Michael T. Lotze, MD

1. Introducing NK and DC into Cancer Care. Interestingly, cellular immunology is now becoming synthetic with the initiation of the primary and the secondary immune response dictated by these cell:cell interactions. We have developed a Program Project grant based on these notions including 5 projects and 5 cores; it was submitted October 1, 2003 and reviewed March 3, 2003. Although not funded, we received an enthusiastic response and were requested to respond to criticisms and resubmit.

Cellular Therapy of Cancer: Immunologic Considerations. Requirements for immune control of cancer include the need to attain T-cells capable of migrating across endothelial barriers, surviving at the tumor site, and mediating a prolonged memory response. Multiple studies in murine models predict that the induction and delivery of therapeutic T-cells mediates important antitumor effects. A number of studies, particularly in the setting of patients with melanoma have now demonstrated that adoptive transfer of such cells mediates antitumor effects. Substantial host modification is required including the use of cytotoxic agents to "open space" for the adoptively transferred cells or to inhibit nominal suppressor T-cells. An alternative strategy, and one promoted in the context of this application, is the notion that the chronic inflammatory response to cancer, is already mediated by T-cells and DCs. Furthermore that the T-cells found within tumors are ineffective and with expansion, of limited utility. The fundamental premise of this proposal, based on an integration of the available insights into the biology, is to test the delivery of NK and DCs coordinately to elicit new tumor-specific effector cells, driving the acute inflammatory response with DC1s matured with NK cells in vitro or in vivo. Although substantial information has been attained about the signals eliciting the adaptive immune response mediated by T-cells, the classic Bretscher and Cohn < A theory of self and non-self discrimination. Bretscher PA, Cohn M. Science 1970;69:104 > model of two signals published now over 30 years ago did not adequately define the biology. Although we recognize that MHC/peptide or Signal 1 and the provision of costimulation, Signal 2 are requisite to drive naïve T-cells, it was not until the last decade that it became apparent that this was the unique set of signals provided by DCs to either immunize or tolerize. Furthermore, Janeway suggested that there needed to be a so-called Signal 0. This was initially believed to be signals coming directly from bacteria, viruses, or other forms of injury or "danger" which prompted the initiation of the immune response. We now recognize that these signals largely drive Toll-like receptors on macrophages recruiting acutely polymorphonuclear leukocytes, NK cells, and DC to the site. Subsequently Signal 3, reflecting the nature of the environmental stimuli provides a polarizing signal, initiating the TH1 or TH2 type of response. At some point after continued inflammation, signals associated with healing, containment, or persistence of chronic inflammation are found, often in the setting of chronic viral infections [hepatitis, HPV, herpes, etc], toxins [smoking, arsenic], or irritation [esophagitis, solar irradiation, etc.]. Many states of chronic inflammation are the setting in which cancer arises. Interfering with this by using conventional vaccination strategies, recapitulating Signals 1-3 are believed to be unlikely to succeed because of the established chronic inflammatory state with factors such as TGFß, nitric oxide, IL-10, prostanoids, and others which limit the ability to launch an acute inflammatory response. One of the central hypotheses of this proposal, tested in both transplantable and spontaneous tumor models, is that inducing an acute inflammatory response with the adoptive transfer of NK and DC will allow induction of a new T-cell response. Data coming from several centers now including our own suggests that this NK/DC interaction plays a primary role in polarizing the DC and enhancing IL-12 and possibly IL-23 production.

Hypothetical Impact of Inducing an Acute Inflammatory Response to Cancer with NK and DCs. We hypothesize that NK will enhance DC function and vaccine efficacy in driving the expansion of tumor-reactive T cells that may be recruited into the tumor micro-enviroment and mediate tumoricidal effects. We also hypothesize that NK delivered intra-tumorally, particularly in combination with DCs, will result in the apoptotic (via Fas Ligand, TNF, and granzyme) and necrotic (perforin) death of tumor cells that will enhance cross-presentation by DCs and the development of new anti-tumor T cells. In this setting inflammatory mediators, tissue breakdown products, and both inflammatory and noninflammatory cells such as endothelial cells and macrophages will be delivered into the peripheral blood. It is in this setting that we imagine that some of the biomarker and surrogate measures will be most effective. The ability to either suppress macrophage products which limit the nature of the inflammatory response in the setting of chronic infection and or enhance the NK/DC interactions promoting IL-1 family members and/or IL-12 family member production by DC's seems like an plausible strategy to enhance new T-cell reactivity.

PO1 Application - Adjuvant Biologic Therapy of Cancer. A group of experienced clinician/scientists gathered together at the University of Pittsburgh Cancer Institute three years ago [John Kirkwood, Ian Pollack, Hideho Okada, Michael Wong, R. Milburn Jessup, Michael Lotze, and Joseph Baar] to discuss this problem. We recognized that we needed to apply biologic agents for cancer earlier and in patients for whom conventional measures of disease [metastases] were not available to determine success or failure. Measures of time to recurrence were felt to require too large a sample size for such early, unproven reagents. We presented this strategy of early intervention in a series of planned clinical trials to colleagues from Cancer Therapy Evaluation Program and the NCI Extramural Program Review group including Drs. Diane Bronzert and Toby Hecht. The sense was, in the absence of well-defined and available biomarkers, that this kind of effort would be difficult to support. This effort terminated. In many ways it is now being redeveloped but with more tractable strategies using proteomics, genomics, and cellomics to identify and apply biomarkers and surrogates in the setting of cellular therapies for glioma, melanoma, and colorectal cancer. Dr. Lotze has presented twice to the Food and Drug Administration Biomarkers Group [Dr. Joseph Hackett] in the Spring and Fall of 2002. Those present agreed fundamentally that the biology of cancer prompted a strategy of developing "patterns of biomarkers", and that finding these in patients was challenging. They emphasized that having suitable animal models was important for making the linkages through clinical trials plausible.

2. Evaluating Imaging Cytometry as a Measure of Immune Activity in vitro and in vivo. The Cellomics Core is housed in a shared facility of the University of Pitsburgh Medical School approximately 1200sq ft. of space, which has fully equipped ArrayScanII High Content Screening System, Cellomics Store analysis system and computers. The Cellomics assay system was pioneered and developed by "Cellomics Inc" for the purposes of drug discovery processes, and until now it has been mainly used by pharmaceutical industry to measure the effects of potential drugs on complex molecular events such as signal transduction pathways, as well as effects on cell functions like apoptosis, division, cell adhesion, locomotion, exocytosis, and cell-cell communication.
The Cellomics Core was established to provide multiplexed functional imaging, screening, analysis and quality control of samples (including the cell products for clinical use) from the five proposed research projects. The measurements are based on fluorescence imaging of multiple targets in the context of intact cells. Up to six different fluorochromes can be used and analyzed separately in one assay. The various assays are performed and screened by an automated instrument "Array Scan II". The characteristics of "The Array Scan II" include: 1)The instrument accepts various different kind of plates (96-well, 24-well, 12-well, etc) and rapidly screens the plates and measures in excess of 20 cell measurements from up to four separate cellular targets, 2) automatically finds cell fields, sets exposures, focus and thresholds during plate scanning, 3) quantifies multiple fluorescent signals on or in cells, 4) automatically converts raw image data into signal distribution, area, morphology, and activity measurements, 5) stores images and all data collected for each cell or cell field measured for later review. We have been evaluating it for the purpose of advancing biomarkers and surrogates and to evaluate cell:cell interactions.

The Automation Partnership SelecT. The group preparing automated systems for cell handling and culture for the Pharmaceutical industry is located in Royston outside of Cambridge in the United Kingdom. They have developed automated systems for handling the requirement of compound storage and acquisition for the Pharmaceutical industry, robotocized systems for maintenance of cryopreserved DNA, RNA, and cellular material. With several large pharmaceutical firms, they developed and exceeded all requirements for developing equipment for automated cell grow up and expansion enabling the development of T175 flasks which can handle within a Class 100 compartment 168 concurrent and separate cultures with the appropriate liquid handling systems and automated bar coded controls to allow for robust and cost effective cell culture systems. In conjunction with Dr. Lotze, the system has been reviewed and identified as an outstanding system for ushering in the next period of academic based cell culture systems meeting and exceeding all requirements of the FDA, with whom he has met to review this strategy. Dr. David Cumpstey has visited with Drs. Herberman and Lotze and has provided substantial enthusiasm and guidance as to how to most effectively integrate this system into modern cell therapy. The equipment will be run and operated within the MMI and the McGowen Institute for Regenerative Medicine and will be delivered sometime in November and fully operational by the end of the year for automated culture of cells. It is expected that all preclinical experiments requiring culture of cells from leukopheresis and from murine bone marrow and splenocytes will be cultured here with integrated quality controls coming from the Cellomics measure of function and phenotype. This system will be evaluated in conjunction with Core B as a site for developing cell therapies for patients and a dedicated second machine will be obtained at the end of the first 5 year granting period to assume the needs of the Core for development of cell therapies for the region for delivery to patients, once certified by the FDA for such purposes and validated with Core E-B.

3. Developing Clinical Protocols using Cell Therapy for Patients with Cancer.
Arguably, one of the most effective immunotherapies for the treatment of patients with metastatic melanoma is systemic IL-2. In our extensive experience with this therapy we observed significant response rates (15-20%) with small but durable numbers of patients surviving long term (7%). Attempts to increase the response rate to IL-2 by co-administration of ANK/LAK did not demonstrate statistically significant improvement. Recently we have been actively investigating the efficacy of a variety of DC based therapies and again we have observed promising early results but few long term responders. The failure to improve on the durable response rates (over that seen with IL-2 alone) in patients with metastatic melanoma is obviously mutltifactorial, but we believe in a large part is due to our lack of understanding about the important role that the interface between the innate and adaptive immune response plays in generating an effective and specific anti-tumor responses.
Recent studies have revealed that the cells of the innate immune system are critically important to initiate the cascade of events that creates the appropriate environment for T cell maturation. While many of the molecular components of these pathways have yet to be fully characterized, it is clear is that environmental stimuli activate antigen presenting cells (chiefly DC) thorough specific signaling pathways. These lead to the presentation of antigen and elaboration of a cascade of co-stimulatory molecules and soluble cytokines that are able drive the development of specific adaptive immune responses. It has also become apparent that early in the initiation events, there is critical cross talk between the APC and NK cells and then later CD4, CD8 T cells that determine the polarization (Th1 vs Th2), and intensity of the immune response (signals 1-3) . Later signals, what we have termed signal 4, allow either resolution and healing or induce a state that allows the immune system to deal with chronic antigen persistence. The overall goal/central hypothesis of this PPG is to, in a systematic way, elucidate the precise molecular mechanisms involved in the critical cross talk between the innate (NK) and adaptive immune response (DC). Specifically, this proposal seeks to understand on a molecular level the role IL-1 homologues play in the initiation and termination of the immune response through their effects on NK and DC. Because the ultimate goal of this proposal and of this program project grant is to develop more effective therapies for the treatment of human cancers, we will initiate clinical trails that will test our central hypothesis.
3.1. Test ANK + escalating iDC [GM/IL-13] in patients with melanoma lesions.
Based on our preliminary data we have observed that the interaction of activated NK (ANK/LAK) with immature DC in the setting of tumor leads to creation of more efficient T cell responses. We hypothesize that delivery of each of these two cells into the tumor microenvironment in vivo will create a novel vaccine that will lead to induction of an effective T cell response to melanoma. We will test this first hypothesis by directly injecting ANK/LAK and immature DC into metastatic melanoma lesions. We will administer systemic IL-2 in a decresendo dosing schema at the time of injection to support the ANK/LAK (IL-2 dependent) and to allow for activation and expansion of specific T cells. Eligibility: Patients with metastatic melanoma who have lesions that are accessible to direct injection and other evaulable disease, and who meet other standard eligibility requirements.. Protocol:. After informed consent patients will be treated by direct injection of a metastatic melanoma lesion with ANK and DC on the following dose escalation schema: Direct injection of highest feasible dose of ANK/DC (estimated to be 108). ANK will harvested and purified by the Cell Therapy core as previously described. All patients will receive a short course (5 days) of decrescendo interleukin-2 systemically. If no grade 3 or 4 toxicity is observed then a second injection will be performed at two weeks. We will next examine co-administration of immature DCs. DCs will be harvested and purified by our cell therapy core as previously described. Increasing numbers of immature DC will be co- administered with 108 ANK/LAK in the following groups: 1) 107 DC; 3 patients; 2) 108 DC; 3 patients; 3) 109 DC ( maximum feasible dose from a single leukopheresis); 3 patients. Following safe administration (no treatment limiting grade 3 or 4 toxicity) a second injection will be administered two weeks later. All patients will undergo biopsy of the injected lesion at one month as well as clinical and radiographic evaluation of their disease. If we see no evidence of treatment-related toxicity and evidence of local or systemic response we would propose to treat another 10 patients at the MTD. Clinical Endpoints: This will be a phase I trial and primary endpoint will be toxicity secondary endpoint will include measurement of evaluable disesease. Patients will be assessed for response at 1 month and then every three months to progression. Immunological endpoints will include development of specific T cell responses to autologous melanoma when available and/or measument of specific CD4/CD8 T cells responsive to a panel of HLA appropriate melanoma epitopes (by ELISPOT or Tetramer). Sequential blood specimens as will be obtained to measure biomarkers as described in Aim III.
3.2. Randomize best IL-1Fx + IL-2 versus IL-2 driven ANK delivery with DC in patients with melanoma. We would anticipate that concurrent with these clinical studies our preclinical evaluation of IL-1 homologues and their ability to augment or inhibit the development of specific T cell responses through their effect on NK and DC will progress. We will integrate these findings into our clinical protocol as the data is validated in vitro and in preclinical murine models. At present we imagine conducting a second phase II trial that would examine the efficacy of coadministration of ANK activated by a combination of IL-2 and IL-1Fx (most likely IL-18 based on our preliminary data) with immature DC. Endpoints in these follow-on trials would be to assure no added toxicity and effect on clinical disease and immunological readout as described above.

4. Developing Novel Therapies and Measures of Activity for Patients with Liver Diseases. We have begun exploring several research themes with the groups studying the liver at the University of Pittsburgh and just brought our first speaker, Dr. Nicholas Crispe, coiner of the term "elephants graveyard" for the liver as a site for CD8+ T-cell death. Other themes which we are exploring in the context of an expected PO1 application next year are the following:

• Early detection and recognition of hepatoma/metastatic colorectal cancer in the liver using proteomics and genomics focusing on inflammatory cells and proteins.
• A distinctive liver architecture, unusual resident liver lymphocytes, and trafficking pattern of leukocytes are found throughout the liver. This accounts for the unusual pattern of tolerance, immunity, and pathology in this organ.
• Liver pathogens use multiple complementary strategies to avoid T-cell immunity
• The liver induces T-cell tolerance by killing T-cells
• Death receptors on liver cells are instrumental in hepatocyte damage/regeneration
• Liver lymphocytes control liver regeneration and fibrosis
• HMGB1 is central to rounds of necrosis and apoptosis in the liver
• Natural Lymphocytes [NK, NKT] control disease outcome
• Stellate cells regulate hepatic function and regeneration
• Dendritic cells and DC variants are important in hepatic remodeling. Effective treatment of chronic inflammatory diseases of the liver will require induction of regulatory T-cells and new T-cells. DCs are central to any therapeutic strategy.
• Just as TNF inhibition has allowed resolution of arthritis and joint remodeling in RA, appropriate targeted therapy in the liver can allow resolution of inflammation, remodeling and regaining of normal function
• Transplantation of the liver is the treatment of choice for chronic liver disease and means to ablate residual tumor or pathogens represent important targets for expanded application.