Overview Therapeutics - Updated: June 5, 2020

Therapeutics summary

1. The anti-viral and anti-inflammatory section is meant to provide a summary of

the literature. The BWH Infectious Diseases COVID-19 treatment guidelines
(Partners login required) and ID consultation service take precedence over the
information provided in the literature review below. This table is from those
treatment guidelines, which mirror our recommendations laid out in this chapter,
but in a more concise format and with BWH-specific contacts.

Clinical Trials

1. There are a variety of clinical trials ongoing at BWH. Please reference the above
ID treatment guidelines for further information (Partners login required)
2. For clinical trial enrollment, the contact person for the trial can be paged or
emailed to discuss further. Preferred methods of contact for each trial can be
found here (Partners login required)
3. Enrollment criteria for each trial can found here (Partners login required)
4. If a patient is enrolled in a COVID-19 clinical trial, verify that other therapeutic
regimens do not add harmful drug interactions with study agents
Nebulization
1. Nebulization is considered an aerosol generating procedure and may contribute
to disease transmission.

1. Nebulization requires appropriate PPE (e.g., N95) and room (e.g., negative
airflow)
2. Laboratory studies on human patient simulators showed increased dispersion of
particles during jet nebulizations at 6L/min from 0.45 to >0.8m when simulating
normal lung or severe lung injury, respectively. In comparison, there was 0.4m
dispersion with 5L/min nasal cannula and 0.3m with Venturi mask (40%) Hui et
al. Hong Kong Medical Journal. 2020). However, a meta-analysis of aerosol-
generating procedures did not find nebulizer as a risk factor for SARS
transmission, unlike procedures like tracheal intubation (Tran et al. PLOS One.
2020).

Bronchodilator therapy
1. COVID-19 clinical reports do not indicate wheeze as a common symptom, and
not all patients require bronchodilators. Bronchodilators should certainly be
prescribed whenever indicated but should not be ordered as a default on every
patient (Zhou et al, Lancet, 2020; Yang et al, Lancet Respir Med, 2020; Guan et
al, N Engl J Med, 2020; WHO, COVID-19 Interim guidance, March 2020).

Non-intubated patients

1. If the patient is COVID-19 confirmed or Person Under Investigation (PUI):

1. If possible, use metered dose inhalers (MDIs) + spacer rather than nebulizers.
Nebulizers are an aerosol generating procedure.
2. Because MDI supply is limited, only prescribe when needed. Ask patients /
families to bring in their home inhalers if possible and check in home MDI with
pharmacy.
3. However, if the patient requires a nebulizer (e.g., difficulty using MDI),
nebulizers should be used in COVID-19 confirmed or PUI patients.

1. For Covid-19 confirmed or PUI patients, consider a breath actuated nebulizer

(BAN), which may help reduce aerosol generation.
2. Order breath actuated nebulizer (BAN) as free text in the nebulizer order.
3. Note: The patient needs to generate flow to trigger the neb; so patients with
upper airway edema/stridor, weakness or inability to cooperate may be poor
candidates for a breath actuated nebulizer (BAN).
4. If a COVID-19 confirmed or PUI patient requires nebulizers and cannot use
BAN, then a regular nebulizer should be ordered.

2. If the COVID-19 PUI patient is ruled out and considered COVID-19 negative:

1. The patient should use their previously prescribed MDI until it runs out, then
switch to nebulizers.
 3. In patients admitted WITHOUT suspicion for COVID-19:

1. Use nebulizers even if on droplet precautions (e.g., influenza) because MDI

supply is limited.

Intubated patients

1. For intubated COVID-19 confirmed or PUI, it is approved to use nebulizers.

1. The in-line nebulizer container is part of a closed ventilator circuit.

Airway Clearance
Secretion thinning

1. Patients can develop thick secretions from Covid-19 itself or secondary bacterial
pneumonia.
2. Nebulized treatments may help with airway secretion management, but
published evidence is not available.

1. Options include:

1. Normal (0.9%) saline nebulizer BID.

2. N-acetylcysteine (“Mucomyst”) nebulizer BID or TID

1. N-acetylcysteine can cause bronchoconstriction.

2. Pre-treat with albuterol 2.5mg just prior to delivery.

3. Nebulized hypertonic (3-7%) saline once daily

1. Hypertonic saline can cause bronchoconstriction.

2. If using, start with 3% saline to assess response and bronchoconstriction.
3. Pre-treat with albuterol 2.5mg just prior to delivery.

4. Dornase alfa 2.5mg nebulizer once daily

1. Dornase can cause bronchoconstriction, mucosal bleeding, and can clog the
HEPA filter, requiring intermittent replacement by RT.
2. Avoid in the setting of bloody secretions.
3. Pre-treat with albuterol 2.5mg just prior to delivery
4. It would be reasonable to consider other agents, including N-acetylcysteine, first
given the need to change HEPA filters. In addition, a RCT for dornase nebulizer
versus saline will begin shortly at BWH. However, if persistent secretions, it is
reasonable to try dornase nebulizer.

5. Although avoided if possible since it is an aerosol generating procedure,

bronchoscopy for pulmonary toilet can be performed if needed on Covid-19
confirmed or PUI patients.
Mechanical airway clearance

1. Covid-confirmed or PUI patients can utilize the below methods of mechanical

airway clearance if needed. These are aerosol generating procedures (AGP) and
require appropriate PPE (e.g., N95) and a negative airflow room:

1. Oscillating positive expiratory pressure devices (Aerobika or Acapella).

Encourage the patient to use the device when staff are outside of the room with
the door closed.
2. Chest PT can be performed, but is an aerosolizing procedure requiring close
contact with staff, so alternate options should be considered first where possible
3. Chest PT vests if the patient requires at home (e.g., CF patients) should be
continued

2. For Covid-confirmed or PUI patients, the use of cough assist devices (e.g., MIE:
Mechanical Insufflation Exsufflation) must be approved by Respiratory Therapy
leadership prior to initiation. Approval is not required for Covid-negative
patients:

1. Cough assist devices can generate significant aerosols and require staff to be
present in the room
2. However, certain situations may merit discussion of the use of cough assist
device, such as a patient with ALS that use this device at home

Systemic Corticosteroids
Evidence

1. Data on corticosteroids for COVID-19 is mixed.

1. Many studies show negative effects of corticosteroids on similar viruses. There

is no clinical evidence of net benefit from steroids in SARS-CoV, MERS-CoV
or influenza infection, and observational data show increased mortality, more
secondary infections, impaired viral clearance and more adverse effects in
survivors (e.g., psychosis, diabetes, avascular necrosis) (Lee et al, J Clin Virol,
2004; Stockman et al, PLoS Med, 2006; Lansbury et al, Crit Care Med 2019;
Arabi et al, Am J Respir Crit Care Med, 2018; WHO, COVID-19 Interim
guidance, March 2020).
2. However, a retrospective cohort analysis of patients with COVID-19 who
developed ARDS (n=84) noted that methylprednisolone treatment was
associated with a decreased risk of death (Wu et al, JAMA Int Med, 2020).
3. An earlier, non-blinded randomized controlled trial of patients with ARDS (not
COVID-19) suggested a benefit to dexamethasone treatment (Villar et a, Lancet
Resp Med, 2020) but this has not been replicated as of yet.
Recommendations

1. At this time, we do not recommend steroids for COVID-19 except as part of

a clinical trial or if treating another indication such as asthma or COPD
exacerbation. This is in line with WHO guidance (WHO COVID-19 Interim
Guidance, March 2020). The Society for Critical Care Medicine makes a weak
recommendation for the use of steroids in COVID+ ARDS and the American
Thoracic Society makes no recommendation for or against the use of steroids,
but notes that their task force came close to recommending against steroids
(SCCM COVID-19 Guidelines, March 2020; ATS COVID-19 Interim Guidance,
April 2020).
2. If treating another indication, use corticosteroids at a low dose for a short
duration:

1. For asthma or COPD exacerbation, treat with 40mg prednisone PO or 30mg

methylprednisolone IV, once daily x 3-5 days.
2. For any shock with a history of chronic steroid use in excess of 10mg
prednisone daily, treat with 50mg hydrocortisone IV Q6H until improvement in
shock.
3. For multipressor (>2 pressors) shock without history of chronic steroid use,
treat with 50mg hydrocortisone IV Q6H until improvement in shock.

Pulmonary Vasodilators
1. Please see the Respiratory chapter under “refractory hypoxemia”.

Remdesivir
Pharmacology

1. Remdesivir is a nucleotide prodrug metabolized to an analog of adenosine

triphosphate, which inhibits viral RNA-dependent RNA polymerase, causing
premature termination of RNA transcription.

Evidence

1. Animal models have shown reduced lung viral loads when remdesivir is used for
both SARS-CoV-1 and MERS-CoV (Sheahan et al, Sci Transl Med, 2017;
Sheahan et al, Nat Commun, 2020; de Wit et al, Proc Nat Acad Sci, 2020)
2. For the treatment of Ebola, remdesivir did not show favorable outcomes
compared to other investigational agents (MAb114 and REGN-EB3) in a
randomized controlled trial (Mulangu et al, N Engl J Med, 2020)
3. Remdesivir has shown in vitro activity against SARS-CoV-2 (Wang et al, Cell
Res, 2020)
4. A case report was published on the use of remdesivir in a 35-year-old male who
improved one day after remdesivir was initiated, but it is unclear if the use of
remdesivir resulted in this improvement (Holshue et al, N Engl J Med, 2020)
5. A case series of compassionate use remdesivir analyzed 53 patients, of whom
68% had improvement in their oxygen-support class, 47% were discharged, and
 13% passed away. Without a control group, it is unclear if the use of remdesivir
altered the natural progression of COVID-19 disease in these patients (Grein et
al, N Engl J Med, 2020)
6. A randomized, open-label phase 3 trial assessed treatment outcomes with
remdesivir for 5- or 10-day courses in hospitalized COVID-19 patients with
oxygen saturations ≤94%. By day 14, a clinical improvement of ≥2 points on an
ordinal scale occurred in 64% of patients in the 5-day group (n=200) and 54% of
patients in the 10-day group (n=197). Without a placebo control, it is unclear if
remdesivir altered the natural progression of disease. It does however seem
remdesivir for 5 days could be considered as effective as a 10-day course. The
most common adverse events noted in both groups were nausea, worsening
respiratory failure, elevated LFTs, and constipation (Goldman et al, N Engl J
Med, 2020)
7. A randomized, double-blind, placebo-controlled trial evaluated remdesivir IV
for 10 days versus placebo in adults with confirmed severe COVID-19 in 10
hospitals in Wuhan, China. In the intention-to-treat population of 237
participants (41% women, median age 65), remdesivir resulted in no significant
difference in time to clinical improvement versus placebo (hazard ratio 1.23,
95% CI 0.87-1.75). Remdesivir was discontinued due to adverse events in 12%
of patients versus 5% of patients in the placebo group. The trial was stopped
early after only enrolling 52% of the target sample size due to decreasing
incidence of COVID-19 in Wuhan. The authors estimated this yielded 58%
power to detect a hazard ratio of 1.4 or higher. As the trial was underpowered,
its findings are inconclusive (Wang et al, Lancet, 2020)
8. Preliminary results of the adaptive NIAID trial (ACTT-1) evaluated over 1000
patients with COVID-19 in a randomized, placebo-controlled trial. The median
time to recovery in the remdesivir group (n=538) was 11 days (95% CI 9-12) vs.
15 days (95% CI 13-19) in the placebo group (n=521) (recovery rate ratio 1.32,
95% CI 1.12-1.55, p<0.001). The mortality rate by day 14 was 7.1% with
remdesivir compared to 11.9% with placebo (HR 0.7, 95% CI 0.47-1.04).
Serious adverse events were reported in 21.1% of remdesivir patients and 27%
of placebo patients (Beigel et al, N Engl J Med, 2020)

Recommendations

1. Remdesivir is available in limited supply via emergency use authorization

(EUA) or for pregnant patients with severe disease through Gilead’s
compassionate use protocol. Please use this link for Partners’ criteria for the use
of EUA remdesivir

Dosing

Remdesivir is only available as an investigational agent through clinical trials or

emergency use authorization

1. 200 mg IV loading dose, followed by 100 mg IV daily for a total of 5-10 days

Monitoring and Toxicity

Favipiravir
1. If treatment of COVID-19 is being considered, favipiravir trial enrollment
should be discussed with the infectious diseases study team for key inclusion
and exclusion criteria (NCT04358549)

Pharmacology

1. Favipiravir, also known as T-705 or Avigan, is a pyrazine derivative and

guanine analog that acts as an inhibitor of viral RNA-dependent RNA
polymerase, causing chain termination and preventing RNA elongation
2. Favipiravir demonstrates anti-viral activity against a broad array of RNA
viruses, including arenaviruses, bunyaviruses, and filoviruses (Furuta et al, Proc
Jpn Acad Ser B Phys Biol Sci, 2017)
3. In Japan, favipiravir is approved for influenza A resistant to neuraminidase
inhibitors (Hayden et al, Curr Opin Infect Dis, 2019)

Evidence

1. Favipiravir is effective in protecting mice against Ebola virus, although the

EC50 is relatively high at 67 μM (Oestereich et al, Antiviral Res, 2014)
2. Favipiravir inhibits SARS-CoV-2 in vitro (Wang et al, Cell Res, 2020)
3. A non-placebo, open-label, single-center trial of 80 patients in Shenzhen, China
of oral favipiravir compared with a historical cohort of patients receiving
lopinavir/ritonavir for 14 days demonstrated reduced median viral shedding time
(primary endpoint) in the favipiravir group (4 vs. 11 days; p < 0.001).
Radiographic improvement was seen in 91% of favipiravir treated subjects vs.
62% of those on lopinavir/ritonavir (p < 0.004). Of note, all patients also
received inhaled interferon. The trial excluded patients with severe disease or
ICU admission. (Cai et al, Engineering, 2020)
4. A prospective, multicenter, open-label, randomized trial of 236 patients in China
comparing favipiravir and umifenovir demonstrated a higher clinical recovery
rate at day 7 in moderately ill patients receiving favipiravir (71.4% vs. 55.9%, p
= 0.0199), but not among severely or critically ill patients. Clinical recovery was
defined as three or more days of improvement in respiratory rate, oxygenation,
cough, and fever. There was no placebo group for comparison and this study has
not been peer reviewed (Chen et al, MedRxiv, 2020, unpublished report).

Recommendations

1. If eligible, patients can be enrolled in the favipiravir clinical trial for COVID-19
(NCT04358549)

Dosing

1. Favipiravir is only available as an investigational agent through clinical trials

and is being studied at a dose of 1800 mg BID loading dose on day 1, followed
by 1000 mg BID for the next 13 days
Monitoring and Toxicity

1. Teratogenicity
2. Increased serum uric acid (Chen et al, MedRxiv, 2020, unpublished report).
3. In monkeys, toxicities observed have been mild hepatic and hematologic effects

1. Elevations in AST, ALT, total bilirubin, and alkaline phosphatase

2. Anemia

Umifenovir (Arbidol)
Pharmacology

1. Broad spectrum of activity against enveloped and non-enveloped RNA and

DNA viruses
2. Blocks multiple stages of viral life cycle, including fusion (Kadam, Proc Natl
Acad Sci USA, 2017; Blaising et al, Antiviral Res, 2014) and exocytosis
(Blaising et al, Antiviral Res, 2013)
3. Umifenovir has been licensed for use as treatment of and prophylaxis against
influenza in China since 2006 and Russia since the 1990’s (Blaising et al,
Antiviral Res, 2014).
4. Demonstrates in vitro activity against HBV, HCV, and, HHV-8 (Pécheur et al, J
of Virol, 2016), Coxsackievirus B (Herod et al, J Gen Virol, 2019), Ebola and
Lassa viruses (Hulseberg, J Gen Virol, 2019), and Influenza (Blaising et al,
Antiviral Res, 2014)

Evidence

1. There is in vitro evidence that umifenovir suppresses reproduction of the SARS-

CoV-1 virus (Khamitov et al, Vapr Virusol, 2008)
2. In one case series of four patients, three out of four demonstrated clinical
improvement with the use of lopinavir/ritonavir, umifenovir, and a traditional
Chinese medicine (Wang et al, BioSci Trends, 2020)
3. A single-center, retrospective cohort study of 33 patients in Guangzhou, China
compared lopinavir/ritonavir combined with umifenovir vs. lopinavir/ritonavir
alone for 7-14 days. The combination therapy was found to be more strongly
associated with the primary endpoint of negative COVID-19 PCR testing at day
7 (75% vs 35%, p < 0.05) and day 14 (94% vs 53%, p < 0.05) (Deng et al, J
Infect, 2020)
4. A multicenter retrospective study of 50 patients with COVID-19 demonstrated
that seven days of umifenovir monotherapy was associated with higher rates of
negative SARS-CoV-2 PCR testing at 7 and 14 days after hospital admission
than lopinavir/ritonavir (Zhu et al, J Infect, 2020)
5. In another retrospective, observational study at multiple hospitals in Wuhan,
China, 36 patients treated with umifenovir had higher rates of hospital discharge
(33% vs 19%) and lower rates of mortality (0% vs 16%) than patients who did
not receive umifenovir. These findings were not statistically significant. There
was no randomization and the other treatments received (e.g., antimicrobials,
 steroids) were not accounted for in the comparison between umifenovir-treated
and umifenovir-untreated groups (Wang et al, Clin Infect Dis, 2020)
6. A retrospective study of 81 non-ICU patients in Wuhan, China with moderate to
severe COVID-19 demonstrated no difference in the primary outcome of rate of
negative nasopharyngeal PCR test at day 7 (73% vs 78%, p=0.19). Additionally,
the median time from onset of symptoms to negative PCR was 18 days for
patients treated with umifenovir and 16 days in the control group (p=0.42) (Lian
et al, Clin Microbiol Infect, 2020)
7. A prospective, multicenter, open-label, randomized trial of 236 patients in China
comparing umifenovir and favipiravir demonstrated a lower clinical recovery
rate at day 7 in moderately ill patients receiving umifenovir (55.9% vs. 71.4% p
= 0.0199), but not among severely or critically ill patients. Clinical recovery was
defined as three or more days of improvement in respiratory rate, oxygenation,
cough, and fever. There was no placebo group for comparison and this study has
not been peer reviewed (Chen et al, MedRxiv, 2020, unpublished report).
8. Trials are ongoing in China (NCT04255017; NCT04260594; NCT04273763;
NCT04252885), but none have been initiated in the US to date

Recommendations

1. Umifenovir is not recommended outside of clinical trial participation at this time

Dosing

1. Umifenovir 200-400 mg PO TID for up to 10 days

Monitoring and Toxicity

1. No major adverse effects have been reported

2. Umifenovir is an inhibitor of UDP-glucuronosyl-transferase (UGT) 1A9
(UGT1A9), so monitoring for drug-drug interactions is necessary (e.g.,
acetaminophen) (Liu et al, Pharmazie, 2013)

Hydroxychloroquine and Chloroquine

Pharmacology

1. Hydroxychloroquine (HCQ) is an anti-malarial 4-aminoquinoline shown to have

in vitro activity against diverse RNA viruses, including SARS-CoV-1 (Touret
and de Lamballerie, Antivir Res, 2020)
2. HCQ is thought to act against viruses through multiple mechanisms (Devaux et
al, Int J Antimicrob Agent, 2020):

1. Inhibition of viral entry. HCQ inhibits synthesis of sialic acids and interferes
with protein glycosylation, which may disrupt interactions necessary for viral
attachment and entry (Vincent et al, Virol J, 2005; Olofsson et al, Lancet Infect
Dis, 2005)
2. Inhibition of viral release into the host cell. HCQ blocks endosomal
acidification, which activates endosomal proteases. These proteases are required
 to initiate coronavirus/endosome fusion that releases viral particles into the cell
(Yang et al, J Virol, 2004)
3. Reduction of viral infectivity. HCQ has been shown to inhibit protein
glycosylation and proteolytic maturation of viral proteins. Studies on other RNA
viruses have shown a resulting accumulation of non-infective viral particles, or
an inability of viral particles to bud out of the host cell (Savarino et al, J Acquir
Immune Defic Syndr, 2004; Klumperman et al, J Virol, 1994)
4. Immune modulation. HCQ reduces toll-like receptors and cGAS-STING
signaling. It has been shown to reduce release of a number of pro-inflammatory
cytokines from several immune cell types (Schrezenmeier and Dorner, Nat Rev
Rheum, 2020)

Evidence

1. Hydroxychloroquine was found to be more potent than chloroquine in inhibiting

SARS-CoV-2 in vitro (Yao et al, Clin Infect Dis, 2020)
2. An expert consensus group out of China suggested that chloroquine improved
lung imaging and shortened disease course (Zhonghua et al, CMAPH, 2020).
Chloroquine is included in the treatment guidelines from the National Health
Commission of the People's Republic of China, but the specific data on which
this is based is not available (Gao et al, Biosci Trends, 2020)
3. One pre-print report of 62 COVID-19+ patients showed improved time to
clinical recovery in the hydroxychloroquine treatment arm compared to placebo
(Chen et al, unpublished report, 2020)
4. However, a number of additional reports have since shown no positive impact
with the addition of hydroxychloroquine (Tang et al, BMJ, 2020; Geleris et al, N
Engl J Med, 2020; Mahevas et al, BMJ, 2020; Borba et al, JAMA Netw Open,
2020)

1. Tang and colleagues published the first randomized controlled trial assessing
HCQ (dosed at 1200 mg/day x3 days, then 800 mg/day x2-3 weeks) + standard
of care (n=75) compared to standard of care (SOC) alone (n=75) for the
treatment of COVID-19 disease. The probability of negative SARS-CoV-2
testing by day 28 was 85.4% in the HCQ group and 81.3% in the SOC group
(difference 4.1%, 95% CI -10.3 to 18.5%) and adverse effects were more
common in the HCQ group (30% vs. 9%) (Tang et al, BMJ, 2020)
2. Geleris and colleagues retrospectively reviewed 1,376 COVID-19+ patients
hospitalized in New York City, 811 of whom received HCQ (dosed at 600 mg
bid day 1, then 400 mg daily for 5 days). There was no significant association
between HCQ use and intubation or death (HR 1.04, 95% CI 0.82-1.32) (Geleris
et al, N Engl J Med, 2020)

1. Patients receiving HCQ were more severely ill at baseline, so it’s difficult to
draw definitive conclusions from this study, despite multiple sensitivity analyses

3. Mahevas and colleagues published an observational study comparing HCQ 600

mg/day (n=84) compared to SOC without HCQ (n=89) in COVID-19 disease. In
the weighted analysis, the survival rate without ICU transfer by day 21 was 76%
in the HCQ group and 75% in the SOC group (HR 0.9, 95% CI 0.4-2.1) and 8
 patients receiving HCQ had to discontinue treatment due to EKG modifications
(Mahevas et al, BMJ, 2020)
4. Borba and colleagues randomized patients to receive high-dose chloroquine at
600 mg bid x10 days (n=41) or low-dose chloroquine at 450 mg bid day 1, then
once daily for 4 days (n=40). In the published interim analysis at day 13, the
high-dose group had higher rates of mortality (39% vs. 15%) and more instances
of QTc prolongation (18.9% vs. 11.1%) than the low-dose group, which resulted
in halting the high-dose arm of the study. Of note, all patients received
azithromycin (Borba et al, JAMA Netw Open, 2020)

5. The first randomized control trial for COVID-19 post-exposure prophylaxis was
published on June 3, 2020. Asymptomatic patients who had household or
occupational exposures to others with COVID-19 for more than 10 minutes
within 4 days of exposure were randomized to receive either placebo (n=407) or
hydroxychloroquine 800 mg once, 600 mg in 6-8 hours, then 600 mg daily for 4
additional days (n=414). The incidence of new illness compatible with COVID-
19 was 11.8% in the hydroxychloroquine arm and 14.3% in the placebo arm
(absolute difference -2.4%, 95% CI -7 to 2.2%, p=0.35). Side effects were more
common in the hydroxychloroquine arm (40.1% vs. 16.8%), but no serious
adverse reactions were reported (Boulware et al, N Engl J Med, 2020)

Recommendations

1. Hydroxychloroquine is not recommended outside of the context of a clinical

trial

Dosing

1. Hydroxychloroquine: 400 mg PO q12h on the first day, followed by 200 mg

q12h (q8h if concerns for absorption) for a total of 5 days

1. May extend up to 10 days depending on clinical response

2. The half-life of HCQ is over 7 days, so a 5-day treatment course should still
yield therapeutic HCQ levels past day 10 (Yao et al, Clin Infect Dis, 2020).
3. A translational PK-PD model published by a UCSF group found that
hydroxychloroquine doses >400 mg BID for ≥5 days were more likely than
doses ≤400 mg daily to rapidly decrease viral loads, reduce the proportion of
patients with detectable SARS-CoV-2 infection, and shorten treatment courses.
However, hydroxychloroquine doses >600 mg BID were predicted to
significantly prolong the QTc interval (Garcia-Cremades et al, Clin Pharmacol
Ther, 2020)
4. A separate PK-PD model from a French group found that a dose of 800 mg on
the first day, followed by 200 mg q12h for 7 days was the most optimal for
intensive care unit patients to balance both efficacy and toxicity (Perinel et al,
Clin Infect Dis, 2020)

2. Chloroquine phosphate

1. Not available at BWH

Monitoring and Toxicity

1. Hydroxychloroquine is contraindicated in epilepsy and porphyria. Known

adverse effects include:

1. Gastrointestinal symptoms (nausea, cramps, diarrhea)

2. Bone marrow suppression
3. Cardiomyopathy and retinopathy

1. Case series and reports have found this to be a long-term (years) and dose-
dependent phenomenon. Given the anticipated short duration in COVID-19, it is
not an expected risk (Nord et al, Semin Arthritis Rheum, 2004; Yusuf et al, Eye,
2017).

4. QT-segment prolongation and therefore torsades de pointes, especially if

administered in combination with azithromycin or other QTc-prolonging agents

2. Due to risks of cardiac arrhythmias, the FDA has cautioned against the use of
hydroxychloroquine for outpatients with COVID-19 (FDA Drug Safety
Communication, April 2020)
3. Given this, the following monitoring is required for patients being treated with
hydroxychloroquine:

1. Obtain baseline ECG, ECG 3.5 hours after first dose, and daily ECG thereafter
2. Discontinue all other QT-prolonging agents
3. Maintain continuous telemetry while under treatment
4. Do not start if QTc > 500 msec (or 550 msec with pacing or BBB)
5. Discontinue if there is an increase in PVCs or non-sustained polymorphic VT.

4. The ACC has published a risk score for drug associated QTc prolongation which
may help in determining those patients in whom these drugs may be dangerous.
ACC and HCQ Risk Assessment

Azithromycin
Pharmacology

1. Azithromycin is a macrolide antibiotic that inhibits RNA-dependent protein

synthesis by binding to the 50S ribosomal subunit of bacteria, resulting in
blockage of transpeptidation
2. There is no evidence of azithromycin’s direct effects on SARS-CoV-2
3. Theoretical benefit could come in azithromycin’s anti-inflammatory effects seen
in other infectious diseases (Amsden GW, J Antimicrob Chemother, 2005)

Evidence

1. A French study has received international attention over the potential

combination of hydroxychloroquine and azithromycin for the treatment of
COVID-19 and each subsequent publication has increased the study size (6
 patients, 80 patients, and most recently 1061 patients), but without a control
group (Gautret et al, Int J Antimicrob Agents, 2020, Gautret et al, Travel Med
Infect Dis, 2020; Million et al, Travel Med Infect Dis, 2020)
2. A separate French study has shown no benefit of adding azithromycin to
hydroxychloroquine, directly conflicting the work done by Gautret and
colleagues (Molina et al, Med Mal Infect, 2020)

1. No conclusions can be drawn from these studies and until more conclusive
studies are done, we believe that the risks of QTc prolongation and torsades de
pointes outweigh the potential benefits of combination treatment

Recommendations

1. There is not sufficient supporting evidence to use azithromycin for COVID-19

disease, unless concomitant community-acquired pneumonia is suspected and
atypical coverage is desired

Dosing

1. Normal azithromycin dosing for community-acquired pneumonia (for atypical

coverage) is 500 mg daily for 5 days inpatient, or 500 mg x1, then 250 mg for a
total of 5 days

Monitoring and Toxicity

1. While not recommended for treatment of COVID-19 at BWH, if azithromycin is

used in combination with hydroxychloroquine, QTc should be monitored as
described in the hydroxychloroquine section above

1. Numerous studies have raised concerns about the deleterious effects of

hydroxychloroquine and azithromycin combination therapy (Mercuro et al,
JAMA Cardiol, 2020; Bessière et al, JAMA Cardiol, 2020; Chorin et al,
unpublished report, 2020)

Lopinavir/ritonavir
Pharmacology

1. Lopinavir/ritonavir (Kaletra, LPV/r) has been available since 2000 as an

antiretroviral agent in the treatment of human immunodeficiency virus (HIV)
2. Lopinavir and ritonavir are both protease inhibitors, which by inhibiting HIV-1
protease, lead to the formation of immature, noninfectious viral particles.
Ritonavir specifically is a CYP3A4 inhibitor that is used to decrease metabolism
of lopinavir (a CYP3A4 substrate), thereby increasing serum lopinavir levels
3. Lopinavir may theoretically work against coronaviruses like SARS-CoV-2 by
inhibiting 3-chymotrypsin-like protease (3CLpro)

Evidence
 1. Lopinavir was shown to have in vitro activity against both SARS-CoV-1 and
MERS-CoV in some studies (Chu et al, Thorax, 2004; de Wilde et al.
Antimicrob Agents Chemother, 2014), but not in others (Chan et al, J Infect,
2013)
2. Against SARS-CoV-1, LPV/r use (n=75) was associated with a lower overall
mortality and intubation rate in one study. A subgroup analysis showed no
difference in overall mortality or intubation rate however when
lopinavir/ritonavir was used as rescue therapy at a median of 18 days after
symptom onset (n=31) (Chan et al, Hong Kong Med J, 2003)
3. A recent randomized, controlled, open-label trial assessed lopinavir-ritonavir
(n=99) vs. standard of care (n=100) in SARS-CoV-2 patients (Cao et al, N Engl
J Med, 2020)

1. Treatment with LPV/r was not associated with a difference in time to clinical
improvement or mortality
2. Randomization didn’t occur until a median of 13 days after symptom onset
however, so the window for benefit may have already closed, as seen in the
Chan et al paper in SARS-CoV-1

4. A randomized open-label, phase 2 trial assessed the combination of

lopinavir/ritonavir, ribavirin, +/- interferon β-1b (n=52 with IFN, 34 without
IFN) against lopinavir/ritonavir monotherapy (n=41) in Hong Kong hospitals.
When given within 7 days of symptom onset in non-critically ill patients,
combination therapy shortened the time from treatment initiation to negative
nasopharyngeal swab compared to monotherapy (7 days [IQR 5-11] vs. 12 days
[IQR 8-15]). It’s difficult to know LPV/r’s role in this regimen as both groups
received the antiviral (Hung et al, Lancet, 2020)
5. There are still many ongoing trials for the use of LPV/r in COVID-19, but
additional results are not yet available (Yao et al, J Med Virol, 2020)

Recommendations

1. Lopinavir/ritonavir should not be used at BWH as overall evidence is lacking

and we are unable to obtain supply of the medication

Dosing

1. If LPV/r were to be used at other institutions, the dosing regimen is 400/100 mg

by mouth twice daily for up to 10 days

Monitoring and Toxicity

1. Interactions are an incredibly important aspect of LPV/r use as ritonavir is a

potent CYP3A4 inhibitor, so will interact with CYP3A4 substrates (i.e.
apixaban, tacrolimus, amiodarone)
2. Diarrhea, nausea, and transaminitis are common. Other adverse effects include
hyperlipidemia, pancreatitis, asthenia, and hyperglycemia
Nitazoxanide
Pharmacology

1. Nitazoxanide is an antiprotozoal agent originally approved by the FDA in 2004

for use in adults for the treatment of diarrhea caused by Giardia lamblia or
Cryptosporidium parvum
2. It is metabolized to its active metabolite tizoxanide in vivo
3. Antiviral activity of tizoxanide is due to interference with host-regulated
pathways involved in viral replication, rather than a particular virus-targeted
mechanism (Rossignol JF, J Infect Public Health, 2016)

Evidence

1. Nitazoxanide’s metabolite tizoxanide has in vitro activity against a variety of

viruses, including influenza (Rossignol et al, J Biol Chem, 2009), hepatitis B
(Korba et al, Antiviral Res, 2008), hepatitis C (Keeffe et al, World J
Gastroenterol, 2009), and Japanese encephalitis (Shi et al, Virol J, 2014)
2. In a randomized, controlled phase 2b/3 trial for uncomplicated influenza,
nitazoxanide dosed at 600 mg twice daily for 5 days reduced the duration of
symptoms by ~21 hours (Haffizulla et al, Lancet Infect Dis, 2014).

1. Other phase 3 trials in the treatment of influenza have been completed

(NCT01610245, NCT02612922, NCT03336619), but results are not yet
available.

3. Another randomized controlled trial comparing nitazoxanide to placebo in the

treatment of severe influenza-like illness showed no difference in hospital length
of stay (Gamiño-Arroyo et al, Clin Infect Dis, 2019)
4. Nitazoxanide had been shown to have in vitro activity against MERS-CoV
(Rossignol JF, J Infect Public Health, 2016)
5. Most recently, nitazoxanide has been shown to have in vitro activity against
SARS-CoV-2 (Wang et al, Cell Res, 2020)

Recommendations

1. Nitazoxanide should not be used at BWH as overall clinical evidence is lacking

and optimal dosing is not yet known

Dosing

1. If nitazoxanide were to be used, the most likely dose regimen would be 600 mg
twice daily of extended release tablets, which was the dosing used for influenza
without major safety concerns. However, the extended release tablets are not
currently available in the United States

Monitoring and Toxicity

1. Common adverse effects of nitazoxanide include:

 1. Abdominal pain (6.6%)
2. Diarrhea (4.2%)
3. Headache (3.1%)
4. Nausea (3%)

Anti-IL-6 Agents (Tocilizumab, Sarilumab,

Siltuximab)
Pharmacology

1. IL-6 activates T cells and macrophages, among other cell types (see “Cytokine
Storm” section in “Critical Care” chapter). IL-6 inhibitors are approved for
cytokine activation syndrome complications related to Chimeric Antigen
Receptor T cell (CAR-T) therapy (Brudno and Kochenderfer, Blood Rev, 2019;
Rubin et al, Brain, 2019).
2. IL-6 levels are reported to correlate with severe COVID-19 (Ruan et al,
Intensive Care Med 2020; Liu et al, unpublished report). While patients have
peripheral lymphopenia, BAL fluid is often lymphocytic, suggesting that IL-6
inhibition and prevention of T cell activation may be protective.

Evidence

1. A Chinese case series described 21 patients who received tocilizumab 400 mg

once. Three patients received a second dose within 12 hours due to continued
fevers. All patients presented with fevers, but all defervesced within 24 hours of
tocilizumab administration. In addition, oxygen requirement and lymphopenia
seemed to improve post-tocilizumab. (Xu et al, Proc Natl Acad Sci USA, 2020)

1. This was a small, single arm study that didn’t report out on any concomitant
therapy used, so it’s unclear if patients also received steroids and/or antivirals.
While thought-provoking, it isn’t conclusive by any means

2. A similarly sized non-peer-reviewed Italian case series described 21 patients

requiring ventilation (CPAP or NIV) who received siltuximab via compassionate
use at a dose ranging from 700-1200 mg once. Five patients received a second
dose 2-3 days after the initial dose at the clinician’s discretion. Of the 21
patients, the condition of 7 seemed to improve (based on CPAP/NIV settings), 9
seemed to stabilize, and 5 seemed to worsen, including 1 death (Gritti et al,
unpublished report, 2020)
3. Neither of these small case series confirm the efficacy or safety of anti-IL6
agents and large randomized controlled trials are needed, and are currently
ongoing
4. Press releases from ongoing randomized controlled trials are inconclusive thus
far (Assistance Publique Hôpitaux de Paris press release April 27, 2020 on
tocilizumab; Regeneron press release April 27, 2020 on sarilumab)
Recommendations

1. The routine use of anti-IL-6 agents is not recommended outside the context
of a clinical trial
2. For severe cases of COVID-19 with suspicion of cytokine activation syndrome
(see “Cytokine Storm”), consider use in conjunction with Rheumatology and
Infectious Diseases consultation following the treatment algorithm found in
appendix A of the ID treatment guidelines

1. Retrospective reviews in patients with rheumatological disease suggest a

possible increase in serious bacterial infection, so use caution if secondary
infection is clinically suspected

3. Other immunomodulatory agents have been proposed in the management of

COVID-19 disease (i.e. baricitinib, lenzilumab, leronlimab), but none currently
have evidence supporting their use

Dosing

1. Tocilizumab (anti-IL6R mAb): Off-label dosing 400 mg (4-8 mg/kg) IV x1.

Dose may be repeated 12 hours later if inadequate response to the first dose, but
is not necessary in the majority of cases. The total dose should be no more than
800 mg. Tocilizumab should not be administered more than twice.

2. Sarilumab (anti-IL6R mAb): New intravenous formulation and dosing,

available only as part of a clinical trial (NCT04315298).

3. Siltuximab (anti-IL6 mAb): 11 mg/kg IV x1

Monitoring and Toxicity

1. Tocilizumab common adverse effects include:

1. Transaminitis (AST, ALT), >22%

2. Infusion reaction, 4-20%
3. Hypercholesterolemia, 20%
4. Upper respiratory tract infection, 7%
5. Neutropenia, 2-7%

2. Sarilumab common adverse effects include:

1. Transaminitis (AST, ALT), 28-47%

2. Neutropenia, 7-10%
3. Infusion reactions, 7%
4. Upper respiratory tract infections, 4%
5. Urinary tract infections, 3%

3. Siltuximab common adverse effects include:

1. Edema, >26%
 2. Upper respiratory infection, >26%
3. Pruritus / skin rash, 28%
4. Hyperuricemia, 11%
5. Lower respiratory tract infection, 8%
6. Thrombocytopenia, 8%
7. Hypotension, 4%

4. Tocilizumab and sarilumab have black box warnings for a risk of serious
infections, including tuberculosis and other opportunistic infections. Patients
treated with either agent should be tested for latent tuberculosis prior to
discharge from the hospital and followed up in the TB clinic if that testing is
positive.

Anti-IL-1 Agents (Anakinra, Canakinumab,

Rilonacept)
Pharmacology

1. IL-1 activates T cells and macrophages, among other cell types (see “Cytokine
Storm” section in “Critical Care” chapter). Two forms of IL-1, IL-1alpha and
IL-1beta, have similar affinity for the IL-1 receptor, but IL-1beta is typically
dominant in systemic inflammation. IL-1beta inhibitors are approved for
systemic juvenile idiopathic arthritis, a disease that is associated with a cytokine
activation syndrome termed macrophage activation syndrome (MAS) (Ruperto
et al, N Engl J Med, 2012). Observational studies suggest efficacy of high-dose
IL-1 blockade for MAS and related cytokine storm syndromes (Eloseily et al,
Arthritis Rheumatol, 2020; Monteagudo et al, ACR Open Rheumatol, 2020).
Cytokine release syndrome related to Chimeric Antigen Receptor T cell (CAR-
T) therapy is dependent on both IL-6 and IL-1 in murine models, and IL-1
blockade is being tested for this indication in clinical trials (NCT04148430)
(Giavridis et al, Nat Med, 2018; Norelli et al, Nat Med, 2018)
2. Levels of IL-1 in blood are difficult to measure and correlate poorly with disease
activity in IL-1-mediated diseases
3. Three IL-1-blocking agents are available. These are distinguished by mechanism
of action and half-life

1. Anakinra is a human recombinant IL-1 receptor antagonist (IL-1ra), the

endogenous IL-1 inhibitor that competes with IL-1alpha and IL-1beta for
receptor binding. It has a half-life of 4-6 hours after subcutaneous
administration. IV administration is off-label but routinely performed in
hospitalized patients; the half-life of anakinra is ~2 hours after IV administration
(Granowitz et al, Cytokine, 1992; Yang et al, Clin Pharmacol Ther, 2003)
2. Canakinumab is a monoclonal antibody that binds to IL-1beta, but does not
antagonize IL-1alpha. It has a half-life of 26 days after subcutaneous
administration
3. Rilonacept is a synthetic fusion molecule consisting of cytokine binding
domains from the IL-1 receptor fused to a human IgG1 Fc, thereby serving as a
soluble decoy receptor that “traps” both IL-1alpha and IL-1beta. It has a half-life
of 7 days after subcutaneous administration. Since rilonacept also sequesters IL-
 1ra, the natural antagonist of IL-1 signaling, it is considered typically less potent
in net anti-inflammatory effect than anakinra or canakinumab

Evidence

1. There are currently no randomized clinical trials that assess the utility or safety
of IL-1 blockade in COVID-19
2. In a series of 3 COVID-19 patients, gene expression of IL-1-related genes
correlated with disease severity (Ong et al, Cell Host Microbe, 2020)
3. The potential utility of IL-1 blockade for COVID-19 cytokine activation
syndrome is extrapolated from its use in other clinical contexts (Henderson et al,
Arthritis Rheumatol, 2020). Notably, high-dose IV anakinra has been tested in
randomized controlled trials of over 1500 adults with sepsis (Fisher et al, JAMA,
1994; Opal et al, Crit Care Med, 1997). Mortality was numerically lower in
anakinra than placebo arms, but was not clinically significant. However, this
effect reached significance in subgroup analyses of patients with the most severe
illness, including features suggestive of cytokine storm (Fisher et al, JAMA,
1994; Shakoory et al, Crit Care Med, 2016). Further, although chronic IL-1
blockade enhanced infection risk in older adults with comorbidities, there was
no enhancement of tuberculosis risk, and blockade is not known to exacerbate
hepatitis B or C (Ridker et al, N Engl J Med, 2017; Anakinra, LiverTox, 2017)
4. An Italian group retrospectively reviewed 29 COVID-19+ patients who received
high-dose anakinra (5 mg/kg IV bid). At day 21, 72% of patients had improved
respiratory function, while 17% were on mechanical ventilation and 10% had
passed away (Cavalli et al, Lancet Rheum, 2020)
5. Clinical trials of anakinra in COVID-19 are ongoing (NCT04324021,
NCT04330638, NCT04339712, NCT04324047)

Recommendations

1. The routine use of anti-IL-1 agents is not recommended unless part of a

clinical trial
2. For severe cases of COVID-19 with suspicion of incipient or active cytokine
activation syndrome (see “Cytokine Storm” chapter), consider use of IL-1
blockade in conjunction with Rheumatology and Infectious Diseases
consultation following the treatment algorithm found in appendix A of the ID
treatment guidelines. Because of its short half-life, anakinra is preferred if
therapy is warranted.

Dosing

1. Anakinra (human recombinant IL-1ra): 100 mg BID-QID subcutaneous (or IV,

off-label) for a 2-3-day trial. Duration may be extended at the discretion of
rheumatology

1. While off-label, IV administration may be possible and is detailed in a recent

publication on anakinra’s use in hemophagocytic lymphohistiocytosis (HLH) or
macrophage activation syndrome (Mehta et al, Lancet Rheum, 2020)
 2. Exhibits partial renal excretion and is not cleared by dialysis; however, the large
therapeutic window suggests that minimal dose adjustment is usually required
(i.e. trials of anakinra in sepsis employed doses of 1-2 mg/kg/day)

2. Canakinumab (anti-IL-1beta mAb): 4 mg/kg (max 300 mg) subcutaneous once;

the 26-day half-life of canakinumab likely renders repeat dosing during acute
COVID-19 unlikely

1. Clinical trials are studying novel dosing of canakinumab administered

intravenously

3. Rilonacept (IL-1 trap): 320 mg subcutaneous once (two 2 mL injections at

different sites)

Monitoring and Toxicity

1. Anakinra common adverse effects include:

1. Injection site reactions (50-75%); these include both acute stinging and a
subacute but transient reaction, beginning a few weeks into treatment where
welts form at sites of prior subcutaneous injections (Kaiser et al, Rheumatol Int,
2012), but it’s unclear what the risk is with intravenous administration
2. Other potential side effects include headache, GI disturbances, arthralgias,
hepatitis, neutropenia, and the potential for increased risk of infection

2. Canakinumab common adverse effects include:

1. Increased susceptibility to infections, 30-55%

2. GI disturbances, 7-20%
3. Injection site reactions, 10-12%
4. Cytopenias, 6-10%
5. Transaminitis (AST, ALT), 4%

3. Rilonacept common adverse effects include:

1. Injection site reactions, 48%

2. Increased susceptibility to infections, 34-48%

1. Upper respiratory tract infection, 26%

3. Hyperlipidemia

1. Mean increases in total cholesterol, HDL, LDL, and triglycerides are 19, 2, 10,
and 57 mg/dL, respectively

4. Neutropenia, transient, <1%

4. Anakinra, canakinumab and rilonacept labeling do not include a black box

warning for a risk of tuberculosis or other opportunistic infections. Testing for
latent tuberculosis, hepatitis B, and hepatitis C is recommended for patients
 receiving therapy, but it is not necessary to delay treatment for patients with
COVID-19 while awaiting these results.

Convalescent Plasma
Pathophysiology

1. Convalescent plasma originates from patients who have previously recovered

from a viral infection and are now able to donate their immunoglobulin-
containing blood
2. The presumed mechanism of action is that antibodies present in convalescent
plasma may suppress viremia
3. In previous viral infections antibody administration is likely more effective as
pre- or post- exposure prophylaxis than treating established infection (Cheng et
al, Eur J Clin Microbiol Infect Dis, 2005)

Evidence

1. A meta-analysis of 32 studies assessing the efficacy of convalescent plasma for

the treatment of severe viral acute respiratory infections (coronaviruses and
influenza) showed a reduction in mortality following treatment (Mair-Jenkins et
al, J Infect Dis, 2015)

1. Of note, studies included were of low or very low quality and generally lacked
control groups

2. SARS-CoV-1

1. In one Hong Kong report, patients who received convalescent plasma had a
12.5% mortality rate (n=80), while the overall mortality rate was 17% in March-
May 2003 (Cheng et al, Eur J Clin Microbiol Infect Dis, 2005)
2. Patients treated with convalescent plasma before day 14 (n=48) were more likely
to be discharged than those treated after day 14 (n=32) (58.3% vs. 15.6%,
p<0.001), and the mortality rates in the two groups were 6.3% vs. 21.9%
respectively (p=0.08)

3. SARS-CoV-2

1. Multiple case series have been published to-date

1. Shen and colleagues published on 5 critically ill COVID-19 patients who

received convalescent plasma, 4 of 5 who saw temperature normalization within
3 days and 3 of 5 who had been discharged at the time of publication (days 51,
53, and 55), with the other two in stable condition (Shen et al, JAMA, 2020)
2. Zhang and colleagues published on 4 critically ill COVID-19 patients who
received convalescent plasma (including one pregnant woman), all of whom
recovered. The time from transfusion to negative PCR test results ranged from
3-22 days (Zhang et al, Chest, 2020)
 3. Duan and colleagues published on 10 severe COVID-19 disease patients who
received convalescent plasma and tolerated it well (Duan et al, Proc Natl Acad
Sci USA, 2020)
4. The most recently published case series evaluated 25 COVID-19 patients
hospitalized in Houston who received convalescent plasma. All patients
tolerated the infusion well, with 76% having at least a 1-point improvement in
clinical status by day 14 post-transfusion (Salazar et al, Am J Pathol, 2020)

1. It’s important to note that all of the above studies do not prove efficacy of
convalescent plasma, as the patients in these case series displayed relatively the
same clinical course as many COVID-19 patients who do not receive the
treatment. Randomized clinical trials are needed

2. The first multicenter randomized clinical trial was published on June 3, 2020, in
which 103 COVID-19 patients were randomized to receive convalescent plasma
(n=52) or standard of care (n=51). In this open label trial, clinical improvement
occurred within 28 days in 51.9% of the convalescent plasma patients compared
to 43.1% of the standard of care patients (HR 1.4, 95% CI 0.79-2.49). There was
also no significant difference in mortality between groups, with 15.7% mortality
in the convalescent plasma group compared to 24% with standard of care (OR
0.65, 95% CI 0.29-1.46). The findings of this study are limited however due to
early termination of the trial. To provide 80% power, 200 patients were required
in the analysis, but only half of that number were enrolled. Further studies are
still warranted (Li et al, JAMA, 2020)

Recommendations

1. Convalescent plasma is only available through clinical trial enrollment at BWH

2. Outside of BWH, providers can request use of convalescent plasma through a
single-patient Emergency Investigational New Drug (eIND) application if a
blood center has plasma available. The Mayo Clinic is leading the charge for
expanded access, more information can be found on their dedicated website
3. If used, earlier administration is more likely to be effective
4. More detailed information regarding convalescent plasma donation can be found
on the Michigan State University website.

Dosing

1. Optimal therapeutic dosing is not yet known. In one case series vs. SARS-CoV-
2, all 5 patients received 400 mL of plasma, but all with varying receptor
binding domain (RBD)-specific antibody titers. In the second study, volumes
ranged from 200 to 2400 mL of plasma, and another case series used a volume
of 200 mL for each patient. The most recent case series used a volume of 300
mL for each patient, with one patient requiring a second transfusion
2. Most ongoing studies are assessing an infusion of 1-2 units (200-500 mL) once

Monitoring and Toxicity

1. Plasma transfusions in general are safe and well-tolerated in most patients.

Potential side effects however include:
 1. Mild fever
2. Allergic reactions, including serum sickness on rare occasions
3. Transfusion-related acute lung injury (Gajic et al. Am J Respir Crit Care Med.
2007)
4. Potential risk of another infectious disease from donor, although risk is
incredibly low with modern blood bank techniques

2. There is a theoretical concern that convalescent plasma may lower a patient’s

INR if on warfarin, similar to (but to a lesser degree than) fresh frozen plasma

Interferons
Pharmacology

1. Interferons are a diverse set of signaling proteins released by host cells in

response to viral infection and are separated into three groups, based on receptor
type (Type I, Type II, and Type III) as part of innate immune function
2. Type I interferons (IFN-α/β) are endogenous compounds secreted by a variety of
virus-infected dendritic cells, macrophages, lymphocytes, fibroblasts, and
endothelial cells
3. Agonism of the IFN-α/β receptor leads JAK-STAT, MAPK, and PI3K-Akt
signaling pathways, leading to transcription of over 2000 genes (Schreiber and
Piehler, Trends Immunol, 2015)

1. Activate and attract other immune cell types

2. Inhibit viral replication
3. Regulate apoptosis of host cells

4. Type I interferons exert a broad-spectrum of antiviral effects, including against

SARS-CoV-1 in humans (Cinatl et al, Lancet, 2003)
5. Type III interferons (IFN-λ) trigger JAK-STAT pathway signaling, leading to
expression of genes

1. Antiviral and antiproliferative effects

2. May have evolved to protect tissue barriers

6. Type III interferons are induced at lower viral burden than type I IFNs and are
believed to induce antiviral activity without leading to significant tissue
destruction by neutrophils (Galani et al, Immunity, 2017)
7. Administration of recombinant or pegylated IFN-λ has been demonstrated to
suppress viral replication and abrogate “cytokine storm” development, making it
an attractive potential therapeutic agent for COVID-19 (Andreakos and
Tsiodras, EMBO Mol Med, 2020)

Evidence

1. The combination of remdesivir and interferon-β demonstrates antiviral activity

against MERS-CoV in vitro (Sheahan et al, Nat Commun, 2020)
 2. The combination of interferon-α, ribavirin, and lopinavir/ritonavir demonstrated
effectiveness in a case report of use in MERS (Kim et al, Antivir Ther, 2016)
3. The MIRACLE trial, the first clinical trial for MERS, consists of recombinant
interferon-ß1 and lopinavir/ritonavir, and is currently underway (Arabi et al,
Trials, 2018; NCT02845843)
4. In vitro, type I interferons inhibit SARS-CoV-2 replication at concentrations
similar to those used for treatment of hepatitis B and C (Mantlo et al, Antiviral
Res, 2020)
5. A study in preprint of 50 patients with severe COVID-19 who had not received
anti-inflammatory therapy demonstrated type I IFN deficiency in serum,
providing a possible rationale for recombinant IFN-α/β in severe disease
(Hadjadj et al, medRxiv, 2020, unpublished report)
6. A case study of two asymptomatic patients treated with lopinavir/ritonavir,
umifenovir, Lianhuaqingwen granules, and IFN-α2b demonstrated improved
chest imaging and clinical improvement after 10 days of therapy (Want et al,
Expert Rev Anti Infect Ther, 2020)
7. A case study of a symptomatic patient who received seven days of inhaled IFN
treatment demonstrated decrease in ground glass opacities on chest CT (Duan
and Qin, Radiology, 2020)
8. A randomized open-label, phase 2 trial assessed the combination of
lopinavir/ritonavir, ribavirin, +/- interferon β-1b (n=52 with IFN, 34 without
IFN) against lopinavir/ritonavir monotherapy (n=41) in Hong Kong hospitals.
IFN-β1b was administered subcutaneously at a dose of 8 million units every
other day for 1-3 doses. When given within 7 days of symptom onset in non-
critically ill patients, combination therapy shortened the time from treatment
initiation to negative nasopharyngeal swab compared to monotherapy (7 days
[IQR 5-11] vs. 12 days [IQR 8-15]) (Hung et al, Lancet, 2020)
9. The DisCoVeRy trial is an ongoing multicenter, randomized phase III clinical
trial in Europe of interferon β-1a with lopinavir/ritonavir compared to
remdesivir and hydroxychloroquine (NCT04315948)
10. A randomized phase 2b study of peginterferon-λ1a in HCV demonstrated similar
rate of virologic response with fewer adverse effects (Muir et al, J Hepatol,
2014)
11. A randomized, singe-blind Phase II clinical trial of pegylated IFN-λ1a for mild
COVID-19 is underway, though data are unpublished (NCT04331899)
12. Unpublished data suggest that IFN-λ production is associated with compromised
epithelial barrier in the lung, which may predispose to bacterial superinfections
in mice (Broggi et al, BioRxiv, 2020)

Recommendations

1. Recombinant type I interferon (IFN-α/β), with or without other therapy, is not

recommended outside of clinical trial participation at this time given lack of
supportive published data and significant side effect profile
2. Recombinant type III interferon (IFN-λ), with or without other therapy, is not
recommended outside of clinical trial participation at this time given potential
risk for bacterial superinfection without proven efficacy
Dosing

1. Interferon β-1a 44 mcg subcutaneous on days 1, 3, and 6 for 3 doses

2. Interferon-β-1b 0.25 mg (8 million units) subcutaneous QOD for 3 doses
3. Pegylated interferon-λ 180 mcg subcutaneous once

Monitoring and Toxicity

1. Neuropsychiatric disorders (e.g., depression, insomnia, headache, suicidal

ideation)
2. MSK (e.g., myalgias, arthralgias)
3. Influenza-like symptoms (e.g., fevers, chills)
4. Hematologic (e.g., anemia, thrombocytopenia)
5. Gastrointestinal (liver injury, nausea, vomiting, diarrhea)
6. Cardiac effects (HTN, SVTs, MI)
7. Aggravation of autoimmune disorders (e.g., thyroiditis, RA, SLE, ITP)
8. Increased risk of T2DM
9. Ocular toxicity (e.g., retinopathy, retinal detachment and hemorrhage, optic
neuritis, papilledema)

Vitamin C
1. While this idea has been popular on social media, there is currently no evidence
to support low- or high-dose vitamin C in COVID-19 patients. There is a trial
currently recruiting for high-dose vitamin C trial in COVID-19 patients in China
slated to be complete in the fall of 2020 (NCT04264533).

1. The use of Vitamin C as a treatment for sepsis is beyond the scope of this
document. A 96-hour infusion of vitamin C did not demonstrate significant
improvement of organ dysfunction, vascular injury or alter inflammatory
markers in sepsis patients with ARDS, although a reduction in 28-day mortality
was exhibited (Difference -0.17, p=0.03). (Fowler et al, JAMA, 2019). This
study does not look at COVID-19 ARDS patients.

Angiotensin Converting Enzyme Inhibitors (ACE-I)

and Angiotensin II Receptor Blockers (ARB)
Pathophysiology and Evidence

1. SARS-CoV-2, the virus that causes COVID-19, enters via the same cell-entry
receptor as SARS-CoV, namely angiotensin-converting enzyme II (ACE2)
(Paules et al, JAMA, 2020). SARS-CoV-2 is thought to have a higher affinity for
ACE2 than SARS-CoV.
2. ACE2 is expressed in the heart, lungs, vasculature, and kidneys. ACE-inhibitors
(ACEi) and angiotensin-receptor blockers (ARBs) in animal models increase the
expression of ACE2 (Zheng et al, Nat Rev Cardiol, 2020), though this has not
been confirmed in human studies. This has led to the hypothesis that ACEi and
ARBs might worsen myocarditis or precipitate ACS. It has also been
 hypothesized that the upregulation of ACE2 is therapeutic in COVID-19 and
that ARBs might be protective during infection (Gurwitz D, Drug Dev Res,
2020).
3. The evidence that currently exists favors continuing these medications unless
otherwise indicated to stop them because their abrupt discontinuation,
particularly in those who have heart failure or have had a myocardial infarction,
may lead to clinical instability and adverse outcomes (Vaduganathan et al, N
Engl J Med, 2020). The American College of Cardiology, American Heart
Association and Heart Failure Society of America joint statement recommends
against discontinuing ACEi and ARBs in patients with COVID-19 (Bozkurt et
al, HFSA/ACC/AHA Statement Addresses Concerns Re: Using RAAS
Antagonists in COVID-19, 2020).

Recommendations

1. For outpatients, we recommend against discontinuing outpatient ACEi/ARBs.

2. For inpatients, we recommend against routine discontinuation of ACEi/ARBs,
unless otherwise indicated (e.g., acute kidney injury, hypotension, shock, etc).

Non-steroidal anti-inflammatory drugs (NSAIDs)

Pathophysiology

1. SARS-CoV-2 binds to cells via ACE2. ACE2 is upregulated by ibuprofen in

animal models, and this might contribute to increased pathology (see
“Angiotensin Converting Enzyme Inhibitors (ACE-I) and Angiotensin II
Receptor Blockers (ARB)” section of this chapter).

Recommendations

1. Reports from France indicate possible increase in mortality with ibuprofen in

COVID-19 infection, but these reports have not been corroborated (Fang et al,
Lancet Respir Med, 2020; Day M, BMJ, 2020). WHO clarified on 3/20/20 it
does not recommend avoiding NSAIDs as initially stated 3/18/20 (WHO,
COVID-19 Interim guidance, March 2020).

1. Concern has been raised that NSAIDs may worsen COVID-19 disease. This has
not been proven clinically to-date, so we cannot make a recommendation for or
against their use at this time.

Blood Products
Considerations

1. In general, treat bleeding rather than numbers.

2. We recommend a restrictive transfusion strategy (Hct > 21, Hgb > 7)
3. If hemodynamically stable, transfuse 1 unit at a time and reassess needs.
1. Parsimony is encouraged given risks associated with blood product transfusions,
limited supply (blood drives are limited by social distancing), and volume
overload being of particular concern in COVID patients.

4. FFP or 4 factor-PCC (lower volume) should be given for active bleeding in the
setting of known or suspected coagulation abnormalities.
5. For warfarin reversal, use 4 factor-PCC given longer effect and lower volume.
6. Massive transfusion protocol, as a very limited resource, will need to be
activated only by the ICU attending.
7. Tranexamic acid: only for ongoing oozing/bleeding with over DIC and
hyperfibrinolysis.
8. Procedures: If the patient is at high bleeding risk, the most experienced operator
should perform the procedure to minimize trauma.

1. We recommend avoiding subclavian lines when placing central venous catheters

in coagulopathic patients.
Transfusion Thresholds

Transfusion Thresholds
Patient DVT ppx
Transfuse 1 unit at a time
RBC Platelets Cryo FFP
No bleeding, LMWH daily or Hgb < 7, If
Fibrinogen INR >
ACS,** Hgb n/a
< 100 10
Plts > 30k SC UFH TID > 10
Heparin gtt
Hgb < 7, If
No bleeding, but patient Fibrinogen INR >
ACS,** Hgb Plts < 30k
requires anticoagulation PTT goal depends on > 10 < 100 10
indication
No bleeding, SCDs* Hgb < 7, If
Fibrinogen INR >
ACS,** Hgb Plts < 10k
< 100 10
Plts < 30k Hold pharmacologic > 10
Continue
pharmacologic ppx in
Minor Procedures Hgb < 7, If
most patients Fibrinogen
ACS,** Hgb Plts < 10k INR > 3
< 100
(a-lines, CVCs) > 10
SCDs* if not using
pharmacologic
Continue
pharmacologic ppx in
Mild Bleeding or Rigors most patients Hgb < 7, If
Fibrinogen
(increases risk of ICH in ACS,** Hgb Plts < 20k INR > 3
< 100
thrombocytopenia) > 10
SCDs* if not using
pharmacologic
+ SCDs*
Hgb < 7, If
Fibrinogen INR >
Intra-Cranial Hemorrhage ACS,** Hgb Plts < 75k
Hold pharmacologic > 10 < 100 1.7
if able
INR > 2
Serious Bleeding#, Trauma or + SCDs*
Major Procedure Transfuse for
Plts < 50k Fibrinogen
active (INR >
Hold pharmacologic or higher < 100
bleeding 1.4 for
(includes LP) if able
LP)

* SCDs = sequential compression devices = “pneumoboots”

** ACS = Acute Coronary Syndrome

# Intracranial hemorrhage and massive bleeding are not included here. The massive transfusion protocol
must be activated by the attending, given blood product shortages.

Blood donation

1. We encourage all staff who are healthy and eligible to donate to make an appointment to donate
blood or platelets at the Kraft Family Blood Donor Center at DFCI and BWH, either by phone
(617.632.3206) or online.

Please send suggestions and questions: covidprotocols@bwh.harvard.edu