Doomsday for Allakos

If lirentelimab reduces eosinophils in diseases characterised by eosinophil accumulation, why doesn’t it work?

Yesterday Allakos was worth $4.4bn. Today its valuation is a minute fraction of that after the catastrophic failure of lirentelimab, the group's sole clinical asset, in rare inflammatory gastrointestinal disorders slashed its share price by 92%.

The results from two late-stage studies make, frankly, little sense, and the group’s failure to offer any explanation for the findings has left Allakos in extremis. And, since the anti-siglec 8 MAb was considered one of the most valuable pipeline projects in biopharma, its loss can only exacerbate the sector’s dismal climate.

Some investors believe that the warning signs were there, with some remarkably consistent data emerging in phase 2. Others point to the lack of fast-track designation after those ostensibly positive mid-stage results, which might suggest a lack of faith from the FDA.


The latest data come from two trials: the phase 3 Enigma 2 study in patients with eosinophilic gastritis and/or eosinophilic duodenitis, and Kryptos, a phase 2/3 trial in eosinophilic oesophagitis. These disorders are characterised by eosinophilic infiltration of the gastrointestinal tract, as well as gastrointestinal symptoms varying from dyspepsia and obstruction to diarrhoea and ascites.

The trials were aptly named: their results are mysterious to say the least. Lirentelimab significantly reduced eosinophil counts in the areas of the GI tract the studies were targeting, but failed to make a dent in symptoms as reported by patients.

Data on Allakos's lirentelimab (AK002) in eosinophilic gastrointestinal diseases
Enigma 2 topline results
Co-primary endpoints Lirentelimab (n=91) Placebo (n=89)
Proportion of responders as determined by gastric or duodenal tissue eosinophil counts 84.6% 4.5%
     P value vs placebo 0.0001  
Absolute mean change in patient reported total symptom score* -10.0 (baseline 29.5) -11.5 (baseline 27.7)
     P value vs placebo 0.343  
Kryptos topline results
Co-primary endpoints Lirentelimab high dose (n=91) Lirentelimab low dose (n=93) Placebo (n=92)
Proportion of responders as determined by oesophageal tissue eosinophil counts 87.9% 92.5% 10.90%
     P value vs placebo 0.0001 p<0.0001  
Absolute mean change in patient reported dysphagia symptom questionnaire -17.4 (baseline 34.2) -11.9 (baseline 36.4) -14.6 (baseline 35.2)
     P value vs placebo 0.237 0.247  
*Includes abdominal pain, nausea, early satiety, loss of appetite, abdominal cramping & bloating. Source: company release.

By publication time, Allakos had not responded to a request for an explanation of how these apparently contradictory findings might have come about. 

It is possible that, though accumulation of eosinophils is seen in eosinophilic gastrointestinal diseases (EGIDs), these are not in fact a cause of the condition – merely a consequence. 

Sanofi and Regeneron's Dupixent has shown significant reductions in eosinophil counts and disease symptoms, versus placebo, in two phase 3 trials in eosinophilic oesophagitis.

Theoretically, then, Dupixent’s anti-IL4/IL13 mechanism is doing something beyond eosinophil reduction, and that action is responsible for the symptom improvement. Dupixent is expected to be filed in eosinophilic oesophagitis in the coming year. 

The same cannot be said for lirentelimab. True, there is another phase 3 EGID trial yet to report: Eodyssey in eosinophilic duodenitis is expected to yield data in mid-2022. But any hope here might have been quashed by Leerink analysts, who wrote that there do not seem to be any notable differences in study design or mechanistic rationale for believing that results from Eodyssey will differ substantially from Enigma 2.


Allakos said it was mulling the path forward for the MAb in EGIDs, and that it would continue to assess subcutaneous lirentelimab in atopic dermatitis, chronic spontaneous urticaria and asthma. The phase 2 Atlas trial in atopic dermatitis is underway.

But Leerink removed EGID sales from their model for lirentelimab, and reduced the value attributed to its other indications. In November they had forecast peak revenues of $3.4bn in eosinophilic gastritis and/or eosinophilic duodenitis and $855m in eosinophilic oesophagitis. 

Forecasts of this magnitude are what put lirentelimab at the top of Evaluate Vantage’s list of the most valuable pipeline projects in its 2022 preview report, with an NPV of $6.2bn. It was also high up in this year’s analysis of the biggest unpartnered assets. How are the mighty fallen.

Share This Article