Interest in bispecific antibodies is growing – and with the potential to hit a combination of immuno-oncology targets or act in a similar way to CAR-T therapy it is no wonder that many of the big players are getting involved in the space. A beneficiary of this attention is the private UK group F-star, which hopes that a new approach will help it corner some of the market.
The company believes that, as its bispecifics have a different structure versus competitors, they could be more stable and efficacious – though with all of these assets at the preclinical stage this remains unproven. But the group’s chief business officer, Jane Dancer, points to deals with the likes of Bristol-Myers Squibb (NYSE:BMY) and Abbvie (NYSE:ABBV) as evidence of interest in its approach.
F-star claims to be the only company creating bispecifics that incorporate a binding site on the constant Fc region of the antibody, resulting in what it calls an Fcab, in addition to separate binding regions at the antibody's usual antigen-binding arms or Fab fragments.
Conventional bispecifics tend to involve two antibody halves merged together, each with a different Fab fragment.
Manufacturing traditional bispecific antibodies can be difficult, and F-star highlights a risk of mismatch between the two antibody halves “that could give rise to selection/purification issues and thus impact other core formulation properties”.
As well as avoiding this issue, Ms Dancer believes that F-star’s bispecifics have improved stability over conventional products thanks to not needing to change their structure significantly.
Whether this translates to better clinical performance is still up for debate, and it will be a while before F-star’s claims can be put to the test: the group’s lead project, FS102, is not actually a bispecific. It is an Fcab antibody fragment targeting Her2 and is partnered with Bristol-Myers Squibb; Phase I trials in Her2-positive breast and gastric cancer are underway.
This might seem a long time coming – F-star had been progressing FS102 towards the clinic last time EP Vantage spoke to the company in October 2013, and the deal with Bristol was signed in October 2014 (F-star forges immuno-oncology spin off in latest test of asset-centric model, October 23, 2013). But progress in the bispecific arena in general has been slower than might have been expected after an initial flurry of interest.
The first of F-star's bispecific antibodies to yield human data looks likely to be FS118, an internal immuno-oncology project targeting PD-L1 and Lag3, already the subjects of combination approaches using separate therapies. F-star plans an IND filing by the end of this year and first-in-human trials in 2018.
The company has a broader preclinical I-O pipeline, some of which is partnered with Abbvie, but is secretive about what targets are involved. Ms Dancer says the group is not chasing novel targets – rather, it is the combination approach that is its unique selling point.
As well as potentially reducing costs over conventional combos, using a bispecific antibody could have clinical and safety benefits, she believes, for example by allowing therapy to be targeted to tumor tissue, or hitting two targets in close proximity, which could improve efficacy.
It could be argued, however, that dosing two antibodies separately allows for easier dosing modifications than with a bispecific. Ms Dancer disagrees: "Our modular platform allows [us] to tune the potency of each part of the [bispecific], which could potentially reduce off-tumor toxicity.”
There will have to be some specific benefit with bispecific antibodies if F-star is to succeed: the I-O combo space is extremely hot, and companies already looking at Lag3/PD-1 combos include Bristol-Myers Squibb and Merck & Co.
Another way in which F-star differs from many other start-ups is its structure, which includes various smaller separate companies under the broader group’s umbrella.
This asset-centric set-up “gives us flexibility to do different types of deals”, Ms. Dancer says. It means that a product, or group of products, can easily be separated from the broader group and licensed or sold, so an acquirer does not have to buy the entire company. The structure also means that shareholders can pick and choose which part or parts of the business they want to invest in – although, so far, the investors across the groups have remained broadly similar.
F-Star has partnered each of its companies with a bigger player: F-star Alpha, which covers FS102, with Bristol; F-star Beta, which houses the I-O assets, some of which are partnered with Abbvie; and F-star Gamma, which involves CNS applications, with Denali. It also has a fourth vehicle, F-star Delta, for future entities.
F-star says Bristol has an option to acquire F-star Alpha for $475m before Phase IIb trials begin, and Denali has the option to purchase F-star Gamma for $450m before Phase I.
F-star also has a partnership with Merck KGaA dating back to 2011, but will not say much about this other than that it targets inflammatory disease.
Crossing the BBB
With the majority of F-star’s portfolio, and bispecifics in general, focused on oncology, the Denali collaboration highlights the potential of this approach in CNS disorders.
The idea is that the technology could allow antibodies to cross the blood-brain barrier, a problem that has scuppered other CNS candidates. The Fcab component of the bispecific antibody binds to a transporter, which gets it across the blood-brain barrier; once there, the Fab region targets the disease in question, Ms Dancer explains.
F-star is not disclosing what CNS disorders it is looking at with Denali, but Ms Dancer says its products could have utility in “any disease where you need to get across the blood-brain barrier” – ranging from neurodegenerative disorders to brain cancers. This could be one way for the company to differentiate itself in a crowded oncology field.