Colchicine as COVID-19 treatment suffers difficult year, but some hope remains
Click Here to Manage Email Alerts
If colchicine and its prospects as a COVID-19 treatment entered 2021 with a bang, then it decidedly exited the year with a whimper.
In January 2021, researchers with the COLCORONA trial rang in the new year with promising news — colchicine, with its established safety and inexpensive cost, showed promising, if mixed, benefits as treatment for COVID-19. However, hopes that the anti-inflammatory could prove effective in reducing mortality and hospitalization, buoyed as they were by COLCORONA, soon found themselves in rough and unpredictable seas.
Trials examining colchicine’s efficacy in patients with COVID-19 have repeatedly found themselves halted or stopped early. Meanwhile, several that were completed have shown little to no efficacy at all. In December, these data culminated in a meta-analysis that found adding colchicine to current standard of care COVID-19 treatment was not associated with any benefit, and potentially associated with more adverse events.
Is this the final word on colchicine and COVID-19, or is there still hope for the drug?
Understanding the Rationale
“The benefits of immunosuppressive drugs have been one of the main surprises in COVID-19 therapy,” Leon Peto, MD, RECOVERY trial clinical lecturer and consultant in infectious diseases at the Nuffield Department of Population Health at the University of Oxford, told Healio Rheumatology.
Peto noted the rheumatology armamentarium that underwent investigation in COVID-19, from corticosteroids to interleukin-6 inhibitors. “However, mortality for hospitalized patients remains high despite the use of corticosteroids and anti-IL-6 drugs, so there is continued interest in other immunosuppressive drugs with distinct mechanisms action such as colchicine.”
Tuhina Neogi, MD, PhD, chief of rheumatology and Alan S. Cohen Professor of rheumatology at Boston University School of Medicine/Boston Medical Center, dug deeper into the mechanism. “The major consequence of COVID-19 infection that leads to morbidity and mortality is due to a hyperinflammatory response to the virus that can damage organs such as the lungs,” she said in an interview. “A number of inflammatory cytokines are responsible for this hyperinflammation, some of which are released by activation of the NLRP3 inflammasome. Colchicine acts as an anti-inflammatory in part through its effects on the inflammasome.”
It is for these reasons that clinicians began treating COVID-19 with colchicine in the first place. As the numbers grew, data emerged.
Early Signs from COLCORONA
Findings from COLCORONA were published by Tardif and colleagues in The Lancet, including 2,235 patients with COVID-19 who were randomly assigned colchicine and 2,253 who received placebo. Mortality or hospitalization due to COVID-19 within 30 days of randomization served as the primary endpoint.
Results showed that 4.7% of patients in the colchicine arm and 5.8% of those in the placebo arm reached the primary endpoint (OR = 0.79; P = .081). Similarly, findings for 4,159 patients with PCR-confirmed COVID-19 showed that 4.6% of 2,075 patients in the colchicine group and 6% of 2,084 patients in the placebo group died or were hospitalized from the virus (OR = 0.75; P = .042).
“In community-treated patients, including those without a mandatory diagnostic test, the effect of colchicine on COVID-19-related clinical events was not statistically significant,” the researchers concluded.
“The challenge with any COVID-19 therapeutic trial has been in the study design, including inclusion criteria, dosing regimen, outcome measures and timing of primary endpoint, given the newness of the disease and rapidly evolving understanding as new data pours in,” Neogi said. “In COLCORONA, the primary composite endpoint, components of the primary endpoint, and secondary endpoint were all in a consistent direction, though they did not meet statistical significance; several effect estimates were close to significance.”
It is important to note that when the analysis was limited to patients with PCR-confirmed infection, the effect estimates remained similar, but the results were more precise, resulting in statistically significant findings, according to Neogi. “The authors noted that the patients who had PCR-confirmed infections had a higher event rate than those without PCR confirmation,” she said. “So I would not say these were mixed results, but rather the effect estimates were consistent across analyses, and there were issues of precision and potential differences in risk of outcomes in individuals enrolled without PCR-confirmed infection.”
Peto offered a similar assessment but noted that the trial had to stop recruiting before it could definitively answer the question it asked. “I remain skeptical about any effects of colchicine in early COVID, as the treatments that have been found to work in early infection are antivirals, whereas the main effect of anti-inflammatories that work seems to be later in infection, in patients who are hospitalized,” he said.
Despite these findings, Peto suggested that more information may ultimately prove the utility of colchicine in COVID-19. “If the effect were real, it could have a significant public health benefit, so it would be nice to see a larger trial along the lines of COLCORONA that could establish whether or not there is any role for colchicine in early infection,” he said.
It seems likely that such a data set could be forthcoming, given the flood of studies being conducted in COVID-19. However, the RECOVERY trial results failed to rally hope in colchicine as an effective treatment.
No RECOVERY
The randomized, controlled, open-label RECOVERY trial included 11,340 patients who were eligible to receive colchicine, 94% of whom were receiving corticosteroids as part of routine care. Researchers assigned 49% of that population to the study drug and 51% to continue with standard treatment regimens.
The 28-day mortality rate was 21% for both colchicine and controls (rate ratio = 1.01; P = .77). Similarly, hospital discharge rates at 28 days were 70% for both study arms (RR = 0.98; P = .44). Moreover, rates for the composite endpoint of invasive mechanical ventilation or death were 25% in both arms (RR = 1.02; P = .47).
“The RECOVERY trial results were very disappointing for colchicine and have really ruled out any worthwhile benefit of colchicine by the time people are admitted to hospital with COVID-19, at least in patients who are already receiving corticosteroids,” Peto said. “However, we know that some treatments can work in early infection but be less effective by the time patients are admitted to hospital.”
For Neogi, ongoing attention to clinical trial design is critical to getting accurate answers for any potential COVID-19 treatment. “Any trial results are interpretable and applicable only to the people that were studied,” she said. “Thus, careful consideration of appropriate inclusion criteria, dosing, endpoints and duration of study are needed.”
One month following the publication of the RECOVERY trial, another study emerged. This time, it was a meta-analysis seeking to pool all of the available data that had been published up to that point, and evaluate the efficacy and safety of colchicine in COVID-19 across multiple studies.
Enter the Meta-analysis
In the meta-analysis published in RMD Open, Mehta and colleagues assessed 69 full texts and six studies comprising 16,148 patients with COVID-19. Results showed that colchicine failed to reduce a number of outcomes, including mortality risk (risk difference = –0.00; 95% CI, –0.01 to 0.01) ventilatory support (risk ratio = 0.67; 95% CI, 0.38-1.21), ICU admission (risk ratio = 0.49; 95% CI, 0.19-1.25), length of hospital stay (mean difference = –1.17; 95% CI, –3.02 to 0.67) and serious adverse events (risk difference = –0.01; 95% CI, –0.02 to 0), compared with patients who received supportive care only.
For Neogi, the same issues with study design as those she noted for COLCORONA apply in the interpretation of these findings. “The meta-analysis included 6 trials that had different inclusion criteria, different severities of disease, different doses and different lengths of follow-up,” she said. “Four of the trials were open label. Five were in hospitalized patients, whereas one — COLCORONA — was in outpatients. Thus, there are numerous differences across the trials.”
Despite these concerns, there may be reasons for continued hope for additional studies examining colchicine as a COVID-19 therapy, according to Neogi, namely the sheer numbers of patients who are eligible for analysis.
“With much more data available now than at the beginning of the pandemic, sample size calculations based on contemporary expected event rates will also help ensure appropriate power,” she said. “A challenge early in COVID-19 trials was that the standard of supportive care improved quite quickly, such that event rates in trials were lower than originally anticipated, and thus some trials ended up being unable to detect differences.”
There are also more practical reasons to for ongoing faith in colchicine. “Given its low toxicity profile and low cost, and some suggestive findings, it is possible that it may be considered for use by some, but FDA has not issued an EUA for colchicine based on the currently available data,” she said.
However, while further study may elucidate a role for colchicine, Neogi stressed that the drug does not top the list of rheumatology drugs in the COVID-19 setting at the current moment. “There are other immunomodulating agents that do have EUA for COVID-19 that would therefore be preferred,” she said.
For more information:
Tuhina Neogi, MD, PhD, can be reached at 650 Albany Street, Suite X200, Boston, MA, 02118; email: tneogi@bu.edu.
Leon Peto, MD, can be reached at the Nuffield Department of Population Health, Richard Doll Building, Old Road Campus, Oxford, OX3 7LF; email: leon.peto@ndph.ox.ac.uk.
Collapse
Mehta KG, et al. RMD Open. 2021;doi:10.1136/rmdopen-2021-001746.
RECOVERY Collaborative Group. Lancet. 2021;doi:10.1016/S2213-2600(21)00435-5.
Tardif JC, et al. Lancet. 2021;doi:10.1016/S2213-2600(21)00222-8.
Click Here to Manage Email Alerts