Screen­ing de­bate

Can can­cer scans lead to over­diag­no­sis?

Modern Healthcare - - THE WEEK IN HEALTHCARE - Steven Ross John­son

The de­ci­sion last week by the U.S. Preven­tive Ser­vices Task Force rec­om­mend­ing an­nual lung can­cer screen­ings for heavy smok­ers us­ing low-dose com­puted to­mog­ra­phy scans has reignited the con­tro­versy among physi­cians about the trade-off be­tween ben­e­fits and risks for a num­ber of can­cer-screen­ing pro­ce­dures.

Con­duct­ing broad screen­ings for var­i­ous can­cers, such as for the breast and lung, has gen­er­ally been cred­ited for im­prov­ing sur­vival rates by catch­ing the dis­ease in its early stages.

But the prac­tice has come un­der in­creas­ing fire from ex­perts who con­tend screen­ing has been overused. It has led to over­diag­no­sis of pre-can­cer­ous le­sions that may never evolve to full-blown can­cers, and sub­jected peo­ple to un­nec­es­sary biop­sies and treat­ments, the crit­ics say.

“Over­diag­no­sis is a huge is­sue in can­cer and is un­der-rec­og­nized among physi­cians,” said Dr. Otis Braw­ley, chief med­i­cal of­fi­cer for the Amer­i­can Can­cer So­ci­ety. “If it’s un­der-rec­og­nized among physi­cians, you can’t ex­pect it to be some­thing pa­tients ap­pre­ci­ate a great deal.”

Braw­ley said he ap­plauded the rec­om­men­da­tions a group of doc­tors re­cently pub­lished on the Jour­nal of the Amer­i­can Med­i­cal As­so­ci­a­tion’s web­site, which sug­gested redefin­ing how physi­cians di­ag­nose can­cer by ex­clud­ing the term when only pre­ma­lig­nant con­di­tions were found.

In such cases, what’s usu­ally found—known as in­ci­den­talo­mas—are so slow-grow­ing it is un­likely they would ever be­come a health con­cern within a pa­tient’s life­time. This has be­come a ma­jor con­cern for el­derly men di­ag­nosed with prostate can­cer.

Some ex­perts, how­ever, con­tend the ben­e­fits of screen­ing some can­cers out­weigh the risks. The task force es­ti­mated that for ev­ery 320 peo­ple screened, one life was saved in the case of lung can­cer, while one life was saved for ev­ery 900 to 1,900 mam­mo­grams taken in breast can­cer cases.

Screen­ing pro­po­nents say the is­sue of over­diag­no­sis is not real be­cause no doc­tor can defini­tively tell a pa­tient whether a can­cer will progress to be­come life-threat­en­ing. “I think I would be very cau­tious about chang­ing ter­mi­nol­ogy that may con­vey the sense to the pa­tient that they will not de­velop an in­va­sive can­cer when we can’t be sure that they won’t,” said Dr. Larry Nor­ton, med­i­cal di­rec­tor for the Eve­lyn H. Lauder Breast Cen­ter at Me­mo­rial Sloan-Ket­ter­ing Can­cer Cen­ter in New York.

But prob­lems arise when screen­ings pro­duce false pos­i­tives, which oc­curred in 96% of CT lung can­cer tests that ini­tially tested pos­i­tive, ac­cord­ing to the panel’s draft re­port. Fully 24% of the screens for 53,000 heavy smok­ers were pos­i­tive for can­cer. But the 96% false-pos­i­tive rate means 24 out of 25 of those cases would lead to fur­ther tests, ex­pos­ing the pa­tients to more ra­di­a­tion and could lead to more in­va­sive and costly pro­ce­dures that car­ried higher risks of com­pli­ca­tions.

De­spite those con­cerns, the task force still rec­om­mended the use of CT scans, which could re­duce the num­ber of lung can­cer deaths among high-risk pa­tients by 16%. The task force ten­ta­tively gave it a B rat­ing, with a fi­nal rul­ing due in sev­eral months. The Pa­tient Pro­tec­tion and Af­ford­able Care Act re­quires in­sur­ers to pro­vide first­dol­lar cov­er­age for preven­tive ser­vices given an A or B rat­ing by the govern­ment ad­vi­sory body.

While Braw­ley ac­knowl­edged the ben­e­fits of broad screen­ings for some can­cers, he said for oth­ers, such as prostate can­cer, the re­sults have been less en­cour­ag­ing. He es­ti­mated 60% or more of pos­i­tive screen­ings are ac­tu­ally ex­am­ples of over­diag­no­sis that do not re­quire ad­di­tional treat­ment.

Ac­cord­ing to the JAMA study, which was funded by the National Can­cer In­sti­tute, even though screen­ing has in­creased the over­all num­ber of early can­cer de­tec­tions dur­ing the past 30 years, it has not pro­duced a pro­por­tional de­cline in the num­ber of can­cer-re­lated deaths.

The prob­lem has been the as­sump­tion that all le­sions that are de­tected dur­ing a screen­ing have the po­ten­tial to be life-threat­en­ing, said Dr. Laura Esser­man, lead author of the JAMA re­port and di­rec­tor of the Univer­sity of Cal­i­for­nia, San Fran­cisco’s Carol Franc Buck Breast Care Cen­ter. She said such as­sump­tions have led to over­diag­no­sis in 20% to 50% of can­cer cases.

With the added health risks that can re­sult from an over­diag­no­sis comes the added cost of treat­ing a pa­tient.

Braw­ley and Esser­man said the costs re­lated to over­diag­no­sis pos­si­bly amount to bil­lions of dollars.

As more is learned about the dis­ease, Esser­man said she felt con­fi­dent more physi­cians would be­gin to de­velop a more nu­anced ap­proach in the way they ap­proached a can­cer di­ag­no­sis.

“We have to work hard to ed­u­cate the pub­lic to let them know that we’re not play­ing Rus­sian roulette with their life—that it’s com­mon to have con­di­tions that are not se­ri­ous,” Esser­man said. “Just like when you’re try­ing to test a new treat­ment to see if you can im­prove the chance of sur­vival, we have to start test­ing how we can safely do less.”

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.