Main Session
Sep 27
QP 01 - Multidimensional Toxicity in Breast Cancer Radiotherapy: Integrating Clinical, Patient-Reported, and Biological Outcomes

1002 - Validation of NSABP B-39/RTOG 0413 Patient-Reported Measures of Breast Cosmesis

03:15pm - 03:20pm ET
Room 253

Presenter(s)

Mylin Torres, MD, FASTRO - Emory University School of Medicine, Atlanta, GA

M. Torres1, R. S. Cecchini2, P. A. Ganz3, J. R. White4, F. A. Vicini5, B. McCormick6, D. W. Arthur7, R. A. Rabinovitch8, R. R. Kuske9, T. B. Julian10, H. Bear7, I. Germain11, J. T. Dilworth12, K. Butler13, D. Shumway14, P. R. Anne15, M. Scheier16, E. P. Mamounas17, B. Movsas18, and R. Jagsi1; 1Department of Radiation Oncology, Emory University School of Medicine, Winship Cancer Institute, Atlanta, GA, 22NRG Oncology Statistical and Data Management Center and the Department of Biostatistics and Health Data Science School of Public Health University of Pittsburgh,, Pittsburgh, PA, 3Department of Medicine (Hematology/Oncology), David Geffen School of Medicine at UCLA; Department of Health Policy and Management, UCLA Fielding School of Public Health; UCLA Jonsson Comprehensive Cancer Center, Los Angeles, 4University of Kansas Comprehensive Cancer Center, Kansas City, KS, 5Michigan Health Professionals, Farmington Hills, MI, 6Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 7Virginia Commonwealth University and Massey Comprehensive Cancer Center, Richmond, VA, 8University of Colorado, Aurora, CO, 9Retired, Scottsdale, AZ, 10Allegheny Health Network Cancer Institute, Pittsburgh, PA, 11CHU de Québec – Université Laval, Québec, QC, Canada, 12Corewell Health William Beaumont University Hospital, Royal Oak, Royal Oak, MI, 13Cancer Research Consortium of West Michigan NCORP, Grand Rapids, MI, 14Department of Radiation Oncology, Mayo Clinic, Rochester, MN, 15Thomas Jefferson University, Philadelphia, PA, 16Carnegie Mellon University, Pittsburgh, PA, 17AdventHealth Cancer Institute, Orlando, FL, 18Department of Radiation Oncology, Henry Ford Health, Detroit, MI

Purpose/Objective(s):

Previously reported exploratory psychometric evaluation of a 27-item symptom checklist (SCL) in NSABP B-39/RTOG 0413 found 2 subscales aligned with acute breast radiotherapy (RT) adverse effects (AEs) among chemotherapy-naïve (CTN) patients (pts): 5-item RT-related breast Skin Changes (RSC) and 8-item localized Pain and Breast Symptoms (PBS). A single-item pt-reported (PR) global cosmetic score (GCS) was also used. To support use of these measures in future research, we sought to determine reliability of RSC and PBS in a broader pool of pts, including B-39/0413 CT pts, and to validate RSC, PBS, and PR GCS.

Materials/Methods:

B-39/0413 quality of life (QOL) sub-study pts were assessed at baseline, end of treatment (EOT), then 4 wks, 6 mos, 1 yr, 2 yrs, and 3 yrs post-RT. EOT PROs were completed by CTN pts (n=220 whole breast irradiation [WBI]); n=256 accelerated partial-breast irradiation [APBI]), and CT pts (n=199 WBI; pts receiving CT+APBI omitted, as their EOT assessment did not coincide with RT completion). Consistency of SCL subscales was explored by factoring the 13 items from RSC plus PBS in the CT/WBI group and total group (CTN+CT) using unweighted least squares factor analysis with promax rotation. We calculated Cronbach’s a to assess reliability and evaluated responsiveness of RSC and PBS over time. Spearman correlations were used to assess validity of RSC, PBS, and GCS, by correlating scores with validated Breast Cancer Treatment Outcomes Scale (BCTOS) cosmesis and pain subscales.

Results:

Factor analysis resulted in consistent item-factor correlations for all factors in the RSC plus PBS subscales in CT/WBI pts alone (all >0.58) and total group (all >0.57) at EOT, indicating RSC and PBS capture acute RT toxicities regardless of RT volume or CT. Using the total group, Cronbach’s a was RSC 0.90 and PBS 0.88, indicating high internal consistency. Total group mean PBS scores and WBI group mean RSC scores peaked at EOT and decreased over the 3-yr period. Mean RSC for APBI pts peaked 4 wks post EOT. Scales were sensitive to changes in pt condition, as repeated measures analysis over all timepoints resulted in statistically significant changes over time (p<0.0001). RSC correlated with BCTOS cosmesis subscale (r=0.42) at EOT. PBS significantly correlated with BCTOS pain subscale at EOT (r=0.76) and across time (r>0.71 for all subsequent timepoints). EOT RSC mildly correlated with 1-yr (r=0.21) but not 3-yr BCTOS cosmesis. EOT PBS correlated with both 1- and 3-yr BCTOS cosmesis and pain (1 yr: r=0.39 and r=0.50, respectively; 3 yr: r=0.34 and r=0.42, respectively). PR GCS significantly correlated with the BCTOS cosmesis subscale across all time points (r>0.58 acutely and r>0.64 at 1, 2, and 3 yrs post-RT).

Conclusion:

NRG Oncology has identified 2 SCL subscales for measuring acute breast RT AEs, which appear valid. Our results also support the validity of the PR GCS. Future studies may find these measures useful to assess the direct impact of RT.

Grants: U10CA180868, -180822, UG1CA189867