Oral Abstracts

Intratumoral Expansion of Tumor-Specific B Cells in Invasive Ductal Carcinoma of the Breast

Julia A. Coronella, Matthew Welch, Cathy Spier, Leah Tatum, Alison Stopeck, Hugo Villar, Richard Junghans and Evan M. Hersh.
Arizona Cancer Center, University of Arizona, Tucson AZ

Background: In the search for novel or improved immunotherapeutic approaches to breast cancer, we seek to understand in vivo antitumor humoral responses, and to clone naturally occurring antibodies and their cognate antigens. Naturally occurring humoral responses to breast cancer are evidenced by tumor-reactive serum Ig and tumor-reactive lymph node B cells. In addition, most breast tumors contain tumor-infiltrating lymphocytes. While many studies have been made of local B cell reactions in pathologic autoimmune states, little study has been done in solid tumors, where understanding such reactions might assist in formulating more effective immunotherapy strategies.
Objective: The objective of this study was to determine if tumor-infiltrating B cells (TIL-B) of invasive ductal carcinoma of the breast (IDC) represent a local tumor-specific humoral immune response.
Results: Immunohistochemical analysis of IDC tumors shows that B cells infiltrating IDC tumors cluster in intratumoral follicles, consistent with an ongoing in situ response. These follicles contain B and T cell zones and follicular dendritic cells, similar to a nodal germinal center. A total of 29, 31, 58 IgG1 heavy chains were cloned and sequenced from three IDC tumors as well as sequences from tumor-draining lymph node and peripheral blood (PBL) of a healthy donor. Intratumoral proliferation of TIL-B as evidenced by multiple clones derived from common progenitor B cells was seen. Of TIL-B sequences, 44-67% were clonal relatives. Repeated clones were not observed in PBL samples, and are known to occur at only low frequency. In addition, TIL-B clones contained unique mutations, indicating somatic mutation concurrent with proliferation in intratumoral germinal centers. Somatic mutation levels and patterns were extremely similar to those observed for node IgG, and at much higher frequency than peripheral blood IgG, suggesting antigen selection. Preliminary data also suggests that phage-displayed antibodies from TIL-B bind autologous tumor.
Discussion: From these data, we conclude that TIL-B constitute a local tumor-antigen driven B cell response. In addition to providing novel anti-tumor antibodies for possible use as diagnostic or therapeutic agents, this study will in the future identify reactive tumor antigens that may be a source of immunologically available targets for vaccine-based immunotherapy.

Signaling defects in CD8+CD27-CD45RO- circulating effector cells in patients with cancer

Iris Kuss, Albert Donnenberg, Theresa L.Whiteside
University of Pittsburgh Cancer Institute Pittsburgh, PA

Background: The analysis of naïve, memory and effector CD8+ T cells is important in patients with cancer, because changes in these subsets might be linked to disease progression or responses to immune therapies. To date, the discrimination between naïve and memory cells was based on CD45 isoform expression. However, the CD8+CD45RA+ population of naïve T cells was shown to contain cells with the surface phenotype of CD45RA+CD27- representing cytotoxic effector cells.

Methods: To quantify subpopulations of CD8+ T cells, we performed 4-color flow cytometry with anti-CD8, -CD45RA, -CD45RO and -CD27 antibodies using peripheral blood mononuclear cells obtained from 26 patients with head and neck cancer (HNC) aged from 36-82 and 30 age-matched normal controls (NC). Also, expression of the T-cell receptor associated z chain was evaluated by flow cytometry in 13 patients and 14 NC.

Results: The percentages of CD8+CD45RA+ and CD8+CD45RO+ T cells were not different in patients vs. NC. The population of unprimed CD8+CD45RO-CD27+ T cells decreased as a function of age in patients as well as NC, and it was significantly lower in HNC patients than in NC ( 27 vs. 57%; p<0.0001). In contrast, the population defined as CD8+CD45RO-CD27- effector T cells increased with age in both patients and NC, and it was significantly higher in HNC patients vs. NC (55 vs. 32%; p<0.0009). Within the memory CD8+CD45RO+ subset, the proportions of CD27+ and CD27- T cells were comparable in patients and NC. When z expression in CD8+ cells was examined, the mean fluorescence intensity (MFI) was found to be 152 in patients vs. 553 in NC (p=0.0002). In CD8+CD45RO-CD27- T cells, MFI for z was 128 in patients vs. 527 in NC (p=0.0001).

Discussion: The results indicate that although the content of antigen-primed effector T cells (CD8+CD45RO-CD27-) in the circulation was higher in HNC patients than in NC, these T cells were anergized. This is consistent with the hypothesis that functional defects in effector T cells of cancer patients might contribute to disease progression and that immunotherapy needs to be targeted toward the reversal of these defects.

CD22-DIRECTED IMMUNOTHERAPY OF LYMPHOMA

John P. Leonard, M.D., Center for Lymphoma and Myeloma and Division of Hematology/Oncology, Weill Medical College of Cornell University and New York Presbyterian Hospital, New York, New York

Monoclonal antibody therapy with rituximab, directed against the CD20 antigen, has become a widely accepted therapeutic modality for non-Hodgkin’s lymphoma. Alternative targets for immunotherapy, utilizing potentially different mechanisms of action, offer the possibility of activity in rituximab-resistant patients as well as the opportunity for synergy with rituximab and/or chemotherapy. The CD22 antigen is expressed on normal and malignant B-cells, with a distribution similar to that of CD20. Epratuzumab (LL2) is a humanized IgG1 monoclonal antibody, which targets the CD22 antigen, and is under evaluation as an naked antibody as well as a radioimmunotherapy for lymphoma. We have conducted a phase I/II trial with this agent, and dose limiting toxicity has not been observed with doses of 120 – 1000 mg/m2/weekly for 4 treatments. Virtually all toxicities were grade 1, primarily infusion reactions such as fevers, rigors, and hypotension, which are infrequent despite an administration time of 30-60 minutes. Depletion of B cells is observed in some patients, with no change in hematologic parameters, blood chemistries or serum immunoglobulins. Measurable antibody levels are present in the serum 3-4 months after completion of therapy and immunogenicity appears to be rare. Anti-tumor activity has been observed in patients with follicular NHL and in diffuse large B-cell NHL. Six of 13 follicular patients treated at the optimal dose levels (240 mg/m2/week or greater) achieved objective responses, 3 of which were complete and extended 1-2 years. In a heavily pretreated group of patients with relapsed diffuse large B cell NHL (median age 60, 65% with elevated LDH, median 3 prior regimens), 5 of 22 (23%) achieved objective responses, three of which were complete, one ongoing over 3 years. Several relapsed patients have undergone retreatment, some with evidence of a second response. Evaluation of additional patients is ongoing. Epratuzumab is currently under investigation in patients with rituximab-refractory indolent NHL as well as in combination with rituximab in follicular NHL and diffuse large B-cell NHL.

Humanized Anti–HLA-DR Monoclonal Antibody Hu1D10 (Remitogen) for the Treatment of Relapsed Non-Hodgkin’s Lymphoma: A Phase I Trial

Brian K. Link, M.D.
Holden Comprehensive Cancer Center University of Iowa

Purpose: A phase I trial was conducted to determine the safety, pharmacokinetics, and antitumor activity of the humanized antibody Hu1D10, which recognizes a polymorphic determinant of human leukocyte antigen (HLA)-DR, in patients with B-cell lymphoma.
Patients and Methods: Patients with 1D10+ relapsed B-cell lymphoma received 4 weekly 2-hour infusions of Hu1D10 at 4 dose levels: 0.15, 0.5, 1.5, and 5 mg/kg. A daily crescendo regimen (1.5, 3.5, 5, 5, and 5 mg/kg/day on 5 consecutive days) was tested in 6 additional patients. Patients were monitored for toxicity, pharmacokinetics, and tumor response. Two responders were evaluated for evidence of an active humoral immune response to their lymphoma.
Results: Twenty patients were treated; all were evaluable for safety and 18 for response. The weekly regimen was associated with transient, grade 1/2, infusion-related adverse events, most commonly fever, chills/rigors, hypotension, nausea/vomiting, headache, rash/urticaria, flushing, and myalgia. Grade 3/4 adverse events were infrequent at doses £ 1.5 mg/kg and increased in frequency at higher doses. The median terminal serum half-life of Hu1D10 was 11 days. The daily dosing regimen was poorly tolerated. Five objective tumor responses were documented among patients with follicular lymphoma treated on the weekly schedule. Responses occurred late, 105 to 296 days posttreatment, and 4 responses have continued to improve at 15 months posttreatment. One of 2 responders demonstrated autologous antilymphoma immunoglobulin G in serum on Day 107.

Anti-Cytotoxic T-Lymphocyte-Associated Antigen-4 (anti-CTLA-4) Monoclonal Antibody (MDX-CTLA-4) in Patients with Metastatic Melanoma

Hodi FS; Adult Oncology, Dana-Farber Cancer Institute, Boston, MA 02115

The use of monoclonal antibody directed CTLA4 blockade to enhance the immune mediated rejection of tumors has been previously studied in murine models. CTLA4 blockade enhances the rejection of B7-transfected tumors and non-transfected immunogenic colon carcinoma cells. Striking therapeutic synergies between CTLA4 blockade and GM-CSF secreting tumor cell vaccines have been observed in murine models of melanoma, breast cancer, and prostate cancer. The augmented anti-tumor effects have been associated with loss of tolerance to normal tissue antigens, manifested by vitiligo and prostatitis in the melanoma and prostate models respectively. Based on these pre-clinical data, we initiated a phase I clinical trial of single anti-CTLA4 antibody administration (3 mg/kg of MDX-CTLA4) in patients with metastatic melanoma. Preliminary results will be discussed.

Early-Phase Clinical Development of the Immunocytokine EMD 273066 (huKS-IL2): Laboratory and Safety Findings in Patients With Prostate Cancer

Finke LH1, Haunschild J2, Kovar A3, Dahl T4, Gold D5, Redfern C5, Thompson T1, Weber R6, Bubley G7, Gillies S4
1. EMD Pharmaceuticals, Durham, NC, USA
2. Merck KGaA, Grafing, Germany
3. Merck KGaA, Darmstadt, Germany
4. Lexigen Pharmaceuticals, Lexington, MA, USA
5. Sidney Kimmel Cancer Center, San Diego, CA, USA
6. Saint Francis Memorial Hospital, San Francisco, CA, USA
7. Beth Israel Hospital, Harvard University, Boston, MA
Background: The immunocytokine EMD 273066 is a member of a novel class of biologics with dual antibody-cytokine anti-tumor mechanisms involving targeted delivery of cytokines to the tumor microenvironment and synergy between antibody and cytokine in triggering a cellular immune cascade directed against the tumor. EMD 273066 is composed of a humanized antibody directed against human adenocarcinoma-associated antigen genetically fused with 2 interleukin-2 (IL-2) molecules.
Objective: A Phase I study was conducted to evaluate the safety, biological pharmacodynamics, and pharmacokinetics of EMD 273066 in patients with advanced prostate cancer.
Methods: Twenty-two patients (7 ECOG 0 and 15 ECOG 1; mean age = 68.5 years) with advanced hormone-refractory prostate cancer were administered EMD 273066 in a dose-escalation protocol (0.35 mg/m2 to 8.5 mg/m2 via 4-hour, once-daily intravenous infusions for 3 days repeated every 28 days for a maximum of 4 cycles) in an open-label, 3-center trial.
Results: The maximum tolerated dose for EMD 273066 administered as a 4-hour infusion for 3 consecutive days was 6.4 mg/m2/day. At doses up to 4.3 mg/m2/day, adverse events were generally reported as mild or moderate (Grade 1-2). The most common adverse events were fatigue, fever, rash, and chills, all of which are characteristic of antibody and/or IL-2 exposure. Four patients (1 at 2.8 mg/m2/day, 1 at 6.4 mg/m2/day, and 2 at 8.5 mg/m2/day) experienced dose-related toxicities classified as Grade 3. No patient required treatment for an acute hypersensitivity or anaphylactic reaction to EMD 273066. Immunologic evaluations show that lymphocyte counts demonstrated known IL-2 effects of lymphopenia followed by lymphocytosis. Natural killer cell number increased after administration of EMD 273066. Immune-cell rebound was not dose-dependent above 2.8 mg/m2/day for up to 4 weeks. Pharmacokinetic evaluations show that EMD 273066 had a half-life of approximately 5 hours and a Cmax <1 hour after completed infusion. The concentration-time profiles were best described by a 1-compartment model.
Conclusions: EMD 273066 was well-tolerated above a level of demonstrable biological activity in these patients with advanced prostate cancer. Tolerability, biological activity, and pharmacokinetic data support further clinical exploration of EMD 273066. Studies optimizing exposure and feasibility via various drug schedules are ongoing.