Medications repurposed to treat COVID-19 varied based on ‘need to do something’
A study assessing inpatient use of medications repurposed to treat patients with COVID-19 including hydroxychloroquine, corticosteroids and tocilizumab demonstrated quick prescribing changes in response to emerging reports.
“Early on in the pandemic, clinicians were eager to help their sick COVID-19 patients but had almost no evidence to guide their actions,” Sameer S. Kadri, MD, MS, FIDSA, head of clinical epidemiology Section in the NIH’s Critical Care Medicine Department, told Healio.
Credit: Adobe Stock
As a result, medications in use for other conditions — for example, hydroxychloroquine for rheumatologic conditions, azithromycin for bacterial infections, and corticosteroids and tocilizumab for their anti-inflammatory and immunomodulatory effects —"were repurposed for COVID-19 based on questionable signals from in vitro studies, extrapolations from analogous clinical scenarios like the flu, mass media influences, emergency use authorization and peer pressure,” Kadri said.
“The need to do something for deteriorating patients bred considerable treatment empiricism; instead of waiting on results of large randomized clinical trials (RCTs), clinicians opted to use these agents anyway out of sheer desperation,” he said.
To help understand how and when these repurposed agents were being used against COVID-19 in hospitals — and how, when and why use changed — Kadri and colleagues studied the patterns of off-label use of medications among patients hospitalized with COVID-19 in one of 318 U.S. hospitals between March and May 2020.
In total, 35,259 inpatients with COVID-19 coding were admitted to these hospitals, of whom 5,950 (16.9%) received mechanical ventilation. According to the study, medication use varied across hospitals, although overall, 16,164 (45.8%) patients received hydroxychloroquine at 302 hospitals, 18,164 (51.5%) received azithromycin at 311 hospitals, 7,570 (21.5%) received corticosteroids at 299 hospitals and 2,005 (5.7%) received tocilizumab across 188 hospitals.
They found that the use of hydroxychloroquine and azithromycin was higher for mechanically ventilated patients compared with nonventilated patients, although the relative difference was much greater for corticosteroids and tocilizumab. Additionally, compared with nonventilated patients with COVID-19, mechanically ventilated patients with COVID-19 were threefold more likely to receive corticosteroids (22% vs. 61.8%) and sixfold more likely to received tocilizumab (3% vs. 18.9%). The study also showed that the use of corticosteroids among mechanically ventilated patients with COVID-19 (62%) was comparable to its use among pre-COVID-19, mechanically ventilated influenza patients (68%).
Further assessment of the repurposed drugs showed that inpatient hydroxychloroquine use declined by 80% around the time reports about potential cardiac arrhythmias associated with its emerged, whereas corticosteroids and tocilizumab were initiated 2 days earlier in May vs. March 2020. Additionally, two-thirds of ventilated patients with COVID-19 were already receiving corticosteroids from March to May, resembling pre-COVID use in mechanically ventilated influenza patients.
According to Kadri, corticosteroids “save lives in COVID-19,” especially among patients needing oxygen and mechanical ventilation. He said hydroxychloroquine is not helpful to patients with COVID-19 — a fact that has been “hammered in” by multiple RCTs.
Kadri added that the efficacy of tocilizumab remains unclear, with mixed signals of potential benefit and lack of benefit across RCTs, and remaining concerns about the risk for secondary infections due to its immune suppressive effects, whereas secondary infection risk has been a concern with long corticosteroids courses for years.
The “sweet spot” in terms of duration of corticosteroid therapy that provides an optimal and durable effect in COVID-19 respiratory failure while minimizing risk for secondary infections is still being worked out, according to Kadri.
“Although our findings are historical, they provide a cautionary tale and offer important lessons for the future,” Kadri said. “During scenarios such as deadly viral pandemics, usual care can be easily influenced by suboptimal evidence, mass media, anecdotal experience coupled with need to ‘do something.’ Fortunately, inpatient clinicians appear to be quick in changing prescribing behavior in response to signals of harm.”
Moving forward, multiple stakeholders, including clinicians, trialists, regulators and guideline panelists, will need to weigh earlier adoption of candidate treatments against the potential for direct harms, which is often unknown at the time of adoption, Kadri said. Additionally, he said the urge to quickly bring out therapies before they are vetted in RCTs can cause fatal delays in the systematic generation of definitive evidence as seen with convalescent plasma.
“Ongoing appraisal of real-world use of candidate therapies for COVID-19 can tell us if practice guidelines are being followed,” Kadri said. “They may also detect changes in usual care patterns, which are important to incorporate into future RCTs that intend on comparing novel COVID-19 therapies to a real-world standard of care.”
Collapse
Read more about