Accelerated PBI Not Equal to WBI in Ipsilateral Breast Tumor Recurrence

Frank Vicini, MD

Accelerated partial breast irradiation (PBI) following lumpectomy was marginally not found to be equivalent to whole breast irradiation (WBI) to control ipsilateral breast tumor recurrence (IBTR), according to findings from the phase III NSABP B-39/RTOG 0413 trial presented during the 2018 San Antonio Breast Cancer Symposium (SABCS).

Additionally, 10-year estimates for recurrence-free interval (RFI) showed that PBI was statistically significantly inferior to WBI at a 1.6% difference. However, since the differences in both IBTR and RFI are relatively small, lead study author Frank Vicini, MD, said that accelerated PBI may still be an alternative option for patients undergoing breast-conserving surgery.

“Because the differences relative to both IBTR and RFI were very small, PBI may be an acceptable alternative to WBI for a proportion of women who are undergoing breast-conserving surgery,” said Vicini, principal investigator at the MHP Radiation Oncology Institute/21st Century Oncology in Pontiac, Michigan, in a press conference during the meeting. “Even though we weren’t able to demonstrate equivalence, it’s nice to see that even in this large population of patients, an extended follow-up, the differences are quite small.”

Standard breast-conserving surgery consists of lumpectomy plus 5 to 7 weeks of WBI, Vicini explained. Accelerated PBI targets and treats the tumor bed area rather than the entire breast, as is done with WBI. PBI also reduces radiation treatment time from the 3 to 6 weeks needed with WBI to just 5 to 8 days. This study is currently the largest clinical trial evaluating PBI.

“In an effort to improve quality of life of our patients, we studied whether or not we could reduce the overall treatment time significantly in these patients down to 1 week or less by doing a technique referred to as accelerated partial breast irradiation by limiting the radiation only to the lumpectomy cavity region and accelerating the treatment to 5 days or less,” Vicini explained.

In 2016, the American Society for Radiation Oncology issued an update to their guideline for APBI for patients with early-stage breast cancer, which included evidence that a greater number of patients can benefit from the accelerated therapy, and also provided direction for the use of intraoperative radiation therapy for PBI.

In the randomized, phase III NSABP B-39/RTOG 0413 trial, 4216 patients were randomized to receive either 5 to 6 weeks of WBI after adjuvant chemotherapy at 50 Gy (2.0 Gy/fraction) or 50.4 Gy (1.8 Gy/fraction) to the whole breast, followed by an optional boost to ≥60 Gy or receive PBI prior to adjuvant chemotherapy for 10 treatments given on 5 days over 5 to 10 days at 34 Gy in 3.4 Gy fractions via interstitial brachytherapy or 38.5 Gy in 3.85 Gy fractions via 3D conformal external beam radiotherapy.

Patients were stratified by disease stage: ductal carcinoma in situ (DCIS), invasive N0, and invasive N1 disease; menopausal status; hormone receptor (HR) status (estrogen receptor [ER]— and/or progesterone receptor [PR]–positive or ER- or PR-negative); and intent to receive chemotherapy.

The primary endpoint was IBTR, in both invasive and DCIS subgroups, as a first recurrence. Secondary endpoints included distant disease-free interval (DDFI), RFI—which was defined as the time from randomization to the development of first event either local, regional, or distant recurrence, regardless of any intervening contralateral or second breast cancer—and overall survival (OS). In order to declare accelerated PBI and WBI equivalent in terms of IBTR risk per a protocol-defined margin, the 90% confidence interval had to entirely lie between 0.667 and 1.5.

Results showed that IBTRs were observed as first events in 161 participants, 90 of which had received PBI, while 71 had received WBI (HR, 1.22; 90% CI, 0.94-1.58). There was an absolute difference of 0.7% (4.6% vs 3.9%) in the 10-year cumulative incidence of IBTR between accelerated PBI and WBI, respectively.

Based on the upper limit of the hazard ratio confidence interval, accelerated PBI did not meet the criteria for equivalence to WBI in controlling IBTR.

Additionally, the 10-year RFI rate was higher with WBI at 93.4% compared with accelerated PBI at 91.8% (HR, 1.33; 95% CI, 1/.04-1.69; P = .02); this 1.6% difference was determined to be statistically significant.

However, there were no statistically significant differences observed in DDFI, OS, and DFS rates. At 10 years, DDFI was 96.7% and 97.1% with accelerated PBI and WBI, respectively (HR, 1.31; 95% CI, 0.91-1.91; P = .15). OS was slightly higher at 10 years with WBI at 91.3% versus 90.6% with PBI (HR, 1.10; 95% CI, 0.90-1.35; P = .35).

“Basically, the intent-to-treat and as-treated analyses did not refute the hypothesis that partial breast irradiation is inferior and could not declare that whole breast irradiation and partial breast irradiation were equivalent in controlling low breast tumor recurrence,” Vicini said. “However, it was pointed out that the absolute difference between the techniques at 10 years was 0.7%. The [RFI] was statistically significantly higher for partial breast irradiation compared with whole breast irradiation, but again the absolute difference at 10 years was also small, at 1.6%.”

At a median follow-up of 10.2 years in this population, the overall breast cancer event rates showed an IBTR rate of ~4.5%, a diabetes mellitus rate of ~3%, and a mortality rate of ~2%.

Grade 3 adverse events were higher in the PBI arm (9.6%) compared with the WBI arm (7.1%), and grade 4/5 rates were also slightly higher at 0.5% and 0.3%, respectively. Vicini added that the rate of secondary malignancies was similar between the 2 arms. Additional analyses are being performed to assess quality of life and cosmesis.

In a recent statement, SABCS co-director Virginia Kaklamani, MD, leader of the Breast Cancer Program at UT Health San Antonio, commented on the NSABP B-39/RTOG 0413 findings and endpoints.

“It’s important when we design the trials to look at clinically meaningful differences, because we don’t want to harm our patients, but at the same time, we are also harming them by giving them more treatment,” Kaklamani explained. “So, if you’re designing a trial where a 0.7% difference is statistically significant, we probably would have been able to get away with much fewer patients and [if the] difference of 1.5% to 2% [was not] significant, then I think everyone would have been happy about it.”

Vicini F, Cecchini R, White J, et al. Primary results of NSABP B-39/ RTOG 0413 (NRG Oncology): A randomized phase III study of conventional whole breast irradiation (WBI) versus partial breast irradiation (PBI) for women with stage 0, I, or II breast cancer. In: Proceedings from the 2018 San Antonio Breast Cancer Symposium; December 4-8, 2018; San Antonio, Texas. Abstract GS4-04.

<<< 2018 San Antonio Breast Cancer Symposium