Efficacy and safety of blinatumomab for CD19+ acute leukemias in patients historically excluded from clinical trials due to comorbidities

Abstract

Background:

Blinatumomab has proven efficacy in the frontline, consolidation, and relapsed/refractory settings in B cell acute lymphoblastic leukemia. Its efficacy and safety among patients commonly excluded from clinical trials are unknown.

Patients and Methods:

This single center, retrospective cohort study included patients treated for acute leukemia with blinatumomab with the following pre-existing conditions: liver dysfunction, renal impairment, central nervous system (CNS) conditions, autoimmune disease, solid organ transplantation, or uncontrolled infection. Rates of blinatumomab completion, efficacy outcomes, cytokine release syndrome (CRS), neurotoxicity (ICANS), and adverse events specific to the impaired organ leading to inclusion were assessed.

Results:

Thirty-four patients were included, and 88% completed at least one cycle of blinatumomab. One patient stopped prematurely due to toxicity and three due to lack of response in the relapsed/refractory setting. Response rates and survival were similar by treatment setting to those treated in clinical trials. The 60-day cumulative incidence of grade ≥ 2 CRS and ICANS were 23.5% and 20.8%, respectively. No excess liver injury, ICANS, autoimmune flares, organ rejection, or infections were observed in cohorts defined by impairment of each system.

Conclusion:

Blinatumomab was successfully administered to the vast majority of patients with a baseline condition that would have led to clinical trial exclusion. Adverse events generally occurred at rates similar to prior clinical trials. These data provide the basis to consider removing certain exclusion criteria from future studies involving blinatumomab, allowing more patients to benefit from its efficacy.

Graphical Abstract

1. Introduction

The CD3-CD19 bispecific engager blinatumomab has helped improve outcomes for patients with B-cell acute lymphoblastic leukemia (B-ALL) in various treatment contexts. In the upfront [1-5], measurable residual disease (MRD) positive [6-8], MRD negative [9], and relapsed/refractory settings [10,11], blinatumomab has produced improved or similar outcomes with generally reduced toxicity compared to prior standards of care for Philadelphia chromosome (Ph) positive or negative disease.

Patients with certain comorbidities were routinely excluded from prospective blinatumomab clinical trials; therefore, it is uncertain how effective or safe this agent is for these patients. Grade 3-4 toxicities on the initial clinical trial of blinatumomab in MRD+ B-ALL included hepatotoxicity, neurologic toxicities, hypogammaglobulinemia, neutropenia, lymphopenia, and infections [7], forming the basis for excluding patients with baseline hepatic impairment, central nervous system (CNS) pathology, and uncontrolled infections from subsequent trials. Those with pre-existing renal impairment have been excluded as blinatumomab appears to be renally eliminated, raising concern for unpredictable pharmacokinetics in these patients [12]. Patients with autoimmune disease or solid organ transplantation requiring systemic immunosuppression have been excluded due to concern for both lack of efficacy due to baseline suppression of CD3+ T cells and toxicity from increased blinatumomab-mediated immunosuppression.

Previously, the efficacy and tolerability of blinatumomab in patients ≥ 65 years who have perceived, relatively reduced functional reserve was questioned. Apart from an increased incidence of serious neurologic events, prospective evaluation showed blinatumomab to yield disease responses with a similar toxicity profile as in younger patients, with no blinatumomab-related deaths [5,13]. Due to blinatumomab’s relative tolerability compared to alternative therapeutic options, it is used widely for this older population. For this reason, blinatumomab may also serve as a preferred option relative to alternatives for CD19+ acute leukemias in those whose comorbidities have led to clinical trial exclusion. There have been reports of four cases of blinatumomab use in patients with liver dysfunction, which demonstrated effective and safe administration [14,15]. A prior analysis of the U.S. Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) linked blinatumomab-induced neurologic adverse events with use of concurrent, broadly defined neurologic medications, but the clinical significance of these events was unclear due to lack of event grade reporting and lack of increased neurotoxicity-related deaths in this subgroup. For this reason and the broad definition of neurologic medication, no clear connection between underlying CNS pathology and neurotoxicity risk could be made [16]. In a pilot study of blinatumomab in refractory rheumatoid arthritis and a case report in rapidly progressive systemic sclerosis, improvement rather than flares of these autoimmune conditions and minimal toxicity were observed [17,18]. Successful use of blinatumomab for B-ALL in the setting of invasive fungal infections have been reported in nine pediatric patients [19,20]. No reports of blinatumomab’s tolerability in patients with underlying renal dysfunction or solid organ transplantation have been published. In order to systematically assess blinatumomab’s utility in those commonly excluded from clinical trials, we report the efficacy and safety of blinatumomab administered to these patients at our institution.

2. Patients and methods

2.1. Patients

Prospective clinical trials of blinatumomab were analyzed to determine the common reasons for study exclusion. Those that investigated prophylactic administration of blinatumomab post-allogeneic stem cell transplantation (alloSCT) were excluded from further review [21,22]. The eight remaining studies were used to determine common trial exclusion criteria [1,3,5,6,8-11]. For inclusion in this retrospective cohort study, a patient must have been age ≥ 18 years and received blinatumomab outside of a clinical trial for a CD19+ acute leukemia. Additionally, a patient must have had at least one of the following markers of prospective trial exclusion on receipt of blinatumomab: (1) impaired liver function defined as total bilirubin > 2 x upper limit of normal (ULN), unless due to Gilbert’s syndrome, or aspartate transaminase (AST) or alanine transaminase (ALT) > 5 x ULN; (2) serum creatinine > 1.5 mg/dL; (3) history or presence of clinically relevant CNS pathology such as epilepsy, childhood or adult seizure, paresis, aphasia, stroke, severe brain injuries, dementia, Parkinson’s disease, cerebellar disease, organic brain syndrome, or psychosis; (4) history of autoimmune disease requiring systemic immunosuppressive treatment; (5) history of solid organ transplantation; or (6) active, uncontrolled, serious infection. Prior systemic immunosuppression for graft-versus-host disease was not sufficient for inclusion. After this study was deemed exempt from human subjects oversight under Department of Health and Human Services (HHS) regulations following review by the Johns Hopkins Institutional Review Board, the medical records of patients who received blinatumomab within this institution were screened for study inclusion.

2.2. Data collection

Information was tabulated regarding patient demographics, disease and treatment characteristics, underlying organ function, blinatumomab cycle(s) completion, response, relapse, survival, and adverse event occurrence. Response was defined according to 2024 European LeukemiaNet criteria [23]. Complete hematologic response (CHR) was defined as meeting all criteria for a complete remission (CR) without bone marrow assessment. MRD was assessed via multicolor flow cytometry (MFC) from the bone marrow with sensitivity of 10⁻⁴. Cytokine release syndrome (CRS); hepatic injury defined by AST, ALT, or bilirubin increase; renal injury defined by creatinine increase and acute kidney injury; autoimmune flares; and sepsis were graded according to the HHS Common Terminology Criteria for Adverse Events (CTCAE), version 5.0, which grades adverse events relative to baseline organ impairment [24]. Neurotoxicity (ICANS) was graded per the American Society for Transplantation and Cellular Therapy ICANS consensus grading for adults [25]. Adverse events were only recorded and analyzed if they were grade ≥ 2, except for sepsis in which grade ≥ 3 events with causative bacteria identified by blood culture were required, as these were considered clinically significant. Data collection was censored on February 19, 2025.

2.3. Statistical analysis

Included patients were analyzed in cohorts defined by treatment context and reason for study inclusion. Follow-up time was determined by the reverse Kaplan-Meier method. For categorical patient and disease response characteristics, differences between groups were determined by Fisher’s exact test. For continuous variables, the Shapiro-Wilk and Levene’s test were used to assess normality and the equality of variances between groups, respectively. If these assumptions were not rejected, one-way analysis of variance (ANOVA) was performed to assess global differences between groups, and if significant, Tukey’s test was used to assess which specific groups differed. As normality assumptions were rejected for some groups’ white blood cell count (WBC) and blinatumomab cycle number distributions, the non-parametric Kruskal-Wallis test was used to assess group differences in these variables’ distributions. Time-to-event analyses were performed from blinatumomab initiation with death; relapse or death; and lack of response, relapse, or death considered events for overall survival (OS), relapse-free survival (RFS), and event-free survival (EFS) investigations, respectively. 1-year OS, RFS, and EFS were calculated using the Kaplan-Meier estimator. The cumulative incidence of blinatumomab-specific non-relapse mortality (NRM) with competing risks of refractory disease, relapse, or alloSCT was determined via the Fine-Gray method. The same method was used to assess the cumulative incidences of CRS, ICANS, hepatic toxicity, renal toxicity, and sepsis with competing risk of death. All reported p-values are two-sided with those ≤ 0.05 considered statistically significant. Statistical analysis and visualization were performed using R, version 4.4.3 (R Foundation for Statistical Computing, Vienna, Austria).

3. Results

3.1. Demographics

One hundred ninety-seven patients received blinatumomab between 2015-2025. Of these, 40 were excluded for exclusively receiving blinatumomab on a clinical trial while 123 were excluded for not meeting inclusion criteria. Demographic and disease characteristics by treatment setting for the 34 included patients are shown in Table 1. Median follow-up was 4.7 years (interquartile range: 1.4-5.8 years). In the frontline setting, cytoreductive therapy prior to blinatumomab included cyclophosphamide and corticosteroids (44.4%), corticosteroids alone (33.3%), and corticosteroids and tyrosine kinase inhibitors (TKIs) (22.2%). In the consolidation setting, 2/16 (12.5%) received blinatumomab in second or greater remission while the remainder were in first remission. In the relapsed/refractory setting, cytoreductive therapy before blinatumomab was administered in 55.6% and consisted of corticosteroids alone (22.2%); cyclophosphamide and corticosteroids (22.2%); and cyclophosphamide, corticosteroids, and TKI (11.1%). Among patients with Ph+ disease, blinatumomab was co-administered with TKIs in 14/16 (87.5%); the two not receiving concurrent TKIs received blinatumomab in 2015 and 2016 and had disease progression or intolerance to prior TKIs.

Table 1.

Baseline demographic and treatment characteristics of included patients based on treatment setting in which blinatumomab was administered. CML, LBC: chronic myeloid leukemia in lymphoid blast crisis. MPAL: mixed phenotype acute leukemia. P-values are calculated for each variable between the three groups defined by treatment setting. *: Tukey’s test revealed that there were significant differences in age between the frontline and consolidation cohorts (p=0.002) and between the frontline and relapsed/refractory cohorts (p=0.00003) but not between the consolidation and relapsed/refractory cohorts (p=0.08). ^a: Defined according to National Comprehensive Cancer Network version 1.2025 ALL guidelines [26].

	All patients (n=34)	Frontline (n=9)	Consolidation (n=16)	Relapsed/ refractory (n=9)	p-value
Age, years, median (range)	53.8 (22.0-76.6)	71 (55.2-76.6)	51.6 (27.8-71.1)	33.8 (22.0-58.8)	4.3x10−5*
Female sex, n, %	20 (58.8)	5 (55.6)	12 (75)	3 (33.3)	0.14
Race, n, %					0.34
White	17 (50)	5 (55.6)	8 (50)	4 (44.4)
Black	8 (23.55)	3 (33.3)	4 (25)	1 (11.1)
Asian	3 (8.8)	1 (11.1)	0	2 (22.2)
Other	6 (17.65)	0	4 (25)	2 (22.2)
Hispanic/latino ethnicity, n, %	4 (11.8)	0	3 (18.8)	1 (11.1)	0.57
Disease					0.11
B-ALL	30 (88.2)	9 (100)	15 (93.75)	6 (66.6)
CML, LBC	2 (5.9)	0	1 (6.25)	1 (11.1)
MPAL	2 (5.9)	0	0	2 (22.2)
Philadelphia chromosome positive, n, %	16 (47.1)	3 (33.3)	11 (68.8)	2 (22.2)	0.06
Cytogenetic and molecular risk stratificationa, n, %					0.60
Standard	7 (20.6)	2 (22.2)	4 (25)	1 (11.1)
Poor	17 (50)	6 (66.7)	7 (43.75)	4 (44.4)
Unknown	10 (29.4)	1 (11.1)	5 (31.25)	4 (44.4)
Therapy-related disease, n, %	6 (17.6)	4 (44.4)	2 (12.5)	0	0.045
Prior allogeneic hematopoietic stem cell transplant, n, %	6 (17.6)	0	0	6 (66.6)	1.2x10−4
White blood cell count at time of blinatumomab initiation, /nL, median (range)	6.3 (0.05-16.0)	1.25 (0.5-13.3)	7.1 (2.1-13.1)	7.2 (0.05-16.0)	0.09
CNS disease at time of blinatumomab initiationa, n, %					0.12
CNS-1	24 (70.6)	4 (44.4)	14 (87.5)	6 (66.6)
CNS-2	9 (26.5)	4 (44.4)	2 (12.5)	3 (33.3)
CNS-3	0 (0)	0	0	0
Unknown	1 (2.9)	1 (11.1)	0	0
MFC MRD status, n, %		N/A		N/A	-
Positive	10 (29.4)		10 (62.5)
Negative	5 (14.7)		5 (31.25)
Unknown	1 (2.9)		1 (6.25)
Reason for inclusion, n, %					0.22
Hepatic impairment	8 (23.5)	2 (22.2)	3 (18.75)	3 (33.3)
Renal impairment	7 (20.6)	3 (33.3)	1 (6.25)	3 (33.3)
CNS pathology	14 (41.2)	3 (33.3)	6 (37.5)	5 (55.5)
Autoimmune disease	6 (17.6)	1 (11.1)	5 (31.25)	0
Organ transplant	2 (5.9)	2 (22.2)	0	0
Infection	1 (2.9)	0	1 (6.25)	0
Two reasons for inclusion, n, %	4 (11.8)	2 (22.2)	0	2 (22.2)	0.07

3.2. Overall efficacy and safety

Blinatumomab completion and response to treatment are detailed in Table 2. Overall, 88.2% completed a full, 28-day cycle with early discontinuations due to nephrotoxicity (n=1) and lack of response in the relapsed/refractory setting (n=3). OS, RFS, and EFS by treatment context are shown in Figure 1. Of 14 patients with refractory disease or relapse after blinatumomab, 64.3% retained CD19 expression. Of ten patients who received blinatumomab for MRD + consolidation, eight (80%) achieved MRD negativity by MFC. The cumulative incidence of blinatumomab-specific NRM is displayed in Figure 2. Reasons for mortality within one year of receiving blinatumomab, other than leukemia or post-alloSCT complications, were an unclear cause at day 44 in a patient with subdural hematoma at blinatumomab initiation, multiorgan failure at day 68 believed to be a sequela of liver and renal injury present before blinatumomab and attributed to prior asparaginase, and sepsis at day 169 in a patient with baseline alcoholic cirrhosis.

Table 2.

Rates of completion and disease response in patients stratified by treatment setting in which blinatumomab was administered. CR: complete remission. CRi: Complete remission with incomplete count recovery. MLFS: Morphologic leukemia free state. P-values are calculated for each variable between the three groups defined by treatment setting. ^a: refers to the development of relapsed disease on first response assessment. All patients treated in the consolidation setting who had relapsed disease on first response assessment had a CNS-only relapse.

	All patients (n=34)	Frontline (n=9)	Consolidation (n=16)	Relapsed/ refractory (n=9)	p-value
Full cycle of blinatumomab completed, n, %	30 (88.2)	8 (88.8)	16 (100)	6 (66.6)	0.04
Blinatumomab cycle completed at full dose, n, %	26 (76.5)	7 (77.7)	15 (93.8)	4 (44.4)	0.02
Blinatumomab stopped before completion due to toxicity, n, %	1 (2.9)	1 (11.1)	0	0	0.53
Number of blinatumomab cycles completed, median (range)	2 (0-6)	2 (0-6)	2 (1-4)	1 (0-2)	0.07
Best response to blinatumomab					0.22
CR	17 (50)	5 (55.5)	10 (62.5)	2 (22.2)
CRi	5 (14.7)	2 (22.2)	2 (12.5)	1 (11.1)
MLFS	1 (2.9)	0	0	1 (11.1)
CHR	2 (5.9)	1 (11.1)	1 (6.25)	0
Relapse/refractory	9 (26.5)	1 (11.1)	3 (18.75)a	5 (55.5)

Figure 1.

A. OS from time of blinatumomab initiation by treatment setting. B. RFS from time of blinatumomab initiation in patients treated in the consolidation setting. C. EFS from time of blinatumomab initiation in patients treated in frontline or relapsed/refractory settings.

Figure 2.

Blinatumomab-specific NRM among all patients who had a historical reason for exclusion from blinatumomab-related clinical trials, regardless of treatment setting. Competing risks were confirmation of refractory disease, disease relapse, or alloSCT.

The cumulative incidences of grade ≥ 2 CRS and ICANS for the entire cohort are displayed in Figure 3. There were no grade ≥ 3 CRS events. No significant difference in grade ≥ 2 CRS incidence was observed between those treated with morphologic disease (in the frontline or relapsed/refractory settings) versus those receiving consolidative treatment in remission (60-day incidence of 27.8% [95% confidence interval (CI): 9.7-49.5%] vs. 18.8% [4.3-41.0%], p=0.5). A significantly higher cumulative incidence of grade ≥ 2 ICANS was observed in those treated for morphologic disease than in the consolidation setting (60-day incidences of 38.9% [16.8-60.7%] and 0%, respectively, p=0.03). The 60-day cumulative incidence of grade ≥ 3 ICANS was 22.2% (6.6-10.1 %) for those treated for morphologic disease and 0% in the consolidation cohort (p=0.052).

Figure 3.

Cumulative incidence of grades ≥ 2 A. CRS and B. ICANS after blinatumomab initiation among all patients who had a reason for exclusion from blinatumomab-related clinical trials, regardless of treatment setting. Competing risk was death.

3.3. Patients with hepatic dysfunction

Eight patients were included due to pre-existing liver disease or drug-related impairment at time of blinatumomab initiation. Liver dysfunction was attributable to asparaginase (n=4), dasatinib (n=1), ponatinib (n=1), ALL involvement (n=1), or alcoholic cirrhosis (n=1). Patients were included due to elevations in bilirubin (n=3), ALT (n=3), bilirubin and ALT (n=1), or bilirubin and AST (n=1). For those included due to elevated bilirubin, ALT, or AST, median baseline values were 3.1 mg/dL (range: 2.5-8.0), 188 international units (IU)/L (158-230), and 242 IU/L, respectively. The 60-day cumulative incidence of grade ≥ 2 hepatic injury from baseline in these patients was 37.5% (95% CI: 7.2-69.4%) of which one reached grade 3 toxicity. Relative injury was demonstrated by bilirubin increase in three patients with concurrent AST rise in one. The incidence of grade ≥ 2 liver injury from baseline was not found to be significantly different between those included for pre-existing hepatic dysfunction and those included for other reasons (60-day incidence of 23.6% [9.3-41.5%] in the latter group, p=0.53). Bilirubin, AST, or ALT improved in 4/8 patients while on blinatumomab.

Patients with renal impairment

Seven patients were included due to abnormal creatinine in the setting of chronic kidney disease (n=4) or an acute renal insult from septic shock (n=1), methotrexate toxicity (n=1), or an idiopathic cause (n=1) at time of blinatumomab initiation. Median creatinine was 2.7 mg/dL (range: 0.7-5.0); the patient with creatinine of 0.7 mg/dL was included due to ongoing renal replacement therapy. The 60-day cumulative incidence of grade ≥ 2 renal injury relative to baseline was 14.3% (95% CI: 0.5-49.1%) in those with baseline renal impairment versus 0% in the remainder of included patients (p=0.0495). For those with baseline impairment, one event of grade ≥ 2 kidney injury occurred, which was grade 4 renal failure leading to permanent renal replacement, attributed to tacrolimus toxicity. Renal improvement was noted in 5/7 of these patients while on blinatumomab. No difference in grade ≥ 2 CRS (60-day incidence of 42.9% [95% CI: 7.6-75.7%] versus 18.5% [6.6-35.2%], p=0.2) or ICANS (60-day incidence of 28.6% [3.1-63.6%] versus 18.8% [6.7-35.2%], p=0.7) could be proven between those with baseline renal impairment and the remainder of included patients.

3.4. Patients with CNS pathology

Fourteen patients with underlying CNS pathology were included. Baseline CNS pathologies were history of stroke (n=5), seizure (n=5), subdural hematoma causing paresis (n=2), stroke and seizure (n=1), and subdural hematoma deemed clinically significant due to associated hypertensive emergency requiring continuous intravenous anti-hypertensives (n=1). The 60-day cumulative incidence of grade ≥ 2 ICANS was 22.1% (95% CI: 4.9-47.0%) in those with baseline CNS pathology versus 20.0% (6.0-40.0%, p=0.60) in the remainder of included patients while the incidence of grade ≥ 3 ICANS was 14.9% (2.1-38.9%) and 10.0% (1.6-27.8%, p=0.68), respectively. In this cohort with baseline CNS pathologies, ICANS of any grade occurred exclusively in patients receiving anti-epileptic or anti-psychotic medications for seizure prevention or bipolar disorder. One patient with a history of seizures on anti-epileptics experienced a seizure during blinatumomab therapy, leading to temporary drug cessation before completion at the lower dose of 9 μg/day.

3.5. Patients with autoimmune disease

Six patients with previously diagnosed autoimmune disease requiring systemic immunosuppression were included. Autoimmune conditions included inflammatory bowel disease (IBD, n=3), systemic lupus erythematosus (SLE, n=1), sarcoidosis (n=1), and anterior uveitis (n=1). One patient, receiving hydroxychloroquine for SLE, was on systemic immunosuppression at time of blinatumomab initiation. Two patients had last received immunosuppression of prednisone for anterior uveitis or vedolizumab for IBD within 5 months while the remaining three patients had been off of systemic immunosuppression for > 1 year for IBD or sarcoidosis. Among these patients, 1/2 patients treated for MRD+ consolidation achieved MRD negativity with the other having disease relapse on first assessment. The latter patient was on hydroxychloroquine at blinatumomab initiation. Both patients treated for MRD- consolidation retained MRD negativity. One patient treated for CHR consolidation has since remained in a CHR. A patient treated in the frontline setting achieved a CR with MRD negativity by MFC. Within the four months following blinatumomab initiation in this cohort, one grade 3 infection with Serratia marcescens sepsis leading to hospitalization was recorded, in the patient who discontinued prednisone for anterior uveitis within 5 months of blinatumomab initiation. No episodes of NRM were identified. No autoimmune flares were observed within 60 days of blinatumomab initiation in this group while one patient without known autoimmune disease in the overall cohort developed anterior uveitis on blinatumomab.

3.6. Patients with history of organ transplantation

Two patients with a history of solid organ transplantation, of either liver or heart and actively on calcineurin inhibitors, were included in this analysis. No episodes of organ rejection were observed within 60 days of blinatumomab initiation in either patient, and they both achieved a CR with MFC MRD negativity after treatment in the frontline setting. One grade 3 sepsis event occurred within four months of blinatumomab initiation, at day 47 from Streptococcus viridans. Neither patient experienced NRM.

3.7. Patient with uncontrolled infection

One patient was included due to necrotizing lung infection attributed to Zygomycosis by microbiology culture for which she received amphotericin B while on blinatumomab. This infection did not worsen on blinatumomab.

4. Discussion

This retrospective cohort study demonstrates the efficacy and feasibility of blinatumomab in the treatment of CD19+ acute leukemias in patients with underlying comorbidities that would have led to clinical trial exclusion. Overall, 88.2% completed at least one cycle of blinatumomab with only one early discontinuation due to toxicity. The majority were able to escalate to the full dose. This is comparable to reported rates of completion of at least one cycle in clinical trials [1,3,5,7,8,10,22]. As in other studies, the rate of completion of at least one full cycle was lower among relapsed/refractory patients than those treated in the frontline or consolidation setting. One-year OS rates of 66.7%, 91.7%, and 22.2% observed in the frontline, consolidation, and relapsed/refractory settings were similar to rates seen in clinical trials of 70-95% [1,3,5], 80->90% [8,9], and 30-40% [10,11], respectively.

Overall, significant blinatumomab-related toxicities in this study did not differ in incidence from previously published rates among clinical trial participants. The 60-day cumulative incidence of grade ≥ 2 CRS was 23.5% (95% CI: 10.9-38.9%), which was similar to the reported rates of 3.4-19% across disease settings [1,3,8,10], and importantly, no grade ≥ 3 CRS were observed. For clinical trials, ICANS is typically only reported if grade ≥ 3. Potentially due to differences in attribution patterns, the rates of ICANS grade ≥ 3 reported in clinical trials vary greatly across treatment settings from 1.6-53% [1,5,8,9]. In this study, the incidence of grade ≥ 3 ICANS fell within this range at 12.0% (95% CI: 3.7-25.5%) in the entire cohort and 14.9% (95% CI: 2.1-38.9%) in those with baseline CNS pathology, suggesting a lack of increased ICANS risk in those with baseline CNS pathology. Grade ≥ 2 ICANS in the latter cohort exclusively occurred in those taking anti-epileptic or anti-psychotic medications, which could reconcile this study’s findings with a FAERS analysis that reported an association between neurologic medication use and ICANS after blinatumomab [16]. Seizure activity has been associated with blood-barrier permeability [27], and due to this in conjunction with the increased cytokine burden after blinatumomab, patients with a history of recurrent seizures could have an increased risk of blinatumomab-induced ICANS. This suggests that only baseline CNS pathology severe enough to require one of these medications may increase the risk of blinatumomab-mediated ICANS, but this needs further validation. More definitively, higher disease burden was associated with ICANS development after blinatumomab in this analysis with it occurring more often in those with morphologically evident disease. A relationship between disease burden and ICANS risk has previously been suggested among patients treated with blinatumomab [28], and this relationship has been found with another T cell-activating product, chimeric antigen receptor modified-T (CAR-T) cells [29,30]. Higher disease burden has been shown to trigger greater cytokine release from activated CAR-T cells [31], which may trigger blood-brain barrier disruption and increase CNS permeability to inflammatory factors [29]. As a similar mechanism of cytokine release is observed after blinatumomab [12], it is mechanistically logical that high disease burden would also lead to a higher ICANS rate after its administration.

In addition to ICANS, this study demonstrated that blinatumomab largely did not increase the risks of toxicity specific to the impaired organ system that led to patients’ inclusion. The similar rates of relative liver injury shown in those with or without pre-existing liver disease or liver toxicity from prior therapy is consistent with prior case reports that reported safe use in patients with a bilirubin up to 20.6 mg/dL [14,15]. The current analysis demonstrated an increased risk of grade ≥ 2 renal injury from baseline in those with pre-existing renal impairment, although this was secondary to a single case in which an alternative cause was present; therefore, this relationship’s validity is unclear. Any grade acute kidney injury has historically been rare, reported in ≤ 2% of clinical trial participants receiving blinatumomab [1,3,8,10]. Blinatumomab’s presumed renal clearance may lead to increased systemic toxicity in those with pre-existing impairment [12], but this was not seen in this analysis with similar rates of CRS and ICANS in those with and without kidney impairment. Among patients with a history of systemic immunosuppression use for an autoimmune condition or solid organ transplantation, responses were seen in all but one patient without any cases of NRM, allaying concerns that blinatumomab has blunted effect and increases toxicity in these patients. This study also demonstrated no cases of autoimmune flare or organ rejection in these patients, consistent with prior reports of blinatumomab decreasing rather than increasing autoimmune activity [17,18]. Only five patients were either on active immunosuppression or discontinued immunosuppression within 5 months of blinatumomab, of whom one had relapsed disease on first response assessment and two had grade 3 sepsis within four months; therefore, there is limited statistical power to make conclusions regarding the relative safety and efficacy of blinatumomab in those actively or recently on immunosuppression. This analysis did not demonstrate any negative effects to infection control in one patient with uncontrolled infection at time of blinatumomab receipt, echoing reports from the pediatric literature [19,20]. Blinatumomab-induced hypogammaglobulinemia theoretically could increase the risk and severity of severe infections; however, the previously determined median of 85-260 days to immunoglobulin nadir from blinatumomab allows time to pursue leukemia control and blood count improvement via this agent while an active infection is present [32,33]. Overall, these data support that with caution, blinatumomab can be administered effectively and safely when comorbidities exist that would have precluded clinical trial enrollment.

As the addition or substitution of blinatumomab has demonstrated improved efficacy with significant survival benefit in nearly every treatment scenario compared to historical standard-of-care approaches, nearly every B-ALL patient should now receive it as a component of their therapy. This study demonstrates that patients who were largely excluded from clinical trials can be safely treated with blinatumomab. Based on these outcomes, less stringent exclusion criteria for future studies involving blinatumomab should be considered to provide prospective data in these patient populations.

This is the first systematic investigation of the efficacy and safety of blinatumomab in patients with pre-existing comorbidities that would have led to exclusion from clinical trials. Eighty-eight percent of patients were able to complete at least one cycle of blinatumomab with 1-year OS rates comparable to those among clinical trial participants. Generally, excess adverse events were not observed in this cohort. Due to limited patient numbers, caution should be used in applying these conclusions to those with pre-existing renal impairment, CNS pathology severe enough to require anti-epileptic or anti-psychotic medication, or autoimmune disease or solid organ transplantation requiring systemic immunosuppression concurrently with blinatumomab. Additional studies, ideally in the prospective setting, are needed to confirm these conclusions.

Highlights.

• Blinatumomab has been safely given to patients excluded from clinical trials.

• Efficacy outcomes were similar to clinical trial cohorts by treatment setting.

• Common blinatumomab toxicities occurred at rates similar to other populations.

• Disease burden was associated with neurotoxicity.

Data sharing statement

Deidentified individual participant data collected during this study will be made available indefinitely following publication of this article upon request to the corresponding author by researchers who provide a methodologically sound proposal.