COVID-19 “Rebound” associated with nirmatrelvir/ritonavir pre-hospital therapy

Dear editor,

The efficacy of nirmatrelvir/ritonavir, an orally active chymotrypsin-like cysteine protease inhibitor in combination with a CYP3A4 inhibitor, in reducing hospitalisation or death from COVID-19 infection has been demonstrated in a high-risk adult population.¹ Treatment within five days of COVID-19 symptoms was associated with a progressive reduction in viral load compared to the placebo arm; the duration of COVID-symptoms and changes in symptom severity were not reported.¹

The effectiveness of nirmatrelvir/ritonavir in vaccinated individuals or against novel SARS-CoV-2 variants remains uncertain. Case reports are emerging of a recurrence (“rebound”) of COVID-19 symptoms during or shortly after nirmatrelvir/ritonavir therapy without evidence of infection with an alternative variant in vaccinated, immunocompetent individuals.3, 4, 5 No universally agreed definition of “rebound” COVID-19 associated with antiviral treatment currently exists, however, the United States Centres for Disease Control and Prevention describes rebound as a recurrence of COVID-19 symptoms or a new positive viral test after negative testing within two to eight days of initial recovery.⁵

Nirmatrelvir/ritonavir has conditional marketing authorisation in the UK for the treatment of COVID-19 in adults not requiring oxygen therapy and who are considered at risk of hospitalisation and/or death from COVID-19.⁶ In December 2021, the Cardiff and Vale University Health Board was commissioned to provide a National Antiviral Service (NAVS) with responsibility for the clinical assessment of and supply of oral antiviral medicines to clinically extremely vulnerable patients in Wales who test positive for COVID-19. The first patients began receiving nirmatrelvir/ritonavir from NAVS in February 2022 in accordance with a UK wide clinical access policy.⁷ As of 5 June 2022, the National Antiviral Service (NAVS) has recommended treatment with nirmatrelavir/ritonavir, molnupiravir and neutralising monoclonal antibodies (Casirivimab/Imdevimab or Sotrovimab) for 939, 647 and 1498 patients respectively. To date, NAVS has received three spontaneous reports from service users of apparent recurrence of COVID symptoms associated with new positive antigen lateral flow tests, Table 1 .

Table 1.

Cases describing relapse of COVID-19 symptoms during out of hospital Paxlovid treatment.

Case number	Age (years)	Sex	Comorbidities	Medications	Symptom onset	LFT +ve	Paxlovid duration	Symptoms improved	LFT -ve	Symptoms reoccurred	LFT +ve	Hospitalised
1	55	Male	Rheumatoid arthritis, vasculitis	Rituximab Prednisolone	Day 1	Day 1	Day 2 to Day 6	Day 5 to Day 7	Day 6	Day 8	Day 8	No
2	44	Female	Crohn's disease	Ustekinumab	Day 1	Day 1 Day 2 Day 3	Day 4 to Day 9	Day 5 to Day 11	Day 5 to Day 11	Day 14	Day 14 (x3) Day 15	No
3	59	Female	Vasculitis	Rituximab Mycofenolate Prednisolone Hydroxychloroquine	Day 1	Day 1	Day 2 to Day 7	Day 3 to Day 5	Day 6	Day 19	Day 19	Assessed but not admitted

Two cases were female and one male. Ages ranged from 44 to 59 years. All three cases had underlying active immune-mediated inflammatory disease, for which they were receiving treatment, and had received a full course of COVID-19 vaccination. Nirmatrelvir/ritonavir was started within one to three days of the onset of typical COVID-19 symptoms and a positive lateral flow test. All three cases reported taking nirmatrelvir/ritonavir, as prescribed, for five days.

Symptomatic improvement was reported from three to five days after initial onset and persisted for two to six days until symptoms were reported to reoccur. Lateral flow tests were newly positive eight to 19 days from initial onset. Two out of three cases did not require hospitalisation but the third was referred to her local Medical assessment Unit to exclude a complication of COVID, such as venothromboemobolism or secondary infection, and received sotrovimab for on-going symptoms associated with positive lateral flow tests.

It is unclear if COVID-19 “rebound” associated with nirmatrelvir/ritonavir is a distinct clinical phenomenon. The number needed to treat in the pivotal nirmatrelvir/ritonavir was approximately 18,¹ suggesting that failure to respond to nirmatrelvir/ritonavir therapy may be an explanation. Alternatively, “rebound” may reflect the natural history of certain SARS-CoV-2 variants.⁸ Investigators have reported second peaks in viral load in “rebound” cases,^{2 , 3} however, these do not correlate well with either symptoms or antigen test positivity. The inclusion of untreated controls and other COVID-19 therapies in future, formal, observational studies would address both alternative hypotheses.

In summary, we describe three cases of apparent recurrence of COVID symptoms associated with new positive lateral flow tests in immunosuppressed adults at high risk of severe COVID-19 treated with nirmatrelvir/ritonavir. The number of reported cases represents a small proportion of all those treated (<1%) although we cannot account for unreported cases. The strengths of our conclusions are limited by the strong risk of reporting bias and a lack of viral load or viral genomic data to exclude early re-infection with an alternative or mutant SARS-CoV-2 subvariant. Our adverse reaction information for patients letter has been amended to encourage patients to report recurrence of COVID symptoms during or following treatment.