ASSESSMENT OF METASTATIC TRAITS OF THE CELLS WITH HYBRID PHENOTYPE IN BREAST CANCER

Mukhamedzhanov R.K.; Grigoryeva E.S.; Tashyreva L.A.; Perelmuter V.M.; Zavyalova M.V.; Savelieva O.E.

Mukhamedzhanov RK14 Grigoryeva ES2, Tashireva LA2, Perelmuter VM2, Zavyalova MV12 Savelieva OE2-3 и

1 Siberian State Medical University, Tomsk, Russia

2 Cancer Research Institute, Tomsk National Research Medical Center, Russian Academy of Sciences, Tomsk, Russia

3 Saint-Petersburg State Pediatric Medical University, Saint-Petersburg, Russia

4 Regional Children's Clinical Hospital, Vladimir, Russia

Nowadays, great attention is paid to the study of circulating tumor cells (CTCs) due to their key role in distant metastasis. At the same time there is little data on the properties of circulating cells showing simultaneous expression of the leukocyte and epithelial markers and their possible role in tumor progression and chemotherapy resistance. The study was aimed to assess subpopulations of cells with hybrid epithelial/leukocyte phenotype and estimate the features of stemness, epithelial-mesenchymal transition (EMT), and integrin interface, which determine the cells' possible metastatic properties in breast cancer (BC). The survey data from 128 patients with invasive breast carcinoma of no special type were included. Multicolor flow cytometry was used to assess the population structure and metastatic potential of the cells circulating in blood and primary tumor cells with hybrid phenotype. The primary tumor cell suspension was prepared by mechanical disaggregation. The high degree of heterogeneity was noted in the population of cells with hybrid phenotype, including the combination of the stemness and EMT features, and diverse integrin interface. Cells with hybrid phenotype are involved in the mechanisms underlying lymph node and distant metastasis. In lymph node metastasis, metastatic potential of these cells is associated with the stemness features (р = 0.0422) and co-expression of p3-, p4-, and aVp5-integrins (р = 0.0338). In distant metastasis, metastatic potential of hybrid cells is associated with the stemness features (р = 0.015) and is not associated with the EMT features and integrin expression.

Keywords: hybrid cells, circulating tumor cells, metastasis, stemness, epithelial-mesenchymal-transition, integrins Funding: the study was supported by the RSF grant № 21-15-00140.

Author contribution: Mukhamedzhanov RK — literature analysis, data acquisition and statistical processing, manuscript writing; Grigoryeva ES — data acquisition and analysis, manuscript writing; Tashireva LA — study design, manuscript editing; Perelmuter VM — data interpretation, manuscript editing; Zavyalova MV — study planning, manuscript editing; Savelieva OE — study planning and management, data analysis and interpretation, manuscript writing.

Compliance with the ethical standards: the study was approved by the Ethics Committee of the Cancer Research Institute, Tomsk National Research Medical Center (protocol № 10 of 24 April 2015) and conducted in accordance with the Federal Laws of the Russian Federation (№ 152, 323, etc.), the Declaration of Helsinki (1964) and all subsequent amendments and supplements regulating the research involving human biomaterials. The informed consent to study participation was submitted by all subjects.

[23 Correspondence should be addressed: Olga E. Savelieva Litovskaya, 2, Saint-Petersburg, 194100; olga_chechina@mail.ru

Received: 04.07.2023 Accepted: 22.08.2023 Published online: 31.08.2023

DOI: 10.24075/brsmu.2023.033

ОЦЕНКА МЕТАСТАТИЧЕСКИХ ХАРАКТЕРИСТИК КЛЕТОК С ГИБРИДНЫМ ФЕНОТИПОМ ПРИ РАКЕ МОЛОЧНОЙ ЖЕЛЕЗЫ

Р. Х. Мухамеджанов1,4, Е. С. Григорьева2, Л. А. Таширева2, В. М. Перельмутер2, М. В. Завьялова1'2, О. Е. Савельева2,3^

1 Сибирский государственный медицинский университет, Томск, Россия

2 Научно-исследовательский институт онкологии, Томский национальный исследовательский медицинский центр Российской академии наук, Томск, Россия

3 Санкт-Петербургский государственный педиатрический медицинский университет, Санкт-Петербург, Россия

4 Областная детская клиническая больница, Владимир, Россия

Изучению циркулирующих опухолевых клеток (ЦОК) в последнее время уделяют большое внимание, благодаря их ведущей роли в формировании отдаленных метастазов. В то же время мало данных о свойствах циркулирующих клеток с одновременной экспрессией лейкоцитарных и эпителиальных маркеров и их возможной роли в опухолевой прогрессии и резистентности к химиотерапии. Целью исследования было изучить субпопуляции клеток с гибридным эпителиально-лейкоцитарным фенотипом, а также оценить признаки стволовости, эпителиально-мезенхимальный переход (ЭМП) и интегриновый интерфейс, обусловливающие их возможные метастатические свойства при раке молочной железы (РМЖ). В работу включены данные исследования 128 больных инвазивной карциномой неспецифического типа молочной железы. Для оценки популяционного состава и метастатического потенциала циркулирующих в крови клеток и клеток первичной опухоли с гибридным фенотипом использовали метод многоцветной проточной цитометрии. Суспензию клеток первичной опухоли готовили методом механической дезагрегации. В популяции клеток с гибридным фенотипом отмечена высокая степень гетерогенности, включая комбинацию признаков стволовости, ЭМП и разнообразный интегриновый интерфейс. Клетки с гибридным фенотипом принимают участие в механизмах лимфогенного и гематогенного метастазирования. При лимфогенном метастазировании метастатический потенциал этих клеток ассоциирован с признаками стволовости (р = 0,0422) и коэкспрессией р3-, р4- и aVp5-интегринов (р = 0,0338). При гематогенном метастазировании метастатический потенциал гибридных клеток ассоциирован с признаками стволовости (р = 0,015) и не связан с признаками ЭМП и экспрессией интегринов.

Ключевые слова: гибридные клетки, циркулирующие опухолевые клетки, метастазы, стволовость, эпителиально-мезенхимальный переход, интегрины Финансирование: работа выполнена при финансовой поддержке гранта РНФ 21-15-00140.

Вклад авторов: Р. Х. Мухамеджанов — анализ литературы, получение и статистическая обработка результатов, написание статьи; Е. С. Григорьева — получение и анализ данных, написание статьи; Л. А. Таширева — дизайн исследования, редактирование статьи; В. М. Перельмутер — интерпретация результатов, редактирование статьи; М. В. Завьялова — планирование исследования, редактирование статьи; О. Е. Савельева — планирование и руководство исследованием, анализ и интерпретация результатов, написание статьи.

Соблюдение этических стандартов: исследование одобрено этическим комитетом НИИ онкологии Томского НИМЦ (протокол № 10 от 24 апреля 2015 г.), проведено в соответствии с федеральными законами Российской Федерации (№ 152, 323 и др.) и Хельсинкской декларацией 1964 г. и всеми последующими дополнениями и изменениями, регламентирующими научные исследования на биоматериале, полученном у людей. Все участники подписали информированное добровольное согласие об участии в исследовании.

[>3 Для корреспонденции: Ольга Евгеньевна Савельева

ул. Литовская, д. 2, г. Санкт-Петербург, 194100; olga_chechina@mail.ru

Статья получена: 04.07.2023 Статья принята к печати: 22.08.2023 Опубликована онлайн: 31.08.2023 DOI: 10.24075/vrgmu.2023.033

Nowadays, great attention is paid to the study of circulating tumor cells (CTCs). This is due to their key role in distant metastasis and, therefore, in adverse outcomes of cancer. Today, the data are available on their subpopulation composition

[I], stem-like properties [2], epithelial-mesenchymal transition (EMT) [3], chemotherapy resistance [4]; the genomic profiling data of CTCs have been also published [5]. Furthermore, the importance of integrin expression in CTCs for metastasis in breast cancer (BC) has been shown; the CTC integrin interface can be associated with the location of distant metastases [6-7]. The data on the correlation of peripheral blood levels of these cells with the survival rate and the risk of distant metastasis in BC [8], lung cancer [9], ovarian cancer [10], colorectal cancer

[II], etc. are provided. Thus, in 2019 it was shown that the presence of CTCs showing stemness and partial EMT features was associated with adverse disease outcomes and reduced overall survival rate. Cells with the same phenotype turned out to be resistant to chemotherapy [12].

When studying various CTC populations, we have found unusual cells showing simultaneous expression of the CD45 leukocyte and CD326 (EpCAM) epithelial markers. It turned out that other researchers obtained similar results. Thus, cells with the CD45+CK+EpCAM+ phenotype were found in blood of patients with BC, and in 90% of cases expression of the CD68 macrophage marker was also noted [13]. Cells with the CD45+EpCAM+ phenotype were found in primary tumors and pleural effusion of all surveyed patients with non-small cell lung cancer. Moreover, the higher percentage of such cells was associated with adverse outcome [14].

Hybridization (cell fusion) between tumor cells and macrophages or leukocytes is the most probable mechanism underlying generation of cells showing simultaneous expression of the leukocyte and epithelial markers. It has been noted that formation of hybrid cells is associated with many body's physiological processes, such as muscle and bone tissue formation, wound healing [15]. The CD45+EpCAM+ cells' physiological role is confirmed by detection of these cells in blood of healthy donors. However, biological value of these cells together with their role in physiological and disease processes is poorly understood.

Due to the lack of knowledge about the cells showing simultaneous expression of the leukocyte and epithelial markers and the data on their possible role in tumor progression and chemotherapy resistance [16], our study was aimed to assess subpopulations of cells with hybrid epithelial/ leukocyte phenotype and estimate the features of stemness, EMT, and integrin interface, which determine the cells' possible metastatic properties in breast cancer (BC).

METHODS

Patients

Survey results of 128 patients treated in the clinics of the Cancer Research Institute, Tomsk National Research Medical Center, in 2015-2020 were included in the study. Inclusion criteria: invasive breast carcinoma of no special type; age 29-76 years (average age: 52.56 ± 11.57; T1-4N0-3M0-1). Exclusion criteria: breast cancer of other histological type; multiple primary malignant tumors; exacerbation of chronic inflammatory disorder. To assess the population structure and metastatic traits of the circulating cells with hybrid phenotype, venous blood was collected from 108 patients before neoadjuvant chemotherapy. The association between cells with hybrid phenotype and lymph node or distant metastasis was estimated in the group of

patients without neoadjuvant chemotherapy (n = 79). To assess properties of the cell populations with the primary tumor hybrid phenotype, surgical material obtained from 35 patients during surgical treatment (radical mastectomy or sectoral resection) was studied. These patients were not prescribed neoadjuvant therapy. The basic clinical and morphological parameters are provided in Table. 1.

Sample preparation for flow cytometry

The patients' venous blood was collected before the course of neoadjuvant chemotherapy and surgical treatment in the morning in the fasting state: 12 mL in the EDTA vacuum tubes. The entire volume of blood was used to prepare cell concentrate by sedimentation at 37 °C for 90 min at an angle of 45° with subsequent collection of buffy coat with cells on the boundary between the erythrocyte sediment and the separated blood plasma, as well as the entire supernatant, in accordance with the method by R.A. Pospelova [17].

Fresh cancer tissue samples were mechanically disaggregated using the BD Medimachine System (BD; USA) for cell suspensions. The total cell count of the resulting suspensions was determined using the Luna-II Cell Counter system (Logos Biosystems; Korea).

Flow cytometry of samples and data processing

After Fc blocking with the Human TruStain FcX™ Fc Receptor Blocking Solution (Biolegend; USA), 5 |jL of the BV570 antihuman CD45 (clone HI30; Sony Biotechnology, USA), BV650 anti-human CD326 (EpCAM) (clone 9C4; Sony Biotechnology, USA), BV510 anti-human CD44 (clone G44-26; BD Horizon, USA), PerCP/Cy5.5 anti-human CD24 (clone ML5; Sony Biotechnology, USA), PE/Cy7 anti-human N-Cadherin (clone 8C11; Sony Biotechnology, USA) monoclonal antibodies and 7-AAD Viability Staining Solution (Sony Biotechnology; USA) were added to the primary tumor cell concentrate and/or cell suspension and incubated in the dark at room temperature for 20 min. Cell concentrate was also supplemented with 5 jL of the BV 421-anti-p3 integrin (clone VI-PL2; BD Biosciences, USA), Alexa Fluor 488-anti-£4 integrin (clone 422325; R&D Systems, USA), BV Alexa Fluor 750-anti-aV^5 integrin (clone P5H9; R&D Systems, USA) monoclonal antibodies. The unstained control was processed in parallel. After incubation, 500 |L of the OptiLyse C buffer (Beckman Coulter; France) were added to the samples for erythrocyte lysis, then the samples were washed in 2 mL of the CellWASH solution (BD Biosciences; USA) for 10 min at 300 g with subsequent removal of supernatant. During the intracellular staining phase each stained sample was supplemented with 250 jL of the BD Cytofix/Cytoperm solution (BD Biosciences; USA), incubated in the dark for 30 min at 4 °C, and then twice washed in 1 mL of the BD Perm/Wash buffer (BD Biosciences; USA) when centrifuged at 300 g for 6 min. A total of 50 jL of the BD Perm/Wash buffer (BD Biosciences; USA) were added to the samples, the stained sample was supplemented with 5 jL of the AF647-anti-human CK7/8 antibody (clone CAM5.2; BD Pharmingen, USA) and incubated at 4 °C for 20 min. After that each sample was washed in 1 mL of the CellWASH buffer (BD Biosciences; USA) by centrifugation at 300 g for 6 min. In the final phase 500 jL of the Cell Staining Buffer (Sony Biotechnology; USA) were added to precipitate, and the sample was resuspended.

Nonspecific staining was controlled using appropriate isotype antibodies. The BC MCF-7 cell line was used as a positive control for antibodies against epithelial markers EpCAM

Table 1. Characteristics of patients with invasive breast carcinoma of no special type

Parameter Parameter value Frequency, % (abs.)

Age 52.56 ± 11.57

Molecular subtype Luminal А 32.8% (42/128)

Luminal В 47.7% (61/128)

Triple negative 15.6% (20/128)

HER2neu+ 3.9% (5/128)

Stage I 25.0% (32/128)

48.4% (62/128)

IB 17.2% (22/128)

IIW 1.6% (2/128)

IIB 4.7% (6/128)

IIIG 3.1% (4/128)

Lymph node metastasis Yes 34.4% (44/128)

No 65.6% (84/128)

Distant metastasis Yes 9.4% (12/128)

No 90.6% (116/128)

Estrogen receptors Positive 80.5% (103/128)

Negative 19.5% (25/128)

Progesterone receptors Positive 71.9% (92/128)

Negative 28.1% (36/128)

HER2 receptors Positive 51.6% (66/128)

Negative 48.4% (62/128)

Ki67 expression < 20% 35.9% (46/128)

>20% 64.1% (82/128)

Tumor size < 2 cm 28.1% (36/128)

2-5 cm 67.2% (86/128)

> 5 cm 4.7% (6/128)

Menstrual status Postmenopausal 41.4% (53/128)

Premenopausal 54.7% (70/128)

Perimenopausal 3.9% (5/128)

Neoadjuvant chemotherapy Yes 38.3% (49/128)

No 61.7% (79/128)

Tumor grade 1 17.2% (22/128)

2 63.3% (81/128)

3 19.5% (25/128)

Recurrence Yes 5.5% (7/128)

No 94.5% (121/128)

and CK7/8 and a negative control for CD45. The histiocytic lymphoma U937 cell line was used as a negative control for antibodies against the above epithelial markers and a positive control for CD45.

The samples were analyzed in the Novocyte 3000 flow cytometer (ACEA Biosciences; USA) using NovoExpress 1.3.0 (ACEA Biosciences; USA). The concentration of circulating cells per 1 mL of blood and the concentration of primary tumor cells per 1000 tumor cells were calculated. When gating cells with hybrid phenotype, the cells were first analyzed in the FSC vs. SSC mode, then singlets were isolated in the FSC-A vs. FSC-H mode. After that viable cells were isolated based on the 7-AAD negative stain, and then cell fluorescence parameters were analyzed in appropriate channels. The gating strategy for cells with hybrid phenotype is provided in Fig. 1.

Statistical analysis

Statistical processing of the results was performed using the IBM SPSS Statistics 22 (Armonk; USA) and GraphPad Prism 8.3.1 (GraphPad Software; USA) software packages. All the studied parameters were tested for normality using the Shapiro-Wilk test. Parameters were described using median (Me) and interquartile range (Q1-Q3). The differences in parameters were assessed using the MannWhitney U test and the Wilcoxon signed-rank test. Fisher's exact test was used to estimate the differences in the traits' frequency. Spearman's rank correlation coefficient was calculated to determine the relationships between the traits. The differences were considered significant at p < 0.05 (5%).

А

B

Fig. 1. Gating strategy for the populations of cells with hybrid phenotype and expression of the leukocyte, epithelial, stem and EMT markers and Integrin receptors (B) exemplified by the cells circulating in blood

Logistic regression was used as a multivariate method to assess the relationships between the traits and build the prognostic models. When building mathematical models, the threshold values were determined by ROC analysis. The probability of an event was calculated using the following formula: P = ev'/(1+ey), where

P was the probability of a trait; Y was a regression equation value; e was a mathematical constant equal to 2.72. When the probability P > 50%, the risk of the event was considered to be high; when the probability P < 50%, the risk was considered to be low. The differences were considered significant at p < 0.05 (5%).

Table 2. Frequency of cells with hybrid phenotype in patients with breast cancer

№ Phenotype Frequency, % Significance level

Blood

1 CD45+EpCAM+CK7/8- 93.5% (101/108) p1-2 = 0.1137; p1-3 = 0.0045

2 CD45+EpCAM+CK7/8+ 86.1% (93/108) p2-3 = 0.2785

3 CD45+EpCAM-CK7/8+ 79.6% (86/108)

Primary tumor

4 CD45+EpCAM+CK7/8- 94.6% (35/37) P4-1 = 1.0000; p4_5 = 0.0123;

P4-6 = 0.0001

5 CD45+EpCAM+CK7/8+ 70.3% (26/37) p5-2 = 0.0452; p5-6 = 0.2305

6 CD45+EpCAM-CK7/8+ 54.1% (20/37) P6-3 = 0.0046

RESULTS

iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.

Subpopulation composition of cells with hybrid phenotype

Flow cytometry was used to estimate the expression of the CD45 leukocyte marker and the EpCAM and CK7/8 epithelial markers in the circulating cells and primary tumor cells. The

А

CD45+ cell populations showing co-expression of two epithelial markers (CD45+EpCAM+CK7/8+) and mono-expression of one epithelial marker (CD45+EpCAM+CK7/8-, CD45+EpCAM-CK7/8+) were found in blood and primary tumors of the majority of patients (Table 2).

The largest population most often found in both blood and primary tumor was represented by cells with the CD45+EpCAM+CK7/8- phenotype (Table 2; Fig. 2A).

B

С

200

150 100 50

Number of cells per 1000 tumor cells

Tumor

p = 0.0233 ^-1

-CZI

60

—i—

40

CD45+EpCAM-CK7/8+

CD45+EpCAM+CK7/8+

CD45+EpCAM+CK7/8-

T

200 400 600

Number of cells per 1 mL of blood

Blood

800

CD44+CD24-CD44-CD24-

£Hh o-

p = 0.0306

H

20

Number of cells per 1000 tumor cells Tumor

p = 0.0001

r h

p = 0.0386

L H

p = 0.0003

200

—I—

150

—I—

100

—I—

50

CD45+EpCAM-CK7/8+

CD45+EpCAM+CK7/8+

CD45+EpCAM+CK7/8-

—I-1-1-1-

200 400 600 800 Number of cells per 1 mL of blood

Blood

1000

H

p = 0.0003

N-Cadherln+ N-Cadherln-

m-

p = 0.0171

H

-1—

200

Number of cells per 1000 tumor cells

400 600 800 Number of cells per 1 mL of blood

1000

Fig. 2. Number of cells with hybrid phenotype In blood and primary tumors of patients with breast cancer. А. Number of cells with various combinations of the CD45 leukocyte and EpCAM and CK7/8 epithelial markers expression. B. Number of cells showing stem features. C. Number of cells showing features of epithelial-mesenchymal transition

0

Table 3. Frequency of cells with hybrid phenotype showing stem features In patients with breast cancer

№ Phenotype Frequency, % Significance level

CD45+EpCAM+CK7/8-

1 CD44+CD24- 85.7% (54/б3) р1-2 = G.G54G

2 CD44CD24- 9б.8% (б1/б3)

Blood CD45+EpCAM+CK7/8+

3 CD44+CD24- 77.8% (49/б3) Р3-4 = G.4179

4 CD44CD24- б9.8% (44/б3)

CD45+EpCAMCK7/8+

5 CD44+CD24- 47.б% (3G/63) Р5-б = °.47б°

б CD44CD24- 55.б% (35/б3)

CD45+EpCAM+CK7/8-

7 CD44+CD24- б2.2% (23/37) р7-8 = G.2G3G; р7-1 = G.G126

8 CD44CD24- 78.4% (29/37) Р8-2 = G.GG48

CD45+EpCAM+CK7/8+

Primary tumor 9 CD44+CD24- 51.4% (19/37) Р9-ю = G.3497; Р9-3 = G.GG81

Ю CD44CD24- 37.8% (14/37) P1G-4 = o.ooзo

CD45+EpCAMCK7/8+

11 CD44+CD24- 43.2% (1б/37) Р11-12 = G.G811; Р11-5 = G.6842

12 CD44CD24- 21.б% (8/37) Р12-б = G.GG15

No significant correlations were revealed for the number of cells with hybrid phenotype in blood and primary tumors of BC patients.

Assessment of the metastatic traits in cells with hybrid phenotype

Acquisition of the stemness and EMT features by tumor cells, including CTCs, is associated with their capability of self-renewal, anticancer therapy resistance, and metastatic

5000

potential increase [3, 12, 18]. Furthermore, the integrin expression in CTCs plays an important role in metastasis and is likely to promote targeting distant organs by these cells, thereby determining the location of prospective metastases [6-7]. The same properties can be possessed by CTCs with hybrid phenotype. In this regard we have analyzed metastatic potential of the cells with hybrid phenotype circulating in blood and primary tumor cells by assessing the features of stemness, EMT, and expression of integrin receptors.

"O

о _o

_Q

4—

о E

ф

Ф _Q

E

iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.

3

4GGG

3GGG

2GGG

1GGG

p = G.GGG2 I I

+ 5 ю ю ю ю ю ю

m m m m m m m

> > > > > > >

<с + <с <с <с <с <c

4 - - if if -

m m m m m m m

СО СО СО со СО CO СО

m m m m m m m

Fig. 3. Number of cells with hybrid phenotype showing expression of ß3, ß4, and aVß5 Integrin molecules In blood of patients with breast cancer

G

Table 4. Frequency of cells with hybrid phenotype showing EMT features In patients with breast cancer

№ Phenotype Frequency, % Significance level

CD45+EpCAM+CK7/8-

1 N-cadherin+ 92.1% (58/63) р1-2 = 0.7175

2 N-cadherin- 95.2% (60/63)

CD45+EpCAM+CK7/8+

Blood 3 N-cadherin+ 65.1% (41/63) P3-4 = 0.0235

4 N-cadherin- 84.1% (53/63)

CD45+EpCAM-CK7/8+

5 N-cadherin+ 52.4% (33/63) P-6 = 0.5909

6 N-cadherin- 58.7% (37/63)

CD45+EpCAM+CK7/8-

7 N-cadherin+ 45.9% (17/37) р7-8 = 0.0004; р7-1 = 0.0000

8 N-cadherin- 86.5% (32/37) р8-2 = 0.1419

CD45+EpCAM+CK7/8+

Primary tumor 9 N-cadherin+ 29.7% (11/37) р9_ю = 0.0340; р9_3 = 0.0009

10 N-cadherin- 56.8 (21/37) Р10-4 = 0.0042

CD45+EpCAM-CK7/8+

11 N-cadherin+ 5.4% (2/37) р11-12 = 0.0001; р11-5 = 0.0000

12 N-cadherin+- 45.9% (17/37) Р12-6 = 0.2988

Detection of the stemness features Detection of the EMT features

The results of the analysis of the stemness features in the populations of cells with hybrid phenotype in blood and primary tumor by flow cytometry are provided in Table 3 and Fig. 2B. It was determined that the stemness features were found in all populations of cells with hybrid phenotype, in both blood and primary tumor. However, no significant differences in the frequency of cells possessing and not possessing stem-like properties were revealed. Cells with the CD45+EpCAM+CK7/8-CD44+CD24- and CD45+EpCAM+CK7/8+CD44+CD24- phenotypes were less common in primary tumor than in blood (p = 0.0126 and p = 0.0081, respectively) (Table 3).

Quantification of the cell populations with hybrid phenotype demonstrated that the CD45+EpCAM+CK7/8-CD44-CD24-cells showing mono-expression of the EpCAM epithelial marker and no stemness features prevailed in blood (p = 0.0306); there were significantly more hybrids possessing stem-like properties among the CD45+EpCAM-CK7/8+ cells of the tumor (p = 0.0233) (Fig. 2B).

The EMT features in the populations of cells with hybrid phenotype were estimated via detection of N-cadherin in the cells by flow cytometry (Table 4; Fig. 2C).

Expression of N-cadherin was found in all populations of cells with hybrid phenotype, in both blood and primary tumor. However, N-cadherin-positive cells were significantly less common than N-cadherin-negative cells (Table 4).

Quantification of cells showing EMT features demonstrated that the number of cells with hybrid phenotype showing N-cadherin expression in both blood and primary tumor was significantly lower than the number of cells with no N-cadherin expression (Fig. 2C).

Integrin interface assessment

The results of analysis of the рэ, p4, and aVp5 integrin expression in circulating cells with hybrid phenotype by flow cytometry are provided in Table 5 and Fig. э. Estimation of the studied cells' frequency showed that cells with the p3+p4+aVp5+ phenotype were the least common (р = 0.0003) (Table 5).

Table 5. Frequency of cells with hybrid phenotype showing integrin expression in patients with breast cancer

№ Phenotype Frequency, % Significance level

1 ß3+ß4+aVß5+ 64.1% (25/39) p1-8 = 0.0003

2 ß3+ß4+aVß5- 82.1% (32/39) р2_8 = 0.0564

3 ß3+ß4aVß5+ 92.3% (36/39) P3-8 = 0.6151

4 ß3+ß4aVß5- 87.2% (34/39) Р„ = 0.2002

5 ß3ß4+aVß5+ 82.1% (32/39) P5-8 = 0.0564

6 ß3ß4+aVß5- 97.4% (38/39) Рб-8 = 1.0000

7 ß3ß4aVß5+ 97.4% (38/39) р7_8 = 1.0000

8 ß3-ß4"aVß5" 97.4% (38/39)

3000

2000

1000

p = 0.0

p = 0.0422

I-1

p = 0.0100

p= 0.0103

li Uà T

002

LM -LM +

p= 0.0031

p= 0.0338

I-1

_

CD44+CD24-

ITG-

ITG+

B4+B4+AvB5+

Fig. 4. Stem features and integrin expression in circulating cells with hybrid phenotype in breast cancer patients with lymph node metastasis

The number of circulating cells with hybrid phenotype showing expression of the 03- and/or 04- and/or aV05 integrin receptors (ITG+) was significantly higher p = 0.0002) than the number of cells showing no expression of these molecules (ITG-) (Fig. 3).

Association between metastatic traits of cells with hybrid phenotype and lymph node metastasis

Comparative analysis of the metastatic traits in cells with hybrid phenotype revealed the association of the cells showing the stemness features and the expression of integrin molecules with lymph node metastasis (LM).

400

8 300

Thus, the level of "stem" CD45+EpCAM+CK7/8-CD44+CD24-cells in blood turned out to be significantly higher (p = 0.0422) in patients with LM than in patients with no LM (Fig. 4).

The levels of ITG- and ITG+ cells with hybrid phenotype were significantly higher in patients with no LM than in patients with LM (p = 0.0103 and p = 0.0031, respectively) (Fig. 4). At the same time, patients with LM has a significantly larger number of ITG+ cells (p = 0.0002) relative the number of ITG- cells. The increase in the number of 03+04+aV05+ cells (p = 0.0338) compared to patients with no metastasis was found in blood of patients with LM (Fig. 4).

The logistic regression models confirmed the role in lymph node metastasis played by cells with hybrid phenotype.

HM -

HM +

p < 0.0001 I-1

p = 0.033

p = 0.0079 Î

p = 0.043

I-1

p= 0.0079

I-1

N-CaH+ CD44CD24- CD44+CD24- 1\1-Сяг|-

N-CaH+ CD44 CD24- CD44+CD24- CD44 CD24- CD44+CD24-

EpCAM+CK7/8- EpCAM+CK7/8+

Fig. 5. Stem and EMT features in circulating cells with hybrid phenotype in breast cancer patients with distant metastasis

EpCAMCK7/8+

0

2000 ■

T3

o _o .£2

iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.

1000'

HM-HM+

p = 0.0009

p = 0.0100

ï

•••

p = 0.0394

ITG-

ITG+

B3B4+AVB5

Fig. 6. Integrin expression by cells with hybrid phenotype in breast cancer patients with distant metastasis

Thus, the risk of LM in BC patients turned out to be associated with the presence of cells with the CD45+EpCAM+CK7/8+ and CD45+EpCAM-CK7/8+ phenotypes showing stemness features in blood. The mathematical model is as follows:

Y = -2.4 + 2.7X1 - 1.0X2,

where Y is the regression equation value; -2.4 is the regression coefficient of the constant term; X1 is the level of CD45+EpCAM+CK7/8+ hybrid cells in blood (X1 = 1 when the frequency is less than 14.94 cells per 1 mL of blood, X1 = 2 when the frequency exceeds 14.94 cells per 1 mL of blood); 2.7 is the regression coefficient of this trait; X2 is the level of CD45+EpCAM-CK7/8+CD44+CD24- cells in blood (X2 = 1 when the frequency exceeds 2.49 cells per 1 mL of blood, X2 = 2 when the frequency is less than 2.49 cells per 1 mL of blood); -1.0 is the regression coefficient of this trait.

The model's sensitivity is 79% and specificity is 85% (x2 = 18.49; p = 0.0001).

Thus, the study has shown that cells with hybrid phenotype have such properties, as stemness and co-expression of 03-, P4-, and aVp5-integrins, that are likely to contribute to the mechanisms underlying lymph node metastasis in BC.

Association between metastatic traits of cells with hybrid phenotype and distant metastasis

Comparative analysis of the metastatic traits of cells with hybrid phenotype showed that cells with no features of EMT or stemness were associated with distant metastasis (DM).

Thus, patients with DM showed a significant increase in the frequency of CD45+EpCAM+CK7/8-N-cadh- and CD45+EpCAM+CK7/8+N-cadh- cells (p = 0.021 and p = 0.033, respectively) compared to patients with no DM, and the decrease in the frequency of CD45+EpCAM+CK7/8-N-cadh+ and CD45+EpCAM+CK7/8+N-cadh+ cells (p = 0.0079 and p = 0.0079, respectively) relative to N-cadherin-negative cells (Fig. 5).

The number of CD45+EpCAM-CK7/8+CD44+CD24- cells in individuals with DM was significantly lower than the number of CD45+EpCAM-CK7/8+CD44-CD24- cells (p = 0.015) (Fig. 5).

No significant differences in the frequency of cells with hybrid phenotype showing and not showing features of stemness and EMT were found in primary tumors of individuals with DM.

Comparative analysis of the properties possessed by cells with hybrid phenotype in individuals with DM revealed an increase in the frequency of cells showing integrin expression (Fig. 6).

The number of ITG- cells was significantly lower than the number of ITG+ cells in patients with no DM or patients with DM (p = 0.0100 and p = 0.0009, respectively) (Fig. 6). The decrease in the number of 03-p4+aVp5- circulating cells (p = 0.0394) associated with DM relative to no DM was also found (Fig. 6).

The risk of DM in BC patients turned out to be associated with the presence of CD45+EpCAM+CK7/8+ cells in blood, regardless of the presence of the features of stemness or EMT, and CD45+EpCAM-CK7/8+ cells showing stemness features. The mathematical model is as follows:

Y = 63.5 - 31.8X1 - 30.0X2,

where Y is the regression equation value; 63.5 is the regression coefficient of the constant term; X1 is the frequency of CD45+EpCAM+CK7/8+ hybrid cells in blood (X1 = 1 when the frequency is less than 14.94 cells per 1 mL of blood, X1 = 2 when the frequency exceeds 14.53 cells per 1 mL of blood); -31.8 is the regression coefficient of this trait; X2 is the frequency of CD45+EpCAM-CK7/8+CD44+CD24- cells in blood (X2 = 1 when the frequency exceeds 2.49 cells per 1 mL of blood, X2 = 2 when the frequency is less than 2.49 cells per 1 mL of blood); -30.0 is the regression coefficient of this trait.

The model's sensitivity is 100.0% and specificity is 98.3% (X2 = 29.52; p = 0.0000004).

DISCUSSION

A biological phenomenon of hybridization in cancer is still a source of debate. Despite numerous in vitro and in vivo studies conducted in the recent decades, there is still no evidence that hybrid tumor cells can cause tumor progression.

Fusion of normal cells and cancer cells is considered to be the most probable mechanism underlying hybrid cell

0

generation. Thus, in vitro studies revealed spontaneous fusion of normal breast epithethelial cells and cancer cells, cancer cells only, epithelial tumor cells and endothelial cells, epithelial tumor cells and stromal cells [19]. It was also noted that the processes of cell fusion and hybrid cell generation were enhanced after using radiotherapy and chemotherapy due to local inflammation in the tumor microenvironment and tissue regeneration processes [20-21]. Discovering the biological nature of cells with hybrid phenotype is still a pressing issue.

Subpopulation analysis of cells with hybrid phenotype has shown that these cells can express one epithelial marker (EpCAM or CK7/8) or both markers. According to modern concepts, the EpCAM and CK7/8 markers are expressed mainly by epithelial cells [22]. However, there is evidence of the EpCAM expression in the bone marrow-derived precursor cells, such as early erythroid precursors [23]. Under physiologic conditions, precursor cells are recruited from the bone marrow during reparative regeneration if needed [24]. In tumor process, these cells are involved in generation and maintenance of the tumor and premetastatic niches and, therefore, contribute to the emergence of metastatic foci in distant organs [25].

Thus, the cells showing expression of CK7/8 and CD45 (CD45+EpCAM+CK7/8+ and CD45+EpCAM-CK7/8+) are likely to by hybrids of leukocytes/macrophages and epithelial tumor cells, while the CD45+EpCAM+CK7/8- cell population can be represented by both leukocyte-epithelial hybrids and bone marrow-derived hematopoietic progenitor cells.

Assessment of the metastatic traits of cells with hybrid phenotype has shown that these cells are involved in the mechanisms underlying LM and DM. Thus, the logistic regression data suggest that both metastasis types are associated with the same patterns: the increase in the number of circulating CD45+EpCAM+CK7/8+ cells and the decrease in the number of CD45+EpCAM-CK7/8+ cells showing stemness features. Furthermore, in LM, the increase in blood levels of CD45+EpCAM+CK7/8~ cells showing stemness features and CD45+EpCAM+ cells showing co-expression of 03-, 04-, and aV05-integrins is observed. In DM, there is an increase in the number of CD45+EpCAM+CK7/8- and CD45+EpCAM+CK7/8+ cells showing no EMT features along with the decrease in the number of the same cells showing EMT features and CD45+EpCAM+ cells showing mono-expression of p4-integrin.

Acquisition of stem-like properties by hybrid cells can (by analogy with CTCs) contribute to their anticancer therapy resistance and metastatic potential enhancement. The increase in the number of 03-, 04-, and aV05-expressing cells with hybrid phenotype observed in LM also suggests enhancement of their metastatic potential. Integrin 03, expressed mainly by platelets, hematopoietic cells, and angiogenic endothelial cells, is responsible for adhesion in hemostasis, wound healing, and angiogenesis. The association of integrin 03 with tumor growth, lymph node and bone marrow metastasis, as well as reduced patient survival, has been shown [26]. Integrin 04 is expressed mainly by epithelial cells. In BC, integrin 04 promotes tumor invasion, increases tumor cell viability, and contributes to angiogenesis [27]. Integrin aV05 is a positive regulator of the tumor cells' stemness, it contributes to their growth and invasion [28]. As we have earlier reported, CTCs express the CXCR4 pro-migratory marker [29]. Elevated expression of CXCR4 and integrin molecules in tumor cells can ensure their high migration potential and promote dissemination to various organs. As a result, the hybrid cells can acquire properties that are necessary for metastasis [30].

Today, it is unknown whether the hybrids of leukocytes and epithelial tumor cells can divide indefinitely to form tumors in distant organs, i.e. play the role of tumor seeds. It is not known whether hybrid circulating cells can have a function of niche formation. However, it is already clear that hybrid cells of the tumor and peripheral blood are associated with both LM and DM.

CONCLUSIONS

Thus, the population of cells with hybrid phenotype, just like CTCs, is characterized by high degree of heterogeneity, including the combination of the features of stemness, EMT, and diverse integrin interface. Cells with hybrid phenotype are involved in lymph node and distant metastasis. LM is associated with such metastatic traits of the circulating cells with hybrid phenotype, as the stemness features and co-expression of 03-, 04-, and aV05-integrins. In DM, metastatic traits of cells with hybrid phenotype are associated with the stemness features, but not with the EMT features and integrin expression. Understanding of the involvement of cells with hybrid phenotype in metastasis can be useful in terms of improving anti-metastatic therapy.

References

1. Ruiz-Rodríguez AJ, Molina-Vallejo MP, Aznar-Peralta I, González Puga C, Cañas García I, González E, et al. Deep Phenotypic Characterisation of CTCs by Combination of Microfluidic Isolation (IsoFlux) and Imaging Flow Cytometry (ImageStream). Cancers (Basel). 2021; 13 (24): 6386.

2. Zhang Q, Kong D, Yang Z, Li G, Cheng S, Feng L, et al. Prognostic value of stem-like circulating tumor cells in patients with cancer: a systematic review and meta-analysis. Clin Exp Med. 2023. DOI: 10.1007/s10238-023-01009-0.

3. Orrapin S, Udomruk S, Lapisatepun W, Moonmuang S, Phanphaisarn A, Phinyo P, et al. Clinical Implication of Circulating Tumor Cells Expressing Epithelial Mesenchymal Transition (EMT) and Cancer Stem Cell (CSC) Markers and Their Perspective in HCC: A Systematic Review. Cancers (Basel). 2022; 14 (14): 3373.

4. Grigoryeva ES, Tashireva LA, Alifanov VV, Savelieva OE, Vtorushin SV, Zavyalova MV, et al. Molecular subtype conversion in CTCs as indicator of treatment adequacy associated with metastasis-free survival in breast cancer. Sci Rep. 2022; 12 (1): 20949.

5. Fina E, Cleris L, Dugo M, Lecchi M, Ciniselli CM, Lecis D, et al. Gene signatures of circulating breast cancer cell models are a

source of novei moiecuiar determinants of metastasis and improve circuiating tumor ceii detection in patients. J Exp Ciin Cancer Res. 2G22; 41 (1): 7B.

6. Grigoryeva ES, Tashireva LA, Saveiieva OE, Zavyaiova MV Popova NO, Kuznetsov GA, et ai. The Association of Integrins ß3, ß4, and aVß5 on Exosomes, CTCs and Tumor Ceiis with Locaiization of Distant Metastasis in Breast Cancer Patients. Int J Moi Sci. 2G23; 24 (3): 2929.

J. Hoshino A, Costa-Siiva B, Shen TL, Rodrigues G, Hashimoto A, Tesic Mark M, et ai. Tumour exosome integrins determine organotropic metastasis. Nature. 2G15; 527 (757b): З29-Зб.

S. Guan X, Ma F, Li C, Wu S, Hu S, Huang J, et ai. The prognostic and therapeutic impiications of circuiating tumor ceii phenotype detection based on epitheiiai-mesenchymai transition markers in the first-iine chemotherapy of HER2-negative metastatic breast cancer. Cancer Commun (Lond). 2G19; 39 (1): 1.

9. Wankhede D, Grover S, Hofman P. Circuiating Tumor Ceiis as a Predictive Biomarker in Resectabie Lung Cancer: A Systematic Review and Meta-Anaiysis. Cancers (Basei). 2G22; 14 (24): B112.

10. Jou HJ, Ling PY, Hsu HT. Circuiating tumor ceiis as a "reai-time

liquid biopsy": Recent advances and the application in ovarian cancer. Taiwan J Obstet Gynecol. 2022; 61 (1): 34-39.

11. Li H, Liu Q, Liang S, Yao P, Lv J, Wang G, et al. Circulating tumor cells and neutrophil-lymphocyte ratio are predictive markers for metastatic colorectal cancer patients. Transl Cancer Res. 2021; 10 (1): 288-97.

12. Papadaki MA, Stoupis G, Theodoropoulos PA, Mavroudis D, Georgoulias V, Agelaki S. Circulating Tumor Cells with Stemness and Epithelial-to-Mesenchymal Transition Features Are Chemoresistant and Predictive of Poor Outcome in Metastatic Breast Cancer. Mol Cancer Ther. 2019; 18 (2): 437-47.

13. Lustberg MB, Balasubramanian P, Miller B, Garcia-Villa A, Deighan C, Wu Y, et al. Heterogeneous atypical cell populations are present in blood of metastatic breast cancer patients. Breast Cancer Res. 2014; 16 (2): R23.

14. Ishizawa K, Yamanaka M, Saiki Y, Miyauchi E, Fukushige S, Akaishi T, et al. CD45+CD326+ Cells are Predictive of Poor Prognosis in Non-Small Cell Lung Cancer Patients. Clin Cancer Res. 2019; 25 (22): 6756-63.

15. Dittmar T, Zänker KS. Introduction. In: Dittmar T, Zänker KS. Cell Fusion in Health and Disease. Advances in Experimental Medicine and Biology, vol 950. Dordrecht: Springer, 2011; p. 1-3.

16. Laberge GS, Duvall E, Haedicke K, Pawelek J. Leukocyte-Cancer Cell Fusion-Genesis of a Deadly Journey. Cells. 2019; 8 (2): 170.

17. Pospelova RA. Lejkokoncentraciya v klinicheskoj praktike: diagnosticheskoe znachenie. M.: Medicina, 1973; 88 s. Russian.

18. Agnoletto C, Corra F, Minotti L, Baldassari F, Crudele F, Cook WJJ, et al. Heterogeneity in Circulating Tumor Cells: The Relevance of the Stem-Cell Subset. Cancers (Basel). 2019; 11 (4): 483.

19. Noubissi FK, Ogle BM. Cancer Cell Fusion: Mechanisms Slowly Unravel. Int J Mol Sci. 2016; 17 (9): 1587.

20. Kaur E, Rajendra J, Jadhav S, Shridhar E, Goda JS, Moiyadi A, et al. Radiation-induced homotypic cell fusions of innately resistant

glioblastoma cells mediate their sustained survival and recurrence. Carcinogenesis. 2015; 36 (6): 685-95.

21. Yan B, Wang J, Liu L. Chemotherapy promotes tumour cell hybridization in vivo. Tumour Biol. 2016; 37 (4): 5025-30.

22. Gires O, Pan M, Schinke H, Canis M, Baeuerle PA. Expression and function of epithelial cell adhesion molecule EpCAM: where are we after 40 years? Cancer Metastasis Rev. 2020; 39 (3): 969-87.

23. Fidanza A, Stumpf PS, Ramachandran P, Tamagno S, Babtie A, Lopez-Yrigoyen M et al. Single-cell analyses and machine learning define hematopoietic progenitor and HSC-like cells derived from human PSCs. Blood. 2020; 136 (25): 2893-904.

24. Rodrigues M, Kosaric N, Bonham CA, Gurtner GC. Wound Healing: A Cellular Perspective. Physiol Rev. 2019; 99 (1): 665706.

25. Psaila B, Lyden D. The metastatic niche: adapting the foreign soil. Nat Rev Cancer. 2009; 9 (4): 285-93.

26. Pan B, Guo J, Liao Q, Zhao Y. p1 and p3 integrins in breast, prostate and pancreatic cancer: A novel implication. Oncol Lett. 2018; 15 (4): 5412-6.

27. Abdel-Ghany M, Cheng HC, Elble RC, Pauli BU. The breast cancer beta 4 integrin and endothelial human CLCA2 mediate lung metastasis. J Biol Chem. 2001; 276 (27): 25438-46.

28. Liu F, Wu Q, Dong Z, Liu K. Integrins in cancer: Emerging mechanisms and therapeutic opportunities. Pharmacol Ther. 2023; 247: 108458.

29. Savelieva OE, Tashireva LA, Buldakov MA, Mukhamedzhanov RK, Kaigorodova EV, Denisov EV, et al. CXCR4 expression in different subsets of CTCs and single (detached) breast cancer cells. Siberian journal of oncology. 2018; 17 (4): 75-80.

30. Ramakrishnan M, Mathur SR, Mukhopadhyay A. Fusion-derived epithelial cancer cells express hematopoietic markers and contribute to stem cell and migratory phenotype in ovarian carcinoma. Cancer Res. 2013; 73 (17): 5360-70.

Литература

1. Ruiz-Rodriguez AJ, Molina-Vallejo MP, Aznar-Peralta I, González Puga C, Cañas Garcia I, González E, et al. Deep Phenotypic Characterisation of CTCs by Combination of Microfluidic Isolation (IsoFlux) and Imaging Flow Cytometry (ImageStream). Cancers (Basel). 2021; 13 (24): 6386.

2. Zhang Q, Kong D, Yang Z, Li G, Cheng S, Feng L, et al. Prognostic value of stem-like circulating tumor cells in patients with cancer: a systematic review and meta-analysis. Clin Exp Med. 2023. DGI: 10.1007/s10238-023-01009-0.

3. Grrapin S, Udomruk S, Lapisatepun W, Moonmuang S, Phanphaisarn A, Phinyo P, et al. Clinical Implication of Circulating Tumor Cells Expressing Epithelial Mesenchymal Transition (EMT) and Cancer Stem Cell (CSC) Markers and Their Perspective in HCC: A Systematic Review. Cancers (Basel). 2022; 14 (14): 3373.

4. Grigoryeva ES, Tashireva LA, Alifanov VV, Savelieva GE, Vtorushin SV, Zavyalova MV, et al. Molecular subtype conversion in CTCs as indicator of treatment adequacy associated with metastasis-free survival in breast cancer. Sci Rep. 2022; 12 (1): 20949.

5. Fina E, Cleris L, Dugo M, Lecchi M, Ciniselli CM, Lecis D, et al. Gene signatures of circulating breast cancer cell models are a source of novel molecular determinants of metastasis and improve circulating tumor cell detection in patients. J Exp Clin Cancer Res. 2022; 41 (1): 78.

6. Grigoryeva ES, Tashireva LA, Savelieva GE, Zavyalova MV Popova NG, Kuznetsov GA, et al. The Association of Integrins ß3, ß4, and aVß5 on Exosomes, CTCs and Tumor Cells with Localization of Distant Metastasis in Breast Cancer Patients. Int J Mol Sci. 2023; 24 (3): 2929.

T. Hoshino A, Costa-Silva B, Shen TL, Rodrigues G, Hashimoto A, Tesic Mark M, et al. Tumour exosome integrins determine organotropic metastasis. Nature. 2015; 527 (7578): 329-35.

S. Guan X, Ma F, Li C, Wu S, Hu S, Huang J, et al. The prognostic and therapeutic implications of circulating tumor cell phenotype detection based on epithelial-mesenchymal transition markers in the first-line chemotherapy of HER2-negative metastatic breast cancer. Cancer Commun (Lond). 2019; 39 (1): 1.

9. Wankhede D, Grover S, Hofman P. Circulating Tumor Cells as a Predictive Biomarker in Resectable Lung Cancer: A Systematic Review and Meta-Analysis. Cancers (Basel). 2022; 14 (24): 6112.

10. Jou HJ, Ling PY, Hsu HT. Circulating tumor cells as a "real-time liquid biopsy": Recent advances and the application in ovarian cancer. Taiwan J Obstet Gynecol. 2022; 61 (l): 34-39.

11. Li H, Liu Q, Liang S, Yao P, Lv J, Wang G, et al. Circulating tumor cells and neutrophil-lymphocyte ratio are predictive markers for metastatic colorectal cancer patients. Transl Cancer Res. 2021; 10 (1): 288-97.

12. Papadaki MA, Stoupis G, Theodoropoulos PA, Mavroudis D, Georgoulias V, Agelaki S. Circulating Tumor Cells with Stemness and Epithelial-to-Mesenchymal Transition Features Are Chemoresistant and Predictive of Poor Outcome in Metastatic Breast Cancer. Mol Cancer Ther. 2019; 18 (2): 437-47.

13. Lustberg MB, Balasubramanian P, Miller B, Garcia-Villa A, Deighan C, Wu Y, et al. Heterogeneous atypical cell populations are present in blood of metastatic breast cancer patients. Breast Cancer Res. 2014; 16 (2): R23.

14. Ishizawa K, Yamanaka M, Saiki Y, Miyauchi E, Fukushige S, Akaishi T, et al. CD45+CD326+ Cells are Predictive of Poor Prognosis in Non-Small Cell Lung Cancer Patients. Clin Cancer Res. 2019; 25 (22): 6756-63.

15. Dittmar T, Zänker KS. Introduction. In: Dittmar T, Zänker KS. Cell Fusion in Health and Disease. Advances in Experimental Medicine and Biology, vol 950. Dordrecht: Springer, 2011; p. 1-3.

16. Laberge GS, Duvall E, Haedicke K, Pawelek J. Leukocyte-Cancer Cell Fusion-Genesis of a Deadly Journey. Cells. 2019; 8 (2): 170.

17. Поспелова Р. А. Лейкоконцентрация в клинической практике: диагностическое значение. М.: Медицина, 1973; 88 с.

18. Agnoletto C, Corra F, Minotti L, Baldassari F, Crudele F, Cook WJJ, et al. Heterogeneity in Circulating Tumor Cells: The Relevance of the Stem-Cell Subset. Cancers (Basel). 2019; 11 (4): 483.

19. Noubissi FK, Ogle BM. Cancer Cell Fusion: Mechanisms Slowly Unravel. Int J Mol Sci. 2016; 17 (9): 1587.

20. Kaur E, Rajendra J, Jadhav S, Shridhar E, Goda JS, Moiyadi A, et

al. Radiation-induced homotypic cell fusions of innately resistant glioblastoma cells mediate their sustained survival and recurrence. Carcinogenesis. 2015; 36 (6): 685-95.