It is time to stop the routine use of MRI before surgery for non-high-risk breast cancer, according to an editorial published online January 6 in the Journal of Clinical Oncology.
The use of MRI may be more "contentious or emotionally debated" than any other issue in modern breast care, writes editorialist Richard Bleicher, MD, from the Fox Chase Cancer Center in Philadelphia.
He asserts that routine preoperative MRI is not appropriate because the evidence now includes a meta-analysis, published online along with the editorial, that found that the imaging does not reduce either the local ipsilateral or distant return of disease.
The findings about recurrence are important, in part, because there is "little suggestion, even among strong routine-MRI advocates," that MRI improves survival, Dr. Bleicher says.
The meta-analysis adds to the "growing body of data that suggests we are unlikely to ever see...improvement in outcomes by using the modality routinely," he writes.
In today's cost-conscious heathcare world, "we must demonstrate improved outcomes...to justify costs," Dr. Bleicher asserts.
In the meta-analysis, an international team of researchers looked at more than 3000 women with non-high-risk disease in previously published studies. They found that the 8-year local recurrence-free survival did not differ between the MRI and no-MRI groups (97% vs 95%; P = .87).
In addition, the 8-year distant recurrence-free survival did not differ between the MRI and no-MRI groups (89% vs 93%; P = .37).
The study, led by Nehmat Houssami, MD, from the Sydney Medical School in Australia, is the first meta-analysis to compare local and distant recurrence in women who had a preoperative MRI and those who did not; all of the women underwent an attempt at breast conservation for non-high-risk disease.
There are no randomized controlled trials that examine MRI and recurrence, the researchers point out.
The meta-analysis is also an "effort to consolidate and more definitively investigate a field of data that is comprised of small retrospective studies and prospective trials," Dr. Bleicher explains.
The researchers do not directly state that routine use of MRI in this setting is not recommended. Instead, they call the results "the best available evidence on the association of MRI and local recurrence (or distant recurrence)," and believe it "should be used to guide or change clinical practice."
MRI Has a Role to Play
The results are seemingly vexing, Dr. Bleicher notes, because of the power of this high-tech imaging. MRI is "more accurate for judging lesion size and more sensitive for detecting otherwise subclinical cancer foci," he writes.
But these attributes might not ultimately matter clinically, he explains.
Most significantly, the routine use of radiation therapy and systemic agents after surgery do a solid job treating the multiple foci that are so common in breast cancer.
"These foci have been treated reasonably well for decades, long before we could so easily see them with MRI," Dr. Bleicher writes, adding that it is frightening to actually see the extra cancer on film. The high-tech images are "detailed and beautiful."
MRI use in breast cancer appears to be "dramatically increasing," Dr. Bleicher said in an email toMedscape Medical News. For instance, at his center, 13% of breast cancer patients had an MRI in 2004; in 2006, 27% did (he noted, however, that these numbers are not broken out by risk).
MRI has a role to play in breast disease, Dr. Bleicher emphasizes. Clinicians should focus its use for "entities such as Paget's disease, occult primaries, and in deleterious BRCA mutation carriers, where it has been demonstrated to be definitively advantageous," he writes.
Challenges Expected
The meta-analysis researchers — from Australia, Canada, the United Kingdom, and the United States — selected 4 studies that met eligibility requirements, including the need to have MRI and no-MRI cohorts. The team used individual patient-level data for the 3169 women in the studies instead of pooled data to reduce the possibility of publication bias.
"Individual patient data meta-analyses tend to be better than using pooled data because they utilize the specific granular data and combine it all together," Dr. Bleicher explains.
In addition to using survival analysis to investigate recurrence, the researchers used the same analysis to estimate the hazard ratio (HR) for MRI. Again, they found no significant advantage with MRI over 8 years.
In terms of local recurrence, the HR for MRI (vs no MRI) was 0.88 (95% confidence interval [CI], 0.52 - 1.51; P = .65). Age, margin status, and tumor grade were associated with local recurrence-free survival (P < .05 for all).
In terms of distant recurrence, the HR for MRI (vs no MRI) was 1.18 (95% CI, 0.76 - 2.27; P = .48).
Sensitivity analyses were also performed for both types of recurrence, and were based on women who had breast conservation with radiotherapy. Neither analysis indicated a significant benefit with MRI.
Dr. Bleicher expects that these results will be challenged because the researchers were unable to obtain data from all eligible studies and the follow-up was not long.
"These caveats suggest that continued evaluation of the modality remains appropriate," he writes. "But considerable data on MRI to date have not shown the promise in the routine preoperative setting for which many have hoped."
Dr. Bleicher also expects quibbles about the inclusion of older studies in the analysis. "One could be dismissive of this meta-analysis for including studies using older MRI machines, for having radiologists who may not be so-called MRI experts, and for localizations that might have been suboptimally performed," he writes.
But this makes the analysis more a real-world study, he notes. "These studies likely reflect the widespread reality of the technology and expertise as it has recently existed or currently exists."
The study was supported in part by National Health and Medical Research Council Program, a National Breast Cancer Foundation Practitioner Fellowship, and Yorkshire Cancer Research. The authors and Dr. Bleicher have disclosed no relevant financial relationships.
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου