Target condition being diagnosed
Ovarian cancer is the sixth most common cancer among women (GLOBOCAN 2008). Worldwide there are more than 200,000 new cases of ovarian cancer each year, accounting for around 4% of all cancers diagnosed in women. A woman's risk of developing cancer of the ovaries by age 75 years varies between countries, ranging from 0.5% to 1.6%. This corresponds to an age-standardised rate of ovarian cancer from 5 to 14 cases per year in 100,000 women under 75 years. In Europe, 37 to 41% women with ovarian cancer are alive five years after diagnosis (EUROCARE 2003). The poor survival associated with ovarian cancer is largely because most women are diagnosed when the cancer is already at an advanced stage (Jemal 2008).
Surgery is the first step in the initial diagnosis and staging of ovarian cancer. Early-stage ovarian cancer [the International Federation of Gynecology and Obstetrics (FIGO) stage 1 and 2] (FIGO 2009) has a combined incidence of less than 35% with five-year survival rates of 92% and 55% for stage 1 and stage 2, respectively (Cancer Research UK 2012) (Appendix 1).
Currently, women presenting with pelvic mass suspected of ovarian cancer are triaged according to risk of malignancy index (RMI) (Jacobs 1990; Bailey 2006). RMI is calculated by incorporating ultrasound features of pelvic mass, menopausal status and serum CA125 levels allowing women with high RMI (> 200) to undergo laparotomy at cancer centres. The main goal of imaging is to differentiate malignant from benign tumours and to determine surgical strategy. In cases with suspected early stage ovarian cancer, frozen section analysis further informs intraoperative findings and aids decision-making about the choice and extent of surgery conducted by the surgeon.
Standard histopathological diagnostic practice is with paraffin section reporting of the surgical specimen, which is often reported a few days after the surgery. If confirmed to have ovarian cancer, some of these women will be considered for post-operative adjuvant chemotherapy. Long-term outcomes from a randomised controlled trial of surgical staging and treatment of early ovarian cancer have confirmed that optimal surgical staging in early stage epithelial ovarian cancer is an independent predictor of improved overall and recurrence free survival (Trimbos 2010).
Therefore, incorrect intraoperative assessment of the mass and inadequate surgical staging can mean that patients with stage 1 ovarian cancer require adjuvant chemotherapy. A Cochrane systematic review on adjuvant (post-surgery) chemotherapy for early stage epithelial ovarian cancer showed no benefit of adjuvant chemotherapy in optimally staged early ovarian cancer (Winter-Roach 2009).
Intraoperative frozen section analysis of the pelvic mass allows intraoperative histopathological diagnosis and appropriate surgical staging. This is not routinely performed in most gynaecological cancer units in the United Kingdom. In centres where frozen section is not available, final histology on paraffin section is used to guide the need for either further laparotomy to complete the surgical staging and in some cases adjuvant chemotherapy.
Intra-operative frozen section analysis allows appropriate selection of cases in which to perform optimal staging. This would include inspection and palpation of peritoneal cavity and organs, peritoneal washings, salpingoophrectomy of affected side, omentectomy, bilateral pelvic node dissection, selective para-aortic lymphadenectomy and contralateral salpingoophrectomy and hysterectomy if retention of fertility was not an issue. Accurate staging in stage I disease is an independent prognostic factor for survival. It also informs ongoing management and can predict individual patient outcomes.
Should the frozen section report the suspicious pelvic mass as benign, a full staging procedure would not be performed, thereby reducing associated morbidity and optimising theatre time. In the event of a borderline or malignant frozen section report, optimal staging is performed. Studies have reported high sensitivity, specificity and overall accuracy when compared with the gold standard of histology on paraffin section reporting (Gol 2003; Medeiros 2005; Naik 2006; Brun 2008; Cross 2012).
Many women will have had pre-operative imaging in the form of an ultrasound, MRI or CT scan. Tumour markers, such as Ca125 may have been performed. Interpretation of the histology slides at frozen section is made independently of these prior tests and so these will bear no relevance to the analysis of frozen section results.
Role of index test(s)
Intra-operative frozen section analysis allows appropriate selection of cases in which to perform optimal staging.
The role of intraoperative frozen section analysis in early stage ovarian cancer diagnosis and management is particularly topical at present, with many recent studies reporting high sensitivity and specificity of this diagnostic test. Frozen section analysis may allow the surgeon to accurately identify cases which will benefit from optimal surgical staging and therefore avoid the need for a restaging procedure or for adjuvant chemotherapy. We decided to review those studies which reported use of intra-operative frozen section in the management of suspected ovarian cancer to evaluate its accuracy, validity and feasibility.
To establish the accuracy and other diagnostic parameters (sensitivity, specificity, positive predictive value and negative predictive value) of intraoperative frozen section analysis in histopatholgical diagnosis of ovarian cancer, in comparison to paraffin section reporting.
To establish if intraoperative frozen section analysis allows the surgeon to accurately identify those cases of early stage ovarian cancer which would benefit from optimal surgical staging, therefore avoiding need for restaging procedure or need for adjuvant chemotherapy
To establish the accuracy of frozen section diagnosis in metastatic tumours, mucinous tumours and large thin-walled tumours, where frozen section may miss areas of malignancy.
Criteria for considering studies for this review
Types of studies
We aim to include studies published in any language. However, we will not include non-English articles for which a full-text translation or evaluation cannot be obtained.
Studies will be eligible if:
- both frozen section analysis and paraffin section analysis are performed in the same patient;
- the absolute numbers of observations of true positives, false positives, false negatives and true negatives are available or derivable from the data reported in the primary studies
We will include both prospective and retrospective studies.
We will exclude studies in which frozen section analysis is performed for conditions other than ovarian malignancy.
Any female patients presenting to a secondary or tertiary care setting with a pelvic mass suspicious of ovarian cancer, in whom frozen section analysis is employed prior to paraffin section analysis.
Intraoperative frozen section histological analysis.
Ovarian malignancy, not obviously surgically or radiologically advanced stage.
Paraffin section histological analysis.
Search methods for identification of studies
We will identify eligible studies by searching the following electronic databases:
The Cochrane Gynaecological Cancer Group Specialised Register
Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library) - current issue
MEDLINE - Ovid (from January 1946 to current)
EMBASE - Ovid (from January 1980 to current)
Database of Abstracts of Reviews of Effects (DARE) - current issue
Health Technology Assessments (HTA) Database - current issue
The MEDLINE search strategy will include both subject headings (MeSH terms) and text words for the target condition (ovarian malignancy) and the histological technique under investigation (Frozen Section analysis). We will not apply language restrictions. We will adapt the MEDLINE search to search EMBASE, CENTREAL, DARE and HTA databases. In particular, we will adapt the MEDLINE MeSH terms into the corresponding terms available in the EMTREE vocabulary. Full details of the MEDLINE and EMBASE search strategies together with a brief summary of the MEDLINE search strategy are presented in Appendix 2. We will import all citations identified by the MEDLINE and EMBASE search strategies into the Reference Manager bibliographic database. All potentially eligible articles will be identified on PubMed and using the 'related articles' feature a further search will be carried out for newly published searches.
Searching other resources
Unpublished and grey literature
Ongoing trials will be searched for in the following trial registers and experts in the field will be contacted to identify any further ongoing trials.
- metaRegister of Controlled Trials (mRCT) (http://www.controlled-trials.com/mrct/)
- Physicians Data Query (PDQ
- ClinicalTrials.gov (http://clinicaltrials.gov/)
- National Cancer Institute (http://www.cancer.gov/clinicaltrials/search)
The citation lists of included studies and key textbooks will be checked through handsearching. When relevant studies are retrieved, their references will also be searched in order to minimise missing relevant studies. Reports of conferences will be handsearched in the following sources:
- Gynecologic Oncology (Annual Meeting of the American Society of Gynecologic Oncologists)
- International Journal of Gynecologic cancer (Annual Meeting of the International Gynecologic Cancer Society)
- British Gynaecological Cancer Society
- European Society of Gynaecological Oncology
- Society of Gynaecological Oncologists
Data collection and analysis
Selection of studies
All titles and abstracts retrieved by electronic searching will be downloaded to a reference management database, duplicates will be removed and the remaining reference titles and abstracts will be examined by two authors (NR and AP) independently to retrieve all potentially relevant reports in full. The same two authors will independently review all relevant reports according to the pre-defined inclusion criteria to determine eligibility. Any disagreements will be resolved by arbitration by two other authors (KG and RN). One author (NR) will extract data from the selected reports. Reasons for exclusions will be documented.
Data extraction and management
We have designed a data extraction form specifically to collect details from selected studies. One author (NR) will record the relevant information for each individual study, without concealing the study authorship of publication details. This information will include: journal name, lead author, year of publication, study design, method of recruitment, setting, number and characteristics of participants, any additional pre-operative investigations performed, the reference standard used, any comparator tests used, follow-up, and information related to the pathologists interpreting the specimens (background specialty, level of expertise).
Both adjusted and unadjusted statistics will be extracted, if reported.
Assessment of methodological quality
Two review authors (NR and AP) will independently assess the methodological quality of each study included using the latest version of the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool, QUADAS-2. The QUADAS tool is structured in a series of questions in four domains which should be answered 'yes', 'no' or 'unclear', and aims to evaluate the spectrum of bias. We will resolve any disagreements by arbitration. A summary of the methodological quality of each study will be presented in a graph.
We will assess the following core QUADAS items in the following domains; patient selection, index test, reference standard, flow and timing:
Could the selection of patients have introduced bias?
- Was a consecutive or random sample of patients enrolled?
- Yes; a study ideally should enrol a consecutive or random sample of eligible patients with suspected disease to prevent the potential for bias
- No, a non-consecutive sample of patients was used.
- Was a case–control design avoided?
- Yes; studies enrolling participants with known disease and a control group without the condition may exaggerate diagnostic accuracy
- Did the study avoid inappropriate exclusions?
- Yes; studies that make inappropriate exclusions (for example, not including "difficult-to-diagnose" patients) may result in overestimation of diagnostic accuracy.
- Were the patients selected representative of the patient population the index test would apply to?
- Yes; Patients with a high risk of malignancy index (RMI >200) are usually the subjects who would benefit from this index test.
- No; Use of the index test in patients at low risk of malignancy; or those with incidental finding at laparotomy for other condition may bias the results.
Could the conduct or interpretation of the index test have introduced bias?
- Were the index test results interpreted without knowledge of the results of the reference standard?
- Yes; Knowledge of the reference standard may influence interpretation of index test results. The potential for bias is related to the subjectivity of interpreting index test and the order of testing. If the index test is always conducted and interpreted before the reference standard, this item can be rated 'yes'.
- No; If there was a previous histological diagnosis of malignancy made during investigation of the same cyst, this item can be rated 'no'.
- Were the index test results interpreted by a pathologist specialising in gynaecological oncology?
- Yes; Specialist centres employing dedicated gynaecological oncology pathologists may perform better in interpreting frozen section slides and thereby sensitivity and specificity of the test.
Could the reference standard, its conduct, or its interpretation have introduced bias?
- Is the reference standard likely to correctly classify the target condition?
- Yes; Estimates of test accuracy are based on the assumptions that the reference standard is 100% sensitive and that specific disagreements between the reference standard and index test result from incorrect classification by the index test.
- Were the reference standard results interpreted without knowledge of the results of the index test?
- Yes; Knowledge of the index test results may influence interpretation of the reference standard results. Potential for bias is related to the potential influence of previous knowledge on the interpretation of the reference standard.
Flow and timing
Could the patient flow have introduced bias?
- Did all patients receive a reference standard, and if so did they receive the same reference standard?
- Yes; Verification bias occurs when only a proportion of the study group receives confirmation of the diagnosis by the reference standard, or if some patients receive a different reference standard. If the results of the index test influence the decision on whether to perform the reference standard or which reference standard is used, estimated diagnostic accuracy may be biased. It is accepted best practice to verify all frozen section diagnoses with paraffin section histology.
- Were all patients included in the analysis?
- Yes; All participants recruited into the study should be included in the analysis. A potential for bias exists if the number of patients enrolled differs from the number of patients included in the 2x2 table of results, because patients lost to follow-up differ systematically from those who remain.
Statistical analysis and data synthesis
The diagnostically important distinction to make is between malignant/borderline and benign frozen section. This is because although only women with malignant disease require surgical staging, the risks of borderline frozen section returning as malignant have shown to be high (Cross 2012) and to inadequately stage these women at primary laparotomy may be deemed unacceptable ( Table 1). However, many would argue that performing unnecessary staging on women with borderline disease is unacceptable and Table 2 investigates whether managing borderline frozen section tumours as benign during laparotomy might reduce the number of false positives, that is women undergoing unnecessary surgical staging procedures.
These data will be inputted into the Cochrane Collaboration's statistical software, Review Manager 2011, for the calculations of sensitivity and specificity for each study (we will also present 95% confidence intervals of these point estimates in a forest plot). We will present individual study results graphically by plotting estimates of sensitivities and specificities in receiver operating characteristic (ROC) space. If more than one threshold is reported, we will choose the 2 x 2 table for the threshold most widely reported in included studies to incorporate in the meta-analysis. We will use the statistical package SAS 9.2 to meta-analyse pairs of sensitivity and specificity using a bivariate random-effects approach (Reitsma 2005). This approach enables us to calculate summary estimates of sensitivity and specificity, while correctly dealing with any correlation that might exist between sensitivity and specificity as well as the following sources of variation:
- imprecision by which sensitivity and specificity have been measured within each study;
- variation beyond chance in sensitivity and specificity between studies.
Covariates can be incorporated in the bivariate model in order to examine the effect of potential sources of heterogeneity on sensitivity and specificity. The results of the bivariate model will be used to calculate likelihood ratios.
Investigations of heterogeneity
We will construct a ROC plot of true positive versus false positive rate and explore the heterogeneity of the sensitivity and specificity estimates by examining both the ROC plot and forest plot.
We will consider separate heterogeneity analysis for the following situations:
- preoperative investigation including a combination of imaging and tumour markers (CA 125 +/- HE4);
- preoperative imaging including CT or magnetic resonance imaging (MRI) scan;
- RMI score >200;
- study population high risk, for example a tertiary referral centre;
- size of ovarian cyst;
- ovarian cyst histological type, for example mucinous or serous;
- expertise of pathologist reporting.
We will restrict our sensitivity analyses to studies without verification bias and without unaccountable missing data.
Assessment of reporting bias
Data regarding loss to follow up and any loss of data from pre-specified outcomes will be documented. Funnel plots corresponding to meta-analysis of the primary outcome will be examined to assess the potential for small study effects such as publication bias.
We thank Jo Morrison for her clinical advice. We thank Jane Hayes for designing the search strategy and Gail Quinn and Clare Jess for their contribution to the editorial process. We are also very grateful to the Cochrane Diagnostic Test Accuracy team for their invaluable comments as well as the peer referees.
Appendix 1. International Federation of Gynecology and Obstetrics (FIGO) staging of ovarian cancer
- Stage I: Growth limited to the ovaries
- Stage IA: Growth limited to 1 ovary, no tumour on the external surface, capsule intact, no ascites present containing malignant cells
- Stage IB: Growth limited to both ovaries, no tumour on the external surfaces, capsules intact, no ascites present containing malignant cells
- Stage IC: Tumour either stage IA or IB, but with tumour on surface of 1 or both ovaries with capsule ruptured,* with ascites present containing malignant cells, or with positive peritoneal washings
- Stage II: Growth involving 1 or both ovaries with pelvic extension
- Stage IIA: Extension and/or metastases to the uterus and/or tubes
- Stage IIB: Extension to other pelvic tissues
- Stage IIC: Tumour either stage IIA or IIB, but with tumour on surface of 1 or both ovaries, with capsule(s) ruptured, with ascites present containing malignant ovaries, or with positive peritoneal washings
- Stage III: Tumour involving 1 or both ovaries with histologically confirmed peritoneal implants outside pelvis and/or positive retroperitoneal or inguinal nodes; superficial liver metastasis; tumour limited to true pelvis, but with histologically proven malignant extension to small bowel and omentum
- Stage IIIA: Tumour grossly limited to the true pelvis, with negative nodes, but with histologically confirmed microscopic seeding of abdominal peritoneal surfaces or histologically proven extension to small bowel mesentery
- Stage IIIB: Tumour of 1 or both ovaries with histologically confirmed implants, peritoneal metastasis of abdominal peritoneal surfaces ≤ 2 cm in diameter; nodes are negative
- Stage IIIC: Peritoneal metastasis beyond the pelvis >2 cm in diameter and/or positive retroperitoneal or inguinal nodes
- Stage IV: Growth involving 1 or both ovaries with distant metastases; if pleural effusion is present, positive cytology must be apparent to allot a case to stage IV; parenchymal liver metastasis qualifies as stage IV disease
Appendix 2. MEDLINE (Ovid) search strategy
- exp Ovarian Neoplasms/
- (ovar* adj5 (cancer* or tumor* or tumour* or adenocarcinoma* or carcinosarcoma*or cystadenocarcinoma* or carcinoma* or malignan* or neoplas* or carcinogen* or teratoma* or metasta* or mass or masses)).tw,ot.
- (thecoma* or luteoma*).tw,ot.
- 1 or 2 or 3
- Frozen Sections/
- (FS or FSA or IFS or IFSA).tw,ot.
- (frozen or quick) adj5 section*.tw,ot.
- ((intraoperative or intra-operative) adj5 (consultation* or histolog* or diagnos* or patholog*)).tw,ot.
- (cryosection* or cryogenic*).tw,ot.
- (fresh or frozen) adj5 tissue*).tw,ot.
- 5 or 6 or 7 or 8 or 9 or 10
- 4 and 11
- exp animals/ not humans.sh.
- 12 not 13
key: tw=textword, ot=original title
Protocol first published: Issue 2, 2013
Contributions of authors
- Guarantor of the review: NR
- Conceiving the idea: RN, KG, PMH
- Designing and coordinating the review: NR, AP
- Data Collection for the review; Designing search strategies; Undertaking searches; Screening search results: NR, AP
- Organising retrieval of papers: NR, AP
- Screening retrieved papers against inclusion criteria: NR, AP
- Appraising quality of papers: NR, AP, KG, RN, PC
- Extracting data from papers: NR, AP
- Writing to authors of papers for additional information: NR
- Providing additional data about papers: NR
- Obtaining and screening data on unpublished studies: NR, AP
- Data management of the review: NR
- Entering data into RevMan5: NR
- Analaysis and interpretation of data: AB, NR, AP
- Providing a methodological perspective; Providing a clinical perspective; Providing a policy perspective; Providing a consumer perspective: KG, RN, NR, AP, PC
- Writing the review: NR
- Providing general advice on the review: KG, RN, NR, AP, PMH, PC
- Secure funding for the review: NR, AP, KG, RN
Declarations of interest
PC, AP, KG and RN were authors in a study which will meet the inclusion criteria in the review.
Sources of support
- No sources of support supplied
- Department of Health, UK.NHS Cochrane Collaboration programme Grant Scheme CPG-10/4001/12