A phase I study of veliparib in combination with metronomic cyclophosphamide in adults with refractory solid tumors and lymphomas

Abstract

Purpose: Oral administration of the alkylating agent cyclophosphamide at low doses, metronomic dosing, is well tolerated, with efficacy in multiple tumor types. PARP inhibition potentiates effects of cyclophosphamide in preclinical models. We conducted a phase I trial of the PARP inhibitor veliparib and metronomic cyclophosphamide in patients with refractory solid tumors and lymphoid malignancies.

Experimental design: Objectives were to establish the safety and maximum tolerated dose (MTD) of the combination; characterize veliparib pharmacokinetics (PK); measure poly(ADP-ribose) (PAR), a product of PARP, in tumor biopsies and peripheral blood mononuclear cells (PBMC); and measure the DNA-damage marker γH2AX in PBMCs and circulating tumor cells (CTC). Cyclophosphamide was administered once daily in 21-day cycles in combination with veliparib administered once daily for 7, 14, or 21 days.

Results: Thirty-five patients were enrolled. The study treatment was well tolerated, and the MTD was established as veliparib 60 mg with cyclophosphamide 50 mg given once daily. Seven patients had partial responses; an additional six patients had disease stabilization for at least six cycles. PAR was significantly decreased in PBMCs (by at least 50%) and tumor biopsies (by at least 80%) across dose levels (DL); γH2AX levels were increased in CTCs from seven of nine patients evaluated after drug administration.

Conclusions: The combination of veliparib with metronomic cyclophosphamide is well tolerated and shows promising activity in a subset of patients with BRCA mutations. A phase II trial of the combination compared with single-agent cyclophosphamide is ongoing in BRCA-positive ovarian cancer, triple-negative breast cancer, and low-grade lymphoma.

Figure 1

Antitumor activity. A, maximum change in target lesion size from baseline assessed according to RECIST 1.0 (n=26). Six patients had non-measurable, evaluable disease only; one patient was taken off study for clinical progression. Dotted line indicates 30% decrease as defined by RECIST 1.0 for partial response. Patient numbers are displayed along the x-axis. *Patients with BRCA mutation. ^CPatients who had prior oral (patients 21 and 25) or intravenous cyclophosphamide. Patient 21 received prior veliparib and patient 3 received prior ifosfamide. B, computed tomography scans from patient 4, a 57-year-old woman with BRCA2-positive ovarian cancer at baseline (left) and after four cycles of treatment on DL 2 (right); patient achieved a partial response. C, magnetic resonance imaging scans from patient 20, a 44-year-old woman with ER-positive, BRCA2-positive breast cancer on DL 7 at baseline (left) and after two cycles (right); patient achieved a partial response (status post doxorubicin/cyclophosphamide, taxane, letrozole, trastuzumab, fulvestrant, gemcitabine, and bevacizumab). Arrows indicate pre- and post-treatment tumor sites.

Figure 2

A, mean PAR levels relative to baseline (100%) by DL in PBMC samples collected at baseline and 2, 4, 6, and 24 hours post-dose on day 1. Error bars represent the standard deviation. B, PAR levels relative to baseline (100%) in tumor biopsy samples collected 2 to 6 hours post-dose on day 7 for patient 1 (DL 1), 4 to 6 hours post-dose on day 21 for patients 16 (DL 6), 22 (DL 7), and 20 (DL 7), and 24 hours after the day 21 dose for patient 34 (DL 8). *Patients with BRCA mutation. Total number of CTCs (C) and percent γH2AX-positive CTCs (D) isolated from nine patients during cycle 1 at baseline and post-treatment on days 2 and 5. Color coding indicates DL; DL 3 (Pt 7-8), DL 5 (Pt 13), DL 6 (Pt 17), DL 7 (Pt 20-22, 24), DL 8 (Pt 34). Dotted lines indicate samples from patients with BRCA mutations.

Figure 3

Pharmacodynamic data from a 57-year-old man with small lymphocytic lymphoma/chronic lymphoid leukemia on DL 1 during cycle 1 (dose escalated to DL 7 in subsequent cycles) who experienced prolonged stable disease and received a total of 42 cycles of therapy. A, PAR levels in serial PBMC samples collected on day 1 of cycle 1. PAR levels were undetectable at 2, 4, and 6 hours post-dose. B, PAR levels in tumor at baseline and on day 7 of cycle 1. C, γH2AX levels measured as percent nuclear area positive (%NAP) in serial PBMC samples collected on day 1 of cycle 1. D, representative γH2AX staining in PBMCs at baseline (left) and 4 hours after combined treatment during cycle 1 (right).