{"paper_id":"e565bb85-1975-4e68-b954-fa8234cd3a19","body_text":"Ovarian tumours represent a wide range of neoplastic conditions due to apoplastic epithelial, stromal and germ-cell components, with\neach having its own biological, behavioural and prognostic characteristics [ 1 ]. Ovarian cancer is\na major health concern in the world and it is the second most common female genital tract malignancy [ 2 ].\nIts high mortality rate is greatly explained by the delayed diagnosis because most patients remain asymptomatic or have an unspecified\nabdominal symptom until the disease is advanced, forming in its advanced period. Ovarian cancer is the fifth most frequent cause of\ncancer mortality in women, with an annual global incidence of 239,000, resulting in approximately 152,000 deaths. The lifetime risk of\ndeveloping ovarian cancer is 1 in 75 and the lifetime risk of dying is estimated at 1 in 100 [ 2 ].\nThe stage of the disease upon diagnosis has a great impact on prognosis. The five-year survival rates are high in early-stage disease\n(stage I), with more than 80% survival, but decrease significantly to less than 30% when women have advanced disease (stages III-IV)\n[ 3 ]. The clinical difficulty is presented by a deep location of the tumour in the pelvis, the\nlack of early warning signs and the similarity of gastrointestinal or gynecologic symptoms, including bloating, abdominal pain and early\nsatiety [ 4 ]. As a result, preoperative diagnosis is usually incidental or delayed and it is\nnecessary to have effective and convenient tools for preoperative diagnosis. The first-line imaging modalities used to examine adnexal\nmasses include transabdominal and transvaginal ultrasonography [ 5 ].\nFacets of morphology, such as mixed echogenicity, septations, solid components, papillary projections and ascites, which can help\ndifferentiate between malignant and benign pathology, can be evaluated using ultrasound. Despite the fact that ultrasonography is\nnoninvasive, it is widely available and inexpensive; its diagnostic value can be affected by the experience of the operator and\nmisinterpretations [ 4 ,  6 ]. The most used tumour marker in\nthe assessment of ovarian cancer is serum CA-125, a high molecular weight glycoprotein that is a byproduct of the coelomic epithelium.\nHigh levels of CA-125 have been widely linked with epithelial ovarian malignancies and can assist in the risk stratification of\nmalignancies. Although useful, CA-125 lacks cancer specificity and can be released in benign tumours, including endometriosis, pelvic\ninflammatory disease and benign cysts, which lead to low specificity [ 7 ]. Nonetheless, studies\nhave shown that when CA-125 is used together with ultrasound results, the accuracy of diagnosis is significantly increased, particularly\nin postmenopausal women who present with suspicious lesions [ 8 ]. The integrative approach not only\nmaximises the diagnosis process but also can inform the process of clinical decision-making and surgical planning. Because of the\nconstraints of individual diagnostic methods and the practical use of tools in the low-resource environment, sonographic evaluation in\nconjunction with serum CA-125 could be the best way to maximise preoperative prediction of benign and malignant ovarian tumours.\nTherefore, it is of interest to show the diagnostic efficacy of ultrasonography and CA-125 individually and in combination with each\nother in distinguishing ovarian tumors by using histopathology as the reference standard.\n\nThis analytical cross-sectional study was conducted in the Department of Gynaecology and Obstetrics at Rangpur Medical College and\nHospital, Bangladesh. The study was carried out over 12 months from January 2017 to December 2017. A total of 62 women clinically\ndiagnosed with ovarian tumours were included. Following surgical intervention and histopathological assessment, participants were\ncategorised into two groups: Group I (benign ovarian tumours, n=48) and Group II (malignant ovarian tumours, n=14).\n\n[1] Patients with suspected ovarian tumor diagnosed by history and clinical examination.\n[2] Patient diagnosed with an ovarian tumour based on imaging (USG & /or CT scan).\n\n[1] Previously diagnosed and treated ovarian tumour (recurrent cases and patients on chemotherapy).\n[2] Patients did not have an ultrasonography or serological testing of CA 125 preoperatively.\n[3] Patients without histopathological examination.\n\nEthical approval was obtained from the institutional review committee and permission was granted by the Department of Gynaecology and\nObstetrics. After providing detailed information about the objectives, risks and benefits, written informed consent was obtained from\nall participants. Data collection included structured interviews for demographic and clinical information, general and pelvic examinations\nand transabdominal ultrasonography assessing consistency, septation, papillary projections and ascites. Serum CA-125 estimation was\nperformed using a cut-off value of 35 U/ml. Patients then underwent definitive surgical management and histopathological examination\nserved as the diagnostic gold standard. Statistical analysis was conducted using SPSS version 23.0. Continuous variables were presented\nas means and standard deviations and compared using the Student t-test, while categorical variables were analysed using chi-square tests.\nDiagnostic indices, including sensitivity, specificity, accuracy, positive predictive value and negative predictive value, were calculated\nfor ultrasonography, CA-125 and the combined approach. Confidentiality was properly maintained throughout the study.\n\nWomen with malignant ovarian tumours are older on average (49.14±16.59 years) than those with benign tumours (35.85±11.20\nyears) and this age difference is statistically significant (t=3.48, p=0.001). Use of oral contraceptive pills, ovulation-inducing drugs\nand family history did not differ significantly between benign and malignant groups (all p>0.05), suggesting these risk factors were\nnot discriminative in this sample.  Table 1  presents the demographic and baseline clinical\ncharacteristics of women diagnosed with benign and malignant ovarian tumours, including age distribution and key reproductive risk\nfactors. Out of 62 ovarian tumour cases, 48 (77.41%) are benign and 14 (23.59%) are malignant, indicating that benign tumours constitute\nthe majority in this hospital-based series.  Figure 1  illustrates the overall distribution of benign\nand malignant ovarian tumours confirmed by histopathology in the study population. A total of 62 cases are included in this study. Among\nthem 48 (77.41%) are benign and 14 (22.59%) are malignant. Ultrasonography correctly identifies malignancy with a sensitivity of 57.42%\nand a specificity of 89.50%, showing it is much better at ruling out malignancy than detecting all malignant cases. The overall\ndiagnostic accuracy of USG is 82.17%, with a positive predictive value (PPV) of 61.84% and a negative predictive value (NPV) of 87.42%,\nindicating that a negative USG finding is relatively reliable, but positive findings still carry a notable false positive rate.\nTable 2  shows the validity of ultrasonography in identifying benign and malignant ovarian\ntumours, based on true-positive, false-positive, false-negative and true-negative results. Serum CA-125 (cut-off 35 U/ml) shows higher\nsensitivity (71.42%) and specificity (93.75%) than USG alone, indicating better ability to detect malignancy and to correctly identify\nbenign cases. CA-125 achieves an accuracy of 88.70%, with a PPV 76.92% and an NPV 91.83%, suggesting it is a stronger single test than\nUSG for differentiating benign from malignant ovarian tumours in this cohort.  Table 3  summarises\nthe diagnostic performance of serum CA-125 levels in detecting ovarian malignancy using a cut-off value of 35 U/ml. When USG findings\nare combined with CA 125, sensitivity increases further to 78.57% and specificity remains high at 93.75%, improving detection of\nmalignant cases while still accurately classifying benign lesions. The combined approach achieves the highest diagnostic accuracy of\n91.32%, with both PPV and NPV at 78.57% and 93.75%, respectively; demonstrating that integrating imaging with a tumour marker provides\nthe most reliable diagnostic performance among the evaluated strategies.  Table 4  demonstrates the\nimproved diagnostic accuracy achieved when ultrasonography findings are interpreted together with serum CA-125 levels.\n\nThe present study evaluated the diagnostic performance of ultrasonography and serum CA-125, individually and in combination, in\ndifferentiating benign from malignant ovarian tumours, using histopathology as the reference standard. The findings highlighted important\ndiagnostic patterns that align with and in several aspects of previously published research. The mean age of the malignant ovarian\ntumours was considerably more than that of the women who had benign tumours. This age distribution is consistent with findings reported\nby Mahale and others, who also noted a higher mean age in women with malignant tumours of the ovaries, indicating that age progression\nis one of the most important epidemiologic factors in the risk of ovarian cancer [ 5 ]. These same\ntrends were observed in the study of Pegu  et al.  especially in postmenopausal women who had adnexal mass surgery\n[ 9 ]. These observations support the overall epidemiological assumption that the rate of increasing\nprevalence of ovarian cancer with age is steadily increasing, probably due to increased genomic instability and an increasing number of\nyears of exposure to environmental risk factors [ 10 ]. Even though there is no significant difference\nin the use of oral contraceptive pills in the current research, prior pooled analyses have revealed protective associations. This trend\nis condoned by large-scale analyses of biomarker algorithms, which educated contraceptive use into the risk stratification systems, owing\nto its long-term protection [ 11 ]. The non-significant difference in the current data could be due\nto a small sample size, but not to the biological insignificance of the effect. In terms of clinical presentation, the malignant cases\nexhibited more abdominal distension and major weight loss. The findings are in line with case-based and clinical cohort reports of malignant\novarian tumours as more probable to manifest systemically through ascites, tumour burden, or endocrine changes [ 12 ,\n 13 ]. Abdominal pain was present in both benign and malignant conditions, meaning that it is\nnonspecific, which is also found in the literature, which states the overlapping appearances of symptoms in adnexal masses\n[ 14 ]. Ultrasonographic morphology was important in the differentiation of tumour type. Malignant\ntumours more often had complicated aspects such as mixed cystic-solid areas, septations, papillary projections and ascites. These\ncharacteristics are very much consistent with the high-risk sonography findings described in the IOTA Simple Rules and O-RADS systems,\nboth of which are regularly characterised by the occurrence of irregular solid structures and multi-locular masses as predictors of\nmalignancy [ 7 ]. In the study, the sensitivity of ultrasonography was 57.42 and the specificity\nwas 89.50. Although the specificity is concordant with that of colour Doppler-enhanced sonography, the sensitivity was lower and this\ncould be attributed to the use of transabdominal instead of transvaginal imaging, as mentioned by Mahale  et al.  who\nfound higher sensitivity figures using enhanced Doppler evaluation [ 5 ,  10 ].\nThe variation can be due to the quality of a machine, experience and transabdominal over transvaginal ultrasound, which is normally more\nresolute. The mean difference in serum CA-125 levels was significant between the malignant group and normal cases and the elevation of\nserum CA-125 levels was significant in malignant cases. This diagnostic use of CA-125 reported in the current results is in line with\nprevious large cross-sectional and case-series studies, as they are uniformly reporting moderate sensitivity but high specificity of the\npresent marker in epithelial ovarian malignancy [ 6 ,  15 ].\nLimitations associated with CA-125 were also apparent, as there are some cases of benign occurrences that demonstrated high levels.\nFalse-positive increases such as these have been widely reported in noncancerous gynaecological conditions, such as functional cysts,\nbenign lesbians where leiomyomas, endometriosis and hormonally active lesions of the ovaries [ 12 ,\n 16 ]. A combination of the ultrasonographic morphology and CA-125 to achieve the best diagnostic\nperformance, the sensitivity improved to 78.57 percent and the specificity was 93.75 per cent. This improved accuracy is consistent with\nrecent multimodal diagnostic studies, reports of studies that combine CA-125 with new biomarkers or algorithmic models like ROMA and\nCPH-I [ 3 ,  17 ]. Similarly, Agrawal & Gunjan reported\nthat when ultrasound findings suggestive of malignancy combined with elevated CA-125 levels, the specificity increased to 88%, with a\npositive predictive value of 92% and a negative predictive value of 85%. This combined approach significantly improved diagnostic\naccuracy, reducing the number of unnecessary surgeries for benign conditions [ 18 ]. Even though\nmore recent biomarkers, including HE4 and TK1 protein, might be more effective on their own, several comparative studies suggest that\nCA-125 still has much potential when used in combination with ultrasound, especially when resources to more sophisticated assays are\nstill limited [ 11 ,  17 ]. The combination methodology\novercomes the shortcomings of each modality and the ultrasound is added, providing spatial and morphological evaluations and CA-125 is\nused to corroborate the overall diagnostic finding [ 19 ]. The ROC analysis in this study also\nsupported the usefulness of CA-125, with an AUC of 0.931, which is statistically excellent in terms of discriminative ability. Similar\nAUCs have been achieved in algorithm tests of ovarian cancer detection based on biomarkers, which supports the clinical significance of\nCA-125 despite the advent of novel markers [ 3 ,  19 ]. The\nhistopathological distribution in the present study, where serous cystadenoma is the most frequent benign tumour and serous\ncystadenocarcinoma is the most widespread malignant form, fits the conventional patterns in ovarian pathology. The dominance of\nepithelial tumours is similarly reported in experimental and clinical studies, which supports the validity of the current results in the\nextended spectrum of epidemiology [ 20 ,  21 ]. The study\noverall supports the clinical utility of both CA-125 and ultrasonography in preumber of ovarian tumours, although it suggests a much\nbetter diagnostic accuracy when the two are used together. Despites the fact that ultrasonography is an important method because it is\naccessible and specific CA-125 has an important biochemical dimension. The results confirm the previous evidence and highlight the\nnecessity of considering both morphological and biochemical measurements in daily practice, especially in settings where sophisticated\nimaging technologies cannot be regularly used. There were some limitations in the study. The study was conducted in a single centre with\na small sample size, which may limit the generalizability of the findings. Only transabdominal ultrasonography was used; more advanced\nmethods such as transvaginal or Doppler studies were not available. Other tumour markers besides CA-125 could not be assessed due to\nlimited laboratory resources, restricting broader diagnostic comparison.\n\nThere was high specificity with ultrasonography and increased sensitivity with CA-125, with diagnostic accuracy being improved\nsignificantly with a combination of both. Both modalities assessed histopathology, which supported their importance in preoperative\nassessment. The integrated method provides a combination of a viable and convenient technique for early detection of malignant ovarian\ntumours in the resource-limited environment.","source_license":"CC-BY-4.0","license_restricted":false}