357P - Prediction of pathologic response to neoadjuvant chemotherapy (NAC) using diffuse optical breast scanner (DOB-Scan) in patients with locally advanced breast cancer (LABC)

Background

No modality exists to accurately predict pathologic response to chemotherapy prior to surgery in patients with LABC receiving NAC. The objective of this study is to use the novel DOB-scan to evaluate the pathologic response of breast cancer to NAC before surgery.

Methods

The study aims to enroll 90 patients and recruitment is ongoing. An eligible patient had a diagnosis of LABC and was recommended NAC followed by resection. DOB-Scan used near-infrared light to examine healthy and breast cancer tissue prior to NAC, at clinic visit prior to each WNAC cycle, and after NAC was completed. Palpated tumour size was measured at the same time; ultrasound (US) was performed prior to NAC start, halfway, and after final NAC. After surgery, the specimens were reviewed by reference pathologist and assigned the pathologic Residual Cancer Burden (pRCB) class (Symmans et al. JCO 2007 ; Yau et al. SABCS 2019 ). We calculated the optical RCB (oRCB) class by determining the minimum reflectance that correlated with maximum absorption on DOB-Scan after the last NAC prior to surgery. Correlation between pRCB and oRCB was modelled to fit a linear regression.

Results

60 patients have been enrolled; data from 30 patients who completed NAC and surgery were included in this interim analysis. Median age was 58 (range 29-72); all but 1 had ductal carcinoma; 22 had axillary lymph node(s) involved. Median longest diameter of the primary tumour on US was 2.3 cm (range 0.7-18). Final size by palpation or US indicated response to NAC, but did not correlate with pRCB. The oRCB clustered linearly with pRCB (pRCB = 4.4 x oRCB), with a sensitivity of 97% and a specificity of 100%. Artificial-intelligence rendered noise reduction of oRCB increased the sensitivity to nearly 100%.

Conclusions

Early results suggest that the oRCB can predict final pathologic response more accurately than palpation or US during NAC. Lower oRCB indicated better response to NAC, with the ability to predict pRCB and pathologic complete response prior to resection. DOB-Scan can be used as a point-of-care hand-held instrument to assess tumour response and assist with treatment decisions in patients receiving NAC.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Michael Smith Foundation; Crystal Gala Foundation.

Disclosure

J. Ko: Financial Interests, Personal, Advisory Board: AstraZeneca, Astellas, Pfizer, Takeda; Financial Interests, Personal, Invited Speaker: BMS, Merck; Financial Interests, Personal, Research Grant: Janssen, Bayer. All other authors have declared no conflicts of interest.