Yesterday Lilly struck a blow against its Covid-19 antibody rival Regeneron, with promising results for its LY-CoV555 plus LY-CoV016 doublet. Based on admittedly sparse data so far, Lilly’s combo looks to have the edge over Regeneron’s antibody combo.
However, it was not long before Regeneron was back on the front foot, announcing that it had filed for US emergency use authorisation for REGN10933 plus REGN10987. Regeneron will need to make the most of its head start on filing and manufacturing – plus a direct endorsement from President Trump – if it is to prevail in the battle of the anti-Covid-19 antibodies.
True, Lilly also announced an EUA submission yesterday, but this was for LY-CoV555 alone. The combo is not set to be filed for EUA until November. And, based on data from two different cohorts of the Blaze-1 study, the combo looks a lot more efficacious in terms of reducing viral load.
The primary endpoint of Blaze-1 was change in viral load from baseline to day 11, versus placebo. This was met with the combo, Lilly disclosed yesterday; in a previous update the group said that only one dose of monotherapy, 2,800mg, had hit this endpoint (Lilly leads the Covid antibody charge, September 16, 2020).
The combo used a 2,800mg dose of both LY-CoV555 and LY-CoV016, which are also known as LY3819253 and LY3832479 respectively.
Evercore ISI’s Umer Raffat called the viral load reduction with the combo “a big deal”, noting that Gilead’s antiviral Veklury had been unable to show this; Veklury received an EUA in May.
However, Mizuho analysts noted that the clinical meaningfulness of viral load reductions was up for debate.
On the harder endpoint of Covid-19-related hospitalisations and emergency room visits, the combo did slightly better, showing an 85% relative risk reduction versus placebo; the risk reduction with monotherapy was 72%.
Another advantage of the combo could be avoiding the development of resistance, and this was borne out by the Blaze-1 data. In the monotherapy cohort, putative LY-CoV555-resistance variants were found in 8% of treated patients; with LY-CoV555 plus LY-CoV016 no putative resistance variants have so far been seen.
Stacking LY-CoV555 plus LY-CoV016 up against Regeneron’s combo is harder. Like Blaze-1, the latter’s phase I/II/III trial enrolled non-hospitalised Covid-19 patients, but used a different definition of a medical visit, and the company only provided viral load data at seven days (Regeneron joins Lilly in validating the Covid antibody approach, September 30, 2020).
Still, with the usual caveats about cross-trial comparisons, and based on the limited evidence so far, the efficacy of Regeneron’s doublet looks to fall somewhere between Lilly’s monotherapy and combo.
|Cross-trial comparison of Covid-19 antibody candidates|
|Trial||Blaze-1 mild-to-moderate pts, NCT04427502*||Ph1/2/3, non-hospitalised pts, NCT04425629**|
|Project||LY-CoV555 + LY-CoV016||LY-CoV555^||REGN10933 + REGN10987|
|Viral load chg vs placebo (day 11)||-0.56 (p=0.011)||0.03 (p=0.870)||N/A|
|Viral load chg vs placebo (day 7)||-1.18 (p<0.001)||-0.33 (p=0.063)||-0.60 high dose (p=0.03^^); -0.51 low dose (p=0.06^^)|
|Medical visits vs placebo||0.9% vs 5.8% (p=0.049)||1.6% vs 5.8% (p=0.020)||7.7% high dose; 4.9% low dose; 15.2% placebo|
|*Medical visits defined as Covid-19-related hospitalisation/ER visits.
**Data in seronegative pts; medical visits defined as hospitalisations, ER, urgent care or telemedicine visits.
^Pooled data across 700mg, 2,800mg & 7,000mg doses.
^^Nominal p values.
So, assuming LY-CoV555 monotherapy and the Regeneron combo receive EUA at around the same time, Regeneron should have the edge until Lilly’s doublet gets the nod.
When and if both combos get the go-ahead, availability could become an important part of the equation. Both companies have been racing to ramp up manufacturing, no mean feat given the high doses of antibody involved.
Regeneron yesterday said it had 50,000 doses of its combo ready to go, and expects to have 300,000 “within the next few months”. Lilly is set to make 100,000 doses of its monotherapy available in October, increasing to one million in the fourth quarter.
Interestingly, Lilly has chosen to focus on the lowest studied dose of LY-CoV555, 700mg, despite its apparently lower efficacy, perhaps with an eye on manufacturing constraints. Regeneron has not said which dose of its combo it will take forward, but both involve eye-popping levels of MAbs: the high dose 8g and the low dose 2.4g.
Meanwhile, Lilly expects to have 50,000 doses of its combo available by the fourth quarter.
Free of charge?
All of this looks like good news for healthcare systems battling the pandemic: the aforementioned data might not be stellar but the antibodies should at least keep some Covid-19 patients out of hospital. And, given the lack of other decent options, this should be enough to spur high demand.
Lilly’s stock closed up 3% yesterday, while Regeneron’s shares opened up 3% this morning. Other companies involved in Covid-19 antibody development include Glaxosmithkline and Vir, whose candidate VIR-7831 (GSK4182136) this week entered phase III; and Astrazeneca, whose combo AZD7442 (AZD8895 + AZD1061) is in phase I.
The other good news, at least in the US, is that the Regeneron antibody cocktail could be provided free of charge if President Trump’s recent missive is anything to go by.
Of course, the president might not come through on his promise, but if he does the pledge could end up being a double-edged sword for Regeneron: it is unclear who will pay for Mr Trump's largesse, potentially leaving the company to absorb any losses.
Leerink analysts earlier this week said Regeneron could charge $3,000 per vial in the US and $2,000 per vial elsewhere, leading to peak worldwide sales of $1.8bn next year.
This story is an updated version of a snippet published yesterday. It has also been updated to include the alternative names for the antibodies.