SEARCH

SEARCH BY CITATION

Sir,

We read with interest this article by Braeckman et al.[1] about the use of HistoScanning for locating and characterizing prostate cancer, but some points merit critical analysis. First, in our opinion the term ‘HistoScanning’ is misleading. HistoScanning can detect specific changes in the tissue morphology by extracting and quantifying statistical features from back-scattered ultrasonographic data, which might further allow differentiation between benign and malignant tissue. However, for the diagnosis of prostate cancer a traditional important prognostic factor is the Gleason score, determined by histology, which is very important for prostate cancer grading [2]. Unfortunately HistoScanning sounds if it might offer an exact histological assessment, which if based on the present data is untrue.

Second, the authors describe that characterization algorithms used for HistoScanning exploit the physical changes to sound waves that result from the interaction of the ultrasound beam and the cancer tissue. As the axial and lateral resolution of ultrasonography depend on the transmitting frequency, the authors failed to state what frequency was used for the TRUS studies.

Third, in patient no. 21, prostate cancer was diagnosed by histology of tissues removed by TURP for BPH. HistoScanning was able to detect multiple cancer foci in this case, where the largest tumour had a diameter of 3.2 mm on HistoScanning, and 2.0 mm on histopathology. As 3.2 mm was the largest tumour in this case, and the other detected tumours were smaller, it would be interesting to know what was the minimum tumour size detected by HistoScanning. An ultrasound probe with a central frequency of 5.0 MHz, which is commonly used for TRUS, allows a maximum axial resolution of 0.35 mm with a lateral resolution of 1.2 mm. Therefore we are unsure how HistoScanning can detect tumours of 1–2 mm with complete accuracy.

Fourth, the authors failed to describe how many cancers were detected in the peripheral (PZ) and/or transition zone (TZ). Compared with PZ cancers, TZ cancers show a different anatomical pattern of positive biopsies, with lower rates in the middle and base of the prostate. Furthermore, the correlation between needle biopsy Gleason scores and those of the radical prostatectomy specimens was clearly lower in TZ cancers. Therefore, to know the rate of PZ and/or TZ cancers would add important additional information [3]. Furthermore, the prostate volume of the patients is missing, and it is well known that cancer detection is more difficult in larger glands [4]. Therefore based on these limitations, the value of HistoScanning for prostate cancer detection is questionable.