A phase 2 single center open label randomised control trial for convalescent plasma therapy in patients with severe COVID-19

Abstract

A single center open label phase 2 randomised control trial (Clinical Trial Registry of India No. CTRI/2020/05/025209) was done to assess clinical and immunological benefits of passive immunization using convalescent plasma therapy. At the Infectious Diseases and Beleghata General Hospital in Kolkata, India, 80 patients hospitalized with severe COVID-19 disease and fulfilling the inclusion criteria (aged more than 18 years, with either mild ARDS having PaO2/FiO2 200-300 or moderate ARDS having PaO2/FiO2 100-200, not on mechanical ventilation) were recruited and randomized into either standard of care (SOC) arm (N = 40) or the convalescent plasma therapy (CPT) arm (N = 40). Primary outcomes were all-cause mortality by day 30 of enrolment and immunological correlates of response to therapy if any, for which plasma abundance of a large panel of cytokines was quantitated before and after intervention to assess the effect of CPT on the systemic hyper-inflammation encountered in these patients. The secondary outcomes were recovery from ARDS and time taken to negative viral RNA PCR as well as to report any adverse reaction to plasma therapy. Transfused convalescent plasma was characterized in terms of its neutralizing antibody content as well as proteome. The trial was completed and it was found that primary outcome of all-cause mortality was not significantly different among severe COVID-19 patients with ARDS randomized to two treatment arms (Mantel-Haenszel Hazard Ratio 0.6731, 95% confidence interval 0.3010-1.505, with a P value of 0.3424 on Mantel-Cox Log-rank test). No adverse effect was reported with CPT. In severe COVID-19 patients with mild or moderate ARDS no significant clinical benefit was registered in this clinical trial with convalescent plasma therapy in terms of prespecified outcomes.

Fig. 1. Convalescent donor characteristics.

A Correlation between the anti-SARS-CoV-2 spike IgG content of CP and age of donors. B Correlation between the anti-spike IgG content of CP and the efficiency to neutralize spike protein RBD-ACE2 interaction. C A scatter plot showing the relative abundance of the constituents of CP proteome with colour coding for major functional families. D Statistically significant correlation (Pearson correlation, p-value < 0.05, one tailed) of CP constituents with the time since RT-PCR positivity of the donors. Spearmann or Pearson correlation was computed and the corresponding ρ or R and two-tailed P-values are noted when significant correlation is found.

Fig. 2. Clinical trial design and patient recruitment.

Diagram representing the design of the randomized control trial.

Fig. 3. Key pathogen and host characteristics and primary clinical outcomes.

A The whole-genome phylogenetic tree of SARS-CoV-2 viral isolates from SOC and CPT groups. Black and purple tips represent the samples from SOC and CPT groups. B Correlation between the neutralizing antibody content of plasma at T1 and day since the patients were tested positive on RT-PCR. C Correlation between the neutralizing antibody content of plasma at T1 and CT values obtained from SARS-CoV-2 RT-PCR from nasopharyngeal swabs collected at T1 is shown. D Venn diagram showing distribution of different co-morbid conditions among the patients in SOC and CPT arms.

Fig. 4. Primary outcomes compared between two arms of the trial.

A Survival of patients in the two arms from the day of enrolment till day 30 post-enrolment are compared in a Kaplan–Meier curve, for all age groups. Surviving patients were censored on day 30 post-enrolment. For all outcomes Mantel-Cox log-rank test was performed and corresponding P-values are only shown for statistically significant differences. Panels B and C describe the correlation network for 36 cytokines measured in plasma of severe COVID-19 patients, from both SOC and CPT group at T1 and T2 timepoints, respectively. The diameter of the nodes represents extent of enriched abundance compared against a median value derived from patients with mild disease. Edges are shown only for Spearman correlation value of 0.7 and above.

Fig. 5. Secondary and exploratory clinical outcomes compared between two arms.

A Total hospital stay duration of the patients from both arms are plotted in an ascending Kaplan–Meier curve, for all age groups. Deaths and non-remission at day 35 post-admission were censored. B Hospital stay duration of the patients from both arms since the day of enrolment are plotted in an ascending Kaplan–Meier curve, for all age groups. Deaths and non-remission at day 30 post-enrolment were censored. C Representative plot of SpO₂/FiO₂ kinetics (shaded as per the colour scale depicted) of individual patients during hospitalization. The blue box denotes day of diagnosis through RT-PCR for SARS-CoV-2 and the red boxes denote time of death. D The ratio between saturation of O2 in blood (SpO₂) and fraction of O2 received (FiO2), or S/F ratio is plotted for patients in SOC (black line) and CPT (purple line) arms from the day before enrolment till the 7th day post-enrolment, among all age groups. Purple arrows indicate the days when convalescent plasma was transfused. 95% confidence interval is shown for each group.