Abstract P6-01-01: Immunohistochemical staining and in vitro analysis of HER2-positive breast cancer using trastuzumab and pertuzumab to develop an appropriate tracer in image-guided surgery

Background: Pre- and intraoperative visualization of cancer cells using monoclonal antibody-based multimodal tracers can demarcate tumor margins in breast-conserving surgery. We focused on trastuzumab (Tmab) to develop a tracer for HER2-positive breast cancer by conjugation with a fluorescent dye. However, whether Tmab can be used as an imaging tracer for patients who receive Tmab as neoadjuvant therapy (NAD) is unclear, because tumor-cell HER2 could be bound by NAD Tmab at the time of surgery. This study evaluated immunohistochemical (IHC) staining and in vitro experiments with Tmab and pertuzumab (Pmab) as primary antibodies to find a suitable tracer. Pmab has a different antigen epitope than Tmab.

Methods: We included 43 patients with HER2-positive breast cancer, who were treated between 2010 and 2016. The NAD cohort (n=10, with 26 lesions) received chemotherapy and Tmab before surgery. The Tmab-naïve cohort (n=33, with 95 lesions) did not receive chemotherapy or Tmab before surgery. We excluded NAD patients with pathological complete responses. We evaluated the lesions, using IHC with Tmab and Pmab. We also performed flow cytometry and fluorescent microscopic analysis (FMA) using Tmab and Pmab conjugated with a fluorescent agent for the HER2-negative MCF7 and HER2-positive HCC1954 cell lines, which were pre-treated with Tmab to model NAD conditions.

Results: IHC with anti-HER2 antibody showed positive staining in all patients. Tmab staining was less intense, but positive Tmab IHC reactions were detected on tumor cell membranes in 77.8% of lesions in the naïve cohort and 70.8% in the NAD cohort. Pmab staining was seen in 46.3% of naïve cohort lesions and 22.2% of NAD lesions. Notably, we observed cytoplasmic staining in 87.8% of Pmab-negative cases in the NAD cohort. Flow cytometry showed less Tmab binding than Pmab binding in HCC1954 cells. After 24 hours' pretreatment incubation with Tmab, FMA showed a clear decrease in Tmab binding from 73.3% (without pretreatment) to 5.7%(after pretreatment), and a smaller decrease in Pmab binding from 72.7% (without pretreatment) to 66.4% (after pretreatment) in HCC1954 cells.

Conclusions:Pmab might be a suitable tracer for image-guided surgery after NAD, but identifying a suitable tracer for HER2-positive breast cancer will require further study.

Citation Format: Otsubo R, Kuppen PJ, Bhairosingh S, Vahrmeijer AL, Smit VT, Nagayasu T, van de Velde CJ, Sier CF. Immunohistochemical staining and in vitro analysis of HER2-positive breast cancer using trastuzumab and pertuzumab to develop an appropriate tracer in image-guided surgery [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-01-01.

Alternatively spliced tissue factor synergizes with the estrogen receptor pathway in promoting breast cancer progression

BACKGROUND

Procoagulant full-length tissue factor (flTF) and its minimally coagulant alternatively spliced isoform (asTF), promote breast cancer (BrCa) progression via different mechanisms. We previously showed that flTF and asTF are expressed by BrCa cells, resulting in autoregulation in a cancer milieu. BrCa cells often express hormone receptors such as the estrogen receptor (ER), leading to the formation of hormone-regulated cell populations.

OBJECTIVE

To investigate whether TF isoform-specific and ER-dependent pathways interact in BrCa.

METHODS

Tissue factor isoform-regulated gene sets were assessed using ingenuity pathway analysis. Tissues from a cohort of BrCa patients were divided into ER-positive and ER-negative groups. Associations between TF isoform levels and tumor characteristics were analyzed in these groups. BrCa cells expressing TF isoforms were assessed for proliferation, migration and in vivo growth in the presence or absence of estradiol.

RESULTS

Ingenuity pathway analysis pointed to similarities between ER- and TF-induced gene expression profiles. In BrCa tissue specimens, asTF expression was associated with grade and stage in ER-positive but not in ER-negative tumors. flTF was only associated with grade in ER-positive tumors. In MCF-7 cells, asTF accelerated proliferation in the presence of estradiol in a β1 integrin-dependent manner. No synergy between asTF and the ER pathway was observed in a migration assay. Estradiol accelerated the growth of asTF-expressing tumors but not control tumors in vivo in an orthotopic setting.

CONCLUSION

Tissue factor isoform and estrogen signaling share downstream targets in BrCa; the concomitant presence of asTF and estrogen signaling is required to promote BrCa cell proliferation.

Specific activity of cyclin dependent kinase 1 as a novel predictor of recurrence risk in stage II colon cancer

402

Background: Altered cell cycle dynamics and check points are typical features of solid tumors, and cyclin dependent kinases (CDKs) play pivotal roles in these processes. Previously we have demonstrated that CDK-based analysis, composed of CDK1 and CDK2, is useful in the prediction of outcomes in early breast cancer patients (Ann Oncol. 19(1):68-72, 2008, Br J Cancer. 100(3):494-500, 2009). Clinically, there is a need for risk stratification in patients with stage II colon cancer who have a recurrence risk of 20 to 30%. Therefore we investigated the use of CDK-based analysis for recurrence prediction of stage II colon cancer patients.

Methods: Fresh frozen tissue samples of 254 patients with histologically confirmed adenocarcinoma of the colon, UICC stage II, who received primary tumor resection in Munich (217 cases), and Leiden (37 cases) were used. Protein expression and activity of CDK1 and CDK2 were determined by in vitro assays as previously described. Specific activity (SA) of CDKs was calculated as kinase activity in relation to its corresponding mass concentration.

Results: Development of distant metastasis was observed in 27 patients (10.6%) after a median follow up of 86 months. We found that predictive performance of CDK1SA, but not CDK2SA, for the metastasis was substantial and almost constant for long-term event prediction (average area under the curve (AUC) = 0.69). Tumor recurrence risk analysis in association with CDK1SA identified a low- (41% of population) and high- risk group (59%). Cox proportional hazard model analysis retained the CDK-based patient classification as an independent prognostic factor for distant metastases-free survival (low vs. high-risk group: Hazard ratio = 6.2, 95% CI: 1.45 to 26.9, p=0.0049). Clinical parameters such as grading, T-categories, age, and sex were excluded as confounding factors for CDK1SA-risk.

Conclusions: CDK1SA allows stratification of different risk subgroups of stage II colon cancer patients. CDK1SA-based analysis is useful for predicting patients with high risk of distant recurrence, who should be treated with chemotherapy.

No significant financial relationships to disclose.

Development and validation of a robust prognostic and predictive signature for colorectal cancer (CRC) patients

4036

Background: Between 25 and 35% of stage II CRC patients will experience a recurrence of their disease and may benefit from adjuvant chemotherapy. Official guidelines give suggestions but no clear recommendation for best risk stratification. Here we describe the development a robust signature that predicts disease relapse and can assist in treatment decisions. Methods: Fresh frozen tumor tissues from 180 patients with stage I, II and III colorectal cancer undergoing surgery were analyzed using high density Agilent 44K oligonucleotide arrays. Median FU was 70.2 months; 85% of patients did not receive adjuvant chemotherapy. Unsupervised hierarchical clustering based on full-genome gene expression measurement indicated the existence of 3 main colon molecular subclasses. Survival analysis of the 3 classes showed that subtype C (n= 27) had a poor outcome and subtype A (n= 48) good outcome. Only the intermediate group B (n=104) was used to develop a signature by using a cross validation procedure to score all genes for their association with 5-yr distant metastasis free survival (DMFS) and subsequently applied to all samples (n=180). The obtained gene signature was further validated on an independent cohort of 178 stage II + III colon samples. Results: A set of 38 prognosis related gene probes showed robust DMFS association in over 50% of all iterations in the Training Set of 180 samples. The gene signature was validated on an independent cohort of 178 samples from stage II + III colon cancer patients. The profile classified 61% of the validation samples as low-risk and 39% as high-risk. The low- and high-risk samples showed a significant difference in DMFS with a HR of 3.19 (P= 8.5e-4). Five-year DMFS rates were 89% (95%CI 83–95) for low-risk and 62% (95%CI 50–77) for high-risk samples. Moreover, the profile showed a significant performance within stage II (P=0.0058) and III (P=0.036) only samples. The performance of the profile was significant for both untreated (P=0.0082) and treated patients (P=0.016) suggesting that its power is independent of treatment benefits. Conclusions: ColoPrint is able to predict the prognosis of stage II and III colon cancer patients and facilitates the identification of patients who would benefit from adjuvant chemotherapy.

[Table: see text]

Validation study of the prognostic value of cyclin-dependent kinase (CDK)-based risk in Caucasian breast cancer patients

In a Japanese study, cyclin-dependent kinase (CDK) based risk determined by CDK 1 and 2 activities was associated with risk of distance recurrence in early breast cancer patients. The aim of our study was to validate this risk categorization in European early breast cancer patients. We retrospectively analyzed frozen breast cancer specimens of 352 Dutch patients with histologically confirmed primary invasive early breast cancer. CDK-based risk was determined in tumour tissues by calculating a risk score (RS) according to kinases activity and protein mass concentration assay without the knowledge of outcome. Determination of CDK-based risk was feasible in 184 out of 352 (52%) tumours. Median follow-up of these patients was 15 years. In patients not receiving systemic treatment, the proportions of risk categories were 44% low, 16% intermediate, and 40% high CDK-based risk. These groups remained significant after univariate and multivariate Cox-regression analysis. Factors associated with a shorter distant recurrence-free period were positive lymph nodes, mastectomy with radiotherapy, and high CDK-based risk. There was no significant correlation with overall survival (OS). CDK-based risk is a prognostic marker of distance recurrence of patients with early breast cancer. More validation would be warranted to use of CDK-based risk into clinical practice.

Role of NK cells in adoptive immunotherapy of metastatic colorectal cancer in a syngeneic rat model

This article reviews our immunotherapy research with natural killer (NK) cells in a syngeneic rat colorectal cancer liver and lung metastasis model. Using adoptive transfer of interleukin (IL)-2-activated NK cells, NK cells were shown to selectively infiltrate the tumors. More NK cells were found in tumors when the NK cells were directly injected into tumor-draining blood vessels than when the cells were injected in systemic blood vessels. Under optimal conditions, a limited, though significant, effect of adoptively transferred NK cells on tumor growth was shown. We observed that both endogenous and adoptively transferred NK cells were predominantly present in the stroma surrounding the tumor cell nodules. It is possible that they did not penetrate the nodules containing the tumor cells because of the presence of a basal membrane-like structure around these nodules. Adoptively transferred NK cells may initiate elimination of tumor cells by activating other effector cells, whereas some may eliminate tumor cells by direct cell-cell contact. A diverse array of molecules was shown to be involved in this process. CD45 on NK cells was found to be important in initiating the lysis-inhibitory signal upon binding of 'self' major histocompatibility complex (MHC) class I on potential target cells. Our results indicate that NK-cell cancer therapy is still promising and needs improvement.

The development of novel mouse monoclonal antibodies against the CC531 rat colon adenocarcinoma

In this paper we describe 4 new monoclonal antibodies to be applied in rat models for cancer. The monoclonal antibodies were obtained by immunizing Balb/c mice with CC531 rat colon adenocarcinoma cells. Hybridomas were produced and 4 were selected for their reactivity with CC531 in vitro (MG1, 2, 3 and 4). All 4 antibodies recognized other rat tumour cell lines and showed limited cross-reactivity with normal rat tissues. Intraperitoneally injected MG1, 2 and 4 homed to in vivo growing, artificially induced CC531 liver metastases. In these in vivo experiments, limited cross-reactivity with normal rat tissues, predominantly of the gastro-intestinal tract, was found. MG4 was found to enhance lysis of CC531 tumour cells mediated by IL-2 activated, cultured natural killer cells. These antibodies are potentially useful for antibody-based laboratory techniques and for investigation of antibody-based immunotherapy of cancer in a rat model.

Remodeling of the actin cytoskeleton of target hepatocytes and NK cells during induction of apoptosis

Natural Killer cells are immune cells that recognize and eliminate altered and non-self cells from the circulation. To study the interaction between NK cells and target cells, we set up an experimental system consisting of rat Interleukin-2 activated Natural Killer cells (A-NK cells) and rat hepatocytes with a masked Major Histocompatibility Complex (MHC). The masking of the MHC induces recognition of the hepatocytes by the NK cells as non-self. We showed that in vitro apoptosis is rapidly induced in the hepatocytes [Blom et al., 1999] after co-incubation with A-NK cells. Now we describe the morphological changes that occur during and after interaction of A-NK cells with hepatocytes. Confocal laser scanning microscopy showed that the actin cytoskeleton of the NK cells was remodeled during attack of hepatocytes. Some NK cells were in close contact with the hepatocytes while others had formed actin-containing dendrites of varying length that made contact with the hepatocytes. However, dendrite formation is not obligatory for induction of apoptosis because cells that were unable to form these did induce FAS-dependent apoptosis in hepatocytes. Apparently both direct as well as distant contact resulted in apoptosis. Formation of the dendrites was calcium-dependent as EGTA largely prevented it. Importantly, chelation of the calcium also suppressed killing of the hepatocytes. Within 1 h after addition of the A-NK cells, morphological changes in hepatocytes that are characteristic of apoptosis, such as the formation of apoptotic bodies and fragmented nuclei, became apparent. Specifically, the actin cytoskeleton of the hepatocytes was remodeled resulting in the formation of the apoptotic bodies. Inhibition of caspase activity by z-Val-Ala-DL-Asp-fluoromethylketone (100 microM) partly protected against the rearrangement of the actin filaments in the hepatocytes.

Rat hepatic natural killer cells (pit cells) express mRNA and protein similar to in vitro interleukin-2 activated spleen natural killer cells

Pit cells are liver-specific natural killer (NK) cells that can be divided into high- (HD) and low-density (LD) subpopulations. The characteristics of pit cells were further investigated in this report. LD and HD pit cells express the specific NK-activation markers gp42, CD25, and ANK44 antigen. LD cells and IL-2-activated NK cells have a high mRNA expression of perforin, granzymes, interferon-gamma, and tumor necrosis factor-alpha. LD pit cells, unlike spleen NK cells, have a weak response to IL-2 with regard to proliferation, cytotoxicity, and production of NK-related molecules. The characteristics of HD cells are intermediate between LD and spleen NK cells. These results show that pit cells, especially LD cells, possess characteristics similar to IL-2-activated NK cells. This is the first evidence on a molecular level that pit cells could be considered in vivo activated NK cells.

Caspase-3 activity as a prognostic factor in colorectal carcinoma

Several techniques to determine apoptotic frequencies in tumors have been described. In this study, we report that biochemical detection of enzymatic caspase-3 activity is a simple and quantitative technique to measure apoptosis in colorectal tumor cells. The relevance of the level of apoptosis in colorectal cancer for the clinical course remains unclear. Therefore, we studied the correlation between caspase-3 activity and prognosis of the disease in relation to different factors known to be involved in apoptosis induction. High caspase-3 activity significantly correlated with a higher risk of recurrence and was preferentially found in tumors of the right side of the colon. No correlation was detected between high caspase-3 activity and altered protein expression of p53, beta-catenin, or proteins of mismatched repair genes. This indicates that high caspase-3 activity has no evident correlation with the genetic Wnt-signaling or the mismatch repair mutational pathways. The caspase-3 activity significantly correlated with CD57(+) tumor infiltrating cells. Therefore, high caspase-3 activity in right-sided tumors might be induced by a specific lymphocytic reaction.

The microscopic anatomy of experimental rat CC531 colon tumour metastases: consequences for immunotherapy?

The colon adenocarcinoma cell line CC531 was adopted as a model for immunotherapeutical treatment of experimental colorectal metastases in a syngeneic rat model. We studied the presence and localization of T and natural killer cells, vessels and matrix proteins in in vivo growing CC531 tumours by immunohistochemistry. CC531 tumours were induced either in the lungs by injecting CC531 tumour cells into a tail vein or in the liver by injection of CC531 tumour cells under the liver capsule or into a mesenteric vein. All 3 tumour types were composed of islets of tightly apposed tumour cells surrounded by abundantly present tumour-stroma which contained tumour vessels and matrix proteins. Some of these matrix proteins, especially laminin and collagen IV formed a basal membrane-like structure around the tumour nodules. This structure was most pronounced in mesenteric vein-induced liver tumours and less prominent in subcapsular-induced liver tumours and tail vein-induced lung tumours. Tumour-infiltrating lymphocytes of both T and natural killer cell origin were found in the tumours, but predominantly in the tumour stroma, separated from the islets of tumour cells by the basal membrane-like structure. We hypothesize that the matrix proteins of these tumours play an ambivalent role: they may provide a substratum for migration of effector cells into the tumour stroma but may also provide a barrier preventing direct contact between tumour target cells and immune effector cells.

Administration of BiMAb-retargeted T cells in a rat hepatic metastases colon tumour model results in T-cell tumour infiltration independent of the route of administration

In this study, cultured T cells, pre-incubated with the bispecific monoclonal antibody (BiMAb) R73IgG1 x CC52IgG1 were adoptively transferred, via systemic and regional routes, to rats bearing day 10 hepatic metastases of the CC531 adenocarcinoma of the colon to investigate the role of the route of administration in tumour infiltration by these BiMAb-retargeted effector cells. The BiMAb, directed against the T-cell receptor and the tumour-associated antigen CC52, were used to crosslink CC531 tumour cells and T cells to induce tumour cell lysis. Retargeted T cells were administered via the jugular vein, hepatic artery or the portal vein. The number of BiMAb-retargeted T cells that reached the liver tumours was independent of the route of administration. There was also no difference between the number of T cells that reached the portal tracts, central veins of parenchyma of the liver, after loco-regional or systemic administration. These findings are in contrast to the interleukin (IL)-2 activated NK (A-NK) cells biodistribution studies earlier performed in the same animal model in our laboratory. Compared with A-NK cells, a lower number of BiMAb-retargeted T cells reached the tumours, irrespective of their route of administration while for A-NK cells, there was an advantage of administration via the hepatic artery.

The phenotypic heterogeneity of human natural killer cells: presence of at least 48 different subsets in the peripheral blood

Peripheral blood natural killer (NK) cells are usually defined as a homogeneous cell population. However, NK cells show heterogeneous expression of a diversity of cell surface molecules, which might reflect the diversity of NK-cell functions. Therefore, a more specific phenotypic definition of NK cells is necessary. In this study, we made an inventory of phenotypic subsets that are present within the peripheral blood NK-cell population of healthy donors based on differential expression of nine cell-surface markers. Using three-colour flow cytometric analysis we were able to define at least 48 different CD56(+) NK-cell subsets within the peripheral blood. This phenotypic heterogeneity appeared to be stable among healthy individuals, and was also steady within CD56(dim) and CD56(bright) NK populations, indicating a possible role for these subsets in NK-cell function or differentiation.

Tumor structure and extracellular matrix as a possible barrier for therapeutic approaches using immune cells or adenoviruses in colorectal cancer

In this article we report about the role that tumor structure and extracellular matrix (ECM) may play in immunotherapy and in gene therapy using adenoviruses. We performed studies in a rat model for colorectal cancer, CC531, and in specimens of human colorectal cancer. The tumors were composed of two compartments, tumor cell nests surrounded by stromal cells. ECM proteins were expressed in the stromal part, where the blood vessels were also located. Furthermore, in several tumors, the tumor cell nests were surrounded by basal membrane-like structures. Therefore, in vascular approaches to treat cancer, therapeutic agents on their route to tumor cells may be hampered by ECM to reach tumor cells. We found that immune cells were abundantly present in tumors from colorectal origin. These cells were, however, not found in direct contact with tumor cells, but mainly in the stromal part of the tumor. Adenoviruses, when intravascularly injected, did not reach tumor cells in the CC531 rat model. Tumor cells were only infected, and even then in limited numbers, in cases of intratumoral injection. We hypothesize that ECM in a tumor is a barrier both for immune cells and for adenoviruses to make direct contact with these tumor cells, and thus limits colorectal tumor therapy.

Long lasting p53-specific T cell memory responses in the absence of anti-p53 antibodies in patients with resected primary colorectal cancer

Colorectal carcinoma is commonly associated with mutation and overexpression of p53, making this antigen a potential target for immune intervention. We analyzed humoral and proliferative immunity against p53 in the blood of patients with resected primary colorectal cancer. The majority of these patients displayed anti-p53 T helper (Th) immunity in the absence of measurable p53 specific antibody levels. The Th responses were long-lasting since they could be detected up to several years after resection of the primary tumor. In a number of cases the Th responses were highly sensitive, reflected by the recognition of naturally processed p53 protein. Our data argue that boosting of these responses in patients with minimal residual disease through p53-specific vaccination, may be employed for improving the chance of disease-free survival of these patients.

T cell responses in colorectal cancer patients: evidence for class II HLA-restricted recognition of shared tumor-associated antigens

Few cases of anti-colon cancer specific T lymphocytes have been described so far. Moreover, the majority of these effectors were generated in vitro by stimulating PBMC from patients or healthy donors with peptides that were derived from proteins expressed and/or secreted by colon cancer tissue such as CEA, Mucin or Her-2/neu. The aim of our study was to evaluate the immunogenicity of colorectal carcinomas in an autologous setting. We exploited the antigen processing and presentation capacity of dendritic cells (DC) to establish an in vitro autologous system that can bypass the need of obtaining cultured tumor cells. DC were generated from the adherent monocyte fraction of PBMC taken from stage II/III colorectal cancer patients. A single cell suspension was prepared by mechanical and enzymatic disruption of the surgical specimens immediately after resection. DC were loaded with autologous tumor lysate, obtained by repeated freezing and thawing, before being used as stimulators for autologous PBL. HLA-class II restricted T cells that recognize the autologous tumor could be generated in a proportion of patients. The fine specificity of the anti-tumor T cells indicates that differentiation as well as tumor restricted antigens are expressed in colon cancer and that these antigens can evoke a class II HLA-restricted response in an autologous setting. Altogether these findings may open a new perspective for a DC based vaccination of colon cancer patients.

NK cells modulate MHC class I expression on tumor cells and their susceptibility to lysis

Cytotoxicity and production of cytokines are two important functions of NK cells. These two different NK functions were studied in a syngeneic rat model in relation to MHC class I expression. We focussed on the mechanism by which NK cells modulate MHC class I expression on target cells and how this interferes with NK cell-mediated lysis. Using transfection experiments an inhibitory role on NK cell cytotoxicity for expression of target cells of RT1.A, rat MHC class I, was found. Co-culturing syngeneic tumor cells and NK cells resulted in enhanced MHC class I expression on the surviving tumor cell fraction, which was less susceptible to NK lysis. Increased tumor cell MHC class I was due to production of a soluble factor by NK cells, most likely interferon gamma. The regulatory function of NK cells shows here, that the enhancing of MHC class I expression on tumor cells in vitro and in vivo, results in downregulation of their target cell killing, but at the same time may facilitate the cytotoxic T cell function.

Recombinant adenoviral vectors have adjuvant activity and stimulate T cell responses against tumor cells

The host-immune response against adenoviruses forms a major obstacle for their use as gene therapy vectors for treatment of genetic defects. None the less, they are the preferred vectors for in vivo gene transfer in experimental gene therapy protocols for cancer. In this article we demonstrate the antitumor efficacy of adenovirus-mediated transfer of human interleukin-2 cDNA in the rat-CC531 model for hepatic metastases of colorectal cancer: intratumoral administration of 10 plaque-forming units of the hlL-2-expressing adenoviral vector, AdCAIL-2, resulted in a cessation of tumor growth in 80% of the injected tumors. In control groups receiving AdCnull, a vector with the same viral backbone, but lacking transgene expression, none of the tumors responded. However, intratumoral treatment with this vector significantly enhanced tumor regression induced by systemic IL-2 protein treatment, which was used as a positive control. In addition we show, by performing delayed-type of hypersensitivity assays, that AdCnull when injected intratumorally enhances recognition of tumor antigens by T lymphocytes to the same extent as intratumoral treatment with the IL-2-expressing vector. The replication-deficient adenoviruses appear to have a therapeutic advantage in cytokine-mediated immunotherapy: even adenovirus vectors that do not express a transgene, show adjuvant activity and stimulate an antitumor immune response.

Secreted and membrane-associated matrix metalloproteinases of IL-2-activated NK cells and their inhibitors

We have previously documented that rat IL-2-activated NK (A-NK) cells produce matrix metalloproteinase-2 (MMP-2) and MMP-9. In this study, we describe mouse A-NK cell-derived MMPs, including MT-MMPs, and also TIMPs. RT-PCR analysis from cDNA of mouse A-NK cells revealed mRNA for MMP-2, MMP-9, MMP-11, MMP-13, MT1-MMP, MT2-MMP, TIMP-1, and TIMP-2. MMP-2 and MMP-9 expression was confirmed by gelatin zymography. Moreover, we report for the first time that MT-MMPs are expressed by NK cells, i.e., large granular lymphocytes as determined by both RT-PCR and Western blots. TIMP-1 expression was detected as a 29-kDa protein in Western blots. It is intriguing that TIMP-2 protein from A-NK cells was also detected as a 29-kDa protein, which is clearly different from the previously reported molecular mass of 21 kDa in mouse and human cells. In addition, inhibition of MMPs by BB-94, a selective inhibitor of MMP, significantly inhibited the ability of mouse A-NK cells to migrate through Matrigel, a model basement membrane. Taken together, these findings suggest that A-NK cells may therefore use multiple MMPs in various cellular functions, including degradation of various extracellular matrix molecules as they extravasate from blood vessels and accumulate within cancer metastases following their adoptive transfer.

Tumour immunotherapy: the adjuvant treatment of the 21st century?

In the course of a century, tumour immunology has revealed a picture of a very complex immune system involving the recognition and eradication of malignancies. Many tumours evade the immune system, and understanding of tumour escape mechanisms is the key to a successful immunotherapy for cancer. A wide array of tumour immunotherapy modalities have been developed, many of which have reached the phase of clinical trials, with some satisfactory results. Based on the available clinical, data and the techniques available for further improvement, we analyse the prospects for the different treatment modalities, and predict an important role for tumour immunotherapy in the near future.

Increased local cytostatic drug exposure by isolated hepatic perfusion: a phase I clinical and pharmacologic evaluation of treatment with high dose melphalan in patients with colorectal cancer confined to the liver

A phase I dose-escalation study was performed to determine whether isolated hepatic perfusion (IHP) with melphalan (L-PAM) allows exposure of the liver to much higher drug concentrations than clinically achievable after systemic administration and leads to higher tumour concentrations of L-PAM. Twenty-four patients with colorectal cancer confined to the liver were treated with L-PAM dosages escalating from 0.5 to 4.0 mg kg(-1). During all IHP procedures, leakage of perfusate was monitored. Duration of IHP was aimed at 60 min, but was shortened in eight cases as a result of leakage from the isolated circuit. From these, three patients developed WHO grade 3-4 leukopenia and two patients died due to sepsis. A reversible elevation of liver enzymes and bilirubin was seen in the majority of patients. Only one patient was treated with 4.0 mg kg(-1) L-PAM, who died 8 days after IHP as a result of multiple-organ failure. A statistically significant correlation was found between the dose of L-PAM, peak L-PAM concentrations in perfusate (R = 0.86, P< or =0.001), perfusate area under the concentration-time curve (AUC; R = 0.82, P<0.001), tumour tissue concentrations of L-PAM (R = 0.83, P = 0.011) and patient survival (R = 0.52, P = 0.02). The peak L-PAM concentration and AUC of L-PAM in perfusate at dose level 3.0 mg kg(-1) (n = 5) were respectively 35- and 13-fold higher than in the systemic circulation, and respectively 30- and 5-fold higher than reported for high dose oral L-PAM (80-157 mg m(-2)) and autologous bone marrow transplantation. Median survival after IHP (n = 21) was 19 months and the overall response rate was 29% (17 assessable patients; one complete and four partial remissions). Thus, the maximally tolerated dose of L-PAM delivered via IHP is approximately 3.0 mg kg(-1), leading to high L-PAM concentrations at the target side. Because of the complexity of this treatment modality, IHP has at present no place in routine clinical practice.

A novel drug delivery system using IL-2 activated NK cells and Zyn-linked doxorubicin

Adoptively transferred IL-2 activated NK (A-NK) cells selectively accumulate within tumor metastases which recommends them as vehicles for locoregional drug delivery. Zyn-Linkers are membrane-binding lipophilic dyes which can be coupled by a variety of conjugation chemistries to therapeutic agents. We have previously demonstrated that A-NK cells labeled with PKH26 are able to accumulate within established B16 melanoma pulmonary metastases by 16 h at a concentration of over 600 cells/mm2 of tumor tissue (Basse et al. J. Exp. Med. 174: 479 1991). Zyn-205 is a prodrug in which doxorubicin is attached to a similar Zyn-Linker through an acid-sensitive bond. We have optimized the ex vivo labeling conditions and found that a 10 min incubation with 25 microM Zyn-205 results in the uptake of over 10(8) drug molecules per cell with no effect on either cell viability or cytolytic activity up to 24 h after labeling. Given these parameters, the amount of drug which may be carried to and concentrated in metastatic lesions represents a local concentration of approximately 15 microM. In addition, A-NK cells carrying Zyn-Linked doxorubicin at an equivalent dose of 25 micrograms/kg was therapeutically comparable to a systemic dose of 8 mg/kg (320x more) in the 3LL model of experimental metastasis. These data indicate that A-NK cells bearing Zyn-Linked chemotherapeutic agents represent a unique and feasible method to target chemotherapeutic agents to cancer metastases and that therapeutic doses can be attained without unwanted systemic exposure.

Matrix metalloproteinases of human NK cells

We have previously reported that MMP-2 and MMP-9 are present in rat A-NK cells, and have recently documented that additional MMPs are present in rodent A-NK cells. To our knowledge only proMMP-9 has previously been reported for human NK and A-NK cells. Herein, we report for the first time the presence of MMP-2 and MT1-MMP in human NK cells. The importance of these enzymes for the migration of A-NK cells into tumor metastases is of great potential relevance. MMPs may be rate limiting in A-NK cells, following their adoptive transfer, to traverse basement membrane and accumulate within established cancer metastases, a likely pre-requisite to their cytolytic function. Human NK cells express and produce MMP-2, MMP-9, MT1-MMP and the inhibitor TIMP-1. Moreover, human A-NK cells degrade the extracellular matrix equivalent (Matrigel) in a seemingly IL-2 dependent manner. It is therefore likely that A-NK cell MMPs play crucial roles in contributing to A-NK cell localisation and positioning the cells in vivo to allow for triggering their cytolytic potential.

Expression profile of saccharide epitope CaMBr1 in normal and neoplastic tissue from dogs, cats, and rats: implication for the development of human-derived cancer vaccines

CaMBr1 is a blood group-related tumour-associated antigen, whose pattern of expression provides a therapeutic window for passive or active immunotherapy and points to the promise of a vaccine against carcinomas overexpressing this antigen. In this context, an animal model that closely mimics the human situation would be extremely useful. We, therefore, utilised the murine monoclonal antibody MBr1, which defines CaMBr1, as a useful probe to detect the molecule targeted for vaccine development on canine and feline spontaneous breast and uterus tumours and on their normal counterparts, and on rat normal tissues and carcinoma cell lines. Immunoperoxidase staining of cryostat sections revealed homogeneous CaMBr1 expression only in normal feline uterus and a uterus papilloma, whereas MBr1 reactivity was very weak and heterogeneous in normal (1/3 and 1/3) and tumour (1/10 and 1/6) breast tissues from dogs and cats, respectively. In contrast, the data obtained in rat tissues were reproducible in the strains tested and showed that CaMBr1 was expressed in all epithelial tissues of the digestive tract, although with variable intensities. Monoclonal antibody staining appeared to correspond to membrane-bound structures as well as mucinous secretions. Similarly, secretion products of lactating mammary glands expressed CaMBr1. The spectrum of expression on rat digestive tract was broader than that in humans but the specificity of MBr1 reactivity was confirmed by competition assay with a synthetic tetrasaccharide that mimics the CaMBr1 antigen. On FACS analysis, only one of two clonal derivatives of the rat breast carcinoma line RAMA 25 expressed CaMBr1, and a negative cell subset was evident in repeated experiments. By contrast, both colon carcinoma lines, DHD/K12 and CC531, showed staining with MBr1, albeit at different levels of intensity, and no evidence of a negative subset. The cell line CC531 maintained or even increased CaMBr1 expression levels following transplantation in syngeneic immunocompetent animals. Our data suggest the usefulness of the rat as a test model for vaccines against human cancers overexpressing the CaMBr1 antigen.

Bispecific antibodies in cancer therapy, from the laboratory to the clinic

Bispecific monoclonal antibodies (BiMAb) have been shown to be able to contribute to an immunological approach in cancer therapy. In this review, essential aspects regarding the production of BiMAb and modes to apply them in immunotherapy for cancer are discussed. The pros and cons of BiMAb are considered, and the development from application in animal models to clinical studies is reviewed. The most important clinical trials are summarized, and the different problems encountered are discussed. Provided some crucial problems can be overcome, BiMAb will have a place in the treatment of cancer, especially in the setting of minimal residual disease.

Involvement of LFA-1 in hepatic NK cell (pit cell)-mediated cytolysis and apoptosis of colon carcinoma cells

BACKGROUND/AIMS

Previous studies have shown that hepatic natural killer (NK) cells, also called pit cells, have a higher cytotoxicity against certain tumor cells and have a higher expression of the cell adhesion molecule CD11a as compared with blood NK cells. We further investigated the involvement of the adhesion molecules, reported to be involved in target cell killing by blood NK cells, in pit cell-mediated colon carcinoma cell killing.

METHODS

51Cr-release and DNA fragmentation were used to quantify target cell lysis and apoptosis, respectively. Adhesion of pit cells to CC531s monolayers was quantitated.

RESULTS

Flow cytometric analysis showed that pit cells expressed CD2, CD11a, CD18 and CD54. CC531s cells expressed only CD54. Treatment of freshly isolated pit cells with monoclonal antibodies (mAbs) to CD11a and CD18 inhibited not only the pit cell-mediated CC531s cytolysis but also the pit cell-induced apoptosis of CC531s cells. The combination of mAbs to CD11a, CD18 and CD54 further increased the inhibition of pit cell-mediated CC531s cytolysis and apoptosis. Anti-CD2 mAb did not affect these processes. The binding of pit cells to CC531s cells was also inhibited by anti-CD11a, and CD18 mAbs, but not by anti-CD2 mAb. Anti-CD54 mAb reduced the target cell killing and the binding only slightly.

CONCLUSIONS

These results indicate that CD11a/CD18 (LFA-1) present on pit cells plays an important role in pit cell-mediated target cell adhesion, lysis and apoptosis. This finding might explain why pit cells, which have a higher expression of LFA-1 as compared to blood NK cells, are more cytotoxic against tumor cells as compared to blood NK cells.

The regulatory role of CD45 on rat NK cells in target cell lysis

To investigate the role of CD45 in rat NK cell function, we developed new mAbs directed against rat CD45. mAb ANK12 binds to a high molecular isoform of CD45 and mAb ANK74 binds to the common part on all known CD45 isoforms, as has been described for the anti-rat CD45 mAb OX1. The ability of these mAbs to affect NK cell-mediated lysis was tested using the Fc receptor-positive target cell line P815. mAb ANK12 was found to significantly enhance the lysis of P815, whereas ANK74 and the anti-CD45 mAb OX1 did not. In addition, cross-linking of the CD45 isoform by ANK12 induced tyrosine phosphorylation of specific proteins in NK cells. Subsequently, the involvement of CD45 in the negative signaling after "self" MHC class I recognition by rat NK cells was investigated. The anti-CD45 mAbs were found to affect NK cell-mediated lysis of syngeneic tumor cell lines, depending upon the expression level of MHC class I on target cells. mAbs ANK74 and OX1 only inhibited lysis of the syngeneic tumor cell lines that expressed low levels of MHC class I. Furthermore, both mAbs caused an inhibition of NK cell-mediated lysis of these tumor cell lines when MHC class I molecules on the tumor cell lines were masked by an Ab. These results suggest that CD45 regulates the inhibitory signal pathway after self MHC class I recognition, supposedly by dephosphorylation of proteins.

Isolated lung perfusion with melphalan prolongs survival in a rat model of metastatic pulmonary adenocarcinoma

INTRODUCTION

Survival after isolated lung perfusion (ILuP) with melphalan was tested in a model of unilateral pulmonary adenocarcinoma.

METHODS

On day 0, rats were randomized into four groups: Group 1 (n = 9) received tumor cells intravenously for induction of bilateral lung metastases, whereas groups 2-4 (n = 21) underwent a 10-min occlusion of the right pulmonary artery during tumor cell injection for induction of unilateral left lung metastases. On day 7, groups 1 and 2 received no treatment. Group 3 underwent left ILuP with melphalan (2.0 mg/kg) while group 4 received melphalan intravenously (0.5 mg/kg). The end point of the study was death from metastatic disease.

RESULTS

Median survival of ILuP-treated animals (81 +/- 12 days) was significantly longer compared to group 1 (18 +/- 1 days; p = 0.0001), group 2 (28 +/- 3 days; p = 0.0002) and group 4 (37 +/- 6; p = 0.0004).

CONCLUSIONS

ILuP with melphalan prolongs survival in the treatment of experimental metastatic pulmonary carcinoma.

Acute-phase response patterns in isolated hepatic perfusion with tumour necrosis factor alpha (TNF-alpha) and melphalan in patients with colorectal liver metastases

BACKGROUND

In this study, we have evaluated hepatotoxicity, secondary cytokine production and hepatic acute-phase response (APR) in patients who underwent isolated hepatic perfusion (IHP) with tumour necrosis factor (TNF) alpha and melphalan for irresectable colorectal liver metastases.

DESIGN

An extracorporeal veno-venous bypass was used to shunt blood from the lower body and intestines to the heart. Inflow catheters were placed in the hepatic artery and portal vein, and an outflow catheter in the inferior caval vein. The liver was perfused for 60 min with 0.4 mg of TNF-alpha plus 1 mg kg-1 melphalan (IHPTM group, n = 6) or 1 mg kg-1 melphalan (IHPM group, n = 3). The liver was washed with macrodex before restoring vascular continuity.

RESULTS

After the washout procedure, a TNF-alpha peak (169 +/- 38 pg mL-1) was demonstrated in the IHPTM group only. Both groups demonstrated peak levels of interleukin 6 (IL-6) in the perfusate as well as systemically. These were significantly higher in the IHPTM group. Acute-phase protein (APP) levels followed a similar pattern as has been demonstrated after major surgery, with no significant differences between both groups. The addition of TNF-alpha to the perfusate did not lead to a significant difference in APP levels and the time course between groups.

CONCLUSIONS

IHP with TNF and melphalan is followed by a transient systemic peak of TNF directly after liver washout. Secondary IL-6 induction was seen in the present study after IHP with and without TNF, which was highest when TNF was added. This phenomenon cannot be extrapolated to APP induction, which appeared unaffected by the addition of TNF, presumably because the surgical procedure itself already causes maximal stimulation of APP production.

Specific tumor-cell killing with adenovirus vectors containing the apoptin gene

Specificity is an essential prerequisite for cancer gene therapy. Recently we described that apoptin, a protein of 121 amino acids which is derived from the chicken anemia virus, induces programmed cell death or apoptosis in transformed and malignant cells, but not in normal, diploid cells (Danen-van Oorschot AAAM et al, Proc Natl Acad Sci USA 1997; 94: 5843-5847). This protein has an intrinsic specificity that allows it to selectively kill tumor cells, irrespective of the p53 or Bcl-2 status of these cells. Hence, it is attractive to explore the use of the apoptin gene for therapeutic applications, viz cancer gene therapy. In this paper, we describe the generation and characterization of an adenovirus vector, AdMLPvp3, for the expression of apoptin. Despite the fact that apoptin ultimately induces apoptosis in the helper cells, which are transformed by the adenovirus type 5 early region 1 (E1), the propagation kinetics and yields of AdMLPvp3 are similar to those of control vectors. Infection with AdMLPvp3 of normal rat hepatocytes in cell culture did not increase the frequency of apoptosis. In contrast, in the hepatoma cell lines HepG2 and Hep3b, infection with AdMLPvp3, but not with control vectors, led to a rapid induction of programmed cell death. Experiments in rats demonstrated that AdMLPvp3 could be safely administered by intraperitoneal, subcutaneous or intravenous injection. Repeated intravenous doses of AdMLPvp3 were also well tolerated, indicating that the apoptin-expressing virus can be administered without severe adverse effects. In a preliminary experiment, a single intratumoral injection of AdMLPvp3 into a xenogeneic tumor (HepG2 cells in Balb/Cnu/nu mice) resulted in a significant reduction of tumor growth. Taken together, our data demonstrate that adenovirus vectors for the expression of the apoptin gene may constitute a powerful tool for the treatment of solid tumors.

Endogenous and adoptively transferred A-NK and T-LAK cells continuously accumulate within murine metastases up to 48 h after inoculation

In murine models, therapeutic efficacy of adoptive immunotherapy (AIT) of cancer with lymphokine activated killer (LAK) cells is seen only when applied together with substantial doses of interleukin-2 (IL-2), probably because this cytokine is imperative for both motility and viability of the LAK cells. We wanted to investigate whether IL-2 in addition mediates an immunostimulatory activation and expansion of endogenous effector cells contributing to tumor regression. Using an immunoperoxidase technique, we have been able to longitudinally analyze the accumulation of tumor infiltrating lymphocytes expressing the pan-T cell/activated lymphocyte phenotype (Thy1.2), the natural killer (NK) cell phenotype (AsGM,) as well as the cytotoxic T (CD8) cell phenotype within experimental established B16 pulmonary melanoma metastases in C57BL/6 mice during the first 48 h after high dose IL-2 monotherapy. Whereas a substantial and selective infiltration of AsGM1+ lymphocytes in tumor tissue was seen (262 and 937 cells per sq.mm malignant tissue at 0 and 48 h, respectively), only a minor increase in accumulation of CD8+ cells was seen (106 and 171 cells per sq.mm tumor tissue at 0 and 48 h, respectively). The addition of adoptive transfer with lymphokine-activated adherent NK (A-NK) cells to the high-dose IL-2 treatment resulted in more than a 1.5 fold increase in infiltrating AsGM1+ cells compared to IL-2 therapy alone (1520 compared to 937 AsGM1+ cells per sq.mm malignant tissue). No substantial accumulation of CD8+ cells was observed in this setting either. In contrast, the treatment with high dose IL-2 together with adoptive transfer of mitogen-stimulated, lymphokine-activated T killer (T-LAK) cells increased the infiltration of CD8+ cells 10-fold compared to IL-2 monotherapy (2078 compared to 171 CD8+ cells per sq.mm malignant tissue, respectively). Interestingly, infiltration of both endogenous and exogenous cells continued over time, since the effector-to-tumor cell ratio in metastatic tissue dramatically increased from 1:8 and 1:6 at 16 h to 1:3 and 1:2 at 48 h after adoptive transfer of A-NK and T-LAK cells, respectively. These data underline the longevity of LAK cells in vivo and highlight the importance of IL-2 treatment in recruiting endogenous immune cells to tumor areas.

Interleukin-2-activated natural killer cells can induce both apoptosis and necrosis in rat hepatocytes

Natural killer (NK) cells play a crucial role in the elimination of virus-infected or transformed cells in the liver. In this article, we describe the mechanism by which liver cells are killed by NK cells. Interleukin-2-activated natural killer (A-NK) cells from the rat induced apoptotic cell death in 30% of freshly isolated rat hepatocytes within 60 minutes. Recognition by the A-NK cells of the hepatocytes as nonself was established by masking the major histocompatibility complex (MHC) class I molecules on the hepatocytes with the OX18 antibody. During the killing process, a decrease of the mitochondrial membrane potential (MMP), formation of blebs, phosphatidyl serine (PS) externalization, chromatin condensation, and nuclear fragmentation were observed. The hepatocytes became apoptotic before permeabilization of the plasma membrane occurred, suggesting that the observed cytolysis was caused by secondary necrosis. The apoptotic process was completely abolished by the caspase inhibitors, Z-Val-Ala-DL-Asp fluormethylketone (zVAD-fmk) and Ac-Asp-Glu-Val-aldehyde (DEVD-cho). However, under these conditions, A-NK cells killed a smaller fraction of the hepatocytes by (primary) necrosis. These results indicate that apoptosis is the major cytotoxic process induced by A-NK cells in hepatocytes. If apoptosis is prevented, a more limited necrotic effect is induced. Therefore, this study shows that NK cells are fully equipped to induce both apoptosis and necrosis in hepatocytes, but appear to prefer the apoptotic route.

The development of a bi-specific anti-CD161A x anti-tumor antibody for rat NK cell targeting

In order to improve the therapeutic efficacy of adoptive immunotherapy of cancer using IL-2-activated NK (A-NK) cells, we developed a bi-specific monoclonal antibody (BimAb) 3.2.3xCC52. One specificity of the BimAb (mAb 3.2.3) was directed against rat CD161A (NKR-P1A) which has been shown to be an activation structure on rat NK cells involved in lysis of target cells and cytokine secretion. The other specificity (mAb CC52) was directed against a tumor associated antigen on the rat colon adenocarcinoma cell line CC531. The hybridomas producing 3.2.3 and CC52 were fused, resulting in a quadroma producing the desired 3.2.3xCC52 BimAb. The hybridomas produced antibodies of different isotypes (IgG2b and IgG1 respectively) which enabled us to pre-select quadromas with a high likelihood for production of BimAb, through testing for the production of bi-isotypic antibodies. Production of functional BimAb by the selected quadromas was demonstrated in an assay showing enhanced conjugate formation between CD161A+ cells and CC531 tumor cells. We also tested the 3.2.3xCC52 BimAb for its capacity to enhance NK cell-mediated lysis of CC531 tumor cells in 4 h and 19 h 51Cr release assays; in a prolonged (2 day) tumor neutralization assay using a tetrazolium salt (MTT)-based assay; and in tests for apoptosis using Annexin V-FITC. Although this BimAb was not demonstrated to cause enhanced lysis of CC531 cells by CD161A+ effector cells in vitro, it might be a useful tool to enhance the number of NK cells at the tumor site and/or prolong contact between tumor cells and NK cells in vivo, thereby probably enhancing the therapeutic efficacy of NK cells.

Characteristics of tumor infiltration by adoptively transferred and endogenous natural-killer cells in a syngeneic rat model: implications for the mechanism behind anti-tumor responses

Interleukin-2-activated, cultured NK cells (A-NK) cells were adoptively transferred into a syngeneic rat liver-tumor model. The kinetics of tumor infiltration by NK cells, originating either from adoptively transferred or from endogenous sources, the localization of these cells in the tumor, and their interactions with extracellular-matrix proteins were studied by immunohistochemistry and transmission-electron microscopy. The adoptive transfer of A-NK cells via the hepatic artery and s.c. injections of IL-2 into rats bearing subcapsularly induced CC531 liver tumors, but also IL-2 monotherapy, resulted in a significant increase of the number of NK cells both at the tumor border and in the tumor center. The majority of tumor-infiltrating NK cells was present in the tumor stroma and only occasionally was an NK cell observed in a tumor nodule in direct contact with tumor cells. Observations by electron microscopy suggested that matrix proteins, abundantly present in the tumor stroma but absent in the tumor nodules, provide a substrate for migration of infiltrating cells, whereas tight structures of matrix proteins surrounding tumor nodules provide a barrier for establishment of direct NK-cell-to-tumor-cell-contact. Our results suggest that direct NK-cell-to-target-cell-contact-mediated lysis is of minor importance for attaining an anti-tumor effect in this model. We hypothesize that treatment of tumor-bearing rats with A-NK cells and/or IL-2 initiates a cascade of events (e.g., secretion of tumor-killing cytokines and/or infiltration of other immune cells) ultimately leading to tumor regression.

Delivery of anticancer drugs via isolated hepatic perfusion: a promising strategy in the treatment of irresectable liver metastases?

The prognosis of patients with irresectable liver metastases derived from colorectal cancer is invariably poor; unfortunately, these tumours show only minor responses to conventional anticancer agents. The best responses have been obtained by fluoropyrimidines delivered as continuous infusion into the hepatic artery (HAI): their rapid uptake and detoxification by liver cells results in relatively low systemic drugs levels. This approach increases mean survival duration from 17 to 26 months and, in few patients, causes "down-staging" that may result in resectability. To improve opportunities for chemotherapy, the technique of 1-hour recirculating perfusion of the vascularly isolated liver (isolated hepatic perfusion, IHP) was developed. If leakage to the systemic circulation is negligible-and the compounds used do not readily cause hepatotoxicity-IHP allows usage of drug doses that would be fatal if delivered systemically. Because alkylating agents generally have steep dose-response curves, mitomycin C (MMC) and melphalan (L-PAM) entered phase I/II studies on IHP. Using these drugs, IHP was performed in principle as a single procedure in 60 otherwise untreated patients at our institution. However, despite preliminary data that indicate impressive clinical responses are obtained, improvement over HAI will probably be minor. Because IHP is a complicated way of drug delivery, one could argue that its use is justified only when it has the potential to kill all tumour cells in the liver. We critically discuss the possibilities of IHP and/or the use of gene therapy in an IHP setting.

Novel monoclonal antibodies against membrane structures that are preferentially expressed on IL-2-activated rat NK cells

Interleukin-2 (IL-2)-activated natural killer (NK) cells are known to mediate specific functions such as cytolytic activity and tumor infiltration more efficiently than nonactivated NK cells. To investigate whether these characteristics are associated with induction or up-regulation of expression of membrane structures after IL-2 activation, we selected four hybridomas (mAbs ANK44, ANK66, ANK7, and ANK123) derived from mice immunized with rat IL-2-activated NK cells and compared the expression of the epitopes recognized on IL-2-activated NK cells versus unstimulated NK cells. We found that ANK44-expression was induced after activation with IL-2. The antigens recognized by ANK66, ANK7, and ANK123 were expressed selectively on subsets of nonactivated NK cells. After activation with IL-2 the level of expression of these antigens was increased. Moreover, the majority of NK cells then expressed these antigens after IL-2 activation. Biochemical characterization of the membrane structures recognized on IL-2-activated NK cells showed that they have not previously been described. The precise function of these membrane structures is not yet known, however, the selective nature of their expression implies their involvement in the specific functions of IL-2-activated NK cells.

Regional administration of natural killer cells in a rat hepatic metastasis model results in better tumor infiltration and anti-tumor response than systemic administration

A syngeneic rat liver metastasis model, the CC531 colon carcinoma cell line in Wag rats, was used to study the homing properties and anti-tumor effects of adoptively transferred, interleukin-2 (IL-2)-activated, cultured natural killer (A-NK) cells. To identify the route of administration that gives the highest tumor infiltration, 1.5 x 10(8) A-NK cells were dyed with fluorescent rhodamine and injected via 4 different routes into rats, bearing subcapsularly induced (day 10) liver metastases. The routes chosen were: jugular vein, portal vein, hepatic artery and directly into the peritoneal cavity (i.p). The rats were sacrificed 20 hr after administration of A-NK cells. The highest (p < 0.05) infiltration of tumors by A-NK cells was found both at the tumor border and in the tumor center after injection via the hepatic artery: 65 +/- 7 A-NK cells/mm2 at the tumor border and 26 +/- 14 A-NK cells/mm2 in the center of the tumor (jugular vein infusion: 32 +/- 10 and 9 +/- 5 A-NK cells/mm2, respectively; portal vein infusion: 36 +/- 13 and 7 +/- 4 A-NK cells/mm2, respectively). No A-NK cells were detected in the liver after i.p. injection. Rats bearing day 5 tumors were injected with 1.5 x 10(8) A-NK cells via the hepatic artery or via the jugular vein (n = 5 and n = 6 respectively). Regional administration of A-NK cells via the hepatic artery resulted in a significant (p < 0.05) lower weight (35 +/- 23 mg) of tumors than did systemic administration (70 +/- 10 mg). Our results suggest that both the level of tumor infiltration by adoptively transferred A-NK cells and the therapeutic outcome depend on the route of administration.

Isolated hepatic perfusion with tumor necrosis factor alpha and melphalan: experimental studies in pigs and phase I data from humans

We report our experience with isolated hepatic perfusion (IHP) with tumor necrosis factor (TNF) and melphalan in an experimental pig study and of a phase I study in humans. IHP was performed with inflow catheters in the hepatic artery and portal vein and an outflow catheter in eh caval vein. An extracorporeal venovenous bypass was used. IHP consisted of a 60-min perfusion with hyperthermia (> 41 degrees C). For the pig protocol rhTNF alpha 50 micrograms/kg alone (n = 5) or rhTNF alpha 50 micrograms/kg plus melphalan 1 mg/kg (n = 3) were added. In two control pigs no drugs were added. In the phase I study, patients received melphalan 1 mg/kg with 0.4 mg rhTNF alpha (n = 8) or 0.8 mg rhTNF alpha (n = 1). After the perfusion the liver was washed with Macrodex before vascular continuity was restored. All pigs but one survived the procedure. Systemic leakage was less than 0.02%. Transient, mild liver toxicity was seen in all pigs, including controls, as demonstrated by liver enzyme assays and histology. There was no significant hemodynamic, cardiopulmonary hematological, or renal toxicity. In the phase I clinical study there was leakage in one patient (cumulative leakage 20%). There were three perioperative deaths (one possibly drug-related). All patients demonstrated significant hepatotoxicity. Survival ranged from 6 to 26 months (median 10.3 months). All patients demonstrated a tumor response (partial response 5/6, 1/6 stable disease) with a median duration of 18 weeks. In contrast to our pig program, many problems were encountered in the phase I study. By using both the hepatic artery and portal vein for IHP we encountered more toxicity than expected based on data from the pig program, resulting in fatal coagulative disturbances in one patient who received the second rhTNF alpha dose. Furthermore, local control after one IHP with TNF alpha and melphalan is only temporary.

Characterization of three new membrane structures on rat NK cells which are involved in activation of the lytic machinery

In this study, three membrane structures on rat NK cells which activate lysis of target cells were characterized. Furthermore, the role of adhesion molecules in this activation process, in particular the CD18-associated integrins, was investigated. Three rat NK-activation structures were identified which have not been previously described. These structures are apparently unique as they differed in molecular weight from known NK-activation structures. Cross-linking of these activation structures with specific mAbs and a Fc gamma R-positive tumor cell line (P815) resulted in enhanced killing of these target cells by NK cells. If the CD18-associated integrins were masked by the anti-CD18 mAb WT.3, the redirected killing of P815 was completely blocked. This indicates that the CD18-associated integrins play a crucial role in activation of NK cells. Furthermore, our results show that rat NK cells possess multiple activation structures.

Identification of potential HLA-A *0201 restricted CTL epitopes derived from the epithelial cell adhesion molecule (Ep-CAM) and the carcinoembryonic antigen (CEA)

The altered expression pattern of the Epithelial Cell Adhesion Molecule (Ep-CAM) and the Carcinoembryonic Antigen (CEA) on tumor cells of epithelial origin as compared to normal epithelia may permit T cells to preferentially recognize and lyse these tumor cells. The binding affinity for human leucocyte antigen A2.1 (HLA-A*0201) and the capacity to form stable peptide-major histocompatibility complex (MHC) interactions with this molecule were tested for 410 Ep-CAM-derived sequences, including an overlapping set of 9 amino-acid-long peptides, and 73 CEA-derived peptides fulfilling the HLA-A*0201 motif. Peptides with a high binding affinity and a low peptide-MHC dissociation rate were subsequently tested for their immunogenicity in HLA-A*0201Kb transgenic mice. One Ep-CAM-derived peptide and 1 CEA-derived peptide were able to reproducibly induce peptide-specific cytotoxic T cells (CTL) in these mice. This indicates that EpCAM and CEA are potential target antigens for CTL-mediated immunotherapy of epithelial cancers.

Liver and tumour tissue concentrations of TNF-alpha in cancer patients treated with TNF-alpha and melphalan by isolated liver perfusion

In this study we determined the level of tumour necrosis factor alpha (TNF-alpha) in liver and tumour tissue samples obtained from patients with colorectal metastases confined to the liver, who were treated with isolated liver perfusion with TNF-alpha and melphalan. We adapted a standard enzyme-linked immunosorbent assay kit for the quantification of TNF-alpha in serum to measure the amount of this cytokine in solid tissue. For this purpose, we developed a buffer that lysed the tissues without affecting the TNF-alpha present. The minimum detection level was about 2 pg of TNF-alpha per mg tissue. Using this technique, we found a significant increase in the TNF-alpha level after perfusion in the liver tissue of all evaluable patients, which may explain the transient liver toxicity we observed in all patients. In tumour tissue, a significant TNF-alpha increase was observed in one out of five patients. The level of TNF-alpha in all liver tissue samples and some of the tumours after treatment by isolated liver perfusion was much higher than the peak serum concentrations obtained after systemic administration of the maximum tolerated dose of TNF-alpha. Furthermore, we demonstrated that the level of TNF-alpha in the liver tissue samples was about seven to eight times higher than in tumour tissue. We concluded that regional liver treatment resulted in a relatively high local level of TNF-alpha, but also that this cytokine did not preferentially accumulate in tumour tissue.

The role of adhesion molecules in the recruitment of hepatic natural killer cells (pit cells) in rat liver

Previous studies showed that blood large granular lymphocytes (LGL), which possess natural killer (NK) activity, develop within rat liver sinusoids into high-density (HD) and subsequently into low-density (LD) pit cells which show an increasing level and spectrum of tumor cytotoxicity. In this study, we investigated the role of adhesion molecules, such as CD2, CD11a, CD18, and CD54 in the recruitment of pit cells to the liver. Immunostaining for electron microscopy, and two color flow cytometry showed that most pit cells expressed CD2, CD11a, CD18, and CD54. After intravenous injections into rats with anti-CD2, anti-CD11a, and anti-CD18 antibodies, the number of pit cells per square millimeter in frozen sections of liver tissue decreased. Treatment of rats with zymosan increased the number of pit cells fivefold, whereas subsequent treatment with anti-adhesion-molecule antibodies resulted in approximately 60% lower number of pit cells. Anti-CD54, supposed to block CD54 expression on sinusoidal endothelial cells, also decreased the number of pit cells. The number of blood LGL was, however, not affected by these antibodies. These results indicate that blocking of CD2, CD11a, CD18, and CD54 antigens on blood LGL and/or liver endothelium decreased the number of pit cells in the liver. These adhesion molecules therefore play an important role in the recruitment of pit cells in the liver.

The role of MHC class I expression in rat NK cell-mediated lysis of syngeneic tumor cells and virus-infected cells

In this study the role of MHC class I antigen expression in rat natural killer (NK) cell-mediated lysis was investigated. Various rat tumor cell lines and two Adenovirus (Ad)-transformed rat cell lines were tested for their expression levels of total MHC class I and two MHC class I alleles, RT1.A and RT1.C, by flow cytometry. Their susceptibility to NK cell-mediated lysis in relation to MHC class I expression was determined by 51Cr release assays. IFN-gamma is know to increase the expression of MHC class I. Therefore target cell with and without prior IFN-gamma treatment were examined for MHC class I expression and its effect on NK lysis. An significant inverse exponential relationship was found. To investigate the effect of virus infection on MHC class I expression and target cell lysis by NK cells, rat embryonal fibroblasts (REF) were infected with cytomegalovirus (CMV) and used as target cells for NK cell-mediated lysis. Results showed that these virus-infected cells were less susceptible to NK lysis than non-infected cells. Moreover, the non-infected cells expressed less MHC class I than the infected cells, indicating that also in this case, there was an inverse correlation between MHC class I expression and susceptibility to lysis by NK cells. Subsequently, we showed that sorted subsets of predominantly CD8-positive and CD8-negative NK cells lysed a MHC class I-positive tumor cell line at the same level. This suggests that CD8 is not likely to participate as a receptor for MHC class I in NK cell-mediated lysis in a syngeneic rat model.

The use of Epstein-Barr virus-transformed B lymphocyte cell lines in a peptide-reconstitution assay: identification of CEA-related HLA-A*0301-restricted potential cytotoxic T-lymphocyte epitopes

In the development of cytotoxic T lymphocyte (CTL)-mediated immunotherapy, the identification of CTL epitopes is of crucial importance. Binding of a peptide to major histocompatibility complex (MHC) class I molecules is one of the prerequisites for its function as a CTL epitope. We describe the technique, validation, and application of a simple cellular assay, intended for the screening of peptides for binding, that can be applied to any human leukocyte antigen (HLA) allele. Reconstitution of peptides in MHC class I molecules after elution by acid treatment was previously shown to be possible in specially engineered cell lines expressing only one type of MHC class I, and was applied for the HLA-A*0201 allele. We now report the optimal conditions for application of this type of binding assay to the HLA-A*0301 allele. The adaptations that were necessary to make the technique operational for HLA-A*0301 are shown in detail. These consisted of lowering the pH during acid treatment to 2.9 and lengthening the duration of elution to 90 s. Furthermore, immediate aspiration of eluted peptides appeared to be essential for this allele. We found also that the use of Epstein-Barr virus (EBV)-transformed B cell lines (B-LCL) yields results similar to those of the use of cell lines expressing only one specific MHC class I allele. Homozygosity for the desired HLA allele improves the sensitivity of the assay, but heterozygous cells can also be employed. Finally, we applied this technique to a search for HLA-A*0301 binding peptides derived from carcinoembryonic antigen (CEA). Of a set of 34 CEA-specific peptides that fit with a specified HLA-A*0301-binding motif, we identified a set of six peptides with high binding affinity to this allele. These peptides can be regarded as potential CTL epitopes.

Tumor heterogeneity and immunotherapy of cancer

Tumor heterogeneity is the presence of intercellular differences, either from clonal origin or present within subpopulations of tumor cells. Recent advances in immuno-histology, flow cytometric analysis, molecular biological techniques, and tissue culture methods makes it possible to investigate tumor heterogeneity in detail. In this review data are presented to document that this hallmark of neoplastic disease results from DNA-instability and the interactions of tumor cells with their environment. The present inability to treat most patients effectively with immunotherapy may partly be due to the occurrence of tumor heterogeneity. Therefore, the heterogeneity of the tumor phenotype is discussed in conjunction with the various immunotherapeutic treatment modalities. In addition to local cytokine production by immune cells and tumor cells, and limited access of either antibodies or immune cells into the tumor, tumor heterogeneity is an important factor that determines the progress of immunotherapy of cancer. Therefore, accurate quantitative methods using antibodies and molecular probes to identify HLA-associated target peptides, tumor-associated antigens and accessory molecules, to predict which patients will have a high probability of responding to treatment, are needed.

The number and distribution of hepatic natural killer cells (pit cells) in normal rat liver: an immunohistochemical study

Pit cells are a unique population of cells in sinusoids and peripheral blood, which can be considered natural killer (NK) cells with large granular lymphocyte (LGL) morphology. The aim of this study was to investigate the use of the monoclonal antibody (MAb) 3.2.3 as a specific marker of rat pit cells to detect their number and distribution in the liver. The number of 3.2.3-positive cells was comparable to the number of LGL in liver low-density (LD) and high-density (HD) pit cell fractions and in blood lymphocytes (P > .05). Immunoelectron microscopy showed that nearly all LGL in hepatic LD and HD and in blood fractions were 3.2.3 positive. Using MAb 3.2.3 immunoperoxidase staining, the mean number of pit cells in liver frozen sections was determined to be 13.7 +/- 1.1/mm2. The number of pit cells was similar in the different liver lobes (P > .05). Reduced nicotinamide-adenine dinucleotide (NADH) oxidase histochemical staining, to visualize a portal to central vein gradient, combined with immunostaining was used to analyze the lobular distribution of pit cells. We found that 61.3% (17.1/mm2) of pit cells were in the periportal area and 38.7% (10.8/mm2) in the central area. We conclude that MAb 3.2.3 can be used as a specific marker of rat pit cells and therefore can be used to quantify rat pit cell number in various experimental models.

The infiltration of experimentally induced lung metastases of colon carcinoma CC531 by adoptively transferred interleukin-2-activated natural killer cells in Wag rats

The number of IL-2-activated natural killer (A-NK) cells reaching the tumor site in vivo may be crucial for their anti-tumor effect following adoptive immunotherapy. We investigated in a syngeneic rat model the infiltration of established lung metastases by adoptively transferred A-NK cells. The Wag rat colon carcinoma CC531 was injected via a tail vein to induce pulmonary metastases. Syngeneic A-NK cells were labeled with the fluorescent dye rhodamine (TRITC) and next injected via a tail vein in rats bearing day-12 lung tumors. The number of A-NK cells in tumor and in normal tissue per rat was counted in sections after administration of A-NK cells. At all time points tested, a significant linear relationship between the cross-section area of the tumor and the number of infiltrating cells was observed, but small tumor areas became fully infiltrated earlier than larger areas. At 24 hr after injection, approximately 10% of the injected cells were found in the tumor tissue and the average A-NK-cell-to-tumor-cell ratio was estimated to be 1:3. A-NK cells were found in the liver too, although the number of cells per mm2 tissue was low compared with the pulmonary tumor tissue. Very low numbers of A-NK cells were found in kidney, adrenal gland, spleen, and blood. We conclude that, in this syngeneic rat model, adoptively transferred A-NK cells are able to find and specifically infiltrate pulmonary metastases in a time-dependent fashion.

Soluble factors produced by macrophages/monocytes inhibit lymphokine-activated killer activity in rat splenocyte cultures

In the present study we investigated the inhibition of interleukin-2(IL-2)-induced lymphokine-activated killer (LAK) activity in rat splenocyte cultures in relation to the presence of 2-mercaptoethanol and macrophages/monocytes. The presence of 2-mercaptoethanol is necessary for induction of LAK activity in rat splenocyte cultures. Removal of macrophages/monocytes from rat splenocytes by plastic or nylon-wool adherence, or iron ingestion resulted in LAK induction by IL-2 in the absence of 2-mercaptoethanol. The effect of macrophages/monocytes on LAK activity was also studied in transwell co-cultures. In the absence of 2-mercaptoethanol, the induction of LAK activity was very low in macrophage/monocyte-depleted splenocytes with macrophages/monocytes in the upper compartment of a transwell culture. In contrast, in the presence of 2-mercaptoethanol a high level of LAK activity was induced in these transwell cultures, showing that 2-mercaptoethanol abolished the LAK-inhibiting capacity of macrophages/monocytes. In addition, established LAK activity was strongly inhibited when, after LAK induction, splenocytes were cultured with supernatant of unfractionated splenocytes, which were cultured with IL-2 but in the absence of 2-mercaptoethanol. Addition of 2-mercaptoethanol abrogated the inhibiting effect of the supernatant completely. These experiments demonstrate that rat macrophages/monocytes produce 2-mercaptoethanol-sensitive soluble LAK-inhibiting factors. Ultrafiltration of conditioned culture medium of macrophages/monocytes revealed the presence of LAK-inhibiting factors larger than 10 kDa. We concluded that 2-mercaptoethanol-sensitive soluble factors produced by macrophages/monocytes determine the level of LAK induction in rat splenocyte cultures.

Rat interleukin-2-activated natural killer (A-NK) cell-mediated lysis is determined by the presence of CD18 on A-NK cells and the absence of major histocompatibility complex class I on target cells

The precise mechanism by which target cells are recognized and subsequently lysed by interleukin-2-activated natural killer (A-NK) cells is poorly understood. In this study the role of major histocompatibility complex (MHC) class I and adhesion molecules in the recognition and lysis of tumor cells was investigated in a syngeneic Wag rat model. Preincubation of tumor cells with F(ab')2 fragments of anti-MHC class I monoclonal antibody (mAb) OX18 strongly enhanced the A-NK cell-mediated lysis. Also normal syngeneic cells such as T cells and A-NK cells became highly sensitive for lysis by A-NK cells after preincubation with mAb OX18. Two other mAb against MHC class I had no effect on lysis of target cells. These data indicate that masking of MHC class I on syngeneic tumor and normal cells by mAb OX18 is sufficient for A-NK cells to recognize target cells as non-self, resulting in lysis. In addition, we found that the presence of mAb against the beta 2 (CD18)-integrins blocked the lysis of all tumor cell lines by A-NK cells in 51Cr-release assays, also when target cells were preincubated with mAb OX18. Because of the absence of CD18 on most tumor cells we concluded that a CD18-associated integrin on A-NK cells is essential for lysis of target cells. These results show that in this syngeneic rat model CD18 on A-NK cells together with MHC class I on tumor cells determine A-NK cell-mediated lysis. Furthermore, we hypothesize that the anti-MHC class I OX18 recognizes an epitope on rat MHC class I which is, or is very close to, the restriction element determining A-NK cell-mediated lysis.