Texte intégral de l'article
 
   
  Version PDF

Tissue prognostic markers for adoptive immunotherapy in melanoma

European Journal of Dermatology. Volume 17, Number 4, 295-301, July-August 2007, Investigative report

DOI : 10.1684/ejd.2009.0203

Summary  

Author(s) : Gaëlle Quereux, Marie-Christine Pandolfino, Anne-Chantal Knol, Amir Khammari, Christelle Volteau, Jean-Michel Nguyen, Brigitte Dreno , Unit of skin oncology, CHU, Place A. Ricordeau 44035 Nantes, France, Clinical Research Department, Methodological Unit, CHU Hôtel Dieu, Nantes, France, Unit INSERM U601.

Summary : Adoptive immunotherapy for melanoma appears to be a promising approach. The aim of our study was to discuss the role of “tumour tissue” immunogenicity in response to adoptive immunotherapy. We thus studied the potential correlation between the expression of some melanocyte differentiation antigens, adhesion molecules and cytokines expressed by the tumour cells and the survival of patients receiving an adoptive immunotherapy with autologous Tumour Infiltrating Lymphocytes (TIL). An immunohistochemical study was performed on the lymph node samples obtained from 38 patients who received autologous TIL plus interleukin-2. Frozen sections were immunostained for melanocyte differentiation antigens (Melan-A and gp100), MHC molecules (Class I and II), adhesion molecules (ICAM-1, LFA-3) and suppressive cytokines (IL-10, TGF-β and α-MSH). Expression levels of each marker were evaluated using a semi-quantitative visual scale. Using a multivariate analysis, a low expression level of TGF-β by tumour cells was significantly associated with a prolonged relapse-free survival. A low expression level of TGF-β, IL-10, ICAM-1 and α-MSH by tumour cells was significantly associated with a longer overall survival. This work suggests that a weak expression of immunosuppressive cytokines (IL-10, TGF-β and α-MSH) could be a favourable prognostic marker for patients receiving autologous TIL.

Keywords : immunotherapy, lymphocyte, melanoma, prognostic marker, tumour infiltrating, survival

Pictures

ARTICLE

Auteur(s) : Gaëlle Quereux1,3, Marie-Christine Pandolfino1,3, Anne-Chantal Knol1,3, Amir Khammari1,3, Christelle Volteau1, Jean-Michel Nguyen2, Brigitte Dreno1,2,3

1Unit of skin oncology, CHU, Place A. Ricordeau 44035 Nantes, France
2Clinical Research Department, Methodological Unit, CHU Hôtel Dieu, Nantes, France
3Unit INSERM U601

accepté le 6 Mars 2007

Standard therapies for metastatic melanoma such as chemotherapy are often unsuccessful. Rather good evidence for the efficacy of immunotherapy has been obtained during recent years in humans [1-3]. The goal of such a therapy is to reverse the immune tolerance state that enables the uncontrolled tumour growth. Adoptive immunotherapy is an immunotherapy based on the infusion of autologous tumour antigen specific T lymphocytes, after ex vivo expansion.Steven Rosenberg’s group initially proposed adoptive immunotherapy protocols in humans, based on the infusion of tumour infiltrating lymphocytes (TIL) [1]. It has since been demonstrated that the adoptive transferred T cell clones prefer to localize at tumour sites [4, 5] and survive in vivo in response to low dose IL-2 [5].The reasons behind the limited success of the immunotherapeutic approach are still partially unknown. One hypothesis is related to the involvement of CD4+CD25 high regulatory T cells (Treg) [6]. Thus, by depleting the Treg cell subset, self-reactive CD8+ T cells could mediate a more efficient anti-tumour immunity against tumours expressing self-antigens and enable longer overall survival in treated cancer patients [7]. Another hypothesis is that the tumour itself may develop immune escape strategies, including impairment of antigen presentation (tumour antigens, MHC molecules or adhesion molecules) or secretion of immunosuppressive molecules such as α-MSH, IL-10 or TGF-β [8].Our study was based on the aforementioned strategies. We used metastatic lymph node samples collected from melanoma patients with metastatic lymph nodes stage IIIb and IIIc AJCC (6th edition), who were included in a therapeutic protocol based on the transfer of autologous TIL [9, 10]. In this protocol, half of the 88 patients (44) received TIL combined with IL-2 Proleukin® (Chiron) and the other half of the patients received IL-2 alone. This study demonstrated that only patients with one invaded lymph node responded to TIL immunotherapy. In those patients the overall survival was increased (p = 0.026) and the estimated relapse rate was significantly lower (p = 0.019) in the TIL+IL-2 group compared to the IL-2 group [9]. In patients with more than one invaded lymph node, no difference was noted between the two groups, neither in the relapse free survival nor in the overall survival. In order to determine whether the efficacy of TIL as adjuvant therapy in patients with lymph node invasion (stage III AJCC) was related to the expression level of some antigens by tumour cells or to the secretion of immunosuppressive cytokines by these same tumour cells, we evaluated the potential correlation between expression levels of these antigens or cytokines and the relapse free survival and overall survival of patients who received TIL. We focused on the following antigens: melanocyte differentiation antigens (Melan-A and gp100), MHC molecules (class I and II), adhesion molecules (ICAM-1, LFA-3) and suppressive cytokines (IL-10, TGF-β and α-MSH). The main objective was to assess how each marker expression correlated with the relapse free survival and with the overall survival. The primary endpoint was disease-free interval, and the secondary endpoint was overall survival. Furthermore, in order to understand the different clinical response obtained with TIL, we also compared the expression levels of these markers in lymph node biopsies from patients with one invaded lymph node versus patients having more than one.

Material and methods

Material

This work was performed on lymph node samples that were used to produce autologous TIL from 38 patients receiving TIL plus IL-2 as part of a clinical research protocol [9]. Patients underwent a sterile lymph node resection. The majority of the invaded lymph node was used to produce TIL and for each patient a biopsy fragment was collected for in situ immunochemistry and for the establishment of the autologous tumour cell line. The remaining biopsy was addressed to the pathology laboratory.

Patients

The study was conducted according to the Helsinki declaration and submitted and approved by the French Health products Safety Agency and local ethic committee. An informed consent was obtained from each patient. The median age of these 38 patients was 51 years; the gender ratio was 1.7:1. Seventy six percent of the patients presented a Breslow thickness of more than 1.5 mm of their primary melanoma and the Clark level median was III (table 1).

The patients were treated between February 1994 and December 1998. All the patients who are alive are still followed up, and the median of follow up is 793 days.
Table 1 Relevant clinicopathological data of the 38 patients with lymph node metastasis

Patients with only one invaded lymph node (n: 13)

Patients with more than one invaded lymph node (n: 25)

Median age

52 years (34 to 74)

49 years (22 to 74)

Sex ratio

9 males/4 females

15 males/10 females

% of patients with Breslow index more than 1.5 mm

92% (12/13)

68% (17/25)

Median Clark level

III

III

Median number of invaded lymph nodes

1

2

% capsular breakthrough

38% (5/13)

56% (14/25)

Relapse rate

31% (4/13)

80% (20/25)

Methods

Immunohistochemistry

Immunohistochemistry was performed using the streptavidin/peroxidase technique as previously described [11] on deep-frozen sections obtained from metastatic lymph nodes. Cryostat sections (5 μm) sections were incubated for 30 minutes in a moist chamber at room temperature with the primary antibodies. The following monoclonal antibodies were used: anti-Melan-A (Dako, Glostrup, Denmark), anti-gp100 (Dako, Carpinteria USA), anti- CMH class I (A, B, C) (BD Pharmingen San diego, USA), anti-CMH class II (DR, DP, DQ) (BD Pharmingen San diego, USA), anti-CD54 (ICAM-1) (Immunotech, Marseille, France), anti-CD58 (LFA-3) (Serotec, Oxford, UK), anti-α-MSH (polyclonal)(TEBU bio, Santa Cruz biotechnology), anti-IL-10 (Diaclone research, Besançon, France) and anti-TGF-β (Serotec, Oxford, UK). Secondary biotinylated rabbit anti-mouse antibody was added for 30 minutes and then peroxydase-conjugated streptavidin (Dako detection kit peroxydase/AEC, Rabit/mouse, Glostrup, Danmark). The sections were then incubated with AEC for 2 minutes. Finally the sections were counterstained with Mayers’s hemalun and mounted with aqueous mounting medium. Between the different steps tissue sections were washed in TBS (Tris buffered saline)(SIGMA)-BSA(bovine serum albumin) (SIGMA) 0.1%. The primary antibodies were omitted in negative controls.

For α-MSH and IL-10, the signal was weak so it was amplified using the biotinyl tyramide (kit TSA indirect; Dupont NEN, Boston, USA) [11].

Slides were read with a Leica microscope (magnification × 40). The expression of the molecule in the tumour cell was assessed using a visual semi-quantitative scale from 0 to 3 for each studied lymph node biopsy (n = 38). All the slides were read by two independent physicians. For each section, the protein level was taken as the mean of three fields. A weak expression (0-25% of positive cells), moderate (26-50%), intermediate (51-75%) and strong (> 75%) were represented respectively by levels 0, 1, 2 and 3 (figures 1-4).

Statistical analysis

Survival analyses were performed. Univariate and multivariate Cox models were used for each variable studied. Each model was used on the following baseline characteristics: number of invaded lymph nodes (1 versus > 1), age at onset of treatment, gender, Breslow thickness (< 1.5 mm versus > 1.5 mm), Clark level (1/2/3 versus 4/5), ulceration and significant variables in the univariate Cox model.

To compare each expression level for all the markers studied according to the number of invaded lymph nodes (one versus more than one) the chi square was used.

Interactions between the number of invaded lymph nodes and the previously mentioned variables were also analysed. Survival models were based on the assumption of proportional hazards along time. This hypothesis was graphically checked with survival curves (Kaplan-Meier model) and on multivariate models with Schoenfeld’s remnants model.

Results

Expression levels

The expression levels of the different markers are summarized in table 2. Eighty percent (32/38) of lymph nodes expressed gp100 antigen and 82% (31/38) Melan-A antigen, with a median expression level of 2 for both. MHC class I expression was absent in 6 lymph nodes (16%) and MHC class II in 3 lymph nodes (8%).

Table 3 summarizes results for relapse-free survival and overall survival according to a univariate model and table 4 according to a multivariate model.
Table 2 Expression levels of the different markers in the lymph node biopsies

Expression levels of the markers

Patient number

Number of invaded lymph nodes

GP-100

Melan-A
  • CMH
  • Class I
  • CMH
  • Class II

ICAM-1

LFA-3

α-MSH

Il-10

TGF-β

 1

5

0

0

1

1

2

1

1

1

0

 2

38

1

1

1

2

2

0

0

0

0

 3

6

3

3

2

3

3

0

0

0

1

 4

2

2

1

0

0

0

0

0

1

0

 5

10

1

1

0

1

1

1

1

1

0

 6

2

3

2

0

1

1

0

0

0

0

 7

2

2

2

2

3

1

1

1

0

3

 8

3

1

2

0

2

0

0

0

1

1

 9

6

3

2

1

2

0

0

0

1

0

10

2

2

1

2

2

1

1

1

0

0

11

4

1

1

2

2

1

0

0

0

0

12

2

2

3

3

3

1

0

2

2

1

13

2

2

2

2

2

1

0

1

2

0

14

2

3

3

2

2

2

0

0

0

1

15

7

1

2

3

2

3

1

2

2

1

16

2

1

3

3

2

0

0

0

1

0

17

4

1

3

2

3

2

0

0

2

1

18

2

0

0

1

1

1

1

1

1

0

19

3

0

0

0

0

0

0

0

0

1

20

3

2

2

0

0

1

0

1

1

0

21

2

1

1

2

1

0

0

3

2

0

22

3

2

3

1

2

1

0

2

3

0

23

2

1

0

1

2

2

1

0

1

0

24

2

0

0

3

3

3

0

0

0

2

25

2

3

3

2

1

2

2

1

1

0

Median

1

2

2

2

1

0

0

1

0

ARRAY(0x387e18)

26

1

3

2

2

3

3

0

0

0

1

27

1

0

0

3

3

2

1

0

0

0

28

1

3

3

1

2

3

1

3

3

0

29

1

1

1

3

3

3

0

2

2

2

30

1

2

3

1

2

2

0

0

0

0

31

1

3

3

1

1

0

1

0

2

0

32

1

1

3

1

2

3

1

2

2

0

33

1

2

2

3

3

1

0

2

3

1

34

1

3

2

1

3

1

1

1

1

0

35

1

2

2

2

2

1

1

0

0

0

36

1

3

2

3

2

0

0

2

2

0

37

1

3

3

2

2

1

1

0

0

2

38

1

0

0

1

2

2

0

2

2

1

Median

2

2

2

2

2

1

1

2

0

Table 3 Relapse-free survival and overall survival analysis according to an univariate model

Relapse free survival

β

S(β)

RR

ICRR95%

p-value

Invaded lymph node (1 vs > 1)

1.57

0.55

4.80

[1.63-14.1]

0.0043

Age

– 0.027

0.019

0.97

[0.94-1.01]

0.15

Sex

– 0.31

0.41

0.74

[0.33-1.64]

0.45

Breslow thickness (< 1.5 vs > 1.5)

0.043

0.47

1.04

[0.42-2.62]

0.93

Clark level (1-2-3 vs 4-5)

– 0.22

0.41

0.81

[0.36-1.80]

0.60

Ulceration (0 : no vs 1 : yes)

0.47

0.44

1.60

[0.68-3.77]

0.28

Capsular breaking (0: no vs 1: yes)

0.92

0.42

2.51

[1.11-5.71]

0.028

ARRAY(0x3a2630)

GP 100a

– 0.25

0.41

0.78

[0.35-1.73]

0.54

Melan-Aa

– 0.00

0.42

0.99

[0.44-2.05]

0.99

MHC class Ia

0.63

0.43

1.89

[0.82-4.32]

0.13

MHC class IIa

– 0.27

0.45

0.76

[0.32-1.85]

0.55

ICAM-1a

– 0.41

0.43

0.67

[0.29-1.53]

0.34

LFA-3a

1.28

1.05

3.59

[0.46-28.3]

0.22

α-MSHa

– 0.08

0.47

0.93

[0.37-2.34]

0.87

Il 10a

– 0.03

0.43

0.97

[0.41-2.26]

0.94

TGF-βa

0.13

0.62

1.14

[0.34-3.85]

0.83

Overall survival

Invaded lymph node (1 vs > 1)

1.66

0.62

5.28

[1.57-17.7]

0.0072

Age

– 0.010

0.018

0.99

[0.96-1.03]

0.60

Sex

– 0.25

0.42

0.78

[0.34-1.79]

0.57

Breslow thickness (< 1.5 vs > 1.5)

– 0.082

0.47

0.92

[0.37-2.32]

0.86

Clark level (1-2-3 vs 4-5)

– 0.36

0.42

0.70

[0.30-1.59]

0.39

Ulceration (0 : no vs 1 : yes)

– 0.59

0.46

1.80

[0.74-4.38]

0.20

Capsular breaking (0: no vs 1: yes)

1.04

0.44

2.83

[1.20-6.68]

0.017

ARRAY(0x3b0d10)

GP 100a

– 0.038

0.42

0.96

[0.42-2.19]

0.93

Melan-Aa

0.50

0.45

1.64

[0.67-4.00]

0.28

MHC class Ia

0.76

0.44

2.14

[0.90-5.06]

0.084

MHC class IIa

0.039

0.08

1.04

[0.41-2.65]

0.93

ICAM-1a

– 0.47

0.44

0.62

[0.26-1.48]

0.28

LFA-3a

1.41

1.06

4.10

[0.51-32.8]

0.18

α-MSHa

– 0.11

0.51

0.90

[0.33-2.43]

2.43

Il-10a

– 0.00

0.45

0.99

[0.41-2.42]

0.99

TGF-βa

0.32

0.63

1.38

[0.41-4.69]

0.61
a0 or 1 vs 2 or 3.


Table 4 Relapse-free survival and overall survival analysis according to a multivariate model

Relapse free survival

β

S(β)

RR

ICRR95%

p-value

GP 100a

– 0.08

± 0.47

0.92

[0.37-2.31]

0.86

Melan-Aa

– 0.29

± 0.51

0.75

[0.27-2.05]

0.57

MHC class Ia

0.38

± 0.59

1.46

[0.46-4.61]

0.52

MHC class IIa

0.08

± 0.60

1.08

[0.33-3.53]

0.90

ICAM-1a

0.93

± 0.54

2.53

[0.88-7.29]

0.09

LFA-3a

1.71

± 1.24

5.51

[0.48-62.8]

0.17

α-MSHa

0.73

± 0.60

2.08

[0.64-6.75]

0.22

Il-10a

0.90

± 0.58

2.45

[0.79-7.58]

0.12

TGF-βa

1.62

± 0.79

5.05

[1.07-23.8]

0.041

Overall survival

GP 100a

0.30

± 0.56

1.35

[0.45-4.03]

0.59

Melan-Aa

0.50

± 0.52

1.65

[0.59-4.03]

0.34

MHC class Ia

0.94

± 0.61

2.55

[0.76-8.50]

0.13

MHC class IIa

0.18

± 0.62

1.19

[0.35-4.03]

0.78

ICAM-1a

0.63

± 0.54

1.87

[0.65-5.40]

0.00042

LFA-3a

2.60

± 1.34

13.42

[0.97-186]

0.053

α-MSHa

1.76

± 0.81

5.82

[1.20-28.3]

0.029

Il-10a

1.73

± 0.66

5.63

[1.55-20.4]

0.0086

TGF-βa

2.50

± 0.91

12.15

[2.04-72.4]

0.0061
a0 or 1 vs. 2 or 3.

Correlation between expression levels and relapse-free survival

At the clinical level, the number of invaded lymph nodes and capsular breakthrough were significantly correlated to the relapse-free survival using univariate analysis (table 3). However none of the studied tissue markers were correlated to the relapse-free survival.

Using multivariate model (table 4), only the expression of TGF-β was significantly correlated to relapse: weak expression of TGF-β was associated with a longer relapse-free survival.

Correlation between expression levels and overall survival

The number of invaded lymph nodes and capsule breakthroughs were significantly correlated to the overall survival using univariate analysis (table 3), but none of the studied tissue markers was correlated to the overall survival.

Using multivariate model (table 4), expression of cytokines (TGF-β, IL-10, α-MSH) and ICAM-1 were significantly correlated to overall survival with a weak expression associated with a longer survival.

Comparison of expression levels according to the number of invaded lymph nodes

There was no correlation between the expression level of melanocyte differentiation antigens, MHC molecules, adhesion molecules and cytokines and the number of invaded lymph nodes.

The 6 lymph nodes with loss of expression for MHC class I came from patients with more than one invaded lymph node, but it was not significant (p = 0.07). Similarly the lymph nodes with a loss of expression for MHC class II came from patients with more than one invaded lymph node.

Discussion

In order to determine whether the lack of efficiency of TIL was related to tumour escape mechanisms to adoptive immunotherapy, we studied the expression level of antigens and the secretion of immunosuppressive cytokines by the tumour cells. This study was performed on lymph node samples that were used to produce autologous TIL extracted from 38 patients receiving TIL plus IL-2 as part of a clinical research protocol [9]. A weak point of this study was due to the technique that was used. Immunohistochemistry is an imprecise technique for quantification and largely depends on the technical conditions. Furthermore, to avoid the subjectivity of the reading all the slides were read by two independent physicians. We differentiated tumour melanoma cells and other immunocompetent cells by their morphologic aspect, which is quite different. We focused on melanocyte differentiation antigens, MHC molecules, adhesion molecules and suppressive cytokines. Concerning the number of lymph nodes, there was no significant difference regarding expression levels for these different markers according to the number of invaded lymph nodes. However, all melanoma tumour cells with loss of MHC class I or class II expression came from patients with more than one invaded lymph node. This could explain at least in part the differential therapeutic response obtained from patients having one or several invaded lymph nodes.

Concerning the level of expression of melanocyte differentiation antigens by tumour cells, we did not, in our study, identify an antigen whose expression was correlated with the relapse free survival or the overall survival of patients receiving TIL. These results are in accordance with those noted in the literature, but in those studies none of the patients had been treated by immunotherapy [12, 13]. Only Berset et al. [14] reported significantly lower overall survival from patients who did not express Melan-A. Their study was conducted on 73 primary melanomas with a 4.5 year follow-up but without adoptive immunotherapy.

Regarding expression of MHC class I and class II molecules, their prognostic value on relapse or overall survival remains debatable [15]. Hofbauer et al. [12] did not report any prognostic value for MHC class I molecules on relapse-free survival and overall survival, while Kageshita et al. [16] reported a correlation between a decreased expression level for MHC class I molecules within primary tumours and a decreased overall survival or relapse-free survival. But none of these studies had been carried out during an immunotherapy. Concerning MHC class II molecules, it has been shown that melanoma cells were stained by anti-class II CMH molecules [15, 17]. Furthermore Ruiter et al. showed that the percentage of class II CMH-bearing melanoma cells is higher in metastasis than in primary lesions [18]. In our study, we found no prognostic value of MHC molecule expression in the response to TIL immunotherapy.

Regarding adhesion molecules using a multivariate model, the overall survival was significantly increased following immunotherapy when the expression level for ICAM-1 was weak, (0 or 1) (p = 0.00042). Similarly, Natali et al. [19] reported that ICAM-1 was correlated to a worse prognosis for 31 primary melanomas. Patients in this study were treated by surgery and received chemotherapy or isolated limb perfusion but no immunotherapy. Regarding integrin LFA-3, its expression level was weak (0 or 1) for almost all of the studied samples (97%: 37/38). Therefore LFA-3 does not appear to play a significant role in the interaction between TIL and the tumour cell. To our knowledge, the relationship between LFA-3 expression and survival in patients with melanoma has not yet been studied.

Regarding cytokine expression levels, our work focused on 3 main immunosuppressive cytokines: α-MSH, TGF-β and IL-10. Using a multivariate analysis, a significantly increased relapse-free survival and overall survival for patients receiving TIL was associated to a weak expression level for these cytokines in the invaded lymph node.

α-MSH has been extensively studied for the last ten years. It down-regulates pro-inflammatory cytokines (IL-1, IL-2, IL-6, TNF-α, IFN-γ), and co-stimulation molecules (CD86, CD40) [20, 21]. In parallel, α-MSH up-regulates the immunosuppressive IL-10 [22]. Nagahama et al. [22] reported an increased α-MSH production in melanomas and in melanoma metastases compared to nævi, suggesting a potential role for α-MSH in melanoma progression, probably related to its suppressive activity. TGF-β also has a suppressive activity on immunosurveillance. IL-10 inhibits IFN-γ production, and disables antigen-presenting cells from processing the antigen by decreasing MHC class II expression levels. Thus, our results indicate that the production of immunosuppressive cytokines by the tumour cells could be one of the mechanisms inhibiting the activity of TIL at the tumour site.

Conclusion

This work suggests that immunosuppressive cytokine production (IL-10, TGF-β and α-MSH) may represent a prognostic marker for TIL-receiving patients. Their expression level is inversely correlated to the overall survival of the patients. Moreover, the differential efficacy of TIL according to the number of invaded lymph nodes (one versus more than one) does not appear to be related to the expression levels of melanocyte differentiation antigens, adhesion molecules nor to the expression levels of immunosuppressive cytokines. However, a loss of MHC class I or II expression was found only in lymph node samples from patients with more than one invaded lymph node.

Acknowledgement

This work has been supported by Ligue contre le cancer 2005 and “Cancer Immunotherapy” 6th Framework Programme and INSERM

References

1 Rosenberg SA, Packard BS, Aebersold PM, et al. Use of tumor-infiltrating lymphocytes and interleukine-2 in the immunotherapy of patients with metastatic melanoma. N Engl J Med 1988; 319(25): 1676-80.

2 Van den Eynde BJ, Van der Bruggen P. T Cell defined tumor antigens. Curr Opin Immunol 1997; 9(5): 684-93.

3 Seledtsov VI, Shishkov AA, Surovtseva MA, et al. Xenovaccinotherapy for melanoma. Eur J Dermatol 2006; 16(6): 655-61.

4 Dudley ME, Wunderlich JR, Yang JC, et al. A phase I study of nonmyeloablative chemotherapy and adoptive transfer of autologous tumor antigen-specific T lymphocytes in patients with metastatic melanoma. J Immunother 2002; 25(5): 243-51.

5 Yee C, Thompson JA, Byrd D, et al. Adoptive T cell therapy using antigen-specific CD8+ T cell clones for the treatment of patients with metastatic melanoma: in vivo persistence, migration, and antitumor effect of transferred T cells. Proc Natl Acad Sci USA 2002; 99(25): 16168-73.

6 Viguier M, Lemaitre F, Verola O, et al. Foxp3 expressing CD4+CD25 (high) regulatory T cells are overrepresented in human metastatic melanoma lymph nodes and inhibit the function of infiltrating T cells. J Immunol 2004; 173(2): 1444-53.

7 Dudley ME, Wunderlich JR, Robbins PF, et al. Cancer regression and autoimmunity following clonal repopulation with anti-tumor lymphocytes and non-myeloablative conditioning. Science 2002; 298(5594): 850-4.

8 Woods GM, Malley RC, Muller HK. The skin immune system and the challenge of tumour immunosurveillance. Eur J Dermatol 2005; 15(2): 63-9.

9 Dreno B, Nguyen JM, Khammari A, et al. Randomized trial of adoptive transfer of melanoma tumor infiltrating lymphocytes as adjuvant therapy for stage III melanoma. Cancer Immunol Immunother 2002; 51(10): 539-46.

10 Labarriere N, Pandolfino MC, Gervois N, et al. Therapeutic efficacy of melanoma-reactive TIL injected in stage III melanoma patients. Cancer Immunol Immunother 2002; 51(10): 532-8.

11 Chebassier N, El Houssein O, Viegas I, Dréno B. In vitro induction of matrix metalloproteinase-2 and matrix metalloproteinase-9 expression in keratinocytes by boron and manganese. Exp Dermatol 2004; 13(8): 484-90.

12 Hofbauer GF, Burkhart A, Schuler G, Dummer R, Burg G, Nestle FO. High frequency of melanoma-associated antigen or HLA class I loss does not correlate with survival in primary melanoma. J Immunother 2004; 27(1): 73-8.

13 Kageshita T, Kawakami Y, Hirai S, Ono T. Differential expression of MART-1 in primary and metastatic melanoma lesions. J Immunother 1997; 20(6): 460-5.

14 Berset M, Cerottini JP, Guggisberg D, et al. Expression of Melan-A/MART-1 antigen as a prognostic factor in primary cutaneous melanoma. Int J Cancer 2001; 95(1): 73-7.

15 Van Duinen SG, Ruiter DJ, Broecker EB, et al. Level of HLA antigens in locoregional metastases and clinical course of the disease in patients with melanoma. Cancer Res 1988; 48(4): 1019-25.

16 Kageshita T, Ishihara T, Campoli M, Ferrone S. Selective monomorphic and polymorphic HLA class I antigenic determinant loss in surgically removed melanoma lesions. Tissue Antigens 2005; 65(5): 419-28.

17 Brocker EB, Suter L, Sorg C. HLA-DR antigen expression in primary melanomas of the skin. J Invest Dermatol 1984; 82(3): 244-7.

18 Ruiter DJ, Bergman W, Welvaart K, et al. Immunohistochemical analysis of malignant melanomas and nevocellular nevi with monoclonal antibodies to distinct monomorphic determinants of HLA antigens. Cancer Res 1984; 44(9): 3930-5.

19 Natali P, Nicotra MR, Cavaliere R, et al. Differential expression of intercellular adhesion molecule 1 in primary and metastatic melanoma lesions. Cancer Res 1990; 50(4): 1271-8.

20 Eves P, Haycock J, Layton C, et al. Anti-inflammatory and anti-invasive effects of alpha-melanocyte-stimulating hormone in human melanoma cells. Br J Cancer 2003; 89(10): 2004-15.

21 Luger TA, Scholzen TE, Brzoska T, Bohm M. New insights into the functions of alpha-MSH and related peptides in the immune system. Ann N Y Acad Sci 2003; 994: 133-40.

22 Nagahama M, Funasaka Y, Fernandez-Frez ML, et al. Immunoreactivity of alpha-melanocyte-stimulating hormone, adrenocorticotrophic hormone and beta-endorphin in cutaneous malignant melanoma and benign melanocytic naevi. Br J Dermatol 1998; 138(6): 981-5.

Copyright © 2007 John Libbey Eurotext - Tous droits réservés