Poster Board III-1015
Ligation of inhibitory receptors renders natural killer (NK) cells inactive against autologous tumors. Recently, bortezomib (bort) was shown to sensitize a variety of hematological malignancies and solid tumors to NK cell cytotoxicity in vitro and in tumor bearing mice by augmenting TRAIL and perforin/granzyme-mediated caspase-8 activity. Based on these findings, we initiated a clinical trial to explore the anti-tumor effects of escalating doses of adoptively infused ex vivo expanded autologous NK cells against malignancies in pts treated with bortezomib to sensitize to NK Cell TRAIL cytotoxicity. Pts age 18-70 with treatment refractory metastatic tumors or hematological malignancies with progressive disease refractory to conventional therapy were eligible for study. Pts underwent a 15-20 liter apheresis that was enriched for NK cells using immuno-magnetic beads to deplete CD3+ T-cells followed by CD56+ selection. Enriched NK cells (5 × 107) were expanded in vitro over 2 weeks using irradiated EBV-LCL feeder cells. Three days prior to NK cell infusion, subjects received a single injection of pentostatin (4mg/m2) to deplete lymphocytes followed by a single injection of bort (1.3 mg/m2) 24 hours prior to the infusion of autologous ex vivo expanded NK cells given on day 0. To maintain NK cell viability and TRAIL surface expression, IL-2 (2 million IU/m2) was given s.c. for 7 days (initially daily, subsequently increased to b.i.d.) after each NK cell infusion. Pts were restaged at 3 weeks and those with a disease responses or stable disease were eligible to receive additional cycles. Cohorts of 3-6 pts are currently being enrolled into escalating NK cell dose levels (5 × 106, 1× 107, 5×107 and 1×108 NK cells/kg). The study is currently accruing pts into NK cell dose level 3, with 9 pts having received a total of 19 NK cell infusions. Following a 2 week culture, NK cells expanded a median 191 fold (range 61-502) with 19/20 cultures expanding successfully to achieve the target NK cell dose. At harvest, expanded NK cell cultures contained a median 99% (range 96-100) CD3-/CD56+ NK cells that had a median 86% (range 71-92) viability by 7AAD/Annexin V staining. To date, no grade III/IV toxicities attributable to NK cell infusions have occurred. The most common adverse events were related to IL-2 and included grades I-II fever, renal insufficiency, edema and hypotension. Two pts had unexplained and transient grade I and II elevations in hepatic enzymes. One pt (dose level I) developed zoster leading to a protocol amendment for subsequent pts to receive valacyclovir prophylaxis. The absolute lymphocyte count nadired on day 0 at a median 0 cells/ul (range 0-150) and peaked on day 10 at a median 1700 cells/ul (range 646-7530). On day 10, PBMCs contained a median 48% of CD3+ T-cells, 13% CD3-/CD56+ NK cells and 10% CD3+/CD4+/CD25+/CD127-/Foxp3 + T-regs. An ELISA analysis of IL-2 levels in sequential serum samples from 2 patients given daily IL-2 (2 million IU/m2) showed levels peaked at 2-4 hours in the range of 300-800 pg/ml but declined rapidly to undetectable levels by 24 hours after each injection. Subsequently the protocol was amended to increase IL-2 dosing to q12 hours. Best clinical response to date includes 4 pts with progressive disease and 5 pts with stable disease. All 5 pts with stable disease have received subsequent cycles of NK cell therapy (range 2-5 cycles). Although no pt met criteria for a PR using RECIST criteria, two pts with metastatic tumors (colon cancer and non small cell lung cancer) who had stable disease had a >30% fall in CEA levels following each NK cell infusion. Conclusions: Using EBV-LCL feeder cells, pure populations of highly activated NK cells can be successfully expanded ex vivo under GMP conditions to evaluate the anti-tumor effects of escalating numbers of adoptively infused NK cells. This regimen utilizing bortezomib to potentiate the antitumor effects of adoptively transferred autologous NK cells continues to enroll patients and to date has been well tolerated and has provided early clinical evidence for anti-tumor activity.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.