Gilead’s remdesivir use for Covid-19 could spill over beyond hospitalised patients yet to require mechanical ventilation

By Reynald Castaneda in London

With Gilead Sciences’ remdesivir expected to lock in Emergency Use Authorization (EUA) for COVID-19 from the FDA, broad use beyond its ideal patients is plausible, which could strain supply. Use in patients on invasive ventilation or even those with milder symptoms but at risk of progressing could negatively impact the supply and demand equilibrium for the drug.

Excitement for remdesivir was at its fever pitch on Wednesday (29 April), when data of varying detail from three different trials dropped. While the Phase III trial (NCT04280705) sponsored by the National Institute of Allergy and Infectious Diseases (NIAID)—considered to be critical for FDA support—reported preliminary positive results, granular patient profile information has yet to be revealed. Gilead also announced data from its Phase III trial (NCT04292899), while the Lancet published data from the China-based Phase III trial (NCT04257656). All three studies recruited severe patients but with varying definitions. Today (1 May), the EMA announced it would start a rolling review based on NIAID’s results.

While data so far suggest remdesivir is ideal to use soon after hospitalisation and prior to invasive ventilation, its utility is likely to be broader—but the balancing act between risk-benefit and drug supply is criticalGilead’s data showing remdesivir’s five-day dosing being comparable to the 10-day schedule could open up use outside of the hospital for the shorter course, while the longer course may be considered for more severe patients who have a higher viral load detected.

While there is agreement there is an ideal therapeutic window, there is the risk of patients being hospitalised too late, which would diminish remdesivir’s efficacy potential. Nonetheless, remdesivir’s potential for EUA is supported by available data so far, but the NIAID data only supports symptom relief for the intravenous therapy, with mortality data less convincing. Yet, the eight-point primary endpoint ordinal scale of time to recovery does have the advantage of offering quicker results—critical in a pandemic—and captures nuances on how patients progress through the disease.

Gilead did not respond to an interview request, while the NIAID pointed to its remdesivir media release.

No clear demarcation between groups allows for wide-ranging use

Because of the ongoing pandemic, there needs to be extra thought put into prescribing remdesivir to the right patients, not just to ensure benefit but also considering the drug’s supply rates, noted Dr Djillali Annane, head, critical care department, Raymond Poincaré University Hospital, Garches, France. Gilead said, as of late March, it has increased its supply to more than 30,000 patient courses, assuming a 10-day course. It aims for 140,000 treatment courses by the end of May and one million by December.

Remdesivir may not work optimally for patients already on mechanical ventilation, noted Dr Olav Dalgard, professor of infectious diseases, Akershus University Hospital, Norway. Yet, on the off chance that they may still benefit, the drug is unlikely to be withheld from such patients, said Dr Robert Shafer, research professor, infectious diseases, Stanford University, California. While patients requiring invasive ventilation were excluded from Gilead’s trial, they were included in the NIAID study and the China-based trial.

However, in the later stage of the disease, targeting the cytokine storm that SARS-CoV-2 infection triggers becomes the most critical therapeutic concern, explained Dalgard, a primary investigator in an investigator-led trial studying hydroxychloroquine’s load reduction potential. Because of remdesivir’s antiviral mechanism, the severe patient’s viral load information may be needed to confirm if the therapy would be beneficial, noted Annane, who has prescribed remdesivir under expanded access.

But viral load checks are mainly relevant in a research setting, said Dr Nelson Lee, professor, Division of Infectious Diseases, University of Alberta, Canada. Patient management is the same regardless of viral load, Shafer noted. Also, in the China-based trial, there was no significant viral load reduction difference between the remdesivir and control arms. This is somewhat concerning because remdesivir is designed to impact the virus’ life cycle and it would have been ideal to prove its mechanism, Dalgard added.

Another potential reason that remdesivir supply could be strained is if the therapy is used as a 10-day course. While the Gilead-run trial demonstrated that five-day dosing is as efficacious as a 10-day course, the latter treatment regimen may be used, particularly for sicker patients who may have longer-term viral replication, Lee added.

On the flip side, the encouraging results that five-day dosing appears as efficacious as 10-day dosing could open up remdesivir use to aged-care facilities or as an outpatient approach for high-risk patients, Dalgard and Shafer said.

However, there needs to be a trial focusing on these specific patients, considering elderly patients have comorbidities, Dalgard added. The NIAID and China-based trials recruited patients 18 years and over, with the Gilead-run trial recruiting 12 years and over. In the Gilead trial, time to clinical improvement in 50% of patients on the five-day schedule was 10 days, and 11 days for those on the 10-day schedule, and more than half of patients in both schedules were discharged from the hospital at day 14 (p=0.14). The NIAID trial uses the 10-day schedule.

Remdesivir ideal when used prior to invasive ventilation

The best therapeutic window for remdesivir seems to be as soon as patients present at the hospital and prior to requiring invasive ventilation, said Dalgard. Remdesivir would be ideal to stop patients who may be on some form of noninvasive oxygen support from progressing to mechanical ventilation use or ICU care, added Shafer. The three aforementioned remdesivir trials all recruited patients who had SpO2 of ≤94%, a group that needs at least noninvasive oxygen support. But SpO2 ≤94% is just a line in the sand, and so remdesivir would likely be used in confirmed COVID-19 patients as long as they have pneumonia symptoms, Lee noted.

Dalgard said he was basing his assessment of remdesivir’s ideal therapeutic window on the China-based trial. While the Gilead-run trial has detailed results, the drawback is that it does not have a control arm, while the opposite is true for the NIAID trial, he added. The China-based data show that for patients with symptom duration of 10 days or less, the remdesivir arm had a numerically faster time to clinical improvement versus the control arm.

However, the China-based trial was terminated early and so is underpowered, Dalgard and Lee noted. The study aimed to enrol 453 patients but stopped at 237, reportedly due to the slowing of COVID-19 cases in that country. Additionally, the study allowed concomitant use of AbbVie’s HIV drug Kaletra (lopinavir/ritonavir), interferons and corticosteroids, said Lee.

Yet, the China-based data should not be discounted totally, as there is a positive efficacy trend, said Dalgard. It seems to have inclusion criteria comparable with the NIAID trial, so there is potential for meta-analysis, noted remdesivir NIAID trial investigator Dr Thomas Lars Benfield, professor, Department of Infectious Diseases, University of Copenhagen, Denmark.

While there is no breakdown of patient profiles for the NIAID trial yet, it seems that a notable chunk of patients recruited are those who are not critically ill, based on reported mortality data, Benfield said. Data based on 1,063 patients shows an 8% mortality rate versus 11.6% for remdesivir versus the control arm, respectively (p=0.059). In Benfield’s site, which recruited 12 patients, all were on a mechanical ventilator at enrolment and the mortality rate was 30%, he noted.

There is still the risk patients may be treated too late, Dalgard said. There are cases where hospitalised patients go straight to mechanical ventilation, Lee noted. Late treatment, though, is more of an issue of awareness about disease symptoms and potential lack of resources in a pandemic setting, Benfield noted. At-risk patients may opt to stay away from hospitals until the last minute due to overcrowding, Shafer noted. The ideal scenario is that patients are in touch with their general practitioner for advice on when to go to hospital, but this is not the case for most, especially in the US where there is fragmented care, he added.

On the other end of the disease spectrum, remdesivir’s peak efficacy potential may not even be in hospitalised patients but those with more moderate symptoms, Dalgard said. Diagnosis is typically based on PCR testing of nasopharyngeal or oropharyngeal swabs. Certain elements of SARS-CoV-2’s viral replication are still unknown, specifically in the lower respiratory tract, Dalgard added.

Remdesivir is relatively well tolerated, but renal and liver function as well as white blood cell count monitoring are needed, making it less ideal for relatively milder patients, Annane and Shafer noted. Yet, with monitoring, remdesivir could still be dosed in patients who have over 94% SpO2 but are at high risk of becoming critically ill, Shafer added.

EUA likely but ordinal scale endpoint leaves knowledge gaps

Even though there is still no definite proof of remdesivir’s efficacy in specific patient groups, EUA is still likely, as there are no red flags regarding safety and there are enough accumulated data suggestive of some benefit, said Annane, Dalgard, Schafer and Benfield.

Yet, there is some reservation with regards to the ordinal scale used to measure time to recovery, the primary endpoint in all three trials, Dalgard said. This scale is less robust than directly measuring mortality rate, which is more of a concern for hospitalised patients than only improving symptoms, he added. NIAID trial results indicate remdesivir patients had 31% faster time to recovery than those who received placebo (11 vs. 15 days, p<0.001) on an eight-point scale.

The NIAID trial’s mortality rate is not statistically significant enough, Annane said. But Dalgard noted the p-value of p=0.059 is encouraging, as the gap between the two arms is likely not by chance. There seems to be a trend toward a mortality benefit, added Dr Krutika Kuppalli, clinical assistant professor, infectious diseases, Stanford University, Palo Alto, California.

Still, there are scant data showing remdesivir can keep patients from progressing to mechanical ventilation or ICU admission or demonstrating it can help patients get off of invasive ventilation, which are important efficacy metrics in a pandemic setting with stretched healthcare resources, Annane added. And so, the FDA could still wait for more mortality data, like from the SOLIDARITY trial (ISRCTN83971151) run by the World Health Organization, which has all-cause mortality as a primary endpoint, Dalgard said.

Another caveat with the ordinal scale is that it leaves room for interpretation, Benfield said. While all patients may start at the same point upon recruitment, movement between arms within the scale may not be comparable, he noted.

Still, the advantage of the ordinal scale is that it needs fewer patients to spell out a difference between two arms, Dalgard said. A mortality endpoint may need 50–100% more patients to spell out a difference compared to the ordinal scale, Benfield explained. The scale also provides information on different scenarios of patient progress, which is important as disease severity does not just mean survival or death, added Lee, a SOLIDARITY trial investigator.

Gilead has a $99.28bn market cap.

Reynald Castaneda is a Senior Reporter for Clinical Trials Arena parent company GlobalData’s investigative journalism team. A version of this article originally appeared on the Insights module of GlobalData’s Pharmaceutical Intelligence Center. To access more articles like this, visit GlobalData.