Abstract
Purpose
Chemokines influence both tumor progression and anti-tumor immune response. A 32-bp-deletion polymorphism in the chemokine receptor 5 gene (CCR5Δ32) has been shown to result in a non-functional protein. This study was aimed at evaluating the potential impact of this gene polymorphism on disease progression and treatment outcome in patients with melanoma.
Patients and methods
CCR5 genotyping was performed by PCR on DNA extracted from serum samples of 782 cutaneous melanoma patients with known disease history and long-term clinical follow-up. Genotypes were correlated with patient survival and types of treatment.
Results
Of 782 melanoma patients, 90 (11.5%) were heterozygous and 12 (1.5%) were homozygous for CCR5Δ32. Analyzing the complete cohort, the disease-specific survival from date of primary diagnosis was not influenced by CCR5 status. Similarly, no significant impact could be detected on the treatment outcome of stage III patients. In 139 stage IV patients receiving immunotherapy, CCR5Δ32 was associated with a decreased survival compared to patients not carrying the deletion (median 12.5 vs. 20.3 months, P = 0.029). Multivariate analysis revealed the CCR5 genotype as an independent factor impacting disease-specific survival in this patient population (P = 0.002), followed by gender (P = 0.019) and pathological classification of the primary (pT; P = 0.022).
Conclusion
The presence of the CCR5Δ32 polymorphism in patients with stage IV melanoma results in a decreased survival following immunotherapy and may help to select patients less likely to benefit from this type of treatment.
Keywords: CCR5, Polymorphism, Melanoma, Immunotherapy, Survival
Introduction
Chemotactic cytokines (chemokines) form a family of structurally related, small soluble proteins which are known for their ability to induce leukocyte migration and cellular homing. For example, the ligands for the CC-chemokine receptor 5 (CCR5), namely CCL3, CCL4 (macrophage inflammatory protein (MIP)-1α and MIP-1β), and CCL5 [regulated-upon-activation, T cell expressed and secreted (RANTES)] facilitate CCR5-expressing immunocompetent dendritic cells and CD4+ T cells to migrate towards inflammatory stimuli [1, 2].
The role of chemokines and their receptors in progression and metastasis of cancer, however, is controversial. In fact, chemokines are not only produced and secreted by cells of the immune system but also by tumor cells, including malignant melanoma cells. These chemokines may be involved in the regulation of immune cell migration and activation, local tumor angiogenesis, tumor cell proliferation and metastasis [3–5]. Increased expression of the CCR5 ligand CCL5 was found to be associated with advanced stage disease in breast cancer patients [6], and treatment with a CCR5 antagonist decreased tumor growth in a breast cancer model [7]. Furthermore, CCL5 increased the proliferative and invasive capacities of prostate cancer cells in vitro, with both effects being abolished by the CCR5 antagonist TAK-779 (7). CCL5 also promoted melanoma formation in nude mice [8]. On the other hand, chemokines, including CCR5 ligands, have been shown to promote adaptive anti-tumor immune responses and inhibit tumor progression. In this regard, plasmid-driven overexpression of intra-tumoral CCL5 was associated with increased numbers of natural killer (NK) cells, CD4+ and CD8+ T cells at the tumor site, and resulted in delayed tumor growth; and this effect was lost in CCR5-deficient mice [9]. Decreased tumorigenicity of a hepatocellular carcinoma model was also observed in CCL3-/CCR1-deficient mice and the anti-tumor effect was associated with markedly diminished tumor angiogenesis [10].
A 32 bp deletion in the CCR5 gene (CCR5Δ32) abolishes receptor expression in homozygous individuals and results in a decreased receptor expression in heterozygous carriers compared to wild-type homozygotes [11]. Since the allele is young from an evolutionary point of view and may have originated from Scandinavia [12], the geographic distribution of the CCR5Δ32 polymorphism shows large variation among different populations. For example, in central Europe high frequencies (around 10%) of CCR5Δ32 are observed, whereas the deletion is virtually absent in African, Asian, Middle Eastern and American Indian populations [13]. This phenomenon has been attributed to a selective rise of CCR5Δ32 during the bubonic plague and the smallpox pandemic since individuals with reduced CCR5 expression are less susceptible to these infections [14]. As CCR5Δ32 also confers resistance to HIV-1, it is currently under intense selection pressure in populations with a high prevalence of HIV infection [12].
The high frequency of the CCR5Δ32 polymorphism in the Caucasian population permits evaluation of the impact of functioning versus impaired CCR5 on the course of neoplastic disease. Essner and coworkers recently investigated the CCR5 genotype in patients with metastatic melanoma on a therapeutic vaccination protocol [15]. They reported that individuals who were hetero- or homozygous carriers for CCR5Δ32 were characterized by a significantly decreased overall survival compared to patients with two functional alleles. This observation might be explained by the negative prognostic impact of CCR5Δ32 on the survival of melanoma patients through lack of CCR5-mediated signaling pathways needed to maintain tumor immune surveillance. Alternatively, the survival disadvantage seen in patients carrying the CCR5Δ32 allele might be due to a decreased immune response of these patients to vaccination therapy. To further elucidate these interactions, we analyzed CCR5Δ32 polymorphisms in a large cohort of almost 800 cutaneous melanoma patients with completely documented course of disease, defined treatment regimens and long-term follow-up with the intent of revealing potential associations of CCR5 status with disease-specific survival and response to immunotherapy.
Patients and methods
Patients and sera
Serum samples were selected from frozen serum banks hosted by the Skin Cancer Unit, Mannheim and the Department of Dermatology, Würzburg. All serum samples were obtained and processed following a standardized protocol. Blood was drawn from the patients’ cubital vein into gel coated serum tubes (Sarstedt, Nuembrecht, Germany) and allowed to clot at room temperature for at least 30 min, but not longer than 60 min. Thereafter, the tubes were centrifuged at 2,500g for 10 min. The serum phase was harvested and subsequently frozen without any additives at 1-ml aliquots at −20°C, and not thawed until immediately prior to analysis. Serum samples to be included in the analysis were selected as follows: (1) histological confirmation of primary cutaneous melanoma, (2) patient of Caucasian origin, and (3) complete documentation of the medical history, primary tumor characteristics, therapeutic interventions as well as patient follow-up available. Sera from patients with secondary malignancies were excluded from the study. Disease staging was performed according to the current AJCC criteria from 2001 for malignant melanoma [16]. The pathological classification of the primary (pT) was used to combine the parameters depth of invasion and ulceration, the classification of distant metastasis (M) was used to combine the parameters distant metastasis localization and serum LDH, both according to AJCC 2001 staging criteria [16]. Patients were grouped together for therapeutic interventions as follows: (1) patients receiving immunotherapy [interferons (IFN), interleukin-2 (IL-2), or vaccination] with or without additional chemotherapy, and (2) patients without immunotherapy within these disease stages. The collection of sera and clinical data were performed after patients’ informed consent with Institutional Review Board approval.
Determination of the CCR5 genotype
The CCR5 polymorphism status of the patients was determined by PCR. Serum (10 μl) was subjected to microwave irradiation on high power for 7 min. Pre-amplification was carried out using primers TTT ACC AGA TCT CAA AAA GAA G (forward) and GGA GAA GGA CAA TGT TGT AGG (reverse) and a mastermix consisting of oligo-dNTPs, TAQ, buffer and water. Samples were amplified for 25 cycles at 95°C (30 s)/ 58°C (30 s)/ 72°C (30 s) with a final step at 72°C for 5 min. The second PCR was performed with 1 μl of pre-amplification product, the mastermix and the following amplification primers: AGG TCT TCA TTA CAC CTG CAG C (forward) and CTT CTC ATT TCG ACA CCG AAG C (reverse). The samples were amplified 40 cycles at 95°C (30 s)/54°C (30 s)/72°C (30 s) followed by a final step at 72°C for 5 min. Subsequently, the PCR products were visualized on ultraviolet transilluminated, ethidium bromide stained 3% agarose gels after electrophoresis, yielding PCR products of 169 and 137 bp for the CCR5 wt and Δ32 alleles, respectively (see Fig. 1). The validity of the PCR based assay was proven by exemplarily sequencing of different genotypes.
Fig. 1.
Detection of CCR5Δ32 polymorphism by PCR. Representative results of 12 consecutively tested cutaneous melanoma patients, numbered from 100 to 111. Patient 103 shows heterozygocity, and patient 108 homozygocity for the Δ32 allele. Products were visualized on ultraviolet transilluminated, ethidium bromide stained 3% agarose gel after electrophoresis, yielding PCR products of 169 and 137 bp for the CCR5 wt and Δ32 alleles, respectively
Stastistical analysis
Disease-specific survival curves and median survival times were calculated beginning with the date of primary diagnosis, diagnosis of stage III or IV disease, respectively, until either death from melanoma or last patient contact, and are graphically presented using the Kaplan–Meier method for censored failure time data. The log rank test was used for comparing survival probabilities. The multivariate proportional hazards regression of Cox was used to assess the impact of multiple prognostic factors on survival. The approximation of Efron [17] was used to treat ties correctly and P values were calculated using the Wald statistic. To examine the assumption of proportional hazards in a Cox regression model the scaled Schoenfeld residuals [18] and a chi-squared test on each covariate were utilized. The factors tested were CCR5 genotype, gender, age at primary diagnosis, pathological classification of the primary (pT), and classification of distant metastasis (M). Statistical analyses were performed using R software (http://www.r-project.org, package survival) and P < 0.01 was considered significant.
Results
Seven hundred and eighty-two patients were enrolled in the trial and included 438 (56%) men and 344 (44%) women. The median age was 56.4 (range 6.9–98.3). The most common histologic diagnoses were superficial spreading melanoma (351 patients) and nodular melanoma (177 patients); 302 patients were initially diagnosed with stage III disease or progressed to stage III during clinical follow-up; 261 patients were diagnosed with stage IV disease. Detailed patient characteristics and staging information are presented in Table 1. Sera from all 782 melanoma patients were examined for the CCR5Δ32 polymorphism. The median follow-up time was 75.2 months.
Table 1.
Patient characteristics
| All patients | Stage III patients | Stage IV patients | ||
|---|---|---|---|---|
| 782 (100.0%) | 302 (100.0%) | 261 (100.0%) | ||
| Gender | Male | 438 (56.0%) | 172 (57.0%) | 153 (58.6%) |
| Female | 344 (44.0%) | 130 (43.0%) | 108 (41.4%) | |
| Median age at primary diagnosis | Years (range) | 56.4 (6.9–98.3) | 56.0 (16.5–86.6) | 54.2 (16.5–86.6) |
| Median follow-up | Months (range) | 75.2 (0.1–304) | 62.3 (1.0–301.4) | 57.9 (0.1–210.5) |
| Histological type of primary | ALM | 44 (5.6%) | 27 (8.9%) | 22 (8.4%) |
| LMM | 43 (5.5%) | 8 (2.6%) | 7 (2.7%) | |
| NM | 177 (22.6%) | 101 (33.4%) | 68 (26.1%) | |
| SSM | 351 (44.9%) | 83 (27.5%) | 88 (33.7%) | |
| Other | 14 (1.8%) | 7 (2.3%) | 3 (1.1%) | |
| Not classifiable | 26 (3.3%) | 9 (3.0%) | 9 (3.4%) | |
| n.a. | 127 (16.3%) | 67 (22.3%) | 64 (24.6%) | |
| pT (AJCC 2001) | T1a | 218 (27.9%) | 31 (10.3%) | 31 (11.9%) |
| T1b + T2a | 195 (24.9%) | 64 (21.2%) | 49 (18.8%) | |
| T2b + T3a | 175 (22.4%) | 85 (28.1%) | 70 (26.8%) | |
| T3b + T4a | 126 (16.1%) | 76 (25.2%) | 65 (24.9%) | |
| T4b | 30 (3.8%) | 19 (6.3%) | 19 (7.3%) | |
| n.a. | 38 (4.9%) | 27 (8.9%) | 27 (10.3%) | |
| M (AJCC 2001) | M1a | 60 (7.7%) | 0 (0.0%) | 60 (23.0%) |
| M1b | 57 (7.3%) | 0 (0.0%) | 57 (21.8%) | |
| M1c | 124 (15.9%) | 0 (0.0%) | 124 (47.5%) | |
| n.a. | 20 (2.6%) | 0 (0.0%) | 20 (7.7%) | |
| CCR5 genotype | Δ32/Δ32 | 12 (1.5%) | 4 (1.3%) | 5 (1.9%) |
| wt/Δ32 | 90 (11.5%) | 39 (12.9%) | 27 (10.4%) | |
| wt/wt | 680 (87.0%) | 259 (85.8%) | 229 (87.7%) |
ALM Acrolentiginous melanoma, LMM lentigo maligna melanoma, NM nodular melanoma, SSM superficial spreading melanoma, pT pathological classification of the primary combining depth of invasion and ulceration, AJCC American Joint Committee on Cancer, n.a not available
Twelve patients (1.5%) of the analyzed melanoma patients were homozygous and 90 (11.5%) were heterozygous for the CCR5Δ32 polymorphisms (Fig. 1, Table 1). In patients with stage III melanoma four patients (1.3%) were homozygous and 39 (12.9%) were heterozygous for the CCR5Δ32 polymorphism. In patients with stage IV melanoma 5 (1.9%) patients were homozygous and 27 (10.4%) were heterozygous for the polymorphism. Thus, genotype frequencies were found to be similar in the stage III and IV subgroups of melanoma patients (Table 1).
No difference could be detected in the disease-specific survival (DSS) from the date of primary diagnosis between patients with fully functional CCR5 and patients harboring CCR5Δ32 (median 180 vs. 193 mo, P = 0.89; Fig. 2). There was also no difference in DSS when survival was analyzed from the date of initial diagnosis of stage III or IV disease, respectively, until last observation or death from melanoma (stage III median 51.9 vs. 50.0 mo, P = 0.99; stage IV median 15.6 vs. 15.0 mo, P = 0.89; Figs. 3a, 4a). With regard to therapeutic interventions, 107 (35.4%) from 302 stage III patients received at least one regimen containing immunotherapy during stage III disease (Table 2). Within this group of patients, the CCR5 genotype did not reveal a significant impact on overall survival starting from the time of diagnosis of stage III disease, neither in patients treated with immunotherapy (median 60.0 vs. 34.8 mo, P = 0.61; Fig. 3b) nor in patients without this type of treatment (median 48.0 vs. 55.0 mo, P = 0.66; Fig. 3c). Of the 261 stage IV patients, 139 (53.3%) patients were treated with at least one immunomodulatory regimen during stage IV disease (Table 2). Within this patient subgroup, patients carrying one or both alleles of CCR5Δ32 showed a reduced overall survival starting from the time point of diagnosis of stage IV disease compared to patients without the Δ32 deletion (median 20.3 vs. 12.5 mo, P = 0.029; Fig. 4b). A multivariate analysis using the proportional hazards model of Cox revealed CCR5 genotype (P = 0.002), patient gender (women surviving longer than men; P = 0.019), and the pathological classification of the primary (pT; P = 0.022) as independent factors influencing overall survival of stage IV melanoma patients treated with immunotherapy. The age of these patients at primary diagnosis (P = 0.58) as well as the classification of distant metastasis (M; P = 0.63) had no independent prognostic impact. In contrast, the existence of the CCR5Δ32 allele was associated with a trend towards a more favorable survival in stage IV patients who did not receive immunotherapy (median 11.2 vs. 26.1 mo, P = 0.12; Fig. 4c).
Fig. 2.
Kaplan–Meier survival estimation of melanoma patients by CCR5 status. Curves showing the overall survival of 782 cutaneous melanoma patients from the time of primary diagnosis. Survival probabilities were compared by the CCR5Δ32 polymorphism status using the log rank test; wt/wt, 680 patients; wt/Δ32 or Δ32/Δ32, 102 patients; P = 0.89. Vertical bars indicate censored observations
Fig. 3.
Kaplan–Meier survival estimation of stage III melanoma patients by CCR5 status. Curves showing the overall survival of cutaneous melanoma patients starting with the time of first diagnosis of stage III disease. Survival probabilities were compared by the CCR5Δ32 polymorphism status using the log rank test. a All 302 patients; wt/wt, 259 patients; wt/Δ32 or Δ32/Δ32, 43 patients; P = 0.99; b 107 patients treated with at least one immunotherapy regimen during stage III disease; wt/wt, 96 patients; wt/Δ32 or Δ32/Δ32, 11 patients; P = 0.61; c 195 patients without immunotherapy during stage III disease; wt/wt, 163 patients; wt/Δ32 or Δ32/Δ32, 32 patients; P = 0.66. Vertical bars indicate censored observations. The number of patients at risk is given underneath each graph
Fig. 4.
Kaplan–Meier survival estimation of stage IV melanoma patients by CCR5 status. Curves showing the overall survival of cutaneous melanoma patients starting with the time of first diagnosis of stage IV disease. Survival probabilities were compared by the CCR5Δ32 polymorphism status using the log rank test. a All 261 patients; wt/wt, 229 patients; wt/Δ32 or Δ32/Δ32, 32 patients; P = 0.89; b 139 patients treated with at least one immunotherapy regimen during stage IV disease; wt/wt, 123 patients; wt/Δ32 or Δ32/Δ32, 16 patients; P = 0.029; c 122 patients without immunotherapy during stage IV disease; wt/wt, 106 patients; wt/Δ32 or Δ32/Δ32, 16 patients; P = 0.12. Vertical bars indicate censored observations. The number of patients at risk is given underneath each graph
Table 2.
Advanced melanoma patients receiving immunotherapy
| Stage III patients | Stage IV patients | |
|---|---|---|
| Immunotherapy | 107/302 (35.4%)a | 139/261 (53.3%)a |
| Non-specific | ||
| IFNα alone | 100 (93.5%)b | 14 (10.1%)b |
| IFNγ alone | 2 (1.9%)b | 1 (0.7%)b |
| IL-2 alone | 2 (1.9%)b | 11 (7.9%)b |
| IFNα + IL-2 | 5 (4.6%)b | 19 (13.7%)b |
| IFNα + chemotherapy | 0 (0.0%)b | 42 (30.2%)b |
| IL-2 + chemotherapy | 0 (0.0%)b | 2 (1.4%)b |
| IFNα + IL-2 + chemotherapy | 0 (0.0%)b | 15 (10.8%)b |
| Specific (vaccination) | ||
| Melanoma-associated peptides | 1 (0.9%)b | 59 (42.4%)b |
| Peptide-loaded dendritic cells | 0 (0.0%)b | 33 (23.7%)b |
| Whole tumor cells/tumor lysate | 0 (0.0%)b | 5 (3.6%)b |
Disease staging was performed according to the AJCC staging system
IFN Interferon, IL interleukin
a Number and percentage of patients receiving any kind of immunotherapy in the respective stage
b For the patients receiving immunotherapy the number and percentage of patients treated with the indicated immunotherapy during stage III and stage IV disease, respectively is given. Multiple entries possible due to multiple regimens applied sequentially in single patients
Discussion
Melanoma is the prototypic immunogenic tumor and various immunotherapeutic approaches for its treatment have been developed over the past decades. While stage III patients are most often treated with non-specific immunotherapy, in general using different regimens of interferon-alpha, stage IV patients are more likely to receive IL-2 or active specific immunologic regimens such as vaccination with tumor-derived proteins or peptides combined with a variety of different vehicles and adjuvants in clinical trials. However, response rates to immunotherapy are often low and vary among patients [19–21]. Although different tumor escape mechanisms such as antigen loss or defects in MHC expression have been identified, the primary resistance of metastatic melanoma to immunologic treatment still requires further explanation [22].
In order to evaluate the relevance of functional CCR5 in patients with cutaneous melanoma we studied a cohort of almost 800 individuals. Thirteen percent of the investigated patients showed homo- or heterozygocity for the CCR5Δ32 polymorphism. This allele frequency is consistent with the ranges previously described for the Caucasian population in central Europe [13]. No significant changes could be observed in CCR5Δ32 frequency with regard to different disease stages of melanoma. Moreover, the DSS starting from primary diagnosis revealed no difference between patients carrying the Δ32 allele and patients not harboring this polymorphism. These findings indicate that the impact of a functional CCR5 system on disease progression, metastasis and overall survival for melanoma patients is very small or does not exist. In contrast, when looking at advanced metastatic melanoma patients receiving immunotherapy, we found a strong association between the CCR5Δ32 genotype and poor outcome following immunotherapy in terms of overall survival. This prognostic effect was even stronger than that of well-defined impact factors such as gender, depth of invasion or ulceration of the primary, localization of distant metastasis, and serum LDH, as shown by multivariate analysis. These findings are in line with the report by Essner et al. [15] who also observed a favorable outcome of CCR5Δ32-carrying stage IV melanoma patients treated with vaccine immunotherapy. In addition, from 160 stage IV melanoma patients which had been made NED (no evidence of disease) by surgical resection, patients with wild type CCR5 status demonstrated significantly higher 5-year survival than those with one or two CCR5Δ32 alleles [23].
Interestingly, when restricting the analysis to the group of stage IV patients who did not receive immunotherapy, we found that individuals carrying CCR5Δ32 showed a trend towards a longer median survival than those without this polymorphism, suggesting that a reduced or absent CCR5 function might be associated with a survival benefit in these patients. This observation might be related to reactions to chemotherapy or other treatment regimens used in these patients. Stress by genotoxic agents is likely to induce CCR5 ligands in the tumor cells, which in turn promotes angiogenesis, proliferation and metastasis. Patients with a reduced or non-functional CCR5 would be less susceptible to these effects, and possibly result in a survival advantage for this patient group. Taking into account that about 70% of the stage IV patients in our series received immunotherapy and chemotherapy, either together or one after the other, it is remarkable that one regimen containing immunotherapy can invert the favorable prognostic impact of CCR5Δ32 into a negative one. These findings support the hypothesis that an effective increase in anti-tumor immunity by functional CCR5 is of higher importance for a prolongation of survival of stage IV melanoma patients than the role of intact CCR5 in overcoming the negative effects of chemotherapy. However, these findings need further validation in larger patient cohorts with well-defined, homogenous treatment regimens.
In conclusion, our data provide evidence that the status of the CCR5 chemokine receptor system has a substantial impact on the outcome of immunotherapy in patients with metastatic melanoma. The data also suggest, to a lesser extent, that CCR5 status may influence the response to non-immunological therapy regimens, reflected by treatment outcome in terms of patient survival. Notably, these effects were antithetic, i.e. functional CCR5 is associated with an improved response to immunotherapy and inferior response to chemotherapy, whereas for CCR5Δ32 homo/heterozygotes immunotherapy outcome was worse but chemotherapy effects were improved. Future investigation will likely identify other polymorphisms in immune responsive genes which may also play a role in shaping the clinical response to immunotherapy and other treatments in patients with melanoma. Thus, individual CCR5 genotype and other polymorphisms should be taken into account when choosing therapeutic modalities for metastatic melanoma patients, and might function as stratification factor in future clinical trials.
Acknowledgments
Grant support: This study was supported by the Wilhelm Sander Stiftung (# 2006.054.1).
Footnotes
Selma Ugurel and David Schrama have contributed equally to this work.
References
- 1.Cyster JG. Chemokines and cell migration in secondary lymphoid organs. Science. 1999;286:2098–2102. doi: 10.1126/science.286.5447.2098. [DOI] [PubMed] [Google Scholar]
- 2.Murphy PM, Baggiolini M, Charo IF, et al. International union of pharmacology. XXII. Nomenclature for chemokine receptors. Pharmacol Rev. 2000;52:145–176. [PubMed] [Google Scholar]
- 3.Dias S, Choy M, Rafii S. The role of CXC chemokines in the regulation of tumor angiogenesis. Cancer Invest. 2001;19:732–738. doi: 10.1081/CNV-100106148. [DOI] [PubMed] [Google Scholar]
- 4.Coussens LM, Werb Z. Inflammation and cancer. Nature. 2002;420:860–867. doi: 10.1038/nature01322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Payne AS, Cornelius LA. The role of chemokines in melanoma tumor growth and metastasis. J Invest Dermatol. 2002;118:915–922. doi: 10.1046/j.1523-1747.2002.01725.x. [DOI] [PubMed] [Google Scholar]
- 6.Luboshits G, Shina S, Kaplan O, et al. Elevated expression of the CC chemokine regulated on activation, normal T cell expressed and secreted (RANTES) in advanced breast carcinoma. Cancer Res. 1999;59:4681–4687. [PubMed] [Google Scholar]
- 7.Robinson SC, Scott KA, Wilson JL, Thompson RG, Proudfoot AE, Balkwill FR. A chemokine receptor antagonist inhibits experimental breast tumor growth. Cancer Res. 2003;63:8360–8365. [PubMed] [Google Scholar]
- 8.Mrowietz U, Schwenk U, Maune S, et al. The chemokine RANTES is secreted by human melanoma cells and is associated with enhanced tumour formation in nude mice. Br J Cancer. 1999;79:1025–1031. doi: 10.1038/sj.bjc.6690164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lavergne E, Combadiere C, Iga M, et al. Intratumoral CC chemokine ligand 5 overexpression delays tumor growth and increases tumor cell infiltration. J Immunol. 2004;173:3755–3762. doi: 10.4049/jimmunol.173.6.3755. [DOI] [PubMed] [Google Scholar]
- 10.Yang X, Lu P, Fujii C, et al. Essential contribution of a chemokine, CCL3, and its receptor, CCR1, to hepatocellular carcinoma progression. Int J Cancer. 2006;118:1869–1876. doi: 10.1002/ijc.21596. [DOI] [PubMed] [Google Scholar]
- 11.Huang Y, Paxton WA, Wolinsky SM, et al. The role of a mutant CCR5 allele in HIV-1 transmission and disease progression. Nat Med. 1996;2:1240–1243. doi: 10.1038/nm1196-1240. [DOI] [PubMed] [Google Scholar]
- 12.Galvani AP, Novembre J. The evolutionary history of the CCR5-Delta32 HIV-resistance mutation. Microbes Infect. 2005;7:302–329. doi: 10.1016/j.micinf.2004.12.006. [DOI] [PubMed] [Google Scholar]
- 13.Martinson JJ, Chapman NH, Rees DC, Liu YT, Clegg JB. Global distribution of the CCR5 gene 32-basepair deletion. Nat Genet. 1997;16:100–103. doi: 10.1038/ng0597-100. [DOI] [PubMed] [Google Scholar]
- 14.Galvani AP, Slatkin M. Evaluating plague and smallpox as historical selective pressures for the CCR5-Delta 32 HIV-resistance allele. Proc Natl Acad Sci USA. 2003;100:15276–15279. doi: 10.1073/pnas.2435085100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Essner R, Lee HT, Itakura T, Huynh Y, Zogakis T, Morton DL (2005) Somatic gene mutations predict outcome of patients receiving adjuvant active specific immunotherapy after resection of stage IV melanoma. ASCO annual meeting 2005
- 16.Balch CM, Buzaid AC, Soong SJ, et al. Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. J Clin Oncol. 2001;19:3635–3648. doi: 10.1200/JCO.2001.19.16.3635. [DOI] [PubMed] [Google Scholar]
- 17.Efron B (1977) The efficiency of Cox’s likelihood function for censored data, 72 edn. pp 557–565
- 18.Therneau TM, Grambsch PM, Fleming TR. Martingale-based residuals for survival models. Biometrika. 1990;77:147–160. doi: 10.1093/biomet/77.1.147. [DOI] [Google Scholar]
- 19.Thurner B, Haendle I, Roder C, et al. Vaccination with mage-3A1 peptide-pulsed mature, monocyte-derived dendritic cells expands specific cytotoxic T cells and induces regression of some metastases in advanced stage IV melanoma. J Exp Med. 1999;190:1669–1678. doi: 10.1084/jem.190.11.1669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Schadendorf D, Ugurel S, Schuler-Thurner B, et al. Dacarbazine (DTIC) versus vaccination with autologous peptide-pulsed dendritic cells (DC) in first-line treatment of patients with metastatic melanoma: a randomized phase III trial of the DC study group of the DeCOG. Ann Oncol. 2006;17:563–570. doi: 10.1093/annonc/mdj138. [DOI] [PubMed] [Google Scholar]
- 21.Luiten RM, Kueter EW, Mooi W, et al. Immunogenicity, including vitiligo, and feasibility of vaccination with autologous GM-CSF-transduced tumor cells in metastatic melanoma patients. J Clin Oncol. 2005;23:8978–8991. doi: 10.1200/JCO.2005.01.6816. [DOI] [PubMed] [Google Scholar]
- 22.Komenaka I, Hoerig H, Kaufman HL. Immunotherapy for melanoma. Clin Dermatol. 2004;22:251–265. doi: 10.1016/j.clindermatol.2003.12.001. [DOI] [PubMed] [Google Scholar]
- 23.Essner R, Lee JH, Itakura H, Huynh Y, Scheri R, Morton D. Chemokine receptor 5 (CCR5) {Delta}32 base pair somatic gene deletions from PBM predict survival of patients with advanced stage (IV) melanoma. J Clin Oncol (Meeting Abstracts) 2006;24:10056. [Google Scholar]