No abstract

Background The presence and extent of residual invasive cancer after neoadjuvant treatment (NACT) is a strong prognostic factor for risk of recurrence, especially in triple-negative (TN) and human epidermal growth factor receptor 2-positive (HER2+) breast cancer (BC). Recent advances in the standard-of-care NACT improved pathological complete response (pCR) rates in published clinical trials. We evaluated the pCR rates, defined as ypT0-is ypN0, in our real-world BC population and in estrogen receptor-positive [ER+] HER2-, HER2+ and TN subgroups and their association with tumour, patients' characteristics and disease-free survival (DFS). Methods We retrospectively identified early BC patients receiving NACT between January 2013 and December 2017. Demographics, patient and disease characteristics, pathological responses, toxicities, dose delays and reductions were recorded. Simple statistics, Fisher's exact test, chi-squared method and Cox regression were used as appropriate. Results 794 patients identified had median age of 50 years (range 24-87) and 93.9% (745 patients) ECOG performance status (PS) 0. 3.0% (24) had clinical stage I disease, 68.0% (540) stage II and 29.0% (230) stage III. 71.0% (564) had grade 3 disease and 91.8% (729) ductal histology. 33.7% (257) had ER+/HER2-, 25.8% (205) had TN and 38.0% (301) HER2+ disease. Overall, 6.8% (54) patients received platinum. 36.6% (291) patients had dose reductions and 24.3% (193) dose delays. Along with NACT, 51.6% (147) of the HER2+ patients received Trastuzumab and Pertuzumab and 48.4% (138) Trastuzumab alone. pCR rate was 33.1% in the overall population and significantly different in ER+/HER2-, HER2+ and TN subgroups (12.84% versus 52.0% versus 28.43% respectively, p<0.001). pCR was influenced by grade (1: 0%; 2: 24.3%; 3: 36.1%, p 0.005) and histology (ductal: 34.2%; lobular: 10.0%; mixed 25.0%; p 0.01). In the HER2+ subgroup, there was a trend for improved pCR rates for patients receiving Pertuzumab and Trastuzumab (57.0%) versus Trastuzumab alone (51.0%). No statistically significant differences were seen based on patients' characteristics including age and PS or in case of treatment dose reductions and delays. Early discontinuation of NACT was associated with lower pCR rates (20.5% vs 36.29%, p <0.001). Of interest, pCR rates remained consistent across the period 2013-2017 in the overall population. We observed a trend for improved pCR in the HER2+ (2013: 47.5%; 2014: 44.4%; 2015: 66.7%; 2016: 51.0%; 2017: 51.4%) and TN cohorts (2013: 23.5%; 2014: 25.0%; 2015: 25.0%; 2016: 33.3%; 2017: 34.1%) but not in the ER+/HER2- group. Median DFS was 83.8 months (95% CI 62.0-NR) in the overall population. Although not reached in the TN cohort, median DFS was different according to disease subgroups (HER2+: 83.78 months; TN: NR; ER+/HER2-: 62.0 months, p <0.0001). Conclusions In our analysis pCR rates are consistent with data published in literature and higher in HER2+ and TN disease. The impact of new agents had a relatively low impact on pCR rates in our overall population over the last 5 years, although they produced gradual improvements in the HER2+ and TN subgroups. Citation Format: Battisti NML, True V, Chaabouni N, Chopra N, Lee K, Shepherd S, Shapira-Rotenberg T, Joshi R, Mohammed K, Allen M, Ring A. Pathologic complete response rates following neoadjuvant systemic therapy in 794 patients with early breast cancer: The Royal Marsden experience [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P1-15-08.

1/1/15-31/12/17 were included. Clinical records were reviewed. Those without a CT, inoperable when referred, with a diagnosis other than breast cancer or with loco-regional recurrence were excluded. Rates of metastases by tumour size (greatest by MRI, USS and mammography), lymph node stage (MRI and USS), receptor status and multifocality were calculated. Results: 185 patients were identified and 22 were excluded (no CT scan (6), inoperable (5), referred for adjuvant therapy ( 7), locoregional recurrence (3) and angiosarcoma (1)). 104/163 (64%) patients had a normal CT scan. 7/163 (4%) patients had definite metastases identified on their staging CT and 52/163 (32%) had an equivocal scan with further imaging required. Following additional imaging a further 10 patients had metastatic disease confirmed giving an overall incidence of of 10% (17 patients). The incidence of metastatic disease by stage (M1/total), (N þ ¼ node positive): T0Nþ 0/1; T1N0 0/11; T1Nþ 0/12; T2N0 1/42 (2%); T2Nþ 9/53 (17%); T3N0 1/8 (12.5%);. T3Nþ 4/32 (12.5%); T4N0 1/1 (100%); T4Nþ 1/3 (33%). T2N0M1 and T3N0M1 patients both had triple negative disease. Multifocal disease: 7/163 of which 4/7 (57%) were metastatic. All 4 of multifocal M1 were node positive. For patients with an equivocal CT scan, 72 additional scans were arranged: ultrasound (16), MR (25), CT (20), bone scan (6), PETCT (5). Conclusions: The incidence of metastatic disease detected on imaging during work-up for NACT was 10%. The rate of metastatic diagnosis increases with T stage in patients with clinical N0 disease. Staging can be omitted in patients T1/2N0 at diagnosis. Omission of staging CT scans in the T1-2N0 cohort would have reduced the imaging frequency across all modalities by 30% when both initial and additional imaging are considered.