BIOMARKERS FOR DISEASE PROGRESSION IN MELANOMA

Abstract

The present invention relates inter alia to therapeutic agents for use in the treatment of melanoma, methods of diagnosing an increased risk of metastasis in a subject suffering from melanoma, methods of treating such subjects, diagnostic assays and kits. More particularly, in certain embodiments the invention relates to identifying whether a subject suffering from melanoma has an increased risk of metastasis by determining the expression of Ambra-1 and Loricrin in a tissue sample obtained from the subject.

Claims

1. A method comprising: a) obtaining tissue overlying a primary melanoma from a subject suffering from melanoma; b) determining the expression of Ambra-1 and Loricrin in the tissue overlying a primary melanoma; c) comparing the expression obtained in (b) with a reference tissue or levels obtained therefrom, and d) (i) following a normal recognized care pathway if expression of Ambra-1 and Loricrin is normal or increased, or (ii) treating the subject with a systemic anti-cancer treatment regime if expression of Ambra-1 and Loricrin is decreased or lost.

2. The method of claim 1, wherein the method comprises detecting a co-occurrence of a decrease in the expression of Ambra-1 and a decrease in Loricrin in the tissue sample compared to the reference levels, or a loss of expression of Ambra-1 and a loss of expression of Loricrin in the tissue sample.

3. The method according to claim 1, wherein the systemic anti-cancer treatment regime is for preventing, inhibiting or delaying metastasis or decreasing the risk of metastasis in the subject.

4. The method according to claim 1, wherein the systemic anti-cancer treatment regime comprises administering a therapeutic agent to the subject.

5. A method of treating a subject suffering from melanoma, the method comprising administering a therapeutic agent to the subject, wherein the subject has been identified as having decreased or loss of expression of Ambra-1 and Loricrin in a tissue sample obtained from the subject.

6. The method according to claim 5, wherein the method is for preventing, inhibiting or delaying metastasis or decreasing the risk of metastasis in the subject.

7. The method according to claim 5, wherein the subject, prior to identification, was ineligible for therapeutic agent treatment.

8. The method according to claim 4, wherein: (i) the therapeutic agent is administered to the subject no more than 12 weeks after the subject has been identified as having decreased or loss of expression of Ambra-1 and Loricrin in the tissue sample; or (ii) the therapeutic agent is a chemotherapeutic agent such as Dacarbazine (DTIC), Temozolomide, Nab-paclitaxel, Paclitaxel, Carmustine (BCNU), Cisplatin, Carboplatin, Vinblastine, interleukin 2, interferon alpha, antibodies or B-Raf inhibitors.

9. A method of treating melanoma in a subject, the method comprising: (a) determining the expression of Ambra-1 and Loricrin in tissue overlying a primary melanoma from the subject; (b) comparing the expression obtained in (a) with a reference tissue or levels obtained therefrom, and if there is a decrease in the expression of Ambra-1 and Loricrin in the tissue sample compared to the reference tissue or levels, or a loss of expression of Ambra-1 and Loricrin, administering a therapeutic agent to the subject.

10. The method of claim 9, wherein the method comprises detecting a co-occurrence of a decrease in the expression of Ambra-1 and a decrease in Loricrin in the tissue sample compared to the reference levels, or a loss of expression of Ambra-1 and a loss of expression of Loricrin in the tissue sample.

11. The method according to claim 9, wherein: (i) the therapeutic agent is administered to the subject no more than 12 weeks after the step of determining a decrease or loss of expression of Ambra-1 and Loricrin in the tissue sample; or (ii) the therapeutic agent is a chemotherapeutic agent such as Dacarbazine (DTIC), Temozolomide, Nab-paclitaxel, Paclitaxel, Carmustine (BCNU), Cisplatin, Carboplatin, Vinblastine, interleukin 2, interferon alpha, antibodies and B-Raf inhibitors.

12. The method according to claim 1, wherein the reference levels are levels of Ambra-1 and Loricrin expression that are characteristic of normal tissue.

13. The method according to claim 1, wherein the reference tissue comprises normal tissue.

14. The method according to claim 13, wherein the normal tissue is epidermis from a site which does not include a primary melanoma.

15. The method according to claim 13, wherein the reference tissue is an internal reference.

16. The method according to claim 13, wherein the normal tissue is from a site adjacent to the primary melanoma.

17. The method according to claim 1, wherein the tissue sample comprises tissue overlying a primary melanoma and a portion of normal epidermis adjacent to the primary melanoma.

18. The method according to claim 1, wherein the expression of Ambra-1 and Loricrin in the tissue sample is from about 25% to about 75% of the respective reference level or less than about 25% of the respective reference level.

19. The method according to claim 1, wherein the expression of Ambra-1 and Loricrin in the tissue sample is determined by visual assessment or by an automatic slide scanner.

20. The method according to claim 1, wherein determining the expression of Ambra-1 and Loricrin in tissue overlying a primary melanoma comprises: contacting tissue overlying a primary melanoma with a first ligand specific for Ambra-1, wherein the presence of Ambra-1 creates an Ambra-1-ligand complex; contacting the tissue overlying a primary melanoma with a second ligand specific for Loricrin, wherein the presence of Loricrin creates a Loricrin-ligand complex; and detecting and/or quantifying the Ambra-1-ligand complex and the Loricrin-ligand complex.

21. The method according to claim 20, wherein the method comprises contacting a first section of the tissue sample with the first ligand and contacting a second section of the tissue sample with the second ligand.

22. The method of claim 20, wherein the first ligand comprises an anti-Ambra-1 antibody or aptamer and the second ligand comprises an anti-Loricrin antibody or aptamer.

23. The method according to claim 1, wherein the tissue sample comprises keratinocytes overlying the primary melanoma and the method comprises determining the expression of Ambra-1 and Loricrin in the keratinocytes.

24. The method according to claim 1, wherein the subject is suffering from American Joint Commission on Cancer (AJCC) stage 1a, stage 1b, stage 2a, stage 2b or stage 2c melanoma.

25. The method according to claim 1, wherein the subject is a human or animal.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0115] Embodiments of the invention will now be described by way of example only, and with reference to the accompanying Figures in which:

[0116] FIGS. 1A-1D are light microscopy images showing Ambra-1 expression in normal epidermis (FIG. 1A) and peri-tumoural epidermis (FIGS. 1B-1D) overlying AJCC stage I melanomas. Scale bars: 50 m.

[0117] FIGS. 2A-2C are immunofluorescent microscopy images showing Loricrin expression in normal epidermis (FIG. 2A), and peri-tumoural epidermis (FIGS. 2B-2C) overlying AJCC stage I melanomas. Scale bars: 50 m).

[0118] FIG. 3A shows univariate analysis of epidermal Ambra-1 expression in all AJCC stage melanomas. Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in primary melanoma epidermises with no alteration in Ambra-1 expression (top line), decreased or absent Ambra-1 expression (dashed line) or ulcerated tumours (bottom line). Log-Rand (Mantel-Cox) Test P<0.0001. DFS=100% no Ambra-1 loss, 72.1% decreased or absent Ambra-1, 35.7% ulcerated. Direct analysis of No loss Ambra-1 and decreased/absent Ambra-1 Log-Rank (Mantel-Cox) Test P=0.0270, HR 3.56 (95% CI 1.16-10.93); FIG. 3B shows multivariate analysis of Ambra-1 expression in pilot cohort of AJCC stage 1 melanomas. Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in AJCC stage 1 primary melanoma epidermises that revealed no alteration in Ambra-1 expression (top line) compared to stage 1 tumours with absent epidermal Ambra-1 (bottom line). DFS=100% no Ambra-1 loss, 83.3% decreased or absent Ambra-1. Log-Rank (Mantel-Cox) Test P=0.0575, HR 4.29 (95% CI 0.95-19.25);

[0119] FIG. 4A is a Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in epidermises in which Ambra-1 expression was maintained (top line) compared to those in which expression of Ambra-1 was lost (bottom line), across all tumour types. P=0.0007, HR 10.1 (95% CI 2.65-38.5);

[0120] FIG. 4B is a Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in epidermises in which Loricrin expression was maintained (top line) compared to those in which expression of Loricrin was lost (bottom line), across all tumour types. P=0.0006, HR 18.4 (95% CI 3.5-96.2);

[0121] FIG. 4C is a Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in epidermises in which both Ambra-1 and Loricrin expression was maintained (top line) compared to those in which expression of Ambra-1 and Loricrin was lost (bottom line), across all tumour types. P=<0.0001, HR 39.6 (95% CI 6.4-243.9);

[0122] FIG. 5A is a Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in AJCC stage 1 primary melanoma epidermises in which Ambra-1 expression was maintained (top line) compared to those in which Ambra-1 expression was lost (bottom line). P=0.001, HR 24.12 (95% CI 3.6-161);

[0123] FIG. 5B is a Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in AJCC stage 1 primary melanoma epidermises in which Loricrin expression was maintained (top line) compared to those in which Loricrin expression was lost (bottom line). P<0.0001, HR 210 (95% CI 16.86-2624);

[0124] FIG. 5C is a Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in AJCC stage 1 primary melanoma epidermises in which both Ambra-1 and Loricrin expression was maintained (top line) compared to those in which Ambra-1 and Loricrin expression was lost (bottom line). P=0.0002, HR 93.5 (95% CI 8.67-1008);

[0125] FIG. 6 shows the amino acid sequence of Homo sapiens Ambra-1 from UniProtKB (primary accession number Q9COC7-1, isoform 1) (SEQ ID NO. 1); and

[0126] FIG. 7 shows the amino acid sequence of Homo sapiens Loricrin from UniProtKB (primary accession number P23490) (SEQ ID NO. 2).

[0127] FIGS. 8A-8D show a range of microscopy images of Ambra-1 (anti-Ambra-1 antibody (Abcam Biochemicals, Cambridge, UK; 69501; 1:200)) expression in normal epidermis (a), or overlying stage 1 melanomas (b). Ambra-1 maintained, (c). Ambra-1 decreased and (d) Ambra-1 completely lost).

[0128] FIGS. 9A-9C show a range of microscopy images of loricrin (anti-Loricrin antibody (Abcam Biochemicals, Cambridge, UK; 24722; 1:500)) expression in normal epidermis (a), or overlying stage 1 melanomas (b) Loricrin maintained, (c) Loricrin completely lost.

[0129] FIG. 10 shows the association between the visual scoring given to the change of peri-tumoural Ambra-1 (no loss, some loss or complete loss) compared to the quantitative analysis results of the percentage decrease in staining positivity in normal epidermis compared to the peri-tumoural epidermis. Horizontal lines represent the mean scoring percentage standard error of the mean (0=12.05, 1=25.16, 3=46.92). One way ANOVA P<0.0001****

[0130] FIG. 11 shows the association between the visual scoring given to the change of peri-tumoural Ambra-1 with no or some loss of staining compared to complete loss of staining. Horizontal lines represent the mean scoring percentage standard deviation (No/some loss mean=18.12 SD 12.97, complete loss mean=46.92 SD 15.34). Mann-Whitney P<0.0001****

[0131] FIG. 12 shows univariate analysis of epidermal Ambra-1 expression in all AJCC stage 1 melanomas. Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in primary melanoma epidermises with no loss or reduced Ambra-1 expression (black line), versus absent Ambra-1 expression (red line). DFS=97.7% no loss/decreased Ambra-1 (n=44), 94.3% absent Ambra-1 (n=35). Log-Rank (Mantel-Cox) Test P=0.411, HR 2.59 (95% CI 0.26-25.05).

[0132] FIG. 13 shows univariate analysis of epidermal Ambra-1 and Loricrin combined expression in all AJCC stage 1 melanomas. Kaplan-Meier curve showing 7-year Disease Free Survival (months until first metastasis detected) in primary melanoma epidermises maintenance of some epidermal Ambra-1 or Loricrin 10 expression (Low risk black line), versus loss of both Ambra-1 and loricrin epidermal expression (High risk red line). DFS=98.46% Low risk (n=65), 86.67% High risk (n=15). Log-Rank (Mantel-Cox) Test P=0.025, HR 9.29 (95% CI 1.49-558.0).

[0133] Further details of certain embodiments of the invention are provided below.

Definitions

[0134] The term antibody as used herein is intended to include monodonal antibodies, polyclonal antibodies, chimeric antibodies, humanised antibodies, bi-specific antibodies, antibody-drug conjugates, and domains and fragments of antibodies including Fab, Fab, F(ab)2, scFv, dsFv, ds-scFv, dimers, minibodies, diabodies, and multimers thereof. Antibodies can be fragmented using conventional techniques. Antibodies may be from any animal origin, including birds and mammals (e.g., human, murine, donkey, sheep, rabbit, goat, guinea pig, camel, horse, or chicken), transgenic animals, or from recombinant sources. Antibodies may be prepared using methods known to those skilled in the art.

[0135] As used herein, the term primary melanoma refers to a malignant tumour on the skin at the site of origin, regardless of thickness, in patients without clinical or histologic evidence of regional or distant metastatic disease.

[0136] As used herein, the wording tissue overlying a primary melanoma, refers to epidermal tissue situated between a primary melanoma and the surface of the skin.

[0137] As used herein, the term normal tissue includes for example normal epidermis, which is healthy (i.e. disease-free) epidermis. In some embodiments, the normal tissue is epidermis that lies adjacent to the primary melanoma, for example within a cuff of normal skin taken with the primary melanoma sample.

[0138] As used herein, peri-tumoural epidermis refers to epidermal tissue overlying or situated around a tumour.

[0139] As used here in, metastasis is defined as the recurrence or disease progression that may occur locally (such as local recurrence and in transit disease), regionally (such as nodal micrometastasis or macrometastasis), or distally (such as brain, lung and other tissues). In some embodiments, the term metastasis is used to refer to metastatic disease following a primary melanoma. Typically, metastasis originating from a primary melanoma may spread to the lungs and/or brain of the subject as well as other locations.

[0140] It is to be understood that the term comparing and compare as used herein usually refers to a comparison of corresponding parameters or levels, e.g., an absolute amount is compared to an absolute reference amount while a concentration is compared to a reference concentration or a signal intensity signal obtained from the tissue sample is compared to the same type of signal intensity obtained from the reference. The comparison may be carried out manually, for example by visual assessment, or it may be automated (e.g. using an automated scanner or computer-assisted). Thus, the comparison may be carried out by a computing device.

[0141] The stage of a melanoma is a description of how widespread it is. This includes its thickness in the skin, whether it has spread to nearby lymph nodes or any other organs, and certain other factors. The stage is based on the results of the physical exam, biopsies, and any imaging tests (CT or MRI scan, etc.) or other tests that have been done. Such tests will be known to those skilled in the art. The system most often used to stage melanoma is the American Joint Commission on Cancer (AJCC) TNM system. Table 1 describes the features identifying each stage.

TABLE-US-00001 TABLE 1 Stage 1 1a Tumour <1.00 mm without ulceration; no lymph node involvement, no distant metastases. 1b Tumour <1.00 mm with ulceration or Clark level IV or V tumour 1.01-2.0 mm without ulceration; no lymph node involvement; no distant metastases. Stage 2 2a Tumour 1.01-2.0 mm with ulceration; tumour 2.01-4.0 mm without ulceration; no lymph node involvement; no distant metastases. 2b Tumour 2.01-4 mm with ulceration. 2b Tumour >4.0 mm without ulceration; no lymph node involvement; no distant metastases. 2c Tumour >4.0 mm with ulceration; no nodal involvement; no distant metastases. Stage 3 3a Tumour of any thickness without ulceration with 1 positive lymph node and micrometastasis or macrometastasis. 3b Tumour of any thickness without ulceration with 2-3 positive lymph nodes and micrometastasis or macrometastasis. 3c Tumour of any thickness and macrometastasis OR in- transit met(s)/satellite(s) without metastatic lymph nodes, OR 4 or more metastatic lymph nodes, matted nodes or combinations of in-transit met(s)/satellite(s), OR ulcerated melanoma and metastatic lymph node(s). Stage 4 4 Tumour of any thickness with any nodes and any metastases

DETAILED DESCRIPTION OF CERTAIN EMBODIMENTS

Examples

Materials and Methods

Example 1

Immunohistochemistry for Ambra-1 and Loricrin

[0142] Analysis of primary melanoma tissue from the patient cohort shown in Table 2 was performed by immunohistochemical staining of formalin-fixed paraffin-embedded sections. Tissue sections of 5-6 m thickness were baked onto X-tra microscope slides (Leica Microsystems, Milton Keynes, UK) at 56 C. overnight. They were then incubated in Histoclear (AGTC Bioproducts, Hessle, UK) for 20 minutes before rehydration in 100%, 75%, 50% ethanol and then distilled water for 5 seconds each. Antigen retrieval was undertaken by microwave heating in pre-heated antigen retrieval buffers (Ambra-1 (10 mM Tris-HCl (pH 7.6)), Loricrin (10 mM Na-Citrate (pH 6.0)) for 12 minutes before being allowed to cool for 20 minutes. Each section was allowed to dry and the tissue isolated with an ImmEdge hydrophobic pen (Vector Laboratories Inc., Burlingame, USA). Sections were then incubated with PBS/0.05% Tween (PBS/T) for 3 minutes to allow rehydration before incubation with 0.2% Triton X-100 (Sigma, St. Louis, USA) in PBS/T for 10 minutes. After washing with PBS/T sections were incubated in 3% H.sub.2O in water for 10 minutes to block endogenous peroxidise. Endogenous Avidin was blocked with the Avidin solution of an Avidin/Biotin Blocking kit (Vector Laboratories Inc., Burlingame, USA) for 15 minutes, before further washing with PBS/T and incubation with the Biotin portion of the kit for 15 minutes, with a following PBS/T wash. Protein blocking was undertaken by incubating sections in 2% blocking serum from an animal specific Vectastain Elite kit (Vector Laboratories Inc., Burlingame, USA).

[0143] Following a further PBS/T wash, sections were incubated with primary antibody for 1 hour at room temperature with anti-Ambra-1 antibody (Abcam Biochemicals, Cambridge, UK; 69501; 1:200) or anti-Loricrin antibody (Abcam Biochemicals, Cambridge, UK; 24722; 1:500). After 3 washes with PBS/T primary antibody was detected with biotinylated animal specific secondary antibody for 30 minutes at room temperature before 3 further washes with PBS/T. Staining was performed through incubation for 30 minutes with the ABC reagents from the Vectastain Elite kit (pre-mixed 30 minutes prior to use), followed by 3 washes with PBS/T and then a 10 minute incubation with VIP solution (Vector Laboratories Inc., Burlingame, USA). Slides were rinsed in tap water for 5 minutes before counter staining with haematoxylin (Sigma Diagnostics, St. Louis, USA) for 2 minutes followed by a final 10 minute wash in tap water with frequent changes. After dehydration through 75% and 100% ethanol for 5 seconds sections were cleaned for 2 minutes in histoclear, allowed to dry, then coverslips mounted with DPX mountant (VWR International Ltd., Poole, UK).

Determination of Expression

[0144] The difference in expression levels of Ambra-1 and/or Loricrin between the normal epidermis and peri-tumoural epidermis was initially determined by consensus agreement of 3 Dermatologists and a Histopathologist. Expression was quantified by assessing the percentage of positively stained cells in the peri-tumoural as a percentage of the Ambra-1/Loricrin expression determined in the internal control reference of the adjacent normal epidermis using Leica QWin image analysis software (Leica Microsystems). These observations were categorized as either maintained (>75% of normal expression), decreased (25-75% of normal expression) or lost (<25% of normal expression). Assessment of each section was undertaken without prior knowledge of eventual disease outcome.

TABLE-US-00002 TABLE 2 Patient Cohort Number of Patients 129 Male:Female 66:62 Median age at diagnosis (range) 58 (17-87) AJCC stage at diagnosis 1a 40 1b 36 2a 22 2b 18 2c 12 Eventual AJCC stage (8 years follow-up) 1a 38 1b 27 2a 12 2b 7 2c 4 3 15 4 25 Median Breslow depth (range) 1.55 mm (0.3-15) Ulcerated primary tumours 28

Statistics

[0145] All statistical analysis and image generation was undertaken using GraphPad Prism 5 (GraphPad Software; San Diego, USA) statistical analysis software.

[0146] All univariate and multivariate analysis of study variables for Disease Free Survival were undertaken using Kaplan-Meier curve constructions against 8-year follow-up, as well as log-rank (Mantel-Cox) analysis of the comparative data.

Example 1

Results and Discussion

[0147] The present inventors identified that a decrease of Ambra-1 expression in the peri-tumoural epidermis overlying melanomas, in particular in AJCC stage I melanomas, significantly correlates with decreased Disease Free Survival over 7 years. As shown in FIG. 1A, Ambra-1 expression is increased from the basal layer towards the stratum corneum in the normal epidermis situated adjacent to an AJCC stage 1 melanoma consistent with maintained differentiation. However, Ambra-1 expression is maintained (FIG. 1B), decreased (FIG. 1C) or even lost (FIG. 1D) in the epidermis overlying a range of AJCC stage 1 tumours.

[0148] Referring to FIG. 3, the loss or decrease of epidermal Ambra-1 expression correlates with an increased risk of metastasis. Across all AJCC melanomas (FIG. 3A), 100% of patients displaying no loss of Ambra-1 expression were disease free after 7 years. This decreased to 72.1% in patients with decreased or absent expression of Ambra-1. Only 35.7% of patients with ulcerated melanomas were disease free after 7 years. This highlights a stepwise increase in disease risk with loss of Ambra-1 and eventual associated frank ulceration of the tumour.

[0149] Taking AJCC stage 1 melanomas only (FIG. 3B), the percentage of patients who were disease free after 7 years was 100% for those with no loss of Ambra-1 expression, reduced to 83.3% for those with decreased or absent Ambra-1 expression.

[0150] FIGS. 4A and 5A show the correlation between expression of Ambra-1 and disease free survival of subjects in a smaller cohort in which both Ambra-1 and Loricrin expression were assessed, over 7 years for all tumour types (FIG. 4A) or stage 1 only (FIG. 5A). Across all tumour types (FIG. 4A), all patients in which expression of Ambra-1 was maintained were disease free after 7 years. For those in which Ambra-1 expression was lost, only 18% were disease free after 7 years. For stage 1 tumours, again 100% of patients with maintained expression of Ambra-1 did not develop metastasis, while the disease-free survival rate was 17% for those with loss of Ambra-1 expression.

[0151] Contrary to the current publications in the art which implicate Ambra-1 in the control of autophagy, Ambra-1's role in this context is thought to be in the down-regulation of differentiation with the normal epidermis resulting in a loss of integrity. It has been demonstrated by the inventors that the down-regulation of other autophagy proteins, such as ATG1, do not affect the differentiation process, supporting the hypothesis that this process is not related to autophagy.

[0152] Unexpectedly, it was found that a loss of Loricrin expression also correlates with an increased risk of metastasis. FIGS. 2A-C show representative images of normal Loricrin expression in the stratum comeum (FIG. 2A) as well as in peri-tumoural epidermis in which Loricrin expression is lost (FIG. 2B or maintained (FIG. 2C).

[0153] Referring to FIGS. 4B and 5B, the loss or decrease of epidermal Loricrin expression correlates with decreased disease-free survival over 7 years. Across all AJCC melanomas (FIG. 48), 64% of patients for whom Loricrin expression was maintained were disease free after 7 years. However, no patients with loss of Loricrin expression were disease-free after 5 years. For AJCC stage 1 melanomas (FIG. 5B), 100% of patients in which Loricrin expression was maintained were disease free after 7 years. No patients with loss of Loricrin expression were disease-free after 5 years. This demonstrates a statistical significant correlation between Loricrin expression and disease-free survival rates in melanoma patients. However, as with Ambra-1, the loss of Loricrin alone is not predictive of disease progression with 100% accuracy, either for AJCC stage 1 melanoma or all tumour stages (see Table 3).

[0154] However, the inventors have determined that determination of expression of both Ambra-1 and Loricrin is strongly representative of disease progression to metastasis. The results of these experiments are shown in FIGS. 4C and 5C, and summarized in Table 3.

TABLE-US-00003 TABLE 3 Marker Sensitivity Specificity PPV.sup.1 NPV.sup.2 All tumour stages n = 15 Loss of Ambra-1 100% 80% 83.3% 100% Loss of Loricrin 63.6% 100% 100% 69.2% Loss of Ambra-1 + Loricrin 100% 100% 100% 100% Stage 1 melanoma n = 12 Loss of Ambra-1 100% 88.9% 83.3% 100% Loss of Loricrin 80% 100% 100% 90% Loss of Ambra-1 + Loricrin 100% 100% 100% 100% .sup.1Positive predictive value .sup.2Negative predictive value

[0155] Thus, it was found that the combination of loss of Ambra-1 and loss of Loricrin identifies with 100% accuracy patients which went on to develop metastasis.

Example 2

[0156] Further analysis was carried out on 80 retrospective AJCC stage 1 melanoma patients' samples recruited to an independent James Cook University Hospital melanoma cohort (Table 4). The analysis reveals data in keeping with the findings in the initial retrospective cohort detailed above.

TABLE-US-00004 TABLE 4 Number of Patients 80 Male:Female 27:53 AJCC stage at diagnosis 1a 54 1b 26 Eventual AJCC stage (7 years follow-up) 1a 53 1b 24 4 3 Mean Breslow depth (range) 0.83 mm (0.14-1.9)

[0157] Immunohistochemical staining was undertaken using DAB counterstain as the most widely used specialist stain in clinical use. All samples were digitally imaged using automated scanning of slides on a Leica SCN400 digital slide scanner (Leica Biosystems) within the Newcastle University Biobank (FIGS. 8 and 9).

[0158] Visual analysis of epidermal Ambra-1 loss was undertaken by two independent dermatologists and scores assigned for each slide based on the degree of loss of epidermal Ambra-1 in the peri-tumoural epidermis compared to normal epidermis within the same section. There was 95% concordance in scores given between dermatologists, and further review of these slides resulted in an agreed score for each.

[0159] Similarly, the same two dermatologists undertook visual analysis of epidermal Loricrin loss independently. Any break in the continuity of the loricrin stain in the peri-tumoural epidermis that was not due to direct tumour invasion of the upper epidermis was scored as Loricrin loss. Concordance was 97.5% with agreement reached for all samples.

[0160] Quantitative analysis of epidermal Ambra-1 was undertaken using the Leica Slidepath systems' previously validated analysis software. Five representative areas of normal epidermis were selected at 200 magnification and the mean percentage of DAB positive pixels obtained. This was compared to the mean percentage of DAB positive pixels in ten representative areas of pen-tumoural epidermis at 200 magnification. The overall percentage decrease in Ambra-1 expression between peri-tumoural expression and that of the normal epidermis was then calculated.

[0161] Comparison of visual and quantitative scores (FIG. 10) reveals a statistically significant (P<0.0001) stepwise increase in quantitative score with decreased peri-tumoural Ambra-1 as analyzed visually. This validates visual scoring as a robust and reliable method for analyzing Ambra-1 epidermal staining.

[0162] To determine a cut-point for survival analysis, visual and quantitative scores were re-analyzed with no or some loss of peri-tumoural Ambra-1 compared to complete loss (FIG. 11). This shows a statistically significant increase in qualitative score in samples scored visually as having a complete loss of Ambra-1 (P<0.0001). This further validates the appropriateness of visual scoring to identify samples with complete loss of peri-tumoural Ambra-1, and one standard deviation below the mean for complete Ambra-1 loss (46.92 mean SD 15.34) gives an appropriate cut off of 30% to determine further qualitative analysis of Ambra-1 loss.

[0163] Univariate analysis of peri-tumoural Ambra-1 loss in all patients revealed no overall difference in disease free survival between High risk (tumours with complete peri-tumoural Ambra-1 loss as determined by a qualitative decrease in expression of >30%) and Low risk tumours (qualitative expression decrease <30%) (FIG. 5). DFS=97.7% Low risk tumours (n=44), 94.3% High risk tumours (n=35). Log-Rank (Mantel-Cox) Test P=0.411, HR 2.59 (95% CI 0.26-25.05). These results do not support Ambra-1 as a prognostic biomarker in this subset of patients.

[0164] To assess the validity of the combination of epidermal Ambra-1 and Loricrin expression as a prognostic biomarker, univariate analysis was undertaken in all samples. High risk samples were determined as having complete peri-tumoural Ambra-1 loss (>30% decrease quantitatively) AND a loss of loricrin. All other tumours, with either loss of Ambra-1 OR Loricrin, were deemed Low risk. These results showed a statistically significant increased risk of metastases in the High risk tumour group, even though the total number of metastatic events was low; further reinforcing the utility of the combination of Ambra-1 and Loricrin as a combined prognostic biomarker in AJCC stage 1 disease. DFS=98.46% Low risk (n=65), 86.67% High risk (n=15). Log-Rank (Mantel-Cox) Test P=0.025, HR 9.29 (95% CI 1.49-558.0).

TABLE-US-00005 TABLE 5 Marker (n = 80) Sensitivity Specificity PPV NPV Loss of Ambra 67% 57% 6% 97.7% Loss of Loricrin 67% 70% 8% 98.1% Combined loss 67% 83% 13% 98.4% of Ambra and Loricrin

[0165] The final analysis of the combined Ambra-1/Loricrin biomarker highlights increased specificity (83%), positive and negative predictive values (13% and 98.4% respectively) of Ambra-1/Loricrin combined over and above either Ambra-1 or loricrin alone (Table 5). This indicates that the combined biomarker would add prognostic value in identifying high-risk patients for increased surveillance, as well as identifying low-risk patients that could be reassured regarding their prognosis with more certainty.

[0166] This is an important finding as a decrease or loss of expression of these two proteins may indicate a breakdown of the epidermis overlying and the vasculature underlying the tumour, meaning that cancer cells may have already migrated from the primary tumour at the time of tumour excision.

[0167] Certain embodiments of the present invention thus provides a means for determining whether a subject suffering from melanoma is at increased risk of metastasis. This allows a treatment regime to be tailored accordingly, thereby reducing the risk of the subject developing metastasis and improving their prognosis.

[0168] Throughout the description and claims of this specification, the words comprise and contain and variations of them mean including but not limited to and they are not intended to (and do not) exclude other moieties, additives, components, integers or steps. Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

[0169] Features, integers, characteristics or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of the features and/or steps are mutually exclusive. The invention is not restricted to any details of any foregoing embodiments. The invention extends to any novel one, or novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

[0170] The reader's attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.