Abstract
Introduction:
Cervical cancer ranks the fourth prevalent cancer in women at the global level, and the second in poor countries. The main objectives of the present study were to investigate the risk factors associated with cervical cancer and to study their possible association with the decision to take a cervical biopsy.
Aim:
The main objectives of this study were to investigate the risk factors associated with cervical cancer and to study their possible association with the decision to take a cervical biopsy.
Methods:
It was cross-sectional study and we analyzed an online data posted on Kaggle. This Dataset is obtained from UCI Repository. A list of risk factors for cervical cancer leading to biopsy examination was included, such as age, number of sexual partners, first sexual intercourse, number of pregnancies, smoking variables, hormonal contraceptives, IUD, and sexually transmitted disease variables, Hinselmann, Schiller, Cytology, and Biopsy. The dataset was prepared for appropriateness through filtering invalid cases with missing data.
Results:
The results of the study showed that the following variables were significantly associated with cytological examination: STD-Condylomotosis (p=0.035), STD-Pericondylomotosis (p=0.029), STD_HIV (p=0.006), Hinselmann (p<0.001), Schiller (p<0.001), and biopsy (p<0.001). the results also showed that the following variables were significantly associated with cytological examination.
Conclusion:
Taken together, cytological variables or biopsy examination variables if carried out at an early stage, lead to better diagnostic and therapeutic options.
Keywords: cervical cancer, risk factors, dataset, Kaggle, HPV, sexually transmitted disease
1. INTRODUCTION
Cancer is the main cause of death internationally and is responsible for approximately 9 million deaths in 2015 (1). Cancer is developed as normal cells are transformed into tumor cells passing through various stages that involve the transformation of precancerous cells into malignant cells. Early identification of cancer leads to better therapeutic responses with higher survival rates, less morbidity, and less treatment costs (1).
Cervical cancer ranks the fourth prevalent cancer in women at the global level, and the second in poor countries (2). It has been estimated that there are approximately 600000 new cases of cervical cancer, and about 300,000 deaths yearly, and the majority of these cases are likely to be encountered in poor countries (3, 4).
However, cervical cancer can be controlled through vaccination against papillomavirus infection (HPV) and routine screening programs including cytology (5). Moreover, surgical removal of affected tissue at an early stage can cure cervical tumor (3, 5).
Cervical cancer examination involves cytology screening, colposcopy, and biopsy. Cytology and colposcopy depend on image screening. Cytological examination requires stained smears (6). The high costs of these techniques make their use not applicable in countries with low income (6).
Colposcopic examination implies 4 stages. In the first stage, a normal saline solution is applied followed by examining both of the squamous and columnar epithelium with a magnifier lens to identify if there are signs of transformation. The general appearance of the squamous epithelium is likely to be smooth with pinkish tone. On the other hand, the columnar epithelium is almost red with villous appearance (7). In the second stage, a green filter is used to improve the contrast of the vessels, here reticular and hairpin-shaped capillaries are identified (7). In the third stage, 5% acetic acid solution is applied and followed by observation of cervix tissues for both squamous and columnar epithelium which leads to detection of precancerous lesions. This process is known as Hinselmann. The fourth stage is known as the Schiller’s test, and is characterized by applying Lugol’s iodine solution to the cervix. Normal cells stain as brown or black, while abnormal cells are stained or partially stained (7, 8).
2. AIM
The main objectives of this study were to investigate the risk factors associated with cervical cancer and to study their possible association with the decision to take a cervical biopsy.
3. METHODOLOGY
Data source
An online data posted on Kaggle (9). This Dataset is Obtained from UCI Repository. A list of risk factors for cervical cancer leading to biopsy examination was included.
Dataset included the following cervical risk factors: Age, Number of sexual partners, First sexual intercourse, Number of pregnancies, Smokes, Smokes (years), Smokes (packs/year), Hormonal Contraceptives, Hormonal Contraceptives (years), IUD, IUD (years), STDs, STDs (number), STDs: condylomatosis, STDs: cervical condylomatosis, STDs: vaginal condylomatosis, STDs: vulvar-perineal condylomatosis, STDs: syphilis, STDs: pelvic inflammatory disease, STDs: genital herpes, STDs: molluscum contagiosum, STDs: AIDS, STDs: HIV, STDs: Hepatitis B, STDs: HPV, STDs: Number of diagnosis, Hinselmann, Schiller, Cytology, and Biopsy.
Data preparation
The data in its raw source was composed of 858 entries of patients. Data contained missing data due to the right of privacy of participants in filling data. Inappropriate data was deleted. The remaining valid data included 676 entries of patients.
Statistical analysis
The analysis of data was made using SPSS version 21. Descriptive analysis was used to describe data including frequencies and percentages for categorical variables, and means with standard deviations for continuous variables. The impacts of cervical cancer risk factors on biopsy or cytology examination were evaluated using One Way ANOVA. The significance was assessed at α≤0.05.
4. RESULTS
General characteristics of study participants
The general characteristics of the study participants are summarized in Table 1. The mean age is 27.23±8.7 years. The mean of sexual partners is 2.50±1.63. The mean age for first sexual intercourse is 17.14±2.85 years old. The mean of pregnancies number is 2.32±1.6. Smoking was reported by 14.2% of study participants. The mean of Smoking years 8.85±7.71, while the mean of smoking packets per year was 3.32±5.48. The use of hormonal contraceptives was reported by about 65% of study participants. For those women who used hormonal contraceptives, the mean number of hormonal contraceptives was 3.53±4.12 per year. The use of IUD was reported by about 11% of study participants with a rate of 4.72±4.006.
Table 1. General characteristics of study participants.
| Variable | Description |
|---|---|
| Age (M±SD) years | 27.23±8.70 |
| Number of sexual partners (M±SD) | 2.50±1.63 |
| First sexual intercourse (M±SD) | 17.14±2.85 |
| Number of pregnancies (M±SD) | 2.32±1.6 |
| Smoking (N, %): Yes No |
95 (14.2%) 572 (85.8%) |
| Smoking years (M±SD) | 8.85±7.71 |
| Smoking packets/year | 3.32±5.48 |
| Hormonal contraceptives (N, %): Yes No |
431 (64.6%) 236 (35.4%) |
| Number of hormonal contraceptives per year | 3.53±4.12 |
| IUD (N, %): Yes No |
75 (11.2%) 592 (88.8%) |
| IUD per year (M±SD) | 4.72±4.006 |
Sexually transmitted diseases and related variables
As seen in Table 2, sexually transmitted diseases (STD) were reported in about 10% of the study participants. The mean number of STDs was 1.70±0.70. Condylomatosis was reported in about 6% of the study participants. Vaginal condylomatosis was reported in 0.6% of study participants. Pericondylomatosis was reported in 5.4% of study participants. Syphilis was reported in 2.2% of the study participants. Pelvic inflammatory diseases, genital herpes, and Molluscum contagiosum were reported in 0.1% of study participants for each. A total of 13 (1.9%) cases were positive for HIV. Hepatitis was reported in 0.1% of cases, and HPV was reported in 0.3% of cases. The frequency of diagnosis for STD was in the following patterns: no diagnosis was reported for 91.2% of cases, one time for 8.6% of cases, two diagnoses were reported for 0.1% of cases, and three diagnoses were reported for 0.1% of cases. Hinselmann was carried out for 4.5% of cases, Schiller test was carried out for 9.3% of study participants. Cytology was reported for 5.7% of cases, and biopsy was taken from 6.7% of study participants.
Table 2. Sexually transmitted diseases (STD).
| Variable | Description |
|---|---|
| STD (N, %): Yes No |
65 (9.7%) 602 (90.3%) |
| No of STD (M±SD) | 1.70±0.70 |
| SDT- Condylomatosis (N, %): Yes No |
37 (5.5%) 630 (94.5%) |
| STD-vaginal condylomotosis (N, %): Yes No |
4 (0.6%) 667 (99.4%) |
| STD-pericondylomotosis (N, %): Yes No |
36 (5.4%) 631 (94.6%) |
| STD-Syphlis (N, %): Yes No |
15 (2.2%) 652 (97.8%) |
| STD-pelvic inflammatory disease (N, %): Yes No |
1 (0.1%) 666 (99.9%) |
| STD- genital herpes (N, %): Yes No |
1 (0.1%) 666 (99.9%) |
| STD- Molluscum contagiosum (N, %): Yes No |
1 (0.1%) 666 (99.9%) |
| STD-HIV (N, %): Yes No |
13 (1.9%) 654 (98.1%) |
| STD- Hepatitis (N, %): Yes No |
1 (0.1%) 666 (99.9%) |
| STD-HPV (N, %): Yes No |
2 (0.3%) 665 (99.7%) |
| STD-No of diagnosis (N, %): Not diagnosed 1 2 3 |
608 (91.2%) 57 (8.6%) 1 (0.1%) 1 (0.1%) |
| Hinselmann (N, %): Yes No |
30 (4.5%) 637 (95.5%) |
| Schiller (N, %): Yes No |
62 (9.3%) 605 (90.7%) |
| Cytology (N, %): Yes No |
38 (5.7%) 629 (94.3%) |
| Biopsy (N, %): Yes No |
45 (6.7%) 622 (93.3%) |
The impact of study variables on biopsy taking decision
We investigated the impact of study variables on biopsy taking decision using One Way ANOVA test. As seen in Table 3, the following variables were significantly associated with biopsy taking decision: smoking (p=0.0483), hormonal contraceptives per year (p=0.023), STD (p=0.003), condylamotosis (p=0.002), pericondylomotosis (p=0.002), genital herpes (p<0.001), AIDS (p=0.018), No of diagnosis (p=0.017), Hinselmann (p<0.001), Schiller (p<0.001), and cytology (p<0.001).
Table 3. The impact of study variables on biopsy taking decision.
| Variable | Significance |
|---|---|
| Smoking | 0.0483 |
| Hormonal contraceptives per year | 0.023 |
| STD | 0.003 |
| Condylamotosis | 0.002 |
| Pericondylomotosis | 0.002 |
| Genital herpes | <0.001 |
| HIV | 0.018 |
| No of diagnosis | 0.017 |
| Hinselmann | <0.001 |
| Schiller | <0.001 |
| Cytology | <0.001 |
5. DISCUSSION
The following cervical risk factors were significantly associated with biopsy examination: smoking, hormonal contraceptives per year, STD, condylamotosis, pericondylomotosis, genital herpes, AIDS, STD-No of diagnosis, Hinselmann, Schiller, and cytology. Several studies have confirmed the significant relationship between cervical cancer and smoking (12-14). We think that smoking plays an important roles in the pathogenesis of diseases through increasing the potential to develop free radicals that harm tissues and accelerate disease progress. The results of this study showed that the use of hormonal contraceptives for a long time is a risk factor of cervical cancer. Our results are in agreement with other studies such as the study of Chichareon et al (12) who found that the use of hormonal contraceptives for >4 years is considered a risk factor for cervical cancer. Hormones contraceptives are believed to increase the susceptibility to HPV infection, an issue that improves the pathogenesis of cervical cancer (15). Sexually transmitted diseases in general have been shown in various studies to increase the likelihood of developing cervical cancer because viruses interfere with the genetic material of host cells and accelerate the development of cervical cancer (16, 17). HPV is a essential sexually transmitted disease that increases the likelihood of occurrence of cervical cancer (18-20).
6. CONCLUSION
The results of the present study revealed several risk factors of cervical cancer that should be taken into account to carry out the biopsy examination. These factors include smoking, hormonal contraceptives per year, sexually transmitted diseases, Hinselmann, Schiller, and cytology.
Authors contribution:
All authors were included in all steps of preparation this article. Final proof reading was made by the first author.
Conflict of interest:
None declared.
Financial support and sponsorship:
Nil.
REFERENCES
- 1.Unlersen Muhammed Fahri, Sabanci Kadir, Özcan Muciz. Determining Cervical Cancer Possibility by Using Machine Learning Methods. International Journal of Latest Research in Engineering and Technology. 2017;3(12):65–71. [Google Scholar]
- 2.Conceição Teresa, Braga Cristiana, Rosado Luís, Vasconcelos Maria João M. A Review of Computational Methods for Cervical Cells Segmentation and Abnormality Classification. Int. J. Mol. Sci. 2019;20:5114. doi: 10.3390/ijms20205114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Fernandes K, Cardoso J. S, Fernandes J. Cham, Switzerland: Springer; 2015. Temporal segmentation of digital colposcopies in Pattern Recognition and Image Analysis; pp. 262–271. [Google Scholar]
- 4.WHO, World Health Organization. Human Papillomavirus (HPV) and Cervical Cancer, Fact Sheet. [17 September 2019]. Available online: https://www.who.int/news-room/fact-sheets/detail/human-papillomavirus-(hpv)-and-cervical-cancer.
- 5.Control C. F. D, et al. Cervical cancer screening among women aged 18-30 years—United states, 2000-2010. MMWR Morbidity Mortality Weekly Rep. 2013;61:51–52. 1038. [PubMed] [Google Scholar]
- 6.Bengtsson E, Malm P. Screening for cervical cancer using automated analysis of PAP-smears. Comput. Math. Methods Med. 2014. Art. no. 842037. [DOI] [PMC free article] [PubMed]
- 7.Sellors J. W, Sankaranarayanan R. New Delhi, India: Diamond Pocket; 2003. Colposcopy and Treatment of Cervical Intraepithelial Neoplasia: A Beginner’s Manual. [Google Scholar]
- 8.Kauffman R. P, Griffin S. J, Lund J. D, Tullar P. E. Current recommendations for cervical cancer screening: Do they render the annual pelvic examination obsolete? Med. Princ. Pract. 2013;22(4):313–322. doi: 10.1159/000346137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. [1/12/2019]; https://www.kaggle.com/loveall/cervical-cancer-risk-classification. [Google Scholar]
- 10.Kashyap N, Krishnan N, Kaur S, Ghai S. Risk Factors of Cervical Cancer: A Case-Control Study. Asia-Pacific journal of oncology nursing. 2019;6(3):308–314. doi: 10.4103/apjon.apjon_73_18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Misra JS, Srivastava S, Singh U, Srivastava AN. Risk-factors and strategies for control of carcinoma cervix in India: hospital based cytological screening experience of 35 years. Indian J Cancer. 2009;46:155–9. doi: 10.4103/0019-509x.49155. [DOI] [PubMed] [Google Scholar]
- 12.Chichareon S, Herrero R, Muñoz N, Bosch FX, Jacob MV, Deacon J, et al. Risk factors for cervical cancer in Thailand: a case-control study. J Natl Cancer Inst. 1998;90:50–7. doi: 10.1093/jnci/90.1.50. [DOI] [PubMed] [Google Scholar]
- 13.Tebeu PM, Major AL, Rapiti E, Petignat P, Bouchardy C, Sando Z, et al. The attitude and knowledge of cervical cancer by Cameroonian women; a clinical survey conducted in Maroua, the capital of Far North Province of Cameroon. Int J Gynecol Cancer. 2008;18:761–65. doi: 10.1111/j.1525-1438.2007.01066.x. [DOI] [PubMed] [Google Scholar]
- 14.Paul BS. Studies on the Epidemiology of Cervical Cancer. Southern Assam. Assam University Journal of Science & Technology: Biological and Environmental Sciences. 2011:36–42. [Google Scholar]
- 15.Murphy N, Xu L, Zervoudakis A, et al. Reproductive and menstrual factors and colorectal cancer incidence in the Women’s Health Initiative Observational Study. British Journal of Cancer. 2017;116(1):117–125. doi: 10.1038/bjc.2016.345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Baldwin SB, Wallace DR, Papenfuss MR, et al. Condom use and other factors affecting penile human papillomavirus detection in men attending a sexually transmitted disease clinic. Sex Transm Dis. 2004;31(10):601–607. doi: 10.1097/01.olq.0000140012.02703.10. [DOI] [PubMed] [Google Scholar]
- 17.Ngoma M, Autier P. (2019). Cancer prevention: cervical cancer. Ecancermedicalscience. 13:952. doi: 10.3332/ecancer.2019.952. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sigurdsson K, Sigvaldason H. Longitudinal trends in cervical histological lesions (CIN 2–3+): a 25-year overview. Acta Obstet Gynecol Scand. 2006;85(3):359–365. doi: 10.1080/00016340500432481. [DOI] [PubMed] [Google Scholar]
- 19.Tan N, Sharma M, Winer R, et al. Model-estimated effectiveness of single dose 9-valent HPV vaccination for HIV-positive and HIV-negative females in South Africa. Vaccine. 2018;36(32 Pt A):4830–4836. doi: 10.1016/j.vaccine.02.023. 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.van der Aa MA, Pukkala E, Coebergh JW, et al. Mass screening programmes and trends in cervical cancer in Finland and the Netherlands. Int J Cancer. 2008;122(8):1854–1858. doi: 10.1002/ijc.23276. [DOI] [PubMed] [Google Scholar]