P2-10-02: Assessment of Tumor Response to Neoadjuvant Chemotherapy Using Ultrasound-Guided Near Infrared Light.

Abstract

Abstract Background Neoadjuvant chemotherapy has been increasingly used in treating breast cancers. Because breast cancer is a heterogeneous disease, it is important to effectively monitor the tumor response to assist in tailoring treatments to response. In our early study, we have introduced a novel ultrasound (US)-guided diffused light imaging in the near infrared (NIR) spectrum to monitor tumor vascular changes which correlated to tumor response. The objectives of this study are: (a) to validate the initial findings with a larger patient pool, and (b) to assess vascular changes at every treatment cycle and to correlate early vascular changes with the tumor pathological response. Methods: 33 patients who underwent neoadjuvant treatment were recruited from Dec. 2007 to May 2011 and their tumor vascular content was assessed with a combined imager consisting of a commercial US system coupled to a NIR imager. Patients were imaged before their treatment, at the end of each treatment cycle and before their surgery. The co-registered US was used to localize the tumor and the NIR imager was used to map the tumor vascular distribution which was assessed based on a percentage total hemoglobin (%tHb) concentration normalized to the pre-treatment level. 18 patients were treated with AC followed by Taxol. This group of patients was monitored every two weeks at the end of each treatment. The remaining patients were on 3-week cycles of chemotherapy and monitored every 3 weeks. 6 patients were treated with TC without Adriamycin (TC), or with Adriamycin (TAC), 6 HER2 positive patients were treated with TC and Herceptin (TCH); and 3 patients were treated with AC/Bevacizumab. Pathologic response was graded based on Miller and Payne system as grade 1: non-responders (A); grades 2 and 3: partial responders (B); 4: near-complete and 5: complete responders (C). Results: In the AC/Taxol group (n=18), there were 5 responders (C), 9 partial (B) and 4 non-responders (A). The statistical significance based on %tHb between groups A and C was achieved at the end of cycle 5 and the rest of the treatment cycles (p<0.05), however, the statistical significance between A and B was only obtained at end of cycle 5 (p<0.05) and not maintained for cycles 6–8. The statistical significance between B and C was only achieved at the end the treatment (p<0.05). For the TC,TAC and TCH group (n=12), there were 6 responders(C) and 6 partial responders (B). The statistical significance between these two groups was achieved at the end of cycle 3 and the rest of the treatment cycles (p<0.05). For the 3 patients who were treated with AC/Bevacizumab, 2 patients achieved complete response and one partial with grade 2. The complete responders had more than 50% reduction in %tHb at the end of cycles 3–4; while the partial one showed only 10–15% reduction during the entire treatment course. Discussion: Our findings indicate that tumor vascular changes assessed by %tHb can be used to predict the tumor pathological response. This is a powerful tool to help predict responsiveness to therapy. Interestingly, dramatic and early responses were noted in the patients who received the biologic agents (bevacizumab and herceptin) and this may be very valuable in following responses using these agents. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P2-10-02.