12 Sept 2022 Episode 286 Update and real world experience with Remdesivir, Paxlovid and Molnupiravir

Mon, 09/12/2022 - 11:52

Episode 286: Update and real world experience with Remdesivir, Paxlovid and Molnupiravir

As we know, three “small molecule” antivirals are in clinical use for COVID: Remdesivir, Molnupiravir and Paxlovid, further abbreviated as RDS, MOL and PAX. After a reminder of their efficacy and safety in the seminal phase 3 trial (PAR 1) , I summarize some of the published data on their effectiveness in clinical practice and on evidence of emerging resistance (PAR 2-4). I will also compare the effects of the most used antivirals Paxlovid and Molnupiravir alone and in combination (PAR 5). Finally I will review the evidence of spontaneous and pot-treatment viral rebound (PAR 6). At the end of each paragraph, you find a short conclusion.

PAR 1: INTRODUCTION: Summary of the 3 groundbreaking trials, based on two reviews (Ref 1 Atluri et al J Clin Med July 2022 and Ref 2 Chih-Cheng Lai in Viruses Aug 2022)

REMDESIVIR (RDS) (Veklury®) is a “chain terminator” polymerase inhibitor to be applied intravenously.

It is mainly used in hospitalized subjects.

Most randomized controlled trials (RCT) showed a small but significant beneficial effect against disease progression and mortality.

The number needed to treat (NNT = how many you need to treat to prevent a bad outcome) = 29
The “hazard ratio” (HR = a comparison between the probability of events in a treatment group, compared to the probability of events in a control group) in this case the outcome was clinical progression and/or death at day 28. HR = 0.73, with a wide confidence interval (CI = 0.52-1.03)

NIH recommendation: Five-day total course of 200 mg IV on the first day, then 100 mg IV on each subsequent day prior to discharge, for a maximum of four additional days.

For those with severe COVID-19, remdesivir is frequently used in conjunction with dexamethasone (and/or other anti-inflammatory drugs such as ant-IL-6 or Baricitinib (JAK inhibitor).

Remdesivir has been used in moderately severe non-hospitalized patients with favorable effects on prevention of hospitalization.

Key RCT Gottlieb NEJM Feb 2022 https://www.nejm.org/doi/pdf/10.1056/NEJMoa2116846?articleTools=true

Efficacy Covid-19–related hospitalization or death from any cause occurred in 2 patients (0.7%) in the remdesivir group and in 15 (5.3%) in the placebo group HR = 0.13 (.03 – 0.59)

NNT = 20

NIH recommendation: 200 mg IV on Day 1 of symptom onset, along with 100 mg IV once daily on Days 2 and 3 in non-hospitalized patients, within the first 7 days of symptom onset (SO)

https://www.covid19treatmentguidelines.nih.gov/therapies/antiviral-therapy/remdesivir/

Note: the development of an orally available Remdesivir derivate has been halted by the company Gilead, but is being pursued by Shanghai JunTop Biosciences Co

Nirmatrelvir/ritonavir (Paxlovid®) (PAX) is an orally available combination of 300 mg Nirmatrelvir (= the SARS-CoV-2 main protease inhibitor) + 100 mg Ritonavir (inhibits liver cytochrome P450 enzymes, preventing the metabolism of protease inhibitors).

Key RCT: Hammond NEJM April 2022 https://www.nejm.org/doi/pdf/10.1056/NEJMoa2118542?articleTools=true)

Efficacy Paxlovid reduced the risk of hospitalization or death 89% if taken within 3 days of symptom onset:

0.77% (3 of 389 patients) hospitalized in the Nirmatrelvir group, with 0 deaths,
as compared with 7.01% (27 of 385 patients) in the placebo group, with 7 deaths.

NNT = 18

NIH Recommendation: Paxlovid twice daily for 5 days within 5 days of SO

https://www.covid19treatmentguidelines.nih.gov/therapies/antiviral-therapy/ritonavir-boosted-nirmatrelvir--paxlovid-/

MOLNUPIRAVIR (MOL) (Lagevrio®) is also orally available and acts on the polymerase, but rather by inducing so many mutations in the viral RNA genome resulting in replication-incompetence.

Because of its action mechanism there is some fear for mutagenicity in host cells (which has only been shown in cell lines in vitro)

Key RCT Jayk Bernal NEJM Feb 2022: https://www.nejm.org/doi/pdf/10.1056/NEJMoa2116044?articleTools=true )

Efficacy: Hospitalization or death at day 29= 6.8% [48 of 709] in MOL group vs. 9.7% [68 of 699] in placebo

NNT = 33

NIH recommendation: 800 mg orally, twice daily for 5 days within 5 days of SO,

only when Paxlovid and Remdesivir are unavailable

https://www.covid19treatmentguidelines.nih.gov/therapies/antiviral-therapy/molnupiravir/

Clearly, in all these cases, the target group are high-risk, nonhospitalized patients with mild to moderate COVID.

Until now, all these antivirals remain equally active to all SARS-CoV-2 variants

Van Geel Antvir Res Feb 2022 https://www.sciencedirect.com/science/article/pii/S0166354222000201?via%3Dihub

Deeper analysis and comparison of the 3 Key RCT with RDS (Gottlieb), PAX (Hammond) and MOL (Jayk Bernal)

With regard to efficacy: PAX > RDS >>>> more efficacious than MOL

RDS

PAX

MOL

The ranking of treatment that is an analogue of the surface under the cumulative ranking (SUCRA) was calculated

using the P-score; a higher probability indicated better treatment

With regard to safety: all three very favorable profile RDS, PAX and MOL each compared to placebo:

Similar risk of any Adverse Effect AEs (OR, 0.90; 95% CI: 0.79–1.03; p = 0.14) and drug-related AEs (OR, 1.43; 95% CI: 0.85–2.41; p = 0.18).
Significantly lower risks of serious AEs (OR, 0.35; 95% CI: 0.16–0.78; p = 0.01) and study drug discontinuation due to AEs (RR, 0.48; 95% CI: 0.32–0.72; p = 0.004)

Conclusion:

RDS, PAX and MOL are safe and efficacious to prevent hospitalization in high risk outpatients with COVID, if started within 5-7 b days after symptoms onset.
RDS and PAX appear more efficacious than MOL. RDS has the disadvantage of IV application.
All 3 are equally effective against variants of concern.

PAR 2: Meta-analysis of RCT and real world experience with REMDESIVIR (RDS)

2.1 Meta-analysis of clinical trials in hospitalized patients

Ref 3: Todd Lee in CMI May 2022 Meta-analysis of effect of Remdesivir on mortality in hospitalized patients from Jan 21 to May 2022 : 8 RCT totaling > 10,000 patients

The risk ratio for mortality comparing RDS vs. control was

0.77 (95% CI, 0.5 - 1.19) in the patients who did not require supplemental oxygen;
0.89 (95% CI, 0.79 - 0.99) for nonventilated patients requiring oxygen;
1.08 (95% CI, 0.88 - 1.31) in the setting of mechanical ventilation.

Conclusion: RDS reduces mortality in patients who require oxygen but are not yet critically ill.

Ref 4: Beckerman Sci Rep Medicine June 2022: Meta-analysis of > 5000 patients in 6 RCT in hospitalized patients requiring oxygen.

Subdivided in Any O2, LFO2 (low flow) and HFO2 (high flow) at early (day 14/15) and late (day 28/29) time points.

RDS improved mortality: mainly early in LFO2 with some additive effect of Baricitinib (anti-inflam. JAK kinase inhibitor)

RDS improved early and late recovery in LFO2

RDS reduced the need for additional oxygen support or mechanical ventilation in all subgroups.

2.2 “Real world “evidence in hospitalized and outpatients

Ref 5: Mehta Int J Infect Dis Feb 2021: A shorter symptom onset to remdesivir treatment (SORT) interval is associated with a lower mortality in moderate-to-severe hospitalized COVID-19 in a real world setting

Retrospective analysis of well-matched 290 “early” (< 9 days) and 86 “late” (> 9 days) RDS : all-cause mortality was significantly lower in patients with SORT interval < 9 days (18 %) vs SORT interval >9 days (n = 33.7%;) p = 0.004.

Ref 6: Piccicacco JAC July 2022 Real-world effectiveness of early remdesivir and sotrovimab in the highest-risk COVID-19 outpatients during the Omicron surge. (Sotrovimab = monocmonal Ab witn remaining activity against omicron)

Retrospective study on high-risk but non-severe COVID treated with RDS or SOTRO (monoclonal Ab) within 7 days of Symptoms Onset, compared to non-treated controls. Primary outcome = hospitalization or emergency department (ED) admission.

Conclusion: Early sotrovimab or remdesivir→ similarly lower likelihoods of a hospitalization and/or ED visit.

2.3. Combination of RDS with Interferon beta (IFN-b) IV in hospitalized patients: timing crucial?

Ref 7: Andre Kalil Lancet Resp Dis Oct 2021 : No benefit of IFN-b in patients with COVID pneumonia

IFN-b + RDS was not superior to RDS alone in hospitalised patients with COVID-19 pneumonia.
Patients who required high-flow oxygen at baseline had worse outcomes after treatment with IFN-b.

Note: treatment started late (8.6 +/- 4.5 days SOT)

Ref 8: AR Tam CID June 2022: Quicker recovery in early treated IFN-B + RDS vs RDS only hospitalized high risk patients

Treatment started early (around day 3 after SO) in patients with high risk (age ≥65 years, history of chronic illnesses, and patients with pneumonia)

Quicker complete alleviation of symptoms ( 4 vs 6.5 days)
Quicker to negative nasopharyngeal swab (6 vs 8 days)

2.4. Drug resistance?

Ref 9: Agnieszka M. Szemiel PLoS Path Sept 2021 In vitro selection for RDS resistance E802D

E802D mutation in NSP12 (= RNA-dependent polymerase):

Associated with partial resistance
Mutant virus remains replication competent, but lower fitness in competition with wild-type
Very low frequency in circulating viruses = 36/844,000 = 0.004 % (while RDS is already extensively used in hospitals, where many sequences are geberated)

Ref 10: Ghandi Nat Comm March 2022 Case report of immunocompromised patient with chronic infection, acquiring RDS resistance mutation E802D

After RDS: viral load increased (upper), the E802D variant emerges (lower left), which is partly resistant (lower right)

Afterwards: Casivirimab +Imdevimab (monoclonal antibodies) resulted in a drop of viral load and improvement of symptoms.

Ref 11: Andreas Heyder Cell Rep Med 2022: Remdesivir-induced emergence of SARS-CoV2 variants in patients with prolonged infection.

Sequencing and variant analysis in 112 longitudinal respiratory samples from 14 SARS-CoV2-infected patients with severe disease progression:

Patients receiving exclusively anti-inflammatory treatment: sporadic and transient novel variants.
RDS treatment increased intra-host genomic diversity and result in the emergence of novel major variant species harboring fixed mutations

Of these provisionally only V166L is suspected to be associated with resistance

Conclusion:

In all these studies RDS seems somewhat less effective in preventing disease progression, especially in outpatients (Piccicacco HR = 0.4) than in the Gottlieb RCT (HR = 0.13)
Early application is crucial. If applied early, maybe additive effect of interferon beta 2 IV
RDS may not be effective in critically ill (high flow oxygen or mechanical ventilation): maybe inflammation dominates the pathogenesis at that point and antivirals less useful?
It is possible that resistant mutations occur in chronically infected immunocompromised patients.

PAR 3: Experience with PAXLOVID

3.1 Real world effectiveness

Ref 12: Scott Dryden-Peterson medRxiv 17 June Paxlovid for early COVID in a large US cohort

Cohort = 30,322 non-hospitalized adults (87.2% vaccinated) over 50 years old with COVID-19 in Massachusetts and New Hampshire during Omicron wave (January 1 to May 15, 2022)

Paxlovid prescribed in 20 % of patients: those were older, had more co-morbidities and were less vaccinated than those not prescribed (groups not matched).

Hospitalization in 40 (0.66%) PAX patients and 232 (0.96%) non-PAX patients:

The adjusted risk ratio was 0.55 (0.28-0.80
Risk reduction greater in unvaccinated and obese patients

3.2 Avoiding drug interactions with Ritonavir component of Paxlovid

Ref 13: François Girardin gives a nice overview in Clin Micr Infect Aug 2022

Ref 14: Ross medRxiv March 2022 studies the frequency of potential drug-drug interaction with PAX in older adults (age >65) with polypharmacy (>5 usual home medications) in Canadian hospitals:

68 % were taking a drug that could be influenced by Ritonavir
21 % with high risk

3.3 Resistance to Nirmatrelvir?

Several groups have induced resistance in vitro with various results

Ref 15: Jochmans bioRxiv June 22: the combination of L50F E166A L167F confers 50 X resistance

Ref 16: Yoyung Zhou bioRxiv June 22 also finds that L50F + E166V confers high (80 X) resistance

Ref 17: Heilmannn bioRxiv July identifies Y54C, G138S, L167F, Q192R, A194S and F305L confer resistance

Ref 18: Iketani bioRxiv Aug 22 finds that

T21I, P252L, or T304I are “precursor mutations”, conferring low level resistance.
Additional mutations such as E166V confer high level (100 X) resistance
But “compensatory mutations such as L50F and T21I are needed to regain fitness

Ref 19: Kai Yang J Med Chem July provides a more structural and evolutionary analysis and identifies M165, L167, P168, R188, and Q189 as crucial amino-acids for interaction with Nirmatrelvir.

Ref 20: Derek Lowe points to the fact that a large number of mutations has been associated with PAX use:: A7S/T/V (n=3), L30F (n=3), M82I/R (n=3), G109E/R/V (n=3), P132L/S (n=4), C145F/R/Y (n=3), D153H/Y (n=3), E166V (n=3), T196A/K/M/R (n=4), W207L/S/del (n=5), A260D/T/V (n=8), D263E (n=3), A266P/V (n=3), and V297A/F/del (n=3); Mpro ORF1ab cleavage site substitutions: Q5324H/R (n=3), A5328P/S (n=6), and T6449I/P (n=3).

None of these substitutions in Mpro gene or cleavage regions occurred in PAXLOVID-treated participants who also experienced hospitalization.

Note that only the E166 mutation has also been seen in the in vitro resistance induction experiments

Thus, the clinical significance of these substitutions is unknown.

Conclusion:

PAX effective in outpatients, but maybe less than in RCT
Care should be taken to avoid potential drug-drug interactions due to Ritonavir.
There are multiple possible pathways to development of resistance and a large number of mutations has been associated with PAX use, but provisionally the clinical significance is not clear.

PAR 4: MOLNUPIRAVIR during Omicron wave

Ref 21: Robert Flisiak Pharmacol Rep Aug 2022: Real‑world experience with molnupiravir during the period

of SARS‑CoV‑2 Omicron variant dominance in Poland

Retrospective set-up of 590 patients Jan-April 2022: 203 MOL treated and 387 no antiviral according to physician’s indication:

Both groups were similar in terms of sex, BMI and age allowing for direct comparisons. BUT: co-morbidities not known.
MOL group less likely to receive anti-inflammatory Dexamethasone, anti-IL-6 and Baricitinib (JAK inhibitor)

Results:

Significantly less 28 day mortality: (9.9 vs. 16.3%), especially in 80+ (14.6 vs. 35.2%).
Less frequent need for oxygen supplementation (31.7 vs. 49.2%).
No difference in mechanical ventilation or hospital duration.

Ref 22: Rongron Zhou Front Pharmacology June 2022: Small RCT of MOL in 77 mild/moderate omicron patients vs 31 controls in China (within 5 days of symptoms)

Clearly: not the normal target group with high risk. Primary aim was virological effect.

Results: Viral RNA clearance more rapid in MOL group

On day 5: 18.4 % in MOL vs 0 % in control
On Day 8: 40.8 % vs 6.4 %

PAR 5: Paxlovid versus Molnupiravir in “real world”

5.1 Comparing effectiveness: PAX seems superior to MOL

Ref 23: Gentile medRxiv 25 Aug 2022: PAX vs MOL in mild/moderate COVID with high risk during omicron in Naples

Set-up: retrospective analysis of 257 patients (95 % vaccinated) of whom 57 % MOL and 43 % PAX treated

During the 14-day follow-up: PAX seemed better outcome

Four hospitalizations in MOL (2.1%) and 1 in PAX (0.9%) group.
Only one patient (MOL) died.
Median time-to-negativity of nasal swab was 8 days in PAX vs. 10 days in MOL p<0.01).
37 adverse drug reactions (mainly dysgeusia, diarrhea, and nausea) (12.1%). 2 MOL terminated treatment

However: patients not matched: MOL patients older, lower BMI and more chronic heart disease

Ref 24: Terry Yip CID June 2022 : Impact of the use of Paxlovid or Molnupiracir on the risk of hospitalization in community during omicron (Feb-March) in Hong-Kong

Set-up: Retrospective cohort of almost COVID 94,000 patients of whom 5,808 used MOL and 4,921 PAX.

MOL users were older, had more comorbidities, lower complete vaccination rate, and more hospitalizations in the previous year than PAX users. Therefore careful matching with resp controls was done.

Outcome (compared to matched non-treated patients)

Primary endpoint = risk on hospitalization: PAX lower HR 0.79 ( CI 0.65-0.95) than untreated, but not MOL 1.17 (CI 0.99-1.39).

Secondary endpoint: a composite of intensive care unit admission, invasive mechanical ventilation use, and/or death: no difference between PAX, MOL and non-treated.

Conclusion: Use of nirmatrelvir/ritonavir but not molnupiravir was associated with a reduced risk of hospitalization in real-world non-hospitalized COVID-19 patients

Ref 25: Carlos Wong medRxiv 26 May 2022 Real world effectiveness of Monupiravir and Paxlovid in ambulatory COVID patients during omicron BA.2.2 wave in Hong-Kong (27 Feb – 3 May 2022)

Set-up:

Early initiation within 5 days of symptoms: about 5,500 started on Molnupiravir (MOL) or Paxlovid (PAX) amongst over 1 million COVID cases.
Observational retrospective 1 case vs 10 matched controls.
Mainly elderly (80 % > 65) with comorbidities and mostly unvaccinated. (see Table 1)

Results:

All cause mortality reduced by 39 % with MOL and 75 % with PAX, each compared to matched controls
Hospitalization risk reduced by 31 % by PAX, but not significant by MOL
In-hospital progression reduced by 36 % with MOL and 53 % with PAX
The need for invasive ventilation might be reduced among MOL users, but not for PAX compared to controls

Ref 26: Wong medRxiv 13 July 2022 A similar study in hospitalized non-oxygen requiring patients: Real-world effectiveness of early molnupiravir and paxlovid among hospitalized, non-oxygen-dependent COVID-19 patients on admission during Hong Kong’s Omicron BA.2 wave

Set-up: Observational case-control amongst > 40,000 hospitalized patients of whom 1,856 MOL users, 890 PAX users and 2,746 control patients: mainly very old (mean 80 years); 90 % unvaccinated

Results: Use of antivirals associated with

Lower all-cause mortality: reduction of 66 % by PAX and 52 % by MOL
Less disease progression: reduction of 40 % by both MOL and PAX
Lower need for oxygen: reduction about 30 % by both
Quicker reduction of viral burden: 30 % both

5.2 Combo of PAX + MOL superior?

Ref 27: Kyle Rosenke bioRxiv 5 Sept 22: Favorable effects of combined MOL + PAX against delta in Rhesus macaques

Combined treatment resulted in

Milder disease progression,
Stronger reduction of virus shedding from mucosal tissues of the upper respiratory tract,
Stronger reduction of viral replication in the lower respiratory tract, and
Reduced lung pathology.

Combination reduced clinical scoring

Combination reduced viral replication in upper and lower respiratory tract

Combination therapy reduced lung pathology and SARS-570 CoV-2 antigen load

Conclusion: Most evidence indicates that Paxlovid is more effective than Molnupiravir, especially in the outpatient setting. The macaque experiment clearly suggests that a combination of PAX + MOL may be beneficialL

PAR 6 Evidence of viral Rebound spontaneous and after treatment ?

6.1. Spontaneous

Ref 28: James Hay medRxiv 22 June 22 Impact of medical history on “spontaneous rebound

Cohort of National Baskett Association: 7/1032 (0.7 % ) rebounded

Low pre-booster antibody titers predictive and Omicron BA.1 lower clearance rate

Ref 29: Rinki Deo medRxiv 2 Aug 22 Viral and Symptom Rebound in Untreated 1 COVID-19 Infection

Set-up: participants of placebo in ACTIV study (evaluating several monoclonals): mild/moderate COVID

Result: 12% of participants had viral rebound.

Viral rebounders were older than non-rebounders (median 54 vs 47 years, P=0.04).
Symptom rebound occurred in 27% of participants after initial symptom improvement
The combination of high-level viral rebound to 5.0 log10 RNA copies/mL and symptom rebound after initial improvement was observed in 1-2% of participants.

6.2. After Paxlovid

Ref 30: Raganath CID June 2022: Rebound after Paxlovid for high-risk patients

In a cohort of 483 high-risk patients treated with PAX four (0.8%) experienced rebound of 3 symptoms, which were generally mild, at median of 9 days after treatment, and all resolved without additional therapy.

Ref-31: Eric Dai medRxiv 4 Aug 22 Viral rebound in vaccinated omicron infected subjects

Set-up: 36 participants with previous vaccination and actual BA.2 infection, of whom 11 PAX treated

Results: Rebound in 3/11 PAX treated and 1/24 untreated

However: PAX treated subjects were older ( mean 44 range 25-48) than untreated (mean 16 range 6-66)

PAX treated more comorbidities, more smokers and more obese.

6.3. Rebound after PAX or MOL

Ref 32: Lindey Wang medRxiv June 22 Rebound after PAX or MOL during Omicron in US

Set-up: Starting from 92 million patients from a nationwide US database: 11,270 PAX treated and 2,374 MOL treated

Result: Rebound risk after PAX or MOL is similar after matching

Patients with rebound had more underlying conditions than without rebound

However: no comparison with untreated patients …

Conclusion: Rebound (of viral load and/or symptoms) occurs in a minority of patients, but it is presently hard to say whether it occurs more frequently in PAX or MOL treated vs untreated subjects. Figures vary and studies were not properly controlled. It is, however, clear that rebound occurs more in subjects with underlying conditions.

REFERENCES