Advances in clinical NK cell studies: Donor selection, manufacturing and quality control.

Natural killer (NK) cells are increasingly used in clinical studies in order to treat patients with various malignancies. The following review summarizes platform lectures and 2013-2015 consortium meetings on manufacturing and clinical use of NK cells in Europe and United States. A broad overview of recent pre-clinical and clinical results in NK cell therapies is provided based on unstimulated, cytokine-activated, as well as genetically engineered NK cells using chimeric antigen receptors (CAR). Differences in donor selection, manufacturing and quality control of NK cells for cancer immunotherapies are described and basic recommendations are outlined for harmonization in future NK cell studies.

Infusion of the allogeneic cell line NK-92 in patients with advanced renal cell cancer or melanoma: a phase I trial.

BACKGROUND: Renal cell cancer and malignant melanoma are two types of cancer that are responsive to immunotherapy. In this phase I dose-escalation study, the feasibility of large-scale expansion and safety of administering ex vivo-expanded NK-92 cells as allogeneic cellular immunotherapy in patients with refractory renal cell cancer and melanoma were determined. METHODS: Twelve patients (aged 31-74 years) were enrolled, three per cohort at cell dose levels of 1x10(8)/m(2), 3x10(8)/m(2), 1x10(9)/m(2) and 3x10(9)/m(2). One treatment course consisted of three infusions. Eleven patients had refractory metastatic renal cell cancer; one patient had refractory metastatic melanoma. RESULTS: The NK-92 cells were expanded in X-Vivo 10 serum-free media supplemented with 500 U/mL Proleukin recombinant human interleukin-2 (rhIL-2), amino acids and 2.5% human AB plasma. Final yields of approximately 1x10(9) cells/culture bag (218-250xexpansion) over 15-17 days were achievable with >or=80% viability. Infusional toxicities of NK-92 were generally mild, with only one grade 3 fever and one grade 4 hypoglycemic episode. All toxicities were transient, resolved and did not require discontinuation of treatment. One patient was alive with disease at 4 years post-NK-92 infusion. The one metastatic melanoma patient had a minor response during the study period. One other patient exhibited a mixed response. DISCUSSION: This study establishes the feasibility of large-scale expansion and safety of administering NK-92 cells as allogeneic cellular immunotherapy in advanced cancer patients and serves as a platform for future study of this novel natural killer (NK)-cell based therapy.

Targeted cellular therapy with natural killer cells.

Natural killer (NK) cells are believed to be important contributors to a patient's immune armamentarium to fight cancer. However, cancer patients have reportedly defective NK cells and the malignant target frequently has developed mechanisms to escape detection of NK cells. Our research is aimed at overcoming this NK cell paralysis through three different approaches: instead of using autologous NK cells for adoptive immunotherapy, allogeneic NK cells are used that are not inhibited by self histocompatibility antigens. Further, NK cells, selected for a high affinity Fc receptor can, together with monoclonal antibodies, kill targets through ADCC irrespective of any inhibitory receptors. Finally, the genetic engineering of NK cells to express chimeric antigen receptors recognizing antigens on tumor target can overcome inhibitory mechanism and effectively lyse tumor cells.

Natural killer cell-based immunotherapeutic strategies.

Not until recently have Natural killer (NK) cells stepped out of the shadow of T-cells to be considered for cellular therapy of malignant diseases. This evolution has been facilitated by the discovery of specific receptors on NK cells that interact with HLA molecules on target cells but also the discovery of specific activating receptors. Since NK cells represent only about 10% of the lymphocyte population in blood, separation and enrichment are important steps if NK cells are to be used clinically. This is of particular consideration in the setting of allogeneic NK cell infusion where contaminating T-cells could potentially induce graft versus host disease. This review will describe the requirements for NK cells to recognize target cells, their ex vivo expansion and potential therapeutic applications.

Activated platelets rapidly up-regulate CD40L expression and can effectively mature and activate autologous ex vivo differentiated DC.

Background DC are a promising immunotherapeutic for treatment of infectious/malignant disease. For clinical trials, immature DC generated from precursor cells such as monocytes, using serum-free media containing GM-CSF and IL-4, can be matured with specific cytokines/factors. CD40 ligand (CD40L) plays an important role in DC activation/maturation but is not available for clinical applications. These studies evaluated the feasibility of using activated platelets with elevated CD40L surface expression to stimulate autologous DC maturation. Methods Pilot and clinical scale studies were executed using magnetic/centrifugal separation. Monocyte precursors were differentiated to immature DC with GM-CSF and IL-4 and the ability of activated autologous platelets to mature these cells was evaluated on the basis of phenotype and function. Results In small-scale studies, DC cultures stimulated with activated autologous platelets (CD40L-AP), tumor necrosis factor-alpha (TNF-alpha) or soluble CD40L (sCD40L) up-regulated expression of phenotype markers indicative of activation and maturation. CD86 expression was significantly enhanced (P<0.05) by stimulation with either CD40L-AP (75.5+/-14.5%) or sCD40L (80.5%+/-5.3%) compared with immature DC (55.2+/-14.8%), as were CD80 and CD83. Similarly, in large-scale studies using Isolex 300I to enrich for monocytes and platelets for DC generation/maturation on a clinical scale, stimulation with CD40L-AP increased CD86 and CD80 expression as well as the ability to stimulate allogeneic lymphocytes compared with control cultures. Discussion These results demonstrate that thrombin-activated platelets express CD40L and are effective at inducing matured DC with potent immunogenic activity. Furthermore, these studies demonstrate the feasibility of this approach for clinical immunotherapeutic interventions.

Ex vivo expansion of natural killer cells for clinical applications.

BACKGROUND: Immunotherapy with NK cells has been limited by the inability to obtain sufficient numbers of pure NK cells suitable for manipulation and expansion. The goal of this study was to isolate CD56(+) cells (CD3(-)/CD56(+), CD3(+)/CD56(+)) and expand them under culture conditions compliant with current good manufacturing practices. METHODS: Magnetic cell-selection technology, using paramagnetic CD56 microbeads and cell selection columns, was used to isolate a CD56(+) population containing both CD3(-)/56(+) NK (60.6+/-10.8%) and CD3(+)/56(+) NK T cells (30.4+/-8.6%) to initiate the expansion studies. The isolated CD56(+) cells were cultured in X-Vivo10 serum-free media supplemented with 10% human AB serum and 500 U/mL recombinant human IL-2 or 500 U/mL IL-2 plus 10 ng/mL recombinant human IL-15 for 14 days. Cultures were fed fresh media and cytokines every 3-4 days, and were evaluated for cell expansion, phenotype, and cytotoxicity at the end of the culture period. RESULTS: Significant expansion of CD56 cells occurred only during the second week of culture. Although an average of two log expansions was observed, there was substantial cell-expansion variability, depending on the donor, and even when the same donor was tested on different occasions. The cytotoxicity of selected and expanded CD56(+) cells at a low E:T ratio was significantly higher than the starting population, but was comparable to non-separated PBMC expanded for 2 weeks under the same conditions. IL-15 (in combination with IL-2) induced higher killing at the 1:1 E:T ratio than IL-2 alone. Since CD3 cells were not depleted upfront, the expansion of CD3(+)CD56(+) cells was 2-3 times that of CD3(-)CD56(+) cells. NK cells that express the FcgammaRIII (CD16) can mediate Ab-dependent cellular cytotoxicity, and can contribute to enhanced efficacy of MAb treatment. Under the given culture conditions, only moderate expansion of CD56(+)/CD3(-)/CD16(+) cells occurred, with the majority of cells being CD56(+)/CD3(+)/CD16(+) cells. DISCUSSION: Our studies suggest that the positive magnetic cell-separation method provides a good basis for obtaining enriched CD56(+) cells but expansion conditions need to be optimized.

Lipid formulations of amphotericin B preserve and stabilize renal function in HSCT recipients.

The current study assessed renal function based on medical records in adult hematopoietic stem cell transplant (HSCT) recipients with proven or probable invasive fungal infection (IFI) transplanted between 1995 and 2000. We confirm that amphotericin B deoxycholate (AmB-d) is nephrotoxic in a large percentage of HSCT recipients. Due to nephrotoxicity, defined as serum creatinine (SCr) >2.5 mg/dl or a 100% increase in SCr from baseline, 88% of patients treated with AmB-d were switched to a lipid formulation of amphotericin B (LFAB). In total, 53% of patients initiated on AmB-d were switched within the first week of therapy. Significantly more patients (70.6%) treated with AmB-d experienced a 100% increase in SCr from baseline compared to patients treated with either AmBisome (44.4%) or Abelcet (41.2%). A Cox Proportional Hazards Model revealed that, compared to patients initiated on AmBisome or Abelcet, the risk of nephrotoxicity (RR=1.5 vs AmBisome; RR=1.7 vs Abelcet), dialysis (RR=2.4 vs AmBisome; RR=1.4 vs Abelcet), and death (RR=2.0 vs AmBisome; RR=1.1 vs Abelcet) were all increased for patients initiated on AmB-d. Study results suggest that renal function improves and mortality declines when an LFAB is given to HSCT patients as initial therapy rather than as second-line therapy, the current practice.

Ex vivo expansion of the highly cytotoxic human natural killer-92 cell-line under current good manufacturing practice conditions for clinical adoptive cellular immunotherapy.

BACKGROUND: Adoptive transfer of ex vivo expanded cytotoxic immune cells has become a viable strategy for treatment of malignant disease. Natural killer (NK)-92, a highly cytotoxic, IL2-dependent human NK cell-line, is an excellent candidate as an immunotherapeutic agent, being active for prolonged periods following irradiation and IL2 deprivation, non-toxic and non-immunogenic, and easily expanded. A number of clinical trials using NK-92 for different indications are currently underway. The aim of this study was to develop current good manufacturing practice (cGMP)-compliant expansion methodology for NK-92. METHODS: The ability to expand NK-92 ex vivo was evaluated. Serum-free culture media, as well as media supplements (IL2, serum/plasma/albumin), culture containers and feeding regimens were compared for their ability to support expansion, viability and cytotoxicity of NK-92 cells. RESULTS: NK-92 cells can be expanded in X-Vivo 10 serum-free media with 500 U/mL of rhIL2 (Proleukin), and 2.5% human serum/plasma to achieve concentrations sufficient to treat patients with >5210(10) cells. The protocol involves cultures initiated at 2.5210(5) cells/mL in 25 mL in 1 L Vuelife culture bags, with addition of fresh media plus IL2 every 3 days to maintain an optimal density of NK-92 cells for expansion. Daily disruption of cell aggregates enhances NK-92 cells expansion and viability during the culture period. Final yields of approximately 1.1-1.3210(6) cells/mL in a 1.2 L volume (1.36-1.56210(9) cells; 218-250 fold expansion) over 15-17 days is achievable under cGMP-compliant conditions with >85% viability. The feasibility of this approach has been shown in ongoing clinical trial with NK-92. DISCUSSION: We describe a protocol that allows for >200-fold expansion of NK-92 cells within a 2-2.5 week period under GMP standards, in quality and quantity suitable for clinical adoptive immunotherapy.

Ex vivo purging with NK-92 prior to autografting for chronic myelogenous leukemia.

Although controversial, purging of the autograft may be necessary to optimize transplant outcome, especially if better treatments become available to eliminate or control residual disease that may be left after the conditioning regimen. The intent of this study was to show that immunological purging with the cytotoxic cell line NK-92 effectively reduces the number of clonogenic cells and that the method can be performed in compliance with GMP. Owing to the easy quantification of bcr-abl transcripts, chronic myelogenous leukemia (CML) was used as a model disease for proof of principle. A detection level of 10(-7) bcr-abl+ cells and purging efficiency of four logs were achievable for the bcr-abl+ cell line, K562. Leukapheresis products collected from CML patients after stem cell mobilization were then tested. For all patients tested, residual CML cells were highly sensitive to purging by NK-92 with a purging efficacy of several logs. No adverse effect on hematopoietic progenitor cell function was noted. These results demonstrate the efficacy of NK-92 as a purging agent to decrease or eliminate malignant contamination of autologous stem cell grafts and establish proof of principle for ex vivo purging of CML autografts using cytotoxic effector cells.

Differential killing of pre-B acute lymphoblastic leukaemia cells by activated NK cells and the NK-92 ci cell line.

The use of NK cells in adoptive therapy for malignant disease is an area of great potential. Currently the only NK cell line in clinical trials is NK-92, an activated NK cell line with a broad range of cytotoxicity against malignant cells. The activity of NK-92 against pre-B acute lymphoblastic leukaemias, however, is highly variable. In this study we compare the cytotoxic mechanisms and signalling pathways utilized by NK-92 ci and IL-2 activated NK cells to mediate killing of pre-B acute lymphoblastic leukaemia cell lines. Deficiencies in TNF family mediated apoptosis, phosphoinositide-3 kinase dependent and phosphoinositide-3 kinase independent killing limit the efficiency of NK-92 ci against pre-B acute lymphoblastic leukaemia cells. Importantly, treatment of the poorly killed leukaemia cells with TNF-alpha augmented both phosphoinositide-dependent and -independent cytolysis.

Alcohol and drug abusers' perceived reasons for self-change in Canada and Switzerland: computer-assisted content analysis.

Although many people recover from substance-use associated problems on their own, little is known about this phenomenon. The paper had two objectives: to use a new research method, computer-assisted content analysis, to understand alcohol and drug abusers' perceived reasons for self-change and to undertake a comparative evaluation across substances and cultures to validate previous findings about subjective appraisal processes. Three studies of natural recoveries of alcohol and drug abusers in two countries conducted tape-recorded interviews with 216 respondents. The taped responses were coded based on a content analytic dictionary approach using a computerized content analysis program. All three studies found several processes mediating the decision to change substance use. The computer content analysis confirmed a cognitive appraisal process regardless of the cultural setting or substance. The findings suggest that several procedures might have benefit in clinical interventions.

Factors regulating the cytotoxic activity of the human natural killer cell line, NK-92.

NK-92, a highly cytotoxic, interleukin-2 (IL-2)-dependent human natural killer (NK) cell line, has been of interest for basic and translational research. We report on a comprehensive analysis of NK-92 for factors implicated in NK cytotoxicity to elucidate factors underlying NK-92's high cytolytic activity and target range. Thus, we hope to develop a method to identify patients best suited to NK-92 immunotherapy. In addition, as a model system, we hope to increase understanding of the basis for the elevated activity exhibited by activated NK (ANK) cells. NK-92 exhibits an unusual receptor expression profile, expressing a relatively large number of activating (NKp30, NKp46, 2B4, NKGD, E, CD28) receptors. Conversely, it expresses few inhibitory receptors (NKGA/B, low levels of KIR2DL4, ILT-2), lacking most of the killer inhibitory receptors (KIRs) clonally expressed on normal NK cells. In addition, NK-92 expresses high levels of molecules involved in the perforin-granzyme cytolytic pathway as well as additional cytotoxic effector molecules including tumor necrosis factor (TNF)-superfamily members FasL, TRAIL, TWEAK, TNF-alpha, indicating the ability to kill via alternative mechanisms. NK-92 also expresses other molecules implicated immune effector cell regulation (CD80, CD86, CD40L, TRANCE) whose relevance in NK killing is unclear. This study provides initial data to develop a method to identify NK-92 susceptible cells (cells expressing ligands for NK-92 activating receptors ie CD48 for 2B4 and CD80/86 for CD28). Furthermore, this work suggests mechanisms that may contribute to ANK cell activity, including modulation of receptor expression to favor activation, up-regulation of cytotoxic effector molecules, and acquisition of new cytolytic pathways.