Evaluation of hTERT Gene Expression and Chromosome 7 Copy Number Variation in Anal Squamous Intra-Epithelial Lesions: A Pilot Study.

Background: Akin to cervical squamous intra-epithelial neoplasia (CIN), anal squamous intra-epithelial lesion (a-SIL) is attributed to persistent infection with high-risk human papilloma virus infection. Amplification of human telomerase reverse transcriptase (hTERT) gene and aneuploidy are known to correlate with CIN evolution. It is plausible that the underlying genetic events in a-SIL are similar. We conducted this cross-sectional analytical study with the objective of determining expression of hTERT gene expression and chromosome 7, as marker of cell ploidy in a-SIL.
Methods: Conventional anal cytology was performed in 86 adult consenting subjects with history of receptive anal intercourse (RAI) and 4 controls without history of RAI. Cases with a-SIL and controls were subjected to fluorescent in-situ hybridization (FISH) to evaluate hTERT gene and chromosome 7 expression, as marker of cell ploidy. Results were expressed as number of abnormal nuclei (≥3 respective signals), maximum degree of amplification, mean signals/nucleus and proportion of cases showing abnormal nuclei.
Results: Twenty cases showed a-SIL; with 15 atypical squamous cells of undetermined significance (ASCUS), 3 low grade squamous intra-epithelial lesion (LSIL) and 2 cases of high-risk cytology. Expression of both hTERT gene and chromosome 7 increased from controls to ASCUS to LSIL with concomitant increase in proportion of cases having abnormal hTERT gene and chromosome 7 expression. Conclusions: Positive association of hTERT gene with a-SIL suggests its possible role in evolution of anal squamous abnormalities. Increase in chromosome 7 also correlated positively with a-SIL. These findings corroborate the similarities between squamous carcinogenesis in CIN and a-SIL. Copyright:

INTRODUCTION

Anal cytology is a cornerstone screening investigation for detection of anal squamous intra-epithelial lesions (a-SIL) in the high-risk population.[1] Individuals engaging in unprotected receptive anal intercourse (RAI), such as men who have sex with men (MSM) are most vulnerable.[2] Akin to cervical intra-epithelial neoplasia (CIN), a-SIL is etio-pathologically attributed to persistent infection with high-risk (hr) human papilloma virus (HPV) infection. Human immunodeficiency virus (HIV) infection is a major determinant of progression of a-SIL to invasive squamous cell carcinoma (SCC); increasing the risk by up-to 78 folds.[23]

The hr-HPV driven SCCs harbor several recurrent genetic alterations acquired in different phases of progression from pre-neoplastic to invasive stage.[4] Gene amplification, i.e., increase in number of gene copies is one of the common genetic alterations in solid organ cancers; it is associated with increased protein production and functional output (dosage effect).[56] Amplification of telomerase subunit genes, namely human telomerase RNA component (hTERC) and human telomerase reverse transcriptase (hTERT) has been consistently described in HPV driven cervical oncogenesis.[7891011121314] The hTERT is considered to be the limiting factor in telomerase activity, as it is not expressed in healthy cells. Aneuploidy is a generic genetic marker of chromosomal instability in cancer cells and is known to associate with CIN progression. Copy number variation (CNV) of chromosome 7 is an uncommon event in SCC. Thus, enumeration of centromeric region (cen) of chromosome 7 (cen7) by centromere enumeration probe (cep) is plausible reflection of overall cell ploidy. Positive association between hTERT expression and CNV of chromosome 7 with worsening CIN has been reported.[151617]

Expression and significance of hTERT gene and CNV of chromosome 7 in a-SIL, the HPV driven squamous carcinogenesis of anal canal, are hitherto unreported. We conducted this cross-sectional analytical study with the objective of determining hTERT gene expression and CNV of chromosome 7 in a-SIL and to evaluate if there was any association with cytologic grade of abnormality. Fluorescent in-situ hybridization (FISH) technique was applied in view of its ability to determine gene expression in individual cell nuclei and sensitivity.

MATERIAL AND METHODS

Eighty-six adult (>18 years of age) consenting subjects with history of RAI were recruited as a part of post-graduate thesis research work conducted in department of Pathology in collaboration with department of Dermatology in Delhi, India. Personal and sexual details were noted. HIV status was available/determined for all. Subjects with anal fissure/fistula/ulceration/history of anal or other cancer were excluded. Four adult subjects with no history of RAI were recruited as controls. Conventional anal cytology smears were prepared and reported in accordance with Bethesda 2014.[18] Written informed consent was obtained from all participants; subject anonymity and confidentiality was maintained. The study was approved by Institutional Ethical Committee - Human Research.

FISH was performed on a-SIL cases and controls. The cells obtained by a cyto-brush were suspended in an alcohol-based preservative, washed once and mixed with density gradient solution to facilitate cell sedimentation. Cell lysis was done using hypotonic solution (0.075 molar potassium chloride). After 2 washes with fixative solution (3:1 methanol and glacial acetic acid respectively), cells were stored at -80 degree Celsius (°C). HTERT gene expression and chromosome 7 CNV were determined using respective fluorescent probes [Zytolight, dual coloured probe binding to hTERT gene at 5p15 (green fluorochrome) and 5q31.2 for centromeric region (orange fluorochrome) respectively and Zytovision cep7 (orange fluorochrome) which binds to cytoband 7p11.1-q11.1 in cen7]. Standard two-day procedure was followed. Briefly, the cells were allowed to thaw and rejuvenated by resuspension in fresh fixative, transferred to chilled glass slide, fixed using 70% acetic acid followed by dehydration through increasing concentrations of ethanol. Cell digestion was performed with pepsin (Sigma pepsin powder ≥250 units/mg solid). Slide and probe denaturation were performed in formamide solution (76°C for 7 minutes) with rinses in phosphate buffered saline as applicable. The slides were dehydrated in chilled absolute ethanol and 10 μl probe applied to the marked area, cover-slipped and sealed with rubber sealant. Probe hybridization was allowed to proceed in dark by overnight incubation at 37°C. Next day, cover slips were removed and slides were washed in post-hybridization detergent buffer solutions. After dehydration and air drying, counterstaining with 10 μl 4’,6-diamidino-2-phenylindole (DAPI) was done, the slides were cover-slipped and sealed with nail varnish.

Image analysis was performed using Leica DM6000B fluorescent microscope. At-least twenty well-spread well-visualized non-overlapping interphase nuclei were evaluated under oil immersion. The number of abnormal nuclei were converted to per 20 nuclei when higher number of nuclei were counted. For hTERT gene, only those nuclei which showed at least two cep5 signals (orange fluorochrome) were considered. Three indices each were calculated for both hTERT and cen7: (a) number of abnormal nuclei (nuclei with ≥3 respective fluorescent signals)/20 nuclei (labelled ab-nuc-hTERT and ab-nuc-cen7 respectively), (b) mean number of signals/nucleus (labelled m-sig-hTERT and m-sig-cen7 respectively) derived from sum total of signals counted/number of nuclei counted and (c) maximum degree of amplification (highest number of fluorescent signals counted in a given preparation) (labelled Amax-hTERT and Amax-cen7 for cen7 polysomy respectively). We also determined the proportion of cases having abnormal hTERT and cen7 expression in all cytologic categories.

Statistical analyses were performed with Statistical Package for Social Sciences (SPSS) version 20 (SPSS, Chicago, Illinois, USA). Pearson correlation coefficient (r) was used to measure the strength of association of above indices with various cytologic categories. Differences between mean expressions of all indices across various cytologic categories were analyzed by one-way ANOVA test followed by Tukey's test. The differences were considered significant at P < 0.05.

RESULTS

Conventional anal cytology smears were satisfactory in 85 subjects and all controls; one case with low squamous cellularity was excluded. All controls and 65 RAI practicing subjects had cytologic diagnosis of negative for intra-epithelial lesion/malignancy (NILM). Twenty RAI practicing subjects had a-SIL. These included 5 transgenders (4 HIV sero-positive), 13 males (11 HIV sero-positive) and 2 females (1 HIV sero-positive). Mean age of subjects with and without squamous cell abnormality was 30.9 ± 6.8 and 31.5 ± 8.2 years respectively. The distribution of a-SIL was as follows: atypical squamous cells of undetermined significance (ASCUS) (15: 5 transgenders, 8 males and 2 females), low-grade squamous intra-epithelial neoplasia (LSIL) (3: all males), atypical squamous cells cannot exclude high-grade squamous intra-epithelial neoplasia (ASC-H) (1 male) and high-grade squamous intra-epithelial neoplasia (HSIL) (1 male) respectively. ASC-H and HSIL results were combined, in view of their high-risk association and small numbers and labeled as high-risk cytology.

Using FISH, hTERT gene expression and CNV of chromosome 7 were analyzed in a-SIL cases and controls [Table 1]. Representative FISH images are shown in Figure 1. Dot-plots are depicted in Figure 2. Nine (45%) and 10 (50%) a-SIL cases showed abnormal nuclei with ≥3 hTERT and cen7 signals respectively. None of the controls had abnormal nuclei; all showed maximum of 2 hTERT gene and cen7 signals. Cases had higher index values compared to controls; with significant inter-group differences between m-sig-hTERT (2.06 versus 1.9, P = 0.003) and m-sig-cen7 indices (2.09 versus 1.9, P = 0.012). The proportion of cases with ≥3 signals for hTERT in ASCUS, LSIL and high-risk cytology were 40%, 100% and none respectively. The corresponding proportions for cen7 were 46.7%, 100% and none respectively. There was positive correlation between m-sig-hTERT index with worsening cytologic grade (r = 0.8663, P < 0.0001). One-way ANOVA test revealed significantly higher levels of hTERT gene and cen7 expression in LSIL compared to controls (P < 0.01 for all indices comparisons). Differences between LSIL versus ASCUS were also significant (P ≤ 0.05) for all indices except m-sig-cen7. ASCUS versus control comparison showed significant difference in m-sig-hTERT only (2.03 versus 1.9, P < 0.05).

Table 1

Comparison of hTERT gene and cen7 expression indices in study subjects stratified according to anal cytology result

Cytology/FISH indices (mean±standard deviation)	Controls (n=4)	All cases of a-SIL (n=20)	ASCUS (n=15)	LSIL (n=3)	ASC-H and HSIL combined (n=2)
Ab-nuc-hTERT	0±0	1.01±1.4	0.59±0.9	3.77±0.33*&	0±0
m-sig-hTERT	1.9±0.08	2.06±0.09*	2.03±0.05*	2.23±0.05*&	2.0±0
Amax-hTERT	2.0±0	2.60±0.73	2.47±0.62	3.67±0.47*&	2.0±0
Ab-nuc-cen7	0±0	1.36±1.73	0.85±1.05	4.8±0.28*&	0±0
m-sig-cen7	1.9±0.08	2.09±0.13*	2.07±0.13	2.24±0.03*	2.0±0
Amax-cen7	2.0±0	2.65±0.73	2.53±0.62	3.67±0.47*&	2.0±0

*P≤0.05, comparison with respect to controls; &P≤0.05, comparison with respect to ASC-US; #P≤0.05, comparison with respect to LSIL; ASC-H and HSIL group excluded from comparison due to small numbers

Figure 1

Representative FISH images in a-SIL cases. Expression of hTERT gene in control (a), ASCUS (b) and LSIL (c). The hTERT gene and chromosome 5 centromere signals are green and red respectively (Oil immersion, 1000x). Cen7 expression (red signal) in in ASCUS (d) and LSIL (e). (Oil immersion, 1000x)

Figure 2

Dot-plots depicting correlation between cytologic diagnoses and hTERT (blue dots) and cen7 (orange dots) indices: (a) ab-nuc (b) m-sig and (c) A-max respectively. In case of indices having same values, both dots overlap

DISCUSSION

The incidence of SCC of anal canal has risen in the at-risk population in the post ART era, especially in high income countries.[19] Upon comparison of 5-year incidence of anal SCC in five continents between 2008-2012 with 1988-1992; Kang et al.[19] found significant increase in standardized rate ratio (SRR) in anal SCC in both men and women; highest increases were seen in those aged <60 years (SRR = 2.34 in men and 2.76 in women). Longer life-span of people living with HIV and incomplete immune reconstitution at genital mucosa despite ART are likely contributory factors.[2021] Anal cytology screening is advocated in high-risk population for detection of a-SIL which has easier management and better prognosis than invasive SCC.[122] The concept of anal squamous carcinogenesis is modelled on the lines of cervical oncogenesis; persistent infection with hr-HPV being the main underlying event in both.[1] Hence it is logical to hypothesize that the genetic events in a-SIL are likely to be similar to CIN. Amplification of hTERT gene is seen in high frequency in SCC and is associated with high telomerase activity.[10] Furthermore, hTERT gene amplification has been found to correlate positively with grade and progression of CIN.[11121314] Increment in ploidy has also been reported in CIN. Chromosome 7 is said to be uncommonly affected in cervical squamous carcinogenesis and has been used as a surrogate marker of ploidy in CIN.[151617] However, there is no literature on expression of either biomarkers in a-SIL.

In the present study, abnormal anal cytology results were found in 23.5% of satisfactory smears. The proportion of HIV sero-positive status was higher in those with SIL compared to those without (80% versus 58.4%). The reported frequency of a-SIL is highly variable; HIV infection being the most important determinant. Wang et al.[23] recently reported significant difference in a-SIL prevalence in HIV positive MSM compared to HIV negative counterparts (20.8% versus 4.8%, P < 0.001). Notably, there is dearth of data on anal cytology from India; with only a handful studies. An earlier study from our institution found a-SIL in 35% and 20% of HIV positive and negative MSM respectively.[24] From eastern India, Gautam et al.[25] have reported 63.15% of HIV positive men to have a-SIL. Factors significantly associated with a-SIL were anal intercourse and higher HIV clinical stage. One study in Indian women found 8% anal smears to have cytologic abnormalities.[26] While the Indian medical fraternity acknowledges RAI practices in MSM and women; there are no current national guidelines on screening/management of cases with a-SIL.

We determined hTERT gene and cen7 expression by FISH in subjects with a-SIL detected on conventional anal cytology. The results were expressed as several indices. Abnormal nuclei having ≥3 hTERT gene and cen7 signals were found in 45% and 50% a-SIL cases respectively, compared to none in controls (both). Furthermore, within the a-SIL group, the proportions increased from ASCUS to LSIL. The results of high-risk cytology group were unexpected since the indices were lower than LSIL and ASCUS. This discrepancy between cytology and FISH results prompted us to revisit our cases. The conventional cytology slides were subjected to independent cyto-morphologic review, which concurred with the original diagnoses. We also reviewed the FISH slides, and failed to find any abnormal nuclei with either ≥3 hTERT gene or cen7 signals. These discrepant results could have one or more underlying explanations. One, severe reactive changes in metaplastic cells mimicked the high grade cyto-morphology, thereby leading to such morphologic diagnosis (false positive cytology diagnosis). This cyto-morphologic resemblance is well-known in cervical cytology; up-to 50% of cervical HSIL return as < CIN2, i.e., low grade CIN histology on diagnostic loop electrosurgical excision.[27] Although the severity of these genetic aberrations and proportion of abnormal cells is generally known to increase with worsening cyto-morphology, not all abnormal looking cells may carry the said genetic aberrations. In this context, other possible could be actual lack of representative cells harboring the genetic abnormalities in the FISH slides (technical factor) or absence of changes in hTERT gene and cen7 expression in the true morphologically abnormal cells present in FISH slides (true negative). Visnovsky et al.[11] too did not find hTERT gene amplification in 16.7% of their cervical HSIL cases. In an ideal setting, this anomalous result would warrant repeating both anal cytology and FISH and performing biopsy to help resolve this dilemma. Unfortunately, both subjects did not return for repeat sampling despite repeated requests. Although, we could not establish the reason/s, our best guess would be there was cytologic overcall for these cases. In similar situations in cervical cytology, biomarkers like p16 are commonly used to address discrepant findings following cervical cytology.

In view of the very low number of high-grade cases and also the divergent FISH results, we re-analyzed our results by excluded these cases as outliers. By doing this, the positive correlation of hTERT gene and cen7 expression with worsening cytologic diagnosis (till LSIL) extended to all indices (unlike only m-sig-hTERT index, as mentioned earlier in results) and the correlation became stronger (r ≥ 0.6213 and P ≤ 0.002 for all indices). This positive correlation was evident as progressive increase in number of cells with ≥3 hTERT gene signals and incremental gene copy number suggesting clonal expansion and evolution respectively. The inter-group comparisons showed significant differences between hTERT gene expression indices (all) of controls versus LSIL and ASCUS versus LSIL. These findings reiterate the survival advantage of the cells harboring hTERT gene amplification. Similar to hTERT results, cen7 indices also showed positive association with worsening cytology (till LSIL) and several significant inter-group differences.

In view of lack of literature on both hTERT gene and cen7 expression in a-SIL, we compared our results with those reported in cervical squamous cell abnormalities. Visnovsky et al.[11] have reported positive correlation between hTERT gene amplification and severity of cervical cytologic diagnosis (P < 0.001). Our results of 1.9, 2.03 and 2.23 m-sig-hTERT in controls, ASCUS and LSIL respectively are comparable to findings of Visnovsky et al., who noted 2.03, 2.04 and 2.14 mean hTERT signals/cell in cervical NILM, ASCUS and LSIL respectively. Using a receiver operating curve derived cut-off value of 2.08 for mean count of hTERT signals per cell, they found 1/9 NILM, 0/5 ASCUS, 0/1 ASC-H, 4/8 LSIL, 10/12 HSIL and 4/4 carcinoma to be positive for hTERT. Costa et al.[12] investigated expression of several genes by FISH in 53 cervical dysplastic lesions. Their results implicated hTERT gene in evolution of CIN to invasive SCC. Wang et al.[13] found hTERT messenger ribonucleic acid expression in 88.9% SCC, 100% HSIL, 100% ASC-H, 37.6% LSIL, 40.0% ASCUS, and 0% negative cases respectively. The difference between high-grade (SCC, HSIL, and ASC-H) vs low-grade cervical lesions (LSIL, ASC-US, and normal) were found to be significant (P < 0.0001). Branca et al.[14] have reported linear relationship between intensity of hTERT protein expression and worsening CIN grade, with major transition at CIN3/invasive cancer (odds ratio = 18.81, P = 0.001). In their study, none of the negative histology case showed increased hTERT expression compared to 15% in CIN1, 30% in CIN2, 74.2% in CIN3 and 85.3% in SCC. They suggested hTERT to be a possible marker for prediction of high-grade lesions.

Ploidy assessment in CIN using cen7 has been reported by few authors. Heselmeyer et al.[15] evaluated hTERC gene expression and ploidy by cen7 in CIN and found cases with severe cytologic findings to have significantly higher frequency of tetraploid cells. They however, concluded that polyploidy and hTERC gene amplification were independent events. Hopman et al.[16] also found progressive increase in frequency of aneuploidy from CIN2/3 (18%) to invasive SCC (42%) through CIN3/micro-invasive carcinoma (39%). Luhn et al.[17] too found increase in median number of cells with ≥3 cen7 signals with histologic severity of cervical lesions (p-trend = 0.02). However, they attributed this increase to general genomic instability and suggested that it may not be an appropriate control for ploidy. Accumulation of cells with increased ploidy results from failure of mitotic mechanisms essential for healthy cell division. Such cells are chromosomally unstable and prone to further genetic events, coincident with increase in severity of morphologic changes.

Our study is limited by small number of cases, a constraint due to the limited time period and absence of any previous study on hTERT and/or cen7 expression in a-SIL to determine adequate sample size. Information about HPV status of the study population would have added weightage to our study. However, the purpose of this study was to determine expression of hTERT gene and cen7 and to find association, if any, with grade of a-SIL. The lower values of hTERT gene and cen7 indices in high-risk cytology may appear divergent; the possible scientific reasons have been explained for, but unfortunately could not be resolved. Histology is considered a gold standard for confirmation of the cytology results, which was not possible in the present study. The current guidelines for ASCUS and LSIL are cytologic surveillance; high-risk cytology cases require high resolution anoscopy and biopsy from suspect areas, a facility not available in our institution. Periodic cytologic surveillance is offered to such subjects in resource limited settings.

To conclude that, we found positive association between hTERT gene and cen7 expression with presence and increasing severity of anal squamous precursor lesions (till LSIL). These results recapitulate the HPV driven carcinogenesis events of CIN. These findings should be tested in larger sized studies with available histology for furthering our understanding of anal squamous oncogenesis.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.