Sightera Biosciences raises €3M pre-seed to reverse the logic of AI drug discovery

Sightera Biosciences, an Antwerp-based biotech spin-off from the University of Antwerp (UA) and Antwerp University Hospital (UZA), has raised €3 million in a pre-seed round. The round was led by Entourage, Anacura, and QBIC. Sightera was founded in January 2025 by Hendrik Vercammen (CEO), Maxim Le Compte (CSO), Lars Vanlommel (CBO), and Christophe Deben (CDO); the company formally spun out of UA and UZA in January 2026 and is based at the Darwin Incubator in Niel, Belgium.

The €3 million will fund team expansion, platform development, pre-clinical asset advancement, and pharma partnership work — specifically advancing the company's lead molecular glue oncology programme, SIGHT001, to preclinical candidate selection.

Starting where others stop

Most AI drug discovery platforms run chemistry first. A company identifies a biological target, designs or screens small molecules with attractive properties, and then validates whether those molecules actually do anything useful in human disease biology. Sightera reverses the sequence. It begins by taking biological samples from patients whose cancer or fibrosis no longer responds to standard treatment — what the company calls extreme biology — and growing those samples into organoids that preserve the resistance profile. It then runs high-throughput drug screening campaigns on these patient-derived models, generating datasets that capture how treatment-resistant disease responds to different therapies over time. The AI platform is trained on this data. It then designs novel small molecules whose predicted biological behaviour is calibrated to produce a specific response in patient-derived systems — rather than designing molecules with chemical promise and hoping they translate.

"We reverse that paradigm," Vercammen says. "We start with complex patient-derived biology, learn directly from therapeutic responses in patient-derived biology, and use those insights to design entirely new small molecules." The argument is that engineering patient biology into the molecule-design loop from the outset reduces the translational failure rate that accounts for most clinical attrition in AI-derived drug candidates.

Vercammen brings a PhD in medical sciences from UA and a background in translational drug discovery and AI. Le Compte's PhD is in oncology; Vanlommel has a corporate finance background at Lotus Bakeries and CMB.TECH; Deben, a UA professor, established the Antwerp Tumoroid Biobank — the tissue infrastructure that underlies Sightera's biology-first model.

A drug, a deadline, and an unusual investor

Sightera's lead programme is SIGHT001, a molecular glue compound targeting oncology. Vercammen is explicit that the pre-seed is not the company's main capital event: "We are currently preparing a larger financing round to take our lead program SIGHT001 into clinical trials." The €3 million buys the scientific milestone — preclinical candidate selection — that is required before any clinical-stage investor can engage. That sequencing is unusually honest about what early capital is actually purchasing.

The cap table contains one anomalous investor. Entourage, a Ghent early-stage fund, typically backs B2B software companies. Its decision to lead a molecular glue oncology pre-seed is an explicit departure. "Sightera doesn't ask AI to predict biology. It lets biology teach AI," the firm said, describing the investment thesis as "a fundamentally different approach to drug discovery." QBIC, a co-lead, is one of Europe's larger inter-university spin-off funds and has deep ties to the Flemish university system. Anacura brings life sciences operations expertise.

The company's business model — advance programmes to key inflection points, then out-license to pharma — is standard early-stage biotech. The thesis that distinguishes Sightera is the data source, not the exit structure.