Antibody response to three‐dose anti‐SARS‐CoV‐2 mRNA‐vaccination in treated solid cancer patients

Solid cancer patients are at higher risk of SARS‐CoV‐2 infection and severe complications. Moreover, vaccine‐induced antibody response is impaired in patients on anticancer treatment. In this retrospective, observational, hypothesis‐generating, cohort study, we assessed the antibody response to the third dose of mRNA vaccine in a convenience sample of patients on anticancer treatment, comparing it to that of the primary two‐dose cycle. Among 99 patients included, 62.6% were ≥60 years old, 32.3% males, 67.7% with advanced disease. Exactly 40.4% were receiving biological therapy, 16.2% chemotherapy only and 7.1% both treatments. After the third dose, seroconversion rate seems to increase significantly, especially in non‐responders to two doses. Heterologous vaccine‐type regimen (two‐dose mRNA‐1273 and subsequent tozinameran or vice versa) results in higher antibody levels. This explorative study suggests that repeated doses of mRNA‐vaccines could be associated with a better antibody response in this population. Furthermore, heterologous vaccine‐type three‐dose vaccination seems more effective in this population. Since this is a hypothesis‐generating study, adequately statistically powered studies should validate these results.

therapy, 16.2% chemotherapy only and 7.1% both treatments.After the third dose, seroconversion rate seems to increase significantly, especially in non-responders to two doses.Heterologous vaccine-type regimen (two-dose mRNA-1273 and subsequent tozinameran or vice versa) results in higher antibody levels.This explorative study suggests that repeated doses of mRNA-vaccines could be associated with a better antibody response in this population.Furthermore, heterologous vaccine-type three-dose vaccination seems more effective in this population.Since this is a hypothesis-generating study, adequately statistically powered studies should validate these results.COVID-19, immunogenicity, third dose, vaccine   What's new?
People with cancer have a higher risk of SARS-CoV-2 infection and severe complications, and cancer treatments can reduce the protective effect of vaccines.Here, the authors tested the antibody response to the third dose of mRNA vaccine in 99 patients undergoing treatment for Davide Dalu and Maciej Tarkowski equally contributed as first authors.
Agostino Riva and Nicla La Verde equally contributed as senior authors.
Grant/Award Number: COVID-2020-12371640; Fondazione Cariplo-Fondazione Umberto Veronesi, Grant/Award Number: 2020-1376; Intramural Program of Fondazione Humanitas per la Ricerca cancer.They found that the third dose stimulated antibody production in nearly all the patients, including those who failed to produce antibodies after the first two doses.Varying the vaccine type for the third dose, from tozinameran to mRNA-1273 or vice versa, increased the antibody response.

| INTRODUCTION
Cancer patients are particularly susceptible to SARS-CoV-2 infection and COVID-19 complications and, therefore, have an increased lethality rate. 1 Immunocompromised patients were excluded from COVID-19 vaccine trials.Thus, since the approval of novel mRNA-vaccines, observational studies were conducted in cancer patients to assess vaccine immunogenicity and efficacy. 24][5][6][7][8] We previously reported that the seroconversion rate after the primary two-dose regimen was impaired in treated cancer patients except for previously virusexposed subjects. 9In this study, we evaluated the immunogenicity of heterologous and homologous vaccine-type three-dose mRNA-vaccination in the same population, focusing on two-dose non-responders and subjects receiving chemotherapy.

| METHODS
This is a retrospective, single-center, observational study based on a convenience sample of patients receiving treatment for solid cancer that was conducted at the Luigi Sacco Hospital in Milan, Italy, between March 2021 and April 2022.The study followed the "Strengthening the Reporting of Observational Studies in Epidemiology" (STROBE) guidelines for cohort studies.The Luigi Sacco Hospital is a secondary referral center.We recruited subjects who attended our outpatient clinic only.We proposed patients to adhere to the study during routine visits.Blood samples were collected at the time of the visit.
Cancer treatment was defined as an ongoing anticancer medical therapy, or an anticancer medical treatment received within the previous 12 months.Obesity was defined as body mass index (BMI) of at least 30.Comorbidity was defined as the presence of at least one relevant disease other than cancer.Hematological malignancies and HIV-positivity were exclusion criteria as known causes of immune depression.All data were retrospectively collected in a standardized format, including cancer diagnosis, cancer stage, anticancer therapy, and clinical features before vaccination.Sex was classified as male or female since all patients considered themselves to belong to only one of the two prespecified gender categories.Data on race and ethnicity were excluded from the analysis since more than 95% of the population was Caucasian.All patients received the primary two-dose cycle of tozinameran or mRNA-1273 following drug-specific recommendations and the third dose was administered not earlier than 4 months after the second dose.The primary two-dose cycle was administered before starting anticancer therapy or on the first day of the treatment cycle.SARS-CoV-2 antibody testing was performed before (pre-V1), 30 days after (post-V2) the primary cycle and 30 days after the booster dose (post-V3).Heterologous vaccine-type regimen was defined as the use of different mRNA vaccine types as primary twodose cycle and booster dose (primary tozinameran and booster mRNA-1273 or vice versa).

| Serological parameters evaluation
We used an FDA-approved fluorescence bead-based multiplex assay (Luminex xMAP ® , Luminex Corp.) to assess serum titer of anti-SARS-CoV-2 immunoglobulin G (IgG).The assay identifies IgGs directed to spike protein (S1), receptor-binding domain (RBD) and nucleo-capsid (N) of SARS-CoV-2.The data were analyzed using xMAP-MULTI-IgG-CoV-2 Assay Software (Luminex Corp., Austin, Texas).Threshold values for N, S1 and RBD were set to 700 median fluorescence intensity (MFI) for all three antigens and 300 for the background, as per manufacturer specifications.

| Statistical analysis
Analyses were performed using the R statistical software, version 4.0.2(R Foundation for Statistical Computing).Median and interquartile range (IQR) was used to describe anti-S1 and anti-RBD levels across groups defined by socio-demographics, clinical features, and information related to the vaccination.Wilcoxon Rank Sum Test or the Kruskal-Wallis Rank Sum Test was used to compare anti-S1 and anti-RBD across groups.Subgroups analysis was conducted using the median and interquartile range (IQR).Multivariate analyses were conducted using median regression with anti-S1 and anti-RBD as dependent variables in separate models.The significance threshold was set at P < .05.

| RESULTS
We enrolled 99 consecutive patients.Considering the major risk factors for reduced antibody response, we report that 62.6% were ≥60 years old, 32.3% were males, 67.7% were in advanced stage of disease and 14.1% were on treatment for lung cancer.
To stratify the iatrogenic risk of infection, we report that the 40.4% of subjects were receiving biological therapy (immunotherapy, cyclin-dependent kinase 4/6 inhibitors, tyrosine-kinase inhibitors and monoclonal antibodies), 16.2% were receiving chemotherapy only and 7.1% a combination of both treatments.The 30.4% of the patient population on chemotherapy were treated with a regimen at intermediate-high risk of febrile neutropenia.
Only in a minority of patients (30.4%) the third dose was injected within 48 hours from the oncologic therapy cycle.Finally, a heterologous vaccine regimen was administered in 25.3% of the subjects.One primary-vaccination non-responder patient developed COVID-19 between the second and the third dose.For complete demographic and clinical characteristics of cancer patients, see Table S1.
Seroconversion after the third dose was reached in 99% of individuals.Antibody levels at post-V3: anti-RBD IgG mean level [min,   max] 18,419 [35, 24,270] MFI; anti-S1 IgG mean level [min, max]   10,289 [29, 17,813] MFI (Figure 1).After the primary vaccination, 22 patients had failed to seroconvert.Among them, 21 seroconverted after the third dose (anti-S1 IgG mean level 8491 [IQR 4652, 12,567]   MFI at post-V3) (Figure 1A).For a better comparison, mean antibody levels at post-V2: anti-RBD IgG mean level 13,902 MFI; anti-S1 IgG mean level 2806 MFI.Among primary cycle non-responders: at post-V2, anti-RBD IgG mean level 87 MFI, anti-S1 IgG mean level 330 MFI; response, but multivariate analysis could not be performed due to the small sample size.A homologous vaccine-type scheme was significantly associated to lower anti-S1 IgG levels (P = .04)(Table 1).In this regard, in multivariate analysis, a primary cycle with mRNA-1273 and tozinameran as booster dose was significantly associated with increased anti-S1 levels (vs homologous, P = .01;vs primary tozinameran/booster mRNA-1273, P = .001)(Table 1, Figure 1B).To favor the interpretation of the main results of this study, in Figure 1 we reported anti-S1 IgG levels only, since anti-RBD IgG levels did not demonstrate as statistically significant in multivariate analyses.
T A B L E 1 Patient variables associated with anti-S and anti-RBD IgG Ab level after the third vaccine dose.
A further analysis showed that non-responders to primary vaccination had a higher increase in their relative amount of anti-S1 and anti-RBD IgG after the third dose if compared to the responder group.

| DISCUSSION
The third dose of mRNA-vaccines seemed to be immunogenic in most of the patients, including subjects who failed to seroconvert after the two-dose primary cycle.The substantial increase in the immune response of primary cycle non-responders could be solely attributable to the effect of the booster dose.At the explorative analysis, none of the considered factors seemed to affect the levels of antibody production except for the use of a heterologous vaccine-type regimen.
To date, few studies reported antibody response to three-dose mRNA-vaccination in treated solid cancer patients using both tozinameran and mRNA-1273. 4,5Considering the results of this explorative study, we hypothesize that a heterologous vaccine-type mRNA-vaccination regimen could enhance the antibody response of treated solid cancer patients compared to a homologous one, as previously reported in the general population. 10This study has limitations.First, the test used was set for the assessment of primary vaccination response.In the literature, antibody levels demonstrated to be significantly higher after the booster dose, if compared to the primary two-dose vaccine cycle, in cancer patients.Indeed, seroconversion rate could have been overestimated due to an abnormally low threshold.Moreover, the more prominent methodological weakness is that the study is retrospective and based on a convenience sample; therefore, it is powered to be hypothesis-generating only.Furthermore, the sample size was believed to be sufficient to allow meaningful analysis; however, the study may not be powered enough to evaluate associations within subgroups due to a limited sample size.In addition, major limitations are, first, the absence of the evaluation of cell-mediated immunity and the antibody neutralizing capacity against new SARS-CoV-2 VOCs (Variants of Concern).However, since anti-S1 IgG levels can be considered as a surrogate of the breadth of neutralization against variants, 11 we report that 20E (EU1), Alfa (B.1.1.7 and B.1.1.7 + E484K), Delta   (B.1.617.2, 21K and 21J) and Omicron (BA.1) were circulating in Italy in the study period. 12Another limitation is the absence of follow-up data about the waning of antibody response after the third dose.Finally, this study lacked a control cohort.Considering historical data in healthy individuals, antibody response to the third dose seems similar compared to cancer patients. 13is study could expand the knowledge regarding the immunestimulation capacity of mRNA vaccines in cancer patients.These informations could be important in the future development of mRNAbased antitumor vaccines.Our study suggests the possibility of enhancing the immune stimulation with the use of different types of mRNA-based antitumor vaccines within the same therapeutic course.These findings could support the design of new experimental studies testing the efficacy of antitumor vaccines.Since our study included patients with a previous SARS-CoV-2 infection, the data reported could provide additional informations regarding "hybrid immunity."This could be an immunological phenomenon of particular relevance in the study of the antitumor immunity, since the immune system of cancer patients has been exposed to tumor antigens in a manner that is similar to what occurs with SARS-CoV-2 antigens in individuals with a previous exposure to the virus. 14 this observational, explorative study of patients treated with anticancer therapy, the third SARS-CoV-2 mRNA vaccine dose appeared to induce an antibody response in previously nonresponders.The small sample size could have hampered the evaluation of possible predictors of response.Heterologous vaccine-type mRNA vaccination seemed to enhance the humoral immune response compared to homologous vaccination.However, since the design of this study is powered to be hypothesis-generating only, prospective and larger studies are needed to validate these results.
Fondazione Umberto Veronesi (Proposal 2020-1376 to DM) and by Intramural Program of Fondazione Humanitas per la Ricerca.The authors gratefully acknowledge their patients and the respective caregivers enrolled in this study.The authors especially thank Joanna Landi for her technical assistance and the volunteers of "Salute Donna Onlus" for their

1
at post-V3, anti-RBD IgG mean level 10,589 MFI, anti-S1 IgG mean level 8491 MFI.In primary vaccination responders: at post-V2, anti-RBD IgG mean level 14,334 MFI, anti-S1 IgG mean level 3514 MFI; at post-V3, anti-RBD IgG mean level 18,664 MFI, anti-S1 IgG mean level 10,803 MFI.Univariate analyses reported that antibody levels after the third dose were independent from age, gender, BMI, comorbidities, cancer stage, cancer treatment, neutropenic potential of chemotherapy, previous viral exposure, timing of third dose and third dose vaccine type.Breast cancer (P = .01)and granulocyte colony stimulating factor (G-CSF) administration (P = .039)reduced the 3rd-dose anti-RBD Change in anti-S1 IgG levels at the three time points in non-responders (A) and in responders (B) to the primary vaccination regimen.Black dots or orange dots: Anti-N1 ≥700 MFI.Black line: homologous regimen.Green dotted line: threshold for seroconversion.MFI, median fluorescence intensity.Orange line: heterologous regimen.Post-V3: after booster dose.Pre-V1: before primary vaccination cycle, Post-V2: after primary vaccination cycle.