Beam intensity modulation for penumbra enhancement in the treatment of lung cancer

Purpose: A treatment planning study was performed for patients with lung ca ncer in order to investigate the extent to which doses to critical structur es could be reduced by penumbra enhancement at the superior and inferior fi eld edges, using beam intensity modulation (BIM) with a multileaf collimato r. By applying two independent published models for the prediction of the i ncidence of normal tissue complications, the potential for dose escalation without increasing the incidence of pneumonitis was estimated. Methods and Materials: For 12 patients, the standard treatment technique wa s compared with the BIM technique using the Cadplan 3D planning system (Var ian-Dosetek). Dose distributions in the healthy lung tissue were evaluated by considering both lungs minus the tumor as one functional unit, The follo wing parameters were compared: (i) the average normalized total dose (NTD), (ii) the lung volume receiving an NTD of more than 20 Gy, and (iii) the ca lculated normal tissue complication probability (NTCP), Results: Due to the applied BIM technique, the field lengths could be reduc ed by 1.4 cm for all patients, while achieving a minimum dose at the superi or and inferior parts of the target of 95% of the isocenter dose. Compared to the standard technique, BIR;I reduced the patient mean of the average NT D for the healthy lung tissue from 16.5 to 15.3 GS, The volume of healthy l ung tissue receiving an NTD of 20 Gy or more was reduced by 9.7% (range 2.2 to 23.1%). The calculated NTCP reduced from 10.7% to 7.6% on average. The length of the esophagus that received a dose of 60 Gy or more could be redu ced for 5 of the 6 stage III patients in this study, Based on equal lung NT CPs for the standard technique and the BIM technique, a mean dose escalatio n of 5.7 Gy (range 1.1 to 16.0 Gy) was possible for the 12 patients in this study. Based on equal average NTDs for the two techniques, the patient mea n of the allowed dose escalation was 6.5 Gy (range 1.1 to 18.2 Gy), All dos e escalations would be possible without exceeding the spinal cord tolerance dose. Conclusions: The BIM technique reduced the dose delivery to critical tissue s. Two published methods for estimating the incidence of pneumonitis both p ainted to a potential for dose escalation of 6 to 7 Gy on average with the BIM technique, without increasing the incidence of pneumonitis. For 2 of th e 12 patients in this study the estimated allowed dose escalation even exce eded 15 Gy, (C) 1999 Elsevier Science Inc.