Kenneth A Kern
Lymphovascular invasion (LVI) describes intravasated tumor cells within intratumoral lymphatic or blood vessels. LVI is reported routinely on pathologic examination because it is prognostic for reduced disease-free and overall survival across many solid tumors. In contrast, it is not considered a predictive biomarker because its presence cannot discriminate which patients benefit from a given therapy. To become predictive, a biomarker must reflect the aberrant, specific biology a drug targets. Bispecific antibodies with functional arms against immunologic and vascular targets are a recent class of biologics that simultaneously engage the malignant immunologic and angiogenic components of a tumor. In this setting, deconstructing LVI into its invasive lymphatic and vascular components, and expressing their ratio, is hypothesized to add predictive capability to this histologic finding. LVI-positive tumors are enriched for parameters relevant to both arms of this class of bispecific antibodies. Immunologically, they carry higher tumor mutational burden, more frequent PD-L1 expression, and TGF-β signaling. Angiogenically, blood vessel invasion reflects a VEGF-A-driven, disordered, T-cell-excluding vasculature. The author proposes the LVI Component Hypothesis, which is that tumor cell intravasation into lymphatic versus blood vessels can be expressed as a ratio, the lymphatic-to-vascular invasion ratio (LVR). This ratio may predict benefit from PD-L1×VEGF-A bispecific antibodies. No clinical data currently link the LVR to response, so it is presented as a hypothesis to be tested, first in archived tissue from completed trials of PD-L1×VEGF-A bispecific antibodies and then in prospective randomized trials. If validated, the LVR might help to select those patients most likely to respond, raising the response rate among those treated and possibly sparing those unlikely to benefit.