Abstract P6-01-05: Radiological evaluation of neo-adjuvant endocrine therapy in hormone-receptor positive early breast cancer

Abstract

Recently, there is increased attention to neo-adjuvant endocrine therapy in breast cancer. Especially for selected tumour types and fragile elderly patients this might be a promising alternative to chemotherapy. Monitoring treatment effect during neo-adjuvant endocrine therapy is crucial to allow a timely switch to chemotherapy in case of a non-successful treatment. Most trials evaluating neo-adjuvant endocrine therapy use palpation as primary outcome and multiple radiological modalities as secondary outcomes. The aim of this study was to determine which evaluation corresponds best to pathological resection size after 6 months of neo-adjuvant endocrine therapy. This analysis was conducted in the TEAM-IIA trial in which 102 patients with early breast cancer (>2 cm and >50% ER expression) were treated with neo-adjuvant exemestane. In total, 83 patients were treated for 6 months and 19 for 3 months. Therapy response evaluation was performed using repeated palpation (mostly by the same clinician), mammography, ultrasound and MRI. Only measurements within 2 months before surgery were considered. After surgery, the size of the remaining tumour was reported and used as reference. In total, 93 resection size measurements were available. From the 93 patients for whom resection size was available, 69 patients were evaluated by palpation, 53 by ultrasound, 42 by mammography and 29 by MRI. Overall, palpation showed to be the most reliable predictor for resection size (correlation of 49%), followed by mammography (31%), ultrasound (14%) and MRI (1%). Mammography showed the smallest mean absolute error (MAE, 8.7 mm), followed by ultrasound (9.2 mm), palpation (11.4 mm) and MRI (12.3 mm). The low correlation of MRI with resection size was mostly due to a relative high number of radiological complete remissions (14%, n=4), of which only one was a true pathological complete response (pCR), while the other tumours were up to 80 mm at resection. Although of less influence on the correlation to resection size, false complete remissions were observed in all other modalities. Time to surgery was an important factor for all modalities. After correcting for non-predictive radiological complete responses and limiting the measurements to one month before surgery, all correlations increased significantly (mammography=72%, palpation=70%, ultrasound=58%, MRI=50%) with a concomitant decrease in mean absolute error. The low correlation of MRI with resection size was mostly due to non-visible measurements, interpreted as radiological complete remissions of which only one in four was a true pathological complete response (pCR) while the other tumours were 25, 65, and 80 mm at resection. This is the first study to report on the reliability of radiological evaluation during neo-adjuvant endocrine therapy. In this study, mammography was the most reliable radiological method, with stronger correlation and small mean absolute error. Non- visible observations in neo-adjuvant endocrine therapy did not always reflect pCR. Hence, in the neo-adjuvant endocrine therapy setting, radiological complete responses should be interpreted carefully, especially in MRI, and use of other modalities or improved image processing methods may be considered. Citation Format: Blok EJ, Charehbili A, Kroep JR, Seynaeve CM, van de Velde CJH, Kuppen PJK. Radiological evaluation of neo-adjuvant endocrine therapy in hormone-receptor positive early breast cancer. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P6-01-05.

Estrogen Receptor Pathway Activity Score to Predict Clinical Response or Resistance to Neoadjuvant Endocrine Therapy in Primary Breast Cancer

“…A contributing factor may have been inaccuracy in assessment of tumor size by manual palpation, which was performed by different MDs at nonfixed time points during treatment. Nevertheless, a detailed comparison of tumor size at resection versus estimated size by palpation, mammography, US, and MRI in the TEAM IIa cohort suggested palpation as most accurate measurement method (40). In addition, in the NEWEST study, threedimensional ultrasound assessment of tumor volume was used to assess clinical response, and again, a low ERPAS was associated with PD, while ERPAS could not statistically separate PR and SD.…”