Startup Spotlight: Firefly Bio

Firefly Bio aims to bring degrader-antibody conjugates (DACs) to the market. A $94 million Series A round will help to get things rolling.

A good rule in life is not to use a sledgehammer to crack a nut. You don’t need 12 meetings to fix a PowerPoint slide, you don’t need a Scag Cheetah II to mow your front yard (as cool as it would be), and you really, really don’t need to use a vacuum to get rid of the spider in your bedroom. In medicine, and specifically cancer, however, we oftentimes don’t have the luxury of choosing. Chemotherapy drugs, as many lives as they save and extend every day, are in many ways molecular sledgehammers: they kill or severely impair cells that they encounter, and they don’t tend to discriminate much between healthy cells and tumor cells¹.

Over the past few decades, tremendous efforts have been undertaken to make cancer drugs more specific. One popular example for this is the advent of antibody-drug conjugates (ADCs) in cancer therapy. ADCs are chemotherapy drugs that are linked (i.e. conjugated) to an antibody which has high affinity towards a protein that is enriched on the surface of specific cancer cells. Take Pfizer’s and Genmab’s Tivdak, for instance (image below). As all ADCs, it is made up of three parts:

1. A cytotoxic small molecule. In this case, it is monomethyl auristatin E (MMAE). MMAE exerts its cytotoxic function by binding to a structural protein called tubulin and thereby disrupting the functionality of cellular structures and processes.

2. A linker. In this case, it is a valine-citrulline dipeptide linker with a para-aminobenzyl carbamate (PABC) spacer module. Valine-citrulline linkers are cleavable through a group of proteins called cathepsins. When in contact with cathepsins², valine-citrulline linkers release the cytotoxic small molecule from the antibody in order to exert their function on the tumor cells.

3. A cancer cell-targeting antibody. In this case, it is a monoclonal antibody against the tissue factor (TF), which is overexpressed in many cancers including cervical cancer, for which Tivdak is approved.

Overview of ADC drugs. From: Tong et al., 2021

Tivdak and other ADCs have created a true paradigm shift in cancer. With 13 FDA-approved ADCs and over 100 in various stages of clinical development, it is clear that the market is heading towards more targeted strategies of fighting cancer. Last year alone saw a wave of deals in the ADC space with large valuations. Abbvie bought out ImmunoGen for over $10 billion, Merck started a collaboration with Daiichi Sankyo worth up to $22 billion, and Pfizer acquired Seagen for the hefty price tag of $43 billion. ADCs are clearly here to stay, and many assume that with improved antibodies and linker chemistries, the next generation of ADCs will become even more effective and safe.

However, ADCs don’t come without their fair share of limitations. The original idea was that for chemotherapy drugs which are too toxic at effective concentrations (e.g. MMAE), conjugation to the cancer-cell targeting antibody would enable their use at higher concentration. As it turns out, however, in many cases the tolerated concentration is not much higher for the ADC than for the respective small molecule. The main reason for this is that because the small molecule is still highly potent, the cost of non-specificity is very high³.

An alternative idea that gained traction in recent years was to use a protein degrader instead of a cytotoxic small molecule, resulting in what is referred to as a degrader-antibody conjugate (DAC). Protein degraders themselves had some therapeutic success with the development of so-called proteolysis-targeting chimeras (PROTACs). PROTACs solve for the age-old issue that some proteins⁴ are not easily targeted with small molecules⁵ because they do not have readily accessible binding pockets or active sites that are too broad. Instead of trying to inhibit the disease-associated protein, PROTACs tag it for degradation by the ubiquitin-proteasome system (UPS).

PROTAC Examples. The target-binding region is shown in green, the linker is shown in orange, and the E3 ligase-binding region is shown in blue. From: Békés et al., 2022, Nature Reviews Drug Discovery

Pfizer’s and Arvinas’ ARV-471⁶ (image above), for instance, targets the estrogen receptor on one end (green) and the cereblon (CRBN) E3 ligase on the other end (blue) of a flexible linker (orange). It is the E3 ligase recruitment which “tags” the estrogen receptor for degradation by the UPS. As the estrogen receptor plays a key role in cancer development (specifically in breast cancer), this strategy has proven to be very attractive. A key advantage of PROTACs is that they are generally not cytotoxic to healthy cells, merely targeting a single protein (e.g. the estrogen receptor). However, an obvious improvement would be the ability to target cancer cells specifically and degrade the tumor-associated protein exclusively where needed.

This is where DACs come in. A DAC is essentially an ADC with a PROTAC instead of a cytotoxic small molecule. There are two important advantages that a DAC would have over an ADC:

1. Non-specific target cells (i.e. healthy cells) would be less susceptible as the active compound is not cytotoxic but would merely attack the protein that the PROTAC component targets. This is less likely to be detrimental.

2. The presence of a PROTAC opens up a wide range of new targets. Instead of targeting a general cellular structure such as microtubule, DACs could engage with specific cancer-causing proteins. This opens up many opportunities for targeted approaches.

DACs have so far been largely an academic endeavour, maybe with the exception of Genentech⁷. Now, Firefly Bio is the first startup to specifically aim for developing DACs and bringing them to the clinic. Founded in Switzerland by a team of very experienced academics and bioentrepreneurs, including Nobel laureate Carolyn Bertozzi and former Merck and Novartis CTO Bernhard Geierstanger, Firefly Bio aims to first focus on solid tumors and previously established targets.

The now completed $94 million Series A financing round, co-led by Versant and MPM BioImpact with additional cash infusion from Decheng Capital and Eli Lilly, will help to progress a program that apparently already has proof-of-concept data established.

There is good reason to be excited: DACs seem like a good candidate approach to become the most elaborate nutcrackers to date, and Firefly Bio might be at the forefront of their development. Good luck to everyone involved!

1. There are dozens of possible examples for this, but a good one is the widely used doxorubicin.

2. Optimally, as with newer ADCs, linkers contribute to specificity by being sensitive to degradation from cancer-specific factors. Valine-citrulline linkers are more specific to cathepsin B, which is commonly enriched in cancers. However, they also have affinity for other cathepsins. There is a whole lot of research on optimal linkers, well-summarized here.

3. Here’s two reviews about this that go into some more detail. Antibody–drug conjugates: Recent advances in payloads - ScienceDirect and Mechanisms of ADC Toxicity and Strategies to Increase ADC Tolerability (nih.gov)

4. This is unfortunately oftentimes true with cancer-associated proteins and/or mutations.

5. You will often hear the word “undruggable” for these proteins.

6. ARV-471 recently received the FDA Fast Track Designation for use in patients with metastatic breast cancer.

7. Genentech has pioneered work on an antibody-directed BET degrader. Here’s the paper: https://pubmed.ncbi.nlm.nih.gov/31674143/