Note to our followers: Our nearly 13-year run of daily publication of new content on HealthNewsReview.org came to a close at the end of 2018. Publisher Gary Schwitzer and other contributors may post new articles periodically. But all of the 6,000+ articles we have published contain lessons to help you improve your critical thinking about health care interventions. And those will be still be alive on the site for a couple of years.

Ovarian cancer screening: shop for the story emphasis of your choice

Let’s take a brief spin through the choices that news consumers (and patients) can make among a range of stories about a new study on ovarian cancer screening.

Should the emphasis be:

  • “Early Detection of Ovarian Cancer May Become Possible,” as the New York Times headlined it.  Although it followed with strong cautions:

“We need to follow up to confirm that this is absolutely significant throughout,” said Dr. Usha Menon, an author of the Lancet article and head of the gynecological cancer center at University College London. She said, “This is almost there, but not yet.” Her reason for caution was that the study passed only two of three tests of statistical significance, which means that the apparent benefits of screening might have been due to chance. She said a few more years of monitoring the participants would resolve that question.

  • “This Formula Might Help Catch Ovarian Cancer Earlier,” reported NBC News, followed by the “first real hint that there might be a way to screen women for ovarian cancer,” though it is “not ready for prime time yet.”

“At first blush, it seemed that screening failed again: women receiving CA-125 plus ultrasound had a 15 percent lower risk of dying from ovarian cancer than unscreened women from 2001 through 2014 — 29 percent versus 34 percent. But the benefit was so small it could have been due to chance. That is, it was not statistically significant. But digging deeper turned up something more encouraging.”

 

Screen Shot 2015-12-18 at 8.40.14 AMAnd then there was this headline – different than all the others:  “Ovarian Cancer Screening Study Falls Short: Significant mortality benefit only in subgroup analysis.”  That was the MedPage Today story.  Headlines matter.  They lure us in and set us up for what’s to follow. And this headline hit harder than any other that we saw.

It also matters whom you choose to interview in the body of the story.  MedPage Today’s first expert quote was more skeptical than those that appeared first in most other stories:

“The results will do little to inform the debate on screening average-risk women, said Don Dizon, MD, of Massachusetts General Hospital Cancer Center and a clinical expert for the American Society of Clinical Oncology.

“I’m underwhelmed by the results,” Dizon told MedPage Today. “I think the summary of the study that was distributed in advance was a bit misleading. It’s a hopeful study, regarding the benefits of screening, but the picture is still incomplete. If anything, it should spur on research, but it is by no means a green light to start screening the general population.”

What I’ve just outlined is part of the roulette that we play every day in perusing health care news.  Where will the ball land? Which news story will you happen to see?  And how will its framing compare with everything else out there? And while it’s not a game – not when people may make choices based on what they read in the news – the roulette analogy does hold up if people do actually make choices based on what they read in the news.

Back to the CBS story at the top.  Its first line was:  “A simple blood test for ovarian cancer can save women’s lives.”  I always say:  whenever you hear someone say “simple blood test,” run for the hills because there’s no such thing.  That should be clear by now.

You might also like

Comments (4)

Please note, comments are no longer published through this website. All previously made comments are still archived and available for viewing through select posts.

Joe Rojas-Burke

December 18, 2015 at 1:25 pm

Only a few stories made clear that for each cancer detected, two women had false-positive surgery in this trial, i.e., their healthy ovaries were removed. The screened population more than doubled their risk of false-positive surgery:

“During the course of this trial, the ratio of women who had surgery for which ovaries had benign pathology or were normal to those diagnosed with ovarian and peritoneal cancer was 2.3-times higher in the MMS group…than in the no screening group,” noteds the authors of The Lancet paper.

Many of the media reports also implied that CA-125 is a new idea, and didn’t bother to note the long and disappointing history of this biomarker since its discovery back in the 1980s.

But most surprising and disappointing was the failure to even mention overdiagnosis – which is *not* the same as false-positive error. Screening tests inevitably detect some tumors that do not need to be diagnosed. By definition, an overdiagnosed cancer is one that was destined to never cause harm because it is slow-growing, for instance, or unable to metastasize. After all we’ve learned from PSA testing and mammography, I thought by now it would be de rigueur to ask questions about overdiagnosis when reporting on cancer screening tests.

And you know, this study only addressed disease-specific mortality, i.e., deaths from ovarian and peritoneal cancer. We have no idea how this screening test affects overall mortality.

Paul Alper

December 21, 2015 at 8:53 am

The complete Lancet article is found at
http://www.thelancet.com/pdfs/journals/lancet/PIIS0140-6736(15)01224-6.pdf
It has over 40 authors. From page 11 of the study, I counted nine who listed some sort of conflict of interest pertaining to the presumably proprietary and rather mysterious algorithm known as ROCA, Risk of Ovarian Cancer, and its connection to Abcodia. “a specialist company that focuses on biomarkers for cancer screening.”
https://en.wikipedia.org/wiki/Abcodia

Juan Gérvas

December 29, 2015 at 6:01 am

-Joe Rojas-Burke, you can go to the supplementary appendix and found more information, even about total mortality and total cancer mortality; the screening has no impact at all
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736%2815%2901224-6/supplemental
-they noted no evidence of a difference in deaths because of other causes between the MMS, USS, and no screening groups; total deaths, 6.7 per 1,000 MMS, 6.4 USS and 6.6 in the women who were not screened; total cancer mortality: 3.4 per 1,000 MMS, 3.3 USS and 3.3 in the women who were not screened
-a summary of the main results:
-the RCT ends with not significant impact in mortality, BUT a significant mortality benefit in a subgroup analysis (most of the mortality benefit occurred during the later years of follow-up: 8% during years 0 to 7 versus 23% during years 7 to 14 in the multimodality group and 2% versus 21% in the ultrasound group (what it was unexpected)
-the data showed that 338 (0.7%) women in the multimodality arm had ovarian cancer, 314(0.6%) in the ultrasound arm, and 630 (0.6%) of the women who were not screened; 148 (0.29%) women in the multimodality arm (MMS) died of ovarian cancer, 154 (0.30%) in the ultrasound arm (USS), and 347 (0.34%) of the women who were not screened.
-true positive were 199 (59%) in the multimodality arm and 161 (51%) in the ultrasound arm
-false negative were 79 (23%) in the multimodality arm (MMS) and 106 (34%) in the ultrasound arm (USS)
-screening-related complication rate of 8·6 per 100,000 in the MMS group and 18·6 per 100,000 in the USS group
-they noted no evidence of a difference in deaths because of other causes between the MMS, USS, and no screening groups; total deaths, 6.7 per 1,000 MMS, 6.4 USS and 6.6 in the women who were not screened; total cancer mortality: 3.4 per 1,000 MMS, 3.3 USS and 3.3 in the women who were not screened
-for each ovarian and peritoneal cancer detected by screening, an additional two women in the MMS group and ten women in the USS group had false-positive surgery
-women in the MMS group had a complication rate of 3·1% and those in the USS group had a rate of 3·5%
-the ratio of women who had surgery for which ovaries had benign pathology or were normal to those diagnosed with ovarian and peritoneal cancer was 2·3-times higher in the MMS group and 5·3-times higher in the USS group than in the no screening group
-so, my question: could you recommend a screening test which has no impact on ovarian cancer mortality, no impact on cancer mortality, no impacto on total mortality, and has heavy harms)
-un saludo juan gérvas @JuanGrvas

Juan Gérvas

December 29, 2015 at 11:34 am

-in my summary, a correction
-they noted no evidence of a difference in deaths because of other causes between the MMS, USS, and no screening groups; total deaths, 66.9 per 1,000 MMS, 64.4 USS and 65.7 in the women who were not screened; total cancer mortality: 33.9 per 1,000 MMS, 32.89 USS and 33.4 in the women who were not screened
-un saludo juan gérvas @JuanGrvas