Yahara et al. have implemented a new method for quantifying bone metastases in patients with prostate cancer, as an indicator of the response to therapy and disease progression. Bone lesions are outlined and the %PABS quantified using a software package. This technique was compared with counting the number of lesions (extent of disease or EOD) and PSA in 42 patients after androgen deprivation therapy. A decline of ≥ 25% in %PABS correlated with longer patient survival.

The authors should be applauded for seeking new means to evaluate outcome in a group of patients most commonly entered into clinical trials because they are generally incurable. PSA has not historically been an accepted endpoint, but rather clinical findings, bone scintigraphy and survival are better acknowledged. Furthermore, while PSA is regulated through its androgen response element by both androgen and other factors, its level of expression is not always directly correlated with prostate cancer growth.

However, I question several points; the authors suggest that quantifying EOD is not simple, although counting the number of metastatic lesions would seem to be simpler than outlining each lesion by hand on a graphic computer program to measure it. This apart, evidence is cited for less variance with %PABS than with counting EOD. There was a strong correlation between EOD grade and %PABS at diagnosis, but neither correlated well with PSA level. PSA was not compared with the imaging methods for survival, which leaves the reader not knowing whether PSA is still superior for predicting survival. The definition and time-point selected for the overall response will show discordance between PSA and bone scans, as the improvement in bone scan will lag behind the rise in PSA as the patients’ disease progresses. This is obvious in Figs 2A-C. We evaluated PSA and bone metastases, and reported that the proportional decrease in PSA level at 6 months is the best determinant of survival in patients with metastatic prostate cancer [1]; it correlated more closely than EOD. Furthermore, EOD has been shown in numerous studies to only correlate with survival when separating Grade 1 from Grade 2–4.

The present clinical utility of %PABS thus appears premature, especially in patients on intermittent androgen deprivation or receiving bisphosphonates. Hormone-refractory patients on protocol might be the best group to study further and more validating studies will help to define an appropriate role for this type of quantitative imaging technique.

  • 1
    Evans CP, Gange S, Khalid A, Eason A, Narayan P. The proportional decrease in prostate specific antigen level best predicts the duration of survival after hormonal therapy in patients with metastatic carcinoma of the prostate. BJU Int 1996; 78: 42631