Impact of the COVID-19 outbreak on prostate cancer care in the Netherlands.

INTRODUCTION: The COVID-19 outbreak has affected care for non-COVID diseases like cancer. We evaluated the impact of the COVID-19 outbreak on prostate cancer care in the Netherlands. METHODS: Prostate cancer diagnoses per month in 2020-2021 versus 2018-2019 were compared based on preliminary data of the Netherlands Cancer Registry (NCR) and nationwide pathology network. Detailed data was retrieved from the NCR for the cohorts diagnosed from March-May 2020 (first COVID-19 wave) and March-May 2018-2019 (reference). Changes in number of diagnoses, age, disease stage and first-line treatment were compared. RESULTS: An initial decline of 17% in prostate cancer diagnoses during the first COVID-19 wave was observed. From May onwards the number of diagnoses started to restore to approximately 95% of the expected number by the end of 2020. Stage at diagnosis remainedstable over time. In low-risk localised prostate cancer radical prostatectomy was conducted more often in week 9-12 (21% versus 12% in the reference period; OR=1.9, 95% CI; 1.2-3.1) and less active surveillance was applied (67% versus 78%; OR=0.6, 95% CI; 0.4-0.9). In the intermediate-risk group, a similar change was observed in week 13-16. Radical prostatectomy volumes in 2020 were comparable to 2018-2019. CONCLUSION: During the first COVID-19 wave the number of prostate cancer diagnoses declined. In the second half of 2020 this largely restored although the number remained lower than expected. Changes in treatment were temporary and compliant with adapted guidelines. Although delayed diagnoses could result in a less favourable stage distribution, possibly affecting survival, this seems not very likely.

Automated detection of cribriform growth patterns in prostate histology images.

Cribriform growth patterns in prostate carcinoma are associated with poor prognosis. We aimed to introduce a deep learning method to detect such patterns automatically. To do so, convolutional neural network was trained to detect cribriform growth patterns on 128 prostate needle biopsies. Ensemble learning taking into account other tumor growth patterns during training was used to cope with heterogeneous and limited tumor tissue occurrences. ROC and FROC analyses were applied to assess network performance regarding detection of biopsies harboring cribriform growth pattern. The ROC analysis yielded a mean area under the curve up to 0.81. FROC analysis demonstrated a sensitivity of 0.9 for regions larger than [Formula: see text] with on average 7.5 false positives. To benchmark method performance for intra-observer annotation variability, false positive and negative detections were re-evaluated by the pathologists. Pathologists considered 9% of the false positive regions as cribriform, and 11% as possibly cribriform; 44% of the false negative regions were not annotated as cribriform. As a final experiment, the network was also applied on a dataset of 60 biopsy regions annotated by 23 pathologists. With the cut-off reaching highest sensitivity, all images annotated as cribriform by at least 7/23 of the pathologists, were all detected as cribriform by the network and 9/60 of the images were detected as cribriform whereas no pathologist labelled them as such. In conclusion, the proposed deep learning method has high sensitivity for detecting cribriform growth patterns at the expense of a limited number of false positives. It can detect cribriform regions that are labelled as such by at least a minority of pathologists. Therefore, it could assist clinical decision making by suggesting suspicious regions.