This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mazzaferro, V. Articles by Parmiani, G. Search for Related Content PubMed PubMed Citation Articles by Mazzaferro, V. Articles by Parmiani, G.

Vaccination with Autologous Tumor-derived Heat-Shock Protein Gp96 after Liver Resection for Metastatic Colorectal Cancer¹

Units of Hepatobiliary and Gastro-Pancreatic Surgery [V. M., J. C., M. S., E. R.], Immunotherapy of Human Tumors [M. G. C., L. R., G. P.], Biomedical Statistics [L. M., T. C.], Radiology [A. M.], and Pathology [S. A.], Istituto Nazionale per lo Studio e la Cura Tumori, 20133 Milan, Italy; Sigma-Tau S.p.A., 00040 Rome, Italy [R. C., M. C.]; Antigenics Inc., New York, New York 10111 [J. J. L.]; and Department of Microbiology and Immunology, University of Connecticut, Farmington, Connecticut 06030 [P. K. S.]

ABSTRACT

Purpose: Heat shock proteins (HSP) from tumor cells contain the gp96 polypeptide associated with cancer-specific antigenic peptides. Mice that are immunized with HSP/peptide-complex (HSPPC) derived from cancer tissue reject tumor from which HSPs are purified. We tested in humans whether vaccination with HSPPC-gp96 (Oncophage) from autologous liver metastases of colorectal carcinoma induces cancer-specific T-cell responses in patients rendered disease free by surgery.

Experimental Design: Twenty-nine consecutive patients underwent radical resection of liver metastases [Memorial Sloan-Kettering Cancer Center (MSKCC) score 1–3 (good prognosis), 18 patients; score 4–5 (bad prognosis), 11 patients] and received autologous tumor-derived HSPPC-96. Two vaccine cycles were administered (four weekly injections followed by four biweekly injections after 8 weeks). Class-I HLA-restricted, anti-colon cancer lines T-cell response was measured by ELISPOT assay on peripheral blood mononuclear cells (PBMCs) obtained before and after vaccination. Feasibility, safety, and possible clinical benefits were also evaluated.

Results: Either a de novo induced or a significant increase of preexisting class I HLA-restricted T-cell-mediated anti-colon cancer response was observed in 15 (52%) of 29 patients. Frequency of CD3+, CD45RA+, and CCR7- T lymphocytes increased in immune responders. No relevant toxicity was observed. As expected, patients with good prognosis had a significantly better clinical outcome than those with poor prognosis [2-year overall survival (OS), 89 versus 64%, P = 0.001; disease-free survival (DFS), 46 versus 18%, P = 0.001]. Patients with immune response had a statistically significant clinical advantage over nonresponding subjects (2-year OS, 100% versus 50%, P = 0.001; DFS, 51% versus 8%, P = 0.0001). Occurrence of immune response led to better tumor-free survival, whatever the predicted prognosis was (hazard ratio, 0.11–0.12 with/without stratification; P = 0.0012–0.0003).

Conclusions: HSPPC-96 vaccination after resection of colorectal liver metastases is safe and elicits a significant increase in CD8+ T-cell response against colon cancer. In this limited number of patients, two-year OS and DFS were significantly improved in subjects with postvaccination antitumor immune response, independently from other clinical prognostic factors.

INTRODUCTION

CRC,³ one the most common malignancies in humans, has frequent metastatic spread to the liver. Indeed, liver metastases are detectable in 15–25% of patients at the time of first diagnosis, progressing to about 60% of patients during their disease course (1 , 2) . Occurrence of liver metastases parallels the stage of primary CRC and represents the leading cause of death in advanced stages, with autopsy series showing liver metastases in more than 80% of patients who died from CRC (3) .

Surgical resection remains the primary treatment, although complete tumor removal from the liver is achieved in only 20–40% of the patients (4) resulting in a 5-year OS of 40% (1 , 5) and tumor recurrence of 59–81% according to different stages at the time of surgery (4 , 6) . A large retrospective analysis has identified several independent determinants of prognosis after liver resection of metastatic CRC; the size and number of tumor nodules, the clearance of surgical margins, the status of liver hilum lymph nodes; and the CEA level are the most significant ones (7) .

Adjuvant systemic or locoregional chemotherapy improves DFS and time-to-progression in some patient categories (1 , 8) . Subjects with CRC at high-risk of metastatic spread to the liver have also been treated with active nonspecific or specific immunotherapy (9, 10, 11) . One randomized trial that used a vaccine made of irradiated autologous tumor cells and Bacillus Calmette-Guérin after resection of the primary colon cancer suggested a potential benefit for stage II but not for stage III patients (10) .

T lymphocytes recognizing epitopes derived from CRC-associated proteins like CEA, Ep-CAM, Her2-Neu have been identified in PBMCs from CRC patients (12) . Moreover, clinical and immunological responses have been observed in CRC patients vaccinated with DCs pulsed with a CEA-derived analogue (13) .

Despite these results, antigens expressed by CRC cells and suitable for vaccine preparation remain to be identified. An alternative approach is offered by tumor-derived HSP, a group of glycoproteins that are the most common and abundant proteins in all forms of life. Previous work has demonstrated that purified homogeneous preparations of gp96, HSP-70, HSP-90, and other HSPs, are actually noncovalent complexes of HSP and peptides (14) . The peptides derive from the proteins expressed in the cells from which the HSPs are purified and include normal self peptides as well as antigenic peptides. On immunization, the HSP bind to the CD91 HSP-receptors expressed by APCs, including DCs (15) . The HSP-chaperoned peptides are then processed within the DCs and re-presented to T lymphocytes by MHC class-I molecules (15) . This pathway has been demonstrated in murine (15) and human systems (16) .

Studies with animal models indicate that vaccination with autologous tumor-derived HSP results in both prophylactic and therapeutic antitumor activity without the need to identify tumor-specific antigenic epitopes (17, 18, 19) . The most potent antitumor activity of HSP was observed in animals rendered disease-free by surgery, but at high risk of recurrence of metastatic cancer (19) . Moreover, vaccination with autologous tumor-derived HSP was found to be individually tumor-specific and not tumor histotype-specific, leading to the conclusion that the relevant tumor-antigenic, immunoprotective peptides are derived not from shared but from unique antigens that characterize each individual neoplasm (19 , 20) . In addition to these preclinical findings, the present study was predicated on a pilot human trial showing that vaccination with HSPPC-96, isolated from various types of autologous gastrointestinal, pancreatic, and renal cancers, was feasible and free of relevant toxicity (21) . More recently, vaccination of metastatic melanoma patients with autologous HSPPC-96 resulted in 50% tumor-specific T-cell response and 18% clinical response (22) .

In the present study, feasibility, safety, immunogenicity, and possible clinical benefits of an adjuvant vaccination with autologous tumor-derived HSPPC-96 were investigated in a consecutive series of patients rendered disease free from metastatic CRC through complete resections of liver metastases.

PATIENTS AND METHODS

Eligibility, Patient Characteristics, and Surgical Procedure
From February to December 1999, at the Hepatobiliary and Gastro-Pancreatic Unit of the Istituto Nazionale Tumori of Milan, a consecutive series of 29 patients (median age, 59 years; range, 37–74 years) with histologically confirmed diagnosis of liver metastases from CRC underwent potentially curative liver resection. All of the subjects underwent extensive preoperative work-up (total body helical-computed tomography, bone scan, colonoscopy, and positron emission tomography scan in case of multiple liver metastases) to rule out any detectable extrahepatic lesion. Different types of liver resections were performed, depending on tumor location and extension within the liver, with the goal of complete removal with clear margins of all neoplastic deposits. Intraoperative ultrasound of the liver was obtained at the beginning and at the end of each operation, to identify any possible metastases; hilar lymph node dissection was performed in all cases.

Eligibility criteria for vaccination included: (a) colon carcinoma metastases with resectability to yield an appropriate amount of nonnecrotic neoplastic tissue (at least 3 g for vaccine preparation plus 1 g for immunological assays); (b) a performance status (Zubrod) of 2 or less; (c) life expectancy of at least 16 weeks; (d) normal WBC and platelet count, hemoglobin level >10 g/liter; (e) bilirubin <1.5 times normal, ALT <4 times normal, and adequate renal function with serum creatinine of <2 times normal; (f) full recovery from prior anticancer therapy with at least a 4-week interval from the last administration of prior anticancer treatment; and, (g) positive response to common recall antigens as assessed by the Multitest Mérieux (Imtix, Milan, Italy) as a sign of sufficient immune function.

Patients were excluded from the study if they: (a) had active brain metastases; (b) had concomitant autoimmune or malignant diseases or were receiving concurrent anticancer or immunosuppressive drugs; (c) had primary or secondary immunodeficiencies; (d) had history of serious intercurrent medical illnesses. Women of child-bearing potential required negative serum pregnancy test before entry into the study and agreed to use an effective method of contraception while on treatment. All of the patients gave written informed consent to participate in the study.

Particular reference was made to the fact that no chemotherapy would be proposed after a complete tumor resection in the absence of cancer recurrence during the follow-up. The entire study was approved by the Ethics and Scientific Committees of the Istituto Nazionale Tumori of Milan.

HSP Preparation
Autologous HSPPC-96 vaccine was prepared from fresh tumor samples of each patient as described previously (23) . Briefly, macroscopically nonnecrotic tumor tissue was obtained in sterile condition from the operating room, weighed, immediately frozen in liquid nitrogen, shipped to the Antigenics Inc. facility (Woburn, MA), and processed under good medical practice conditions as described previously (21) . Preparations were considered to be of acceptable quality only if all of the following conditions were met: (a) the major protein band on SDS-PAGE was M_r 96,000; (b) this band could be immunoblotted with an anti-gp96 mAb; and (c) no other band detected by SDS-PAGE could be immunoblotted with anti-gp96 mAb. All of the HSPPC-96 preparations met such criteria and were consequently shipped back to Milan to be used to vaccinate patients.

Vaccination Schedule and Toxicity Assessment
Patients were treated with two cycles of intradermal vaccination courses at one of the three defined dose of HSPPC-96 (11 cases at 2.5 µg, 11 cases at 25 µg, and 7 cases at 100 µg) depending on tissue availability. The site of vaccination was rotated weekly and included: anterior deltoid region, median inguinal region and subclavicular regions. The first cycle lasted 1 month (four injections, once a week) and started 4–6 weeks after surgery, whereas the second cycle included four injections given at two-week intervals, starting 8 weeks after the end of the first cycle. Vaccine injection as well as clinical monitoring before and after each cycle were conducted in the Hepatobiliary Surgery outpatient clinic. At any patient’s visit, toxicity and possible relationship to vaccination were registered and graded according to the National Cancer Institute toxicity scale (24) .

Clinical Monitoring and Prognostic Grouping
Physical examination, blood testing, liver ultrasound and/or computed tomograph scan and chest X-ray were performed monthly. In case of tumor recurrence, the vaccination was stopped, and the patient was assigned to other therapies, either re-resection or conventional chemotherapy for CRC, depending on tumor extension. In all cases, patients remained on permanent follow-up, and data on 20 clinical and pathological prognostic factors associated with prognosis after liver resection for CRC metastases were collected. Table 1 summarizes these factors and their relationship to patient prognosis. Patients were divided in two prognostic groups (A and B) based on our previous experience with over 400 liver resections for liver metastases and on the MSKCC score predictive of recurrence after hepatic resection for metastatic CRC (7) . Briefly, Group A (good prognosis, i.e., with MSKCC score 1–3) included 18 patients with clear margins, 3 metastases, tumor-free hilar lymph nodes, and CEA 10 ng/ml; Group B included 11 patients with poor prognosis (i.e., with MSKCC score 4–5) and carrying any other risk factor not included in Group A.

Patients’ immune response to vaccination was defined as a significant postvaccination increase in the number of PBMCs responding with IFN- release (as detected by ELISPOT assay, see below) to either autologous or allogeneic class I HLA-matched colon cancer cells as compared with prevaccination PBMCs. The time point of the assay was different in 6 of the 29 subjects because of the unavailability of enough PBMCs (see below). In some cases, PBMCs could be obtained at multiple time points and, therefore, a kinetics of the antitumor immune response could be determined.

Cell Samples for Immunological Monitoring
PBMCs.
Eighty to 100 ml of heparinized blood for immunological assays were obtained by preparation of a buffy coat from each patient, before and after vaccine administration. PBMCs were isolated by Ficoll (Pharmacia & Upjohn, Kalamazoo, MI) gradient centrifugation, washed twice with PBS and cryopreserved, with the addition of RPMI 1640, plus 30% human serum and 10% DMSO (Fluka, Buchs, Switzerland) in liquid nitrogen. Before assaying, frozen cells were thawed, and functional analysis was performed after overnight resting in RPMI 1640 supplemented with 10% human serum. At the time of testing, PBMCs were at least 95% viable.

HLA Typing.
Serological typing for HLA-A, -B, and -C was performed by the standard two-step NIH complement-dependent microcytoxicity test. For molecular typing, genomic DNA was purified from protease-treated PBMCs with QIAamp DNA kit (Qiagen, Hilden, Germany), and HLA typing was performed by amplification with PCR with sequence-specific primes (PCR-SSPs; Dynal, Bromborough, United Kingdom).

Tumor Cells.
Tumor cell suspensions were obtained by mechanical processing of liver metastases from colon carcinoma. Enzymatic digestion, with a pool of DNase (Sigma-Aldrich, Milan, Italy), collagenase (Sigma-Aldrich) and trypsin (Bio-Whittaker Europe, Verviers, Belgium) was additionally performed to achieve higher cell recovery. Tumor cells were then aliquoted, frozen, and stored in liquid nitrogen. Because of the poor viability of autologous tumor cells after thawing, such targets could be tested for PBMC recognition in ELISPOT assay only in two patients (patients 3 and 20). For the other patients, colon carcinoma cell lines matching at least one HLA-A or -B allele with each patient’s PBMCs were used as targets. These lines either were purchased from ATCC [Manassas, VA (SW480: HLA-A2, -24, -B7, and -B15; Co115: HLA-A1, -A3, -B35, and -B5; HT29: HLA-A1, -24, -B35, and -B44; SW1463: HLA-A1, -A2, -B8, and -B44; SW116: HLA-A2, -A23, -B44, and -B60)] or were kindly provided by Dr. C. Scheibenbogen (B. Franklin University, Berlin, Germany; Colo206, CX94, HCT116, all three typing as HLA-A2). In detail, 10 patients were tested with HLA-A2-matched (Colo206, HCT116, SW480), 5 with HLA-A24-matched (HT29, SW480), 4 with HLA-A1-(HTC116, Colo206, HT29), 4 with the HLA-A3-matched (Co115), 3 with HLA-B7-(SW480, Colo206), and 3 with HLA-B44-matched (SW1463, SW116) lines, respectively. In addition, the melanoma line 501mel (HLA-A2+, -A1+; Ref. 25 ) was used as an unrelated target.

ELISPOT Analysis

This assay allows direct testing of antigen recognition by patients’ T cells and has been used in melanoma patients vaccinated with different peptides (26 , 27) . As recently described (22 , 27) , sterile 96-well nitrocellulose plates (Milititer; Milliporre, Bedford, MA) were coated with 50 µl/well mouse antihuman IFN- mAbs (Mabtech, Nacka, Sweden) at a concentration of 15 µg/ml in sterile coating buffer) and were incubated overnight at 4°C. The following day, plates were extensively washed with PBS to remove unbound coating mAbs. PBMCs, previously thawed and incubated overnight at 37°C, were added to the plates at the concentration of 1.66 x 10⁵/well in 100-µl/well volume. A total of 10⁶ PBMCs/patient were analyzed. After 30-min incubation at 37°C, target cells were added at the concentration of 1.6 x 10⁴/well. To check for total IFN- release, cytokine release in response to pokeweed mitogen (1 µg/ml; Sigma, Munich, Germany) was included as a positive control. Blocking experiments were performed by incubating monocyte-depleted PBMCs with target cells preincubated for 30 min at 37°C with the anti-HLA class I mAb W6.32 (ATCC) or the anti-HLA class II mAb L243. To control the ELISPOT assay, the anti-Melan-A/MART-1_27–35 T-cell clone A42 (28) , in the presence or absence of relevant targets (T2 lymphoma cells pulsed with Melan-A/MART-1_27–35 peptide and/or the Melan-A/MART-1+ melanoma line 501mel), was always included at the concentration of 400 cells/well in triplicate. Interassay variability, as evaluated by these two target controls was indeed low, producing 120 ± 18 and 297 ± 21 spots/well, respectively, in 56 assays considered. After an additional incubation for 20 h at 37°C and 5% CO₂, plates were washed with PBS. Wells were then incubated for 2–4 h at room temperature with 50 µl/well biotynilated mouse antihuman IFN- mAbs (Mabtech, Nacka, Sweden) at the concentration of 1 µg/ml in PBS with 0.5% FCS. Wells were then washed and 100 µl/well streptavidin alkaline phosphatase (Mabtech, Nacka, Sweden), diluted 1 in 1000 in PBS-0.5% FCS, was then added. After 1-h incubation at room temperature, plates were washed and 100 µl/well substrate (Bio-Rad Laboratories, Hercules, CA) were added. Color development was stopped by washing in tap water when dark spots emerged (up to 30 min). Plates were then left to dry overnight at room temperature, and spots were counted by a computer-assisted ELISPOT reader (Bioline AID, Turin, Italy). Spot number per seeded PBMCs represents the mean value of three or six replicates. To calculate the number of PBMCs responding to tumor cells by IFN- release, a background (i.e., the number of IFN- spots obtained with PBMCs alone: median, 6.5 spots/1.67 x 10⁵ PBMCs; range, 0–36; n = 29 samples) was subtracted.

Immunophenotyping of Effector Lymphocytes

PBMCs were thawed and stained with the murine antihuman CCR7 mAb (clone 2H4; BD PharMingen, San José, CA) for 30 min at 4°C, were washed twice, and were incubated with the biotinylated rat antimouse IgM (BD PharMingen) for 20 min at 4°C. After two washings, PBMCs were incubated with streptavidin-change in phycoerythrin (Molecular Probe, Eugene, OR) and were washed three times. Cells were then stained with the anti-CD3-APC and the CD45RA-FITC mAbs (BD PharMingen). As negative controls, an isotype-matched (IgM) mAb followed by the biotinylated antirat IgM and isotype-matched (APC or FITC conjugated) mAb were used. Fluorescence was then analyzed by FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA) and CellQuest software (Becton Dickinson).

Statistical Analysis

The induction of detectable immune responses against CRC cells after vaccination with autologous tumor-derived HSPPC-96 in the absence of detrimental toxic effects was the primary end point of this cohort study. To evaluate this, pre- and postvaccine difference in the number of IFN- spots on colon cancer cells were compared in each patient by Student’s t test. There was no result adjustment of spot numbers, and the SDs were evaluated on three- or six-well replicates. T-cell responses were considered positive if: (a) a minimum of 10 tumor-induced spots/1.67 x 10⁵ PBMCs were detected (after subtracting the spots of unstimulated PBMCs): (b) a 2-fold increase of tumor recognition in postimmunization samples as compared with preimmunization samples was achieved; and (c) the differences in the number of tumor-induced spots between PBMCs obtained before and after vaccination were statistically significant with a P of <0.05 using the Student’s t test.

OS and DFS were the secondary end point of the study. They were estimated according to the Kaplan-Meier method (29) and compared in univariate analysis with the use of the two-sided log-rank tests, with all survival intervals calculated starting from the date of liver resection. Follow-up was calculated up to 24 months. Because of the small sample size (29 consecutive cases), the analysis of prognostic factors was confined to the comparison of favorable versus unfavorable prognostic groups according to clinical MSKCC scores (7) and to the possible combined effect of patient immune response on recurrence-free survival. The Cox proportional hazard model was used, whether or not subjects were stratified for the prognostic score. Any variable, such as immune response to vaccine, with a hazard ratio of <1 with a P of <0.05 was considered to be significantly associated with a protective effect of the analyzed factor on prognosis (i.e., DFS).

Data on whether or not an immune response was achieved in each patient were acquired at least 1 year after the conclusion of the vaccination schedule. Thus, investigators were blinded with regard to the possible effects of immunization until the end of the study.

RESULTS

Immunological Results
Induction of Anti-Colon Carcinoma-specific T Lymphocytes after Vaccination with HSPPC-96.
T-cell-dependent antitumor-specific immune reactivity in fresh PBMCs obtained before, and at different times after, HSPPC-96 vaccination was analyzed by IFN- ELISPOT assay. Because of the difficulty in establishing cell lines from fresh tumor samples, allogeneic HLA-A- or -B-matched (i.e., sharing at least one class I HLA allele with the patient) colon carcinoma lines available in our laboratory were additionally used as targets in the ELISPOT assay. The decision to use matching class I HLA was based on the evidence that most of the HSPPC-96-induced immune responses detected both in animals and in vaccinated patients were mediated by CD8+ T cells and, thus, were class I HLA-restricted (18 , 19 , 21 , 22) .

Postvaccine values refer to PBMCs collected at V5 (22 cases), V6 (2 cases), V7 (3 cases), V9 (one case), and V10 (one case), i.e., weeks 4–18 after the onset of vaccination. In 22 cases (exceptions are patients 2, 4, 7, 8, 15, 25, and 27), PBMCs were obtained also at time points V7 and V10, i.e., at the beginning and at the end of the second cycle of vaccination. In two subjects, antitumor reactivity was evaluated also at V11, i.e. 28 weeks from surgery. Overall, 17 (59%) of the 29 patients displayed a statistically significant increase in postvaccination frequency of PBMCs that released IFN- in response to either autologous (patients 3 and 20) or allogeneic HLA-matched colon carcinoma cells (Fig. 1A) . In these 17 subjects, the postvaccination time refers to weeks 4 (V5), 8 (V6), 12 (V7), 16 (V9), and 18 (V10) from the onset of vaccination in 10, 2, 3, 1, and 1 cases, respectively. Thus, the vast majority of these subjects was tested at the same postvaccine time. When considered according to the dose of vaccine, the frequency of patients with an increased immune response was 4 of 7, 4 of 11, and 9 of 11 for subjects who received 100, 25, or 2.5 µg of vaccine, respectively. This may suggest, as reported in the murine system (30) , that the lowest dose of HSPPC-96 was more immunogenic than the intermediate or high doses, although the number of cases tested was limited.

View larger version (33K): [in this window] [in a new window]   Fig. 1. Effect of HSPPC-96 vaccination on IFN- release by PBMCs in response to autologous/allogeneic class I HLA-matched colon carcinoma or to histologically unrelated tumor cells. The antitumor immune response before and after HSPPC-96 (Oncophage) vaccination was investigated by IFN- ELISPOT in PBMCs of 29 CRC patients with completely resected liver metastases. Postvaccine values refer to PBMCs collected at visit (V) 5 (22 cases), V6 (2 cases), V7 (3 cases), V9 (1 case), and V10 (1 case), i.e., weeks 8–22 after surgery. In 22 cases (exceptions are patients 2, 4, 7, 8, 15, 25, and 27), PBMCs were obtained also at time points V7 and V10, i.e., at the beginning and the end of the second cycle. In two subjects, antitumor reactivity was evaluated also at V11, i.e., 28 weeks after surgery. Black columns, prevaccine values. White columns, nonsignificant postvaccine increases. Gray columns, statistically significant postvaccine increases (P < 0.05, evaluated as described in "Patients and Methods"). Triplicate values were used for each patient and time point; error bars, SDs. A, recognition of autologous or allogeneic, class I HLA-matched CRC cells. As targets, allogeneic CRC lines (matching at least one HLA-A or -B allele with patient PBMCs, as described in detail in "Patients and Methods" section) were used. Recognition of autologous tumor cells is reported for patients 3 and 20 (gray boxed numbers). In 59% (17 of 29) of the patients, a significant increase in IFN- release against either class I HLA-matched or autologous CRC cell lines was detectable after vaccination. B, lack of recognition of histologically unrelated tumor cells. Over two-thirds of the patients (24 of 29) failed to display increased recognition of melanoma cell line (501mel), suggesting a specific recognition of colon cancer cells after vaccination of HSPPC-96. However, a significant increment in postvaccine PBMCs antitumor reaction was detectable in five patients (patients 1, 14, 15, 18, and 26).

View larger version (32K): [in this window] [in a new window]   Fig. 2. Class I HLA-restricted recognition of CRC lines by PBMCs from patients vaccinated with HSPPC-96 (Oncophage). Postvaccine PBMCs from the 17 subjects who developed an anti-colon carcinoma lymphocyte response after HSPPC-96 vaccination were tested against class I HLA-matched CRC cell lines in the presence (black columns) and in the absence (white columns) of the anti-HLA-class I mAb W6.32 (ELISPOT assay). Fifteen of the 17 responsive patients (exception was patient 9, although in patient 11, the inhibitory effect was borderline) displayed significant (P < 0.05) class I HLA-restricted recognition of HLA-class I matched CRC cell lines, proving CD8+ T-cell-specific involvement in recognition of colon carcinoma cells.

To better characterize such CD8+ T-cell responses, we performed a phenotypic analysis of PBMCs, particularly for CCR7 expression, which, together with CD45RO and CD45RA, defines subsets of naive and memory CD8+ T lymphocytes with different functional activity (32 , 33) . PBMCs from patients showing significant enhancement of antitumor T-cell reactivity after vaccination, displayed an increased percentage of the CD45RA+/CCR7 subset (Table 2) , which is known to define final cytotoxic effector T cells that are endowed with the ability to migrate into inflamed or neoplastic tissues (32 , 33) . On the contrary, no significant increase in such T-cell subsets was found in patients who showed no increase in antitumor T-cell responses (Table 2) .

No reactions were observed in any patient at sites of HSPCC-96 vaccination.

Clinical Results

Vaccination-Related Side Effects.
Twenty-two of the 29 patients completed the whole vaccination schedule, whereas 7 subjects (patients 2, 4, 7, 8, 15, 25, and 27) had only the first cycle (6 cases because of cancer recurrence and one case because of refusal). Neither severe toxicity nor serious adverse events were registered during vaccination, although there were 9 cases (31%) of grade 0–1 toxicity (3 hypertensions, 2 flu-like syndromes, 1 case each of fever, diarrhea, headache, and abdominal pain). In all instances, such toxicity was easily managed with common medications. Patients’ compliance with the vaccine schedule and follow-up visits was high, with no patient being lost or delayed during the study period.

Patients’ Outcome.
All of the patients survived the liver resection procedure and left the hospital after a median of 7 days. Major postsurgical complications were not observed, and all subjects followed the vaccination protocol on schedule. After a median follow-up of 24 months, 23 patients (79%) are alive whereas 6 (21%) died because of disease progression. Altogether, tumor recurrence occurred in 18 patients (62%), and 11 (38%) are currently tumor free. Median time for recurrence was 7 months after liver resection (range, 2–24 months). Tumor locations of recurrences were: liver (11 cases), lung (5 cases), bone and pelvis (one case each). In 5 subjects, recurrence occurred during the second cycle of vaccine administration (<6 months from liver resection), whereas in the remaining 13 patients at least 6 months elapsed from the end of vaccine cycles and the evidence of relapsing disease. In 5 of 11 subjects with liver recurrence, repeated surgical resections were performed followed by conventional systemic chemotherapy, whereas in the remaining 6 cases, only chemotherapy was given. As of April 2002, 2 of 18 patients with early recurrences were still tumor free, and 10 patients were alive with disease; the remaining 6 patients died because of cancer progression.

The actuarial 24 months OS of the entire study group is 79%, with a DFS of 33% (Fig. 3) . When prognostic factors were considered according to MSKCC-score (7) , 18 patients fell into Group A (good prognosis, score 1–3) and 11 in Group B (adverse prognosis, score 4–5) as indicated in the "Patients and Methods" section and in Table 1 . As expected, Group A had a significantly better outcome than Group B. In fact, at 2 years, OS was 89% versus 64% (P = 0.001), and DFS was 46% versus 18% (P = 0.001) in group A versus B, respectively.

View larger version (13K): [in this window] [in a new window]   Fig. 3. Survival of 29 consecutive patients treated with autologous HSPPC-96 vaccination after complete liver resection of metastatic CRC. OS and DFS at 2 years are 79 and 33%, respectively; the median follow-up is 24 months.

However, as shown in Fig. 4 , patients with a postvaccination immune response had a statistically significant survival advantage compared with nonresponding patients. This was both on OS (100 versus 50%; P = 0.001; Fig. 4A ) and DFS (51 versus 8%; P = 0.0001; Fig. 4B ) at 2 years. More specifically, when tumor recurrence rate was plotted in a factorial table, cross-classifying patients based on immune response (absent versus present) and clinical prognostic score (favorable versus unfavorable), the occurrence of postvaccine antitumor immune response always led to better tumor-free survival, whatever the predicted prognosis was.

View larger version (17K): [in this window] [in a new window]   Fig. 4. Different clinical outcome between immune responders and nonresponders vaccinated with HSPPC-96 after complete liver resection of metastatic CRC. Patients with a postvaccination immune response (17 cases) had a significant survival advantage at 24 months compared with that of nonresponding patients (12 cases) both on OS (A; 100 versus 50% at 2 years) and on DFS (B; 51 versus 8% at 2 years).

This study demonstrates the de novo induction or augmentation of patients’ antitumor-specific T-cell response against CRC cells on vaccination with autologous HSPPC-96 in an adjuvant setting, namely in cases with advanced disease (i.e., liver metastases) rendered tumor free by surgical resection. In fact, preclinical and clinical data suggest that the efficacy of such an approach against cancer is increased if minimal residual disease has to be targeted (34) .

Currently, the clinical setting of curatively resected CRC liver metastases is commonly used for adjuvant locoregional and/or systemic chemotherapy (1 , 5 , 8) . Such a model seems particularly suitable for our vaccination approach because: (a) enough material for vaccine preparation could be collected from each patient; (b) HSPPC-96 was extracted in all instances; (c) postsurgical vaccination was not associated with serious adverse events; and (d) full patients’ compliance was achieved. This clinical setting seems suitable for additional studies on adjuvant vaccination against CRC in humans. On the contrary, in metastatic melanoma patients, a similar approach was less successful when only 70–80% of subjects could receive the complete vaccination schedule because of tumor progression (22) .

The dose of vaccine, although apparently without effect on clinical outcome, indicated that the lowest dose (i.e., 2.5 µg) caused the highest frequency of increased antitumor immune reactions in vaccinated subjects (9 of 11). This observation, although requiring confirmation with a larger sample, is in keeping with data from animal studies that indicate that vaccination with HSP may be more effective when low/intermediate doses of vaccine are used (30) .

The main end point of the present study was the de novo induction or augmentation of antitumor T-cell response in patients immunized with HSPPC-96 as defined by a significant postvaccination increase in the number of PBMCs responding with IFN- release to class-I HLA-matched allogeneic and/or to autologous (available in two cases) colon cancer cells. The rationale for the usage of autologous vaccine is to see whether the in vivo induction of tumor-specific T-cell responses can be mediated, as in animal models (18) , by unique antigens. Thus, the autologous tumor-derived HSPPC-96 vaccine was prepared on a strictly individual basis, as part of a comprehensive clinical strategy that included standardized surgical and medical care.

Unfortunately, the demonstration of in vivo induction of T-cell responses against individual antigens could not be obtained because of the poor viability of fresh tumor cell suspensions and the difficulty of establishing cell lines from colon cancer metastases by which one could compare the T-cell recognition of autologous versus allogeneic, HLA-compatible colon cancer cells. However, the availability of a large panel of allogeneic HLA-class I matched colon carcinoma cell lines and the use of class I HLA blocking antibodies allowed the evaluation of CD8+ T-cell-mediated recognition of common tumor antigens and its modulation by HSPPC-96 vaccine. This conclusion is also supported by experiments showing recognition of well-defined CRC T-cell epitopes (e.g., CEA, epithelial-cell adhesion molecule) by some of the HLA-A*0201 patients vaccinated with HSPPC-96.⁴

A significant increase of class I HLA-restricted T-cell-mediated recognition of colon cancer cells was observed in 52% (15 of 29) of patients. In two additional cases, this increased reaction did not appear to be HLA-restricted and, thus, should be attributed to NK cells or to MHC-unrestricted T-cell activity (35) . Such a measurable effect of HSPPC-96 vaccination has been shown to be related to the time course of the disease, because the activation of the immune response was significantly associated with prognosis, an effect that rarely occurs with other types of vaccines (34) . As in animal models, the intensity of vaccine-elicited antitumor immune response could discriminate between control of tumor growth and control of tumor progression (18) . This observation confirms results obtained in metastatic melanoma patients (22) .

The present study also suggests that HSPPC-96 vaccination can expand the effector compartment of tumor-specific T lymphocytes, with special reference to CD8+ T cells recognizing CRC antigens in a HLA-class I restricted fashion. Corroborating this conclusion is an increase in the CD45RA+/CCR7- effector T lymphocytes that was observed in patients with enhanced immune response to tumor cells (32) .

The number of patients studied is too small for any definite conclusion on the possible clinical benefit of adjuvant vaccination with tumor-derived HSPPC-96 administered after complete resection of CRC liver metastases. At first glance, the 29 consecutive patients whose liver metastases were resected had an OS and DFS after vaccination that was similar to those described in a much larger series of patients operated on with the same intent of cure (6 , 7 , 36 , 37) . Although this is a pilot study, the results obtained in vaccinated patients are of interest when compared with our own experience with over 400 patients who never showed a 2-year OS and DFS better than 75% and 35%, respectively, particularly when one considers the subset of patients who mounted an immune response after vaccination. In fact, more than 50% of our patients who were vaccinated with HSPPC-96 developed a detectable T-cell reaction against CRC cells and showed a significantly increased OS and DFS with respect to immunologically nonresponding subjects (Fig. 4) . Indeed, regression analysis (Table 4) demonstrated that an immune response to vaccine is an independent factor predicting improved recurrence-free survival in all cases, whether or not the patients had a score, including performance status, predictive of good or bad prognosis. These data suggest that HSPPC-96 vaccine not only generated an immune response in more than one-half of our patients but may also have contributed to improved OS and DFS.

Such results must be interpreted cautiously because the small number of patients prevented a multivariate analysis on the possible influence of other prognostic factors. Also, an intrinsic patient selection cannot be excluded, although cases were collected consecutively. One factor that strengthens the results in favor of the relevance of the effect of vaccine immunization is the protective effect on cancer recurrence that was evaluated in a blinded fashion until the end of follow-up and that then was compared with patient prognosis generated at the beginning of the study using a validated prognostic scoring system.

By comparing our findings with recent results of adjuvant chemotherapy (8 , 38) , the data are not sufficient to recommend HSPPC-96 vaccination as an alternative to conventional chemotherapy after resection of CRC liver metastases. Nevertheless, the results of the combination of tumor resection plus vaccine could justify a prospective clinical trial, especially for patients with good prognosis, particularly in light of the lack of toxicity observed. The use of HSP-derived vaccines should be further investigated, either alone or in combination with chemotherapy, in prospective clinical trials.

In conclusion, HSPPC-96 vaccination represents a feasible and safe approach for adjuvant treatment after curative resection of CRC liver metastases. A statistically significant increase in postvaccination, class-I HLA-restricted T-cell response against CRC cells was detectable in more than 50% of the patients studied for the first time in such a clinical setting. Moreover, such immunization effect involved effector CD8+ T lymphocytes, reinforcing the conclusion that these lymphocytes were instrumental in destroying CRC cells in vivo. Two-year OS and DFS were improved in subjects showing an antitumor immune response, and, although prognostic scores (based on preoperative tumor characteristics) correctly predicted survival, the immune response after vaccination exerted a significant impact on tumor-free survival.

Although the number of patients was small, the present prospective cohort study suggests a possible clinical benefit of HSPPC-96 vaccination in a homogeneous series of patients at high risk of recurrence after complete surgical removal of CRC hepatic metastases. This vaccination approach, therefore, warrants additional studies with a larger number of subjects.

ACKNOWLEDGMENTS

We thank Drs. L. Ascani and W. Gianessi (Pharmacy), C. Lombardo and F. Arienti (Blood Bank) of the Istituto Nazionale Tumori for their invaluable help during the study; Dr. C. Scheibenbogen (University of Berlin, Berlin, Germany) for providing valuable suggestions and colon cancer lines; A. Cova, P. Squarcina, and F. Giardino for excellent technical and clinical work; and G. Barp and D. Guarneri for editorial assistance.

FOOTNOTES

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ Supported by a grant from Sigma-Tau i.f.r., S.p.A., Rome. V. M. is recipient of the grant "New therapeutic strategies in liver tumors" by the Italian Association for Cancer Research (AIRC). P. K. S. is supported by a NIH grant and through a sponsored research agreement with Antigenics Inc., in which he has a substantial financial interest. R. C. and M. C. are employees of the Sigma-Tau i.f.r. S.p.A., and J. J. L. is an employee of Antigenics Inc. Partially presented at the 37th Annual Meeting of the American Society of Clinical Oncology (ASCO), San Francisco, CA, May 14, 2001.

² To whom requests for reprints should be addressed, at Department of Surgery, Hepatobiliary and Gastro-Pancreatic Unit, Istituto Nazionale Tumori, Via Venezian 1, 20133 Milano, Italy. Phone: 39-02-2390-2760; Fax: 39-02-2390-.3259; E-mail: vincenzo.mazzaferro{at}istitutotumori.mi.it

³ The abbreviations used are: CRC, colorectal carcinoma; APC, antigen-presenting cell; CCR7, chemokine receptor 7; CEA, carcinoembryonic antigen; EP-CAM, epithelial cell adhesion molecule; DC, dendritic cell; DFS, disease-free survival; HSP, heat shock protein; HSPPC, HSP/peptide-complex; mAb, monoclonal antibody; NK, natural killer; OS, overall survival; PBMC, peripheral blood mononuclear cell; MSKCC, Memorial Sloan-Kettering Cancer Center; ELISPOT, enzyme-linked immunospot; ATCC, American Type Culture Collection.

⁴ L. Rivoltini and W. Hansjochen, unpublished observations.

Received 1/ 2/03; revised 2/21/03; accepted 2/24/03.

REFERENCES

1. Kemeny N., Huang Y., Cohen A. M., Shi W., Conti J. A., Brennan M. F., Bertino J. R., Turnbull A-D., Sullivan D., Stockman J., Blumgart L. H. Hepatic arterial infusion of chemotherapy after resection of hepatic metastases from colorectal cancer. N. Engl. J. Med., 341: 2039-2048, 1999.[Abstract/Free Full Text]
2. Landis S. H., Murray T., Bolden S., Wingo P. A. Cancer statistics, 1999. CA Cancer J. Clin., 49: 8-31, 1, 1999.[Abstract/Free Full Text]
3. Weiss L., Grundmann E., Torhorst J., Hartveit F., Moberg I., Eder M., Fenoglio-Preiser C. M., Napier J., Horne C. H., Lopez M. J., Shaw-Dunn R. I., Sugart J., Davies J. D., Day D. W., Harlus J. P. Haematogenous metastatic patterns in colonic carcinoma: an analysis of 1541 necropsies. J. Pathol., 150: 195-203, 1986.[CrossRef][Medline]
4. Lorenz M., Staib-Sebler E., Hochmuth K., Heinrich S., Gog C., Vetter G., Encke A., Muller H. H. Surgical resection of liver metastases of colorectal carcinoma: short and long-term results. Semin. Oncol., 27: 112-119, 2000.[Medline]
5. Nordlinger B., Rougier P. Liver metastases from colorectal cancer: the turning point. J. Clin. Oncol., 20: 1442-1445, 2002.[Free Full Text]
6. Figueras J., Valls C., Rafecas A., Fabregat J., Ramos E., Jaurrieta E. Resection rate and effect of postoperative chemotherapy on survival after surgery for colorectal liver metastases. Br. J. Surg., 88: 980-985, 2001.[CrossRef][Medline]
7. Fong Y., Fortner J., Sun R. L., Brennan M. F., Blumgart L. H. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann. Surg., 230: 309-318, 1999.[CrossRef][Medline]
8. Kemeny M. M., Adak S., Gray B., Macdonald J. S., Smith T., Lipsitz S., Sigurdson E. R., O’Dwyer P. J., Benson A. B., III. Combined-modality treatment for resectable metastatic colorectal carcinoma to the liver: surgical resection of hepatic metastases in combination with continuous infusion of chemotherapy—an intergroup study. J. Clin. Oncol., 20: 1499-1505, 2002.[Abstract/Free Full Text]
9. Wolmark N., Fisher B., Rockette H., Redmond C., Wickerham D. L., Fisher E. R., Jones J., Glass A., Lerner H., Lawrence W., Prager D., Wexler M., Evans J., Cruz A., Dimitrov N., Jochimsen P., et al Postoperative adjuvant chemotherapy or BCG for colon cancer: results from NSABP protocol C-01. J. Natl. Cancer Inst. (Bethesda), 80: 30-36, 1988.[Abstract/Free Full Text]
10. Vermoken J., Claessen A., van Titeren H., Gall H. E., Ezinga R., Meijer S., Scheper R. J., Maijer C. J. L-M., Bloemena E., Ranson J. H., Hanna M. G., Jr., Pinedo H. Active specific immunotherapy for stage II and stage III human colon cancer: a randomised trial. Lancet, 353: 345-350, 1999.[CrossRef][Medline]
11. Harris J., Ryan L., Hoover H., Stuart R. K., Oken M. M., Benson A. B., III, Mansour E., Haller D. G., Manola J., Hanna M. G. Adjuvant active specific immunotherapy for stage II and III colon cancer with an autologous tumor cell vaccine: Eastern Cooperative Oncology Group Study E5283. J. Clin. Oncol., 18: 148-157, 2000.[Abstract/Free Full Text]
12. Nagorsen D., Keilholz U., Rivoltini L., Schmittel A., Letsch A., Asemissen A. M., Berger G., Buhr H-J., Thiel E., Scheibenbogen C. Natural T-cell response against MHC class I epitopes of epithelial cell adhesion molecule, her-2/neu, and carcinoembryonic antigen in patients with colorectal cancer. Cancer Res., 60: 4850-4854, 2000.[Abstract/Free Full Text]
13. Fong L., Hou Y., Rivas A., Benike C., Yuen A., Fisher G. A., Davis M. M., Enfleman E. G. Altered peptide ligand vaccination with Flt3 ligand expanded dendritic cells for tumor immunotherapy. Proc. Natl. Acad. Sci. USA, 98: 8809-8814, 2001.[Abstract/Free Full Text]
14. Ishii T., Udono H., Yamano T., Ohta H., Uenaka A., Ono T., Hjizuta A., Tanaka N., Srivastava P. K., Nakayama E. Isolation of MHC class I-restricted tumor antigen peptide and its precursors associated with heat shock proteins hsp70, hsp90, and gp96. J. Immunol., 162: 1303-1309, 1999.[Abstract/Free Full Text]
15. Basu S., Binder R. J., Ramalingam T., Srivastava P. K. CD91 is a common receptor for heat shock proteins gp96, hsp90, hsp70, and calreticulin. Immunity, 14: 303-313, 2001.[CrossRef][Medline]
16. Castelli C., Ciupitu A. M., Rini F., Rivoltini L., Mazzocchi A., Kiessling R., Parmiani G. Human heat shock protein 70 peptide complexes specifically activate anti-melanoma T cells. Cancer Res., 61: 222-227, 2001.[Abstract/Free Full Text]
17. Ullrich S. J., Robinson E. A., Law L. W., Willingham, Appella E. A mouse tumor-specific transplantation antigen is a heat shock-related protein. Proc. Natl. Acad. Sci. USA, 83: 3121-3125, 1986.[Abstract/Free Full Text]
18. Udono H., Srivastava P. K. Heat shock protein 70-associated peptides elicit specific cancer immunity. J. Exp. Med., 178: 1391-1396, 1993.[Abstract/Free Full Text]
19. Tamura Y., Peng P., Liu K., Daou M., Srivastava P. K. Immunotherapy of tumors with autologous tumor-derived heat shock protein preparations. Science (Wash. DC), 278: 315-318, 1997.[CrossRef]
20. Srivastava P. K., Menoret A., Basu S., Binder R. J., McQuade K. L. Heat shock proteins come of age: primitive functions acquire new roles in an adaptive world. Immunity, 8: 657-665, 1998.[CrossRef][Medline]
21. Janetzki S., Palla D., Rosenhauer V., Lochs H., Lewis J. J., Srivastava P. K. Immunization of cancer patients with autologous cancer-derived heat shock protein gp96 preparations: a pilot study. Int. J. Cancer, 88: 232-238, 2000.[CrossRef][Medline]
22. Belli F., Testori A., Rivoltini L., Maio M., Andreola G., Sertoli R., Gallino G., Piris A., Cattelan A., Lazzari I., Carrabba M., Scita G., Santantonio C., Pilla L., Tragni G., Lombardo C., Arienti F., Marchianò A., Queirolo P., Bertolini F., Cova A., Lamaj E., Ascani L., Camerini R., Corsi M., Cascinelli N., Lewis J. J., Srivastava P., Parmiani G. Vaccination of metastatic melanoma patients with autologous tumor-derived Heat Shock Protein Peptide-Complex-96: clinical and immunological findings. J. Clin. Oncol., 20: 4169-4180, 2002.[Abstract/Free Full Text]
23. Srivastava P. K., Jaikaria N. S. Methods of purification of heat shock protein-peptide complexes for use as vaccines against cancers and infectious diseases. Methods Mol. Biol., 156: 175-186, 2001.[Medline]
24. Green S., Weiss G. R. Southwest Oncology Group standard response criteria, endpoint definitions and toxicity criteria. Investig. New Drugs, 10: 239-253, 1992.[CrossRef][Medline]
25. Kawakami Y., Eliyahu S., Sakaguchi K., Robbins P. F., Rivoltini L., Yannelli J. R., Appella E., Rosenberg S. A. Identification of the immunodominant peptides of the MART-1 human melanoma antigen recognized by the majority of HLA-A2-restricted tumor infiltrating lymphocytes. J. Exp. Med., 180: 347-352, 1994.[Abstract/Free Full Text]
26. Lewis J. J., Janetzki S., Schaed S., Panageas K. S., Wang S., Williams L., Meyers M., Butterworth L., Livingston P. O., Chapman P. B., Houghton A. N. Evaluation of CD8(+) T-cell frequencies by the ELISPOT assay in healthy individuals and in patients with metastatic melanoma immunized with tyrosinase peptide. Int. J. Cancer, 87: 391-398, 2000.[CrossRef][Medline]
27. Scheibenbogen C., Romero P., Rivoltini L., Herr W., Schmittel A., Cerottini J-C., Woelfel T., Eggermont A. M. M., Keilholz U. Quantitation of antigen-reactive T cells in peripheral blood by IFN--ELISPOT assay and chromium-release assay: a four-centre comparative trial. J. Immunol. Methods, 244: 81-89, 2000.[CrossRef][Medline]
28. Kawakami Y., Eliyahu S., Delgado C. H., Robbins P. F., Rivoltini L., Topalian S. L., Miki T., Rosenberg S. A. Cloning of the gene coding for a shared human melanoma antigen recognized by autologous T cells infiltrating into tumor. Proc. Natl. Acad. Sci. USA, 91: 3515-3519, 1994.[Abstract/Free Full Text]
29. Kaplan E. L., Meier P. Nonparametric estimation from incomplete observations. J. Am. Stat. Assoc., 53: 457-481, 1958.[CrossRef]
30. Chandawarkar R. Y., Wagh M. S., Srivastava P. K. The dual nature of specific immunological activity of tumor-derived gp 96 preparations. J. Exp. Med., 189: 1437-1442, 1999.[Abstract/Free Full Text]
31. Wells A., Malkovsky M. Heat Shock proteins, tumor immunogenicity and antigen presentation: an integrated view. Immunol. Today, 21: 129-132, 2000.[CrossRef][Medline]
32. Sallusto F., Lenig D., Forster R., Lipp M., Lanzavecchia A. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature (Lond.), 40: 659-660, 1999.
33. Valmori D., Scheibenbogen C., Dutoit V., Nagorsen D., Asemissen A. M., Rubio-Godoy V., Rimoldi D., Guillaume P., Romero P., Schadendorf D., Lipp M., Dietrich P-Y., Thiel E., Cerottini J-C., Liénard D., Keilholz U. Circulating tumor-reactive CD8+ T cells in melanoma patients contain a CD45RA+CCR7- effector subset exerting ax vivo tumor-specific cytolytic activity. Cancer Res., 62: 1743-1750, 2002.[Abstract/Free Full Text]
34. Parmiani G., Castelli C., Dalerba P., Mortarini R., Rivoltini L., Marincola F. M., Anichini A. Cancer immunotherapy with peptide-based vaccines: what have we achieved, where are we going?. J. Natl. Cancer Inst. (Bethesda), 94: 805-818, 2002.[Abstract/Free Full Text]
35. Multhoff G., Botzler C., Jennen L., Schmidt J., Ellwart J., Issels R. Heat shock protein 72 on tumor cells. A recognition structure for natural killer cells. J. Immunol., 158: 4341-4350, 1997.[Abstract]
36. Nordlinger B., Guiguet M., Vaillant J. C., Balladur P., Boudjema K., Bachellier P., Jaeck D. Surgical resection of colorectal carcinoma metastases to the liver. A prognostic scoring system to improve case selection, based on 1568 patients. Association Française de Chirurgie. Cancer (Phila.), 77: 1254-1262, 1996.[CrossRef][Medline]
37. Minagawa M., Makuuchi M., Torzilli G., Takayama T., Kawasaki S., Kosuge T., Yamamoto J., Imamura H. Extension of the frontiers of surgical indications in the treatment of liver metastases from colorectal cancer: long-term results. Ann. Surg., 231: 487-499, 2000.[CrossRef][Medline]
38. Dizon D. S., Kemeny N. Locoregional chemotherapy for metastatic colorectal cancer to the liver Bleiberg H. Kemeny N. Rougier P. Hansjochen W. eds. . Colorectal Cancer: A Clinical Guide to Therapy, 367-374, Martin Dunitz London 2002.

This article has been cited by other articles:

				A. Testori, J. Richards, E. Whitman, G. B. Mann, J. Lutzky, L. Camacho, G. Parmiani, G. Tosti, J. M. Kirkwood, A. Hoos, et al. Phase III Comparison of Vitespen, an Autologous Tumor-Derived Heat Shock Protein gp96 Peptide Complex Vaccine, With Physician's Choice of Treatment for Stage IV Melanoma: The C-100-21 Study Group J. Clin. Oncol., February 20, 2008; 26(6): 955 - 962. [Abstract] [Full Text] [PDF]

				G. Parmiani, A. De Filippo, L. Novellino, and C. Castelli Unique Human Tumor Antigens: Immunobiology and Use in Clinical Trials J. Immunol., February 15, 2007; 178(4): 1975 - 1979. [Abstract] [Full Text] [PDF]

				E. Schmitt, M. Gehrmann, M. Brunet, G. Multhoff, and C. Garrido Intracellular and extracellular functions of heat shock proteins: repercussions in cancer therapy J. Leukoc. Biol., January 1, 2007; 81(1): 15 - 27. [Abstract] [Full Text] [PDF]

				Y. Enomoto, A. Bharti, A. A. Khaleque, B. Song, C. Liu, V. Apostolopoulos, P.-x. Xing, S. K. Calderwood, and J. Gong Enhanced Immunogenicity of Heat Shock Protein 70 Peptide Complexes from Dendritic Cell-Tumor Fusion Cells J. Immunol., November 1, 2006; 177(9): 5946 - 5955. [Abstract] [Full Text] [PDF]

				R. H. Wilson Novel Therapeutic Developments Other Than EGFR and VEGF Inhibition in Colorectal Cancer Oncologist, October 1, 2006; 11(9): 1018 - 1024. [Abstract] [Full Text] [PDF]

				X.-Y. Wang, H. Arnouk, X. Chen, L. Kazim, E. A. Repasky, and J. R. Subjeck Extracellular Targeting of Endoplasmic Reticulum Chaperone Glucose-Regulated Protein 170 Enhances Tumor Immunity to a Poorly Immunogenic Melanoma J. Immunol., August 1, 2006; 177(3): 1543 - 1551. [Abstract] [Full Text] [PDF]

				A. Rosato, A. Zoso, S. D. Santa, G. Milan, P. Del Bianco, G. L. De Salvo, and P. Zanovello Predicting Tumor Outcome following Cancer Vaccination by Monitoring Quantitative and Qualitative CD8+ T Cell Parameters J. Immunol., February 1, 2006; 176(3): 1999 - 2006. [Abstract] [Full Text] [PDF]

				J. Qian, S. Wang, J. Yang, J. Xie, P. Lin, M. E. Freeman III, and Q. Yi Targeting Heat Shock Proteins for Immunotherapy in Multiple Myeloma: Generation of Myeloma-Specific CTLs Using Dendritic Cells Pulsed with Tumor-Derived gp96 Clin. Cancer Res., December 15, 2005; 11(24): 8808 - 8815. [Abstract] [Full Text] [PDF]

				E. V. PIERPAOLI The Role of Hsp70 in Age-Related Diseases and the Prevention of Cancer Ann. N.Y. Acad. Sci., December 1, 2005; 1057(1): 206 - 219. [Abstract] [Full Text] [PDF]

				S. Liu, H. Wang, Z. Yang, T. Kon, J. Zhu, Y. Cao, F. Li, J. Kirkpatrick, C. V. Nicchitta, and C.-Y. Li Enhancement of Cancer Radiation Therapy by Use of Adenovirus-Mediated Secretable Glucose-Regulated Protein 94/gp96 Expression Cancer Res., October 15, 2005; 65(20): 9126 - 9131. [Abstract] [Full Text] [PDF]

M. W. Graner and D. D. Bigner Chaperone proteins and brain tumors: Potential targets and possible therapeutics Neuro-oncol, July 1, 2005; 7(3): 260 - 278. [Abstract] [PDF]