Fluorescence in situ hybridization (FISH) or other in situ hybridization (ISH) testing of uterine cervical cells to predict precancer and cancer final
While most studies evaluated an ISH assay examining multiple probes, this scan served to focus the subsequent detailed evidence review on the most commonly studied ISH tests, which included probes for the telomerase RNA component gene (TERC [3q26]), the myelocytomatosis oncogene (MYC [8q24]), HPV 16...
| Main Author: | |
|---|---|
| Corporate Authors: | , |
| Format: | eBook |
| Language: | English |
| Published: |
Rockville, Maryland
Agency for Healthcare Research and Quality
2013, February 16, 2013
|
| Series: | Technology assessment report
|
| Subjects: | |
| Online Access: | |
| Collection: | National Center for Biotechnology Information - Collection details see MPG.ReNa |
| Summary: | While most studies evaluated an ISH assay examining multiple probes, this scan served to focus the subsequent detailed evidence review on the most commonly studied ISH tests, which included probes for the telomerase RNA component gene (TERC [3q26]), the myelocytomatosis oncogene (MYC [8q24]), HPV 16, or HPV 18, alone or in combination with other probes. For KQ2, about the analytic validity of ISH testing, we included any study that used ISH with any of these four probes in cervical cytology or histology specimens and compared the ISH test with a non-ISH reference test. For KQ3, about the clinical validity of ISH testing on cervical cytology for high-grade CIN on histology, cervical cancer or clinical outcomes related to morbidity and mortality from cervical cancer, we included any study using ISH with any of the four probes in cervical cytology specimens to detect high-grade CIN or cervical cancer (or clinical outcomes). OBJECTIVES: Screening for cervical cancer has the potential to detect precancerous lesions and cancers in early stages, which can be effectively treated. Screening tests currently used in the United States on cervical cell samples include the Papanicolaou (Pap) test to detect cellular changes, as well as tests for high-risk human papillomavirus (HPV) genotypes. A particular challenge is the management of women with test results of atypical squamous cells of unknown significance (ASCUS) or of low-grade squamous intraepithelial lesions (LSIL) on cytology or those with a normal Pap test but a positive test for high-risk HPV genotypes, since only a fraction of these women will have a finding on colposcopically directed tissue biopsy that warrants treatment (e.g., high-grade cervical intraepithelial neoplasia [CIN]). For KQ3, 10 studies provided information on the clinical validity of FISH tests for CIN2+ or CIN3+. Of these, eight provided results for FISH using a TERC probe (with three using probes for both TERC and MYC); three studies provided results for ISH using a probe for HPV 16 or 18 (one study was of FISH with all four probes). HPV status was not known except in one study of women who were all HPV positive (type not specified). Meta-analysis was performed for studies of ISH for TERC in women with LSIL cytologic findings. For CIN2+, with data from seven studies, the summary sensitivity was 0.76 (95 percent confidence interval [CI] 0.60, 0.86) and the summary specificity was 0.79 (95 percent CI 0.50, 0.93). For CIN3+, with data from five studies, the summary sensitivity was 0.78 (95 percent CI 0.65, 0.87) and the summary specificity was 0.79 (95 percent CI 0.51, 0.93). For KQ1, 135 articles described use of ISH on cervical specimens (cytologic or histologic), and 116 involved ISH using one of the four probes of interest: 31 used an ISH probe for TERC, with 7 of these also using probes for MYC; and 91 studies used an ISH probe for HPV 16, with 87 of these also using a probe for HPV 18. (Five studies used both a TERC probe and an HPV 16 or 18 probe). For KQ2, 14 studies provided data on agreement between ISH tests with an HPV 16 or 18 probe (among other HPV probes) and HPV reference tests (polymerase chain reaction [PCR] or Hybrid Capture 2). (None compared a FISH test for TERC or MYC with a DNA-based reference test.) The agreement between each ISH-non-ISH test pair was variable, reflecting differences in measurement techniques between the ISH tests and reference tests as well as the use of nonoverlapping panels of probes. Assessment of study quality showed deficiencies in reporting. Further research is needed to standardize techniques; compare clinical validity, thresholds, and combinations across different ISH tests; and compare the clinical utility of combinations of probes as add-on tests to HPV and cytology tests We rated the strength of evidence as low for KQ3, failing to show consistently better sensitivity or specificity with FISH testing for identification of CIN2+ or CIN3+ than would be expected by chance. There were no standard thresholds for test positivity across KQ2 or KQ3 studies of ISH for TERC or MYC. For other questions related to preanalytic issues impacting analytic validity, the data were sparse or not informative. For KQ3, no study in the specified contexts examined the association of FISH test results with clinical outcomes. For KQ4, no study compared patient care strategies resulting from different tests, thresholds, or combinations of ISH and/or non-ISH tests. CONCLUSIONS: Overall, the evidence of the analytic and clinical validity of ISH tests in screening for cervical cancer was limited. We aimed to examine the role of in situ hybridization (ISH) tests, including fluorescence ISH (FISH), to detect chromosomal abnormalities or DNA from high-risk oncogenic HPV genotypes on cervical cytologic specimens to increase the clinical validity of identification of precancerous lesions or cervical cancer. DATA SOURCES: MEDLINE(r) (from inception to October 2011, week 2), the Cochrane Central Trials Registry (through the fourth quarter of 2011), and Scopus (including Embase) on November 7, 2011, with no language exclusion. The searches were updated on July 12, 2012. REVIEW METHODS: We used established systematic review methods to identify articles on the basis of predetermined eligibility criteria: studies of ISH tests in cervical tissue from at least 10 women. We addressed four Key Questions (KQs). For KQ1, a horizon scan of what ISH tests have been examined with what frequency, we included any study that tested ISH in cervical cytology or histology. Also for KQ3, two studies compared combinations of FISH tests with reference tests, with both defining positivity on combination testing as positivity of either FISH or the reference test. In one, FISH testing alone, for TERC, showed lower sensitivity but higher specificity than did combined testing with FISH and Hybrid Capture 2. The other study showed that FISH testing for TERC or MYC had a lower sensitivity but higher specificity than did FISH for TERC, MYC, or HPV and Hybrid Capture 2 for high-risk HPV. For other KQ3 comparisons, the number of studies was limited. Only three studies had data on FISH for TERC in ASCUS specimens, and only three had data on ISH for HPV in LSIL or ASCUS samples. Across all 10 KQ3 studies, there was a trade-off between sensitivity and specificity, which may be indicative of a threshold effect. There was also large clinical heterogeneity across populations and test probes. Assessment of risk of bias suggested low study quality and incomplete reporting. Cervical cytology had to be classified according to the Bethesda classification as ASCUS or LSIL or as normal in combination with a positive HPV test. Histology outcomes had to be defined as CIN and had to be expressible as either CIN3+ (i.e., CIN3 or cervical cancer) or CIN2+ (i.e., CIN2, CIN3, or cervical cancer). Studies had to provide data that allowed for calculation of sensitivity and specificity. For KQ4, about the clinical utility and possible harms of ISH testing, we reviewed studies that compared patient management strategies using different screening or testing algorithms, including ISH testing. RESULTS: The literature search yielded a total of 1462 abstracts, of which we screened 227 in full text. |
|---|---|
| Item Description: | Title from PDF title page |
| Physical Description: | 1 PDF file (various pagings) illustrations |