Thinking more about therapy with convalescent plasma for COVID-19 patients

The severe acute respiratory virus coronavirus 2 (SARS-CoV-2), the etiological pathogen of coronavirus disease 2019 (COVID-19), is spreading in the world. There is currently no specific treatment for COVID-19, and the case fatality rate is as high as 6.8%.¹ Passive antibody therapy, through transfusion of convalescent plasma (CP) from recovered patients with COVID-19, appeared to be a possibly effective treatment for COVID-19. To date, totally seven original studies reported the efficacy of CP therapy in a total of 34 COVID-19 patients (Table 1); six studies reported that CP was effective,^2,7 and the other one appeared to be ineffective.⁸ The US Food and Drug Administration also approved the use of CP in critical COVID-19 patients.⁹ However, we consider that more thinking is required to evaluate the efficacy of CP based on the currently available data.

Table 1. Summarized information of convalescent plasma (CP) used in 34 COVID-19 patients

Reference	Sex/age (years)	Patient condition	Days from onset to CP (ml)	SARS-CoV-2 IgG before CP	SARS-CoV-2 IgG after CP
2	M/70 s	ARDS, MOF	24 (400)	+	+
	M/60 s	ARDS	14 (400)	+	+
	F/50 s	ARDS	22 (400)	+	+
	F/30 s	ARDS	21 (400)	+	+
	M/60 s	ARDS	23 (400)	+	+
3	M/46	Severe	11 (200)	1:160	1:640
	F/34	Severe	11 (200)	N/A	N/A
	M/42	Severe	19 (200)	1:320	1:640
	F/55	Severe	19 (200)	1:160	1:640
	M/57	Severe	14 (200)	1:640	1:640
	F/78	Severe	17 (200)	1:640	1:640
	M/56	Severe	16 (200)	1:320	1:640
	M/67	Severe	20 (200)	1:640	1:640
	F/49	Severe	10 (200)	1:160	1:640
	M/50	Severe	20 (200)	1:640	1:640
4	M/69	Persistent SARS-CoV-2	33, 36, 39 (each 200)	180 AU/mL	184–199 AU/mL
	F/75*	Pulmonary consolidation	33, 37 (each 200)	106 AU/mL	271–258 AU/mL
	M/56^†	Extensive lung lesions	32, 33, 36 (each 200)	72 AU/mL	143 AU/mL
	F/63^¶	With Sjögren syndrome	40 (200)	205 AU/mL	217–213 AU/mL
	F/28	Reappearance RNA +	42 (200)	162–170 AU/mL	213 AU/mL
	M/57^§	Ground glass opacities	49 (200)	217 AU/mL	N/A
5	F/69	ARDS, septic shock	23 (200); 33 (400); 34 (300)	N/A	N/A
	M/55	ARDS	16 (200)	N/A	N/A
	M/73	ARDS, MOF, septic shock	20–46 (2400, 8 times)	N/A	N/A
	F/31, pregnancy	ARDS, MOF, septic shock	24 (300)	N/A	N/A
6	M/71	ARDS	22 (500)	N/A	N/A
	F/67	Severe	7 (500)	N/A	N/A
7	F/64	Hypertension and diabetes	20 (200)	N/A	N/A
8	5 males; 32, 61^††, 62^††, 76^††, 83^††	Critical conditions (all respiratory failure, 5 ARDS, 4 secondary infection, 3 septic shock)	18–23^§§ (median 300)	N/A	N/A
	F/30^††, pregnancy	Respiratory failure	27^§§ (400); 29^§§ (200)	N/A	N/A

COVID-19, coronavirus disease 2019. SARS-CoV-2, severe acute respiratory virus coronavirus 2. ARDS, acute respiratory distress syndrome. MOF, multiple organ failure. N/A = not available.

*SARS-CoV-2 RNA negative 13 days before CP.

^†SARS-CoV-2 RNA negative 12 days before CP.

^¶SARS-CoV-2 RNA negative 16 days before CP.

^§SARS-CoV-2 RNA negative 4 days before CP.

^††Fatal.

^§§Days from positive SARS-CoV-2 RNA to CP administration.

So far the six studies showing that the CP therapy in COVID-19 patients is effective just enrolled a total of 28 COVID-19 patients with various clinical conditions (Table 1), with all 28 patients survived.^2,7 Of the 28 patients, 20 were already positive for anti-SARS-CoV-2 IgG before transfusion of CP, and 8 others were unknown because the antibodies were not tested (Table 1). In another study that enrolled six patients treated with CP, five patients died (Table 1).⁸ Since nearly all patients received CP at the late stage of illness and anti-SARS-CoV-2 IgG develops within median 14 days after the illness onset,¹⁰ it was likely that those patients without antibody test before administration of CP were also anti-SARS-CoV-2 IgG positive. Usually, the neutralizing antibodies in CP are considered to be the active components, and transfusion of CP in individuals with negative or very low level of antibodies can prevent or treat infectious disease. Since CP was administered in the presence of actively produced anti-SARS-CoV-2 IgG in the patients,^2,7 the efficacy of CP should be carefully evaluated. Generally, an adult has 2000–3000 ml plasma (accounting for 4–5% of body weight). Compared with the amount of actively produced anti-SARS-CoV-2 in the patients, the specific IgG amount in 200–500 ml CP used in the patient (Table 1) is considerably small. Indeed, the measurement of anti-SARS-CoV-2 IgG levels after the transfusion of CP in most patients showed no or very little increment in most patients (Table 1),^2,4 and 2–4 fold increment of neutralizing antibody in some patients after transfusion of CP^2,4 was more likely resulted from the active production during the natural course of disease.

Although the six studies concluded that CP therapy is effective in treating COIVD-19,^2,7 none contained a control or comparison group of patients without use of CP, which appeared to be unreasonable as there were a large number of patients in each of the six hospitals. Because of no control or comparison patients without use of CP in these studies, it is impossible to determine whether the effectiveness in these patients resulted from the use of CP or from the self-resolution in the natural course, as these patients had the antibodies before the use of CP. The reported effectiveness of CP in patients with SARS, pandemic 2009 influenza A (H1N1), avian influenza A (H5N1), or Ebola cannot be simulated in COVID-19 patients because none of the studies in treating these diseases with CP was randomized. Therefore, the randomized trials are essential to attain reliable results to prove the effectiveness of CP in treating COVID-19.

The lessons in applying hyperimmune globulin against cytomegalovirus (CMV) to prevent congenital CMV infection in fetuses showed us that randomized clinical trials are necessary to evaluate the efficacy of CP in treating COVID-19. Numerous case or case series studies reported that CMV-specific hyperimmune globulin applied in pregnant women with primary CMV infection is effective to prevent congenital CMV infection in fetuses.¹¹ However, the randomized, placebo-controlled, double-blind study demonstrated that maternal treatment with CMV-specific hyperimmune globulin dose not modify the course of primary infection during pregnancy and reduce the congenital infection in fetuses but increases the obstetrical adverse events.¹²

As SARS-CoV-2 is still being widely spread and a large number of COVID-19 patients have recovered, currently it is feasible to conduct randomized trials to prove the effectiveness of CP in treating COVID-19. If the therapeutic efficacy of CP is true, other blood components, in addition to the neutralizing antibodies, should also be investigated, because the antibodies can only neutralize virions outside the cells and do not have activity to eliminate the virions inside the cells.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.