Deep Neural Networks Outperform the CAPRA Score in Predicting Biochemical Recurrence After Prostatectomy

Evaluation of machine learning strategies for imaging confirmed prostate cancer recurrence prediction on electronic health records

BACKGROUND

The main screening parameter to monitor prostate cancer recurrence (PCR) after primary treatment is the serum concentration of prostate-specific antigen (PSA). In recent years, Ga-68-PSMA PET/CT has become an important method for additional diagnostics in patients with biochemical recurrence.

PURPOSE

While Ga-68-PSMA PET/CT performs better, it is an expensive, invasive, and time-consuming examination. Therefore, in this study, we aim to employ modern multivariate Machine Learning (ML) methods on electronic health records (EHR) of prostate cancer patients to improve the prediction of imaging confirmed PCR (IPCR).

METHODS

We retrospectively analyzed the clinical information of 272 patients, who were examined using Ga-68-PSMA PET/CT. The PSA values ranged from 0 ng/mL to 2270.38 ng/mL with a median PSA level at 1.79 ng/mL. We performed a descriptive analysis using Logistic Regression. Additionally, we evaluated the predictive performance of Logistic Regression, Support Vector Machine, Gradient Boosting, and Random Forest. Finally, we assessed the importance of all features using Ensemble Feature Selection (EFS).

RESULTS

The descriptive analysis found significant associations between IPCR and logarithmic PSA values as well as between IPCR and performed hormonal therapy. Our models were able to predict IPCR with an AUC score of 0.78 ± 0.13 (mean ± standard deviation) and a sensitivity of 0.997 ± 0.01. Features such as PSA, PSA doubling time, PSA velocity, hormonal therapy, radiation treatment, and injected activity show high importance for IPCR prediction using EFS.

CONCLUSION

This study demonstrates the potential of employing a multitude of parameters into multivariate ML models to improve identification of non-recurring patients compared to the current focus on the main screening parameter (PSA). We showed that ML models are able to predict IPCR, detectable by Ga-68-PSMA PET/CT, and thereby pave the way for optimized early imaging and treatment.

Post-Operative Radiotherapy in Prostate Cancer: Is It Time for a Belt and Braces Approach?

Approximately 30% of patients treated with radical prostatectomy (RP) for prostate cancers experience biochemical recurrence (BCR). Post-operative radiation therapy (RT) can be either offered immediately after the surgery in case of aggressive pathological features or proposed early if BCR occurs. Until recently, little data were available regarding the optimal RT timing, protocol, volumes to treat, and the benefit of adding androgen deprivation therapies to post-operative RT. In this review, we aim to pragmatically discuss current literature data on these points. Early salvage RT appears to be the optimal post-operative approach, improving oncological outcomes especially with low prostate-specific antigen (PSA) levels, as well as sparing several unnecessary adjuvant treatments. The standard RT dose is still 64–66 Gy to the prostate bed in conventional fractionation, but hypofractionation protocols are emerging pending on late toxicity data. Several scientific societies have published contouring atlases, even though they are heterogeneous and deserve future consensus. During salvage RT, the inclusion of pelvic lymph nodes is also controversial, but preliminary data show a possible benefit for PSA > 0.34 ng/ml at the cost of increased hematological side effects. Concomitant ADT and its duration are also discussed, possibly advantageous (at least in terms of metastasis-free survival) for PSA rates over 0.6 ng/ml, taking into account life expectancy and cardiovascular comorbidities. Intensified regimens, for instance, with new-generation hormone therapies, could further improve outcomes in carefully selected patients. Finally, recent advances in molecular imaging, as well as upcoming breakthroughs in genomics and artificial intelligence tools, could soon reshuffle the cards of the current therapeutic strategy.

Pharmacophore based in silico study with laboratory verification-environmental explanation of prostate cancer recurrence

The prediction of harmful effects can guide research on the health hazards of environmental pollutants. The development of a computer-aided drug design in pharmacological technology and omics database remarkably facilitates the prediction of the possible harmful effects of hazardous substances. In this study, the pharmacophore target database based on molecular structure served as a bridge between pollutants and genes and combined with the omics database and molecular pathway enrichment technology to predict the potential prostatic cancer-promoting effect of dibutyl phthalate. Cell experiments and gene expression were carried out to verify the previous prediction, and the characteristics of harmful effects were further explored. Low concentrations of dibutyl phthalate may have androgen-independent prostate cancer-promoting effects, which may put patients receiving androgen deprivation therapy in danger. This study suggests the potential negative effects of phthalates on prostate cancer and a method for predicting harmful effects on the basis of pharmacology technology.

Single-Center Experience of Focal Thermo-Ablative Therapy After Pelvic Radiotherapy for In-Field Prostate Cancer Oligo-Recurrence

PURPOSE

In-field prostate cancer (PCa) oligo-recurrence after pelvic radiotherapy is a challenging situation for which metastasis-directed treatments may be beneficial, but options for focal therapies are scarce.

METHODS

We retrospectively reviewed data for patients with three or less in-field oligo-recurrent nodal, bone and/or locally recurrent (prostate, seminal vesicles, or prostatic bed) PCa lesions after radiation therapy, identified with molecular imaging (PET and/or MRI) and treated by focal ablative therapy (cryotherapy or radiofrequency) at the Institut Bergonié between 2012 and 2020. Chosen endpoints were the post-procedure PSA response (partially defined as a >50% reduction, complete as a PSA <0.05 ng/ml), progression-free survival (PFS) defined as either a biochemical relapse (defined as a rise >25% of the Nadir and above 2 ng/ml), radiological relapse (on any imaging technique), decision of treatment modification (hormonotherapy initiation or line change) or death, and tolerance.

RESULTS

Forty-three patients were included. Diagnostic imaging was mostly 18F-Choline positron emission tomography/computerized tomography (PET/CT) (75.0%), prostate specific membrane antigen (PSMA) PET/CT (9.1%) or a combination of pelvic magnetic resonance imaging (MRI), CT, and 99 mTc-bone scintigraphy (11.4%). PSA response was observed in 41.9% patients (partial in 30.3%, complete in 11.6%). In the hormone-sensitive exclusive focal ablation group (n = 31), partial and complete PSA responses were 32.3 and 12.9% respectively. Early local control (absence of visible residual active target) on the post-procedure imaging was achieved with 87.5% success. After a median follow-up of 30 months (IQR 13.3-56.8), the median PFS was 9 months overall (95% CI, 6-17), and 17 months (95% CI, 11-NA) for PSA responders. Complications occurred in 11.4% patients, with only one grade IIIb Dindo-Clavien event (uretral stenosis requiring endoscopic uretrotomy).

CONCLUSION

In PCa patients showing in-field oligo-recurrence after pelvic radiotherapy, focal ablative treatment is a feasible option, possibly delaying a systemic treatment initiation or modification. These invasive strategies should preferably be performed in expert centers and discussed along other available focal strategies in multi-disciplinary meetings.