Impact of delay of local control in nonmetastatic extremity primary osteosarcoma.

Abstract

The timing of local control (LC) is not well studied in osteosarcoma. We assessed the impact of the delay of LC on the survival outcome of patients with osteosarcoma. We conducted a retrospective analysis of children (≤18years) with nonmetastatic extremity primary osteosarcoma at King Hussein Cancer Center from January 2005 until March 2020. Patients' demographics, disease characteristics, and outcomes were collected. Events were defined as death, progression, or relapse. Cox proportional hazards regression was used for univariable and multivariable comparisons of different covariates. Eighty-two patients were included; 41 (50%) were females; the median age was 12.5years (range: 5.9-18). Sixty-four patients (78%) underwent LC by limb-salvage surgery. Fifteen patients (18%) had a delay of LC greater than or equal to 18weeks. After a median follow-up of 54months (range: 8-188), the 5-year event-free survival (EFS) and overall survival (OS) were 55.3% ± 5.8% and 66.6% ± 6%, respectively. On univariable analysis, LC greater than or equal to 18weeks, progression before LC, amputation, and poor histologic response were associated with worse EFS (p=.007, .007, .006, .002) and OS (p=.01, .001, .006, .004). On multivariable analysis, LC greater than or equal to 18weeks, progression before LC, and poor histologic response were associated with worse EFS (p<.001, .007, .002); and OS (p<.001, .007, .008). Male gender was associated with worse OS on univariable and multivariable analysis (both p=.02). After exclusion of patients with early progression before Week 11, good histologic response and LC less than 18weeks were associated with better EFS (p=.04 and .03). While good histologic response was associated with better OS (p=.02), LC less than 18weeks was not significant (p=.2). Our findings suggest that delay in LC could have an impact on the outcomes in patients with osteosarcoma. However, further investigations involving larger sample sizes and longer follow-up are necessary to fully comprehend the extent of this influence.