PORTAL IMAGING PROTOCOL FOR RADICAL DOSE-ESCALATED RADIOTHERAPY TREATMENT OF PROSTATE-CANCER

Purpose: The use of escalated radiation doses to improve local control in conformal radiotherapy of prostatic cancer is becoming the focus o f many centers. There are, however, increased side effects associated with increased radiotherapy doses that are believed to be dependent on the volume of normal tissue irradiated. For this reason, accurate pat ient positioning, CT planning with 3D reconstruction of volumes of int erest, clear definition of treatment margins and verification of treat ment fields are necessary components of the quality control for these procedures. In this study electronic portal images are used to (a) eva luate the magnitude and effect of the setup errors encountered in pati ent positioning techniques, and (b) verify the multileaf collimator (M LC) held patterns for each of the treatment fields. Methods and Materi als: The Phase I volume, with a planning target volume (PTV) composed of the gross tumour volume (GTV) plus a 1.5 cm margin is treated confo rmally with a three-field plan (usually an anterior field and tao late ral or oblique fields). A Phase II, with no margin around the GTV, is treated using two lateral and four oblique fields. Portal images are a cquired and compared to digitally reconstructed radiographs (DRR) and/ or simulator films during Phase I to assess the systematic (CT plannin g or simulator to treatment error) and the daily random errors. The ma tch results from these images are used to correct for the systematic e rrors, if necessary, and to monitor the time trends and effectiveness of patient immobilization systems used during the Phase I treatment co urse. For the Phase II, portal images of an anterior and lateral field (larger than the treatment fields) matched to DRRs (or simulator imag es) are used to verify the isocenter position 1 week before start of P hase II. The Portal images are acquired for all the treatment fields o n the first day to verify the MLC field patterns and archived for reco rds. The final distribution of the setup errors was used to calculate modified dose-volume histograms (DVHs). This procedure was carried out on 36 prostate cancer patients, 12 with vacuum-molded (VacFix) bags f or immobilization and 24 with no immobilization. Results: The systemat ic errors can be visualized and corrected for before the doses are inc reased above the conventional levels. The requirement for correction o f these errors (e.g., 2.5 mm AP shift) was demonstrated, using DVHs, i n the observed 10% increase in rectal volume receiving at least 60 Gy. The random (daily) errors observed showed the need for patient fixati on devices when treating with reduced margins. The percentage of field s with displacements of less than or equal to 5.0 mm increased from 82 to 96% with the use of VacFix bags. The rotation of the pelvis is als o minimized when the bags are used, with over 95% of the fields with r otations of less than or equal to 2.0 degrees compared to 85% without. Currently, a combination of VacFix and thermoplastic casts is being i nvestigated. Conclusion: The systematic errors can easily be identifie d and corrected for in the early stages of the Phase I treatment cours e. The time trends observed during the course of Phase I in conjunctio n with the isocenter verification at the start of Phase II give good p rediction of the accuracy of the setup during Phase II, where visibili ty of identifiable structures is reduced in the small fields. The acqu isition and inspection of the portal images for the small Phase II fie lds has been found to be an effective way of keeping a record of the M LC field patterns used. Incorporation of the distribution of the setup errors into the planning system also gives a clearer picture of how t he prescribed dose was delivered. This information can be useful in do se-escalation studies in determining the relationship between the loca l control or morbidity rates and prescribed dose. (C) 1998 Elsevier Sc ience Inc.