Diagnostic Value of PET/CT for Ovarian Cancer Recurrence or Metastasis in Postoperative Patients With Elevated Serum CA125 Levels: A Systematic Review and Meta-Analysis.

Zuowei Zou: data curation (equal), formal analysis (equal), investigation (equal), methodology (equal), resources (equal), software (equal), validation (equal), visualization (equal), writing – original draft (equal). Luhua Xia: data curation (equal), formal analysis (equal), funding acquisition (equal), investigation (equal), methodology (equal), resources (equal), software (equal), validation (equal), writing – original draft (equal). Saikang Tang: data curation (equal), formal analysis (equal), investigation (equal), methodology (equal), resources (equal), software (equal), validation (equal), visualization (equal), writing – original draft (equal). Lin Lin: formal analysis (equal), investigation (equal), software (equal), validation (equal). Quanyang Wu: formal analysis (equal), investigation (equal), software (equal), validation (equal). Donghui Hou: formal analysis (equal), investigation (equal), methodology (equal), validation (equal). Shijun Zhao: conceptualization (equal), data curation (equal), funding acquisition (equal), project administration (equal), resources (equal), supervision (equal), visualization (equal), writing – review and editing (equal).

The authors have nothing to report.

The authors have nothing to report.

A literature search was conducted in PubMed, Embase, Cochrane Library, Web of Science, and Chinese databases (CNKI, VIP database, Wan Fang Data, CBM) for studies published from January 1, 2000, to December 31, 2023. The following search terms were used: ovarian cancer subject terms + free words connected by OR, CA125 subject terms + free words connected by OR, PET/CT subject terms + free words connected by OR, and finally AND to connect the three for searching. The reference lists in the identified articles were manually searched to locate other relevant publications. The search approach for Chinese literature was comparable to that for English papers, with the publication language limited to English or Chinese. The search strategy was as follows: Step 1: Ovarian cancer subject terms + free words concatenated with OR, search formula I (ovarian neoplasms OR neoplasm, ovarian OR ovarian neoplasm OR ovary neoplasms OR neoplasm, ovary OR neoplasms, ovary OR ovary neoplasm OR neoplasms, ovarian OR ovary cancer OR cancer, ovary OR cancers, ovary OR ovary cancers OR ovarian cancer OR cancer, ovarian OR cancers, ovarian OR ovarian cancers OR cancer of ovary OR cancer of the ovary). Step 2: CA125 subject words + free words joined by OR, search formula II (CA‐125 antigen OR antigen, CA‐125 OR antigen CA‐125 OR antigen CA 125 OR CA 125 antigen OR antigen, CA 125). Step 3: PET/CT subject words + free words joined by OR, search formula III (positron emission tomography computed tomography OR PET‐CT scan OR PET‐CT scans OR scan, PET‐CT OR scans, PET‐CT OR PET CT scan OR CT scan, PET OR CT scans, PET OR PET CT scans OR scan, PET CT OR scans, PET CT OR CT PET OR positron emission tomography‐computed tomography OR PET‐CT OR CT PET scan OR CT PET scans OR PET scan, CT OR PET scans, CT OR scan, CT PET OR scans, CT PET). Step 4: An additional search was performed as follows: search formula I AND search formula II AND search formula/span > III. A comprehensive search was conducted for gray literature. No articles meeting the inclusion/exclusion criteria were found. Our meta‐analysis was carried out in accordance with the Preferred Reporting Items for Systematic Review and Meta‐analysis (PRISMA) guidelines. A PROSPERO registration number was obtained for this study: CRD42023472702. The inclusion criteria were as follows: (1) ovarian cancer was identified through pathological analysis of surgical specimens; (2) clinical suspicion of ovarian cancer recurrence or metastasis; (3) cases that underwent both serum CA125 examination and 18 F‐FDG PET/CT examination; (4) cases showed elevated CA125 levels; (5) studies that included cases with both normal and elevated CA125 levels, with the elevated CA125 group included in the analysis; (6) diagnostic studies that provided data on TP, FP, FN, and TN results; (7) studies with clear criteria for follow‐up of recurrence or metastasis; and (8) studies with at least 15 cases. Exclusion criteria were as follows: (1) cases without pathologic confirmation of ovarian cancer; (2) case studies with normal serum CA125 levels; (3) cases with missing 18 F‐FDG PET/CT or CA125 examination results; (4) there is no clear pathological or clinical follow‐up standard for the recurrence or metastasis of the included cases; (5) literature in which TP, FP, FN, and TN could not be directly or indirectly obtained; (6) reviews, conference abstracts, case reports; (7) duplicate studies; (8) the full‐text and incomplete data could not be acquired; (9) the study was a non‐diagnostic experiment; (10) studies with fewer than 15 cases; and (11) animal and cellular studies. Recurrence or metastasis was confirmed by the following: (1) pathologic findings, for instance, reoperation, biopsy, or cytology of ascites; or (2) clinical signs of relapse, such as positive routine imaging or increasingly elevated CA‐125 concentrations during the 6‐month follow‐up interval. Two independent examiners collected data from the studies using a pre‐designed form. Disagreements were addressed through discussion. The following information was collected from each study: first author, year of publication, patient age range, preoperative staging, and TP, FP, FN, and TN results. All the included studies were evaluated for methodological quality by two authors applying the QUADAS‐2 tools. If there was disagreement between the two authors, a third author adjudicated, and the final results were determined by consensus. None of the authors participated in the incorporated research. The QUADAS‐2 checklist mainly covers four aspects: patient screening, indicator test, reference criterion, and flow and timing. The above dimensions were evaluated for risk of bias based on several questions using the Review Manager 5.4 software. Sensitivity (SEN), specificity (SPE), 95% confidence intervals (CI), positive likelihood ratios (+LR), negative likelihood ratios (−LR), and diagnostic odds ratios (DOR) were obtained from the studies and then pooled to assess diagnostic veracity. Summary receiver operating characteristic (SROC) lines were formed, and the area under the receiver operating characteristic curve (AUC) was computed to determine the reliability of the diagnosis. Inter‐study heterogeneity was analyzed using the index of inconsistency ( I 2 ) statistic, a value calculated from the Q ‐statistic of the chi‐square test (25%–50%, 51%–75%, and > 75% for low, moderate, and high heterogeneity, respectively). I 2  > 50% was considered obvious heterogeneity using a stochastic‐effects model, and I 2  < 50% was used in a fixed‐effects model using the Meta‐Disc 1.4 software. Sensitivity analysis, publication bias test, and clinical utility evaluation were conducted using STATA 14.0 software.

The study selection process is shown in Figure  1 . A preliminary search identified 899 records, 280 of which were excluded after screening of titles and duplicates, including 239 duplicate studies, 40 reviews and reports, and one animal study. Abstract screening led to the exclusion of 539 records. A further 67 articles were excluded following a comprehensive full‐text evaluation. Finally, 13 studies were chosen for the descriptive analysis. The study selection process. The detailed characteristics of the 13 studies are listed in Table  1 . The 13 studies included 421 female patients, with an age range of 27–85 years. Among the 13 studies, seven were in Chinese, and six were in English. Summarized data of the 13 included studies. Note: Methods of follow‐up for recurrence or metastasis: (1) postoperative metastasis; (2) postoperative pathological. The clinical and other imaging data were followed up. Abbreviations: FN, false negative; FP, false positive; TN, true negative; TP, true positive. Literature quality was appraised using the QUADAS‐2 tool in Review Manager 5.4 software. The results indicated that all studies were of comparatively high caliber. In addition, the criterion standard of 13 studies, all of which followed up with recurrence or metastasis, did not accept the same gold standard, so there was a high concern about follow‐up. Some studies did not clearly describe whether they included consecutive or randomized cases, so there was a concern regarding patient selection. There was no concern regarding any of the index tests. Overall, from the results of the QUADAS‐2 assessment, the 13 included studies were considered suitable for inclusion in the meta‐analysis (Figure  2A,B ). Assessment of literature quality: (A) methodological quality graph; (B) methodological quality summary. Heterogeneity of threshold impact: After analysis, the Spearman association parameter between sensitivity log and (1‐specificity) log was 0.366. This indicated that there was no dissimilarity from the threshold impact in this study. Heterogeneity of non‐threshold: The Cochran‐Q test of DOR showed that Cochran‐Q = 1.90, p  = 0.9654 > 0.001, indicating that there was no heterogeneity that resulted from non‐threshold influence in this study. The I 2 % of DOR was less than 50%, and the fixed effect model was used to integrate the above five effects. The Galbraith plot, generated using STATA 14.0 software, showed that all studies were positioned both sides of the red effect line, implying that there was no heterogeneity in this study. The forest plot showed that the sensitivity I 2  = 16.4, specificity I 2  = 0%, positive likelihood ratio I 2  = 0%, and negative likelihood ratio I 2  = 0% (Figure  3A–E ). Using a fixed‐effects model, the pooled sensitivity was 0.94 (95% CI: 0.91–0.97), pooled specificity was 0.83 (95% CI: 0.71–0.91), pooled positive likelihood ratio was 4.59 (95% CI: 2.81–7.51), pooled negative likelihood ratio was 0.09 (95% CI: 0.05–0.15), pooled DOR was 64.22 (95% CI: 27.21–151.57), and AUC was 0.9379. The results of the meta‐analysis are summarized in Table  2 . Meta‐analysis diagram: (A) the forest plot for sensitivity; (B) the forest plot for specificity; (C) the plot for the positive likelihood ratio; (D) the plot for the negative likelihood ratio; (E) the plot for the DOR; (F) the SROC curve. Meta‐analysis data results. Abbreviations: DOR, diagnostic odds ratio; LR, likelihood ratio [ 31 ]. Observing Analysis of the SROC curve demonstrated that the AUC was 0.9379 (Figure  3F ). This indicates that the veracity of 18 F‐FDG PET/CT in detecting relapse or dissemination of ovarian cancer cases with postoperative CA125 elevation was as high as 94%. This demonstrates that 18 F‐FDG PET/CT has a high diagnostic veracity for this population. Sensitivity analysis was conducted using STATA 14.0 software (Figure  4A ). None of the 13 included studies significantly influenced the pooled results, demonstrating that the outcomes of the meta‐analysis are comparatively steady. Sensitivity analysis and publication bias chart: (A) sensitivity analysis chart; (B) publication bias test chart. Deeks' publication bias examination was conducted using STATA 14.0 software (Figure  4B ). The funnel plot displayed that the studies were symmetrically scattered on both sides of the regression line, indicating that there was no publication bias in this study ( p  = 0.07). Fagan's nomogram was used to characterize the ability of 18 F‐FDG PET/CT in detecting relapse or dissemination in cases with elevated serum CA125 levels after surgery for ovarian cancer. When the pre‐test odds were set at 20%, the post‐test odds rose to 58% with a positive likelihood ratio of 5. This indicates that an affirmative 18 F‐FDG PET/CT result is five times more probable to occur in a case with ovarian cancer than in a healthy woman. Conversely, a negative 18 F‐FDG PET/CT result would decrease the probability of metastatic relapse to 2% with a negative probability ratio of 0.07. This indicates that 18 F‐FDG PET/CT is an effective test for detecting dissemination and recurrence in patients with ovarian cancer who have elevated CA125 after surgery.

Our meta‐analysis provides a comprehensive summary of the diagnostic effect of 18 F‐FDG PET/CT for identifying recurrence or metastasis in postoperative ovarian cancer cases with elevated serum CA125 concentrations. The pooled sensitivity, specificity, and veracity were 94%, 83%, and 94%, respectively. These findings suggest that 18 F‐FDG PET/CT is a significant imaging technique for monitoring recurrence or metastasis in this population and provides valuable information for clinical decision‐making. The CA125 detection index is a commonly used serum indicator in clinical practice for the follow‐up of ovarian cancer cases; however, CA125 levels may also increase in the presence of benign conditions such as peritoneal inflammatory lesions and endometriosis. Therefore, while CA125 is included as a routine screening item for postoperative metastasis or recurrence of ovarian cancer, it lacks specificity and has a substantial rate of false positives [ 31 , 32 , 33 , 34 ]. Sun et al. [ 35 ] investigated the utility of CA125 levels (> 35 U/mL) in identifying ovarian cancer relapse during case follow‐up. The sensitivity, specificity, and accuracy were 77.78%, 86.67%, and 52.00%, respectively. The positive and negative predictive values were 79.71% and 95.45%, respectively. These outcomes indicate that CA125 levels have moderate diagnostic performance in detecting ovarian cancer recurrence. The current treatments for ovarian cancer include aggressive cytoreductive operation and platinum‐based combination chemotherapy, with approximately 80% of patients achieving clinical remission, indicated by normalization of CA125 levels and imaging findings. Karam and Karlan [ 36 ] reported that patients with continuously increased CA125 levels (> 35 U/mL) after primary therapy may experience recurrence within 6 months and have a higher risk of death. For patients who achieve clinical remission with CA125 < 35U/mL, the absolute level of CA125 is negatively associated with survival. However, most of these women will experience disease relapse within 2–5 years. Therefore, monitoring for recurrence or metastasis by various imaging methods and tumor markers is critical to facilitate timely treatment. The 2022 revised NCCN guidelines also recommend using effective methods to assess whether patients are suitable for secondary tumor cell ablation, with PET/CT being one of the effective evaluation methods on the basis of its unique detection advantages [ 37 , 38 ]. PET/CT provides metabolic information for the timely detection of recurrence or metastasis in patients after ovarian cancer surgery. Since its introduction, PET/CT has been frequently used in clinical practice, especially in tumor detection, staging, and efficacy evaluation [ 39 , 40 ]. Clinical evidence shows that most malignant ovarian tumors exhibit high FDG uptake, so 18 F‐FDG PET/CT has a high acuity in identifying ovarian malignancies, particularly in cases where CA125 levels are elevated but CT shows negative results or where CA125 levels are normal but CT shows positive results [ 41 ]. Combined PET/CT evaluation has become a new trend. However, PET/CT cannot detect microlesions with a diameter smaller than 0.5 cm [ 42 ]. This suggests that patients with negative imaging results should be closely observed and followed up clinically to avoid delayed diagnosis and treatment. The combination of two or more detection methods may thus compensate for the shortcomings of a single detection method. PET/CT combined with HE4 and CA125 levels greatly enhances the diagnostic accuracy of ovarian cancer, with high reliability, validity, and specificity. PET/CT is of crucial importance for the identification, prognosis evaluation of ovarian cancer and is an efficient diagnostic method in the follow‐up of patients after ovarian cancer surgery [ 43 ]. For cases with abnormal CA125 levels but negative B‐ultrasound, CT, and MRI results, PET/CT examination can enhance the precision of the diagnosis of postoperative metastasis and recurrence, providing accurate molecular imaging data and information for clinical treatment [ 44 , 45 ]. For patients with abnormal CA125 levels after ovarian cancer surgery, it is critical to be highly vigilant about recurrence or metastasis and undergo PET/CT imaging examination in a timely manner to detect and treat it as soon as possible. Most previous studies on PET/CT for diagnosing ovarian cancer have focused on a combined analysis of patients with normal and elevated postoperative CA125 levels. However, these studies have not specifically addressed the subset of cases with postoperative CA125 elevation. The innovation of the current research lies in its exclusive focus on postoperative ovarian cancer cases with abnormally increased CA125 levels. This approach fills a gap in the systematic evaluation and meta‐analysis of the diagnostic value of 18 F‐FDG PET/CT for relapse or dissemination in patients with postoperative CA125 elevation in ovarian cancer using evidence‐based medicine methods. However, this study has some limitations. First, the 13 studies considered in this analysis included cases with elevated CA125 levels, excluding patients with normal CA125 levels, which may have led to bias in the outcomes. Second, the follow‐up of ovarian cancer recurrence or metastasis included the use of postoperative pathology and other imaging data. The lack of a standardized follow‐up method may have affected the accuracy of the results. Third, the follow‐up time for recurrence or metastasis was not standardized, leading to further limitations in case selection, case process, and progression. Additionally, among the 13 studies included, five studies had 0 true negatives, which caused the analysis software to automatically ignore their effect size. Future studies should focus on increasing the number of relevant studies, expanding sample sizes, and addressing the shortcomings of this study.

As can be seen, the meta‐analysis provides compelling evidence that 18 F‐FDG PET/CT is a valuable imaging modality for identifying relapse or dissemination in postoperative ovarian cancer cases with increased serum CA125 concentrations. For this specific patient population, 18 F‐FDG PET/CT should be regarded as an essential monitoring strategy. This imaging modality enables clinicians to accurately identify metastatic lesions, guiding appropriate management decisions and potentially improving patient outcomes. As novel therapeutic strategies emerge, the treatment landscape for ovarian cancer continues to evolve. Consequently, there is a growing need for further research to optimize the integration of 18 F‐FDG PET/CT into the therapeutic algorithms for ovarian cancer. We anticipate that future studies will provide valuable insights into the role of 18 F‐FDG PET/CT in guiding personalized therapy approaches and ultimately improving the outcome of ovarian cancer patients.

Ovarian cancer has a high incidence rate. Despite the application of operation and postoperative chemotherapy in primary stage patients, the post‐surgery recurrence rate is approximately 50%–75%, with a 5‐year survival rate of 30% [ 1 , 2 , 3 ]. Recurrent ovarian cancer refers to signs of tumor recurrence after satisfactory tumor cytoreduction and after 6 months of completion of a regular and sufficient amount of chemotherapy to achieve complete clinical remission. Recurrent ovarian cancer is mostly advanced and incurable; it has a severe influence on the quality of life and overall survival [ 4 , 5 ]. Therefore, early diagnosis of relapsing ovarian cancer and timely intervention are crucial for improving patient prognosis and prolonging survival. Several factors influence the prognosis of recurrent ovarian cancer, including pathological type, preoperative clinical staging, postoperative re‐staging, serum tumor marker levels, postoperative residual cancer foci size, and the standardization of postoperative chemotherapy [ 6 , 7 , 8 , 9 ]. Carbohydrate antigen 125 (CA125) is one of the indicators that can reflect the cell load of ovarian cancer. CA125 is frequently used in clinical practice as a serum test indicator for ovarian cancer follow‐up to determine recurrence or metastasis in the postoperative period of ovarian cancer [ 10 , 11 ]. CA125 has good sensitivity, but it cannot provide information on the anatomical localization of the tumor [ 12 , 13 ]. Therefore, it needs to be combined with other examination methods to aid in improving diagnosis. The main methods for monitoring the prognosis of ovarian cancer after surgery include various imaging modalities (ultrasound, CT, MRI, PET/CT) and serum biomarkers (HE4, CA125). Among the imaging modalities used to postoperatively monitor patients with ovarian cancer, ultrasound has the advantages of being simple and radiation‐free, making it a frequently used imaging approach for gynecological diseases. However, early ovarian cancer tumors may be small in volume and limited to the ovaries, with no changes in ovarian morphology; imaging thus may result in false negative results. CT and MRI offer detailed anatomical information about the ovaries and adjacent tissues, which is of important clinical value in formulating the extent of invasion and formulating surgical plans. MRI has better soft tissue imaging capabilities than CT but may not be sensitive enough to detect tumors under 5 mm [ 14 ]. However, PET/CT imaging is a functional imaging technique that integrates anatomical localization and functional metabolism imaging. PET/CT imaging can identify and detect metabolic changes at the molecular level before changes in tissue and organ structures are apparent. This modality can also show the different distributions of radiopharmaceutical uptake in tumor tissues, reflecting the metabolic activity of tumor cells, tissue perfusion, and cell proliferation. It provides reliable imaging for the early detection of lesions and clinical staging and helps in evaluating the efficacy of tumor staging and subsequent treatment [ 15 ]. PET/CT has the advantages of noninvasiveness, sensitivity, and high diagnostic veracity in the comprehensive evaluation of tumors and shows unique value in the identification and localization of relapse or dissemination foci of tumors in ovarian cancer cases after surgery [ 16 ]. Given the high risk of recurrence and metastasis after ovarian cancer surgery, close follow‐up monitoring is recommended. CA125 alone has high sensitivity in diagnosing ovarian epithelial carcinoma, but poor specificity and a high false‐positive rate [ 17 ]. The combined evaluation of CA125 and PET/CT has become a new trend. In this meta‐analysis, we evaluated the diagnostic effectiveness of 18 F‐FDG PET/CT or identifying relapse or metastasis in ovarian cancer cases with postoperative elevated levels of serum CA125.

The authors declare no conflicts of interest.