Xoft® Intraoperative Radiotherapy (IORT) for Patients With Early-Stage Breast Cancer

Details of Radiation Therapy: Both breast conserving surgery and IORT will be delivered as per standard practice. Lumpectomy and sentinel lymph node biopsy will be performed by the breast surgeon using standard techniques and as per standard of care. IORT with Xoft Axxent System will be given at the time of lumpectomy for early stage breast cancer. During the same operative procedure, a single fraction of radiation will be given to the lumpectomy bed with IORT. The Xoft Axxent treatment device size will be customized to the patient's anatomy based on the volume of lumpectomy cavity. The prescription dose will be 20 Gy prescribed to the surface of the lumpectomy cavity. Postoperative, external beam, whole breast radiation therapy will be given that the discretion of the treating physician based on the presence of high risk clinicopathologic features. Based on data from existing, published, clinical trials, it is estimated that 15% of patients will require postoperative radiation in addition to IORT. As with standard practice, all patients planned for IORT will be discussed pre- and post-surgery at the breast cancer multidisciplinary conference for pathology and radiology review.

Study outline: After patients complete lumpectomy and IORT, their acute and late toxicities and cosmetic outcome will be monitored on protocol. Patients will complete clinical follow up in radiation oncology at 4-6 weeks, 6 months, 1 year, 18 months, 2 years and then yearly for a total of 5 years. In accordance with standard clinical practice, patients will have toxicity assessment, physical exam, and cosmetic scoring by the physician. Physician toxicity assessment will be performed with Common Terminology Criteriae for Adverse Events (CTCAE) version 5.0, Late Effects Normal Tissue Task Force-Subjective, Objective, Management, Analytic (LENT-SOMA), and Radiation Therapy Oncology Group (RTOG) skin and toxicity scales. Physician cosmetic assessment will be performed with the Harvard Cosmetic Scale. Patients will also be asked to complete Breast Cancer Treatment Outcome Scale (BCTOS) self-assessment for patient reported cosmetic, functional and pain symptoms. Breast-Q version 2.0 will also be used to evaluate patient-reported cosmetic outcome, physical well being and adverse events.

A prospective database will be created and maintained. Patient data will be deidentified in the database with patients assigned unique study codes. Their assigned study code will be kept in a single encrypted file by the principal investigator.

This is an observational prospective study. To reduce selection bias, all eligible consecutive patients will be enrolled. The investigators expect the sample size to be around 20 per year with a total of 60 in three years. Multiple adverse radiation outcomes (including breast pain, radiation dermatitis, breast edema, seroma, infection) will be evaluated and compared with historical controls to detect if prevalence differs by 20% and 16% in two- and three-years accrual, respectively (n=40 in two years and n=60 in three years) at alpha=0.05 and power =80%.

Descriptive statistics (mean, median, standard deviation, minimum and maximum for continuous measures, and frequency and percentages for ordinal measures) related to participant characteristics, treatment and risk factors will be produced for the entire cohort and subsets of cohort. Differences in adverse radiation effects in the IORT protocol participants and historic controls will be assessed using the Cochran-Armitage test for trend or chi-squared test, as appropriate. Univariate analysis will be used to detect associations between cosmetic outcome and patient-specific and treatment-related factors by use of either a chi-squared test (categorical variables) or t-test (continuous variables). The freedom from ipsilateral breast tumor recurrence, progression free survival {PFS) and overall survival (OS) will be estimated by the Kaplan-Meier method. Differences in survival between the protocol cohort and historic controls will be evaluated through the estimated hazard rates using the unweighted log-rank tests. PFS and OS will be estimated along with 95% precise confidence intervals. In order to examine the significant factors, Cox proportional hazards regression models will be used in both uni-variable and multi-variable settings. Other factors to be analyzed are ethnicity, gender, age, and pathological subtype. The various factors will be placed into categorical variables. Simple logistic regression will be used to study association between a dichotomous outcome measure and any predictor. Once again, for any dichotomous outcome measure, the effect of other covariates (as listed above) will also be explored in addition to primary predictor using a multiple logistic regression analysis.

All results will be declared significant at significance level of 5%. Since there are multiple outcomes compared, the investigators will adjust significance level for multiple comparisons.

The investigators will maintain a prospectively collected database of patients who undergo treatment, the details of which are kept in their electronic medical record. This EMR is managed in the usual and customary manner throughout treatment and follow-up. An excel spreadsheet of patients' MRN and their assigned study code will be kept in a single encrypted file by the principal investigator. This encrypted file will remain on campus, secured in the principal investigator's office, and held under lock and key.