Ep187: Eric Fischer on Creating a New Class of Medicines

What It Covers

Eric Fischer, structural biologist at Dana-Farber Cancer Institute, explains how targeted protein degraders and molecular glues work as a new drug class, tracing the field from thalidomide's rediscovered mechanism through the 2014 breakthrough papers, the $80M Deerfield-Dana-Farber Center for Protein Degradation, and spinout companies including Neomorph now partnered with Novartis, Biogen, and AbbVie.

Key Questions Answered

• •Thalidomide as clinical proof of concept: The 2014 discovery that thalidomide and lenalidomide work by binding ubiquitin ligase cereblon and degrading transcription factors IKZF1 and IKZF3 provided overnight clinical validation for an entirely new drug class. Millions of multiple myeloma patients were already safely taking these drugs, eliminating the need to prove the mechanism could work in humans — a shortcut that compressed years off the field's development timeline.
• •Molecular glues vs. PROTACs — size determines tractability: PROTACs use two warheads connected by a linker, often exceeding 1,000 daltons, creating challenges with solubility, permeability, and plasma protein binding that complicate dose prediction. Molecular glues like thalidomide, at roughly 260 daltons, face none of those barriers. When selecting a degrader modality, smaller targets with no existing binding pocket favor molecular glues; well-characterized targets with known binders are more compatible with PROTAC approaches.
• •Mass spectrometry cataloging to unlock molecular glue targets: Because molecular glues lack predictable design rules, Fischer's lab uses mass spectrometry to systematically identify which proteins become susceptible to degradation when glue derivatives are modified. Diversifying the chemical scaffold and mapping resulting target shifts builds a catalog researchers can then optimize against a specific disease target — converting a historically serendipitous process into a structured, repeatable discovery workflow.
• •Composite protein surface mimics antibody CDR logic: Molecular glues work by presenting a small chemical moiety on the surface of a ubiquitin ligase, creating a composite binding interface between drug and ligase that recruits the target protein. This mirrors how antibody CDRs — just a few diversified amino acids on a conserved IgG scaffold — achieve high specificity. Recognizing this analogy helps chemists design glues by focusing diversification on the small presented chemical group rather than the whole molecule.
• •Academic-industry hybrid structure preserves discovery breadth: The Center for Protein Degradation at Dana-Farber, funded by Deerfield Management at $80M in 2018, was structured as a separate team inside the institution rather than a standalone biotech. This preserved open communication between academic and translational work, allowed technology advancement that small biotechs cannot resource, and enabled spinning out companies like Neomorph only after achieving sufficient mechanistic clarity — avoiding the premature focus pressure that comes with early-stage venture timelines.
• •Target selection criteria for spinout viability: Fischer applies three filters before committing a project to a startup: a strong biological rationale that a specific target drives disease and that degrading it would shift clinical outcomes measurably within the first few dozen patients; a clear, tractable chemistry path to a drug-like molecule; and an honest assessment of whether the team and setting are genuinely best positioned to execute. Projects where trials require pharma-scale resources are routed to large company partnerships rather than biotech formation.