Our current research focuses on: (i) understanding the relationship of Epstein-Barr virus (EBV) with cancer immunity and immunotherapy; (ii) developing and translating cytotoxic CD4+ T cell (CD4 CTL)-based therapeutic approaches for B cell malignancies and other cancers; (iii) elucidating the molecular pathogenesis of B cell lymphomas, particularly that of diffuse large B cell lymphoma (DLBCL). Our guiding principle is making discoveries and mechanistic insights in animal models and validating the findings in humans.

EBV is a human herpesvirus, which preferentially infects B cells but can also infect other cell types. It has coevolved with our species for millions of years, infecting and taking up residency in about 95% of humans worldwide. EBV can, in concert with predisposing factors, cause cancers in a tiny minority of infected people, and has been regarded as “the first identified human tumor virus”. Strikingly, recent studies from us and others indicate that EBV, through its signaling protein LMP1, induces potent cytotoxic CD4 and CD8 T cell responses against a wide range of tumor-associated antigens (TAAs), in addition to various EBV-encoded antigens. The induced TAA-specific T cells not only work together with EBV-specific T cells to prevent EBV-related cancers, but they can also recognize and attack EBV-unrelated cancers via shared TAAs, as demonstrated in vitro. This opens the door for further research to address whether EBV-induced T cells can indeed prevent EBV-unrelated cancers in some people, how they may fail to do so in other people, and whether they can be reactivated for cancer therapy.

Exploiting the unique immunostimulatory function of LMP1, we developed a novel approach to produce multi-specific CD4 CTLs for treating B cell malignancies. This enables targeting a wide range of endogenous tumor antigens, such as TAAs and neoantigens, without the need to identify them in each patient. Efforts to translate the CD4 CTL-based adoptive cell therapy to clinical testing are underway. Meanwhile, we endeavor to develop LMP1-based immune approaches for other cancers.

DLBCL is the most common lymphoid malignancy in adults. The integration of gene expression and various sequencing technologies has dramatically accelerated our understanding of the disease. Various DLBCL subtypes and their featured genetic lesions have been revealed. To understand the functional significance and operation mechanisms of these lesions, we recapitulate the genetic events, individually or in combination, in mice. In this regard, our previous work has contributed to the understanding of the roles of BCR, NF-kB, BLIMP1 and BCL6 pathways in DLBCL pathogenesis, and our ongoing efforts are focusing on other genes, including the dual-specificity phosphatase 2 (DUSP2).