Abstract P3-11-17: A novel biodynamic imaging assay predicts success or failure of neoadjuvant chemotherapy in breast cancer patients

Background: Use of neoadjuvant chemotherapy (NAC) in breast cancer patients has increased significantly over the past decade. The clinical benefits of NAC, including potential to downstage disease, facilitation of breast conserving surgery and use of pathologic response as a prognostic marker, are well established. However, with multiple regimens approved and recommended for NAC, choosing the optimal therapy for individual patients remains a challenge. Biodynamic imaging (BI), a novel technology that captures cellular motility in living tissue via Doppler spectroscopy, could be used ex-vivo to prospectively evaluate the efficacy of systemic therapies in patient tumor samples prior to treatment. This study aimed to determine whether BI could accurately predict likelihood of response to NAC in breast cancer patients.

Methods: Fresh core biopsies were obtained from 84 patients prospectively enrolled in an IRB-approved clinical trial at 5 institutions between 1/5/17 and 8/3/2018. Patient tumor tissue was collected at time of routine diagnostic biopsy and sent to a central laboratory where it was divided into intact tumor fragments measuring approximately 1mm in diameter. Fragments were placed into 96 well plates and imaged using the BI assay (Onco4D™), while being challenged by various cytotoxic agents for up to 20 hours. Cellular characteristics and motility signatures were evaluated and compared to pathologic NAC response established upon surgical resection (mastectomy or lumpectomy).

Results: At the time of this analysis, centrally-confirmed pathologic response data were available for 17 patients treated with doxorubicin/cyclophosphamide + taxane (AC). Pathologic outcomes are pending for an additional 8 AC patients. The remainder of the 84 patients initially enrolled in the study either did not receive NAC (n=10), have not yet selected a course of therapy (n=12), or received NAC regimens other than AC (n=37). Of the 17 currently evaluable AC-treated patients, 4 had triple negative (TN) disease, 12 were hormone receptor positive, and 1 hormone receptor negative patient showed equivocal HER2 results. Two of the TN patients were known to harbor pathogenic BRCA1 mutations and received carboplatin in addition to AC. Seven of 17 patients (40%) displayed resistance to AC (2 with progressive and 5 with stable disease) while 10 experienced objective response (8 partial and 2 complete response). A multilinear regression model using 10 BI markers accurately classified 16 of 17 patients (94%) while producing 1 false prediction of partial response for a patient with stable disease clinically (R-squared=0.9994, p<.0001). The positive predictive and negative predictive values of BI to AC response were 100% and 91%, respectively.

Table 1.Performance Characteristics  Clinical   OutcomeResponseNon-ResponseTotalClassifierResponse10111 Non-Response066 Total10717

Conclusion: BI was able to accurately predict patient response to neoadjuvant AC, demonstrating the potential for the platform to support personalized patient therapy. This clinical trial is ongoing and will report out results for TC (docetaxel/cyclophosphamide), TCHP (docetaxel/carboplatin/trastuzumab + pertuzumab), and additional AC patients as outcome data are accrued.

Citation Format: Whitacre E, Manahan E, Burak W, Morgan T, An R, Loesch D. A novel biodynamic imaging assay predicts success or failure of neoadjuvant chemotherapy in breast cancer patients [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 P3-11-17.

Abstract P3-13-10: Minimally-invasive (percutaneous) stereotactic and ultrasound-guided lumpectomy for DCIS and small breast cancers

Background

In spite of advances characterizing the biology of individual breast cancers, current analytics cannot adequately identify truly indolent lesions. As a result many indolent breast cancers and DCIS are over-treated in order to avoid under-treatment of truly life-threatening cancers. Reducing over-treatment for DCIS and small cancers is a current priority for breast cancer management, to optimize both patient care and resource utilization. Intact minimally-invasive excision (percutaneous lumpectomy) with radiofrequency basket capture can eliminate the need for open excision of many atypical ductal hyperplasia (ADH) breast lesions (Whitworth, et al, Ann Surg Oncol, 2011). Percutaneous lumpectomy provides standard margin and lesion analysis of the specimen, accomplishing the same goals as open excision. Here we report results of the IPEX study, a prospective phase 4 trial evaluating the same minimally-invasive strategy when employed to manage small breast cancers and DCIS.

Methods

94 women age 31-86 had minimally-invasive (percutaneous) removal of DCIS or small breast cancers using a 20 mm radiofrequency basket capture technique with stereotactic or ultrasound guidance. Tissue elasticity permitted removal through a small incision (less than 10 mm). A second 20 mm basket capture was used to shave and further evaluate margins. Procedures were conducted with adequate local anesthesia and P.O. sedation in the imaging department. Patient tolerance scores were recorded. Standard radiologic evaluation (specimen and breast) and histologic criteria were applied in all cases. Standard open sentinel node biopsy was offered to patients with invasive breast cancers. All patients provided informed consent for the study.

Results

Final histologic lesion size was 1-20 mm. Of 40 DCIS and 54 invasive lesions, 18 (19%) had positive margins and/or open re-excision, including one patient who had re-excision in spite of negative margins because of inadequate shaved-margin quality (no residual disease on pathology). 58 patients completed APBI (balloon or strut-based brachytherapy) and 16 had whole breast radiation. Of 54 patients with invasive cancers, 34 (age 66-86) had no sentinel node biopsy based on patient preference and low risk features. Patient pain scores averaged 1.9 out of 10 (range 0-7).

Conclusions

Minimally-invasive (percutaneous) lumpectomy provides satisfactory management of DCIS and small breast cancers, accomplishing the same goals as open surgery. The specimen is intact so the pathologist can conduct standard histologic analysis of the lesion, distinguishing DCIS, ADH and invasive cancer. Margin analysis is standard. Compared to open surgery, less tissue can be removed because of small lesion size with more precise targeting and resection. Minimally-invasive (percutaneous) lumpectomy reduces morbidity, distress, discomfort and expense associated with overtreatment of small breast cancers.

Citation Format: Whitworth PW, Graham C, Schonholz S, Manahan E, Phillips R, Robertson Y, Hardwick MK. Minimally-invasive (percutaneous) stereotactic and ultrasound-guided lumpectomy for DCIS and small breast cancers. [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 P3-13-10.

Third place winner of the Conrad Jobst Award in the gold medal paper competition. Prevention of spinal cord dysfunction in a new model of spinal cord ischemia

Paraplegia or paraparesis caused by temporary cross-clamping of the aorta is a devastating sequela in patients after surgery of the thoracoabdominal aorta. No effective clinical method is available to protect the spinal cord from ischemic reperfusion injury. A small animal (rat) model of spinal cord ischemia is established to better understand the pathophysiological events and to evaluate potential treatments. Eighty-one male Sprague-Dawley rats weighing 300 g to 350 g were used for model development (45) and treatment evaluation (36). The heparinized and anesthetized rat was supported by a respirator following tracheostomy. The thoracic aorta was cannulated via the left carotid artery for post-clamping intra-aortic treatment solution administration. After thoracotomy, the aorta was freed and temporarily clamped just distal to the left subclavian artery and just proximal to the diaphragm for different time intervals: 0, 5, 10, 15, 20, 25, 30, 35, and 40 minutes (five animals per group). The motor function of the lower extremities postoperatively showed consistent impairment after 30 minutes clamping (5/5 rats were paralyzed), and this time interval was used for treatment evaluation. For each treatment, six animals per group were used, and direct local intra-aortic infusion of physiologic solution (2 mL) at different temperatures with or without buffer substances was given immediately after double cross-clamp to protect the ischemic spinal cord. Arterial blood (2 mL) was infused in the control group. The data indicate that the addition of HCO3-(20 mM) to the hypothermic (15 degrees C) solution offered complete protection of the spinal cord from ischemic injury.(ABSTRACT TRUNCATED AT 250 WORDS)