Diversity of the fine specificity displayed by HLA-A*0201-restricted CTL specific for the immunodominant Melan-A/MART-1 antigenic peptide

HLA-A*0201 melanoma patients often develop a CTL response to an immunodominant peptide derived from the melanocyte lineage-specific protein Melan-A/MART-1. We have shown previously that the antigenic peptide most often involved is the decapeptide Melan-A(26-35) (EAAGIGILTV). We also observed some clonal diversity in the fine specificity of Melan-A-specific CTL. To substantiate this observation, we have now tested a series of Melan-A(26-35) variant peptides containing single alanine substitutions for binding to HLA-A*0201 and recognition by polyclonal and monoclonal Melan-A-specific CTL. Substitution of several residues with alanine reduced peptide binding activity by > 10-fold. In contrast, substitution of E26 with alanine (AAAGIGILTV) resulted in a 5-fold higher binding activity as well as in stronger stability of the corresponding HLA-A*0201/peptide complexes. Interestingly, the peptide variant AAAGIGILTV was recognized more efficiently than the natural decapeptide by short term cultured, tumor-infiltrated lymph node cell cultures and a number of Melan-A-specific CTL clones derived from different individuals. Moreover, this analysis revealed that the fine specificity of the CTL response to the Melan-A immunodominant epitope is quite diverse at the clonal level. At least three distinct patterns of fine specificity were identified. This diversity appears to reflect the diversity of the TCR repertoire available for this Ag, since similar results were obtained with a panel of Melan-A-specific CTL clones derived from a single melanoma patient. These findings have important implications for the formulation of Melan-A peptide-based vaccines as well as for the monitoring of Melan-A-specific CTL responses in melanoma patients.     

Cell population changes during atrophy and regeneration of rat parotid gland

Melan-A/MART-1 is a melanoma differentiation antigen that is recognized by a high proportion of cytolytic T lymphocyte (CTL) clones derived from human leukocyte antigen (HLA)-A2+ melanoma patients. Whereas peptide Melan-A/ MART-1(27-35) was originally defined as the immunodominant CTL epitope, we have previously reported that peptide Melan-A/MART-1(26-35) was recognized more efficiently by the majority of tumor-reactive CTL clones. As demonstrated here, CTL populations generated from blood lymphocytes of either melanoma patients or healthy individuals after in vitro stimulation with peptide Melan-A/MART-1(26-35) killed specifically HLA-A2+ Melan-A+ allogeneic melanoma cells, thus suggesting their potential use in adoptive immunotherapy. We characterized the surface phenotype of the circulating CTL precursors (CTLp), which respond to in vitro stimulation with peptide Melan-A/MART-1(26-35). In melanoma patients, these CTLp predominantly expressed the CD45RO memory marker. In contrast, they were mainly, although not exclusively, found in the CD45RA subpopulation of CD8 T cells in healthy individuals. The demonstration that Melan-A/MART-1-specific CTLp in peripheral blood lymphocytes from HLA-A2+ patients with metastatic melanoma express a memory phenotype provides direct evidence that in vivo priming of this antigen may occur during tumor progression.     

Binding and presentation of peptides derived from melanoma antigens MART-1 and glycoprotein-100 by HLA-A2 subtypes. Implications for peptide-based immunotherapy

Cellular immune responses to melanoma-associated Ags are the focus of ongoing studies aimed at developing immunotherapies for treatment of malignant melanoma. Melanoma predominantly affects Caucasians, a population in whom expression of HLA-A2 is prevalent. Among HLA-A2 subtypes, HLA-A*0201 is widely expressed, and HLA-A*0201-restricted, tumor-reactive CTL responses are well studied. We have observed in a group of melanoma patients an unexpectedly high frequency (approximately 20%) of non-HLA-A*0201 subtypes (*0202, *0204, and *0205), and little is known regarding antimelanoma response profiles in patients expressing such subtypes. We analyzed non-HLA-A*0201 peptide response profiles using HLA-A*0201-restricted epitopes from melanoma Ags MART-1/Melan A and glycoprotein 100. Most of these peptides bound to the majority of subtypes tested with 50% inhibitory concentrations less than 500 nM. Recognition of cells pulsed with different peptides (MART-1(27-35), G9(154), and G9(280) Flu M1(58-66)) and expressing different subtype molecules by HLA-A*0201-restricted CTL was limited to only a subset of non-HLA-A*0201 molecules, and the peptide/subtype complexes recognized varied among the effector populations tested. CTL responses elicited from PBL of patients and healthy donors expressing subtypes HLA-A*0202 and HLA-A*0205 suggested significant differences among HLA-A2 subtype function in the context of melanoma Ag presentation. These observations imply the necessity of subtyping patients considered for peptide-based protocols and highlight the need for further study of melanoma-directed cellular responses among patients expressing non-HLA-A*0201 subtypes.     

Cytolytic T lymphocyte recognition of the immunodominant HLA-A*0201-restricted Melan-A/MART-1 antigenic peptide in melanoma

The Melan-A/MART-1 gene product is frequently recognized by tumor-specific HLA-A2-restricted CTL. An immunodominant nonapeptide has been localized to the region spanning residues 27-35. However, the decapeptide including residues 26-35 (the nonapeptide extended NH2 terminally by one residue) appeared to be recognized as efficiently as the nonapeptide. In this study, we show that the optimal length immunodominant peptide appears to correspond to the decapeptide 26-35, as assessed by quantitative analyses of both 4 polyclonal and 13 monoclonal populations of specific CTL. Functional assays of peptide binding to HLA-A2 indicate that the decapeptide is significantly a more efficient binder than the nonapeptide. Moreover, analogues of the decapeptide including substitutions at a secondary HLA-A2 peptide anchor further improve decapeptide binding. Finally, we show that the functional (9 CTL clones analyzed) and structural TCR repertoire (7 CTL clones) of a group of specific CTL clones is rather diverse. The findings reported here may have important implications for future peptide-based melanoma vaccination trials as well as for the monitoring of specific CTL responses in vivo.     

Human melanoma patients recognize an HLA-A1-restricted CTL epitope from tyrosinase containing two cysteine residues: implications for tumor vaccine development

To identify shared epitopes for melanoma-reactive CTL restricted by MHC molecules other than HLA-A*0201, six human melanoma patient CTL lines expressing HLA-A1 were screened for reactivity against the melanocyte differentiation proteins Pmel-17/gp100, MART-1/Melan-A, and tyrosinase, expressed via recombinant vaccinia virus vectors. CTL from five of the six patients recognized epitopes from tyrosinase, and recognition of HLA-A1+ target cells was strongly correlated with tyrosinase expression. Restriction by HLA-A1 was further demonstrated for two of those tyrosinase-reactive CTL lines. Screening of 119 synthetic tyrosinase peptides with the HLA-A1 binding motif demonstrated that nonamer, decamer, and dodecamer peptides containing the sequence KCDICTDEY (residues 243-251) all reconstituted the CTL epitope in vitro. Epitope reconstitution in vitro required high concentrations of these peptides, which was hypothesized to be a result of spontaneous modification of cysteine residues, interfering with MHC binding. Substitution of serine or alanine for the more N-terminal cysteine prevented modification at that residue and permitted target cell sensitization at peptide concentrations 2 to 3 orders of magnitude lower than that required for the wild-type peptide. Because spontaneous modification of sulfhydryl groups may also occur in vivo, tumor vaccines using this or other cysteine-containing peptides may be improved by amino acid substitutions at cysteine residues.     

HLA-independent heterogeneity of CD8+ T cell responses to MAGE-3, Melan-A/MART-1, gp100, tyrosinase, MC1R, and TRP-2 in vaccine-treated melanoma patients

An important element in melanoma vaccine construction is to identify peptides from melanoma-associated Ags that have immunogenic potential in humans and are recognized by CD8+ T cells in vivo. To identify such peptides, we evaluated HLA-A*02+ melanoma patients immunized to a polyvalent vaccine containing multiple Ags, including MAGE-3, Melan-A/MART-1, gp100, tyrosinase, melanocortin receptor (MC1R), and dopachrome tautomerase (TRP-2). Using a filter spot assay, we measured peripheral blood CD8+ T cell responses, before and after immunization, to a panel of 45 HLA-A*0201-restricted peptides derived from these Ags. The peptides were selected for immunogenic potential based on their strong binding affinity in vitro to HLA-A*0201. Vaccine treatment induced peptide-specific CD8+ T cell responses to 22 (47.8%) of the peptides. The most striking finding was the HLA-independent heterogeneity of responses to both peptides and Ags. All responding patients reacted to different combination of peptides and Ags even though the responding patients were all A*0201+ and the peptides were all A*0201-restricted. From 9 to 27% of patients developed a CD8+ T cell response to at least one peptide from each Ag, but no more than 3 (14%) reacted to the same peptide from the same Ag. This heterogeneity of responses to individual peptides and Ags in patients with the same haplotype points to the need to construct vaccines of multiple peptides or Ags to maximize the proportion of responding patients.     

The 2,4-dichlorophenol hydroxylase of Alcaligenes eutrophus JMP134 is a homotetramer

Several epitopes in the human melanoma antigens recognized by HLA-A2-restricted CTLs have a relatively low MHC-binding affinity and as a result may be expressed at very low densities on the cell surface, indicating that these epitopes may not be efficient immunogens. To express these epitopes at higher densities on the surface of antigen-presenting cells and therefore improve their immunogenicity, a DNA construct in which a cDNA fragment encoding the melanoma epitope MART-1(27-35) or gp100(280-288) was inserted between sequences encoding the leader and the HLA-A*0201 protein. Cells transfected with these epitope-HLA fusion constructs were recognized by HLA-A2-restricted melanoma-reactive CTLs specific for the MART-1 or gp100 epitope. In addition, tumor-reactive CTLs could be induced from PBMCs of patients with metastatic melanoma by in vitro stimulation with HMY-C1R B-cell lines expressing the MART-1 or gp100 epitope-HLA-A*0201 fusion protein. These epitope-HLA fusion constructs may be useful for the development of immunotherapies for patients with melanoma.     

Induction of HLA-A2-restricted CTLs to the mucin 1 human breast cancer antigen

HLA-A*0201/Kb transgenic mice were immunized with oxidized mannan-mucin 1 (MUC1) as a fusion protein (containing five repeats of the 20-amino-acid MUC1 VNTR (variable number of tandem repeats) that generated highly active CD8+ CTLs to MUC1 peptides. In a direct CTL assay, the MUC1 peptides could be presented specifically by both the transgenic murine HLA-A*0201/Kb and human HLA-A*0201 molecules. The 9-mer MUC1 peptide sequences (APDTRPA and STAPPAHGV) were presented by HLA-A*0201, although they did not contain L at P2 and L/V at P9, the preferred motifs; as a consequence, the binding was of relatively low affinity when compared with a high affinity-binding HIV peptide (ILKEPVHGV). In addition, when mice were immunized separately with the HLA-A*0201-binding peptides (STAPPAHGV or APDTRPAP-containing peptides-keyhole limpet hemocyanin-mannan), direct lysis of MCF-7 (HLA-A*0201+, MUC1+) also occurred. The findings are of interest for tumor immunotherapy, particularly as the CTLs generated to low affinity-binding peptides were highly active and could specifically lyse an HLA-A*0201+ human breast cancer cell line without further in vitro stimulation. The findings demonstrate that the range of peptides that can generate CTLs is broader than formerly considered.     

Lack of strong immune selection pressure by the immunodominant, HLA-A*0201-restricted cytotoxic T lymphocyte response in chronic human immunodeficiency virus-1 infection

Despite detailed analysis of the HIV-1-specific cytotoxic T lymphocyte response by various groups, its relation to viral load and viral sequence variation remains controversial. We analyzed HLA-A*0201 restricted cytotoxic T lymphocyte responses in 17 HIV-1-infected individuals with viral loads ranging from < 400 to 221,000 HIV RNA molecules per milliliter of plasma. In 13 out of 17 infected subjects, CTL responses against the SLYNTVATL epitope (p17 Gag; aa 77-85) were detectable, whereas two other HLA-A*0201 restricted epitopes (ILKEPVHGV, IV9; and VIYQYMDDL, VL9) were only recognized by six and five individuals out of 17 individuals tested, respectively. Naturally occurring variants of the SL9 epitope were tested for binding to HLA-A*0201 and for recognition by specific T cell clones generated from five individuals. Although these variants were widely recognized, they differed by up to 10,000-fold in terms of variant peptide concentrations required for lysis of target cells. A comparison of viral sequences derived from 10 HLA-A*0201-positive individuals to sequences obtained from 11 HLA-A*0201-negative individuals demonstrated only weak evidence for immune selective pressure and thus question the in vivo efficacy of immunodominant CTL responses present during chronic HIV-1 infection.     

The effect of epitope variation on the profile of cytotoxic T lymphocyte responses to the HIV envelope glycoprotein

To address the relationship between viral and host factors during HIV infection, we analyzed the effect of viral mutations on T cell responses in seropositive, asymptomatic HLA-A2+ individuals using four envelope (env)-specific peptides with the HLA-A*0201 binding motif. We showed that the natural sequence variation was frequent within epitopes located in the C-terminal region of the env glycoprotein and was largely responsible for a lower env-specific cytotoxic T lymphocyte (CTL) activity in the peptide-stimulated cultures. The highest CTL responses in vitro were induced with conserved epitopes D1 and 4.3 that mapped to the N-terminal region of the env glycoprotein. These peptides exhibited high binding affinity for HLA-A*0201 molecules and stimulated CD8+ T cells of relatively limited TCR Vbeta chain repertoire. Decreased CTL activities to the D1 epitope were observed in the absence of any detectable viral mutation, and were associated with lower proliferative responses and expression of the CD28 antigen. Results of this study demonstrate that the degree of sequence variation within a stimulatory epitope of the viral quasispecies, as well as proliferative potential of the effector cells, are among the factors underlying decreased CTL activity in HIV-infected patients. These experiments also provide evidence that the D1 peptide might be useful for the development of vaccines and immune-based therapy.     

Identification of new melanoma epitopes on melanosomal proteins recognized by tumor infiltrating T lymphocytes restricted by HLA-A1, -A2, and -A3 alleles

To isolate melanoma Ags recognized by T cells, cDNA libraries made from melanoma cell lines were screened with four CTLs derived from tumor infiltrating lymphocytes (TIL) that were able to recognize melanoma cells in a HLA-A1, -A2, or -A3 restricted manner. Although cDNAs encoding the previously identified melanoma Ags, tyrosinase and gp100, were isolated, these TIL were found to recognize previously unidentified peptides. An HLA-A1-restricted CTL, TIL1388, was found to recognize a tyrosinase peptide (SSDYVIPIGTY), and an HLA-A3-restricted CTL, TIL1351, recognized a gp100 peptide (LIYRRRLMK). CTL clones isolated from the HLA-A2-restricted TIL1383 recognized a gp100 peptide (RLMKQDFSV). HLA-A2-restricted CTL, TIL1200, recognized a gp100 peptide (RLPRIFCSC). Replacement of either cysteine residue with alpha-amino butyric acid in the gp100 peptide, RLPRIFCSC, enhanced CTL recognition, suggesting that the peptide epitope naturally presented on the tumor cell surface may contain reduced cysteine residues. Oxidation of these cysteines might have occurred during the course of the synthesis or pulsing of the peptide in culture. These modifications may have important implications for the development of efficient peptide-based vaccines. These newly identified peptide epitopes can extend the ability to perform immunotherapy using synthetic peptides to a broader population of patients, especially those expressing HLA-A1 or HLA-A3 for whom only a few melanoma epitopes have previously been identified.     

Enhanced in vitro potency and in vivo immunogenicity of a CTL epitope from hepatitis C virus core protein following amino acid replacement at secondary HLA-A2.1 binding positions

Since the natural immune response to hepatitis C virus (HCV) is often unable to clear the infection, to enhance immunogenicity we studied substituted peptides from an HCV cytotoxic T lymphocyte (CTL) epitope (C7A2) from a conserved region of the HCV core protein (DLMGYIPLV) recognized by CTL lines from HLA-A2.1(+) HCV-infected patients and HLA-A2.1 transgenic mice. HLA-A2.1 binding, human and murine CTL recognition, and in vivo immunogenicity (using mice transgenic for human HLA-A2 in lieu of immunizing humans) were analyzed to define peptides with enhanced immunogenicity. Peptides substituted at position 1 showed enhanced HLA-A2 binding affinity, but paradoxically poorer immunogenicity. A peptide with Ala substituted at position 8 (8A) showed higher HLA-A2 binding affinity and CTL recognition and was a more potent in vivo immunogen in HLA-A2-transgenic mice, inducing higher CTL responses with higher avidity against native C7A2 than induced by C7A2 itself. These results suggest that peptide 8A is a more potent in vitro antigen and in vivo immunogen than C7A2 and may be useful as a vaccine component. They provide proof of principle that the strategy of epitope enhancement can enhance immunogenicity of a CTL epitope recognized by human CTL.     

Shared epitopes for HLA-A3-restricted melanoma-reactive human CTL include a naturally processed epitope from Pmel-17/gp100

Human CD8+ CTL recognize peptides bound to class I MHC molecules on the surface of melanoma cells. Several peptides derived from melanocyte lineage-specific proteins have been identified as epitopes for HLA-A2 restricted melanoma-reactive CTL. Because less than half of melanoma patients express HLA-A2, it is important to identify CTL epitopes restricted by other common MHC molecules including HLA-A1 and -A3. We have generated HLA-A3-restricted human CTL that recognize one or more shared melanoma Ags. All of the melanomas recognized by one of these CTL lines express Pmel-17/gp100, and those that fail to express this Ag are not lysed. This CTL line also specifically recognizes the lymphoblastoid line C1R-A3 following infection with a recombinant vaccinia encoding the melanocyte lineage-specific protein Pmel-17/gp100. Thus, at least one Pmel-17/ gp100 peptide is an epitope for this CTL line. We have identified ALLAVGATK (Pmel-17/gp100 residues 17-25) as an epitope for this CTL line and have shown that it is naturally processed and presented by HLA-A3 on melanoma cells. A second HLA-A3-restricted melanoma-reactive CTL line recognizes at least one additional shared epitope. These findings suggest that cellular immune responses directed against multiple shared melanoma epitopes exist in the 20 to 25% of melanoma patients who express HLA-A3. In addition, immunotherapy directed against Pmel-17/gp100 and other shared melanoma Ags may be useful in a large subset of these patients.     

A hsp70-2 mutation recognized by CTL on a human renal cell carcinoma

We performed T cell cloning experiments with a tumor-infiltrating lymphocyte subpopulation derived from a renal cell carcinoma tumor site (RCC-7) in which the TCR clonotypic repertoire had been analyzed in terms of TCRBV complementarity-determining region 3 size distribution. We report in this work the characterization of one of the five RCC-specific MHC class I-restricted CTL clones isolated in RCC-7. This TCRBV6J1S1 CTL recognized only the autologous RCC-7 tumor cell line in the context of HLA-A*0201, and the Ag is encoded by a mutated form of the hsp70-2 gene found in the tumor cells, but not in autologous PBLs nor in 47 other tumors. The identification of this gene was achieved by cotransfecting into COS cells a cDNA library of RCC-7 together with HLA-A*0201. Transfectants expressing the Ag were identified by their ability to stimulate TNF release by the CTL clone. The antigenic peptide is a decamer with a mutated residue at position 8. Half-maximal lysis was obtained with only 5 x 10(-11) M of decapeptide in target sensitization assays compared with 5 x 10(-8) M for the wild-type decapeptide. This difference in recognition was not related to difference in binding HLA-A*0201-presenting molecules, as assessed in an immunofluorescence-based peptide-binding assay using T2 cells. Constitutive hsp70 expression in various tumors suggests that this stress-induced protein may be recognized in situ by tumor-infiltrating lymphocytes. The finding in the tumor of a mutated form of the stress-induced hsp70-2 gene whose product is specifically recognized by TILs with high avidity is discussed in view of the present use of mycobacteria or heterologous heat-shock proteins as immunomodulators or as subunit vaccine candidates.     

Tumor defense by murine cytotoxic T cells specific for peptide bound to nonclassical MHC class I

Cytotoxic T Cells (CTLs) can exhibit considerable antitumor activity. Thus far, the characterized tumor peptide antigens recognized by CTLs are all presented by classical MHC class Ia molecules [human lymphocyte antigen A (HLA-A), HLA-B, and HLA-C in humans and H-2K, H-2D, and H-2L in mice]. Here we show that CTLs recognized peptides presented by nonclassical MHC class Ib molecule Qa-1b expressed by tumor cells. These CTLs conferred in vivo protection by delaying the growth of Qa-1b-expressing B78H1 melanoma cells pulsed with Qa-1b-binding peptides Cw4L or B35L and injected s.c. in C57BL/6 mice. A hierarchy of the peptides was found with regard to their ability to trigger CTLs; Cw4L stimulated a strong CTL response. The closely related and cross-reactive peptide B35L induced a weaker CTL response but was still efficient in sensitizing the target cells. Finally, Qa-1b-expressing melanoma cells without exogenous peptides were not immunogenic but possibly expressed endogenous cross-reactive antigenic peptides. The data are compatible with earlier findings that CTL activation requires relatively strong peptide antigens, whereas subsequent effector functions are also mediated by weak peptide analogues. In conclusion, CTLs mediated tumor immunity through the recognition of peptides presented by nonclassical MHC class Ib molecules. The identification of similar CTLs in humans may facilitate the vaccination of cancer patients because MHC class Ib/peptide complexes are much less polymorphic than MHC class Ia/peptide complexes.     

Functional dissociation between local and systemic immune response during anti-melanoma peptide vaccination

Peptide vaccination against tumor Ags can induce powerful systemic CTL responses. However, in the majority of patients, no tumor regression is noted. To study this discrepancy, we analyzed CTL reactivity in a melanoma patient (F001) vaccinated with g209-2M peptide, a single residue variant of gp100(209-217). G209/g209-2M-reactive CTL were identified in post- but not prevaccination PBL. Limiting dilution analysis identified one predominant CTL clone (C1-35), with TCR Vbeta6s2, recognizing g209/HLA-A*0201-expressing targets. Additionally, two autologous melanoma lines (F001TU-3 and -4) and 20 separate tumor-infiltrating lymphocyte cultures were generated from a fine needle aspirate of a metastatic lesion progressing after initial response to vaccination. Both F001TU did not express gp100 and were not recognized by C1-35. Loss of gp100 by F001TU correlated with a marked reduction of gp100 expression in the same metastatic lesion compared with prevaccination. Thus, ineffectiveness of C1-35 and tumor progression could be best explained by loss of target Ag expression. Interestingly, 12 of 20 tumor-infiltrating lymphocyte cultures recognized F001TU, but none demonstrated g209/g209-2M reactivity, suggesting a functional dissociation between systemic and local immune response. This study suggests that vaccination effects must be analyzed in the target tissue, rather than in the systemic circulation alone.     

Whole blood production of thromboxane, prostacyclin and leukotriene B4 after dietary fish oil supplementation in man: effect of vitamin E

Antigenic peptides derived from several differentiation antigens of the melanocyte lineage were recently identified in human melanomas as targets for HLA-A2.1-restricted cytotoxic T lymphocytes (CTLs). To examine their potential role in tumour-directed immune responses in vivo, we determined CTL reactivity against seven antigenic peptides derived from the Melan A/MART-1, tyrosinase and gp100/Pmel17 antigens in the peripheral blood of 10 HLA-A2+ healthy controls and 26 HLA-A2+ melanoma patients. The influenza matrix peptide (GILGFVFTL) presented by HLA-A2.1 was used as a control peptide. CTL reactivity was assessed in a mixed lymphocyte 'peptide' culture assay. Reactivity against Melan A/MART-1-derived peptide antigens was readily detectable in both melanoma patients and controls. Reactivity directed against tyrosinase-derived peptide antigens was also detected in both melanoma patients and healthy individuals, but less frequently. A measurable response against gp100/Pmel17-derived antigens was found in 1/10 controls and in 1/26 of the melanoma patients. Reactivity against the influenza matrix peptide was common in both melanoma patients and controls. Our findings show that precursor CTLs against melanocyte differentiation antigens can be detected in peripheral blood of melanoma patients and healthy individuals. The pattern of CTL reactivity directed against melanoma-associated antigens does not seem to be altered in melanoma patients. Despite antigen-specific CTL reactivity, tumour growth was not prevented in melanoma patients and autoimmune phenomena were not detected in healthy individuals. It remains to be determined whether precursor CTLs recognizing melanocyte differentiation antigens can be activated by immunization and lead to effective tumour rejection in vivo.     

Analysis of MAGE-3-specific cytolytic T lymphocytes in human leukocyte antigen-A2 melanoma patients

The MAGE-3 gene is a member of a multigene family that is selectively expressed by subsets of different human tumor types, including malignant melanoma, but not by normal tissues except for testis and placenta. A cytolytic T lymphocyte (CTL)-defined MAGE-3 antigen, corresponding to the MAGE-3 peptide 271-279 associated with the human leukocyte antigen (HLA)-A2 molecule, has been recently identified using T lymphocytes from a normal individual stimulated in vitro with peptide-pulsed autologous antigen-presenting cells. Because MAGE-3 is expressed in 76% of metastatic melanomas, the HLA-A2-restricted MAGE-3 antigen should be expressed by approximately 37% of Caucasians bearing a metastatic melanoma tumor, thus representing an attractive candidate for the elicitation of specific CTL immune responses in vivo. In this study, we determined the proportion of HLA-A2+ melanoma patients displaying detectable MAGE-3 peptide 271-279-specific CTL precursors in peripheral blood. Peptide-specific CTL populations were obtained from at least 4 of 11 HLA-A2+ patients. Peptide-specific CTL lines derived from these populations readily lysed HLA-A2-positive target cells that were pulsed with MAGE-3 peptide 271-279 at nanomolar concentrations yet were unable to recognize (as assessed by cytolysis and cytokine production) MAGE-3-expressing autologous or allogeneic HLA-A2-positive melanoma lines. Similarly, the CTL lines failed to recognize MAGE-3-negative HLA-A2-positive tumor lines after transfection with the MAGE-3 gene, although they were able to recognize COS-7 cells transfected with MAGE-3. In contrast, HLA-A1-positive melanoma lines transfected with MAGE-3 were efficiently recognized by CTL lines directed against the MAGE-3 peptide 168-176, a known HLA-A1-restricted CTL epitope. These results suggest that the expression level of the MAGE-3 peptide 271-279, unlike that of MAGE-3 peptide 168-176, in melanomas may be too low to allow efficient recognition by specific CTLs. Thus, it appears that despite the presence of CTL precursors against MAGE-3 peptide 271-279 in some HLA-A2+ melanoma patients, the usefulness of this peptide for specific immunotherapy of melanoma may be limited.     

Cell-mediated immunological responses in cervical and vaginal cancer patients immunized with a lipidated epitope of human papillomavirus type 16 E7

Human papillomavirus (HPV) infection has been causally associated with cervical cancer. We tested the effectiveness of an HLA-A*0201-restricted, HPV-16 E7 lipopeptide vaccine in eliciting cellular immune responses in vivo in women with refractory cervical cancer. In a nonrandomized Phase I clinical trial, 12 women expressing the HLA-A2 allele with refractory cervical or vaginal cancer were vaccinated with four E786-93 lipopeptide inoculations at 3-week intervals. HLA-A2 subtyping was also performed, and HPV typing was assessed on tumor specimens. Induction of epitope-specific CD8+ T-lymphocyte (CTL) responses was analyzed using peripheral blood leukapheresis specimens obtained before and after vaccination. CTL specificity was measured by IFN-gamma release assay using HLA-A*0201 matched target cells. Clinical responses were assessed by physical examination and radiographic images. All HLA-A*0201 patients were able to mount a cellular immune response to a control peptide. E786-93-specific CTLs were elicited in 4 of 10 evaluable HLA-A*0201 subjects before vaccination, 5 of 7 evaluable HLA-A*0201 patients after two vaccinations, and 2 of 3 evaluable HLA-A*0201 cultures after all four inoculations. Two of three evaluable patients' CTLs converted from unreactive to reactive after administration of all four inoculations. There were no clinical responses or treatment toxicities. The ability to generate specific cellular immune responses is retained in patients with advanced cervical cancer. Vaccination with a lipidated HPV peptide epitope appears capable of safely augmenting CTL reactivity. Although enhancements of cellular immune responses are needed to achieve therapeutic utility in advanced cervical cancer, this approach might prove useful in treating preinvasive disease.     

Identification of HLA-A3-restricted cytotoxic T lymphocyte epitopes from carcinoembryonic antigen and HER-2/neu by primary in vitro immunization with peptide-pulsed dendritic cells

The human carcinoembryonic antigen (CEA) and HER-2/neu are potential target antigens for CTL specific immunotherapy for common malignancies such as breast, lung, colon, and gastric carcinomas. Several CTL epitopes restricted by HLA-A2, the most common human histocompatibility molecule, have been previously reported. However, to develop CTL-based immunotherapies for the general population, it is necessary to identify epitopes restricted by other common histocompatibility alleles. Here, we describe two HLA-A3-restricted CTL epitopes from the CEA and HER-2/neu antigens. HLA-A3 binding synthetic peptides from CEA and HER-2/neu were tested for immunogenicity by in vitro primary CTL induction protocol using peripheral blood mononuclear cells from normal healthy volunteers. One peptide from CEA (CEA[9(61)]: HLFGYSWYK) and one peptide from HER-2/neu (HER2[9(754)]: VLRENTSPK) were shown to induce CTL that was capable of killing a tumor cell line expressing HLA-A3 and the corresponding tumor-associated antigen. Additional MHC binding studies with the most common HLA molecules belonging to the HLA-A3 superfamily (HLA-A*1101, -A*3101, -A*3301, and -A*6801), demonstrated that CEA[9(61)] binds five of five A3 supertype molecules with high affinity, and the HER2[9(754)] epitope was able to bind to four of the same five alleles. These results indicate that these two new CTL epitopes should be immunogenic in individuals expressing either HLA-A3, or other members of the HLA-A3 superfamily.