CTA: You found that 3 years after treatment, approximately 14% of C-509T-positive women experienced post-radiation breast fibrosis, compared to 4% C-509T-negative women. What else did you find?
Dr Smith: We’d previously concluded from the study that a shorter, hypofractionated 4-week course of whole-breast radiation is not inferior to a 6-week regimen. [In this analysis, we] also found that 3-year fibrosis was predicted by this allele and also by [a] poor post-operative cosmetic outcome. I had not anticipated that.
CTA: Why do you think poor healing was associated with fibrosis?
Dr Smith: If your breast does not heal well after surgery, that could indicate that you won’t heal well from radiation either. It might also be the case that a breast that healed poorly from surgery does not respond well to radiotherapy. One hit from surgery and a second hit from radiotherapy, yielding more scarring.
In an exploratory analysis, we also found a trend that patients with this variant allele were reporting more shoulder stiffness 3 years after radiation. That’s pretty interesting because we usually tell our patients that with whole-breast radiation alone, we don’t anticipate much shoulder stiffness. But there was this small group of patients who did experience it, and that was associated with this allele. I posit that they have more scarring, which can occur in the pectoralis muscle.
CTA: What variables did your team include in the multivariate analysis, to correct for potentially confounding factors?
Dr Smith: We included everything we could easily measure as candidate covariables: age, race, surgery type, chemotherapy history, tumor stage, tumor ER/PR/HER-2 status. But only 2 variables were significant in the multivariate analysis: adverse post-operative cosmetic outcome and the C-509T allele. We forced study arm [hypofractionated or conventionally-fractionated radiotherapy] into the model but it was not significant and did not interact with the allele.