HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

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

Short-Term Kinetics of Tumor Antigen Expression in Response to Vaccination

Galen A. Ohnmacht^*, Ena Wang^*, Simone Mocellin, Andrea Abati, Armando Filie, Patricia Fetsch, Adam I. Riker^*, Udai S. Kammula^*, Steven A. Rosenberg^* and Francesco M. Marincola^1,*,

^* Surgery Branch, Department of Transfusion Medicine, Clinical Center, and Department of Cytopathology, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
The melanoma patient’s immune response to tumor has been extensively studied. Yet, the frequently observed coexistence of tumor-associated Ag (TAA)-specific T cells with their target cells in vivo remains unexplained. Loss of TAA expression might contribute to this paradox. We studied TAA expression in metastases by obtaining fine-needle aspirations from 52 tumor lesions in 30 patients with melanoma before and soon after immunotherapy. Limitations due to low amounts of starting material were overcome with a high fidelity antisense RNA amplification method. TAA expression was measured by quantitative real-time PCR of anti-sense RNA. Decrease in gp100/Pmel-17 TAA preceded tumor disappearance in several instances and could be best explained by immune selection because most patients had received gp100/Pmel-17-specific vaccination. Conversely, immune selection was absent in nonregressing lesions. These observations suggest that vaccination, when successful, triggers a broad inflammatory reaction that can lead to tumor destruction despite immune selection. Additionally, lack of clinical response might be attributed to lack of this initiating event rather than immune escape. This study provides an insight into the natural history of tumors and defines a strategy for the characterization of gene expression in tumors during therapy.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The kinetics of expression of tumor-associated Ags (TAA),² particularly melanoma differentiation Ags (MDA) or cancer testis (CT) Ags (1), has been the focus of great interest because it could provide an explanation for the paradoxical coexistence of cancer cells in the tumor-competent host (2). Indeed, the expression of MDA can be lost or decreased in melanoma metastases (2), perhaps in association with Ag-specific immunization (3, 4, 5, 6, 7), and some have suggested that it might explain tumor survival (8). In contrast, the short-term kinetics of expression of CT in response to immunologic manipulation has been less extensively characterized, perhaps in part because of the lack of suitable serologic reagents. We have recently noted that the expression of MDA (MART-1/MelanA) may also mark genetic drift among melanoma cell populations (9), suggesting that the biologic significance of TAA expression might trespass beyond the limits of their immunologic relevance into the realm of genetic determination of the neoplastic processes. Because MDA are not related to the oncogenic process and represent a remnant of the melanocytic origin of melanoma cells, maintenance of their expression confers no survival advantage to tumor cells. CT (10), on the other hand, are expressed in association with a progressive genome-wide demethylation associated with tumor progression (11, 12). Therefore, it is possible that, as tumor cells dedifferentiate during the neoplastic process and independently from immune pressure, MDA expression decreases while CT expression remains stable or even increases. Thus, it is presently unknown whether TAA loss during the neoplastic process is the specific result of immune selection, nor is it known whether it bears a causative relationship with the final outcome of the disease.

The expression of TAA throughout the disease process has not been accurately documented. Most published studies compare different primary or metastatic lesions surgically removed from patients at different time points throughout the natural course of the disease or in response to therapy (2). With this approach, we and others have been able to demonstrate a decrease in TAA in some patients that appears to be specific to the Ag adopted for active immunization (3, 7, 13). In particular, Jager et al. could demonstrate an inverse relationship in TAA expression in melanoma metastases and CD8⁺ cytotoxic T cell responses that suggested immune selection of Ag-loss variants in vivo (13). The excision of tissue for correlative studies with clinical outcome presumes that metastases are representative of each other. However, even synchronous metastases can be quite heterogeneous (6, 14, 15), and the experimental noise created by such heterogeneity can most suitably be eliminated by studying the kinetics of expression of given genes within the same metastasis by the use of serial fine-needle aspirations (FNA) (2). We were impressed by a case in which serial FNA demonstrated that tumor recurrence in response to MDA-specific vaccination was associated with maintenance of the expression of CT and loss of expression of the MDA targeted by the vaccine (16, 17). Based on these considerations, we have recently proposed that tumor-host interactions could be best followed by serial gene expression analysis of identical lesions through time by repeated FNA (18).

Until recently, the serial analysis of identical tumor deposits by repeated FNA has been hampered by the limited amount of material obtainable, which allows analysis of the expression of only a few gene products. Furthermore, traditional use of this technique generally involved the assessment of expression by immunochemical methods using the few markers for which appropriate reagents were available. We have recently developed a method of high fidelity anti-sense RNA (aRNA) amplification that uses a combination of template switching and in vitro transcription (19). This method yields 1:10,000–1:100,000 amplification of transcripts from conventional total RNA preparations that maintain gene expression profiles comparable to that of conventional poly(A) and total RNA based sources in cDNA microarrays.

Therefore, we tested whether aRNA could be used as a source material for accurate measurement of relative gene expression among different samples using quantitative real-time PCR (qRT-PCR). The aRNA was amplified from sequential FNA of 52 metastases from 30 patients with metastatic cutaneous melanoma. After amplification, aRNA was used as the template for qRT-PCR-based measurement of MDA and CT expression in a short-term (1–3 mo) follow-up period. This made it possible to study a larger number of genes compared with conventional methods. Furthermore, it was possible to study the expression of genes for which no serologic markers were available. The second goal of this study was to explore the change of expression of TAA during Ag-specific immunotherapy and to correlate the observed findings with therapy.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Sample collection

Patients who presented to the Surgery Branch of the National Cancer Institute of the National Institutes of Health (Bethesda, MD) for treatment of metastatic malignant melanoma were enrolled in treatment protocols after signing an informed consent. Upon identification of suitable candidates, metastatic melanoma masses were biopsied with a 23-gauge needle coupled to a 10-cc syringe. To ensure that the FNA was representative of the whole lesion, aspirations were obtained from four quadrants and pooled in a 15-ml Falcon tube (Sarstedt, Newton, NC) containing refrigerated RPMI 1640 medium (Biofluids, Rockville, MD). Sequential FNA of the same metastatic tumor site were performed. Part of the FNA material was forwarded to cytopathology for confirmation of the diagnosis of melanoma. Only specimens containing at least 100 cancer cells per cytospin preparation were collected further and used for this study. These specimens were also assessed for the quantity of expression of gp100/Pmel-17 and MART-1/MelanA by immunochemical methods. The remainder of the aspirate was placed in RNA lysis buffer (Qiagen, Santa Clarita, CA) and stored at -180°C until ready for processing.

RNA isolation, cDNA synthesis, and aRNA preparation

RNA isolation, RNA amplification, and cDNA transcription from the aspirate material were performed in batches containing patient’s pre- and posttherapy samples to minimize variability. RNA isolation was performed with RNeasy mini kits (Qiagen). RNA concentrations were determined by OD₂₆₀ reading in 50 mM sodium hydroxide (GeneQuant, Clamart, France). The aRNA was prepared from total RNA in 9 µl diethyl pyrocarbonate (DEPC)-treated H₂O containing 1 µg/µl oligo(dT) (15)-T7 primer (5'-AAA CGA CGG CCA GTG AAT TGT AAT ACG ACT CAC TAT AGG CGC-3'). Total RNA was denatured at 70°C for 3 min and primed while cooling to room temperature. T7 bacteriophage promoter was incorporated into cDNA synthesis in a reverse transcription reaction by adding 4 µl of first-strand reaction buffer, 2 µl 0.1 M DTT (Life Technologies, Rockville, MD), 2 µl 10 mM dNTP, 1 µl RNsin (Promega, Madison, WI), 1 µg/µl template switch primer (5'-AAG CAG TGG TAT CAA CGC AGA GTA CGC GGG-3'; Clontech Laboratories, Palo Alto, CA), and 2 µl SuperScript II reverse transcriptase (Life Technologies). The cDNA synthesis was completed at 42°C for 1 h. Full-length double-stranded cDNA was synthesized by adding 106 µl of DNase-free water, 15 µl Advantage PCR buffer (Clontech Laboratories), 3 µl 10 mM dNTP, 1 µl RNase-H (Promega), and 3 µl Advantage cDNA polymerase (Clonetech Laboratories). The following temperature cycle was used: 2 min at 37°C for RNA digestion, 3 min at 94°C for denaturation, 3 min at 65°C for priming, and 30 min at 75°C for extension. Reactions were terminated by incubation in 7.5 µl 1 M NaOH with 2 mM EDTA at 65°C for 10 min. The cDNA was phenol-chloroform-isoamyl extracted and ethanol precipitated in the presence of 0.1 µg/µl linear acrylamide (Ambion, Austin, TX). The cDNA, resuspended in 16 µl DEPC H₂O, was passed through a Bio-6 chromatography column (Bio-Rad, Cambridge, MA) and washed three times with 700 µl DEPC-treated H₂O. Samples were lyophilized to 16 µl. For the second round of amplification, 16 µl of purified full-length double-stranded cDNA was incubated with 4 µl of each 75 mM NTP (ATP, GTP, CTP, and UTP), 4 µl 10x reaction buffer, and 4 µl transcription enzyme mixture (MEGAscript T7 kit number 1334; Ambion) in 40 µl vol at 37°C for 5 h. RNA recovery and removal of template DNA was achieved by TRIzol purification (Life Technologies). The aRNA was prepared using 1 µg or less of one-amplification aRNA prepared from source total RNA that was reverse transcribed into cDNA using 2 µg random hexamer with 5 µl first-strand buffer, 2 µl 0.1 M DTT, 1 µl RNAsin, 2 µl of 10 mM dNTP, and 2 µl of SuperScript II. The reaction mixture was heated to 65°C for 10 min before adding SuperScriptII, and then synthesis was continued at 42°C for 1 h. Second-strand cDNA synthesis was initiated by 1 µg oligo(dT) T7 primer in the conditions used in the first round. In vitro transcription of aRNA was conducted as for the first round. Reverse transcription reaction from T RNA and aRNA was accomplished by adding 2 µg random hexamer, 4 µl first strand-reaction buffer, 2 µl 0.1 M DTT (Life Technologies), 2 µl 10 mM dNTP, and 2 µl SuperScript-II reverse transcriptase (Life Technologies). The cDNA synthesis was completed at 42°C for 1 h, and cDNA was stored at -30°C until ready for qRT-PCR analysis.

qRT-PCR anlaysis

Measurement of gene expression was performed using the ABI Prism 7700 sequence detection system (PerkinElmer, Foster City, CA) as previously described (20, 21). Primers and TaqMan probes (Custom Oligonucleotide Factory, Foster City, CA) were designed to span exon-intron junctions to prevent amplification of genomic DNA and to result in amplicons <150 bp to enhance efficiency of PCR amplification. Only the primers and probes for MAGE-3 and MAGE-12 did not span exon-intron junctions due to the extreme 5' position of these boundaries in these genes; alternatively, these primers and probes were designed to incorporate areas of the known antigenic regions of these genes. Genomic DNA amplification was further reduced by using aRNA, which selectively amplifies mRNA, in addition to multiple aRNA purification steps. TaqMan probes were labeled at the 5' end with the reporter dye molecule 6-carboxy-fluorescein (FAM; emission _max = 518 nm) and at the 3' end with the quencher dye molecule 6-carboxytetramethyl-rhodamine (TAMRA; emission _max = 582 nm). The cDNA standards were generated by reverse-transcriptase primer-specific amplification of mRNA of the relevant genes using a technique identical with the one used for the preparation of test cDNA. The resulting cDNA was then purified and quantified by spectrophotometry (OD₂₆₀). Copies were calculated using the m.w. of individual gene amplicons. RT-PCR of cDNA specimens and cDNA standards were conducted in a total volume of 25 µl with 1x TaqMan Master Mix (PerkinElmer) and primers and probes at optimized concentrations. We have previously presented the sequences of the primer probe pairs for gp100 (22), -actin (22), MAGE-12 (17), tyrosine-related protein (TRP)2, MAGE-3, and MART-1/MelanA (23). The sequences for the other genes are: Fas-associated death domain-like IL-1-converting enzyme-like inhibitory protein (FLIP), FAM-TCAAACGTATCTTGAAGATGGACAGAAAAGCTG-TAMRA, AGAGTGAGGCGATTTGACCTG, and AAGGTGAGGGTTCCTGAGCA; and TRP-1, FAM-TGATGAATGGCTGAGGAGATACAATGCTGATATA-TAMRA, TCCTGCACACCTTCACAGATG, and TGGCACCATGTTGTATTGTCTATTATG. Thermal cycler parameters included 2 min at 50°C, 10 min at 95°C, 40 cycles involving denaturation at 95°C for 15 s, and annealing/extension at 60°C for 1 min. Real-time monitoring of fluorescent emission from cleavage of sequence-specific probes by the nuclease activity of Taq polymerase allowed definition of the threshold cycle during the exponential phase of amplification (20). Standard curves were generated for each gene and found to have excellent PCR amplification efficiency (90–100% with 100% meaning that, in each cycle, the amount of template is doubled) as determined by the slope of the standard curves. Linear regression analysis of all standard curves was 0.99. Standard curve extrapolation of copy number was performed for the gene of interest as well as an endogenous reference gene for each sample. To correct for cellularity and concentration of starting material, normalization of samples was performed by dividing the copies of the gene of interest by copies of the reference gene.

Immunohistochemistry (IHC)

Immunocytochemistry was performed on cytospins and frozen sections by standard methods (24). Without exception, multiple cytospins could be obtained from each FNA, and comparisons of expression of MDA within each lesion were determined from cytospins prepared from the same aspiration. Cytospins and frozen sections were fixed in acetone for 10 min, then rinsed in PBS and blocked in 3–5% goat serum in PBS for 20 min. The following primary mAbs were used: HMB-45 for assessment of gp100/Pmel-17 expression (Biogenex, San Ramon, CA), M2-7C10 for the assessment of MART-1/MelanA expression (6, 25), and KS-1 for the assessment of HLA-A2 Ag expression (26). Primary Ab was applied for 2 h at room temperature. After washing, the slides were incubated with goat anti-mouse IgG secondary Ab followed by avidin-biotin-peroxidase complex (Vector Laboratories, Burlingame, CA). For color development, 3,3'-diaminobenzidine solution (Sigma, St. Louis, MO) was applied for 6 min at room temperature. For tumor samples that were heavily pigmented, 3-amino-9-ethylcarbazole, which stains red, was used for color development to provide better contrast. The slides were rinsed in H₂0, counterstained with hematoxylin, and cover-slipped. All specimens were processed and stained by the same researcher (P. Fetsch) and read in a blinded fashion by two pathologists (A. Fillie and A. Abati).

The level of TAA and/or HLA Ag expression within a tumor population was measured by two criteria: 1) the percentage of tumor cells within a tumor population that were positive for the relevant Ag and grouped as negative, < 25%, 25–50%, 50–75%, and > 75%; and 2) the intensity of tissue staining scored from 0 to 4+, with 0 being equivalent to a negative control using nonspecific myeloma-associated protein for background staining and 4+ staining equivalent to the relevant mAb of a control positive cell line. For this analysis, all the malignant cells included in a cytospin were counted. The number of cells present in each cytospin varied; however, at least 100 cells were present and evaluated in each sample. The IHC score (sIHC) consisted of the prevalence of cells expressing a given Ag (percentage) multiplied by the overall intensity of expression for each specimen using the methods described above.

In vitro sensitization assay

Parallel in vitro sensitization of cryopreserved pre- and postvaccination PBMC was performed as previously described (27, 28). Briefly, PBMC were thawed into complete medium consisting of Iscove’s modified DMEM with 25 mM HEPES buffer (Biofluids), 10% heat-inactivated human AB serum (Pel-Freez Biologicals, Brown Deer, WI), and 100 µg/ml of streptomycin (Sigma). Cells were resuspended at 1.5 x 10⁶ cells/ml in 2 ml containing 1 µM sensitizing peptide. After 1–2 days, 300 IU/ml rIL-2 (Chiron, Emeryville, CA) was added to the culture. On day 5 and/or when cultures became acidic, complete medium (1 ml) was withdrawn and replaced with fresh complete medium containing IL-2. After 10–13 days from the original stimulation, PBMC were tested by 24-h coculture with T2 cells pulsed with 1 µM peptide. Supernatants were tested for IFN- concentration by ELISA (Endogen, Cambridge, MA).

Statistical analysis

To emphasize the central tendency of the variables presented independently of the nonsymmetric nature of expected changes in TAA expression with time, data are reported nonparametrically using median values. However, statistical values obtained with nonparametric vs parametric methods were similar. Wilcoxon signed rank test and Student’s t test for paired samples were used to compare values derived from pre- and posttreatment FNA samples. Correlation between values obtained with different source materials or between various methods of assessment of gene expression was performed using the Spearman correlation test. Regression analysis was used for parametric assessment of linearity of the relationship between two parameters.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Efficiency and fidelity of primary and secondary aRNA amplification and fidelity of tertiary qRT-PCR amplification

The efficiency of aRNA amplification was tested from limiting dilutions of source total RNA prepared from a melanoma cell line (A375; Table I). An 10²- to 10³-fold higher amount of aRNA from source mRNA (1% of total RNA) was obtained with one round of amplification. With a second amplification, an additional 10- to 100-fold increase in aRNA was obtained. Because of the very low amounts of starting total RNA yielded by some FNA, it was decided to apply two rounds of amplification to all specimens used in this study. This strategy minimizes differences secondary to possible aRNA amplification bias and provides consistency among different specimens.

View larger version (21K): [in this window] [in a new window]   FIGURE 1. Design and location of primer and probes for the different genes analyzed with the TaqMan-based qRT-PCR system. Their sequences are reported separately in the text when unpublished or referenced if previously published. Gene names are reported according to the Human Gene Nomenclature Committee. White arrows indicate the primer location, whereas red boxes indicate the location of the probe. Vertical lines within the genes demonstrate the location of introns. Green inserts within the genes identify sequences associated with T cell epitopes. Yellow regions represent large nontranslated regions. Exons are indicated with the letter "e." Gene sequences for MART-1/Melan A and MAGE genes have been previously published (23 ). -actin sequence was obtained by the GenBank entries X00351 and M10277. Primer/probe pairs for TRP-1 and TRP-2 were also designed according to published information (46–48). FLIP gene structure is inferred from the characterized functional subunits of the caspase and related gene sequences (DED, death effector domains; Ref. 34 ).

View larger version (22K): [in this window] [in a new window]   FIGURE 2. A, Scatter plot of estimated TAA transcripts using as a template for qRT-PCR either aRNA or total RNA. Data is presented as log10 of transcript copy number over 105 copies of -actin. Spearman’s correlation values are presented in the figure. Similar R2 values were obtained when linear correlation was assessed parametrically (not shown). The trend equation is displayed as a power-based regression formula because the data are expressed on a logarithmic scale. B, Insert, estimated gp100/Pmel-17 transcript copy number over 105 copies of -actin (y-axis) and expression of gp100/Pmel-17 according to sIHC (x-axis). This score represents the product of (the percentage of cancer cells within a cytospin staining for gp100/Pmel-17) x (the subjectively assigned intensity of staining of the same cells graded from 1 to 4 by a blinded pathologist). Main graph, the data shown are identical with those shown in the dashed-margins box in the insert and demonstrate the linearity of the relationship between aRNA-derived data and sIHC.

Gene expression was compared with expression of the respective protein product according to IHC for gp100/Pmel-17 and MART-1/MelanA, for which reliable mAbs were available to us. TAA transcript levels adjusted to -actin levels were compared in pairs to the sIHC of expression by the same lesion. In this fashion, the estimated overall amount of protein in a given specimen was correlated with the overall amount of transcript estimated to be present in the same specimen independently from the level of expression in individual cells and the frequency of cell subpopulations in each specimen. gp100 aRNA expression was compared with sIHC in 149 consecutive FNA from melanoma metastases (Fig. 2B, insert). Nonparametric ranking of the two data sets suggested a tight link between them (Spearman’s = 0.5, p < 0.0001), but there was not a linear correlation because aRNA values reached a plateau in association with sIHC >50. However, if FNA samples with sIHC values 50 were tested, a strong linear correlation was noted (Fig. 2B; Spearman’s = 0.7, p < 0.0001; and R² = 0.5, p value < 0.001). This suggests that aRNA-based RT-PCR yields informative data at low levels of protein expression but is not quantitatively informative above this threshold. This limitation was not intrinsic to the aRNA amplification method because a nearly identical pattern was noted when total RNA was used as template for qRT-PCR (data not shown). Thus, in samples with high transcript abundance, TaqMan-based qRT-PCR may have a limited ability to discriminate variations in levels of gene expression. Because the purpose of the study was to evaluate the kinetics of TAA expression throughout the natural history of melanoma or in response to immunological treatment, this model fit the requirements for such analysis. Comparative analysis of MART-1/MelanA expression using aRNA vs sIHC yielded comparable results (data not shown). Furthermore, tumor cells in metastases from HLA-A*0201 Ag-expressing patients who received vaccination with a TAA including a T cell epitope associated with HLA-A2 presentation were tested for HLA-A2 Ag expression. Loss of HLA-A2 Ag was not detected in any case. These findings excluded, in this cohort of patients, loss of the HLA class I allele associated with the vaccine epitope as a factor affecting the interpretation of the results.

Heterogeneity of tumors

Analysis of 116 synchronous metastases from 58 simultaneous FNA biopsies in 58 patients was then performed to evaluate the heterogeneity of expression of various MDA and CT at a given time point. The expression of all TAA appeared to be heterogeneous in a large percentage of synchronous lesions. If 1 log₁₀ difference in expression was considered significant, all TAA were heterogeneously expressed in 40–50% of the lesions with the exception of MART-1/MelanA, which was heterogeneously expressed in only 28% of them. If a more stringent parameter was used to define differences of expression, more variability in the pattern of expression of various TAA was noted. For instance, if differences of at least 3 log₁₀ were considered, then heterogeneity was identified in 16, 14, 12, 10, 10, and 9% of lesions for MAGE-3, MAGE-12, gp100/Pmel-17, TRP-1, MART-1/MelanA, and TRP-2, respectively.

Short-term changes in expression of MDA and CT in melanoma metastases

The expression of gp100/Pmel-17, MART-1/MelanA, TRP-1 and TRP-2, and MAGE-3 and MAGE-12 was assessed in 49 metastases from 28 patients who were undergoing different immunological treatments (Tables II and III). Three additional lesions from two patients regressed during treatment so rapidly (42–56 days) that no tumors suitable for biopsy were observed when the patients returned for follow-up. Therefore, 31% (16 of 52) of the lesions exhibited a complete regression, and 40% of the patients (12 of 30) followed prospectively had at least one lesion that regressed. This is a proportion of responses comparable to historical results, suggesting that repeated FNA does not substantially affect the ability of tumor masses to respond to immunologic manipulation. (We should emphasize that these rates refer to individual lesions and should not be confused with clinical response rates, in which the overall clinical outcome is accounted). FNA were obtained before treatment and during treatment in an interval ranging from 1–4 mo from the initiation of treatment. At the initial time point, the various TAA were expressed with the following frequencies: 85, 85, 67, 63, 90, and 76% for gp100/Pmel-17, MART-1/MelanA, TRP-1, TRP-2, MAGE-3, and MAGE-12, respectively. Neither the frequency nor their level of expression predicted response of a given lesion to treatment. We then analyzed changes with time in TAA expression in all lesions independently of treatment received. Fig. 3A (left panel) shows the kinetics of expression of the TAA studied. We noted a significant decrease in expression of gp100/Pmel-17 (Wilcoxon singed rank test, p₂ < 0.0; Student’s t test, p < 0.05) and no significant change in any of the other TAA studied. Of the metastases followed by this study, 42 were from patients who had received a gp100/Pmel-17-specific vaccine (Table III). Thirty-six lesions were in patients who had received repeated s.c. administrations of a peptide modified from the natural gp100_209–217 epitope by substituting a methionine for a tyrosine in position 210 to induce stronger immunogenicity in vitro and in vivo (28). The remaining six lesions were from patients who received the s.c. administration of an HLA-A*0301-restricted gp100 epitope (ALLAVGATK; Ref. 33) or intramuscular administration of a pox-virus encoding the full-length gp100/Pmel-17 protein. Therefore, we stratified the data according to treatment and noted that the significant decrease in gp100/Pmel-17 gene expression could be noted in particular in patients who had been vaccinated against this TAA (Wilcoxon singed rank test, p < 0.05; Fig. 3A, right panel). No significant change in gp100/Pmel-17 expression was noted in lesions from patients not immunized against this TAA. However, because only 8 lesions were included in this subgroup, the power of the analysis is not sufficient to allow definitive conclusions.

View larger version (19K): [in this window] [in a new window]   FIGURE 3. A, Percentage of change in expression of four representative TAA in a short-term follow-up of the same melanoma metastases (n = 49, left panel) and melanoma metastases from patients undergoing gp100/Pmel-17-specific vaccination (n = 42, right panel). The units represent the percentage of change from pre- (100% by definition) to posttreatment values. Data are presented as median of estimated TAA transcript abundance to better represent the skews in the population studied (*, Wilcoxon signed rank test; p2 < 0.05; Student’s t test for paired sample, p <0.05). B, Percentage of change in expression of four representative TAA in a short-term follow-up of the identical melanoma metastases undergoing regression upon immunological treatment (n = 13, left panel) and progressing (n = 36, right panel). (**, Wilcoxon signed rank test, p2 < 0.01; Student’s t test for paired sample, p <0.01).

We then analyzed the kinetics of TAA expression in lesions stratified according to their clinical behavior. A total of 16 lesions from 12 patients underwent complete regression in response to treatment. The median interval between initiation of treatment (and harvest of the pretreatment FNA) and documentation of lesion disappearance was 118 days with a range spanning between 42 and 243 days (Fig. 4). Three lesions from two patients disappeared so quickly that no posttreatment FNA could be performed. A total of 13 lesions from 10 patients that underwent a complete regression in response to treatment could be serially sampled during treatment, and of those, changes in TAA expression were documented. FNA biopsies were usually obtained at 6-wk intervals. These intervals sometimes varied in response to the clinical needs of the patient and the healthcare team. The median time for posttreatment FNA harvest in responding lesions was 70 days with a range spanning between 21 and 116 days from the beginning of treatment. Similar time points were compared for nonresponding lesions whose FNA were obtained within a median interval from initiation of treatment of 71 days (range of 25–140 days).

View larger version (15K): [in this window] [in a new window]   FIGURE 4. Time elapsed between treatment initiation and disappearance of metastases undergoing complete regression. Three metastases regressed so promptly that no FNA could be obtained before their disappearance. In 13 metastases, FNA could be obtained when tumor deposits were still present, and cancer cells could be identified in the FNA specimens by IHC evaluation. The time elapsed for the accrual of the 13 posttreatment FNA is shown as a dashed line.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Probes and primers to be used for measurement of gene expression by quantitative PCR were selected, when possible, according to the following criteria: specificity and uniqueness for the given gene, intron-spanning range, and linkage to epitopic sequences. These criteria could be fully satisfied for some of the genes but not others. For instance, primers and probes for gp100/Pmel-17 and MART-1/MelanA shared little homology with any other known genes, were intron spanning, and were relatively close to described epitopic sequences. In contrast, MAGE-3 primer and probes shared 90–95% homology with other MAGE genes (particularly MAGE-6), although they were epitope spanning. Because epitopic sequences in the MAGE genes appear to be clustered in the central region of exon 3 (30), it was impossible to design intron-spanning primer/probe pairs including immunologically relevant regions. As a consequence, the MAGE-12 primer/probe pair was also not intron-spanning, although it did not share homology with other MAGE genes and spanned a recently described HLA-Cw*0702-associated epitope (17, 33). Primer/probe pairs for other genes such as TRP-1 and TRP-2 were designed to be intron spanning and without homology with other known genes; however, their linkage with epitopic sequences was low. The concern that selection of non-intron-spanning primer/probes for the MAGE genes might affect qRT-PCR measurements because of genomic contamination is excluded by the aRNA amplification strategy that is highly specific for mRNA. Finally, a criterion not sought in this study was the preferential selection of 5'-located primer and probes. This criterion was not adopted because the method applied here limits the measurement of gene expression to full-length transcripts insensitive to the location of the primer/probe set by exploiting the template-switching effect at the 5' end (19). Although transcript size bias could be introduced by this method, previous work with cDNA microarrays does not support this concern (19).

The information obtained in this study suggests that, given an effective treatment, immune selection can rapidly occur in some of the target tissues, although some lesions regressed without evidence of selection, and some lesions that exhibited decrease in Ag expression did not regress. Because most patients had been actively immunized against gp100/Pmel-17 throughout the duration of this study, it is likely that the preferential loss of expression of the same TAA in responding lesions is more than coincidental. Interestingly, loss of gp100/Pmel-17 did not interfere with regression of the metastases. This suggests that the interaction between gp100/Pmel-17-specific T cells and their targets could represent only an initiating event toward tumor regression. More importantly, in the group of lesions that did not respond to therapy, overall, gp100/Pmel-17 did not decrease significantly (although few exceptions were noted), suggesting that loss of target TAA expression is not in itself the primary reason for failure of these treatment protocols. This result is not a complete surprise, as we have already noted that, in a small cohort of patients receiving TAA-specific vaccination, evidence of vaccine-elicited T cell reactivity could be detected in tumors expressing target TAA. However, the dialogue initiated between tumor cells and vaccine-elicited T cells was not sufficient per se to induce clinical regression (22, 34). The possibility that lack of clinical response is due predominantly to insufficient inflammatory response at the tumor site rather than tumor cell escape from a vigorous Ag-specific CTL response is also supported by a recently concluded study.³ In this study, comparison of molecular phenotypes in pre- vs posttreatment FNA samples of lesions that responded compared with those that did not respond identified significantly different gene expression patterns in posttreatment samples from responding lesions and no significant changes among nonresponding lesions.

These findings do not exclude that immune selection may lead to immune escape later on in the long-term natural history of cancer, as previously shown by others (3, 13). Loss of TAA or HLA Ags might play a later role in the metastatic process by allowing TAA-deprived cells that have survived the short-term effects of treatment to grow undisturbed and reconstitute new lesions (16). However, this phenomenon can be documented only with a longer-term analysis that was beyond the purposes of this study.

No correlation was noted between evidence of systemic sensitization against gp100 and regression of metastatic lesions or loss of gp100 expression. This phenomenon has been previously described (28) and suggests that factors other than the presence of circulating immune cells play a role in the localization and effector function of vaccine-elicited T cells at the tumor site. For consistency with previous reports from our group (28), the monitoring data reported was obtained by a comparative single in vitro sensitization of pre- and postvaccination PBMC. We have previously shown that this method yields comparable results to other methods directly assessing precursor T cells’ number and function, such as enumeration with tetrameric HLA/epitope complexes or evidence of specific IFN- expression by intracellular cytokine staining and/or qRT-PCR assessment (35, 36). However, these studies have shown a strong correlation of results obtained with in vitro sensitization that maintains the highest sensitivity in the detection of vaccine-responsive cells. These data emphasize the necessity of studying tumor host interactions where they are more likely to occur within the tumor microenvironment as we have recently emphasized (18, 22).

Difficult to explain is the decreased expression of the other MDA noted in responding lesions in some patients. Theoretically, immune selection operates by destroying cells bearing a specific target Ag and should be insensitive to the expression of irrelevant Ags. However, the mechanism(s) responsible for the reduced expression of gp100/Pmel-17 may be common to other MDA. In addition, we cannot exclude that the patients may have developed an immune response to the other MDA whose expression was found to be reduced. Therefore, broader effects of systemic IL-2 administration or epitope spreading in the tumor microenvironment might have been at the basis of the extended loss of MDA (37, 38). However, this explanation would leave unanswered the question of why epitope spreading would preferentially affect MDA compared with CT, whose expression was not affected in responding lesions.

In lesions responding to therapy, a statistically significant reduction in expression of three MDA was noted (gp100/Pmel-17, MART-1/MelanA, and TRP-2) with no simultaneous reduction in CT. This loss of MDA expression was not always directly related to the specificity of the vaccine because most patients had been vaccinated against gp100/Pmel-17 in combination with IL-2 and not against the other two MDA. Thus, it could be hypothesized that, in response to the inflammatory insult provided either by the specific or the nonspecific immune stimulation, cells more sensitive to death signals might have cleared more rapidly. Alizadeh et al. (39) have recently noted over-expression of genes whose products inhibit programmed cell death in B cell lymphomas most resistant to conventional treatment. For instance, they identified FLIP as an antiapoptosis factor that has been associated with poor prognosis and responsiveness to treatment of large cell lymphomas (39). It is possible that more anaplastic cells might resist immune-related insult in the melanoma model as well, and in this case, MDA and CT would represent only bipolar markers of the level of differentiation of various cell populations within a metastasis. However, decreased expression of gp100/Pmel-17 was highly specific, and FLIP expression was no significantly different between pretreatment lesions that did or did not regress with treatment (data not shown). Finally, it is possible that the decreased expression of MDA transcript observed in this study could have been related to a decreased number of cancer cells in the posttreatment specimens. However, the unchanged expression of CT Ags expressed only by cancer cells and not normal cells (40) suggests that this is not a likely explanation.

Whatever the explanation for the decreased expression of MDA in some responding lesions, this study provides the surprising demonstration that, before a tumor metastasis disappears, a profound change in expression of target TAA can occur, whether immunologically mediated or not. Yet this loss of epitopic determinants does not seem to affect the final outcome, perhaps because of epitope spreading secondary to the primary insults specific to the vaccine or in relation to a more generalized anti-neoplastic effect of the systemic administration of IL-2. Also, interestingly, minimal and insignificant changes in TAA expression (particularly those targeted by the vaccination) were observed in lesions that did not respond to treatment suggesting that, at least in this human model, this is not the primary reason for the lack of effectiveness of these treatments. Perhaps lack of effective immunization against the target epitopes is the culprit in these cases, or the loss or lack of a secondary inflammatory signal within these nonresponding lesions (41). In addition, we cannot totally exclude that changes in MDA expression might have no effect on the recognition of melanoma cells by CTL. Finally, although addressed in a more controlled system based on sequential biopsies of the same lesion before and after treatment, the studies of this paper are not, per se, totally surprising. Still, the major question of the mechanism(s) directly involved in tumor regression remains unanswered, and such question warrants continued investigation in this vein.

	Footnotes

 
¹ Address correspondence and reprint requests to Dr. Francesco M. Marincola, Surgery Branch, Department of Transfusion Medicine, Clinical Center, National Cancer Institute, Building 10, Room 2B42, 10 Center Drive, MSC 1502, Bethesda, MD 20892-1502. E-mail address: marincola{at}nih.gov

² Abbreviations used in this paper: TAA, tumor-associated Ag; MDA, melanoma differentiation Ag; aRNA, anti-sense RNA; IHC, immunohistochemistry; FNA, fine needle aspiration biopsy, MDA, melanoma differentiation Ag; qRT-PCR, quantitative real-time PCR; sIHC, immunohistochemistry score; CT, cancer testis; DEPC, diethyl pyrocarbonate; FAM, 6-carboxy-fluorescein; TAMRA, 6-carboxytetramethyl-rhodamine; FLIP. Fas-associated death domain-like IL-1-converting enzyme-like inhibitory protein; TRP, tyrosine-related protein.

³ E. Wang et al. Submitted for publication.

⁴ E. Wang et al. Submitted for publication.

Received for publication March 13, 2001. Accepted for publication May 29, 2001.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Old, L. J., Y. T. Chen. 1998. New paths in human cancer serology. J. Exp. Med. 187:1163.[Free Full Text]
2. Marincola, F. M., E. M. Jaffe, D. J. Hicklin, S. Ferrone. 2000. Escape of human solid tumors from T cell recognition: molecular mechanisms and functional significance. Adv. Immunol. 74:181.[Medline]
3. Jager, E., M. Ringhoffer, M. Altmannsberger, M. Arand, J. Karbach, D. Jager, F. Oesch, A. Knuth. 1997. Immunoselection in vivo: independent loss of MHC class I and melanocyte differentiation antigen expression in metastatic melanoma. Int. J. Cancer 71:142.[Medline]
4. de Vries, T. J., A. Fourkour, T. Wobbes, G. Verkroost, D. J. Ruiter, G. N. van Muijen. 1997. Heterogeneous expression of immunotherapy candidate proteins gp100, MART-1, and tyrosinase in human melanoma cell lines and in human melanocytic lesions. Cancer Res. 57:3223.[Abstract/Free Full Text]
5. Scheibenbogen, C., I. Weyers, D. Ruiter, M. Willhauck, A. Bittinger, U. Keilholz. 1996. Expression of gp100 in melanoma metastases resected before or after treatment with IFN and IL-2. J. Immunother. 19:375.[Medline]
6. Cormier, J. N., Y. M. Hijazi, A. Abati, P. Fetsch, M. Bettinotti, S. M. Steinberg, S. A. Rosenberg, F. M. Marincola. 1998. Heterogeneous expression of melanoma-associated antigens (MAA) and HLA-A2 in metastatic melanoma in vivo. Int. J. Cancer 75:517.[Medline]
7. Riker, A., U. S. Kammula, M. C. Panelli, E. Wang, A. Abati, P. Fetsch, K.-H. Lee, S. M. Steinberg, S. A. Rosenberg, F. M. Marincola. 1999. Immune selection following antigen specific immunotherapy of melanoma. Surgery 126:112.[Medline]
8. Anichini, A., A. Molla, R. Mortarini, G. Tragni, I. Bersani, M. Di Nicola, A. M. Gianni, S. Pilotti, R. Dunbar, V. Cerundolo, G. Parmiani. 1999. An expanded peripheral T cell population to a cytotoxic T lymphocyte (CTL)-defined, melanocyte-specific antigen in metastatic melanoma patients impacts on generation of peptide-specific CTLs but does not overcome tumor escape from immune surveillance in metastatic lesions. J. Exp. Med. 190:651.[Abstract/Free Full Text]
9. Bittner, M., P. Meltzer, Y. Chen, E. Jiang, E. Seftor, M. Hendrix, M. Radmacher, R. Simon, Z. Yakhini, A. Ben-Dor, et al 2000. Molecular classification of cutaneous malignant melanoma by gene expression: shifting from a continuous spectrum to distinct biologic entities. Nature 406:536.[Medline]
10. Coulie, P. G., B. J. van den Eynde, P. van der Bruggen, A. Van Pel, T. Boon. 1997. Antigens recognized by T-lymphocytes on human tumours. Biochem. Soc. Trans. 25:544.[Medline]
11. De Smet, C., O. De Backer, I. Faraoni, C. Lurquin, F. Brasseur, T. Boon. 1996. The activation of human gene MAGE-1 in tumor cells is correlated with genome-wide demethylation. Proc. Natl. Acad. Sci. USA 93:7149.[Abstract/Free Full Text]
12. Liteplo, R. G., R. S. Kerbel. 1987. Reduced levels of DNA 5-methylcytosine in metastatic variants of the human melanoma cell line MeWo. Cancer Res. 47:2264.[Abstract/Free Full Text]
13. Jager, E., M. Ringhoffer, J. Karbach, M. Arand, F. Oesch, A. Knuth. 1996. Inverse relationship of melanocyte differentiation antigen expression in melanoma tissues and CD8⁺ cytotoxic-T-cell responses: evidence for immunoselection of antigen-loss variants in vivo. Int. J. Cancer 66:470.[Medline]
14. Natali, P. G., R. Cavaliere, A. Bigotti, M. R. Nicotra, C. Russo, A. K. Ng, P. Giacomini, S. Ferrone. 1983. Antigenic heterogeneity of surgically removed primary and autologous metastatic human melanoma lesions. J. Immunol. 130:1462.[Abstract]
15. Hicklin, D. J., F. M. Marincola, S. Ferrone. 1999. HLA class I antigen down-regulation in human cancers: T-cell immunotherapy revives an old story. Mol. Med. Today 5:178.[Medline]
16. Lee, K.-H., M. C. Panelli, C. J. Kim, A. Riker, M. Roden, P. A. Fetsch, A. Abati, M. P. Bettinotti, S. A. Rosenberg, F. M. Marincola. 1998. Functional dissociation between local and systemic immune response following peptide vaccination. J. Immunol. 161:4183.[Abstract/Free Full Text]
17. Panelli, M. C., M. P. Bettinotti, K. Lally, Y. Li, G. A. Ohnmacht, P. A. Robbins, A. Riker, S. A. Rosenberg, F. M. Marincola. 2000. Identification of a tumor-infiltrating lymphocyte recognizing MAGE-12 in a melanoma metastasis with decreased expression of melanoma differentiation antigens. J. Immunol. 164:4382.[Abstract/Free Full Text]
18. Wang, E., F. M. Marincola. 2000. A natural history of melanoma: serial gene expression analysis. Immunol. Today 21:619.[Medline]
19. Wang, E., L. Miller, G. A. Ohnmacht, E. Liu, F. M. Marincola. 2000. High fidelity mRNA amplification for gene profiling using cDNA microarrays. Nat. Biotechnol. 17:457.
20. Heid, C. A., J. Stevens, K. J. Livak, P. M. Williams. 1996. Real time quantitative PCR. Genome Res. 6:986.[Abstract/Free Full Text]
21. Kruse, N., M. Pette, K. Toyka, P. Rieckmann. 1997. Quantification of cytokine mRNA expression by RT PCR in samples of previously frozen blood. J. Immunol. Methods 210:195.[Medline]
22. Kammula, U. S., K.-H. Lee, A. Riker, E. Wang, G. A. Ohnmacht, S. A. Rosenberg, F. M. Marincola. 1999. Functional analysis of antigen-specific T lymphocytes by serial measurement of gene expression in peripheral blood mononuclear cells and tumor specimens. J. Immunol. 163:6867.[Abstract/Free Full Text]
23. Riker, A., U. S. Kammula, M. C. Panelli, E. Wang, G. A. Ohnmacht, S. M. Steinberg, S. A. Rosenberg, F. M. Marincola. 2000. Threshold levels of gene expression of the melanoma antigen gp100 correlate with tumor cell recognition by cytotoxic T-lymphocytes. Int. J. Cancer 86:818.[Medline]
24. Marincola, F. M., Y. M. Hijazi, P. Fetsch, M. L. Salgaller, L. Rivoltini, J. N. Cormier, T. B. Simonis, P. H. Duray, M. Herlyn, Y. Kawakami, S. A. Rosenberg. 1996. Analysis of expression of the melanoma associated antigens MART-1 and gp100 in metastatic melanoma cell lines and in in situ lesions. J. Immunother. 19:192.
25. Kawakami, Y., J. K. Battles, T. Kobayashi, W. Ennis, X. Wang, J. P. Tupesis, F. M. Marincola, P. F. Robbins, V. J. Hearing, M. A. Gonda, S. A. Rosenberg. 1997. Production of recombinant MART-1 proteins and specific anti-MART-1 polyclonal and monoclonal antibodies: use in the characterization of the human melanoma antigen MART-1. J. Immunol. Methods 202:13.[Medline]
26. Tsujisaki, M., K. Sakaguchi, M. Igarashi, P. Richiardi, F. Perosa, S. Ferrone. 1988. Fine specificity and idiotype diversity of the murine anti-HLA- A2, A28 monoclonal antibodies CR11-351 and KS1. Transplantation 45:632.[Medline]
27. Cormier, J. N., M. L. Salgaller, T. Prevette, K. C. Barracchini, L. Rivoltini, N. P. Restifo, S. A. Rosenberg, F. M. Marincola. 1997. Enhancement of cellular immunity in melanoma patients immunized with a peptide from MART-1/Melan A. Cancer J. Sci. Am. 3:37.[Medline]
28. Rosenberg, S. A., J. C. Yang, D. Schwartzentruber, P. Hwu, F. M. Marincola, S. L. Topalian, N. P. Restifo, E. Dufour, L. Schwartzberg, P. Spiess, et al 1998. Immunologic and therapeutic evaluation of a synthetic tumor-associated peptide vaccine for the treatment of patients with metastatic melanoma. Nat. Med. 4:321.[Medline]
29. Kawakami, Y., S. A. Rosenberg. 1997. Immunobiology of human melanoma antigens MART-1 and gp100 and their use for immuno-gene therapy. Int. Rev. Immunol. 14:173.[Medline]
30. DePlaen, E., K. Arden, C. Traversari, J. J. Gaforio, J. P. Szikora, C. De Smet, F. Brasseur, P. van der Bruggen, B. Lethe, C. Lurquin, et al 1994. Structure, chromosomal localization and expression of 12 genes of the MAGE family. Immunogenetics 40:360.[Medline]
31. Engzell, U., P. A. Jakobsson, A. Sigurdson, J. Zajicek. 1971. Aspiration biopsy of metastatic carcinoma in lymph nodes of the neck: a review of 1101 consecutive cases. Acta. Otolaryngol. 72:138.[Medline]
32. Behm, F. G., G. J. O’Dowd, W. J. Frable. 1984. Fine-needle aspiration effects on benign lymph node histology. Am. J. Clin. Pathol. 82:195.[Medline]
33. Heidecker, L., F. Brasseur, M. Probst-Kepper, M. Gueguen, T. Boon, B. Van den Eynde. 2000. Cytolytic T lymphocytes raised against a human bladder carcinoma recognize an antigen encoded by gene MAGE-A12. J. Immunol. 164:6041.[Abstract/Free Full Text]
34. Panelli, M. C., A. Riker, U. S. Kammula, K.-H. Lee, E. Wang, S. A. Rosenberg, F. M. Marincola. 2000. Expansion of tumor/T cell pairs from fine needle aspirates (FNA) of melanoma metastases. J. Immunol. 164:495.[Abstract/Free Full Text]
35. Kammula, U. S., F. M. Marincola, S. A. Rosenberg. 2000. Real-time quantitative polymerase chain reaction assessment of immune reactivity in melanoma patients after tumor peptide vaccination. J. Natl. Cancer Inst. 92:1336.[Abstract/Free Full Text]
36. Nielsen, M.-B., V. Monsurro, S. Miguelse, E. Wang, A. Perez-Diez, K.-H. Lee, U. S. Kammula, S. A. Rosenberg, F. M. Marincola. 2000. Status of activation of circulating vaccine-elicited CD8⁺ T cells. J. Immunol. 165:2287.[Abstract/Free Full Text]
37. el-Shami, K., B. Tirosh, E. Bar-Haim, L. Carmon, E. Vadai, M. Fridkin, M. Feldman, L. Eisenbach. 1999. MHC class I-restricted epitope spreading in the context of tumor rejection following vaccination with a single immiunodominant epitope. Eur. J. Immunol. 29:3295.[Medline]
38. Disis, M. L., K. H. Grabstein, P. R. Sleath, M. A. Cheever. 1999. 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. 5:1289.[Abstract/Free Full Text]
39. Alizadeh, A. A., M. B. Eisen, R. E. Davis, C. Ma, I. S. Lossos, A. Rosenwald, J. C. Bedrick, H. Sabet, T. Tran, Y. Xin, et al 2000. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 403:467.[Medline]
40. Driscoll, M. C., A. Chu, M. W. Hilgartner. 1996. Heteroduplex analysis in hemophilia B: detection of two novel factor IX gene mutations. Am. J. Hematol. 51:324.[Medline]
41. Fuchs, E. J., P. Matzinger. 1996. Is cancer dangerous to the immune system?. Semin. Immunol. 8:271.[Medline]

This article has been cited by other articles:

				M. Sensi and A. Anichini Unique Tumor Antigens: Evidence for Immune Control of Genome Integrity and Immunogenic Targets for T Cell-Mediated Patient-Specific Immunotherapy Clin. Cancer Res., September 1, 2006; 12(17): 5023 - 5032. [Abstract] [Full Text] [PDF]

				C. F. Basil, Y. Zhao, K. Zavaglia, P. Jin, M. C. Panelli, S. Voiculescu, S. Mandruzzato, H. M. Lee, B. Seliger, R. S. Freedman, et al. Common cancer biomarkers. Cancer Res., March 15, 2006; 66(6): 2953 - 2961. [Abstract] [Full Text] [PDF]

				M. A. Valasek and J. J. Repa The power of real-time PCR Advan Physiol Educ, September 1, 2005; 29(3): 151 - 159. [Abstract] [Full Text] [PDF]

				F. Avogadri, C. Martinoli, L. Petrovska, C. Chiodoni, P. Transidico, V. Bronte, R. Longhi, M. P. Colombo, G. Dougan, and M. Rescigno Cancer Immunotherapy Based on Killing of Salmonella-Infected Tumor Cells Cancer Res., May 1, 2005; 65(9): 3920 - 3927. [Abstract] [Full Text] [PDF]

				C. J. Wheeler, A. Das, G. Liu, J. S. Yu, and K. L. Black Clinical Responsiveness of Glioblastoma Multiforme to Chemotherapy after Vaccination Clin. Cancer Res., August 15, 2004; 10(16): 5316 - 5326. [Abstract] [Full Text] [PDF]

				G. Parmiani, C. Castelli, P. Dalerba, R. Mortarini, L. Rivoltini, F. M. Marincola, and A. Anichini Cancer Immunotherapy With Peptide-Based Vaccines: What Have We Achieved? Where Are We Going? J Natl Cancer Inst, June 5, 2002; 94(11): 805 - 818. [Abstract] [Full Text] [PDF]

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