Antibody formation against SARS-CoV-2 in imatinib-treated COVID-19 patients.

Dear editor:

The article by Gobbi and colleagues exemplifies the general interest in understanding the dynamics of immune response to SARS-CoV-2 vaccine, which is currently being widely investigated. However, the study of this response after COVID-19 infection should not be left aside, particularly in those patients who have been treated with drugs that may alter the normal functioning of the immune system. This is the case of imatinib, the principal BCR-Abl inhibitor for the treatment of chronic myeloid leukaemia (CML), which has been proposed as a potential treatment for SARS-CoV-2 infection, as it may exhibit both antiviral and immunomodulatory properties. Indeed, a recent randomised controlled trial on the use of imatinib in severe COVID-19 has yielded encouraging results. However, concerns have been raised regarding possible interferences of this kind of tyrosine kinase inhibitors (TKI) with B-cell response.

Given the increasing attention on imatinib in COVID-19, it is a matter of interest to assess the presumed negative effect of this drug in the production of antibodies against SARS-CoV-2 in real-world patients. Thus, we have analysed the first 30 consecutive subjects from the ongoing COVINIB clinical trial, which evaluates the activity of imatinib and baricitinib in COVID-19 inpatients (NCT04346147), who were treated with the former drug. According to the study protocol, patients assigned to this group received imatinib 400 mg daily for 7 days as part of their treatment for COVID-19. Nuclear acid amplification test (transcription-mediated amplification or real-time polymerase chain reaction) for SARS-CoV-2 in nasopharyngeal swab was positive and pneumonia was confirmed by chest radiography in all patients before entering the clinical trial. Epidemiological, clinical, radiological and laboratory data were collected, as well as the main concomitant treatments for COVID-19 during follow-up. SARS-CoV-2 serology was performed at the last control visit, which was scheduled at 70 ± 5 days from inclusion. In this regard, serum antibodies against SARS-CoV-2 were measured using Maglumi 2019-nCoV chemiluminescence immunoassay for specific IgG against viral spike and nucleocapsid antigens (Snibe Diagnostic) and/or Access SARS-CoV-2 IgG Antibody Test for IgG against the receptor binding domain of the S1 protein subunit of the viral spike (Beckman Coulter). Immunoglobulin concentrations were obtained in AU/mL, and the result was considered positive or negative according to manufacturer's instructions.

The main baseline characteristics of the patients are included in Table 1 , as well as clinical and laboratory data regarding SARS-CoV-2 infection. To describe quantitative data, mean and standard deviation or median and interquartile range were used as appropriate, while percentages were calculated for qualitative data. 76% of patients were male, and the mean age reached 54.6 ± 7.8 years. Three subjects (10.0%) were current smokers and 16.7% had a previous history of alcohol consumption. The prevalence of dyslipidaemia, hypertension and diabetes mellitus was 40.0%, 30.0% and 16.7%, respectively, and the most common preadmission medications were statins (23.3%), angiotensin-converting enzyme inhibitors (16.7%) and oral hypoglycaemic drugs (16.7%). Four patients had been previously diagnosed with obstructive sleep apnea syndrome, and one subject had asthma. The mean of days from symptom onset to initiation of imatinib was 7.2 ± 2.0. Median C-reactive protein and interleukin-6 levels at imatinib start were 8.5 mg/dL (interquartile range 12.5–4.6 mg/dL) and 26.4 pg/mL (interquartile range 53.9–16.0 pg/mL), respectively. Mean lymphocyte count at that moment was 1.1 ± 0.3 × 109/L, while mean D-dimer value was 501.3 ± 208.5 ng/mL. Regarding concomitant therapies for COVID-19, dexamethasone, tocilizumab and remdesivir were administered to 70.0%, 30.0% and 6.7% of patients, respectively. Prophylactic enoxaparin was received by 93.3% of subjects. Only one patient (number 16) required intensive care unit admission, and all subjects were alive at the conclusion of the study. Anti-SARS-CoV-2 IgG positivity reached 100% at the end of follow-up; no patient had received any dose of any SARS-CoV-2 vaccine. Neither readmissions related to reactivation of COVID-19 nor reinfections were documented.

Table 1

Baseline characteristics, clinical data and main laboratory results.

Patient	Age (yr)	Sex	Day of symptoms*	Oxygen therapy		Laboratory results at imatinib onset				Other therapies	Days of hospitalization
				Day 1 of imatinib	Highest†	C-reactive protein (mg/dL)	Interleukin-6 (pg/mL)	Lymphocyte count (x109/L)	D-dimer (ng/mL)
1	47	M	9	No need	No need	4.1	3.2	1.3	392	–	8
2	56	M	8	LFNC	LFNC	16.4	53.9	0.8	349	DXM, TCZ	9
3	57	M	3	No need	LFNC	14.4	–	0.7	328	DXM	7
4	61	M	10	No need	No need	9.8	26.8	1.2	729	–	8
5	53	M	6	No need	No need	2.5	–	0.6	952	DXM	7
6	49	M	5	LFNC	LFNC	4.6	16.0	1.1	302	DXM	7
7	57	F	8	No need	No need	2.7	6.2	1.6	306	–	6
8	64	M	9	LFNC	LFNC	8.5	10.2	1.0	910	–	5
9	56	M	5	No need	LFNC	10.6	21.7	1.4	445	–	7
10	53	M	8	LFNC	LFNC	2.1	37.4	1.4	453	DXM	8
11	50	M	10	No need	No need	6.4	25.7	1.0	409	–	4
12	39	M	8	LFNC	Mask	6.5	26.9	1.5	307	DXM, TCZ	13
13	56	F	6	LFNC	LFNC	14.4	3.6	1.0	709	DXM	7
14	61	F	5	LFNC	LFNC	11.1	1.5	1.2	475	DXM, RDV	9
15	53	M	3	LFNC	Mask	30.6	237.0	0.9	630	DXM, TCZ	18
16	65	M	5	LFNC	IMV	15.1	91.0	1.0	396	DXM, TCZ	41
17	40	M	7	No need	No need	6.2	24.5	1.3	911	–	7
18	65	F	8	No need	LFNC	13.2	29.3	1.2	882	DXM	11
19	53	F	10	No need	No need	7.7	18.6	1.2	571	–	5
20	53	M	9	No need	No need	4.6	12.6	1.0	313	DXM	7
21	50	M	7	LFNC	LFNC	8.6	27.0	1.4	561	–	5
22	51	M	10	No need	LFNC	13.4	109.2	0.6	238	DXM, TCZ	9
23	64	F	8	No need	LFNC	4.5	19.0	1.4	449	DXM	7
24	54	M	9	LFNC	LFNC	10.9	106.0	1.1	445	DXM, TCZ	6
25	60	M	5	No need	LFNC	6.0	18.0	1.2	551	DXM, RDV	7
26	51	M	6	LFNC	LFNC	9.2	145.5	0.8	458	DXM, TCZ	6
27	60	M	6	No need	LFNC	2.1	26.4	1.3	150	DXM, TCZ	12
28	69	F	9	LFNC	LFNC	12.3	–	0.7	505	DXM	8
29	57	M	7	LFNC	Mask	8.4	35.1	0.8	390	DXM, TCZ	12
30	35	M	6	No need	No need	7.1	391.0	1.2	522	DXM	6

Abbreviations: yr = years; M = male; F = female; LFNC = low-flow nasal cannula; Mask = Venturi mask or nonrebreather face mask; IMV = invasive mechanical ventilation; DXM = dexamethasone; TCZ = tocilizumab; RDV = remdesivir. Reference ranges for C-reactive protein, interleukin-6 and D-dimer: < 0.5 mg/dL, < 7 pg/mL and < 500 ng/mL, respectively.

Days from symptom onset to treatment with imatinib.

Highest oxygen therapy support needed after imatinib onset.

Previous evidence from in vitro studies supports the putative deleterious effect of BCR-Abl inhibitors on humoral response, as both decreased serum levels of immunoglobulins and a reduced number of memory B-cells have been observed in CML patients treated with this kind of TKI. , Moreover, treatment with such drugs has also been linked with a suppressed B-cell activation, as well as with impaired isotype switching in a murine model. Although there is not robust correlation between these observations and an increased risk of infection, concerns have been raised about the potential interference of these TKI with antibody formation after vaccination or some infections, including COVID-19. Regarding serological response to SARS-CoV-2 vaccine, studies of hematological patients receiving BCR-Abl inhibitors have shown conflicting data: while anti-SARS-CoV-2 immunoglobulins are detected in most of these subjects after vaccination, , antibody synthesis may be negatively affected in some degree, as one study found that median IgG antibody concentration in TKI-treated patients was around half of that reported in healthy controls.

To our knowledge, this is the first report of non-hematological COVID-19 patients who received treatment with imatinib for this infection and in whom the subsequent production of IgG against SARS-CoV-2 has been assessed. Even though the short duration of imatinib treatment might have prevented this TKI to display its hypothetical detrimental effects on humoral immunity, the finding that anti-SARS-CoV-2 antibodies were present in all analysed subjects after infection contributes to encouraging the research on the role of imatinib in COVID-19.

In conclusion, although our results are preliminary and related to a low number of cases, they suggest that a short course of imatinib does not interfere with IgG formation in hospitalised COVID-19 patients. Further investigation with this and other BCR-Abl inhibitors would be needed to elucidate the potential impact of these drugs on humoral immunity against SARS-CoV-2.

Funding

None

Ethical statement

This study was approved by the Regional Ethics Committee (Comité de Ética de la Investigación con Medicamentos Regional de la Comunidad de Madrid)

CRediT authorship contribution statement

Alejandro Morales-Ortega: Conceptualization, Visualization, Data curation, Formal analysis, Writing – original draft, Writing – review & editing. Ana Isabel Farfán-Sedano: Data curation, Writing – review & editing, Writing – original draft. Aida Izquierdo-Martínez: Data curation, Writing – review & editing, Writing – original draft. Cristina Llarena-Barroso: Data curation, Writing – review & editing, Writing – original draft. Beatriz Jaenes-Barrios: Data curation, Writing – review & editing, Writing – original draft. Nieves Mesa-Plaza: Data curation, Writing – original draft. María Toledano-Macías: . Guillermo Soria Fernández-Llamazares: . Laura Molina-Esteban: . Jaime García de Tena: . Santiago Prieto-Menchero: . Sonia Gonzalo-Pascua: . Juan Víctor San Martín-López: . David Bernal-Bello: Conceptualization, Visualization, Data curation, Formal analysis, Writing – original draft, Writing – review & editing.

Declaration of Competing Interest

David Bernal-Bello is the principal investigator of the COVINIB trial, a non-sponsored randomised trial investigating the therapeutic role of imatinib and baricitinib in COVID-19 patients (NCT04346147; EudraCT 2020–001321–31). The rest of the authors are sub-investigators in this project. All authors declare no other competing interests.