Prevalence of hypogammaglobulinemia and its management with subcutaneous immunoglobulin supplementation in patients after allogeneic hematopoietic stem cell transplantation- a single- center analysis

Ewa Marta Karakulska-Prystupiuk (  ewa.prystupiuk@uckwum.pl ) Warsaw University Hospital: Samodzielny Publiczny Centralny Szpital Kliniczny https://orcid.org/0000-0002-0201-5041 Jadwiga Dwilewicz-Trojaczek Medical University of Warsaw: Warszawski Uniwersytet Medyczny Joanna Drozd-Sokołowska Medical University of Warsaw: Warszawski Uniwersytet Medyczny Ewelina Kmin Warsaw University Hospital: Samodzielny Publiczny Centralny Szpital Kliniczny Marcin Chlebus University of Warsaw: Uniwersytet Warszawski Karolina Szczypińska Warsaw University Hospital: Samodzielny Publiczny Centralny Szpital Kliniczny Piotr Boguradzki Medical University of Warsaw: Warszawski Uniwersytet Medyczny Agnieszka Tomaszewska Medical University of Warsaw: Warszawski Uniwersytet Medyczny Krzysztof Mądry Medical University of Warsaw: Warszawski Uniwersytet Medyczny Jarosław Bilinski Medical University of Warsaw: Warszawski Uniwersytet Medyczny Grzegorz Władysław Basak Medical University of Warsaw: Warszawski Uniwersytet Medyczny Wiesław Wiktor Jędrzejczak Medical University of Warsaw: Warszawski Uniwersytet Medyczny


Introduction
Despite a considerable reduction in the incidence of non-relapse mortality following allogeneic hematopoietic stem cells transplantation (allo-HSCT) in recent years, infection-related mortality remains a challenge in the post-transplantation care [1,2]. It is related to secondary immunode ciencies that are frequently observed and are multifactorial in etiology. According to NCCN (National Comprehensive Cancer Network) guidelines allo-HSCT recipients (with neutrophil recovery) who require intensive immunosuppressive therapy for graft-versus-host disease (GHVD) are an example of non-neutropenic patients at great risk for infections [7]. Humoral immunode ciencies may result in a quantitative, qualitative, and functional shortage of immunoglobulins [3][4][5][6].
Immunoglobulin supplementation is commonly used in allo-HSCT recipients to prevent and treat infections, but also to modulate acute GvHD. However, there is an ongoing controversy about the bene t of such therapy, dosing regimens, and treatment monitoring of the immunoglobulin supplementation [7][8][9][10][11]. Additionally, immunoglobulin concentrates for subcutaneous administration are being more frequently administered instead of intravenous preparations, and to our best knowledge, there is no report on this subject in the available literature. Therefore, we nd it important to report our experience with subcutaneous immunoglobulins.

Study population
Patients treated at the Outpatient Transplantation Service, transplanted in the years 2012-2019 were included. Patients who died shortly after allo-HSCT, and had not been referred to the Outpatient Service, were not included.

Allo-HSCT
Conditioning was chosen on the discretion of treating physician and depended on the underlying hematological disease. Immunosuppressive treatment was combined of calcineurin inhibitor (cyclosporin or tacrolimus) and antiproliferative drug -either short course of methotrexate or mycophenolate mofetil. All patients with unrelated or mismatched donors received anti-T-cell globulin (2,5-5mg/kg daily) for 2-3 days in conditioning regimen. Immunosuppressive therapy was discontinued after 6-8 months following allo-HSCT if GVHD did not occur. Diagnosis and grading of acute and chronic GVHD were performed based on clinical symptoms and/or biopsies according to established criteria [6]. Grading of acute GvHD was performed according to Glucksberg score, while the severity of chronic GvHD-according to National Institutes of Health (NIH) Consensus Criteria 2014 [12][13][14]. All patients received anti-infective prophylaxis and vaccinations according to the NIH guidelines [8,15] as described in the Supplement.
Doses and types of immunoglobulin, the indications for supplementation therapy All patients were systematically tested (once every 2-3 weeks) and immunoglobulin concentration was determined. Serum IgG was analyzed by the immunoturbidimetric method (Roche biochemical analyzer Cobas 8000, reagent Tina-quant IgG Gen.2) at the Central Laboratory of the Medical University of Warsaw (reference normal level: 700-1600mg/L).
The patients received immunoglobulin supplementation based on the CDC (Centers for Disease Control and Prevention) recommendation and the NIH guidelines [6,8]. Due to the di culty in accessing intravenous preparations, subcutaneous preparations were mostly used. Since the vast majority of patients had transient immunoglobulin de ciencies, IgG was administered solely if the pre-administration IgG concentration was beyond the prede ned threshold.
The doses and frequencies of immunoglobulin supplementation were given according to the individual patients' requirements. Patients with IgG level below 500mg/dL were classi ed as signi cantly de cient and received prophylactic immunoglobulins at a dose of 0.4-0.8g/kg. Patients with episodes of severe infection within rst 12 months after transplantation who had IgG level below 700mg/dL received IgG supplementation at a dose of 0.5-1.0g/kg (divided into 2 doses). Preparations used were: immunoglobulin facilitated subcutaneously by recombinant human hyaluronidase (10% infusion)-HyQvia (Baxalta Innovations GmbH, Austria) or/and solution 165mg/mL -Gammanorm (Octapharma, United Kingdom). The patients, who had received intravenous IgG (IVIg) at the beginning of the study were switched to subcutaneous preparations of immunoglobulin (after at least 28 days of the last IVIg dose). Subcutaneous preparations were always administered under the abdominal skin using a variable rate portable pump according to the producer instructions. Immunoglobulin administration was preceded by hyaluronidase administration in case of HyQvia.

Statistical analysis
This is a retrospective analysis. Medians, means and standard deviations were calculated for continuous and ordered variables. In the search of predictors that increase the need for IgG supplementation, the logistic regression model was used. The following variables were tested: gender, diagnosis being an indication for transplantation, type of conditioning, type of donor, GvHD occurrence (acute and chronic GVHD were considered only if they were documented prior to the landmark of the rst IgG supplementation), age of the recipient on the day of allo-HSCT, age of the recipient at the time of diagnosis, use of corticosteroids before the rst IgG supplementation. Likelihood Ratio test was calculated and presented along with standard error. If not mentioned otherwise, the level of statistical signi cance was set up at 5%.The analyses were conducted using R version 3.6.1. All time-to-event endpoints were computed from the day of allo-HSCT. The Kaplan-Meier estimator and log-rank test were used for overall survival.

Treatment after transplantation
Thirty-four patients (27%) received additional therapy after transplantation (mainly azacitidine) aimed at preventing relapse. The detailed information on this treatment is provided in the Supplement.
IgG was administered to all patients in whom IgG level was below 500mg/dL (41 patients (32.5%)). In 25 of them, it was only prophylactic, while in the remaining 16 also therapeutic because of recurrent infections. Additionally, 12 patients with IgG in the range 500-700mg/dL with accompanying severe and recurrent infections also received IgG supplementation. Therefore, the entire group of patients with infection receiving IgG supplementation comprised 28 patients (16 with IgG below 500mg/dL and 12 with IgG in the range 500-700mg/dL). The median number of IgG administrations was 3. While it is a retrospective and not a prospective study aiming at assessing regular pharmacokinetics, patients were controlled for IgG concentration at various time points after IgG subcutaneous administration. Nevertheless, subcutaneous IgG signi cantly increased IgG concentration that was maintained for 6-8 weeks (Fig.1). Patients originally suffering from ALL(82%) and lymphoma(58%) most frequently met the criteria for IgG supplementation and were accordingly treated. The characteristics of patients treated with IgG are shown in Table S3 (supplement).

Survival analysis
The  7) in the group that never required immunoglobulin supplementation, in the prophylactic IgG group, in the group with recurrent infections treated with IgG because of IgG level below 500mg/dL, and nally, in the group treated with IgG because of recurrent infections and IgG 500-700mg/dL respectively. Relapse was the primary cause of death in 7 out of 7 patients, 4 out of 5 patients, 4 out of 4 patients and 3 out of 4 patients respectively.
The 1-y OS differences were not statistically signi cant with one exception; in the group with recurrent infections treated with IgG because of IgG level within 500-700mg/dL 1-y OS was signi cantly lower compared to the group never required immunoglobulin supplementation (p=0.009). The survival curves for all groups using Kaplan-Meier estimator are presented in Fig.2. Detailed results of the statistically signi cant 1-y OS differences for all groups are shown in Table S5.
Despite the fact that relapse contributed to death of 90% of all deceased patients, various infections complicated the course of relapse (mainly pneumonia or sepsis). Nine out of 10 patients (90%) with BKV infection receiving IgG improved, and thus achieved successful control of hemorrhagic cystitis, while 1 patient died (10%) due to relapse of the underlying disease. Detailed causes of death and diagnosed infections that occurred during the last 14 days of patients' lives are presented in Table S4.

Factors predictive for IgG supportive treatment
In the logistic regression model, the diagnosis of ALL or CLL and the use of systemic corticosteroid therapy were associated with the need for IgG supplementation. The average probability that a patient diagnosed with ALL or CLL will require IgG supplementation was 83.8%, while for patients with other diagnoses the average probability was 39.3% (p=0.0001). The use of corticosteroids was associated with the average probability of hypogammaglobulinemia requiring IgG supplementation in 64.2%, while no corticosteroid use with 31.5% (p=0.005) only.

Toxicity
Adverse events were observed in 53% patients treated with fscIgG, including 49%-grade 1 and 4%-grade 2. In two male patients (5%) transient scrotal oedema was observed, which resolved without any medical intervention. Malaise was reported in 2 patients (4%) and discomfort at the injection site in 10 patients (22%) patients. In 6 patients (13%) who suffered from chronic kidney disease-a clinically insigni cant transient decrease in creatinine clearance was observed. In 3 patients (6%) of all treated patients, the change of the type of IgG products was required due to an allergic skin reaction at the application site.

Discussion
Apart from relapse of the underlying malignancy, secondary immunode ciency is the most common complication of allo-HSCT. All patients are affected to some degree. It results in the occurrence of infectious complications, but also in an increased risk of relapse by altered effectiveness of graft-versusleukemia reaction [3][4][5][6]17]. The occurrence of immunode ciency, due to the altered and insu cient recovery of the immune system, is in uenced by many factors including pre-transplant, like patient's age, type of the underlying medical condition and its treatment (including a growing number of therapeutics targeting B cells), peri-transplant factors related to the transplantation procedure, e.g. the type of the donor, type of conditioning, ATG use and post-transplant factors i.e. multi-stage course of reconstitution, complications especially GvHD, immunosuppressive treatment and preventive treatment used after transplantation [18][19][20][21][22].
While this immunode ciency is complex involving various parts and mechanisms of the immune system with limited possibility for intervention, the de ciency of immunoglobulins, particularly IgG can be corrected with exogenous preparations. Nevertheless expert opinions regarding IgG supplementation after allo-HSCT remain ambiguous [23][24][25][26][27].
In our group allo-HSCT recipients, IgG de ciency was usually transient and the supplementation of IgG was interventional in most cases. Approximately 32% of recipients required incidental IgG supplementation while 10%-regular. Patients received immunoglobulins when they met the prede ned criteria i.e. when either the IgG level was below 500mg/dL, or below 700mg/dL (but above 500mg/dL) with frequent infections or both.
The IgG level used in this study as the indication for preventive IgG supplementation was higher than the one (<400mg/dL) recommended by the ASBMT (American Society for Blood and Marrow Transplantation) and EBMT (The European Society for Blood and Marrow Transplantation) guidelines [6,8]. The reason for this modi cation was the use of subcutaneous preparation of IgG, which e cacy in this population of patients was unknown.
Given possible qualitative and functional shortages of immunoglobulins, we used IgG replacement as well in patients with higher IgG concentration as support therapy in infections. It is hard to assess the effectiveness of IgG reliably because we administered causal treatment, usually at the same time. Nevertheless, it should be emphasized, that IgG supplementation was very effective in the treatment of symptomatic BK Polyomavirus-associated hemorrhagic cystitis (grade II or III according to the ECIL-6 Guidelines) [28,29]-90% of those patients achieved resolution of symptoms without any antiviral treatment.
The median time to the rst IgG administration (4.1 months) in our analysis appears to be consistent with the course of B-cell reconstitution, described in the literature [4,5]. Most of supplemented patients suffered from GvHD. Both acute and chronic GvHD are associated with delayed B cell reconstitution, reduction or lack of B cell precursors in the bone marrow and delays in IgG2 and IgG4 production [5,30,31]. Whether IgG supplementation may be useful to correct de ciencies of different IgG subclasses was beyond the scope of this study.
Cordonnier et al reported an increased risk of fatal veno-occlusive disease/sinusoidal obstruction syndrome (VOD/SOS) in patients with acute GvHD receiving IgG replacement [32]. In our work, the shortest time to rst administration was 34 days and we did not observe this complication.
Survival analysis revealed inferior survival in the group with infections than the prophylactic group but the vast majority of patients who eventually died, succumbed primarily to relapse of their leukemia with infection being only an accompanying condition. It needs another studies to con rm that such groups of patients with lower levels of IgG relapses most often than patients without IgG immunode ciency.
In our study, the occurrence of IgG de ciency was found to be mainly dependent on the disease being an indication for transplantation, with ALL and CLL being the most common, and on systemic use of corticosteroid throughout the treatment process. In patients with B-cell malignancies total and/or functional hypogammaglobulinemia results from disease-related effects on the immune system and from side effects of the treatment [19,26,27]. Anti-B-cell therapy used before transplantation may be responsible for the development of post-transplantation complications. Low pre-allo-HSCT IgG level is a signi cant risk factor for hypogammaglobulinemia after transplantation [18][19][20][21].
In our study, no correlation for IgG replacement was found for GvHD (acute and/or chronic). The results of our analysis may be due to small number of analyzed patients, but also may indicate a stronger impact of steroid therapy than the diagnosis of GvHD on the occurrence of hypogammaglobulinemia.
Corticosteroids are the current gold-standard for the treatment of GvHD [6]. According to published studies [5,31] it is di cult to determine the causality of antibody de ciency following allo-HSCT in a situation where both GvHD and the corticosteroid therapy can lead to antibody de ciency. Greinix et al. described that impairment of reconstitution, observed in patients with GvHD (either acute or chronic) cannot be separated from the possible in uence of corticosteroids [33].
To our best knowledge, this study is the rst to describe the effectiveness and safety of subcutaneous preparations instead of intravenous. This treatment was well tolerated, and we did not observe any signi cant toxicity. According to the literature, IgG has a half-life of about 21 to 29 days following intravenous administration [34]. However, especially in patients with immunode ciencies, interindividual variation has been reported. Taking into consideration variability of the patient's situation (different time after transplantation, different initial diagnosis, the severity of GVHD) and also the fact that they had endogenous IgG production (albeit reduced) these data have shown that subcutaneous IgG signi cantly increased IgG level that was maintained for 6-8 weeks.
While there was no systematic pharmacokinetic experiment, our data collected from patients more frequently tested may unexpectedly suggest that the half-life of subcutaneously administered IgG is longer than intravenously, at least in this setting, in which some endogenous IgG production is present [34][35][36]. This would require veri cation in a prospective study but is interesting and potentially useful. Promising results of the scIgG treatment may grant a possibility of at-home treatment for selected patients after allo-HSCT.
To conclude, over 40% of the adult stem cell recipients may require immunoglobulin supplementation. A vulnerable group of patients to the development of hypogammaglobulinemia after allo-HSCT are patients with B-cell neoplasms, such as acute lymphoblastic leukemia or chronic lymphocytic leukemia, as well as patients treated with corticosteroids after transplantation. Subcutaneous immunoglobulins replacement seems to be a safe and e cacious alternative to intravenous IgG preparations in patients after allo-HSCT.

Declarations Funding
The authors did not receive support from any organization for the submitted work.
Other Authors declare no nancial interests.

Availability of data and material
All patient data was collected based on available medical records. All of the data was analyzed anonymously.
Code availability Not applicable.

Authors' contribution
All authors contributed to the study conception. The rst draft of the manuscript was written by Ewa Karakulska-Prystupiuk and all authors commented on previous versions of the manuscript. All authors read and approved the nal manuscript. All contribution data was attached to the title page.

Ethics approval
The ethical consent was approved by Medical University of Warsaw, Bioethical Commission.

Consent to participate
This is a retrospective analysis of anonymous data contained in medical records.

Consent for publication
Not applicable because there is no identifying information about patients. Overall survival of patients with and without IgG supplementation