| Summary: | 博士 === 臺北醫學大學 === 醫學科學研究所 === 96 === Accumulating evidence shows that tumor DNA can be found in the circulation of patients with cervical cancer. The presence of such tumor DNA in the blood may be of diagnostic and prognostic value. HPV DNA has been found in serum or plasma samples from cervical cancer patients with detection rates varying from 7% to 45%. The discrepancy may be due to different target materials (serum or plasma), method of extracting DNA, tools of analysis (conventional PCR, real-time PCR, or PCR-enzyme immunoassay), and differing primers used (L1, E6, E7). Therefore, information regarding the comparison of detection rates of HPV DNA in circulating blood is limited.
The first part of this study provides a prospective study of HPV DNA detection at a single diagnostic time point. Real-time PCR is used to detect the low viral loads of HPV DNA in blood. The results show that more than one-fourth (27%) of patients with invasive cervical cancer had HPV DNA detected in their blood samples. Approximately 50% of patients with confirmed HPV 16, 18 or 52 positive cervical cancers had HPV DNA detected in their blood. This study also used serial follow-up data on HPV DNA viral load among cervical cancer patients after treatment to understand its clinical significance. Six cervical cancer patients with HPV DNA viral loads undetectable in their blood after treatment showed no recurrence during follow-up. In longitudinal follow-up, eight out of ten cervical cancer patients with viral loads of HPV DNA detectable in the blood at 3 months after treatment were associated with recurrence. Among these, seven of eight patients had distant metastases. Although the study was limited to a small number of patients and a short period of follow-up, it is worth pointing out that detection of circulating HPV DNA after treatment could predict recurrence. It is postulated that blood HPV DNA might be a useful marker to select subsets of patients who need more aggressive treatment. The presence and quantity of HPV DNA in blood are likely to be a reflection of metastasis and may be of prognostic value.
The second part of this study focuses on the role of integration of HPV type 52 and 58 in cervical cancer patients. The integration of HPV DNA into the host genome is thought to occur early in cancer development and to be an important event in malignant transformation of cervical cancer. However, most studies on the integration of HPV DNA focus on type 16 and a few on type 18. While HPV type 52 and 58 are oncogenic types with relatively low prevalence in cervical cancer in the Americas, Europe, Africa and Southeast Asia, they are as prevalent as the known high-risk (for cervical cancer) HPV types 16 and 18 in Taiwan and other Asian countries. To analyze whether integration or high viral loads of human papillomavirus (HPV) are essential for malignant transformation of HPV type 52 and 58 as well as type 16 and 18, cervical swabs from 178 consecutive patients, including 81 with invasive cervical cancers and 97 with cervical intraepithelial neoplasias (CIN) II-III, were collected and examined to determine the prevalence, physical status and viral load of HPV type 16, 18, 52 and 58 DNA using genechip and real-time PCR (polymerase chain reaction) analysis.
The infrequent integration of HPV 52 and 58 DNA in cervical cancer suggests that it is not a prerequisite for progression to cervical cancer. By contrast, integration appears to be a critical step for carcinogenesis of HPV 16 and 18 DNA. High viral loads (E6) of HPV 16, 18 and 52 DNA may be predictive of the transition of CIN II-III to cervical cancer. The results indicate that both viral DNA physical status and viral loads of HPV are important factors in the carcinogenesis of different HPV types. This study successfully used the median log of viral loads of HPV 16, 18 and 52 DNA to predict the presence of cervical cancer. The selected cut-off values of the median log of viral loads in HPV 16, 18 and 52 DNA achieved 62.5-83.3% sensitivity and a 0-25% false positive rate in predicting the presence of cervical cancer. The ROC curve analyses indicated that the model could accurately predict the diagnostic group of CIN II-III or cervical cancer in 73.8%, 92.9%, and 88.5% of patients with positive HPV 16, 18 and 52, respectively.
The third part of this study focuses on low-grade squamous intraepithelial lesions (LSILs). Approximately 50% of atypical squamous cells of undetermined significance (ASCUS) and 80% of LSILs are infected by oncogenic types of HPV. HPV DNA testing for patients with ASCUS provides useful information and allows referral of patients for immediate colposcopy to detect high-grade squamous intraepithelial lesions (HSILs) and cancer. By contrast, oncogenic HPV DNA testing is not informative for triage of patients with LSILs because a high percentage of LSIL patients are HPV positive. A repeat Pap smear in 3 to 6 months or direct biopsy under colposcopy is generally used in clinical practice. Development of alternative triage strategies for women with LSILs would be valuable in distinguishing women with LSILs that have high probabilities of progression to HSILs from women with LSILs that have spontaneously regressed.
The 2-year cumulative risks were evaluated for HSIL attributable to HPV 16, 18, 52, and 58, the most common oncogenic types in pre-invasive cervical lesions including LSILs and HSILs in Asia, and questioned as to whether the integration of HPV oncogenes into the host genome contributed to the risk of LSILs progressing to HSILs. In addition, it was determined whether or not E6 viral load and its change contributed to the risk of LSILs progressing to HSILs during the interval between baseline diagnosis of LSIL by Pap smear and a 6-month follow-up visit by repeat Pap smear. It was found that women with LSILs whose viral loads increased between baseline and 6 month follow-up had a 45% risk of developing HSIL, which was seven-fold greater than those without increased viral loads (OR = 7.6, 95% CI = 1.9 to 29.4, p < 0.01), as evaluated by real-time PCR. The risk was calculated at 44%, a six-fold greater risk than those without increased viral loads (OR = 6.1, 95% CI = 1.6 to 22.7, p < 0.01), as evaluated by HC2. The two viral load measures correlated well (Person’s coefficient, r = 0.687, p < 0.001). The results indicate that evaluation of viral load changes (increased or not increased) through repeat HPV DNA testing could predict progression of disease in LSIL cases of HPV types 16, 18, 52, and 58, which correlates to clinical implications.
In summary, this research strives to understand the role of HPV DNA viral loads and integration in the carcinogenesis of cervical cancer by searching for a useful marker applicable in clinical practice to predict disease progression in pre-invasive and invasive cervical cancer.
|
|---|
