Virtual Library

Start Your Search

B. Stam

Author of

  • +

    O10 - Stereotactic Ablative Body Radiotherapy (ID 104)

    • Event: WCLC 2013
    • Type: Oral Abstract Session
    • Track: Radiation Oncology + Radiotherapy
    • Presentations: 1
    • +

      O10.07 - Dose-response analysis of radiation induced rib fractures after SBRT for NSCLC (ID 2690)

      16:15 - 17:45  |  Author(s): B. Stam

      • Abstract
      • Presentation
      • Slides

      Symptomatic rib fractures occur in approximately 5% of patients treated with SBRT for early stage NSCLC. Only in small patient cohorts has the dose-effect relation of radiation induced rib fractures been determined. Recent developments in automatic rib segmentation allow determining the dose-effect relation in a large patient cohort, which is the aim of this study.

      From 2006-2012 453 patients with early stage NSCLC were treated with SBRT (3x18 Gy). Follow-up (FU) consisted of a physical examination and a CT scan 4 months after treatment and every 6 months up to two years and yearly thereafter. For the first 101 patients with FU>6 months, all ribs were automatically segmented using 15 atlases of manually delineated ribcages. A non-rigid registration followed by a multi-level label fusion produced for each patient a set of ribs. The physical dose distributions were NTD (Normalized Total Dose) corrected with α/β=3 Gy. Cox proportional hazard regression analysis, which takes into account the time to event with patient as random intercept, was used to find the optimal dose parameter. Evaluated were the dose received by x% of the rib D~x~ (x ranged 1-30%) and equivalent uniform dose (EUD) (volume effect 1/n ranged 0.1-60). The Lyman-Kutcher-Burman (LKB) model based on this optimal dose parameter was used to model the dose-effect relationship. Using maximum-likelihood estimation, parameters were median toxic dose (TD~50~), steepness parameter m and 1/n were optimized.

      In 354 patients with FU>6 months (median 22 months), 38 patients(11%) were diagnosed with a total of 49 rib fractures, symptomatic (grade 2) for 9 patients(2.5%). Included in the dosimetric analysis were 2410 ribs (14 ribs outside field-of-view). 26 ribs in 15 patients(15%) were fractured, symptomatic for 4 patients(4%). In the univariate analysis, all dose parameters significantly correlate with rib fracture (p-values<0.001). Hazard ratios (95%CI) for the parameters with highest log likelihood: D~1~=1.022 (1.017-1.027) and EUD~0.033~=1.021 (1.016-1.026). Multivariate analysis identified EUD as the predictor with the highest log-likelihood and was used in the LKB model. The optimal LKB parameters to predict rib fracture in this dataset were (95% CI): TD~50~=395.5 Gy (244.3-555.1), m=0.348 (0.311-0.384) and 1/n=32.3 (4.82-inf). The risk of rib fracture was <5% in case the NTD-corrected EUD<170 Gy.Figure 1

      In this subgroup of NSCLC patients treated with 3x18Gy, the risk of rib fracture was significantly correlated to the dose, and was <5% in case the biological dose is kept under 170 Gy.

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.