Pathological Features of Ovarian Tumors, Diagnosed at a Tertiary Care Hospital in Afghanistan: A Cross-sectional Study.

Cancer will be the leading cause of mortality and one of the most important causes of decreasing life expectancy all over the world in twenty-first century, that would be due to the rapid growth and aging of the population and socioeconomic development. Accordingly, it has been demonstrated that mortality rates from stroke and coronary heart disease have decreased considerably in comparison to cancers in many countries. 1 Globally, the eighth most common cancer among women is ovarian cancer, that constitutes 4% of cancer in female population. The mortality and morbidity related to ovarian cancer, when compared to other reproductive system malignancies, is high. According to recent global estimates, 225,000 new cases are identified each year, and 140,000 patients die every year from ovarian cancer. 2 American Cancer Society (ACS) has revealed that there were 22,530 new cases with 13,980 deaths reported due to ovarian cancers, in 2019 alone. 3 Even though ovarian tumors are not common among children and teenage girls they form a significant section of the gynecological malignancies in this age group, constituting 1% of all malignant neoplasms in children and 8% of the entire childhood abdominal neoplasms. Around 10–30% of all ovarian tumors occurring in girls less than 17 years of age are estimated to be malignant. 4 The ovaries are pelvic organs located on both sides of the uterus adjacent to the lateral pelvic wall, behind the broad ligament and anterior to the rectum. The location of the ovary posterior to the broad ligament and a similar relationship of the ovarian ligament to the ipsilateral fallopian tube helps in the determination of the laterality of a salpingo-oophorectomy specimen. 5 The ovary’s sex and mesenchymal cells are totipotent and multipotent, respectively, thus resulting in a wide spectrum of neoplastic transformations. Therefore, accurate diagnosis of ovarian neoplastic and non-neoplastic lesions is crucial but challenging. 6 Ovarian malignant tumors constitute the third most common type of malignancy in women, surpassed only by cervical and endometrial cancers. 7 On the basis of tumor origin, ovarian neoplasms have been classified by World Health Organization (WHO) into surface epithelial, germ cell, sex cord – stromal and metastatic tumors. The most frequently occurring malignant tumors comprise surface epithelial tumors, which are further classified based on the cell type as serous, mucinous, and endometrioid carcinomas. Considering tumor behavior, ovarian neoplasms can be sub-classified into benign, borderline, and malignant. 8 It must be noted, that around 90% of ovarian neoplasms are benign. 9 Benign ovarian tumors may arise at any age, however, they are commonly found during the childbearing ages, especially between 20 and 45 years of age. Conversely, malignant tumors are more common in advanced ages, usually between 40 and 65 years of age. 10 Although, the risk of developing ovarian cancer during a woman’s lifetime is pretty low with one in every 71 women, most of the deaths occur from complications of the disease. Luckily, 75% of ovarian cancer patients survive at least one year after diagnosis. Sadly, same percentage of patients die from the disease within five years of diagnosis. In general, lifetime chance of dying from invasive ovarian malignant tumor is nearly 1 in 95. 2 In 2012, Afghanistan’s estimated ovarian cancer incidence was 346 while mortality rate was 1.6 (per 100,000) for all ages. 2 Ovarian tumors have various histopathological patterns and considering their prognostic implications, correct diagnosis guides in better treatment plans. 11 To the best of our knowledge, there is no published data available for ovarian neoplasms in Afghanistan. Identification of common benign and malignant neoplasms is essential before appropriate health guidelines for management and funding for the disease are ever planned by the country’s health sector. Pathology and Clinical Laboratory of French Medical Institute for Mothers and Children (FMIC) is one of the country’s very few diagnostic centers, which receives notable numbers of pathological specimens, including those of ovarian origin, from almost every region of the country. Our study provides baseline data for the future research related to ovarian neoplasms in Afghanistan that can be used for further planning of health policies.

A total of 198 cases were included in our study with the mean age of 34.4 (±SD=13.4) at diagnosis. Abdominal mass was the main complaint and almost half of the participants were multiparous, as shown in Table 1 . A considerable number of neoplasms were seen in the right side and most of the benign tumors were cystic. Most of the tumors were benign followed by malignant and borderline. Mature cystic teratoma and serous cystadenocarcinoma were the frequently occurring benign and malignant tumors respectively. Table 1 Demographics and Clinical Characteristics of Women with Ovarian Neoplasms (n=198) Characteristics Frequency Percentage (%) Age (in years) Mean ±standard deviation 34.42±13.4 1 Age groups (in years) Less than 20 21 10.6 21 to 40 116 58.6 41 to 60 55 27.8 Above 61 6 3 Ethnicity Pashtun 55 27.8 Tajik 95 48.0 Hazara 47 23.7 Uzbek-Turkmen 1 0.5 Signs and Symptoms Abdominal pain 76 38.4 Abdominal mass 106 53.5 Abdominal distension 8 4.0 Ascites 2 1.0 Menstrual irregularities 1 0.5 Asymptomatic 5 2.5 Menarche in years (n=195) Mean ±standard deviation 36.6 ± 1.02 Menopause in years (n=40) Mean ±standard deviation 48.7 ± 2.46% Parity Uniparous 38 19.2 Multiparous 113 57.1 Patient with no childbearing age or has no children 47 23.7 Survival status Dead 1 0.5 Alive 197 99.5 Demographics and Clinical Characteristics of Women with Ovarian Neoplasms (n=198) Considering the surgical procedures, total abdominal hysterectomy, and bilateral salpingo-oophorectomy (n=21, 10.6%) was the most common procedure applied for resection of ovarian tumor, followed by bilateral salpingo-oophorectomy (n=11, 5.6%), right salpingo-oophorectomy (n=47, 23.7%), left salpingo-oophorectomy (n=30, 15.2%), right oophorectomy (n=30, 15.2%) and left oophorectomy (n=27, 13.6%). However, there were no surgical information available for (n=32, 16.2%), as shown in Table 2 . Table 2 Surgical Procedure, Laterality, Size and Consistency of Ovarian Neoplasms Characteristics Frequency Percentage (%) Surgical procedure Total abdominal hysterectomy and bilateral salpingo-oophorectomy 21 10.6 Bilateral salpingo-oophorectomy 11 5.6 Right salpingo-oophorectomy 47 23.7 Left salpingo-oophorectomy 30 15.2 Right oophorectomy 30 15.2 Left oophorectomy 27 13.6 Unknown surgical procedure 32 16.2 Laterality Right 84 42.4 Left 66 33.3 Bilateral 15 7.6 Unknown laterality 33 16.7 Tumor size (in cm) Less than 5 16 8.1 5–10 60 30.3 10–20 103 52.0 More than 20 16 8.1 Tumor size cannot be assessed 3 1.5 Tumor consistency Cystic 144 72.7 Solid 22 11.1 Complex 31 15.7 Diffuse pattern (metastatic tumor) 1 0.5 Association with endometriosis Yes 2 1.0 No 196 99.0 Surgical Procedure, Laterality, Size and Consistency of Ovarian Neoplasms A considerable number of the neoplasms were seen in the right side (n=84, 42.4%), followed by left side (n= 66, 33.3%), bilateral (n=15, 7.6%); and for 33 (16.7%) cases, the tumor laterality was unknown, as shown in Table 2 . The tumor size was as small as 0.3 cm and as large as 35 cm and was categorized into four groups. Tumors less than 5 cm constituted 8.1% (n=16), 5 cm to 10 cm, 30.3% (n=60), 10 cm to 20 cm, 52% (n=103) and tumor size more than 20 cm formed 8.1% (n=16). In three of the cases (1.5%), the tumor had either diffuse pattern or the tumor found incidentally where their gross size could not be assessed, as shown in Table 2 . In consideration of tumor consistency, 72.7% (n=144) were cystic, 11.1% (n=22) solid, 15.7% (n=31) complex and in one case, tumor consistency showed diffuse pattern which could not be assessed grossly. Only one patient died of the disease and two patients had concomitant ovarian endometriotic cysts, a precursor lesion to ovarian cancer, as shown in Table 2 . Table 3 describes tumor type and frequency of ovarian neoplasms. Benign tumors were 81.8% (n=162), borderline 1.5% (n=3) and malignant 16.7% (n=33) cases. Table 3 Natural Behavior and Frequency of Ovarian Neoplasms Characteristics Frequency Percentage (%) Benign 162 81.8 Borderline 3 1.5 Malignant 33 16.7 Natural Behavior and Frequency of Ovarian Neoplasms As shown in Table 4 , the majority of the diagnosed cases were of surface epithelial in origin, consisting of 52% (n=103) of cases, followed by germ cell tumors 39.9% (n=79), sex cord stromal tumors 7.1% (n=14), and finally, there was only one metastatic tumor (0.5%) and a single poorly differentiated neoplasm (0.5%), for which, histologic classification was not possible without performing immunohistochemical stains. Out of 162 benign tumors, the most common diagnosis was mature cystic teratoma, constituting 34.8% (n=69) of cases, followed by serous cystadenoma 25.2% (n=50). Out of 33 malignant cases, serous cystadenocarcinoma, and adult granulosa cell tumors each comprised 3% (n=6), were the commonest malignant neoplasms followed by endometrioid adenocarcinoma 2.5% (n=5) and mucinous cystadenocarcinoma 2% (n=4). Table 4 Histologic Types, Classification and Diagnostic Category of Ovarian Neoplasms Classification Benign (n=162/81.8%) Borderline (n=3/1.5%) Malignant (n=33/16.7%) Total (%) Surface epithelial cells Serous cystadenoma (50/25.2%), mucinous cystadenoma (29/14.6%), benign Brenner tumor (2/1%) Mucinous borderline tumor (1/0.5%), serous borderline tumor (2/1%) Serous cystadenocarcinoma (6/3%), mucinous cystadenocarcinoma (4/2%), clear cell carcinoma (1/0.5%), malignant Brenner tumor (1/0.5%), endometroid adenocarcinoma (5/2.5%), adenocarcinoma (1/0.5%), squamous cell carcinoma arising in mature teratoma (1/0.5%) 103 (52) Germ cell Mature cystic teratoma (69/34.8%) stroma ovarii (4/2%) Dysgerminoma (3/1.5%), Immature teratoma (1/0.5%), malignant mixed germ cell tumor (2/1%) 79 (39.9) Sex cord stromal cells Fibroma (2/1%), fibro-thecoma (5/2.5%), leiomyoma (1/0.5%) Adult granulosa cell tumor (6/3%) 14 (7.1) Metastatic Krukenburg tumor (1/0.5%) 1 (0.5) Unclassified Malignant neoplasm (1/0.5%) 1 (0.5) Histologic Types, Classification and Diagnostic Category of Ovarian Neoplasms

A descriptive cross-sectional study was conducted to determine the histopathological patterns of ovarian neoplasms diagnosed at FMIC, Kabul, Afghanistan. The study was approved by the Ethical Research Committee (ERC) of FMIC (64-FMIC-ER-19) and 198 patients diagnosed with ovarian tumors were consecutively included in the study from July 2017 to August 2020. Formal consents were taken upon participation from the patients and a well-organized questionnaire was developed that consisted of all the variables with well-defined inclusion criteria for the study. The consents were taken from the patients’ legal guardians when they were younger than 18 years old or were not able to give consent due to cultural concerns. Histopathological reports were retrieved from medical records of the Department of Pathology of FMIC. All the cases were diagnosed through histopathological examination by qualified and credentialed histopathologists. Biopsy tissues were submitted in 10% formalin solution prior to further processing. The tumor specimens less than 3 cm in the greatest dimension were entirely submitted, however, the solid tumors larger than 3 cm were sliced in 1 cm intervals and random sections per 1 cm of the greatest dimension of the tumors were taken in separate blocks. Paraffin embedded tissue sections were stained with hematoxylin and eosin (H&E) and then the slides were examined under a microscope for diagnosis. The cases were diagnosed and subclassified in accordance with the fourth edition of WHO Classification of Female Genital Tumors. Statistical Package for Social Sciences (SPSS) application, version 25, was used for data analysis. Considering the type of data, frequencies and proportions were used for calculating categorical variables, while mean, median and standard deviation were used to describe continuous variables.

To the best of the researcher’s knowledge, this study was the first of its kind to be conducted in Afghanistan. Due to the decades of continuous war in the country, research is still in its infancy in Afghanistan; and the researchers do believe that the current study will pave the way for large scale studies in the future, regarding ovarian tumors. Despite the abovementioned strengths, the present study was conducted at a private tertiary care hospital setting; therefore, it extrapolated the study findings and hence the generalizability of study is questionable. Moreover, due to the lack of immunohistochemical stains, a considerable number of cases were excluded from the study which might have affected the internal validity of research.

The current study demonstrated significant information regarding demographic, histopathological patterns, and clinical characteristics of ovarian tumors in Afghanistan. The study revealed that ovarian neoplasms had nonspecific signs and symptoms, occurring mainly in reproductive age groups and the majority of them were benign. The proportion of malignant ovarian neoplasms was far less than benign ones. While many of the patterns related to ovarian neoplasms in Afghanistan were similar to the patterns demonstrated in other regions of the world, there were peculiar differences that need further elaboration. Furthermore, considering the study setting, our data represented only the tip of the iceberg and it is indispensable to conduct further multi-institutional studies, targeting larger sample size, for detailed scrutiny into the subject, that will eventually lead to the development of effective health policies.

The findings of our study were similar to studies conducted in international and regional countries and described the histopathological patterns of ovarian neoplasms in Afghanistan. In the current study, we attempted to fill an important literature gap regarding the patterns and proportions of ovarian neoplasms diagnosed based on histopathology in Kabul, Afghanistan. One hundred and ninety-eight subjects with histopathological diagnosis of ovarian neoplasms with an age range from 4 to 80 years were included in the study. The mean age at diagnosis was 34.4 years and most of the cases occurred between 21 and 60 years of age which was similar to studies carried out in Lahore, Pakistan, and in Ahmedabad, India. 12 , 13 The current study revealed that 81.8% of the neoplasms were benign, while only 16.7% malignant and 1.5% borderline tumors. This finding was in concordance with the studies conducted in western region of Saudi Arabia, Pakistan, and in eastern India. 14–16 In our study, surface epithelial tumors accounted for 52%, germ cell tumors were 39.9%, sex cord stromal tumors were 7.1% and metastatic tumors constituted only 0.5% of the cases. This finding was similar to studies conducted in Iraq, Pakistan and Iraq. 15 , 17 , 18 However, a study conducted in Ghana showed that germ cell tumors were slightly more than surface epithelium. 19 The most common benign neoplasm was mature cystic teratoma (34.8%). This finding was inconsistent with previous studies where serous cystadenoma was demonstrated to be the commonest benign tumor. 6 , 13 , 16 However, studies conducted in Nepal, Ghana and Nigeria showed that mature cystic teratoma was the commonest benign ovarian neoplasm. 19–21 The most common malignant tumor was serous cystadenocarcinoma and adult granulosa cell tumor, each constituting 3% of the all malignant cases followed by endometrioid adenocarcinoma (2.5%) and mucinous cystadenocarcinoma (2%) which apart from relatively higher incidence of adult granulosa cell tumor in our study, shows a concordance with the literature 20 , 22 that could be explained related to our study’s small sample size, genetic or environmental factors. In this study, information for laterality for 33 cases was not available. One-sided (75.7%) tumors were more common than bilateral (7.3%) and right ovarian involvement (42.4%) was more common than the left (33.3%), which was in concordance with findings of studies conducted in India, Iraq, and Bangladesh. 10 , 17 , 22 , 23 In terms of tumor consistency, 72.7% were cystic, 11.1% solid and 15.7% complex and only one case (0.5%) was not assessable due to diffuse pattern of the tumor. Tumors with cystic consistency were mostly benign constituting 88.3%, while malignant tumors were either solid or complex, each being 42.4% and 51.5%, respectively. These finding are in concordance with studies conducted in Bangladesh and India. 23 , 24 Our study revealed that 53.4% of the patients presented with abdominal mass, which was the most common presenting symptom followed by abdominal pain. These findings were in line with a study carried out in India, showing abdominal mass followed by abdominal pain as the major presenting symptoms in patients suffering from ovarian tumors. 6 Although the ovarian tumors were subclassified in accordance with fourth edition of the WHO Classification of Female Genital Tumors at the time of diagnosis in our study, the updated fifth edition comprises five principal histological types: high-grade serous carcinoma, low-grade serous carcinoma, mucinous carcinoma, endometrioid carcinoma, and clear cell carcinoma. 25 In addition to that, although the identification of TP53 mutation in high grade serous carcinomas and missense point mutation in FOXL2 in granulosa cell tumors is vital for confirmative diagnosis of these tumors, 26 , 27 the cases in our study were exclusively diagnosed on the basis of morphological findings with the help of H&E stains. No immunohistochemical and molecular studies were done for the diagnosis in the current study. In Afghanistan, research work is still in its infancy with no reliable tumor registry at national level that could provide a good picture of prevalence and distribution of various tumors and their possible linked factors to the environment, lifestyle and genetics of the Afghan population. 28 , 29 In addition, low level of education, poverty and lack of specialized diagnostic and therapeutic facilities for neoplastic disorders adds to predicaments of the issue. 30