CEO SUMMARY: Based on an impressive number of 4,230,129 vials collected from 437,937 biopsies, the new study is expected to add fuel to the fire of the ongoing debate about the appropriate number of prostate biopsies physicians should collect and refer to pathology labs for cancer testing. The researchers compared the number of prostate biopsies sent to a national reference laboratory with the number collected by urologists and self-referred to their own in-office pathology labs.
UNDER ATTACK FROM MANY QUARTERS because of alleged patterns of utilization from their in-office pathology labs, researchers collaborated with the Large Urology Group Practice Association (LUGPA) to conduct their own study into this matter. The results, endorsed by three urology professional associations, were announced last month.
If numbers matter, then the study, titled “Utilization and cancer detection by U.S. prostate biopsies (2005-2011),” should have credibility. It involved 4.2 million specimens collected from 440,000 prostate biopsies.
The study’s findings were presented at a poster session during a symposium conducted last month in Orlando, Florida, by the American Society of Clinical Oncology Genitourinary Cancers.
In their abstract of the study, the authors boldly concluded, “The increased cancer detection rate correlated significantly with the increased number of specimens examined. Segregation of prostate biopsy cores into 10-12 unique specimen vials has been adopted by urologists across sites of service and can be considered the de facto national standard of care.”
At the heart of the matter is the 20-year debate over what number of prostate biopsy cores represent a standard of care that contributes to optimal diagnostic accuracy. Community pathologists, national pathology lab companies, and urologists have all weighed in at different times on this issue. The debate has often been rancorous because of the differing opinions.
“Identifying the proper number of cancer appropriately is a patient care issue that is at the heart of the dispute,” stated Deepak A. Kapoor, M.D., in an exclusive interview with THE DARK REPORT. A board-certified urologist, Kapoor is President of the Large Urology Group Practice Association, and Chairman and CEO of Integrated Medical Professionals (IMP), in Melville, New York.
Integrated Medical Professionals is a multispecialty practice serving Long Island, parts of New York City, and Westchester and Rockland counties in New York. One of the largest practices of its kind in the United States, IMP runs one of the nation’s largest in-office pathology laboratories. It has 106 physicians and six multi-purpose outpatient treatment facilities.
At the 2013 Genitourinary Cancers Symposium last month in Orlando, Florida, Kapoor and other experts on this issue presented research they conducted to assess the positive biopsy rate and core sampling pattern in patients. The researchers collected data on prostate biopsies from a national reference laboratory and from pathology laboratories integrated into urology group practices. The research produced two significant results, Kapoor said.
Number of Specimen Vials
“The first significant result involved the relationship between positive biopsy rates and the number of specimen vials per biopsy (sv/b),” observed Kapoor. “In a subsequent analysis of similar data, our research team hopes to determine the appropriate number of specimen vials per biopsy needed to identify cancer in prostate patients.
“We believe our study is the largest of its kind ever done,” he declared. “It involved more than 2,000 urologists who collected 437,937 biopsies and 4.2 million cores over six years.”
In addition to Kapoor, the researchers included three of his IMP colleagues: Carl A. Olsson, M.D., Lattimer Professor and Chairman of Urology, emeritus, at the Columbia University Medical Center; and two pathologists, Savvas E. Mendrinos, M.D., and Ann E. Anderson, M.D. Also participating and providing data from a national lab company was David G. Bostwick, M.D., Chief Medical Officer of Bostwick Laboratories in Uniondale, New York.
The researchers collected data on the total number of specimen vials submitted per prostate biopsy and the final diagnosis for each case from urologists and urology practices referring samples to a national reference laboratory (NRL). Over the same period, they also collected similar data from urology practices with in-house laboratories performing global pathology services.
Positive Biopsy Rates
For each year studied, the positive biopsy rate and number of specimen vials per biopsy were calculated in aggregate and separately for each site of service, according to an abstract presented at the symposium.
The results showed that from 2005 to 2011, 437,937 biopsies were submitted in 4.2 million vials (meaning 9.4 specimen vials per biopsy or sv/b). The overall positive biopsy rate of 40.3% was identical at both the NRL and in-office pathology labs, the abstract said. Interestingly, the results showed urologists tended to collect more specimens in the last three years of the study, when the sv/b rate increased sharply from a mean of 8.8 during 2005 to 2008 to 10.3 from 2009 to 2011, the researchers reported.
“The reason urologists collected more specimens per biopsy over the last three years studied than they did in the previous four years resulted from changes in urologists’ practice patterns,” explained Kapoor. “This is after they saw a correlation between a positive biopsy rate and sv/b.
“In 2005, the average number of cores per biopsy sent to the NRL was 7.2 cores, and the positive biopsy rate was 38%,” he continued. “By 2011, the number of cores per biopsy was about 10 to 11 cores, and the positive biopsy rate was 42.5%.
“When doing a prostate biopsy, the goal is to optimize the yield while minimizing costs and complications,” added Kapoor. “Therefore, it is significant that there was a near linear correlation between the number of biopsy cores collected and the positive biopsy rate.
“When the physician collects seven to eight biopsy cores, the positive cancer rate is about 38%,” he noted. “The collection of 10 to 11 cores is associated with positive cancer rates of about 41% to 42%.
“That is a difference of about 10%, which is significant in terms of patient care because it may help us identify the appropriate rate of biopsy specimens to collect,” Kapoor explained.
“This issue will be the subject of another study,” he said. “Through extrapolation, we want to determine that a specific number optimizes the yield in terms of positive biopsy versus cores collected. I believe that number will almost certainly be somewhere between 10 and 12.”
More Research Needed
Kapoor pointed out that there was a second significant result from the study. This result involved the number of specimens sent to the national reference laboratory compared with the number of specimens reviewed at in-office laboratories. “From 2009 to 2011, the sv/b rate for samples sent to the NRL was 10.0, which was statistically identical to the rate of 10.6 sv/b done at in-office labs. The notion that physician ownership drives utilization of pathology services is not sustainable since positive biopsy rates as well as numbers of sv/b are so similar across sites of service,” Kapoor said.
Research by Kapoor and colleagues was reported only in abstract form at the symposium. But a peer-reviewed medical journal is reviewing the full study and Kapoor hopes the full manuscript will be published soon.
Battling Studies over Biopsy Utilization Rates
FOR PATHOLOGISTS AND LAB EXECUTIVES watching this ongoing battle over what number of cores is appropriate for prostate cancer testing, the release of this study by three national urology associations is the urologists’ turn at bat.
This study, more than 10 times larger and covering over twice as long an interval, is designed to answer the findings of a study published last year in the journal Health Affairs. Titled, “Urologists’ Self- Referral For Pathology Of Biopsy Specimens Linked To Increased Use And Lower Prostate Cancer Detection,” the Health Affairs article reported that self-referring urologists billed Medicare for 4.3 more specimens per prostate biopsy (s/pb) when compared with the adjusted mean of six s/pb that non-self-referring urologists sent to independent pathology providers—a difference of almost 72%.
In Health Affairs, the study authors wrote that, “Additionally, the regression-adjusted cancer detection rate in 2007 was 12 percentage points higher for men treated by urologists who did not self-refer. This suggests that financial incentives prompt self-referring urologists to perform prostate biopsies on men who are unlikely to have prostate cancer. These results support closing the loophole that permits self-referral to ‘in-office’ pathology laboratories.”
The Health Affairs study had its critics who pointed out that the funding for the research had been provided by the College of American Pathologists (CAP) and the American Clinical Laboratory Association (ACLA), two organizations which represent those pathologists who would prefer to see the in-office pathology laboratories owned by specialty physicians go away.
As well, there were criticisms about the Medicare data sets chosen by the study’s author, as well as criticisms about the design and the methodology of the study itself.