J Clin Oncol 22: by American Society of Clinical Oncology INTRODUCTION

Transcription

1 VOLUME 22 NUMBER 10 MAY JOURNAL OF CLINICAL ONCOLOGY O R I G I N A L R E P O R T Effect of Dose on Immune Response in Patients Vaccinated With an HER-2/neu Intracellular Domain Protein Based Vaccine Mary L. Disis, Kathy Schiffman, Katherine Guthrie, Lupe G. Salazar, Keith L. Knutson, Vivian Goodell, Corazon dela Rosa, and Martin A. Cheever From the Tumor Vaccine Group, Oncology, University of Washington; Fred Hutchinson Cancer Research Center; and Corixa Corporation, Seattle, WA. Submitted September 3, 2003; accepted March 2, Supported by grants from the Cancer Research Treatment Foundation and the National Institutes of Health (NIH), National Cancer Institute grant No. U54- CA (M.L.D), by NIH training grant No. T32 (HL07093; L.G.S.), and by a fellowship from the Department of Defense Breast Cancer Program (DAMD ; K.L.K.). Patient care was conducted through the Clinical Research Center Facility at the University of Washington, which is supported through NIH grant No. MO1-RR Authors disclosures of potential conflicts of interest are found at the end of this article. Address reprint requests to Mary L. Disis, MD, Tumor Vaccine Group, Oncology, Box , University of Washington, Seattle, WA ; ndisis@u.washington.edu by American Society of Clinical Oncology X/04/ /$20.00 DOI: /JCO A B S T R A C T Purpose To evaluate the safety of an HER-2/neu intracellular domain (ICD) protein vaccine and to estimate whether vaccine dose impacts immunogenicity. Patients and Methods Twenty-nine patients with HER-2/neu overexpressing breast or ovarian cancer and with no evidence of disease after standard therapy received a low- (25 g), intermediate- (150 g), or high-dose (900 g) HER-2/neu ICD protein vaccine. The vaccine was administered intradermally, monthly for 6 months, with granulocyte-macrophage colony-stimulating factor as an adjuvant. Toxicity and both cellular and humoral HER-2/neu specific immunity was evaluated. Results The vaccine was well tolerated. The majority of patients (89%) developed HER-2/neu ICD-specific T-cell immunity. The dose of vaccine did not predict the magnitude of the T-cell response. The majority of patients (82%) also developed HER-2/neu specific immunoglobulin G antibody immunity. Vaccine dose did not predict magnitude or avidity of the HER-2/neu specific humoral immune response. Time to development of detectable HER-2/neu specific immunity, however, was significantly earlier for the highversus low-dose vaccine group (P.003). Over half the patients retained HER-2/neu specific T-cell immunity 9 to 12 months after immunizations had ended. Conclusion The HER-2/neu ICD protein vaccine was well tolerated and effective in eliciting HER-2/neu specific T-cell and antibody immunity in the majority of breast and ovarian cancer patients who completed the vaccine regimen. Although the dose of vaccine did not impact the magnitude of T-cell or antibody immunity elicited, patients receiving the highest dose developed HER-2/neu specific immunity more rapidly than those who received the lowest dose. J Clin Oncol 22: by American Society of Clinical Oncology INTRODUCTION The HER-2/neu oncogenic protein is a tumor antigen in patients with breast and ovarian cancer. 1 Multiple investigations have shown that some cancer patients, whose tumors overexpress the HER-2/neu oncoprotein, have both detectable antibody and T-cell immunity directed against HER-2/neu. 2-4 Furthermore, experiments in rodents suggest that the HER- 2/neu protein, particularly the intracellular domain (ICD), is a tumor rejection antigen. The generation of an immune response specific for neu ICD will provide animals some protection from a tumor challenge, 5 and T cells specific for the ICD will mediate an antitumor response when infused into animals bearing neu-overexpressing tumors. 6 For these reasons, several vaccine strategies have been developed to target the HER-2/neu oncogene and are being evaluated for safety and immunogenicity in phase I clinical trials. One of the most common methods of immunizing cancer patients against weakly immunogenic proteins expressed in their tumors is via peptide vaccination. Peptide vaccines are composed of fragments of the tumor antigen specifically designed to elicit 1916

2 HER-2/neu ICD Protein Based Vaccine HLA class I (cytotoxic T cell) and/or HLA class II (T helper) antigen-specific T-cell responses. Peptide vaccines targeting the HER-2/neu oncogenic protein have been studied extensively in phase I studies and have been found to be both safe and immunogenic. 7-9 However, there are disadvantages to the use of peptides as vaccine antigens. Development of peptide vaccines necessitates some preclinical assessment as to the most useful immune response desired, and the choice of epitope must be made before clinical studies. 10 Furthermore, peptides are designed for exogenous binding in major histocompatability antigen molecules; therefore, HLA restrictions are of utmost importance in predicting vaccine efficacy. Finally, HLA class I specific epitope vaccines, in particular, may require a concomitant T helper response to promote a durable cytotoxic T-cell response. 11 Whole protein vaccines are composed of both HLA class I and class II epitopes and are not subject to specific HLA restrictions to predict efficacy. However, little is known about the use of protein-based vaccines to elicit immunity to self tumor antigens, and unlike the case of peptide-based vaccines, 12 appropriate immunogenic doses of tumor protein vaccines have not been defined. The study described here is the first clinical trial of a recombinant HER-2/neu ICD protein based vaccine designed to stimulate HER-2/neu specific immunity in patients with HER-2/neu overexpressing breast and ovarian cancers. The goal of the study was to determine the safety of the approach as well as assess the immunogenicity of the vaccine. Patients were vaccinated in the adjuvant setting after optimal cytoreductive therapy. Recently, the clinical application of cancer vaccines is being focused on patients with minimal disease. Extrapolating from application of active immunization in infectious diseases, vaccines, such as those targeting chicken pox or influenza, are given in the absence of infection and have only limited efficacy if administered after exposure to the pathogen. The goal of vaccination against cancer may be to elicit significant immunologic memory capable of rapidly expanding an antigen-specific T-cell population in the presence of low levels of antigen encountered during cancer onset or early relapse. Indeed, vaccination against cancer may likely only be effective when disease is either absent or below the limit of detection. 13 Investigation of recombinant protein vaccines in infectious disease models suggests that the dose of protein administered affects the antigen-specific immune response elicited. Studies were undertaken to evaluate the effect of dose of a recombinant HER-2/neu ICD protein vaccine on the incidence and magnitude of HER-2/neu specific cellular and antibody immunity generated with active immunization. PATIENTS AND METHODS Vaccine Formulation and Administration The HER-2/neu ICD construct consisted of amino acids 676 to 1255 of the full-length HER-2/neu. The first 20 amino acids were unrelated to the ICD including an 8 histidine tag followed by a Ser-Ser-Gly linker and an enterokinase site. For this reason, proteins derived from Leishmania major and Mycobacterium tuberculosis were constructed in the same fashion and purified in a similar manner as the HER-2/neu ICD vaccine to be used as negative controls in evaluating immunogenicity. The HER-2/neu ICD protein was produced in Escherichia coli (pet 28 vector; Novagen, Madison, WI; BLRplyS strain) as inclusion bodies by the Alberta Research Council (Edmonton, Canada). The inclusion body pellets were purified, and the final product was frozen at 20 C until formulated (Corixa Corp, Seattle, WA). Disulfide bond formation was minimized by first reducing with 10 mmol/l dithiothreitol before purification followed by a second reduction before dialysis in the final buffer. The material was then stored at 20 C to prevent further oxidation. The material was not chemically modified. The molecular weight of the final product was approximately 80 kd. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of the final product showed only one band at 80 kd. The final formulation consisted of 500 g HER-2/neu ICD in 1 ml of 20 mmol/l Tris (ph 9.0), with 5% mannitol and 0.05% Tween-80. Patients were enrolled sequentially to receive intradermal vaccinations monthly for 6 months with 25 g (low dose), 150 g (intermediate dose), or 900 g (high dose) of HER-2/neu ICD protein admixed with 100 g of granulocytemacrophage colony-stimulating factor (supplied by Immunex Corp, Seattle, WA) as an adjuvant. The vaccines were administered to the same draining lymph node site for each patient. All subjects received a tetanus toxoid (tt) booster vaccination, as a positive control, within 2 weeks before beginning immunization with the HER-2/neu vaccine. Blood was obtained for assessment of immunologic responses and toxicity evaluations at baseline, 1 month after the first, third, fifth, and sixth vaccinations, 2 to 3 months after completion of the sixth vaccination, and then, if possible, every 3 months for 1 year. Patient Population Patients with stage II, III, or IV HER-2/neu overexpressing breast cancer with no evidence of disease after standard therapy or stage IIC or III HER-2/neu overexpressing ovarian cancer in complete remission were eligible for enrollment onto a phase I study of a HER-2/neu ICD protein based vaccine between April 2000 and January The study was approved by both the United States Food and Drug Administration and the University of Washington Institutional Review Board. To be enrolled, patients had to meet the following criteria: documentation of HER- 2/neu overexpression in their primary tumor or metastasis; Karnofsky performance status 90%; WBC count 3,000/mL and platelet count more than 100,000/mL; normal renal and hepatic function; fully treated for their disease by conventional standards and in clinical complete remission; and off chemotherapy and other immune-suppressive drugs, such as steroids, for a minimum of 30 days before starting the vaccination series. The clinical characteristics of the patient population are listed in Table 1. Evaluation of T-Cell Immunity After Vaccination Antigen-specific T-cell proliferation was measured using a modified limiting dilution assay that has been previously described. 12 Results are reported as a stimulation index (SI), defined as the mean of 24 experimental wells divided by the mean of 24 control wells. Phytohemagglutinin incubated with patient T cells at a concentration of 5 g/ml was used as a positive control to assess the ability of T cells to respond to mitogen and resulted in an

3 Disis et al Table 1. Patient Characteristics for Patients at Each Dose Level Dose of HER-2/neu ICD Protein (No. of patients) Characteristics Low Intermediate High Diagnosis Breast Stage II Stage III Stage IV (CR) Ovarian IIC Total patients per dose Age, years Median Range Time from last chemotherapy, months Median Range Abbreviations: ICD, intracellular domain; CR, complete remission. SI greater than 2 in all cases (data not shown). Patients peripheralblood mononuclear cells were evaluated for immunologic response against the following antigens; recombinant HER-2/neu domain proteins, extracellular domain (ECD) and ICD (1 g/ ml), 9 tt (5 g/ml) evaluated as a vaccinated control antigen, and proteins derived from L major and M tuberculosis (1 g/ml) served as negative control protein antigens for an E coli and vectorrelated response, having been produced under identical manufacturing methods as the HER-2/neu ICD protein. In addition, because the negative control proteins were manufactured in an identical vector as the HER-2/neu construct, it is likely any immunogenicity elicited by the components involved in purification, such as histidine, would also be detected to these constructs. A reference population of 30 age-matched volunteer donors was used to establish baseline response to the HER-2/neu antigens. 9 The mean and standard deviation (SD) of this population was g/ml for HER-2 ICD and g/ml for HER-2 ECD. A subset of 10 volunteer donors was assessed for the control proteins, and the mean and SD of this population was g/ml for L major and g/ml for M tuberculosis.ansi of more than 2 to antigen defined a positive response (mean and two SDs greater than the reference population). If subjects had an SI of greater than 2 to the HER-2/neu ICD before vaccination, a postvaccination T-cell response was defined as positive if the value was twice the prevaccine value. Patients with a pre-existent immune response to the HER-2/neu protein had to meet this requirement to be included as responders in the analysis. Evaluation of Antibody Immunity After Vaccination Antibody immunity to HER-2/neu protein was assessed as previously described using a human breast cancer cell line that overexpresses HER-2/neu as a source of HER-2/neu protein. 14 Some patients did have pre-existent detectable antibody immunity to the HER-2/neu protein and were only considered to have responses if they boosted the antibody level to twice the baseline value. Pre- and postvaccination samples for each patient were analyzed simultaneously on the same plate. A positive sample was defined as an antibody concentration greater than the mean of a volunteer blood donor population two SDs ( g/ ml, n 200). The specificity of the assay was 78%, and the sensitivity was 90%. The assay was linear over the range of antibody concentrations tested and had a correlation coefficient of On the basis of repeated measures of 20 volunteer and 20 experimental sera over a 6-month period, the intra- and interassay coefficients of variation were 15% and 9%, respectively. Positive responses were verified by Western blot analysis. 15 Antibody responses to E coli were also measured as a control. Briefly, 96-well Immulon 4HBXmicrotiter plates (Dynex Technologies, Inc, Chantilly, VA) were coated with E coli lysates diluted (1:50) in carbonate buffer, alternating rows with carbonate buffer alone. Diluted purified human immunoglobulin (Ig) G (Sigma Chemical Co, St Louis, MO) provided a standard curve on each plate. Treated plates were incubated overnight at 4 C. All wells were then blocked with 100 L/well of 10% phosphate-buffered saline (PBS)-1% bovine serum albumin solution (Sigma) and incubated at room temperature with agitation for 1 to 2 hours. Plates were washed four times with 10% PBS and 0.5% Tween-20 before the addition of patient sera at dilutions of 1:100, 1:200, 1:400, and 1:800. Sera were diluted with 10% PBS, 1% FCS, 1% bovine serum albumin, and 25 g/ml mouse IgG (Sigma), 50 L/well, and incubated at room temperature on a rocker for 1 hour. After serum incubation, plates were washed four times as described above, and goat antihuman IgG-horseradish peroxidase conjugate (diluted 1:100,000; Zymed Laboratories, San Francisco, CA) was added, 50 L/well, and incubated for 45 minutes at room temperature. After a final wash, TMB (Kirkegaard and Perry Laboratories, Gaithersburg, MD) developing reagent was added (75 L/well), and color reaction was read at 640 nm until the well containing the standard at a concentration of 0.16 g/ml read 0.3 optical density (OD). The reaction was then stopped with 75 L/well of1nhcl and read at 450 nm. Pre- and postimmunization samples were assayed on the same plate, and patients were considered to have boosted E coli protein specific antibodies if the mean of four replicates of their postvaccine sample was two SDs above the mean of four replicates of their prevaccine sample. Evaluation of Antibody Avidity After Vaccination Assessment of HER-2/neu specific antibody avidity is based on a modification of previously described methods. 16 Two-fold serial dilutions of serum samples were plated in duplicate and incubated for 3 hours at room temperature. After four washes, 200 L of wash buffer (described earlier) containing 8 mol/l urea was added to one of the duplicate wells and incubated at room temperature for 3 minutes. All wells were washed four times. For the detection of antibody binding the plates, goat antihuman IgG conjugated with horseradish peroxidase (diluted 1:10,000; Zymed Laboratories) was added for 45 minutes at room temperature. After washing, 3,3,5,5 -tetramethylbenzidine substrate solution (Kirkegaard & Perry Laboratories) was added, and the reaction was allowed to proceed for 3 to 5 minutes and was stopped by adding 75 L of 1 N HCl per well. For a given patient serum dilution, the avidity index (AI) was calculated using the following formula: (change of OD with urea/change of OD without urea) 100; and the AI was expressed as a percentage. An identical assay was developed to evaluate tt-specific antibodies, with the modification of tt (1 g/ml) plated in alternating rows with carbonate buffer. An AI less than 30% is considered low avidity, 31% to 50% is considered intermediate avidity, and greater than 51% is considered high avidity. 16 Avidity is reported here at the 1:25 dilution of sera JOURNAL OF CLINICAL ONCOLOGY

4 HER-2/neu ICD Protein Based Vaccine Statistical Analysis To estimate the significance of differences in median immune responses, the data were compared using a two-tailed Student s t test, with a level of significance set at 0.05 (GraphPad InStat v.3.05; GraphPad Software, San Diego, CA). The probability of developing detectable immunity was summarized using a cumulative incidence estimate. 17 RESULTS The HER-2/neu ICD Protein Based Vaccine Was Well Tolerated at All Doses Twenty-nine subjects were enrolled onto the trial; 10 were enrolled at the lowest vaccine dose (25 g), 10 were enrolled at the intermediate vaccine dose (150 g), and nine were enrolled at the highest vaccine dose (900 g) of HER- 2/neu ICD protein. Twenty-seven of the 29 subjects enrolled completed all six vaccinations; two subjects who were enrolled at the lowest dose withdrew from study because of progressive disease after receiving only the first vaccination. The ages of patients, stages of disease, and time from completing prior chemotherapy were similar for the subjects in each group (Table 1). Adverse events are reported on all 29 subjects, and immune responses are reported on the 27 subjects who completed six vaccinations (eight for the lowdose, 10 for the intermediate-dose, and nine for the highdose group). The vaccine was well tolerated at all doses. There were no grade 2, 3, or 4 adverse events. Three subjects, one at each dose level, reported grade 1 fatigue; three subjects, two in the low-dose group and one in the high-dose group, reported grade 1 myalgia; and 24 subjects, seven in the lowdose group, eight in the intermediate-dose group, and nine in the high-dose group, experienced grade 1 local injection site erythema. Majority of Patients Immunized With an HER-2/ neu ICD Protein Based Vaccine Developed HER- 2/neu Specific T-Cell Immunity The majority of patients (24 of 27 patients, 89%) who completed all six vaccines developed HER-2/neu ICD specific T-cell immunity with active immunization regardless of the vaccine dose. Eight (100%) of eight patients who received the lowest dose of HER-2/neu ICD protein (Fig 1A), nine (90%) of 10 patients who received the intermediate dose (Fig 1B), and seven (78%) of nine patients who received the highest dose of the vaccine (Fig 1C) developed T-cell immunity to the HER-2/neu ICD protein domain. All responses obtained during the duration of the trial for each group are shown. Figure 2 demonstrates the temporal development of immunity to the ICD during the time of active immunization and immediately after immunization in the low-dose (Fig 2A), intermediate-dose (Fig 2B), and high-dose (Fig 2C) populations. Two patients in the highdose group had a pre-existent immune response to the HER-2/neu ICD protein (one patient had an SI of 3.5 and one had an SI of 4.4). Both patients boosted immunity during the course of vaccination, one at a distant time point. None of the patients developed a significant immune response to the L major or M tuberculosis proteins, indicating the specificity of the HER-2/neu ICD response to the HER- 2/neu protein and not to any potentially contaminating proteins residual to the vaccine production process (Fig 1). All patients developed significant immunity to tt, the control vaccination (Fig 1). Those patients with a pre-existent detectable T-cell response to tt boosted immunity over twice baseline after the control immunization, indicating the immune competence of the study population. Only one (4%) of 27 patients developed immunity to the HER-2/neu ECD, which demonstrates epitope spreading (Fig 1). 9 Dose of Protein in the Vaccine Does Not Predict the Magnitude of HER-2/neu Specific T-Cell Immunity Elicited With Active Immunization The magnitude of the T-cell response to HER-2/neu ICD after immunization was similar at all doses (Fig 3). The median SI for patients receiving the low-dose vaccine was 3.4 (range, 2.2 to 9.2), the median SI for patients receiving the intermediate-dose vaccine was 4.4 (range, 1.4 to 15.8), and the median SI for patients receiving the high-dose vaccine was 5.7 (range, 1.6 to 15.7). There was no statistical difference in the median responses of each group (low v intermediate, P.41; intermediate v high, P.79; and low v high, P.32). Higher Dose of a HER-2/neu ICD Protein Based Vaccine Was Associated With Earlier Development of Detectable HER-2/neu Specific T-Cell Immunity Although the dose of HER-2/neu ICD protein based vaccine was not associated with a difference in the magnitude of HER-2/neu specific T-cell or antibody immunity, the higher dose vaccine was associated with a more rapid development of detectable immunity (Fig 4). The percentage of patients achieving a HER-2/neu ICD specific T-cell response, as defined by an SI of more than 2.0, is shown over the time of active immunization as a cumulative incidence curve for each of the dose groups (Fig 4). The time to the development of a detectable T-cell immune response was significantly earlier for the group receiving the higher dose vaccine compared with the lowest dose vaccine (P.003). HER-2/neu Specific T-Cell Immunity Persists After Active Immunizations Have Ended The majority of the patients were able to be observed for persistent immunity after their vaccine regimen was completed. There was no difference in the persistence of immunity between any of the dose groups. Figure 5 shows HER-2/neu ICD T-cell immunity persisting as long as 1 year from the end of active immunization. At 1 month after the

5 Disis et al Fig 1. The majority of patients who were immunized developed T-cell immunity. T-cell responses, (A) low-dose, (B) intermediate-dose, and (C) high-dose groups. Preimmunization (E) and maximal response during immunization (F). The dashed line is the mean and two standard deviations of the reference population for all antigens except tt. SI, stimulation index; ICD, intracellular domain; ECD extracellular domain; tt, tetanus toxoid; Leish, Leishmania major; Mtb, Mycobacterium tuberculosis. vaccinations were finished, 18 (72%) of 25 assessable patients had detectable HER-2/neu ICD T-cell immunity (median SI, 2.6; range, 1.0 to 7.6). Six months after the end of immunizations, 12 (55%) of 22 assessable patients had detectable HER-2/neu ICD T-cell immunity (median SI, 1.9; range, 0.5 to 15.7). Finally, between 9 months and 1 year after the end of the vaccine regimen, eight (53%) of 16 assessable patients still had detectable HER-2/neu ICD T-cell immunity (median SI, 2.6; range, 0.4 to 14.0). Majority of Patients Immunized With an HER-2/neu ICD Protein Based Vaccine Develop HER-2/neu Specific Antibody Immunity The majority of patients (22 of 27 patients, 82%) who completed all six immunizations developed HER-2/neu specific IgG antibody immunity. Results are shown as the HER-2/neu antibody level determined 1 month after the last vaccine. Figure 6A demonstrates that five (63%) of eight patients in the low-dose group developed HER-2/neu antibody immunity. One patient had a pre-existent HER-2/ neu specific antibody response (3.0 g/ml, which boosted to 6.0 g/ml after vaccination). Ten (100%) of 10 intermediate-dose patients developed new HER-2/neu specific IgG antibody immunity (Fig 6B). Finally, seven (78%) of nine patients developed HER-2/neu specific antibody immunity after receiving the highest dose of vaccine (Fig 6C). One patient had a significant HER-2/neu antibody response of 6.0 g/ml before receiving any immunizations, which decreased over the course of immunization to 0.9 g/ml (a negative response). No patient had boosted antibody immunity to E coli lysate during the course of the immunizations. The mean E coli antibody response ( OD at 450 nm) at a 1:5000 sera dilution before immunization 1920 JOURNAL OF CLINICAL ONCOLOGY

6 HER-2/neu ICD Protein Based Vaccine Fig 2. Intracellular domain T-cell immunity over the time of active immunization. T-cell immunity as stimulation index (SI) for each patient at each time. (A) Low-dose, (B) intermediate-dose, and (C) high-dose groups. The dashed line is the mean and two standard deviations of the reference population. was The mean E coli antibody response ( OD at 450 nm) after the end of active immunizations at a sera dilution of 1:5000 was (P.111). HER-2/neu specific antibody responses were documented by Western blot analysis, and examples are shown in Figure 7. Figure 7A shows antibodies directed against HER-2/neu, which were immunoprecipitated from SKBR3, a human breast cancer cell line that overexpresses HER-2/neu, in a breast cancer patient who developed HER-2/neu antibodies detected by enzymelinked immunosorbent assay. 15 Figure 7B is a similar immunoblot from a patient who did not develop detectable antibodies after active immunization. The dose of protein in the vaccine did not predict the magnitude or avidity of HER-2/neu specific antibody immunity elicited with active immunization. In the low-dose group, the median antibody response was 1.6 g/ml (range, 0 to 9.3 g/ml); in the intermediate-dose group, it was 2.5 g/ml (range, 1.5 to 7.6 g/ml); and in the highdose group, it was 2.3 g/ml (range, 0.58 to 5.42 g/ml). There was no statistical difference in the median response between any group (low v intermediate, P.41; intermediate v high, P.29; or low v high, P.96). Of the 22 patients who developed HER-2/neu specific antibody immunity, 17 (77%) of 22 demonstrated low-avidity antibody immunity (range, 0% to 29%; Fig 3D), three (14%) of 22 demonstrated intermediate-avidity antibody immunity (range, 38% to 43%), and two (9%) of 22 demonstrated high-avidity antibody immunity (64% and 65%). The intermediate- and high-avidity responses were detected in patients at all dose levels (Fig 6D). All patients who completed the study had high-affinity tt antibodies (data not shown). The patients with intermediate- and high-avidity HER-2/neu specific antibodies were able to be observed for persistence of T-cell immunity over time, and all five patients had persistent HER-2/neu specific T-cell immunity at 12 months after vaccination. DISCUSSION The study described here is the first report of a clinical trial evaluating an HER-2/neu protein based vaccine and is also one of the first evaluations of the effect of the dose of a tumor protein antigen vaccine on the immune response elicited in cancer patients. The data presented demonstrate

7 Disis et al Fig 3. Dose of protein did not predict magnitude of HER-2/neu specific T-cell immunity elicited. Maximal intracellular domain (ICD) responses are shown for the low, intermediate, and high doses. Solid bars denote median value. The dashed line is the mean and two standard deviations of the reference population. SI, stimulation index. Fig 5. HER-2/neu specific T-cell immunity persists after immunizations have ended. T-cell immunity 1, 6, and 9 to 12 months after immunizations. The dashed line depicts the mean and two standard deviations of the reference population response. Solid bars denote the median value. SI, stimulation index. the following: the majority of patients who were immunized developed HER-2/neu protein specific T-cell immunity, and the dose of the vaccine did not impact the magnitude of the antigen-specific T-cell response; the majority of patients who were immunized developed HER-2/neu protein specific antibody immunity, and the dose of the vaccine did not impact the magnitude or avidity of the antigen-specific antibody response; the highest dose of vaccine was associated with earlier development of detectable HER-2/neu specific T-cell immunity; and tumor antigen specific immunity persisted after active immunization had ended. The HER-2/neu ICD protein based vaccine elicited both cellular and humoral immunity in the majority of patients. Preclinical development of HER-2/neu protein based vaccines indicates that the ICD may be a better immunogen for the generation of HER-2/neu specific immunity than the ECD. The HER-2/neu oncogenic protein is a Fig 4. Higher dose was associated with earlier development of detectable HER-2/neu specific T-cell immunity. Time (months) to development of detectable HER-2/neu specific T-cell immunity for the percentage of patients receiving the highest dose (solid black line), intermediate dose (gray dashed line), and low-dose (black dotted-dashed line). ICD, intracellular domain. self protein, and in animal models, vaccination using the whole protein or the ECD portion of neu to stimulate immunity was found to be ineffective secondary to immunologic tolerance. 18,19 Rodents immunized with the ICD, however, can develop significant neu-specific T-cell responses. 5 Inability to immunize to the ECD may be a result of the extracellular nature of this portion of the HER-2/neu protein, and thus, previous exposure to the immune system results in tolerance. Because the ICD is sequestered from the immune system by its intracellular location, it is potentially more immunogenic. Furthermore, T-cell immunity directed against the HER-2/neu ICD has been shown to mediate an antitumor effect in animal models of neuoverexpressing breast cancer. 5,8 Of note, although the T-cell immunity detected was specific for HER-2/neu ICD, because of the amount of clinical material available for analysis, it is unknown what component of that response is CD8 versus CD4 HER-2/neu specific T cells. Peptidebased vaccines are specifically designed to elicit cellular immunity. 10 Vaccines formulated from protein, however, are successful in eliciting both humoral and cellular immunity. 20 Studies in animals have demonstrated that rejection of HER-2/neu overexpressing tumors after vaccination may require both tumor-specific T cells and antibodies. 21 Although antibodies to the ICD protein may not have the direct antitumor effects associated with antibodies specific for the ECD, such as trastuzumab, antibodies binding to ICD protein present in a tumor bed or shed by dying cells may serve to amplify the cellular immune response via induction of antibody-dependent cell-mediated cytotoxicity. Thus, immunizing cancer patients, whose tumors overexpress HER-2/neu, with an HER-2/neu protein based vaccine may be an effective method to elicit both tumor JOURNAL OF CLINICAL ONCOLOGY

8 HER-2/neu ICD Protein Based Vaccine Fig 6. The majority of patients who were immunized developed antibody immunity. HER-2/neu antibodies before (E) and after (F) immunization: (A) low dose, (B) intermediate dose, and (C) high dose. Dashed line is the mean and two standard deviations of controls. (D) Antibody avidities ( ) of dose groups (1:25 dilution). IgG, immunoglobulin G; ICD, intracellular domain

9 Disis et al Fig 7. Antibody immunity to HER-2/neu. (A) Immunoblot of a positive response. Lanes A and B are post- and prevaccination sera against immunoprecipitated HER-2/neu, respectively; lane C is the molecular weight marker. (B) Immunoblot of a negative response. Lane A is the molecular weight marker; lane B is preimmunization sera, and lane C is postimmunization sera against immunoprecipitated protein. specific T-cell and antibody immunity in the majority of vaccinated individuals. Potentially, the dose of a protein-based vaccine can impact the development of an antigen-specific immune response in many ways. First, some studies in infectious disease systems have suggested that higher doses of recombinant protein vaccines are associated with more robust antigen-specific T-cell responses. 22 A trial performed in animals, immunizing against a herpesvirus, demonstrated higher herpes-specific T-cell proliferative responses in those animals receiving the 40- g versus the 20- g dose of recombinant protein-based vaccine. In the study reported here, the magnitude of the T-cell response was not dependent on the dose. Second, several trials of infectious disease vaccines have suggested that higher doses of protein result in more subjects developing antigen-specific immunity. An evaluation of a recombinant human papillomavirus protein vaccine in volunteers suggested that a 533- g dose was superior to a 128- gor26- g dose in generating interferon gamma producing human papillomavirus specific T cells in the greatest number of vaccinated subjects. 23 Similarly, higher doses of a modified tick-borne encephalitis vaccine resulted in a greater number of seroconversions in volunteers. 24 Unlike these foreign antigen vaccines, the dose of the HER-2/neu ICD protein based vaccine did not influence the number of patients developing detectable immunity. The lowest dose and highest dose were equally effective in stimulating HER-2/neu specific antibody and T-cell responses. Finally, key to the development of any vaccine is the ability of immunization to elicit immunologic memory. Memory responses are particularly important for cancer vaccines that may have the greatest clinical utility in preventing disease relapse months to years after primary therapy has been completed. The development of T-cell memory responses after active immunization has been associated with higher antigen doses in vaccines. 25 Similarly, highavidity antigen-specific antibody immunity, a potential predictor of immunologic memory, has been associated with high-dose vaccines. In the current study, there was no difference in the persistence of T-cell immunity after active immunizations had ended in the low-, intermediate-, or high-dose groups receiving the HER-2/neu ICD vaccine. Furthermore, intermediate- and high-avidity HER-2/neu specific antibody responses were found in all dose groups. The dose of HER-2/neu ICD protein based vaccine did impact the length of time needed to develop a detectable HER-2/neu specific T-cell response. HER-2/neu specific T-cell immunity could be detected within 2 months of starting the vaccine regimen in the majority of patients receiving the highest dose of vaccine compared with 6 months in the majority of patients receiving the lowest dose vaccine. The more rapid induction of measurable immunity with the highest dose of protein may have importance in the clinical application of cancer vaccines. For example, if a vaccine was being developed for a cancer that was associated with rapid time to relapse after primary therapy, a higher dose of a protein vaccine may allow a more rapid induction of potentially protective immunity. Furthermore, for vaccine strategies applied in the neoadjuvant setting where definitive therapy cannot be delayed for the extended period of time needed for multiple immunizations, higher dose protein-based vaccines may allow induction of tumor antigen specific immunity in a shorter period of time. Recombinant protein-based vac JOURNAL OF CLINICAL ONCOLOGY

10 HER-2/neu ICD Protein Based Vaccine cines are much more expensive to formulate than peptide- or plasmid-based vaccines. Therefore, it is important to choose clinical situations where the higher, more expensive dose would be preferred to the lower, less expensive dose. HER-2/neu, particularly the ICD portion of the protein, is a biologically relevant tumor antigen that has become a model for testing a variety of vaccination strategies from peptides, 7,9 to dendritic cells, 26 to plasmidbased vaccines. 27 Immunizing cancer patients with recombinant HER-2/neu protein has several advantages when compared with other vaccine formulations; the vaccine is immunogenic across all HLA types, can elicit immunity in the majority of patients vaccinated, can stimulate both tumor-specific cellular and humoral immunity, and will result in the development of immunologic memory. Detailed evaluations of the immunogenicity of cancer vaccines in the setting of phase I trials will better define the clinical utility and application of tumorspecific immunization. Acknowledgment We are grateful for the nursing care of Donna Davis, BSN, and the assistance in manuscript and data preparation by Chalie Livingston and Robert Schroeder. We also thank all the patients who agreed to participate in this study. Authors Disclosures of Potential Conflicts of Interest The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Owns stock (not including shares held through a public mutual fund): Martin A. Cheever, Corixa. Served as an officer or member of the Board of a company: Martin A. Cheever, Corixa. REFERENCES 1. Disis ML, Knutson KL, Schiffman K, et al: Pre-existent immunity to the HER-2/neu oncogenic protein in patients with HER-2/neu overexpressing breast and ovarian cancer. Breast Cancer Res Treat 62: , Scardino A, Gross DA, Alves P, et al: HER-2/neu and htert cryptic epitopes as novel targets for broad spectrum tumor immunotherapy. J Immunol 168: , Bremers AJ, Andreola S, Leo E, et al: T cell responses in colorectal cancer patients: Evidence for class II HLA-restricted recognition of shared tumor-associated antigens. Int J Cancer 88: , Disis ML, Pupa SM, Gralow JR, et al: High titer HER-2/neu protein specific antibody immunity can be detected in patients with early stage breast cancer. J Clin Oncol 15: , Foy TM, Bannink J, Sutherland RA, et al: Vaccination with Her-2/neu DNA or protein subunits protects against growth of a Her-2/neuexpressing murine tumor. Vaccine 19: , Knutson KL, Disis ML: Expansion of HER2/neu-specific T cells ex vivo following immunization with a HER-2/neu peptide-based vaccine. Clin Breast Cancer 2:73-79, Zaks TZ, Rosenberg SA: Immunization with a peptide epitope (p ) from HER-2/ neu leads to peptide-specific cytotoxic T lymphocytes that fail to recognize HER-2/neu tumors. Cancer Res 58: , Knutson KL, Schiffman K, Disis ML: Immunization with a HER-2/neu helper peptide vaccine generates HER-2/neu CD8 T-cell immunity in cancer patients. J Clin Invest 107: , Disis ML, Gooley TA, Rinn K, et al: Generation of T-cell immunity to the HER-2/neu protein after active immunization with HER-2/neu peptide-based vaccines. J Clin Oncol 20: , Disis ML, Knutson KL, McNeel DG, et al: Clinical translation of peptide-based vaccine trials: The HER-2/neu model. Crit Rev Immunol 21: , Knutson KL, Schiffman K, Cheever MA, et al: Immunization of cancer patients with a HER- 2/neu, HLA-A2 peptide, p , results in short-lived peptide-specific immunity. Clin Cancer Res 8: , Disis ML, Grabstein KH, Sleath PR, et al: Generation of immunity to the HER-2/neu oncogenic protein in patients with breast and ovarian cancer using a peptide-based vaccine. Clin Cancer Res 5: , Finn OJ: Cancer vaccines: Between the idea and the reality. Nat Rev Immunol 3: , Ward RL, Hawkins NJ, Coomber D, et al: Antibody immunity to the HER-2/neu oncogenic protein in patients with colorectal cancer. Hum Immunol 60: , Disis ML, Calenoff E, McLaughlin G, et al: Existent T cell and antibody immunity to HER-2/ neu protein in patients with breast cancer. Cancer Res 54:16-20, Goldblatt D, Vaz AR, Miller E: Antibody avidity as a surrogate marker of successful priming by Haemophilus influenzae type b conjugate vaccines following infant immunization. J Infect Dis 177: , Gooley TA, Leisenring W, Crowley J, et al: Estimation of failure probabilities in the presence of competing risks: New representations of old estimators. Stat Med 18: , Disis ML, Gralow JR, Bernhard H, et al: Peptide-based, but not whole protein, vaccines elicit immunity to HER-2/neu, an oncogenic selfprotein. J Immunol 156: , Bernards R, Destree A, McKenzie S, et al: Effective tumor immunotherapy directed against an oncogene-encoded product using a vaccinia virus vector. Proc Natl Acad Sci U S A 84: , Casal J, Tarrago D: Immunity to Streptococcus pneumoniae: Factors affecting production and efficacy. Curr Opin Infect Dis 16: , Reilly RT, Machiels JP, Emens LA, et al: The collaboration of both humoral and cellular HER-2/neu-targeted immune responses is required for the complete eradication of HER-2/ neu-expressing tumors. Cancer Res 61: , Martina BE, van de Bildt MW, Kuiken T, et al: Immunogenicity and efficacy of recombinant subunit vaccines against phocid herpesvirus type 1. Vaccine 21: , de Jong A, O Neill T, Khan AY, et al: Enhancement of human papillomavirus (HPV) type 16 E6 and E7-specific T-cell immunity in healthy volunteers through vaccination with TA- CIN, an HPV16 L2E7E6 fusion protein vaccine. Vaccine 20: , Ehrlich HJ, Pavlova BG, Fritsch S, et al: Randomized, phase II dose-finding studies of a modified tick-borne encephalitis vaccine: Evaluation of safety and immunogenicity. Vaccine 22: , Kundig TM, Bachmann MF, Oehen S, et al: On the role of antigen in maintaining cytotoxic T-cell memory. Proc Natl Acad Sci U S A 93: , Brossart P, Wirths S, Stuhler G, et al: Induction of cytotoxic T-lymphocyte responses in vivo after vaccinations with peptide-pulsed dendritic cells. Blood 96: , Disis M, Shiota F, McNeel D, et al: Soluble cytokines can act as effective adjuvants in plasmid DNA vaccines targeting self-tumor antigens. Immunobiology 207: ,