Is intraoperative nomogram-based overplanning of prostate implants necessary?
PURPOSE: Several investigators have described intraoperative planning of prostate implants based on a nomogram. The aim of this work was to investigate the adequacy of the nomogram in predicting the total activity necessary for optimal dosimetry. METHODS AND MATERIALS: Eighty CT-based postimplant treatment plans were performed for patients who underwent ultrasound guided I-125 permanent implants alone between April 2000 and March 2001. The cohort of 40 patients had early stage (T1-T2) prostatic carcinoma and pre-treatment prostate volumes of 19-50 cc. I-125 seeds (0.391 mCi/seed) were implanted to achieve a distribution of 75% of the activity peripherally and 25% centrally. The CT studies were obtained on the day of (CT1) and at 1 month (CT2) after implant. All patients were catheterized at CT1, and 28 patients were catheterized at CT2 to visualize the urethra. For each patient, the percentage difference (dA) between the total implanted and nomogram predicted activity for a known prostate volume was calculated. The V200 (volume receiving 200% of the prescribed dose), V150, V100, V90, D100 (maximum dose received by 100% of the volume), D90, and D80 were measured for the prostate at CT1 and CT2. For the urethra, V275, V250, V200, and V150 were evaluated, and V100 and V70 were evaluated for the rectum. The Pearson test was used to correlate the dosimetric parameters with dA. Linear regression was used to fit the correlation of the volume and dose parameters with dA. RESULTS: The median V100 at CT1 and CT2 was 91.8% and 94.2%, respectively. The Pearson test was significant for the prostate V100 and dA measured at CT1 (p = 0.005) but not at CT2 (p = 0.106). A similar correlation was found for the prostate D90 at CT1 (p = 0.002), but not at CT2 (p = 0.076). D100 (maximum dose received by 100% of volume) for prostate did not correlate with dA at CT1 (p = 0.094) and CT2 (p = 0.148). The volume of the prostate receiving higher doses (greater than 150% and 200% of the prescribed dose) correlated with dA. There were no significant correlations between V275, V250, V200, and V150 at CT1 and CT2 as a function of dA for the urethra. V100 and V70 for the rectum correlated significantly with dA; for V100, p = 0.041 at CT1 and p = 0.014 at CT2 and for V70, p = 0.041 at CT1 and p = 0.026 at CT2. A linear regression model fitted to the prostate data obtained from CT1 with the goal of achieving a V100 of 90% and D90 of 145 Gy suggests that no increase in the number of seeds may be warranted using intraoperative planning. The implants examined showed no concomitant increase of urethral doses with increase in activity relative to the nomogram, but showed an increase in the rectal doses for the same increase in activity. CONCLUSION: The doses evaluated at CT1 represent an underestimate, whereas those obtained at CT2 represent an overestimate of the actual delivered protracted permanent implant dose. Based on these results and consideration of the dynamic nature of the dose distribution, target coverage obtained with intraoperative planning using the nomogram predicted activity is consistent with published guidelines for a quality implant and critical structure doses are within tolerance.