The use of IFN-α in clinical oncology has generally been based on the rationale of exploiting its antiproliferative and antiangiogenic activities. However, IFN-α also exhibits enhancing effects on T-cell and dendritic cell functions, which may suggest a novel use as a vaccine adjuvant. We have carried out a pilot phase I-II trial to determine the effects of IFN-α, administered as an adjuvant of Melan-A/MART-1:26-35(27L) and gp100:209-217(210M) peptides, on immune responses in stage IV melanoma patients. In five of the seven evaluable patients, a consistent enhancement of CD8+ T cells recognizing modified and native MART-1 and gp100 peptides and MART-1+gp100+ melanoma cells was observed. Moreover, vaccination induced an increase in CD8+ T-cell binding to HLA tetramers containing the relevant peptides and an increased frequency of CD45RA+CCR7− (terminally differentiated effectors) and CD45RA−CCR7− (effector memory) cells. In all patients, treatment augmented significantly the percentage of CD14+ monocytes and particularly of the CD14+CD16+ cell fraction. An increased expression of CD40 and CD86 costimulatory molecules in monocytes was also observed. Notably, postvaccination monocytes from two of the three patients showing stable disease or long disease-free survival showed an enhanced antigen-presenting cell function and capability to secrete IP10/CXCL10 when tested in mixed leukocyte reaction assays, associated to a boost of antigen and melanoma-specific CD8+ T cells. Although further clinical studies are needed to show the adjuvant activity of IFN-α, the present data represent an important starting point for considering a new clinical use of IFN-α and new immunologic end points, potentially predictive of clinical response. (Cancer Res 2006; 66(9): 4943-51)
IFN-α is a cytokine belonging to type I IFNs, which has been most frequently used in patients with certain types of cancer, including some hematologic malignancies and solid tumors, such as melanoma, renal carcinoma, and Kaposi's sarcoma. Despite many years of work in preclinical as well as in clinical settings, the mechanisms underlying the IFN-induced antitumor response are not well understood. For a long time, it was thought that the direct inhibitory effects on tumor cell growth/function were the major mechanisms of the IFN-mediated antitumor responses in patients. However, early experiments in mouse tumor models have shown that IFN-α plays an important role in the activation of a long-lasting antitumor response (1). Subsequent studies have also provided evidence for a role of type I IFNs in the differentiation of the Th1 subset, as well as in the generation of CTL and in the promotion of the in vivo proliferation and survival of T cells (reviewed in ref. 2). In mouse models, type I IFNs have been shown to be powerful adjuvants when administered with soluble proteins or with the human influenza vaccine (3, 4). Interestingly, in both these studies, the co-injection of type I IFNs with the antigen or the vaccine followed by additional IFN injections, both 1 and 2 days later, proved to be the optimal schedule for the induction of the antibody response (3, 4).
Recently, several studies have shown that type I IFNs promote in vitro the differentiation of monocytes into dendritic cells and can markedly enhance dendritic cell activities (refs. 5–11 and reviewed in ref. 12). However, the pathways by which different subsets of monocytes can differentiate into distinct types of dendritic cells in response to IFNs in vivo remain poorly characterized. Human blood monocytes represent a heterogeneous cell population and several subsets can be distinguished on the basis of the expression of different membrane markers. Monocytes expressing the CD16 lymphocyte marker constitute the main monocyte subpopulation (13–15). Of note, CD14+CD16+ monocytes expressing low or high levels of CD14 seem to develop into dendritic cells or macrophages, respectively (16). In addition, several studies reported an increased number of circulating CD14+CD16+ in the course of infections, inflammation, and malignant diseases (17–20). Another subpopulation of circulating monocytes expresses the CD2 marker and these cells have been described as dendritic cells (21, 22). In particular, we have recently reported that peripheral blood monocytes may differentiate into highly active antigen-presenting cells and CD14+CD2+ monocytes can rapidly acquire the expression of the dendritic cell maturation marker CD83 after a short time (i.e., 4 hours) of incubation in the presence of IFN-α and granulocyte macrophage colony-stimulating factor (GM-CSF; ref. 23). As dendritic cells represent professional antigen-presenting cells for the generation of an immune response, these studies strongly suggest that IFN-α can play a pivotal role in linking innate and adaptive antitumor immunity and can be used as an adjuvant for cancer vaccines.
In spite of the evidence on the immune adjuvant activity of type I IFNs in viral and tumor models (1, 2), these cytokines have never been used as vaccine adjuvants in cancer patients. The main purpose of this pilot trial was to evaluate frequency, phenotype, and function of peptide-specific CD8+ T lymphocytes as well as phenotype and antigen-presenting cell activity of monocytes from stage IV melanoma patients vaccinated with MART-1 and gp100 peptides in combination with IFN-α as a vaccine adjuvant.
This was an open-label, two-institution study to test the feasibility and the toxicity of Melan-A/MART-1 and gp100 peptides given in association with IFN-α as an adjuvant. Ten stage IV (American Joint Committee on Cancer) pretreated metastatic melanoma patients, molecularly typed as human leukocyte antigen HLA A*0201, were enrolled in the study. Patients were enrolled at the Istituto Nazionale per lo Studio e la Cura dei Tumori of Milan (n = 7) and the Istituto Dermopatico dell'Immacolata of Rome (n = 3). Of these patients, seven received at least one vaccine cycle and were considered assessable. Patients' characteristics are reported in Table 1. Additional inclusion criteria were life expectancy of >6 months, Eastern Cooperative Oncology Group performance status of 0 to 1, and adequate hematopoietic, renal and hepatic functions. Patients were excluded if they had (a) brain metastases; (b) a concomitant malignant disease; (c) a severe cardiovascular disease; (d) chronically active infections; (e) a clinically significant autoimmune disease; (f) any illness requiring immunosuppressive therapy, previous chemotherapy, radiotherapy, or biological therapy received within 4 weeks before starting vaccination; (g) were pregnant or lactating; or (h) had a psychiatric illness that would interfere with patient compliance and informed consent. The study was approved by Ethics Committee of the Istituto Nazionale per lo Studio e la Cura dei Tumori and of the Istituto Dermopatico dell'Immacolata; approval and informed written consent were given by all patients. Patients included in the study were vaccinated according to a regimen consisting of two cycles of four vaccinations each (cycle 1: 250 μg of each peptide given i.d. every 2 weeks and 3 MU IFN-α given s.c. on days −1, 0, and +1 with respect to peptides, i.e., day 0; cycle 2: 250 μg of each peptide given monthly with a simultaneous single IFN-α dose of 3 MU). Both peptides and IFN-α were injected in close but separate sites next to local lymph nodes, changing the site of injection every two administrations. Toxicity was documented by evaluating the frequency and intensity of local adverse events as well as the clinically relevant changes in the laboratory variables categorized according to the WHO common toxicity criteria. Ophthalmologic examinations were carried out at 4, 8, and 12 weeks to assess possible autoimmune reactions caused by melanoma/retina cross-reacting differentiation antigens. Tumor evaluation was done at pretreatment visit, before the first (T73) and the fourth (T163) vaccine injection of the second cycle, and thereafter every 3 months and as clinically indicated. All time points of response were recorded from the time of the first peptide injection. This was done by the same sequential diagnostic imaging method. Clinical responses were defined according to Response Evaluation Criteria in Solid Tumors (24).
* All the enrolled patients had stage IV melanoma. Disease status at the time of vaccine start is indicated; NED, no evidence of disease. SC, s.c. lesions; LN, lymph node metastases.
† S, surgery on metastatic lesions; CT, chemotherapy; BT, biotherapy (including IFN, IL-2, or vaccines).
‡ SD, stable disease; PD, progressive disease.
Two melanoma-associated peptides were used in this study: gp100:209-217(210M), IMDQVPFSV; Melan-A/MART-1:26-35(27L), ELAGIGILTV. The peptides were prepared under Good Manufacturing Practice conditions by Clinalfa (Laufelfingen, Switzerland) and were supplied as a water-soluble white powder in vials containing 250 μg of peptide. IFN-α (human leukocyte IFNα; Alfaferone) was supplied by Alfawassermann (Bologna, Italy) in commercial ampoules containing 3 MU of Alfaferone. The native Melan-A/MART-1:27-35 (AAGIGILTV) and gp100:209-217 (ITDQVPFSV) peptides, synthesized as previously described (25), together with the HLA-A*0201-binding peptides derived from HIV-NEF (VLEWRFDSRL), Flu A matrix M1 (GILGFVFTL), gp100 (280-288, YLEPGPVTA), and NY-ESO-1 (157-165V, SLLMWTTQV) proteins, were also used for in vitro studies.
Molecular class I typing was done on genomic DNA by PCR with sequence-specific primers (PCR-SSP), following reported conditions and international guidelines (26, 27). For immunologic monitoring, 30 mL of heparinized blood were obtained from each patient before vaccination (pre-Tx and T0) and 28 and 73 days after the first vaccination. For some patients, samples from the T117 vaccination time point were also available for testing. Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll gradient centrifugation, frozen in aliquots by a computer-assisted freezing system (Microgitcool, IMV Technologies Italia, Piacenza, Italy), and stored in liquid nitrogen. For immunologic assays, PBMCs obtained before and at different times during vaccination were simultaneously thawed and incubated overnight at 37°C in RPMI 1640 (BioWhittaker Europe, Verviers, Belgium) supplemented with 10% AB human serum (BioWhittaker Europe) before doing functional and phenotypic assays. PBMC recovery after thawing was ≥60% whereas viability after overnight incubation was ≥ 95% as assessed by trypan blue staining.
Delayed type hypersensivity (DTH) reaction to the two vaccine peptides was done at the baseline (T0), at T42 (before the fourth vaccine injection), at T117 (between the sixth and the seventh vaccine injection), and then at 2 months after the end of the treatment. At least 5 mm of induration or erythema read 48 hours after intradermal injection was required to score a gp100 or Melan-A/MART-1 skin test as positive. However, no positive DTH reaction was observed.
IFN-γ ELISPOT was done at the Istituto Nazionale per lo Studio e la Cura dei Tumori for all patients as previously described (28) and according to the instructions manufacturer (MabTech, Nacka, Sweden). Data were evaluated by a computer-assisted ELISPOT reader (Bioline, AID, Turin, Italy). PBMCs (1.67 × 10 5 per well) were tested in triplicates against the TAP-deficient line T2 (1.67 × 10 4 per well) pulsed with Melan-A/MART-1 and gp100 native and modified peptides, or with an HIV-derived epitope (NEF), the Flu A matrix M1 peptide, gp100:280-288, and NY-ESO-1:157-165V, used as controls. The HLA-A*0201+ Melan-A/MART-1+gp100+ melanoma cell line 501mel (29) was also included in the assay for evaluating T-cell recognition of endogenously processed antigens. The HLA-A*0201+ colon carcinoma cell line Colo 206 (ref. 30; purchased from American Type Culture Collection, Manassas, VA) was used as negative control. To assess interassay variability, IFN-γ production by the anti–Melan-A/MART-1(27-35) HLA-A*0201-restricted T-cell clone A42 (31), releasing 120 ± 9 spots/500 cells (n = 21) in response to T2 cells pulsed with Melan-A/MART-1(27-35) peptide, was included in each test. HLA blocking experiments were carried out by preincubating target cells with the anti–class I HLA (A, B, and C) immunoglobulin M antibody (clone A6-136; kindly provided by Dr. Daniela Pende, INT, Genoa, Italy) or with the anti-HLA-DR monoclonal antibody (mAb) L243, as negative control (32). For statistical evaluation, a t test for unpaired samples was used to compare prevaccine and postvaccine spots of the same patient or to evaluate statistical significance of HLA blocking experiments. P< 0.05 was considered statistically significant.
HLA-A*0201 tetramers containing Melan-A/MART-1:26-35(27L), Melan-A/MART-1:27-35, gp100:209-217 (210M), or gp100:209-217 were purchased from Beckman Coulter (San Diego, CA). Tetramer binding was evaluated after staining with the iMASC Gating Kit (Beckman Coulter), containing FITC-anti-CD8 mAb, together with PC5-anti-CD4, CD13, and CD19 mAbs. Data were reported as percentage of tetramer+CD8+, CD4/CD13/CD19− cells. As negative control, we used iTAgTM HLA class I human negative tetramers SA-PE (Beckman
