COMBINATION OF ANTI-CD303 AND ANTI-HER2 ANTIBODIES

Abstract

Disclosed is a pharmaceutical composition including, in a pharmaceutically acceptable carrier: a) at least one anti-CD303 antibody; and b) at least one anti-HER2 antibody. Also disclosed is the combination of the two aforementioned antibodies a) and b) to form combination products to be used simultaneously, separately or spread out over a period of time, for the prevention or treatment of HER2-positive cancer.

Claims

1-11. (canceled)

12. Pharmaceutical composition comprising, in a pharmaceutically acceptable carrier: a. at least one anti-CD303 antibody; and b. at least one anti-HER2 antibody.

13. A method for preventing or treating HER2-positive cancer in a subject in need thereof, comprising administering to said subject products containing: c. at least one anti-CD303 antibody; and d. at least one anti-HER2 antibody, which are combination products for simultaneous, separate or sequential use.

14. The pharmaceutical composition according to claim 12, wherein the antibodies a) and b) are selected from among mouse antibodies, chimeric antibodies, humanized antibodies and human antibodies.

15. The pharmaceutical composition according to claim 12, wherein the anti-CD303 antibody is a monoclonal antibody directed against the ectodomain of the human CD303 antigen.

16. The pharmaceutical composition according to claim 12, wherein the anti-CD303 antibody comprises the following CDRs: TABLE-US-00008 CDR1-H-IMGT-122A2 GYTFTDYS(SEQIDNO:13) CDR2-H-IMGT-122A2 ISTYYGDS(SEQIDNO:14) CDR3-H-IMGT-122A2 ARNGNFYVMDY(SEQIDNO:15) CDR1-L-IMGT-122A2 QDISNY(SEQIDNO:16) CDR2-L-IMGT-122A2 YTS(SEQIDNO:17) CDR3-L-IMGT-122A2 QQGNTLPWT(SEQIDNO:18) orelse: CDR1-H-IMGT-102E9 GYTFTDYS(SEQIDNO:19) CDR2-H-IMGT-102E9 INTETGEP(SEQIDNO:20) CDR3-H-IMGT-102E9 TRNGYYVGYYAMDY(SEQIDNO:21) CDR1-L-IMGT-102E9 SSVIY(SEQIDNO:22) CDR2-L-IMGT-102E9 STS(SEQIDNO:23) CDR3-L-IMGT-102E9 QQRRSYPFT(SEQIDNO:24) orelse: CDR1-H-IMGT-104C12 GYTFTDYS(SEQIDNO:25) CDR2-H-IMGT-104C12 ISPYYGDT(SEQIDNO:26) CDR3-H-IMGT-104C12 ARNDDYYRFAY(SEQIDNO:27) CDR1-L-IMGT-104C12 QDINNY(SEQIDNO:28) CDR2-L-IMGT-104C12 YTS(SEQIDNO:29) CDR3-L-IMGT-104C12 QQGKTLPWT(SEQIDNO:30) orelse: CDR1-H-IMGT-114D11 GYTFTDSS(SEQIDNO:31) CDR2-H-IMGT-114D11 INTETGGP(SEQIDNO:32) CDR3-H-IMGT-114D11 ARNGYYVGYYALDY(SEQIDNO:33) CDR1-L-IMGT-114D11 SSVFY(SEQIDNO:34) CDR2-L-IMGT-114D11 STS(SEQIDNO:35) CDR3-L-IMGT-114D11 QQRRSYPYT(SEQIDNO:36) orelse: CDR1-H-IMGT-104E10 GYTFTDYS(SEQIDNO:37) CDR2-H-IMGT-104E10 INTETGEP(SEQIDNO:38) CDR3-H-IMGT-104E10 ARNGYYVGYYAMDY(SEQIDNO:39) CDR1-L-IMGT-104E10 SSVIY(SEQIDNO:40) CDR2-L-IMGT-104E10 STS(SEQIDNO:41) CDR3-L-IMGT-104E10 QQRRSYPYT(SEQIDNO:42)

17. The pharmaceutical composition according to claim 12, wherein the anti-HER2 antibody is a composition of trastuzumab, said composition comprising glycan structures at the Fc glycosylation sites having a galactose content higher than 60%.

18. The pharmaceutical composition according to claim 12, wherein the anti-HER2 antibody is a composition of trastuzumab, said composition comprising glycan structures at the Fc glycosylation sites having a degree of fucosylation of at least 50%.

19. The pharmaceutical composition according to claim 12, wherein the anti-HER2 antibody is trastuzumab produced in the milk of transgenic goats.

20. The pharmaceutical composition according to claim 12, wherein the anti-CD303 antibody and/or anti-HER2 antibody is selected from among Fab, F(ab)2, Fd, scFv and multimers of scFv.

21. A method for preventing or treating HER2-positive cancer in a subject in need thereof, comprising administering to said subject a pharmaceutical composition according to claim 12, wherein the HER2-positive cancer is selected from among uterine, ovarian, breast and stomach cancer.

22. A method for preventing or treating HER2-breast positive cancer in a subject in need thereof, comprising administering to said subject a pharmaceutical composition according to claim 12.

23. The pharmaceutical composition according to claim 12, wherein the anti-HER2 antibody is a composition of trastuzumab, said composition comprising glycan structures at the Fc glycosylation sites having a galactose content higher than 70%.

24. The pharmaceutical composition according to claim 12, wherein the anti-HER2 antibody is a composition of trastuzumab, said composition comprising glycan structures at the Fc glycosylation sites having a degree of fucosylation of at least 60%.

25. The method according to claim 13, wherein the antibodies a) and b) are selected from among mouse antibodies, chimeric antibodies, humanized antibodies and human antibodies.

26. The method according to claim 13, wherein the anti-CD303 antibody is a monoclonal antibody directed against the ectodomain of the human CD303 antigen.

27. The method according to claim 13, wherein the anti-CD303 antibody comprises the following CDRs: TABLE-US-00009 CDR1-H-IMGT-122A2 GYTFTDYS(SEQIDNO:13) CDR2-H-IMGT-122A2 ISTYYGDS(SEQIDNO:14) CDR3-H-IMGT-122A2 ARNGNFYVMDY(SEQIDNO:15) CDR1-L-IMGT-122A2 QDISNY(SEQIDNO:16) CDR2-L-IMGT-122A2 YTS(SEQIDNO:17) CDR3-L-IMGT-122A2 QQGNTLPWT(SEQIDNO:18) orelse: CDR1-H-IMGT-102E9 GYTFTDYS(SEQIDNO:19) CDR2-H-IMGT-102E9 INTETGEP(SEQIDNO:20) CDR3-H-IMGT-102E9 TRNGYYVGYYAMDY(SEQIDNO:21) CDR1-L-IMGT-102E9 SSVIY(SEQIDNO:22) CDR2-L-IMGT-102E9 STS(SEQIDNO:23) CDR3-L-IMGT-102E9 QQRRSYPFT(SEQIDNO:24) orelse: CDR1-H-IMGT-104C12 GYTFTDYS(SEQIDNO:25) CDR2-H-IMGT-104C12 ISPYYGDT(SEQIDNO:26) CDR3-H-IMGT-104C12 ARNDDYYRFAY(SEQIDNO:27) CDR1-L-IMGT-104C12 QDINNY(SEQIDNO:28) CDR2-L-IMGT-104C12 YTS(SEQIDNO:29) CDR3-L-IMGT-104C12 QQGKTLPWT(SEQIDNO:30) orelse: CDR1-H-IMGT-114D11 GYTFTDSS(SEQIDNO:31) CDR2-H-IMGT-114D11 INTETGGP(SEQIDNO:32) CDR3-H-IMGT-114D11 ARNGYYVGYYALDY(SEQIDNO:33) CDR1-L-IMGT-114D11 SSVFY(SEQIDNO:34) CDR2-L-IMGT-114D11 STS(SEQIDNO:35) CDR3-L-IMGT-114D11 QQRRSYPYT(SEQIDNO:36) orelse: CDR1-H-IMGT-104E10 GYTFTDYS(SEQIDNO:37) CDR2-H-IMGT-104E10 INTETGEP(SEQIDNO:38) CDR3-H-IMGT-104E10 ARNGYYVGYYAMDY(SEQIDNO:39) CDR1-L-IMGT-104E10 SSVIY(SEQIDNO:40) CDR2-L-IMGT-104E10 STS(SEQIDNO:41) CDR3-L-IMGT-104E10 QQRRSYPYT(SEQIDNO:42)

28. The method according to claim 13, wherein the anti-HER2 antibody is a composition of trastuzumab, said composition comprising glycan structures at the Fc glycosylation sites having a galactose content higher than 60%.

29. The method according to claim 13, wherein the anti-HER2 antibody is a composition of trastuzumab, said composition comprising glycan structures at the Fc glycosylation sites having a degree of fucosylation of at least 50%.

30. The method according to claim 13, wherein the anti-HER2 antibody is trastuzumab produced in the milk of transgenic goats.

31. The method according to claim 13, wherein the anti-CD303 antibody and/or anti-HER2 antibody is selected from among Fab, F(ab)2, Fd, scFv and multimers of scFv.

Description

EXAMPLES

[0125] Advantageously, in the present invention the anti-CD303 antibody used is a chimeric or humanized antibody comprising the CDR 122A2 sequences. The anti-HER2 antibody preferably used below is a trastuzumab antibody produced in transgenic goat milk e.g. such as described in application WO2014/125377.

Example 1: Impact of pDC Depletion on ADCC and Phagocytosis Activities of Trastuzumab (Herceptin)

[0126] 1.1Principle of the Study

[0127] pDCs are responsible for the differentiation of regulatory T cells (Moseman E A, 2004; Martin-Gayo E, 2010) inter alia via ICOS/ICOSL interaction (Ito, T., 2007; Faget, J. 2012; Faget, J. 2013). The regulatory T cells thus differentiated exert immunosuppressive mechanisms on the functions of the other cells in the immune system, NK cells in particular, via cell/cell contact but also via the secretion of immuno-modulating cytokines such as IL-10, IL-35 and TGF- (Liu, C 2016).

[0128] Through cascade effect, it is therefore possible to examine the impact of pDC depletion on the action of an anticancer agent specific to a tumour involving in situ activation of pDCs. In particular, the protective effect of anti-CD303 antibodies targeting pDCs can be shown in a correlative study on the reduction or removal of IL-10 and TGF- cytokines in the tumour environment, and the impact thereof on the effector functions of an administered anticancer agent (anti-HER2 antibody such as Herceptin) specific to the tumour. According to this model, the anti-CD303 antibodies depleting the pDCs lead to limiting the immunosuppressive effects of regulatory T cells, thereby limiting their secretion of IL-10 and TGF-. This limiting of IL-10 and TGF- secretion is correlated with better ADCC of the NK cells, validating the indirect stimulating effect of the anti-CD303 antibodies on the anticancer action of anti-HER2 antibodies.

[0129] 1.2Effect of IL-10 and TGF- on ADCC Activity of Trastuzumab

[0130] To test the ADCC of trastuzumab in a tumoral context involving either activation of pDCs or depletion of pDCs, the following protocol was set up:

[0131] NK effector cells transfected with CD16 (NK-CD16) were placed in culture at 310.sup.5 cells/ml in a culture medium containing IL-2.

[0132] At D-1, the NK-CD16 cells were placed in a culture medium without IL-2, in the absence or presence of IL-10 (5, 50 or 120 ng/ml) and TGF- (5, 50 or 120 ng/ml).

[0133] At D0, BT-474 cells (breast tumour cells) (35000 cells/well) expressing HER-2 were incubated in a flat bottom 96-well plate with said NK-CD16 cells, in an E/T ratio (EffectorNK/TargetBT-474) of 5:1 and 5000 ng/ml of anti-HER-2 antibody (trastuzumab). After an incubation time of 16 hours at 37 C., the supernatant was collected.

[0134] The negative control followed the same protocol wherein trastuzumab was replaced by a chimeric antibody inhibiting the anti-FVIII antibody, produced in YB2/0.

[0135] Lysis of the target cells induced by the anti-HER-2 antibodies was measured chromogenically by quantifying the intracellular lactate dehydrogenase enzyme (LDH) released into the supernatant by the lysed target cells (Roche DiagnosticsCytotoxicity Detection Kit LDH)

[0136] Percent lysis was calculated with the following formula:

% lysis=[(ERSR)/(100SR)][(NCSR)/(100SR)]

[0137] Where ER and SR respectively represent experimental (ER) and spontaneous (SR) release of LDH, and NC represents the natural cytotoxicity of NK cells.

[0138] The results (% lysis) are expressed as a percentage, 100% being the value taken as reference obtained with the NK-CD16 cells in the presence of trastuzumab and in the absence of IL-10 and TGF- (i.e. Trastuzumab alone).

[0139] ADCC results are detailed in following Table 1:

TABLE-US-00006 TABLE 1 Trastu- Negative Trastu- zumab + IL- Negative control + IL- zumab 10/TGF- control 10/TGF- % lysis 21 32 12 22 0 0 0 0 100% 100 100 57 69 0 0 0 0 corresponds to the reference: Trastuzumab alone Mean 100 63 0 0

[0140] These results show that the ADCC activity of trastuzumab is inhibited in the presence of IL-10 and TGF-, compared with the control without these cytokines.

[0141] Based on an arbitrary value of 100% for trastuzumab alone, the mean percentage of ADCC is 63% in the presence of IL-10 and TGF- at a concentration of 5000 ng/ml of trastuzumab.

Conclusion

[0142] Through cascade effect, comparison of percent lysis observable in the absence of IL-10 and/or TGF- cytokines, with percent lysis observable in the presence of these same cytokines, allows evaluation of the impact of pDC depletion at the tumour site. It can thus be shown that anti-CD303 antibodies allow indirect potentializing of the effect of the anti-HER2 anticancer antibodies.

[0143] 2Effect of IL-10 and TGF- on Phagocytosis Induced by Trastuzumab

[0144] The monocytes were differentiated into CD16+ microphages (M2 like) for 2 days in RPMI 1640+10% SVF+M-CSF 50 ng/ml for 48 h.

[0145] The SKBR3 cells and macrophages were labelled with PKH-67 (green fluorescence) and PKH-26 (red fluorescence), respectively.

[0146] The SKBR3 cells were opsonized with 10 g/ml of Trastuzumab antibody or with a control antibody and then incubated with the macrophages (1.105 of each cell/well) in the absence or presence of different concentrations of IL-10 (5, 50 or 120 ng/ml) alone, of TGF- (5, 50 or 120 ng/ml) alone, and IL-10+ TGF-.

[0147] After incubation for 3 h at 37 C., the cells were placed in a counting chamber (Mallassez) and observed under a fluorescence microscope.

[0148] Percent phagocytosis was evaluated by counting the number of macrophages (at least 100 macrophages) containing SKBR3 cells.

Conclusion

[0149] Through cascade effect, comparison of percent phagocytosis observable in the absence of IL-10 and/or TGF- cytokines, with percent lysis observable in the presence of these same cytokines, allows evaluation of the impact of pDC depletion at the tumour site. It is thus possible to show that the anti-CD303 antibodies allow indirect potentializing of the effect of the anti-HER2 anticancer antibodies.

Example 2Effect of Anti-CD303 Antibody on Activation of Regulatory T Cells (Treg)

[0150] 2.1Role of Anti-CD303 on the Phenotype and Expansion of Tregs in PBMC

[0151] The mononuclear cells (PBMC) were isolated from a tube of blood taken on anti-coagulant. The PBMCs did not contain pDCs. The Treg cells were identified and subjected to phenotype characterization via flow cytometry on the basis of 3 markers: CD4, CD25 and Fox-P3.

[0152] Different quantities of anti-CD303 antibody (from 1 ng to 10 g/ml) were added to the PBMCs in the presence of IL-2 (500 U/ml). The number of Tregs and their phenotype were monitored over time (1 to 4 days).

[0153] Under the same conditions, beads coated with anti-CD3/anti-CD28 to stimulate T proliferation were added in a Treg/beads ratio of 4:1 to verify Treg activation.

Conclusion

[0154] It can therefore be shown that the anti-CD303 antibodies, in the absence of pDCs, do not have any direct impact on the expansion and immunosuppressive phenotype of regulatory T cells.

[0155] 2.2Role of Anti-CD303 on the Phenotype and Expansion of Purified Tregs in the Presence of pDCs

[0156] The Treg cells (CD4.sup.+, CD25.sup.+) were purified from PBMCs using a 2-step process: depletion of negative CD4 cells (positive cells for the CD8, CD14, CD15, CD16, CD19, CD36, CD56, CD123, TCR/ and CD235a markers) followed by positive selection of CD25.sup.+ cells.

[0157] Purified pDCs or pDC lines e.g. obtained with the method described in Maeda T et al., Int J Hematol. 2005 February; 81(2):148-54 (such as the CAL-1 line or a new line generated by retransfection of CD303 in Cal-1 and selection of a line stably expressing more than 40 000 CD303 antigen sites per cell, preferably between 40 000 and 50 000) were added in a Treg/pDC ratio of 100, 10 and 1. Different quantities of anti-CD303 antibody (from 1 ng to 10 g/ml) were added to the Treg/pDC mixture in the presence of IL-2 (500 U/ml). The number of Tregs and their phenotype were monitored over time (1 to 4 days). Under the same conditions, beads coated with anti-CD3/anti-CD28 to stimulate T proliferation were added in a Treg/beads ratio of 4:1 to verify Treg activation. A negative control in the absence of pDCs was prepared, to verify the direct impact (expected to be neutral) of anti-CD303 on Tregs.

Conclusion

[0158] Observation over several days of the expansion and differentiation of purified Tregs in the presence of pDCs, after administration of anti-CD303 antibody, can show that administration of anti-CD303 is effective in reducing or suppressing the immunosuppressive properties of pDCs.

Example 3: Depletion of Human pDCs Via an Anti-CD303 Antibody In Vivo in the Treatment of Breast Cancer

[0159] 3.1-Generation of a Reproductive Study Model of a Pathological Situation in Man

[0160] To study the effect of an anti-CD303 antibody in the treatment of breast cancer, a humanized tumour mouse model (HTM) can be used. It is characterized by the development of a mature human immune system and the growth of human breast cancer cells previously co-transplanted with the human hemopoietic stem cells. Very recently, the effectiveness of treatment with Trastuzumab/Herceptin was proven in this model (Wege A K et al., Oncotarget 2017, 8(2): 2731-2744).

[0161] This model advantageously allows the grouping together of several relevant elements for reproducibility under in vivo conditions: presence of human pDCs which alone express the target CD303 on their surface, presence of infiltration by human Treg cells, presence of human tumour cells expressing HER2 on their surface, molecule targeted by trastuzumab/Herceptin, and an immunocompetent murine host (effector cells of NK type for ADCC activity). The model previously described in the article by Wege et al. (Int. J. Cancer 2011: 129, 2194-2206) combines all these characteristics.

[0162] This model was adapted to make it compatible with BRGSF or BRGSF-A2 mice (BALB/c, Rag2tm1Fwa, IL-2Rctm1Cgn, SIRPNOD, Flk2tm1Irl, Tg(HLA-A/H2-D/B2M)1Bpe) characterized by the absence of T, B, and NK mouse cells, and solely expressing the HLA of human class 1, HLA-A2.1. Said BRGSF mice can be generated following the procedure described by Legrand N, Huntington N D, Nagasawa M et al. (Functional CD47/signal regulatory protein alpha (SIRP(alpha)) interaction is required for optimal human T- and natural killer- (NK) cell homeostasis in vivo. Proc. Natl. Acad. Sci. U.S.A 2011; 108:13224-13229). They acquire the genotype (HLA-A/H2-D/B2M).sup.1Bpe via transgenesis (Pascolo, S., N. Bervas, J. M. Ure, A. G. Smith, F. A. Lemonnier, and B. Perarnau. 1997. HLA-A2.1-restricted education and cytolytic activity of CD8(+) T lymphocytes from beta2 microglobulin (beta2m) HLA-A2.1 monochain transgenic H-2Db beta2m double knockout mice. J Exp Med 185:2043-2051).

[0163] 3.2Method Able to be Used to Test the Activity of the Anti-CD303 Antibody

[0164] In brief, new-born mice derived from the line of immunodeficient BRGSF-A2 mice (BALB/c Rag2.sup.tm1Fwa IL-2R.sub.c.sup.tm1Cgn SIRP.sup.NOD Flk2.sup.tm1Irl Tg(HLA-A/H2-D/B2M).sup.1Bpe are irradiated (3 Gy) over the first 192 hours of their lifetime. Twenty-four hours later they are transplanted by intra-hepatic injection with 1.510.sup.5 human CD34+ cells isolated from umbilical cord blood (CB) in the presence or absence of 310.sup.6 COV434-AMHRII-Luc tumour cells (expressing luciferase for bioluminescent tracking). BT474 cells are cells of human origin derived from breast carcinoma expressing the HER2/Neu receptor on their surface: the target of the Trastuzumab antibody. In this particular model, the BT474 cells, before administration thereof, were modified by lentiviral transduction to cause them to become luminescent through the constituent expression of luciferase. This modification provides transverse tracking of tumour penetration over time by means of bioluminescence analysis, whilst not sacrificing the animals and thereby best adjusting the window for start of treatment. Eleven to twelve weeks after co-administration of human cells, the mice were tested for their extent of humanization by analysing the cell composition of their blood (human and murine) by flow cytometry, and divided into five different groups (cf. Table 2 below): a control group without injection of BT474 cells, treated with an isotype antibody (i.e. an anti-Factor VIII inhibiting antibody) (this group was used as negative control for tumour penetrationGroup 1); a control group with injection of BT474 cells, treated with an isotype antibody (i.e. an anti-Factor VIII inhibiting antibody) (Group 2); a group treated with transgenic trastuzumab (TTG)Group 3); a group treated with the anti-CD303 antibody (Group 4); and a group given the combination of anti-CD303 antibody and TTG treatments (Group 5).

[0165] The doses administered and treatment frequencies are given in Table 2 below:

TABLE-US-00007 Groups BT474-luc.sup.+ Tested product(s) Number 1 No Isotype n = 10 (anti-Factor VIII inhibiting antibody) 2 Yes Isotype n = 10 (anti-Factor VIII inhibiting antibody) 3 Yes TTG n = 10 4 Yes Anti-CD303 n = 10 5 Yes TTG + anti-CD303 n = 10

[0166] Treatment started 14 weeks after humanization (i.e. injection of human CD34+ cells isolated from umbilical cord bloodCB) in the presence or absence of 310.sup.6 BT474-Luc tumour cells, and lasted 19 weeks.

[0167] For treatment, the tested products were injected via intravenous route at a dose of 30 mg/kg bodyweight for the anti-CD303 antibody every 3 days, and once a week at 10 mg/kg bodyweight for the TTG antibody. The bodyweight of the mice was determined 3 days before the start of treatment for individual dose adjustment.

[0168] Blood samples were frequently taken to test the efficacy of human pDC depletion, by flow cytometry. In addition, bioluminescent analysis at the start of treatment and subsequently every two weeks was performed to compare the efficacy of the different tested products. Finally, tumour analysis in three animals per group at week 18 after humanization was carried out by flow cytometry to verify the presence or absence of human pDCs infiltrating the tumours.

[0169] Results: [0170] The impact of the treatment on the human pDC sub-population, and the other populations of lymphoid cells (B lymphocytes B, T lymphocytes . . . ) present in the blood and spleen, was determined by flow cytometry at different times: D1, D3 and D7. The results show that treatment with the anti-CD303 antibody at a dose of 30 mg/kg in humanized BRGSF-HIS mice induces depletion of human pDCs for at least 7 days in the blood and spleen. In the blood, the depleting activity on human pDCs was rapid (>80% on Day 1) but occurred later in the spleen. In the blood and spleen, depletion of human pDCs was always efficient (>90%) on Day 7, i.e. 3 days after the last injection of the monoclonal anti-CD303 antibody. It is to be noted that the depleting activity of the anti-CD303 antibody was highly specific since it did not significantly affect the other sub-populations of human hemopoietic cells in the tested organs. [0171] Additionally, it is to be noted that this model proves to be a particularly suitable model for the study of the invention, since the detecting of pDCs in the tumours of sacrificed mice for the first time shows the presence of pDCs infiltrating these tumours (cf. FIG. 1 giving the percentage of human pDcs in the livertumour sitein the negative control (mice not injected with BT474 cells(CT)), and in mice injected with BT474 cells (BT474)).

Conclusion

[0172] The adapted HTM mouse model can advantageously be used to evaluate the indirect impact of anti-CD303 antibody administration on the effect of the anti-breast tumour agent: the anti-HER2 antibody (trastuzumab), under conditions reproducing a physiological situation in vivo, in particular by comparing the results obtained with the different tested groups.

[0173] This model is therefore useful in evaluating the benefit, and advantageously the synergic effect, of administering an anti-CD303 antibody in combination with administration of an anti-HER2 antibody in a breast tumour.