Asco-GI – Pancreatic cancer field awaits sparse data

After a year in which the pancreatic cancer pipeline lost five phase III agents to clinical failure, investigators are anxiously looking for signs of progress in this notoriously difficult field.

Discussion at the Asco-GI symposium centred on why so many agents had failed recently, as well as which of the new strategies could be the most promising (see table below). At a keynote lecture on Friday Dr Volker Ellenreider, of the University of Goettingen, highlighted two relatively new phase III entrants: Halozyme’s pegPH20 and the little-known napabucasin, being developed by Daiichi Sankyo’s Boston Biomedical unit.

That said, a glance at the pipeline suggests that physicians are in for a relatively lean period, with only a handful of phase II and III pancreatic cancer trials set to read out this year. Dr Ellenreider’s keynote shed little light on pancreatic cancer’s numerous failures.

Agents that dropped out from the pipeline in 2016 have included vastly differing approaches, such as Newlink Genetics’ vaccine algenpantucel-L, small molecules like Incyte’s Jakafi, and radiolabelled antibodies such as Immunomedics’ Y-90 clivutuzumab. As such there is no inference that can easily be drawn, other than that pancreatic cancer is incredibly hard to treat.

Halozyme, whose pegPH20 is only suitable for the roughly 30% of patients with high hyaluronan levels, recently published detailed phase II data in pancreatic cancer.

However, these have proven difficult to interpret and caused some concern among investors as Halozyme had already initiated the Halo-301 trial before they were available. The second stage of the two-part study did show solid improvements in progression-free and overall survival, but when combined with the first part – which might have been affected by a clinical hold – the OS benefit largely vanished.

Meanwhile, Daiichi’s napabucasin, a Stat3 inhibitor that acts on cancer stemness pathways, has entered a very large phase III study on the basis of highly promising but early results. Daiichi reported at Asco last year that a phase I study with 29 evaluable patients showed disease control in 27, 10 of which amounted to partial responses.

Kyrie eleison – Lord have mercy

Pancreatic cancer specialists do have two phase III results to look forward to this year: Eleison’s study of glufosfamide in the second-line setting and Astrazeneca’s Lynparza in first-line maintenance of patients with gBRCA mutations, which are thought present in around 14% of pancreatic cancers.

Glufosfamide is a glucose conjugated prodrug of ifosfamide. An earlier phase III study, conducted by its originator, Threshold Pharmaceuticals, showed a 25% improvement in OS in the second-line setting but this failed to achieve statistical significance.

Meanwhile, Astra will soon see if Lynparza can extend its indicated range from BRCA-mutated ovarian cancer. The only prior data on Lynparza in pancreatic cancer come from a phase I study in which an overall response rate of 21.7% was reported in 23 heavily pretreated patients.

There are also a number of phase II trials that could render results in pancreatic cancer in the first half of this year, and could result in relatively quick moves into phase III.

Perhaps most important is Oncomed’s Yosemite trial of demcizumab, an anti-DLL4 antibody. This is important as it is one of the two main datasets on which Celgene will base its decision on whether to opt into a licensing deal. Celgene, which owns Abraxane, is one of the two main players in pancreatic cancer.

Other companies reporting pancreatic data at Asco-GI included Armo Biosciences, which has a phase III trial with AM0010 (pegylated IL-10). Armo’s 21-patient phase Ib study showed a 16% objective response rate, with two complete responses.

If anything, Asco-GI showed there remains a consensus among key opinion leaders that pancreatic cancer, which is usually diagnosed late and with a highly aggressive phenotype, has a number of features that make it uniquely difficult to treat among solid tumours.

It is the eleventh-most commonly diagnosed cancer, but ranks fourth by cancer-related death – and is expected to rise to second place soon, given the improving prognosis for other tumour types. If one thing remains clear it is that progress in treatment cannot come too soon.

To contact the writer of this story email Robin Davison, reporting from Asco-GI in San Francisco, at [email protected] or follow @RobinDavison2 on Twitter