45P - The challenge of evaluating new targeted therapies: Opportunities in stratifying the therapeutic response per tumour location

Abstract

BackgroundCompared with conventional cancer chemotherapy, tyrosine kinase inhibitors show novel patterns of radiology response, therefore using chemotherapy-based response criteria can be suboptimal. Coupling the tumor’s volume as an imaging biomarker with a stratified analysis of the therapeutic response per sites of disease may offer more insightful information in drug efficacy. We propose to compare the therapeutic response between Varlitinib plus Capecitabine (VC) and Lapatinib plus Capecitabine (LC) in analyzing the changes of tumor’s volume per tumor’s locations in patients with HER2-positive metastatic breast cancer after trastuzumab therapy. MethodsWe retrospectively analyzed data from the ASLAN001-003 study (NCT02338245). At baseline, 42 patients displayed at least one target lesion, representing a set of 88 target lesions: Lung (31%), Breast (26%), liver (23%), nodes (17%), miscellaneous (3%). Responses were evaluated at week 12 on 35 patients (14 VC; 21 LC). Between two arms, we compared responses using non-parametric tests and compared the proportion of tumor at each location using the Chi-square test. We used the R software for statistics, p < 0.05 was considered a significant difference. ResultsThe distribution of tumor size at baseline was different at the different tumor locations (p < 0.001). Between the two arms the proportion of tumor at each location were not equivalent with p = 0.126 and 0.06 for breast and lung tumors respectively. The global average change from baseline was VC=-64.1% and LC=-26.59% (p = 0.13) The average change of breast tumors was significantly different in the VC (-84.2%) and the LC (-24.9%) (p < 0.001) arm. Between breast and liver tumors, the average response was significantly different (p = 0.007) in the VC arm, while not in the LC arm (p = 0.94). Also, between nodal and liver tumors with p = 0.06 in the VC and p = 0.99 in LC arm. ConclusionsThe therapeutic response differed across tumor locations, the differential response was drug-dependent. For a fair comparison, the proportion of tumor at each tumor location must be balanced between arms. The stratification of the therapeutic response can be used for assessing new therapies, helping designing cocktails or refining drug’s indications. Clinical trial identificationASLAN001-003 - NCT02338245. Legal entity responsible for the studyMedian Technologies/Aslan Pharmaceuticals. FundingHas not received any funding. DisclosureH. Beaumont: Full / Part-time employment: Median Technologies. N. Faye: Full / Part-time employment: Median Technologies. C. Klifa: Full / Part-time employment: Median Technologies. C-Y. Hsieh: Full / Part-time employment: Aslan Pharmaceuticals. All other authors have declared no conflicts of interest.