The RECOVERY trial just released a preprint demonstrating benefit from tocilizumab in COVID-19. As with the prior RECOVERY studies, this is a multicenter, open-label, pragmatic trial. A robust mortality benefit was detected (the fragility index calculates to 17, which isn’t too shabby). Several important secondary endpoints were positive as well (e.g., reduced intubation rate, reduced requirement for dialysis):
how can we reconcile this study with prior studies?
Early studies involving tocilizumab monotherapy were negative. Over time, as more patients in the studies received steroids, tocilizumab began to seem more effective. In the most recent trial (REMAP-CAP), steroid use increased to 88% and tocilizumab was found to improve mortality.
As discussed earlier here, this suggests that tocilizumab may work only in combination with steroid. Tocilizumab inhibits only one cytokine, which may not be sufficient to work amid a storm of numerous cytokines. Steroid affects more inflammatory pathways, allowing it to work as monotherapy against COVID. Nonetheless, it’s plausible that adding tocilizumab to 6 mg of dexamethasone could improve the overall potency.
82% of patients in the RECOVERY trial received steroid, since this study was largely performed following the publication of positive results regarding dexamethasone. Tocilizumab caused a 6% reduction in mortality when combined with dexamethasone, but had no impact on mortality when given alone (red box, figure below). The difference is statistically significant (p=0.01).
This fits perfectly with prior evidence suggesting that tocilizumab works only in combination with steroid. Perhaps most tantalizing, it suggests that the actual effect size of tocilizumab when combined with dexamethasone is larger than reported in the main study results (e.g., absolute mortality benefit of 6%, rather than 4%).
who should receive tocilizumab?
The RECOVERY trial included hypoxemic patients admitted to the hospital with COVID who had a C-Reactive Protein (CRP) >75 mg/L. Benefits were seen across all patient subgroups (figure above). Additionally, there was no difference on benefit when comparing patients treated with tocilizumab within <2 days of hospital admission versus >2 days after hospital admission.
Targeting an immunomodulator towards patients with elevated inflammatory markers makes biological sense. The REMAP-CAP results are consistent with this concept, showing greatest benefits among patients with the highest tercile of CRP levels:
In contrast, there was no dependence on timing of tocilizumab. This may reflect that the precise timing of hospitalization or ICU admission in COVID is often arbitrary (e.g., depending on variable ICU transfer criteria, or when patients with asymptomatic hypoxemia choose to present to the hospital). Furthermore, the requirement that the CRP be >75 mg/L itself inherently selects for patients experiencing an inflammatory phase of their illness – so additional selection criteria based on timing of hospitalization or ICU admission may be unnecessary.
Some additional aspects of patient selection bear mention. Patients with active non-COVID infection were excluded. Additionally, in some cases tocilizumab was not given because the managing physician felt that tocilizumab was contraindicated. Thus, an additional layer of clinical judgement is warranted (rather than administering tocilizumab to every single patient with COVID and hypoxemia).
(It also bears mention that some patients develop bacterial or fungal superinfection leading to re-elevation of CRP some weeks into their hospital course. Such patients won’t benefit from tocilizumab.)
what implications does this have for early hospitalization of COVID patients?
We now have two therapies for COVID-19 – steroid and tocilizumab. Patients with hypoxemia (room air saturation <92%) should generally be treated with dexamethasone. Those who also have a CRP >75 mg/L additionally qualify for tocilizumab (roughly half of admitted patients, based on the ISARIC4C database!). Timely initiation of these therapies may reduce mortality and also avoid intubation (offering benefits to the patient and also to the medical system as a whole – since ICU beds and ventilators are often a scarce resource).
The existence of effective therapies for early COVID emphasizes the importance of identifying and hospitalizing patients early in the course of the illness (e.g., using home pulse oximetry to detect early deterioration). This also highlights the importance of accurate determination of oximetry among patients with darker skin pigmentation – potentially with the use of arterial blood gas analysis if pulse oximetry results are borderline.
what if we don’t have access to tocilizumab?
A paradox of pandemic medicine is that an effective medication will often immediately be used up. Tocilizumab is no exception to this. As an expensive monoclonal antibody, tocilizumab is available at only a limited number of hospitals. It may be difficult to rapidly scale up production of tocilizumab, so medication shortages are likely. Finally, tocilizumab is expensive, so many hospitals may be unable to afford it. Thus, for a variety of reasons, tocilizumab may not be a treatment option for most patients infected with COVID throughout the world.
Fortunately, tocilizumab isn’t the only immunomodulator available:
- JAK-1/2 inhibitors (especially baricitinib) inhibit several cytokines, including IL-6. Baricitinib has been shown to be effective in COVID-19, although the combination of steroid plus baricitinib hasn’t been prospectively investigated (more on this here).1–4
- Steroid itself will suppress IL-6 and CRP levels (albeit in a less targeted fashion than tocilizumab). Several studies have suggested that higher doses of steroid are safe and effective in ARDS generally and COVID specifically (e.g., dexamethasone 20 mg/day or methylprednisolone ~125 mg/day).5–7
Thus, if tocilizumab isn’t effective, either baricitinib or higher doses of steroid could be a rational therapy for patients with COVID and systemic inflammation. Unfortunately, the evidentiary basis for these choices is currently less robust than for tocilizumab, so they may not be a preferred treatment. One possible approach is as follows:
- Among hypoxemic COVID patients with CRP >75 mg/L, adding tocilizumab on top of dexamethasone 6 mg/day improves mortality and other important endpoints (e.g., reductions in intubation and hemodialysis).
- Evidence suggests that tocilizumab monotherapy is ineffective, but tocilizumab provides benefit when added to steroid. This may explain the success of RECOVERY and REMAP-CAP, in comparison to prior studies of tocilizumab which didn’t detect benefit.
- The key factor in selecting patients who may benefit from tocilizumab seems to be the presence of systemic inflammation (CRP >75 mg/L), rather than illness severity or timing.
- If tocilizumab is unavailable, other immunomodulators may be considered in its place. Higher doses of steroid, or JAK-inhibitors combined with steroid are supported by some evidence in COVID and might be rational therapies here (albeit with less robust evidentiary support).
- 6 mg dexamethasone daily is inadequate for sicker patients with COVID pneumonia.
Image credits: Photo by Gabriel Crismariu on Unsplash
references
- 1.Rodriguez-Garcia J, Sanchez-Nievas G, Arevalo-Serrano J, Garcia-Gomez C, Jimenez-Vizuete J, Martinez-Alfaro E. Baricitinib improves respiratory function in patients treated with corticosteroids for SARS-CoV-2 pneumonia: an observational cohort study. Rheumatology (Oxford). 2021;60(1):399-407. doi:10.1093/rheumatology/keaa587
- 2.Kalil A, Patterson T, Mehta A, et al. Baricitinib plus Remdesivir for Hospitalized Adults with Covid-19. N Engl J Med. Published online December 11, 2020. doi:10.1056/NEJMoa2031994
- 3.Cantini F, Niccoli L, Nannini C, et al. Beneficial impact of Baricitinib in COVID-19 moderate pneumonia; multicentre study. J Infect. 2020;81(4):647-679. doi:10.1016/j.jinf.2020.06.052
- 4.Stebbing J, Sánchez N, Falcone M, et al. JAK inhibition reduces SARS-CoV-2 liver infectivity and modulates inflammatory responses to reduce morbidity and mortality. Sci Adv. 2021;7(1). doi:10.1126/sciadv.abe4724
- 5.Villar J, Ferrando C, Martínez D, et al. Dexamethasone treatment for the acute respiratory distress syndrome: a multicentre, randomised controlled trial. Lancet Respir Med. 2020;8(3):267-276. doi:10.1016/S2213-2600(19)30417-5
- 6.Tomazini B, Maia I, Cavalcanti A, et al. Effect of Dexamethasone on Days Alive and Ventilator-Free in Patients With Moderate or Severe Acute Respiratory Distress Syndrome and COVID-19: The CoDEX Randomized Clinical Trial. JAMA. 2020;324(13):1307-1316. doi:10.1001/jama.2020.17021
- 7.Edalatifard M, Akhtari M, Salehi M, et al. Intravenous methylprednisolone pulse as a treatment for hospitalised severe COVID-19 patients: results from a randomised controlled clinical trial. Eur Respir J. 2020;56(6). doi:10.1183/13993003.02808-2020
- PulmCrit Blogitorial – Use of ECGs for management of (sub)massive PE - March 24, 2024
- PulmCrit Wee: Propofol induced eyelid opening apraxia – the struggle is real - March 20, 2024
- PulmCrit wee: Why I like central lines for GI bleed resuscitation - March 13, 2024
Does thay mean Toci + steroids should/will work in all cytokine storm states?
Probably not. For one thing, we currently lack any clear definition of exactly what a cytokine storm is (at least to my knowledge). There has been a ton of argument about whether COVID-19 is or isn’t a “cytokine storm” but this is a very loosely used terminology with no clear definition. For example, defining this solely on the basis of circulating cytokine levels in the plasma is probably missing a *lot* of the picture (e.g. it misses paracrine effects of cytokines within the tissue and also the appropriateness or inappropriateness of those cytokine levels in the context of the clinical… Read more »
If “Toci” works, we just need to enroll 2000 patients in a standard old RCT without Bayesian methods and prove it the old-fashioned way. There are way way too many researcher degrees of freedom in a Bayesian trial, not only the selection of the prior probability value but its distribution. I am unmoved by REMAP-COVID. Great way to find signal, but I don’t think it “confirms” the signal.
I shared your reservations after publication of the REMAP-CAP trial, and in my last post on tocilizumab I concluded that the RECOVERY trial would be needed to validate the results of REMAP-CAP. And now RECOVERY has done *exactly* that. So we now have *two* large, independent, multi-center trials showing the same thing (RECOVERY & REMAP-CAP). For me, that’s sufficient evidence to change practice. Skepticism and rigor in medicine are good, but beyond a certain point, if we demand an unrealistic burden of proof then we’re simply therapeutically paralyzed. If you want to design and run a 2000-patient placebo-controlled double-blind trial… Read more »
“Bayesian method” is great, how could not be?
Does the value of interleukin 6 direct the use of tocilizumab?
Most hospitals lack the ability to measure IL-6 levels rapidly and reliably, so it doesn’t seem to play a role in actual clinical management. CRP functions as a surrogate for IL6 levels.
Is there a possibility that dexa 6mg + tocilizumab is just a more expensive version of higher dose dexa (20mg)? I know some trials have shown higher dose is usually not helpful, but the patients were not selected by CRP levels.
Yes, I think that’s entirely possible.
Please look at the mortality rates inside your red rectangle, in the Usual Care arm, corticosteroids vs norticosteroids and tell me if that makes any sense: a reduction from 34.6% to 32.8% (which, by my calculations, doesn’t reach statistical significance) in a group of patients for which their previous trial arm, with dexamethasone vs no dexamethasone, had a mortality RR of 0.64-0.82.
This headline is misleading. Rather, too many questions were raised about this trial (as above) to come to any definitive conclusions about toci. The overall effect was small, barely statistically significant. Not RCT, but open-label study. One other obvious question: why no benefit in women?
“Thus, if tocilizumab isn’t effective, “. Did you mean, if tocilizumab isn’t available, go straight to dex 20 mg daily for this population, or do you mean if a patient worsens on tocilizumab, try dex 20 mg daily?
Just curious, what publication are you citing as evidence the assertion that “6mg daily of dexamethasone is inadequate for sicker patients with COVID pneumonia”? RECOVERY trial used that dose for ventilated patients and that study was the EBM basis for using steroids in COVID. Also, where are does the potential suggested dexamethasone 20mg daily dose come from? CODEX trial evaluated ARDS and didn’t use a dose that high, but still found clinical benefit. In regards to toci, shouldn’t we maintain some healthy skepticism if 1) monotherapy continuously fails to demonstrate benefit for mortality, 2) studies that show benefit include composite… Read more »
Re-looking at RECOVERY trial results for toci; does anyone find it odd that in figure 3 (forest plot) that breaks down mortality by baseline O2 need, that none of the cohorts were statistically different between arms, but overall there’s a mortality benefit? Then the takeaway from the secondary and subsidiary outcome analyses (table 2) is that if you combine cohorts of patients not on IMV at baseline, there’s a mortality benefit for toci arm. However, in those starting on lowest O2 cohort, there’s no benefit from toci in terms of progression to further ventilation (including HFNC or IMV). This might… Read more »
Patient with COVID-19, initial CRP 81, admitted on NC 4 liters, then deteriorates. On day #2 is requiring HFNC 60L/min 70% Fio2, clinically stable, able to self prone, looks comfortable. Now CRP is 45 So, is toci still indicated? CXR showing some progression of the infiltrates. No evidence of vol up (Vexus 0).
How long does a Covid-19 cytokine storm last? As the inflammatory markers of hospital-acquired infection are similar to a CRS when does or when should we think that C-19 CRS should be stopped by now and a hospital-acquired infection has started?