FR2992427A1 - METHOD OF PROGNOSIS OF A LYMPHOMA AND KIT FOR ITS IMPLEMENTATION - Google Patents

Abstract

The invention relates to a method for diagnosing and predicting the progression of lymphoma, particularly intraocular lymphoma, comprising the measurement of soluble CD25, and to a kit designed for the implementation of this method.

Description

The invention relates to a new soluble CD25 assay kit, usable in a diagnostic method and prognosis for the progression of lymphoma, in particular intraocular lymphoma. The term intraocular lymphoma includes a heterogeneous set of tumor pathologies affecting the ear and is manifested in most cases by posterior uveitis resistant to corticosteroid treatment. In all cases, it is a monoclonal lymphoid proliferation with an invasion of a targeted structure of the ear and according to the starting point of lymphoproliferation, there are primitive or secondary intraocular lymphomas. Primary (L10) intraocular lymphomas (or PIOLs for Primary Infra Ocular Lymphoma) are among the primary lymphomas of the central nervous system (PCNSL). Ocular involvement occurs with or without concomitant brain involvement and lymphoma cells preferentially invade the retina or vitreous. In the majority of cases, it is non-Hodgkin's lymphoma, diffuse large B-cell, high grade malignancy. PIOL is a rare lymphoma associating ocular and cerebral forms, of difficult diagnosis and putting the patient's vital prognosis at risk. In approximately 80% of cases it is chronic bilateral ocular involvement with subretinal or vitreous infiltration, but unilateral initial presentations are possible. A complication by a cerebral lymphoma is found in about 80% of the cases in the absence of treatment, and when the invasion is bifocal (eye and brain), the term oculo-cerebral lymphoma (LOC) is retained (Peterson et al. al, Cancer, 1993 Aug 1, 72 (3): 843-9). Clinical symptoms of PIOL are not specific, and may mislead the clinician to chronic posterior uveitis. The diagnosis is therefore difficult, often late, and requires a high degree of clinical suspicion on the part of the clinician who will have to be alerted in case of uveitis resistant to treatment with corticosteroids. Ideally, the treatment of ocular lymphomas should serve two purposes: eradicating the intraocular tumor and preventing relapses and brain complications. No optimal therapeutic management has been identified for the moment, the treatments being radiotherapy (ocular irradiation, the tumor being sensitive to the treatment, but not healing: relapses are frequent and a subsequent brain invasion is possible) or the chemotherapy (the main difficulty resides in the lack of penetration of anti-cancer molecules through the blood-eye barrier), in particular local chemotherapy (intravitreous administration of methotrexate). which allows a long-term remission and can be repeated many times. As seen above, because of the absence of specific clinical signs, the diagnosis of ocular lymphoma is not easy. Ophthalmological examination will reveal abnormalities in the anterior or posterior segment of the ear. These abnormalities will be explored by anterior chamber puncture (PCA) and the diagnosis will be confirmed by vitrectomy. The anterior chamber puncture is minimally invasive and performed first-line when a PIOL is suspected. The yield is low, the volume of the sample being about 100p1. It is mainly performed as part of the monitoring of treated PIOLs and for the detection of relapses because the sampling can be done at short time intervals. Vitrectomy is a surgical procedure performed under anesthesia, intended to partially or totally remove the vitreous body for diagnostic purposes. In practice, it is performed when the results of the PCA examination evoke a PIOL and must be done at a distance from a corticosteroid therapy in order to avoid a false negative result by lysis of the tumor cells. The vitreous is fragmented thanks to a vitreotome placed inside the eye, and the suction is compensated by a fluid intake (hypertonic saline solution, Borate Buffer Saline) to avoid the risk of ocular hypotonia. The first aspiration is considered to be "pure" vitreous, which should be preferred for the diagnostic cytokine assay. The following aspirations report "diluted" vitreous, which can be used for cytological examination. Retinal biopsy is performed when vitrectomy is not contributive. The samples thus taken are then examined, in particular by cytological analysis (performed on the vitreous, taking into account the small volume of the PCA), which makes it possible to make a definitive diagnosis of PIOL and to start a treatment rapidly. The T or B monoclonal character of the tumor cells can also be demonstrated by molecular analysis (polymerase chain reaction, PCR) making it possible to identify rearrangements in the majority of intraocular lymphoma cases (translocation between chromosomes 14 and 18 or [t (14; 18)], involving IGH and BOL-2 genes, reported in more than 67% of cases). Interleukin-10 (1L-10) and interleukin-6 (1L-6) assays in aqueous vitreous and aqueous samples are now part of the systematic reviews and provide an almost indispensable a diagnosis when intraocular lymphoma is suspected. Indeed, tumor B cells secrete relatively high levels of IL-10 whereas inflammatory reaction cells produce IL-6. The determination of these two cytokines and the determination of their ratio therefore give a good idea of the tumor and inflammatory component involved in the pathological process (Merle-Béral et al., Br. J. Haematol, 2004 Feb; 124 (4): 469-73). IL-10 acts as a factor of tumor growth by promoting the proliferation of B-cell lymphomas (Beatty et al., J. Immunol 1997 May 20 1; 158 (9): 4045-51), and is produced by lymphoma cells at very high levels. Intraocular levels of IL-6 are elevated in uveitis and may be very important in progressive eye infections (toxoplasmic chorioretinitis, acute retinal necrosis related to varicella zoster virus) (Ongkosuwito et al., Invest Ophthalmol, Vis, Sci 1998 Dec, 39 (13): 2659-65). IL6 is an important mediator of intraocular inflammation (Hoekzema et al., Curr Eye Res 1990; 9 Supp1: 207-11) because of its active role in the amplification of leukocyte recruitment, and its dosage in ocular fluids make it possible to make the differential diagnosis between an infectious, inflammatory or autoimmune panuideitis and a pseudo-uveitis due to a PIOL (Cassoux et al., Invest Ophthalmol, Sci., 2007 Jul; 48). (7): 3253-9). In most published studies, the IL-10 and IL-6 assay is based on a sandwich ELISA. Several ready-to-use kits are marketed. Assays can be performed on aqueous humor or vitreous.

The threshold of IL-10 in the aqueous humor for which the probability of PIOL is probable is 50 ng / mL, with a sensitivity of 89% and a specificity of 93% (assay by ELISA test, Quantikine® kit, R & D System). For vitreous samples, the cutoff is 400 μg / mL with a sensitivity of 80% and a specificity of 99% (Cassoux et al, op.cit.). On the other hand, the dilution of the vitreous humor taking place during the vitrectomy is a parameter to be taken into account for the interpretation of the results. The determination of the ratio between IL-10 and IL-6 makes it possible to overcome the effect of the dilution, each cytokine being diluted in the same proportions. An IL-10 / IL-6 ratio greater than 1 suggests a lymphomatous process (VVhitcup et al., Ophthalmol Arch, 1997 Sep; 115 (9): 1157-60). Thus, according to a study of 35 cases of PIOL and 64 uveitis, the value of the ratio greater than 1 confirmed the diagnosis of PIOL in 74% of cases, with a sensitivity of 74.3% and a specificity of 74.7%. The main disadvantage of this technique is the large volume of sample needed for each test. More recently, the detection of these cytokines has been implemented using a flow cytometry technique, called Cytometric Bead ArrayTM (CBATM marketed by BD Biosciences). This technique brings a real progress in the diagnosis of PIOL, because it offers the possibility of simultaneously detecting and quantifying several cytokines in a single, low volume sample, unlike the ELISA technique that is usually used as a sample-consuming device. It is a "sandwich" type detection, using fluorescent polystyrene microbeads coated with an anti-cytokine antibody (primary antibody), serving as a capture reagent, and a specific detection antibody ( secondary antibody) coupled to a fluorochrome, phycoerythrin (PE), to reveal the complex. The beads are homogeneous in size and structure but have a specific fluorescence intensity, defined by the manufacturer, easily identifiable with the cytometer by a point cloud well individualized on a dot plot representing the fluorescence in APC (Allophycocyanine) / APCCy7 (Allophycocyanin bound with cyanin-7 (Cy7), emitting by FRET, Fluorescence Resonance Energy Transfer). For the APCCy7, the APC is excited at 633nm. It emits, alone at 665nm, and functions as a donor for the Cy7 when it is linked to it. The Cy7 emits at a wavelength of 767nm. Thus, depending on the amount of fluorescence in the beads and APC / APCCy7 ratio, each point cloud corresponding to a type of ball is associated with an alphanumeric position.

Each type of ball is coupled to a cytokine-specific primary antibody, so that each given position corresponds to a cytokine. For example, in the BD-Bioscience Human IL-10 Flex Set kit, the beads coupled to primary antibodies to capture IL-10 are positioned at B7.

After binding between the ball and the cytokine, the complex is revealed by the secondary antibody coupled to PE, whose fluorescence intensity allows the quantification of the cytokine of interest. This quantification requires the parallel production of a standard range established with a recombinant cytokine. Thus, the addition to a sample of several beads of different specificities will make it possible to simultaneously dose several cytokines in a single reaction from a very low sample volume. Currently, there are no clinico-biological elements can give a prognostic value. Such a tool could prove useful to the clinician, especially to anticipate the therapeutic management of patients whose prognosis would be bad from the outset. The Applicant has thus evaluated the prognostic value of several intraocular soluble factors. The assay of these biomarkers was combined with those of IL-10 and IL-6 using the same flow cytometry analysis technique (CBATm), so that all these analytes can be assayed simultaneously. a constant and low sampling volume irrespective of the number of analytes to be assayed, ie a volume of 50p1. Four cytokines have been selected: CD255 (or soluble IL-2 receptor), MI P-1a (Macrophage Inflammatory Protein 1 alpha), and RANTES (Regulated upon Activation, Normal T cell Expressed and Secreted, as well as Each of these cytokines seems to be of interest for the diagnosis and / or prognosis, which is demonstrated for the first time, but the cytokine CD25s is the most interesting.The interleukin-2 receptor ( 1L-2) exists in two forms: a membrane form, located on activated T, B and NK (Natural Killer) lymphocytes, and a free, detectable form in serum.

The membrane form is a heterotrimer consisting of three protein subunits: an IL-2Ra subunit (also called Ag Tac or CD25 or p55), an IL-2R subunit [3 (p75) and an IL subunit -2Ry (p64). The α chain is a 55 kDa glycoprotein consisting of 251 amino acids, 219 of which are extra cellular, and constitutes the specific recognition site for IL-2. This α chain is the result of an open reading phase leading to the production of a protein of 272 amino acids (SEQ ID No. 1), whose signal peptide of 21 amino acids is cleaved (Kuziel and Greene, J Invest Dermatol 1990 Jun; 94 (6 Suppl): 27S-32S). The free form of the receiver consists only of the chain a. This soluble portion, designated II-2Ras or CD25s, is a 45kDa protein, detectable in body fluids and capable of binding specifically to IL-2. This protein has three epitopic regions A, B and C. The Applicant has shown that the amount of CD25s in a patient's ocular sample was a factor in anticipating the likelihood of a clinical course of the disease, and could also be used to refine the diagnosis in some cases. Moreover, the Applicant has also shown that certain other factors (IFNy, MIP-1a, RANTES) can also be used for diagnostic purposes and prognosis of PIOLs. In view of the constraints related to the amount of sample available in these diseases, the Applicant has therefore developed a kit for detecting CD25s, which is particularly useful by flow cytometry. In particular, the Applicant envisages that this detection kit can be used in parallel with the detection of other analytes, including IL-10 and IL-6. It is indeed important that the detection of many factors can be performed simultaneously, because of the small volume present in the samples taken from the patients. The Applicant has, for the first time, and without the results reported in this application have been expected or predictable, demonstrated that CD25s was an important analyte to be assayed to make a prognosis for PIOL evolution. The Applicant was therefore the first to ask the question of the interest of developing a test for the detection of this compound, which can be used for small sample volumes, and simultaneously with the detection of other compounds. Thus, the Applicant has, in particular, developed a detection system that can be used with the CBATM system developed by BD Bioscience.

The invention thus relates to a detection kit for CD25s comprising a set of beads emitting fluorescence under excitation at an excitation wavelength, said fluorescence emitted by said beads being at a wavelength of emission different from the excitation wavelength, each of the beads comprising, coupled to its surface, a plurality of antibodies directed against the CD25s protein. In a particular embodiment, the wavelength of the fluorescence emitted, after excitation, by each of the balls of the assembly present in the kit is identical. In a particular embodiment, the antibodies of the plurality of antibodies directed to the CD25s protein are all identical: this means that each of the beads present in the set of beads contains, on its surface, a plurality of copies of the same antibody directed against CD25s (i.e. having the same sequence). Thus, in this embodiment, the plurality of antibodies directed against the CD25s protein is a plurality of monoclonal antibodies, i.e., the same monoclonal antibody recognizing CD25s is present, in many instances, at the surface of each of the balls. In another embodiment, the antibodies of the plurality of CD25s directed antibodies are not all identical, but all recognize the same epitope (or domain) of the CD25s protein: this means that each of the beads present in the CD25s The bead set contains, on its surface, several identical antibodies against CD25s (i.e. having the same sequence). In a preferred embodiment, the excitation wavelength is less than or equal to 650 nm. Preferably, this excitation wavelength is equal to 633 nm. In this embodiment, the emission wavelength (which is detected after excitation) is greater than 655 nm. In particular, it may be 660-665nm or 767-773nm. The 633nm wavelength corresponds to a wavelength permitting the excitation of allophycocyanin. This emits at a wavelength around 660nm, detectable with a 665nm filter. When coupled to cyanine-7, the cyanine 7 is indeed excited by the fluorescence emitted by the allophycocyanin and then emits at 767nm or 773nm, detected at 767nm with a suitable filter. This radiation is based on fluorescence resonance energy transfer (FRET) fluorescence resonance energy transfer. In a particular embodiment, said beads emit fluorescence at two different wavelengths. This is the case when the beads contain two fluorophores emitting at these two wavelengths. Such beads are described in particular in patents US Pat. No. 7,445,844, US Pat. No. 6,929,859, US Pat. No. 6,632,526, US Pat. No. 6,599,331 and US Pat. No. 6,514,295. In a particular case, said emission wavelengths are 660-665 nm and 767-773 nm, which corresponds to beads containing both allophycocyanin and allophycocyanin coupled to cyanine-7. In this embodiment, and as seen above, said beads contain both allophycocyanin and allophycocyanin coupled to cyanine-7, the amount and ratio being the same for each of the beads. Thus, the passage of the beads in a flow cytometer, the excitation at a wavelength of 633 nm, and the collection of the amount of fluorescence emitted by the beads at 655 nm and 767 nm, makes it possible to observe, on a 2D diagram. (abscissa APC fluorescence amount, ordered amount of APCCy7 fluorescence), the beads at a specific position of this diagram. In a preferred embodiment, said beads are polystyrene microbeads. They preferably have a diameter of between 5 and 10 μm, more preferably 7.5 μm. In another embodiment, said beads are silica beads (see US 20090068639). In a particular embodiment, said detection kit also contains secondary antibodies capable of binding to CD25s. In the preferred embodiment, these secondary antibodies should be capable of binding to CD25 after binding of this protein to an antibody present on the surface of said beads. It is preferred that these secondary antibodies are bound to a fluorescent compound. The selected fluorochrome preferably has excitation and emission wavelengths different from those of fluorochromes present in the beads.

Thus, a fluorochrome having excitation and emission wavelengths of less than 600 nm is preferably chosen. We can choose phycoerythrin (PE), or any other fluorochrome. Those skilled in the art know other usable fluorochromes (listed in particular on wikipedia).

As seen above, it is preferable that the antibodies present on the surface of the beads all recognize the same epitope of the CD25s protein. In particular, it is preferred that they recognize the B epitope of the CD25s protein. Similarly, the secondary antibodies all recognize the same epitope or domain of CD25s. In particular, it is preferred that they recognize the epitope (domain) A of the CD25s protein. Finally, it is preferred that the kit according to the invention also contains a recombinant human CD25s solution, which makes it possible to perform more precise controls and quantifications. In a preferred embodiment, the detection kit comprises a) polystyrene microspheres, having in particular a diameter of about 7.5 μm, containing a mixture of cyano-7-linked allophycocyanine-associated fluorophores, said beads comprising, coupled to their surface, a plurality of murine monoclonal antibodies directed against the human CD25s B epitope b) a solution of murine monoclonal antibodies directed against the human CD25s epitope A, linked to phycoerythrin This kit comprises optionally also a recombinant human CD25s solution. In a particular embodiment, the beads appear in A4, after passing through a BD FACSCant0 ™ IL flow cytometer (BD Bioscience), and analysis by appropriate software such as FACSDiva (BD Bioscience). This means that the fluorescence emitted by the beads is similar or identical to the fluorescence emitted by the reference beads 558578 sold by BD Bioscience. However, any other beads can be chosen, in particular the balls A5, B4, B5, E5, E8, C8, D8 (BD Bioscience classification). All these beads are available from BD Bioscience. They differ only in the amount and ratio of APC and APCCY7 fluorochromes.

In a preferred embodiment, the detection kit is used in flow cytometry to assay CD25s in a sample obtained from a patient. In a particular embodiment, this sample is a vitreous or other intraocular sample, and other cytokines are concurrently dosed. Thus, the detection kit may also contain at least one other set of beads, said beads comprising, coupled to their surface, a plurality of antibodies recognizing another protein than the CD25s protein, said beads emitting fluorescence under excitation at the same excitation wavelength than that exciting the beads for assay CD25s, said fluorescence emitted by said beads being a) at an emission wavelength different from said emission wavelength of the beads covered with the plurality of antibodies directed against the CD25s protein, and / or b) at an intensity different from the intensity observed for the covered beads of the plurality of antibodies directed against the CD25s protein. In particular, said other protein as the CD25s protein is selected from the group consisting of interleukin 6 (IL-6) and interleukin 10 (IL-10). In particular, beads containing the same fluorochromes (APC / APCCy7) are chosen as the beads intended for the detection of CD25s, but in different proportions and / or quantities. Thus, these beads of the second set will be located on the 2D diagram (abscissa APC fluorescence quantity, ordered amount of APCCy7 fluorescence), the beads at a specific position of this diagram which is different from that of the anti-antibody coated beads. -CD25s. The kit can thus contain several sets of beads, each intended for detecting and assaying a different protein, the beads of each set having a clean fluorescence profile and different from that of the beads of another set. It is preferred when the kit allows simultaneous assay of IL-6 and IL-10 (in addition to CD255). In particular, other sets of beads can be obtained from BD 35 Bioscience (San Jose, California, USA), for example under the reference 560484 (Human Th1 / Th2 / Th17 CBA Kit which groups together beads for the simultaneous detection of IL-2, IL-4, IL-6, IL-10, IL-17A, IFN-γ, and TNF), or cytokine by cytokine. For example, mention may be made of reference 558274 (IL-10, beads B7) or 558276 (IL-6, beads A7) from BD Bioscience.

The kit thus described, although it can be used in other applications, has been specifically designed for the implementation of the in vitro measurement (assay) of the amount (concentration) of CD25s in an eye sample in a patient. . Indeed, as seen above, the Applicant has specifically developed this kit, in order to perform simultaneous assays of several cytokines in a sample of very low volume, because of the interest of CD25s in the prognosis of primary intraocular lymphoma. The invention also relates to a prognostic method for intraocular lymphoma in a patient, comprising the step of measuring (in vitro) the amount of CD25s in an eye sample in said patient, a concentration of CD25s greater than 200 pg / ml in said sample being indicative of an unfavorable prognosis of progression of the disease in said patient. In a preferred embodiment, this eye sample is a vitreous sample. The concentration measured in vitro is then that actually present in vivo. When the sample is not pure vitreous, it is then necessary to know the dilution to recalculate the effective concentration present in vivo. Thus, if the specimen is vitreous diluted to 50%, the concentration measured in vitro should be multiplied by two to be able to conclude. It is understood that the method thus described is carried out in vitro on a sample which has been obtained after sampling from a patient. This method does not include the eye sample collection step. This measurement (assay) is especially carried out using the kit described above, which is particularly well suited to such use, having been developed for this purpose. Thus, this assay is preferably carried out by flow cytometry. Commercially available apparatuses such as two-laser cytometers developed by BD Bioscience can be used. Instrument Detection of Antibodies BD FACSVerseTM Secondary B Detection Parameters for PE CBA Red and CBA NI R BD AccuriTM Cytometer 06 Flow FL2 F L4 (675/25) and F L3 Cytometer * (780/60) BD FACSArray TM Yellow Red and NIR BD FACSCantoTM II Bioanalyzer for AP APC Cyconomer and APC-CyTM7 BD LSRFortessaTM for AP APC and APC-Cy7 Cytometer BD FACSAriaTM III APC APC Cell and APC-Cy7 BD FACSCaIiburTM BD Flow Sorter FL2 FL4 and FL3 Cytometer With software such as BD FACSDiva Software, it is possible to isolate the beads corresponding to a particular APC / APCCy7 pair, and to display the PE fluorescence of the population identified on a histogram representing the fluorescence intensity on the abscissa and the number of events on the ordinate. The software renders an average fluorescence intensity (MIF) for each tube, corresponding to a concentration determined from a standard range made in parallel. By "unfavorable prognosis of disease evolution" is meant a high probability (greater than 50%) that the patient's disease presents at least one following evolution: - relapse in the year following discontinuation of treatment - no Treatment response (treatment refractory) This means in particular that more than 50% of patients for whom the concentration of CD25s is greater than 200 μg / ml will exhibit the clinical course mentioned above. Similarly, among patients with such a clinical course, more than 50% have a CD25 concentration greater than 200 μg / ml. In particular, an amount of CD25s greater than 500 μg / ml in the ocular sample is indicative of a high probability of having a relapse within one year of stopping treatment.

Similarly, a CD25 concentration of less than 125 μg / ml, or even 150 μg / ml in the ocular sample is indicative of a high probability of complete remission. In order to refine this prognosis aspect, it may also be of interest to measure the amount of IL-6 in said sample. Indeed, the presence of high levels of these two inflammatory cytokines is indicative of a poor prognosis for the evolution of the disease. Thus, an amount of IL-6 greater than 50 μg / ml is indicative of an unfavorable prognosis of disease progression. In particular, an amount of L-6 greater than 100 μg / ml in the ocular sample is indicative of a high probability of having a relapse within one year of stopping treatment. Finally, it is possible and possible to dose and measure the amount of CD25s in a cerebrospinal fluid sample of a patient, in particular by using the kit thus described, in diagnostic methods and / or prognosis of a lymphoma of the system. central nervous system. This assay is of course performed in vitro. In general, in the methods described above, the in vitro measurement step of the amount of CD25s (CD255 assay) is preferably carried out using a kit as described above. Description of the Figures Figure 1: Representation of the alphanumeric position of the beads carrying the antibodies to detect cytokines according to the standard developed by BD Bioscience for the CBATM technique, in APC / APC-Cy7 representation Figure 2: Summary of clinical information obtained for 25 patients with IOL. Figure 3: Intravitreal concentration of CD25s measured in patients in complete or refractory treatment-refractory (A.) who had a relapse 30 within one year after stopping treatment (relapse <1 year) or more than one year after stopping treatment (relapse> 1 year) (B.), figure C. synthesizing figures A. and B. RC: complete remission Examples Example 1. Preparation of a kit for the detection of CD25s by CBATM a) Preparation of the Beads Coupled to Anti-CD25 Antibodies The CD25s assay in CBATM can be performed by carrying out a coupling of beads with specific antibodies. For this purpose, non-coupled beads were selected, a pair of monoclonal antibodies each recognizing an epitope different from CD25s, one purified, intended to be conjugated to the beads, and the other labeled with PE. A standard recombinant of CD25s allows to realize the standard range. The beads were chosen such that they have a position on the APC / APCCy7 dot plot far away from the other beads that can be used in parallel for the determination of the other cytokines. The choice of beads in A4 position (BD Biosciences, ref 558578) was retained because this position is quite far from that of IL-6, IL-10, MIP1-a, RANTES and IFN-γ positioned respectively at A7, B7, B9, D4 and E7. The predictive position of the cytokines is shown in FIG. 1. The primary antibody selected is a purified monoclonal antibody of human anti-CD25 NA / LE mice (clone M-251) directed against the B domain (epitope) of the protein. The secondary antibody is a human anti-CD25 mouse monoclonal antibody (clone 2A3), directed against the protein (epitope) A domain, coupled to phycoerythrin (excited at a wavelength of 488nm, and emitting a flourescence at a wavelength of 580nm). Both antibodies can be purchased from BD Biosciences (San Jose, California) as 555429 and 341011 respectively. These antibodies or other antibodies are also available from other suppliers (eg Stemcell, or Miltenyi Biotec). The anti-CD25 primary NA / LE antibody was coupled to the A4 beads according to the manufacturer's recommendations. The principle of the conjugation reaction is to reduce the disulfide bridges on the surface of the polystyrene beads by means of a reducing agent, dithiothreitol. In parallel, the primary amino functions of the antibodies react with a hetero-bifunctional conjugating agent, sulfo-SMCC, and form a highly reactive maleimide derivative. The latter can then form a stable thioether linkage with the thiol functions of the beads via a radical reaction, thus resulting in the covalent anchoring of the antibodies. B) Labeling procedure The beads coupled to the primary antibody were used for the detection of CD25s, according to the following labeling procedure: for each test, 50 μl of sample were incubated for 1 hour in the dark with 1 μl of conjugated A4 beads previously washed and diluted in washing buffer.

The anti-CD25 secondary antibody coupled to PE diluted 1/20 was then added. After a 2 hour incubation and a final wash, the sample can be analyzed by cytometer. c) Analysis on FACSConomer IITM cytometer - Standard ranges - Linearity Each tube containing the reaction medium was analyzed on a FACSCONT0IITM 8-color cytometer (BD Bioscience, Becton Dickinson), in manual mode, by acquiring at least 300 events at a "medium" speed. . The data was analyzed using BD FACSDiva Software (Becton Dickinson) computer software.

The analysis strategy first includes conditioning all the beads in a size / structure analysis window. They are then fenestrated in APC and APC-Cy7 so as to identify the maximum A4 beads. This last selection makes it possible to eliminate from the analysis window all the impurities related to the conservation or the doublets of balls generated during the coupling step. The PE fluorescence of the population surrounded by A4 can be visualized on a histogram representing the fluorescence intensity on the abscissa and the number of events on the ordinate. The software renders an average fluorescence intensity (MIF) for each tube, corresponding to a concentration determined from a standard range made in parallel.

On the basis of the measurements made on the standard range, it has been observed that there is good linearity for the values between 0 and 10000 μg / ml. Thus, in subsequent assays, the end point of the range should not exceed 10000 pg / ml.

As the CD25s assay had never been tested on intraocular samples, it was not possible to predict with certainty what the expected concentration range would be prior to testing patient samples. However, it was considered that the results should be of the same order of magnitude as those of IL-10 and IL-6, for which the maximum value detected in CBA is 5000 μg / ml. Thus, the upper point of the calibration range of CD25s has been set at 8000 pg / ml: this point is in the range of linearity of the curve and makes it possible to scan a very wide concentration range. The results of assaying CD25s greater than 8000 pg / ml were thus made "> 8000 pg / ml".

A 12-point standard range with more diluted CD25 concentrations ranging from 7.8 μg / ml to 8000 μg / ml was performed. The results of this range show that the fluorescence intensity corresponding to the very low concentrations of between 0 and 30 μg / ml do not vary proportionally. The detection threshold was thus set at 30 μg / ml. d) Multiplex assay test In order to test the compatibility of the beads thus designed with other commercial beads, to demonstrate the possibility of performing a detection of all cytokines via a single experiment, the reagents allowing the detection of CD25s, IL-10, IL-6, and IFN-γ were combined to see if all these cytokines could be assayed simultaneously, without interference their detection. For this, a standard range was made from a mixture of recombinant CD25s, IL-10, IL-6, and IFN-γ. Each cytokine is initially at a concentration of 5000 μg / ml in the first tube which constitutes the upper point of the range. It was diluted cascade to obtain the different points of range, each cytokine then being diluted in the same proportions. Finally, the mixture of all the reagents allowing the detection of each cytokine was added to each tube corresponding to a point of range. It has been observed that the balls do not interfere with each other for their detection since the linearity is good for each calibration line. It can be noted, however, that the signal intensity is lower for the CD25s, even though the linearity remains quite acceptable. Example 2. Labeling for Multiplex Assay of CD25s, IL-10, IL-6, MIP1-a, RANTES and IFN-γ in CBATM in Ocular Samples This procedure describes the cytokine labeling technique which has been applied. Ready-to-use kits (Flex Set BDTM CBA kits) were used for IL-10, IL-6, IFN-y, MIP-la and RANTES assays. Each kit makes it possible to carry out 50 tests and contains the capture beads coupled to the primary antibody, the secondary antibody, and a standard recombinant in freeze-dried form to be reconstituted at the time of preparation of the standard range. The CBATM kits used were developed by BD Biosciences and are sold under the references 558274 (IL-10), 558276 (1L-6), 558324 (RANTES), 558449 (MIP-1a) and 561515 (IFNy). For CD25s, the functional A4 beads described above, the secondary anti-CD25 antibody and the standard recombinant in liquid form were used.

Buffers required for the dosage recommended by the manufacturer are provided in the CBATM Human Soluble Protein Master Buffer Kit, BD Bioscience. a) Preparation of the capture bead solution The beads are to be used at the rate of 1 μL per test, ie 50 μL of each bead for a series of 50 tests. They are first washed in 500 μl of VVash Buffer BD wash buffer, centrifuged for 5 min at 1500 rpm. The supernatant is gently aspirated and the beads are resuspended in 2500 μl of Capture Bead Diluent for Serum / Plasma dilution buffer, or 50 μl per test. The solution is incubated for 15 min at room temperature and in the dark to avoid exciting the fluorochromes present on the beads. b) Extemporaneous Preparation of the PE Secondary Antibody Solution For IL-10, IL-6, IFN-γ, MIP-1α and RANTES, 1 μl of secondary antibody is required for assay, or 50 μL of each antibody for a series of 50 tests. As the minimum necessary amount for a test that was not accurately assessed for CD25s, the same amount of secondary antibody that was chosen for technical development, ie 5 μl per test, was used. A total of 250 μl of anti-CD25 antibodies are required for 50 tests. The mixture of all the necessary antibodies for a series of 50 tests represents a volume of 500 μl. This volume is diluted quarterly in 2 ml of Diluent Detector buffer and 200 μl of this dilution are necessary for each test. C) Marking of the samples The marking is carried out on a volume of 50 μl of sample. A volume of 50 μl of the mixture of beads, previously vortexed, is added to the sample and then incubated for 1 hour in the dark and at room temperature. After this incubation, 50 μl of the PE secondary antibody mixing solution is added to the sample. The mixture is incubated for 2 hours in the dark and at room temperature. At the end of this step, the samples are washed to remove excess secondary antibodies. A volume of 1000 μl of Buffer buffer VVash BD is added, the sample is centrifuged for 5 min at 1500 rpm. The supernatant is removed gently and the bead pellet is resuspended in 300 μl of VVash Buffer BD buffer. Samples are ready for cytometer analysis. d) Cytometer Analysis Before each cytometer run, the sample should be vortexed to resuspend the beads. The analysis should be done at the end of the labeling step to avoid decoupling between the beads and cytokines that may lead to signal loss. The samples are analyzed by hand in the cytometer by acquiring at least 300 events at a "medium" speed. The data is analyzed using the BD FACSDiva SoftwareTM computer software and switched to the FCAP-ArrayTM results processing software. e) Technical validation of the test on intraocular samples The cytokine multiplex assay on 15 intraocular samples from patients with PIOL (7 PCA and 8 vitreous) and 18 intraocular samples from 25 patients with uveitis was performed on preserved specimens. at -80 ° C for about 2 years. IL-10 and IL-6 were measured initially and simultaneously with the CBATM technique. The multiplexed assay of the 6 cytokines was performed after thawing the samples, and the results of IL-10 and IL-6 obtained before and after freezing were compared to see if, on the one hand, the preservation is impairing the quality of the sample and secondly to see if the simultaneous assay of 6 cytokines has an impact on their respective results. The results are well correlated for IL-10 (R2 = 0.9511) as well as for L1 (R2 = 0.9873) which led to the conclusion that freezing did not occur. 35 resulted in degradation of the material and on the other hand, that the simultaneous determination of 6 cytokines does not interfere with the results of each cytokine. The test thus developed is therefore technically valid and provides reliable results. Example 3. Assay of Different Cytokines in Patients Apart from IL-10, IL-6 and IFN-γ which are regularly routinely assayed in intraocular samples, the dosage of CD25s, RANTES and MIP1 -a had never been practiced in this type of levy. After having developed the combined assay of these 6 cytokines in CBATM technique, it was applied to a cohort of patients with intraocular lymphoma (L10) to see, on the one hand, whether these cytokines are detectable in intraocular samples, and on the other hand, to study the clinical significance of the results. a) Patients and samples The study looked at a retrospective cohort of 36 patients with IOL, all from the Pitié-Salpêtrière ophthalmology department and for whom the initial diagnosis was made by cytological examination on a patient. vitreous humor sampling and / or IL-10 and IL-6 assay (assayed in CBATM technique, except for samples taken before January 2009 for which the assay was performed using ELISA).

The multiplex assay of the 6 cytokines was performed on the vitreous humor samples corresponding to the diagnostic sample for each of the 36 patients. All specimens were stored at -80 ° C and in sufficient quantity to perform a new assay, ie a collection volume of at least 50p1. All these samples were taken between April 2004 and January 2012. The cohort included 17 men and 19 women. The average age was 65, with extreme values ranging from 44 to 88 years. For 25 patients, clinical information was obtained from the medical records made available by the Ophthalmology Department.

In parallel, a "non-PIOL" control cohort was established, comprising 16 vitreous humor samples from patients with a cytokine profile of uveitis (raised intra-ocular IL-6 concentration, low IL-10) and two patients with which IL-10 and IL-6 were not detectable. Clinical information could not be collected for these 18 patients. b) Results of IL-10 and IL-6 Levels of IL-10 in the vitreous of patients with IOL are on average 1787 pg / ml whereas they are of 5 pg / ml in the samples patients in the control group. The median concentrations of each cytokine were compared between the two groups with a non-parametric Mann-Whitney test, adapted to small series. Median IL-10 concentrations were significantly higher in the IOL group (P <0.05). In contrast, median IL-6 concentrations were higher in the uveitis group (P <0.05). The summary results of the assays are shown in Table 1.

IL-10 (μg / ml) IL-6 (μg / ml) Ratio I L-10 / 1L-6 IOL Uveitis PIOL Uveitis IOL Uveitis Mediane 1110 2 52 207 17.7 0.003 Mean 1787 5 265 1406 37 0.008 Standard deviation 1774 4.7 824 2092 44 0.2 Minimum 80 1 6 2 0.9 0 Maximum 5000 14 5000 5000 211 0.5 Table 1 Results of IL-10 and IL-6 in the group of intraocular lymphomas (n = 36) and the uveites (n = 16) According to a study carried out in the applicant's laboratories, involving 56 vitreous glasses in which IL-10 and IL-6 were measured in CBA, the threshold of 37 pg / ml IL-10 has been proposed to carry the diagnosis of IOL. The minimum IL-10 value detected in this IOL series is 80 μg / ml, which is consistent with this proposed diagnostic threshold. In addition, the calculation of the I L-10 / IL-6 ratio is often considered relevant for providing diagnostic assistance when its value is greater than 1 (VVhitcup et al, op.cit.). This ratio has been determined, and it has been observed that it is consistently greater than 1, with the exception of a patient for whom it is 0.9. These results thus confirmed the diagnosis of IOL or uveitis initially made in each group and made it possible to confirm the technical validity of the test that we have developed, since the initial diagnoses remain unchanged.

The average IFN-γ concentration in the IOL group is 8 μg / ml whereas it is 3 μg / ml in the uveitis group. Medians are significantly different (P <0.05). The dispersion of results is low in both groups, and only 3 patients in the IOL group have higher concentrations at 21, 37 and 124 pg / ml. However, given the significant overlap of concentration values between the two groups, IFN-γ does not appear to be a relevant parameter for the diagnosis of IOL. c) Results of CD25s, RANTES and MIP-la The concentrations of RANTES and MIP-1a are higher in IOLs (P <0.05 for RANTES and for MIP-1a) but there is a significant overlap of values between the two groups over a range of between 0 and 50 μg / ml for RANTES and between 0 and 40 μg / ml for MIP-1a. The few patients for whom the values of RANTES and M IP-1a are high correspond to already high IL-10 values. These markers do not provide additional diagnostic assistance. CD25s concentrations were not significantly different in the two groups of studies, but there was significant heterogeneity of the results in the IOL group. Values ranged from 30 to 8000 pg / ml in the IOL group, ranging from 30 to 837 pg / ml in the uveitis group. High concentrations of CD25s tend to be associated with elevated IL-10 concentrations, but there is no correlation between the two parameters (R2 = 0.4265).

Conclusion CD25s, RANTES, MIP-la and INF-y do not appear to be more relevant than IL-10 and IL-6 for the diagnosis of IOLs. However, they could be used as second-line to refine a prognosis, but the dosage of IL-30 10 and IL-6 and calculation of the ratio remain a priori the best elements of the diagnosis. Example 4. Interpretation of results: prognostic slope For each patient, a limited number of parameters (when they were available) were studied: diagnosis, unilateral or bilateral ocular presentation, the presence of concomitant brain damage to the diagnosis , time to onset of secondary cerebral involvement, type of response to first-line treatment (complete response greater than one year, partial response or no response), time to relapse, and death after announcement diagnostic. a) Clinical description of 25 patients from the IOL cohort Clinical information could only be obtained for 25 patients from the IOL cohort. All of the clinical information obtained is summarized in Figure 2.

Of these 25 patients, 13 have PIOL (no brain damage at diagnosis), 7 have oculo-cerebral lymphoma (LOC), 2 have an ocular location of a relapse of primary brain lymphoma, and 3 Ocular lymphomas secondary to systemic lymphoma (1 follicular lymphoma, 1 ethmoid lymphoma and 1 large cell B lymphoma).

Of the 12 patients with PIOL, 8 had unilateral, and 5 bilateral, ocular presentations. Three patients developed cerebral invasion within 10 months, 23 months and 24 months after the diagnosis was announced. Five patients developed an ocular relapse of their lymphoma in less than 1 year, whereas only one patient relapsed in more than 1 year. Three patients have been in complete remission for more than one year from the first line of treatment, and 5 patients are refractory to initial treatment (partial response requiring second-line treatment or no response). Of the 7 LOCs, 2 have unilateral and 4 bilateral ocular involvement. A patient has been in complete remission for more than a year from the first line of treatment, and 2 patients have developed an ocular relapse of their treated cerebral lymphoma in a time greater than one year. Two patients had an ocular relapse of cerebral lymphoma in remission for 7 years for one and 1 year for the other. The latter died 4 to 30 months after the onset of eye relapse. Finally, 3 patients have intraocular lymphoma secondary to systemic lymphoma (ethmoidal lymphoma, follicular lymphoma and large cell lymphoma B). For these 3 patients, systemic lymphomas were in remission at the time of ocular relapse. Among them, a patient has concomitant cerebral involvement. In general, it was difficult to know if the patients were still alive at the time of the study. 4 patients had died, but clinical information linking death to lymphoma was not found except for one patient. b) Univariate analysis of the results Given the small proportion of clinical information obtained to perform a significant multivariate analysis, a non-parametric Mann-Whitney comparison test was performed to see if the increase of a cytokine could be associated with a particular clinical situation. The concentrations of IL-10, IL-6, CD25s, RANTES, MIP-1a and IFNy were compared in the following groups: PIOL versus LOC: the 3 patients with secondary ocular lymphoma were excluded to perform this comparison, which was therefore performed on 13 patients with PIOL and 7 patients with LOC. unilateral ocular lesions with respect to bilateral lesions, without distinction in the diagnosis of PIOL, LOC or secondary ocular lymphoma, which represents 12 unilateral ocular lesions and 11 bilateral lesions. patients in complete remission from the first line of treatment for at least one year (7 patients) compared to patients refractory to treatment, that is to say, partially responding to treatment or requiring multiple lines of treatment (12 patients ). - patients with a relapse delay of less than 1 year (5 patients), compared to patients with a relapse delay greater than 1 year (5 patients). C) Comparison between PIOL and LOC The median concentrations of IL-10, RANTES and IFNy are significantly higher in PIOL compared to LOC. This difference is much more marked for IL-10, since the median is 1535 pg / ml in PIOL and 428 pg / ml in LOC. For RANTES and IFNy, even if the concentrations are significantly different, the values remain very low in PIOL and LOC. IL-6 and CD25s show a slight tendency to be increased in PIOL but not significantly. Finally, MIP-1a concentrations are equivalent in both groups (Table 2). D) Comparison between unilateral and bilateral lesions There are no significant differences for the different cytokines according to whether the ocular involvement is unilateral or bilateral. IL-10, IL-6 and CD25s appear to tend to be increased in unilateral forms (Table 2).

Concentration pg / ml PIOL LOC Achievement Bilateral unilateral injury IL-10 1535 428 NS 1189 862 NS IL-6 128 22 NS 202 34 NS CD25s 430 183 NS 314 188 NS RANTES 43 13 * 33 35 NS MIP-la 8 7 NS 10 7 NS IFNy 3 1 * 3 1 NS NS: Not Significant *: P <0.05 Table 2: Median Comparison of Intravitreal Concentrations in Different Groups of Patients Studied e) Comparison Between Patients in Complete Remission and Patients Refractory to Treatment There are no statistically significant differences between patients in complete remission for more than one year from the first line of treatment and patients refractory to treatment. However, concentrations of IL-10, IL-6 and CD25s tend to be higher in treatment-refractory patients. RANTES also appears to be increased but to a lesser extent since the concentrations are generally quite low (Table 3). f) Comparison between patients with a relapse delay of less than 1 year and a relapse delay greater than 1 year Again, there are no significant differences in cytokine concentrations between patients who relapse within one year. after discontinuation of treatment and those who relapse later, except for IFNy, for which the concentrations are, however, very low. There is, however, a trend of nonsignificant increase in IL-6, CD25s, RANTES and I FNy in patients relapsing early whereas patients in complete remissions for more than one year and patients who relapse late seem to be in comparable cytokine (Table 3, and Figure 3).

Concentration pg / ml Relapse Recurrence Complete remission Refractory at <1y> 1y treatment IL-10 693 532 NS 200 1134 NS IL-6 195 53 NS 20 72 NS CD25s 651 198 NS 115 281 NS RANTES 43 19 NS 18 33 NS M IP 8 8 NS 7 10 NS IFNy 7 2 * 1 3 NS NS: Not Significant *: P <0.05 Table 3: Median Comparison of Intravitreal Concentrations in Different Groups of Patients Studied g) Summary of Results Given the low In each case, it is difficult to observe significant differences between the groups we have formed. However, the order of magnitude of the cytokine concentrations must be taken into account in interpreting these results. Indeed, RANTES, MIP-1a and IFNy are generally still less than 100 μg / ml whereas IL-10, IL-6 and CD25s can show a significant dispersion of the results with values that can often reach more than 1000 pg / ml. Taking into account these elements and with all the reservation related to the limited number of patients studied, it is necessary to interpret the cytokine profiles as a whole, rather than each marker taken one by one. It is noted that IL-6, and CD255 (IFNy and RANTES to a lesser extent) vary in the same direction. They tend to be increased in patients with early onset of relapse (1 <year) compared to those with a delayed onset. In addition, these same cytokines also tend to be increased in patients responding less well to treatment, compared to those whose complete remission is obtained quickly and sustainably. If we combine these results, we also find that the median concentrations of these cytokines in patients with a late relapse delay and in patients in complete remission are close.

Thus, the increase in inflammatory cytokines at diagnosis seems to give a good indication of the response to treatment. A tumor environment with a significant inflammatory component would rather be associated with a poorer response to treatment. Thus, the cytokines RANTES, IFNγ, MIP-1α and CD25s do not appear to be more relevant than IL-10 and IL-6 for the diagnosis of intraocular lymphomas, but may be useful in Complex second-line cases, in particular CD25s, to support a complex diagnosis. On the other hand, IL-6 and CD25s tend to be increased in 10 patients who relapse within one year after stopping treatment, compared to those whose relapse delay is greater than one year. . This difference is not statistically significant given the small size studied, but the difference between the medians is significant. RANTES and IFNy are also increased but to a lesser extent. In addition, patients whose complete remission was obtained quickly and sustainably have IL-6 and CD25 concentrations comparable to those whose relapse is later. In other words, early relapsing patients would experience increased impregnation with pro-inflammatory cytokines, particularly IL-6 and CD25s. It is therefore questionable whether an increased inflammatory response at the time of diagnosis would not play a deleterious role in the host and condition a poorer long-term therapeutic response. In addition, IL-10, IL-6, CD25s and RANTES tend to be increased in PIOL compared to LOC. This tendency is accentuated if the higher points of IL-10, IL-6 and CD25s are removed from the LOC group, which correspond to the same patient whose cytokine profile is rather atypical. This increasing trend is also found, but to a lesser extent, for IL-10 and IL-6 in unilateral ocular forms compared to bilateral disease. Cytokine concentrations, therefore, tend to vary depending on the unique ocular localization or bilateral or bifocal secondary localization. These results give the impression that the cytokine impregnation is greater when the tumor is compartmentalized with the eye, or even at the level of only one eye, and that it decreases when the attack extends to another site, as if the immune reaction tended to run out or move to another compartment. To confirm this hypothesis, these cytokines should be assayed in vitreous and CSF at different times of disease progression, at diagnosis and at the time of cerebral invasion to follow the kinetics of these markers as the disease moves to another compartment.

Very interestingly, very high concentrations of CD25s were detected in vitreous patients of PIOL. In general, the origin and role of secretion of CD25s are not fully understood. Some authors attribute a tumor origin while others consider that it reflects the activation of immune cells, including Treg lymphocytes (CD4 + / CD25 + / FOXP3 regulatory T cells) which strongly express the membrane CD25. It is therefore not impossible that the intraocular concentration of CD25s indirectly reflects infiltration by Tregs, which tend to favor tumor progression by inhibiting the immune response. If this hypothesis is true, one can easily understand that high concentrations of CD25s are associated with a poor prognosis for the patient. Patients who relapsed within 1 year of discontinuing treatment tended to have higher IL-6, CD25s, RANTES and 20 IFNy levels than patients with late relapse. Without being bound by this theory, it can be hypothesised that an increase in these cytokines would reflect a significant and deleterious inflammatory response to a good therapeutic response. This inflammatory response would be accompanied by recruitment of regulatory T cells that would promote tumor progression. Levels of IL-10, IL-6 and CD25s also tend to be increased in purely ocular cases, which could be explained by a more aggressive tumor target and a concentration of immune mechanisms at the level of the tumor. oail.

Claims (19)

REVENDICATIONS1. CD25s protein detection kit comprising a set of beads emitting fluorescence under excitation at an excitation wavelength, said fluorescence emitted by said beads being at an emission wavelength different from the length of d excitation wave, each of the beads comprising, coupled to its surface, a plurality of antibodies directed against the CD25s protein. 2. Detection kit according to claim 1, characterized in that said excitation wavelength is less than 650 nm. 3. Detection kit according to claim 1 or 2, characterized in that said emission wavelength is greater than 655nm. 4. Detection kit according to one of claims 1 to 3, characterized in that said beads emit fluorescence at two emission wavelengths after excitation at the excitation wavelength. 5. Detection kit according to one of claims 1 to 4, characterized in that said emission wavelengths are detected at 665nm and 767nm. 6. Detection kit according to one of claims 1 to 5, characterized in that said beads are polystyrene microbeads. 25 7. Detection kit according to one of claims 1 to 6, characterized in that the same monoclonal antibody recognizing CD25s is present in many copies on the surface of each of the balls. 30 8. Detection kit according to one of claims 1 to 7, characterized in that it also contains secondary antibodies capable of binding to CD25s after binding of this protein to said antibodies present on the surface of said beads. 35 9. Detection kit according to one of claims 1 to 8, characterized in that the antibodies present on the surface of the beads recognize the epitope B of the CD25s protein. 10. Detection kit according to one of claims 1 to 9, characterized in that said secondary antibodies recognize the epitope A of the CD25s protein. 11. Detection kit according to one of claims 1 to 10, comprising a) polystyrene microspheres containing a mixture of cyano-7-linked allophycocyanine-allophycocyanine fluorophores, said beads comprising, coupled to their surface, a plurality of murine monoclonal antibodies directed against the B-epitope of human CD25s b) a solution of murine monoclonal antibodies directed against the human CD25s epitope A, bound to phycoerythrin c) optionally a recombinant human CD25s solution 12. Detection kit according to one of claims 1 to 11, characterized in that it also contains at least one other set of beads, said beads comprising, coupled to their surface, a plurality of antibodies recognizing another protein than the CD25s protein, said beads emitting fluorescence under excitation at said excitation wavelength, said fluorescence emitted by said beads being a) at an emission wavelength different from said wavelength of emission of the covered beads of the plurality of antibodies directed against the CD25s protein, and / or b) at an intensity different from the intensity observed for the covered beads of the plurality of antibodies directed against the CD25s protein. 13. Detection kit according to claim 12, characterized in that said other protein as the CD25s protein is selected from the group consisting of interleukin 6 (IL-6) and interleukin 10 (IL-10). 14. Method of prognosis of intraocular lymphoma in a patient, comprising the step of measuring in vitro the amount of CD25s in an eye sample in said patient, a concentration of CD25s greater than 200 μg / ml in said sample being indicative an unfavorable prognosis of progression of the disease in said patient. 15. The method according to claim 14, characterized in that the adverse prognosis of progression of the disease in said patient is the probability of having a relapse in the year following discontinuation of treatment (greater than 500 pg / ml). 16. The method according to claim 14, characterized in that the unfavorable prognosis of evolution of the disease in said patient is the probability of being refractory to the treatment. 17. The method of claim 14, characterized in that one also measures the amount of IL-6 in said sample. 18. A method of diagnosing central nervous system lymphoma of a patient comprising the step of in vitro measuring the amount of CD25s in a cerebrospinal fluid sample of said patient. 19. Method according to one of claims 14 to 18, characterized in that said step of in vitro measurement of the amount of CD25s is carried out using a kit according to one of claims 1 to 13.20.