Dear Members of the Selection Committee,

I am excited to apply for the Hong Kong Laureate Forum. As a PhD student specializing in bioinformatics and breast cancer research, I am eager to engage with leading scientists and fellow young researchers in this esteemed academic exchange.

My research focuses on integrating multi-omics data to identify biomarkers for breast cancer prognosis and treatment response. Specifically, I study the evolution of HER2-positive breast cancer under targeted therapies and investigate immune metabolism in the tumor microenvironment. My interdisciplinary approach has been enriched through collaborations with oncologists and bioinformaticians.

Attending the Forum would provide a unique opportunity to interact with world-renowned laureates and explore cutting-edge advancements in life sciences and medicine. The event’s emphasis on cross-disciplinary exchange aligns with my belief that innovation in biomedical sciences thrives at the intersection of diverse expertise. Furthermore, the Forum’s international and cross-cultural environment would help broaden my scientific network and refine my research vision.

I am eager to contribute to and learn from this inspiring gathering. Thank you for your time and consideration.

Kang Wang, MD, Ph.D

Karolinska Institutet

HER2-targeting antibody-drug conjugates (ADCs) have become standard of care for HER2-positive breast cancer, yet clinically useful predictive biomarkers of response are lacking. Here we report the multi-omics characterization of fresh-frozen pre-treatment biopsies from a prospective, randomized trial (PREDIX HER2, n = 197), comparing neoadjuvant docetaxel, trastuzumab, and pertuzumab (DHP) with trastuzumab emtansine (T-DM1) monotherapy in HER2-positive breast cancer. Using exome sequencing, shallow whole-genome sequencing by CUTseq, RNA sequencing, mass spectrometry-based proteomics, whole slide image analysis and spatial transcriptomic, we correlated multi-omics features with treatment response at surgery. We identified favourable biomarkers specifically associated with response to T-DM1, including pathogenic TP53 mutations, germline HLA-A01 supertype, tumor dormancy, and ADC-trafficking HER2-enriched tumor cells. In contrast, genome instability and high infiltration of immune cells associated with type 2 immunity were associated with resistance to T-DM1. We integrated all data sources using a comprehensive machine learning analysis workflow to predict responses to T-DM1 (area under the curve (AUC) = 0.75), DHP (AUC = 0.79), and both treatments (AUC = 0.81). These findings reveal distinct molecular and cellular determinants of response to ADCs versus dual HER2 blockade, paving potential avenues for individualized anti-HER2 treatment strategies.