Search

Radiation therapy (RT) of thoracic cancers may cause severe radiation dermatitis (RD), which impacts on the quality of a patient's life. Aim of this study was to analyze the incidence of acute RD and develop normal tissue complication probability (NTCP) models for severe RD in thoracic cancer patients treated with Intensity-Modulated RT (IMRT) or Passive Scattering Proton Therapy (PSPT). We analyzed 166 Non-Small-Cell Lung Cancer (NSCLC) patients prospectively treated at a single institution with IMRT (103 patients) or PSPT (63 patients). All patients were treated to a prescribed dose of 60 to 74 Gy in conventional daily fractionation with concurrent chemotherapy. RD was scored according to CTCAE v3 scoring system. For each patient, the epidermis structure (skin) was automatically defined by an in house developed segmentation algorithm. The absolute dose-surface histogram (DSH) of the skin were extracted and normalized using the Body Surface Area (BSA) index as scaling factor. Patient and treatment-related characteristics were analyzed. The Lyman-Kutcher-Burman (LKB) NTCP model recast for DSH and the multivariable logistic model were adopted. Models were internally validated by Leave-One-Out method. Model performance was evaluated by the area under the receiver operator characteristic curve, and calibration plot parameters. Fifteen of 166 (9%) patients developed severe dermatitis (grade 3). RT technique did not impact RD incidence. Total gross tumor volume (GTV) size was the only non dosimetric variable significantly correlated with severe RD (p = 0.027). Multivariable logistic modeling resulted in a single variable model including S20Gy, the relative skin surface receiving more than 20 Gy (OR = 31.4). The cut off for S20Gy was 1.1% of the BSA. LKB model parameters were TD50 = 9.5 Gy, m = 0.24, n = 0.62. Both NTCP models showed comparably high prediction and calibration performances. Despite skin toxicity has long been considered a potential limiting factor in the clinical use of PSPT, no significant differences in RD incidence was found between RT modalities. Once externally validated, the availability of NTCP models for prediction of severe RD may advance treatment planning optimization.

In radiation oncology, the need for a modern Normal Tissue Complication Probability (NTCP) philosophy to include voxel-based evidence on organ radio-sensitivity (RS) has been acknowledged. Here a new formalism (Probabilistic Atlas for Complication Estimation, PACE) to predict radiation-induced morbidity (RIM) is presented. The adopted strategy basically consists in keeping the structure of a classical, phenomenological NTCP model, such as the Lyman-Kutcher-Burman (LKB), and replacing the dose distribution with a collection of RIM odds, including also significant non-dosimetric covariates, as input of the model framework. The theory was first demonstrated in silico on synthetic dose maps, classified according to synthetic outcomes. PACE was then applied to a clinical dataset of thoracic cancer patients classified for lung fibrosis. LKB models were trained for comparison. Overall, the obtained learning curves showed that the PACE model outperformed the LKB and predicted synthetic outcomes with an accuracy >0.8. On the real patients, PACE performance, evaluated by both discrimination and calibration, was significantly higher than LKB. This trend was confirmed by cross-validation. Furthermore, the capability to infer the spatial pattern of underlying RS map for the analyzed RIM was successfully demonstrated, thus paving the way to new perspectives of NTCP models as learning tools.