| Part? Epidemiological Survey and Analysis of 308 Patients withOsteosarcomaObjective:To investigate the epidemiological and clinical features of patients with osteosarcoma in Zhejiang Province.Methods:All patients from Zhejiang Province diagnosed with osteosarcoma in the second affiliated hospital of Zhejiang University during 2010.3-2016.3 were entered into a retrospective analysis of demographic,tumor and treatment-related variables.Results:The study consisted of 308 patients with osteosarcoma.Their mean age was 20.47±13.93 years old,people of 10-30 and 40-55 years old accounted for 71.74%and 14.61%.There were 180 male patients and 128 female patients,male to female ratio was 1.41:1.Tumor occurred more often around knee joint,the most common location were distal femur(47.73%)and proximal tibia(19.48%).About 68.91%patients reported tumor-related pain as the first symptom.The average duration of symptoms prior to diagnosis was 3.72±9.71 months,the longest time was 10 years.The most common subtype of osteosarcoma was osteoblast bone sarcoma(40.02%).Distant metastases were found in 12.34%patients by using preoperative imaging examinations.Limb preserved rate was 84.74%.3-year and 5-year survival rate were 74.79%and 63.64%,respectively.The incidence of osteosarcoma was estimated to be 0.79/million/year.Conclusion:This study offers demographic,tumor and treatment-related data of 308 patients with osteosarcoma of Zhejiang Province.It may help to learn about the epidemiological and clinical features of patients with osteosarcoma in Zhejiang Province.Part? Multivariate Analysis for Prognostic Factors of OsteosarcomaObjective:To determine the prognostic factors that influence survival of patients with non-metastatic highly malignant osteosarcoma of the extremity.Methods:The clinical data of 119 incident patients with primary osteosarcoma in the second affiliated hospital of Zhejiang University were retrospectively reviewed between 2009.3 and 2014.3,which included clinical,hematological,imaging,pathological,treatment-related variables and so on.Kaplan-Meier method was used to calculate the 3-year event-free survival rate,and the univariate analysis was used to determine prognostic factors related with event-free survival rate by Log-rank test.And Cox model multivariate analysis was used to identify independent prognostic factors.P value<0.05 was considered statistically significant.Results:Event-free survival rate of 3 years was significantly related to tumor site,mass size,biopsy,lactate dehydrogenase level after preoperative chemotherapy,tumor volume before and after preoperative chemotherapy,tumor volume change and tumor necrosis(Log-rank,P<0.05),and tumor volume change was the only independent prognostic factor(Cox model,P<0.05).Conclusion:Tumor volume change played an important role in prognosis for survival in osteosarcoma patients.It shows the potential by using imaging methods to evaluate tumor response to chemotherapy and prognosis in patients with osteosarcoma.Part? Development and Validation of a Preoperative Risk-Scoring Model in Osteosarcoma PatientsObjective:In order to screen the high-risk osteosarcoma patients with poor-prognosis,this study aimed to predict poor prognosis preoperatively by using available clinical data of incident osteosarcoma patients,and to establish a risk scoring model.Methods:The preoperative clinical data of 92 incident patients diagnosed with osteosarcoma in the second affiliated hospital of Zhejiang University were retrospectively reviewed between 2008.3 and 2013.3,which included clinical,hematological,imaging,pathological variables and so on.Death,local recurrence,and distant metastasis were observed as the primary outcome.All patients were randomly divided into two groups,one was training group used to develop the prediction model and the other was validation group used to examine the predictive validity.Logistic regression model was used to establish model on the basis of the clinical data in training group.The discrimination and calibration of prediction model was tested by using the area under the receiver operator characteristic curve according to the clinical data in validation group,described in terms of sensitivity and specificity.The risk-scoring model was set up on the basis of the coefficient and rank of variables in the prediction model.Results:The study consisted of 92 incident patients with osteosarcoma and 35 patients were poor-prognosis in 3years.Tumor site and tumor volume change were included in risk-predictive model at last by using Logistic regression model.The prediction model=1.367×tumor site(=0 or 1)+1.791 ×tumor volume change(=0 or 1)-1.386.The area under the receiver operator characteristic curve of prediction model in training group was 0.790(95%CI(0.661-0.918)),while in validation group was 0.753(95%CI(0.516-0.990)),the sensitivity was 83.9%and the specificity was 66.7%.Risk-scoring model was set up as:tumor located in extremities was 0 point,tumor located in other bones was 1 point,tumor volume decreased was 0 point and tumor volume had no significant change or increased was 2 points.It was divided into four groups according to the final risk scores of 0 point,1 point,2 points and 3 points,respectively.Conclusion:We developed an accurate and practical risk-scoring model to predict poor-prognosis preoperatively by using available clinical data of incident osteosarcoma patients.It may help to screen the high risk osteosarcoma patients with poor response to chemotherapy and poor prognosis and make clinical decisions. |
PDF Full Text Request