The Correlation Between Platelet Count and Survival in Prostate Cancer

Improving platelet-RNA-based diagnostics: a comparative analysis of machine learning models for cancer detection and multiclass classification

Liquid biopsy demonstrates excellent potential in patient management by providing a minimally invasive and cost-effective approach to detecting and monitoring cancer, even at its early stages. Due to the complexity of liquid biopsy data, machine-learning techniques are increasingly gaining attention in sample analysis, especially for multidimensional data such as RNA expression profiles. Yet, there is no agreement in the community on which methods are the most effective or how to process the data. To circumvent this, we performed a large-scale study using various machine-learning techniques. First, we took a closer look at existing datasets and filtered out some patients to assert data collection quality. The final data collection included platelet RNA samples acquired from 1397 cancer patients (17 types of cancer) and 354 asymptomatic, presumed healthy, donors. Then, we assessed an array of different machine-learning models and techniques (e.g., feature selection of RNA transcripts) in pan-cancer detection and multiclass classification. Our results show that simple logistic regression performs the best, reaching a 68% cancer detection rate at a 99% specificity level, and multiclass classification accuracy of 79.38% when distinguishing between five cancer types. In summary, by revisiting classical machine-learning models, we have exceeded the previously used method by 5% and 9.65% in cancer detection and multiclass classification, respectively. To ease further research, we open-source our code and data processing pipelines (https://gitlab.com/jopekmaksym/improving-platelet-rna-based-diagnostics), which we hope will serve the community as a strong baseline.

Correlation Between Long-Term Acetylsalicylic Acid Use and Prostate Cancer Screening with PSA. Should We Reduce the PSA Cut-off for Patients in Chronic Therapy? A Multicenter Study

Purpose

To evaluate the prostate cancer (PCa) detection rate in men with chronic use of Aspirin and to compare it with the detection rate of non-users.

Patients and Methods

Prospectively maintained database regarding patients undergoing prostate biopsy over the last 10 years in five institutions. Patients were divided into two groups according to their exposure to Aspirin. We relied on multivariable linear and logistic regression models to test whether Aspirin administration was associated with lower PSA values at prostate biopsy, higher PCa diagnosis, and higher Gleason Grade Grouping (GGG) at biopsy.

Results

Were identified 1059 patients, of whom 803 (76%) did not take Aspirin vs 256 (24%) were taking it. In multivariable log-linear regression analysis, Aspirin administration was associated with lower PSA levels (OR 0.83, 95% CI 0.71–0.97, p = 0.01), after controlling for age, prostate volume, smoking history, associated inflammation at prostate biopsy, presence of PCa at biopsy, and GGG. In multivariable logistic regression analysis, Aspirin administration was not found to be a predictor of PCa at prostate biopsy (OR 1.40, 95% CI 0.82–2.40, p = 0.21) after controlling for age, PSA, smoking history, prostate volume, findings at digital rectal examination and the number of biopsy cores. In patients with PCa at prostate biopsy (n = 516), Aspirin administration was found to predict higher GGG (OR 2.24, 95% CI 1.01–4.87, p = 0.04).

Conclusion

Aspirin administration was found to be a predictor of more aggressive GGG. These findings suggest that a lower PSA threshold should be considered in patients taking Aspirin, as, despite low PSA levels, they might harbour aggressive PCa.

The prognostic role of pre-cystectomy thrombocytosis in invasive bladder cancer

PURPOSE

We aim to evaluate the impact of preoperative thrombocytosis on oncological outcomes in patients with bladder cancer (BC) who undergo radical cystectomy (RC).

METHODS

Retrospective data collection of 1092 patients managed by RC for BC from 2 tertiary-care centers was performed. Elevated platelet count (PLT) was defined as > 450 × 10⁹/L. Univariable and multivariable logistic regression analyses were used to investigate the impact of thrombocytosis on oncological outcomes. These outcomes were also compared using Kaplan-Meier survival analysis.

RESULTS

The median follow-up was 50 months (32-64 months). Thrombocytosis was detected in 18.6% of the patients. The 3-year cancer-specific survival (CSS) for patients with normal PLT count was 92% which was higher than those with elevated PLT count (55%, P < 0.001). Similar results were found for the 6-year CSS with 82% for the no thrombocytosis group and 27% for the thrombocytosis group. Thrombocytosis was still significantly associated with poor prognosis for overall survival and recurrence-free survival (P < 0.001). In the multivariate analysis, CSS was significantly lower in patients with thrombocytosis (HR = 1.71, 95% CI = 1.22-2.39, P = 0.002). Patients with elevated PLT counts were also significantly more likely to receive adjuvant chemotherapy, to have a T stage > pT2b (P = 0.024), to have a positive lymph node, to have variant histology and positive resection margins, and to have concomitant carcinoma in situ (CIS) on final pathology (all P < 0.001).

CONCLUSIONS

Preoperative thrombocytosis was valuable for predicting the oncological outcomes of patients undergoing RC for BC.