DIAGNOSTIC ANTI-CD95L ANTIBODY

Abstract

The present invention relates to a specific CD95L antibody and to the use thereof in the diagnosis of a cancer disease.

Claims

1. A monoclonal anti-CD95L antibody or a functional fragment thereof characterized in that the antibody specifically binds to an epitope comprising amino acids 13-19 of human CD95L.

2. The monoclonal antibody according to claim 1, which is a full-length immunoglobulin or a functional immunoglobulin fragment selected from the group consisting of Fab, Fab, F(ab)2, Fv, single chain antibodies (scFv) and single domain antibodies.

3. The monoclonal antibody according to claim 1, comprising a heavy chain comprising an amino acid sequence including a CDRH1 as shown in SEQ ID NO: 1, or an amino acid sequence differing in 1 or 2 amino acids therefrom, a CDRH2 as shown in SEQ ID NO: 2, or an amino acid sequence differing in 1 or 2 amino acids therefrom, and a CDRH3 as shown in SEQ ID NO: 3, or an amino acid sequence differing in 1 or 2 amino acids therefrom, and a light chain comprising an amino acid sequence including a CDRL1 as shown in SEQ ID NO: 4, or an amino acid sequence differing in 1 or 2 amino acids therefrom, a CDRL2 as shown in SEQ ID NO: 5, or an amino acid sequence differing in 1 or 2 amino acids therefrom, and a CDRL3 as shown in SEQ ID NO: 6, or an amino acid sequence differing in 1 or 2 amino acids therefrom.

4. The monoclonal antibody according to claim 1, comprising a heavy chain variable region comprising the amino acid sequence as shown in SEQ ID NO: 7, or an amino acid sequence having a sequence identity of at least 90% thereto, and/or a light chain variable region comprising the amino acid sequence as shown in SEQ ID NO: 8, or an amino acid sequence having a sequence identity of at least 90% thereto.

5. The monoclonal antibody according to claim 1, comprising a heavy chain comprising an amino acid sequence as shown in SEQ ID NO: 9, or an amino acid sequence having a sequence identity of at least 90% thereto, and/or a light chain comprising an amino acid sequence as shown in SEQ ID NO: 10, or an amino acid sequence having a sequence identity of at least 90% thereto

6. The monoclonal antibody according to claim 1, wherein a label group is covalently attached to the antibody.

7. An isolated nucleic acid molecule comprising a nucleic acid sequence selected from the group consisting of: (a) a nucleic acid sequence encoding an antibody or a functional fragment thereof as defined in claim 1, preferably a nucleic acid sequence as shown in any one of SEQ ID NOs: 11 and/or 12, (b) a nucleic acid sequence complementary to any one of the sequences in (a), and (c) a nucleic acid sequence capable of hybridizing to (a) or (b) under stringent conditions.

8. A vector comprising the nucleic acid sequence as defined in claim 7, preferably an expression vector, wherein the nucleic acid sequence is operably linked to a control sequence.

9. A host comprising the nucleic acid sequence as defined in claim 7.

10. The host of claim 9, which is a human, bacteria, animal, fungal, amphibian or plant cell, or a non-human transgenic animal.

11. A process of manufacturing an antibody according to claim 1, comprising the step of culturing a host under conditions that allow synthesis of said antibody and recovering said antibody from said culture, wherein the host comprises a nucleic acid sequence selected from the group consisting of: (a) a nucleic acid sequence as shown in any one of SEQ ID NOs: 11 and/or 12, (b) a nucleic acid sequence complementary to any one of the sequences in (a), and (c) a nucleic acid sequence capable of hybridizing to (a) or (b) under stringent conditions.

12. A pharmaceutical composition comprising the antibody according claim 1, optionally together with a pharmaceutically acceptable carrier.

13. Use of an antibody according to claim 1 in an immunohistochemical method for detecting CD95L.

14. A method for diagnosing a cancer disease comprising (a) determining the expression of CD95L in a cancer sample using an antibody according to claim 1, and (b) classifying the cancer disease according to the level of CD95L expression.

15. The method of claim 14, wherein in step (a) the expression of CD95L is determined by an immunohistochemical method.

16. Use of the antibody of claim 1, in the diagnosis of a cancer disease by classifying the cancer disease according to the level of CD95L expression.

17. Use of the antibody of claim 1, for prognosing the overall survival time or/and the relapse-free survival time in a cancer patient by classifying the cancer disease of the patient by the level of CD95L expression.

18. The use of claim 16, wherein the level of CD95L expression is determined by an immunohistochemcial method.

19. A method of prognosing the overall time or/and the relapse-free survival time in a cancer patient, comprising: (a) determining CD95L expression in a cancer sample using an antibody of claim 1, and (b) prognosing the survival time or/and the relapse-free survival time of the patient by the level of CD95L expression, wherein the CD95L expression is negatively correlated with the survival time of the patient.

20. The method of claim 19, wherein in step (a) CD95L expression is determined by an immunohistochemical method.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0122] This application contains at least one drawings executed in color. Copies of this patent or patent application publication with color drawings will be provided by the Office upon request and payment of the necessary fee.

[0123] FIG. 1: Survival analysis (Kaplan-Meier plot) of GBM patients randomized to the radiation therapy (RT) only group (control group) and the radiation therapy plus APG101 group (test group). (1) Test group. (2) Control group. Abscissa: survival time (days). Ordinate: survival rate. A survival rate of 1.0 indicates 100% survival.

[0124] FIG. 2: Comparison of the survival data of the subgroup of CD95L positive patients taken from the patients described in FIG. 1 (Kaplan-Meier plot). (1) Test group, radiation therapy plus APG101. (2) Control group, radiation therapy only. Abscissa: survival time (days). Ordinate: survival rate. A survival rate of 1.0 indicates 100% survival.

[0125] FIG. 3: Comparison of the survival data of the subgroup of CD95L negative patients taken from the patients described in FIG. 1 (Kaplan-Meier plot). (1) Test group, radiation therapy plus APG101. (2) Control group, radiation therapy only. Abscissa: survival time (days). Ordinate: survival rate. A survival rate of 1.0 indicates 100% survival.

[0126] FIG. 4: Examples of CD95L expression levels in glioblastoma tissue sections according to Example 3. Examples of high expression (CD95L positive) and low CD95L expression (CD95L negative) are given. The panel shows two examples for each expression level.

[0127] FIG. 5: Comparison of purified anti-CD95L monoclonal rabbit antibodies. A) Non reduced SDS-Page/Silver staining of purified anti-CD95L rabbit monoclonal antibodies. Lane 1 shows recombinant antibody (APG1181) purified after expression in HEK-cells. Lane 3 shows purified antibody purified form the supernatant of the rabbit hybridoma clone 24-8.

[0128] B) ELISA analysis of purified antibodies: APG1181 and antibody from clone 24-8 were analysed for their specificity to recognise the immobilized CD95L-peptide (CD95L-NT_1-21) used for immunization of the rabbits. Both antibodies specifically bind to the immobilized peptide. The specific recognition for both antibodies was confirmed by competition by the addition of the epitope specific peptide (+specific peptide). However, an unspecific peptide has no significant competing effect for both antibodies.

[0129] No difference in specificity or sensitivity could be seen between APG1181 and clone 24-8

[0130] FIG. 6: Specificity of anti-CD95L rabbit monoclonal antibody in IHC-based analysis. Human tonsil sections, known to express CD95L in a subset of plasma cells were analyzed with the purified antibody from clone 24-8 in the absence or presence of the specific peptide (CD95L-NT_1-21) as indicated. The specific signal seen in the absence (clone 24-8) of the peptide is entirely missing in the presence of the competing peptide confirming the specificity of the antibody.

[0131] FIG. 7: Comparison of purified antibodies anti-CD95L monoclonal rabbit antibodies. Purified antibodies APG1181 and antibody from clone 24-8 were analysed for their specificity to recognise CD95L in IHC-analysis on glioblastoma sections. For the analysis two glioblastoma sections were chosen that were previously analysed as CD95L-negative or CD95L-positive, respectively, using a commercial polyclonal anti-CD95L antibody. As seen in the figure no difference in specificity or sensitivity could be seen between APG1181 and clone 24-8 in IHC-based analysis of glioblastoma tissue.

[0132] FIG. 8: Peptides used for epitope mapping of the CD95L-antibody.

[0133] The peptides shown were used for the identification of the epitope of the anti-CD95L rabbit monoclonal antibody. The peptides encode an N-terminal part of the human CD95L. The first peptide (CD95L-NT_1-21) was used for the immunization of rabbits for the generation of rabbit monoclonal antibodies. The shorter, truncated versions (peptides 2-7) and the respective permutated peptides (8-11) were used for fine mapping of the epitope of individual antibodies.

[0134] FIG. 9: Epitope mapping of anti-CD95L rabbit monoclonal antibody (APG1181).

[0135] For the epitope mapping the respective peptides (see FIG. 8) were coupled to BSA, the peptide-BSA-conjugates (as indicated) were immobilized to an ELISA plate and incubated with the recombinant anti-CD95L rabbit monoclonal antibody (APG1181) followed by a specific peroxidase-conjugated anti-rabbit secondary antibody. As shown in the figure APG1181 specifically recognises an epitope that encodes the N-terminal aa13-19 of human CD95L.

[0136] FIG. 10: Fine mapping of the epitope for anti-CD95L rabbit monoclonal antibody (APG1181).

[0137] Peptide-BSA-conjugates (as indicated) were immobilized to an ELISA plate and incubated with the recombinant anti-CD95L rabbit monoclonal antibody (APG1181) followed by a specific peroxidase-conjugated anti-rabbit secondary antibody.

[0138] (A) As shown APG1181 specifically recognizes an epitope that encodes the N-terminal aa13-19 of human CD95L. Using permutated peptides (8-11 FIG. 8) for fine mapping of the epitope reveals the Y(tyrosine) in position 13 is absolutely required for defining the specificity of the respective antibody APG1181/24-8. The W(tryptophan) in position 14 and the A (alanin) in position 19, respectively, do only have a marginal influence on the binding specificity (epitope) of the antibody.

[0139] (B) A different antibody (clone 39-9) that recognizes a similar epitope but does not have the absolute requirement for the Y in position 13, does surprisingly show no specific reactivity towards CD95L in IHC-based analysis.

[0140] FIG. 11: Comparison of the specificity of anti-CD95L rabbit monoclonal antibodies in IHC-based analysis.

[0141] Purified antibodies from clone 24-8 and clone 39-9 were analyzed for their ability to detect CD95L in a subset of plasma cells on human tosil sections. Although both antibodies share a similar epitope, clone 39-9 does surprisingly show no reactivity towards CD95L in IHC-based analysis. Based on this result it can be concluded that the Y in position 13 is a prerequisite for IHC reactivity of clone 24-8.

EXAMPLE 1

APG101 (CD95 Ligand Inhibitor) for Treatment of Cancer

A Phase II Study

[0142] The phase II study was a randomised, open-label, multi-centre study of weekly APG101 plus re-irradiation (APG101+RT group, test group) versus re-irradiation alone (RT group, control group) in the treatment of patients with first or second relapse or progression of glioblastoma. In total, 91 patients were included in the study. The ITT population consisted of 84 patients, 58 in the APG101+RT group and 26 in the RT group. Male and female patients with glioblastoma at first or second relapse either not being eligible for tumor resection or having macroscopic residual tumor after resection were candidates for this study. Eligible patients must have failed standard treatment that must have included radiotherapy (60Gy) and temozolomide and had to be a candidate for re-irradiation. Patients included in the study received either re-irradiation alone or re-irradiation plus 400 mg APG101 administered as a 30 min i.v. infusion once every week until progression.

[0143] Archived tumor tissue for histological analyses was available from 81 patients (of the ITT population) to confirm glioblastoma diagnosis. For the central histology review, biopsy or surgery material embedded in paraffin must have been available for central review in order to confirm the primary diagnosis of glioblastoma. All samples that were obtained were archived samples originating from the initial surgery and disease diagnosis or pre-study surgical resections. Following registration of the patient, a paraffin embedded tumor sample was collected by the study centre. Tissue sections were prepared from these tumor samples to assess various markers including CD95L using immunohistochemical staining (IHC).

[0144] The glioblastoma multiforme (GBM) patients participating in this clinical study were randomized into the control group and a test group.

Results

[0145] FIG. 1 describes the survival rate of the patients of the control group and the test group (overall survival rate, OS). The figure indicates that the patients of the test group receiving APG101 and radiation therapy have an improved survival rate compared with the control group receiving radiation therapy only (see Table 1).

TABLE-US-00001 TABLE 1 overall survival in all patients RT only (OS rates) APG + RT (OS rates) 6 m 77% 90% 12 m 50% 50% 18 m 23% 34% 24 m 7.3% 22%

[0146] Sections of GBM tumor samples of the control group patients and the test group patients were analysed for the presence of CD95 ligand (CD95L) by immunohistolochemical staining (see Example 2).

[0147] It was surprisingly found that the level and distribution of CD95L expression strongly differed among the patients. About one third of patients exhibited a strong positive CD95L expression in the tumor tissue. Yet another third of patients exhibited a CD95L negative phenotype. The remaining patients showed a slight expression of CD95L. This phenotype could be termed low positive phenotype.

[0148] In the test group patients receiving APG101 on top of radiation therapy (N=58), 18 patients exhibited the strong positive CD95L expression phenotype in the tumor tissue. 18 patients exhibited the CD95L negative phenotype. In the control group (N=26), 10 patients showed the strong positive CD95L phenotype, and 9 patients showed the CD95L negative phenotype (Table 2).

TABLE-US-00002 TABLE 2 Overall survival times of the test group (APG101 on top of radiation therapy, APG + RT) and in the control group (radiation therapy only, RT only). CD95Lpos. CD95L neg. N 95% (/+) median N 95% (/+) median APG + RT 18 200/500 350 18* 140/670 410 RT only 10 59/440 250 9 190/740 460 * 3 patients censored in the APG group (still alive) N: number of patients. The median is given in days. 95% (/+) indicates the 95% confidence interval.

[0149] Surprisingly, the overall survival time of the control group GBM patients was negatively correlated with the level of CD95L expression. The patients showing the strong CD95L positive phenotype of GBM had a worse prognosis than patients with CD95L negative GBM. CD95L positive patients survived 250 days, wherein CD95L negative patients survived 460 days (see Table 2).

[0150] In the test group, the patients showing the strong CD95L positive phenotype of GBM benefited from the treatment. CD95L positive patients survived 350 days.

[0151] The Kaplan-Meyer plot of FIG. 2 compares the survival data obtained in CD95L positive patients in the control group and the test group. Upon treatment with APG101, the curve is shifted to larger survival times. This means that patients showing the CD95L positive phenotype of GBM benefited from the administration of APG101. The median overall survival time increased from 250 days (control group) to 350 days.

[0152] In the test group, median overall survival in CD95 negative patients was 410 days. In comparison to the control group patients showing the negative CD95L phenotype, the median overall survival decreased from 460 to 410 days upon treatment with APG101. The data are described in the Kaplan-Meyer plot of FIG. 3. This figure indicates that the curves of the test group and the control group are largely overlapping and not statistically significant.

SUMMARY

[0153] This example demonstrates that the expression of CD95L in cancer is associated with a reduced survival of the patients. This means that CD95L is suitable as a prognostic factors in cancer. This is important in particular in those cancer types which express CD95L to a variable extent, for example glioblastoma. Therefore, the present invention enables a diagnostic differentiation of cancer by expression of CD95L. This leads to a new diagnostic differentiation of cancer, for example glioblastoma, in CD95L positive cancer and CD95L negative cancer.

[0154] Furthermore, this example surprisingly demonstrates that a CD95L inhibitor, for example APG101, can improve the survival of cancer in patients suffering from a cancer expressing CD95L. In contrast, patients suffering from a cancer which does not express CD95L, do not benefit from the treatment with a CD95L inhibitor. This example enables a therapeutic strategy comprising determining the CD95L expression in a cancer sample, and treating the patient if CD95L is expressed. This strategy is advantageous, because the CD95L inhibitor can be administered to those patients in which a therapeutic success can be expected. This is not a disadvantage for patients suffering from a cancer not expressing CD95L, because these patients probably do not benefit from a treatment with a CD95L inhibitor.

EXAMPLE 2

Immunohistochemical Determination of CD95L

[0155] CD95L can be determined in a tissue section (for example a tumor tissue section) by performing the following procedure: [0156] Tissue samples were fixed in 4% buffered formaldehyde and embedded in paraffin. [0157] 3 m thick sections were dewaxed:

1. Xylol 10 min.

2. Xylol 10 min.

3. Xylol 10 min.

1. Ethanol 100% 5 min.

2. Ethanol 100% 5 min.

Ethanol 96% 5 min.

Ethanol 90% 5 min.

Ethanol 70% 5 min.

[0158] 1. H.sub.2O dest. 1 min
2. H.sub.2O dest. 5 min. [0159] High temperature antigen demasking procedures:
Citrat pH 6.0 (Target Retrieval Solution), 99 C., (DAKO, Cat.No. S1699), 25 min. [0160] Cooling 15 min at room temperature [0161] Washing 2 5 min in PBS [0162] For tissue with high endogenous biotin: Biotin-Block (DAKO, Biotin Blocking System, X0590), for example for samples obtained from pancreas, liver, kidney, [0163] .fwdarw.Avidin 10 min. room temperature (RT) [0164] .fwdarw.Rinse in PBS 1 5 min [0165] .fwdarw.Biotin 10 min. RT [0166] .fwdarw.Rinse in PBS 1 5 min [0167] To block unspecific antibody binding, sections were incubated with blocking solution 1 (PBS, BSA 20 mg/ml; Serva, Germany, Cat.No.11924), human IgG 1 mg/ml, (Gammunex 10%, Talecris) for 20 min. [0168] 1.sup.st Antibody diluted in blocking solution 1 [0169] .fwdarw.Anti CD95L polyclonal rabbit (Dianova, Cat.No. DLN-14047) 1:100 [0170] .fwdarw.Rabbit IgG isotype control (Invitrogen (Zytomed), Cat.No. 08-6199) [0171] Incubation 60 minutes at room temperature. [0172] Washing 2 5 min in PBS [0173] Blocking with blocking solution 2 for 20 minutes. Blocking with blocking solution 2=blocking with blocking solution 1+20% normal goat serum, Dianova, Cat.No. 005-000-121. [0174] 2.sup.ndary antibody diluted in blocking solution 2, for example Goat F(ab).sub.2 Anti_Rabbit IgG (H+L chain) biotinylated, (SouthernBiotech, Cat.No. 4052-08) 1:100, Incubation 30 minutes at room temperature [0175] Washing 2 5 min in PBS [0176] Streptavidin-alkaline phosphatase (Concentrated AP label, BioGenex, Cat:No. HK321-UK), diluted in blocking solution 1, incubation 30 minutes at room temperature [0177] Washing 2 5 min in PBS [0178] Incubation with alkaline phosphatase substrate (DAKO Liquid Permanent Red, DakoCytomation GmbH, Glostrup, Denmark, Cat.No. K0640) [0179] Counterstaining with hematoxylin (Merck, Mayers Hmalaunlsung Cat.No. 1.09249.0500) [0180] Determination of the number of stained cells with respect to the total number of cells (% CD95L positive cells), or/and the size of the stained area with respect to the total area (% CD95L positive area). If tumor tissue is examined: determination of the number of stained tumor cells with respect to the total number of tumor cells (% CD95L positive tumor cells), or/and the size of the stained area of tumor tissue with respect to the total area of tumor tissue (% CD95L positive area of tumor tissue). Non-tumor tissue is disregarded.

EXAMPLE 3

Evaluation of CD95L Immunohistochemically Stained Slides

1. General Aspects

[0181] All CD95L-stained slides (see Example 2) were evaluated by a single board-certified neuropathologist with extensive experience in the field of neuro-oncology (C.H.). The whole evaluation was performed slide-by-slide in a single session. The longest intermission in the whole evaluation process was not longer than 60 min in between. The neuropathologist was blinded for the clinical data of the patients. The same neuropathologist had generated the reference histology diagnosis based on H&E- and silver staining combined with Mib1-, GFAP- and IDH1 R132H immunohistochemistry before. However, at the time of CD95L-evaluation these diagnosis-related slides were not available. The neuropathologist only had personal notices available about the diagnosis. Furthermore, the neuropathologist had evaluated the MGMT status and results from this analysis were known when the CD95L slides were assessed.

2. Generation of Calibration Figures

[0182] The CD95L immunohistochemistry staining was established by the company Apogenix. The evaluating neuropathologist (C.H.) was not involved in the technical aspects of the stainings, he was blinded regarding the protocols that were applied. After finalization of the CD95L immunohistochemistry establishing process a small number of CD95L-stained slides were committed to the neuropathologist to check for the quality. The neuropathologist confirmed that the applied immunohistochemistry allows high quality results and agreed to define this protocol to be the standard for further CD95L stainings. Based on these slides that were generated for quality assessment calibration figures were created that allowed comparison of CD95L immunohistochemistry intensities. Pictures were taken on a Carl Zeiss AxioPlan2 microscope with an AxioCamHR digital camera using the software AxioVision 4.8. The unmodified pictures were directly incorporated in CorelDraw v14.0 and arranged. Two pictures from different samples of similar CD95L immunohistochemistry intensities were set in one line. In summary, 2 lines of pictures representing the CD95L immunohistochemistry intensities CD95L positive (high) or CD95L negative (low) were generated (see FIG. 4).

3. Overview and Tissue Selection

[0183] In a first step the slide was scanned under the microscope (Olympus BX46) with low resolution to evaluate the amount of solid tumor tissue. Areas of normal brain parenchyma and areas of tumor infiltration were excluded from further analysis. Tumor tissue was selected for further evaluation if the following criteria were fulfilled: [0184] Tumor tissue that appeared to be vital without signs for hypoxic damage. [0185] Tumor tissue without resection-induced fresh bleedings. [0186] Tumor tissue without signs of cutting-induced artifacts.

4. Single Slide EvaluationCriterion Intensity

[0187] The suitable tumor tissue of each slide was evaluated regarding the CD95L positive or CD95L negative staining intensities. The CD95L calibration figure was used to standardize the evaluation. Because most tumors showed different staining intensities in separate areas the percentage of these summed areas were counted. Each tumor was assigned a specific value in percent representing the area showing CD95L positive or CD95L negative staining intensity:

TABLE-US-00003 Intensity Percentage Absent (CD95L negative) <2% CD95L positive >5%

[0188] Finally, the results were noticed in central pathology review form.

5. Single Slide EvaluationCriterion Distribution Pattern

[0189] A glial tumor (for example a glioblastoma) is composed of tumor cells and a fibrillary matrix of tumor cell processes. Using a light microscope these processes usually cannot be assigned to a particular tumor cell. If CD95L antibody binding was predominately seen in such cellular compartment of the matrix the staining pattern was defined to be diffuse. If instead clearly tumor cells were CD95L-labeled the staining pattern was defined to be focal. If essentially no CD95L labeling was observed in the tumor tissue, or the CD95L positive area was below 2% of tumor tissue, the distribution pattern was not evaluated. Finally, the results were noticed in central pathology review form.

6. Reporting of CD95L Results

[0190] Results from the CD95L evaluation were noticed in the central pathology review form (see above). The form was dated and signed by the responsible neuropathologist.

7. Generation of a Specific CD95L-Antibody Suited for IHC-Based Analysis

[0191] CD95L specific rabbit monoclonal antibodies employing immunization of rabbits with a synthetic N-terminal peptide, encoding the first 21aa of human CD95L (CD95L_NT 1-21: MQQPFNYPYPQIYWVDSSASS) were developed. For the selection of an appropriate antibody suited for the detection of CD95L in an IHC-based assay, a thorough characterization of candidate anti-CD95L rabbit monoclonal antibodies including IHC-analysis was performed. The screening process identified numerous antibodies that specifically recognized the N-terminal peptide used for immunization. However, one rabbit monoclonal antibody clone 24-8 was in particular suited for the detection of CD95L in IHC-based analysis.

[0192] To ensure constant antibody supply recombinant expression of the antibody derived from clone 24-8 was employed in HEK cells. The recombinant antibody derived from clone 24-8 is based on the sequence of the respective IgG heavy and light chain and was designated APG1181. The specificity of the recombinant antibody is identical to the antibody derived from the hybridoma clone 24-8 (see FIG. 5 and FIG. 7). The specificity of the antibody was characterized by ELISA, IHC-analysis and mapping of the specific epitope.