Diagnostic accuracy of intraoperative haptoglobin point-of-care testing in ovarian tumor fluid versus frozen section for epithelial ovarian cancer.

Ovarian tumor fluid samples were collected intraoperatively using sterile techniques. The cystic component of the tumor was identified based on macroscopic appearance and aspirated using a sterile syringe while minimizing spillage. Haptoglobin levels were assessed using OvaCis ® Research Use Only (RUO) (INNEX Innovate), a rapid qualitative immunochromatographic point-of-care assay. Approximately 10 µL of ovarian tumor fluid was applied to the test cassette according to the manufacturer’s instructions. Results were read after 10 min. A positive result was defined by the appearance of both control (C) and test (T) lines, while a negative result showed only the control line. Frozen section examination was performed intraoperatively according to standard protocols. Pathologists interpreting frozen sections and final histopathology were blinded to haptoglobin results, and operators performing haptoglobin testing were blinded to histopathological outcomes. The ovarian tissue sample is then processed to histopathological examination following the standard operating procedures of the Anatomical Pathology Department of Dr. Hasan Sadikin General Hospital fixed in formalin, embedded in paraffin, sectioned at 4–5 μm, and stained with hematoxylin–eosin (H&E). Histopathological evaluation was performed by board-certified anatomical pathologists. Statistical analysis was performed using MedCalc Statistical Software. Continuous variables were analyzed using the independent t-test or Mann–Whitney U test, depending on data distribution. Categorical variables were analyzed using the chi-square test or Fisher’s exact test. Diagnostic accuracy parameters—including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the haptoglobin point-of-care test (Hp-POC) and frozen section were calculated against final paraffin histopathology. Paired comparisons between Hp-POC and frozen section were performed using McNemar’s test. 95% confidence intervals (95% CIs) were calculated using the Wilson score method. This study was approved by the Health Research Ethics Committee of Dr. Hasan Sadikin General Hospital/Faculty of Medicine Universitas Padjadjaran. Written informed consent was obtained from all participants prior to enrollment.

A total of 55 women with suspected malignant ovarian tumors were recruited between March and August 2023 at Dr. Hasan Sadikin General Hospital (Fig.  1 ). There were 11 patients were excluded: 3 due to borderline ovarian tumors and 8 due to non-epithelial ovarian malignancies. A total of 44 patients were included in the final analysis. Fig. 1 Flow diagram of patient recruitment, inclusion, and analysis Flow diagram of patient recruitment, inclusion, and analysis Baseline characteristics are summarized in Table  1 . There were no statistically significant differences in age, body mass index (BMI), or CA-125 levels between the two groups. However, the RMI-2 score was significantly higher in the malignant group compared to the benign group ( p  = 0.025). Table 1 Comparison between study subject characteristics in malignant and benign group Variable Histopathology P Value Malignant Benign Age (years) Mean ± SD Median Range (min-max) 51. 57 ± 13.35 53,00 48(30–78) 45.96 ± 15.499 46,00 58(20–78) 0,207 a BMI (kg/m 2 ) Mean ± SD Median Range (min-max) 25.26 ± 4,86 25,00 17.10(17.80–34.90) 24.72 ± 4,48 24,10 18.10(16.00-34.10) 0,703 a CA-125 (U/mL) Mean ± SD Median Range (min-max) 926.8733 ± 1153.424 600 4411 (42.40-4453.40) 166.7471 ± 154.82 110 585(14.80–600.00) 0,073 b RMI 2 (score) Mean ± SD Median Range (min-max) 9814.453 ± 10301.87 4732.80 31125.20(494-31619) 1145.17 ± 1491.16 442.40 4185.60 (16.00-4201.60) 0,025 b Note: P-value numerical data were determined based on the independent T- test (a) or Mann-Whitney test (b). Statistical significance was set at P  < 0.05. * Statistically significant difference ( P  < 0.05) Comparison between study subject characteristics in malignant and benign group Age (years) Mean ± SD Median Range (min-max) 51. 57 ± 13.35 53,00 48(30–78) 45.96 ± 15.499 46,00 58(20–78) BMI (kg/m 2 ) Mean ± SD Median Range (min-max) 25.26 ± 4,86 25,00 17.10(17.80–34.90) 24.72 ± 4,48 24,10 18.10(16.00-34.10) CA-125 (U/mL) Mean ± SD Median Range (min-max) 926.8733 ± 1153.424 600 4411 (42.40-4453.40) 166.7471 ± 154.82 110 585(14.80–600.00) RMI 2 (score) Mean ± SD Median Range (min-max) 9814.453 ± 10301.87 4732.80 31125.20(494-31619) 1145.17 ± 1491.16 442.40 4185.60 (16.00-4201.60) Note: P-value numerical data were determined based on the independent T- test (a) or Mann-Whitney test (b). Statistical significance was set at P  < 0.05. * Statistically significant difference ( P  < 0.05) Frozen section examination and histopathology showed perfect concordance, with sensitivity, specificity, PPV, and NPV of 100%. The OvaCis rapid haptoglobin test demonstrated high diagnostic performance, with a sensitivity of 90.48% (95% CI 69.62%–98.83%), specificity of 91.30% (95% CI 72.00%–98.90%), PPV of 90.48%, and NPV of 91.30%. There were 2 false-positive and 2 false-negative results identified in the haptoglobin test. False-positive results may be associated with elevated haptoglobin levels in benign inflammatory conditions, whereas false-negative results may reflect tumor heterogeneity or variable haptoglobin expression. Detailed diagnostic performance is presented in Table  2 . Table 2 Diagnostic analysis between frozen section and haptoglobin compared to histopathology as the gold standard Variable Histopathology P Value Malignant N (%) Benign N (%) Frozen Section Malignant Benign 21(100) 0(0) 0(0) 23(100) 0,000 Sensitivity : 100% (83,89%-100,00%) Specificity : 100% (85,18%-100,00%) Positive Predictive Value: 100% (83,89%-100,00%) Negative Predictive Value: 100% (85,18%-100,00%) Accuracy : 100% (91,96%-100%) Haptoglobin Malignant Benign 19(90,5) 2(8,7) 2(9,5) 21(91,3) 0,000 Sensitivity: 90,48% (69,62%-98,83%) Specificity: 91,30% (71,96%-98,93%) Positive Likelihood Ratio : 10,40(2,75 − 39,40) Negative Likelihood Ratio : 0,10 (0,003 − 0,39) Positive Predictive Value : 90,49% (71,50%-97,30%) Negative Predictive Value : 91,30% (73,63%-97,53%) Accuracy : 90,91% (78,33%-97,47%) Diagnostic analysis between frozen section and haptoglobin compared to histopathology as the gold standard Frozen Section Malignant Benign 21(100) 0(0) 0(0) 23(100) Sensitivity : 100% (83,89%-100,00%) Specificity : 100% (85,18%-100,00%) Positive Predictive Value: 100% (83,89%-100,00%) Negative Predictive Value: 100% (85,18%-100,00%) Accuracy : 100% (91,96%-100%) Haptoglobin Malignant Benign 19(90,5) 2(8,7) 2(9,5) 21(91,3) Sensitivity: 90,48% (69,62%-98,83%) Specificity: 91,30% (71,96%-98,93%) Positive Likelihood Ratio : 10,40(2,75 − 39,40) Negative Likelihood Ratio : 0,10 (0,003 − 0,39) Positive Predictive Value : 90,49% (71,50%-97,30%) Negative Predictive Value : 91,30% (73,63%-97,53%) Accuracy : 90,91% (78,33%-97,47%)

This study was a prospective observational analytical cross-sectional study conducted at Dr. Hasan Sadikin General Hospital, Bandung, Indonesia, between March and August 2023. Although conducted at a single tertiary referral center, this hospital serves as a major referral center, receiving a diverse patient population. All eligible subjects underwent intraoperative frozen section examination as part of clinical standards, followed by haptoglobin testing using a rapid point-of-care assay. The index test was the haptoglobin point-of-care assay, the comparator was frozen section examination, and the reference standard was final paraffin histopathology. Both tests were evaluated independently. Sample size was calculated using Buderer’s formula for diagnostic test studies based on expected sensitivity and specificity. Assuming an expected sensitivity of 95%, specificity of 90%, disease prevalence of 40%, α = 0.05, and desired precision of 10%, the minimum required sample size was 45 subjects. To account for possible dropouts or incomplete data, 10% was added, resulting in a final target sample size of 50 subjects. Patients with suspected malignant ovarian tumors who were scheduled for cystectomy, salpingo-oophorectomy, or total hysterectomy with salpingo-oophorectomy via laparotomy at Dr. Hasan Sadikin General Hospital were consecutively recruited. Laparotomy was selected as the surgical approach because it allowed safe intraoperative tumor handling and controlled aspiration of cyst fluid while minimizing the risk of spillage, which is critical for both oncologic safety and accurate sampling. Suspicion of malignancy was based on on preoperative assessment using the IOTA (International Ovarian Tumor Analysis) criteria and/or a Risk of Malignancy Index (RMI) score > 200. All included patients underwent haptoglobin testing, frozen section examination, and final histopathological assessment. No indeterminate test results were observed. Patient recruitment was conducted by certified gynecologic oncologists. Final inclusion was determined postoperatively based on histopathological confirmation. Only patients with epithelial ovarian tumors (benign or malignant) were included. Demographic and clinical data—body mass index (BMI), comorbidities, family history of cancer, menopausal status, preoperative CA-125 levels, and RMI scores—were documented. Exclusion criteria included non-epithelial malignancies, borderline ovarian tumours, dermoid cysts, pelvic inflammatory disease, tumors without cystic components, samples unsuitable for analysis (e.g., visibly blood-contaminated or too viscous for pipetting), and prior neoadjuvant chemotherapy followed by interval debulking surgery. Borderline tumors were excluded due to their intermediate biological behavior and potential to confound diagnostic classification.

Haptoglobin point-of-care testing using ovarian tumor fluid demonstrated high diagnostic accuracy for differentiating benign and malignant epithelial ovarian tumors in this cohort. The test may provide a rapid and practical adjunctive intraoperative diagnostic option, particularly in settings where frozen section services are limited. However, histopathological examination remains the reference standard, and larger multicenter studies are needed to confirm these findings and define its role in routine clinical practice.

This study demonstrated that haptoglobin point-of-care testing using intraoperatively aspirated ovarian tumor fluid has high diagnostic accuracy for differentiating benign and malignant epithelial ovarian tumors, with sensitivity and specificity exceeding 90%. Importantly, these findings highlight the potential of a rapid, intraoperative, and minimally resource-dependent diagnostic tool that may complement frozen section analysis, particularly in settings with limited access to pathology services. Ovarian cancer is typically age-related malignancy, with a higher incidence in women over 65 years. However, it may also occur in younger women. In this study, the mean age of patients diagnosed with malignant tumors was 51.57 years, which is comparable to previous studies by Sumeya et al. and Zheng et al [ 14 , 15 ]. No significant age difference was observed between benign and malignant groups, suggesting that age alone may not reliably distinguish tumor behavior in certain populations, although some studies have reported higher malignancy rates in older women [ 16 ]. The relationship between obesity and ovarian cancer remains controversial. Several epidemiological studies have suggested that excess adiposity may increase ovarian cancer risk through chronic low-grade inflammation, hyperinsulinemia, altered adipokine signaling, and increased peripheral estrogen production [ 17 ]. However, in the present study, BMI did not differ significantly between benign and malignant groups. This may reflect the multifactorial nature of ovarian carcinogenesis, the modest sample size, and the limited utility of BMI alone as a discriminator between benign and malignant ovarian tumors. The diagnostic performance observed in this study is consistent with previous research evaluating haptoglobin as a biomarker using ELISA and other rapid detection methods [ 11 ]. However, unlike prior studies, this study specifically assessed haptoglobin in ovarian tumor fluid using a point-of-care approach in an intraoperative setting, thereby providing more direct clinical applicability. This distinction strengthens the potential translational value of the findings. Although frozen section demonstrated perfect concordance with histopathology in this study, previous literature has reported variable diagnostic accuracy, particularly in borderline tumors [ 7 ]. Frozen section analysis is also dependent on pathologist expertise, subject to sampling error, and requires processing time that may prolong surgical procedures. In contrast, haptoglobin point-of-care testing offers a rapid and simpler alternative, with minimal infrastructure requirements, making it particularly attractive for resource-limited environments. The occurrence of false-negative and false-positive results in the haptoglobin test warrants consideration. False-negative results in this study may be explained by tumor heterogeneity, particularly in multiloculated tumors where cystic fluid from a single compartment may not represent the entire tumor [ 18 ]. Conversely, false-positive results may be associated with inflammatory conditions such as endometriosis, in which increased levels of proinflammatory cytokines (e.g., IL-6) can stimulate haptoglobin expression. Additionally, haptoglobin-like proteins reported in endometriosis, historically described as ENDO-I, may contribute to cross-reactivity in certain cases [ 19 ]. Clinically, these findings suggest that haptoglobin point-of-care testing may serve as a complementary intraoperative diagnostic tool to support surgical decision-making. Its rapid turnaround time and ease of use may help reduce operative delays and improve workflow efficiency. However, this test is not intended to replace histopathological examination, but rather to function as an adjunct, particularly in settings where frozen section is unavailable or limited. This study has several strengths. It evaluates a novel application of haptoglobin testing directly in ovarian tumor fluid using a point-of-care approach, reflecting real-world intraoperative conditions. Additionally, the use of histopathology as the reference standard and the implementation of blinding between the index and reference tests strengthen the internal validity of the findings. Several limitations should be acknowledged. First, it was conducted in a single center with a relatively small sample size, which may limit generalizability and precision. Second, borderline tumors were excluded, which may restrict the applicability of findings across the full spectrum of ovarian neoplasms. Third, the study did not include direct comparative diagnostic performance with other biomarkers such as CA-125 or IOTA models. Fourth, some diagnostic accuracy estimates showed relatively wide confidence intervals, reflecting the limited sample size. Future multicenter studies with larger sample sizes and inclusion of borderline tumors are needed to validate these findings and further define the role of haptoglobin testing in clinical practice.

Ovarian cancer is one of the leading causes of cancer-related death in women worldwide, affecting more than 200,000 women annually [ 1 ]. Approximately 90% of malignant ovarian tumors are of epithelial origin, arising from the ovarian surface epithelium or the epithelium of the fallopian tubes [ 2 ]. Its high mortality is largely due to late-stage diagnosis, as most patients present with advanced disease [ 2 ], [ 3 ] Early diagnosis remains challenging, as initial symptoms are vague and nonspecific, including abdominal discomfort, bloating, urinary urgency, and early satiety [ 2 ]. Several preoperative models have been developed to improve diagnostic accuracy, such as the Risk of Malignancy Index (RMI), which combines CA-125 levels, ultrasound findings, and menopausal status, and ultrasound-based models developed by the International Ovarian Tumor Analysis (IOTA) group [ 4 , 5 ]. Other biomarkers and algorithms, such as human epididymis protein 4 (HE4) and the Risk of Ovarian Malignancy Algorithm (ROMA), have also been introduced. However, these modalities have limitations. Serum CA-125 lacks specificity and may be elevated in several benign conditions, including endometriosis and pelvic inflammatory disease [ 6 ]. Intraoperatively, frozen section examination is widely used in routine practice for distinguishing benign, borderline, and malignant ovarian tumors. Its reported overall accuracy ranges from approximately 86% to 97%, with higher performance for benign and malignant tumors, but lower diagnostic performance for borderline tumors. [ 7 ] Furthermore, frozen section analysis is time-consuming, requires specialized pathology expertise, and may not be readily available in resource-limited settings. Sampling errors and tissue processing artifacts can also affect diagnostic accuracy. [ 8 ]. Given these limitations, there is a need for a rapid, reliable, and accessible intraoperative diagnostic tool. Haptoglobin (Hp) is an acute-phase protein involved in binding free hemoglobin and preventing oxidative damage, has emerged as a potential biomarker in various malignancies, including ovarian cancer. Previous studies have demonstrated elevated haptoglobin levels in ovarian tumor fluid and suggested its potential utility in differentiating benign, borderline, and malignant tumors [ 9 , 10 ]. Several methods are available to detect haptoglobins, such as surface-enhanced Raman spectroscopy (SERS) and rapid colorimetric assays (RCA), which can be used for intraoperative point-of-care testing. Both tests had good sensitivity and specificity, 94% and 91% for SERS and 97,3% and 92% for RCA, respectively [ 11 – 13 ]. However, these approaches are not yet widely implemented in routine clinical practice, particularly in intraoperative settings. To date, limited studies have evaluated the clinical applicability of haptoglobin point-of-care testing directly in ovarian tumor fluid as a rapid intraoperative diagnostic tool, particularly in comparison with frozen section analysis. Therefore, this study aims to evaluate the diagnostic accuracy of a haptoglobin point-of-care test in ovarian tumor fluid compared with frozen section for epithelial ovarian cancer, and to assess its potential role as a complementary intraoperative diagnostic modality.