SAN ANTONIO -- For patients with so-called "good-risk" ductal carcinoma in situ (DCIS) who did not have radiation therapy after breast-conserving surgery, adjuvant tamoxifen reduced their overall risks for invasive recurrence, but not their risks for recurrence in either the same or contralateral breast.
These findings come from an exploratory analysis of combined data from two clinical trials. They suggest that for this select group of patients, the choice to forgo radiation following definitive surgery may be an acceptable option, assuming that they follow a full course of endocrine therapy.
"In the absence of survival impact for adjuvant therapy, the decision to recommend radiation therapy or endocrine therapy should be part of a shared decision process, and I think that this data helps us provide clearer data points to our patients to help them make choices between endocrine therapy and radiation therapy in the setting of good-risk [ductal carcinoma in situ]," said Jean L. Wright, MD, from the University of North Carolina at Chapel Hill.
She presented the findings in an oral abstract session and media briefing at San Antonio Breast Cancer Symposium (SABCS) 2024.
Wright and colleagues looked at pooled data from two clinical trials that enrolled patients with low- or intermediate-grade DCIS with tumor size no larger than 2.5 cm, grade 1 or 2 lesions, and with surgical margins ≥ 3 mm.
The trials included NRG/RTOG 9804, with 317 patients who fit the "good-risk category," and ECOG-ACRIN E5194, which included a cohort of 561 patients that met the good-risk definition used for the exploratory analysis.
In each trial, tamoxifen use was optional, and choices were tracked. In the NRG/RTOG trial, 66% of patients used tamoxifen and 34% did not. The respective percentages in the ECOG/ACRIN trial were 30% and 70%.
The majority of patients were adherent to the 5-year prescribed course of tamoxifen, Wright said.
In the combined data, the median age of patients who used tamoxifen vs who did not use tamoxifen was 58 vs 61 years.
In all, 23% of women in both the tamoxifen yes or no groups were premenopausal, with the remainder either postmenopausal or of unknown menopausal status.
After a median follow-up of 14.85 years, the rate of 15-year ipsilateral breast recurrence (IBR) was 19% for patients who did not receive tamoxifen vs 11.4% for those who did. This translated into a hazard ratio for IBR on tamoxifen of 0.52 (P = .001).
Tamoxifen also reduced the risk for invasive recurrence in the same breast, with a 15-year invasive IBR rate of 11.5% in the no tamoxifen group vs 6% in the yes tamoxifen group.
However, as noted before, tamoxifen use was not associated with significant reduction in the risk for noninvasive DCIS recurrence in the same breast, as evidenced by a 15-year DCIS IBR rate of 8.1% without tamoxifen and 5.5% with tamoxifen, a difference that did not reach statistical significance.
One finding from the data that seemed to defy clinical wisdom was that tamoxifen use did not appear to significantly reduce the risk for events in the other breast. The 15-year rate of contralateral breast events was 8.8% in the no-tamoxifen group vs 5.6% in the yes tamoxifen group, a difference that was not statically significant.
"It was surprising that there was so little effect on contralateral disease," commented Elinor Sawyer, MBBS, PhD, the invited discussant.
"But I think it's really important, this decrease in ipsilateral invasive recurrence [with tamoxifen] because there are studies such as the Sloane study from the UK that show that if you develop an invasive recurrence after DCIS, you have a worse survival than those who develop a pure DCIS recurrence," she said.
At the media briefing held prior to Wright's presentation, moderator Virginia Kaklamani, MD, leader of the breast cancer program at UT Health San Antonio, also said that she found it surprising that tamoxifen did not reduce risk for contralateral breast cancer "since every study that we've done has shown that."
"I also found that result a little bit surprising," Wright agreed.
"I think the main feature that we want to focus on is that this was a group of patients with a very clear inclusion criteria of this good-risk DCIS, and even though the definition of good-risk DCIS included patients with [tumors] up to 2.5 cm in DCIS, we saw that, in reality, the patients enrolled had very small DCIS. So I'm wondering if it's perhaps that it's related to the fact that patients that were enrolled in these studies had really low-risk features and perhaps just had a lower risk of contralateral breast events as compared to a broader population of patients with DCIS," she said.
The analysis by Wright and colleagues was supported by grants from the National Cancer Institute. Wright reported receiving honoraria from ASTRO and PER. Sawyer disclosed receiving grants/research support from Pfizer, Seagen, and IQIVIA. Kaklamani disclosed serving as a speaker and/or consultant for AstraZeneca, Celldex Therapeutics, Daiichi Sankyo, Genentech, Gilead, Lilly, Menarini, and Novartis and receiving research support from Eisai.