RAB5A expression is a predictive biomarker for trastuzumab emtansine in breast cancer

2021 
ABSTRACT PURPOSE Targeted therapeutics strongly depends on validated biomarkers in order to select patients most likely to benefit from the treatment. HER2 serves as a predictive biomarker for HER2-targeted tyrosine kinase inhibitors and monoclonal antibodies. HER2 may, however, also be utilized as a transport gate for delivery of cytotoxic agents into the cell, such as for HER2-targeted antibody drug conjugates (ADCs; e.g. trastuzumab emtansine (T-DM1)). The predictive biomarkers for such ADCs may be more complex, also reflecting the intracellular transport. METHODS Five HER2-positive breast and ovarian cancer cell lines were evaluated with respect to T-DM1 sensitivity and correlated to the expression levels of proteins involved in endocytic trafficking including RAB4A, RAB5A and RAB11A, with possible impact on ADC pharmacology. The results were confirmed in a clinical cohort consisting of patients from the adaptive breast cancer clinical trial I-SPY2 where pathological complete response (pCR) was correlated to the RNA expression level of RAB4A, RAB5A and RAB11A. A subset of the clinical KAMILLA trial including 19 patients was used as a verification cohort where semi-quantitative IHC of RAB5A was correlated to progression free survival (PFS). RESULTS The early endosome marker RAB5A, was found to correlate positively to T-DM1 sensitivity in the cell line panel. Correlation between RAB5A expression and T-DM1 sensitivity (pCR) was confirmed in patients treated with trastuzumab emtansine/pertuzumab in the I-SPY2 trial, but not in the trastuzumab/paclitaxel control arm. The clinical correlation was verified in the patients from the KAMILLA trial where semi-quantitative RAB5A IHC staining correlated significantly positive to PFS. CONCLUSION The present results indicate that RAB5A is a predictive biomarker for T-DM1 and outline, for the first time, proteins involved in endocytic trafficking as predictive biomarkers for ADCs.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    39
    References
    0
    Citations
    NaN
    KQI
    []