In the article entitled, "Permanent Prostate Brachytherapy: Is Supplemental External-Beam Radiation Therapy Necessary?" Merrick and coauthors address the prostate cancer controversy, "Which treatment is best—and for whom?" Whereas brachytherapy has assumed its proper place on the menu of options currently available to treat localized prostate cancer, the selection of optimal candidates for this modality remains controversial. With the many therapeutic choices available, it would behoove us to rely on the evidence-based approach whenever possible, as a lack of data can lead us to make decisions based on personal preference and opinion.

Ends of the Risk Spectrum

With regard to prostate brachytherapy, there is little contention about treatment choices for patients at the ends of the spectrum—ie, those at lowest and highest risk. Several studies have shown that good quality brachytherapy alone for patients with low-grade, low-stage, and low-prostate-specific antigen (PSA) prostate cancer consistently produces good results, equivalent to those achieved with both external-beam radiation therapy (XRT) and radical prostatectomy. Most experts agree that there is no need for the potential added financial cost and medical morbidity of supplemental XRT in this group of patients.

At the other end of the risk spectrum, in men with advanced-stage, high-grade, and high-PSA tumors—who have a significant risk of seminal vesicle and lymph node involvement—there is a need for a more regional approach, as suggested by the authors, as well as for a more systemic approach, as evidenced by studies documenting the benefit of hormonal therapy in this patient group. Therefore, patients in the intermediate-risk category remain the most controversial with regard to treatment approach.

Dilemma of the Intermediate Group

Part of the dilemma in dealing with this middle group is the fact that it is extremely heterogeneous. Using a large external-beam dataset with long-term follow-up, Thames and colleagues showed that the classic National Comprehensive Cancer Network-defined intermediate-risk group contained patients with very different outcomes. On the other hand, they found that recursive partitioning analysis chose multiple subcategories that grouped patients with like biochemical and clinical disease-free survival more homogeneously.[1] This work also showed that risk groupings were very dependent on dose: The higher the dose, the more adverse the risk factors associated with the same outcome. Analysis of a large brachytherapy dataset with long-term follow-up indicated that the quality of the implant, as represented by the amount of radiation received by 90% of the prostate volume (D90), does indeed have an impact on outcome, as Merrick et al suggest.[2]

There are various intermediate-risk group definitions, such as the one used by these authors (Gleason score ≥ 7, PSA ≥ 10 ng/mL, or clinical stage ≥ T2c). Unlike some other risk-group stratifications, more than one of these factors would put patients into the high-risk group, removing those with more adverse features. While some patients considered to have intermediate risk by this definition—for example, a patient with stage T1c, Gleason score = 7, PSA level = 5, and one biopsy core showing tumor—would likely do well with implant alone, do we really know where to draw that line? Do we have the evidence to feel certain that brachytherapy alone can measure up for the entire group of intermediate-risk patients?

Opposing Viewpoint

The authors present reasons for a positive response to these questions but do not objectively discuss the evidence for the opposite viewpoint. The potential for compromised outcome with brachytherapy is largely attributed to inadequate technique, which leads to less-than-ideal dose and dose distribution, especially when extracapsular margins are considered. While this may be a contributing cause, convincing evidence that it is the only factor related to outcome and that it can be overcome is lacking.

Figure 5 is used to illustrate the authors’ opinion that high-risk patients can also be treated with implant alone, if the technique is correct. However, I interpret this graph as showing that the two groups not receiving supplemental XRT fared considerably worse than the two groups who did receive such therapy, although the patients with margins and no XRT certainly did better than those without either margins or XRT.