FDA Approval Summary: Blinatumomab for the Treatment of B-cell Precursor Acute Lymphoblastic Leukemia in the Consolidation Phase of Multiphase Chemotherapy

Abstract

On June 14, 2024, the FDA approved blinatumomab (Blincyto; Amgen, Inc) in the consolidation phase of treatment for CD19-positive, Philadelphia chromosome-negative B-cell precursor acute lymphoblastic leukemia (BCP ALL). FDA reviewed results from three randomized trials using blinatumomab in consolidation: Study E1910 (adult patients with newly diagnosed BCP ALL without minimal residual disease), Study 20120215 (pediatric patients with high-risk BCP ALL in first relapse), and Study AALL1331 (pediatric and young adult patients with BCP ALL in first relapse). Study E1910 demonstrated a significant improvement in overall survival (OS) when comparing alternating blinatumomab and chemotherapy cycles versus chemotherapy cycles alone (HR 0.42; 95% CI 0.24, 0.75; p = 0.003). In Study 20120215, the efficacy of blinatumomab as an alternative to a third cycle of chemotherapy was primarily supported by descriptive analyses of the secondary endpoint of OS (HR 0.35; 95% CI 0.17, 0.7) and a post-hoc analysis of relapse-free survival (RFS) (HR 0.38; 95% CI 0.22, 0.66). AALL1331 did not meet its primary objectives for the randomizations in the high/intermediate-risk arm comparing blinatumomab vs chemotherapy or in the low-risk arm comparing blinatumomab cycles and chemotherapy cycles versus chemotherapy cycles alone. A meta-analysis of OS performed using the above studies and the infant study NL59901.078.17 was consistent with a treatment effect of blinatumomab during consolidation. The safety profile of blinatumomab cycles was consistent with previous studies. The benefit of blinatumomab during the consolidation phase of therapy is consistent across line of therapy (newly diagnosed versus relapsed) and patient age (adult versus pediatric).

INTRODUCTION

B-cell precursor acute lymphoblastic leukemia (BCP ALL) is the most common cancer in children¹ and the second most common acute leukemia in adults.² Once considered a uniformly fatal disease, outcomes for both pediatric^3,4 and adult⁵ patients with BCP ALL have improved over time. For pediatric patients, sequential introduction of cytotoxic drug combinations, multiphase therapy, use of high-doses of cytotoxic drugs, optimization of CNS prophylaxis, risk-stratified treatment plans, and risk-adapted therapy correlated with incremental improvements in survival and emergence of a substantial cure fraction.³ Thus, the current approach to treatment of BCP ALL with curative intent generally consists of induction, postremission intensification or consolidation, and maintenance phases of therapy.²

Consolidation therapy for pediatric patients with BCP ALL was introduced in the 1960s⁶ and has been optimized over decades with increasing complexity of the combinations and maximization of dose intensity.^3,4 Uptake of consolidation for adult patients with ALL was more gradual due to inconsistent benefit seen in the initial randomized trials,^7,8 but more recent experience has confirmed that the pediatric-inspired treatment regimens are advantageous for adults who can tolerate the toxicity.⁹ In the current era, interest is focused not only on improving remission rates and long-term survival but also on decreasing treatment-related adverse reactions, including late toxicities from cytotoxic chemotherapy.^10,11 Additional intensification of the treatment regimens frequently results in excess toxicity while not improving outcomes;^12,13 however, attempts to reduce toxicity by de-intensification have been unsuccessful largely due to increased rates of relapse.¹⁴ These results suggest that we have reached the limit of the utility of cytotoxic chemotherapy in consolidation for BCP ALL.

BCP ALL is characterized by the expression of CD19, a potential therapeutic target, on lymphoblasts.¹⁵ Blinatumomab is a bispecific CD19-directed CD3 T-cell engager with a mechanism of action distinct from that of cytotoxic chemotherapy. Blinatumomab was first approved by the FDA in 2014 for treatment of relapsed or refractory CD19-positive BCP ALL;¹⁶ it is the only targeted therapy that has demonstrated a survival advantage over chemotherapy in this setting.¹⁷ Subsequently, blinatumomab was approved for treatment of CD19-positive BCP ALL in first or second complete remission (CR) with minimal residual disease (MRD) ≥ 0.1%.¹⁸ Herein, we provide a summary of the FDA review of the marketing application for blinatumomab in the consolidation phase of multiphase treatment of Philadelphia (Ph) chromosome-negative CD19-positive BCP ALL.

CLINICAL DEVELOPMENT PROGRAM

The clinical development program consisted of three randomized controlled trials (Studies E1910, 20120215, and AALL1331) testing blinatumomab cycles plus chemotherapy cycles (blin + chemo) vs chemotherapy cycles alone (chemo) or testing blinatumomab cycles (blin) as a replacement for chemo in the consolidation phase of multiphase treatment of CD19-positive BCP ALL (Table 1; Supplementary Figures S1–S3). FDA also considered the reported results from NL59901.078.17 (EUDRACT 2016–004674-17), an investigator-sponsored single-arm trial with an external control evaluating blin + chemo in infants with newly diagnosed KMT2A-rearranged BCP ALL (Supplementary Figure S4)¹⁹ in the context of a meta-analysis. Overall, there were 6 comparisons made in the 4 trials (Table 1).

Table 1:

Summary of Clinical Trials Reviewed

Study	Design	Population	OS Endpoint	Treatment Groups	Consolidation Arms	FDA Review Consideration
E1910	Phase 3 Randomized	Adults 30 to 70 years old with newly-diagnosed Ph-negative B-ALL	Primary	MRD−	Blin + chemo (n=112) vs chemo (n=112)	Efficacy
				MRD+	Blin + chemo (n=40) vs chemo (n=22)	Supportive
20120215	Phase 3 Randomized	Pediatric patients 28 days to 18 years old with first relapse of Ph-negative B-ALL	Key Secondary	High-risk	Blin (n=54) vs chemo (n=57)	Efficacy
AALL1331	Phase 3 Randomized	Pediatric and young adult patients 1 to < 31 years old with first relapse of Ph-negative B-ALL	Key Secondary	Low-risk	Blin + chemo (n=127) vs chemo (n=129)	Supportive
				High- or intermediate-risk	Blin (n=107) vs chemo (n=109)	Supportive
NL59901.078.17	Single-Arm with Historical Control	Infant patients < 1 year old with newly-diagnosed KMT2Ar (Ph-negative) B-ALL	Secondary	KMT2Ar Infant B-ALL	Blin + chemo (n=30) vs chemo historical control (n=214)	Supportive

Source: FDA Analysis

Abbreviations: Ph: Philadelphia chromosome; B-ALL: B cell precursor acute lymphoblastic leukemia; MRD: minimal residual disease; KMT2Ar: KMT2A-rearranged

See Supplementary Figures S1–S4 for additional details on the trial designs

Dose Selection

The recommended blinatumomab dosage for consolidation regimens is 28 μg/day (15 μg/m²/day for those < 45 kg) as a continuous IV infusion for 4 weeks followed by a 2 week treatment-free interval administered as monotherapy. The proposed dosage provided steady state exposures in adult and pediatric patients with BCP ALL in the consolidation phase that were comparable across the body weight categories and across pediatric subgroups by age (Supplementary Figure S5). The disposition of blinatumomab in consolidation also did not differ by whether patients were being treated for newly diagnosed or first relapse of BCP ALL. Because only a single dosage was studied, the data did not enable identification of an exposure-response relationship; however, the steady state exposures observed were comparable to those for adult and pediatric patients treated for R/R ALL, at approximately 616 pg/mL and 533 pg/mL, respectively (Supplementary Figure S5).²⁰

REVIEW OF EFFICACY

Study E1910

Design:

E1910 was a randomized controlled study in adult patients (30 to 70 years old) with newly diagnosed Ph-negative BCP ALL in hematologic complete remission (CR) or CR with incomplete peripheral blood count recovery (CRi) following induction and intensification chemotherapy (Supplementary Figure S1).²¹ Consolidation chemotherapy was based on a BFM-like regimen adapted from Study E2993/UKALLXII. Patients were randomized to blin + chemo (total 8 cycles) or chemo alone (total 4 cycles) (Supplementary Figure S1). Patients who did not proceed to allo-HSCT received maintenance therapy through 2.5 years after the start of intensification.

E1910 was designed originally with a primary objective of comparing the overall survival (OS) of blin + chemo versus chemo in the subgroup of participants who were MRD-positive (> 0.01% by the clinical trial assay) at randomization. If the results were significant, OS would then be tested in the MRD-negative (≤ 0.01%) subgroup and in the total population. However, following the accelerated approval of blinatumomab for MRD-positive BCP ALL in 2018, the protocol was modified so that MRD-positive patients were nonrandomly assigned to blin + chemo, and only the MRD-negative patients were randomized (Supplementary Figure S1).²¹ After the protocol modification, the stratification factors included age (< 55 years vs ≥ 55 years), CD20 status (positive vs negative), rituximab use (yes vs no), and intent to undergo allo-HSCT (yes vs no). The primary endpoint was OS. With a target accrual of 190 patients and 94 events, the study would have 80% power to detect an OS hazard ratio of 0.55 with one-sided alpha of 0.025. A secondary objective included a descriptive analysis of relapse-free survival (RFS). Because the submission did not include the data needed for independent adjudication of RFS, the results of this objective could not be evaluated.

Efficacy Results for the E1910 MRD-Negative Cohort:

The randomized MRD-negative cohort included 112 patients randomized to blin + chemo and 112 to chemo. The baseline characteristics are shown in Supplementary Table S1. There were 28 (25%) patients in the blin + chemo arm and 33 (29%) patients in the chemo arm who underwent allo-HSCT. At the third interim analysis (data cutoff date 7/19/2022), the prespecified efficacy boundary for the MRD-negative cohort was crossed; the blin + chemo arm had significantly improved OS when compared to the chemo arm (HR 0.42, 95% CI 0.24, 0.75, p = 0.003) (Table 2, Figure 1A, Supplementary Table S2). This was a positive trial. In a descriptive analysis of updated OS data with a later data cutoff date of 6/23/2023 and a median follow-up of 4.5 years, the 5-year OS rate was 82.4% (95% CI 73.7, 88.4) in the blin + chemo arm and 62.5% (95% CI 52.0, 71.3) in the chemo arm (HR 0.44, 95% CI 0.25, 0.76). The treatment effect was consistent when allo-HSCT was included as a time-dependent variable (OS HR 0.44, 95% CI 0.26, 0.77).

Table 2:

Overall Survival and Relapse-free Survival Results

Study	Treatment Groups	Consolidation Arms	OS HR (95% CI)	RFS HR (95% CI)
E1910	MRD−	Blin + chemo (n=112) vs chemo (n=112)	0.42 (0.24, 0.75) p = 0.003*	N/A
	MRD+	Blin + chemo (n=40) vs chemo (n=22)	0.40 (0.14, 1.13)	N/A
20120215	High-risk	Blin (n=54) vs chemo (n=57)	0.35 (0.17, 0.70)	0.38 (0.22, 0.66)
AALL1331	High- or intermediate-risk	Blin (n=105) vs chemo (n=103)	0.71 (0.47, 1.08)	N/A
	Low-risk	Blin + chemo (n=127) vs chemo (n=128)	0.46 (0.24, 0.88)	N/A

Source: FDA analysis

Abbreviations: OS: overall survival; RFS: relapse-free survival; MRD: minimal residual disease; Blin: blinatumomab; chemo: chemotherapy

^*Only comparison that met the prespecified criteria for statistical significance

Figure 1. Outcomes in the Clinical Trials.

Kaplan-Meier curves for OS (A, B, C, E, F) and RFS (D) for Study E1910 (A, B), Study 20120215 (C, D), and Study AALL1131 (E, F). The hazard ratios (HR) from a stratified Cox regression, using the stratification factors described in the text, are shown for each comparison. Abbreviations: HR, hazard ratio; OS, overall survival; RFS, relapse-free survival. Source: FDA Analysis.

Efficacy Results for the E1910 MRD-Positive Cohort:

There were 62 patients in the MRD-positive cohort, 40 treated on the blin + chemo arm and 22 treated on the chemo arm. Of note, 18 of the 40 patients (45%) who received treatment on the blin + chemo arm were not randomized but were assigned to this arm after the March 2018 protocol amendment. In an ad hoc analysis of OS, the HR was 0.40 (95% CI 0.14, 1.13) (Table 2, Figure 1B, Supplementary Table S2).

Study 20120215

Design:

20120215 was a randomized controlled study in pediatric patients 1 month to 17 years old with high-risk first relapse of Ph-negative BCP ALL (Table 1, Supplementary Figure S2).²² All patients had received induction therapy and two cycles of consolidation chemotherapy prior to screening and randomization. For the third cycle of consolidation, patients were randomized to either receive one cycle of blin or one cycle of chemo (IntReALL HC3 regimen) prior to proceeding to allo-HSCT.²² Randomization was stratified by age, marrow status at randomization, and MRD status at the end of induction.

The primary endpoint of the study was event-free survival (EFS), defined as the time from randomization to relapse (> 5% bone marrow blasts) after achieving CR, failure to achieve a CR at the end of treatment, second malignancy, or death from any cause. With a target accrual of 202 patients and 95 events, the study would have 84% power to detect an EFS HR of 0.63 with a 2-sided alpha level of 0.05. There were two interim analyses for EFS planned, when ~50% and ~75% of the EFS events had occurred, a primary analysis when all 94 events had occurred, and a final analysis when all patients completed 36 months of follow-up after allo-HSCT or discontinued earlier. OS was a prespecified key secondary endpoint. RFS, defined as the time from randomization to relapse or death from any cause, was added as an ad hoc analysis.

Efficacy Results for 20120215:

At the time of the first interim analysis (data cutoff date 7/17/2019), 54 patients were randomized to each arm. Three additional subjects were accrued by the final analysis. The baseline characteristics are shown in Supplementary Table S3. At the time of the final analysis, 51 (94%) patients in the blin arm and 47 (82%) patients in the chemo arm underwent allo-HSCT, and 13 (24%) in the blin arm and 28 (49%) in the chemo arm underwent chimeric antigen receptor (CAR) T cell therapy. At the time of the first planned interim analysis, the reported EFS HR was 0.36 (95% CI: 0.19, 0.66; p < 0.001), and enrollment was terminated early as prespecified for the positive outcome.

Because EFS as defined by the protocol is not considered a clinical benefit for treatments of consolidation, FDA’s review focused on OS and RFS. OS was a key secondary endpoint, and alpha control was prespecified for the primary analysis of OS; however, the statistical analysis plan did not specify how alpha would be used for OS if an interim analysis for EFS became the primary analysis. Therefore, the OS results from Study 20120215 were considered descriptive. At the final analysis (data cutoff date 11/21/2022), the OS HR was 0.35 (95% CI 0.17, 0.70). The treatment effect was consistent when allo-HSCT was included as a time-dependent variable (OS HR 0.35, 95% CI 0.17, 0.72). The 5-year OS was 78.4% (95% CI 64.2, 87.4) in the blin arm and 41.4% (95% CI 26.3, 55.9) in the chemo arm (Table 2, Figure 1c, Supplementary Table S2). At the final analysis, the RFS HR was 0.38 (95% CI 0.22, 0.66); the 5-year RFS was 61.1% (95% CI 46.3, 72.9) in the blin arm and 27.6% (95% CI 16.2, 40.3) in the chemo arm (Table 2, Figure 1d, Supplementary Table S2).

Study AALL1331

Design:

AALL1331 was a randomized controlled study in pediatric and young adult patients (1 to 30 years old) with first relapse of Ph-negative BCP ALL.^23,24 Patients were randomized within two independent cohorts, high/intermediate-risk (HR/IR) and low-risk (LR) (Supplementary Figure S3). For both randomizations, the treatment was based on the United Kingdom Medical Research Council ALLR3 study (mitoxantrone arm). The primary efficacy endpoint in both cohorts was disease-free survival (DFS), defined as time from randomization to relapse (marrow blasts > 25%), treatment failure, second malignancy, or death. OS was the key secondary endpoint. Because the submission did not include the data needed for independent adjudication of RFS, an analysis of RFS was not performed.

For the HR/IR cohort, randomization was conducted after induction and was stratified by risk group, site of relapse, duration of first remission, and End Block 1 MRD. Patients on Arm A received two blocks of chemo, and those on Arm B received two blocks of blin prior to proceeding to allo-HSCT. There were four interim analyses planned, and a primary analysis 2 years after completion of enrollment. The study was designed to have at least 80% power to detect a DFS HR of 0.58 with one-sided alpha 2.5% in the HR/IR cohort.

For the LR cohort, randomization was conducted after Block 1 and was stratified by site of relapse and End Block 1 MRD (<0.01% vs ≥ 0.01% and <0.1%). Patients on Arm C received four blocks of chemo (total four blocks); patients on Arm D received three blocks of chemo and three blocks of blin (total six blocks) prior to initiation of maintenance therapy. There were five planned interim analyses, and a primary analysis when approximately 80 events occurred or 3 years following completion of enrollment, whichever was earlier. The study as-designed had at least 80% power to detect a DFS HR of 0.55 with one-sided alpha 5% in the LR cohort.

Efficacy Results for AALL1331 HR/IR Cohort:

The HR/IR randomization was stopped early at a planned interim analysis (data cutoff date 6/30/2019), citing less toxicity in the blinatumomab arm but without meeting stopping rules for efficacy or futility. At that time, there were 107 patients randomized to the blin arm and 109 to the chemo arm. The baseline characteristics are shown in Supplementary Table S4. Because the analysis of the primary endpoint was not positive, the OS results are only descriptive. With a data cutoff date of 12/31/2022, the OS HR was 0.71 (95% CI: 0.47, 1.08) for the HR/IR group (Figure 1e, Table 2, Supplementary Table S2). Following study treatment, 85 (79%) patients in the blin arm and 67 (62%) patients in the chemo arm underwent allo-HSCT, and 21 (20%) in the blin arm and 33 (30%) in the chemo arm underwent chimeric antigen receptor (CAR) T cell therapy. With allo-HSCT as a time-dependent covariate, the OS HR was 0.79 (0.52, 1.20).

Efficacy Results for AALL1331 LR Cohort:

The LR randomization read out to completion but did not meet the primary DFS objective (data cutoff date 12/31/2020). This was a negative trial. At the time of analysis, there were 127 patients randomized to the blin + chemo arm and 129 to the chemo arm. The baseline characteristics are shown in Supplementary Table S4. In a descriptive analysis with a data cutoff date of 12/31/2022, the OS HR was 0.46 (95% CI 0.24, 0.88) (Figure 1f, Table 2, Supplementary Table S2). Following study treatment, 28 (22%) patients in the blin + chemo arm and 31 (24%) patients in the chemo arm underwent allo-HSCT, and 34 (27%) in the blin + chemo arm and 42 (33%) in the chemo arm underwent chimeric antigen receptor (CAR) T cell therapy. With allo-HSCT as a time-dependent covariate, the OS HR was 0.45 (95% CI 0.23, 0.87).

Literature Search

A search of the literature yielded one additional study which quantitated the treatment effect of blinatumomab in consolidation. Study NL59901.078.17 was a single-arm study from the Interfant Network that assessed one cycle of blinatumomab postremission followed by the Interfant-6 regimen for infant patients with newly diagnosed CD19-positive, Ph-negative, KMT2A-rearranged BCP ALL (Supplementary Figure S4).¹⁹ Thirty infants were treated on the trial. OS was compared with that of 214 patients selected from 484 Interfant-06 historical controls based on meeting the eligibility criteria used in NL59901.078.17 and availability of MRD at the End of Induction (EOI). The reported analysis showed an OS HR of 0.15 (95% CI 0.04, 0.62) (adjusted for risk group, age at diagnosis, WBC at diagnosis, and MRD EOI); the 2-year OS rate was 93.3% (95% CI 75.9, 98.3) in the NL59901.078.17 cohort and 65.8% (95% CI 58.9, 71.8) in the Interfant-6 historical controls.¹⁹

Meta-Analyses

While all OS HR point estimates in the comparisons discussed above favored the blinatumomab-containing arm (Table 2), several of the six comparisons represent essentially negative trials statistically. The question arose as to whether the positive result for E1910 was a chance outcome among multiple trials, or whether there was a true treatment effect on OS that could not be demonstrated in the individual comparisons due to reduced power or due to the inappropriate early study termination. A meta-analysis of all six comparisons was performed using a common effects model (Supplementary Figure S5). The results were consistent with a treatment effect for blinatumomab (OS HR 0.50; 95% CI 0.38, 0.64). A similar outcome was obtained using a random effects model. The applicant also reported similar outcomes when assessed by disease status (OS HR 0.38; 95% CI 0.24, 0.61 for the three first-line treatment comparisons and OS HR 0.54; 95% CI 0.38, 0.77 for the three first-relapse treatment comparisons).

REVIEW OF SAFETY

The overall safety profile of blinatumomab has been well-characterized in the reviews of blinatumomab for treatment of R/R and MRD+ BCP ALL.^16–18 To focus the safety relevant to consolidation, the current analyses compared the adverse event profile of blin + chemo versus chemo (Study E1910) and of blin versus chemo (Study 20120215).

Blinatumomab + Chemotherapy versus Chemotherapy Alone (Study E1910)

The safety population for E1910 included 111 patients treated with blin + chemo and 112 patients treated with chemo as consolidation. The median (range) of treatment cycles was 8 (1–8) in the blin + chemo arm and 4 (1–4) in the chemo arm. Fatal adverse reactions occurred in 2 patients during blinatumomab cycles (one infection and one coagulopathy). Adverse reactions resulted in discontinuation of blinatumomab in 2% of patients, interruptions in 5% of patients, and dose reductions in 28% of patients. The most common adverse reactions during consolidation cycles on the blin + chemo arm were neutropenia (81%), thrombocytopenia (74%), anemia (58%), leukopenia (43%), headache (41%), infection (35%), nausea (32%), lymphopenia (31%), diarrhea (28%), musculoskeletal pain (22%), and tremor (22%). The most common grade 3–5 adverse reactions during consolidation cycles on the blin + chemo arm were neutropenia (76%), thrombocytopenia (56%), leukopenia (40%), infection (30%), lymphopenia (30%), and anemia (29%).

Table 3 shows the all-grade adverse reactions with a risk difference of ≥ 10 percent between the arms in all blocks of consolidation. CRS, CRS-related, and neurologic adverse reactions were higher in patients on the blin + chemo arm, but grade 3–5 febrile neutropenia and hematologic adverse reactions were lower, despite patients in both arms receiving identical chemotherapy cycles during treatment (Table 3).

Table 3:

Safety Results from E1910 Consolidation Cycles

Adverse Reaction *	Blinatumomab + Chemotherapy N=111	Chemotherapy N=112	Risk Difference (%)
	N (%)	N (%)
All Grade with Risk Difference ≥ 10%
Tremor	25 (23)	3 (3)	20
Musculoskeletal pain	25 (23)	6 (5)	17
Cytokine release syndrome	18 (16)	0	16
Diarrhoea	32 (29)	17 (15)	14
Infection	39 (35)	25 (22)	13
Headache	46 (41)	33 (30)	12
Nausea	36 (32)	25 (22)	10
Pyrexia	17 (15)	6 (5)	10
Aphasia	11 (10)	0	10
Leukopenia	48 (43)	64 (57)	−14
Grade ≥ 3 with Risk Difference ≥ 5%
Infection	34 (31)	23 (21)	10
Aphasia	9 (8)	0	8
Hypertension	11 (10)	3 (3)	7
Lymphopenia	33 (30)	26 (23)	7
Febrile neutropenia	21 (19)	28 (25)	−6
Anaemia	32 (29)	43 (38)	−11
Neutropenia	85 (77)	100 (89)	−13
Thrombocytopenia	63 (57)	79 (71)	−14
Leukopenia	45 (41)	63 (56)	−16

Source: FDA Analysis

^*Includes Grouped Terms (see Supplementary Table S5)

Blinatumomab vs Chemotherapy (Study 20120215)

The safety population for 20120215 included 54 patients treated with blinatumomab and 52 patients with chemotherapy for the last cycle of consolidation. There were no fatal adverse reactions in patients treated with blinatumomab. Adverse reactions resulted in discontinuation of blinatumomab in 4% of patients, and interruptions in 11%. Neurotoxicity accounted for the majority of the adverse reactions in these cases. The most common adverse reactions in the blin arm were pyrexia (76%), nausea (43%), headache (37%), hypogammaglobulinemia (24%), anemia (24%), and rash (22%). Table 4 shows the all-grade adverse reactions with a risk difference of 10 percent or more between the arms. CRS or neurologic toxicities, including fever and headache, were higher in patients treated with blin for consolidation, while both all-grade and grade 3–5 pancytopenia, infection, febrile neutropenia, and stomatitis were higher in patients treated with chemo.

Table 4:

Safety Results from 20120215 Consolidation Cycle

Adverse Reaction *	Blinatumomab N=54	Chemotherapy N=52	Risk Difference (%)
	N (%)	N (%)
All Grade with Risk Difference ≥ 10%
Pyrexia	41 (76)	10 (19)	57
Headache	20 (37)	8 (15)	22
Hypogammaglobulinemia	13 (24)	6 (12)	12
Rash	12 (22)	6 (12)	12
Nausea	23 (43)	16 (31)	12
Abdominal pain	7 (13)	12 (23)	−10
Hemorrhage	6 (11)	12 (23)	−12
Infection	7 (13)	15 (29)	−16
Neutropenia	10 (19)	18 (35)	−16
Liver function test abnormal	5 (9)	14 (27)	−20
Musculoskeletal pain	5 (9)	15 (29)	−20
Anemia	13 (24)	24 (46)	−22
Febrile neutropenia	1 (2)	13 (25)	−23
Thrombocytopenia	8 (15)	20 (39)	−24
Stomatitis	6 (11)	31 (60)	−49
Grade ≥ 3 with Risk Difference ≥ 5%
Pyrexia	3 (6)	0	6
Aplasia	0	4 (8)	−8
Liver function test abnormal	3 (6)	9 (17)	−13
Neutropenia	9 (17)	16 (31)	−14
Thrombocytopenia	8 (15)	18 (35)	−20
Febrile neutropenia	1 (2)	13 (25)	−23
Stomatitis	2 (4)	15 (29)	−25
Anaemia	8 (15)	22 (42)	−27

Source: FDA Analysis

^*Assessed from the start of study therapy through 30 days from the end of the assigned treatment cycle, or until the start of the allo-HSCT preparative regimen, whichever occurred first.

Includes Grouped Terms (see Supplementary Table S5).

Adverse Events of Special Interest Across Studies

For blinatumomab, the adverse events of special interest (AESIs) include CRS, neurologic toxicities, fever, and risk of infection. To assess consistency across trials, AESIs were enumerated in E1910, AALL1331, and 20120215 (Supplementary Table S6). In the clinical trials, Common Terminology Criteria for Adverse Events (CTCAE) version 4 criteria were used to grade CRS. The rates of CRS and neurological events were consistent across trials with multiple cycles of blin and lower in 20120215 using a single cycle of blin. The incidence of grades 3 or higher CRS were consistently low (0–4%) across all studies. Immune effector cell-associated neurotoxicity syndrome (ICANS) was not assessed as an entity in these trials; encephalopathy, the adverse reaction closest to ICANS, was reported in only 0–4% of patients. All-grade neurotoxicity, however, was high (37–63%) across studies, consistent with the neurologic toxicity profile that was seen in prior approvals. The incidences of febrile neutropenia and infection were also consistent across trials, but fever appeared to occur more frequently in the pediatric patients (AALL1331 and 20120215) than in adult patients (E1910).

REGULATORY INSIGHTS

The clinical development program for blinatumomab in the consolidation phase of multiphase chemotherapy included 3 randomized controlled trials in patients with Ph-negative BCP ALL. For regular approval of an indication for use in consolidation for acute leukemia, OS or RFS would be acceptable measures of clinical benefit. Study E1910 had a positive analysis for OS, albeit with some uncertainty due to the major revision in trial design after initiation. Study 20120215 might also have been positive for OS with statistical rigor but for lack of clarity in the statistical analysis plan; however, we do note that Study 20120215 was a positive study for its predefined primary EFS endpoint. In Study AALL1331, failure of the primary analyses precluded firm conclusions about OS. Acknowledging that the OS meta-analysis pools effects across a variety of settings and regimens and is exploratory, the results were consistent with a treatment effect for blinatumomab in consolidation. Thus, anchored by the results of Study E1910 with a statistically significant and clinically meaningful improvement in OS, FDA concluded that the totality of the results provided substantial evidence of effectiveness for use of blinatumomab in consolidation for patients with Ph-negative BCP ALL (Table 5).

Table 5:

Benefit-Risk Assessment

Dimension	Evidence and Uncertainties	Conclusions and Reasons
Analysis of Condition	The expected survival of patients with untreated newly diagnosed or relapsed BCP ALL is weeks.	BCP ALL is a fatal disease.
Current Treatment Options	The current standard of care for treatment of newly diagnosed or relapsed Ph-negative BCP ALL with curative intent is intensive multiphasic combination chemotherapy including at least induction and consolidation. With treatment for newly diagnosed disease, the survival is about 50% for adults and 85% for pediatric patients. With treatment with curative intent for relapsed disease, the survival is about 10% for adults and 50% for pediatric patients.	New treatments are needed to improve survival of patients with Ph-negative BCP ALL.
Benefit	E1910 was a randomized trial comparing chemotherapy alone to chemotherapy plus blinatumomab as consolidation in adults with new diagnosed Ph-negative BCP ALL. In the population identified as “MRD-negative”, the OS HR was 0.42 (95% CI: 0.24, 0.75; p=0.003). 20120215 was a randomized trial comparing chemotherapy vs blinatumomab as the third cycle of consolidation in pediatric patients with high risk first relapse of Ph-negative BCP ALL. The OS HR was 0.35 (95% CI: 0.17, 0.70). In a study-level meta-analysis of 6 comparisons from 4 trials for Ph-negative BCP ALL, the OS HR was 0.46 (95% CI: 0.33, 0.64), and the treatment effect was consistent across disease status and trial design. Although there was substantial heterogeneity in these trials, the results are considered supportive.	Use of blinatumomab was associated with a survival advantage when used in consolidation for the regimens employed in E1910 and 20120215 for treatment of Ph-negative BCP ALL.
Risks and Risk Management	The overall safety profile in patients treated with blinatumomab in consolidation was similar to that seen in patients with MRD-positive ALL or with advanced ALL treated with blinatumomab. While efficacy could be extrapolated across age groups, safety was supported by extant data available for patients at least 1 month old.	The safety profile of blinatumomab is acceptable for the intended population. Serious risks can be managed with labeling.

Source: FDA Analysis

Consistency in the treatment effect across endpoints within a trial provides supporting information. A weakness in the clinical development program was that none of the studies included hypothesis testing for RFS. Further, two studies (E1910 and AALL1331) did not include the data needed to adjudicate the RFS endpoint at all. Fortunately, the results of the post hoc analysis of RFS in Study 20120215 were supportive of a treatment effect of blinatumomab. While the major focus of cooperative group trials may be on advancing the standard of care, partnering with industry to generate the data to support marketing applications for new drugs would seem consistent with that intent, so the lack of attention to the RFS endpoint in the blinatumomab trials is perplexing, especially since the remedy would be to simply plan the statistical analyses, without requiring any additional patient testing or intervention. This endpoint need not be the primary endpoint of cooperative group studies, but a prospective plan for ensuring the appropriate patient-level data are collected to test for it would have allowed for more robust analysis of these studies. The ACCELERATE Fit for Filing Working Group²⁵ provided recommendations to improve the success of industry partnerships with cooperative groups in order to maximize leverage of clinical trial data, and we wholly support their position.

Although improvement in OS with blinatumomab in consolidation was shown only for first-line treatment in adults and for treatment of relapse in pediatric and young adult patients, FDA granted a broader indication across age groups and disease settings by extrapolation. The framework for extrapolation is based on the similarity of disease, drug pharmacology, and response to treatment between a reference population and a target population.²⁶ In the US, extrapolation accounts for the majority of approvals in the pediatric population and is used commonly outside the pediatric population as well.^27,28 FDA’s review of this blinatumomab application concluded that there was adequate similarity across age groups in Ph-negative BCP ALL disease biology with regard to the target CD19 molecule to enable drug activity,^17,18 similarity in the treatment paradigm when limited to multiphase intensive chemotherapy, and adequate clinical data to establish doses that provide safe and effective exposure across age groups (Supplementary Figure S6).²⁰ This was supported by the experience with blinatumomab monotherapy in the pediatric and adult population with relapsed/refractory and MRD+ BCP ALL, in addition to the studies reviewed as part of this application^16–18. It should be noted, however, that this degree of extrapolation was largely enabled by the robust and productive clinical trial experience with blinatumomab and would not likely be possible for new molecular entities with small single-trial programs.

Additionally, the extent to which extrapolation is applied may vary by global regulatory agency. The first approval of blinatumomab for consolidation in the European Union and associated areas in 2021 was limited to pediatric patients 1 year or older with high-risk first relapsed Ph-negative CD19-positive BCP ALL as part of consolidation therapy based on the results of Study 20120215. The current blinatumomab application, inclusive of Studies E1910, 20120215, and AALL1331, was reviewed under the Oncology Center of Excellence’s initiative Project Orbis, allowing for a collaborative but independent review process among international regulatory partners, including the United Kingdom (MHRA), Brazil (ANVISA), Canada (Health Canada), and Switzerland (SwissMedic).²⁹ Because each country’s decision is based on its own regulations and local convention, the approved indications and intended populations differ by country (Supplementary Table S7). As a consequence, the standard of care for BCP ALL may differ by country as well, which may subsequently complicate the design or interpretation of future global trials.

Shortly after the approval, the ongoing Children’s Oncology Group Study AALL1731, which included a randomization to blin + chemo versus chemo alone for pediatric patients with standard risk newly diagnosed BCP ALL, underwent a planned interim analysis for the primary efficacy endpoint of DFS.^30,31 Results were presented for patients at average risk of relapse (SR-Avg) and high risk of relapse (SR-High), with both cohorts showing an improvement in 3-year-DFS in the blin + chemo arm (SR-Avg HR 0.33, 95% CI 0.16–0.69; SR-High HR 0.45, 95% CI 0.24–0.85). However, we note that DFS is not an accepted regulatory endpoint and that at the time of publication, there was no difference in OS between arms. This exemplifies the difficulty in improving OS in a population with high rates of OS, and we await the results of RFS analysis and the final OS results to confirm the positive DFS results of this study.

FDA identified no new safety issues for blinatumomab when used in the consolidation phase. Two potential hazards were anticipated for the postmarket setting. First, there should be no a priori expectation of benefit when blinatumomab is simply added to a different regimen. When used in regimens that are very different from those in E1910 or 20120215, the incidence or intensity of the expected toxicities may differ from those listed in the USPI. Second, the safety data in the USPI is for blinatumomab monotherapy alone. Use of blinatumomab in a multiagent combination cycle would be investigational, and selection of a safe dose would require the standard pathway for development of a new combination regimen. To mitigate these risks in practice, labelling is specific about the use of blinatumomab as monotherapy in the consolidation cycles and specifies the exact chemotherapy used to generate the safety and efficacy data.

Conclusions

Given the observed survival improvement, and with adequate labelling in place to mitigate risks, the clinical benefit of blinatumomab appears to outweigh the risks for treatment of CD19-positive Ph-negative BCP ALL in the consolidation phase of multiphase chemotherapy in adult and pediatric patients one month and older.