A fully automatic framework for evaluating cosmetic results of breast conserving therapy.

The breast cosmetic outcome after breast conserving therapy is essential for evaluating breast treatment and determining patient's remedy selection. This prompts the need of objective and efficient methods for breast cosmesis evaluations. However, current evaluation methods rely on ratings from a small group of physicians or semi-automated pipelines, making the processes time-consuming and their results inconsistent. To solve the problem, in this study, we proposed: 1. a fully-automatic Machine Learning Breast Cosmetic evaluation algorithm leveraging the state-of-the-art Deep Learning algorithms for breast detection and contour annotation, 2. a novel set of Breast Cosmesis features, 3. a new Breast Cosmetic dataset consisting 3k+ images from three clinical trials with human annotations on both breast components and their cosmesis scores. We show our fully-automatic framework can achieve comparable performance to state-of-the-art without the need of human inputs, leading to a more objective, low-cost and scalable solution for breast cosmetic evaluation in breast cancer treatment.

Clinical Effectiveness of an Adaptive Treatment Planning Algorithm for Intensity Modulated Radiation Therapy Versus 3D Conformal Radiation Therapy for Node-Positive Breast Cancer Patients Undergoing Regional Nodal Irradiation/Postmastectomy Radiation Therapy.

PURPOSE: Clinical trials support adjuvant regional nodal irradiation (RNI) after breast-conserving surgery or mastectomy for patients with lymph node-positive breast cancer. Advanced treatment planning techniques (eg, intensity modulated radiation therapy [IMRT]) can reduce dose to organs at risk (OARs) in this situation. However, uncertainty persists about when IMRT is clinically indicated (vs 3-dimensional conformal radiation therapy [3DCRT]) for RNI. We hypothesized that an adaptive treatment planning algorithm (TPA) for IMRT adoption would allow OAR constraints for RNI to be met when 3DCRT could not without significantly changing toxicity and locoregional recurrence (LRR) patterns. METHODS AND MATERIALS: Since 2013, all RNI patients also underwent an adaptive TPA that began with 3DCRT and then changed to IMRT when OAR constraints (mean heart dose ≤500 cGy; ipsilateral lung V20 ≤35%) could not be met. Patients received 2 Gy/d to the prospectively contoured target volumes (including internal mammary nodes). We retrospectively evaluated the dosimetry and clinical outcomes of the treatment groups (IMRT vs 3DCRT). The primary endpoint was the cumulative incidence of LRR as the site of first recurrence, and we specifically address patterns of failure based on dose to the posterior supraclavicular nodal region (SCL-post). RESULTS: Two hundred forty patients (60% stage III; mean 4.0 + nodes) underwent an adaptive-TPA for RNI after mastectomy (74%) or breast-conserving surgery (26%), resulting in 168 patients treated with 3DCRT and 72 patients treated with IMRT. There were 7 LRRs (2 IMRT, 5 3DCRT) resulting in 4-year LRR of 2.8% for IMRT versus 1.8% for 3DCRT (P = .99). Three patients (2 IMRT, 1 3DCRT) had SCL nodal failures (1 in the SCL-post). CONCLUSIONS: An adaptive TPA for use of IMRT when 3DCRT does not meet critical OAR constraints resulted in rare high-grade toxicity and no difference in failure patterns between patients treated with IMRT and 3DCRT. These data should provide reassurance that IMRT maintains the therapeutic ratio by preserving cancer control outcomes without excess toxicity when 3DCRT fails to meet OAR constraints.

Race Does Not Affect Tumor Control, Adverse Effects, or Quality of Life after Proton Therapy.

PURPOSE: To compare 5-year biochemical control, toxicity, and patient-reported quality of life (QOL) outcomes for African American and White patients treated with proton therapy (PT) for prostate cancer. MATERIALS AND METHODS: We reviewed the medical records of 1,066 men with clinically localized prostate cancer. Patients were treated with definitive PT between 2006 and 2010. Patients received a median radiation dose of 78 Gy (RBE) with conventional fractionation (1.8- 2 Gy [RBE] per fraction). Sixty-eight (6.4%) men self-identified as African American and 998 (93.6%) self-identified as White. Five-year rates of biochemical control, grade 3 genitourinary and gastrointestinal toxicity, and patient-reported QOL are reported and compared between African American and White patients. RESULTS: Median biochemical follow-up was 5.0 years for both African American and White patients. Median follow-up for toxicity was 5.0 and 5.2 years, respectively. On multivariate analysis, race was not a significant predictor for 5-year freedom from biochemical failure (HR 0.8, p=0.55). No significant association was found between race and grade 3 genitourinary toxicity on multivariate analysis at 5 years (HR 2.5, p=0.10). Patient-reported QOL using median EPIC bowel, urinary incontinence, and irritative summaries scores were not significantly different between the groups. African Americans had higher median sexual summary scores at 2 years than White patients (75 vs. 54, p=0.01) but by 5+ years, the sexual summary scores were no longer significantly different (63 vs. 53, p=0.35). CONCLUSION: With a median follow-up of 5 years, there were no racial disparities in biochemical control, grade 3 toxicity, or patient-reported QOL after PT for prostate cancer.

Five-Year Biochemical Results, Toxicity, and Patient-Reported Quality of Life After Delivery of Dose-Escalated Image Guided Proton Therapy for Prostate Cancer.

PURPOSE: To report clinical outcomes in patients treated with image guided proton therapy (PT) for localized prostate cancer. METHODS AND MATERIALS: The medical records of 1327 men were reviewed. Each man was enrolled on an outcomes tracking study. Dual enrollment on a prospective clinical trial was allowed. Each patient was treated for localized prostate cancer with PT at our institution between 2006 and 2010. Ninety-eight percent of patients received 78 Gy (radiobiological equivalent [RBE]) or higher; 18% received androgen deprivation therapy (ADT). The 5-year freedom from biochemical progression (FFBP), distant metastasis-free survival, and cause-specific survival rates are reported for each risk group. Data on patient-reported quality of life and high-grade toxicities were prospectively collected and reported. A multivariate analysis was performed to identify clinical predictors of biochemical failure and urologic toxicity. RESULTS: The median follow-up time was 5.5 years. The 5-year FFBP rates were 99%, 94%, and 74% in low-risk, intermediate-risk, and high-risk patients, respectively. The actuarial 5-year rates of late grade 3+ Common Terminology Criteria for Adverse Events, version 4.0, gastrointestinal (GI) and genitourinary (GU) toxicity were 0.6% and 2.9%, respectively. Multivariate analysis showed a significant correlation between grade 3+ GU toxicity and pretreatment prostate reductive procedures (P<.0001), prostate volume (P=.0085), pretreatment α-blockers (P=.0067), diabetes (P=.0195), and dose-volume histogram parameters (P=.0208). The median International Prostate Symptom Scores pretreatment scores and scores at 5 years after treatment were 7 and 7, respectively. The mean Expanded Prostate Cancer Index Composite (EPIC) scores significantly declined for sexual summary for patients not receiving ADT (from 67 to 53) between baseline and 5 years. CONCLUSIONS: Image guided PT provided excellent biochemical control rates for patients with localized prostate cancer. The actuarial rates of high-grade toxicity were low after PT. From pretreatment to 5 years of follow-up, a significant decline was found only in mean EPIC sexual summary scores. Prospective clinical studies are needed to determine the comparative effectiveness of PT and other radiation treatment strategies.

Patient-Reported Quality of Life in Men with Transurethral Resection of the Prostate Undergoing Proton Therapy for Management of Prostate Cancer.

PURPOSE: We report on quality of life (QOL) and early toxicity among men with prostate cancer who underwent transurethral resection of the prostate (TURP) before proton therapy. MATERIALS AND METHODS: Between 2006 and 2010, 1,289 patients were treated definitively with proton therapy for prostate cancer at our institution and enrolled on a prospective outcomes-tracking protocol. Ninety-six of the men had received a TURP before proton therapy, while 1,193 men had not. Baseline comorbidities, medications, expanded prostate index composite (EPIC) score, international prostate symptom score (IPSS), and CTCAE vs.3 toxicity assessment were collected prospectively. The Kaplan-Meier product limit method was used to estimate freedom from toxicity. RESULTS: Men who had TURP before proton therapy had lower baseline EPIC scores for urinary incontinence, bowel summary, and sexual summary compared with the non-TURP group, but no significant difference in urinary obstructive score was observed. After controlling for baseline scores, there was no significant difference in bowel summary or sexual summary scores between the two groups over time. There were, however, differences for urinary irritation/obstruction scores and urinary incontinence scores favoring those patients who did not have a TURP-like procedure. Toxicity assessment showed that the 2-year and 5-year rates of grade 3 genitourinary toxicity in the pretreatment TURP group were 12.3% and 17.2%, respectively. CONCLUSIONS: Pretreatment TURP was associated with both a high incidence of physician-assessed toxicity and inferior patient-reported QOL scores both before and after proton therapy treatment. Studies investigating QOL and toxicity after specific prostate cancer therapies should stratify patients by pretreatment TURP. Longer follow-up is needed to confirm if these differences ever resolve.

Angiosarcoma after breast-conserving therapy: long-term disease control and late effects with hyperfractionated accelerated re-irradiation (HART).

BACKGROUND: Secondary angiosarcoma is a malignant cancer that develops in approximately 1% of patients treated with breast-conserving therapy (BCT) for primary breast cancer. Most treatments for secondary angiosarcoma have been unsuccessful and no consensus has been reached on what is the best therapeutic strategy. We report long-term outcomes of patients with secondary angiosarcoma treated with hyperfractionated and accelerated re-irradiation (HART). MATERIAL AND METHODS: We retrospectively reviewed the medical records of, and established direct contact with, 14 consecutive patients with secondary angiosarcoma after BCT with axillary lymph node dissection who were treated at our institution with HART with or without surgery from November 1997 to March 2006. With HART, patients received three radiation therapy treatments each day, with a minimum interfraction interval of four hours, five days a week, at 1 Gy per fraction, to total doses of 45 Gy, 60 Gy, and 75 Gy for areas with a moderate risk for subclinical disease, a high risk for subclinical disease, and gross disease, respectively. The minimum follow-up for these patients was six years. RESULTS: Median survival was 7.0 years (range 0.4-14.7 years), with five- and 10-year overall survival rates of 79% [95% confidence interval (CI), 51-93%] and 63% (95% CI 37-84%), respectively, and five- and 10-year cause-specific survival rates of 79% (95% CI 51-93%) and 71% (95% CI 44-89%), respectively. Toxicity was minimal. CONCLUSION: Our long-term study provides evidence that patients with secondary angiosarcoma after BCT can frequently be cured. Patients treated with HART have higher overall survival, progression-free survival, and cause-specific survival rates than patients who receive only surgery, conventional radiation therapy, or chemotherapy. HART is well tolerated.