Utilization of machine learning methods for prediction of acute and late rectal toxicity due to curative prostate radiotherapy.

OBJECTIVE: Rectal toxicity is one of the primary dose-limiting side effects of prostate cancer radiotherapy, and consequential impairment on quality of life in these patients with long survival is an important problem. In this study, we aimed to evaluate the possibility of predicting rectal toxicity with artificial intelligence model which was including certain dosimetric parameters. MATERIALS AND METHODS: One hundred and thirty-seven patients with a diagnosis of prostate cancer who received curative radiotherapy for prostate +/- pelvic lymphatics were included in the study. The association of the clinical data and dosimetric data between early and late rectal toxicity reported during follow-up was evaluated. The sample size was increased to 274 patients by synthetic data generation method. To determine suitable models, 15 models were studied with machine learning algorithms using Python 2.3, Pycaret library. Random forest classifier was used with to detect active variables. RESULTS: The area under the curve and accuracy were found to be 0.89-0.97 and 95%-99%, respectively, with machine learning algorithms. The sensitivity values for acute and toxicity were found to be 0.95 and 0.99, respectively. CONCLUSION: Early or late rectal toxicity can be predicted with a high probability via dosimetric and physical data and machine learning algorithms of patients who underwent prostate +/- pelvic radiotherapy. The fact that rectal toxicity can be predicted before treatment, which may result in limiting the dose and duration of treatment, makes us think that artificial intelligence can enter our daily practice in a short time in this sense.

The expression of stanniocalcin-1, estrogen receptor and progesterone receptor in endometrioid endometrial cancer.

Objective: To evaluate the expression of stanniocalcin-1 (STC-1) and to investigate the correlation of STC-1 with expression of estrogen receptor (ER), progesterone receptor (PR) and clinical parameters, histopathological findings and prognostic factors in endometrioid endometrial cancer (EEC). Materials and Methods: In this retrospective study, STC-1 (cytoplasmic), ER (nuclear), and PR (nuclear) stainings were applied to tissue microarray sections of 89 EEC, 27 endometrial intraepithelial neoplasia (EIN), and 21 normal endometrium (NE). Prognostic factors such as age, tumor size, depth of myometrial invasion, lymphovascular invasion, perineural invasion, and lymph node metastasis were compared with the expression of these markers. Results: ER showed significantly higher positivity in grade 1 EEC. PR expression was also higher in grade 1 EEC, but these findings were not statistically significant. Strong expression of STC-1 was observed in EIN and EECs compared with NE. STC-1 showed low staining in the NE, and high staining was also noted in the EIN foci adjacent to the NE. STC-1 expression was positively correlated with grade 1 EECs. Conclusion: STC-1 expression was positively correlated with low histologic grade in EECs. STC-1 can be used for distinguishing low-grade endometrioid tumors and high -grade endometrioid tumors in curretage specimens. Since STC-1 is related to well differentiated tumors, it can also be regarded as a good prognostic factor in EECs.

The prognostic value of decrease in prognostic nutritional index in stage III non-small cell lung cancer patients during curative thoracic radiotherapy.

BACKGROUND: Curative thoracic radiotherapy (CTRT) with concurrent chemotherapy has been considered as standard treatment approach for stage-III non-small cell lung cancer (NSCLC). The hematological and esophageal toxicities that have been encountered during CTRT would affect the immunonutritional status of the patients. The aim of this study is to evaluate the prognostic value of the change in pre- and post-treatment prognostic nutritional index (PNI) in stage-III NSCLC patients. METHODS: Eighty seven consecutive stage III NSCLC patients' data were collected. Pre-radiotherapy (RT) and post-RT PNI values were calculated and the impact of prognostic value of PNI change on overall survival (OS) was evaluated by univariate and multivariate Cox regression analyses. A cutoff value of PNI change was obtained by receiver operator characteristic (ROC) curve analysis. RESULTS: The cutoff value was found to be a 22% decrease in PNI by ROC curve analysis in terms of effect on OS. The median OS of low and high PNI decrease groups were 22.5 and 16.5 months respectively (P = 0,001). In univariate and multivariate analyses PNI decrease of ≥ 22% was found to be an independent poor prognostic factor for OS (P = 0.012) and hazard ratio (95% confidence interval)= 2.05 (1.16-3.62). CONCLUSION: The PNI change would be a convenient parameter to assess the immunonutritional status of the patient at the end of CTRT. A decrease of more than 22% of PNI value may predict poor prognosis.

Correlation of 18F-FDG/PET SUVmax, SUVmean, MTV, and TLG with HIF-1α in Patients with Colorectal Cancer

Objectives: Post-hypoxia hypoxia-inducible factor (HIF)-1α activation plays a vital role in colorectal cancer (CRC) angiogenesis. Although glucose metabolism is induced in some cancer types via HIF-1α, the prognostic significance of HIF-1α in CRC and its correlation with 18fluorinefluorodeoxyglucose (18F-FDG) uptake in positron emission tomography (PET) remain controversial. This study aims to investigate the association between 18F-FDG/PET parameters and HIF-1α expression in CRC. Methods: Thirty-six histopathologically confirmed patients with CRC who had 18F-FDG/PET scans before surgery were enrolled in the study. The correlations between the maximum standardized uptake value (SUVmax), SUVmean, metabolic tumor volume (MTV), total lesion glycolysis, HIF-1α overexpression, and histopathological features were evaluated. Results: The tumor location, tumor diameter, perineural invasion, lymphovascular invasion, T and N stage were not significantly correlated with HIF-1α overexpression. In contrast, the tumor differentiation was negatively correlated with HIF-1α expression (r=-0.332, p=0.048). None of the 18F-FDG/PET parameters was significantly correlated with HIF-1α overexpression. A significant relationship was found between tumor differentiation, tumor necrosis percentage, and MTV (p=0.030, p=0.020). Conclusion: The expected association between HIF-1α overexpression and 18F-FDG/PET parameters was not found in this study. However, there was a relationship between MTV, tumor differentiation, and tumor necrosis percentage. Hence, further studies are required to predict the pathological and prognostic courses of CRC using a diagnostic 18F-FDG/PET evaluation.