Evaluation of Radiation Dosimetric Parameters and its Association with Acute Skin Toxicity in Whole Breast Radiation

Purpose: This study was designed to evaluate the radiation dosimetric parameters and its association with acute skin toxicity following whole breast radiation using conformal radiotherapy in early breast cancer. Methods and materials: Computed tomography scan images of 50 early breast tumour patients who had undergone breast conservation surgery were selected for this study. Contouring of skin was done as a structure with 5mm thickness from the surface of the body to the anterior margin of the breast planning target volume (PTV). Constraints to the skin were given, without compromising the dose to the PTV (Planning Target Volume). Dose delivered was 50 Gy/25# to the whole breast with tumour bed boost of 12 Gy/6#. Assessment of reactions of the skin was done for each patient weekly by using the Common Terminology Criteria for Adverse Events (CTCAE), version 4.03. All patients in this study were followed up weekly by clinical examination during RT and post RT every month until 3 months. During follow- up the worst grade of skin toxicity was recorded. Correlation between dosimetric parameters of skin and PTV with a grade of skin toxicity was done using the Spearman method. Results: Mean dose received by the skin was 36.3±7.6 Gy. Maximum dose received by the skin was 65.6±2.92. There was no significant correlation between the skin constraints and PTV parameters given with the highest grade of skin toxicity. During RT, 54 % had grade 1 skin reaction, 32 % had grade 2 skin reaction and 14 % had grade 3 skin reaction. Grade 2 and 3 skin toxicities peaked at 1 or 2 weeks after RT despite stratification of RT techniques (3DCRT/IMRT/RapidArc) Conclusion: Taking skin as an organ at risk, and limiting the dose to the skin, by contouring the irradiated part of the skin without disturbing the PTV coverage, showed no significant association with skin toxicity but the rate of grade 3 and 4 skin toxicities were reduced. Thus that there is no statistical significance in g