Accelerated partial-breast irradiation: does the evidence stack up?

With regard to multicatheter interstitial brachytherapy (MIB), the review notes high infection rates, citing rates of 0% to 11%. However, an analysis found an infection rate of 10.3% with whole-breast irradiation (WBI), comparable to that noted in the MIB series.[1] In addition, several clinical studies with large numbers of patients and long-term follow-up (including a matched-pair analysis of MIB and WBI at 12 years) found no differences in outcomes with brachytherapy.[2-4] When discussing single-entry intracavitary brachytherapy catheters, the review focuses on two recent claims-based analyses; the first study, by Smith et al, suggested that brachytherapy-based APBI was associated with higher rates of mastectomy as well as infectious and noninfectious complications compared with WBI.[1] A second claims analysis by Presley et al found higher rates of wound and skin complications with brachytherapy-based APBI.[5] Significant limitations to the above studies prevent them from being practice changing: they were observational retrospective analyses with short follow-up (3.8 years and 1 year, respectively), and both had the potential for treatment-era bias (2003–2007); they also used Medicare data, which allowed for misclassification bias (ie, mastectomy coding cannot differentiate which breast was removed). Further, a billing code was used as a surrogate for clinical outcome; and no or limited information was provided on clinical factors (stage, histology, systemic therapy, etc), the quality of initial surgery (ie, margin status), and comorbidities.[6,7] In addition, it is believed that observational studies are especially prone to methodological and statistical biases that can render results unreliable.[8] In support of this opinion, Dr. Stanley Young of the US National Institute of Statistical Sciences stated that one can “troll the data, slicing and dicing it any way you want...a great deal of irresponsible reporting of results is going on.”[8] Other major issues with such studies are the assumption that association is equivalent to causality, and the lack of reproducibility. For example, two recent observational studies that used the exact same data set came to completely opposite conclusions.[8] Finally, it is interesting that one of the coauthors of the Smith et al study has previously published data demonstrating that observational studies can lead to “improbable results” and concluded that “results from observational studies of treatment outcomes should be viewed with caution.”[9] On the other hand, actual clinical data from prospective studies have found the infection rate with brachytherapy-based APBI to be < 5% and the noninfectious complication rate to be lower than that presented in either claims analysis.[10] Clearly, the majority of prospective clinical data confirm the safety and efficacy of APBI. These highly flawed observational analyses should not discourage treatment with APBI on and off protocol, particularly in light of interim analyses of the Groupe Europen de Curiethrapie and European Society for Therapeutic Radiology and Oncology (GEC-ESTRO) and National Surgical Adjuvant Breast and Bowel Project (NSABP) B-39 phase III trials, which found no such concerns with respect to toxicity or mastectomy rates for patients treated with APBI. When discussing external beam–based APBI, the review focuses on the preliminary results of the RAPID trial (as yet unpublished data) and summarizes the discussion by noting that “these findings should be concerning to those considering offering this treatment to patients.” It must be pointed out that the RAPID trial used no boost in the WBI arm (a major factor affecting cosmesis) and further,