Clinical scorecard for COVID-19: lessons from the first wave of controlled trials
An analysis of 25 controlled trials for COVID-19
Breaking through the noise of poor-quality clinical studies are the first clear signals of a treatment paradigm for COVID-19, indicating how and when to use the two major therapeutic approaches -- attacking the virus and modulating the immune system.
The most recent example came June 16, when University of Oxford’s RECOVERY trial -- a large master protocol study -- found the anti-inflammatory steroid dexamethasone effective in reducing mortality, particularly in patients on ventilators (see “UK Approves Steroid”).
Together with findings on remdesivir and a handful of other antivirals, the readouts support the prevailing hypothesis that antivirals are most effective early in disease, when viral replication is the primary driver, and that immunosuppressants are most effective later on, when an over-aggressive immune response takes over, often being the cause of death.
Figure: Mapping COVID-19 readouts to disease stage
The results have also begun to support and rule out specific agents, a picture that should fill out more over this summer as over 200 trials have completion dates before Sept. 1, according to ClinicalTrials.gov and the Chinese database www.chictr.org.cn (see “Over 200 COVID-19 Trials Nearing Completion”).
High-profile master protocols such as RECOVERY will command most of the attention, but searching for patterns among the more numerous small studies can also yield insights.
BioCentury identified readouts from 27 controlled trials on pubmed.gov, the preprint server medRxiv, and in press releases that collectively tested 16 agents across all stages of the disease. The analysis excludes retrospective studies, single-arm trials, case reports and reviews.
While the controlled studies vary in size and quality, all hold at least some lessons.
Figure: Clinical readouts from major COVID-19 mechanisms
Remdesivir stands out among the antivirals for producing positive results in the most patients, but it remains an open question as to which patients are best suited to receive it.
Although its effect size has been modest, the nucleotide analog appears to shorten the time to recovery by a few days in patients with moderate disease, according to the 584-patient Phase III SIMPLE study by Gilead Sciences Inc. (NASDAQ:GILD), and in patients with severe disease, according to a 1,059-patient trial by NIH which enrolled mild to severe patients.
Remdesivir received Emergency Use Authorization (EUA) from FDA on May 1, based on the NIH data.
However, the SIMPLE trial suggested remdesivir was more effective as a five-day course of treatment in moderate patients, with no significant improvement in patients who received a 10-day course (see “More Data Needed”). Gilead’s second SIMPLE study, which was in severe patients, showed no difference between a five-day and 10-day course, but because it lacked a control arm, it is impossible to know whether either regimen had benefit in the setting.
A randomized, double-blind study of the antiviral in the severe setting conducted in 237 patients in China failed to demonstrate faster recovery.
Figure: Antiviral readouts
There’s little support left for hydroxychloroquine and chloroquine diphosphate, which have been tested in more controlled trials and across more diseases stages than any other agents. Positive results have been observed in just two small studies of hydroxychloroquine, one of which was not randomized and suffered from multiple sources of bias; whereas negative results have been seen in five controlled studies of hydroxychloroquine or chloroquine, including the RECOVERY master protocol, involving about 40 times as many patients in total.
The U.S. and France have rescinded EUAs for the drugs, and on June 19, Novartis AG (NYSE:NVS; SIX:NOVN) announced that it was ending its Phase III trial of hydroxychloroquine.
Data are available from controlled trials of at least five other antivirals, but most of these studies have been small and their controls weak. Only 20% compared the therapy against placebo, and the other 80%, which evaluated antivirals against a different drug or SOC, lacked blinding.
The result is that while the studies in aggregate tend to support use of antivirals earlier in the course of disease they aren’t sufficient to rule these other agents in or out.
Figure: Antiviral trial design
COVID-19: an immune disease
Severe COVID-19 can be thought of as an immune disease as much as a viral infection, and immune modulation has emerged as the leading approach to COVID-19 in terms of number of clinical trials in public databases (see BioCentury’s COVID-19 Clinical Trial Dashboard).
Though COVID-19 requires a robust immune response to clear the viral infection from the body, excess inflammation can arise that becomes pathogenic itself, driving the lung damage that leads to ARDS (acute respiratory distress syndrome).
Several observational studies have shown that IL-6, TNFα and CRP all rise steeply in the transition from moderate to severe disease.
Dexamethasone has underscored the case.
In the RECOVERY master protocol, the steroid reduced 28-day mortality by 41% in patients who required ventilation and by 25% in patients on other forms of oxygen support. The drug had no benefit in patients who did not require oxygen support. The trial enrolled 2,104 patients into the treatment arm and 4,321 controls.
Table: Immunosuppressant progress
Early results from IL-6 inhibitors have been mixed. Two investigator-run studies of Actemra tocilizumab demonstrated a survival benefit: one in moderate and severe patients, and the other in severe and critical patients. A third study did not find a survival benefit in a “recent onset” population not on ventilation.
Across all published studies, including retrospective and single-arm designs, at least 11 have claimed a survival benefit of tocilizumab, whereas four have not.
The Genentech Inc. unit of Roche (SIX:ROG; OTCQX:RHHBY) is collaborating with Massachusetts General Hospital on a Phase III trial of Actemra to treat patients with severe COVID-19. According to ClinicalTrials.gov, the primary completion date is June 30.
At least 11 clinical trials of Actemra have primary completion dates by the end of summer, making it one of the most popular immune approaches.
Figure: Tocilizumab tally
In contrast to the Actemra findings, Regeneron Pharmaceuticals Inc. (NASDAQ:REGN) reported that IL-6 inhibitor Kevzara sarilumab failed to improve survival or the need for ventilation at day 14 in severe and critically ill patients in a Phase II trial. The biotech discontinued treatment of severe patients in its ongoing Phase III trial, but is continuing to treat critical patients who showed a trend toward improvement in Phase II.
Hints of efficacy have been seen in small, controlled trials of two other immunosuppressants: Arava leflunomide, a DHODH inhibitor that Sanofi (Euronext:SAN; NASDAQ:SNY) markets for rheumatoid arthritis; and mavrilimumab, an investigational GM-CSF inhibitor that CSL Ltd. (ASX:CSL) is developing for inflammatory disorders.
In investigator-led studies, leflunomide increased the number of subjects who were virus negative and the hospital discharge rate at day 14; mavrilimumab increased the number of subjects with clinical improvement at day 28. Both studies suffered from a small N size and lack of placebo (both used SOC as the comparator), but they suggest follow-up studies are warranted.
University of Chicago has leflunomide in a Phase I trial to treat COVID-19, and Immunic AG and Panoptes Pharma GmbH each have a DHODH inhibitor in Phase II testing in the indication.
Kiniksa Pharmaceuticals Ltd. (NASDAQ:KNSA) is running a Phase II study of mavrilimumab in COVID-19, and five other companies including GlaxoSmithKline plc (LSE:GSK; NYSE:GSK) have mAbs against GM-CSF in the clinic.
Ultimately, immune boosters may prove useful early in the disease along with antivirals, while late disease may benefit from immunosuppressants or combinations of different types of immune modulators that can rebalance the immune response by inhibiting some cell types and molecular pathways and boosting others.
Figure: Vaccines start to read out
Vaccines start to yield results
At least 12 vaccines designed to prevent infection with COVID-19 have made it to the clinic, as well as a handful repurposed vaccines. So far, companies have reported early data from three.
Sinovac Biotech Ltd. (NASDAQ:SVA) announced June 13 that over 90% of the 600 volunteers in the Phase II portion of a Phase I/II trial generated neutralizing antibodies (NAb) against SARS-CoV-2 at 14 days after prime-boost administration of CoronaVac.
The study is the most notable because of its size; its rate is on par with the eight subjects in the 45-subject study reported by Moderna Inc. (NASDAQ:MRNA) in May, and higher than the 75% rate of the high dose in the 108-subject study from CanSino Biologics Inc. (HKEX:6185) the same month.
Of the three, only CanSino disclosed neutralizing antibody titers, which -- at 34 -- were lower than FDA’s 160 guideline for neutralizing antibodies in convalescent plasma. Although Moderna did not report titers, it said mRNA-1273 achieved titers similar to convalescent plasma.
CanSino’s titer data, and its T cell response data, fell into bimodal distributions based on whether the subjects had pre-existing antibodies against the adenovirus vector used in Ad5-nCoV. About half of the study population had such antibodies, suggesting vector-directed antibodies could be a substantial issue for the vaccine (see “Making Sense of Early COVID-19 Vaccine Data”).
Xuefeng Yu, co-founder, chairman and CEO of CanSino Biologics Inc. (HKEX:6185), told BioCentury on a webinar that use of a higher dose or two doses may get around the weaker immune responses seen in subjects with pre-existing antibodies (see “Preparing for COVID-19 Vaccine Success”)
At least three other companies -- AstraZeneca plc (LSE:AZN; NYSE:AZN), Inovio Pharmaceuticals Inc. (NASDAQ:INO) and Novavax Inc. (NASDAQ:NVAX) -- have said publicly that they expect early vaccine data this summer.
Novavax reported on June 9 neutralization titers in non-human primates for its NVX CoV2373 subunit vaccine that were at least one order of magnitude higher than three other vaccines, including Sinovac’s CoronaVac and AstraZeneca’s AZD1222 (see “Novavax Announces High Titers” and “Daily Chart”).