Particularly in the unfavorable subtypes—triple-negative, HER2-positive, and high-grade hormone receptor–positive patients—having positive nodes does not bode well in terms of locoregional recurrence.
—Eleftherios (Terry) P. Mamounas, MD, MPH, FACS
Women who achieve a pathologic complete response (pCR) to neoadjuvant chemotherapy rarely have local or regional recurrence of breast cancer, but this largely depends on tumor subtype, which remained an independent predictor of locoregional recurrence when pathologic response was taken into account in the CTNeoBC analysis.
These findings—which were derived from the pivotal meta-analysis of pCR studies by Cortazar et al,1 involving almost 12,000 women—were presented at the 2014 Breast Cancer Symposium by Eleftherios (Terry) P. Mamounas, MD, MPH, FACS, Medical Director of the Comprehensive Breast Program at the UF Health Cancer Center in Orlando, Florida. Dr. Mamounas is also Chairman of the Breast Committee of NRG Oncology, Philadelphia.
“There is limited information on rates and patterns of [locoregional recurrence] in patients treated with neoadjuvant chemotherapy, and this has resulted in controversy on the use of sentinel lymph node biopsy and postoperative radiotherapy,” Dr. Mamounas noted.
He said the various associations between pathologic complete response and locoregional recurrence, by tumor subtype, “can potentially help to better identify patients at higher risk for recurrence who may benefit from the addition of radiotherapy, and those at low risk who may not need it.”
The data presented found that pathologic complete response and tumor subtype were both strong predictors of locoregional recurrence. These two predictors may be more informative than tumor stage at diagnosis, Dr. Mamounas reported.
In an interview with The ASCO Post, Dr. Mamounas said, “This confirms what we have known from previous trials about the meaning of [pathologic complete response]. Having a pCR puts the patient at low risk for recurrence, but within that group, you do see differences by subtype. In the worse subtypes, such as triple-negative, women who achieve a pCR have recurrence rates in the single digits, even those with locally advanced disease. For good-prognosis subtypes, pCR doesn’t matter as much, as locoregional recurrence rates were low for all groups irrespective of pCR.”
“We saw that a bad prognostic factor becomes less important if you achieve a pCR,” Dr. Mamounas said.
While both pathologic complete response and breast cancer subtypes have been shown to independently predict for locoregional recurrence, the combined effect of both these factors on locoregional recurrence has not been established. By examining predictors of locoregional recurrence after neoadjuvant chemotherapy, the CTNeoBC aimed to elucidate this issue.
The main definition of pathologic complete response used was ypT0/is, ypN0 (ie, absence of invasive cancer in the breast and axillary nodes; ductal carcinoma in situ allowed). The endpoints were time to first locoregional recurrence in the absence of prior distant recurrence (locoregional recurrence concurrent with distant recurrence was included as a locoregional recurrence event), and cumulative incidence of locoregional recurrence. Risk factors evaluated in the model included age (≥ 50 or < 50 years), tumor subtype, pCR status, baseline stage, and surgery type (lumpectomy or mastectomy).
Postoperative radiotherapy was delivered to virtually all lumpectomy patients and to about one-third of mastectomy patients. Trastuzumab (Herceptin) was given to HER2-positive patients neoadjuvantly in 38% of cases and adjuvantly in 45%.
The investigators assessed locoregional recurrence rates among 11,995 women with stage I to III breast cancer treated with neoadjuvant chemotherapy. After a median follow-up of 5.4 years, locoregional recurrences occurred in 992 (8.3%) patients.
“In patients treated with neoadjuvant chemotherapy, the overall 5-year rates of [locoregional recurrence] were generally low—less than 10%,” Dr. Mamounas noted.
Data on all covariates in the multivariable analysis were available for 5,252 patients, of whom 391 had locoregional recurrences. The median follow-up for this group was 3.5 years.
pCR as Predictor for Recurrence
The 5-year cumulative incidence of locoregional recurrence by breast pCR and nodal status post-treatment was 6.2% for women attaining pathologic complete response (ypT0/is, ypN0), 6.7% for those with residual disease in the breast only (ypT1-3, ypN0) and 11.8% for those with positive nodes (ypTany, ypN+). Compared to women with a pCR, those with residual disease in the breast and none in the nodes had a 1.6 times increased risk of locoregional recurrence, whereas those with cancer in the nodes had a 2.7 times higher risk of recurrence (P < .0001).
“Particularly in the unfavorable subtypes—triple-negative, HER2-positive, and high-grade hormone receptor–positive patients—having positive nodes does not bode well in terms of locoregional recurrence,” he added. These patients may need more aggressive local therapy and radiotherapy.
The effect of pathologic complete response on locoregional recurrence was evident in patients treated with either mastectomy or lumpectomy plus breast radiotherapy.
Tumor Subtype as Predictor of Recurrence
The rate of locoregional recurrence varied considerably by tumor subtype (P < .001), with the highest rate (14.8%) observed among patients with hormone receptor–negative/HER2-positive tumors and the lowest rate (4.2%) among women with grade 1 or 2 hormone receptor–positive/HER2-negative tumors. Locoregional recurrence rates were 12.2% for patients with triple-negative disease, 9.7% for patients with hormone receptor–positive/ HER2-positive tumors, and 9.2% for those with grade 3 hormone receptor–positive/HER2-negative tumors.
The odds ratios give an idea of the increased risk by subtype. For example, compared to patients with grade 1 or 2 hormone receptor–positive/HER2-negative tumors, patients with hormone receptor–negative/HER2-positive tumors had a five times increased risk for locoregional recurrence.
Age, baseline stage (3 vs 1 or 2), and surgery type were not independent predictors of recurrence. However, Dr. Mamounas added that in women undergoing lumpectomy, age “does matter.” Higher risk is incurred in younger women with poor-prognosis subtypes, especially those lacking pathologic complete responses. In this younger age group, locoregional recurrence occurred in 35% of hormone receptor–negative/HER2-positive women who did not achieve a pCR, vs 11% of those with a pCR, he reported.
“Age is an independent predictor of [locoregional recurrence] in lumpectomy patients, but not in mastectomy patients,” he said. “They may need more intensive treatment.”
He also noted that the study’s findings support the conduct of ongoing clinical trials attempting to individualize the use of radiotherapy and surgery in patients treated with neoadjuvant chemotherapy: NSABP B-51/RTOG 1304 and ALLIANCE 11202. ■
Disclosure: Dr. Mamounas reported no potential conflicts of interest.
1. Mamounas EP, Cortazar P, Zhang L, et al: Loco-regional recurrence after neoadjuvant chemotherapy: Pooled analysis results from the Collaborative Trials in Neoadjuvant Breast Cancer (CTNeoBC). Breast Cancer Symposium. Abstract 61. Presented September 4, 2014.
2. Cortazar P, Zhang L, Untch M, et al: Pathological complete response and long-term clinical benefit in breast cancer: The CTNeoBC pooled analysis. Lancet 384:164-172, 2014.
At the Breast Cancer Symposium, William M. Sikov, MD, Associate Professor of Medicine at the Warren Alpert Medical School of Brown University, Providence, Rhode Island, gave a talk on the use of pathologic complete response in the clinic and summarized the CTNeoBC findings for The ASCO Post.