BACKGROUND: Prostate cancer (PCa) is the leading male neoplasm in South Africa with an age-standardised incidence rate of 68.0 per 100,000 population in 2018. The Gleason score (GS) is the strongest predictive factor for PCa treatment and is embedded within semi-structured prostate biopsy narrative reports. The manual extraction of the GS is labour-intensive. The objective of our study was to explore the use of text mining techniques to automate the extraction of the GS from irregularly reported text-intensive patient reports.
METHODS: We used the associated Systematized Nomenclature of Medicine clinical terms morphology and topography codes to identify prostate biopsies with a PCa diagnosis for men aged > 30 years between 2006 and 2016 in the Gauteng Province, South Africa. We developed a text mining algorithm to extract the GS from 1000 biopsy reports with a PCa diagnosis from the National Health Laboratory Service database and validated the algorithm using 1000 biopsies from the private sector. The logical steps for the algorithm were data acquisition, pre-processing, feature extraction, feature value representation, feature selection, information extraction, classification, and discovered knowledge. We evaluated the algorithm using precision, recall and F-score. The GS was manually coded by two experts for both datasets. The top five GS were reported, with the remaining scores categorised as \"Other\" for both datasets. The percentage of biopsies with a high-risk GS (≥ 8) was also reported.
RESULTS: The first output reported an F-score of 0.99 that improved to 1.00 after the algorithm was amended (the GS reported in clinical history was ignored). For the validation dataset, an F-score of 0.99 was reported. The most commonly reported GS were 5 + 4 = 9 (17.6%), 3 + 3 = 6 (17.5%), 4 + 3 = 7 (16.4%), 3 + 4 = 7 (14.7%) and 4 + 4 = 8 (14.2%). For the validation dataset, the most commonly reported GS were: (i) 3 + 3 = 6 (37.7%), (ii) 3 + 4 = 7 (19.4%), (iii) 4 + 3 = 7 (14.9%), (iv) 4 + 4 = 8 (10.0%) and (v) 4 + 5 = 9 (7.4%). A high-risk GS was reported for 31.8% compared to 17.4% for the validation dataset.
CONCLUSIONS: We demonstrated reliable extraction of information about GS from narrative text-based patient reports using an in-house developed text mining algorithm. A secondary outcome was that late presentation could be assessed.

The Gleason score is an important grading factor of prostate cancer. Gleason scores can be extracted from pathology report texts using regular expressions, but previously developed programmes have targeted only relatively simple Gleason score expressions. We developed a programme capable of extracting also complex expressions. The programme is relatively easy to adapt to other languages and datasets.
We developed and evaluated our regular expression-based programme using manually processed pathology reports of prostate cancer cases diagnosed in Finland in 2016-2017. Both simple and complex Gleason score expressions were targeted. We measured the performance of our programme using recall, precision, and the F1. The proportion of complex Gleason score expressions was estimated as the complement of the recall when only addition expressions (e.g. \"Gleason 3 + 4\") were targeted.
The detection of values (scores and score components) is based on mandatory keywords before or after the value. The programme favours precision over recall by primarily allowing for lists of optional expressions between keyword-value pairs and only secondarily allowing for arbitrary expressions. The programme is straightforward to adapt to new datasets by modifying the lists of mandatory and optional expressions. The full and addition-only programmes had 92% (95% CI: [90%, 95%]) and 65% ([61%, 70%]) recall and high precision (98% [97%, 99%] and 100% [99%, 100%]), respectively. The estimated proportion of complex Gleason score expressions was 100-65 = 35%.
Even complex Gleason score expressions can be extracted with high recall and precision using regular expressions. We recommend implementing automated Gleason score extraction where possible by adapting our validated programme.

Benign prostate hypertrophy (BPH) could be associated with low urinary symptoms requiring medical treatment: 5-alpha-reductase inhibitors (5-ARI) or ɑ-blockers. Two clinical trials investigating 5-ARI use in prostate cancer (PCa) primary prevention highlighted a potential safety signal with an increased risk of high-grade PCa. Later observational studies failed to show similar results but have some limits. This paper focuses on describing the protocol of the CANARI study and its feasibility, as regards the matching process of two pseudo-anonymous databases. The study concerned patients living in the Brittany region (France) between 2010 and 2013. We designed a case-control study nested within a cohort of men treated by medical drugs licensed for symptomatic BPH between 2010 and 2011. Cases were patients with incident PCa diagnosed between 2012 and 2013 identified through French Health database (SNIIRAM). Gleason score was searched through Brittany pathology laboratories. Controls were patients without PCa diagnosis. Local pathology laboratories database was constituted in Brittany, gathering Gleason scores. No unique identification number is available in France; linkage of SNIIRAM and Brittany pathology laboratories database was made by deterministic matching. We matched 859 cases to Gleason grading (119 had Gleason score ≥8 and 740 had Gleason <8); around 22% of cases received 5-ARI and 78% α-blockers or phytotherapy. The CANARI study investigated in a population of men treated for BPH the risk of PCa with 5-ARI, according to Gleason grade thanks to SNIIRAM database enriched by local pathological results.

In utero and early-life exposures are suspected to modulate the risk of prostate cancer. This study examines the influence of certain perinatal and childhood-related factors on prostate cancer risk overall and by Gleason score at biopsy.
MCC-Spain is a multicase-control study where 1088 histologically-confirmed incident prostate cancer cases (aged 42-85years) and 1345 population-based controls (aged 38-85years), frequency matched by age and province of recruitment, were recruited in 7 Spanish provinces. Self-reported perinatal and childhood-related characteristics were directly surveyed by trained staff. The association with prostate cancer risk, globally and according to Gleason score at biopsy, was evaluated using logistic and multinomial regression mixed models, adjusting for age, family history of prostate cancer, educational level and body mass index one year before the interview, and including the province as a random effect term.
Most perinatal factors were not related to prostate cancer risk, with the exception of middle-high socioeconomic level at birth (OR for high grade tumors=1.36; 95%CI=1.09-1.68). Regarding puberty, risk rose by 6% for each year of delayed onset (OR=1.06; 95%CI=1.01-1.10; p trend=0.016), with a clear excess of risk in men who reached puberty after age 15 (OR:1.35; 95%CI=1.08-1.68). A borderline significant positive association with prepubertal height was also observed (p trend=0.094).
Some exposures experienced in utero and during adolescence, when the prostate is still maturing, might be relevant for prostate cancer risk in adulthood.

It remains unanswered whether and how intraoperative frozen section analysis contributes to the surgical margin status on radical prostatectomy specimens. We aimed to determine whether frozen section analysis during radical prostatectomy reduces the incidence of positive surgical margins. We retrospectively analyzed a consecutive series of patients undergoing robot-assisted laparoscopic radical prostatectomy performed at our institution between 2004 and 2011. We identified 2608 cases, including 1128 (43.3%) where intraoperative frozen section analysis was performed to assess surgical margins. Of the cases with positive (n = 60; 5.3%)/negative (n = 1029; 91.2%)/atypical or indeterminate (n = 39; 3.5%) frozen section analyses, 22 (36.7%)/83 (8.1%)/4 (10.3%) were found to have positive surgical margins on radical prostatectomy specimens, respectively. Thus, 109 (9.7%) of 1128 cases with frozen section analysis had positive surgical margins, compared with 163 (11.0%) of 1480 cases with no frozen section analysis (P = .264). When the patients were subgrouped by histopathologic characteristics, frozen section analysis led to a considerable reduction in the rate of positive surgical margins in cases with biopsy Gleason score 7 (12.4% → 8.7%; P = .087)/8 (28.6% → 16.3%; P = .048)/≥7 (15.3% → 10.1%; P = .012) tumor or pT3b (36.6% → 23.2%; P = .075)/≥pT3b (38.1% → 25.4%; P = .091) disease. Multivariate analysis further revealed that performing frozen section analysis in biopsy Gleason score 7 or higher tumors was an independent predictor of negative surgical margins (odds ratio, 0.61; P = .018). In addition, frozen section analysis of the distal urethra or apex of the prostate (7.5%, P = .035) as well as multiple negative frozen section analyses (≥2: 6.2%, P = .001; ≥4: 2.2%, P = .007) correlated with significantly lower rates of positive surgical margin, compared with no frozen section analysis. Overall, intraoperative frozen section analysis did not dramatically change surgical margin status of radical prostatectomy. Nonetheless, it could be useful in preventing incomplete tumor resection, especially in men with high-grade (Gleason score ≥7) tumor at the apex.