Identifying independent predictors of locoregional recurrence after neoadjuvant chemotherapy could help determine whether a patient with breast cancer will need adjuvant radiotherapy, according to research reported at the Breast Cancer Symposium. The meeting is sponsored by the American Society of Breast Disease, the American Society of Breast Surgeons, the American Society of Clinical Oncology, the American Society for Radiation Oncology, the National Consortium of Breast Centers, and the Society of Surgical Oncology.
Pathologic complete response to neoadjuvant chemotherapy and breast cancer subtypes are known independent predictors of local and regional recurrence, but the combined effect of the two in patients treated with neoadjuvant chemotherapy has not been completely understood.
The new study (Abstract 61), shows that pathologic complete response in the breast and pathologic axillary nodal status after neoadjuvant chemotherapy are independent predictors of locoregional recurrence, and that breast cancer subtypes continue to be independent predictors of locoregional recurrence even when pathologic response is taken into account. Age was also found to be an independent predictor, but only in patients receiving lumpectomy, not in those who have mastectomy.
However, more advanced stage at diagnosis did not adversely affect the rates of locoregional recurrence when pathologic complete response and tumor subtype information were taken into account.
"Inclusion of independent predictors of locoregional recurrence such as age, pathologic response to neoadjuvant chemotherapy, and breast cancer subtypes minimized the effect of baseline stage on locoregional recurrence," said the lead investigator, Eleftherios P. Mamounas, MD, Professor of Medicine at the University of Central Florida, describing the study during a presscast for reporters before the meeting.
The Collaborative Trials in Neoadjuvant Breast Cancer (CTNeoBC) study reviewed 12 large trials of neoadjuvant chemotherapy in a pooled analysis that included 11,955 patients with pathologic complete response information. Median follow-up was 5.4 years.
"Overall five-year rates of locoregional recurrence are generally low-under 10 percent," he noted. The overall five-year incidence of locoregional recurrence for the 5,252 women included in the multivariate analysis was 8.8 percent. The rate for lumpectomy patients was somewhat lower than the overall rate, 7.8 percent, and slightly higher for mastectomy patients, at 10.4 percent.
As expected, the differences in breast cancer subtypes were significant: 4.2 percent for HR+/HER2- tumors grade 1 and 2, which was the lowest five-year incidence of locoregional recurrence by subtype, versus 14.8 percent for HR-/HER2+ tumors of any grade. In between were 9.2 percent for HR+/HER2- grade 3; 9.7 percent for HR+/HER+ of any grade; and 12.2 percent for HR-/HER2- of any grade.
Mamounas cautioned, however, that the groups are not entirely comparable because patients with more aggressive disease are more likely to have mastectomy.
Also, only about one-third of patients with HER2+ tumors received adjuvant trastuzumab, since the data include cases from before the trastuzumab era.
Age was an independent predictor of locoregional recurrence for lumpectomy patients (hazard ratio 1.41), but not for mastectomy patients (hazard ratio 0.97).
"This relation of recurrence to age differing between lumpectomy and mastectomy patients confirms some observations shown in NSABP trials in the past," he said.
For both surgical subtypes, tumor subtypes remained independent predictors of locoregional recurrence even when pathologic complete response was taken into account-"So now we have an additional factor that predicts locoregional recurrence."
Mamounas said the data support the conduct of ongoing clinical trials that attempt to tailor locoregional therapy in the neoadjuvant chemotherapy setting.
A Bit 'Inside Baseball'
Commenting on the results in his role as moderator of the presscast, Harold J. Burstein, MD, PhD, a member of the ASCO Cancer Communications Committee, Associate Professor of Medicine at Harvard Medical School, and a breast cancer specialist at Dana-Farber Cancer Institute, said, "We've known for a long time that this clinical endpoint of complete pathologic response is a powerful predictor of recurrence elsewhere in the body. What these new data show is that it is also a predictor for not having the cancer come back within the chest wall or in the breast itself."
It also points to breast cancer subtype as very important for predicting outcomes, he said, adding that the study results-"although perhaps a bit 'inside baseball'-speak to a very complicated matrix that helps us understand the risk of local recurrence in a woman who has a greater than average risk of breast cancer recurrence."