Bispecific antibodies in lymphoma, myeloma and acute myeloid leukemia

Allogeneic (allo-) stem cell transplant (SCT) is a procedure performed in the treatment of several hematological malignancies including lymphoma, acute myeloid leukemia (AML) and multiple myeloma (MM). Post-SCT relapse remains a prominent issue, especially following haploidentical, T-cell depleted transplants. Whilst there has been progress made in reducing the rates of graft-versus-host disease (GvHD), less progress has been made in increasing tumor control and reducing relapse rates.¹ Novel agents, such as bispecific antibodies (bsAbs) and chimeric antigen receptor (CAR) T-cell therapy, are being investigated as agents to reduce the tumor load pre-SCT or post-SCT, with the aim of decreasing relapse rates.¹ This article provides an overview of bsAbs based on an article by Einsele et al.¹, and summarizes the latest data on bsAbs in lymphoma, AML and MM.

What are bsAbs?

bsAbs are bispecific immunotherapies that redirect immune effector cells to tumor cells. This redirection can occur by genetically engineering cells to express CARs, tumor specific T-cell receptors (TCRs) or bsAbs. These approaches are typically applied to T-cells and natural killer (NK)-cells. bsAbs are therefore designed to bind and activate T-cells, via CD3, or NK-cells, via CD16. However, enhancing the immune response in this manner may lead to significant toxicity requiring specific management.¹

Immunotherapies in hematological malignancies¹

Tumor cells accumulate DNA mutations which disrupt the normal regulatory processes that cells undergo. These mutations cause the expression of antigens that the adaptive immune system recognizes as non-self, and as such an immune response is elicited. However, this relies on the expression of major histocompatibility complex (MHC) class I and/or class II proteins. Many tumor cells have downregulated MHC expression, preventing this immunosurveillance.

Immunotherapies do not rely on the presence of specific antigens in the T-cell receptor (TCR) repertoire, rather they introduce new specificities via genetic modifications e.g. CAR T-cell therapy, or through the administration of bsAbs. bsAbs allow tumor-directed T-cell activation even when there is low expression of the target antigen, by redirecting the cytotoxic effect of T- and NK-cells, causing tumor cell lysis.

The clinical efficacy and safety of bsAbs depend on:

1. Biological rationale
2. Format of bispecific antibody
3. Potential addition and property of variable fragment (Fc) domain

Response to bsAbs¹

A lack of response to bsAbs has been associated with:

• High disease burden
• High frequency of circulating regulatory T cells
• Concurrent or prior history of extramedullary disease
• Upregulation of immune checkpoint molecules in tumor cell

Bispecific T-cell engager (BiTE) antibodies¹

Most of the currently approved bsAbs are BiTE format antibodies. They weigh between 50–60 kDa and link a single chain antibody Fc (scFv), which binds to the invariant CD3 part of the TCR, to another scFv that binds to surface antigen on the tumor cell, such as CD19. This brings the T-cell and tumor cell into close proximity, inducing T-cell activation and proliferation and causing lysis of the tumor-antigen-expressing target cell. This is independent of MHC I and II presentation.

What are the issues with bsAbs?¹

One main complications of treatment with bsAbs is cytokine release syndrome (CRS). CRS is caused by the release of:

• High levels of inflammatory cytokines released by effector memory T-cells due to the bsAb binding to CD3
• Markers of inflammation such as C-reactive protein

Some cases of CRS are self-limiting and do not require intervention, whilst others are handled by the administration of tocilizumab or interleukin (IL)-1 antibodies. bsAbs have a short half-life, meaning the infusion can be stopped to reduce acute toxicity.

The main risk factors for the development of CRS are 1) tumor load and 2) the starting dose of the bispecific antibody. Therefore, reducing the burden of disease prior to treatment with a bsAb reduces the risk of CRS.

bsAbs targeting CD19 can also be associated with neurotoxicity, though the etiology is unclear. Tocilizumab is unable to control this toxicity as it does not cross the blood-brain barrier and cannot control inflammation in the central nervous system. For this reason, corticosteroids are often given.

How can we improve the efficacy of bsAbs?¹

scFvs used in bsAb constructs are from murine mAbs which can induce human anti-mouse antibodies (HAMA). To reduce HAMA responses, Einsele et al., recommend using humanized mAbs or generating mAbs based on human sequences.

Antigen loss variants emerge in CD19⁺ lymphomas and CD19⁺ precursor B-ALL and splice variants may also be present in cells which have lost the extracellular domain (CD19). To overcome antigen loss, Einsele et al., recommend the generation of trispecific mAbs or the infusion of two bsAbs targeting two different surface antigens on malignant cells.

bsAbs in lymphoma¹

Lymphoma is the only disease indication for which a bsAb is currently approved by the United States (US) Food & Drug Administration (FDA) and European Medicines Agency (EMA). Blinatumomab is a CD3/CD19 BiTE approved for the treatment of adult and pediatric patients with R/R B-cell precursor acute lymphoblastic leukemia (B-ALL) in first or second complete remission (CR) with minimal residual disease (MRD) ≥0.1%.^2,3 Accelerated approval was granted to blinatumomab based on the results of the BLAST study (NCT01207388):^2,4

• Phase II study (N= 116)
• Included adults with MRD-positive ALL in remission after chemotherapy
• 78% achieved MRD response with blinatumomab (88/113)
• Of 110 patients with Philadelphia chromosome (Ph)-negative B-cell ALL in hematologic remission:
  • Estimated relapse-free survival (RFS) at 18-months: 54%
  • Median OS: 36.5 months
• MRD responders:
  • RFS (MRD responder vs non-responder): 23.6 vs 5.7 months, p= 0.002
  • Overall survival (OS, MRD responder vs non-responder): 38.9 vs 12.5 months, p= 0.002
  • Achievement of MRD response was significantly associated with improved overall survival (OS)

There are many ongoing studies investigating blinatumomab in different subtypes of lymphoma, as monotherapy and in combination therapies, and in different populations.

• Results from a phase I study of blinatumomab monotherapy in patients (N= 76) with non-Hodgkin lymphoma (NHL), including diffuse large B-cell lymphoma (DLBCL), mantle cell lymphoma (MCL) and follicular lymphoma (FL), found:⁵
  • Overall response rate (ORR) at target dose of 60 μg/m²/day (n= 35): 69%
    • For patients with DLBCL: 55%
  • Median duration of response (DoR): 404 days
• In pediatric patients with R/R ALL, in a phase I (n= 49) and phase II (n= 44) study, blinatumomab has shown promising efficacy with 39% (27/70) achieving CR within the first two cycles, 52% of whom achieved MRD-negativity.⁶
• Multiple other studies have/are investigating the applications of blinatumomab in lymphoma, including:
  • NCT03023878: Blinatumomab in combination with upfront chemotherapy for high-risk DLBCL⁷
  • NCT02393859: Phase III study in pediatric patients in first relapse with high-risk precursor B-cell ALL: blinatumomab versus standard chemotherapy⁸
  • NCT03114865: Phase I study of blinatumomab as post-allo-SCT remission maintenance in patients with precursor B-cell ALL and B-cell NHL⁹
  • NCT02910063: Phase II/III study in patients with R/R aggressive B-cell NHL. In the phase II portion, participants will receive blinatumomab as monotherapy, and the phase III portion will compare blinatumomab to the investigator’s choice of chemotherapy. In March 2019, a decision was made to not proceed with the phase III portion¹⁰

CRS and neurotoxicity with blinatumomab

For patients receiving blinatumomab, Einsele et al., recommend that patients are admitted as inpatients to be monitored for CRS and neurotoxicity since there is a correlation between tumor load, and incidence of CRS and neurotoxicity. This is further rationale for cytoreduction prior to infusion in patients with a high tumor load.

Einsele et al., recommend a starting dose of 9µg/day followed by 28µg/day for remaining three weeks. If grade three CRS occurs, they recommend stopping the infusion until CRS resolves, and restarting at 9µg/day, however treatment should be discontinued in the case of grade four CRS.

The main preventive treatment for CRS is the administration of steroids. Pretreatment with 20mg of intravenous (IV) dexamethasone prior to each dose and before each intra-cycle dose escalation is recommended. Tocilizumab has been shown to control blinatumomab-induced CRS.

However, tocilizumab does not control neurotoxicity since it cannot cross the blood-brain barrier. The current approach to managing neurotoxicity with blinatumomab is the use of a step-dosing approach whilst monitoring for signs of neurological toxicity via handwriting analysis, and using corticosteroids if symptoms develop.

bsAbs in MM

In MM, studies of bsAbs are ongoing in early phase clinical trials. An early phase I trial (NCT03173430) is investigating the use of blinatumomab after high-dose melphalan and salvage autologous-SCT (ASCT) with the aim of determining whether blinatumomab can control MM post-ASCT, and whether its administration is safe and feasible. The primary outcome measure is number of adverse events (AEs).¹¹

Other bsAbs which target B-cell maturation antigen (BCMA) are in development, such as AMG 420. AMG 420 is a BiTE antibody construct that binds to BCMA on tumor/plasma cells, and CD3 on T-cells. Max Topp, University of Würzburg, Würzburg, DE, recently reported the results from a phase I trial (NCT02514239) of AMG 420 in patients with R/R MM at the American Society of Clinical Oncology (ASCO) annual meeting. AMG 420 was administered by continuous infusion in six-week cycles for up to 10 cycles in the dose escalation phase. Forty-two patients were enrolled and received treatment, with a median age of 65 years, and a median of four prior lines of therapy.¹²

The maximum tolerated dose (MTD) was found to be 400μg/day. CRS occurred in 38% (16/42) of patients with infections in 31% (13/42), ten of which were grade three or higher.  Treatment discontinuations were predominantly due to progressive disease (n= 25), followed by AEs (n= 7), death (n= 4), completion of 10 cycles (n= 3) and withdrawal of consent (n= 1).¹²

At the MTD dose of 400μg/day (n= 10), five MRD-negative CRs were observed, with a 70% response rate lasting for a median of nine months. Dr. Topp and colleagues concluded that AMG 420 had activity in a heavily pretreated population with no major toxicities reported in the dose escalation phase, though three patients subsequently experienced a dose limiting toxicity. Therefore, the risk of infection must be carefully evaluated and managed in future trials.¹²

Read the full article on the MM Hub here.

bsAbs in AML

There are several membrane-expressed antigens that may be viable targets in AML including; CD33, CD123, CD44v6, CLEC12A (CLL1) and FLT3. The use of allo-SCT in AML, even with reduced intensity conditioning (RIC) in patients with high-risk or R/R disease, indicates the disease is susceptible to immunotherapeutic approaches. Currently, there are CD123- and CD33-bsAbs being investigated in ongoing clinical trials for patients with AML.

During the 24^th meeting of the European Hematology Association (EHA), Gail Roboz, Weill Cornell Medicine, New York, US, presented an updated analysis from the first-in-human (FIH) phase I trial of AM564 in patients with R/R AML. AMG564 is a bsAb targeting CD33 (expressed on myeloid derived suppressor cells (MDSCs) and >95% of AML blasts) and CD3. This study evaluated AMV564 in 33 adult patients with AML and included participants who had relapsed after allo-SCT. AMV564 was administered as a 14-day continuous infusion. The most common treatment-emergent AEs (occurring in more than 20% of patients) were CRS and pyrexia. A selective depletion of leukemic blasts in the bone marrow and MDSCs was observed in 17/27 patients. Seven patients received two or more cycles of AMV564; of whom there was one complete remission, one complete remission with incomplete blood count recovery and one partial remission. Dr Roboz and colleagues concluded AMV564 was well-tolerated up to 250mcg with no disease limiting toxicities (DLT) to date. The lead-in dose schedule will be continued up to 450mcg.¹³ AMV564 was previously awarded Orphan Drug Designation (ODD) by the US FDA in November 2017.¹⁴

Read the full article on the AML Global Portal here.

Another bsAb under investigation in AML is AMG 330 which also targets CD33 and CD3. During the 60^th Annual Meeting and Exposition of the American Hematology Society (ASH), Farhad Ravandi, MD Anderson Cancer Center, Houston, US, presented the results of the phase I FIH study of AMG 330 in patients with R/R AML. Forty patients were enrolled in 12 dose cohorts. Patients had received an average of four prior therapies. Discontinuations occurred in 88% of patients due to PD (n= 27), AEs (n= 6) and patient request (n= 2). Serious AEs occurred in 73% of patients. Two patients achieved a CR at 240μg/d. Dr Ravandi and colleagues concluded that these results validated the use of CD33 as a target in the treatment of AML, and provided early evidence that AMG 330 was tolerable, with anti-leukemic activity in heavily pretreated patients with AML.¹⁵

Read the full article on the AML Global Portal here.

Conclusion

Having waited over 50 years for the first bsAb in hematological malignancy to be approved by the FDA, Einsele et al., believe the coming years will see this technology develop at a rapid pace, bringing this promising technology to the clinic in a multitude of patient and disease settings.¹