Identification and Quantification of “Cancer Stem Cells” in Potentially Malignant and Malignant Disorders of Head and Neck Region:A Flow-Cytometric Study

Introduction

Head and neck cancers are group of diversified entities arising from the mucous membrane lining and other various tissues of oro-facial region. The cause for the same has been documented to be certain life style and environmental risk factors. To name a few, habits such as tobacco smoking, alcohol, and certain strains of virus such as Human Papilloma virus, environmental factors such as UV-light, chemicals etc, are collectively called as carcinogens. At the molecular level, these carcinogens set in genetic instabilities in the form of mutations, causing the conversion of the inherent proto-oncogene to an aberrant oncogene. The aggressiveness, metastatic potential, as well as their recurrence is dependent mainly on their cellular genotypic and phenotypic characteristics.¹

Globally, oral cancers stand sixth in the incidence among all the types of cancers. India has the largest number of oral cancer cases and forms one-third of its total burden occurring globally.^1,20 Cancer is a disease of the elderly with a few exceptions. Males are reported to exhibit an increased incidence when compared to females.^1,2 Though the early detection of cancers has been possible due to various invasive and non-invasive research investigational tools, the number of cases detected at an advanced stage is still about 60-80%.¹ Hence in the present scenario, a five year survival rate of detected cancers is very low (about 20%), marking the prognosis still to be very bleak.^3,4

Since a great percentage of these malignancies are still being diagnosed at advanced stages, need of the hour and the only light of hope is early detection and a subsequent expected steady decline in their incidence. Various treatment modalities viz surgery or radio and chemotherapies have not been very successful in dealing with the aggressiveness and metastasis of cancers, thereby increasing the morbidity and mortality rates.^4,5

Hence, even after years of research and understanding of various aspects of this malady, the final verdict of its prognosis has still not improved. Among the various arenas of research, the concept of cancer stem cells (CSCs) and their transition, in the incidence, pathogenesis, metastasis and in turn its implications on newer treatment modalities of cancer is much sought after in the recent times.^6,7

Cancer stem cells are sub population of cells present in the arena of an oral cancer tissue. These cells are mutated stem cells, hijacking the attributes of self-renewal and continued replications native to the characteristics of a normal stem cell. Hence, they are said to be contributory to the crucial clinico-pathological behaviours such as tumerogenicity, metastasis, and recurrence. The concept of CSCs and their role in causing oral cancer is very alarming and is a subject of research to be thoroughly explored for newer methods of early identification as well as treatment modalities. But the identification of the same among the tumor mass is relatively difficult. Their identification is being assessed by trying to analyze their surface protein markers and also their intracellular enzymatic signatures.^8,9

Most researched surface markers used to identify CSC among oral cancers and potentially malignant disorders are CD44, CD133, and ALDH1. The cells showing the presence of these markers were further investigated and isolated by using side population (SP) cells in flow cytometry and cause and effect of these probable SP cancer stem cells were tested on animal models for the formation of similar tumours.

No single surface marker has hitherto been substantially hallmarked to identify a site specific or tissue specific CSCs.^10-12 Hence, a set of surface markers most often identified with CSCs were tested in combinations to identify a CSC in question.

This study and research was tailored to conduct and scientifically prove the presence of sub- population cells called cancer stem cells in Oral cancers (OC) and Potentially Malignant Disorders (PMDs) using CD44, CD133 and ALDH1 surface markers in combinations.

Aims and Objectives

1. Identification and quantification of CSC in head and neck squamous cell carcinomas

2. Identification and quantification of CSC in head and neck PMDs

3. Identification and quantification of CSC in healthy controls

4. Evaluation and comparison of CSC in all the three groups

Materials and Methods

It was a cross-sectional comparative study. A total of 108 patients visiting the outpatient department of Oral Medicine and Radiology were categorized into three groups. Group I and II included 40 subjects each, while group III included 28 subjects. The subjects allocated to the groups were age and gender matched.

After obtaining written informed consent, all the subjects were thoroughly examined and investigated. The study was independently reviewed and approved by the Ethics Committee. Patients suffering from infectious/contagious disease, those who were unable to undergo minor surgical procedures due to systemic health conditions were excluded from the study.

The biopsy specimens were collected depending upon the location of the lesion in pre-cancer and cancer subjects. In OSCC, sites with red or white, painless, non-healing, indurated ulcers were examined and selected. The clinical staging of oral cancer as per AJCC TNM staging was done. Morphologically altered oral mucosal areas with altered colour configuration such as white or red lesions were selected for PMDs. Oral Leukoplakia was staged according to OLEP staging system of Leukoplakia given by Vander Waal et al (1994). OSMF was staged clinically according to Haider et al. (2000), graded histologically as per Kiran Kumar et al. (2007).

Patients reporting for minor oral surgical procedures like pre-prosthetic surgery were taken as controls and the tissue samples were obtained with prior consent.

The tissue samples were grouped as Group I, II and III. Group II & III – Study groups (Histopathologically proven)

•  Group I: Normal controls (Healthy subjects) 
• Group II: Oral cancer 
• Group III: Potentially malignant disorders

Tissue samples were sent for both histopathological evaluations and flow cytometric analyses to establish the cancer stem cells through presence of their surface markers.

Collection of Tumor Specimens

The study was approved by the Institutional Ethical Committee of Bangalore Institute of Dental sciences and was performed according to the guidelines of the study at National Institute of Mental Health and Neurosciences (NIMHANS). Human oral cancer tissues (OC; n=40), pre-malignant cancer specimens (PM; n=28) and tissue from healthy individuals as healthy controls or normal tissues (HC/NT; n=40) were collected. Tissue processing and further experimental procedures were carried out in the facilities provided by NIMHANS, Bangalore, as per the protocol.

Dissociation of Primary Tissues 

Freshly obtained tumor tissues were collected and transported (within 1–2 hours after biopsy) to NIMHANS laboratory in Dulbecco’s modified Eagle medium (DMEM; Sigma- Aldrich, USA) containing 100 U/mL penicillin and 100 mg/mL streptomycin (Invitrogen; Thermo Fisher Scientific, Inc.), 100ug/mL gentamycin (Himedia, TCO26) and 500ng/mL amphotericin B (BionovaA-9528).

Specimens were washed in sterile Dulbecco’s phosphatebuffered saline (PBS) containing antibiotics as mentioned above to remove blood tinge or microbial contaminants if any on the surface. Then they were mechanically minced into small pieces (2–4mm) using sterile scalpel blades and enzymatically dissociated using DMEM supplemented with 10% foetal bovine serum (sigmaAldrich, USA), antibiotics, 1 mg/mL Collagenase (Stem cell Technologies, USA) and 100 U/mL Hyaluronidase (Stem-cell Technologies, USA) in 3mL volume, for 3hrs at 37⁰ C, maintained in a water bath with intermittent mixing.

After incubation, cell suspension was filtered through 70μm sterile cell strainers, centrifuged at 400×g for 5min, and resuspended in DMEM containing antibiotics for cell counting. Finally, viable cells were counted using a haemocytometer by trypan blue staining.

ALDEFLUOR Assay

Aldehyde dehydrogenase (ALDH) activity was detected using the ALDEFLUOR assay kit (Stem-cell Technologies, USA) as described by the manufacturer. Briefly, cell pellet of 1- 0.1x10⁶ cells were resuspended in 500uL of ALDEFLUOR assay buffer and stained with ALDH substrate, Bodipy-aminoacetaldehyde (BAAA). Immediately, 250uL of the cell suspension with ALDH substrate was transferred to the control tube containing 2.5μL Diethylaminobenzaldehyde (DEAB), which is a specific inhibitor of ALDH.

Both the control and test samples were incubated for 40 - 45mins at 37⁰ C, maintained in a water bath protected from light.

Cell surface marker staining

Further, following incubation, the cells were centrifuged at 400×g for 5min, washed with chilled assay buffer and stained with monoclonal antibodies cocktail, anti-human CD133- PE (Biolegend, USA) and anti-human CD44- APC (Biolegend, USA) prepared in chilled assay buffer. The cell pellet was washed and resuspended in 200uL of chilled ALDEFLUOR assay buffer. Cell morphology was evaluated using side scatter (SSC) and forward scatter (FSC) excluding debris. Further, gating was established using ALDEFLUOR/DEAB treated cells as negative gates. Flow cytometry data was acquired and analyzed by BD FACSVerse (BD Biosciences) using FACSuite (BD Biosciences) and FlowJo 10.4 software. Excluding debris, cell clusters were gated during sidescatter and forward-scatter analyses.

Sample Size Calculation

A power analysis was established by G* power version 3.0.1 (Franz Faul universitat, K, Germany). A sample size of 120 (40 per group) yielded 80% power to detect significant differences, with effect size of 0.29 and significance level at 0.05.

Statistical Analysis

Statistical analysis was based on Normalcy test and Shapiro Wilk test. SPSS (Statistical Package for Social Sciences) version 20 (IBM SPASS statistics [IBM corp. released 2011] was used to perform the statistical analysis. Data was entered in the excel spreadsheet. Descriptive statistics of the explanatory and outcome variables were calculated by mean, standard deviation, median and IQR (based on data distribution) for quantitative variables, frequency and proportions for qualitative variables. Inferential statistics like Chi-square test was applied for qualitative variables. ANOVA/Kruskal-wallis test was applied to compare the quantitative variables among the groups with post hoc Bonferroni/Mannwhitney test for inter group comparisons. The level of significance was set at 5%. 

Gated from CD44 to ALDH1

CD44 & ALDH1 Gated separately from singlets

Data was subjected to normalcy test (Shapiro-wilk test). Data showed non-normal distribution. Hence, nonparametric tests (Kruskal-wallis with Post hoc Mannwhitney) were applied. 

OLEP staging system of leukoplakia given by Van der Waal et al. (1994) L0- no evidence of lesion S1- all sites excluding L1<=2cm, FOM & tongue L2-2-4cm, S2- FOM and /or Tongue, L3- >=4cm, Sx—not specified Lx –not specified C1 - Homogeneous, P1- no dys Px - not specified C2 - non-homogeneous, P2- mild dys, C3 - not specified P3-moderate dys, P4 –severe dys, Stage I—any L, S1, C1, P1, or P2, Stage 2--any L, S1, or S2, P1 or P2, Stage 3—any L, S2, C2, P1 or P2, Stage 4—any L, any S, any C, P3 or P4

Inference: This Kruskal-Wallis analysis showed CD44 value to be highest in the healthy group followed by the potentially malignant group and then the oral cancer group; thereby concluding that it is not a statistically significant data as CD44 is a non-standard marker as per the analysis.

Inference: Oral cancer group has a high median value as per the table here. ALDH1 also proved to be a nonsignificant marker, thereby producing a statistically nonsignificant result.

Inference: Kruskal-Wallis test was applied in this as well, which showed statistically significant result in relation to CD133 (p=0.01), thereby concluding that CD133 can be considered as a standard marker.

Inference: Post Hoc Mann-Whitney test was carried out which showed statistically significant difference between oral cancer and potentially malignant group (p=0.005).

Results

Non-parametric tests were applied.

Group I

The control group consisted of 40 subjects. The median age was 46.5 years. Twenty one males (52.5%) and 19 females (47.5%) were included. CD44 expression was 7.58%, CD133 expression was 0.061%, ALDH1 expression was 0.49%.

Group II

The Oral cancer group consisted of 40 subjects. The median age was 59 years. Twenty-two males (55%) and 18 females (45%) were included. Eleven (27.5%) were 

in clinical stage I, 23 (57.5%) were in clinical stage II and 6 (15%) were in clinical stage III. Among these, 15 (37.5%) were in HP grade I, 20 (50%) were in HP grade II and 5(12.5%) in HP grade III. CD44 expression was 7.22%, CD133 expression was 0.03% and ALDH1 showed 1.31% expression.

Group III

The PMDs group consisted of 28 subjects. The difference in sample size from 40 to 28 was inevitable due to the pandemic situation as patients were not available to be included in group III. The median age was 47.5 years. Eleven males (54%) and 17 females (54%) were included. Among the PMDs, Lichen planus were 16 (57.1%), Leukoplakia of different varieties were 6 (21.6 %), whereas other PMDs included were OSMF- 2 (7.2%), verroucous hyperplasia with dysplasia- 3 (10.8 %), Lichenoid reaction- 1 (3.6%). CD44 expression was 5.95%, CD133 expression was 0.08% and ALDH1 showed 0.80 % expression.

When results of all the groups were compared and correlated with regards to the identification of cancer stem cells, based on the presence of surface markers -CD44, CD133 and ALDH1, Group I and II and III showed no statistically significant presence of CD44, ALDH1 positive cancer stem cells, but only CD133 expression was slightly significant.

Discussion

The early diagnosis of oral cancer has been a challenging issue despite many molecular and technological tools available for its prompt detection as well the early management.

In this direction, the presence of certain sub-population cells namely Oral Cancer stem cells and their identification was hypothesized and thought to be a new approach for tackling this malady.^12-15 Many researches and studies have been carried out, to identify these CSCs from the cancer tissue mass based on the presence of certain cell surface markers.^16-25

These Cancer Stem Cells can be identified by specific cell surface markers like CD44, CD133 and also some intra cytoplasmic enzymatic functional markers like ALDH1.^26,27,30

Our study was designed to identify and quantify the expression of CSC markers and thereby positively identify CSCs.

In Oral cancer, the expression of CD44, CD133, and ALDH1 among oral cancer (Group II) was 7.22%, 0.03%, and 1.31 respectively. In PMDs (Group III), the expression of CD44, CD133, and ALDH1 was 5.95%, 0.08% and 0.80% respectively.

As per the objectives set, our study did not show any significant association of CD44, ALDH1 markers between different groups, as well as between different clinical and histological staging in oral cancers. On the contrary, unlike the other studies, our study showed a significant association of CD133. This association should be further analysed.^28-31

Although various researches and studies have documented in affirmation about the pivotal role of oral cancer stem cells in pathophysiology of various oral cancers, there always has been an iota of doubt, where many studies have been questioned for the very existence of cancer stem cells supposedly residing in the stem cell niche.^30,31

Conclusion

In comparison to other studies, our study did not report much difference between the presence or absence of CD44 and ALDH1 in normal tissues, oral cancer and PMDs. This could be either because of the considerably smaller sample size or due to technique sensitivity. Therefore, our study was just a tip of an iceberg to determine the presence or absence of CSCs in patients with oral cancer.

Although the present study did not give a statistically significant result regarding the presence of CSCs in oral cancer, further studies in this regard probably with a larger sample size would be able to give a much clearer picture on this highly debated issue.

Conflicting Interest

None.