A randomized trial of pamrevlumab in patients with COVID‐19 pneumonia

In patients with pneumonia from novel coronavirus SARS-CoV-2, the appearance of interstitial changes on chest high-resolution computed tomography (HRCT) suggests the possible development of fibrosis,1 which has been confirmed in histological samples.2 The persistence of COVID-related interstitial lung abnormalities following hospitalization has been estimated in up to two third of surviving patients3, 4 and can be responsible for residual respiratory impairment.5, 6 As such, there is ground to believe that a combination of anti-inflammatory and antifibrotic therapies could properly address the plethora of pathways involved in the lung damage perpetuated by SARS-CoV-2 infection. A prospective, randomized phase 2/3 study (FIBROCOV) was designed to assess the efficacy and the safety of intravenous administration of pamrevlumab, a recombinant fully human immunoglobulin G1 (IgG) kappa monoclonal antibody binding to Connective tissue growth factor (CTGF), plus standard of care (SOC) versus SOC alone in patients with COVID-19 pneumonia. The study was performed at Fondazione Policlinico Universitario “A. Gemelli” IRCCS and Istituto Nazionale per le Malattie Infettive “Lazzaro Spallanzani”, Rome, Italy (ethics registration number 99/2020). Eligible patients were those with documented SARS-CoV-2 infection, age ≥18 and ≤80 years, evidence of interstitial pneumonia at chest x-ray or HRCT scan and respiratory distress requiring supplemental oxygenation (defined as the ratio of arterial oxygen partial pressure [PaO2] to fractional inspired oxygen [FiO2] of ≤300 mm Hg) and hospitalization. Patients who were intubated and on invasive mechanical ventilation at screening were excluded, while patients with known chronic parenchymal lung disease were allowed in the study. Patients were randomly assigned to receive SOC and pamrevlumab 30 mg/kg by intravenous infusion or SOC only in a 1:1 ratio, according to a pre-generated randomization list. Pamrevlumab was administered on day 1 (day of randomization), day 7 and day 14. According to national guidelines for the treatment of COVID-19 disease, SOC for this study included remdesivir (within 7 days from positive nasopharyngeal swab test for SARS-CoV-2), dexamethasone 6 mg per day for 10 consecutive days (or other dose-equivalent steroid) and low molecular weight heparins. Remdesivir was not recommended in subjects with moderate to severe acute respiratory distress syndrome (ARDS) (PaO2/FiO2 ≤200) undergoing mechanical ventilation or high-flow nasal cannula oxygen therapy. Immunotherapy with tocilizumab could be also considered in patients showing intense inflammatory status and lack of response to steroids and/or remdesivir. Visits were performed every day as per clinical practice until hospital discharge. For all patients, a follow-up visit or phone call for a vital status check was scheduled at week 12. A volumetric chest HRCT was performed at baseline (within 3 days from randomization), at day 14 ± 2 days and week 12 ± 7 days. HRCT scans were visually assessed by two expert radiologists (Giuseppe Cicchetti, Nicoletta Fusco) on consensus, using a validated COVID-19 pneumonia severity score.7, 8 The primary efficacy endpoint for the intention-to-treat population was the proportion of patients alive and not on mechanical ventilation on day 15. The choice of such endpoint was justified by its clinical meaningfulness and for being simple to measure in a study population of hospitalized patients with acute respiratory failure. A sample size of 68 patients was calculated based on Simon's two-stage design9 to detect a treatment difference of 25% in the proportion of responders between the study arms, with 90% power at the two-sided significance of 0.05. An interim analysis was planned after 23 patients in the pamrevlumab arm completed 14 days of treatment (stage 1). Secondary efficacy endpoints included change in the PaO2/FiO2 ratio and change in radiological scores of disease severity at consecutive HRCT scans of the chest, assessed up to 28 days from randomization. Analysis of binary endpoints was conducted using Chi-square test. Analysis of change from baseline in CT score was conducted using a Mixed Model for Repeated Measures (MMRM) model with baseline value, treatment, visit and visit-by-treatment interactions as fixed effects. Change from baseline in PaO2/FiO2 was analysed using analysis of covariance (ANCOVA) with treatment as a fixed effect and baseline value as a covariate. All statistical analyses were performed using the SAS software package, version 9.3. Between 31 May 2020 and 07 June 2021, 42 patients were randomly assigned to receive SOC plus pamrevlumab (n = 19) or SOC alone (n = 23). An interim analysis was performed before the target of 23 patients treated with pamrevlumab was met due to the slowing-down of the recruitment rate, caused by the seasonal flattening of COVID-19 epidemiological curve. Overall, 36 patients (85.7%) completed the study. In the pamrevlumab arm, four patients (21.0%) received only one drug administration and therefore did not complete the 14-day treatment period. Reasons for discontinuation included death in one patient, occurrence of a serious adverse event (not related to the study drug) in one patient, and withdrawal by two patients who were discharged from hospital before completing the treatment period, of which one was also lost to follow-up. In the SOC arm, two patients discontinued the study due to death, while for one patient study completion date was missing. Patients' demographics and baseline clinical characteristics were generally balanced (Table 1). Chronic respiratory comorbidities were infrequent in the study population, as two patients had diagnosis of chronic obstructive pulmonary disease and one patient had diagnosis of interstitial lung disease. The proportion of patients alive and who were not on mechanical ventilation at day 15 was similar in the pamrevlumab group and in the SOC group (17, 89.5% and 20, 87.0%, respectively; 95% CI of difference −17% to 22%, p = 0.8). At baseline, the mean CT severity score for COVID-19 pneumonia was 33.7 (SD 13.7) for the pamrevlumab group and 34.6 (SD 10.9) for the SOC group. On day 14, patients treated with pamrevlumab were more radiologically stable as compared to patients in the SOC arm, who showed an increase in CT severity scores (least squares mean change in CT severity score 0.4 [SE 4.9] and 10.9 [SE 11.8], respectively), although the difference was not significant (−10.5, 95% CI −40.7 to 19.7, p = 0.44). Between day 14 and week 12 there was a significant decrease in CT scores of disease severity in the overall population (difference in least squares mean change in CT severity score: −20.7, 95% CI −39.6 to −1.8, p = 0.03). Radiological improvement was similar in the pamrevlumab and SOC groups (difference in least squares mean change in CT severity score: −2, 95% CI −24.9 to 20.9, p = 0.84). Patients receiving treatment with pamrevlumab showed greater improvements in gas exchange as compared to SOC at day 14 (least squares mean change in PaO2/FiO2 ratio 108.0 [SE 36.1] and 52.9 [SE 26.9], respectively) and at last assessment available (day 7 or after) (least squares mean change in PaO2/FiO2 ratio 103.4 [SE 20.8] and 74.9 [SE 17.8], respectively): however, differences between groups were not statistically significant. Most patients were discharged on or before day 28 (15, 78.9% in the pamrevlumab arm; 18, 78.3% in the SOC arm; difference 0.6%, p = 0.96). No previously unreported, new side effects emerged as compared to previous trials of pamrevlumab. Four deaths were reported during the study (1 [5.3%] in the pamrevlumab group and 3 [13.0%] in the SOC group; difference 7.7%, p = 0.61). Taken together, the findings of this study did not hint at any additional beneficial effect of pamrevlumab in patients with COVID-19 pneumonia as compared to SOC alone, at least not in the short term. Antifibrotic drugs have been so far investigated in chronic fibrotic disorders, and whether they could act fast enough to contrast the rapid evolution of COVID-19 pneumonia is unknown. In a phase 2 trial in patients with IPF, pamrevlumab reduced the decline of forced vital capacity (FVC) starting after 12 weeks of treatment, corresponding to the time taken by blood concentration of pamrevlumab to reach steady state.10 Indeed, the most important limitation of this study is represented by the small population and the recruitment challenges due to the seasonal flattening of the epidemiological curve, which forced the interim analysis being performed before reaching the predefined target. On the other hand, the improvements of radiological involvement, gas exchange and the low mortality rates observed in the overall population across the observation period suggest a favourable course of COVID-19 pneumonia in most patients, which could have further lowered the chances of getting an efficacy signal in such a small population. The treatment and follow-up period were relatively short, as such the effects of pamrevlumab in mitigating the impact of COVID-19 pneumonia on longer term outcomes such as pulmonary function or tolerance to exercise could not be explored. In conclusion, while this study failed to provide conclusive information one way or the other about pamrevlumab, it does highlight the challenges faced in the evaluation, which may be of benefit in the planning of future, more definitive studies. Giacomo Sgalla: Conceptualization (lead); data curation (equal); formal analysis (lead); investigation (equal); methodology (lead); project administration (supporting); writing – original draft (lead); writing – review and editing (supporting). Paolo Maria Leone: Data curation (lead); investigation (lead); project administration (supporting); writing – review and editing (supporting). Gina Gualano: Data curation (equal); investigation (equal); project administration (equal); supervision (supporting); writing – review and editing (supporting). Jacopo Simonetti: Data curation (supporting); formal analysis (supporting); investigation (supporting); methodology (supporting); writing – review and editing (supporting). Alessia Comes: Conceptualization (supporting); data curation (supporting); project administration (supporting); writing – review and editing (supporting). Diana Verdirosi: Conceptualization (supporting); data curation (supporting); formal analysis (supporting); investigation (supporting); methodology (supporting); project administration (lead); writing – review and editing (supporting). Francesco Di Gennaro: Data curation (equal); investigation (equal); project administration (supporting); writing – review and editing (supporting). Anna Rita Larici: Data curation (supporting); formal analysis (supporting); methodology (supporting); software (lead); supervision (supporting); writing – review and editing (supporting). Stefania Ianniello: Data curation (supporting); formal analysis (supporting); investigation (supporting); project administration (supporting); software (supporting); writing – review and editing (supporting). Giuseppe Cicchetti: Data curation (supporting); formal analysis (supporting); investigation (supporting); software (supporting); writing – review and editing (supporting). Nicoletta Fusco: Data curation (supporting); investigation (supporting); software (supporting); writing – review and editing (supporting). Marcello Pani: Project administration (supporting); resources (lead); supervision (supporting); writing – review and editing (supporting). Fabrizio Palmieri: Conceptualization (supporting); methodology (supporting); project administration (supporting); supervision (equal); writing – review and editing (supporting). Luca Richeldi: Conceptualization (equal); methodology (equal); project administration (supporting); supervision (lead); writing – review and editing (lead). This study was supported in part by the generous financial contributions of the Fondazione Carla Fendi, the Ester Fadlun and the Renata Boccanelli families (Rome, Italy). The authors would like to thank the members of the ‘Gemelli Against COVID-19 Group’ and all the healthcare professionals at Fondazione Policlinico Universitario ‘A. Gemelli’ IRCCS and Institute ‘L. Spallanzani’ for their dedication and expertise in treating COVID-19 patients. Finally, the authors would like to thank FibroGen team members Lona Poole, Elias Kouchakji, Xueping Zhang and Jack Li for their support during the conduct of the study. FibroGen Inc., San Francisco, USA supplied the investigational drug (pamrevlumab) for this study. Giacomo Sgalla reports personal fees from Boehringer Ingelheim, outside the submitted work; Luca Richeldi reports personal fees from Biogen, grants and personal fees from Roche, personal fees from ImmuneWorks, grants and personal fees from Boehringer Ingelheim, personal fees from Celgene, personal fees from Nitto, personal fees from FibroGen, personal fees from Promedior, personal fees from Pliant Therapeutics, personal fees from Asahi Kasei, personal fees from Toray, personal fees from BMS, personal fees from RespiVant, personal fees from CSL Behring and outside the submitted work. The other authors have made no disclosures. This study was performed in accordance with the Declaration of Helsinki approved by ethics committee of Istituto Nazionale per le Malattie Infettive ‘Lazzaro Spallanzani’, Rome, Italy—approval: 99/2020. Participant registration took place from May 2020 to June 2021. All adult participants provided written informed consent to participate in this study. Clinical trial registration: NCT05262309 at ClinicalTrials.gov The data that support the findings of this study are available from the corresponding author upon reasonable request.