In addition to picking the MD Anderson study apart, APBI proponents at the press conference criticized researchers for bias and failure to incorporate relevant current information into their data. The MD Anderson study also used data that were up to 10 years old and did not reflect recent advances in APBI technology.
APBI proponents are fearful that results from the MD Anderson study could discourage many women from participating in clinical trials, particularly a large randomized study currently under way that's designed to answer many of the questions regarding which breast cancer patients are best suited for APBI.
The controversy erupted at the San Antonio Breast Cancer Symposium (SABCS), which took place December 6 to 10, when MD Anderson researchers presented data indicating that breast cancer patients had worse outcomes if they received brachytherapy-based APBI instead of whole-breast radiotherapy. The study claimed that APBI brachytherapy led to higher rates of late mastectomies, radiation-related toxicities, and other complications.
A number of professional organizations quickly rose to APBI's defense, including the American Brachytherapy Society (ABS), the American Society for Radiation Oncology (ASTRO), and the American Society of Breast Surgeons (ASBS). The counteroffensive culminated in Tuesday's audioconference in which four experts on APBI explained the shortcomings of the MD Anderson study to journalists.
The four speakers criticized the MD Anderson study as "flawed" and "biased," and pointed out what they said were several shortcomings:
- Researchers selectively presented findings from uncorroborated data that were incomplete and therefore inconclusive.
- They used obsolete information relating to treatment costs.
- They did not adhere to the rigorous research standards for which MD Anderson studies are generally known.
Dr. Rakesh Patel, chairman of the ABS and director of breast cancer services at Western Radiation Oncology, chaired the press conference. He was joined by Dr. Peter Beitsch, co-principal investigator of the ASBS MammoSite Registry; Dr. Jayant Vaidya, PhD, a pioneer in the use of intraoperative breast radiotherapy and principal investigator of several clinical trials using the technology; and Dr. Robert Kuske, co-principal investigator of a clinical trial sponsored by the U.S. National Cancer Institute (NCI) comparing five-day APBI to six-week whole-breast irradiation.
Kuske, a radiation oncologist at Arizona Breast Cancer Specialists, has 20 years of experience with partial-breast irradiation technology and spared no punches with his criticism. He said that APBI is one of the most studied medical treatments in history, with nine clinical trials alone evaluating its performance. The MD Anderson study was a backward look at breast cancer patients whose disease characteristics and general health were unknown, he said.
Unlike in clinical trials, no controls were used to make a comparison of outcomes accurate and unbiased. Kuske pointed out that the MD Anderson researchers, led by radiation oncologist Dr. Benjamin D. Smith, did not know if more patients treated by APBI had aggressive tumors, positive nodes, cancer cells at the surgical margins, or a history of smoking, or whether they were diabetic or more obese than their non-APBI counterparts. Any of these factors can negatively affect outcomes.
Kuske also referenced registry data and clinical trials showing that recurrence, post-treatment complications, and toxicities are comparable between the two treatments. These studies matched patients by age, cancer grade, tumor size, and other important factors, something the MD Anderson study did not do, he said.
The reason for a higher mastectomy rate in the study could be attributed to many other reasons, including testing positive for the BRCA gene at a later date, having a cancer develop in an untreated quadrant of the breast or the opposite breast, having a false-positive breast MRI exam, or "simply out of fear generated by poor studies like this one," he said.
"The consequences of misinformation from this analysis are potentially damaging," Kuske said. "If the NCI randomized clinical trial [NSABP B-39/RTOG 0413] fails to reach its goal of enrolling 4,300 breast cancer patients because of this misinformation, this would be a tragedy, because real answers to all of our questions will come out of this scientifically pure study."
Findings don't match TARGIT-A
Vaidya is the principal investigator of Targeted Intraoperative Radiotherapy Versus Whole-Breast Radiotherapy for Breast Cancer (TARGIT-A), a 29-country multicenter clinical trial with four-year findings published in June. TARGIT-A compares the outcomes of patients who receive a single intraoperative radiotherapy (IORT) treatment while in an operating room immediately following a lumpectomy to the outcomes of patients who receive conventional whole-breast irradiation.
Vaidya said that 2,232 patients have undergone IORT to date in TARGIT-A. Of these, approximately 15% also had whole-breast irradiation because postsurgical pathology analysis determined that they were at higher risk. Randomized evidence showed that local recurrence and toxicities four years after treatment for these patients were similar, and that later studies have found better patient satisfaction with IORT compared with external beam radiotherapy.
Beitsch, a breast oncology surgeon affiliated with the Dallas Surgical Group, criticized the MD Anderson study's failure to explain hospitalization rates for either group. He said that after attending the SABCS session where the results were presented, he contacted many breast surgeons and radiation oncologists, including some who worked at MD Anderson, to ask if any of their patients had been hospitalized as a result of APBI treatment. None had been.
Vaidya also expressed concern about the study's hospitalization statistics, noting that APBI may be the only realistic treatment for older, unhealthy patients considered unable to undergo a six- to seven-week regime of daily radiation. Both Vaidya and Beitsch suggested that hospitalization rates for the patient cohort could be entirely unrelated to treatment.
Beitsch also criticized the fact that while the MD Anderson researchers were willing to provide data about rates of breast pain and fat necrosis, they did not -- and, in fact, could not from available data -- define it. Infection is understandable, he observed, but the researchers did not associate billing codes for infections with breast cancer treatment itself. When did the infections occur? Why? The researchers never presented this information, and they probably could not because the study was an analysis of billing codes incurred from medical treatment between 2000 and 2007 of breast cancer survivors older than 66.
In addition, the study presenters failed to note that APBI itself has dramatically improved since 2007. Much improved treatment plans are used today for APBI, compared with the "obsolete" ones used during the MD Anderson study period, Beitsch emphasized. This omission and other factors "seemed to be part of the presenters' overall strategy to call partial-breast irradiation inferior and inefficacious. There was certainly a spirit of that during their talk."
Patel also noted that the presenters had said APBI was twice as expensive as whole-breast irradiation. "While the cost was higher then, it isn't now," he said. "Since 2009, the Medicare rates of reimbursement show that APBI is the more cost-effective and less-costly treatment option for breast cancer patients who qualify to receive it."
Kuske was blunter.
"I was particularly disappointed that when interviewed after their presentation, the MD Anderson team members would state something that is categorically false," he said. "They know APBI is less expensive than the whole-breast irradiation treatment they perform in Houston. To state this using 10-year-old information makes me as a physician doubt their motives for presenting their data."