OncoMed digs itself out from under Alpine avalanche

The 43% selloff in OncoMed Pharmaceuticals’ shares seems a bit harsh given that tarextumab’s failure was in that most vexing of oncology indications, pancreatic cancer. But in a tough trading environment news that trial investigators called a halt not just because of futility but because of a worsening on key endpoints was not taken well.

It also casts doubt on whether the California-based group’s other phase II asset, demcizumab, will be any more successful as it acts on the same signalling pathway in tumour-initiating cells. If not, two more promising agents will have been removed from a pancreatic cancer pipeline that as always looks inadequate (see table below).

The Alpine trial was testing tarextumab in combination with Abraxane and gemcitabine in first-line pancreatic cancer to determine whether it could improve on the two approved agents alone. Abraxane’s rate of complete or partial response is 23%.

In any event, the trial’s safety board, in reviewing data from 172 patients, found a significant worsening of response rate and progression-free survival and a trend towards lack of overall survival benefit. These findings were true even in a subgroup of patients showing high expression of the Notch 3 receptor, one of the antibody’s two targets.

Aiming for the heights

Alpine was one of two phase II trials for the project, the other being Pinnacle in small cell lung cancer. Data from the lung disease study, which combines tarextumab with etoposide and platinum therapy, will not emerge until 2017. This means that investors might be waiting 18 months before hearing the type of announcement that would encourage partner GlaxoSmithKline to execute its option, which is part of a collaboration that dates back to 2008.

OncoMed’s other phase II asset, the Celgene-partnered demcizumab, is likewise in pancreatic and lung cancer trials. As a delta-like ligand antibody, demcizumab also blocks the Notch pathway; thus the Leerink analyst Paul Matteis questioned whether the positive phase I data this project generated would successfully translate into phase II.

Assuming neither of OncoMed’s agents succeeds in pancreatic cancer, the supply of promising novel clinical-stage candidates looks as thin as ever – although many developers try in pancreatic cancer because patients’ prognosis is so poor with existing treatments.

The main innovation in pancreatic cancer in recent years has been to improve chemotherapies – see Celgene’s Abraxane and Baxalta’s Onivyde.

Phase III trials are under way in kinase inhibitors in Jakafi, momelotinib and Imbruvica; an immunotherapy, NewLink’s algenpantucel-L; a PARP inhibitor in Lynparza and an antibody-radiotherapy conjugate in yttrium Y-90 clivatuzumab tetraxetan.

Surviving phase II candidates include Clovis Oncology’s PARP inhibitor rucaparib and Novartis’s hedgehog inhibitor Odomzo.

OncoMed's stumble should not necessarily give any developer reason for optimism, but with the bar so low trying in pancreatic cancer is often worth the gamble.

To contact the writer of this story email Jonathan Gardner in London at [email protected] or follow @ByJonGardner on Twitter