Curettage of Biopsy-diagnosed Grade 1 Periacetabular Chondrosarcoma

Abstract

Intralesional excision of intraosseous grade 1 periacetabular chondrosarcoma may seem an attractive alternative to the current recommendation of joint-sacrificing, en bloc resection. We report eight patients who initially underwent this treatment to identify if local tumor control can be achieved consistently. All patients had a percutaneous biopsy that was interpreted as grade 1 chondrosarcoma. The final histology after curettage indicated Grade 1 chondrosarcoma in five patients, Grade 2 in two, and dedifferentiated chondrosarcoma in one. Three of the five patients with true grade 1 chondrosarcoma have been free of disease at a median followup of 108 months. The other two patients with grade 1 chondrosarcoma on the curettage specimen recurred as higher-grade tumors. Three patients died of high-grade chondrosarcoma at a median of 23 months (range 17-72 months). Overall, the majority of patients in this cohort had either a misdiagnosis or recurrence of higher-grade disease. Based on these observations, in the absence of a predictable method to identify the true intraosseous grade 1 chondrosarcomas of the pelvis, curettage must be undertaken with the understanding that a percutaneous biopsy may misrepresent the histologic grade and that curettage may result in a higher risk of recurrence.

Level of Evidence: Level II, prognostic study. See the Guidelines for Authors for a complete description of levels of evidence.

Periacetabular pelvic chondrosarcoma is difficult to treat. Wide surgical excision remains the only proven means of cure at this time, as there is no effective adjuvant therapy. Because the anatomic constraints of the pelvis are unique, resection is associated with a high complication rate.^1,4,11,12 Local recurrence is reported more frequently in the pelvis than in the appendicular skeleton.^5,11-13 Moreover, reconstruction of the pelvis and hip after peri-acetabular resection is inconsistently effective^6,7 and has a high morbidity. Functional outcomes are not as good after pelvic reconstruction as they are elsewhere in the long bones.^3,6

Less aggressive surgery in the form of curettage may seem advantageous for small, contained, intraosseous lesions. This is a less morbid procedure than en bloc periacetabular resection. It is not certain, however, whether curettage can predictably control the tumor. Previous studies have demonstrated wide surgical excision and negative margins correlate with better local tumor recurrence-free and disease-specific survival.^9,12,13 Authors emphasize the first surgical procedure represents the best chance for eradication of the disease.^8,11,13 Inadequate margins have been associated with an increased risk of local recurrence, which has been demonstrated to have a negative impact on survival.^4,5,13 Thus, generally an aggressive approach to pelvic chondrosarcoma is recommended.

The efficacy of curettage has been demonstrated for treating low-grade, contained chondrosarcoma in the long bones of the extremities.² There are no known biologic differences between chondrosarcoma occurring in the bones of the pelvis and tumors originating at other bony sites. A similar management strategy for periacetabular chondrosarcoma is appealing, especially in light of the reconstructive challenges.

We asked whether curettage can provide predictable local tumor control as a definitive treatment for biopsy-diagnosed Grade 1 intraosseous periacetabular chondrosarcoma.

MATERIALS AND METHODS

From a prospective database we retrospectively identified eight patients who underwent curettage of contained, intraosseous periacetabular chondrosarcoma between 1980 and 1999. The study was carried out after approval by the Institutional Review Board and was conducted in accordance with its standards.

Six patients were treated at our institution by curettage. Two patients were referred after undergoing curettage at another institution: one was treated by an orthopaedic oncologist and the other by a general orthopaedic surgeon. Medical records were reviewed for patient age, gender, type of biopsy, postbiopsy treatment, presence of local recurrence or metastasis, timing of disease relapse, and patient status at last followup evaluation (Table 1). Pathology reports were reviewed for the tissue diagnosis on biopsy, after curettage, and subsequently after en bloc resection, if performed. Radiographic studies were analyzed to determine the tumor epicenter, local extent of disease, and disease staging and for postoperative surveillance for tumor recurrence.

TABLE 1: Characteristics and Disease Course for Patients Treated With Curettage of Periacetabular Chondrosarcoma

All patients had intraosseous tumors without soft tissue extension as determined by cross-sectioning imaging (CT and/or MRI) and plain radiographs. Two patients had a pathologic fracture. Each patient had undergone a percutaneous biopsy before the curettage. The histologic diagnosis of the biopsy-obtained tissue of the two referred patients was verified by our pathologists to be Grade 1 chondrosarcoma.

Outcome at last followup was assessed by clinical and radiographic evaluation for each patient. Surveillance studies included local imaging with conventional radiography (anteroposterior projection of the pelvis), and cross-sectional imaging (CT and/or MRI) and distant imaging with plain radiographs of the chest (posterior-anterior and lateral projections) every 3 to 6 months as determined by the grade of the chondrosarcoma. The local recurrence-free and disease-specific survival were estimated at 5 years from the date of the index surgical procedure with the Kaplan-Meier method and SPSS® Software for Windows (SPSS Inc, Chicago, IL).

RESULTS

Five of the six patients diagnosed and treated at our institution had histologic confirmation of the biopsy-diagnosis of Grade 1 chondrosarcoma on analysis of the curetted specimen. Three of these five had no subsequent disease relapse at a median followup duration of 108 months (median 33-142 months). The other two patients developed local tumor recurrences (Grade 2 and Grade 3) and underwent en bloc resection (14 months) and hemipelvectomy (48 months) for definitive control of local disease. One patient remains free of disease at 107 months after the resection. The patient with Grade 3 chondrosarcoma developed uncontrollable local disease and died 18 months after the index surgery.

One patient diagnosed and treated at our institution did not have histologic concordance of biopsy and curettage diagnoses. A Grade 2 chondrosarcoma was revealed upon analysis of the entire specimen. Observation was recommended. One year later a local tumor recurrence was detected and en bloc resection was performed. Histologic evaluation revealed a high-grade spindle cell component in the specimen. The diagnosis of dedifferentiated chondrosarcoma was rendered. The patient had a deep wound infection that prompted hemipelvectomy 2 months later. Recurrent tumor was identified in the specimen. One month later pulmonary metastases were detected and the patient died 17 months later.

The histologic grade of the chondrosarcoma for both of the patients referred for evaluation after curettage of a biopsy-diagnosed Grade 1 chondrosarcoma was different on the curettage specimen. One patient had a Grade 2 and the other dedifferentiated chondrosarcoma on review of the curettage specimens by our pathologists. En bloc resection for the former patient and hemipelvectomy for the latter patient were done. The final pathologic analysis of the Grade 2 curettage specimen revealed Grade 3 chon-drosarcoma. Local tumor recurrence and pulmonary metastases resulted in patient death 23 months after surgery. The patient with dedifferentiated chondrosarcoma had had a total hip arthroplasty at the time of curettage. He underwent hemipelvectomy followed by induction chemotherapy. Surveillance scans have revealed no evidence of local or distant disease at 22 months after amputation.

The 5-year estimates for local recurrence-free and disease-specific survival were 44% ± 22% and 83% ± 15%, respectively.

DISCUSSION

Intralesional excision of biopsy-diagnosed Grade 1 chondrosarcoma resulted in inadequate initial treatment for the majority of our patients. Incorrect diagnosis and inadequate excision specifically, and the lack of effective adjuvant therapy generally, contributed to the poor outcomes.

An analysis of this small cohort of patients is limited by the sample size and the lack of statistical analyses to make a definitive statement regarding the role of curettage in the treatment of intraosseous Grade 1 periacetabular chondrosarcoma. Bias is introduced by including the two patients referred for evaluation and treatment after curettage, as no knowledge of successfully treated patients by curettage has been identified. As such, our recommendations must be interpreted in the context of these limitations.

Many of the clinical problems in our series stemmed from incorrect characterization of the primary lesions. The interpretation of biopsy obtained tissue for cartilaginous neoplasms is fraught with errors and has a likelihood of underestimating the true grade of the lesion. Irrespective of the biopsy technique, even when performed by experienced, knowledgeable experts, the biopsy diagnosis of cartilage tumors can be misleading and unreliable. Selecting the area of the lesion to biopsy can be difficult, and the heterogeneous nature of cartilage tumors can compromise representative sampling. Histologic data must be interpreted in concert with radiographic findings. In this cohort, all tumors were deemed to exhibit nonaggressive radiographic features despite three cases of chondrosarcoma with histologic characteristics of intermediate or high grade tumors Based on the results for the patients presented here, the pelvic location of the primary tumor may itself be a feature that trumps radiographic features and histologic data with respect to surgical decision-making.

Even under favorable circumstances-for example, when biopsy results are correct and treatment is appropriately aggressive-pelvic chondrosarcoma has been reported to recur locally with higher frequency than in other sites and with a propensity to upgrade histologically with local recurrence.¹³ The change in histology reflects a more aggressive behavior and is associated with a worse prognosis. Histologic grade and completeness of the tumor resection (wide surgical margins) are important prognostic factors for systemic and local disease outcomes in large published series of pelvic chondrosarcomas^4,5,9,12,13 Moreover, higher mortality rates have been reported for patients with local recurrence of pelvic chondrosarcoma.¹³

In the only other series of patients that addressed the question of the adequacy of intralesional excision of chondrosarcoma of the pelvis, Ozaki et al¹⁰ analyzed 18 patients and noted a higher local recurrence rate than with en bloc excision. Long-term local-recurrence-free survival was substantially diminished over 20 years when intralesional procedures were compared to those in which a wide or marginal margin was obtained in their series that included tumors of the extremities. Overall survival was unchanged when the two types of procedures were compared; however, tumors located centrally had only 57% overall survival compared to 87% overall survival for tumors located in the extremities. The authors of that study concluded intralesional management of chondrosarcoma was inadequate and highlighted the need for long-term surveillance, given the risk of late local recurrences.

In our series, curettage did provide adequate treatment for the majority, but not all patients who had a true Grade 1 chondrosarcoma. There is a cohort of patients for whom curettage represents an appropriate treatment option. As with three of our patients, such a management approach would be expected to control the tumor locally and preserve hip and limb function without a high risk of metastasis. Other clinical variables, such as advanced age and poor medical status, may influence the decision to proceed with this approach, despite the potential attendant risks.

Presently, predicting biological behavior based on radiographic and histologic criteria can be misleading. Contained intraosseous lesions that are truly Grade 1 and lack aggressive biologic potential cannot be identified reliably at this time. Therefore, the results of curettage of grade 1 pelvic chondrosarcoma remains unpredictable. Wide en bloc resection remains favored as the primary treatment for this disease.

Acknowledgments

The authors thank Maribel D. Martinez and Leah A. Dobbins for their assistance with the preparation of this manuscript.