High Dose Celecoxib and Metronomic ‘Low-Dose’ Cyclophosphamide Is Effective Therapy in Patients with Relapsed and Refractory Aggressive Histology NHL.

Background: Relapsed aggressive histology non-Hodgkin’s lymphoma (NHL) has a poor prognosis, and new treatments are needed. Angiogenesis is increased in aggressive NHL and may be a target in these diseases. Low-dose chronic chemotherapy (metronomic chemotherapy, MC) inhibits angiogenesis in vitro and in vivo. Since COX-2 may promote neoplasia and tumor angiogenesis, selective COX-2 inhibitors may have antitumor effects in NHL. We assessed response and toxicity to MC and COX-2 inhibition in patients (pts.) with relapsed aggressive NHL following anthracycline-based chemotherapy.

Methods: Pts. with measurable disease and normal renal function received cyclophosphamide 50 mg po qd + celecoxib 400 mg po bid. Pts. were assessed clinically and radiologically in serial fashion. We measured serial plasma VEGF and TSP-1 levels and pharmacokinetics (PK) were performed in selected pts. Serial circulating endothelial cells (CEC) and their precursors (CEP) were measured by flow cytometry as of October 2003.

Results: To date, 29 of a planned 32 pts. have been enrolled; 3 are ineligible and 25 are evaluable for response. Median age: 62 y (range 22–83). Histology: 17 DLBCL, 1 MCL, 3 transformed (CLL+FL), 2 T cell, 3 anaplastic large cell. Median 2.7 yrs from diagnosis (range 0.5–9.5); Median # of previous chemotherapy regimens: 3 (range 1–6); 31% were chemoresistant to preceding regimen; 8 had undergone prior ASCT. Median time to death or last follow-up: 4 mos (range 0.9–22).

Response:10/25 pts. responded (2 CR/CRu, 8 PR, ORR 40%) at a median time of 4.6 mos, and at last f/up, 6/10 remained in PR at 22, 21, 16, 11, 8 and 7 mos. respectively. Three pts. achieved stable disease (SD) but 2 later progressed at a median time of 6 mos (range 4–14 mos) and 12/25 progressed primarily (PD) at a median time of 1.9 mos (range <1–9); 10 have died of NHL (median time 4 mos). 2 partial responders discontinued treatment in PR at 21 and 16 mos for cumulative cyclophosphamide exposure exceeding 40,000 mg. Both progressed within 2–3 months of drug disontinuation. Median OS and PFS are 8.5 and 3.9 mos, respectively.

Adverse events (# pts): grade 3–4 ANC (2), grade 3 plt (3), grade 3 hypertension (1), grade 3 atrial arrhythmia(1) and grade 3 drug rash (1). Plasma VEGF and TSP-1 levels do not appear to correlate with response but the number of circulating endothelial cells (CEC) and their precursors (CEP) decline in responders. Celecoxib and cyclophosphamide pharmacokinetics will be presented. This regimen has been well tolerated with minimal GI toxicity. 90% of responders had normal LDH compared with 25% with PD. No patient with preceding chemoresistance responded.

Conclusions: MC with high-dose celecoxib and low-dose cyclophosphamide is well tolerated and active in 40% of heavily pre-treated patients with relapsed aggressive NHL. The anti-angiogenic activity of this regimen is still being determined.