Predictive value of PSA velocity over early clinical and pathological parameters in patients with localized prostate cancer who undergo radical retropubic prostatectomy

A Weak and Semi-supervised Segmentation Method for Prostate Cancer in TRUS Images

The purpose of this research is to exploit a weak and semi-supervised deep learning framework to segment prostate cancer in TRUS images, alleviating the time-consuming work of radiologists to draw the boundary of the lesions and training the neural network on the data that do not have complete annotations. A histologic-proven benchmarking dataset of 102 case images was built and 22 images were randomly selected for evaluation. Some portion of the training images were strong supervised, annotated pixel by pixel. Using the strong supervised images, a deep learning neural network was trained. The rest of the training images with only weak supervision, which is just the location of the lesion, were fed to the trained network to produce the intermediate pixelwise labels for the weak supervised images. Then, we retrained the neural network on the all training images with the original labels and the intermediate labels and fed the training images to the retrained network to produce the refined labels. Comparing the distance of the center of mass of the refined labels and the intermediate labels to the weak supervision location, the closer one replaced the previous label, which could be considered as the label updates. After the label updates, test set images were fed to the retrained network for evaluation. The proposed method shows better result with weak and semi-supervised data than the method using only small portion of strong supervised data, although the improvement may not be as much as when the fully strong supervised dataset is used. In terms of mean intersection over union (mIoU), the proposed method reached about 0.6 when the ratio of the strong supervised data was 40%, about 2% decreased performance compared to that of 100% strong supervised case. The proposed method seems to be able to help to alleviate the time-consuming work of radiologists to draw the boundary of the lesions, and to train the neural network on the data that do not have complete annotations.

Do all patients with newly diagnosed prostate cancer need staging radionuclide bone scan? A retrospective study

PURPOSE

Define a group of patients with newly diagnosed prostate cancer, whose risk of bone metastasis is low enough to omit a bone scan staging study.

MATERIALS AND METHODS

From 2003 to 2009, the medical records of patients who were newly diagnosed with prostate cancer were retrospectively reviewed. The data collected included: age, digital rectal examination, serum prostate specific antigen (PSA), Gleason score, clinical T stage, and bone isotope scan. Patients were divided into two groups according to the results of bone isotope scan; positive group and negative group. A univariate and multivariate binary logistic regression was used to analyze the results.

RESULTS

Of the 106 patients, 98 had a complete data collection and were entered into the study. The median age of the patients was 70.5 years and patients with a positive bone scan was 74 years, significantly higher than for patients with negative scans (69 years) (p=0.02). Bone metastasis was detected in 39 cases (39.7%). In all patients with clinical T1-2 stage, a Gleason score of <8 and PSA≤20 ng/mL, the bone isotope scans were negative. In univariate analysis, PSA (>20 ng/mL) and Gleason score (>7) were independently predictive of positive bone scan, while clinical stage was not.

CONCLUSION

Staging bone scans can be omitted in patients with a PSA level of ≤20 ng/mL, and Gleason score<8. Our results suggest that by considering the Gleason score and PSA, a larger proportion of patients with prostate cancer could avoid a staging bone scan.

Computer-aided prostate cancer detection using texture features and clinical features in ultrasound image

In this paper, we propose a new prostate detection method using multiresolution autocorrelation texture features and clinical features such as location and shape of tumor. With the proposed method, we can detect cancerous tissues efficiently with high specificity (about 90-95%)and high sensitivity (about 92-96%) by the measurement of the number of correctly classified pixels. Multiresolution autocorrelation can detect cancerous tissues efficiently, and clinical knowledge helps to discriminate the cancer region by location and shape of the region and increases specificity. The support vector machine is used to classify tissues based on those features. The proposed method will be helpful in formulating a more reliable diagnosis, increasing diagnosis efficiency.

'Insignificant' prostate cancer on biopsy: pathologic results from subsequent radical prostatectomy

'Insignificant' prostate cancer is defined as disease of virulence insufficient to threaten survival. In this review, which describes nine articles and two abstracts discussing almost 800 cases, we discuss the correlation of such 'insignificant' biopsy findings in the context of subsequent radical prostatectomy data. From our review, minimal disease on biopsy does not reliably predict minimal disease in the subsequent prostatectomy specimen, in terms of the size and grade of tumor, extracapsular extension or positive margins. Thus, reasoned accounting should be made of other data before undertaking a course of radiation therapy as monotherapy, particularly prostate-specific antigen kinetics and potential molecular markers.

Patient selection, cancer control, and complications after salvage local therapy for postradiation prostate-specific antigen failure

Among men who experience prostate-specific antigen (PSA) failure after external beam radiation or brachytherapy (RT), many will harbor occult micrometastases; however, a significant minority will have a true local-only failure and, thus, potentially may benefit from a salvage local therapy. Those most likely to have a local-only failure initially have low-risk disease (PSA < 10 ng/mL, Gleason score < or =6, clinical T1c or T2a tumor status), pretreatment PSA velocity < 2.0 ng/mL per year at the time of initial presentation, interval to PSA failure > 3 years, PSA doubling time > 12 months, negative bone scan and pelvic imaging, and positive rebiopsy. In addition, men with presalvage PSA levels > 10 ng/mL, presalvage T3/T4 disease, or presalvage Gleason scores > or =7 on a rebiopsy sample without significant RT effects are unlikely to be cured by salvage local therapy. Based on a review of all series of post-RT salvage prostatectomy, cryosurgery, and brachytherapy published in English since 1990, morbidity can be substantial. Although urinary incontinence appeared to be greater after salvage prostatectomy (41%) or cryosurgery (36%) than after brachytherapy (6%), patients who received salvage brachytherapy faced a 17% risk of grade 3 or 4 genitourinary complications and a fistula risk that averaged 3.4% across all series. From this review, the authors concluded that prospective randomized studies are needed to determine the relative efficacy of the 3 major local salvage modalities and that additional research is needed to identify factors associated with an increased risk of significant complications to improve patient selection and to augment the benefit/risk ratio associated with attempts to cure local-only recurrences after radiation therapy.

Relapse after radical prostatectomy correlates with preoperative PSA velocity and tumor volume: Results from a screening population

OBJECTIVES

To evaluate, in a screening population, the impact of tumor volume and prostate volume on prostate-specific antigen (PSA) velocity (PSAV) and to find predictors of biochemical failure after radical prostatectomy. Longitudinal PSA changes in men with prostate cancer have been reported to be significantly different from those without prostate cancer.

METHODS

PSAV was assessed in 102 men undergoing radical retropubic prostatectomy. The pathologic findings of specimens obtained at radical retropubic prostatectomy and pelvic lymph node dissection were analyzed separately for all patients.

RESULTS

The median preoperative PSA in the 102 patients was 6.4 ng/mL, the median prostate volume was 32.8 cm3, and the median tumor volume was 1.27 cm3. The PSAV correlated significantly with tumor volume (P <0.05) but not with prostate volume (P = 0.142). The median tumor volume in men with biochemical progression after radical retropubic prostatectomy was 2.55 cm3 versus 0.94 cm3 in men who were free of disease 5 years after surgery. The median PSAV in the year before diagnosis in men with relapse after radical prostatectomy was 1.98 ng/mL/yr versus 1.05 ng/mL/yr in men who had no evidence of disease.

CONCLUSIONS

The results of our study have shown that the main factor contributing to the PSAV in patients with prostate cancer is cancer load and that prostate volume is not significantly associated with the PSAV. Men with a PSAV of more than 2 ng/mL/yr in the year before cancer diagnosis are at a high risk of relapse. The PSAV may be helpful in identifying patients with small tumors and thus increase the detection rate of potentially curable prostate cancers.