CD47’s long goodbye

Given the repeated setbacks with Gilead’s anti-CD47 MAb magrolimab, the latest of which was a phase 3 failure in acute myelogenous leukaemia and discontinuation of all work in haematology, it’s remarkable that Gilead still hasn’t canned the entire project. The pipeline includes four mid-stage magrolimab trials in solid tumours, and for now these are continuing.

Magrolimab is the result of Gilead’s $4.9bn acquisition of Forty Seven, and while that move now looks costly perhaps Gilead doesn’t want to throw in the towel until all hope is extinguished. Equally remarkable is the amount of competitor work still ongoing in CD47 blockade; this field’s only major success is last year’s phase 3 win for ALX’s evorpacept in gastric cancer, and full data from this are a key 2024 catalyst.

Even that gastric cancer result, from the phase 2 part of the Aspen-06 trial, caused relatively cautious enthusiasm, given questions about the relevance of the study’ setting in a changing treatment landscape. Moreover, evorpacept itself has had problems, disappointing in head and neck cancer and myelodysplastic syndromes, leading ALX to scrap work in AML.

Final analysis from Aspen-06’s 122-patient phase 2 part is expected in the second quarter of 2024 – ASCO seems a possible venue – with phase 3 starting by the end of this year.

Better than magro

Still, while some might question evorpacept, at least its prospects seem brighter than magrolimab’s. The Gilead MAb has endured several clinical holds and trial discontinuations, of which the ending of Enhance-3 was only the latest. 

This leaves four continuing phase 2 magrolimab studies in Gilead’s pipeline: Elevate TNBC (Trodelvy/chemo combo), Elevate HNSCC (Keytruda/zimberelimab/chemo combo), Elevate CRC (Avastin/chemo combo), and a chemo combo trial in solid tumours. There is no guarantee that these will continue, and Gilead says it is reviewing magrolimab’s “safety ... across all ongoing solid tumour trials, and will provide an update on this assessment as soon as possible”.

Apart from evorpacept one other asset with activity on CD47/SIRPα remains in phase 3: I-Mab’s lemzoparlimab. However, this is in a China trial, and the project received a vote of no confidence when AbbVie pulled out of a licensing deal last September. 

Meanwhile, Surface Oncology was once a prominent CD47 player, but discontinued work after seeing toxicity, and ended up being sold to Coherus last year on the strength of the rest of its portfolio for just $65m. 

The mid-to-late-stage CD47 pipeline

Source: OncologyPipeline.

From one point of view work against CD47 resembles that on TIGIT, in the sense that neither approach has shown much monotherapy activity, but combos have suggested that the mechanism is adding something, though what precisely this is is still unclear.

One CD47 company that did boast monotherapy activity was Trillium, and as a result it was bought by Pfizer for $2.3bn. In a 2022 pipeline update Pfizer touted CD47 as a $3 billion-plus peak year sales opportunity, but subsequent work has proceeded slowly. Mid-stage trials with the Trillium-originated maplirpacept and ontorpacept continue, but neither asset featured in Pfizer’s latest pipeline presentation.

The industry pipeline also includes two projects with a slightly different modality, PD-L1 blockade combined with anti-CD47 activity in a single bispecific. In phase 2 these comprise Mabwell’s 6MW3211 and Innovent’s IBI322, and some of their studies are due to be completed this year.

A similar approach is being pursued by Akeso, which in addition to the MAb ligufalimab recently took into phase 1 AK132, an anti-Claudin18.2 x CD47 MAb. And it’s worth remembering Shattuck Labs, whose own take on CD47 differs by the inclusion of a CD40L domain to activate macrophages; further data from early trials of SL-154 are due this year.

Unfortunately evidence backing much of industry’s CD47 work has yet to emerge, and magrolimab’s repeated stumbles don’t do this field any favours.