Maximum Tolerated Dose (MTD) and Safety Profile from Phase I Study of MT-3724 (CD20 targeted immunotoxin) in Relapsed/Refractory (R/R) Non-Hodgkin B-Cell Lymphoma (NHL)

Background

Novel mechanisms of action (MOA) are needed to treat R/R NHL. CD20 is ubiquitously and persistently expressed in B-cell malignancies. Consequently, there is a strong rationale to develop novel MOAs targeting CD20. However, CD20's non-internalizing nature has impeded the development of novel MOAs against this target. MT-3724 is a recombinant fusion protein consisting of a CD20 binding variable fragment (scFv) fused to the ribosomal inhibitory protein Shiga-like toxin-I A1 subunit (SLT-I A1). Upon scFv binding to surface CD20, SLT-I A1 forces MT-3724 internalization and irreversibly inactivates cell ribosomes triggering cell death. MT-3724 has been shown to specifically bind and kill CD20+ malignant human B-cells in vitro and in in vivo animal models. Final safety data including the MT-3724 MTD confirmed from the 21 enrolled patients (pts) in the phase I monotherapy dose escalation trial are presented.

Methods

This MT-3724 first-in-human, open label ascending dose study (3+3 design) enrolled sequential cohorts of 5, 10, 20, 50, 100, and 75 mcg/kg/dose. Pts who had previously responded to a CD20 monoclonal antibody (MAb) containing therapy followed by relapse received 6 infusions over 2 hours in the first 12 days of a 28 day cycle (cycle 1). Dose escalation was based on < 33% dose limiting toxicities (DLTs) observed during the first 28 day cycle. Pts without disease progression or an adverse event (AE) precluding further MT-3724 dosing could receive up to 4 more 21-day cycles with tumor assessments after cycles 2, 4 and 5. Safety data was collected until all AEs were resolved, unchanged or further tumor therapy was initiated.

Results

Of 27 screened pts (mean number of prior therapies > 4), 21 initiated dosing and 18 completed at least one cycle of treatment, the majority with diffuse large B cell lymphoma (DLBCL). One patient (pt) was withdrawn prior to completing cycle 1 for worsening tumor pain and 2 pts were withdrawn from the 100 mcg/kg/dose cohort for DLTs. Both DLTs were consistent with early signs/symptoms of vascular leak syndrome, a known side effect of some biologics and immunotoxins. These AEs were non-life threatening and reversible upon drug withdrawal. Six pts completed cycle 1 at a step-down dose of 75 mcg/kg/dose without a DLT, confirming this to be the MTD.

There were 3 serious AEs in 2 pts attributed to MT-3724 in cycle 1, both in the 100 mcg/kg/dose cohort: 1 pt with ileus (Grade 2) and severe muscle weakness (Grade 3) and 1 pt with pneumonia (Grade 3). The most common AEs across all cohorts were 14 AEs of peripheral edema (14 </= Grade 2), 9 of fatigue (9 </= Grade 2), 8 of diarrhea (1 > Grade 2) and 8 of myalgia (2 > Grade 2) most of which were attributed to prior therapies. Hypoalbuminemia was reported in 14 pts (14 </= Grade 2) with or without edema, most improving spontaneously. Only 2 pts experienced neutropenic events (both > Grade 2), 1 each in the 5 and 50 mcg/kg/dose cohorts, both of which were attributed to prior therapies by the Data Monitoring Committee. Across all cycles 2 pts experienced at least one Grade 4 AE, and 5 pts experienced at least one Grade 3 AE that were assessed by investigators as possibly or probably related to study treatment. Complete safety data along with cumulative response data including data from the MTD expansion cohort will be presented.

In vitro anti-CD20 MAb competitively inhibits MT-3724 binding to cell surface CD20. PK analyses of pt serum samples suggested that similar competitive binding inhibition was occurring in vivo. MT-3724's greatest effect was observed in more rapidly growing DLBCL consistent with its MOA. Responses to treatment of those pts enrolled with a diagnosis of DLBCL are presented in Table 1 grouped by whether their baseline anti-CD20 MAb level was < 1,000 ng/ml or >/= 1,000 ng/ml at initiation of treatment. Complete anti-drug antibody and PK data will be presented.

Conclusions

The results from this phase 1 first-in-human study of MT-3724 in heavily pre-treated R/R NHL pts supports continued investigation of MT-3724 in R/R DLBCL. Expansion of the 75 mcg/kg/dose (MTD) cohort enrolling only pts with DLBCL to further define efficacy and safety has started. To further explore the effect of residual anti-CD20 MAb on response to MT-3724, enrollment into this expansion cohort will require a lower baseline serum anti-CD20 MAb concentration of < 500 ng/mL at initiation of dosing.

View largeDownload slide

View largeDownload slide

Close modal