Exploring the depths of hypogammaglobulinemia in lymphoid malignancies: pathophysiology, clinical implications, management options, and future directions

Abstract

The treatment of lymphoid malignancies is rapidly advancing with recognition of hypogammaglobulinemia (HG) and increased infection risk with the use of cellular-targeted therapies, such as chimeric antigen receptor (CAR) T cell and bispecific antibody (BsAb) therapy. Underlying adaptive immune dysfunction seen in malignancy, immunoparesis, and prior lines of B cell-targeting therapies also predispose patients to HG and infections prior to more advanced cellular therapies. Clinicians should become familiar with how to evaluate adaptive immunity and causes of HG. Various strategies exist to reduce infections in immunosuppressed patients, including risk-reducing practices, vaccination, prophylactic antibiotics, and IgG replacement therapy (IgG-RT). Future research should focus on identifying biomarkers to help with infectious risk-stratification and identify populations that would best benefit from IgG-RT.

1. Introduction

Hypogammaglobulinemia (HG) in the setting of primary immune deficiency (PI) and chronic lymphocytic leukemia (CLL) has long been established, and although a multitude of chemotherapeutic agents used in the treatment of lymphoid malignancies have been associated with HG, recent advances in immune-effector cell therapy in the form of T cell engagers have brought the clinical implications of HG back to the forefront. Since the first description of agammaglobulinemia by Colonel Ogden Bruton in 1952 and the description of the essential role Bruton’s tyrosine kinase (Btk) plays in B cell development, multiple PIs with some form of B cell absence or impairment, including severe combined immunodeficiency (SCID), common variable immunodeficiency (CVID), and hyper IgM syndromes (HIGM) have been described [1–3]. The treatment approach for these PIs forms the basis of the management of HG seen with lymphoid malignancies and the systemic therapies instituted.

Cellular therapies, such as chimeric antigen receptor (CAR) T cells and bispecific antibodies (BsAbs), have improved clinical outcomes, but have also led to increased rates of secondary HG in hematologic malignancies that parallel those seen in PI. However, heterogenous patient populations and treatment regimens have made this topic challenging to study. This review paper will summarize the general approach to HG with a focus on secondary HG in lymphoid malignancies, mechanisms of immunodeficiency with treatments, and management strategies. Since these novel therapies are increasingly utilized in both academic centers and community practices, this review will expand on the topic of HG seen with immune effector cell therapies (e.g. T cell engagers and CAR T cells).

2. Definition and general approach to HG

From a laboratory perspective, HG is defined as low gamma globulin and can describe any low immunoglobulin component (generally IgG, IgM, and/or IgA). However, in general clinical utility, HG typically refers to low IgG. The American Academy of Allergy, Asthma, and Immunology (AAAAI) Work Group Report in 2022 defined HG as an IgG level < 700 mg/dL [4]. Although HG can be further stratified by level and duration, specific cutoffs for these vary and many guidelines usually consider IgG replacement therapy (IgG-RT) once the IgG level is < 400 mg/dL. From here onwards, HG will also be defined as low IgG, unless otherwise specified.

Immunoglobulin levels (IgG, IgA, IgM, and IgE) are typically measured by nephelometry, where human serum is diluted and then mixed with immunoglobulin-specific antibodies [5]. Quantification of immunoglobulins in the sample can be calculated by comparing the turbidity from precipitated complexes to solutions with known quantities of immunoglobulins. As with any test, laboratory and human error can affect results and an initial finding of HG should be verified with repeat testing.

HG can be primary or secondary (Figure 1). Primary HG is due to underlying genetic defects that lead to PI. Causes of secondary HG can involve malignancy, immunosuppression, protein loss, or various other conditions [4, 6]. Protein loss can include nephrotic syndrome, protein-losing enteropathy (PLE), and lymphangiectasia. Despite appearing as a consumptive issue, nephrotic syndrome and PLE have also been associated with infections [3]. However, the infection risk in nephrotic syndrome is confounded by frequent rituximab use for treatment. The mechanism behind infections in PLE may involve HG and lymphopenia due to gastrointestinal losses. A comprehensive evaluation of various secondary causes of HG can help direct management towards addressing additional underlying contributors of HG.

Figure 1: Causes of hypogammaglobulinemia.

Primary HG is the result of an underlying genetic defect that impairs antibody production. Secondary HG is associated with multiple conditions, medications, immunosuppression, or protein loss.

While HG has been correlated with serious bacterial infections, clinicians should also be aware of other infection risk factors in conjunction with HG. These infectious risks include neutropenia, type of malignancy, prior chemotherapy, hematopoietic cell transplantation, immunosuppressive therapy (e.g. steroids, IL-6 inhibitors), and lymphopenia [7].

3. Definition of infections

The National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) is the main reference used for grading infections in hematologic clinical trials, and grade ≥ 3 infectious adverse events are typically reported [8]. In relation to potential infectious complications that could be reduced with IgG-RT, grade ≥ 3 adverse events include (1) lung infections (e.g. pneumonia) requiring intravenous (IV) antimicrobial intervention and (2) sepsis with positive blood cultures and/or life-threatening consequences necessitating treatment. Other grade ≥ 3 adverse events that may warrant other considerations (i.e. prophylactic antibiotics) include medically significant fungemia and viremia. Infections referenced in this review paper are reported according to CTCAE, unless otherwise specified.

4. Diagnostic testing

Evaluation of humoral immunity begins with checking immunoglobulins (IgG, IgM, IgA) and lymphocyte subset counts (T, B, and NK cells) (Table 1) [4]. Ideally, this should be ordered at baseline prior to immunosuppressive therapy and when the patient is not acutely sick to obtain baseline immune function, but is not always done. Some studies have also shown utility in calculated globulin fraction (total protein minus albumin) to predict the presence of HG. One study reported an 82.5% predictive value for IgG ≤ 570 mg/dL when the calculated globulin was ≤ 2 g/dL [9]. Another study showed a calculated globulin level of 2.4 g/dL had a sensitivity of 86.2% and specificity of 92% for IgG < 600 mg/dL [10].

Table 1: Comprehensive evaluation of hypogammaglobulinemia.

The evaluation of HG should include assessment of adaptive immunity, a careful review of medications, and identifying sources of protein loss.

Evaluation of HG
• Immunoglobulins: IgG, IgM, and IgA • Monoclonal protein study  • Subtract IgG component of M-spike from total IgG to calculate corrected IgG • Lymphocyte subsets: B, T, and NK cells • Vaccine responses (check at least 3 months after last IgG-RT dose if applicable)  • T dependent: tetanus, diphtheria are typically checked in PI, but this could also include responses to Prevnar, hepatitis B, varicella, and Haemophilus influenzae type b  • T independent: Pneumovax 23, salmonella  • If absent, consider revaccination and recheck in approximately 4–8 weeks • Total serum protein • Serum albumin • 24-hour urine collection or spot protein-to-creatinine ratio • 24-hour stool alpha-1-antitrypsin clearance • Careful review of medication list

Although not standardized in secondary HG, vaccine antibody titers (e.g. tetanus, diphtheria, hepatitis B, varicella, Haemophilus influenzae type b, and Streptococcus pneumonia) can also be checked to assess humoral function if the patient is IgG replacement naïve [11, 12]. From this list, Pneumovax 23 is the only non-conjugated polysaccharide vaccine that can elicit a T-independent response that is absent in specific antibody deficiency [11]. The remaining other conjugated or recombinant protein vaccines elicit a T-dependent response that may still be intact even in immune deficient patients. Additionally, the diagnostic evaluation of Pneumovax 23 response (typically checked 4–8 weeks after vaccination) should be cautiously interpreted in the setting of multiple unclear variables, including time from vaccination, laboratory variability, and the increasing use of conjugated Prevnar [13]. Checking for vaccine serology responses must be done prior to IgG-RT initiation or at least 3 months after the last dose of IgG-RT to minimize the risk of false positive results from remaining exogenous IgG.

Checking lymphocyte subsets can show evidence of humoral deficiency with absent B cells or a component of cellular impairment with diminished T cells. Evaluating for other contributors to the patient’s HG should also be considered as above by checking total protein, albumin, 24-hour urine collection or spot protein-to-creatinine ratio, and 24-hour stool alpha-1-antitrypsin clearance. The medication list should also be carefully reviewed, as uncommon medications can also contribute to HG, such as antiepileptic medications, clozapine, and chlorpromazine [4].

Timing of immune evaluation during or after treatment is also important. Various studies have shown prolonged B cell depletion after CD19-targeted CAR T therapy that can last years [14]. Similarly, BCMA-targeted CAR T therapy has been shown to cause B cell aplasia for a median of 70 days (range 23–270) with delayed bone marrow plasma cell recovery at median day 212 [15]. Although the kinetics of B cell and antibody recovery have not been elucidated, patients who have active disease with ongoing treatment are expected to have a higher ongoing risk of HG compared to patients in remission where active therapy has been stopped, although more observation is needed to evaluate the duration of lasting antibody impairment and recommended monitoring duration.

5. HG in hematologic malignancy

5.1. Chronic lymphocytic leukemia (CLL)

HG is the most common inherent immune defect in patients with CLL with an incidence rate that ranges from 26% to 64% [16–18]. Clonal expansion and inadequate response to IL-2 can result in defective functioning of T cells and non-clonal B cells [19, 20]. However, this hypothesis may not fully explain the phenomenon and suggests other mechanisms at play, since treatment of CLL does not generally result in reversal of HG. The severity of HG correlates with duration and stage of disease, but does not consistently correlate with incidence of infections [17, 21].

Multiple studies have evaluated the effect of immunoglobulin heavy-chain variable region gene (IGHV) mutation status on humoral immunity. In a retrospective review of 231 CLL patients, those with unmutated IGVH genes had a shorter time to first infection and higher infection-related mortality than patients with mutated IGVH genes (31 vs. 62 months, p < .001) [22]. Other genetic abnormalities in CLL have not been associated with differences in HG or infection rates.

The most common site of infection in CLL is the respiratory tract (including sinopulmonary infections) and is possibly related to mucosal immune defects. A Danish registry-study estimated the cumulative incidence of severe infection (positive blood culture) in patients with untreated CLL within the first year after diagnosis to be around 12%, and 31% within 5 years after diagnosis [23]. About 10% of seriously infected patients died within 30 days and 25% died within 1 year. Other studies have reported 0.13–0.19 infections per patient-year [19].

5.2. Lymphoma

Various lymphoma subtypes are associated with hypogammaglobulinemia, but to a lesser degree than what is seen with CLL. The most well studied subtypes of lymphoma include diffuse large B cell lymphoma (DLBCL), follicular lymphoma (FL), and marginal zone lymphoma (MZL). The rates of HG (definitions ranged from IgG < 500 mg/dL to IgG < 700 mg/dL) in these lymphoma subtypes range from around 14–22% [24, 25]. All patients with B cell non-Hodgkin lymphomas who require systemic therapy also receive rituximab as part of their treatment plan. Rituximab increases the incidence of HG, and the risk is greatest with maintenance rituximab. In one study, 211 patients with B cell lymphoma treated with rituximab had serial quantitative serum IgG measurements pre- and post-treatment [26]. At baseline, only 15% had low IgG levels, but an additional 38% developed HG after rituximab treatment. Severe HG prompting administration of IVIG developed in 6% of patients treated with rituximab. Limited data exists for Hodgkin lymphoma, but the rate of HG appears to be low or non-existent unless there is an underlying PI.

5.3. Multiple myeloma

Immunodeficiency and HG in myeloma is the result of abnormal and non-functional expansion of malignant plasma cells that leads to immunoparesis (defined as a reduction of 1 or 2 uninvolved immunoglobulin) [27]. Quantitatively identifying HG in initially diagnosed myeloma is challenging, since an IgG M-protein can falsely elevate the measured IgG level in the serum. In a cohort of 1027 myeloma patients, only 8% had an IgG level < 700 mg/dL, however one or more uninvolved immunoglobulins were reduced in 91% of patients [28]. Another study of 5826 patients assessing for polyclonal immunoglobulin levels revealed 80% of non-IgG myeloma to have an abnormal IgG <600 mg/dL with median polyclonal IgG level of 390 mg/dL (range 270–560 mg/dL) [29]. Other studies have also highlighted immunoparesis in myeloma [30]. Higher M-protein levels were also correlated with lower polyclonal immunoglobulin levels. These findings may still underestimate the prevalence of HG, and various studies have corrected for this by subtracting the IgG M-spike component from the total IgG level [31, 32]. However, since infections are so common in myeloma, one recommendation based on expert panel discussion is to suspect that all myeloma patients have specific antibody deficiency regardless of IgG concentration [27].

The largest population-based study of 8,672 symptomatic myeloma patients (most studied before initiation of cellular immunotherapy) showed a 5-fold increased risk of bacterial infection and a 7-fold increased risk of viral and fungal infections when compared to matched controls [33]. The absolute risk of pneumonia and sepsis were 18% and 20% respectively, compared to 4% for both in controls. Infection was the cause of 32% of deaths within the first 3 months of diagnosis and 27% of deaths after 1 year.

5.4. Monoclonal gammopathy of undetermined significance (MGUS) and smoldering myeloma

In a cohort of MGUS patients, 38% of 840 patients had reduced polyclonal immunoglobulins and the median IgG concentration was 580 mg/dL [34]. A subsequent Swedish population study comparing MGUS patients to matched controls showed more than a 2-fold increased risk of any infection in MGUS patients at 5- and 10-year follow up [35]. At 10-year follow up, MGUS patients had increased risk for bacterial infections (pneumonia, osteomyelitis, sepsis, pyelonephritis, cellulitis, meningitis, and endocarditis) and viral infections (influenza and herpes zoster). However, the average number of infections per MGUS patient and per control patient was overall low at 0.34 and 0.17 respectively, and the median time between MGUS diagnosis and first infection was 1928 days.

Immunoparesis was seen in 52% of 93 smoldering multiple myeloma patients, although the specific isotypes were not reported [36]. Currently, no studies have looked at infection rates in smoldering myeloma, however underlying immunoparesis is thought to increase the risk of infection.

5.5. B cell acute lymphoblastic leukemia (ALL)

No data exists on the rates of HG in newly diagnosed B cell ALL. However, multiple treatments are associated with significant HG and B-cell aplasia as discussed later.

6. HG with various chemotherapy and/or immunotherapy treatments

The direct effect of various hematologic treatments on HG is challenging to study due to the vast number of combination therapies. Much of the available data are from other conditions, like autoimmune diseases. The extent of HG may also be different depending on the type of malignancy treated [37].

6.1. Corticosteroids

Oral and intravenous (IV) corticosteroids are widely used in malignancy treatment. Aside from corticosteroid-induced CD4 lymphopenia, the mechanisms of corticosteroid-mediated HG include inhibiting B cell function, inducing B cell apoptosis, and increasing immunoglobulin catabolism [4, 38]. In asthma, cumulative doses of ≥ 5 mg/day of oral prednisolone over the course of ≥ 2 years or > 12.6 mg/day over the course of ≥ 1 year resulted in decreased IgG [39, 40]. Systemic corticosteroids significantly affect IgG in as little as 3 days and effects may persist two to four weeks after steroid discontinuation [39, 41]. IgM and IgA are suppressed to a lesser extent [41, 42]. Some studies have suggested that isolated corticosteroid-associated HG does not affect vaccine antibody responses or increase the risk for infection [4]. Therefore, increased rates of infection may be attributed to additional underlying comorbidity or resulting lymphopenia from corticosteroid use. In patients with malignancy or history of bone marrow transplant, prolonged corticosteroid use above 10 mg prednisone equivalents daily has been associated with increased opportunistic infections [43]. In heart transplant patients, multimodal immunosuppression with prednisone can reduce IgG levels and unmask CMV disease [44].

6.2. Anti-CD20

B cell depletion is a major strategy in hematologic malignancy. The first-generation anti-CD20 monoclonal antibody rituximab is most commonly used. Anti-CD20 monoclonal antibodies target CD20 expressed by pre-B cells, naïve, and memory B cells, but spare hematopoietic stem cells and plasma cells [45]. As a result, rituximab is a well-established cause of secondary HG with the proportion of HG increasing two-to-four-fold on therapy depending on levels of preexisting HG [46]. IgG and IgM levels are typically more affected than IgA levels [47, 48]. Time to onset varies significantly, with a median time of 22.5 months for the IgG to drop < 500 mg/dL in autoimmune patients treated with rituximab [49]. Infections predominantly affect the respiratory tract and pneumococcal titers may be low [49]. Around 4% of patients on rituximab receive IVIG replacement therapy for persistent symptomatic HG, and a higher cumulative IVIG dose can reduce the risk of serious infectious complications [46, 50, 51].

Multiple studies in autoimmune disease show that lower baseline immunoglobulin levels (total pretreatment IgG < 700 to 800 mg/dL) can predispose to worsening or persistent HG after treatment with rituximab [47, 50, 52]. Co-administration of glucocorticoids, cyclophosphamide, and/or mycophenolate mofetil may also increase the risk of HG [49, 51]. Interestingly, patients who do not develop significant HG during initial rituximab treatment do not appear to develop HG on maintenance therapy [47]. Notably, half of moderate to severe HG may improve spontaneously with continued rituximab, although low baseline antibodies reduce likelihood of recovery [48, 50].

Several second- and third-generation anti-CD20 monoclonal antibodies (mAbs) have been developed including obinutuzumab, ocrelizumab, ofatumumab, ublituximab, and veltuzumab with the goal of better safety and efficacy [53, 54]. HG has already been reported with ocrelizumab and obinutuzumab, and further studies should evaluate this risk even as these newer anti-CD20 options gain approval for autoimmune and other diseases [4, 53].

6.3. Anti-CD22

Anti-CD22 therapies target the CD22 molecule primarily found on B cells, leading to B cell depletion and subsequent HG. CD22 can be found at low levels on pre-B cells and higher levels on mature B cells, but is absent on plasma cells [55, 56]. Examples include inotuzumab ozogamicin (InO) (for refractory or relapsed B cell ALL), moxetumomab pasudotox (for refractory or relapsed hairy cell leukemia), and epratuzumab (for B-cell non-Hodgkin lymphoma and SLE) [57–60]. Given their effect on B cell depletion, anti-CD22 therapies carry theoretical risk for HG. However, there remains a lack of data regarding incidence and extent associated with their use at this time [61]. In the B cell ALL setting, InO monotherapy is associated with 17% infection rate, as opposed to 71% when InO is combined with chemotherapy [62].

6.4. Anti-CD38

Anti-CD38 therapies target CD38 primarily found on plasma cells, although CD38 is also expressed by developing B cells, germinal center B cells, and other immune regulatory and inflammatory cells [63]. Another study suggests that anti-CD38 therapy could also affect memory B cells [64]. The most widely used agents of this class include daratumumab (multiple myeloma and light chain amyloidosis) and isatuximab (newly diagnosed and relapsed/refractory multiple myeloma) [65–67].

In patients receiving daratumumab for multiple myeloma, the prevalence of HG may double following therapy [68, 69]. Following daratumumab therapy, severity of HG also increases, and patients with HG show increased mortality compared to those without [69]. In non-IgG multiple myeloma, the proportion of patients with severe infections rises from 12% to 17%, although being on IVIG significantly decreases risk for infection [69].

Studies looking at the association between isatuximab and HG are sparse. One abstract reported HG (corrected polyclonal IgG < 400 mg/dL) rates up to 54% (ICARIA-MM) and 15.5% (IKEMA) during treatment, however also reported that isatuximab monotherapy had little impact on decreasing immunoglobulin levels [70]. Of those without HG at baseline, 70.9% of the Isa-Pd group (ICARIA-MM) and 90.6% of the Isa-Kd (IKEMA) went on to develop HG. 11.8% of the Isa-Pd group (ICARIA-MM) and 19.8% of the Isa-Kd (IKEMA) had grade ≥ 3 infections. Patients with non-IgG myeloma receiving combination therapy also experienced worse HG recovery than patients with IgG myeloma.

6.5. BTK inhibition

Ibrutinib is a first generation BTK inhibitor approved for the treatment of CLL in 2014. Ibrutinib directly causes HG through inhibition of B cell signaling, resulting in increased risk of upper respiratory infections, sinusitis, cellulitis and urinary tract infections [71]. In the relapsed/refractory setting, the risk of serious infections was reported to be around 10–20% [72]. Some of these serious infections were opportunistic, the most common of which was aspergillosis [73–75]. The incidence of serious opportunistic infections is estimated to be around 4% and the relationship with HG is unclear. One study found HG (IgG < 750 mg/dL) in 25 of 39 (64%) of various lymphoma patients treated with ibrutinib [76].

Second generation BTK inhibitors (e.g. acalabrutinib and zanubrutinib) seem to have similar rates of infections (15–20%) [77, 78]. Due to their shorter time on the market, there are fewer reports of opportunistic infections with second generation BTK inhibitors. The rates of HG are seldomly reported in pivotal trials of these agents, so the association between infection risk and HG is unclear.

Non-covalent BTK inhibitors (e.g. pirtobrutinib) are the newest class in this category, and the rates of serious infections seem to be similar to that of covalent BTK inhibitors (e.g. ibrutinib, acalabrutinib) [79].

6.6. Bortezomib

Bortezomib is a proteasome inhibitor used in various malignancies, including myeloma and mantle cell lymphoma. Studies looking at the association between bortezomib and HG in malignancy are limited and challenging, since bortezomib is typically used in combination with other agents. The initial 2005 clinical trial comparing bortezomib to high-dose dexamethasone in myeloma did not report resulting IgG levels [80]. In a small retrospective study of 12 lupus nephritis patients, 4 received a myeloma-based bortezomib regimen [81]. IgG levels declined by an average of 38 ± 14% and 5 of 12 patients (41.6%) developed HG (IgG < 500 mg/dL), however all of these patients also received dexamethasone and were maintained on MMF, prednisone, and hydroxychloroquine. The IgG levels increased back to > 500 mg/dL in 10 of 12 patients by the last follow up (range 9–30 months after first dose of bortezomib). None of the patients had significant infections. Another study looked at bortezomib use in 12 patients with severe lupus [82]. In these patients, bortezomib was also given in conjunction with dexamethasone. 5 of 12 (42%) developed HG (IgG < 670 mg/dL). 2 patients had reported infections that included viral infection, otitis media, and urinary tract infection.

In the phase 3 ATLAS trial at 1 year of maintenance initiation, 9 of 36 patients in the carfilzomib, lenalidomide, and dexamethasone (KRd) arm had at least polyclonal immunoglobulin recovery (defined has normalization of uninvolved immunoglobulins), which was significantly lower when compared to 58 of 66 patients who recovered in the lenalidomide maintenance arm [83]. At baseline, immunoparesis was present in 69.7% of the KRd arm and 69% in the lenalidomide arm. Grade 3 or higher infections were higher in the KRd arm, but not statistically significant (14% compared to 7%, p = 0.12).

Rates of HG or immunoparesis with ixazomib have not yet been reported in the literature.

6.7. Cyclophosphamide

The association between cyclophosphamide and HG has all been derived from non-hematologic literature and findings have not been conclusive. Studies in hematologic malignancy are challenging, because cyclophosphamide is also typically given in conjunction with other chemotherapy. Some studies have associated concomitant use of cyclophosphamide and rituximab with worse HG, however a 2015 study looking at associated risk factors in autoimmune disease treated with rituximab found only a weak association between the total prior dose of cyclophosphamide and nadir IgG post-rituximab [50]. A 2021 study did find that prior cyclophosphamide use before rituximab was associated with increased likelihood of HG IgG < 500 mg/dL and/or IgG-RT use in autoimmune disease, in line with another study showing that cyclosporine caused HG in ANCA-associated vasculitis [49, 84].

6.8. Purine analogs (pentostatin, cladribine, fludarabine)

Purine analogs often do not result in HG, but rather cell-mediated immunodeficiency. Fludarabine and cladribine can induce severe, prolonged immunosuppression with studies showing CD4 cell counts to be depressed 12 months after therapy [85–87]. As a result, the spectrum of infections following these therapies include opportunistic infections like PJP, varicella zoster virus (VZV), CMV, and Listeria monocytogenes [88]. In the modern era, the use of purine analogs is primarily part of the preparation phase for CAR T therapy or allogeneic hematopoietic stem cell transplantation (allo-HSCT). In both of these contexts, they are not used as single agents, but rather in combination with alkylating agents and/or radiation therapy. The resulting immunosuppressive effects are therefore likely a result of multiple agents with variable rates of HG.

6.9. Allogeneic HSCT

HG in allo-HSCT is likely due to multiple factors including underlying primary disease, pre- transplant therapies, and post-transplant immunosuppression. As a result, IgG-RT is often used as primary prevention for infection post-allo-HSCT. Studies have shown that patients with low IgG levels (< 400 mg/dL) during the first year after allo-HSCT had lower survival rates and increased transplant-related mortality (TRM) [89]. Some risk factors for HG include acute graft-versus-host disease (GVHD), low pre-transplant IgG level, previous HSCT, GVHD prophylaxis with cyclosporine, methotrexate, and mycophenolate mofetil [89, 90]. The incidence of HG after allo-HSCT can reach up to 75% [91]. Randomized studies in post allo-HSCT have shown that IgG-RT reduced symptomatic CMV infection and interstitial pneumonia [92, 93]. These studies were conducted in an era where CMV monitoring strategies were less accurate and primary prevention with antiviral agents such as letermovir were not implemented. Subsequent trials with an intensive weekly schedule of IVIG administration have shown lower rates of acute GVHD and gram-negative sepsis/local infection but have failed to demonstrate an overall survival advantage [94]. In a systematic review and meta-analysis of 3934 patients, IgG-RT did not demonstrate a survival benefit, although this older study was published in 2008 and included articles published up to 2007 [95]. Transplant society guidelines vary, but generally recommend IgG-RT for patients with IgG levels < 400 mg/dL or with GVHD on immunosuppressive therapy [96].

6.10. Immune effector cell and bispecific antibody therapies

Several CAR-T cell constructs and BsAbs have been approved by the FDA for the treatment of lymphoid malignancies in various treatment settings. Although these agents have been associated with a substantial risk of HG and infection, reporting on HG in pivotal trials is inconsistent, owing to the fact that the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE), version 5.0 does not specify a threshold for IgG levels to define an adverse event. Study protocols typically define it as IgG < 500 mg/dL (Table 2).

Table 2: Rates of HG and infection post CAR T therapy reported in pivotal trials leading to FDA approval^*.

CAR-T Construct	Disease and Trial Name (Phase)	Median follow-up	Rate of HG	Rates of Infection (grade 3 or higher)
Lisocabtagene maraleucel	CLL, TRANSCEND CLL 004 (Phase 2) [97]	21 months	18/117 (15%)	20/117 (17%)
	DLBCL, TRANSFORM (Phase 3) [99]	18 months	10/92 (11%)	14/92 (15%)
	FL, TRANSCEND-FL (Phase 2) [103]	19 months	6/130 (5%)	7/130 (5%)
Axicabtagene ciloleucel	DLBCL, ZUMA-7 (Phase 3) [100]	25 months	19/170 (11%)	24/170 (14%)
	FL, ZUMA-5 (Phase 2) [104]	23 months	25/148 (17%)	26/148 (18%)
Tisagenlecleucel	DLBCL, JULIET (Phase 2) [127]	14 months	20/115 (17%)	47/115 (41%)
	FL, ELARA (Phase 2) [102]	17 months	9/97 (9%)	5/97 (5%)
Brexucabtagene autoleucel	MCL, ZUMA-2 (Phase 2) [101, 128]	12 months	13/82 (16%)	28/82 (34%)
	B-ALL, ZUMA-3 (Phase 2) [122, 128]	16 months	7/78 (9%)	27/78 (35%)
Idecabtagene vicleucel	Multiple myeloma, KarMMa (Phase 2) and KarMMa-3 (Phase 3) [129]	13 months (KarMMa), 19 months (KarMMa-3)	158/349 (45%)	47/222 (21%)
Ciltacabtagene autoleucel	Multiple myeloma, CARTITUDE-1 (Phase 2) and CARTITUDE-4 (Phase 3) [130]	12 months (CARTITUDE-1), 16 months (CARTITUDE-4)	265/285 (93%)	69/285 (24%)

^*As reported in specified clinical trials or FDA package insert

NR, not reported

6.10.1. CLL

Lisocabtagene maraleucel (liso-cel), a CD19 targeting autologous product, was approved in March 2024 for adult patients with relapsed/refractory CLL who had received at least 2 lines of therapy, including a BTK inhibitor and a BCL-2 inhibitor based on results from the TRANSCEND-CLL-004 trial. In the primary analysis of the phase 1–2 study, the rates of HG and serious (grade 3 or higher) infections were 15% and 17%, respectively [97]. Rates of IgG replacement were not reported. BsAbs are currently under investigation for relapsed/refractory CLL. Results from the CLL cohort of the EPCORE CLL-1 study that evaluated epcoritamab as a single agent reported 4 fatal treatment emergent AEs (out of 23 patients enrolled in the expansion cohort), 3 of which were infections [98].

6.10.2. Non-Hodgkin lymphoma

Four CAR T constructs that target CD19 are FDA-approved for the treatment of various types of lymphoma (Table 2) [99–105]. Furthermore, 3 bispecific antibodies that target CD20 and CD3 are commercially available. These agents are associated with a varying degree of B cell aplasia (BCA) and HG (Table 3). In CAR T therapy, lymphodepleting chemotherapy (an alkylating agent and a purine analogue) is administered prior to CAR T infusion, which exacerbates the degree of HG and infections seen in the post-CAR T setting. For example, in one study of 579 patients receiving CAR T therapy for lymphoma, 60% had baseline HG and that number increased to 91% post-CAR T therapy [106]. Mean IgG levels decreased from 587 mg/dL pre- to 362 mg/dL post-CAR T therapy levels (P < .0001). Post-CAR T, 37% of patients developed a serious infection and use of intravenous immunoglobulin (IVIG) resulted in decreased mortality (HR 0.59). Although the duration of HG is often prolonged after CAR-T cell therapy, pivotal trials have not consistently reported the duration for different constructs with a few exceptions. In the JULIET trial that evaluated tisa-cel for relapsed/refractory DLBCL, the median duration of HG among responders was 742 days (range 0–1132 days) [107].

Table 3: Rates of HG and infection with bispecific antibody therapy reported in pivotal trials leading to FDA approval^*.

Bispecific Ab	Disease and Trial Name (Phase)	Median follow-up	Rate of HG	Rate of infection (grade 3 of higher)
Epcoritamab	DLBCL, EPCORE NHL-1 (Phase 2) [131]	25 months	NR	40/157 (25.5%) (2 fatal infections)
	FL, EPCORE NHL-1 (Phase 2) [132]	17 months	NR	24/128 (19%) COVID-19, 5/128 (4%) urinary tract infection (8 fatal infections)
Glofitamab	DLBCL, NP30179 (Phase 2) [133]	13 months	NR	15/145 (10%) (5% fatal infections)
Mosunetuzumab	FL, GO29781 (Phase 2) [134, 135]	18 months	40%	13/90 (14%) (1% fatal infections)
Teclistamab	MM, MajesTEC-1 (Phase 2) [112]	14 months	74.5% had HG (by report and/or IgG < 500 mg/dL)	44.8% of the 165 patients had grade 3–4 infections, which included pneumonia (18.2%), bronchitis (13.3%), upper respiratory tract infection (10.9%), and Pneumocystis jirovecii pneumonia (PJP) (3.6%)
Elranatamab	MM, MagnetisMM-3 (Phase 2) [113]	15 months	75.5% (IgG < 400 mg/dL)	40% of these patients had grade 3–4 infections, including 6 cases of PJP and fatal infections in 6.5%
Talquetamab	MM, MonumenTAL-1 (Phase 2) [115]	12 months (405 μg dose) 4 months (800 μg dose)	87% (IgG < 500 mg/dL) in pts who received the 405 μg dose and in 71% of those who received the 800 μg dose	7% (1.5% fatal infections)

^*As reported in specified clinical trials or FDA package insert

NR, not reported

BsAb use in lymphoma is associated with less HG, as they are administered without the need for lymphodepleting chemotherapy. In a pooled analysis of 32 trials, all grade and grade ≥ 3 infections occurred at a rate of 38% and 12%, respectively. However, the rates of HG with BsAbs were not consistently reported across trials and the association between HG and infection was unclear [108].

6.10.3. Multiple myeloma

Anti-CD38 monoclonal antibodies, CAR T therapy, and BsAbs aim to selectively target malignant plasma cells and have significantly improved myeloma outcomes, but can also affect non-malignant B cells and cause worsening HG. BCMA is preferentially expressed on mature B cells and differentiated plasma cells [109]. Idecabtagene vicleucel (ide-cel), a BCMA-targeting CAR T therapy, led to HG in 27 of 128 (21%) of patients in the phase 2 KarMMa study, however the specific cutoff for HG was not referenced [110]. In the CARTITUDE-4 trial, 94% of patients developed HG with ciltacabtagene autoleucel (cilta-cel) compared to 71.6% in the standard-care group [111]. A following study showed that 13 of 17 (76%) patients developed HG—6 of 17 patients had an IgG between 300–600 and an additional 7 of 17 had an IgG < 300 by 9–12 months post-BCMA CAR T cell therapy (products included JCARH125, BB2121, BB21217, or JNJ4528) [32].

In the MajesTEC-1 trial for teclistamab (BCMA-targeting BsAb), 74.5% had HG (by report and/or IgG < 500 mg/dL) [112]. 44.8% of the 165 patients had grade 3–4 infections, which included pneumonia (18.2%), bronchitis (13.3%), upper respiratory tract infection (10.9%), and PJP (3.6%). In the phase 2 MagnetisMM-3 trial for elranatamab (BCMA-targeting BsAb), HG (IgG < 400 mg/dL) was seen in 75.5% [113]. 40% of these patients had grade 3–4 infections, including 6 cases of PJP and fatal infections in 6.5%.

G protein–coupled receptor class C group 5 member D (GPRC5D) is predominantly expressed on malignant plasma cells, with little to no expression in other B cells and immune cells [114]. In the talquetamab (GPRC5D-targeting BsAb) registration trial, HG (IgG < 500 mg/dL) was seen in 87% of patients who received the 405 μg dose and in 71% of those who received the 800 μg dose [115]. The Phase1b MonumenTAL-2 study looking at talquetamab and pomalidomide combination therapy showed that 77.1% of patients had post-treatment IgG < 400 mg/dL. 80% had any grade of infection, including 8 of 35 who developed pneumonia [116]. Despite these rates of HG, comparison of the different BsAb trials showed lower grade ≥ 3 infections in 7% with talquetamab, compared to 44.8% with teclistamab, 26.4% with elranatamab, and 18.8% with cevostamab [117]. A recent retrospective study of 229 patients (67% receiving teclistamab, 20% receiving elranatamab, and 13% receiving talquetamab) also found that GPRC5D-targeting BsAbs were associated with lower risk of infection using univariate analysis [118]. This difference in infection risk may be related to GPRC5D being expressed preferentially on malignant plasma cells and not normal plasma cells, while the presence of BCMA on both malignant and normal plasma cells leads to unwanted depletion of normal plasma cells with BCMA-targeting therapies. Of note, rates of infection in these trials should be viewed cautiously if recruitment was done near the start of the COVID-19 pandemic when vulnerable myeloma patients were at higher risk of serious COVID-19 infection [119].

These significant rates of HG and infection in myeloma post-cellular therapy have sparked considerable interest in measures to prevent infection. The 2024 National Comprehensive Cancer Network Guidelines for myeloma now recommends IVIG after CAR T and BsAb therapy based on clinical context. Future studies looking at the differences in HG kinetics in different BsAb therapy may help guide timing and duration of IgG-RT.

6.10.4. B-ALL

Prior to the availability of CAR T cell therapy for B cell ALL, one of the first bispecific antibodies approved for the treatment of hematologic malignancies was blinatumomab, a bispecific T cell engager targeting CD19 and CD3 [120]. In the pivotal trial leading to FDA approval, 6% of patients developed HG (compared to 1% in the chemotherapy arm). The rate of serious infections was 28% [120]. Similarly, inotuzumab ozogamicin and lymphoid directed chemotherapy are associated with higher rates of HG [121].

Brexucabtagene autoleucel (brexu-cel), a CD19 targeting autologous CAR T cell product, received approval October 2021 for adults with relapsed/refractory B-ALL based on results from the ZUMA-3 trial [122]. Subsequent studies have confirmed the efficacy of brexu-cel as standard therapy [123]. In these studies, maintaining B cell aplasia (BCA) was generally considered a desired outcome during the treatment of B cell ALL, and may be a surrogate marker of ongoing CAR T cell persistence, and thereby durable remission [124]. However, this BCA is commonly seen before CAR-T cell therapy. In one study, pre-existing BCA (CD19 numbers < 50/μL) before CAR T cell was seen in 70% of patients [125]. This number increased to 100% post-CAR-T cell therapy.

As blinatumomab makes its way into frontline therapy for B-cell ALL [126], it remains to be seen whether combination BsAb with chemotherapy will result in higher rates of HG and/or infections.

7. Treatment and prevention

Current strategies for reducing infection risk include close monitoring with infection-reducing practices, vaccination, prophylactic antibiotics, and IgG-RT (Table 4) [7, 136, 137]. Although routine prophylactic antibiotic regimens have not been defined for HG, various strategies used in patients with neutropenia and CD4 lymphopenia include anti-bacterial, anti-viral, and anti-fungal prophylaxis. Seroconversion to vaccines while on active anti-CD20 therapy appears low, however myeloma patients can retain serologic response to the MMR vaccine after treatment [138]. Studies also support intact T cell responses to vaccines in antibody-deficient patients [139, 140]. Of note, live vaccines are typically avoided in patients with severe HG, due to risk of infection [141].

Table 4: Strategies to reduce infection in immune suppressed patients.

Management options include infection-reducing practices, close monitoring, vaccination, prophylactic antibiotics, and IgG replacement therapy.

Strategies to reduce infection in immune suppressed patients

• Standard infection infection-reducing practices (e.g. hand hygiene, masking in some circumstances) • Close monitoring and aggressive treatment of infections • Vaccination  • Influenza and COVID-19 still recommended even if on IgG-RT (due to uncertain levels of current circulating pathogen strain-specific titers in IgG-RT products) • Prophylactic antibiotic considerations  • CD4 count < 200 cells/mm3: PJP prophylaxis   • Consider prophylaxis for Coccidioides, Histoplasma, Mycobacterium, and/or Toxoplasma in certain clinical contexts  • Prolonged neutropenia: anti-bacterial (fluoroquinolones), anti-fungal, anti-viral (HSV, VZV coverage)  • Recurrent respiratory infections: azithromycin 3 times per week  • Asplenia: penicillin, amoxicillin • IgG-RT (IV or SC) • Disease-specific immunoglobulin (i.e. hyperimmune globulin)

Initiation of IgG-RT (intravenous or subcutaneous) for HG is not uniform throughout all the malignancies, however various strategies exist for patients with HG and recurrent infections [4, 142, 143]. Gammagard S/D is FDA-approved for CLL, but not for any of the other malignancies. Subcutaneous IgG-RT (SCIG) is not yet approved for any hematologic malignancy, however studies in hematologic malignancy and PI support that SCIG is able to maintain higher trough IgG levels with a favorable side-effect profile [144, 145]. A randomized controlled trial with 46 myeloma patients (24 SCIG and 22 control) showed that SCIG maintaining trough IgG > 500 mg/dL significantly reduced serious infections (mean 62 days/year vs 135 days/year, p < 0.01), days in the hospital for severe infection (mean 8 days/year vs 121 days/year, p < 0.001), and days of antibiotic treatment (mean 28 days/year vs 217 days/year, p < 0.001) [146]. Most side effects were mild, but 3 of 24 patients (13%) discontinued SCIG due to grade 3/4 skin reactions and the duration of SCIG infusions ranged from 10–28 months.

IgG-RT has been shown to reduce serious bacterial infections like pneumonia, with higher IgG trough levels correlated with less episodes of pneumonia [147]. Myeloma patients treated with BsAbs were also found to have 90% lower grade 3–5 infections while receiving IVIG [31]. IgG-RT titrated to an IgG level > 600 mg/dL has been shown to potentially help with recurrent sinusitis with underlying PI, however data is lacking for this specific use in secondary HG [148]. The benefit of IgG-RT in preventing serious viral or fungal infections has also not been established. Pooled antibody in IgG-RT is expected to have antibodies against viruses such as hepatitis A and B that the population has been vaccinated against. But since IgG-RT consists of pooled antibody derived from general population blood donors, specific antibody levels to various pathogens are variable, unstandardized, and there is a lag time between collection and a clinically significant presence of specific antibodies in IgG-RT products against current strains of pathogens like SARS-CoV-2 [149, 150]. One meta-analysis showed that IVIG use in patients hospitalized for coronavirus disease 2019 (COVID-19) had a reduction in mortality odds compared to those not receiving IVIG, however subgroup analysis of patients with severe COVID-19 did not show statistical significance [151]. In contrast to IVIG, a clinical trial published in 2022 showed that convalescent plasma given within 9 days of symptom onset decreased COVID-19-related hospitalizations within 28 days of transfusion (2.9% vs 6.3%, p = 0.005) [152]. In December 2024, high titer COVID-19 convalescent plasma became FDA-approved for immunocompromised patients [153].

In addition to pooled antibody in IgG-RT, passive antibody products (i.e. hyperimmune globulin) exist that have high titers of pathogen-specific antibody as post-exposure prophylaxis for botulinum, cytomegalovirus, hepatitis A, hepatitis B, rabies, measles, respiratory syncytial virus, tetanus, vaccinia, and varicella zoster [154, 155]. Each form of disease-specific immunoglobulin has specific guidelines for timing of administration, such as administering intramuscular measles immunoglobulin within 6 days of exposure and not concomitantly with the MMR vaccine [156]. In light of the recent 2025 measles outbreak, a new study looking at both pediatric and adult patients receiving IVIG showed that patients had median trough measles antibody titers around 1300 mIU/mL after 5–10 doses 21 or 28 days apart (the cutoff for protection against measles has been suggested to be ≥ 120–240 mIU/mL) [157].

8. Future considerations and summary

The medical landscape of antibody deficiency has significantly advanced since the first diagnosis of XLA, however multiple unanswered questions now arise from the comparison between primary HG versus secondary HG in hematologic malignancy. Extrapolating the success of using IgG-RT to restore IgG levels in PI offers hopeful benefit in secondary antibody deficiency, but the heterogeny of hematologic malignancy and variable infection rates suggests that certain differences have yet to be discovered. A deeper look into immune cell subsets, genomics, and even environmental factors may reveal additional risk-factors that explain this contrast. The diverse landscape of hematologic malignancy, treatment, and clinical trials makes a unifying evaluation challenging. More detailed reporting of the incidence, severity, and duration of HG in future pivotal clinical trials would help elucidate the association between HG and infection risk. Risk of infection in hematologic malignancy is likely multi-factorial and future research design would benefit from evaluating the direct correlation between HG severity and rates of grade 3–4 infections. Explanation for why some patients never recover B cell counts after therapy or why some recover sooner than others is also unclear. Some hopeful studies have investigated strategies to restore B cell function, including exploring compounds that could restore B cell function, therapeutic vaccination, and checkpoint inhibition [158, 159].

In sparse cases, various PI conditions associated with HG due to underlying B cell dysfunction/absence have improved after hematopoietic stem cell transplantation (HSCT), such as in severe combined immunodeficiency [160]. HSCT is not currently routinely done in XLA, since the risks of transplantation at this time outweigh the lower risk and success of IgG-RT, however additional therapeutic options may change over time as transplantation methods improve or as more hopeful gene therapy emerges. Theoretically, transplantation in various malignancies could be curative for HG and secondary immunodeficiency if there is successful immune reconstitution after transplantation. However, this is likely also affected by various factors, including anti-rejection immune suppression.

In summary, the use of advanced cellular therapies in hematologic malignancies shows a trend towards severe HG and infections, paralleling similar phenotypes seen in primary antibody deficiency. However, research has not yet revealed exactly why some patient with significant HG appear to be at higher risk of infection, while others get few infections. Nevertheless, clinicians should be aware of these HG risks, how to evaluate humoral immunity, monitoring strategies, and various treatment considerations, including IgG-RT. Future research and collaboration between Hematology/Oncology and Allergy/Immunology will hopefully elucidate additional biomarkers to help with infectious risk-stratification and identify populations that would best benefit from IgG-RT with the ultimate goal of improving patient outcomes.

Practice points

• The prevalence of HG and/or specific antibody deficiency is high in lymphoid malignancies, magnified with treatment, increasing with the use of cellular therapy (i.e. CAR T, BsAbs), and associated with serious grade ≥ 3 infections.

• Evaluation of HG should include assessment of adaptive immunity (immunoglobulin levels, lymphocyte subsets, and vaccine responses) and awareness of other sources of protein loss.

• Strategies to reduce infection in immunosuppressed patients include close monitoring with risk-reducing practices, vaccination, prophylactic antibiotics, hyperimmune globulin, and IgG-RT.

Research agenda

• Identifying additional immune markers that could help monitor immune status and stratify risk of infection in patients with HG on advanced therapies.

• Future clinical trials for IgG-RT in hematologic malignancy-induced HG are needed.

Future considerations

• Novel strategies to identify and assist with long-term recovery of T cell, B cell, and neutrophil counts and function will hopefully reduce the risk of serious infection.