Phase 1 Clinical Safety, Pharmacokinetics (PK), and Activity of Apilimod Dimesylate (LAM-002A), a First-in-Class Inhibitor of Phosphatidylinositol-3-Phosphate 5-Kinase (PIKfyve), in Patients with Relapsed or Refractory B-Cell Malignancies

Introduction : PIKfyve is an endosomal lipid kinase that phosphorylates P1(3)P to phosphatidylinositol 3,5-bisphosphate (PI[3,5]P2). Unlike PI3Kδ-related signaling, PIKfyve regulates endosomal membrane trafficking and has a critical role in autophagy, which exposes damaged proteins to proteases (cathepsins) in autophagolysosomes. Enhanced autophagy can promote cancer cell survival while inhibition can promote tumor cell death. LAM-002A is a highly potent and selective inhibitor of PIKfyve that disrupts lysosomal homeostasis resulting in cytotoxic activity in multiple B-cell cancers, including diffuse large B-cell lymphoma (DLBCL) harboring c-myc or Bcl-2 translocations. LAM-002A monotherapy and combinations (including with anti-CD20 and anti-PDL1 antibodies) demonstrate significant in vitro and in vivo antitumor activity in lymphoma models (Gayle et al., Blood 2017;129(13):1768).

Methods and Patients :This Phase 1 study evaluated the safety, PK, and preliminary clinical activity of orally administered LAM-002A in patients with previously treated B-cell cancers. In a 3+3 dose escalation, sequential cohorts of patients were enrolled to receive LAM-002A administered 2 or 3 times per day (BID or TID) continuously in 28-day cycles. PK was assessed over 8 hours postdose on Days 1 and 8. Antitumor activity was evaluated at 6- to 12-week intervals.

Results : At data cutoff, the study had enrolled 24 patients. The study population comprised 12 males and 12 females with median age (range) of 69.5 (53-89) years. Diagnoses were DLBCL (n=11, including germinal center B-cell (GCB) subtype [n=6], non-GCB subtype [n=3], unknown [n=2]; and transformation from follicular lymphoma [FL] [n=3] or marginal zone lymphoma [MZL] [n=1]); mantle cell lymphoma (MCL) (n=4), chronic lymphocytic leukemia (CLL) (n=4), MZL (n=3), or FL (n=2). The median (range) number of prior therapies was 4 (2-11). The sequence of evaluation of dosing regimens was 50 mg BID (n=3), 100 mg BID (n=8), 150 mg BID (n=5), 75 mg TID (n=4), and 125 mg BID (n=4). Durations of therapy ranged from <1 cycle to 12+ cycles. Regimens of 50 mg BID and 100 mg BID were well tolerated. At 150 mg BID, events of Grade 1-3 nausea and Grade 1-2 diarrhea occurred in 4/4 evaluable patients, resulting in drug discontinuation (n=3) or dose reduction due to diarrhea-related hyponatremia (n=1). The regimen of 75 mg TID was also associated with nausea and vomiting, resulting in drug discontinuation in 2/4 patients. The regimen of 125 mg BID has not shown dose-limiting toxicities in 3 evaluable patients and is being explored in additional subjects. One patient with DLBCL developed Grade 4 tumor lysis syndrome (TLS) shortly after starting LAM-002A at 125 mg BID. Other adverse events or laboratory findings during the study were typically low-grade and not clearly drug-related. PK data showed LAM-002 to be rapidly bioavailable with a mean time of maximum concentration (T_max) of 1 hour. Maximum concentration (C_max) and area under the plasma concentration-time curve (AUC_t) values increased with increasing dose in a generally dose-proportional manner. Repeat dosing did not affect the C_max but the AUC_t increased from Day 1 to Day 8 as steady-state exposure was attained. The mean terminal elimination half life (t_1/2) ranged from ~3.0 to 6.5 hours without dose dependency. Positron emission tomography (PET) demonstrated systemic partial metabolic responses in nodal and extranodal lesions in 3 patients with refractory DLBCL-GCB (treated at 100 mg BID, 75 mg TID, and 125 mg BID) (1 of whom also received radiation for bulky axillary adenopathy); concomitant computed tomography (CT) demonstrated anatomic shrinkage of many lesions. One additional subject with transformed, refractory DLBCL-GCB had TLS, as noted. One patient with MZL and 1 patient with CLL (both receiving 100 mg BID) have had prolonged SD through 9+ and 12+ cycles, respectively.

Conclusions : As the first PIKfyve inhibitor in clinical development for B-cell malignancies, LAM-002A shows a favorable safety profile at doses of ≤125 mg BID and promising early antitumor activity, including in patients with refractory DLBCL. Together with the nonclinical findings, these data in heavily pretreated patients support further clinical evaluation of LAM-002A, both as a monotherapy and together with chemo/immunotherapy.