A staged algorithm that incorporates the CA-125 assay to screen postmenopausal women for ovarian cancer has a near-perfect specificity of 99.9%, according to the results of a single arm, multicenter study that enrolled more than 3,200 women at average risk of the disease.
If confirmed in larger studies, this approach could be used to detect ovarian cancer in its early, more curable stages, lead author Dr. Karen Lu said during a May 20, 2010 press briefing sponsored by the American Society of Clinical Oncology.
“Ovarian cancer is the most lethal gynecologic cancer,” said Dr. Lu, professor of gynecologic oncology at the University of Texas M.D. Anderson Cancer Center, Houston. “Greater than 75% of cases present with advanced stage disease, when cure rates are less than 30%. If caught at an early stage, cure rates are 60%-90%, but at the current time there are no effective screening methods.”
For the 9-year study, scheduled to be presented on June 6 during ASCO’s annual meeting in Chicago, Dr. Lu and her associates enrolled 3,252 women aged 50-74 years with no significant family history of breast or ovarian cancer to be screened with the Risk of Ovarian Cancer Algorithm (ROCA). She described ROCA as a mathematical model that takes into account a woman’s age as well as changes in the values of her CA-125 assay over time.
“From here there are three possibilities,” she explained. “Those individuals who have a low ROCA score are told to come back at 1 year for a repeat CA-125. Those who have an intermediate ROCA score are told to come back at 3 months for another CA-125, and those who have a high ROCA score are triaged to a transvaginal ultrasound and referral to a gynecologic oncologist.”
After following the women for 9 years the researchers found that the average annual rate of referral for CA-125 assays every 3 months was 6.8% and that the average annual rate of transvaginal ultrasound and referral to a gynecologic oncologist was only 0.9%. “Each year the overwhelming majority of women were triaged to the low-risk category – an annual CA-125,” Dr. Lu said.
Cumulatively, 85 women (2.6%) received transvaginal ultrasound and subsequent referral to a gynecologic oncologist. Of these, eight required surgery: three for invasive ovarian cancers (two stage 1C and one stage IIB), two for borderline ovarian tumors, and three for benign ovarian tumors. This translated into a positive predictive value of 37.5%. “This means that three operations would be necessary to detect one case of invasive ovarian cancer,” Dr. Lu said.
The combined specificity of ROCA followed by transvaginal ultrasound was 99.9%, “which means that there were very few false positive.”
Dr. Lu emphasized that while results of the ROCA screening strategy are encouraging, “they are not practice changing at this time. We need to await the results of a definitive ovarian cancer screening trial that uses mortality as an end point, and uses the same ROCA algorithm.” That trial of more than 200,000 women is underway in the United Kingdom, she said. Results are expected in 2015.
ASCO President Dr. Douglas W. Blayney said that the ROCA algorithm “represents yet another example of personalized medicine. Here, we have a personalized screening strategy for a vicious type of cancer. This also represents a more refined application of known technology. The CA-125 is widely available, as is transvaginal ultrasound, which is intrusive and technologically somewhat difficult to interpret. Here, we have a staged application.”
One of the study authors, Dr. Herbert A. Fritsche, disclosed that he received research funding from Roche Diagnostics. Another study author, Dr. Robert C. Bast Jr., disclosed that he serves as a consultant and advisor to Fujiresio Diagnostics Inc. He also receives other remuneration and royalties for helping to invent the CA-125 assay.
Most of the more than 4,000 abstracts to be presented at ASCO have been posted online at www.abstract.asco.org.
Copyright (c) 2009 Elsevier Global Medical News. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.
根据一项单组多中心临床研究结果，一种将CA-125检测纳入其中的分阶段计算公式从绝经后妇女中筛查出卵巢癌患者的特异性近乎完美，为99.9%。该研究的受试妇女超过3,200人，其患卵巢癌的危险度均处于平均水平。
2010年5月20日，第一作者Karen Lu博士在美国临床肿瘤学会(ASCO)主办的新闻发布会上称，如果该方法在更大规模的研究中得以证实，则可用于检出早期卵巢癌，而此阶段卵巢癌的治愈率较高。
Lu博士是位于休斯顿的德克萨斯大学MD安德森癌症中心的妇科肿瘤学教授。她指出：“卵巢癌在妇科癌症中最为致命。超过75%的病例发现时处于晚期阶段，其治愈率低于30%。如果在早期发现该病，则治愈率可达60%~90%，但当前尚无有效的筛查方法。”
Lu博士及其同事预定于7月6日在芝加哥召开的ASCO年会上公布研究结果。该项研究历时9年，涉及3,252名年龄在50~74岁的妇女，她们均无明显的乳腺或卵巢癌家族史。研究者通过卵巢癌危险度计算公式(ROCA)对受试者进行筛查。Lu博士称ROCA是一种将受试妇女年龄和随时间变化的CA-125检测值纳入其中的数学模型。
她解释说：“在此，存在3种可能性。那些ROCA评分较低的个体被告知1年后返回再次进行CA-125检测。那些ROCA评分居中的个体被告知3个月后返回再次进行CA-125监测。而ROCA评分较高者则被分流进行经阴道超声检查，并转至妇科肿瘤医师处进行进一步诊疗。”
对受试妇女进行9年随访后，研究者发现受试妇女中需要转为每3个月进行CA-125检测者的平均年化比例为6.8%，而需要转为进行经阴道超声检查并转至妇科肿瘤医师处进行进一步诊疗者的平均年化比例仅为0.9%。Lu博士指出：“每年，绝大多数受试妇女被分类为低危险度组别，只需每年进行1次CA-125检测。”
累积下来，共85名(2.6%)受试妇女接受经阴道超声检查并随后转至妇科肿瘤医师处进行进一步诊疗。其中8例需要手术治疗：3例为侵袭性卵巢癌(2例为ⅠC期，1例为ⅡB期)，2例为交界性卵巢肿瘤，3例为良性卵巢肿瘤。根据这些数据转化出的阳性预测值为37.5%。Lu博士称：“这意味着必需进行3台手术才可检出1名侵袭性卵巢癌患者。”
应用ROCA后再进行经阴道超声检查，获得的总特异性可达99.9%，“这意味着几乎不会出现假阳性”。
Lu博士强调，虽然应用ROCA筛查策略的结果令人鼓舞，“但当前尚无法改变实践。我们需要等待一项决定性、应用同一ROCA计算公式、将病死率作为终点的卵巢癌筛查试验的结果” 。后者正在英国进行，受试妇女超过200,000人，预计可在2015年获得结果。
ASCO主席Douglas W. Blayney博士称ROCA计算公式“是个体化医疗的另一例证。现在，我们对这种高度恶性的癌症有了个体化筛查策略。这也说明可对已知技术进行深入挖掘。虽然经阴道超声检查是侵入性的，且从技术上来讲其操作有点困难，但同CA-125检测一样，均已得到广泛应用。如今，我们将二者分阶段使用。”
研究作者之一，Herbert A. Fritsche博士称其接受罗氏诊断公司的研究资金支持。另一位作者，Robert C. Bast Jr博士称担任富吉瑞必欧诊断公司的顾问，并接受其他酬金和版税以帮助研制CA-125检测方法。
在ASCO上发布的超过4,000篇论文摘要中的大部分已张贴在以下网站上：www.abstract.asco.org。
爱思唯尔 版权所有
