Improvement in dose escalation using the process of adaptive radiotherapy combined with three-dimensional conformal or intensity-modulated beams for prostate cancer

Purpose: Advances in technology allow the creation of complex treatment pla ns with tightly conforming doses. However, variations in positioning of the organ/patient with respect to treatment beams necessitate the use of an ap preciable margin, potentially limiting dose escalation in many patients. To (1) reduce this margin and (2) test the hypothesis that the achievable lev el of dose escalation is patient dependent, a patient-specific, confidence- limited planning target volume (cl-PTV) was constructed using an adaptive r adiotherapy (ART) process for prostate cancer treatment developed in-house. The potential dose escalation achievable with this ART process is quantifi ed for both conformal radiotherapy (CRT) delivery and intensity-modulated r adiotherapy (IMRT) delivery. Material and Methods: Patients with organ confined prostate cancer were ent ered prospectively into an ART process eve ope in-house. This ART process h as been designed to improve accuracy and precision of dose delivery, conseq uently enhancing dose escalation. In this process, a cl-PTV is constructed for each patient in the second week of treatment based upon on-line portal and CT images acquired during the first week of treatment. The treatment pr escription dose, defined as the minimum dose to the cl-PTV, is selected bas ed on predefined dose-volume constraints for rectum/bladder and derived fro m the pretreatment planning CT image. In addition, the treatment modality ( CRT or IMRT) is determined based on the level of dose escalation achievable and the risk of inaccurate targeting. The potential for both dose escalati on and the application of IMRT was evaluated by comparing the prescription doses delivered using the ART process, with the cl-PTV, to those in the tra ditional treatment process, with a conventional generic PTV. In addition, t he distributions of potential geometric target underdosing and normal tissu e overdosing were also calculated to evaluate the quality of the convention al treatment plans. Results: One hundred and fifty patients have been treated with the ART proc ess. When compared to the treatment dose delivered with the conventional tr eatment process (generic PTV), an average 5% (2.5-10%) more dose could be d elivered using the ART process with CRT, and 7.5% (2.5-15%) more dose could be delivered with IMRT. Of the 150 patients, 70% were treated to a minimum cl-PTV dose greater than or equal to 77.4 Gy (81.3 Gy ICRU isocenter dose) . Dosimetric analysis revealed that 81 Gy to the cl-PTV (or 86.7 Gy ICRU) c ould be prescribed to at least 50% of patients if IMRT was applied using th e ART process. In contrast, IMRT did not yield an obvious dose escalation g ain if patients were treated using the generic PTV. Our results also demons trate that the cl-PTV is significantly smaller than the conventional generi c PTV for most patients, with a mean volume reduction of 24% (range, 5-43%) . Conclusion: These results support our hypothesis that the achievable level of dose escalation using ART is patient dependent. By using the ART process to develop a cl-PTV, one can (1) optimize the dose level, (2) increase the applicability of IMRT, and (3) improve the quality of dose delivery. The A RT process provides the foundation to identify a suitable option (CRT or IM RT) for the delivery of a safe treatment and dose escalation. It is now our standard of practice for prostate cancer treatment. (C) 2001 Elsevier Scie nce Inc.