HDR brachytherapy monotherapy results in excellent long-term local control for intermediate risk prostate cancer. Improvements in biochemical and local disease control for patients treated with 38 Gy in 4 fractions warrants consideration of prospective study of optimal treatment dosing regimens.

There is limited data correlating dosimetry with toxicity for intracavitary accelerated partial breast irradiation (APBI). We hypothesized that radiation dose to skin, dose heterogeneity, and PTV volume significantly correlated with physician assessed acute and chronic toxicities. One hundred ninety one patients treated with balloon catheter high dose rate brachytherapy (BCHDRB) were included in this study. Median follow-up was 7 years. All patients were treated with 34Gy in 10 twice daily fractions. Acute (occurring within 6 months of radiotherapy) and chronic (occurring after 6 months of treatment) toxicities, including dermatitis, breast pain, breast infection, breast edema, hyperpigmentation, hypopigmentation, telangiectasia, volume reduction, fat necrosis, induration, chest wall pain, rib fractures, and cosmesis were assessed at each follow-up visit using the CTCAE (v3.0) criteria. Dosimetric parameters analyzed were maximum point dose to the skin, maximum point dose to a rind of skin 2 mm in thickness, dose to 1-5 cc of a rind of skin 2 mm in thickness, PTV volume, PTV V200%, PTV V150%, PTV V125%, and maximum rib dose. Statistical analyses were conducted using ROC curves, Cox regressions, Kaplan Meier curves, and log rank tests. Acute breast pain was associated with larger PTV volume (HR 5.0; P<0.019) and higher PTV V150% (HR 12.9; P=0.001). Acute induration was significantly correlated with skin maximum point dose (HR 3.8; P=0.025). For chronic toxicities, dose to the skin significantly correlated with chronic telangiectasia (HR 3.5), dermatitis (HR 2.8), breast pain (HR 1.9), hypopigmentation (HR 3.7), hyperpigmentation (HR 5.6), induration (HR 5.3), volume reduction (HR 45.2), and cosmesis (HR 1.7) when assessed as either maximum point dose or volumetrically to a rind of skin 2 mm in thickness (P<0.05 for all comparisons). No dosimetric parameters correlated with breast edema, chest wall pain, or rib fractures. Using maximum skin dose of 34Gy as cut off, rates of chronic toxicity below vs. above 34Gy were 7.1% vs. 17% (any telangiectasia), 36% vs. 51.8% (any breast pain), 9.1% vs. 20.2% (any hypopigmentation), 34.4% vs. 44.1% (any hyperpigmentation), and 32.4% vs. 43.8% (G2+ induration) (P<0.05 for all comparisons). The median time to first incident of chronic toxicities ranged from 8 to 40 months, but there were late first occurrences of chronic toxicities with 20% of patients experiencing a first chronic toxicity 30-137 months after completing radiotherapy. Skin dose calculated as either a maximum point dose or volumetrically to a rind of skin 2 mm in thickness significantly correlated with chronic telangiectasia, dermatitis, breast pain, hypopigmentation, hyperpigmentation, induration, volume reduction, and cosmesis. Attempts should be made to limit maximum skin dose to less than 34Gy (100% Rx) to reduce chronic toxicity.

Prior studies have shown improved dosimetry with lower skin doses using multi-lumen (ML) catheter brachytherapy compared to single lumen (SL) devices for accelerated partial breast irradiation (APBI). However, there are no long-term clinical data to correlate the improved dosimetry with less early and late toxicities. We hypothesized that ML APBI had a more favorable side effect profile and dosimetry compared to SL APBI. A total of 191 patients were treated between 6/30/2000 and 2/22/2013 using either SL MammoSite (N=129; single or multi-dwell), ML MammoSite (N=22), or ML Contura applicators (N=40). All patients were treated with 34 Gy prescribed to a depth of 1 cm from the applicator in 10 twice daily fractions. Only patients with more than 3 years of follow-up data were included in this study. The median follow-up was 8 years. Toxicities including dermatitis, hyperpigmentation, hypopigmentation, breast pain, breast edema, telangiectasia, and induration were assessed at each follow-up visit using the CTCAE (v3.0) criteria. Acute toxicity was defined as <6 months from the date of radiotherapy completion, and chronic toxicity was defined as >6 months. Skin spacing and maximum skin dose were measured for each patient. Differences in acute and chronic toxicities were compared using chi-square and log-rank tests. Mean maximum skin doses were compared using a t-test. Patients treated with ML applicators had significantly less acute dermatitis (grade 1-4: ML 42.4%/SL 66.7%, P=0.003; grade 2-4: ML 8.5%/SL 9.4%, P=0.785), chronic dermatitis (grade 1-4: ML 14.5%/SL 30.5%, P=0.018; grade 2-4: ML 0%/SL 3.9%, P=0.135), and chronic telangiectasia (grade 1-4: ML 19.4%/SL 34.6%, P=0.031; grade 2-4: ML 14.6%/SL 17.3%, P=0.681) compared to SL MammoSite. There were no significant differences between ML and SL in the rates of acute or chronic hyperpigmentation, hypopigmentation, breast pain, breast edema, and induration. Using a log-rank test, patients treated with ML applicators had less chronic dermatitis compared to SL MammoSite (P=0.05). The mean maximum skin dose was significantly lower for ML (34.1 Gy) compared to SL (36.3 Gy) when the skin spacing was between 6-11 mm (P=0.038). However, the mean maximum skin dose was not significantly different between ML and SL when skin spacing was <6 mm (P=0.086) or >12 mm (P=0.419). Regardless of skin spacing, the median maximum skin doses were 32.2 Gy for ML and 33.7 Gy for SL. ML applicators resulted in less acute and chronic toxicities compared to SL MammoSite. ML devices also had significantly lower mean maximum skin doses when skin spacing was between 6-11 mm.

To determine if differing irradiated volumes affect the clinical outcomes and cosmesis with brachytherapy interstitial MultiCatheter-Accelerated Partial Breast Irradiation [MC-APBI] vs that of single-entry Balloon-APBI [BAL-APBI] stratified by the ASTRO-Guidelines [ASTRO-G] from a single institution. From 1993-2008, 484 patients with early-stage breast CA were treated with APBI. Between 1993 and 2000, all eligible patients were treated with either I-125 low-dose-rate or Ir-192 high-dose-rate MC-APBI with dose prescribed at 2 to 3 cm beyond the lumpectomy. Between 2000 and 2008, the balloon applicator (BAL-APBI) was increasingly used. Dose prescription for balloon applicators was 1 cm from the balloon surface. Due to the differing dose prescriptions and thus irradiated volumes, we undertook a propensity-score matched-pair analysis to determine the outcomes of MC-APBI vs BAL-APBI stratified by ASTRO-G. Match criteria were age and ASTRO-G by propensity score; exact match of Tis/T1/T2 stage and ER +/-; and minimal follow-up 3 yrs and tx info. Of the original 484 pts, 293 had information, yielding 90 matched pairs (ratio 1:1) totaling 180 patients for comparative analysis of clinical outcomes via Kaplan-Meier, stratified by ASTRO-G. Median age for MC-APBI was 65 yrs (range 42-84) and BAL-APBI 63 (41-86) (p = 0.82). Median follow-up was 9.0 yrs; MC-APBI 14.9 yrs and BAL-APBI 6.2 yrs (p < 0.001). The majority of tumors were T1 (93%) with median size 1.0 cm. ER + was found in 84% and 68% received endocrine Rx, with 69% MC-APBI and 66% BAL-APBI (p = 0.66). Chemotherapy was delivered in 15%, with 8% MC-APBI and 22% BAL-APBI (p = 0.01). The irradiated volumes differed significantly, with median 154 cc (36-455) for MC-APBI vs 95 cc (61-168) for BAL-APBI (p < 0.001). For disease control, MC-APBI had a 5 and 10 yr actuarial local recurrence rate of 1.2% & 3.8% vs BAL-APBI of 4.5% at both 5 & 10 yrs (p = 0.45); true recurrence/marginal miss revealed 0% & 1.2% at 5 & 10 yrs for MC-APBI vs 1.1% at both for BAL-APBI (p = 0.74). No difference was noted for 10-yr actuarial rates of regional recurrence (0% for both, p = ns) & distant metastases (2.2% & 4.7% for MC-APBI at 5 & 10 yrs, 2.2% for BAL-APBI, p = 0.69). The 10-yr actuarial cause specific survival (97% MC-APBI vs 96% BAL-APBI, p = 0.98), disease-free survival (92% vs 93%, p = 0.75) and overall survival (81% vs 76%, p = 0.68) were similar. When stratified by ASTRO-G, no difference in LR, DM, CSS, or DFS was noted between MC-APBI vs BAL-APBI. Cosmetic outcomes were good/excellent in 95% of all patients. In comparing 2 differing APBI treatment techniques, with significantly different irradiated volumes, no difference in clinical outcomes or cosmesis were discernable, even when stratified by ASTRO-G. Further follow-up will be required for validation of long-term outcomes and cosmesis in such APBI patients.

PurposeTo investigate the discrepancies in catheter tip localization and their effects on dosimetry in transrectal ultrasound (US) -guided real-time prostate high-dose-rate (HDR) brachytherapy. The catheter tips in US images are usually defined by adjusting the free length (distance from the distal end of the catheters to the template) so that the majority of the reconstructed tips match the image. However, from our experience, there are many uncertainties in this method, including the operator's judgment (1 ∼ 2 mm), curvature of the needles (< 0.5 mm), free length and template position mismatch (1 ∼ 2 mm), and bent ultrasound probe (1 ∼ 2 mm).Materials and MethodsFourteen patients were identified with discrepancy between treatment planning assumed needle tip positions and their tips as shown on the US images. While the tips of the anterior catheters are clearly outside the edge of the Foley balloon on US images, they were marked inside the Foley by the planning system due to the tip defining method mentioned above (as seen in Fig. 1). The tip discrepancies in the cranial-caudal direction for each catheter were calculated for all 14 patients. New dosimetry was created with corrected catheter reconstruction and the same corresponding dwell positions and time. Prostate V100, V125, V150, and urethra V100, V110, V115 from original plans were compared to those from the adjusted plans.ResultsThe mean tip discrepancy is 2.2±2.0 and 3.9±1.9 mm for all catheters and the anterior catheters, respectively, with a max of 8.5 mm. The prostate V100, V125 and urethra V100 from the original plans are significantly higher than those from the adjusted plans: 98.8%±0.8% vs. 95.9%±1.9% (p < 0.001), 55.6%±2.4% vs. 51.1%±3.3% (p < 0.001), 85.9%±5.5% vs. 77.2%±11.3% (p < 0.05), respectively. Although there are no significant difference for the prostate V150, urethra V110 and V115 (23.5%±2.3% vs. 21.8%±2.7%, 2.6%±4.5% vs. 3.7% ±5.4%, and 0.0%±0.1% vs. 0.3%±0.9%), two patients have post-adjusted urethra V110 > 15% and two have urethra V115 > 1.5%, which are substantially higher than our in-house criteria.Conclusions PurposeTo investigate the discrepancies in catheter tip localization and their effects on dosimetry in transrectal ultrasound (US) -guided real-time prostate high-dose-rate (HDR) brachytherapy. The catheter tips in US images are usually defined by adjusting the free length (distance from the distal end of the catheters to the template) so that the majority of the reconstructed tips match the image. However, from our experience, there are many uncertainties in this method, including the operator's judgment (1 ∼ 2 mm), curvature of the needles (< 0.5 mm), free length and template position mismatch (1 ∼ 2 mm), and bent ultrasound probe (1 ∼ 2 mm). To investigate the discrepancies in catheter tip localization and their effects on dosimetry in transrectal ultrasound (US) -guided real-time prostate high-dose-rate (HDR) brachytherapy. The catheter tips in US images are usually defined by adjusting the free length (distance from the distal end of the catheters to the template) so that the majority of the reconstructed tips match the image. However, from our experience, there are many uncertainties in this method, including the operator's judgment (1 ∼ 2 mm), curvature of the needles (< 0.5 mm), free length and template position mismatch (1 ∼ 2 mm), and bent ultrasound probe (1 ∼ 2 mm). Materials and MethodsFourteen patients were identified with discrepancy between treatment planning assumed needle tip positions and their tips as shown on the US images. While the tips of the anterior catheters are clearly outside the edge of the Foley balloon on US images, they were marked inside the Foley by the planning system due to the tip defining method mentioned above (as seen in Fig. 1). The tip discrepancies in the cranial-caudal direction for each catheter were calculated for all 14 patients. New dosimetry was created with corrected catheter reconstruction and the same corresponding dwell positions and time. Prostate V100, V125, V150, and urethra V100, V110, V115 from original plans were compared to those from the adjusted plans. Fourteen patients were identified with discrepancy between treatment planning assumed needle tip positions and their tips as shown on the US images. While the tips of the anterior catheters are clearly outside the edge of the Foley balloon on US images, they were marked inside the Foley by the planning system due to the tip defining method mentioned above (as seen in Fig. 1). The tip discrepancies in the cranial-caudal direction for each catheter were calculated for all 14 patients. New dosimetry was created with corrected catheter reconstruction and the same corresponding dwell positions and time. Prostate V100, V125, V150, and urethra V100, V110, V115 from original plans were compared to those from the adjusted plans. ResultsThe mean tip discrepancy is 2.2±2.0 and 3.9±1.9 mm for all catheters and the anterior catheters, respectively, with a max of 8.5 mm. The prostate V100, V125 and urethra V100 from the original plans are significantly higher than those from the adjusted plans: 98.8%±0.8% vs. 95.9%±1.9% (p < 0.001), 55.6%±2.4% vs. 51.1%±3.3% (p < 0.001), 85.9%±5.5% vs. 77.2%±11.3% (p < 0.05), respectively. Although there are no significant difference for the prostate V150, urethra V110 and V115 (23.5%±2.3% vs. 21.8%±2.7%, 2.6%±4.5% vs. 3.7% ±5.4%, and 0.0%±0.1% vs. 0.3%±0.9%), two patients have post-adjusted urethra V110 > 15% and two have urethra V115 > 1.5%, which are substantially higher than our in-house criteria. The mean tip discrepancy is 2.2±2.0 and 3.9±1.9 mm for all catheters and the anterior catheters, respectively, with a max of 8.5 mm. The prostate V100, V125 and urethra V100 from the original plans are significantly higher than those from the adjusted plans: 98.8%±0.8% vs. 95.9%±1.9% (p < 0.001), 55.6%±2.4% vs. 51.1%±3.3% (p < 0.001), 85.9%±5.5% vs. 77.2%±11.3% (p < 0.05), respectively. Although there are no significant difference for the prostate V150, urethra V110 and V115 (23.5%±2.3% vs. 21.8%±2.7%, 2.6%±4.5% vs. 3.7% ±5.4%, and 0.0%±0.1% vs. 0.3%±0.9%), two patients have post-adjusted urethra V110 > 15% and two have urethra V115 > 1.5%, which are substantially higher than our in-house criteria. Conclusions

To assess toxicities and clinical outcomes in prostate cancer patients treated with HDR BT using higher Ir-192 source strength with activity > 8.5 Ci versus lower source strength with activity <5.2 Ci. During 2004-2010, 66 patients were treated during either the first week following source exchange (group 1, n = 30) or the last week prior to source exchange (group 2, n = 36). Median follow up was 2.1 years, and all patients were treated with a single implant in two or four fractions. Patients treated with concurrent external beam or hyperthermia and those treated for recurrent cancers were excluded. Patients whose treatment occurred on the day of source exchange and patients whose treatment overlapped a source exchange were excluded. Median activities of Ir-192 were 9.5 Ci (range 8.6-10.6) and 4.4 Ci (range 3.9-5.2) for groups 1 and 2, respectively. The dose-volume constrains to prostate, urethra, and rectum were similar. Acute (6 mo) genitourinary (GU) and gastrointestinal (GI) toxicities were assessed using NCI Common Criteria, v.3. Biochemical failure was defined using the Phoenix definition. Patient characteristics between the two groups were similar, with: Stage T1c 87% and T2a 13% (p = 0.302); Gleason Score GS-6 68% and GS-7 32% (p = 0.809); median pre-tx PSA 4.8 ng/ml (p = 0.991); median pre-tx AUA 4.5 (p = 0.538); median prostate volume 39cc2 (p = 0.991); median age 62 yrs (p = 0.279); hormones used in 11% (p = 0.131); fractionation - 950cGy x4: 33 patients, 1200cGy x2: 18 patients, 1350cGy x1: 15 patients. There was a significant difference in grade 1 acute GU retention with 13% vs 0% (p-value 0.038) in groups 1 and 2, respectively. There was also a consistent trend toward higher GU toxicity in group 1 vs. higher GI toxicity in group 2. Examples of such trend included: grade 1 acute toxicities: Dysuria 10 vs 3% (p = 0.464); Rectal Pain 0 vs 5% (p = 0.196); grade 1 chronic toxicities: Diarrhea 3 vs 8% (p = 0.412), Rectal Bleeding 0 vs 8% (p = 0.111); grade 2 chronic GU frequency 17 vs 6% (p = 0.258). Biochemical control at 3 years was 95.5 in group 1 vs 84.2% in group 2 (p = 0.363). There was no difference in 3 years OS, CSS, DFS, LC, and DMFS (100%). The Ir-192 source activities in group 1 were as much as twice as those in group 2. This study showed a higher grade 1 acute GU retention in group 1 (13% vs 0%, p = 0.038). There was a consistent trend toward higher GU toxicity in group 1 and higher GI toxicity in group 2 but no statistically significant difference in biochemical control in this small co-hort. Additional patients and follow up will further quantify any differences due to higher versus lower activity of Iridium-192 in toxicity and biochemical control after prostate cancer HDR BT.

To compare the acute and chronic toxicity profiles associated with three high-dose-rate interstitial brachytherapy (HDR) monotherapy fractionation schedules: 38 Gy in 4 fractions (38/4), 24 Gy in 2 fractions (24/2), and 27 Gy in 2 fractions (27/2). 482 patients with low-risk (69%) or intermediate-risk (31%) prostate cancer were treated with Ir-192 following ultrasound-guided, percutaneous implantation of afterloading catheters. Gland volumes and pre-treatment AUA scores were similar among groups. 317 patients received 38/4 (interfraction period: 6 h or 18 h); 72 received 24/2 (interfraction period: 6 h or 14 d); 93 received 27/2 (interfraction period: 6 h or 14 d). Chi-square, t-test, ROC, and univariate analyses were conducted to compare patient characteristics, clinical outcomes, and maximal acute (<6 m following treatment) and chronic (>6 m following treatment) toxicities. Patients were treated in an HIC approved sequential dose schedule protocol with the same eligibility criteria. Clinical outcomes were similar among groups, with 3-y actuarial rates of 91% (biochemical failure-free survival), 99% (overall survival), and 100% (cause-specific survival). Acute gastrointestinal (GI) toxicities (diarrhea, bleeding, proctitis, and pain/tenesmus) were similar among groups; no grade 3 or higher toxicities were reported. Patients receiving 38/4 experienced the highest rate of acute grade 1/2 dysuria (27%), urinary retention (22%), and incontinence (6%) (p = 0.04, 0.04, 0.01). Other acute genitourinary (GU) toxicity rates were similar among all patients, including grade 3 urinary frequency/urgency (2%) and grade 3 urinary retention (<1%). Patients receiving 38/4, 25/2, and 27/2 experienced dissimilar rates of chronic grade 1/2 diarrhea (3%, 11%, 7%; p = 0.02) and rectal bleeding (7%, 4%, 0%; p = 0.02). No grade 3 or higher chronic GI toxicity was reported in any group. Chronic GU toxicity was similar among groups, with <1% of all patients experiencing grade 3 dysuria, frequency/urgency, retention, incontinence, or hematuria. Nine patients (2%) developed a urethral stricture requiring surgical dilation. Patients receiving pre-treatment hormone therapy (HT) experienced higher acute GI and GU toxicities (20% and 70%) than patients not receiving HT (10% and 53%)(p = 0.03 and 0.02). Gland volume did not predict acute or chronic toxicity. The toxicity and preliminary clinical outcomes associated with 38/4, 24/2, and 27/2 are comparable. These data suggest that accelerated, hypofractionated HDR constitutes a well tolerated, convenient, and effective treatment modality. Additional follow-up to characterize long-term outcomes is warranted.

Toxicity resulting from accelerated partial breast irradiation (APBI) using MammoSite has been correlated to minimum balloon-to-skin distance, a surrogate for skin dose. In this study, we evaluate maximum planned skin dose as a predictor for toxicity. One hundred ten patients treated with MammoSite APBI were retrospectively analyzed. All patients were treated with HDR Iridium-192 to a total dose of 3,400 cGy, one cm from the balloon surface, in 10 BID fractions (6-hour minimum inter-fraction time). All treatments were delivered using a single, central source dwell position. Treatment planning and dose calculation were performed using the TG-43 dose calculation formalism within the Pinnacle planning system. For each patient, minimum balloon-to-skin distance, maximum skin dose, and skin doses to surface areas of 1, 2, 3, 4, and 5 cm2 were recorded. Toxicities included the acute (<6 months) morbidities of erythema/dry desquamation, moist desquamation, and induration, and the chronic (≥6 months) sequelae of erythema, hyperpigmentation, hypopigmentation, telangiectasia, induration, and fat necrosis using the NCI CTAE v3.0. Skin spacing and maximum skin dose were analyzed as dichotomous variables to determine threshold levels where toxicity became more prevalent. A Chi-square or Fisher's exact test was used, depending on expected frequency, to determine statistical significance. Median follow-up time was 34 months. There was considerable variation in maximum skin dose among patients with a common skin distance (mean standard deviation of 40 cGy). A significant correlation was found between incidence of acute moist desquamation and a maximum skin dose >410 cGy (120% prescription dose, p = 0.014) or a skin distance <8 mm (p = 0.025). Correlations were found between a maximum skin dose >400 cGy and incidence of chronic induration (p = 0.042) and chronic erythema (borderline significant, p = 0.051). Incidence of telangiectasia was found to increase continuously with closer skin distance (p = 0.0006) and maximum skin dose (p = 0.0003) with no threshold found ≤10 mm or >210 cGy. Maximum planned skin dose was found to correlate with incidence of acute moist desquamation, telangiectasia, chronic erythema, and chronic induration resulting from MammoSite APBI. This correlation supports the empirically validated use of minimum balloon-to-skin distance as a predictor of toxicity from MammoSite treatment. We expect that as more sophisticated brachytherapy dose calculation methods evolve, calculated maximum skin dose will improve correspondingly as a predictor for morbidity.