and was exposed on the mucosal surface in one focus. No vascular invasion was seen.
Figure 4. Pathological Features of the Resected Bony Tumor.
A photograph taken in the operating room (Panel A) shows a segment of bone, 5.5 by 5.5 by 1.3 cm, that is diffusely thickened, with no discrete mass. Microscopical examination shows diffusely infiltrating tumor surrounding a tooth (Panel B; hematoxylin and eosin, x31) and bone formation (Panel C; hematoxylin and eosin, x500).
Dr. Dodson: With this additional information, we can revise the treatment plan. The lesion is less than 10 cm in diameter, so the prognosis for long-term survival is good. Because it is larger than 4 cm, however, there is an increased risk of local recurrence, but it is still low to intermediate in grade. Although the bony margins are negative, they are less than 0.5 cm from the edge of the specimen, and there is tumor exposed on the gingival surface. I would recommend additional surgery and subsequent radiation therapy for local control. The role of chemotherapy is unknown, and there are no good data to support or refute its efficacy. At this point, it is surely time to sit down with the patient and discuss the potential benefits and risks of chemotherapy. Dr. Harmon, can you tell us how you advised the patient?
Dr. David C. Harmon (Hematology–Oncology): In the absence of information from controlled trials to define the role of chemotherapy in this setting, the patient and I had extensive discussions and reviewed some of the available research concerning osteosarcoma of the jaw. Most reports indicated a high mortality rate, but recent reports of treatment involving radical resection seemed more optimistic. In a series at this institution,5 a trend toward improved survival among patients who received chemotherapy was noted. A more recent, larger series from Memorial Sloan-Kettering Cancer Center6 also suggested an improvement with the more aggressive, combined-treatment approach now in use. The few randomized trials of chemotherapy in osteosarcoma involved patients with large lesions in the arms or legs. Adjuvant chemotherapy substantially improves the cure rates in that high-risk setting. The toxicity, inconvenience, and expense of chemotherapy — close to a year of treatment with high-dose methotrexate, doxorubicin, cisplatin, and ifosfamide or similar drugs — are substantial. Nevertheless, even a 10 percent improvement in the cure rate would more than offset the small but real risk of death from chemotherapy. After consideration, this patient thought that chemotherapy was worth a try.
Dr. Dodson: In summary, my treatment recommendation consists of additional surgery, radiation therapy for local control, and chemotherapy. Dr. Clark will summarize the subsequent treatment.
Dr. Clark: After discussion with the patient, we operated again at the original surgical site and resected an additional 2 cm of bone and soft tissue. Because of the large size of the resulting defect and the plan to use adjuvant radiation therapy, Dr. Mark Varvares performed immediate reconstruction with a vascularized, composite flap of bone and soft tissue from the fibula. The bony fragments were stabilized with a new rigid reconstruction plate. The tissue from the fibula was transferred to the site, and arterial and venous anastomoses were constructed with the use of an operating microscope (Figure 5). Pathological examination of the excised tissue revealed no residual tumor.
Figure 5. Intraoperative Photograph of the Mandibular Defect after Reconstruction with a Vascularized Fibular Bone Graft and Titanium Reconstruction Plate.
Dr. Dodson: Dr. Kaban, will you comment?
Dr. Leonard B. Kaban (Oral and Maxillofacial Surgery): This very difficult case illustrates the many challenges clinicians encounter in managing primary jaw tumors of mesenchymal origin. These tumors can occur in both young and old persons and have a spectrum of clinical behavior that ranges from benign to locally aggressive to malignant. The histologic diagnoses include fibro-osseous lesions, atypical fibro-osseous lesions, aggressive fibromatoses, desmoplastic fibroma, myxoma, fibromyxoma, fibrosarcoma, and osteosarcoma. There are no accurate predictors of these tumors' clinical or biologic behavior.
I suspect that the bone of the patient's right mandible was abnormal back in 1993 and certainly in 2000. Had biopsy of the lesion been performed at that time, pathological examination probably would have identified it as a fibro-osseous or atypical fibro-osseous lesion. At some point it evolved into an osteosarcoma.
The infiltrative nature of this tumor was an important problem. The initial workup suggested that it was small and that it was localized to the alveolus around the lateral incisor and canine teeth of the right mandible. Despite a generous resection, the margins were surprisingly close to the tumor, and the tumor was remarkably infiltrative.
This case offers some take-home lessons. First, a biopsy specimen must be obtained from any lesion that results in tooth displacement, even when there are no symptoms and the x-rays look benign. Second, very large margins (2 to 3 cm) should be planned for jaw osteosarcomas. Third, microvascular technology allows radical resection and radiotherapy. Fourth, biologic markers to predict the risk of local recurrence and systemic spread would be useful. The close collaboration of all the specialists involved in this patient's care — including oral and maxillofacial surgeons, otolaryngologists, head and neck surgeons, and radiation and medical oncologists — was critical.
Dr. Dodson: Dr. Varvares, will you comment on the reconstruction?
Dr. Mark A. Varvares (Otolaryngology): Many of us are old enough to remember what used to happen to patients who had a large anterior mandibular defect. A variety of reconstructive techniques were attempted, among them reconstruction with alloplastic implants, nonvascularized bone grafts, and even pectoralis major myocutaneous flaps with vascularized rib. All these approaches were unreliable, and these disfigured patients typically had serious social problems. In the 1970s, transfer of vascularized free tissue for head and neck reconstruction was introduced. In most medical centers, patients now undergo immediate mandibular reconstruction with vascularized bone and soft tissue for anterior and lateral mandibular defects. We now achieve a 95 percent success rate in patients with defects that were previously considered to be irreparable.
The fibula is the usual source of tissue used for oral mandibular reconstruction. The bone can be readily molded to the contour of the mandible, it accepts implants well, and a defect as long as 25 cm can be reconstructed. Soft tissue that is sufficient to cover the vascularized bone can be obtained from the donor site.
Dr. Dodson: Dr. Clark, will you give us an update on the status of the patient?
Dr. Clark: Two months after the reconstructive surgery, the swelling had greatly diminished, the patient's bite appeared reasonable, and all the wounds had healed. Several of his physicians believed that he should have chemotherapy and delay irradiation, whereas others believed that irradiation should be the next step. After extensive consultation, the patient elected to proceed with radiation therapy.
He has regained weight and strength and has returned to his normal activities. He has a removable partial denture to restore his teeth, and he is considered tumor-free. He has moved out of Massachusetts and is expected to begin chemotherapy soon.
Addendum
Dr. Kaban: I saw this patient in follow-up 13 months after the definitive reconstruction. He had completed his chemotherapy and was free of any evidence of disease.
Source Information
From the Departments of Oral and Maxillofacial Surgery (T.B.D.), Radiology (P.A.C.), and Pathology (G.P.N.), Massachusetts General Hospital; and the Departments of Radiology (P.A.C.) and Pathology (G.P.N.), Harvard Medical School.
References
Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based medicine: how to practice and teach EBM. New York: Churchill Livingstone, 1997:2.
Guyatt G, Rennie D, eds. Users' guides to the medical literature: essentials of evidence-based clinical practice. Chicago: AMA Press, 2002:21-6.
Brockstein BE. Head and neck sarcomas. Wellesley, Mass.: UptoDate, 2002. (Also available at http://www.uptodate.com/.)
Fuchs B, Pritchard DJ. Etiology of osteosarcoma. Clin Orthop 2002;397:40-52.
August M, Magennis P, Dewitt D. Osteogenic sarcomas of the jaws: factors influencing prognosis. Int J Oral Maxillofac Surg 1997;26:198-204.
Patel SG, Meyers P, Huvos AG, et al. Improved outcomes in patients with osteogenic sarcoma of the head and neck. Cancer 2002;95:1495-1503.
Delgado R, Maafs E, Alfeiran A, et al. Osteosarcoma of the jaw. Head Neck 1994;16:246-252.
Gadwal SH, Gannon FH, Fanburg-Smith JC, Becoskie EM, Thompson LDR. Primary osteosarcoma of the head and neck in pediatric patients: a clinicopathologic study of 22 cases with a review of the literature. Cancer 2001;91:598-605.
Kassir RR, Rassekh CH, Kinsella JB, Segas J, Carrau RL, Hokanson JA. Osteosarcoma of the head and neck: meta-analysis of nonrandomized studies. Laryngoscope 1997;107:56-61.
Evidence-Based Medicine Working Group. Evidence-based medicine: a new approach to teaching the practice of medicine. JAMA 1992;268:2420-2425.
Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based medicine: how to practice and teach EBM. New York: Churchill Livingstone, 1997:3.
McKibbon A, Hunt D, Richardson WS, et al. Finding the evidence. In: Guyatt G, Rennie D, eds. Users' guides to the medical literature: essentials of evidence-based clinical practice. Chicago: AMA Press, 2002:21-73.
How to ask questions you can answer. In: Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based medicine: how to practice and teach EBM. New York: Churchill Livingstone, 1997:21-36.
McKibbon A, Hunt D, Richardson WS, et al. Finding the evidence. In: Guyatt G, Rennie D, eds. Users' guides to the medical literature: essentials of evidence-based clinical practice. Chicago: AMA Press, 2002:27.
Focusing clinical questions. Oxford, England: Centre for Evidence-Based Medicine. (Accessed December 18, 2003, at http://www.cebm.net/focus_quest.asp.)
Levels of evidence and grades of recommendation. Oxford, England: Centre for Evidence-Based Medicine. (Accessed December 18, 2003, at http://www.cebm.net/levels_of_evidence.asp.)
Bertoni F, Dallera P, Bacchini P, Marchetti C, Campobassi A. The Istituto Rizzoli-Beretta experience with osteosarcoma of the jaw. Cancer 1991;68:1555-1563.(Thomas B. Dodson, D.M.D.,)
Dr. H. Daniel Clark (Oral and Maxillofacial Surgery): A 32-year-old man was admitted to the hospital because of a mandibular sarcoma. He had been well until two months before admission, when he had tenderness over the chin and saw a pea-size lump when he pulled his lower lip down. A dentist in another state confirmed the presence of a swelling on the anterior aspect of the mandibular alveolus, in the area of the right mandibular canine. Two or three weeks later, the patient consulted an oral and maxillofacial surgeon, who also found the firm mass, which was now 1 cm in diameter, labial to the lower incisors. A biopsy was performed, and a diagnosis of chondroblastic osteogenic sarcoma was made.
Dental periapical radiographs that had been obtained nine years before presentation showed that the lower right canine and left lateral incisor were tipped and slightly displaced. The patient had been aware of having "crooked" teeth for at least eight years and thought in recent months that the position of the splayed teeth was becoming worse. He was referred to this hospital.
The patient was well otherwise. His only medication was finasteride. He was an executive, and he drank alcohol but had never smoked. There was no family history of malignant tumors.
The pulse was 65 beats per minute and the blood pressure 135/65 mm Hg. On physical examination, there was no facial asymmetry. The mucosa over the mandible had a deep reddish-blue color over the region from the right canine to the left lateral incisor (Figure 1). Slight swelling and tenderness were observed on the lingual side of the mandible near the biopsy site.
Figure 1. Clinical Photograph of the Mandible on Admission.
Swelling and displacement of mandibular teeth are evident.
Differential Diagnosis
Dr. Thomas B. Dodson: May we review the radiologic studies?
Dr. Paul A. Caruso: Periapical dental radiographs that were obtained nine years before the current presentation show splaying of the right lateral incisor and the canine (Figure 2A). The lamina dura along the margin of the teeth is intact, but the periodontal ligament space is slightly enlarged.
Figure 2. Radiographic Studies.
A periapical radiograph obtained nine years before admission (Panel A) shows splaying of the roots of the lateral incisor and canine of the right mandible (arrow). The periodontal ligament space is slightly widened along the mesial aspect of the canine, but the lamina dura, depicted by a radiodense line along the periodontal ligament space, appears to be intact. A panoramic radiograph (Panel B) taken shortly before the biopsy of the lesion shows progressive splaying of the lateral incisor and canine (arrow), loss of definition of the lamina dura along the margin of the canine, and widening of the periodontal ligament space. An axial CT scan obtained after the biopsy (Panel C) shows postoperative changes in the mandible (arrow) and a sclerotic lesion that extends from the labial to the lingual aspect of the mandible, with enlargement of the periodontal ligament space adjacent to the canine. There is irregularity of the cortical margin and splaying of the lateral incisor and canine inferior to the biopsy site.
A panoramic view obtained elsewhere two years before the current presentation showed an increase in the splaying of the teeth, a widening of the periodontal ligament space, and a loss in the definition of the lamina dura relative to earlier images. A panoramic radiograph obtained at the time of presentation and before the biopsy shows progressive splaying of the teeth and widening of the periodontal ligament space (Figure 2B), both findings that can be observed in cancer.
Computed tomographic (CT) scanning of the paranasal sinuses, mandible, and neck, performed after the biopsy and without the use of intravenous contrast material, shows postoperative changes in the right side of the mandible and a sclerotic lesion that extends from the labial to the lingual aspect of the mandible (Figure 2C), with internal sclerosis and enlargement of the periodontal ligament space adjacent to the canine. There is irregularity of the cortical margin and splaying of the lateral incisor and the canine. There were no enlarged lymph nodes. The paranasal sinuses were normal. CT scans of the thorax, abdomen, and pelvis, obtained after the oral and intravenous administration of contrast material, revealed no evidence of metastatic disease.
Magnetic resonance imaging (MRI) examination showed a lesion, 1 cm in diameter, in the right anterior portion of the mandible; there was abnormal low signal on the T1-weighted images and high signal on the T2-weighted images, findings that may have indicated the presence of postbiopsy edema.
In summary, the imaging findings were progressive splaying of the lateral incisor and the canine, widening of the periodontal ligament space, heterogeneous sclerosis, slight cortical irregularity, and (on MRI) infiltration of the marrow cavity, a finding that was somewhat difficult to interpret, given that a biopsy of the area had recently been performed.
The differential diagnosis of a single, primarily sclerotic lesion of the jaw includes infection and certain dysplasias, such as fibrous or mature cemento-osseous dysplasia. Benign tumors, such as cementoblastomas and ossifying fibromas, can also be manifested as isolated sclerotic lesions in the jaw, as can malignant tumors, such as osteosarcomas.
Dr. Dodson: May we review the biopsy findings?
Pathological Discussion
Dr. G. Petur Nielsen: Examination of the biopsy specimen obtained at the other hospital showed several different histologic patterns. In some areas, the tumor cells are spindle shaped and embedded in an extracellular myxoid matrix (Figure 3A). In other areas, the tumor shows chondroblastic differentiation, with the formation of areas mimicking normal hyaline cartilage (Figure 3B). Focally, the neoplastic cells have produced an extracellular, eosinophilic matrix known as osteoid, which is a precursor to bone, and this finding confirms the diagnosis of osteosarcoma (Figure 3C). The relatively low cellularity of the tissue and the large areas of cartilaginous differentiation suggest that the lesion is a grade 1 or 2 (low-to-intermediate grade) chondroblastic osteosarcoma.
Figure 3. Biopsy Specimen of the Mandibular Lesion (Hematoxylin and Eosin, x500).
The tumor has a variety of histologic patterns. In some areas, the tumor cells are spindle shaped and embedded in an extracellular myxoid matrix (Panel A). In other areas, there is chondroblastic differentiation, with areas that mimic normal cartilage (Panel B). In still other areas, the neoplastic cells have produced an extracellular, eosinophilic matrix known as osteoid, which is a precursor to bone (Panel C) — a finding that confirms the diagnosis of osteosarcoma.
Discussion of Management
Dr. Dodson: Osteosarcoma involving the maxillofacial region is a challenge that requires the attention of many clinical specialists. Because of its rarity, it is hard for a single clinician or institution to have a large amount of experience with it. To formulate a treatment plan for this patient, I applied the principles of evidence-based clinical practice.
Evidence-based clinical practice is a paradigm for clinical decision making that integrates findings from the clinical literature with the clinicians' and the patients' preferences and values to provide informed, compassionate treatment.1 The prospect of performing an extensive literature search can be intimidating, but with practice and experience it is possible to become efficient in this task.
A valuable first step when addressing an unfamiliar topic or unusual clinical condition is to perform a search to obtain background information. The most efficient source for background-level information is secondary publications that abstract and summarize high-quality articles.2 In this case, I used an online service available through the library at this hospital. I searched using the key word "osteosarcoma" and found the heading "head and neck sarcomas." I was quickly able to find and review the following background material.
Osteosarcoma of the jaw is a rare and highly malignant bone tumor. It represents less than 10 percent of all osteosarcomas. There are approximately 110 new cases in the United States each year. The mean age of patients is 36 years (range, 5 to 78), and the age distribution is bimodal, with one peak in the third decade and a smaller peak after the age of 50.3 The condition occurs slightly more often in men than in women. The mandible is involved more frequently than the maxilla. Risk factors include previous radiotherapy, Paget's disease, previous diagnosis of an atypical fibro-osseous lesion, trauma, exposure to chemical agents such as beryllium oxide, viral infection, and hereditary predisposition.3,4
Signs and symptoms on presentation include swelling, sometimes pain, and paresthesia, which is an ominous finding. Patients may report changes in tooth position, loose teeth, or a change in the fit of a dental prosthesis. Many of these signs and symptoms are nonspecific, and there is often a delay of three to four months before the diagnosis is made.5,6 Factors associated with a better prognosis include young age, negative margins after resection, small tumor size (diameters smaller than 10 cm are associated with better survival, and diameters smaller than 4 cm are associated with a decreased risk of local recurrence), and low-to-intermediate histologic grade.5,6,7,8 The long-term prognosis is bleak. In one review, the 2-, 5-, and 10-year survival rates were 58, 37, and 20 percent, respectively.9 Fifty-six percent of the patients had recurrent disease at one or more sites. Local recurrence occurred in 50 percent and distant metastases in 20 percent.
The rarity of the disease and the lack of prospective studies make it difficult to render well-supported treatment recommendations. Most authors agree that the best outcomes are associated with en bloc resection with adequate margins (larger than 5 mm).5 For osteosarcoma of the jaw, unlike osteosarcoma of the appendicular skeleton, for which chemotherapy has revolutionized management, the role of adjuvant radiation therapy or chemotherapy is unclear.
In an attempt to develop a treatment plan, I applied a tool from evidence-based clinical management known as the critical appraisal exercise.10,11,12 The first step is to convert information needs into an answerable question. The method of formulating a good clinical question is known as PICO,13,14,15 where P denotes patient or population, I intervention, C control or comparison group, and O outcome. In this case, the clinical question was: "Among patients (P) with osteosarcoma of the jaw, do those receiving adjuvant therapy (I), as compared with those undergoing surgery alone (C), survive longer (O)?"
The second step in the critical appraisal exercise is to search the literature. An effective practitioner of evidence-based clinical management needs to hone his or her skills in this regard and search the literature with an efficiency bordering on the ruthless. The most efficient way to complete the search process is to search and review citations from secondary publications in which critical appraisal exercises or other synopses of various clinical questions and issues have already been completed. I searched Medline with the help of a search engine available at our hospital library, one that incorporated various filters to assist in the identification of high-quality articles related to natural history, prognosis, diagnosis, etiology, and therapy. These features make it possible to carry out complex searches in a relatively short time. Full-text versions of the key citations were obtained electronically.
The third step is to appraise the quality of the evidence in the literature, select the articles with the highest level of evidence to support clinical decision making, and abstract the articles. To select the best articles for review, I applied what is known as the "hierarchy of evidence" (Table 1).16 Studies derived from randomized clinical trials are considered level 1 evidence and provide the strongest support for clinical decision making. Cohort studies, case–control studies, case series, and "expert opinions" (levels 2 through 5, respectively) are considered progressively weaker sources of support for clinical decision making.
Table 1. Levels of Evidence to Support Treatment Decisions.
For osteosarcoma of the jaw, the quality of evidence that I could find to support clinical decision making was poor. The studies I identified were retrospective case series enrolling a few patients each year from single institutions or studies derived from combining data from many institutions (level-4 evidence). Because of the large time spans of these studies (as long as 37 years) — periods during which diagnostic, grading, and treatment schemes have evolved — it was difficult to evaluate the outcomes.17
Fortunately, one article had abstracted and tabulated the findings of 23 studies, and that article became the prime resource for this critical appraisal exercise.9 I briefly scanned the remaining articles to identify clinical "pearls" that may help guide or support treatment recommendations. Kassir et al.9 identified 23 studies for which treatment and survival data were reported; the studies comprised 163 patients, and the authors added 15 additional patients from their own institution. The median duration of survival among the patients treated with resection alone was 60 months, whereas that among the patients treated with surgery and adjuvant therapy was 24 months. There is a negative treatment bias, since patients with poor prognostic factors, such as positive surgical margins or large or high-grade tumors, are disproportionately selected to receive adjuvant therapy, whereas patients with smaller or lower-grade tumors are treated only with resection. However, there was no compelling evidence that adjuvant therapy contributed any incremental benefit in terms of survival in this or any of the other articles I reviewed. The role of neoadjuvant therapy in this disease, in contrast to osteosarcoma of the appendicular skeleton, is also unclear.
After carefully searching and reviewing the literature, I found that the clinical question remains unanswered. Nonetheless, we must proceed to the fourth step and recommend a treatment strategy. I would make the following recommendations for this patient, who has favorable prognostic factors, since he is young, and his tumor is of low-to-intermediate grade and small. The primary treatment should be surgical, with en bloc resection of the lesion with adequate margins (more than 5 mm). After the operation, I would reassess the prognostic variables on the basis of the findings on pathological examination of the surgical specimen. If the margins are negative, I would consider adjuvant radiotherapy to ensure local control and to decrease the risk of local recurrence. If the margins are positive, I would recommend additional surgery followed by adjuvant radiation therapy for local control and adjuvant chemotherapy for prevention of distant metastatic disease. If the lesion is large (more than 10 cm in diameter) or is a high-grade tumor, I would consider the use of chemotherapy even if margins are negative.
At this point, I would like to invite Dr. McIntyre to comment on the role of radiation therapy.
Dr. James F. McIntyre (Radiation Oncology): Osteosarcoma has a high rate of local recurrence. Since radiotherapy has been shown to control microscopic disease in long-bone osteosarcomas, it seems reasonable to assume that radiation can sterilize microscopic disease in head and neck osteosarcomas. Therefore, I would treat this young patient with aggressive local therapy: wide local resection and, depending on the pathological findings, postoperative radiotherapy.
Dr. Dodson: Dr. Clark will tell us about the subsequent management.
Dr. Clark: Dr. Leonard B. Kaban performed en bloc resection of the mandible, including the overlying muscle and mucosa, from the left premolar region to the right molar region to achieve 1.5-cm margins. After the jaw had been immobilized, a rigid titanium fixation plate was adapted to the contour of the mandible and fixed in place with three screws on either side of the defect. The immobilized jaw was released, and occlusion was verified. Intraoral wound closure was achieved without tension. The postoperative course was uneventful.
Dr. Dodson: Dr. Nielsen, will you comment on the pathological examination of the surgical specimen?
Dr. Nielsen: The resected specimen was a segment of the mandible, 5.5 by 5.5 by 1.3 cm (Figure 4A). On gross examination, there was diffuse thickening of the bone, with no discrete mass identified. Microscopically, however, there was residual osteosarcoma extensively involving the entire specimen (Figure 4B and Figure 4C), with a microscopical appearance identical to that of the biopsy specimen. The bony margins at either end were free of tumor; the distance of the tumor from the resection margin could not be determined but was believed to be less than 0.5 cm. The buccal surface of the specimen was negative for tumor; on the lingual aspect, the tumor had infiltrated the cortex and was exposed on the mucosal surface in one focus. No vascular invasion was seen.
Figure 4. Pathological Features of the Resected Bony Tumor.
A photograph taken in the operating room (Panel A) shows a segment of bone, 5.5 by 5.5 by 1.3 cm, that is diffusely thickened, with no discrete mass. Microscopical examination shows diffusely infiltrating tumor surrounding a tooth (Panel B; hematoxylin and eosin, x31) and bone formation (Panel C; hematoxylin and eosin, x500).
Dr. Dodson: With this additional information, we can revise the treatment plan. The lesion is less than 10 cm in diameter, so the prognosis for long-term survival is good. Because it is larger than 4 cm, however, there is an increased risk of local recurrence, but it is still low to intermediate in grade. Although the bony margins are negative, they are less than 0.5 cm from the edge of the specimen, and there is tumor exposed on the gingival surface. I would recommend additional surgery and subsequent radiation therapy for local control. The role of chemotherapy is unknown, and there are no good data to support or refute its efficacy. At this point, it is surely time to sit down with the patient and discuss the potential benefits and risks of chemotherapy. Dr. Harmon, can you tell us how you advised the patient?
Dr. David C. Harmon (Hematology–Oncology): In the absence of information from controlled trials to define the role of chemotherapy in this setting, the patient and I had extensive discussions and reviewed some of the available research concerning osteosarcoma of the jaw. Most reports indicated a high mortality rate, but recent reports of treatment involving radical resection seemed more optimistic. In a series at this institution,5 a trend toward improved survival among patients who received chemotherapy was noted. A more recent, larger series from Memorial Sloan-Kettering Cancer Center6 also suggested an improvement with the more aggressive, combined-treatment approach now in use. The few randomized trials of chemotherapy in osteosarcoma involved patients with large lesions in the arms or legs. Adjuvant chemotherapy substantially improves the cure rates in that high-risk setting. The toxicity, inconvenience, and expense of chemotherapy — close to a year of treatment with high-dose methotrexate, doxorubicin, cisplatin, and ifosfamide or similar drugs — are substantial. Nevertheless, even a 10 percent improvement in the cure rate would more than offset the small but real risk of death from chemotherapy. After consideration, this patient thought that chemotherapy was worth a try.
Dr. Dodson: In summary, my treatment recommendation consists of additional surgery, radiation therapy for local control, and chemotherapy. Dr. Clark will summarize the subsequent treatment.
Dr. Clark: After discussion with the patient, we operated again at the original surgical site and resected an additional 2 cm of bone and soft tissue. Because of the large size of the resulting defect and the plan to use adjuvant radiation therapy, Dr. Mark Varvares performed immediate reconstruction with a vascularized, composite flap of bone and soft tissue from the fibula. The bony fragments were stabilized with a new rigid reconstruction plate. The tissue from the fibula was transferred to the site, and arterial and venous anastomoses were constructed with the use of an operating microscope (Figure 5). Pathological examination of the excised tissue revealed no residual tumor.
Figure 5. Intraoperative Photograph of the Mandibular Defect after Reconstruction with a Vascularized Fibular Bone Graft and Titanium Reconstruction Plate.
Dr. Dodson: Dr. Kaban, will you comment?
Dr. Leonard B. Kaban (Oral and Maxillofacial Surgery): This very difficult case illustrates the many challenges clinicians encounter in managing primary jaw tumors of mesenchymal origin. These tumors can occur in both young and old persons and have a spectrum of clinical behavior that ranges from benign to locally aggressive to malignant. The histologic diagnoses include fibro-osseous lesions, atypical fibro-osseous lesions, aggressive fibromatoses, desmoplastic fibroma, myxoma, fibromyxoma, fibrosarcoma, and osteosarcoma. There are no accurate predictors of these tumors' clinical or biologic behavior.
I suspect that the bone of the patient's right mandible was abnormal back in 1993 and certainly in 2000. Had biopsy of the lesion been performed at that time, pathological examination probably would have identified it as a fibro-osseous or atypical fibro-osseous lesion. At some point it evolved into an osteosarcoma.
The infiltrative nature of this tumor was an important problem. The initial workup suggested that it was small and that it was localized to the alveolus around the lateral incisor and canine teeth of the right mandible. Despite a generous resection, the margins were surprisingly close to the tumor, and the tumor was remarkably infiltrative.
This case offers some take-home lessons. First, a biopsy specimen must be obtained from any lesion that results in tooth displacement, even when there are no symptoms and the x-rays look benign. Second, very large margins (2 to 3 cm) should be planned for jaw osteosarcomas. Third, microvascular technology allows radical resection and radiotherapy. Fourth, biologic markers to predict the risk of local recurrence and systemic spread would be useful. The close collaboration of all the specialists involved in this patient's care — including oral and maxillofacial surgeons, otolaryngologists, head and neck surgeons, and radiation and medical oncologists — was critical.
Dr. Dodson: Dr. Varvares, will you comment on the reconstruction?
Dr. Mark A. Varvares (Otolaryngology): Many of us are old enough to remember what used to happen to patients who had a large anterior mandibular defect. A variety of reconstructive techniques were attempted, among them reconstruction with alloplastic implants, nonvascularized bone grafts, and even pectoralis major myocutaneous flaps with vascularized rib. All these approaches were unreliable, and these disfigured patients typically had serious social problems. In the 1970s, transfer of vascularized free tissue for head and neck reconstruction was introduced. In most medical centers, patients now undergo immediate mandibular reconstruction with vascularized bone and soft tissue for anterior and lateral mandibular defects. We now achieve a 95 percent success rate in patients with defects that were previously considered to be irreparable.
The fibula is the usual source of tissue used for oral mandibular reconstruction. The bone can be readily molded to the contour of the mandible, it accepts implants well, and a defect as long as 25 cm can be reconstructed. Soft tissue that is sufficient to cover the vascularized bone can be obtained from the donor site.
Dr. Dodson: Dr. Clark, will you give us an update on the status of the patient?
Dr. Clark: Two months after the reconstructive surgery, the swelling had greatly diminished, the patient's bite appeared reasonable, and all the wounds had healed. Several of his physicians believed that he should have chemotherapy and delay irradiation, whereas others believed that irradiation should be the next step. After extensive consultation, the patient elected to proceed with radiation therapy.
He has regained weight and strength and has returned to his normal activities. He has a removable partial denture to restore his teeth, and he is considered tumor-free. He has moved out of Massachusetts and is expected to begin chemotherapy soon.
Addendum
Dr. Kaban: I saw this patient in follow-up 13 months after the definitive reconstruction. He had completed his chemotherapy and was free of any evidence of disease.
Source Information
From the Departments of Oral and Maxillofacial Surgery (T.B.D.), Radiology (P.A.C.), and Pathology (G.P.N.), Massachusetts General Hospital; and the Departments of Radiology (P.A.C.) and Pathology (G.P.N.), Harvard Medical School.
References
Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based medicine: how to practice and teach EBM. New York: Churchill Livingstone, 1997:2.
Guyatt G, Rennie D, eds. Users' guides to the medical literature: essentials of evidence-based clinical practice. Chicago: AMA Press, 2002:21-6.
Brockstein BE. Head and neck sarcomas. Wellesley, Mass.: UptoDate, 2002. (Also available at http://www.uptodate.com/.)
Fuchs B, Pritchard DJ. Etiology of osteosarcoma. Clin Orthop 2002;397:40-52.
August M, Magennis P, Dewitt D. Osteogenic sarcomas of the jaws: factors influencing prognosis. Int J Oral Maxillofac Surg 1997;26:198-204.
Patel SG, Meyers P, Huvos AG, et al. Improved outcomes in patients with osteogenic sarcoma of the head and neck. Cancer 2002;95:1495-1503.
Delgado R, Maafs E, Alfeiran A, et al. Osteosarcoma of the jaw. Head Neck 1994;16:246-252.
Gadwal SH, Gannon FH, Fanburg-Smith JC, Becoskie EM, Thompson LDR. Primary osteosarcoma of the head and neck in pediatric patients: a clinicopathologic study of 22 cases with a review of the literature. Cancer 2001;91:598-605.
Kassir RR, Rassekh CH, Kinsella JB, Segas J, Carrau RL, Hokanson JA. Osteosarcoma of the head and neck: meta-analysis of nonrandomized studies. Laryngoscope 1997;107:56-61.
Evidence-Based Medicine Working Group. Evidence-based medicine: a new approach to teaching the practice of medicine. JAMA 1992;268:2420-2425.
Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based medicine: how to practice and teach EBM. New York: Churchill Livingstone, 1997:3.
McKibbon A, Hunt D, Richardson WS, et al. Finding the evidence. In: Guyatt G, Rennie D, eds. Users' guides to the medical literature: essentials of evidence-based clinical practice. Chicago: AMA Press, 2002:21-73.
How to ask questions you can answer. In: Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based medicine: how to practice and teach EBM. New York: Churchill Livingstone, 1997:21-36.
McKibbon A, Hunt D, Richardson WS, et al. Finding the evidence. In: Guyatt G, Rennie D, eds. Users' guides to the medical literature: essentials of evidence-based clinical practice. Chicago: AMA Press, 2002:27.
Focusing clinical questions. Oxford, England: Centre for Evidence-Based Medicine. (Accessed December 18, 2003, at http://www.cebm.net/focus_quest.asp.)
Levels of evidence and grades of recommendation. Oxford, England: Centre for Evidence-Based Medicine. (Accessed December 18, 2003, at http://www.cebm.net/levels_of_evidence.asp.)
Bertoni F, Dallera P, Bacchini P, Marchetti C, Campobassi A. The Istituto Rizzoli-Beretta experience with osteosarcoma of the jaw. Cancer 1991;68:1555-1563.(Thomas B. Dodson, D.M.D.,)