Hypermethylated ZNF582 and PAX1 are effective biomarkers for detection of oral dysplasia and oral cancer.

OBJECTIVES: This study investigated whether the methylation of ZNF582, PAX1, SOX1, NKX6.1, and PTPRR genes in oral scrapings could be used to detect oral dysplasia and oral cancer and to predict oral cancer recurrence. MATERIALS AND METHODS: Oral scrapings were collected from 65 normal oral mucosa subjects, 107 oral precancer patients, and 95 oral squamous cell carcinoma patients. Methylation levels of the five genes were quantified by real-time methylation-specific PCR after bisulfite conversion. RESULTS: Among the five tested genes, methylated ZNF582 (ZNF582m) and PAX1 (PAX1m) were found to be appropriate biomarkers for oral dysplasia and oral cancers. ZNF582m could detect mild dysplasia or worse oral lesions with the sensitivity and specificity being 0.85 and 0.87, respectively. PAX1m performed better in identifying moderate dysplasia or worse oral lesions with the sensitivity and specificity being 0.72 and 0.86, respectively. Moreover, the methylation levels and positive rates for ZNF582m and PAX1m were increased when disease severity increased. Thus, they may be applicable as a triage tool for patients with abnormal visual oral examinations. After cancer excision, both ZNF582m and PAX1m levels decreased. However, their levels increased again at the subsequently recurrent sites in some patients approximately 3-4 months before cancer recurrence. Finally, areca-quid chewing alone and in combination with cigarette smoking or alcohol drinking were found to be correlated with ZNF582 and PAX1 hypermethylation. CONCLUSION: We conclude that hypermethylated ZNF582 and PAX1 are effective biomarkers for the detection of oral dysplasia and oral cancer and for the prediction of oral cancer recurrence.

Combined clinical and genetic testing algorithm for cervical cancer diagnosis.

BACKGROUND: Opportunistic screening in hospitals is widely used to effectively reduce the incidence rate of cervical cancer in China and other developing countries. This study aimed to identify clinical risk factor algorithms that combine gynecologic examination and molecular testing (paired box gene 1 (PAX1) or zinc finger protein 582 (ZNF582) methylation or HPV16/18) results to improve diagnostic accuracy. METHODS: The delta Cp of methylated PAX1 and ZNF582 was obtained via quantitative methylation-specific PCR in a training set (57 CIN2- and 43 cervical intraepithelial neoplasia ≥grade 3 (CIN3+) women), and the individual and combination gene sensitivities and specificities were determined. The detection accuracy of three algorithms combining gynecologic findings and genetic test results was then compared in a randomized case-control study comprising 449 women referred for colposcopic examination by gynecologists in the outpatient department of Xiangya Hospital between November 2011 and March 2013. RESULTS: Significant association was observed between CIN3+ and methylated PAX1 or ZNF582 in combination with HPV16/18 (OR:15.52, 95 % CI:7.73-31.18). The sensitivities and specificities of methylated PAX1 or ZNF582 combined with HPV16/18 for CIN3+ women were 89.2 and 76.0 %, or 85.4 and 80.1 %, respectively. Of the three algorithms applied to cohort data and validated in the study, two indicated 100 % sensitivity in detecting cervical cancer and a low rate of referrals for colposcopy. CONCLUSIONS: These algorithms might contribute to precise and objective cervical cancer diagnostics in the outpatient departments of hospitals in countries with high mortality and low screening rates or areas with uneven resource distribution.