RU2735982C2 - Method for prediction of radiosensitivity of malignant growths of upper respiratory tract - Google Patents

Abstract

FIELD: medicine.SUBSTANCE: invention refers to medicine, particularly to oncology and radiology, and discloses a method for prediction of radiosensitivity of upper respiratory cancer. Method involves staining a suspension of biopsy cells using monoclonal antibodies to CD44, CD24, CD45 and DNA-binding dye Höchst33342, using flow cytometry, tumor stem cells with immunophenotype CD44+CD24-/loware detected, and their relative amount is determined among nucleated CD45-Höchst33342+cells, then comparing the relative amount of tumor stem cells before and after irradiation in dose of 10 Gy, then by the change of this indicator, the degree of regression of the new growth is predicted under the effect of ionizing radiation.EFFECT: proposed method provides more accurate assessment of indications for radiotherapy of patients suffering VAD cancer and can be used for prediction of individual radiosensitivity of upper respiratory malignancies at the initial stages of radiation and chemoradiation therapy.1 cl, 5 ex, 5 tbl, 5 dwg

Description

The invention relates to medicine, in particular to oncology and radiology, and can be used to predict the individual radiosensitivity of malignant neoplasms (MNO) of the upper respiratory tract (URT) at the initial stages of radiation and chemoradiation therapy.

In industrially developed countries about 70% of cancer patients receive radiation therapy in the form of basic, adjuvant, neoadjuvant and palliative treatment (Andreev VG et al . Therapeutic radiology. A guide for doctors / Edited by AF Tsyba, Yu.S. Mardynsky.M .: Medkniga, 2010.552 p.). The tendency to an increasing role of ionizing radiation in the treatment of malignant neoplasms is due to the high efficiency and organ-preserving orientation of its effect on the affected organ, which makes it possible to achieve recovery against the background of good social and family rehabilitation (Ternovaya S.K. et al. Radiation diagnostics and therapy: Textbook for medical students universities in 3 volumes. M .: Medicine, 2008).

At the same time, it is known that the radiosensitivity of URT cancer (as well as of other localizations) varies greatly at the individual level with the same clinical and morphological parameters of the tumor process (anatomical region, stage, histological type, degree of differentiation of tumor cells). This fact makes the use of radiotherapy ineffective in some patients and determines the need to identify such patients even before the start or at the first stages of treatment in order to optimize it.

Among the reasons for the wide individual variability of cancer radiosensitivity, such well-known biological factors as differences in tumor oxygenation and proliferative activity of tumor cells should be indicated (Dedenkov A.N. et al. Predicting tumor response to radiation and drug therapy // M .: Medicine, 1987, 159 pp .; Diehn M. et al. Association of Reactive Oxygen Species Levels and Radioresistance in Cancer Stem Cells // Nature. 2009. V. 458. N7239. P. 780–783; Bychenkov O. A. et al. Method of determination radiosensitivity of tumors of the oral mucous membranes, RU2387472). In recent years, molecular biological indicators (gene expression profile, various mutations, the presence of papillomavirus infection), which also affect the radiosensitivity of tumors, have been intensively studied (Rieckmann T. et al. HNSCC cell lines positive for HPV and p16 possess higher cellular radiosensitivity due to an impaired DSB repair capacity // Radiotherapy and Oncology. 2013. V.107, N2. P. 242-246; Beck TN et al. Head and neck squamous cell carcinoma: Ambiguous human papillomavirus status, elevated p16, and deleted retinoblastoma 1 // Head Neck. 2017. V.39. N3: E34-E39; Foy JP et al. A 13-gene expression-based radioresistance score highlights the heterogeneity in the response to radiation therapy across HPV-negative HNSCC molecular subtypes // BMC Med. 2017 V. 15. N1: 165).

Known biomarkers and methods for predicting the clinical outcome of head and neck cancer, including URT cancer, based on tumor characteristics such as gene expression profile (Pradhan S. et al. A chip and a method for head & neck cancer prognosis, WO / 2019/220459) , expression of a number of microRNAs (Ko YH. Biomarker microRNA for prediction of prognosis of head and neck cancer, WO / 2017/073862), expression level p16^ink4aand CCND1 (Hayes D. et al. Method for head & neck cancer prognosis, WO / 2013/192089), hypermethylation of promoters of a number of genes (Guerrero-Preston R. et al. Hypermethylation biomarkers associated with poor survival outcomes for head & neck squamous cell cancer , WO / 2015/066170), mutations of the TP53 gene (Lichtarge O. Biological action of missense genotype perturbations on phenotype, WO / 2014/007863), protein expression of cytokeratin 17 (Shroyer K. et al. Keratin 17 as a biomarker for head & neck cancers, WO / 2015/175858). All of these methods make it possible to predict the overall and (or) disease-free survival of patients, but these sources do not provide information on the possibility of predicting the degree of cancer regression, which is a direct indicator of tumor radiosensitivity.

A known method for assessing the radiosensitivity of cancer of the upper respiratory tract (Zamulaeva I.A. and others RU 2505817), based on the determination of the frequency of hematopoietic stem cells with the immunophenotype CD34 ⁺ CD45 ^low in the peripheral blood of patients before treatment.

However, all of the above prediction methods do not take into account the presence of tumor stem cells (CSCs) in the cancer, which are resistant to the effects of many anticancer agents, including ionizing radiation (Zhu P. et al. Cancer stem cells and tumorigenesis // Biophysics Reports. 2018. V 4 No. 4. P. 178-188; Lytle NK et al. Stem cells fate in cancer growth, progression and therapy resistance // Nature Reviews. 2018. V.18. P. 669-680; Battle E., Clevers H. Cancer stem cells revisited // Nature Medicine. 2017. V.23.N.8.P. 1124-1134; Matchuk O.N. et al. Sensitivity of SP cells of the melanoma B16 line to the action of rare and densely ionizing radiation // Radiation Biology Radioecology 2012. Vol.52 No. 3. P. 261-267; Matchuk O.N., Zamulayeva I.A.Quantitative changes in the population of cervical cancer stem cells of the Hela line under the influence of fractionated γ-irradiation in vitro // Radiation and risk. 2019. V.28. No. 2. P.112-123). Taking into account the resistance of CSCs to antitumor therapy, it is believed that it is CSCs that provide repopulation of tumor cells during treatment, and it is these cells that retained their viability and proliferative potential during radiation and chemotherapy that can cause the development of relapses and metastases after the end of treatment.

There are various methods for detecting CSCs, which include the use of lectins of various origins (Carre V., Lacroix A. Method for isolating and detecting cancer stem cells, WO / 2018/224761), (Carre V, Lacroix A. Method for isolating and detecting cancer stem cells, WO / 2019/171010), (Lacroix A. et al. Method for detecting cancer stem cells, WO / 2017/093696), microRNA (Renaud S. et al. MiRNA as biomarkers and regulators of cancer stem cells, WO / 2017/207623), CD133-binding RNA aptamers (Duan W. CD133 aptamers for detection of cancer stem cells, WO / 2014/019024), fluorescent glucose analogs (Satake N., Nitin N. Methods of selecting and isolating cancer stem cells , WO / 2016/073737), markers CD 133, CD13, SSA, ST6 GALNAC 1, GCNT3, MGAT 5 (Miyoshi E et al. Method for isolating cancer stem cells, WO / 2012/039430), SEMA3C, LOX, GPM6A, HGF / SF, ALDH1 (Dekel B. et al. Identification of cancer stem cell markers and use of same for diagnosis and treatment, WO / 2015/198334), Lgr5 il and Lgr6 (Clevers J. et al. A Method for identifying, expanding, and removing adult stem cells and cancer stem cells, WO / 2009/022907), CD20, CD24, CD34, CD38, CD44, CD45, Cod105, CD133, CD166, EpCAM, ESA, SCA1, Pecam, Stro1 (Gupta P. et al. Methods for identification and use of agents targeting cancer stem cells, WO / 2009/126310). Among other things, antibodies to surface markers CD44 and CD24 are used to detect URT cancer stem cells, and immunoreactivity is detected using immunohistochemistry or flow cytometry. It is known that USC VAR are characterized by:

- high expression of CD44 (CD44 ⁺ ) (Okamoto A. et al. Expansion and characterization of cancer stem-like cells in squamous cell carcinoma of the head and neck // Oral Oncol. 2009. V.45. N7. P.633- 639; Prince ME et al. Identification of a subpopulation of cells with cancer stem cell properties in head and neck squamous cell carcinoma // Proc Natl Acad Sci USA. 2007. V.104. N3. P. 973-978; Kokko LL et al. Significance of site-specific prognosis of cancer stem cell marker CD44 in head and neck squamous-cell carcinoma // Oral Oncol. 2011. V. 47. N6. P. 510-516; Chen J. et al. Significance of CD44 expression in head and neck cancer: a systemic review and meta-analysis // BMC Cancer. 2014.V.14: 15; Chai L. et al. CD44 expression is predictive of poor prognosis in pharyngolaryngeal cancer: systematic review and meta-analysis // Tohoku J Exp Med. 2014. V.232. N1. P. 9-19);

- Absence or low expression of CD24 (CD24 ^{- / low} ) (Chen C. et al. Evidence for epithelial-mesenchymal transition in cancer stem cells of head and neck squamous cell carcinoma // PLoS One. 2011. V. 6. N1: e16466 ; Chen YC et al. Aldehyde dehydrogenase 1 is a putative marker for cancer stem cells in head and neck squamous cancer // Biochem Biophys Res Commun. 2009. V. 385. N3. P. 307-313; Facompre N. et al. Stem-like cells and therapy resistance in squamous cell carcinomas // Adv Pharmacol. 2012. V. 65. P.235-265; Han J. et al. Identification and characterization of cancer stem cells in human head and neck squamous cell carcinoma / / BMC Cancer. 2014. V. 14: 173; Modur V. et al. CD24 Expression May Play a Role as a Predictive Indicator and a Modulator of Cisplatin Treatment Response in Head and Neck Squamous Cellular Carcinoma // PLoS One. 2016. V 11. N6: e0156651; Okamoto A. et al. Expansion and characterization of cancer stem-like cells in squamous cell carcinoma of the head and neck // Oral Oncol. 2009. V.45. N7. P.633-639; Albers AA Stem cells in squamous head and neck cancer // Crit Rev Oncol Hematol. 2012. V. 81. N3. P. 224-240).

Therefore, OSC URT can be detected by the CD44 ⁺ CD24 ^{- / low} immunophenotype.

There are numerous data on the association of a high number of CSCs or high expression of markers associated with CSCs before treatment with low survival in patients with URT cancer, which is the basis for creating methods for predicting the aggressiveness of cancer and the effectiveness of treatment in such patients (Chen J. et al. Significance of CD44 expression in head and neck cancer: a systemic review and meta-analysis // BMC Cancer. 2014.V.14: 15; Kokko LL et al. Significance of site-specific prognosis of cancer stem cell marker CD44 in head and neck squamous- cell carcinoma // Oral Oncol. 2011. V.47. N6. P. 510-516; Chai L. et al. CD44 expression is predictive of poor prognosis in pharyngolaryngeal cancer: systematic review and meta-analysis // Tohoku J Exp Med 2014. V.232.N1.P. 9-19; van der Heijden M. et al. Biological Determinants of Chemo-Radiotherapy Response in HPV-Negative Head and Neck Cancer: A Multicentric External Validation // Front Oncol. 2020. V.9: 1470; Qian X. et al. Prognostic significance of AL DH1A1-positive cancer stem cells in patients with locally advanced, metastasized head and neck squamous cell carcinoma // J Cancer Res Clin Oncol. 2014. V. 140. N7. P.1151-1158; Szafarowski T. et al. Assessment of cancer stem cell marker expression in primary head and neck squamous cell carcinoma shows prognostic value for aldehyde dehydrogenase (ALDH1A1) // Eur J Pharmacol. 2020. V. 867: 172837; Athanassiou-Papaefthymiou M. et al. Evaluation of CD44 variant expression in oral, head and neck squamous cell carcinomas using a triple approach and its clinical significance // Int J Immunopathol Pharmacol. 2014. V. 27. N3. P. 337-349; Linge A. et al. HPV status, cancer stem cell marker expression, hypoxia gene signatures and tumor volume identify good prognosis subgroups in patients with HNSCC after primary radiochemotherapy: A multicenter retrospective study of the German Cancer Consortium Radiation Oncology Group (DKTK-ROG) // Radiother Oncol. 2016. V.121. N3. P. 364-373).

However, the methods cited above, based on the quantitative assessment of CSC in the URT malignant neoplasm before treatment, allow predicting the overall and (or) disease-free survival of patients, but not the degree of malignant neoplasm regression, which is a direct indicator of tumor radiosensitivity. In addition, these methods do not take into account the response of the CSC pool to the therapeutic effect.

The prototype of the proposed technical solution is a method for predicting the results of radiochemotherapy for locally advanced head and neck cancer, including cancer of the larynx, laryngopharynx, nasopharynx, oropharynx, parotid gland, according to the expression of the CD44 marker before treatment (Koukourakis MI et al. Cancer stem cell phenotype relates to radio-chemotherapy outcome in locally advanced squamous cell head-neck cancer // British Journal of Cancer 2012 V. 106. P. 846-853). The study group consisted of patients with inoperable tumors or with relapses of the disease who underwent platinum-containing chemotherapy and radiotherapy in the dose hypofractionation mode (2.7 Gy x 20-22 fractions for 4-5 weeks) with daily administration of the cytoprotective agent Amifostin. In biopsy material of 74 patients before treatment, the expression of a number of membrane and cytoplasmic markers of the stem state of cells (CD44, CD24, ALDH1, Oct4, etc.) was assessed using the method of immunohistochemistry. 2-4 months after the completion of radiochemotherapy, the response of the tumor (primary focus and lymph nodes) to treatment was assessed, dividing patients into 2 groups with complete or incomplete response according to computed tomography data. A high proportion of CD44 ⁺ cells before treatment and stage N + were statistically significantly associated with incomplete response (p = 0.04), with both parameters having independent prognostic significance. In addition, the expression of CD44 and a number of other biological and clinical-morphological parameters were predictive of relapse-free survival. Expression of CD24 had no prognostic value either for the immediate or long-term results of treatment.

However, the work did not simultaneously determine the expression of CD44 and CD24 on the same cells; therefore, the prognostic value of the number of cells with the CD44 ⁺ CD24 ^{- / low} immunophenotype before treatment has not been studied, and the prognostic value of the response of these cells during treatment has not been studied either. The work also does not provide information on the sensitivity and specificity of methods for predicting tumor response to treatment or disease-free survival for any of the studied parameters, including CD44 expression. It is important that all estimates of the prognostic value of CD44 ⁺ cells refer to hypofractionated irradiation of inoperable or recurrent tumors. The prognostic value of these cells using standard dose fractionation regimes (2.0 Gy per fraction) for the treatment of primary operable tumors has not been clarified.

The technical result of the claimed invention is to predict the radiosensitivity of the URT malignant neoplasm on the basis of the radiation response of the CSC population after the first sessions of radiation therapy and the formation of a group of patients with an unfavorable prognosis of treatment results by the criterion of a low degree of tumor regression.

This technical result is achieved in that, as in the known method, biopsy samples of tumor tissue are obtained before treatment and the expression of CD44 and CD24 is assessed using monoclonal antibodies.

A feature of the proposed method is that the relative number of tumor stem cells is determined after irradiation in a total focal dose of 10 Gy and compared with that before treatment, then the change in this indicator predicts the degree of regression of the neoplasm under the influence of ionizing radiation, and if:

- the relative number of tumor stem cells increases by more than 1% compared to their initial number before treatment, then a low tumor radiosensitivity is predicted;

- the relative number of tumor stem cells increases by no more than 1% inclusive, remains at the same level or decreases compared to the initial number of these cells before treatment, then a high tumor radiosensitivity is predicted.

The essence of the invention lies in the fact that a biopsy material of tumor tissue is taken from a patient with URT cancer before treatment and after the first five sessions of radiation therapy in a standard dose fractionation mode, i.e. after irradiation in a total focal dose (SOD) of 10 Gy, the relative amount of CD44 ⁺ CD24 ^{- / low} CSC is determined and its change is estimated, according to which the nature of the cancer reaction to radiation or chemoradiation therapy is predicted. So, if in the patient's biopsy material after irradiation in SOD 10 Gy there is a strong increase in the relative amount of CSCs (above 1.0%) compared to the initial level before treatment, then a low radiosensitivity of the tumor is predicted (i.e., a low degree of its regression - less than 50 % - at the end of the first stage of radiation therapy in TDS 45-50 Gy). If there is a decrease, preservation at the initial level or a slight increase (up to 1% inclusive) in the relative amount of TSC, then a high radiosensitivity of the tumor is predicted (i.e., a high degree of its regression - 50% or more).

The invention is illustrated by a detailed description, examples of execution and illustrations, which depict:

FIG. 1 - Dot graph of the distribution of cells of biopsy material of URT cancer according to the intensity of staining with antibodies to CD45, labeled with phycoerythrin conjugate with cyanine-5 (FE-Cy5), and DNA-binding dye Hoechst33342 (R1 is the region of nucleated cells that do not bind antibodies to CD45).

FIG. 2 - Dot plots of the distribution of cells taken from the R1 region (Fig. 1), according to the intensity of binding of antibodies to CD44 and CD24, labeled with fluorescein isothiocyanate (FITC) and PE, respectively. Samples of two tumors with low (A) and high (B) relative numbers of the desired CD44 ⁺ CD24 ^{- / low} cells are ^presented .

R2 - region of cells with immunophenotype CD44 ⁺ CD24 ^low , R3 - region of CD44 ⁺ CD24 ^- cells.

FIG. 3 - Distribution of biopsy samples of URT cancer before treatment by the relative number of CD44 ⁺ CD24 ^{- / low} CSC.

FIG. 4 - Distribution of biopsy samples of URT cancer after irradiation in SOD 10Gy by the relative amount of CD44 ⁺ CD24 ^{- / low} CSC.

FIG. 5 - Distribution of biopsy samples of URT cancer by the change in the relative amount of CD44 ⁺ CD24 ^{- / low} CSCs after irradiation in SOD 10Gy. The abscissa shows the difference: (relative amount of CSC after treatment,%) - (relative amount of CSC before treatment,%). Thus, positive values indicate an increase in the relative amount of CSC after irradiation, negative values indicate a decrease in this indicator.

The method is carried out in several stages.

Stage I.

Collection of samples and staining of cells of biopsy material from patients with URT cancer using monoclonal antibodies (MCAT) labeled with fluorochromes and a DNA-binding dye is performed as follows:

- after local application anesthesia with a 10% solution of lidocaine in the form of a spray, using Blexley biopsy forceps, under video-endoscopic control, the most pronounced fragments of tumor tissue with a volume of at least 1 mm ³ are taken;

- the biopsy material is placed in a vial with complete RPMI culture medium containing 10% fetal calf serum. Using mechanical grinding of biopsy material, a suspension of tumor cells is obtained, to which 0.5 ml of 0.01 M phosphate buffered saline (PBS), pH 7.2, containing 0.15 M NaCl is added, thoroughly mixed and filtered through a nylon filter with a pore diameter 40 microns;

- Determine the concentration of nucleated cells in suspension using a Goryaev chamber according to the standard method. 2 aliquots of 10 ⁵ cells are taken from each sample of the resulting suspension;

- an aliquot of cells and MCATs to CD44 labeled with FITC, to CD24 labeled with PE, and to CD45 labeled with phycoerythrin-Cy5 (Becton Dickinson - BD, USA) are added to a marked test tube at the rate of 20 μl MCAT per 1 million nucleated cells. The second aliquot of cells is used to control nonspecific binding to the MAB to hemocyanin Fissurella of the same isotype and labeled with the same fluorochromes as specific antibodies to the above surface markers. Control MCAT are also added at the rate of 20 μl per 1 million nucleated cells. Both samples are incubated at room temperature in the dark for 30 minutes;

- two-fold washing of suspension cells from unbound antibodies in 1 ml of PBS is carried out by centrifugation (300xg for 5 minutes);

- 200 μl of PBS and DNA-binding dye Hoechst 33342 at a final concentration of 6.25 μg / ml are added to a test tube with a sediment. The mixture is stirred and incubated in the dark for 15 minutes;

- the suspension of stained cells is filtered through a nylon filter with a pore diameter of 40 μm.

Stage II.

Data acquisition using flow cytometry is performed as follows:

- the sample prepared as described in stage I is analyzed on a FACS Vantage flow cytometer (BD, USA) equipped with lasers with a wavelength of 364 and 488 nm, or on other flow cytometers with the specified characteristics;

- to measure the FITC fluorescence, narrow-band filters 530/30 nm are used, for FE - 575/30 nm, for Hoechst 33342 - 424/20 nm;

- up to 100 thousand cells are analyzed in each sample. Data on the intensity of forward and side scattering, fluorescence FITZa, FE, FE-Cy5 and Hoechst 33342 save the file. The results are recorded digitally;

- the saved files are processed using the CellQuestPro program (BD, USA) or other software with the required characteristics.

Stage III.

Processing of data collected by flow cytometry in accordance with the CSC identification algorithm according to the following criteria:

- strong expression of the CD44 marker (CD44 ⁺ );

- absence or weak expression of the CD24 marker (CD24 ^{- / low} );

- the lack of expression of the general leukocyte marker CD45 (CD45 ^- ), which makes it possible to conduct negative selection of leukocytes and increase the accuracy of the quantitative analysis of CSC;

- intensive staining with Hoechstom33342 (Hoechst ⁺ ), which makes it possible to differentiate nucleated cells from debris, erythrocytes, platelet conglomerates, and other nonspecific events, which ultimately ensures high accuracy of quantitative analysis of CSCs.

In particular, analysis of flow cytometric data in order to identify and determine the relative amount of CSCs in a biopsy includes:

- construction of a dotted graph of the distribution of cells according to the fluorescence intensity of MCAT to CD45 and the fluorescence intensity of Hoechst 33342 (Fig. 1). The graph shows the region of nucleated cells (R1), characterized by the absence of fluorescence of the CD45 marker;

- construction of a dotted graph of the distribution of cells according to the fluorescence intensity of MCAT to CD44 and CD24 (Fig. 2). The graph is built only for cells that meet the conditions R1. Regions of cells (R2 and R3) with immunophenotypes CD44 ⁺ CD24 ^- and CD44 ⁺ CD24 ^low , respectively, are distinguished on the graph, taking into account the fluorescence of cells in the control of nonspecific binding. Next, the number of cells in the regions R2 and R3 is determined;

- calculation of the relative number of CSCs, expressed as a percentage, by dividing the total number of cells in R2 and R3 by the number of cells in R1.

Thus, the relative number of CD44 ⁺ CD24 ^{- / low} CSCs among CD45 ^- Hoechst ⁺ nucleated cells is determined.

Stage IV.

Prediction of tumor radiosensitivity is as follows:

- compare the relative amount of CSC, calculated at stage III before treatment and after irradiation in SOD 10 Gy;

- if the relative amount of CSCs in the patient's biopsy after irradiation in SOD

10 Gy is more than 1% higher than that before treatment (i.e., under the influence of radiation exposure, a rather strong increase in the relative amount of TSC occurs), low tumor radiosensitivity is predicted, since a strong increase in the number of TSC occurs in 91% of patients with low tumor regression (<50%) after the completion of the first stage of radiation therapy in SD 45-50Gy;

- if the relative amount of CSCs in the patient's biopsy after irradiation in SOD

10Gy increases relatively weakly (by 1% or less), remains unchanged or decreases in comparison with that before treatment, the probability of high tumor regression (≥50%) after the completion of the first stage of radiation therapy in SD 45-50Gy is 92%, therefore it is predicted high radiosensitivity of the tumor.

The method is illustrated by the following examples of execution.

Example No. 1 - The relative amount of CSCs in the biopsy material of URT cancer before treatment in the general group of patients, as well as in subgroups of patients with different clinical and morphological parameters of the tumor process and the degree of tumor regression.

The relative amount of CSCs was determined in 123 patients with URT cancer before treatment . The patients were admitted to the Department of Radiation and Surgical Treatment of Upper Airway Diseases (Department of Radiation and Surgical Treatment of Head and Neck Diseases of the A.F. The average age (± SE) of the patients was 57.2 ± 0.8 years. Squamous cell carcinoma was histologically verified in all patients, while non-keratinizing cancer was diagnosed in 36% of patients, keratinizing in 43%, and other squamous cell carcinoma histotypes in 18%. Stage T1 was observed in 7, T2 in 41, T3 in 51 and T4 in 24 patients. Regional lymph nodes were involved in the tumor process in 63 patients (N +), in the rest of the patients there were no signs of lymph node involvement (N0). All patients, except one, were in the M0 stage.

In all patients, treatment was started with simultaneous chemoradiation therapy (PF regimen). In the polychemotherapy scheme, two drugs were used - cisplatin and 5-fluorouracil. On the first day of treatment, before the start of radiation therapy, against the background of overhydration, cisplatin was injected intravenously at the rate of 100 mg / m2 ^of body surface area. Then the patient started intravenous administration of 5-fluorouracil at a dose of 3000 mg continuously for 72 hours using an infusion syringe pump. Radiation therapy was started simultaneously with the continuous administration of 5-fluorouracil. The chemotherapy cycle was repeated on the 22nd day of treatment. Patients with incomplete regressions of regional metastases underwent planned excision of lymph nodes 4-6 weeks after completion of chemoradiation therapy. In the case of a residual tumor or local recurrence, the patients underwent surgery, the volume of which was determined by the prevalence of the neoplasm.

The irradiation was carried out using gamma-therapeutic devices. To carry out external beam radiation therapy to the area of the primary tumor, the technique of irradiation of 2 Gy five times a week was applied.

After irradiation in SOD 45-50 Gy, the degree of tumor regression was determined. If the degree of tumor regression was low (less than 50%), the radiosensitivity of the tumor was regarded as low, radiation therapy was interrupted and surgical treatment was performed. With pronounced tumor regression (more than 50%), radiation therapy was continued up to SOD 60-62 Gy. With the development of radiation epithelitis, a forced break was made for 10-12 days. In total, 2 cycles of polychemotherapy were carried out during the course of treatment.

FIG. 3 shows the distribution of the relative amount of CSCs in the general group of patients before treatment. This indicator varied widely from 0.1 to 26.8%. The data on the relative number of CSCs did not obey the normal distribution (according to the Shapiro-Wilk test), therefore, for the descriptive statistics of this indicator, the values of the median and the range of quartiles (Q1-Q3) were used, and the comparison of this indicator in subgroups of patients was carried out using the nonparametric Mann- Whitney. Differences were considered statistically significant at p <0.05.

Table 1 shows the relative number of CSCs in the subgroups of patients with different clinical and morphological parameters of the tumor process. No significant differences were found in the relative number of CD44 ⁺ CD24 ^{- / low} cells in patients before treatment at different stages of the disease, histological types of tumors, lesions of regional lymph nodes (Table 1). The relative number of CSCs before treatment did not statistically significantly differ with different degrees of regression (radiosensitivity) of tumors, constituting a median of 4.1% in cases of a high degree of tumor regression (≥50%) and 2.6% in cases of low regression (p = 0, 07).

Table 1.

Comparison of the relative number of CD44 ⁺ CD24 ^{- / low} cells in biopsies of URT patients before treatment in subgroups with different clinical and morphological parameters of the tumor process

Indicators Number of patients Relative amount
CD44 ⁺ CD24 ^{- / low} cells,%
Median (Q1 ÷ Q3) p Stage I + II 48 3.2 (1.5 ÷ 7.0) 0.73 III + IV 75 3.5 (1.4 ÷ 6.6) Regional lymph node status N0 60 3.2 (1.6 ÷ 6.6) 0.61 N + 63 3.7 (1.3 ÷ 7.2) Histological type of squamous cell carcinoma Keratinizing 52 3.8 (1.4 ÷ 7.0) 0.08 Non-keratinizing 44 2.4 (1.2 ÷ 7.0) Other types 22 4.2 (2.8 ÷ 7.5)

Example No. 2 - The relative amount of CSCs in the biopsy material of URT cancer after irradiation in SOD 10Gy against the background of chemotherapy in the general group of patients, as well as in subgroups of patients with different clinical and morphological parameters of the tumor process and the degree of tumor regression.

The relative amount of CSC was determined in 39 patients after irradiation in SOD 10 Gy . Patients were admitted to the same department for radiation and surgical treatment of upper respiratory tract diseases (as in example 1) in 2010-2014.

The average age (± SE) of the patients was 59.4 ± 1.1% years. Squamous cell carcinoma was histologically verified in all patients, while non-keratinizing cancer was diagnosed in 28.2% of patients, and keratinizing in 35.9%. Stage T1 was observed in 5 patients, T2 in 13, T3 in 10, T4 in 11 patients. Regional lymph nodes were involved in the tumor process in 22 patients, the remaining patients showed no signs of lymph node involvement. All patients were in the M0 stage.

Treatment of patients and assessment of tumor radiosensitivity according to the degree of its regression in response to treatment was performed similarly to example 1.

The distribution of the relative amount of CSCs in the general group of patients after irradiation in SOD 10 Gy is illustrated in Fig. 4. This indicator varied widely from 0.3 to 37%. As before treatment, the data on the relative number of CSCs after irradiation did not obey the normal distribution (according to the Shapiro-Wilk test), therefore, for descriptive statistics of this indicator, the values of the median and the range of quartiles (Q1-Q3) were also used, and the comparison of this indicator in subgroups of patients was carried out according to the nonparametric Mann-Whitney test.

No significant differences were found in the post-radiation number of CSCs at different stages of the disease, histological types of tumors, and lesions of regional lymph nodes (Table 2). It is important that the relative number of CSCs after irradiation did not differ with different degrees of regression (radiosensitivity) of tumors: the median of this indicator was 5.4% in cases of high degree (≥50%) tumor regression and 5.0% in cases of low (<50% ) regression (p = 0.46).

Table 2.

Comparison of the relative number of CD44 ⁺ CD24 ^{- / low} cells in biopsies of patients with URT after irradiation in SOD 10 Gy in subgroups with different

clinical and morphological indicators of the tumor process

Indicators Number of patients Relative number of CD44 ⁺ CD24 ^{- / low} cells after irradiation at a dose of 10 Gy,%
Median (Q1 ÷ Q3) p Stage I + II 18 6.8 (3.6 ÷ 10.0) 0.23 III + IV 21 4.4 (1.8 ÷ 8.1) Regional lymph node status N0 22 5.2 (3.7 ÷ 10.1) 0.61 N + 17 5.7 (1.8 ÷ 8.8) Histological type of squamous cell carcinoma Keratinizing fourteen 6.1 (3.7 ÷ 8.9) 0.40 Non-keratinizing eleven 5.7 (1.8 ÷ 7.1) Other types fourteen 6.7 (3.6 ÷ 11.4)

Example No. 3 - Changes in the relative amount of CSCs in the biopsy material of URT cancer after irradiation in SOD 10Gy in the general group of patients, as well as in subgroups of patients with different clinical and morphological indicators of the tumor process and the degree of tumor regression.

The relative amount of CSCs before treatment and after irradiation with SOD 10 Gy was compared in a group of 39 patients, described in detail in Example 2. Treatment of patients and assessment of tumor radiosensitivity by the degree of its regression in response to treatment was performed in the same way as in Example 1.

The change in the relative amount of CSCs in the general group of patients after irradiation in SOD 10Gy is illustrated in Fig. 5. Data on post-radiation changes in the relative number of CSCs were characterized by a normal distribution (according to the Shapiro-Wilk test), therefore, for descriptive statistics of this indicator, the mean values and standard errors (± SE) were used, and the comparison of changes in the CSC pool in subgroups of patients was carried out according to the Student's test. There was a high individual variability in changes in the relative amount of CSCs in the biopsy material of patients after irradiation in SOD 10Gy: from a decrease by 23.7% to an increase by 36.1%, on average this indicator increased by 2.7 ± 1.3%. In 76.9% of patients (30/39), there was an increase in the relative amount of CSCs after this exposure as compared with the initial level. Moreover, in 66.7% of patients (26/39), the relative number of CSCs increased by more than 1.0%. The magnitude of changes in this indicator after irradiation in SOD 10 Gy did not correlate with clinical and morphological indicators (Table 3).

When comparing changes in the CSC pool in subgroups of patients with different degrees of regression (radiosensitivity) of the tumor, it is shown:

- in a subgroup of 28 patients with a high degree of tumor regression (more than 50%), the mean

the amount of CSC did not change statistically significantly after irradiation, amounting to 5.0 ± 1.1% before treatment and 6.2 ± 0.8% after exposure to radiation at a dose of 10 Gy (p = 0.37);

- in the subgroup of 11 patients with low tumor regression, the average number of CSCs

It increased statistically significantly after irradiation by 6.6 ± 3.4%, being 3.8 ± 1.5% before treatment and 10.4 ± 3.3% after it (p = 0.007).

Thus, as a result of the conducted prospective study, a different response of the CSC pool to the first sessions of irradiation in SOD 10Gy in tumors with a high and low degree of regression, assessed after irradiation in SOD 45-50Gy, was found. With a high degree of tumor regression, the relative amount of CSCs did not change statistically significantly on average, indicating approximately the same radiosensitivity of tumor stem and non-stem cells. At the same time, in tumors with a low radiosensitivity, CSCs showed a higher radioresistance compared to other cells, since the relative amount of CSCs increased on average.

Table 3.

Change in the relative number of CD44 ⁺ CD24 ^{- / low} cells in biopsies of patients with URT cancer after irradiation in SOD 10 Gy in subgroups with different clinical and morphological parameters of the tumor process

Indicators Number of patients Change in the relative number of CD44 ⁺ CD24 ^{- / low} cells after irradiation at a dose of 10 Gy,%
Mean ± SE p Stage I + II 18 1.6 ± 1.9 0.45 III + IV 21 3.7 ± 1.9 Regional lymph node status N0 22 2.1 ± 1.6 0.58 N + 17 3.6 ± 2.3 Histological type of squamous cell carcinoma Keratinizing fourteen 0.2 ± 2.3 0.15 *
0.29 ** Non-keratinizing eleven 6.1 ± 3.4 Other types fourteen 2.6 ± 1.1

* When comparing subgroups of patients with keratinizing and non-keratinizing cancer.

** When comparing subgroups of patients with keratinizing cancer and other histological types.

Example 4 - ROC-analysis of the efficiency (specificity and sensitivity) of the method for predicting the radiosensitivity of the VAR ZNO according to the criterion of the change in the relative amount of CSC after irradiation in SOD 10 Gy.

In the group of 39 patients described in examples 2 and 3, the ROC analysis method established the optimal discriminator (for the indicator "Change in the relative amount of CSC after irradiation in SOD 10 Gy"), dividing patients into groups with a relatively high and low probability of an unfavorable treatment outcome (low radiosensitivity of the tumor). It turned out that with a decrease, maintenance at the same level or a slight increase (up to 1% inclusive) of the initial amount of CSCs, low tumor radiosensitivity is observed only in 1/13 (7.7%) patients, while with a strong increase in this indicator by more than by 1% - in 10/26 (38.5%) patients, i.e. 5 times more often (p = 0.046 according to Fisher's criterion) (table 4).

The sensitivity of this method for predicting low tumor radiosensitivity is 0.91 (95% confidence interval 0.62-0.98), specificity - 0.43 (95% confidence interval 0.28-0.58), positive predictive value - 0, 38, negative predictive value - 0.92. The AUC (area under curve) value, which characterizes the predictive efficiency of the indicator, is 0.67, which makes it possible to regard the change in the number of CSCs after irradiation in SOD 10 Gy as an average classifier in terms of power.

Table 4.

Comparison of the nature of changes in the relative amount of CSCs after irradiation in SOD 10 Gy (compared with the initial value before treatment) in patients with URT cancer with high and low tumor radiosensitivity, determined by the degree of its regression (the table shows the number of patients)

Change in the number of CSCs after irradiation in SOD 10 Gy Tumor radiosensitivity Low (degree of regression <50%) High (degree of regression ≥50%) Strong increase (over 1.0%) ten sixteen Reduction, maintenance at the original level or slight increase (up to 1.0% inclusive) 1 12

Example 5 - Multiple regression analysis of the radiosensitivity of ZNO URT using the program Statistics 6.0 (Stat Soft, Inc., USA).

In the group of 39 patients described in examples 2-4, multiple regression analysis of the dependence of tumor radiosensitivity (degree of regression of the primary tumor focus) on a number of possible predictors, including the patient's age, stage of the disease T1-4, lesion of regional lymph nodes (N + / N0) was performed , histological type (keratinizing / non-keratinizing / other types of squamous cell carcinoma), as well as the relative amount of CSCs before treatment, the relative amount of CSCs after irradiation in SOD 10Gy, change in the relative amount of CSCs after irradiation in SOD 10Gy (strong increase / weak increase, preservation at the initial level or decrease in accordance with table 4). Three indicators that demonstrated the strongest relationship with tumor regression were chosen to construct a multiple regression model: change in the relative amount of CSCs after irradiation in SOD 10Gy, stage T, and histological type of tumor. As shown in Table 5, the model as a whole was characterized by statistical significance at p = 0.03, and only one indicator (change in the relative amount of CSCs after irradiation in SOD 10 Gy) had an independent prognostic value for tumor radiosensitivity.

Table 5.

Results of multiple regression analysis of the dependence of tumor radiosensitivity on changes in the relative amount of CSCs after irradiation in SOD 10Gy, the stage of the disease and the histological type of tumor

Indicators (predictors) Beta The p-value for the predictor R P value for the entire model as a whole Change in the relative amount of CCA after irradiation in SOD 10Gy -0.362 0.02 0.47 0.03 Stage T -0.254 0.10 Histological type of tumor 0.187 0.22

Note: Beta is the standardized regression slope (in SD units), R is the multiple correlation coefficient.

Proof of achievement of the technical result.

As examples 1-3 show, there is a high individual variability in the relative amount of CD44 ⁺ CD24 ^{- / low} CSC before treatment, after irradiation in SOD 10 Gy and, as a consequence, changes in this indicator under the influence of the first sessions of radiation therapy, which is the basis of this method. predicting the radiosensitivity of tumors. Moreover, all 3 indicators are not associated with the clinical and morphological characteristics of the tumor process.

Example 4 proves that with an increase in the relative amount of CSCs by more than 1% after irradiation at a dose of 10 Gy, a low radiosensitivity of the tumor (which is recorded by the low degree of its regression after the completion of the first stage of chemoradiation therapy) is 5 times more likely than with a decrease in the amount of CSCs. , maintaining at the initial level or weakly increasing (up to 1% inclusive) (p = 0.046). The integral indicator (AUC) of sensitivity and specificity of this method for predicting tumor radiosensitivity is 0.67.

Example 5 confirms the predictive value of the indicator "Change in the relative amount of CSC after irradiation in SOD 10 Gy" using multiple regression analysis. Moreover, it is this indicator that has an independent and greater prognostic value than the stage of the disease and the histological type of tumor.

The proposed method increases the accuracy of assessing the indications for radiation therapy in patients with URT cancer and, thus, improves treatment planning. In addition, the identification of patients with cancer who are resistant to standard types of radiation exposure contributes to the development of an optimal strategy for their treatment, including the use of special modes of fractionation of the dose of rare ionizing radiation, the use of hadron therapy, combined methods of treatment, etc.

Claims (3)

A method for predicting radiosensitivity of upper respiratory tract cancer, including obtaining biopsy samples of tumor tissue before treatment and assessing the expression of CD44 and CD24 using monoclonal antibodies, characterized in that the biopsy material before treatment and after irradiation in a total focal dose of 10 Gy is disaggregated, the resulting cell suspension is stained using monoclonal antibodies to CD44, CD24, CD45 and a DNA-binding dye Hoechst33342, using flow cytometry, tumor stem cells with the immunophenotype CD44 ⁺ CD24 ^{- / low are} detected and their relative number among nucleated CD45 ^- Hoechst33342 ⁺ cells is compared, then the relative number is compared tumor stem cells before treatment and after irradiation at a dose of 10 Gy, then the change in this indicator predicts the degree of regression of the neoplasm under the influence of ionizing radiation, and if: - the relative number of tumor stem cells increases by more than 1% compared to their initial number before treatment, then a low tumor radiosensitivity is predicted; - the relative number of tumor stem cells increases by no more than 1% inclusive, remains at the same level or decreases compared to the initial number of these cells before treatment, then a high tumor radiosensitivity is predicted.

Images