Safety and immunogenicity of the PRAME cancer immunotherapeutic in metastatic melanoma: results of a phase I dose escalation study.

PURPOSE: The PRAME tumour antigen is expressed in several tumour types but in few normal adult tissues. A dose-escalation phase I/II study (NCT01149343) assessed the safety, immunogenicity and clinical activity of the PRAME immunotherapeutic (recombinant PRAME protein (recPRAME) with the AS15 immunostimulant) in patients with advanced melanoma. Here, we report the phase I dose-escalation study segment. PATIENTS AND METHODS: Patients with stage IV PRAME-positive melanoma were enrolled to 3 consecutive cohorts to receive up to 24 intramuscular injections of the PRAME immunotherapeutic. The RecPRAME dose was 20, 100 or 500 µg in cohorts 1, 2 and 3, respectively, with a fixed dose of AS15. Adverse events (AEs), including predefined dose-limiting toxicity (DLT) and the anti-PRAME humoral response (ELISA), were coprimary end points. Cellular immune responses were evaluated using in vitro assays. RESULTS: 66 patients were treated (20, 24 and 22 in the respective cohorts). AEs considered by the investigator to be causally related were mostly grade 1 or 2 injection site symptoms, fatigue, chills, fever and headache. Two DLTs (grade 3 brain oedema and proteinuria) were recorded in two patients in two cohorts (cohorts 2 and 3). All patients had detectable anti-PRAME antibodies after four immunisations. Percentages of patients with predefined PRAME-specific-CD4+T-cell responses after four immunisations were similar in each cohort. No CD8+ T-cell responses were detected. CONCLUSIONS: The PRAME immunotherapeutic had an acceptable safety profile and induced similar anti-PRAME-specific humoral and cellular immune responses in all cohorts. As per protocol, the phase II study segment was initiated to further evaluate the 500 µg PRAME immunotherapeutic dose. TRIAL REGISTRATION NUMBER: NCT01149343, Results.

Prospective assessment of a gene signature potentially predictive of clinical benefit in metastatic melanoma patients following MAGE-A3 immunotherapeutic (PREDICT).

BACKGROUND: Genomic profiling of tumor tissue may aid in identifying predictive or prognostic gene signatures (GS) in some cancers. Retrospective gene expression profiling of melanoma and non-small-cell lung cancer led to the characterization of a GS associated with clinical benefit, including improved overall survival (OS), following immunization with the MAGE-A3 immunotherapeutic. The goal of the present study was to prospectively evaluate the predictive value of the previously characterized GS. PATIENTS AND METHODS: An open-label prospective phase II trial ('PREDICT') in patients with MAGE-A3-positive unresectable stage IIIB-C/IV-M1a melanoma. RESULTS: Of 123 subjects who received the MAGE-A3 immunotherapeutic, 71 (58.7%) displayed the predictive GS (GS+). The 1-year OS rate was 83.1%/83.3% in the GS+/GS- populations. The rate of progression-free survival at 12 months was 5.8%/4.1% in GS+/GS- patients. The median time-to-treatment failure was 2.7/2.4 months (GS+/GS-). There was one complete response (GS-) and two partial responses (GS+). The MAGE-A3 immunotherapeutic was similarly immunogenic in both populations and had a clinically acceptable safety profile. CONCLUSION: Treatment of patients with MAGE-A3-positive unresectable stage IIIB-C/IV-M1a melanoma with the MAGE-A3 immunotherapeutic demonstrated an overall 1-year OS rate of 83.5%. GS- and GS+ patients had similar 1-year OS rates, indicating that in this study, GS was not predictive of outcome. Unexpectedly, the objective response rate was lower in this study than in other studies carried out in the same setting with the MAGE-A3 immunotherapeutic. Investigation of a GS to predict clinical benefit to adjuvant MAGE-A3 immunotherapeutic treatment is ongoing in another melanoma study.This study is registered at www.clinicatrials.gov NCT00942162.

A tyrosinase peptide presented by HLA-B35 is recognized on a human melanoma by autologous cytotoxic T lymphocytes.

We previously described different cytotoxic T lymphocyte (CTL) clones isolated from the blood lymphocytes of a melanoma patient after in vitro stimulation with autologous tumor cells. These CTL clones recognized at least 2 distinct antigens on the melanoma cells. Here, we show that one of them consists of a peptide derived from tyrosinase and presented by HLA-B35. The peptide is 9 amino acids long and has the sequence LPSSADVEF. It can be presented by the 2 major B35 allelic subtypes, B*3501 and B*3503. As HLA-B35 is one of the most frequent HLA-B specificities, being present in about 20% of Caucasian individuals, it may be a useful target for peptide-based immunotherapy of melanoma.

Isolation of tyrosinase-specific CD8+ and CD4+ T cell clones from the peripheral blood of melanoma patients following in vitro stimulation with recombinant vaccinia virus.

The identification of Ags expressed by tumor cells and recognized by autologous T cells has led to the prospect of treating cancer by adoptive transfer of tumor-reactive T cells selected for Ag specificity. Tyrosinase is an Ag expressed by normal melanocytes as well as melanoma cells for which responses by autologous T cells have been detected. To evaluate the frequency with which tyrosinase-specific T cells can be isolated from melanoma patients for potential use in therapy, a recombinant vaccinia virus expressing tyrosinase was constructed for infection of autologous APCs that could be used to stimulate T cells reactive with this protein. Eight patients were studied, with peripheral blood serving as the source of both responder T cells and autologous APCs. Tyrosinase-specific CD8+ CTL clones were isolated from five of the eight patients with melanoma. The tyrosinase-specific CTL generated in this manner recognized autologous tumor cells as well as targets expressing the recombinant virus vector. CTL clones from three of the individuals were restricted to HLA-A28, -B8, and -B60, which have not previously been identified as alleles that can present immunogenic tyrosinase peptides. Tyrosinase-specific CD4+ T cell clones were isolated from six of the eight patients by stimulation with autologous APCs infected with recombinant vaccinia virus, and all these CD4+ clones were capable of recognizing autologous tumor cells. These studies demonstrate a high prevalence of CD4+ and CD8+ tyrosinase-specific responses in peripheral blood and support the feasibility of using peripheral blood to generate T cells for tumor therapy without the requirement for isolating T cells that have infiltrated tumor sites.

A peptide recognized by human cytolytic T lymphocytes on HLA-A2 melanomas is encoded by an intron sequence of the N-acetylglucosaminyltransferase V gene.

A cytolytic T lymphocyte (CTL) clone that lyses many HLA-A2 melanomas was derived from a population of tumor-infiltrating lymphocytes of an HLA-A2 melanoma patient. The gene coding for the antigen recognized by this CTL was identified by transfection of a cDNA library. It is the gene which has been reported to code for N-acetylglucosaminyltransferase V (GnT-V). Remarkably, the antigenic peptide recognized by the CTL is encoded by a sequence located in an intron. In contrast to the fully spliced GnT-V mRNA, which was found in a wide range of normal and tumoral tissues, the mRNA containing the intron region coding for the antigen was not found at a significant level in normal tissues. This mRNA was observed to be present in about 50% of melanomas. Our results suggest that a promoter located near the end of the relevant intron is activated in melanoma cells, resulting in the production of an mRNA coding for the antigen.

A tyrosinase nonapeptide presented by HLA-B44 is recognized on a human melanoma by autologous cytolytic T lymphocytes.

The human tyrosinase gene has been reported previously to code for two distinct antigens recognized on HLA-A2 melanoma cells by autologous cytolytic T lymphocytes (CTL). By stimulating lymphocytes of melanoma patient MZ2 with a subclone of the tumor cell line of this patient, we obtained a CTL clone that lysed this subclone but did not lyse other subclones of the same melanoma cell line. The sensitive melanoma subclone was found to express a much higher level of tyrosinase than the others, suggesting that the antigen recognized by the CTL might be encoded by tyrosinase. Transfection of a tyrosinase cDNA demonstrated that the CTL clone indeed recognized a tyrosinase product presented by HLA-B*4403. The relevant antigenic peptide corresponds to residues 192-200 of the tyrosinase protein. Lymphoblastoid cells of the B*4402 subtype were not recognized by the CTL following incubation with the peptide. Nevertheless, by stimulating in vitro lymphocytes of a healthy HLA-B*4402 donor with autologous adherent cells pulsed with the same peptide, we obtained a CTL clone which recognized tumor cells expressing tyrosinase and HLA-B*4402. As HLA-B44 is expressed in 24% of Caucasians, the tyrosinase-B44 antigen may constitute a useful target for specific immunotherapy of melanoma.

New tumor antigens recognized by T cells.

A series of tumor cell antigens that are recognized by cytolytic T lymphocytes has been characterized this year. Besides the antigens derived from proteins specifically expressed in tumors, many melanoma antigens derive from melanocytic differentiation proteins. In addition, antigens unique to individual tumors result from mutations in ubiquitously expressed genes.

Individual differences in the orientation of the cytolytic T cell response against mouse tumor P815.

We reported previously that the mouse tumor P815 expresses four distinct antigens (A, B, C, D) recognized by syngeneic cytolytic T lymphocytes (CTL). A fifth P815 antigen (E) was identified by means of a CTL clone derived from tumor-infiltrating lymphocytes. We compared a number of mice for the orientation of their CTL response with respect to the various P815 antigens. Lymphocytes from mice inoculated subcutaneously with living P815 cells were stimulated in vitro with tumor cells and the resulting CTL were tested against targets expressing either antigens A and B or antigens C, D and E. Many mice had an asymmetrical response, some producing CTL directed almost exclusively against antigens A, B and others producing CTL directed almost exclusively against C, D. E. When mice were inoculated into two separate sites, different orientations in the responses of the two local lymph nodes were often observed, suggesting that individual differences in the orientation of the anti-P815 CTL response do not result from preexisting differences between the animals. Asymmetrical CTL responses persisted in mice that were given a second injection of tumor cells. A possible interpretation of our results is that the major component of the CTL response is made of the progeny of a very small number of CTL precursors that happen to be the first to be stimulated by the tumor antigens.