Roche, seeing MAGE magic wane, makes pair of solid tumor bispecifics disappear

Roche no longer sees the magic in MAGE-A4. The Swiss drugmaker signaled a retreat from the TCR T-cell target in its first quarter results by putting a pair of phase 1 solid tumor programs out to pasture along with hopes of recouping the $490 million paid out for the assets.

Genentech, a subsidiary of Roche, put down $100 million in upfront and near-term milestone payments to co-develop one of the candidates with Immunocore in 2018. The drug candidate, a bispecific T-cell engager targeting MAGE-A4, entered the clinic the following year, but both developers have walked away from the program since the release of phase 1 ovarian cancer data late last year.

In February, Immunocore stopped funding the program, apart from the cost of winding down the phase 1 trial. The action left Roche’s Genentech with an exclusive worldwide license to the MAGE-A4 HLA-A02 soluble TCR bispecific therapeutic candidate compounds. It seems Roche doesn’t want the compounds, either.

The Swiss drugmaker disclosed the axing of RG6290, a drug candidate based on Immunocore’s ImmTAC technology that is also known as IMC-C103C, alongside news of the removal of another MAGE-A4 asset from its pipeline. Roche took the other candidate, HLA-A2-MAGE-A4xCD3 bispecific antibody RG6129, into the clinic months after Immunocore began a phase 1 trial of IMC-C103C.

Early this year, Roche said it had stopped enrolling patients in the study of RG6129, pulling the targeted completion date of the trial forward by more than two years in the process. The removal of the asset from the pipeline recasts the cessation of enrollment as the start of Roche’s retreat from the candidate.

Roche invested in the programs in the belief a molecule that simultaneously binds to CD3 on cytotoxic T-lymphocytes and HLA-A2 MAGE-A4 on tumor cells could orchestrate an anti-cancer immune attack. As MAGE-A4 is overexpressed by a variety of cancer cell types, the mechanism is potentially applicable to a range of tumors.

The attributes of MAGE-A4 have made it a popular target for TCR T-cell biotechs, with Immunocore and its sibling Adaptimmune among the companies to go after the protein. Adaptimmune is closing in (PDF) on a filing for FDA approval of its first-generation MAGE-A4 candidate.

Roche disclosed the MAGE-A4 news as part of a pipeline update that saw it punt two other programs. The list of castoffs includes zinpentraxin alfa, also known as PRM-151 and RG6354. Roche previously stopped a phase 3 trial of the candidate in idiopathic pulmonary fibrosis early because the study looked on course to fail. 

Now, the drugmaker has stopped work on myelofibrosis, too. Roche acquired the drug candidate in the 2019 takeover of Promedior, which cost it $390 million upfront. Removing the myelofibrosis program from the midphase pipeline leaves Roche without active clinical programs involving zinpentraxin alfa.

Finally, Roche also used the update to disclose the removal of a geographic atrophy program RG6312 from its phase 1 pipeline. Since RG6312 entered the clinic in 2020, the Big Pharma has suffered setbacks in its pursuit of the eye indication, failing a phase 3 trial in 2017 and dropping galegenimab last year. But it still has two shots on goal, including a Lineage Cell Therapeutics candidate that recently entered phase 2.