Outcome of advanced, unresectable conventional central chondrosarcoma


  • See related editorial on pages 3103–4, this issue.



For patients who have chondrosarcoma with unresectable disease, because of tumor location, tumor size, or extensive metastatic disease, treatment options are very limited because of their relative resistance to radiotherapy and chemotherapy. The overall survival of this patient population is poor; however, specific studies are lacking, and large series have not been published. Therefore, the authors conducted this retrospective, 2-center study to gain insight into the outcome of patients with advanced, unresectable, conventional central chondrosarcoma.


All patients with unresectable conventional central chondrosarcoma who were diagnosed between January 1, 1980 and December 31, 2011 in 2 major European bone sarcoma centers (Rizzoli Institute, Bologna, Italy and Leiden University Medical Center, Leiden, the Netherlands) were selected. Relevant information was collected from the medical records at both centers.


In total, 171 patients met the selection criteria. The overall survival rate for all patients was 48% at 1 year, 24% at 2 years, 12% at 3 years, 6% at 4 years, and 2% at 5 years. Patients with unresectable, locally advanced disease without distant metastases had a significantly better survival than patients with metastatic disease (P = .0014). Systemic treatment, consisting of either doxorubicin-based chemotherapy or the noncytotoxic drugs imatinib and sirolimus, improved survival significantly compared with no treatment (P = .0487). For patients who had locally advanced disease without metastases, radiotherapy was associated with a survival benefit (P = .0032).


This study provides a standard for overall survival rates after a diagnosis of unresectable conventional central chondrosarcoma. Systemic treatment and radiotherapy may improve survival, although selection bias because of the retrospective nature of this study may have influenced the outcome. The poor survival underlines the need for new therapeutic options for this patient population. Cancer 2014;120:3159–3164. © 2014 American Cancer Society.


Chondrosarcomas are a heterogeneous, diverse group of tumors that share at least the characteristic of chondroid matrix production. After osteosarcoma, it is the second most common primary bone tumor in adults and accounts for 20% of new primary bone cancer cases. Conventional central chondrosarcoma is the most common subtype, and the majority is of low histologic grade, which, to date, is the best prognostic indicator.[1-3] Low-grade chondrosarcoma tumors rarely metastasize, grow slowly, and have a favorable prognosis after surgery; therefore, in the new World Health Organization classification of 2010, grade I chondrosarcomas are reclassified as atypical cartilaginous tumors (available at: www.who.int; accessed December 2013). Approximately 10% of conventional central chondrosarcomas are histologically high grade (grade II or III) with a high risk for distant metastases and/or local recurrence and, thus, a poor prognosis after resection alone.

In the last decades, there has been no significant improvement in the survival of patients with chondrosarcoma. The only treatment option with curative intent is surgical resection.[4] However, for some locations in the body, such as the pelvis or the skull, resection with a wide margin is difficult to achieve; therefore, local recurrence and metastatic disease are more common. Currently, it is not believed that the available systemic therapy options improve outcome, although randomized studies and large series have not been published. Therefore, we conducted this study to gain insight into the outcome and the effectiveness of systemic treatment for patients with advanced, unresectable conventional central chondrosarcoma.


For the current study, we selected all patients who had unresectable central chondrosarcoma diagnosed between January 1, 1980 and December 31, 2011 from 2 major European bone sarcoma centers (Rizzoli Institute, Bologna, Italy and Leiden University Medical Center, Leiden, the Netherlands) in January 2012. Reasons for unresectable disease were: complete resection of primary tumor and/or metastases was not deemed feasible due to of technical inability because of the size or location of the primary tumor or extensive metastatic sites, or because complete resection would lead to unacceptable morbidity as judged by the multidisciplinary teams at the referral sites in Bologna and Leiden. Relevant information was collected retrospectively from the archives at both centers. Baseline information consisted of sex; age at first presentation of disease; date of first disease presentation; disease location, type, grade, and TNM stage of chondrosarcoma at first diagnosis and at recurrence; treatment received; date of development of local recurrence or metastatic disease; pattern of metastases, overall survival (OS) from disease onset, and postrelapse survival.

OS was calculated from the day of nonresectability until death or the last follow-up examination. The day that the decision was made not to perform surgery in the multidisciplinary sarcoma board meeting was recorded as the day of nonresectability. The survival curves were calculated according to the Kaplan-Meier method and were compared using the log-rank test.


In total, 171 patients were diagnosed with unresectable central conventional chondrosarcoma in the given timeframe at the 2 centers, including 49 patients at first presentation and 122 patients after 1 or more relapses. The patient characteristics are listed in Table 1. Of the 171 unresectable patients, 72 (42%) had unresectable metastatic disease in the lungs only, 45 (26%) had only local unresectable disease, 39 (23%) had local and unresectable lung disease, and 15 (9%) had multiorgan involvement. The median OS was 11 months (range, 1-106 months) (Fig. 1).

Table 1. Characteristics of 171 Patients With Unresectable Central Conventional Chondrosarcoma
CharacteristicNo. of Patients
  1. a

    Other primary sites were foot, neck, sternum, and hand.

Age at diagnosis: Mean (range), y53 (17-90)
Primary site 
Limb proximal51
Limb distal19
Grade at diagnosis 
Outcome of primary surgery 
Refused surgery10
Unresectable disease localization 
Local and lung39
Figure 1.

Median overall survival for all included patients was calculated from the time point of nonresectability to death or last patient contact.

The OS rate for all patients was 48% at 1 year, 24% at 2 years, 12% at 3 years, 6% at 4 years, and 2% at 5 years. The OS also was analyzed according to disease location. For patients with only local unresectable disease, the OS rate was 26% after 3 years compared with 7% for patients with only lung involvement and 8% for patients with lung and local disease involvement. Patients with multiorgan involvement had an OS rate of 0% after 2 years and a mean survival of 7 months (Fig. 2). The difference between OS for patients with only local unresectable disease and patients with metastatic disease was statistically significant with a P value of .0014.

Figure 2.

Median overall survival was calculated from the time point of nonresectability to death or last patient contact for patients subdivided into the groups that had local disease involvement only, local and lung disease involvement, lung disease involvement only, and multiple sites of disease involvement.

OS after 3 years was the same for patients who received only cytotoxic drugs and patients who received systemic treatment, probably because of the small group of patients who received with these drugs. It was not possible to calculate progression-free survival (PFS), because patients often were not followed by scans as this is a palliative setting and, if scans were made, then they were not done with standard time intervals.

Next, we investigated which variables may influence survival after unresectable disease. Both age <40 years (P = .001) and grade II tumors (P = .022) were correlated with a better OS, but no correlation with OS was observed for sex, site, or resectable versus nonresectable disease at primary diagnosis (data not shown).

Of 171 patients, 37 received systemic treatment: most patients received doxorubicin-based chemotherapy, but the nonchemotherapy-based agents imatinib and sirolimus also were adopted. This group had an OS rate of 26% after 3 years compared with 8% for patients who did not receive systemic treatment (P ≤ .05) (Fig. 3), with a median OS of 20 months for those who received systemic treatment versus 15 months for those who received no treatment.

Figure 3.

For the included patients, median overall survival was correlated with systemic treatment. Patients were subdivided in those who received systemic treatment versus those who received no treatment.

Patients who had only metastatic disease, excluding patients with local disease or local disease plus metastases, had a better OS if they received any systemic or radiotherapy treatment compared with patients who received no treatment (P = .0082) (Fig. 4). The characteristics of these patient subpopulations are provided in Table 2.

Figure 4.

A subpopulation of patients who had only metastatic disease was defined, excluding the patients who had local disease or local and metastatic disease. Patients were subdivided into those who received treatment (systemic and/or radiotherapy) and those who received no treatment.

Table 2. Comparison of Characteristics in the Subpopulation of Patients With Metastatic Only Disease Between Those Who Received Treatment (Systemic and/or Radiotherapy) Versus Those Who Received No Treatmenta
 Metastatic Only Subpopulation: No. of Patients
CharacteristicSystemic TreatmentNo Systemic Treatment
  1. a

    Patients were compared for sex, age, primary disease site, histologic grade at diagnosis, and outcome of first surgery.

  2. b

    Other sites were sternum, foot, hand, and calcaneus.

Men:women15 :2449:25
Age at diagnosis: Mean (range), y49 (27-68)49 (17-90)
Primary site  
Limb proximal1225
Limb distal115
Grade at diagnosis  
Outcome of primary surgery  
Refused surgery00

Thirty-three patients received radiotherapy mainly with palliative intent. Patients received radiotherapy according to the standard protocol in the centers. A broad range of dose fractionation schedules was used ranging from 66 grays (Gy) in 2-Gy fractions in an attempt to maximize local control to 24 Gy in 8-Gy fractions when the intent was palliative in patients with a poor prognosis. These patients had an OS rate of 27% after 3 years versus 8% for those who did not receive radiotherapy. After separating the 33 patients who did receive radiotherapy into groups according to disease location, a survival benefit for patients with only local disease was demonstrated with a P value of .0032 (Fig. 5). The characteristics of these patient subpopulations are listed in Table 3.

Figure 5.

A subpopulation of patients who had only locally advanced, unresectable disease was defined. The median overall survival for these patients was better after radiotherapy versus no radiotherapy.

Table 3. Comparison of Characteristics in the Subpopulation of Patients With Locally Advanced Only Disease Between Those Who Received Radiotherapy Versus Those Who Received No Radiotherapya
 Locally Advanced Only Subpopulation: No. of Patients
CharacteristicRTNo RT
  1. Abbreviations: RT, radiotherapy.

  2. a

    Patients were compared for sex, age, primary disease site, histologic grade at diagnosis, and outcome of first surgery.

  3. b

    Other sites were in the neck.

Age at diagnosis: Mean (range), y49 (26-81)59 (26-89)
Primary site  
Limb proximal15
Limb distal02
Grade at diagnosis  
Outcome of primary surgery  
Refused surgery36


Conventional central chondrosarcoma is a primary bone tumor that, if the grade is low, can be treated using surgery with curative intent. However, patients who present with or develop unresectable disease have a poor prognosis. In the current study, the outcome of all patients who were diagnosed with unresectable conventional central chondrosarcoma at 2 major bone sarcoma centers was evaluated to explore OS in this condition and to set the standard for future studies.

The OS for all patients was poor, with a 3-year OS rate of 12%. Patients who had only local unresectable disease had a better prognosis compared with patients who had metastatic disease, with an OS of 26% after 3 years. Comparing these figures with the published OS data for resectable chondrosarcoma, the difference in our patient population was striking: the 5-year OS rate was 99% for patients who had grade I tumors and 77% for those who had grade III tumors.[5] The results from our study demonstrate the urgent need for new therapeutic options is in this patient population. Currently, trials with targeted single agents or combinations are being conducted to address this unmet medical need. A phase 2 study in patients with unresectable or metastatic conventional chondrosarcoma who received saridegib had to be terminated, because the interim analyses demonstrated no treatment benefit compared with placebo. A phase 2 study testing the hedgehog inhibitor GDC-0449 in patients with advanced chondrosarcoma demonstrated some activity and was well tolerated.[6] A phase 2 study investigating pazopanib for patients with unresectable or metastatic conventional chondrosarcoma (clinicaltrials.gov identifier NCT01330966) and a phase 2 study of imatinib in patients with advanced conventional chondrosarcoma (clinicaltrials.gov identifier NCT00928525) are still recruiting.

Because of the dismal survival of patients with unresectable disease, it is important to determine prognostic factors that can identify the subpopulation of patients with chondrosarcoma who are at high risk of developing unresectable disease. Several studies already have been conducted to search for prognostic factors.[1, 7, 8] To date, no molecular marker has been identified that is independent from and superior to histologic grading in combination with clear margins in predicting outcome.

In contrast to what is generally stated in the literature, we observed that, for patients with metastatic disease, there was a survival benefit from receiving chemotherapy: various regimes of alkylating drugs were being used. Because of heterogeneity of the chemotherapy treatments used, it was not possible to do a subanalysis on the type of treatment. We did compare the patients who received cytotoxic drugs with those who received noncytotoxic drugs, but the difference was not statistically significant (P = .436). This outcome may not be representative because of the small number of patients and the wide time range in which the patients were treated. The general result of improved survival after chemotherapy is rather unexpected, because the common opinion is that patients with chondrosarcoma do not benefit from nonsurgical treatment. However, some preclinical studies have already produced promising data that contradict this assumption.[9] Moreover, patients with only locally advanced unresectable disease have a survival benefit from radiotherapy, although this benefit disappears when patients have metastatic disease. We also were interested in progression-free survival (PFS) after radiotherapy. However, no routine scans were made after radiotherapy, most likely because of the palliative intent of radiotherapy in many patients. This hampers correct interpretation of the time to progression and, thus, the PFS endpoint in this retrospective study. However, the results from these nonsurgical treatments could be affected by selection bias, which may have played a role in this retrospective study of 2 bone tumor referral centers.

The patients who had only locally advanced disease eventually died because of local problems caused by increasing tumor load, local tumor pressure on important structures, and side effects of the tumor, like increasing need of pain medication (such as morphine) and anemia.

Italiano et al very recently published an article in which they retrospectively describe survival data from patients with advanced chondrosarcoma.[10] One hundred eighteen patients with advanced chondrosarcoma were analyzed. Their median PFS was 4.7 months, and their median OS was 18 months. Performance status, the number of metastatic sites, and palliative surgery all were associated with OS. That population was heterogeneous because of the different subtypes of chondrosarcoma included and, thereby, the different expected outcomes. The authors also mention this because they observed a significant difference in the objective response rate for different histologic subtypes. This is why we decided to study outcomes in a larger cohort of a homogenous population of patients with central conventional chondrosarcoma.

In addition, our study has some limitations. Our retrospective study, despite using data from 2 experienced bone cancer centers in Europe, may have encountered some difference in the definition of unresectability between the centers that may have had limited impact on the outcome of this study. Differences in risk-benefit ratio may be perceived differently between sites, surgeons, and patients. Also, because of the retrospective nature of this study, not all possible prognostic items could be reliably retrieved from the records, such as performance status, which had been related to OS in earlier studies.[10]

In conclusion, our study adds significantly to the limited data available on OS for both locally and metastatic, unresectable conventional central chondrosarcoma. Our data demonstrate that chemotherapy in patients with unresectable chondrosarcoma may increase survival, but further studies are warranted. Radiotherapy provides a survival advantage and is common practice for locally advanced conventional central chondrosarcomas in both reference centers.


The research leading to these results received funding from the European Union Seventh Framework Programme (FP7 of 2007-2013) under grant agreement 278742 (Eurosarc).


The authors made no disclosures.