Hybrid immunity in immunocompromised patients with CLL after SARS-CoV-2 infection followed by booster mRNA vaccination

TO THE EDITOR:

Patients with chronic lymphocytic leukemia (CLL) are immunocompromised.¹ They are at high risk for developing severe COVID-19^2-4 and mount suboptimal immunity after messenger RNA (mRNA) vaccination,^5,6 but they have slightly conflicting results regarding cellular immunity.^7,8 In a “hybrid immunity” setting, immune protection is influenced by prior infection status and vaccination. Here, we studied humoral and cellular hybrid immunity in a CLL cohort of 29 patients with a history of COVID-19² and 3 consecutive mRNA vaccinations against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (2 patients received 2 doses).

The study was approved by Swedish Ethical Review Authority (www.etikprovningsmyndigheten.se). Written informed consent was obtained before sampling. Serum and saliva antibodies were measured prevaccination, after dose 2, and before and after dose 3 vaccinations. SARS-CoV-2 immunoglobulin G (IgG) antibodies in serum were analyzed using Roche Elecsys qualitative anti-SARS-CoV-2 and quantitative anti-SARS-CoV-2 immunoassays as described elsewhere.^2,5 Saliva was collected using a self-sampling technique, and SARS-CoV-2 IgG antibodies were analyzed as described elsewhere.^9-11 T-cell responses were analyzed by activation-induced marker assay as described elsewhere¹² before vaccination and before and after dose 3. We also applied the interferon gamma (IFN-γ) ELISpot as described elsewhere.^2,7 Experimental methods are described in the supplemental Methods, available on the Blood website.

Twenty-nine patients from a previous COVID-19 study were included.² Patient characteristics and mRNA vaccinations are summarized in Table 1 and supplemental Table 1. Twenty-one patients were untreated, 6 had received CD20 monoclonal antibody–containing treatment >12 months before, and 2 received BTK inhibitor (BTKi) therapy. The variability in time since recovery from COVID-19 and between vaccinations and tests is depicted in Table 1. Breakthrough infections were not observed during the study (until 21 December 2021). Immune responses are shown in Figure 1 (entire cohort) and supplemental Figure 1 (patients without missing data points). Serology results are shown in Figure 1A. Ninety-three percent of patients (26/28) were seropositive before the first vaccination with titer levels available in 25 patients (Figure 1A). The median spike-specific antibody titer was 47 U/mL (range <0.8-911), including 3 seronegative patients. After 2 doses, 96% (24/25) were seropositive and the median antibody titer had increased to 17 208 U/mL (range <0.8 to >25 000, interquartile range [IQR] 2793 to >25 000) (P < .0001). This titer fell to a median of 6825 U/mL (range <0.8 to >25 000, IQR 2532 to >25 000) prior to dose 3 and then increased to a median of 24 956 U/mL following dose 3 (range <0.8 to >25 000 U/mL, IQR 4219 to >25 000) (P < .0001) (upper detection limit 25 000 U/mL). Nucleocapsid-specific antibody levels tended to decline over time (Figure 1A).

Next, we analyzed salivary IgG, and 95% of patients (19/20) had spike-specific IgG prior to dose 3 (Figure 1B), with a moderate correlation between spike antibodies in serum and saliva (r = 0.4622, P < .0006) (supplemental Figure 2A). Salivary spike IgG levels appeared with slower kinetics and increased before dose 3 (P < .05) (Figure 1B). The average level before and after dose 3 was comparable to naive healthy controls after 2 vaccine doses.¹⁰ Nucleocapsid reactivities were stable with a transient rise before dose 3 (P < .05) (Figure 1B).

Thereafter, we analyzed T-cell responses against wild-type Wu-Hu1 and Omicron (BA.1) as described elsewhere.¹² Thirteen patients were included, with memory responses shown in Figure 1C. After dose 3, the spike-specific CD4⁺ T cells (P < .05) (Figure 1D) and CD8⁺ T cells (P < .01) increased (Figure 1E) with a similar magnitude for Wu-Hu1 (wild-type) and Omicron (BA.1).

IFN-γ ELISpot analysis showed no significant differences following dose 3 (Figure 1F), although spike-specific responses tended to increase and non-spike-specific membrane, envelope, and nucleocapsid responses declined over time.

Additionally, we also made paired analyses of results in patients without missing data points (supplemental Figure 1). Almost identical and statistically significant changes over time were observed in this group compared with the entire cohort for spike antibodies in serum (n = 19, supplemental Figure 1A) and saliva (n = 14, supplemental Figure 1B) and for CD8⁺ T cells (n = 7, supplemental Figure 1D). The change in CD4⁺ T cells (n = 7, supplemental Figure 1C) no longer reached statistical significance. In contrast, the change in nucleocapsid antibody levels was more pronounced than in the entire cohort, decreasing in serum (n = 13, supplemental Figure 1A) and increasing in saliva (n = 14, supplemental Figure 1B).

Patients with CLL have shown low anti-spike titers following mRNA vaccination against SARS-CoV-2^5,6 even after 3 doses.¹³ Primary COVID-19 infection resulted in higher titers and T-cell responses,² although, as shown here, at low levels at the time of first vaccination. The present results mimic those observed of hybrid immunity in otherwise healthy individuals that a combined effect of infection and vaccination results in robust humoral and cellular anti-SARS-CoV-2 immunity.¹⁴ 

To the best of our knowledge, the present study is the first to report on hybrid immunity in patients with CLL. The median serology titers of 17 208 to 24 956 U/mL after 2 and 3 doses were not affected by the analytical range of the assay even though several patients had serology titers above the upper level 25.000 U/mL (Figure 1A). Titers are markedly higher than the median of <100 U/mL that were found after 2 vaccine doses in non-COVID patients with CLL who participated in an earlier prospective vaccine trial.⁵ This includes abundant local immunity in saliva.⁵ The plateau at dose 3 in the present study is in line with 4 vaccine doses in healthy individuals.¹⁵ Saliva antibodies followed a slower time kinetics compared with serum, not being significantly increased until predose 3. Whether this is due to rebound infection limited to the oral cavity¹⁶ is unknown but partly supported by the increase in nucleocapsid-specific IgG. Robust spike-specific CD8⁺ T-cell responses occurred after vaccine dose 3, in line with a report on patients with multiple sclerosis receiving 3 vaccine doses.¹⁷ T-cell immunity was also assessed against Omicron (BA.1), with higher CD8⁺ T-cell magnitudes after 3 doses in the CLL cohort than in healthy non-COVID-19 individuals after 2 mRNA vaccine doses.¹² 

There are limitations with the present study. The cohort is limited, most patients had early-stage CLL, and few had ongoing therapy, which may affect immune responses favorably. The variation in time between COVID-19 and the start of vaccination and between vaccinations may affect the magnitude of the immune responses. Neutralizing antibodies were not measured even though we found a strong correlation between serum and saliva IgG levels with neutralization in the same patients earlier² and confirmed by others.¹⁸ Finally, there was no control group tested in parallel with this real-world cohort, and the groups studied (healthy donors and CLL patients) were not COVID-19 convalescents and received only 2 vaccine doses at the time of reporting.^5,7,10,12 

Hybrid immunity was recently reported to confer long-lasting protection from severe disease in healthy persons^19-21 even though not preventive against Omicron.²² The serial measurement of systemic B- and T-cell responses were spike restricted. Nucleocapsid-directed immune responses were relatively stable over time, albeit with a slow decline, 9 to 20 months after COVID-19. Also, nucleocapsid antibody levels were stable or showed a slight increase in saliva, which may serve as a first-level defense barrier against reinfection by ancestral SARS-CoV-2 strains.^16,23 In conclusion, we demonstrate robust hybrid immunity in serum, saliva, and the T-cell compartment in patients with CLL who received 3 doses of mRNA vaccine following COVID-19 infection. The results are encouraging in the context of immunocompromised patients who have recovered after COVID-19 and need continuous protection against new SARS-CoV-2 variants of concern. To obtain protection, patients who remain seronegative shall be offered available anti-SARS-CoV-2 preventive therapies.