key: cord-0738427-cmm8wqcf
authors: Buxeda, Anna; Arias-Cabrales, Carlos; Pérez-Sáez, María José; Cacho, Judit; Pelegrin, Sheila Cabello; Melilli, Edoardo; Aladrén, María José; Galeano, Cristina; Lorenzo, Inmaculada; Mazuecos, Auxiliadora; Saura, Isabel María; Franco, Antonio; Ruiz-Fuentes, María del Carmen; Sánchez-Cámara, Luis Alberto; Siverio, Orlando; Martin, María Luisa; González-García, Elena; López, Verónica; Martin-Moreno, Paloma Leticia; Moina, Iñigo; Berrio, Esperanza Moral; Moreso, Francesc; Portolés, José María; Santana-Estupiñán, Raquel; Zárraga, Sofía; Canal, Cristina; Sánchez-Álvarez, Emilio; Pascual, Julio; Crespo, Marta
title: Use and safety of remdesivir in kidney transplant recipients with COVID-19
date: 2021-07-06
journal: Kidney Int Rep
DOI: 10.1016/j.ekir.2021.06.023
sha: 338f628e66a0a0105f85c07b5be4e0de2ce65df8
doc_id: 738427
cord_uid: cmm8wqcf

INTRODUCTION: Remdesivir has demonstrated antiviral activity against coronavirus, shortening the time to recovery in adults hospitalized with moderate/severe COVID-19. Severe adverse events such as acute kidney injury (AKI) have been reported. There is little available data on the use and safety of remdesivir in kidney transplant (KT) recipients. METHODS: We present a multicenter cohort study of 51 KT recipients with COVID-19 treated with remdesivir. Outcomes and safety were assessed. RESULTS: Mean age at diagnosis was 60 years, with a median time since KT of 4.5 years. Mean time since admission to remdesivir was 2 days. Twenty-eight patients (54.9%) required mechanical ventilation (19 non-invasive). Mortality was 18.9%, markedly higher if ≥65 years-old (45% vs. 3.2% in younger patients). AKI was present in 27.7% of patients, but in 50% of cases it was diagnosed before treatment. Remdesivir did not require discontinuation because of adverse events in any case. We did not find significant hepatoxicity or systemic symptoms resultant from the drug. CONCLUSIONS: In our cohort of KT recipients, remdesivir was well tolerated and safe in terms of renal and hepatic toxicity, but randomized trials are needed to assess its efficacy.

Since March 11, 2020, Spain and many other countries have seen a three-wave pattern in reported 75 cases of severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) coronavirus disease 2019 76 (COVID-19) 1,2 . As of February 17, 2021, more than 3,105.000 cases of COVID-19 and 55.000 deaths 77 due to the infection have been reported in Spain, being the seventh most affected country 1,3 . 78

Kidney transplant (KT) recipients appear to be at notable high risk for critical COVID-19 illness due 79 to chronic immunosuppression, age, coexisting comorbidities, and frequent contact with healthcare 80 facilities 4-8 . Among 35.183 prevalent KT recipients in Spain 9 , almost 1700 cases of COVID -19 have 81 been reported by the Spanish Society of Nephrology (S.E.N.) registry. This results in a global incidence 82 of 4.8% 10 , similar to 6.1% reported in the general population 1 . 83

Treatment for COVID-19 has been a matter of debate during the pandemic spread, including both 84 antiviral and anti-inflammatory drugs [11] [12] [13] [14] [15] [16] [17] . Remdesivir is a prodrug of an adenosine analog that has 85 demonstrated antiviral activity against a broad range of RNA virus families, including coronaviruses. 86

This drug has shown nanomolar in vitro activity against SARS-CoV2 in human airway epithelial cells 87 and clinical and virologic efficacy in a primate model 18 . Some clinical trials showed that remdesivir 88 was superior to placebo in shortening the time to recovery in adults hospitalized with COVID-19 16,17 . 89 However, patients with severe acute kidney injury (AKI), chronic kidney disease (CKD) with an 90 estimated glomerular filtrate rate (eGFR) <30 ml/min per 1.73 2 , or those on renal replacement therapy 91 (RRT) have been excluded from all remdesivir trials. There are concerns about the potential toxicity 92 of remdesivir in patients with renal impairment related both to dose-dependent direct tubular damage 93 produced by remdesivir in experimental studies and the potential accumulation of its sulfobutylether-94 ß-cyclodextrin (SBECD) carrier 19 . 95

The most common adverse events described in patients treated with remdesivir are increased hepatic 96 enzymes, diarrhea, rash, renal impairment, and hypotension 20,21 . Particularly, up to 22.8% of patients 97 treated with remdesivir have been described to develop AKI 21, 22 . Nevertheless, randomized clinical 98 trials presented a comparable incidence of AKI between remdesivir and placebo 16, 23 . Moreover, in recent studies remdesivir was well tolerated in patients with AKI and CKD, including those on 100 RRT 24, 25 . Assessment of drug-induced nephrotoxicity is particularly challenging as COVID-19 also 101 bears frequent renal complications 26-28 . 102 Remdesivir has scarcely been used in KT patients with 7, 24, 29 . KT recipients have only one 103 functioning kidney most of the time with reduced eGFR compared with the general population, and 104 frequently receive calcineurin inhibitors which compromise the glomerular blood flow. There are 105 currently minimal available data on the use and safety of remdesivir in this at-risk population 24 

The characteristics and variables included in the COVID-19 registry have been previously reported 8, 30 . 125

More specific and detailed data regarding demographics (recipient age, sex, race, and comorbidity), 126 time after KT, time between onset of symptoms and admission, epidemic waves, respiratory situation 127 (including arterial partial pressure of oxygen/fraction of inspired oxygen ratio (PaO2/FiO2 or PaFI), 128 acute respiratory distress syndrome (ARDS), oxygen saturation (SpO2), pneumonia demonstrated by 129 (defined as PaFI <300) and/or bilateral pneumonia with rapidly increased levels of inflammatory 136 markers (such as C-reactive protein [CRP] , IL-6, D-dimer, ferritin, or lactic acid dehydrogenase 137

[LDH]). Most patients had given oral informed consent before remdesivir administration. Remdesivir 138 dose regimens were based on the protocol of each hospital. The most common dosing regimen was 139 200 mg on day 1, followed by 100 mg on days 2 through 5. 140

Obesity was defined as body mass index (BMI) >30 Kg/m 2 . The immunosuppression regimen 141 corresponds to that received at the time of COVID-19 diagnosis. AKI was defined following KDIGO 142 criteria for stage 1, stage 2, and stage 3 31 . The last serum creatinine (SCr) before admission was 143 considered as baseline to assess the incidence of AKI. Drug-induced liver injury was defined according 144 to a grading system developed by the Drug-Induced Liver Injury Network. Grade 1 or mild was defined 145 as raised serum alanine aminotransferase (ALT) or alkaline phosphatase levels or both, but total serum 146 bilirubin <2.5 mg/dL and no coagulopathy (international normalized ratio or INR <1.5). No patients 147 presented severity grades 2 to 5 32 . Respiratory symptoms included cough, sneezing, and rhinorrhea but 148 excluded dyspnea, which was collected separately. Respiratory status was recorded through 149

PaO2/FiO2 when available (n=21). If data to calculate PaFi index were missing, SpO2 was considered 150 (n=28). 151

Clinical follow-up and laboratory tests were collected at 5 points: admission, when remdesivir was 152 administered, early after remdesivir infusion (median 2 days, IQR 1-4), at discharge and at early 153 follow-up after discharge (median 28 days, IQR 17-57). Additionally, we collected the last SCr value 154 and tacrolimus levels before admission to assess incidence and outcomes of AKI and tacrolimus levels 155 variability. 156

Outcomes were assessed as COVID-19-related mortality or recovery until February 10, 2021. Table 2 summarizes clinical characteristics and laboratory data. Ninety percent of recipients were 190 diagnosed during the second wave. The median time between the onset of symptoms and hospital 191 admission was 3 days (IQR 2-5). There was just one nosocomial SARS-CoV2 infection. The most 192 common clinical manifestations at admission were fever and respiratory symptoms (78.4%), followed 193 by dyspnea (70.6%) and gastrointestinal symptoms (37.3%). Ninety percent of patients had pneumonia 194 on the chest x-ray. Twenty-six patients (51%) presented ARDS or oxygen saturation ≤94% at 195 admission. 196 Analytical parameters at admission showed lymphopenia and elevated inflammatory markers. 197 Regarding anti-COVID-19 therapies, remdesivir was administered at a median time of 2 days after 205 admission. Of the 15 patients who required intensive care unit (ICU) management, remdesivir was 206 generally administered before admission, as the median time since hospital admission to ICU was 4.5 207 days. A trend toward a decrease in the mean duration of hospitalization was seen in patients in whom 208 the drug was initiated within 48 hours of admission (n=27, 12.5 days ) in comparison to 209 those who received remdesivir beyond the first 48h after admission (n=20, 22.5 days [IQR 15-34]) 210 (p=0.07). Similarly, the duration of oxygen therapy in the first group was slightly shorter (10 days [5-211 23] vs. 16 days [10,25], p=0.21). Nevertheless, we did not observe a lower mortality in patients who 212 initiated the drug earlier (11.1% vs. 24%, p=0.21). The most frequent remdesivir regimen was 200 mg 213 on day 1 and 100 mg on days 2 through 5 (56%) (Figure 2) . 214

Most patients received combined treatment with steroid pulses, and up to 70.6% received antibiotics 215 (other than azithromycin) due to confirmed or suspected bacterial coinfections. Tocilizumab was 216 administered in 14 patients, and most of them received the drug after remdesivir treatment, with a 217 median time from admission to the administration of tocilizumab of 5 days. Azithromycin, 218 hydroxychloroquine, and ritonavir/lopinavir were less frequently used (Table 3) . 219 220

Ninety percent of patients required supplemental oxygen. The respiratory failure led to non-invasive 222 mechanical ventilation support in 19 patients (8 of them outside an ICU facility), and 9 required 223 endotracheal intubation (Table 3 ). The fatality rate was 18.9%, though markedly higher in 65 years 224 recipients (45% vs. 3.2%, p<0.001) (Figure 3) . 225 AKI was present in 27.7% of the cohort. Most patients presented with AKI stage 1 (57.1%) and only 226 one patient required RRT with complete recovery of renal function at discharge. Eight of 14 recipients 227 with AKI presented the peak SCr before the initiation of remdesivir, and therefore, renal function 228 abnormalities could not be attributable to the drug. The mean SCr levels at remdesivir administration 229 were 1.4±0.5 mg/dL, with a mean eGFR of 57.5±18.8 mL/min per 1.73 2 . Only 8.5% of patients had 230 eGFR <30 mL/min per 1.73 2 when remdesivir was initiated. AKI after remdesivir initiation happened 231 in 11.7%. Of note, one patient developed T-cell mediated rejection, and another one presented 232 thrombotic microangiopathy. Remdesivir therapy was not required to be discontinued because of renal 233 impairment in any patient. The dynamic profile of SCr after remdesivir treatment is detailed in Figure  234 4a. Tacrolimus trough levels were analyzed before and after remdesivir use in those patients who 235 maintained the treatment during admission (n=22). We found a significant decrease in tacrolimus 236 levels after remdesivir treatment (6.9 [5.7-8.8] vs. 5.2 [4.2-8.2] ng/ml, p=0.04), though tacrolimus 237 dose was reduced in 11 out of 22 patients at admission. 238

Evolution of liver function is depicted in Figure 4 (b, c, d) . Baseline liver function abnormalities 239 (elevated aspartate aminotransferase [AST] and ALT levels) were observed in 4 cases before starting 240 remdesivir. Only two patients were found to have newly occurring grade 1 elevation of AST/ALT and 241 liver function remained stable in 95.7% of cases after the drug administration. In the 4 patients with 242 raised baseline AST and ALT levels, remdesivir therapy was not associated with worsening 243 transaminitis. 244

We observed a significant decrease in CRP levels after treatment (53.6 mg/L [9.1-114.7] vs. 7.2 mg/L 245 [1.4-31.5]), whereas other inflammatory markers remained stable (Table S1 ). Finally, no rash or 246 hypotension episodes were described after remdesivir administration in our cohort. 247

Our study reports the largest cohort of KT recipients with COVID-19 treated with remdesivir. Fifty-249 one patients have been analyzed with an overall mortality of 18.9%, being markedly higher in 65 250 years-old recipients. Most cases were diagnosed during the second wave. Twenty-seven percent of 251 patients developed AKI during the study period, but only 11.7% was potentially attributed to the drug. 252

Remdesivir therapy was not withdrawn because of renal impairment. We observed no clinically 253 significant ALT/AST increase after receiving the drug. 254

The evolution of the pandemic in Spain has drawn a three-wave pattern in reported cases of COVID-255 19, with some differences in clinical features 1,2 . Anti-COVID-19 therapies and epidemiological 256 characteristics of KT with SARS-CoV2 infection have changed between waves due to a better 257 knowledge of the disease 10 . Anti-COVID-19 therapies such as hydroxychloroquine or 258 ritonavir/lopinavir, are not used anymore, whereas pulse steroid therapy increases in consonance with 259 the current scientific data 10-14 as well as remdesivir. Developed initially as a possible treatment for 260

Ebola, remdesivir has shown promising results against SARS-CoV2 in reducing hospital stay and 261 mortality 15-17 . However, a latter study has displayed no effect on hospitalized patients with COVID-262 19, as indicated by overall mortality, initiation of ventilation, and duration of hospital stay 33 . The Food 263 and Drug Administration and the European Medicines Agency granted remdesivir emergency use 264 during the first wave 34,35 , but our study shows that remdesivir treatment in KT has been predominantly 265 administered during the second wave. 266

The most common adverse events described in patients treated with remdesivir are increased hepatic 267 enzymes, diarrhea, rash, AKI, and hypotension 20,21 . Serious adverse events were reported in 24.6% of 268 532 remdesivir treated patients versus 31.6% of 516 placebo cases in a randomized trial 16 . The drug 269 was withdrawn early because of adverse events in 9.8% of patients. No KT patient in our experience 270 required remdesivir interruption due to adverse events. 271

The presumed toxicity in patients with renal impairment is attributed to both remdesivir direct tubular 272 damage in experimental studies and SBECD carrier accumulation. SBECD is a solubility-enhancing agent predominantly excreted through glomerular filtration. Studies in animals have associated 274 SBECD accumulation with renal tubule obstruction due to vacuolation at doses 50-fold higher than 275 those typically administered in humans. However, the short duration of treatment (5-10 days) and 276 relatively low concentration of SBECD carrier used in COVID-19 treatment (100 mg of lyophilized 277 powder of remdesivir contain 3 grams of SBECD and 100 mg solution of remdesivir contain 6 grams 278 of the carrier) would be far below SBECD recommended safety thresholds of 250 mg/kg/day 19 Nonetheless, as the authors postulated, the assessment of drug-induced nephrotoxicity is particularly 289 challenging as COVID-19 also displays frequent renal complications, and randomized clinical trials 290 have presented a comparable incidence of AKI between remdesivir and placebo 16,23,26-28 . 291 When we evaluated the potential hepatic toxicity, only two patients were found to have newly 292 occurring grade 1 elevations of AST/ALT. Moreover, remdesivir therapy was not associated with 293 worsening transaminitis in those patients with raised baseline AST and ALT levels. Similar results 294 have been previously described in patients with AKI, CKD, or RRT 24,25 . 295

The use and safety of remdesivir as part of COVID-19 management has been previously reported in 296 COVID-19 KT patients treated with remdesivir. No renal function abnormalities attributable to the 299 drug were observed 24 . 300

More evidence is available from the usage of remdesivir in non-transplanted patients with AKI, CKD, 301 or RRT, suggesting generally good tolerance 19, 24, 25, 36, 37 . Aiswarya et al. analyzed 48 hemodialysis 302 patients showing no association with severe adverse events after administering 2 to 6 doses of 303 remdesivir. They failed to administrate a loading dose of the drug as no evidence of safety in patients 304 with end-stage renal disease was available 25 . 305

Other side effects reported after remdesivir use such as hypotension or rash 16, 17 were not observed in 306 our cohort. 307

As for immunosuppression, we found it of utmost importance to assess tacrolimus blood through levels 308 variability in those patients treated with remdesivir who maintained the drug during admission. As an 309 inhibitor of CYP3A4 in vitro, remdesivir could potentially interact with tacrolimus. However, its 310 capacity to be a perpetrator to clinically significant drug-drug interactions is limited by its rapid 311 clearance if kidney function is stable 38 . Coadministration of remdesivir with this nephrotoxic agent 312 may increase the plasma concentration of remdesivir metabolite, SBECD carrier, or tacrolimus levels. 313

In our cohort, tacrolimus blood trough levels were significantly lower after remdesivir treatment (6.9 314 vs. 5.2 ng/ml), but most patients underwent dose reduction at admission. More importantly, we found 315 no tacrolimus accumulation in the plasma due to remdesivir. 316

Those patients who initiated remdesivir over 48 hours from admission showed a delayed discharge 317 (12.5 vs. 22.5 days since admission), which suggests that early treatment onset could be associated 318 with a better clinical response. Aiswarya, et al. found early initiation of remdesivir (within 48 hours 319 of hospital admission) to be associated with 5.5 days decrease in the mean duration of hospitalization 25 . 320

These results are consonant with what has been reported in the general population 15, 16, 39 . 321

The mortality rate due to SARS-CoV2 infection in Spain has progressively decreased throughout the 322 different waves. As of February 2020, the fatality rate was 1.77% in the general population and 17% 323 in KT population 1,3,10 . Impaired kidney function, immunosuppression, advanced age, pneumonia and early post-KT period have been identified as risk factors for death 4,30,40-42 . Our study cohort showed 325 similar COVID-19 related mortality to the one previously described in the overall cohort of KT by the 326 S.E.N. registry including both hospital and non-hospitalized infected patients. However, previously 327 published experience with remdesivir in COVID-19 patients with AKI, CKD, or RRT displayed higher 328 mortality rates (20-44%) 24,25,36 . Univariate analysis comparing survivors and non-survivors in our 329 cohort identified recipient age over 65 years-old to be associated with an increased risk of death. In 330 this group, the fatality rate reached 45%. Multivariate analysis could not be performed due to the 331 inherent limitations of the sample size. 332

Our study has several limitations. Due to its retrospective nature, some relevant clinical information 333 might be under-reported and residual confounders may be present. The small sample size and the lack 334 of a comparison group preclude establishing definitive associations about safety or efficacy. 335

Nonetheless, the administration criteria were disease severity, and patients who received the drug 336 presented a worse clinical status than an eventual control group not receiving the drug. Therefore, to 337 avoid selection bias, we did not include a contemporary or historical control group because only a 338 placebo-controlled trial can achieve this aim and very few KT patients with COVID-19 were left 339 untreated when center protocols indicated remdesivir. Some patients received other anti-COVID-19 340 treatments (i.e., tocilizumab), and, therefore, we cannot discard the potential influence of these 341 therapies in outcomes. Finally, the patients included received variable dosing of remdesivir. 342

Notwithstanding, to date, this is the largest series focused on KT with COVID-19 treated with 343 remdesivir, and the extensive clinical and analytical work-up contributes to the understanding of the 344 use and safety of the drug in this at-risk population. 345

In summary, we present the largest cohort so far of KT recipients treated with remdesivir for COVID-346 We are indebted to the many physicians and nurses who take care of these patients and are facing the 359 COVID-19 pandemic in our country. -epidemiolgico-vhc-vhb-vih-1315046. (February 27, 2021, 410 date last accessed) 411

Severe Covid-19. N Engl J Med. 2020; 382(19) 

Serum Creatinine (mg/dl, mean ± SD) 1.4 ± 0.5 eGFR CKD-EPI (ml/min per 1.73 2 , mean ± SD) 57.5 ± 18.8 eGFR <30 ml/min per 1.73 2 (n, %) 4 (8.5%) 

Tocilizumab in Hospitalized Patients with Severe Covid-19

Mortality Benefit of Remdesivir in COVID-19: A 420 Systematic Review and Meta-Analysis

Remdesivir for the Treatment of Covid-19 -Final

Remdesivir in adults with severe COVID-19: a randomised, 425 double-blind, placebo-controlled, multicentre trial

Remdesivir for 5 or 10 Days in Patients with Severe 428

Remdesivir in patients with acute or chronic 430 kidney disease and COVID-19

Compassionate Use of Remdesivir for Patients with Severe 433

Compassionate remdesivir treatment of severe Covid-435 Clinical features at COVID-19 diagnosis Fever

Cough, expectoration and/or rhinorrhea (n, %)

Lymphocytes (x10 3 /uL, mean ± SD) (n=51)

Bilirubin (mg/dL, median

Abbreviations: ALT: alanine aminotransferase, ARDS: acute respiratory distress syndrome, AST: aspartate 529 aminotransferase, CRP: c-reactive protein, IL-6: interleukin-6, IQR: interquartilic range, LDH: lactate dehydrogenase